Journal of Humanitarian Assistance
4.1. Crisis
On 6 April at approximately 8:30 in the evening Kigali time, the plane carrying Habyarimana was shot down as it was returning from Dar-es-Salaam, where the President had finally committed himself to implement the BBTG. What followed in the next 24 hours was the almost simultaneous occurrence of a military coup, renewed civil war, systematic political assassinations, and commencement of genocide. At this stage, early warning with respect to forthcoming events was irrelevant. What was required for rational decision-making was knowledge about the confusion in the present. Who knew, what did they know, and to whom did they communicate what was known?
By the evening of 7 April, the UNAMIR command had put together some major pieces of the picture: the Prime Minister and other key moderate and opposition political leaders had been killed; the well-known extremist, Col. Bagosora, appeared to be in charge of the coup; 10 Belgian UNAMIR soldiers had been overpowered and subsequently murdered by Rwandese government soldiers who accused the Belgians of having shot down the President's plane; the RPF unit in Kigali was breaking out of its quarters while the main RPF force in the north prepared to move down through the DMZ; the militia were manning roadblocks and civilians were being killed.[68] The Force Commander knew that a military coup and politicide (systematic political assasinations) had taken place. He knew civilians were being killed, but did not recognize that genocide was under way. He considered his main objective to be safe evacuation of the expatriates [69] and prevent the resumption of the civil war, while ensuring the safety of his own troops.
With the coup a fait accompli, the immediate options were limited. New York instructed Dallaire and Booh-Booh to talk with the representatives of the "interim government" and the RPF in order to obtain a cease-fire and restart the Arusha process. The UNAMIR Force Commander also had a number of immediate tasks. With inadequate transport, he had to relocate his Ghanaian battalion to Kigali since the RPF had ordered it out of the DMZ. Escort services were provided for evacuation of expatriates, which began to be organized by the morning of 8 April. He had to deal with French, Belgian, and Italian forces who had arrived, not to assist him, but to secure the airport while evacuating their nationals. UNAMIR also provided protection to Rwandese nationals who had fled to the Amahoro stadium and a nearby hospital. In all of this, the force had only one working armoured personnel carrier, a demoralized Belgian battalion and an under-equipped, below-strength unit from Bangladesh.
Between Friday and Monday, most expatriates had managed to leave. With the evacuation problem less pressing, the UN started to focus on UNAMIR's role in the new situation (Lægreid, 1995). During the first week (April 7-13), much of the initiative lay with the Secretariat, which received information from its newly-established Situation Center, was in constant telephone contact with the Force Commander in the field, and was repeatedly asked by the Security Council to provide it with information and options.[70] Until 13 April, when Belgium formally told the Security Council of its decision to withdraw from UNAMIR, the diplomatic situation was fluid, with a range of actions available to the Secretary-General. There were daily, sometimes twice-daily informal consultations between the Security Council and the Secretariat. The week, in other words, was a window of opportunity for the Secretariat to define UN policy towards Rwanda.
The Secretariat proceeded with great caution. After discussing the situation in the Secretary-General's task force, DPKO officials suggested to the Security Council that the most likely alternatives were withdrawal or a Chapter VII enforcement action. Under Secretary-General Kofi Annan told the Council on 9 April and again on 11 April that if UNAMIR were to carry on, additional resources and a new mandate were needed. On 13 April, Assistant Secretary-General Iqbal Riza raised the question of humanitarian intervention to protect civilians, only to dismiss it. Acknowledging that at the time UNAMIR was actually trying to protect civilians in Kigali, Riza cautioned that this course of action would require more resources, and that the Council should consider whether peacekeeping operations should involve such tasks.
The Security Council commented critically on the failure of the Secretariat to come up with options outside the classic framework of Chapter VI versus Chapter VII, or a Somalia-type engagement versus withdrawal, as the British Ambassador said on 13 April .[71] One reason stemmed from the nature of the conflict. In the dynamic interaction of the response to the genocide and the management of the renewed civil war, there was a failure to distinguish between the two. Options to reinforce UNAMIR were always put by the Secretariat in terms of an enforcement operation, suggesting intervention between the two armies, rather than maintaining or increasing troop strength to protect civilians. Even two weeks after the killings commenced, the Secretary-General reported to the Security Council in terms that showed a preoccupation with the war between the Rwandese army and the RPF (S/1994/470, 20 April 1994). An option to deal with the war on civilians was never presented to the Security Council, and the latter did not strike new directions on its own. The decision to withdraw the bulk of UNAMIR on 21 April was taken accordingly in the context of a situation depicted by the Council as civil war with related "mindless violence", rather than organized genocide accompanied by a smaller civil war (Res.912(1994), 21 April 1994).[72] This misinterpretation of the situation occurred despite the earlier warnings about plans for organized, systematic killings received by the Secretariat in communications from UNAMIR and elsewhere.
Despite incomplete information during the seemingly chaotic first days after 6 April, the UN was receiving through its own channels information to the effect that the killings were both extensive and organized. In a perceptive cable, Booh-Booh, the SRSG, reported from Kigali already on 8 April that "a very well planned, organized, deliberate and conducted campaign of terror" was under way.[73] Aggressive action had been taken against the opposition leaders, Tutsi were being massacred, and UNAMIR and the RPF were attacked. There were indications of large-scale killings. On April 11 the DPKO Situation Centre relayed reports of "many thousand deaths". The Nigerian draft resolution, which was circulated in the Security Council on 13 April (but never tabled), expressed shock over the "death of thousands of innocent civilians". In the following two weeks, additional information about the scale and pattern of killings came in.
When the Secretary-General on 20 April formally presented the Security Council with options of how to respond, he failed to recognize the organized and systematic nature of the violence (S/1994/470). Instead, Boutros-Ghali saw anarchy and spontaneous slaughter. It was the same distorted picture that much of the Western media had reported in April (see 4.3 below). Given the significant role of the Secretariat at this juncture, the 20 April report merits closer attention.
Prefacing his options with a discussion of the situation in Rwanda, Boutros-Ghali claimed that "reliable reports strongly indicate that the killings were started by unruly members of the Presidential Guard" (not the Presidential Guard itself, and certainly not the extremists). Stressing the initiative of the RPF in resuming the civil war, Boutros-Ghali reported that the RPF battalion was breaking out of its compound in Kigali and "started to engage the government troops...despite the best efforts of UNAMIR", while other RPF units from the demilitarized zone were moving towards Kigali. There was no mention of an organized conspiracy, and when the killing of moderate cabinet ministers (starting at dawn the morning after the plane was shot down), of the Prime Minister, and of the 10 Belgians were mentioned, those deaths were blamed on "unruly soldiers". There was no mention of the beginning of the organized murder of Tutsi. The provisional government was depicted as disintegrating, as if it fell apart on its own. "Authority collapsed"; (it did not, at least not the authority organizing the conspiracy). Instead of stressing the need to stop the massacre of civilians, Boutros-Ghali declared that, "the most urgent task is the securing of an agreement on a cease-fire in the hopes that this would lead to the resumption of the peace process and reviving the Arusha Accords." Further, UNAMIR could not be left at risk indefinitely when there was no possibility of performing the task for which it was sent. The implication was that its sole task was to observe the cease-fire. The statement made no mention of UNAMIR's efforts on the ground to protect civilians, a subject that had been discussed in both the Secretariat and the Security Council.
A month later, on 25 May, Boutros Boutros-Ghali publicly confessed: "We are all responsible for this disaster, not only the super-powers, but also the African countries, the non-government organizations, the entire international community. There has been a genocide, and the world is talking about what it should do. It is a scandal."[74] In light of the historical record, this statement might appear as an effort to shift the responsibility around. Earlier misperceptions and inaction not only reduced the impact of the statement, but possibly also reduced the subsequent and very significant efforts by the Secretariat to reverse direction.
From 29 April and onwards, the Secretariat played a vigorous and innovative role vis-à-vis the Security Council. Its inability to produce a more differentiated set of alternatives earlier, and the near-paralysis that characterized the first week, is therefore puzzling. It is also important since early action conceivably could have prevented the critical loss of power and credibility brought on by withdrawal.
Evidently several factors were at work. There were institutional limitations. While expanding rapidly between 1990 and 1993 to keep up with burgeoning mission activities, the Department of Peace-keeping Operations in early 1994 had less than 200 officials to oversee 70,000 military and civilians in the field.[75] It had a one-person unit for policy analysis, another one-person unit for evaluation. Moreover, nowhere in the Secretariat had information from the past been accumulated, analyzed and structured to prepare policy options either for pre-emptive action or for crisis management, so important given the general inability of bureaucracy to innovate during a crisis. With limited previous experience in protecting civilians, and doctrinal poverty, the Secretariat's response was framed by the conventional dichotomy of Chapter VI versus Chapter VII action. Those charged with leadership had a distorted view of events: both the Secretariat and the Security Council drew a picture of Rwanda as a failed state in which rogue troops and spontaneous mobs were killing Tutsi. The concept of failed state, of course, suggested the analogy with Somalia, then uppermost in the consciousness of UN officials. Bureaucratic caution reinforced the conclusion drawn from that experience: the UN could not afford another peacekeeping failure, with failure defined as loss of UN peacekeepers in the field. Finally, the striking aspect of the first week of crisis was the physical absence from New York of Boutros-Ghali, who was travelling at a brisk pace in Europe and the former Soviet Union.[76] During the fast-moving and critical first days of crisis, the Secretary-General was unavailable to provide leadership for action.
4.2. Withdrawal
Until the middle of April, there were open divisions in the Security Council on the issue of UNAMIR's future. The non-aligned states, led by Nigeria, argued for strengthening UNAMIR, and on 13 April circulated a draft proposal to that effect. By then, however, it was too late and not backed by offers to contribute. The Belgian decision to withdraw its contingent, following the murder of the 10 Belgian peacekeepers, was communicated to the Secretary-General on 12 April, and to the Security Council the next day. UNAMIR was deprived of its strongest unit, and put the rest of the force in a precarious position. The Secretary-General made the point bluntly in a letter to Security-Council members on 13 April. Belgian withdrawal will make it "extremely difficult for UNAMIR to carry out its task effectively...In these circumstances, I have asked my Special Representative and the Force Commander to prepare plans for the withdrawal of UNAMIR."[77] Having decided to withdraw its own contingent, Brussels lobbied hard to persuade Council members that conditions in Rwanda necessitated withdrawal of UNAMIR as a whole. The stance was widely seen as an attempt to legitimize its own withdrawal, but the Belgians were pushing on an open door.[78] They were strongly backed by the Americans;[79] the UK and France, though less vocal, also favored withdrawal. No country came forward with troop contributions, and the Secretariat claimed later that informal canvassing at the time had produced negative results.[80] It was obviously not a question of capacity; collectively or individually, several UN members had the means to intervene decisively, as France and Belgium had shown by their efficient airborne operations to evacuate expatriates.
The assessment from the field was that, even with the Belgian withdrawal, the rest of the mission of around 1,200 should carry on as long as possible. "My decision to remain was a matter of moral concern," Dallaire later wrote.[81] With the tide running the other way in New York, however, alternate perspectives from the field were not forwarded to the Security Council in a forceful manner. Instead, the options were formulated as full withdrawal versus leaving a token force. In the "stampede to get out", as one ambassador later noted, the Nigerian draft resolution for strengthening UNAMIR was not even tabled.
While the Security Council debated options, UNAMIR's position in the field deteriorated. The Belgian contingent completed its withdrawal on 20 April. Dallaire redeployed the Ghanian battalion from the DMZ to the airport, enabling the Canadian communications unit to operate.[82] Otherwise, his force was not being re-supplied and had to work hard to secure even some equipment from the departing Belgian contingent. UNAMIR was short of water and fuel, and was taking indirect fire. Continuing negotiations with the RPF and FAR to respect a UN-supervised neutrality of the airport did not succeed, rendering the airport and hence the exit route insecure. In a self-fulfilling prophecy, UNAMIR had been weakened to the point where withdrawal increasingly seemed the most rational option. The field mission still refused to concede the point. On the eve of the Security Council's final decision on 21 April to withdraw, Dallaire's "cabinet" of officers recommended as their first option that they stay on at present strength (i.e. without the Belgian battalion), while listing scaling down and withdrawal as less desirable alternatives.
The formal decision to withdraw was taken by the Security Council in the context of a choice between three options formulated by the Secretary-General. In reality, all but one option had already been ruled out in the informal and officially unrecorded consultations in the Council and its communications with the Secretariat, probably de facto almost a week earlier. As presented by Boutros-Ghali on 20 April, the language clearly excluded the option to increase: the situation in Rwanda could only be changed by "immediate and massive reinforcement of UNAMIR... which would require several thousand additional troops and UNAMIR may have to be given enforcement powers under Chapter VII" (S/1994/470. para.13). The alternative of complete withdrawal also seemed difficult. Given the "dimensions of the violence and mass killings over the last two weeks...the consequences of complete withdrawal, in terms of human lives lost, could be very severe indeed" (para 19). That left the middle alternative of reducing UNAMIR to about 270 military personnel. These would take on a diplomatic role by promoting a cease-fire and, when feasible, assist in the resumption of humanitarian relief. On 21 April the Security Council approved this option over the recorded doubt, but with the votes, of its non-aligned members (S/PV.3368, 1994).
The two weeks following the shooting down of the plane carrying President Habyarimana and associates constituted a critical decision-making phase in the international response to the conflict. If there were any chances of stopping the genocide before it fully unfolded, this was the time.[83] Pulling out would mean abandoning the initiative to the forces on the ground. To re-enter would require additional efforts to secure the airport. The implications were fully understood by the UNAMIR officers who discussed options in the middle of the madness that was Kigali on 21 April. Reducing the force to 270 men, which the Security Council was about to decide, meant "we can do little", and "we will really not be able to come back", the Deputy Force Commander scribbled on a notepad.
The Security Council and the Secretariat saw withdrawal as a means to salvage a UN peacekeeping operation that had been tailored to a situation that no longer existed. But by largely absenting itself from the conflict, the UN simultaneously lost leverage to influence its future course - on the ground and diplomatically. External conflict management essentially came to a halt. When the UN subsequently reversed itself, re-entry proved slow, difficult and fundamentally too late.
The civilians, thousands of whom were being killed daily, were largely abandoned to their fate. A symbolic presence at key points in the Kigali area enabled UNAMIR to provide protection for an estimated 20,000 persons (at the Amahoro stadium, the Hotel Mille Collines, the Méridien Hotel and the King Faysal Hospital). Initially, a Belgian platoon effectively protected several thousand persons at the Ecole Technique Officielle; as soon as it was withdrawn, the killers closed in.
Protection of this kind relied on a combination of bluff, inertia or intersecting interests much more than a projection of force. By their presence, ICRC and MSF personnel also provided limited protective "space" in Kigali and elsewhere;[84] they were later joined by UN agencies and NGOs providing humanitarian assistance. Despite these efforts, the limited scope and vulnerability of such protection meant that, for the vast majority of civilians, security as well as humanitarian assistance came to depend primarily upon the advance of RPF troops. These came too late to save many.
Given the importance of the Belgian decision to withdraw from Rwanda, several questions have been raised in this connection. One set concerns the circumstances of the death of the 10 Belgian soldiers and how the UNAMIR command handled the matter. This has been the subject of an official Belgian investigation but remains a source of some controversy.[85] Another question is whether it was wise to have a Belgian unit in UNAMIR, given Belgium's colonial past in Rwanda and recent policies, which had caused hostility on both the government side (for failing to render sustained support during the 1990 offensive) and that of the RPF (for having tried). The Belgians clearly did not meet the formal UN requirement that peacekeepers should come from states that are neutral relative to the country where they will serve. The wisdom of sending Belgians was for that reason questioned at the time. However, no other country came forth to provide a fully equipped unit, and quick deployment, it will be recalled, was considered important.
Others have argued that it was its European and not its Belgian identity that made the contingent targeted and vulnerable to extremists. The latter clearly sought to replicate the American withdrawal from Somalia by capitalizing on the low threshold in Western states for incurring casualties in Africa. By contrast, Ghana did not propose to withdraw its battalion, nor did several African and other states whose military observers were attached to UNAMIR when the old NMOG was folded into the larger structure. For all of them, the decision to withdraw was made by the Security Council.
The small UNAMIR force that remained (it turned out to be 540) had as one of its main tasks the promotion of a cease-fire in the civil war, which had resumed alongside the genocide. For this "obsession", the UN has been severely criticized (African Rights 1995:1121). The argument is that the UN focused on the least devastating conflict - the civil war - rather than the genocide, and, by implication, sought to restrain the advance of the RPF which was the only force on the scene able to stop the mass killing of civilians. The factual points are not here at issue. The Security Council consistently demanded a cease-fire, and the UNAMIR mission in Rwanda continuously sought to negotiate one. Motivations are more difficult to assess. The rationale for demanding a cease-fire may have been politically motivated to halt the RPF, as the RPF claims.[86] It was also a habitual UN response to wars of any kind. When the decision was made in May to redeploy a larger UN force, a cease-fire was considered even more important.
How a cease-fire would have affected the level of killings is speculative. The extremists were determined and well organized; rage and fear prevailed. This suggests that only a military defeat would have put an end to the killings. On the other hand, there were divisions in the army and moderate army factions might have restrained the killers if a cease fire had been obtained. The Chief of Staff in the self-styled interim government, Gen. Bizimungu, told the UN Special Rapporteur that the government would stop the killings only after the RPF agreed to a cease-fire, implying that this was within its power. Also, a cease-fire might have reduced the fear and rage among the Hutu that accompanied the advance of the RPF and possibly intensified the killings.[87] As it was, the war and the killings continued until the middle of July, when the RPF proclaimed a new government in Kigali. For almost three months after 6 April, RPF units were the only significant force seeking to stop the killings as they advanced. The RPF consequently made a cease-fire conditional upon an end to the killings.
OAU efforts in this period also focused on the need to obtain a cease-fire and restore the Arusha process. There was no evident recognition of an organized genocide. An ambassadorial-level meeting of the Central Organ on 14 April referred to "carnage and bloodletting"; Secretary-General Salim's letter to his UN counterpart on 21 April mentioned "massacres and wanton killings". A consistent theme of OAU communications to the UN, however, was that the UN ought not to reduce the mission in Rwanda it so eagerly had sought the previous year.
4.3. The media
Had there been any forewarning in the media? Rwanda is a relatively small country on the periphery of the industrialized world. As might be expected, during the first three months of 1994, there was virtually no Western media coverage of events in Rwanda.[88] In the early part of the year, the main African story was the UN withdrawal from Somalia, which would prove prescient in itself. Massive media attention then switched to the violent build-up to the South African elections.
Surveys of the British, French and US media (Hilsum 1995, Johansson 1995, Livingston and Eachus 1995, Verschave 1994), show that relatively little change occurred in the media coverage after 6 April compared to the paucity before. There was a blip with the shooting down of the plane and the reporting on the slaughters - generally portrayed as ancient tribal feuds - but with the withdrawal of foreign personnel there was a precipitous drop in coverage. When the genocide was accelerating, the Western press virtually ceased to report on Rwanda. The lack of coverage cannot be blamed simply on the relative disinterest in Rwanda. The real danger, the genuine confusion on the ground, the restricted mobility of the reporters, and the inability to fly out photos or videos were major handicaps. In addition, American employers had ordered their reporters out for reasons of safety, and possibly also because of costs.[89] But some stayed and accurately reported events, demonstrating all the more the failure of those who did neither.
There were, by chance, two reporters in Kigali at the time, Katrin van der Schoot, a freelance Flemish reporter for Belgian radio, and Lindsey Hilsum, an experienced freelance journalist on Africa who normally worked for the BBC, The Guardian, and The Observer, but was in Kigali on _a special assignment for UNICEF. With the shooting down of the plane carrying Habyarimana, the slaughter of the moderate Prime Minister and cabinet ministers, the massacre of the 10 Belgian peacekeepers, and the subsequent murder of Tutsi in very large numbers, the number of reporters sometimes grew to a maximum of 15 (in contrast with 2,500 accredited to cover the elections in South Africa at the time).
Mark Doyle of the BBC reported on the RPF advance and the scale of the civilian slaughter by mobs, accurately depicting the murders as being instigated and led by the government and the military. Jean Hélène, the Le Monde correspondent, provided detailed and accurate background information to the Rwandese crisis. While reporting the mutual accusations levelled at each other by the two sides, the implication seemed clear that it was Habyarimana's enemies within the government and armed forces who targeted the plane. On 12 April, the fear was expressed that the Tutsi would all be systematically massacred before the RPF captured the capital.[90] Catherine Bond, The Times correspondent, reported on 12 April from Kigali that most of the killing was probably not random. Michael Taylor, director of the British Christian Aid, wrote in a letter to The Times that groups of armed youth loyal to different political parties were instigators of the violence. If he had written that they were perpetrators of violence instigated by higher-ups, this partial truth would have been more accurate. But it at least counteracted the impression of spontaneous tribal violence that was the general theme of most stories.
Such accurate or partially accurate reportage was the exception, however. The initial reporting in both The Times and New York Times had appallingly misleading reports: the downed plane was a result of a Tutsi attempt to destroy the Hutu leadership in Rwanda and Burundi; "mobs" or a troop rampage killed the Rwandese Premier and 10 Belgian soldiers; anarchy (not interahamwe with roadblocks) reigned in the streets; "rival tribal factions waged vicious street battles". On all the critical points, these early reports were wrong.[91] An interpretive piece in The New York Times on 9 April explained the events as a genocidal orgy (rather than a systematic organized genocide), a continuation of a centuries-old feud. Adding to this fictional tale of Tutsi-Hutu orgiastic killing, the paper on 12 April retribalized the murdered Prime Minister as a Tutsi.
The depth of Western stereotypical perceptions of Africa is indicated by the difficulties encountered by informed NGOs to place op-ed pieces that had a different angle. The director of the US Committee for Refugees, Roger Winter, who had worked with Rwandese refugees for years and returned from Rwanda on 4 April 1994, wrote an article emhasizing the political and organized nature of the violence. To present it as simply another case of "African tribal bloodletting", was a "fatalistically superifical interpretation", he concluded. The piece was turned down by several American papers, but eventually published in the Toronto Globe and Mail on 14 April.
US television coverage and the CNN erred on the side of vagueness, generally referring to "unspeakable atrocities", and "ethnic violence," but picked up the theme of tribal or mutual ethnic slaughter. It would not be until 7 May that ABC correspondent Ron Allen suggested that the events were not a product of spontaneous tribal violence, but were a premeditated political act intended as a final solution.
An exception to this kind of reporting was an op-ed piece by Alison des Forges of Human Rights Watch providing an in-depth background to the Rwandese crisis and indicating Western complicity and responsibility in the genocide. However, this was not the predominant Western version of events, even if it was an authoritative one. The editorial line of The New York Times followed the fictional interpretation of events and concluded that the cure for Rwanda's "blood frenzy" was the business of Rwandese.[92] The Times did the same. Even if obtaining accurate information was difficult - especially at the beginning of the crisis - more accurate analysis of the information that gradually was coming in could easily have been made by fuller consultation with Rwanda experts and human rights and refugee organizations. There was also more accurate reporting in some of the media that could have been utilized by others less well informed.
In the French press, the genocide was by early May a given, and the emphasis began to be on French responsibility for it.[93] Print and television coverage of Rwanda began to increase also in the English language press, partly because the election in South Africa had been held and Mandela was sworn in on 10 May in a relatively peaceful way, and journalists were reassigned to Rwanda. The Times had quoted Oxfam (29 April) that "the systematic killing of the Tutsi minority in Rwanda amounted to a genocide," and reported that the Kagera river filled with floating dead bodies, but the focus by 2 May was still on the unprecedented 250,000 refugees who had flooded into Tanzania in one 24-hour period.[94] The refugees became the predominant story (See Fig II).
Only when reporters reached Lake Victoria and witnessed the logjam of dead bodies floating down from the Kagera River did the Western media refocus on the massive killings. More accurate reporting on the mechanism of the genocide began to appear in the American press (Time Magazine, 18 May; Associated Press, 22 May), although Rwanda soon faded into the background, crowded out by Haiti and the O.J. Simpson story at least until Goma (See Study III).
The misleading media coverage was echoed in the accounts of events in Rwanda given by both the Security Council and the Secretary-General, explaining the withdrawal of UNAMIR on 21 April. As discussed above, these were totally distorted. More generally, the Western media's failure to report adequately on the genocide in Rwanda possibly contributed to international indifference and inaction, and hence the crime itself.