President Juvenal Habyarimana, nearing the end of two decades in power, was losing popularity among Rwandans when the RPF attacked from Uganda on October 1, 1990. At first Habyarimana did not see the rebels as a serious threat, although they stated their intention to remove him as well as to make possible the return of the hundreds of thousands of Rwandan refugees who had lived in exile for a generation. The president and his close colleagues decided, however, to exaggerate the RPF threat as a way to pull dissident Hutu back to his side and they began portraying Tutsi inside Rwanda as RPF collaborators. For three and a half years, this elite worked to redefine the population of Rwanda into “Rwandans,” meaning those who backed the president, and the “ibyitso” or “accomplices of the enemy,” meaning the Tutsi minority and Hutu opposed to him.

In the campaign to create hatred and fear of the Tutsi, the Habyarimana circle played upon memories of past domination by the minority and on the legacy of the revolution that overthrew their rule and drove many into exile in 1959. Singling out most Tutsi was easy: the law required that all Rwandans be registered according to ethnic group. Residents of the countryside, where most Rwandans lived, generallyknew who was Tutsi even without such documentation. In addition, many Tutsi were recognizable from their physical appearance.

But shattering bonds between Hutu and Tutsi was not easy. For centuries they had shared a single language, a common history, the same ideas and cultural practices. They lived next to one another, attended the same schools and churches, worked in the same offices, and drank in the same bars. A considerable number of Rwandans were of mixed parentage, the offspring of Hutu-Tutsi marriages. In addition, to make ethnic identity the predominant issue, Habyarimana and his supporters had to erase—or at least reduce—distinctions within the ranks of the Hutu themselves, especially those between people of the northwest and of other regions, those between adherents of different political factions, and those between the rich and the poor.

From the start, those in power were prepared use physical attacks as well as verbal abuse to achieve their ends. They directed massacres of hundreds of Tutsi in mid-October 1990 and in five other episodes before the 1994 genocide. In some incidents, Habyarimana’s supporters killed Hutu opponents—their principal political challengers—as well as Tutsi, their declared ideological target.

Habyarimana was obliged to end his party’s monopoly of power in 1991 and rival parties sprouted quickly to contend for popular support. Several of them created youth wings ready to fight to defend partisan interests. By early 1992, Habyarimana had begun providing military training to the youth of his party, who were thus transformed into the militia known as the Interahamwe (Those Who Stand Together or Those Who Attack Together). Massacres of Tutsi and other crimes by the Interahamwe went unpunished, as did some attacks by other groups, thus fostering a sense that violence for political ends was “normal.”

Through attacks, virulent propaganda, and persistent political manoeuvering, Habyarimana and his group signficantly widened divisions between Hutu and Tutsi by the end of 1992. During 1993 a dramatic military advance by the RPF and a peace settlement favorable to them—which also stipulated that officials, including the president, could be prosecuted for past abuses—confronted Habyarimana and his supporters with the imminent loss of power. These same events heightened concerns among a broader group of Hutu, including some not previously identified with Habyarimana. Increasingly anxious about RPF ambitions, this growing group was attracted by the new radio Radio Télévision Libre des Mille Collines (RTLM) and by a movement called Hutu Power, which cut across party lines and embodied the ethnic solidarity that Habyarimana had championed for three years. In late October, Tutsi soldiers in neighboring Burundi seized and murdered the Hutupresident, freely and fairly elected only months before. In massacres touched off by the assassination, tens of thousands of Burundians died, both Hutu and Tutsi. The crime, energetically exploited by RTLM, confirmed the fears of many Rwandan Hutu that Tutsi would not share power and swelled the numbers supporting Hutu Power.

Meanwhile the Habyarimana circle was preparing the organization and logistics to attack the minority. During 1993, some loyalists from Habyarimana’s party expanded the recruitment and training of the Interahamwe. But others, perhaps concerned that the militia were too tainted by partisan rivalries, proposed a “civilian self-defense force” which was to recruit young men through administrative rather than party channels. The recruits were to be trained by former soldiers or communal police who would direct them in attacking the “enemy” in their communities. In early 1993, Col. Théoneste Bagosora sketched out elements of the program in his appointment book, the intellectual Ferdinand Nahimana advocated such a force in a letter to friends and colleagues, and administrators began preparing lists of former soldiers who could command its ranks.

Soldiers and political leaders distributed firearms to militia and other supporters of Habyarimana in 1993 and early 1994, but Bagosora and others concluded that firearms were too costly to distribute to all participants in the “civilian self-defense” program. They advocated arming most of the young men with such weapons as machetes. Businessmen close to Habyarimana imported large numbers of machetes, enough to arm every third adult Hutu male.

Aware of these preparations, the RPF anticipated further conflict. They too recruited more supporters and troops and, in violation of the peace accords, increased the number of their soldiers and firearms in Kigali. They understood the risk that renewed combat would pose to Tutsi, particularly those who had come out publically in support of the RPF in the preceding months, and warned foreign observers to this effect.

By late March 1994, Hutu Power leaders were determined to slaughter massive numbers of Tutsi and Hutu opposed to Habyarimana, both to rid themselves of these “accomplices” and to shatter the peace agreement. They had soldiers and militia ready to attack the targeted victims in the capital and in such outlying areas as Cyangugu in the southwest, Gisenyi in the northwest and Murambi in the northeast. But elsewhere they had not completed the arrangements. In the center of the country, they had successfully disseminated the doctrine of Hutu Power, but they were unsure how many ordinary people would transform that ideology intoaction. In other areas, particularly in the south, they had not won large numbers of supporters to the idea, far less organized them to implement it.

On April 6, the plane carrying President Habyarimana was shot down, a crime for which the responsibility has never been established. A small group of his close associates—who may or may not have been involved in killing him—decided to execute the planned extermination. The Presidential Guard and other troops commanded by Colonel Bagosora, backed by militia, murdered Hutu government officials and leaders of the political opposition, creating a vacuum in which Bagosora and his supporters could take control. Soldiers and militia also began systematically slaughtering Tutsi. Within hours, military officers and administrators far from the capital dispatched soldiers and militia to kill Tutsi and Hutu political leaders in their local areas. After months of warnings, rumors and prior attacks, the violence struck panic among Rwandans and foreigners alike. The rapidity of the first killings gave the impression of large numbers of assailants, but in fact their impact resulted more from ruthlessness and organization than from great numbers.

The genocide was not a killing machine that rolled inexorably forward but rather a campaign to which participants were recruited over time by the use of threat and incentives. The early organizers included military and administrative officials as well as politicians, businessmen, and others with no official posts. In order to carry through the genocide, they had to capture the state, which meant not just installing persons of their choice at the head of the government, but securing the collaboration of other officials throughout the system.

Bagosora and his circle sought first to obtain the backing, or at least the acquiescence, of the majority of military commanders. They began negotiating for this support even as troops under their command slaughtered civilians in the streets. Bagosora’s first proposal, to take power in his own right, was rejected by a number of influential officers as well as by the ranking representative of the United Nations in Rwanda. But his next move, to install a regime of extremists masquerading as a legitimate government, was accepted by the soldiers, the U.N. representative, and the international community. The day after Habyarimana’s death, the RPF renewed combat with the government forces, a response to the continuing attacks by the Rwandan army on civilians and on RPF headquarters. With the resumption of the war and the ensuing pressure for solidarity, officers opposed to Bagosora found it increasingly difficult to challenge his actions.

As the new leaders were consolidating control over military commanders, they profited enormously from the first demonstration of international timidity. U.N. troops, in Rwanda under the terms of the peace accords, tried for a few hours tokeep the peace, then withdrew to their posts—as ordered by superiors in New York—leaving the local population at the mercy of assailants. Officers opposed to Bagosora realized that a continuing foreign presence was essential to restricting the killing campaign and appealed to representatives of France, Belgium and the U.S. not to desert Rwanda. But, suspecting the kind of horrors to come, the foreigners had already packed their bags. An experienced and well-equipped force of French, Belgian, and Italian troops rushed in to evacuate the foreigners, and then departed. U.S. Marines dispatched to the area stopped in neighboring Burundi once it was clear that U.S.citizens would be evacuated without their help. The first impression of international indifference to the fate of Rwandans was confirmed soon after, when the Belgians began arranging for the withdrawal of their troops from the U.N. peacekeeping force. Ten of these soldiers, a contingent different from those of the evacuation expedition, had been slain and, as the organizers of the violence had anticipated, the Belgian government did not want to risk any further casualities.

Against the backdrop of Rwandan military acquiescence and foreign flight, Bagosora and his circle moved to recruit administrators and political leaders for the killing campaign. They expected and received support from politicians, prefects and burgomasters associated with Habyarimana’s party, but to expand the killing campaign more broadly they needed the collaboration also of administrators and local leaders from the other parties, those that were predominant in central and southern Rwanda. Adherents of these parties, stunned by the murder of their Hutu colleagues in the first days, were ready to oppose soldiers and militia whom they believed to be fighting to restore exclusive control to Habyarimana’s party. The new authorities hurried to dispel these concerns in a meeting of prefects on April 11 and through radio appeals for Hutu unity broadcast by the minister of defense and influential politicians on April 12. They stressed that partisan interests must be put aside in the battle against the common enemy, the Tutsi.

By April 15, it was clear that the U.N. Security Council would not order the peacekeepers to try to stop the violence and might even withdraw them completely. By this date, the organizers of the genocide had also expanded their ranks considerably and were strong enough to remove opponents and impose compliance with the killing campaign. On April 16 and 17, they replaced the military chief of the staff and the prefects best known for opposing the killings. One prefect was later imprisoned and executed and the other was murdered with his family. Three burgomasters and a number of other officials who sought to stop the killings were also slain, either by mid-April or shortly after. The leaders of the genocide held meetings in the center and south of the country to push hesitant local administrators into collaboration. At the same time, they sent assailants from areas where slaughter was well under way into those central and southern communes wherepeople had refused to kill and they used the radio to ridicule and threaten administrators and local political leaders who had been preaching calm.

By April 20, two weeks after the plane crash, the organizers of the genocide had substantial, although not yet complete, control of the highly centralized state. The administration continued to function remarkably well despite the disruptions in communication and transport caused by the war. Orders from the prime minister were handed down to the prefect, who passed them on to the burgomasters, who called local meetings throughout the communes where they read the instructions to the population. The same language echoed from north to south and from east to west, calling for “self-defense” against “accomplices.” Slaughter was known as “work” and machetes and firearms were described as “tools.” Reports on the situation at the local level and minutes of meetings held by people out on the hills were handed back up through the administrative channels.

By appropriating the well-established hierarchies of the military, administrative and political systems, leaders of the genocide were able to exterminate Tutsi with astonishing speed and thoroughness. Soldiers, National Police (gendarmes), former soldiers, and communal police played a larger part in the slaughter than is generally realized. In addition to leading the first killings in the capital and in other urban centers, soldiers and National Police directed all the major massacres throughout the country. Although usually few in number at sites of massive killing, their tactical knowledge and their use of the weapons of war, including grenades, machine guns, and even mortars, contributed significantly to the death tolls in these massacres. It was only after the military had launched attacks with devastating effect on masses of unarmed Tutsi that civilian assailants, armed with such weapons as machetes, hammers, and clubs, finished the slaughter. In addition, the military encouraged and, when faced with reluctance to act, compelled both ordinary citizens and local administrators to participate in attacks, even travelling the back roads and stopping at small marketplaces to deliver the message.

The administrators were charged with driving Tutsi from their homes and gathering them at places of slaughter, with assembling the masses of assailants, providing transportation and “tools” for the “work,” arranging for the disposal of the corpses, and directing the division of looted property and confiscated land. They transformed administrative practices, benign in themselves, such as obligatory labor for the common good (umuganda) or the use of security patrols, into mechanisms for executing the genocide.

The political leaders provided the militia for attacks, dispatching them around the country as needed. They prodded reluctant administrators and military officers to greater activity, sometimes using party supporters to harass or threaten those who hesitated to participate. Political leaders also incited Hutu to kill in more direct language than that used by officials who often spoke in ambiguous and allusive terms.

Even as leaders of the genocide were exploiting existing hierarchies, they also created a fourth channel dedicated to implementing the “civilian self-defense” program. The system was formalized only late in May, but such key elements as the recruitment of participants by administrators and the reliance on former soldiers to command them were in use during the massacres of early April. With headquarters in Bagosora’s own office, the “civilian self-defense” hierarchy was staffed largely by retired officers-cum-politicians, much like Bagosora himself.

Through these hierarchies, organizers carried out a killing campaign, a perversion of previous campaigns that called on citizens and officials alike to contribute extra efforts for some public good. The urgency and importance of the objective was deemed to justify departing from usual bureaucratic practice. Zeal for killing took on more significance than formal rank: subordinates could prevail over their superiors, in both civilian and military spheres, if they showed greater commitment to the genocide. This flexibility encouraged ambition and initiative among those willing to trade human lives for personal advantage. Actors could also bypass the usual limits set by law or administrative practice, with politicians or soldiers speaking for government officials, militia approving candidates for administrative position, and medical assistants calling in military strikes.

These practices, which promoted rapid and effective execution of the killing campaign, now complicate the task of assessing responsibility for crimes. All who seek accountability for the genocide must take care to ensure that officials of lesser rank but greater power not escape blame for crimes that are wrongly imputed to their superiors alone.

In the first days of killing in Kigali, assailants sought out and murdered targeted individuals and also went systematically from house to house in certain neighborhoods, killing Tutsi and Hutu opposed to Habyarimana. Administrative officials, like the prefect of the city of Kigali, ordered local people to establish barriers to catch Tutsi trying to flee and to organize search patrols to discover those trying to hide.

By the middle of the first week of the genocide, organizers began implementing a different strategy: driving Tutsi out of their homes to governmentoffices, churches, schools or other public sites, where they would subsequently be massacred in large-scale operations.

Towards the end of April, authorities declared a campaign of “pacification,” which meant not an end to killing, but greater control over killing. Sensitive to criticism from abroad—muted though it was—authorities ended most large-scale massacres. They also sought to rein in assailants who were abusing their license to kill, such as by slaying Hutu with whom they had disputes or who were allowing Tutsi to escape injury in return for money, sexual favors or other considerations. They ordered militia and other citizens to bring suspects to officials for investigation and then murder instead of simply killing them where they found them. Authorities used “pacification” also as a tactic to lure Tutsi out of hiding to be killed.

By mid-May, the authorities ordered the final phase, that of tracking down the last surviving Tutsi. They sought to exterminate both those who had hidden successfully and those who had been spared thus far—like women and children—or protected by their status in the community, like priests and medical workers. As the RPF advanced through the country, assailants also hurried to eliminate any survivors who might be able to testify about the slaughter.

Throughout the genocide, Tutsi women were often raped, tortured and mutilated before they were murdered.

The density of the administrative and political hierarchies, characteristic of Rwanda for many years, gave genocidal leaders rapid and easy access to the population, but did not guarantee mass participation in the slaughter. As authorities played on popular fears and greed, some people picked up their machetes and came readily. Others came more slowly and some refused to come, even at the risk of their lives.

Both on the radio and through public meetings, authorities worked to make the long-decried threat of RPF infiltration concrete and immediate. Throughout the country they disseminated detailed false information, such as reports that Tutsi had hidden firearms in the bushes behind the Kibungo cathedral, or that they had prepared maps showing fields to be taken from Hutu in Butare, or that they had killed local administrative officials in Nyakizu. Authorities counted on such news to convince Hutu that their Tutsi neighbors were dangerous agents of the RPF who had to be eliminated. Community leaders and even clergy assured Hutu that they were justified in attacking Tutsi as a measure of “self-defense.”

Authorities offered tangible incentives to participants. They delivered food, drink, and other intoxicants, parts of military uniforms and small payments in cashto hungry, jobless young men. They encouraged cultivators to pillage farm animals, crops, and such building materials as doors, windows and roofs. Even more important in this land-hungry society, they promised cultivators the fields left vacant by Tutsi victims. To entrepreneurs and members of the local elite, they granted houses, vehicles, control of a small business, or such rare goods as television sets or computers.

Many poor young men responded readily to the promise of rewards. Of the nearly 60 percent of Rwandans under the age of twenty, tens of thousands had little hope of obtaining the land needed to establish their own households or the jobs necessary to provide for a family. Such young men, including many displaced by the war and living in camps near the capital provided many of the early recruits to the Interahamwe, trained in the months before and in the days immediately after the genocide began. Refugees from Burundi, in flight from the Tutsi-dominated army of Burundi, had also received military training in their camps and readily attacked Rwandan Tutsi after April 6.

In some regions, particularly those where Habyarimana’s supporters were strongest, authorities needed to do little more than give the signal for Hutu to begin attacking Tutsi. In other areas, such as central and southern Rwanda, where Tutsi were numerous and well integrated and where Habyarimana’s party had little standing, many Hutu initially refused to attack Tutsi and joined with them in fighting off assailants. Only when military and civilian authorities resorted to public criticism and harassment, fines, destruction of property, injury, and threat of death did these Hutu give up their open opposition to the genocide.

In some places, authorities apparently deliberately drew hesitant Hutu into increasingly more violent behavior, first encouraging them to pillage, then to destroy homes, then to kill the occupants of the homes. Soldiers and police sometimes threatened to punish Hutu who wanted only to pillage and not to harm Tutsi. Authorities first incited attacks on the most obvious targets—men who had acknowledged or could be easily supposed to have ties with the RPF—and only later insisted on the slaughter of women, children, the elderly, and others generally regarded as apolitical.

Just as communities were readier to kill some Tutsi than others, so individual Hutu would agree to attack one person and not another or, in an extension of the same logic, would attack one person and save another. Hutu who protected Tutsi ordinarily helped those to whom they were linked by the ties of family, friendship, or obligation for past assistance, but sometimes they also saved the lives of strangers. Even such persons as Colonel Bagosora and leading figures of the interim government saved the lives of Tutsi close to them, testimony to the extent to which ties between Hutu and Tutsi survived even the most persistent efforts toeradicate them. In some cases, former officials now seek credit for saving the lives of a few favored Tutsi, as if having done so reduced their responsibility for directing or permitting the slaying of so many others.

Many Rwandans say that they killed because authorities told them to kill. Such statements reflect less a national predisposition to obey orders, as is sometimes said, than a recognition that the “moral authority” of the state swayed them to commit crimes that would otherwise have been unthinkable.

Itself the chief actor in a masquerade of legitimacy, the interim government gave its officials and citizens the cover of “legitimate” orders to hide from themselves and others the evil they were doing. Administrators broke the genocide down into a series of discrete tasks which they executed without consideration of the ultimate objective of the work. Cultivators turned out for the long-standing practice of communal labor although they knew that they were to cut down people as well as the brush in which they found them. Priests announced public meetings without consideration of the message to be delivered there. Businessmen contributed money to the “self-defense” fund established by the government as they had contributed to similar collections in the past, even though the money was to buy “refreshments” for the militia and fuel to transport them to their places of “work.”

As part of the”pacification” effort in late April, authorities ordered churches, schools, hospitals, and shops to resume their functions, ignoring the absence of Tutsi who used to participate in these various activities. They presumed to create a veneer of “normalcy” in a world where untold numbers of people were violating the laws, religious teachings, and cultural norms that they had always lived by.

Many Tutsi and those Hutu associated with them fought to save their lives. We know of their heroic resistance, usually armed only with sticks and stones, at such places as the hills of Bisesero, the swamps of Bugesera, and the church at Cyahinda, but we have no way of knowing about the countless small encounters where targeted people struggled to defend themselves and their families in their homes, on dusty paths, and in the fields of sorghum.

Some tens of thousands fled to neighboring countries and others hid within Rwanda, in the ceilings of houses, in holes in the ground, in the forest, in the swamps. Some bought their lives once, others paid repeatedly for their safety over a period of weeks, either with money or with sexual services.

Many Tutsi who are alive survived because of the action of Hutu, whether a single act of courage from a stranger or the delivery of food and protection over many weeks by friends or family members.

When organizers of the genocide gained control of the state, they suppressed dissent but did not extinguish it. In May and June, when the interim government was weakened by military losses and by the first signs of international disapproval, Hutu in one community after another began refusing to undertake further searches or to participate in guarding barriers. As the majority of participants withdrew, they left execution of the genocide in the hands of smaller, more zealous groups of assailants, who continued to hunt and kill in hopes of profit or because they were committed to exterminating the last Tutsi.

With the campaign against Tutsi no longer a strong bond, Hutu of different areas and parties once more began to fight against each other. Some revived old battles. Others competed in new rivalries over power or over goods and property taken from Tutsi. Interahamwe and other young men who had been authorized to terrorize Tutsi began robbing, raping, and killing Hutu as the number of Tutsi declined.

Hutu used the discourse of the genocide in conflicts with other Hutu: they accused each other of being Tutsi, of having hidden Tutsi, or of supporting the RPF. Just as some charged enemies with too great lenience towards Tutsi at this time, so others would charge their opponents with violence against Tutsi once the genocide was ended.