Collapse, War & Reconstruction in Rwanda (PDF)

Working Paper No. 28 COLLAPSE, WAR AND RECONSTRUCTION IN RWANDA AN ANALYTICAL NARRATIVE ON STATE-MAKING...

Working Paper No. 28 COLLAPSE, WAR AND RECONSTRUCTION IN RWANDA AN ANALYTICAL NARRATIVE ON STATE-MAKING Frederick Golooba-Mutebi Makerere Institute of Social Research Makerere University February 2008 Copyright © F. Golooba Mutebi, 2008 Although every effort is made to ensure the accuracy and reliability of material published in this Working Paper, the Crisis States Research Centre and LSE accept no responsibility for the veracity of claims or accuracy of information provided by contributors. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means without the prior permission in writing of the publisher nor be issued to the public or circulated in any form other than that in which it is published. Requests for permission to reproduce this Working Paper, of any part thereof, should be sent to: The Editor, Crisis States Research Centre, DESTIN, LSE, Houghton Street, London WC2A 2AE. Crisis States Working Papers Series No.2 ISSN 1749-1797 (print) ISSN 1749-1800 (online)  1 Crisis States Research Centre Collapse, War & Reconstruction in Rwanda: an analytical narrative on state-making Frederick Golooba-Mutebi1 Makerere Institute of Social Research Abstract Rwanda entered independence following a transition marked by violent internecine conflict. The conflict was stoked by the departing colonial rulers as they sought to place control of the levers of state in the hands of an ethnic majority, which they had hitherto marginalised in favour of a minority they now sought to exclude. It carried on into the country’s post-colonial politics. For nearly three decades Rwanda’s post- colonial rulers presided over an ethnocracy that perpetuated the negative colonial legacy of ethnic division. They systematically practiced a politics of exclusion and repression that placed the country’s long-term stability under threat, eventually led to civil war, and culminated in the genocide of 1994. After the genocide and the defeat and overthrow of the ancien regime of ethnic supremacists, the new ruling elite - most of whom had spent nearly three decades in exile or been born there - embarked on re-building a collapsed state and re-ordering the country’s politics. The last fourteen years have witnessed deliberate efforts to re-orient the country away from three decades of politics of division and exclusion under the First and Second Republics, towards a system which privileges national reconciliation and unity, equity, and inclusion. This paper examines developments in post-1994 Rwanda against the background of pre-1994 politics and society, and the factors that led to and facilitated the war that culminated in the genocide and eventual overthrow of the Second Republic. It provides insights into the efforts and achievements made by the new ruling elites in pursuit of long-term peace and stability. A great deal, however, remains inadequately explored, including political organisation and the role of political parties, economic reform and management, and the reform and management of the security sector, all of which are the focus of on-going research. 1 I acknowledge the excellent research assistance of Yvonne Habiyonizeye.  2 Introduction Rwanda gained independence following tragic circumstances. Although Rwandans did not have to wage war to win independence, the years leading to self-rule were marred with conflict. In line with the history of colonialism and anti-colonial struggles in other parts of Africa and elsewhere, Rwandans were expected to seek to free themselves from domination by their colonial masters. Rwanda’s struggle for independence, however, was two pronged. On the one hand, nationalists among Hutu and Tutsi elites were united in seeking to end Belgian rule. On the other, Hutu nationalists sought to ‘liberate’ the majority Hutu population not only from Belgian rule, but also, with the support of the colonial administration, from what they saw as colonialism by the Tutsi minority. It is important to emphasise that inter-ethnic conflict in Rwanda was concentrated in elite circles on both sides, as it was amongst elites that the struggle for ethnic supremacy was located. Ordinary Hutu and Tutsi featured largely as either innocent victims or attackers mobilised and encouraged by elites seeking to acquire or monopolise power and the privileges that went with holding it. Although it is not as widely analysed as inter-ethnic conflict, intra-ethnic rivalry within the wider Hutu population also had its epicentre within elite circles divided along north-south axes (Strauss 2006; Jefremovas 2002; Munyarugerero, 2003). For most of the colonial period the Tutsi were favoured by the Belgians and had been designated by colonial-era historians and anthropologists as foreign elements who, prior to colonial rule, had migrated into the area and subjugated the indigenous Bantu Hutu and Twa. Modern historians have demonstrated that the history of the area was actually much more complex than this (Vansina 2004; Chretien 2003). During most of the pre-second world war period, however, the Belgians administered Rwanda in line with their version of its political and social history. One of the earliest measures they took in their state-building project was to exclude the Hutu, whom they judged to be less intelligent than the Tutsi and incapable of exercising leadership, from chieftainship. They also placed Tutsi chiefs in areas where, prior to colonial rule, Hutu chiefs had been in charge. Chiefs became the immediate face of colonial rule and were responsible for collecting (punitive) taxes and fees, enforcing the building up of food reserves against famine, the cultivation and marketing of cash crops, environmental conservation, organising unpaid community work and other tasks the population found onerous, and meting out punishment to those who did not comply. Whilst in executing their tasks they acted as representatives of the state and faced demotion or dismissal if they failed, the Hutu population who bore the brunt of the exactions alongside poor, ordinary Tutsi (Semujanga 2003), saw Tutsi chiefs through ethnic lenses and therefore as representatives of the Tutsi community rather than those of the colonial administration (Jefremovas 2002). As Strauss (2006) argues with regard to the targeting of Tutsi civilians during the successive episodes of communal violence, “all Tutsis stood in for the actions of a few” (p. 199). Consequently the chiefs – and Tutsis generally – came to represent tyranny and, more so than the colonial administration they served, became objects of popular resentment. The Tutsi monopoly over administrative posts ensured disproportionate access to education and training opportunities, as public and church schools and training institutions enlisted mostly Tutsi students (Baranyizigiye 1999). After completing  3 their studies, Tutsis had greater access to employment opportunities than their Hutu counterparts because of their connections in the administration. This further sharpened inequality and the sense of grievance among Hutu (Nkundabagenzi 1961). Colonial rule therefore created a wide political and socio-economic chasm between Tutsi and Hutu elites. Ordinary Hutu and Tutsi, however, led similar lives amidst poverty, deprivation, and landlessness (Jefremovas 2002; also Semujanga 2003). Hutu elites, however, seized upon the theory propounded by the colonial authorities, the church2 and ethnologists that portrayed the Tutsi as foreign invaders who had to be uprooted. The outcome was a series of bloody political crises that began in the lead up to independence and came to characterise post-colonial Rwanda during the 1960s and early 1970s, and culminated in the 1994 genocide. Rather than take measures to dampen the inter-ethnic animosity that had developed largely as a result of their divisive policies, the Belgian authorities, by acts of omission and commission, deepened the divisions. During the final years of colonial rule, due partly to tensions between Tutsi elites and the colonial administration as well as the Catholic Church, they embarked on promoting the Hutu to take over the post-colonial state by curtailing Tutsi dominance and influence (Strauss 2006; Semujanga, 2003). By the mid-1950s Tutsi hegemony was on the decrease and Hutu supremacy on the rise. By the late 1950s Tutsi hegemony had effectively ended (Lugan, 1997; Strauss 2006). Nonetheless, the decline of their dominance belied their determination to continue playing an important role in the country’s evolution towards independence. The birth of political parties Four political parties emerged in 1959 when the colonial authorities finally acquiesced to their formation. The four main parties that formed as a result were constituted mainly along ethnic lines. Tutsi traditionalists and monarchists founded the Rwanda National Union (UNAR)3 in May 1959, and from its inception campaigned against all forms of discrimination. The party’s official position was that “ethnicity was not a relevant political factor and that democracy meant independence from European rule”. The UNAR was, however, distinguished by its intolerance towards its Hutu and Tutsi opponents whom it denounced as traitors, enemies, and snakes and called for their extermination (Strauss 2006:178). UNAR advocated self- rule by 1960 and total independence by 1962. The Association for the Social Promotion of the Masses (APROSOMA),4 which had been founded earlier in 1957 as a civic group and had been pre-occupied with issues of ethnicity, also became a party in 1959. Its partisans saw Rwanda as a colony of the “Ethiopian Tutsi” and argued that the struggle of the Hutu was a struggle against “Tutsi colonialism”. The party opposed all forms of co-habitation between Hutu and Tutsi, and was the first party to call publicly for the latter’s extermination (Murego 1975). Also founded in 1957 by a Catholic catechist and the future first president of Rwanda, Grégoire Kayibanda, ,was the Muhutu Social Movement (MSM).5 The Belgians encouraged him to convert this 2 See, for example, Gatwa, 2005; Linden, 1999. 3 Union Nationale Rwandais. 4 Association pour la Promotion Sociale de la Masse. 5 Mouvement Social Muhutu.  4 into a political party, the Party of the Movement for Hutu Emancipation6 (PARMEHUTU) (Kakwenzire and Kamukama 2000). The Party’s clearly stated objective was “Hutu emancipation”. Tutsi moderates founded the Rwanda 7 Democratic Assembly (RADER). The party was opposed by extremist Hutu who saw it as the monarchy’s Trojan horse and UNAR’s Tutsi extremists who saw its members as traitors. Over time, the party, the liberalism it represented, and Hutu and Tutsi moderates who called for political compromise and gradual change, were confined to the political margins as extremists took centre stage (Strauss 2006). The impact of party politics The birth of party politics brought Hutu-Tutsi animosity squarely into the public domain and turned ethnicity into a major arena for political mobilisation. The ensuing high-voltage politicking, by then beyond the colonial authorities’ capacity to control, erupted into inter-ethnic violence in November 1959. The violence was ignited by rumours of attacks against Hutu politicians by young Tutsi militants belonging to the UNAR party. Hutu political elites then mobilised mostly Hutu youths to attack members of the Tutsi community and their property in the central and north-western parts of the country. In retaliation, supporters of the UNAR, both Tutsi and Hutu, beat, arrested, tortured and assassinated political leaders from opposing parties, including Tutsi accused of disloyalty towards the monarchy (Strauss 2006: 178-181).8 Colonial administrators did not intervene immediately to restore order but simply stood aside as the violence spread. Generally, however, they supported the ‘Hutu insurgency’, with some making no secret of their desire to see Tutsi dominance end (Rugengamanzi 1999; Karekezi 1982; Munyarugerero 2003). Nonetheless, it was only after military intervention by the colonial administration, which targeted Tutsi authorities more than Hutu, that the violence ended.9 After order had been restored, Tutsi chiefs who had been killed, maimed, or driven into exile were replaced by Hutu.10 As a consequence of the partisan manner in which the colonial authorities had reacted to the violence, Hutu political elites seem to have concluded that they could not only incite violence with impunity, but also benefit from it. It was also during this time that the ideology of ‘Hutu Power’ was born, as Hutu elites sought to supplant Tutsi dominance and replace it with its Hutu variant. By 1960 it was clear that the colonial authorities who had envisaged setting up democratic institutions before granting Rwanda independence would not achieve that objective. They, nonetheless, carried on with preparations to grant the country independence. 6 Parti du Mouvement de l’Emancipation Hutu. It later became the Party of the Movement for Hutu Emancipation – Rwanda Democratic Movement (PARMEHUTU-MDR). 7 Rassemblement Democratique Rwandais 8 According to Strauss (2006) the attacks against members of the Tutsi community by ordinary Hutu were not necessarily motivated by ethnic hatred. Some apparently believed that the king had ordered the attacks supposedly because he was being held prisoner by Tutsi elements. Others believed they were acting on behalf of the colonial administration and even sought paraffin from Belgian officials to help with the torching of Tutsi houses. 9 For example, the most active UNAR supporters among the Tutsi chiefs were dismissed from their posts. Other Tutsi authorities were arrested for crimes they had allegedly committed (Strauss 2006: 181; also Semujanga 2003). 10 See Strauss (2006) for details of the sweeping restructuring of the ethnic composition of Rwanda’s local administration (pp. 181-182).  5 In 1961 the first local elections were held in an atmosphere of calm. With the support of the colonial administration, Hutu-led parties of which PARMEHUTU was the most significant achieved resounding victories. On the other hand Tutsi-led parties each received less than ten percent of the votes cast. PARMEHUTU won over seventy percent of the total vote under a new name, PARMEHUTU/MDR (Democratic Republican Movement), after campaigning on an anti-monarchy platform. Despite their best efforts, the two Tutsi-led parties, UNAR and RADER, were unable to forge an alliance. In the same year, following a plan hatched by the Hutu leaders and the colonial authorities, the monarchy was abolished and Rwanda declared a republic in what became known as the Gitarama coup.11 Legislative elections held the same year were preceded by efforts to narrow the Hutu- Tutsi divide. UNAR and RADER, with the support of the United Nations (which was worried about the divisions in the colony) had sought to have legislative elections held after the ‘total pacification’ of the country. By contrast the Hutu parties, with the support of the colonial authorities, wanted the elections to go ahead to avoid slowing down the march towards independence. In theory at least, both the government and assembly created by the coup were suspended before the elections. In order to level the ground for the contending parties the colonial administration assumed the role of a neutral government. This claim to neutrality, however, was undermined by the use of state resources by party-affiliated government officers and the overt support local officials showed for their parties. The elections and the referendum kamarampaka (the arbiter) on the question of the monarchy took place on 25 September 1961. The election campaigns were marred by anti-Tutsi violence entailing intimidation through acts of arson, murder, internal displacement, and exile. Once again the Belgian authorities took sides. PARMEHUTU scored another resounding victory, winning over seventy-seven percent of the vote. Second was UNAR with less than twenty percent of the vote. The referendum saw eighty percent of voters opting for abolition of the monarchy. Thus Rwanda ceased being a Tutsi monarchy and became a “one-party racial dictatorship” (Strauss 2006: 182). The newly-elected legislative assembly elected a president to also act as head of government in line with the provisions of the constitution put in place by the Gitarama coup. Grégoire Kayibanda became the first president of the Republic of Rwanda, which on 1 July 1962 became an independent country. Over and above the advantage they enjoyed owing to the numerical superiority of their supporters, the Hutu elite were helped in winning power by organised ethnic violence and active support from the colonial authorities and the Catholic Church (Munyarugerero 2003; Gatwa 2005). The wanton anti-Tutsi violence that accompanied Rwanda’s evolution towards independence set the tone for the country’s politics for the next three decades. During that time politics in Rwanda remained highly divisive, entailing violent contestation for power at first mainly between Hutu and Tutsi elites and then among Hutu elites themselves. Over the three decades the Tutsi would become the principal victims of whichever Hutu faction managed to win control of the state. 11 The decision was made by a gathering of Hutu political elites in Gitarama town, Gregoire Kayibanda’s birth place.  6 Post-colonial politics A number of factors laid the foundation for, facilitated, and fed the political violence in Rwanda’s post-colonial politics: colonial-era ethnic divisions and the consequent inter-ethnic animosity; late colonial-era and early post-colonial political violence and the forced exile of hundreds of thousands of Tutsi; the Rwanda governments’ policy of denying them the right to return and the continued discrimination against and persecution of those left behind; the Tutsi’s experience of exile, which consisted of great success for some but discrimination and persecution for others in their countries of exile; Rwanda’s unstable neighbourhood; regionalism and the intra-Hutu conflict it bred; demographic pressures and their impact on access to land and the economy; and in addition, hasty and externally-imposed political liberalisation in the 1990s (Otunnu 2000a; Otunnu 2000b; Gachuruzi 2000; Strauss 2006). Changes instituted since the RPF’s seizure of power in 1994, however, seem to promise long-term, if generally uneasy peace, fragile political stability, and ethnic reconciliation or, at least, accommodation in Rwanda. Colonial-era ethnic division As demonstrated above, the Belgian colonial administration and the Catholic Church bear much responsibility for creating divisions within Rwandese society. The divisions created, nurtured and deepened inter-ethnic animosity between Hutu and Tutsi elites seeking to acquire or monopolise political power and its attendant privileges (Gatwa 2005; Linden 1999). Taking advantage of Belgian support in the years immediately leading up to independence and the numerical strength of their constituency, Hutu elites in and out of government orchestrated episodes of violence that, from 1959 to the early 1970s, led to the death of many members of the Tutsi and Hutu communities and forced hundreds of thousands of Tutsi into exile in neighbouring countries and beyond. State-instigated murder of Tutsi, the exiling of many others and the Rwanda governments’ policy of non-return created a sense of grievance against the Rwandese state and successive governments, which was exploited by Tutsi elites seeking to return to Rwanda through any means, including war. Tutsi Exile and the politics of persecution and exclusion The forcible departure of Tutsi from Rwanda started with the violence of November 1959. While the violence followed a rumoured attack on, and murder by, Tutsi militants of the UNAR of a prominent Hutu politician, it did not, at least in its initial stages assume a decidedly ethnic pattern. For example, as already pointed out, Tutsi and Hutu supporters of the UNAR and the monarchy combined forces and fought Hutu and Tutsi members of other parties and those believed to be disloyal to the king. Nonetheless, as it spread, the violence increasingly took on an ethnic complexion when groups of Hutu youths hunted down Tutsi local officials and chiefs in what one colonial official called “a surgical operation” deemed necessary to facilitate passage from feudalism to democracy (Lugan 1997). Hutu youths were mobilised through the portrayal of Tutsi as foreigners who sought to dominate the Hutu population whose security lay in killing or expelling them from the country. Thousands of Tutsi sought  7 refuge in neighbouring countries sharing common borders with Rwanda. This is important because of the critical role the porosity of the borders eventually played in facilitating the RPF’s transnational recruitment of fighters and invasion of Rwanda. Far from settling down to a quiet life in exile, many refugees embarked on political and military organisation within the borders of their host countries in preparation for forcible return (Munyarugerero, 2003). In 1963 groups based in Burundi and Uganda staged an armed incursion. The Kayibanda government reacted with a campaign to eliminate all remaining Tutsi, whom it accused of conspiring with the exiles. The invasion also handed the government a chance to eliminate what remained of Tutsi political activity and the structures through which it was conducted. RADER and UNAR members of the government and the national assembly were executed, the two parties destroyed, and the Tutsi completely excluded from participation in public life. Other members of the two parties died in prison (La Communauté Rwandaise de France 1990). In what the government termed “uncontrollable mass reaction to Tutsi provocation”, thousands of Tutsi were hunted down and killed by specially mobilised Hutu militants overseen by local officials and government ministers (Willame 1995; Strauss 2006; Munyarugerero 2003). The killings led to a new exodus of between 200,000 and 300,000 Tutsi (Lugan, 1997: 436). In 1966 Burundi-based exiles staged another armed incursion, leading to further massacres. Tutsi men, women, and children were rounded up and executed, and yet more left the country. The massacre of Tutsi remaining in the country in response to armed incursions eventually persuaded the exiles to cease the armed struggle in 1967 and settle into life in exile (Semujanga 2003). Nonetheless, the decision by exiles to cease insurgent activities did not stop attacks on those Tutsi who had stayed behind. In February 1973 a new round of persecution commenced. First, Tutsi were purged from educational, administrative and other public institutions, as well as from the private sector on the grounds that their numerical dominance surpassed the share warranted by their small proportion of the population (Munyarugerero, 2003; Semujanga, 2003). The purges were then followed by physical attacks culminating in a new round of killings. Past massacres had been blamed on provocation by exiles, but this time there had been no incursion. It is, however, important to note that the Kayibanda regime was having to contend with growing unpopularity, rivalries within the ruling party, conflicts ensuing from regional favouritism, and a restive military dominated by Northern Hutu who felt marginalised. Taking advantage of the anxiety created within the Rwandese Hutu population as a result of the 1972 civil war in Burundi in which a Tutsi-dominated military had massacred thousands of Hutu elites, the Kayibanda government instigated the 1973 purge and killings in the hope that it would catalyse a healing of intra-Hutu divisions and a closing of ranks among Hutu (Strauss 2006: 189; Munyarugerero, 2003). The well-planned attacks, which started in educational establishments and spread to the National University in Butare before extending to other employment sectors, sought to establish Hutu dominance throughout the social, economic and political spheres. Government officials claimed that the attacks were aimed at “ethnic re- balancing” in reaction to Tutsi having surpassed the ten to twenty percent quota they had been allocated in various aspects of public life (Vidal 1991; Strauss 2006: 189- 190). More survivors sought refuge outside Rwanda, and this mass exodus contained the seeds of future destabilisation and war.  8 Much has been written about the experiences of Rwandese exiles in their countries of refuge (Waugh 2004; Otunnu 2000b; Munyarugerero 2003; Prunier 1997; Gachuruzi 2000; Mamdani 2001). While many gradually adjusted to life in exile and eventually became socially integrated, others were unable either fully to integrate because of deliberate exclusion or marginalisation, or did not wish to settle down permanently in their host societies because of persecution (Gachuruzi 2000; Otunnu 2000b; Semujanga 2003). Failure to integrate nurtured a desire to return to Rwanda to reclaim their citizenship. However, neither the Habyarimana regime nor that of Kayibanda before it was prepared to countenance their return. Both used reasons such as population density, poor soil productivity, poverty and environmental degradation to justify their reluctance to allow the refugees back into the country. Also, in addition to the Tutsi there were Hutu who had fled from political persecution and they, too, were not wanted back into the country. This inability of refugees either to integrate abroad or return to Rwanda created a large reservoir of hostility towards the Rwandese state. The regime’s withholding of their right of return meant that they had only one option, which was to force their way back. This therefore rendered them potential insurgents and, consequently, a threat to the country’s security. Pressure for return and the security dimensions of the refugee crisis Table 1 provides a picture of the estimated total population displaced by the 1960s anti-Tutsi pogroms. While some of the exiles sought refuge as far afield as Europe and North America, the vast majority settled in countries bordering Rwanda and in its neighbourhood: Burundi, Uganda, Zaïre, Tanzania and Kenya. Irrespective of where they ended up, however, many, including the descendants of refugees who were born outside the country, continued to insist on their right to return and live in Rwanda, an objective they pursued through numerous pressure groups scattered around the world.12 Table 1: Rwandese Refugees in the Great Lakes Region Country Official figure13 Adjusted figure14 Burundi 200 000 200 000 Tanzania 36 000 54 000 Uganda 78 000 156 000 Zaïre 22 000 88 000 Total 336 000 498 000 Source: 1964 UNHCR census (Communauté Rwandaise de France, op.cit.:38) The two attempts by those who had sought refuge in Uganda and Burundi to return by force in 1963 and 1966 hardened the stance of the Kayibanda regime which labelled all refugees ‘inyangarwanda’ (enemies of Rwanda) who did not deserve to be welcomed back into the country. However, the Habyarimana regime modified its stance when in 1982, during the celebrations of the country’s 20th independence anniversary the President announced that refugees who did not pose a threat to the government could return to visit their relatives but not to live in the country permanently. He was in effect not offering the 12 See, for example, Munyarugerero (2003), pp. 217-223) 13 UNHCR & Red Cross censuses conducted in refugee camps. 14 Estimated figures based on census figures but taking into account refugees living outside camps.  9 citizenship rights they were insisting on, but permission to visit as tourists. The tenacity with which the two sides held on to their respective positions rendered the conflict surrounding whether or not to return an indivisible one, thereby sowing the seeds of armed confrontation. The slight change in position by the Habyarimana regime was, however, a reaction to the turbulent political situation in Uganda, its fall- out for the Rwandese refugees, and its possible implications for Rwanda. After Milton Obote won the 1980 general elections in Uganda followed by the declaration of war against his government by Yoweri Museveni’s National Resistance Army, the Uganda government decided to expel Rwandese refugees back into Rwanda on the grounds that they were a threat to its security, as many had joined or were actively supporting the insurgency.15 Also, barely three years before the Habyarimana declaration, there had been regime change in Uganda, involving the deposition by force of arms of the Amin regime by former political refugees who organised and launched their armed struggle from Tanzania, where a large number had lived for many years. The lessons of that war, in which some Rwandese refugees in Uganda had participated, were not lost on the regime in Kigali, which had become increasingly nervous about the large number of potential Rwandese insurgents in Uganda (Otunnu 2000b; Munyarugerero 2003, chapter 5). Ironically, the nervousness felt by President Habyarimana did nothing to change his government’s policy of not granting the refugees the one thing they wanted most, and for which, as we shall see, they were prepared to wage war. Therefore in offering only temporary visitation rights to the refugees President Habyarimana, like Kayibanda before him, was effectively betting on his government’s ability to maintain the status quo over the long term, and disregarding the security dimensions of the refugee question.16 The magnitude of the security threat the refugees posed to the Rwandese state lay in the experience many had gained from participating in Uganda’s wars, some with the express objective of acquiring combat experience that would enable them to reclaim their citizenship by force if necessary.17 Also, others had served or continued to serve in both the armies and security agencies as well as in various state organs of almost all the other neighbouring countries and had acquired the knowledge, contacts, experience, and skills necessary for an insurgency.18 However, in reaction to their insistence on the 15 Among the 27 men who launched the NRA insurgency were two Rwandese refugees, Fred Gisa Rwigema who led the RPF/A’s invasion of Rwanda in 1990, and Paul Kagame who succeeded him after he was killed on the battle field. Both had also been members of Yoweri Museveni’s Front for National Salvation (FRONASA) which had participated in the Tanzania-aided war that toppled Idi Amin. 16 The blasé attitude of President Habyarimana and his government towards the security dimensions of the refugee question can be seen in the fact that, despite specific reports during during the early 1990s that the RPF was preparing to attack the country, the national army was literally caught off-guard when the attack came (see Munyarugerero 2003 pp. 213-225). 17 Many refugees resident in Kenya but with relatives in Uganda crossed into Uganda and joined the insurgency for the purposes of acquiring military training and experience. It is estimated that 3,000 out of a total NRA strength of 14,000 men and women, were Rwandese refugees (Prunier 1997). Yet more young refugees joined the NRA with the same intentions after Museveni seized power in 1986 (personal communication from a serving Rwandese military officer who served in the NRA, April 2007). 18 According to Munyarugerero (2003: 218) the refugees had played important roles in the regimes of Amin, Obote and Museveni in Uganda; Micombero, Bagaza and Buyoya in Burundi; Mobutu in Zaïre; and Nyerere and Mwinyi in Tanzania.  10 right to return, Rwanda’s government and the ruling party, the National Revolutionary Movement for Development,19 hardened their positions. In July 1986 the MRND’s Central Committee stated its position and that of the government: the country could not cope with a large influx of returnees. Nonetheless, the government was ready and willing to welcome back individual refugees in line with the country’s laws and the international conventions to which it was signatory. The national laws in question prohibited the return of refugees who had borne arms against the government, participated in subversive activities, and who had no independent means of self-sustenance (Reyntjens 2004:143). Clearly, the conditions had been designed to prevent return, not to facilitate it, thereby increasing the chances of armed confrontation yet further. Preparing for War: The Rwanda Patriotic Front and the 1990-1994 Insurgency The refugees' desire to return and the determination of the authorities in Rwanda to prevent them from doing so led to increasingly coordinated efforts by the former aimed at finding a solution to their predicament. These efforts culminated in a large network linking the refugees through their various associations and pressure groups in their host countries. In an international conference convened in Washington in August 1988, delegates reiterated their determination to return to Rwanda as the only solution to their plight. Attempts to engage the Rwandese authorities directly in dialogue failed. The Habyarimana regime preferred to talk to governments hosting the refugees in an attempt to convince them to grant the latter permanent residence rights (Otunnu 2000b). The effect of the Rwanda government’s intransigence was to convince a growing section of the refugee community that only armed struggle could guarantee their return. In preparation for possible confrontation, the Rwandese Alliance for National Unity (RANU), which had been founded in Uganda in 1980 after the dissolution of the Rwandese Refugee Welfare Foundation (RRWF),20 had transformed itself into the Rwanda Patriotic Front (RPF). The armed option had been rendered particularly attractive by the events of 1982 in Uganda, when the Obote government had decided to expel the refugees. As they fled from the agents of the Ugandan state, the majority of the refugees were refused entry by the Rwandese authorities. This refusal culminated in the death of some of the refugees at the hands of militants of Uganda’s ruling party at the time, the Uganda People’s Congress. The Habyarimana government justified refusing them entry on the grounds that there was not even an inch of land available for them to settle on. After the Washington conference rejected the government’s stand on the return of exiles, the authorities in Rwanda set up a ‘Special Commission on the Problem of Rwandese Emigrés’ in 1989. In its 1990 report, the Commission envisaged two options: return, seen as the ideal solution, and naturalisation of the refugees by their host countries. Under the ideal solution, the Commission, like the Central Committee of the MRND, saw return as a possibility only if the refugees or the international community could guarantee their welfare without recourse to public resources. Short 19 Mouvement Révolutionnaire National pour le Développement (MRND). 20 The RRWF was founded in 1979.  11 of fulfilling this condition, they were advised to stay where they were and seek naturalisation or diplomatic protection by the Rwandese government. The implication of this impasse for Rwanda’s national security was that neither dialogue nor reconciliation was possible in the short term, leaving war as the only option for the refugee community. However, the entrenched positions of the refugees and the Rwandese authorities were merely an addition to factors already steering Rwanda in the direction of violent conflict. Internally the regime was already grappling with growing opposition, economic decline, and pressure for democratisation and economic reform from the international community. It was within this context that the refugees started to mobilise actively and prepare to return by force (Otunnu 2000a, 2000b; Gachuruzi 2000; Prunier 1997; Munyarugerero 2003). Internal difficulties rendered the country ripe for war, while a number of external factors encouraged the refugees in their preparations. Internal factors Military-backed political exclusion As we saw earlier, the end of armed incursions by refugees in the late 1960s did not stop the persecution of the Tutsi who had opted to remain inside Rwanda. Demagoguery by Hutu politicians characterised by hate speeches and deliberate and systematic discrimination against Tutsi, ironically through the pursuit of a policy of 'ethnic equillirium', which had ostensibly sought to ensure equity, continued to compound ethnic divisions. Semujanga (2003: 167) describes the situation: During its Tenth Congress in 1969, the leaders of this party, dominated by politicians who came for the most part from President Kayibanda’s home region, decided to set up a policy excluding the Tutsi from the civil service and from state schools. The effects of this policy were felt from the beginning of the 1970-71 school term. At the beginning of the 1972-73 term, during the admission exams to secondary schools, not one Tutsi pupil was accepted, and no Tutsi student was matriculated into the National University of Rwanda. Moreover, in February-March 1973, all Tutsi students were expelled from every educational institution in the country; state and private sector employees were fired as well. The pursuit of deliberate exclusion was, however, to boomerang on the Kayibanda regime as it became increasingly unpopular within the broad Hutu community and attention focused more on its failures and shortcomings than on the Tutsi. Increasingly, ruling party leaders came under attack by their own party members for inciting ethnic hatred and victimising innocent Tutsi. Kayibanda and the ruling party reacted to the growing fissures by expelling the dissenters. To ensure greater loyalty in the party, recruitment of new members, especially into its inner circle, became increasingly focused on the president’s home region of Gitarama. Henceforth regionalism (or Gitaramism) became an important political reality. So important did it become that towards the end of the First Republic, with the exception of military officers, the vast majority of leaders in the country were originally from Gitarama. Exclusion on the basis of region and the perception of the regime as corrupt created and deepened schisms within Hutu elite circles, which eventually made it easy for the  12 northern Hutu dominated army to organise the 1973 coup that was to change the balance of power in favour of northern Hutu elites or a sub-section thereof. After the northern Hutu military elite under the leadership of Major-General Juvenal Habyarimana wrested power from Kayibanda and his southern cohorts, the new president denounced their conduct in power and promised to institute change. He promised to end ethnic and regional exclusion, corruption, and injustice. Habyarimana’s rise to power therefore promised a new beginning and raised expectations among the population, especially the Tutsi for whom independence had brought exclusion and persecution (Erny 1994). However, the optimism, especially for the Tutsi, was not to last. In its bid to rectify the injustices of the past, the new regime institutionalised discrimination under a policy formulated but not implemented by the Kayibanda regime (Kimonyo, forthcoming). The policy of ethnic balancing (équilibre éthnique) was extended to include regional balancing. Semujanga (2003: 156) describes the new situation: … General Habyarimana’s coup d’etat of 1973 brought about a new situation. The Tutsi were secure again, since the regime endorsed national unity. … in reality, Habyarimana’s attitude was, from the start, ambiguous. At the same time that he affirmed his will to establish a policy of national unity likely to allow the Tutsi to recover their basic civil rights, he set up a policy that was supposed to promote ethnic and regional equilibrium, but in fact was used to marginalise the Tutsi and Hutu who were not from the northern part of the country. Once his power was established, Habyarimana turned against the Tutsi, designating them “oppressors of the Hutu” because of what he described as their control over the country’s economy. In moving to exclude Tutsi and privilege northern Hutu, especially those from Gisenyi (where he was born) and Ruhengeri, the Habyarimana regime continued “the logic of the bipolarisation of society: Hutu-Tutsi, Nduga-Rukiga”21 (Semujanga 2003: 167) inherited from the Kayibanda era. Senior members of the Kayibanda regime having been liquidated, the new regime brought the state under the control of a section of northern Hutu elites linked by blood and marriage ties in what amounted to a replay of Kayibanda-era winner-takes-all politics.22 As this narrow elite consolidated its hold on power and the state, it resorted to repression to assert and retain its control. When active internal opposition and calls for political pluralism emerged, they came from civil society groups (see Longman 1999), northern Hutu elites outside the President’s circle of friends and relatives, and southern Hutu elites seeking to end their marginalisation (Munyarugerero 2003). Habyarimana in Power From his seizure of power in 1973 up to the late 1980s, Habyarimana presided over a peaceful and stable country whose economy registered high rates of growth (Erny 1994). Furthermore, before 1990 the regime did not instigate or authorise killings of civilians, Tutsi or Hutu (Strauss 2006). In addition, the president’s anti-sectarian declarations of 5 July 1973 had seemed to elevate him above the politics of ethnic and regional exclusion that had marked Kayibanda’s presidency. The good news, 21 The phrase stands for “south-north”. 22 For a detailed examination and explanation of this phenomenon in African politics, see Allen (1995).  13 however, was overshadowed by what one could describe, in relation to the Rwandan context up to that time, as ‘politics as usual’. First, the new regime embraced its predecessor’s practice of institutionalised discrimination, albeit under the guise of making efforts to promote equity. Its pursuit of regional and ethnic balancing derived from the principle that access to education (public and private), employment and political position, should be based on quotas established in accordance with ethnic and regional considerations. The operationalisation of these quotas was facilitated by the long-instituted practice of identification, right from birth, of all Rwandese nationals by ethnicity on official records and identity documentation.23 Ethnic identification in official documents had been consciously adopted by Hutu power elites after independence in order to facilitate discrimination against the Tutsi community (Waugh 2004). The policy played an important role in preventing the emergence of a post-colonial Rwanda in which ethnicity did not play a role in determining how citizens were treated by the state. Significantly, the allocation of education, employment and other quotas was based on statistics of doubtful validity. For example, the results of the 1978 national population census were never published. Nonetheless, the government claimed that the Hutu made up ninety percent of the population, the Tutsi nine percent, and the Twa one percent; figures that never seemed to change with the passage of time. More significantly, even the quotas were never adhered to. For example, out of a total of 143 mayors (bourgmestres), none were Tutsi. All the ten heads of prefectures were Hutu. Out of 70 parliamentarians, only two were Tutsi. There was only one Tutsi officer in the army, reportedly recruited by accident just before independence, after which Tutsi were barred. There was only one Tutsi member of the government (Communauté Rwandaise de France 1990: 25). Discriminatory laws extended from the public and private sectors into the domestic domain, and even religious institutions, which also had to observe ethnic quotas in filling vacant positions. For serving members of the military, inter-ethnic marriages were strictly prohibited. In its determination to enforce ethnic quotas, the state was facilitated by laws that prohibited people from changing their designated ethnic group and the requirement that local officials adhere strictly to their preservation by paying particular attention to record keeping.24 Like discrimination on the basis of ethnicity, which paid little regard to competence, discrimination on the basis of regional origin did the same and elevated northern Hutu, especially from Gisenyi and Ruhengeri prefectures and more narrowly from the families of President Habyarimana and his wife, Agathe Kanziga. Regional favouritism manifested itself in the (over)concentration of development projects in the President’s prefecture of origin, Ruhengeri, and the forcible splitting of the National University at Butare into two universities, with the new University of Arts and Human Sciences going to Nyakinama in Ruhengeri Prefecture. By the mid-1980s, a third of the 85 most important appointments in government and almost the entire leaderships of the army and security agencies were from Gisenyi Prefecture. People from Gisenyi and Ruhengeri also dominated the leaderships of 23 In 1934-35 the colonial administration had officially allotted ethnic affiliation to Rwandans on account of the number of heads of cattle one possessed. 24 For a graphic depiction of the extent to which state agents went to investigate and verify people’s ethnic affiliation, see, for example, Kimonyo (forthcoming, pp. 63-65).  14 public enterprises and access to scholarships to study abroad (Reyntjens 1994: 34). Underlying the exclusion of southern Hutu was the belief that they had been extensively infiltrated by the Tutsi and should therefore not be trusted with the affairs of state in a predominantly Hutu country. This left the way open for the extended families of Habyarimana, his wife and their friends to dominate public life. The politics of ethnic and regional balancing begot feelings of grievance among a considerable section of the population. The Tutsi living in the South suffered double discrimination as Tutsi and as southerners, which further diminished their access to employment in the army, forces of law and order, and in the public sector. On the other hand, discrimination also ensured that those who benefited from the injustice committed against their compatriots were closely aligned with the government and opposed all attempts at questioning the status quo in order to protect their privileges. As had been the case during the Kayibanda years, when exclusion of northern Hutu had destroyed Hutu solidarity, so it was under Habyarimana. Regionalism isolated the regime from southern Hutu and prepared the ground for internal Hutu opposition politicians and groups to make overtures to the RPF in an effort to join forces and oust it. The Birth of the MRND, Single-Party Rule, and Concentration of Power The National Revolutionary Movement for Development25 was founded in July 1973. With it came single-party rule, thereby continuing in the footsteps of PARMEHUTU, the party which, under Kayibanda’s leadership, led Rwanda to independence. In what seems to have been the Habyarimana regime’s version of inclusive politics, every Rwandan had to become a member of the ruling party by law. Ironically, despite the perpetuation of ethnic and regional divisions by the new regime, Habyarimana justified the imposition of one-party rule by reference to the homogeneity of Rwandese society: “single partism is the route Rwanda has chosen because we have the same culture and speak one language” (Munyarugerero 2003: 173).26 Until the creation of the party, Rwanda had been under the control of the military and before that, under that of PARMEHUTU, the former ruling party which, following the coup d’etat, had been proscribed. With the creation of the MRND, however, the country came under the dual control of the party and the army, with power concentrated in the hands of Major-General Habyarimana. In addition to being president of the republic, he became the Founder President of the MRND, Prime Minister, Army Chief of Staff, Minister of Defence, President of the MRND’s Central Committee, and President of the High Judicial Council (NKunzumwami 1996: 74), functions he apparently executed with effectiveness during the first 15 years of his reign (Rusatira 2005). In 1981, following the 1978 general elections in which only he was eligible to contest due to constitutional amendments approved in a referendum held for the purpose, the National Council for Development27, an organ of the MRND, replaced the National Assembly. Members of the Council were elected for a period of 5 years from a list drawn up by the MRND’s Central Committee (Munyarugerero, 2003: 173). 25 Mouvement Révolutionnaire National pour le Développement (MRND). 26 My translation from the French. 27 Conseil National de Développement (CND).  15 In the context of continued, albeit reverse, regionalism, single-party rule gave rise to opposition from marginalised southern Hutu as well as once privileged northern Hutu elites who in the course of time fell out of favour and were confined to the political margins or excluded completely. The regime’s response, as with all exclusivist regimes, was repression and further closing off of all possibilities for participation in the country’s politics by their rivals. As Longman (1999: 342) points out: Habyarimana created parallel state and party structures that reached down to the most local level to facilitate monitoring and control of the population. Social organisations were almost entirely subsumed… By the mid-1980s, however, the population was becoming increasingly disenchanted. The benefits of economic growth were heavily concentrated in the hands of Habyarimana and supporters, particularly those from his family and home region … The MRND, once regarded as a unifying and developing force in the country, seemed increasingly to be an instrument for controlling the population and concentrating wealth in the hands of the party elite. As has been the case elsewhere, “the totalizing project of the party-state and its attempt to control every aspect of social, political and economic life” (Longman 1999: 343) meant that violent contestation became the only credible way through which those opposed to the regime stood a chance of acquiring power.28 Thus, by the time the RPF were organising and mobilising for war in Uganda and the surrounding countries, frustration by the internal Hutu opposition about their inability to peacefully compete for power on equal terms with the MRND, which already employed “irregular tactics” to keep them out, had already built up into tensions that rendered the country ripe for violent conflict internally as well (Strauss 2006). According to Longman (1999: 348): The Interahamwe, the youth militia of the MRND, and the Coalition for the Defence of the Republic (CDR), an extreme Hutu ethnonationalist party allied with the MNRD, disrupted rallies by the opposition political parties, blocking traffic and starting fights. They also actively harassed opposition politicians and other critics of the government. It is reported that with regards to the invasion of the country by the RPF, elements of the internal Hutu opposition were already in contact with the rebel group while some had already formally enlisted (Otunnu 2000b; Munyarugerero 2003). It is probable that awareness of the regime’s vulnerability inside the country through its contacts with the internal opposition encouraged the RPF’s military wing to invade at the time it did. 28 For a discussion of the effects of power monopoly on the calculations of rival elites to those in power in Africa, see Tull and Mehler (2005) and Herbst (2006).  16 Demographic pressures, land scarcity, and lack of opportunity for self- advancement Rwanda is a small country with a large population and possibly the worst problem of land scarcity in the world. This has placed a very high premium on land and explains the link between land scarcity, associated struggles and conflicts, and popular participation in episodes of politically-inspired communal violence and murder (Jefremovas 2002). Indeed, according to Chretien (2003), with each anti-Tutsi pogrom “Tutsi houses were razed and replaced by banana plants” (333). Kimonyo (forthcoming) also shows how the intention of acquiring land and other property abandoned by exiled Tutsi or those that had been killed was an important motivator for Hutu peasants to participate in anti-Tutsi violence. And as Jefremovas (2002: 72) points out, between 1959 and 1961, “anti-Tutsi sentiments took the form of land invasions”. Land scarcity has been intimately linked to the twin problems of poverty and deprivation in the country. Right from independence until 1994, problems associated with land scarcity had been accentuated by concentration of resources – land inclusive – in the hands of small networks of Hutu elites in both the north and south of the country, as well as heavy control and abuse through patronage of access to land by state functionaries.29 This overconcentration of land in a few hands deprived many ordinary Rwandese of a chance to own land and make a living working it. Consequently many landless people, especially youth, were forced to migrate to urban areas in search of work – most of them unsuccessfully. Lack of employment opportunities created a large reservoir of unemployed youths in towns. It was they – street boys, car washers, and the un-domiciled unemployed – that the perpetrators of the 1994 genocide recruited into the ranks of the interahamwe and impuzamugambi militias responsible for killing large numbers of people. For them the genocide offered opportunities for elimination of those from the wrong side of the political and social divide who had valuables for them to steal, and therefore for self-enrichment without fear of punishment.30 How central these factors were to the outbreak of war in 1990 is debatable. What is clear, however, is that they created a context in which anti-Tutsi violence could easily be instigated and sustained by Hutus fearful of being dispossessed by returning Tutsis if the RPF won the war, and by those seeking to take over land and property belonging to deceased Tutsi. As Tutsi persecution intensified, so did the determination by the RPF to end it by defeating government forces and seizing power. As the war intensified, so did the fear and uncertainty it created in the minds of ordinary Hutu, driving them further into the hands of elite extremists who urged them to kill Tutsi – itself presented as an act of war (Strauss 2006: 201-223). Moreover one of the arguments presented by the Habyarimana regime, and the regime before it, for resisting the return to Rwanda of exiled Tutsi was land scarcity (Kimonyo, forthcoming). As discussed earlier, regardless of the validity of the claim in the light of the mass return of exiles post-1994 and the absence of widespread conflicts at that point, land scarcity fed the RPF’s determination to seize power by force and ensure the return of refugees forcibly displaced by state-orchestrated political violence. 29 See, for example, Munyarugerero (2003), pp. 122-125. 30 For a stimulating and sceptical discussion of the view that land scarcity and deprivation played a role in bringing about the genocide, see Strauss (2006).  17 Hate propaganda Over and above poverty and associated phenomena, there were other specific reasons why members of the Hutu community were prepared to take up weapons on behalf of government troops and militia and kill Tutsis, supposedly in self-defence, and in the process fuel the war. One reason which has been widely cited in the literature31 is excessive obedience by the Rwandese towards authority. Analysts assert that it facilitated both the Kayibanda and Habyarimana regimes in their mobilisation of the Hutu population against the Tutsi.32 Also, both in the run-up to independence and after Rwanda became a sovereign state, Hutu elites seeking to acquire and monopolise power had portrayed members of the Tutsi community as enemies who should be killed ‘in self-defence’. While the image of the Tutsi as alien enemies of Rwanda and of the Hutu community (umwanzi ni umwe ni umututsi)33 faded during periods of calm and stability, during times of crisis and uncertainty or whenever Hutu elites wanted to whip up anti-Tutsi sentiment, it would be re-activated. It was, for example, an important basis for the Habyarimana and Kayibanda regimes rejecting demands by the exiles to return to Rwanda and for persecuting or killing Hutu suspected of collaboration (ibyitso) with the RPF after it invaded the country. The turning of the Tutsi community into second-class citizens effectively created a reservoir of potential recruits for insurgencies as evidenced by earlier attempts at armed struggle by exiles. Indeed, as Strauss (2006) shows, pre-war Tutsi-dominated areas of the country such as the Bugesera region witnessed outflows of RPF recruits and were particular targets of anti-Tutsi killer squads during the 1994 genocide (Kimonyo, forthcoming). The Impact of the 1990s Economic Crisis It is widely acknowledged that although the Habyarimana regime was authoritarian, it was also development-oriented with a good track record of economic management during its first decade in power (Reyntjens 1994; Lugan 1997; Prunier 1997; Uvin 1997; Adelman 2000). Measured in terms of Gross National Product (GNP) per capita, and considering the country’s intrinsic disadvantages (landlocked, highly populated, natural-resource-poor) and the relatively poor performance of its neighbours (Table 2), Rwanda under Habyarimana registered considerable progress. At independence in 1962, only two countries in the world had less income per capita than Rwanda. However, by 1987 there were eighteen of them. With an average income per capita of USD300, Rwanda could be said to have been at the same level with China whose average income per capita was USD310 (Prunier 1997). 31 See, for example, Zorbas (2007), who presents evidence to cast serious doubt on the claim. 32 See, for example, Uvin, 1997. 33 “The enemy is one; it is the Tutsi” (Strauss 2006:.225).  18 Table 2: Evolution of GNP/capita in Rwanda and her Neighbours Ranking Year Rwanda Burundi Zaire Uganda Tanzania 1976 7 11 16 33 25 1981 16 14 12 13 19 1985 18 11 9 n.a. 21 1990 19 11 12 13 2 Difference +12 - -4 - 20 - 23 1976-1990 Source: World Bank Reports (Reyntjens 1994, p.35) As Table 2 shows, in 15 years Rwanda had improved its position by 12 places, Burundi had remained static despite having started off ahead of Rwanda at independence (Erny 1994: 61), while other neighbouring countries had become poorer. While in 1976 Rwanda had been the poorest, by 1990 it had become the least poor country in its neighbourhood. At sector level, the agricultural (primary) sector which in 1962 accounted for 80 percent of GNP had declined to 48 percent by 1986. Meanwhile the service sector had grown from 8 percent to 21 percent, with the manufacturing sector increasing from 12 percent to 31 percent. The mortality rate was declining, hygiene and medical care indicators were improving, and the quality of education was rising. Despite high population growth, school enrolment had risen from 49.5 percent in 1978 to 61.8 percent in 1986 (Prunier 1997). In terms of infrastructural development, Rwanda had one of the best road networks in Africa, reliable posts and telecommunications services, with piped water and electricity covering many parts of the country. General economic management standards were also higher than in much of the rest of Africa, which explains why up until the mid-1980s the country owed modest levels of external debt compared to other countries. For example, in 1987 Rwanda’s external debt was only 28 percent of GNP, one of the lowest in Africa at the time (Prunier 1997). Nonetheless, its dependence on foreign aid had grown substantially by the end of the 1980s. According to the OECD, foreign aid – which in 1973 stood at less than 5 percent of GNP – had grown to 11 percent in 1986 and to 22 percent in 1991 (Prunier 1997). The country, however, started experiencing economic problems during the late 1980s, especially after the collapse of the price of coffee, the country’s main source of revenue, in 1988. The economic crisis prompted the World Bank, supported by the donor community, to demand that Rwanda implement a structural adjustment programme (SAP) to address it. The economic crisis and the difficulties associated with implementing a structural adjustment programme coincided with increased internal opposition and growing unrest stemming from years of the MRND’s political monopoly, high levels of corruption, regionalism, and political repression. Growing political instability prompted the country’s main donors such as Canada and France to link continued aid to democratisation consisting of, among other things, opening up to multi-party politics. The coincidence of economic crisis with growing internal dissent and opposition and the imposition of economic and political conditionality created a mood of panic within ruling circles, increased the power and influence of Hutu extremists within the ruling party, and pushed the regime to adopt unorthodox methods to protect its hold on power. As government-allied gangs harrased the opposition, the latter took to  19 forming their own youth brigades to fight back (Munyarugerero 2003). Therefore the intensification of political repression by the state served only to heighten instability and render the country ripe for insurgency (Lugan 1997; Callamard 2000; Adelman 2000). External Factors In addition to the internal factors that laid the ground for political violence and armed conflict, there were external ones, which, rather than working towards averting it, guaranteed Rwanda’s descent into war. These factors included the post-cold war change in attitude towards, and in a sense retreat from, Africa by the world’s great powers, the commitment and friendship some countries felt towards Rwanda and its rulers at the time, Rwanda’s war-infested neighbourhood, the refugee experiences of Tutsi exiles, and the RPA invasion itself. Influence of the End of the Cold War After decades of super-power rivalry ended with the collapse of communism, interest by the world’s great powers in African affairs and events in Africa diminished. With the need to outmanoeuvre each other in Africa removed, the former cold warriors focused their attention on domestic affairs and to goings-on in regions closer to home, such as the Balkans, or areas of more immediate strategic importance than Africa. In the case of tiny and obscure Rwanda and other countries ruled by dictators, this came as a blessing but also a source of risk. It was a blessing for repressive governments, allowing them to commit crimes against their own citizens without attracting much attention or publicity. The risk, however, lay in the fact that this repression invited violent reaction from internal opponents convinced that violence was the only means through which such repression could be ended, as indeed was the case in Rwanda and previously and subsequently elsewhere on the continent (Tull and Mehler 2005). Inattention to the brewing political crisis in 1980s Rwanda by the great powers also partially accounts for the ability by the Museveni government in Uganda to facilitate, by commission or omission, the activities of the Rwanda Patriotic Front as it prepared for and prosecuted the war against the Habyarimana government.34 Rwanda’s and Habyarimana’s International Friends Ironically, alongside the diminished interest in Africa by the great powers ran a high degree of support for and admiration of Rwanda and its President by some international actors. Among these were governments and their leaders, and members of their families. Here France and Zaïre and Presidents Mobutu Sese Seko and François Mitterand are the best examples.35 There were also members of the international aid community working for bilateral and multi-lateral organisations (Uvin 1997). 34 On the Museveni government’s role in facilitating the RPF as well as consequent diplomatic activity aimed at stopping it see Otunnu 2000b; Adelman 2000; and Waugh 2004. 35 For detailed examination of these relationships, see, for example, Waugh, 2004 & Munyarugerero, 2003; also Callamard, 2000.  20 Despite linking continued aid to democratisation in the early 1990s, France had long been an uncritical ally of Rwanda, a situation attributed to “personal and patrimonial ties” and “personal contacts” between French and African governments as well as among their high-level officials. These, an analyst has argued, were underlain by a “poverty of institutional mechanisms” as captured in the “passivity” of the French Parliament (Callamard 2000; also Wallis 2006). It is the nature of this relationship that accounts for the lack of concern, even after the RPF had invaded, with “exactions committed by the Habyarimana regime”, the “absence of French diplomatic interventions against human rights violations committed by the Rwandese regime”, and eventually to the stepping up of French military assistance to prop up the regime (Callamard 2000: 169).36 Mobutu’s Zaïre provided the Habyarimana regime not only with moral support, but also signed a pact with the regime providing for “common security services, the sharing of security information, military co-operation, and interdiction of opposition movements on each other’s territory” (Gachuruzi 2000: 58). It was on the basis of these arrangements that, despite general repression and human rights violations, Zaïre provided the Habyarimana regime with military backing in the form of troops after the RPF invasion (Gachuruzi 2000; Munyarugerero 2003). Meanwhile international aid workers continued naïvely to portray Rwanda as a showcase of success even after it became clear that this was no longer the case.37 A good illustration of this comes from Adelman (2000) according to whom: Canadian development experts involved and committed to Rwanda had no sense of popular unrest even in the late eighties. For them, the anti- Rwandese propaganda efforts were considered to be the product of Tutsis who had been forced out of Rwanda over twenty years ago (p. 189). To the same experts, Adelman continues, President Habyarimana “remained the knight of purity for the vast majority of Rwandans, a man dedicated to the well-being of his people who could do little wrong in the eyes of those he ruled” (p. 189). On the basis of these sorts of assessments Rwanda continued to receive large amounts of aid, which must have served as encouragement to the Habyarimana regime not to change its behaviour towards the internal opposition, let alone the Tutsi population or Tutsi exiles pressing for their right to return to their country of birth. In this way, the country was helped on its inexorable descent into political unrest, armed conflict, and eventually genocide. Rwanda’s War-torn Neighbourhood After the Rwandese Alliance for National Unity (RANU) transformed itself into the Rwanda Patriotic Front during the late 1980s to facilitate the return of exiles by force, recruitment of cadres and fighters started in earnest. In Uganda the Museveni-led National Resistance Army (NRA) insurgency had provided thousands of refugees with the necessary military training and combat experience. After Museveni seized power, many others enlisted in the Ugandan army and served in various capacities, 36 According to Callamard (2000) “French military assistance to Rwanda remained modest for several years”. The RPF invasion, however, “drastically transformed the nature and extent of the military relationship” (pp. 158-159). 37 Uvin (1997) provides the most authoritative account of influence.  21 some in senior and sensitive positions in preparation for eventual invasion of Rwanda. Prunier (1997) and Otunnu (2000b) have so far written the most comprehensive accounts of the RPF’s infiltration of the Ugandan military in preparation for the insurgency they intended to launch against the Rwanda government. While the Uganda government has denied complicity in the RPF’s plans, albeit with individual officials giving oblique indications of its possible role, circumstantial evidence points to far-reaching involvement. Elsewhere, in Zaïre, Burundi, Tanzania and Rwanda itself, according to interviews with participants in the insurgency,38 the RPF exploited a combination of weak border controls, limited vigilance by national security agencies, and political turmoil in some of these countries to recruit and ferret out combatants from local branches of the large Rwandese, especially Tutsi, diaspora for training. Outside the immediate neighbourhood of the Great Lakes region, from as far afield as southern Africa, the Americas and Europe, the Movement sourced volunteer fighters and financial resources. Meanwhile by the early 1990s, because of the Habyarimana regime’s politics of ethnic and regional exclusion, the political situation in Rwanda and the deteriorating economic environment, the government was losing control as internal opposition grew in intensity (Chretien 2003; Strauss 2006). Munyarugerero (2003: 253-254) sums up the situation as it was in early 1992: By April 1992 … the government was trapped. Within a year it had to negotiate a peace settlement, guarantee internal peace, sort out the country’s administration, implement the structural adjustment programme imposed from the beginning of 1991 by the Bretton Woods Institutions, organise a debate about the national conference and when the chance presented itself, find a solution to the refugee problem, and finally, organise general elections. For a government used to working according to its own internal dictatorial and self- assured logic, the combination of these simultaneous pressures and forces it could neither resist effectively, nor control, amounted to a recipe for breakdown. Already, as Munyarugerero (2003: 255) points out, the army was experiencing mutinies and taking out its frustration on members of the public in reaction to the humiliating losses it was suffering on the battle front against the Rwanda Patriotic Army (RPA), the RPF’s military wing. To make matters worse, after the formation of the transitional government bringing together the MRND and internal opposition parties, the army became divided along lines reflecting the political complexion of the new government. This served only to weaken further the state’s capacity for self- preservation. The RPA invasion does in a way fit with the notion of the 'contagion' effect of war in Africa, the so-called ‘war next door’ syndrome, whereby armed conflicts occur in countries or sub-regions that have had previous conflict, and where war in one country usually ignites or fuels war in one or several of its neighbours where conditions favourable to armed conflict already exist. The transformation of the National Resistance Army insurgency in Uganda into the Rwanda Patriotic Front’s invasion of Rwanda testifies to the transmissibility of war from one country to another or others (see De Waal 2000). In this case the situation in Rwanda as captured broadly by the two Hutu/Tutsi and northern Hutu/southern Hutu polarisations was ripe for the transmission of this particular war. 38 These interviews have been conducted at different intervals over the 2005-2007 period.  22 In Rwanda the government’s portrayal of the invasion simply as a decision by the Tutsi refugees to return by force and reverse the gains of the 1959 revolution, ignored the internal and external conditions that made it possible and lent credibility to the RPF’s claim that it was motivated by wider political objectives.39 The RPF presented itself as a multi-ethnic movement and as an alternative to the regime in Kigali, which it accused of corruption, nepotism and violation of human rights. In the context of growing popular disaffection with the regime inside Rwanda itself, the RPF’s actions, ethnic considerations aside, struck a chord with disenchanted members of the general public. It is probable that if Hutu governing elites in their desperation to hold on to power had not stirred anti-Tutsi sentiments and mobilised the Hutu population to kill members of the Tutsi community, the RPF invasion would have been welcomed by many Tutsi and Hutu alike; after all, both had suffered as a result of ethnic and regional exclusion. The Refugee Experiences of the Exile Community Rwanda’s large exile community in their various countries of refuge had had vastly varied experiences of life outside Rwanda. Some had spent their lives in refugee camps segregated from their host societies. Others had melted into local communities and become socially integrated. Yet others, even as they lived outside refugee camps and participated in their host countries’ social, political and economic life, had failed fully to integrate and had often experienced hostility from members of local communities in the areas where they lived. Failure to integrate had been due mostly to labelling,40lingering discrimination and, from time to time, systematic exclusion. In Zaïre, for example, their experiences consisted of being allowed citizenship and then having it withdrawn, along with state protection (Gachuruzi 2000). Failure to integrate, however, did not generally serve as an obstacle to success, especially in commerce. Ironically, it was the combination of failure fully to integrate and resounding success for some that propelled the formation and success of the Rwanda Patriotic Front. Those who had failed to integrate and experienced discrimination longed to return to Rwanda where they felt they belonged. Among these were thousands serving in the Ugandan military, whose experience of discrimination is described by Otunnu (2000a; 2000b) and Prunier (1997). Those who had prospered and may possibly not have felt as keen a desire to return, had the means to make or mobilise significant financial contributions to the rebel movement and the insurgency. Writing about the experiences of rwandophones in Mobutu’s Zaïre Gachuruzi (2000) shows that they made large material and financial contributions to the RPF’s preparations for war and to the war effort after hostilities began. Invasion by the Rwanda Patriotic Army The invasion by the RPA marked the final stage in Rwanda’s descent into armed conflict. Even then, it is plausible that without the invasion the country might have 39 See, for example, the organisation’s 8-point plan. 40 On the psychological and other (especially negative) effects of labelling, see Moncrieffe & Eyben (2007).  23 been spared the war that culminated in the 1994 genocide. Despite growing internal unrest and opposition to the Habyarimana regime, an immediate consequence of the invasion was that a large section of an otherwise fractious Hutu community closed ranks and tried to fight back. The invasion rallied disparate elements of the Hutu population, many of whom had grown to despise the Habyarimana dictatorship and to see it as a threat to their own wellbeing, around the ruling elite. In a different context the government may possibly have had difficulty mobilising an ethnic militia.41 But Rwanda already had a history of armed incursions by refugees that had been portrayed as seeking to exterminate or enslave the Hutu, with the connivance of Tutsi still living inside the country. This particular invasion, like those before it, provided Hutu elite demagogues with an opportunity to instill fear and suspicion among ordinary Hutu of members of the Tutsi community. Besides arresting large numbers of Tutsi civilians accused of collaborating with the RPF/A, the government reacted by recruiting youth into pro-government militia. The invasion complicated further an already turbulent political situation. It coincided with Habyarimana’s acceptance of multi-party politics under international pressure. Political liberalisation led to a virtual explosion in the number of political parties. In addition, it led to the coalescing of Hutu extremist parties, including the MRND,42 into the Alliance for the Strengthening of Democracy (Alliance pour le Renforcement de la Démocratie (ARD)). The emergence of new parties saw a haemorrhage of members of the MRND to the opposition and to the more extremist members of its coalition, the ARD. Some of the ARD’s member parties were more extreme in their views on the question of ethnicity than the MRND, and had been formed by members of the MRND disillusioned by the party’s seeming abandonment of ethnic radicalism. Inside the MRND itself, by 1994 Habyarimana’s authority and power in both the liberal and extremist wings of the party had been considerably whittled away. Not only had he given up his position in the army, but also the party presidency. Meanwhile the northern Hutu-dominated MRND, until then the self-styled voice of the Hutu population, increasingly had to contend with competition from the southern Hutu-dominated Coalition for the Defence of the Republic (CDR).43 The latter, mostly under the influence of Kayibanda-era Hutu extremists, saw itself as both the successor to PARMEHUTU and the main bulwark against “Tutsi feudalism” (Lugan 1997: 481). Gradually the MRND and the grand alliance of Hutu parties came under the influence of these elements. Meanwhile, despite the formal opening up of the political playing field, the government continued to behave as if the MRND was still the only legally recognised party. Its partisans in the countryside, including local government officials and civil servants, continued to obstruct the activities of opposition parties. Even then, the opposition parties acting under the auspices of an opposition consultative committee were exerting strong pressure on a reluctant government to institute reforms in keeping with the multi-party dispensation. Fearful of losing power, the ruling elite of the MRND and their newfound allies who, following Kayibanda’s overthrow had 41 See Strauss (2006, chapter 8) for contextual factors amenable to inter-ethnic violence generally and those that favoured the organisation of militias and death squads in Rwanda, and Herbst (2006) for a discussion of conditions that favour the formation and sustainability of insurgencies. 42 By then rebaptised MRND(D): Mouvement Révolutionnaire National pour le Développement et la Démocratie (Revolutinary National Movement for Development and Democracy). 43 Coalition pour la Défence de la République.  24 been confined to the margins, decided to adopt desperate means to retain Hutu control over the government and country. As the army lost ground to the RPA insurgents, they resorted to ethnic militia and ordinary members of the Hutu community, the willing and the forcefully conscripted, to defend the country by killing Tutsi and Hutu dissidents and members of the opposition. Despite efforts to stop the war through the Arusha peace process, for example, the events sparked off by the invasion made the RPF increasingly determined to seize power, save the Tutsi population from extermination, and bring the country to order. After the War and the Collapse of the Second Republic After it seized power, the RPF had to contend with a number of immediate and long- term challenges. First was how to restore order and ensure security in a context of collapsed state structures, reconstitute the state, and gain credibility. Then there were the twin questions of how to win the acceptance of the majority Hutu population whose minds had been poisoned against the Front in particular and the Tutsi community in general by hate propaganda, or alienated by acts of violence committed by its fighters during the war. There was also the question of how to restore social order and trust in communities torn apart by violence and the mutual suspicion that developed as a result. Finally, the RPF had to decide how to deal with members of the defeated army, the Forces Armées Rwandaise (FAR), as well as its non-armed political rivals and opponents. Restoring order and Re-constituting the state After it captured power, the RPF installed a Transitional Government of National Unity which, in line with the provisions of the Arusha Peace Accords, included all political actors save for the former ruling party, the MRND and its ally, the Coalition for the Defence of the Republic (CDR),44 which stood accused of playing a direct role in fomenting anti-Tutsi sentiment as well as planning and executing the genocide. While political inclusion was in line with the provisions of the Accords, to the RPF leadership, conscious of the effects of political exclusion of which they had been victims for over three decades, it was also a mark of the manner in which they sought to govern the country. The RPF-led government crafted an interim constitution in 1995 and appointed a constitutional commission to work on a new constitution based on wide consultations. The new government, though, started life in dire circumstances. It had inherited a state in ruins and a country with empty coffers. The civil service had fled the country almost in its entirety, leaving such organs as the judiciary as well as health and educational services non-functional. Infrastructure, too, had been destroyed, and businesses, including banks, looted (Reyntjens 2004: 178). To compound its difficulties, the new government was virtually completely isolated, without friends to help it establish itself. Only its Ugandan ally was at hand to assist, and with its ability to do so heavily constrained by its own limited resources. Worse still, France, the principal backer of the Habyarimana regime, was determined to undermine the new government through its activities in the Congo, where the defeated 44 Together, these parties were deemed to be guilty of instigating the genocide.  25 government and remnants of its army and allied militias had sought refuge, and through the use of multi-lateral aid and development organisations. For example, resources mobilised in the early days by international actors such as the European Union were, “at the insistence of Paris”, directed mostly at relief activities in favour of Congo-based refugees, among them perpetrators and executors of the genocide (Waugh 2004: 95-96). In addition to the French, the new government had another enemy: doubts within the international community about its ability to establish control, return the country to order, and survive. Feeding the doubts in the early days, at least in part, were activities by the French, who as well as channelling aid towards refugees in the Congo, as noted above, were involved in retraining and re-arming of the defeated Forces Armées Rwandaises (FAR) and creating the impression that a counter- invasion was imminent. According to Waugh (2004: 97), it is likely that all this created an impression among some donors that in the event of such an invasion, the RPF might flee back to Uganda with their contributions. Consequently, during the first few months in power, the new government was financially crippled. So dire was the situation that the country could not even deploy ambassadors to their countries of posting, the exception being the ambassador to France whose deployment, at the time testimony to the sense of urgency with which the new government wanted to establish diplomatic relations with their main external protagonist, was made possible by financial contributions from individual members of the government.45 With time, though, a few donors led by the British, Rwanda’s main donor, and including the United States, the Netherlands, Belgium, the World Bank and the IMF, embarked on providing much-needed support that eventually enabled the new government to stand on its feet and embark on rebuilding the state virtually from scratch. The remarkable recovery the country went on to make won the praise of even some the RPF government’s severest critics: A mere two years after the extreme human and material destruction of 1994, the state had been rebuilt. Rwanda was again administered from top to bottom, territorial, military and security structures were in place, the judicial system was re-established, tax revenues were collected and spent. The regime was able in a short time to establish total control over state and society. This control was seen in the maintenance of an efficient army, able to operate inside and far beyond the national borders; the establishment of ‘re-education’, ‘solidarity’ and ‘regroupment camps’; the villagization policy (known as ‘imidugudu’ policy) …; and establishment of an important intelligence capacity, with the Directorate of Military Intelligence (DMI) operating inside the country and the External Security Organisation in charge of operations abroad. While many other African countries tend towards state collapse, the Rwandan state has re-affirmed itself vigorously” (Reyntjens 2004). Reyntjens goes on to comment on the nature of Rwanda’s public service: “technocratic governance is apparently satisfactory, with competent and even charming elites articulating an intelligent discourse”. In general terms this resonates with what is observable and regularly remarked on in the country. 45 Interview with a senior member of the RPF and long-serving government official (November 2006).  26 Restoring Order With hundreds of thousands dead and millions having fled into neighbouring countries before its advancing troops, the RPF took over a country in calm contemplation of the events of the previous three years and the tragedy that had befallen it in the months leading up to the collapse of the Second Republic. The newly established calm was to last only for the period it would take elements of the defeated army intent on re-taking the country to re-organise themselves. Within months they launched a number of successive insurgencies in the Northwest of the country, taking advantage of weak border controls and the uncontrolled return of refugees to infiltrate the country. Not surprisingly, given the hate propaganda of the war years, the insurgencies attracted substantial support in the areas where elements of the defeated forces sought to establish themselves. The RPF government has been heavily criticised for conducting a brutal counter-insurgency campaign in which, it is claimed, unarmed civilians were targeted, including when its armed forces invaded the Democratic Republic of Congo to take the war to the insurgents and dismantle the refugee camps where they trained (Africa Rights 1998; Longman 1995). This, and instances of anti- civilian violence by RPA fighters during the war, which some in the RPF acknowledge and explain away as unplanned, unsystematic, and not condoned by the ruling party or the military, forms the basis of claims that there is a culture of impunity in post-genocide Rwanda.46 The weakness of these criticisms47 lies in the failure to place the violence in its proper context. In the “heat of war”, especially counter-insurgency operations against non- uniformed combatants who might mingle with non-combatants, the issue of who bears responsibility for civilian deaths is subject to debate. Accusations of wilful violation of human rights by the army and claims that there is a culture of impunity in the country are therefore open to question, especially given evidence that those guilty of blatant human rights violations have been punished.48 Accusations of human rights violations notwithstanding, there is a case for arguing that it is the vigorous manner in which the government responded to the threat of insurgency, including the forcible closure of the refugee camps and return of their occupants to Rwanda, that accounts for the order that was eventually restored inside the country (Waugh 2004). Also, efforts to restore order were helped by the decision to absorb elements of the defeated army who did not go into exile or who subsequently gave themselves up into the ranks of the RPA, which became the national army, changing its name eventually to the Rwanda Defence Forces (RDF). Even those accused of genocide crimes are tried and, where no evidence is adduced to secure prosecutions, re-integrated into the RDF.49 Outside the army, in sharp contrast with past regimes that had deliberate 46 See, for example, Reyntjens (2004) and Longman (2004). 47 See, also, Africa Rights (1998). 48 In discussing allegations of systematic human rights abuse by their fighters, RPF officials often refer to officers and men of RPA who were tried and imprisoned in Rwanda and who to this day remain in prison because of acts of indiscipline, including random killing of civilians. 49 Two cases illustrate the army’s and government’s approach to dealing with returnee officers. Current Minister of Defense, General Marcel Gatsinzi and another officer, Major General Laurent Munyakazi, were integrated into the RDF upon their return from exile. Both were then subsequently accused by genocide survivors of genocide crimes and tried in Gacaca courts. General Gatsinzi, based in the southern city of Butare during the genocide, was eventually acquitted on grounds of insufficient  27 policies of keeping millions of political exiles outside the country, the RPF government directed efforts at attracting political exiles to return. Underlying this policy seems to be the belief, not entirely uninformed by the 30-year exile of many leaders and members of the RPF, that both the soldiers of the defeated army and civilian exiles were less of a threat at home than they were likely to be from outside the country’s borders. In this the RPF emulates the National Resistance Movement (NRM) government in Uganda, whose approach to dealing with exiled dissidents and former government officials and military officers has been to lure them back into the country and assist them to find employment or re-settle in their communities of origin. In this way one potential source of future insecurity and destabilisation is neutralised. With specific reference to the re-integration of defeated forces into the new army, experience from Uganda had already demonstrated that this strategy helped neutralise or win over potential insurgents and bandits, while that from post- civil-war Ethiopia shows that their exclusion creates a reservoir of potential troublemakers as well as a source of social instability (Kiyaga-Nsubuga 2004; Young 2004). The extension of the arm of reconciliation to men and officers of the FAR and to politicians and officials associated with the First and Second Republics has not always borne the expected fruits. Indeed, some have gone on to flee the country over the years, accusing the RPF government of dictatorship. Failures of the policy as well as the challenges it has encountered along the way, however, do not detract from the government’s deliberate efforts at uniting rather than dividing the country along party, regional, or ethnic lines. Another domain in which early efforts were made to devise strategies to stabilise the polity was land tenure. To appreciate the seriousness of this issue, one has to look at relevant statistics. Rwanda has a land area of only 26,388 sq km. In 1964, two years after independence, it had a population of 3 million, with a population density of 114 people per sq km. By 2003 the population had more than doubled, to 8.6 million, with a population density estimated at an average of 300 people per sq km. It is estimated that density per area of arable land (physiological density), stands at 400 people per sq km in some areas, going up to a spectacular 1000 people per sq km in others. Over 90 percent of the population live by means of peasant agriculture (Gasarasi & Musahara 2004), the majority scratching out insecure livelihoods from increasingly small, fragmented plots of which about 77 percent are less than 1 hectare (Liversage 2003: 11). At any one time about 80 percent of all farming plots are under cultivation (Rurangwa 2004), with implications for the ability of their owners to maintain the fertility and productivity of the soil. Problems linked to land scarcity in Rwanda go way back to the pre-independence period (Jefremovas 2002; Uvin 1997; Pottier 2002). However, with the mass influx in 1994 of old-case load refugees who had fled past anti-Tutsi pogroms, and the return by the new-case load refugees who had fled during and after the genocide, the problem became critical. As they returned from their refuge in neighbouring countries, old-case load refugees claimed and occupied land abandoned by the new-case load refugees, to which some of the former had legitimate claim, as it was the land from which they had originally been driven. The return of new-case load refugees beginning in 1996, however, raised the spectre of land-related conflict likely to pose a new threat to the fragile evidence against him, while Major General Munyakazi who had been based in Kigali was convicted and imprisoned.  28 stability the new government was struggling to maintain. The government introduced measures, pending the formulation of a national land policy and enactment of comprehensive land tenure legislation,50 in which the two categories of land claimants were encouraged to share the available land in the interest of promoting national reconciliation and unity.51 In addition, the transitional government embarked on a villagisation programme known locally as imidugudu. Much criticised by international NGOs,52 the imidugu were intended as an emergency measure to deal with the refugee crisis in the face of land scarcity. Parts of three nature reserves, the Mutara Game Reserve, the Akagera National Park and the Gishwati Mountain Forest Reserve as well as a number of state- owned projects were de-gazetted and distributed to returning refugees, as were parts of communal land, pastures and marshlands (Liversage 2003). This contrasts sharply with the denial of old-case load refugees of the right to return by the Kayibanda and Habyarimana regimes on account of land scarcity. The issue of land tenure generally and land scarcity specifically are far from being resolved and may probably never be resolved completely. Indeed, the prevalence of land-rela