Early Warnirg and Response: Why the International Community Failed to Prevent :the Genocide

HOWARD ADELMAN, York University, Toronto, and ASTRI SUHRKE, Christian Michelsen Institute, Norway

'Western leadership .. . has proclaimed moral indifference to be its decent Christian right' jotin le Carr6 (1995, p. 213).

The enormity of the genocide in Rwanda demands that it be subjected to searching enquiry and that members of the international community, collectfuely and indiaidually, examine their own roles in the eaent. This paper draws extensiaely on Study II of the loint Eualuation, and examines the effectiveness of international monitoring (early warning) and management of the Rwanda conflict. It is not intended to explore all the factors which together contrtbutud ta the genocide that were or might haae been amenable to modifiutioi by the internatianal communi$. The focus is on warning and response beginning uith the start of the civil war in 1990, and culminating in an analysis of tie international response to the genocide in April-lune L994.

INTRODUCTION and organisations. This paper, like the study before it, is based Evaluators and historians have the luxury on a number of sub-studies, secondary sources as well as of hindsight to criticise actions and inac- considerable tions that subsequently appear inappro- primary data collected priate. Any search for lessons from the through interviews and document searches system, tragedy must, however, avoid retrospec- in the UN the NGO community and visits to national capitals tive speculation. Decision.makers were in Europe, North America and Africa. faced with often confusing signals, numer- ous competing demands for their attention and limitations on the resources at their IMPLICATIONS OF THE EVENTS OF 1959.-52 disposal. Critical policy theory was used by the evaluators to weigh the formulation Having administered the country through and execution of policies against their the Tutsi aristocracy for four decades, the stated objectives as well as accepted inter- Belgian colonial authoritiesr turned their national norms for the behaviour of states support towards the majority Hutu in the late-1950s - a switch driven by the realisa- The developments of these last eighteen tion of the inevitability of the de-colonisa- menths have brought about the racial dicta- tion process, the logic of democratic rule torship of one party. .. . An oppressive and Belgian resentment that the strongest system has been replaced by another one. calls for independence were coming from . . . It is quite possible that some day we will witness violent among the favoured Tutsi class. After the reactions on the part of the Tutsi (cited in Prunier, 1995). soBurundi which tration and took the opportunity of placing continued to be governed by a Tutsi Hutu appointees in chief and sub-chief minority. One attack into the Bugasera posts vacated by Tutsi. area in late 1.963 came quite close to The table-turning process was carried reaching Kigali, but mostly the attacks forward by the ovetwhelming victory of were neither well organised nor adequate- the PARMEHUTU party in the local elec- ly resourced. Inside Rwanda, which had tions of 1960, the parliamentary elections become a de facto single-party state, Tutsi ot1967 and the referendum of 1967 which were increasingly excludedfrom public life saw the massive rejection of the monarchy and armed incursions were not only in favour of a republican system of govern- repulsed, but bouts of anti-Tutsi violence ment. The process was accompanied by flared; an estimated 10,fi)0 Tutsi were sporadic fighting between Tutsi and Hutu, slaughtered in the Bugasera area following and the persecution of Tutsi as the Hutu the 1963 incursion (Prunier, 1995). increased their grip on power. By late Another wave of anti-Tutsi violence took 1963, at least 130,000 Tutsi had sought place in L972-3 fuelled by events in Bur- refuge in the neighbouring Belgian Congo undi where tens of thousands of Hutu had (now Zaire), Burundi, Uganda and Tan- been massacred by the Tutsi minority in a gyanika (now Tanzania). Some estimates bid to remain in power. of the numbers involved have been much Such repression and violence swelled higher. the number of Rwandan Tutsi living in The speed of the switch in the configu- exile in neighbouring countries. When ration of Rwandan politics, and an aware- allowance is made for natural population ness that it could leave the country with an growth amont this group, by the 1990s the embittered Tutsi minority, provoked con- total number of refugees was in the order siderable criticism of Belgium's policy in of 500,000-700,000 (Prunier, 7995), the UN General Assembly,2 and repeated although the political and propaganda srg- calls for the encouragement of a process nificance of such estimates and the difficul- of national reconciliation. The Belgian ties of accurately measuring such a widely administration rejected the calls, and distributed population makes any figure a sought to head off further UN 'tinkering' matter of dispute. With the overthrow of by organising a de facto declaration of the Hutu PARMEHUTU dictatorship andits independence in early 196tby the acdama- replacement by a dictatorship from the tion of over 3,0(X) bourgemestres and munici- north-west under Habyarimana in 173, the pal councillors. The country achieved persecution of, though not the discrimi- formal independence in July 1%2. nation against, Tutsi stopped. Habyari- A report by the UN Trusteeship Com- mana, however, steadfastly refused to allow mission in early 1961 noted: the Tutsi refugees to return. In 1981 a small number of the Tutsi flict. Although the issue of Rwandan refugees in Uganda joined one of the refugees in Uganda defied easy solution, groups opposed to the regime of President opportunities that had existed remained Obote. The group was led by Yoweri unexplored or were not aggressively pur- Museveni, who was related (though dis- sued. With the exception of Tanzania, tantly) to a Rwandin Tutsi family. As the those regional states that were not indiffer- guerilla activities increased in intensity ent were part of the problem, while more Museveni's links with Rwandan Tutsi led distant states showed little interest. to increasing rhostility between Ugandan UNHCR, overburdened and understaffed, officials and people towards the Tutsi lacked the political or economicleverage to living there in.exile. This in turn led to help resolve the issue, which, at that time, more Tutsi ioiningwhat was now Museve- seemed minor in any global scale of ni's National Resistance Army (NRA), refugee problems. which eventually succeeded in ousting the Obote regime in 1.985. By then a number of Tutsi had risen to very prominent THE CIVI WAR STARTS positions within the NRA. On 1. October 199O, a well-armed force of Once in power, Museveni's position the military wing of the RPF invaded depended upon accommodating the vari- northem areas of Rwanda from Uganda. ous factions within the country who Much of the force was composed of resented the connection with the Rwandan Rwandan Tutsi soldiers from the Ugandan Tutsi. Seeing the direction of Ugandan army and much of their armaments were politics, the Rwandan Tutsi community in reputedly stolen from Ugandan army Uganda became more determined to stores. Though repulsed, the attack return to their own country. In 1988, the signalled the start of the civil war in Rwandan Patriotic Front (RPF), which also Rwanda: a war which was used to increase included some Hutu opposed to the the intensity of fear and hatred and raise Habyarimana regime, was formally the stakes of political survival. created in Kampala with the objective not The build-up of tension leading to the just of ensuring the return of those Tutsi in 1990 invasion by RPF forces was accom- exile, but changing the political regime in panied by many tell-tale signs, which Rwanda. Within Rwanda the position of were inadequately monitored. When the the government was weakened by a com- invasion was a fait accompli, however, it birration of deteriorating economic con- attracted considerable international ditions (resulting from the dramatic concern and reaction, both in the region decline in the intemational price of coffee, and in Europe. France and Zaire came which was Rwanda's primary export) and to the aid of the Rwandan govern- opposition to regionally based, autocratic ment. Other actors, including Belgium, rule. Habyarimana was also coming under the OAU and key regional state$, pressure from France and other donor initiated diplomatic efforts to defuse the countries to move away from the one- conflict. party system by allowing new political The initial diplomatic efforts even- parties to develop. tually led to a series of peace talks, By failing to deal with the festering initiated and led by the OAU and Tanzania refugee problem before 1990, both the and held at Arusha in Tanzania. The Rwandan and the Ugandan governments, process, begun in August L992, rcceived as well as the wider intemational considerable international attention, and community, set the stage for future con- resulted in a comprehensive settlement, the Arusha Accords, signed in August but concluded that the ongoing democrati 1993. The United Nations assumed formal sation process and the peace talks'required responsibility for overseeing the imple- their continued economic and political mentation of the Arusha Accords, includ- support, particulatly since the peace ing the provision of peace-keepers, and in agreement under negotiation would the process limited the role of the OAU produce a new governmental structure and local African states. As a consequence, that, it was hoped, would address human there was. a disjuncture between the rights abuses. In this way, donors were mediation and implementation phases held hostage to their own policies. which served to undermine the Accords. Human rights organisations and states From 19XJ onwards, civil violence were also at odds over the issue of arms against the minority Tutsi community and supplies. In L992 and 1993, the former critics of the Habyarimana regime had recommended that states (France and gtadually escalated. Observers commonly Uganda were obvious targets) cut off all linked the violence to the civil war, either arms supplies to the parties in the conflict. as retaliatory measures or as warnings to France openly defended its role and the the advancing RPF forces. However, two right of a sovereign state to support a authoritative reports - one by an indepen- beleaguered friendly government. Uganda dent International Commission of Inquiry denied any involvement in helping the (FIDH, 1993), and another Prepared for rebel army, /et its territory constituted the the UN Commission on Human Rights rear base for the RPF forces (UNCHR, 1994) - suggested a more radi- Would aid conditionality and an inter- cal and comprehensive design that fore- national arms embargo in the 199f93 shadowed events to come. Both reports period have defused the conflict and pre- determined that the killings were geno- vented the genocide? Effective use of cidal in nature and that existing authorities human rights conditionality is difficult, were deeply involved. As the war conti- requiring fine-tuned and timely interven- nued and the outlines of a Peace formula tion. Arguably, there were windows of took shape, persistent reports indicated opportunity, particularly in mid-1992, that Hutu extremists were organising and when more pressure could have been arming themselves to de-rail the peace applied to the Habyarimana regime to deal process and massacre 'internal enemies'. with the extremist forces as well as the The creation of documented structures of critical issue of impunity. Also military violence (death squads, death lists, and assistance (direct and indirect) to the later, hate propaganda inciting violence) Rwandan protagonists could have been provided warnings of a potential calibrated better with the continuing peace genocide, process, particularly in dealing with the The UN Commission on Human central issue of extremist forces who Rights took little notice of its own report. opposed the Arusha process and resultant Except for pointed diplomatic protests by Accords. Belgium, the findings of the international Though such speculations are debat- human rights inqurry were mostly buried able, firmer conclusions can be drawn away in national ministries and the UN about what did happen. By not standing system. With the partial exception of firm on human rights conditionality, Canada, no state observed the recommen- donors collectively sent the message that dation to impose strict human rights con- their priorities lay elsewhere. By permit- ditionality on aid transfers. Some states ting arms to reach the Rwandan protago- were indifferent; others were concerned nists, the possibilities for demilitarising the conflict were reduced. Arms supplies in public rallies and at o{ficial cocktail reinforced the determination of both par- parties. There were assassinations and ties.to seek a military and forceful solution organised violence. Detailed intelligence to a political conllict. They strengthened reports were passed to New York and the the RPF's ability to advance militarily. Belgian military authorities by the un- They permitted the government to equip official UNAMIR intelligence unit docu- and expand its almed forces as well as menting the military training of militias, para-military units, both of which became hidden arms caches and plans for violent involved in the genocide. action. Unequivocal warnings reached the When, as a result of the Arusha UN Secretariat on 1L |anuary 1994 regard- process, the Hutu extremists were ing a planned coup, an assault on the UN excluded from the key instruments of the forces to drive them out, provocations to transitional government and marginalised resume the civil war, and even detailed in the political process, alternative strate- plans for carrying out genocidal killings in gies were not developed to 'de-fang'those the capital. The cable was placed in a extremists. The UN force (UNAMIR) sent separate black-coloured file, designed to to oversee the implementation of the peace draw attention to the gravity of its content, agreement was given a mandate tailored and circulated to several departments in to a classic, minimalist peace-keeping the UN Secretariat. However, senior operation. Yet the force faced a situation officials in the Secretariat questioned the considered by many, including some of validity of the information and made no those who planned the operation, as contingency plans for worst-case dangerously unstable, scenarios. Similar intelligence failures As the architects of the Arusha were evident at the state level, particularly Accords had foreseen, conditions in in France and Belgium, both of which had Rwanda suggested a mandate with broad a considerable capacity for overt and powers to protect civilians and seize arms covert information gathering in Rwanda at caches. Further, the UN Security Council the time. established a force that was structured and Thus pieces of information were avail- financed to satisfy a cost-conscious United able that, if combined and analysed, States, increasingly unwilling to support would have permitted policy-makers to UN peace-keeping, rather than to meet the draw the conclusion that both political needs on the ground. The force was assassinations and genocide might occur, inadequately supported (although rela- and that the scale would be different from tiveiy rapid given the normally cumber- past patterns (1959-53; 1991-3) of 'just' some UN procedures), and only slowly hundreds or thousands of victims. Yet this deployed relative to the need for speed analysis was not done. Although some (considered essential to maintain the peace had available fragments of prescient and process). The operation had no flexibility significant information, the enormity of to respond to changing circumstances, in the genocide took virtually all by surprise. particular those caused by the crisis in The failure to anticipate planned and neighbouring Burundi in October L993. targeted mass murder was particularly In the rnonths immediately preceding significant given the political commitment the genocide, many additional signs indi- and actual involvement of the UN in cated that the implementation of the Rwanda, the legal right and moral obli- Arusha Accords was faltering and that gation to act to prevent genocide according massive violence was being planned. The to the Genocide Convention, and the air was full of extremist rhetoric on radio, enormous cost of a miscalculation. While mandated to help implement decision was made to reduce the force the peace agreement, the UN made no drastically. preparations to deal with a breakdown of Once the direction and magnitude of the Accords, except to withdraw. Nor the genocide became undeniabli:, the UN were there contingency preparations to reversed itself and accepted an obligation deal with the plans to scuttle the Accords to protect civilians. However, the realisa: or the massive violence plotted by the tion of this peace-keeping mission extremists. Generally, the UN Secretariat (UNAMIRII) was hampered by the unwill- interpreted UNAMIR's mandate and rules ingness of key mernbers tq pay for or of engagement narrowly, and on several provide troops, and to match troops with occasions denied the Force Commander equipment in an expeditious manaer. The permission to search for and seize arms force was deployed ordy after the genocide caches. When developments in early L994 and the civil war had run their course. further eroded the peace accords, the France's role in Rwanda was signifi- Secretary-General and the Security cant in its many contradictions. While Council threatened to withdraw the UN warning the Security Council in early 1993 force, hence strengthening the hands of that massacres were a real possibility, the extremists. No member of the Security France supported a regime that was Council carne forward to suggest a differ- deeply compromised by human rights ent course of action. On the contrary, the violations. France uged the UN, rather Council kept UNAMIR on a tight leash than the OAU, to take the lead in monitor- with only a three, months' authorisation, ing and implementing the peace agree- accompanied by admonitions of caution ment, but subsequently did little to sup- and emphasising the importance of port UNAMIR I. Nor did France pledge cost-cutting. support for UNAMIR II, even though the French Foreign Minister was the first cabinet member of a government holding a CRISIS AND RESPONSE: APRIL TO IUNE permanent seat on the Security Council to '1994 identify the massacres as gerrocide (16 May In the months before the crisis struck, t994). With the aid of some of its African UNAMIR's presence contributed to a false ex-colonies, France subsequently under- sense of security in Rwanda. When events took a unilateral intervention, Opdration came to a head on 6 April, the UN Turquoise, endorsed by a Chapter VII collectively failed to respond. There was Security Council resolution. The action an absence of leadership at UN head- saved many lives and undoubtedly pre- quarters in New York. The Secretary- vented an additional mass outflow of General, traveiling at a brisk pace through refugees from the south-west of Rwanda, Europe, misread the nature of the conflict. but came very late: two and a half months The understaffed and ovetstretched after the genocide commenced and when Department of Peace-keeping Operations the civil war was almost over. Further, seemed paralysed. In the Security because of France's historical role in the Council, the killing of 10 Belgian peace- country, the intervention was open to keepers created a political surge to with- misinterpretation, and did not serve to draw, although this was not recom- disarm the extremists or prevent sus- mended by UNAMIR's Force Commander pected organisers of the genocide from nor by the African countries contributing escaping. troops. Information on the genocide under During the whole conflict, but way was already available when the final especially after the coup on 6 April, the media, with some notable exceptions, actually occurring seemed beyond belief. failed to report accurately and adequately Maior states with the capacity to on a crir4e against humanity. This signifi- monitor and anticipate the crisis were cantly contributed to international lack of either not interested, or, if interested, interest in the genocide and the conse- were unwilling to undermine a friendly quent inadequate response. Too many govemment. [n such a situation, interna- reports conformed to stereotyPical notions tional organisations exist in part to pick up of tribal violence and orgies of killing the slack, but neither the UN nor the OAU rather than to the power struggle and did so in the Rwanda case. The UN had systematic slaughter of Hutu moderates as poorly developed structures for systemati- well as Tutsi that was under wav.3 cally collecting and analysing inlormation in a manner relevant to preventive diplo- macy and conflict management. The FAILURES OF EARLY WARNING newly formed inter-agency arrangement Whatever the failures in media coverage, for humanitarian early warning (HEWS) prime responsibility for the failure to read was oriented mainly towards humanitar- the signals and respond adequately cannot ian operations; it was not equipped to be placed on the media. Why were the detect or analyse political and military signals that were sent ignored? Why were warning signals. they not translated into effective conflict Within the Secretariat, information management? The failwe of early warning collection and policy analysis was divided is attributable to many factors. The UN among the Department of Political Affairs was poorly organised to collect and flag (DPA), the Department of Peace-keeping information about human rights violations Operations (DPKO) and the Department and certainly genocide. There was a failure of Humanitarian Affairs (DHA). There was in both the UN system and the NGO also a disjuncture between information community to link human rights repods to collection, analysis and the development dynamic analyses of social conflict so as to of strategic policy options. Thus DPA was provide strategic policy choices. There also assigned the responsibility for monitoring existed an internal predisposition by a events in the region, but not for develop- nurnber of the key actors to deny the ing related strategic policy options. One of possibility of genocide because facing up the most significant sources for early to it would have required them to alter warning, the UN humanrights monitoring their course of action. The mesmerisation system, was not part of the information- with the success of Arusha and the failure gathering structure in the Secretariat and, in Somalia together cast long shadows and arguably, became isolated from the distorted an objective analysis of Rwanda. decision-making process. In the field, the The vast quantity of 'noise' from other UN had no formal capacity for collecting crises worldwide preoccupied world intelligence; nevertheless, UNAMIR, leaders. The confusion between genocide through the initiatives of both the as a legal term, referring primarily to an Canadian Force Commander and the Bel- intent, and the popular association of gian Kigali-sectot Commander, succeeded genocide with massive murder in the in running small, if irregular, intelligence order of hundreds of thousands, created operations. The other main organisation confusion. Finally, a general desensitisa- concerned, the OAU, had virtually no tion developed with respect to mass capacity at all for early warning data slaughters in Rwanda and Burundi, and collection and policy analysis. the possibility of. a massive genocide Yet the shortcomings of early warning in the Rwanda case go further. The issue is ience tells us that 'prevention is better not better quantitative data or formal than cure'. The lack of international modelling. The UN lacked a system for investment in early conflict regulation drawing on existing information sources, signified a more fundamental disinterest in the region and outside, from specialists in Rwanda. The UN Security Council in state agencies, academic institutions, authorised only the most minimal peace- human rights monitoring agencies, and keeping force, and the Secretariat insisted the various agencies of the UN itself. The that UNAMIR maintain a low profile. UN lacked a specialised unit, without When the crisis struck, and it became clear operational responsibilities, for analysing that massive genocide was under way, such information and translating that there was still no effective international analysis into evolving strategic options action. that can be channelled directly to the Secretary-General. CONFLICT MANAGEMENT Both the UN and NGOs failed to relate human rights monitoring to analysis of the Throughout, some individual and collec- development of social conflict and, hence, tive actors did the most with the least to assess the direction of everits. When the under difficult or adverse circumstances. UN became involved in a peace-keeping Human rights NGOs monitored the situa- mission, the monitoring of political deve- tion. Tanzania struggled to turn the lopments was not linked with contingency Arusha process into effective preventive preparedness. Without contingency plan- diplomacy. LINAMIR I tried to function ning, the UN was left with a short time proactively, despite tight reins before 6 frame and few re$ources to respond to Aptil. Many remaining units, including sudden changes in the situation. This the ICRC, bravely sought to save civilians point is critical for two reasons: even once the killings started. This could not under the best of circumstances, it is compensate, however, for the overall impossible to pin-point specific future failure of the international community to outcomes of complex social conflict; attempt to prevent or stop the genocide, or second, the absence of contingency plan- its very inadequate efforts to mitigate it. In ning limits both what the decision-makers one sense, the inaction can be seen as a will hear and the options they are willing result of the propensity of states to be to consider. guided by narrow self-interest rather than Despite the shortcomings of early moral obligations to uphold international waming, at the critical stage the relevant norms of iustice. However, this propensity actors dealing with Rwanda knew that the has historically varied over time and place; situation was unstable and dangerous. Yet its prominence in the Rwanda case, there- the sustained and careful attention so fore, requires additional explanation. necessary to successful conllict manage- No state involved in the conflict hap- ment was lacking. [n part, early action is pened at the time to have the optimal problematic and preventive diplomacy is combination of interest, capacity and neu- inherently difficult because outcomes are trality that could have generated appro- uncertain, reflecting the typical complexity priate early warnings and translated them of cause-and-effect relations in social and into conflict-mitigation strategies. More political conflict. Moreover, policy-makers fundamentally, the Rwanda conflict who are continuously faced with actual occurred in a period when the United cises are disinclined to pay attention to Nations was acting in an expansive yet hypothetical ones, even though exper- highly selective fashion, reflecting a struc- l

tural mismatch between the responsibili- lacking. This is the conundrum. Without ties of international institutions and inter- either kind of interest, the UN as a ests of states in the post-Cold War world. collective actor was unable to mount an Revitalised by the end of the Cold adequate peace-keeping force expedi War, the UN in the 1990s rapidly tiously and cut through the Byzantine expanded its peace-keeping operations problems endemic to UN peace-keeping. throughout the world. Rwanda was added The mix-and-match system of deployment to the list in October 1993. The framework was slow and inadequate. Lacking a for peace-keeping was, however, set by powerful patron in the Security Council, the distribution of power in the Security the Rwandan operation was subject to Council, which represented the world as it cumbersome and bureaucratic procedures was half a century ago. Apart from France, that involved delays and inflexibility, and the maior powers on the Council were gave insufficient autonomy to the leader- uninterested in a small Central African ship in the field. country that was marginal to their econ- The international community might omic or political concerns and peripheral have responded better had the early warn- to international strategic rivalries. By their ing systems generated a clearer anticipa- power of veto and finances, the Perma- tion of forthcoming events. On the other nent Five controlled the peace-keeping hand, conflict management is a function of and enforcement operations of the UN. interest and capacity, not only to ensuring The only state with a demonstrated ability that information is collected and commu- to energise the Council in a crisis, the nicated, but to react. In this respect, United States, was haunted by memories regionalism appears as a critical and posi- of Somalia and determined not to get tive force that was not sufficiently recog- involved in another African conflict. It was nised or utilised. Structures of conflict also preoccupied with crises elsewhere, resolution and peace-keeping could have especially in Bosnia and Haiti. The lack of been strengthened by more involvement interest in Rwanda on the part of the maior of regional and sub-regional actors (the Western states left France to define a large OAU and the sub-regional grouping of the part of the policy field; the result was to states in the Great Lakes area) in the magnify the consequences, negative as decision and management structures. well as positive, of unilateralism. After all, these actors had definite interests Within the UN system as a whole, in the conflict and a critical stake in the there was no locus for assessing key policy outcome. Strengthening regional rnecha- questions. How, for instance, can the nisms for conflict resolution and peace- democratisation process be promoted keeping will require financial support from without exacerbating ethnic and regional richer states since most of the world's tensions or creating excuses for human conflicts occur in regions where the parties rights violations? How are extremists to be have the fewest resources to deal with controlled? Moreover, there was too little them. The consequence of these cumulative effort at policy co-ordination when oppor- fault lines in the intemational system was tunities appeared. an inability to stop or significantly mitigate a The rationale for UN peace-keeping is genocide of immense proportions. that it provides a neutral force, indepen- dent of partisan interest. However, parti- san interests can provide motivation and Notes energy to be directed at a problem when a 1.. The area which now forms Rwanda and commitment to conflict resolution per se is Burundi was administered bv German colonial authorities based in Dar-es-Salaam References between 1899 and L916. Belgium occu- pied the area during the First World War F6d6ration International Des Droits De l'Homme (FIDHX1993\ and subsequently administered the terri- Rapprt de la commis- tories under mandates first of the sion internationale d'enqu€te sur les oiolations des droits de l'homme au Rwnda depuis le 7er League of Nations and subsequently of the Octobre 1990. Paris, UN. New York, Washington, London, Ouagadougou, Montreal, 2. In part such criticisms were influenced by (1995) the Eastern Bloc and Non-aligned view that Le Carr6, ,. Our .Game. Hodder and Stoughton, London. the Tutsi political parties represented the most effective and radical opposition to Prunier, G. (1995) The Rwanda Crisis: History of a Genocide 7.959-1994. Company, Belgian colonialism. Hurst & London. 3. On 12 April, for instance, the Nal York Times United Nations Commission on Human Rights stated that the Hutu Prime Minister Agathe (UNHCRX1994I Report on Situation of ' Uwilingiyimana who was killed by hoops the Human Rights Rwanda. Addendum and militia on 7 April, in the same attack in E/CN,4/1994l7lAdd 1, 11 August 1993. which resulted in killing of 10 Belgian soldiers serving in UNAMI& was a Tutsi in an apparent attempt to mould reality to the Address for correspondence: Centre for notion that the events were simply Hutu Refugee Studies, York University, 4700 versus Tutsi. Keele Street, North York, Ontario, Canada M3V 1P3.