Foreword

Is this a lesson that we need to have taught to us a second time? Do we, the members of the international community, really require that more innocent women and children be slaughtered by the thousands to cause a change in our priorities and level of concern? When the sanctity of human rights can be so blatantly violated and remain tolerated by the international community, there is a problem of such seriousness that words alone cannot explain. I remain mystified that human life, the security of noncombatants, and the prevention of such horrors as the genocide in Rwanda are, sadly, not sufficient to act as a catalyst for a swift and determined response from the international community.

I often ponder the possible solutions to the many problems that the international community and the UN faced in the spring and summer of 1994, and am convinced that it is imperative that these solutions be found quickly. It would be immoral if not outright criminal to allow another tragedy to occur by failing in our collective responsibility to humanity at large. The ingredients and recipes for solutions currently exist but remain in want of a sponsor, a leader with moral determination to bring together the political, the humanitarian, and the security structures and disciplines in synergistic applications of innovative thought to this requirement to respond to human dignity rather than national self-interest.

It behooves us to take the horrible lessons of the Rwandan debacle and prevent future genocide by formalizing a pragmatic and cohesive multidisciplinary prevention capability. The killings could have been prevented if there had been the international will to accept the costs of doing so even after the politically difficult losses of peacekeeping in Somalia and the ad hoc confusion of April 1994. We need to use our processes to achieve the aim of assisting humanity, as opposed to preserving our processes at the expense of humanity. The coalition of like-minded free nations, with well-developed doctrines respecting human rights, should form the nucleus of a rapid reaction capability for the United Nations to bolster its ability to keep the peace. The looming threat of overwhelming international retribution is still required to keep in check some of the impulses of hate-filled elements. We, as the international community, must be prepared to come to the aid of humanity in a swift yet effective manner. What remains lacking, what is absent, is the will to implement such solutions. We must all strive toward this goal or continue to repress the collective guilt and wash our hands that have been stained with the blood of so many innocent victims of power-hungry and ruthless extremists.

Romeo A. Dallaire
Lieutenant-General
Force Commander, UNAMIR/UNOMUR
March 24, 1998

Acknowledgments

Preventing Genocide
How the Early Use of Force Might Have Succeeded in Rwanda

Responding to the situation that confronted him at the time, Canadian Major General Romeo Dallaire, Commander of the United Nations Assistance Mission for Rwanda (UNAMIR), commented that with troops and the right mandate, he could have prevented most of the killing.

Senators Paul Simon and James Jeffords, members of the Senate Committee on Foreign Relations, in a May 13, 1994, letter to President Bill Clinton reporting on hearings of the Subcommittee on African Affairs

Repeated attempts by the commander of UNAMIR, Major General Romeo Dallaire, and the special representative of the UN secretary-general, Jacques Roger Booh-Booh, to bring the parties back to the peace process met with failure. The situation spun out of control as UNAMIR was repeatedly weakened, first by the withdrawal of the Belgians, who openly advocated a complete withdrawal of UNAMIR,² and then by the timid response of participating nations. With the notable exception of Ghana, governments instructed their UNAMIR contingents to protect themselves at all costs, even if that meant standing by while lightly armed, drunken thugs hacked women and children to death. (Those who had cash could buy a quick death by firearm; those who could not received a less costly and less sophisticated execution by machete, stoning, or burning.³) The United Nations, stung by the intervention in Somalia, fearful of another mission of ambiguous intent, participation, and support, and hampered by the sovereignty issues raised by member states, did not take decisive action to intervene. Individual member states in a position to act also delayed unilateral measures. Within three months, UNAMIR was reduced to 450 personnel;⁴ between 500,000 and 800,000 Rwandans, mostly Tutsi, were dead; 500,000 Rwandans were displaced within the country; and over two million Rwandans had fled to surrounding countries. More human tragedy was compressed into three months in Rwanda than occurred in four years in the former Yugoslavia.

Much of the violence wracking the world since 1989 has been attributed initially to ethnic causes, rooted in immutable history, or to the unavoidable release of tension or redress of grievances held too long in check by the last vestiges of colonialism or the bipolar international structure. The conflicts in the former Yugoslavia and Rwanda, for example, have had ethnic, historical, and broad social components, but they also have had a strong, immediate political component.⁵ In these cases, the precipitating motivation for conflict stemmed from actions designed to achieve political goals. Leaders within factions steered the conflict toward violence, tapping into long-standing, deep-rooted ethnic tension as an accelerator. Later, those historical and ethnic forces surged out of control, fed by momentum, suffering, and acts of retribution. Indeed, in many cases that is what the initiators may have wanted. But if the proximate causal factors were political, then the violence began as the result of choice, and such choices can be influenced.⁶

With that in mind, the Carnegie Commission on Preventing Deadly Conflict, the Institute for the Study of Diplomacy at Georgetown University, and the United States Army undertook a project to consider whether the introduction of international military force into the situation in Rwanda in 1994 could have had any effect on the situation there and what the nature of such an intervention might have been. An international panel of distinguished senior military leaders (Appendix A) assessed General Dallaire's thesis and addressed the following questions:

• What actions could a military force have taken to forestall violence?

• How large a military force would have been needed and how should such a force have been organized, trained, and equipped?

• When was the most appropriate time for sending a military force, and what was the relationship between the timing of interventions and the kind of force required?

THE HISTORY OF INTERNATIONAL
INVOLVEMENT IN RWANDA

The United Nations Assistance Mission for Rwanda was established by the UN secretary-general and the Security Council in the fall of 1993 to oversee what many hoped would be an end to a four-year civil war. In the name of a "democratic" majority, the Hutu-dominated government in Rwanda had repeatedly reduced the rights of the Tutsi minority and their role in the society. Moreover, systematic corruption and looting of the public purse had prevented economic gain for all but a favored faction of the Hutu majority. Rather than creating serious cleavages within the Hutu group, the leaders of the country had deftly directed frustration toward the Tutsi minority.

The periodic outbursts of ethnic violence that have punctuated the long-standing, conflictive relationship between Hutu and Tutsi increased in frequency after Rwanda's independence from Belgium in the early 1960s (see Appendixes B and C). Many Tutsi fled the country during these periodic outbursts and settled in surrounding countries. In 1986 Tutsi expatriates assisted the victors in a Ugandan civil war and in 1990 began to attack the Rwandan government from bases in that country. In response, the United Nations, proceeding in optimistic incremental stages, supported, but did not lead, a long regional process designed to encourage the Hutu-dominated government to deal with the Tutsi expatriates and the moderate factions in Rwanda.

The UN attempted to support negotiations by establishing an early presence in the region with the United Nations Observer Mission Uganda/Rwanda (UNOMUR). UNOMUR's objective was to monitor the Uganda-Rwanda border from the Ugandan side and assist in the reduction of weapons traffic and violent incidents. Disagreements between the UN and Uganda over the status of forces, however, delayed UNOMUR's deployment, and the mission's operational effectiveness was overtaken by events and the deployment of UNAMIR.

Prodded by the UN, the Organization of African Unity (OAU), and surrounding countries, the government of Rwanda and the RPF finally reached a settlement in August 1993 at Arusha, Tanzania. The parties to the Arusha Peace Accords pledged a cessation of hostilities, repatriation of refugees, and installation of a new broad-based transitional government. They also called for an expanded UN presence to support implementation of the Arusha framework. Provisions were made for demobilizing many of the military forces involved, the integration of the remainder into a new army, and the reentry of the Tutsi minority into a legitimate government. The UN therefore established UNAMIR under General Dallaire, who had commanded UNOMUR. In the months that followed, UNOMUR was integrated into the emerging structure of the peacekeeping force and follow-on forces (primarily Belgian, Ghanaian, and Bangladeshi) deployed through late 1993 and early 1994. The force eventually reached a strength of about 2,500.⁸ National contingents were deployed in Kigali, where they established a weapons-free area; in the demilitarized zone in the north, to demobilize the combatants and train the new army; and with the field forces of the RPF and RGF to observe and report any movements.⁹

The peace process, however, continued to stumble. On December 28, 1993, a battalion of the RPF arrived in Kigali to provide security for members of the opposition as they took their place in the government. On January 5 President Habyarimana was sworn in, but from that point on, obstacles to progress mounted. Among other things, violent incidents occurred throughout the country, usually just as a deadline for achieving a concrete measure of progress approached. These incidents served as a pretext for both parties to balk at implementation steps, to question publicly the sincerity of their opposite number, and to strengthen their factions for continued violent struggle. The upheaval produced by an earlier event in Burundi only added to the pressure to implement the framework while simultaneously destabilizing the situation: Tutsi rebels had staged an abortive coup in October 1993, killing the elected Hutu president. The ensuing violence forced 500,000 refugees from Burundi into southern Rwanda during a drought, adding a significant issue to the number of crises in the making.

In the months leading up to the death of President Habyarimana in April 1994, UNAMIR carried out its deployment operations and, despite the continuing instability, accomplished several intermediate objectives. It established the Kigali Weapons Secure Area (KWSA) which was secured by the Belgian and Bangladeshi battalions; deployed UN military observers to monitor RPF and RGF elements along the demilitarized zone (DMZ); and positioned the Ghanaian battalion between the belligerents. Steps to demobilize the combatants and to reorganize the army in the field were to commence three months after the new transitional government was established. The UNAMIR command, however, was consumed by the daily logistics of sustaining the force itself and the RPF battalion in the capital and by a series of emergencies, such as the coup in Burundi and the resulting refugee crisis. As a result, detailed planning and implementation activities were pushed into the background. Most important, UNAMIR lacked the capability to conduct thorough intelligence analysis. The small contingent of intelligence personnel provided by the Belgians could not discern the degree of political organization or sophistication of those opposed to the peace process, nor could it assess the capabilities of the interahamwe ("those who stand together"--the irregular militia organized, trained, and equipped by units of the RGF and often led by local political leaders) or the degree of support that factions in Rwanda were receiving from outside the country.

Adding to its problems, UNAMIR's supply and sustainment situation never progressed beyond the critical point. There were no stocks of water, food, ammunition, fuels and lubricants, and repair parts, nor were there the skilled mechanics and logisticians to support the force in the field. Civilian contractors provided communications support consisting of a variety of equipment, including hand held, unsecure radios and local telephones. While national and UN bureaucracies were negotiating reimbursement rates, UNAMIR was finding that its logistics arrangements severely constrained its ability to conduct extensive operations in support of the peace process.¹⁰ Even had the command been able to develop adequate warning of impending violence and to develop plans to anticipate and head off violence, there was no guarantee that it had the capability to act in a preemptive manner or to sustain effective operations.

General Dallaire's assessment of the situation in April 1994 has since been substantiated:¹¹ the rapid spread of violence just after the death of the president was primarily a "political decapitation" of Hutu moderates and Tutsi in and around the capital. The killings, probably including that of the president, were directed by extremists within the deceased president's own party and were designed to disrupt the peace process permanently. The perpetrators carried out their attacks by direct assault on opposition targets and then by uprooting the entire population and "straining" the refugees through a system of intermittently established roadblocks. To encourage the Hutu moderates and the Tutsi to leave their communities, the state-controlled radio broadcast clearly inflammatory messages in the months leading up to April 6, demanding the expulsion of the Tutsi and exhorting the Hutu majority to fill the rivers of the country with Tutsi dead (referred to as inyenzi, or cockroaches). The inflammatory broadcasts were in the native language, while much more tempered and conciliatory broadcasts and statements were in French, obviously for foreign consumption. These broadcasts continued throughout the crisis sparked by the president's death and fueled continued killing with a constant undertone of hatred. The perpetrators intensified their series of attacks on local moderate political leaders, even those in the president's own party. Extremists in the government had obtained lists of opposition party members, provided as one of the steps toward determining proportional representation in the new legislature. Thus, throughout the winter and spring of 1994 , apparently random attacks became more focused, and reports of training, arming, and targeting by the Hutu interahamwe continued. The population, primed for flight by the radio broadcasts, began to move as village leaders and politicians from moderate opposition parties were slain and police, neighbors, and even clergy participated in the murders.

When the RPF began its offensive, the panicked Tutsi on the roads were joined by thousands of Hutu who feared reprisals. This population upheaval benefited the extremist leaders, who hid within the mass of refugees, left the country with them, and later seized control of the refugee camps in Zaire. They were supported (unintentionally, but not unknowingly) by aid from humanitarian agencies and began to train forces and plan for a counteroffensive to regain the country. The humanitarian disaster that followed dwarfed the resources of aid agencies and created a festering situation in the refugee camps that was difficult to resolve.¹²

A PROPOSED MANDATE AND INTERVENTION FORCE

In response to the April 1994 crisis in Rwanda, General Dallaire sought unsuccessfully to reverse the defensive orientation of his national contingents, obtain reinforcements, stop the genocide, and bring the parties back to the peace process. This section describes the mandate and force that General Dallaire thought sufficient to quell the violence in the country at the time and to return the participants to the Arusha peace process.¹³ It also outlines the operational plan he would have undertaken had sufficient forces been provided in April 1994. Later sections present an alternative plan based on the discussions with the other conference participants. The reports ends with a description of the characteristics of a generic force and a discussion of how and by whom such a force could be generated.

Threats to any force intervening in Rwanda in 1994 could be expected from both belligerents and armed civilians. The mass violence in Rwanda was a planned, encouraged, and systematic genocide, largely conducted by lightly armed militia and civilians occasionally assisted by members of the gendarmerie and army. This violence preceded conventional combat operations by the RPF and RGF and continued after this combat started. Other factors bearing on military operations were the role of the political parties, the refugee/displaced person crisis, the security of the humanitarian operations under way in the area, and the eventual resumption of the Arusha peace process.

The military component of the operation in Rwanda, while receiving the most attention because of the nature of the crisis, was but one aspect of a comprehensive political, diplomatic, and humanitarian approach. Conference participants noted the significant differences between the situations in Rwanda and Yugoslavia in the strategic environment and the respective directions given to UNAMIR and the Dayton Accord Implementation Force (IFOR). In Rwanda the military operation was established in an ad hoc fashion to support a predesigned political framework. UN military forces preceded UN political staff (including the special representative of the UN secretary-general) into Rwanda by several months. In contrast, the military component of the peace settlement in Yugoslavia was conceived as an integrated part of a comprehensive approach to resolving the conflict. Despite the difficulties faced by civilian agencies in meeting the timetable in the former Yugoslavia, the relationships among the military, diplomatic, and economic components of the framework were addressed in concept.¹⁴

General Dallaire's proposal for a successful intervening force in Rwanda envisioned that the strategic directive for such a force would be adopted under Chapter VII, rather than Chapter VI,¹⁵ of the UN Charter and would comprise five "decisive" or critical elements. The force would be directed to: (1) stop the genocide; (2) conduct a peace enforcement mission; (3) assist in the return of refugees and displaced persons; (4) ensure the successful delivery of humanitarian aid; and (5) assist in a cessation of hostilities. Coordination of the political, diplomatic, and economic components of the strategy would depend on fulfillment of six measures of success: (1) bringing a halt to the genocide; (2) permitting the return of refugees and displaced persons to their homes; (3) ensuring the routine safe delivery of humanitarian aid; (4) establishing a cease-fire; (5) facilitating a return of the responsible parties to the Arusha Peace Accords process; and (6) providing a secure environment for establishment of a broad-based transitional government.

Based on that strategic directive and those measures of success delineated by the competent political authority, General Dallaire proposed the following mission statement for an intervention force in Rwanda:

The Intervention Force will conduct operations in Rwanda under Chapter VII of the United Nations Charter to reestablish peace and security, thereby facilitating a return to the Peace Process of the Arusha Accords and assisting in the establishment of a Broad-Based Transitional Government.

Although the measures of success are broad, General Dallaire envisioned that failure in any one of four critical tasks would result in continued killing or a humanitarian disaster: (1) prevent genocide; (2) protect the people so they can return to their homes; (3) provide security for the flow of humanitarian aid; and (4) provide a secure environment facilitating the cessation of conventional hostilities. To accomplish these critical tasks, the military force would undertake several subordinate, supporting tasks. Rules of engagement would permit the force to take offensive action, including the use of deadly force, to prevent continued genocidal killing. In concert with the application of force where necessary, all the rear-area noncombatants would be disarmed and their weapons collected and controlled by the intervening force. The force (together with a growing UN police support detachment) then would begin to gather information and secure witnesses as a prelude to the prosecution of the perpetrators. The exact uses of this information would be determined as part of the political settlement and the arrangements made for reconciliation and judicial redress of criminal acts.

Security would be provided at specific sites for citizens threatened by violence. This probably would result in the establishment of separate sites for the ethnic groups, as retribution and retaliation could be expected. An essential element in providing security for the population would be stabilizing the movement of refugees and displaced persons because in 1994 the displacement of the target population gave the killers the opportunity to select victims as they passed through roadblocks.

The design of tactics and measures appropriate to the situation are a major consideration in planning such an intervention. The intervening force must not operate under ambiguous concepts that may carry political weight or elicit public condemnation of the mission. Each intervention situation is unique in its details and therefore requires discrete analysis and innovation to achieve success. For example, according to General Dallaire, in Bosnia there were ethnic concentrations, but in Rwanda the situation was much more homogeneous; both ethnic groups were evenly distributed throughout the country. Therefore the concept of safe havens was not as appropriate in Rwanda as in Bosnia.

A safe haven, as the term was used in Bosnia, suggested a preexisting concentration of a targeted population within a geographic area. The provision of security for the concentrated population gives the outlying, at-risk members of the ethnic group a destination toward which they can move to obtain protection. In Rwanda, where the population was thoroughly mixed, the attackers could not target certain villages or wide swaths of terrain. Killing the ethnic minority required more detailed knowledge and a more discretionary application of force. This led to the tactic of displacing whole villages and, as the moving population encountered roadblocks and checkpoints or gathered at large central buildings such as churches, separating villagers marked for death. Under these circumstances, measures to prevent people from leaving their villages would be extremely important; "safe sites," smaller and more easily defended community groupings, would be the best way to stabilize and secure the population in Rwanda.

Such a concept could have been successful. Even the greatly reduced UNAMIR force of 450 provided significant security for thousands of displaced persons in the capital area. The French in Operation Turquoise were able to significantly stabilize the situation in southwestern Rwanda in late June 1994. That intervention was not without its difficulties and detractors, but thousands were saved without the loss of any French lives.¹⁶

The military force also would protect the refugees and displaced persons returning to their homes and provide security for the receipt, storage, transport, and distribution of humanitarian assistance. Convoy escorts, patrols, and secure distribution centers would be required as the fighting drew to a close.

After establishing general order and security in the rear areas where the majority of the killing was taking place, the force would move toward the more traditional mode of peacekeeping, by dealing with the conventional belligerent forces, the RPF and RGF. In his operational plan, General Dallaire viewed the cessation of the genocidal violence as a necessary and sufficient precondition to the end of the civil war and a resumption of the Arusha peace process. In his estimation, the RPF would have no incentive or rationale to continue conventional hostilities once the mass violence had ended. Thus the intervening force had to stop the mass killings and then establish for itself a role as a "conduit" for negotiations between the military forces. This was especially important in 1994 as the two sides were much more willing to talk to military personnel than to diplomats and politicians. As a conduit for negotiations, the intervening force would provide a secure neutral environment for the reestablishment of the peace process and security for the establishment of a broad-based transitional government. As the ethnic violence decreased, the force would be prepared to delineate a new cease-fire line, establish a new demilitarized zone, and place itself between the belligerents.

RESPONSES TO THE INTERVENTION FORCE

How would the warring factions in Rwanda respond to these measures? The intervening force could meet resistance from one or more quarters upon deployment in the country. We should expect that at least one of the belligerents would be opposed to the intervention, and both might see the introduction of a robust force by third parties determined to end the violence quickly as an obstacle to achieving their ultimate goal. The intervening force should anticipate that the population and the belligerents would respond to the force in one of four ways. First, either the RPF or the RGF, or both, would oppose the intervening force with all of their military capability. Second, factions or subordinate units of either the RPF or the RGF would oppose the intervention or block the intervening force at certain times or places as it attempted to carry out its mission. Third, the force would encounter only sporadic opposition by relatively unorganized local groups. Or, fourth, the force would meet no opposition.

In the event of the first possibility--full opposition--General Dallaire thought that the intervening force would have to be led by and organized around a self-contained national contingent, with augmentation from other countries. The force would make a forced entry by air, arrive prepared for offensive combat operations, and be supported by an air bridge to Kigali airport. Only the United States could mount such an operation, General Dallaire observed, but he also noted that the possibility of full opposition would be very remote. The political situation in the country and the estimated capabilities of the RPF and RGF--they were fairly astute at estimating their own capabilities--reduced the probability of full opposition in the field.

The second possible level of opposition--blocking movement, small unit attacks, and intrusions--would be aimed at influencing negotiations or preventing the intervening force from entering certain enclaves. In such a case, a multinational force under UN direction, entering by air or land, could affect the situation. This force would be capable of combat but would not have to make a forced entry. It could be supported by a national logistic capability, and complemented by the U.S. and UN contractors once it was established on the ground.

For the third possibility--sporadic local opposition by small ad hoc, uncoordinated bands--the original UNAMIR force, backed by broad-based and public political commitment by the international community and reinforced by units capable of defensive action to protect population and facilities, would be able to achieve the force objectives. For the fourth possibility--no opposition to the force--the original UNAMIR forces with slight enhancement of certain capabilities in light weapons and mobility would suffice.

Given the nature of the belligerents and the type of fighting occurring in Rwanda in early April 1994, the most likely opposition would be periodic blocking and interference by small organized units from either the RPF or RGF. General Dallaire based his assessment on the capabilities of the two forces, their situation in the country at the time, the examples of the evacuation forces that entered Kigali in April to extract expatriates, and the French experience in Operation Turquoise. (The latter forces, operating with full weaponry and known to the belligerents to brook no interference, were not opposed.) The operation, therefore, could be orchestrated around a multinational force operating in a phased combat operation to remove the pretext for violence and return the parties to the Arusha Peace Accords timetable.

INTERVENTION PHASES

The force employment under Chapter VII of the UN Charter required five battalions of infantry and associated support, including a small number of armored personnel carriers for mobility, security, and "intimidation" value. In Phase I of the operation proposed by General Dallaire (see Figure 1), two battalions would land by air in Kigali and secure the capital and the Kigali Weapons Secure Area (KWSA). Signals intelligence units and special forces teams would locate and silence the extremist radio stations and repeaters. A third battalion would move from Uganda down the eastern side of the country from Gabiro toward Kibungo. The objectives of these forces would be to stop all violence in the capital, return the RGF and RPF forces as far as possible toward positions they held before April 6, and reinforce the demilitarized zone (or establish a revised DMZ) to ensure separation of the conventional forces.

In Phase II of the operation, the two remaining battalions would move to secure the northwestern portion of the country (Figure 2). One battalion focusing on the RGF forces would move south from Uganda through Ruhengeri to prevent the spread of violence outward from the capital. The final battalion would land by air in Kigali and move out from the capital to the west in the direction of Gitarama and Kibuye with the same objective. The battalion that secured the eastern part of the country would then move into Kigali as the force reserve. Thus, of the total five battalions in the country, two would secure Kigali and the KWSA, two would move to secure the northwest and west, and one would provide a reserve ready to move by air to reinforce any engaged force or take on other missions. Three battalions would arrive by air; two would move by ground transport into Rwanda from Uganda. Aviation assets would be positioned at Kigali Airport, and associated logistics, engineering, and support staff would begin to coordinate the security functions for the resumption of the peace process and support of the humanitarian relief effort.

In Phase III, the battalions moving west and south would continue toward Cyangugu, Gikongoro, and Butare (Figure 3). With Kigali relatively secure and the force headquarters established, four battalions would assume duties as sector headquarters. Battalions would be located with the RPF and RGF, in a newly established DMZ, and in the vicinity of the refugee camps. One battalion in the vicinity of Kigali would be designated as a reserve. In Phase IV, the priority would shift to control of refugee movement and support of the humanitarian aid effort. With the end of genocidal killings and mass violence, the conditions would be set for the implementation of a cease-fire, and force headquarters could serve as a main conduit between the military factions involved. The disarmament, demobilization, and integration training plans would be revised and implemented. In Phase V, the force would revert to operations under Chapter VI of the UN Charter--traditional peacekeeping--but would maintain its ability to respond to acts that threaten to disrupt the process or to harm the population, including positioning itself between the belligerents. In Phase VI, the force would hand over responsibility to a peacekeeping force with a more limited mandate.¹⁷

AN ALTERNATIVE OPERATIONAL PLAN

Some members of the panel took exception to the sequential nature of General Dallaire's plan.¹⁸ Their objections concerned (1) the inability of the military force in such a situation to distinguish genocidal violence from conventional civil war violence; (2) the perceptions of support for the RPF that such a sequential plan would create; and (3) the uncertainty about how long it would take the RPF to recognize that the genocide had stopped and whether the RPF would actually cease a successful military offensive in response to a halt in the genocide. Their conceptual objection was that the plan left too much initiative to the belligerents. In their view, leaving political or operational maneuvering room to the objective parties in a Chapter VII peace enforcement scenario posed unnecessary risks to the intervening force itself and placed success in the hands of those who had already demonstrated thorough disregard for the process.

Although Rwanda has been called an "ethnic conflict," even the Rwandan killers had to check identification cards to select their victims. Panelists were concerned that in a highly fluid situation, with troops who could not distinguish any particular side in the conflict and with perpetrators armed with everything from machetes to automatic weapons, members of the force would not be able to distinguish among victims and perpetrators. Classifying violence and motive would be beyond the capability of soldiers on the ground, and taking time to establish such factors would slow the mission and endanger the victims and members of the force. If the mission was to stop violence, then all violence had to be targeted by the intervening force.

The situation in Rwanda called not for action by large formations of centrally directed troops, but for small, independent units to impose "routine and habitual compliance" with specified requirements for behavior. This complex task required individual assessment, decision, and response, simple yet comprehensive rules of engagement, and practice in applying the rules.

In addition, most of the killing took place behind RGF lines and the targeted population was composed of Tutsi and moderate Hutu (the groups supporting the RPF). Thus, action to intervene against only the genocide would have been directed primarily at the government forces and militia, creating the appearance of support for the RPF. Many panelists felt that the only way to achieve objectivity and facilitate long-term success was to stop impartially all violent acts and to control movement of any faction or group, including the advance of the RPF. The panelists were especially sensitive to the paradox that the intervening force would face. In 1994 the leaders of the government forces stated that they were trying to stop the killing but that the diversion of troops to defend against the advancing RPF prevented them from stopping the genocide. The RPF was reluctant at the time to have any outside force intervene because they were winning and saw no reason to settle for a less-satisfactory negotiated settlement with the perpetrators; they had a chance to gain the entire country and bring the killers to victors' justice. Thus, in the view of some panelists, there was no guarantee that the cessation of genocide would necessarily lead to the cessation of the civil war and the rebel offensive. The intervening force, if directed in its operations to stop only the genocide, could run into the unenviable position of having its efforts viewed by the government forces as assisting the rebels, and by the rebel forces as enabling the government to devote more troops to the civil war. For all of these reasons, several panelists argued that while a more robust force (than Dallaire's proposed 5,000) was not required, a more aggressive employment concept was necessary.

Panelists made cogent arguments for the entire force to be deployed into country simultaneously, most preferably by air, with a mandate to shut down completely all acts of violence (see Figure 4). The UN would announce that forces were coming into country with the express purpose of (1) interposing themselves between the two conventional combatants, and (2) securing the capital and countryside by imposing a set of behavioral standards on all elements of the population. Those favoring this approach argued that a total of 5,000 troops would still be about right, but that the mix of units would be different. Tactical and strategic air mobility would be vital, and would require a more mobile core of infantry and helicopters. The opposition that the force might face because of its publicly stated mission was addressed by one panel member who had been in Rwanda in 1994. He had observed the reaction to the evacuation forces in early April and to the French in Operation Turquoise in June and July. It was clear to him that a determined, modern force that advertised its mission and its robust rules of engagement had no difficulty in controlling the level of violence. It was only when the extremist perpetrators sensed that the world was not going to address the crisis and that UNAMIR's contingents were in a self-protection mode that the genocide began in earnest.

Several senior officers on the panel stressed the value of introducing the intervening force and simultaneously seizing critical physical and functional assets (terrain, communications sites, politically significant buildings, and people), using overwhelming force to "shock" the participants and seize the initiative. In their judgment, there would be an inverse relationship between the timing and capability of the force and the numbers required. Fewer numbers might be required if deployed early under such a concept. While participants also generally agreed that the intervening force could provide a secure but perishable environment for a political settlement, delays in achieving political stability would require higher force requirements over time.

IDENTIFYING THE MILITARY FORCE REQUIREMENTS

Although General Dallaire's plan shared conceptual elements with the proposals of the other panelists, this section develops a force from the panelists' starting point and perspective. The planning assumptions are that the intervening force would most likely meet blocking operations conducted by factions of the fighting groups, and its units would have to conduct large-scale operations initially, followed by small-scale patrol and security operations.¹⁹

The most critical aspect of an intervention force would be the state of its training.²⁰ Upon deployment the force had to be capable of employing sophisticated tactics and modern equipment to overcome a dangerous yet ambiguous and unpredictable threat. In more conventional operations, the threat is more easily identified, and an assessment of the adversaries' plans, tactics, and goals form an important part of the military planning process that makes military operations more understandable. Units maneuver, engage, assess, and modify their actions based on outcomes related to their adversaries. In the Rwandan situation, that might have been the case only at the beginning of operations or in one geographic area. However, the Rwandan conventional conflict was intertwined with more complex and low-level actions requiring protection of the populace from either or both belligerent groups. Such situations call for a clear understanding of the overall situation, the ability to discern subtle changes in behavior to foretell the outbreak of violence, and a disciplined, confident approval that can impose order on chaos. The training required for these situations is more challenging because lower-ranking military personnel are asked to make judgments and take actions that may have consequences far beyond the average soldier's normal day-to-day responsibilities. Soldiers need knowledge of the situation and the ability to shift mental gears between full-scale, large-unit military operations and small-unit, even individual, monitoring, intervention, and assistance roles--knowledge and ability that is not incorporated into training in many military organizations. Only in the last five years have the United States and some of its allies begun to incorporate this type of training into standard unit schedules. It is labor- and resource-intensive to set up such scenarios and requires the diversion of resources from the traditional training that many military leaders consider more critical and more relevant to the military role.

Equally important, properly trained troops must be supported by a staff that can integrate military functions with the directives of the political authorities and the interested countries supporting the peace agreements (in this case the UN and the Arusha Peace Accords). Planning for such an operation and commanding and controlling the disparate and complex functions occurring simultaneously pose unique complexities not found in the traditional large-scale conventional military operations. Staffs must have an appreciation not only of the elements that normally go into military planning but also of the significant differences to be found in the criteria for success, the number and goals of the parties, the peculiar methods being used to further their goals, and different concepts of strengths and weaknesses. Most important, the planning for such an operation must take into account the increased importance and effect of day-to-day political direction.

Policymakers will never turn operations of the kind described here over to the military with the admonition to "win it and we'll sort it out later." The decentralized actions of small units and individual soldiers may have consequences seemingly out of proportion to their immediate and narrowly defined military effects. As a result, the force must expect additional oversight, scrutiny, and visibility.

Planning also must include the multinational nature of peacekeeping or peace enforcement missions. Rarely will such operations be conducted by a nation acting alone or without international support, and will almost certainly include the participation or supervision of an international organization. Therefore, planning not only must incorporate the diverse interests of policymakers who have a hand in determining the mission and resources, but also must address the diversity in the forces that may be provided.

Robust communications capabilities, civil-military operations personnel, psychological operations staff, interpreters, and an augmented intelligence analysis and distribution cell are vital elements of an effective staff for intervention operations. Communication in particular, becomes a critical element in the success of such a complex operation: all members of the mission and those observing it must be clear about the goals, means, and methods to be used and the linkages among them.²¹

The communications capabilities also must include secure, compatible systems that allow the commanders and staffs to talk to UN headquarters in New York. UNAMIR depended on contract and civilian staff for much of its communications. When all but four of those civilians departed, the force was left only with a few radios of limited range and one reliable secure communications system with which to contact UN head-quarters. In 1994, commercial radio broadcasts also played a crucial part in inciting the violence. To counter such capabilities the intervening force needed electronic support, countermeasures, and signals intelligence capability that enabled it to monitor broadcasts, locate rogue stations and repeaters, and maintain secure and uninterrupted communications for the force.

Fire support operations would include normal coordination of close air support, attack helicopters, artillery, and mortars. In Rwanda clearance of all fires would be especially difficult, and all-important terrain management (positive identification of who is occupying what ground) would require additional manpower and communications. Units could deploy with organic mortars; the use of any indirect fire weapons in such a densely populated country would be an extremely sensitive issue.

Suppression of the opposition's air defense capability would be essential to retain force mobility and to secure the air bridge to Kigali. Limited air defense assets might be required if belligerents possess combat aircraft or helicopters.

Mobility of the tactical forces and preservation of any existing infrastructure would be essential to ensuring that the population remains stable and close to its sources of food and security. Military engineers would focus on these missions with a combat engineer unit working in each battalion sector. This tactical capability would be complemented by a construction unit at force headquarters able to undertake the critical projects necessary to legitimize the presence of the force and the reestablishment of the peace process. Explosive ordnance disposal and demining surveys would have to commence immediately.

Reconnaissance and surveillance capabilities would also be essential for forces engaged in mobile operations as well as for the security of the fixed-support sites. If the force were successful in restraining population movement, then the extremist perpetrators, to find their targets, would have to move through the countryside and villages. Dismounted observation coupled with patrolling and operations to maintain a visible presence would play a large part in identifying and influencing patterns of movement. Because this activity is akin to police work, a substantial link between civilian agencies and the military would be crucial. Aviation also would play a significant role. Reconnaissance, command and liaison, transport, and attack helicopters would be very useful for gathering information, providing responsive and precise fire support, and enabling the force to achieve a significant advantage in mobility over belligerents.

The units of the intervening force should be self-contained logistically, providing their own food, fuel, ammunition, maintenance, medical support, and movement control support. Initially, the force could be supported by air from Entebbe, Uganda, or other nearby sites. Then, as the situation developed, overland transportation could be used. For movement control and medical support, the need to provide both force sustainment and refugee/humanitarian support functions would require a large commitment and a division of labor for specific units.

All personnel should be capable of chemical defense and decontamination. The use of nonlethal chemicals for crowd control or to subdue isolated small opposition elements might be feasible, but the panelists were very divided on this issue. Not only is the use of nonlethal chemicals a murky area legally and morally given the status of current treaty negotiations, it also poses operational problems. For example, the use of nonlethal chemicals such as tear gas may generate rumors and resentment and certainly would cause counterproductive feelings and misunderstandings among the population.

The capabilities outlined above can be found in a modern reinforced infantry brigade. While a typical brigade would require additional infantry to achieve the five-battalion strength as well as significant aviation, staff, and logistics reinforcement to meet the unique requirements of the situation, such a reinforced infantry brigade with its normal complement of combat and service support would provide a foundation for an effective intervention force.

For example, the Division Ready Brigade of the American 101st Airborne Division (Air Assault) (see Figure 5)²² is a capable, potent, mobile combat force that is accustomed to conducting and linking small-unit operations over a large area. It possesses the firepower, staff capability, and combat, combat support, and logistics functions required by the operational concept outlined by General Dallaire and the concepts put forth by the other conference participants. Additional infantry battalions could be added to the force in place of its organic 105-millimeter artillery and air defense units (Stinger and Avenger crews). Depending on the threat assessment by the operational commander, the number of antitank missile weapons and crew also could be reduced. These modifications would provide space in a 5,000-troop force for the additional intelligence assets, special operations forces, and civil-military operations center staff required.

The Division Ready Brigade as currently organized deploys routinely with its own aviation support and is able to establish itself in a 24-hour period over a distance of about 450 miles. This operational capability would suffice for deployments to Kigali and the surrounding area from a staging base such as Entebbe. A force of this size and composition would require approximately 90-100 strategic aircraft sorties to deploy to theater.²³

GENERATING AN INTERVENTION FORCE

The window of opportunity offering the best chance for success in Rwanda in 1994 was a small one regardless of the employment concept. The conference participants generally agreed that any action after the last week in April 1994 would have required massive amounts of force because the situation had expanded to the countryside. Yet this fleeting opportunity was not seized. Throughout the spring and summer of 1994, there was a notable lack of consensus on just what had to be done in Rwanda and how best to go about it. The rapid introduction of force presupposes some definable end to be achieved and the will to achieve that end in a reasonable amount of time. The participants thought that mandates, without a commitment of resources, are more expressions of moral outrage than political will. In the spring of 1994 several considerations combined to prevent the generation of either individual or collective political will.

First, any peacekeeping force would have to depend on sophisticated transportation and logistics capabilities, which are maintained by few nations in the world. For this operation, conference panelists observed, the participation of the United States would be crucial.²⁴ But the United States and other members of the UN Security Council were hesitant to become engaged again in Africa as they were still trying to absorb the lessons of the recent UN action in Somalia.²⁵ There, casualties, a change in operations, and political pressure prompted U.S. withdrawal. Second, in 1993 and 1994 the UN, regional organizations, and major powers were trying to devise a solution to the ongoing violence in the former Yugoslavia. The strategic situation (some called it fatigue, caution, or fear of "stretch") militated against the formation of political will to do anything in Rwanda in the spring of 1994.²⁶

The participants lamented the role of the UN in this case in particular and in peacekeeping and peacemaking in general. Formed at the end of World War II, the United Nations developed two major aims: to end colonialism and to prevent direct confrontation between the superpowers, the United States and the Soviet Union. In carrying out these two aims, the UN was markedly successful. Sovereignty was a key, inviolable concept at the UN, and deliberation became its institutional hallmark. The UN acted most effectively in slowing the actions of member states (especially those with the most capability), forestalling a precipitous deterioration in a crisis. Today, however, even in the eyes of UN officials, territorial and political integrity are not the impediment to action that they once were. Indeed, the UN finds itself primarily engaged in disputes within countries. This shift in conceptual framework reflects new demands on the institution and requires some adjustments because the UN, in the post-Cold War world, is not equipped to make or implement rapid decisions that require establishing a physical presence on the ground in a crisis. The political machinery and the logistical and financial structure do not exist to make things happen quickly. In fact, according to one conference participant, the Department of Peacekeeping Operations may not even begin contingency planning for a deteriorating situation without Security Council approval. On a more practical level, the UN contracts for transportation, communications, sustainment, and integration functions through a laborious competitive system. As one panelist described the situation, crisis staffing is ad hoc and draws from standing organizations, operating on a volunteer or differentially compensated basis.²⁷

United Nations or Regional Forces

More central than these operational difficulties is the conceptual problem faced by current UN policymakers: How to bridge the gap between Chapter VI missions (classical peacekeeping within an agreed-upon framework) and Chapter VII missions (peacemaking or enforcement). The UN currently lacks the capability to respond rapidly in concrete ways when Chapter VI missions deteriorate into situations requiring Chapter VII actions and forces. The best solution may be a "force-in-being," ready to act at the direction of political authorities. Putting such a capacity at the disposal of political authorities would remove a constraint on their ability to anticipate a crisis and direct a meaningful response with the flexibility required in a dynamic situation. Such a solution, however, may not be achievable in the short term. Developing the right forces for peacemaking is an exercise in determining the possible, which in the post-Cold War world demands consideration of a wide range of options.

One innovative (some would say radical) suggestion is that the United Nations establish a standing force for peace operations. Generating a force-in-being traditionally has been the province of sovereign states and is not an activity that countries view as an ad hoc exercise. They devote considerable intellectual and physical resources to the creation and controlled use of military forces. This author doubts that in the current strategic setting the UN can achieve the command over the resources necessary to establish a standing military force. The sovereign concerns of member states and the current structural capabilities of the UN preclude such innovation.²⁸

There may, however, be opportunities to link the UN staff with standing forces provided either by regional organizations or by individual countries. This arrangement has two advantages. First, support may be more forthcoming if the mission is performed under UN auspices and with UN direction. Second, operations directed by UN staff provide some political cover for the major force provider, especially if troops are sent to a sensitive area. Such an arrangement is a plausible compromise that builds on an existing structure and contains sufficient safeguards, rooted in sovereignty, for the forces of the contributing organization or state.

But critics might argue that the very safeguards of sovereignty are themselves the main impediments to effective use of international force. Contributor states or organizations, having developed and fielded forces to accommodate national defense requirements, are reluctant to a priori designation or dedication of forces to a supranational body without adequate guarantees about their use. These guarantees often focus on estimates of national interest and finances. Although some countries may be accused of possessing military capability in excess of national requirements, few if any are willing to use the forces created to defend national interests for the pursuit of a more diffuse, ambiguous supranational agenda. It is reasonable to assume, then, that countries providing dedicated or earmarked forces for the UN or other organizations would expect compensation. Compensation and "burden sharing" return the question to one of interest and the ability to mobilize support within a global organization for intervention in local crises whose importance for the organization's members is ill-defined.

Use of forces from regional organizations such as NATO could provide a solution when used in their own region. Such forces have a closer relationship with the parties most likely to have a stake in successful peacekeeping or peacemaking; the interests are better defined for all parties; justification for the creation and use of such forces is more obvious; and the parties have less incentive to build excess or inapplicable capabilities. Regional forces may be more familiar with the local problems and languages, have fewer problems adjusting to the climate and culture, and be able to exploit the synergy that comes from working with allies who share cultural traditions and economic interests. Secondary benefits include training, interoperability, and confidence-building measures that may exist among the participants in the force.

There are significant obstacles, however to the use of regional forces. First, the very stake in the outcome that regional participants have in a conflict may lead to misuse of the force for other agendas or introduce obstacles to its effectiveness.²⁹ Disinterested parties from distant countries are sometimes perceived by belligerents as being more important. Second, building and training such a force are costly, and confidence building and interoperability are not achieved easily. NATO is a 50-year work in progress that requires constant maintenance and remains--Yugoslavia provides but one example--subject to significant internal strains. Third, few regions have the military capability to deploy or sustain such a peacekeeping or peacemaking force for the significant period of time it takes to achieve indigenous momentum toward stability. Finally, such regional efforts may be viewed by some countries as an excuse by the developed world to wash its hands of problems (many of which stem from colonial practices) in the developing regions. The final paradox, pointed out by one panelist, is that the countries in the very regions in which a force may be required are in dire financial, social, and political straits. They would be hard-pressed to participate in intervention operations without assistance.³⁰

One solution to the problems facing regional peacekeeping forces is the creation of a "regional plus" force, such as the African Crisis Response Force, which is currently being developed.³¹ Indigenous regional forces can be trained in peacekeeping and combat skills. The capabilities they lack are transportation, intelligence, logistics, and communications, which could be furnished by another nation. Thus the "on-the-ground" resources would be provided by those who stand to gain the most from regional stability, and technical support would be furnished by other countries (even former colonial powers), which then could maintain a discreet distance from any hostilities.

Yet the "regional plus" option has its own set of difficulties. Countries in Latin America, Africa, and Asia, where such forces are proposed, may perceive "regional plus" as a way for the developed world to shift the risk of casualties to poor countries or to maintain a de facto veto over the use of regional force by retaining certain key capabilities. Evidence of an analogous perception is growing at the UN, where, for lack of resources, developing countries cannot maintain personnel in New York to work on the UN military staff. According to one conference participant, many members of the staff are officers from the West and North and may lack the perspective of the regions where they are most likely to implement their plans. Clearly, resistance to intervention is more likely if there is a perception that an operation is designed and run by a staff that has few regional representatives.

Sometimes a single nation with credibility can be as effective as, or more effective than, an international organization. Much depends on the situation and the history of the conflict. Ultimately, a force such as the one proposed for Rwanda may be in the purview of a "lead country." If questions of operational goals, force generation, deployment, employment, and engagement require a timely response by one actor, then the United States, France, and Britain, among others, must recognize that their services will be in demand. This is not to say that only Western democracies can, should, or will perform the bulk of preventive operations, but their capabilities often will be indispensable. To be sure, intervention forces that incorporate wide participation send a strong message to their intended audience that the world community has a stake in successful outcomes. But incorporation of wider diversity in the intervening force as a criterion carries a set of disadvantages, as noted, which must be weighed against the advantages. In the final analysis, a force is effective only if someone is firmly in charge, and those in charge generally have to put up a large percentage of the resources. As a result, the "sanctioning," "licensing," or underwriting of a major power to carry out UN missions has been done on a number of occasions and offers good chances for success.³² The difficulty is that this system does not move the conceptual framework for peace operations very far forward and leaves cases like Rwanda in an orphaned status unless and until the UN or other parties can persuade potential lead countries to get involved.

CONCLUSIONS OF THE CONFERENCE PARTICIPANTS

The hypothetical force described by General Dallaire--at least 5,000 strong, depending on the method of employment, and armed with the equipment and capabilities to employ and sustain a brigade in combat--could have made a significant difference in Rwanda in 1994.

Intervening forces require strategic direction, giving the commander the latitude to increase or decrease the use of force to contain the violence. Rules of engagement must be flexible and understood to permit the intervening force to respond instantly to or preempt violent acts. As one panelist noted, "Rather than really using deadly force, the most important point is to be able to do it."

In Rwanda, a window of opportunity for the employment of such a force extended roughly from about April 7 to April 21, 1994, when the political leaders of the violence were still susceptible to international influence. The rapid introduction of robust combat forces, authorized to seize at one time critical points throughout the country, would have changed the political calculations of the participants. The opportunity existed to prevent the killing, to interpose a force between the conventional combatants and reestablish the DMZ, and to put the negotiations back on track. Additional forces may have been required to solidify the initial success and maintain order.

• Generation of a force with the capabilities to intervene successfully requires the participation of a modern, sophisticated national military--in the case of Rwanda, U.S. participation would have been essential--to lead in supplying resources, carrying out critical functions, and achieving mission goals.

  For the foreseeable future, "lead countries" will provide the best means of achieving peacekeeping mission goals. These lead countries should operate under the mandate of a UN Security Council resolution, but with executive authority to determine intermediate goals, objectives, and methods.

• The roles of the UN and other international organizations must be better defined. Supranational organizations such as the UN, alliances such as NATO, and nongovernmental and private volunteer organizations are now widely involved in preventing or ameliorating conflicts. Operational definitions contained in standing documents such as Chapters VI and VII of the UN Charter, however, do not address the "gray areas" where dynamic situations require military operations in support of diplomatic efforts, or peacekeeping or peace-enforcement efforts. Functional integration of these participants (governments, transnational organizations, private volunteer organizations, and military units) and of the capabilities for generating political will, forces, and mandates will require study and articulation.

• If international organizations are to be involved in crisis situations on an operational basis, they will have to overhaul their intelligence capabilities, staff planning techniques, and decision-making procedures. An underdeveloped intelligence capability--a capability critical to the anticipation of events, formulation of effective options, and the marshaling of resources--is the most glaring deficiency.

• Efforts to create a standing peacekeeping force and the delineation of capabilities, responsibilities, and parameters surrounding the use of force in such situations show promise. Indeed, organizations such as the Baltic Battalion and the Nordic Brigade indicate that there is room to develop standing military organizations and to train them for peacekeeping and peace enforcement. For now, however, those kinds of organizations must be integrated into a larger framework that will continue to be dependent on the capabilities of a lead country.

• In the kinds of operations described in this report, the "uncertainty principle" reigns. As a member of the panel noted, "We must always understand that there are second- and third-order consequences . . . and we must consider potential unintended outcomes and account for them." Troops must go in "as part of a larger concept, with the required `civilian implementation' resourced, organized, energized, and clearly behind." The panelist then concluded, "A force inserted into this situation can quell the violence . . . and sometimes enable a political [solution] to be found, but it does not guarantee it. A military force can produce stability [for a time], but it cannot produce normalcy."

Appendix A

Brigadier General Henry K. Anyidoho
Commander
Headquarters, 2nd Infantry Brigade
Idris Barracks
Kumasi, Ghana

Major General Romeo A. Dallaire
Chief of Staff
Assistant Deputy Minister, Personnel
National Defense Headquarters
Ottawa, Canada

General (Ret.) Vigleik Eide
Personal Representative of the
Chairman-in-Office of the OSCE
Vienna, Austria

Major General Greg L. Gile
Director for Operations, J-3
U.S. Atlantic Command
Norfolk, VA, USA

Major General James T. Hill
Deputy Chief of Staff for Operations
Headquarters, United States
Forces Command
Fort McPherson, GA, USA

Jane E. Holl, Ph.D.
Executive Director
Carnegie Commission on Preventing Deadly Conflict
Washington, DC, USA

Major General (Ret.) John Arch
MacInnis (Chair)
Chief, Mine Clearance and Policy Unit
Department of Humanitarian Affairs United Nations
New York, NY, USA

Major General William Nash,
U.S. Army
Commanding General
1st Armored Division
Bad Kreuznach, Germany

Brigadier General Bruce Scott
Deputy Director, Strategy, Plans, and Policy
Headquarters, Department of the Army
Washington, DC, USA

Major General Franklin van Kappen
Military Advisor
Department of Peace-keeping
Operations
United Nations
New York, NY, USA

Brigadier General Bo Wranker
Commander
UNPREDEP
Macedonia (FYROM)

Casimir Yost
Director
Institute for the Study of Diplomacy
Georgetown University
Washington, DC, USA

Colonel Scott R. Feil, U.S. Army
Senior Service College Fellow
Institute for the Study of Diplomacy
Georgetown University
Washington, DC, USA

Appendix B

Rwanda: A Background Report

Rwanda is a landlocked sub-Saharan country about the size of Vermont, or approximately 26,000 square kilometers (see Figure B.1).³³ The terrain is composed of volcanic rock and soil, with high steep mountains in the west, descending gradually to the east through a central plateau region and then through broken and rolling hills to savannas in the far eastern portions of the country. Rwandans refer to their country as the "land of one thousand hills." Numerous small lakes and marshes dot the country, especially in the east. The highest peak (4,507 meters) is Karisimbi in the west. Kigali, the capital, is located in the center of the country at 1,430 meters; it is surrounded by low hills about 1,700 meters high. Lake Kivu is the highest lake in Africa.

The country is bounded by Uganda to the north, Lake Kivu and the Democratic Republic of Congo (formerly Zaire) to the west, Burundi to the south, and Tanzania to the east. Although Rwanda is one of the most deforested countries on the continent, it has set aside more than 10 percent of its area, primarily on its eastern and western borders, as park, nature, or game preserve. What forest remains is interspersed with communal compounds called *Irugo and associated agricultural fields scattered throughout the hills. The bulk of the population lives on the central plateau.

The high altitudes moderate the equatorial climate. The average annual temperature in Kigali is 19 degrees Celsius, and the annual rainfall averages 85 millimeters on the central plateau. The elevation, combined with the rainfall patterns, contribute to differences in vegetation: mountainous equatorial jungle in the west and savanna in the east. There are two seasons: a rainy season from October through June, and a short dry season from July through September. The rains can be torrential. The geography and climate conditions have had an effect on population distribution.

POPULATION

The 1991 national census reported a population of about 7.5 million for Rwanda, but the U.S. Central Intelligence Agency estimated that the population was significantly larger, 8.6 million, before the civil war began in 1994.³⁴ Ninety-five percent of the population lived in rural areas. Rwanda is one of the most densely populated countries in Africa, with an average of about 271 people per square kilometer (Figure B.2). The range of population density was from 820 people per square kilometer in the north (in and around Ruhengeri) to 62 in the southeastern savanna. There were pockets of extremely dense population (over 500 persons per square kilometer) in and around

Butare in the south and around Cyangugu and Gisenyi at the southern and northern ends of Lake Kivu, respectively.

The ethnic makeup of the population in 1994 was estimated to be 85-90 percent Hutu, 9-14 percent Tutsi, and 1 percent Twa. All speak the same native language, Kinyarwanda, and some also speak French. About one-half of the population was under age 15. The overall literacy rate was 56 percent; one-third of women are literate compared with two-thirds of men. While 71 percent of school-age children attended primary school in 1991, only 8 percent continued onto secondary school, and less than 1 percent went on to obtain a college or university education.

More than 20 percent of the sexually active adults in the urban areas of Rwanda are infected with HIV. By 2000, a projected 100,000 to 200,000 Rwandans will die from AIDS. Most of the other diseases afflicting the population are waterborne and parasitic. Amebic dysentery, schistosomiasis, and respiratory infections such as tuberculosis are prevalent. In 1992, 1.5 million Rwandans were without any access to health care, 2.6 million were without potable water, and 3.2 million had no sanitation facilities. The life expectancy for Rwandan males was 49 years before the civil war.

The cultural conditions and economic activities of the Tutsi (primarily herders) and Hutu (farmers), described in the next section, have led to uneven distribution of the population. To avoid tsetse fly concentrations around the lowlands, thereby preserving the health of their herds, the Tutsi have tended to live in the highlands and central plateau. The Hutu farmers and coffee and tea growers have largely settled in the lowlands where the adequate rainfall and relatively steady climatic conditions can produce up to three growing seasons a year. The rugo or household compounds that dot the hills are home to both Hutu and Tutsi.

Finally, an indication of the number of Rwandans available for military service is given by the following age distribution: ages 13-17, 459,000 men and 471,000 women; ages 18-22, 375,000 men and 387,000 women; and ages 23-32, 542,000 men and 566,000 women.

CULTURE

Despite a common language and shared religious tradition, the Hutu and Tutsi are subject to important cultural distinctions that have political consequences. These cultural distinctions reflect historical rather than ethnic or genetic cleavages. Most of these fissures in the Rwandan culture can be traced to the bases for wealth, class, and status that were apparent when Europeans arrived in Africa in the nineteenth century and that were institutionalized by colonial masters during the early twentieth century. The Twa population subgroup first inhabited the area that is now Rwanda and are still engaged in hunting and gathering activities. Hutu clans began to migrate into the area around a.d. 1000 and established an agricultural base, exploiting the rich volcanic soil. Around a.d. 1500, Tutsi herders began to move into the area in large numbers. During this relatively peaceful migration, interaction between the groups initially was based on comparative advantage and barter; cattle products and use of land were traded for agricultural products. Economic interaction, however, was followed by conquest and a stratification of society. Tutsi warlords gradually assimilated Hutu clans, and one Tutsi clan in particular began to unify the diverse small enclaves. At the same time, a Tutsi-established military and organizational capability, coupled with the remote location of Rwanda, tempered some of the exploitation and slave trading that afflicted the more accessible areas of Africa. Local settlements were mixed, with a complex web of patron-client relationships between members of clans and between clans and family groups and political organizations of neighboring rugo. These relationships could cross Tutsi-Hutu bounds, with responsibilities for farming, land management, and war and security generally, but not always, following the Tutsi-Hutu cleavages. There is evidence that movement was possible between groups, depending on the ability to acquire and hold on to cattle. Thus those Tutsi who lost their herds and reverted to agriculture moved down the social and political scale, and individual Hutu who acquired cattle could move up. Tutsi were generally war chiefs and responsible for leadership and political organization. Hutu were generally farmers and lower on the social ladder. While the somewhat feudal character of social and economic relations required Hutu to be "beholden" to Tutsi, the complicated relationships among family, clan, rugo, and ethnic group led to situations where individual Tutsi or groups of Tutsi were indebted in some ways to Hutu. There always has been significant intermarriage between the two groups, and the tracing of ethnic heritage can be arbitrary, especially considering that Rwandans normally do not carry family names. This little-known fact played an important part in investigations into the 1994 violence.

At the time of European discovery and domination in the mid- and late nineteenth century, the area that is now Rwanda was governed by the Tutsi minority through a monarch. When Europeans began to control the region, they found it convenient to govern through the existing power distribution. The Germans governed Rwanda as a colony, and, after World War I, the Belgians governed Rwanda under a League of Nations mandate and then as a UN trust. The existing divisions became identified, for good or ill, with what some would call the social Darwinist outlook prevalent in the late nineteenth century. The Tutsi were reported at the time (although this is now disputed) to be somewhat taller and lighter skinned than the Hutu, and, based on their more recent migration to the area from the north, were regarded as related to Ethiopians and more akin to Europeans on the racial scale then in vogue. The Hutu, generally shorter, more thickly muscled, and darker, fell lower on this scale. Occupational divisions between cattle raising and agriculture seemed to reinforce the European view of the herder Tutsi as noble and intelligent and the Hutu farmers as deserving peasants. Therefore, it was convenient and intellectually consistent to consider the existing distribution of wealth and power in the country as evidence of a just and "inevitable" system.

Beginning in the 1890s the Germans exploited the organizational and military capability of the Tutsi. The northern regions around Ruhengeri, with a substantial Hutu majority, were not incorporated into Rwanda proper until the Germans, assisted by Tutsi and some Hutu from the central part of the country, attacked and occupied the area, completing this process just before World War I. The Hutu there, with a more recent independent past and memory of subjugation by both Tutsi and southern Hutu, identify themselves as a distinct group and bear historical animosity to southern Hutu. This happened to be the home region of President Juvénal Habyarimana.

Awarded custody of Rwanda in the aftermath of World War I and the establishment of the League of Nations, Belgium introduced a series of identity measures and cards that citizens were required to carry to identify them as Hutu, Tutsi, or Twa. The identity cards, coupled with clear preferential treatment of Tutsi, tended over time to solidify the division of the population and reinforce citizens' identity with a particular subgroup. As the country began the slow process of modernization under European tutelage, education, jobs, and government training and positions were given to Tutsi. Because politics on the continent reflected the politics of the European administration, the Tutsi identified with the Belgians. Indeed, Belgium continued a substantial relationship with Rwanda, but Belgium's identification with the colonial past and Tutsi domination have limited its influence with the Hutu government since independence in 1962. In addition, when war broke out in 1990 between the Rwandan Government Forces (RGF) and the Rwandan Patriotic Front (RPF), the Belgians were required to withdraw because of constitutional constraints. This vacuum was filled by the French, who have come to be identified with the Hutu majority. Francophone Rwandan Hutu also were supported by the Democratic Republic of Congo, while the Tutsi-led RPF was supported by anglophone Uganda. Thus the cleavages in Rwandan society have intertwined with the regional political divisions and reflect the political interests of some of the surrounding countries as well as the local priorities of the Rwandans themselves.

Any description of the politics and culture of Rwanda would be incomplete without mention of the consumption of alcohol in that country. Beer brewing is a major economic activity in Rwanda, and the consumption of beer carries cultural significance. Beer is sipped through straws from a communal bowl at all ceremonies and is used to mark and celebrate everything from betrothals to business deals. Public intoxication is not an occasion for shame but carries connotations of wealth, power, and manhood. This has significant implications for any person or group attempting to influence or control violent behavior in Rwanda.

ECONOMY

Rwanda made significant economic gains from 1976 to 1990. Its world ranking in terms of gross domestic product climbed twelve places. At the same time the rankings of the surrounding countries stayed the same or fell by as many as 20 places. World institutions viewed Rwanda as a model developing country with a manageable debt and the capability to integrate into and progress with the global market economy. Only about 4 percent of Rwandans, however, operate within the modern economic sectors. Over 93 percent are involved in agriculture (more than in other sub-Saharan countries), and most exist at the subsistence level. The next biggest employer is the government, with 50,000 administrators at the national and local levels and 5,000 or so more in the active military.

The primary revenue-generating activity in Rwanda is the cultivation of coffee and tea. The significant tin ore mining operation there recently shut down. There is no large-scale manufacturing. Rwanda produces only small electrical appliances, a few other inexpensive consumer goods, and beer. In 1992 the percentages of GDP by sector were: agriculture, 40.6 percent; total industry, 22.4 percent (of which manufacturing accounted for 16.0 percent;) and services, 37.0 percent. With the collapse of the coffee market in the late 1980s and continued low prices for the only substantial cash-generating activity, Rwanda saw its GDP decline during the early 1990 s (Table B.1).

With an annual population growth rate of about 3.3 percent and an ability to grow only about 85 percent of its required food supply, Rwanda appears to be firmly adhering to a classic pattern: an agriculture-based developing country in a global market economy with no comparative advantage. Rwanda imports food, energy, and capital. Inheritance laws and custom require the division of family land into smaller and smaller plots for sons to cultivate. Social upheaval before 1994 displaced farmers, and the collapse of coffee markets forced more and more land into subsistence farming. Population pressure has pushed cultivation upslope, where erosion has created serious environmental problems. The upshot of these factors has been a decline in the quality of life for most Rwandans. They face uneven progress and performance in the areas of job creation, nutrition, health care delivery, sanitation, and education.

Table B.1. Decline of Rwanda's GDP, 1987-92

YearGDP	Real Growth Rate (% )
1987	0.6
1988	0.5
1989	6.0
1990	0.1
1991	3.3
1992	1.3

INFRASTRUCTURE

Rwanda has no railroads and only eight airstrips.³⁵ Of the eight airstrips, three are paved, but none is over 3,600 meters in length. Only the airport at Kigali is capable of any large volume of air traffic and servicing intertheater airlift. Rwanda has 460 kilometers of paved roads, primarily those connecting Kigali and most of the prefecture capitals. The remaining roads range from gravel and improved earth ( 1,725 kilometers), to unimproved earth (2,700 kilometers), to forest tracks. None of the roads can sustain heavy traffic by armored vehicles, and the dirt roads always are susceptible to the rains. In fact, overland transportation is so rudimentary and the transit costs through Zaire and Tanzania so high that much of Rwanda's commerce is transported by air. Military and humanitarian missions to the area have been carried out from bases in Uganda, usually Entebbe. Rwanda produces little electricity--it has a 30,000-kilowatt capacity, or 15 kilowatt-hours per capita. Buildings are usually constructed of local materials--bricks, wood, or cement--with no dominating structures in any of the towns.

Two AM radio stations (including RTLM, the Hutu extremist station) and one FM station were based in Kigali. The FM station had seven repeaters. Television is not a communications factor. One Indian Ocean INTELSAT and one SYMPHONIE satellite serve Rwanda.

MILITARY FORCES

In the early 1990s Rwanda bought small arms, mortars, and ammunition from Egypt ($6 million in small arms and ammunition in March 1992) and South Africa ($5.9 million in small arms, ammunition, and grenade launchers in October 1992 ). France has provided a significant amount of financial and other support.

As of 1994, the Rwandan Government Forces (RGF), which had battalion-sized units of between 500 and 800 personnel and companies of between 100 and 200 personnel, were organized into: a Presidential Guard (battalion size--best equipment, pay, and training); one commando battalion; one paracommando battalion; one reconnaissance battalion; one artillery battalion; eight infantry companies; and one engineering company. The heavy equipment and weapons of the RGF consisted of twenty-eight reconnaissance vehicles (twelve AML-60, sixteen UBL M- 11), sixteen armored personnel carriers (M-3), eight 81-millimeter mortars, an unknown quantity of 83-millimeter rocket launchers, and nine 105-millimeter howitzers. The military also had two CH-47 helicopters, seven SA 3-116 helicopters, and six SA-342 helicopters. No armed or attack helicopters were available. Two Britten-Norman Islander fixed-wing aircraft, two counterinsurgency aircraft, and two civil aviation aircraft also were available.

The Rwandan military was a minimally capable force, outmanned by and probably not as well trained as the Rwandan Patriotic Front (RPF), which had about 12,000 personnel. Early Western estimates put the formal organization of the RGF at about 5,200 personnel, with an additional 1,200 gendarmerie. There was no civil police force, and the gendarmerie, for all intents and purposes, was under the control of the army. (These figures, from open sources in the United States and United Kingdom, do not correspond with estimates made by the staff of the United Nations Assistance Mission for Rwanda and reports to the UN by the RPF and RGF.) In the spring of 1994, as part of the Arusha Peace Accords on the integration and demobilization of forces, the RPF and RGF reported their military personnel to be 23,000 and 20,000, respectively. During the period leading up to the crisis, the gendarmerie was expanded to between 4,000 and 6,000. The total RGF forces available from all organizations, according to the UNAMIR commander, was 28,000-30,000.

The interahamwe also constituted a significant force and a major military concern. In 1991 the government instituted a policy of providing one gun to units of every ten households. These interahamwe received some training, and many groups maintained a working relationship with local military units over a period of three years. No precise numbers of these armed "militia" exist, but the reports of the Rwandan government as part of the Arusha Peace Accords and the observations of the UN, nongovernmental organizations, and others indicated that the interahamwe numbered between 20,000 and 30,000. They were armed with a variety of weapons, ranging from clubs and machetes to hunting rifles to modern assault rifles and grenades. They appeared to be under no consistent chain of command. Sometimes they cooperated with the formal military, the RGF; sometimes they worked at the direction of local political authorities; and sometimes they reacted to the exhortations of RTLM, the Hutu extremist radio station.

Historically, foreign military support has played a large role in Rwandan conflicts. From independence, Belgium inherited a role as Rwanda's principal military supporter, but this role ended in 1990 when the civil war began. Belgium's constitution prohibits supporting parties to a conflict. The French replaced the Belgians and provided not only financial support and weapons but also training for Rwandan units, advice for the military leadership, and troops. Indeed, 300 French troops were in Rwanda in 1990 at the outbreak of the civil war. The French reduced their forces to 170 during the war, but Rwanda received additional support from Zaire--500 troops. Zaire withdrew its force, however, after allegations of abuse and lack of discipline. The 170 French soldiers were the major foreign military presence in Rwanda when the RPF offensive began in February 1993. The French then increased their strength in the country to 670. In December 1993, after the deployment of UNAMIR, all French forces were withdrawn.

The Rwandan Patriotic Front (the rebel force of Tutsi and Hutu moderates) was better trained, more experienced, and better supported than the RGF. Moreover, it had assisted in the revolt that brought Yoweri Museveni to power in Uganda and had a broader political and ethnic base than the RGF. Despite the more inclusive nature of the RPF, it operated along more disciplined lines, which can in part be attributed to its goal of taking control of the country through conventional combat. The RGF, by contrast, was poorly disciplined, had less combat experience, and was distracted through the arming and training of the interahamwe.

General Military Situation

In the spring of 1994 the most significant concentration of military forces was in and around Kigali (Figure B.3). The RGF's best troops, the Presidential Guard, the commando and reconnaissance battalions, the artillery battalion, and the aviation contingent were stationed in the city, at the airport, and at the Kanombe military base next to the airport. UNAMIR stationed the Belgian battalion in Kigali, along with the UNAMIR headquarters and the special representative of the UN secretary-general (Table B.2). A 600-man battalion of RPF troops, one of the best of the rebel force, was in the capital to begin the integration of the Presidential Guard. The rest of the

RGF forces were mostly concentrated around Ruhengeri in the northwest, monitored by a battalion of Bangladeshi troops from UNAMIR. Farther to the east, the Ghanaian battalion of UNAMIR was stationed with the RPF. The UNOMUR observers in Uganda (by then integrated with UNAMIR) were still on station ensuring that no cross-border movement went unreported. All these troops were primarily infantry. The armored and mobility equipment on both sides and in UNAMIR was in short supply, and maintenance and operator training were poor. Any outbreaks of violence were sure to be small foot-mobile actions, and the response would have to come the same way. This put a premium on intelligence. Although not included in the secretary-general's request to establish UNAMIR and not a specific part of the mandate, a small intelligence section was deployed as part of the Belgian contingent.

A significant consideration for military forces was the large percentage of the population--at times more than 1 million people--constantly on the move to avoid persecution and violence. Based on the political situation, the level of violence in any particular region, and humanitarian assistance, these flows were initiated and then reversed, degrading the economy, providing cover for violent groups, and exposing the refugees to additional exploitation and violence.

In January 1993, 300 people were killed in ethnic violence in the northwest.

This prompted the offensive by the RPF to occupy parts of the Ruhengeri and Byumba prefectures. As a result, almost 1 million people were displaced. The French then increased their forces by 300. A cease-fire was concluded in mid-March, spurring the French to withdraw their reinforcements. In the aftermath of the cease-fire and in response to international efforts, 500,000 displaced people returned to their homes. In mid-October thirty-seven government party (Mouvement republicain national pour la démocratie et le développement, MRND) supporters were killed near Ruhengeri. UN personnel in the area reported the continued operation of armed groups throughout the northwestern portion of the country. In November RPF forces attacked, ostensibly to prevent more killing. The cease-fire was reestablished, but sporadic violence by local armed groups continued up to April 6, 1994, when the president's aircraft was shot down. Much of the killing that occurred in the spring and summer of 1994 took place at roadblocks as people tried to avoid the war between the RPF and RGF or the spreading attacks by interahamwe.

Appendix C
Chronology

1926: Belgium introduces a system of ethnic identity cards.

1933: Census of the Rwandan population is carried out using identity cards that specify the ethnicity of the bearer.

1957: The Party for the Emancipation of the Hutu (PARMEHUTU) is organized.

1959: The Tutsi king dies, and Hutu use the occasion to rise and kill thousands of Tutsi. Many Tutsi flee to surrounding countries.

1962: Ruanda, now renamed Rwanda, gains independence from Belgium. More killing follows independence and more Tutsi flee the country. A Hutu majority government (PARMEHUTU is the controlling party) comes to power.

1963: Further massacres of Tutsi occur, this time in response to military attacks by Tutsi expatriates in Burundi.

1967: The massacre of Tutsi continues.

1973: Tutsi are purged from universities and other institutions, and fresh violence against the Tutsi minority breaks out. The chief of staff of the army, Juvénal Habyarimana, comes to power with a pledge to restore order. He institutes a policy of ethnic quotas for public service employment, the largest sector of the economy after agriculture; Tutsi are restricted to 9 percent of available jobs.

1975: Two years after seizing power, Habyarimana forms another movement, MRND (Mouvement republicain national pour la démocratie et le développement), and begins to funnel government largesse to his Hutu homeland in the north, excluding other Hutu and Tutsi. This continues for twenty years and is used as a wedge between the two disadvantaged groups.

1986: Exiled Rwandan Tutsi assist Yoweri Museveni in the overthrow of President Milton Obote in Uganda. They then form the Rwandan Patriotic Front (RPF) but include moderate Hutu in the organization.

1989: Coffee prices collapse on the world market, gutting the Rwandan economy and further exacerbating tensions between Hutu and Tutsi.

July 1990: Western aid donors pressure Habyarimana to accept the principle of multiparty democracy.

October 1, 1990: The RPF invades Rwanda from Uganda. Hundreds of civilian Rwandan Tutsi are later killed in reprisals for the invasion. French and Zairian troops assist the Rwandan government and push the RPF back to the border area.

October 26, 1990: At a meeting in Zaire the two sides agree to a cease-fire, but it is repeatedly violated.

November 20, 1990: The government of Rwanda and the RPF confirm the initial cease-fire and agree to the presence of observers from the Organization of African Unity (OAU).

1990-1991: The Rwandan army begins to equip and train militias and paramilitary organizations known as interahamwe ("those who stand together"). There is no physical or institutional movement toward establishing democracy. The press and opposition are controlled and intimidated, and killings of Tutsi continue in separate incidents around the country.

February 19, 1991: The OAU, UN High Commissioner for Refugees, and governments of Rwanda, Burundi, Uganda, Tanzania, and Zaire sign the Dar es Salaam accords, which provide for the voluntary repatriation of refugees.

March 29, 1991: Rwanda and the RPF sign the more comprehensive N'sele cease-fire agreement.

September 1991: Increasing cease-fire violations are addressed with the amendment of previous documents.

November 1991: Hostilities and violence increase between Rwandan Hutu and Tutsi.

March 1992: Rwandan Government Forces (RGF) sign an agreement with Egypt for $6 million in small arms, mortars, rockets, grenades, and mines. The deal is underwritten by France.

July 12, 1992: The parties agree to the deployment of a Neutral Military Observers Group to the areas between the Rwandan army and the RPF.

October 1992: The RGF signs a $5.9 million agreement with South Africa for mortars, grenades, and small-arms ammunition.

January 1993: Ten independent human rights organizations report widespread, serious human rights abuses in Rwanda.

February 8, 1993: The RPF launches a new offensive in northern Rwanda. The ongoing negotiations are suspended.

February 21, 1993: The RPF offensive is stopped outside Kigali, the capital, only with the help of French troops. The RPF controls significant territory and announces a cease-fire.

February 22, 1993: The Rwandan government announces a cease-fire. Uganda and Rwanda request UN Security Council observers along their border areas.

March 4-19, 1993: The UN dispatches a goodwill mission/fact-finding tour to Rwanda. During this period the cease-fire is formalized, peace talks resume, and the Security Council passes the request for peacekeepers to the secretary-general.

March 16, 1993: Peace talks resume in Arusha and continue until June 1993.

March 18-25, 1993: The United Nations Department of Humanitarian Affairs conducts a visit to Rwanda and prepares a request for international aid.

April 2-6, 1993: The secretary-general dispatches a technical team to assess the conditions and requirements for possible deployment of observers along the Rwanda-Uganda border.

May 8-18, 1993: The UN continues to investigate human rights abuses and prepares the political ground for the deployment of observers.

May 20, 1993: The secretary-general proposes the establishment of an observer force along the border in Uganda (United Nations Observer Mission Uganda-Rwanda, or UNOMUR).

June 14, 1993: The government of Rwanda transmits a request, signed by the government and the RPF, to the UN for the stationing of a neutral international force in Rwanda upon the signing of a peace agreement.

June 22, 1993: The Security Council passes a resolution establishing UNOMUR for a period of six months.

August 4, 1993: The government of Rwanda and the RPF sign the Arusha Peace Accords. In the agreement, President Habyarimana agrees to share power with the Hutu opposition and the Tutsi minority and establish a broad-based transitional government (BBTG) within thirty-seven days. The BBTG will exist for a period of twenty-two months and hold elections by the end of 1995. The RPF is to be merged with the Rwandan army and the Presidential Guard is to be merged with Rwandan elite forces and reduced. The agreement also calls for the deployment of an international neutral force of 2,500 to maintain peace in Rwanda.

August 11, 1993: The UN receives a special report from its mission of April 8-17 detailing widespread human rights violations and possible genocide.

August 18, 1993: UNOMUR's advance team arrives in Uganda.

August 19-September 4, 1993: The secretary-general sends a team to Rwanda to assess the requirements, such as mission needs and force size, for the force specified in the peace accords. The Security Council and the secretary-general urge the combatants to observe the Arusha Peace Accords even as deadlines for the establishment of the transitional government slip by without either side taking action. The reconnaissance element spends two weeks on the ground making its assessment. The report, however, takes an additional three to four weeks to wind its way through the UN channels in New York.

August 24, 1993: The secretary-general requests deployment of a force inside Rwanda to observe the implementation of the Arusha accords (United Nations Assistance Mission for Rwanda, or UNAMIR) and recommends that UNOMUR eventually be integrated within UNAMIR. The recommended force is to include two infantry battalions of 800 personnel each and a total of 2,500 troops.

September 30, 1993: UNOMUR becomes fully operational with eighty-one personnel. The major contingent is a company of Tunisian troops.

September 1993-March 1994: While no progress is made in implementing the accords, training of the militias continues and radio stations with ties to the government incite Hutu to kill Tutsi. Concurrently, hard-line Hutu within the major political parties isolate and villify moderates. The hard-liners eventually control the parties and target former moderate Hutu members (often including President Habyarimana) as subservient to the RPF.

October 5, 1993: The Security Council passes a resolution establishing UNAMIR for six months and provides that its mandate will expire after elections scheduled for October 1995 but no later than December 1995. The Security Council authorizes the secretary-general to deploy only one of the requested infantry battalions.

October 21, 1993: A coup d'état takes place in Burundi, and hundreds of thousands of refugees flee into Rwanda.

October 22, 1993: The UNAMIR commander, Major General Romeo Dallaire, arrives in Kigali.

October 27, 1993: A UNAMIR advance party of twenty-one personnel arrives in Kigali.

November 23, 1993: The special representative of the secretary-general, Jacques Roger Booh-Booh, arrives in Kigali.

November 30, 1993: The Belgian battalion deploys to Kigali with 403 troops. They are assigned to Kigali with the operational name of KIBAT (Kigali battalion).

December 1993: The RPF battalion of 600 troops reports to Kigali as part of the force integration under the Arusha Peace Accords. In December the first contingent of troops from Bangladesh (485) arrives.

December 10, 1993: The special representative convenes a meeting of the government of Rwanda and the RPF, after which the two parties commit themselves to establishing a broad-based transitional government by December 31, 1993.

December 20, 1993: The Security Council extends the UNAMIR mandate until June 22, 1994.

December 30, 1993: The government and the RPF fail to meet the previously agreed-upon deadline for establishing the transitional government. In a progress report to the Security Council, the secretary-general assesses the situation as extremely fragile and requests authority to deploy the second infantry battalion. The situation is such that any reduction in the planned force level or any change to its planned introduction would signal lack of interest and resolve and deepen the crisis.

January 5, 1994: The incumbent head of state, General Habyarimana, is sworn in as president of Rwanda. No further progress is made on the transitional government because of disputes over the composition of assemblies and other organizations and the deteriorating security situation.

January 6, 1994: The Security Council adopts a resolution authorizing the deployment of the second infantry battalion.

January 1994: Throughout the month the UN applies pressure to adhere to the accords. The Security Council accelerates deployment of UNAMIR forces amid intelligence reports that the interahamwe are planning mass killings of opposition leaders and Tutsi citizens. Additional troops (369) from Bangladesh arrive. The Bangladeshis are known operationally as RUTBAT (positioned in Ruhengeri with the RGF). In January and February UNAMIR prevents the delivery of four planeloads of arms for the RGF and places the weapons under the joint control of UNAMIR and the Rwandan government. In the north, the main road from Kigali to Mulundi (RPF headquarters) is mined, and the mines are removed only after significant pressure is exerted on the government.

January 11, 1994: UNAMIR reports to UN headquarters that intelligence indicates a plot is under way to kill large numbers of Tutsi in Kigali and the existence of arms caches to carry out the plan.

January 12, 1994: The special representative and the UNAMIR commander inform President Habyarimana that they know of the plot, that this would constitute a serious breach of the peace accords, and that the Security Council will be informed if the actions are carried out.

February 3, 1994: The UN Department of Peacekeeping Operations authorizes UNAMIR to assist the government of Rwanda in the recovery of illegal arms.

February 7, 10, and 13, 1994: The special representative convenes meetings with the parties after which a new deadline of February 14 is set for the establishment of transitional institutions.

February 18, 1994: After the February 14 deadline is missed, and with prodding from the Security Council, a new date of February 22 is set.

February 21, 1994: Violence breaks out in Kigali, including the assassination of a government minister, disrupting efforts to meet the February 22 deadline.

February 25-March 25, 1994: Repeated attempts to get the transitional institutions installed meet with failure. Lists of proposed assembly members fail to meet with the approval of all parties concerned.

February 28, 1994: By the end of February, 800 Ghanaian troops have arrived and are labeled operationally BYUBAT (Byumba battalion, monitoring the RPF). UNAMIR is authorized to redeploy 200 troops from the Ghanaian battalion from the demilitarized zone in the north to Kigali to stabilize the situation.

March 22, 1994: UNAMIR completes Phase II deployment with a strength of 2,539.

April 6, 1994: The presidents of Rwanda and Burundi, returning from another round of talks in Tanzania, are killed when their plane is shot down on final approach to Kigali airport. Within one hour sporadic killings begin. The airport is sealed, and UNAMIR forces are prevented from reaching the crash site.

April 7, 1994: Organized bands of Hutu extremists conduct house-to-house searches, set up roadblocks, and kill identified Hutu opposition, human rights advocates, and Tutsi. UN troops stand by and follow their "monitoring" mandate. Prime Minister Agathe Uwilingiyimana and the ten Belgian troopers sent to protect her are disarmed, tortured, and murdered.

April 8, 1994: The RPF launches an offensive in the north and moves toward Kigali to link up with the 600 RPF troops that are in the capital as part of the Arusha peace agreement. The stated RPF objective is to stop the killings. UNAMIR attempts to secure a cease-fire, protect civilians and UN staff, and provide humanitarian assistance.

April 8-12, 1994: An interim government sets up in Kigali but leaves the capital the next day as RPF forces close in. Humanitarian workers are evacuated from Kigali. As a result of the murder of the prime minister and the ten Belgian peacekeepers, Belgium announces on April 12 that it will withdraw its battalion, stating that they are exposed to unacceptable risk. The RPF demands that all foreign troops leave Rwanda.

April 15, 1994: The foreign minister of Belgium recommends that UNAMIR be disbanded and withdrawn. The secretary-general proposes to the Security Council three options for adjusting UNAMIR's mandate. Option 1 (the secretary-general's recommendation) is massive reinforcement of UNAMIR and expansion of the mandate under Chapter VII of the UN Charter to stop the killings. Option 2 is reduction of UNAMIR to 250-270 personnel, who would serve as an intermediary between the RPF and RGF. Option 3 (specifically not recommended by the secretary-general) is the complete withdrawal of UNAMIR.

April 21, 1994: Unable to reach a consensus on a program for action, the UN does not enlarge the UNAMIR mandate. Instead, reflecting the concerns and actions of the member nations on the ground, the Security Council authorizes the reduction in UNAMIR force levels to about 250.

April 30-May 17, 1994: The UN spends six weeks organizing a response to the violence. Two sticking points surface. The first, use of the word "genocide," is sensitive because it obligates the UN to intervene. The second is the composition and funding of the force entering the country. On May 17 the Security Council adopts Resolution 918 authorizing the expansion of UNAMIR, renamed UNAMIR II, to 5, 500 personnel and mandating it to provide protection to displaced persons, refugees, and civilians at risk, while supporting relief efforts.

June 22, 1994: The UN, unable to cobble together and support an appropriate African force, authorizes the French to deploy to southwest Rwanda (Operation Turquoise). The French, commencing operations with Senegalese troops on June 22 and deploying farther into southwest Rwanda on July 9, meet with mixed success in preventing killings.

June-July 1994: When it becomes clear that the French will not assist the RGF in repelling the advance of the RPF from the east, the RGF decides to withdraw to the northwest toward Goma, Zaire. The government-controlled radio continues to inflame the population and spread fear of continued killings by the Hutu and, as the RPF advances, of reprisals by the Tutsi and the RPF. The RGF and the interim government conduct this operation, covered by the mass movement of the population toward the northwest. With the population on the move, there are continued instances of military forces, interahamwe, and ordinary citizens setting up roadblocks and checkpoints, separating out targeted groups and killing them.

July 4, 1994: The RPF takes Kigali.

July 18, 1994: The RPF declares a unilateral cease-fire and the civil war aspect of the conflict ends.

July 19, 1994: The RPF forms a "government of national unity," including Hutu and Tutsi.

July 1994: Hutu continue to flee the country with estimates at Zairian border crossings exceeding 1.5 million in a two-week period. Aid agencies are faced with a dilemma of feeding some Hutu who were responsible for the genocide. Many agencies pull out when refugee camps are organized along military lines by de facto Hutu leaders, and training and arming continue in the camps. Killings and reprisals also continue in the refugee camps. Two and a half months after the Security Council has authorized a UNAMIR II force level of 5,500 troops, no additional troops are on the ground. On July 31 the French begin to withdraw Operation Turquoise forces.

Notes and References

1. UNAMIR situation reports and correspondence among personnel in the field, UN headquarters, and UN agencies are archived in the Department of Peacekeeping Operations, Lessons Learned Office, United Nations, New York. The RPF battalion was in Kigali to secure the opposition leadership during the establishment of a broad-based transitional government. It was not to be demobilized and integrated with the new army prescribed in the Arusha Peace Accords until well into the transitional period.

2. See UNAMIR situation reports. Also see Paul Noterdaeme, Permanent Representative of Belgium to the United Nations, Letter to the president of the Security Council, April 13, 1994, UN Document S/430/1994, Document 44, in The United Nations and Rwanda, 1993-1996 (New York: United Nations, 1996), pp. 258-259. This volume contains all the Security Council and General Assembly resolutions and official correspondence among members of the UN, the Arusha participants, and UN Agencies and member states.

3. First-person accounts and documentation of the violence are contained in African Rights, Rwanda: Death, Despair, and Defiance, 2d ed. (London: African Rights, 1995).

4. UNAMIR's mandate was reduced to 270 personnel but never fell below 450 because of the requirement to secure the airport as a logistical terminus and to secure almost 25,000 refugees and internally displaced persons at the Kigali stadium, the King Faisal Hospital, major hotels, and other sites.

5. Gerard Prunier, "Rwanda: Political Conflicts, not Ethnic Hatreds," Précis 3, no. 1 (Winter 1997). Prunier is also the author of The Rwanda Crisis: History of a Genocide (New York: Columbia University Press, 1995). For an analysis of the political nature of the conflict in the former Yugoslavia, see Laura Silber and Allan Little, Yugoslavia: Death of a Nation, rev. ed. (New York: Penguin Books, 1997). Senior American military officers on the ground in the summer of 1994 emphasize that the sophisticated and prepared political infrastructure that orchestrated the violence in Rwanda cannot be underestimated.

6. Max M. Kampelman and Job L. Dittberner, "Ethnic Conflicts: Old Challenges, New Dimensions," Policy Paper of the Atlantic Council of the United States, Washington, DC, 1995; and Jane E. Holl, Carnegie Commission on Preventing Deadly Conflict: Second Progress Report (Washington, DC: Carnegie Commission on Preventing Deadly Conflict, 1996).

7. It is difficult to demonstrate an empirical link between timing of intervention and reduction in the spread of conflict. Success results in a "nonevent." Practitioners and observers, however, acknowledge an intuitive attraction to the concept. "A timely intervention by a relatively small but highly trained force willing and authorized to take combat risks and representing the will of the international community could make a decisive difference in the early stages of a crisis," stated Brian Urquhart, former UN under-secretary-general for Special Political Affairs, who has extensive peacekeeping and conflict resolution experience. Quoted by Senator Carl Levin in: U.S. Senate Committee on Armed Services, Subcommittee on Coalition Defense and Reinforcing Forces, International Peacekeeping and Peace Enforcement, 103 d Cong., 1st sess., July 14, 1993, p. 38. Scholars, observers, and practitioners have differing views on the rate of the spread of violence in the Rwandan situation and the width of any intervention "window." But even those who have reservations about how much time was available to intervene do not reject the existence of a window of opportunity for action.

8. United Nations and Rwanda, pp. 14-36, and UNAMIR situation reports.

9. UNAMIR situation reports; interview with Major General Romeo Dallaire, Canadian Forces headquarters, Ottawa, December 16, 1996; and General Dallaire's presentation at a panel discussion on military options in Rwanda in the spring and summer of 1994, Georgetown University, Washington, DC, January 23, 1997.

10. Ibid.

11. Steering Committee of the Joint Evaluation of Emergency Assistance to Rwanda, The International Response to Conflict and Genocide; Lessons from the Rwanda Experience, 4 vols. (Stockholm: Steering Committee of the Joint Evaluation of Emergency Assistance to Rwanda, 1996); Prunier, "Rwanda: Political Conflicts, not Ethnic Hatreds" and Rwanda Crisis.

12. Charles Krauthammer, "Fear and Food," Washington Post, November 22, 1996; and Chester A. Crocker, "All Aid Is Political," New York Times, November 21, 1996. Also see Lieutenant General Daniel Schroeder, "Operation Support Hope, 1994: After Action Review," Headquarters, U.S. European Command; and Stephen Metz, "Disaster and Intervention in Sub-Saharan Africa: Learning from Rwanda," Strategic Studies Institute, U.S. Army War College, Carlisle Barracks, Pa., September 9, 1994.

13. This section is adapted from a briefing presented by General Dallaire at the January 23 conference at Georgetown University. Subsequent concepts are reported from conference participants, who participated on a nonattribution basis.

14. Alan J. Kuperman, "The Other Lesson of Rwanda: Mediators Sometimes Do More Harm than Good," SAIS Review 16, no. 1 (Winter/Spring 1996), pp. 221-240, provides a very interesting look at the perceptions and actions taken by the parties and the outside participants in Rwanda. The combination of pressures and the differences in perceived costs and alternatives between the mediators and the extremists contributed to the establishment of measures that went far enough to threaten the ruling elites but not far enough to prevent them from resorting to violence to retain their power.

15. Chapter VI of the UN Charter, "Pacific Settlement of Disputes," enjoins the parties to an international dispute to "first of all, seek a solution by negotiation, enquiry, mediation, conciliation, arbitration, judicial settlement, resort to regional agencies or arrangements, or other peaceful means." The UN Security Council can recommend or implement supportive actions, such as the establishment of UNOMUR or UNAMIR, when one or more of the parties report failure in their peaceful efforts to resolve the dispute. Chapter VII, "Action with Respect to Threats to the Peace, Breaches of the Peace, and Acts of Aggression," provides a basis for greater action. Article 42 enables the Security Council to "take such action by air, sea, or land forces as may be necessary to maintain or restore international peace and security." For the development of the mandate that authorized UNAMIR, see United Nations and Rwanda , Document 19 (letter from the permanent representative of Tanzania transmitting the Arusha Peace Accords), pp. 169-201, especially the role envisioned for the "Neutral International Force," pp. 193-198; Document 21, pp. 217-220; Document 23 (the secretary-gene