Published in Global Change, Peace & Security, Vol. 21, No. 3, October 2009, pp. 275-290
pdf of published article
Brian Martin's publications on peace
Brian Martin's publications
Brian Martin's website
Keywords: genocide; Rwanda; tactics; outrage
According to some estimates, during the twentieth century more people died in genocides than in wars, yet genocide receives relatively little attention compared to war. Widespread awareness of the problem of genocide only occurred after the Holocaust and the UN genocide convention. Yet despite expressions of formal concern, government responses to genocide have been grudging and often ineffectual.
The UN convention defines genocide in a special way whereas many popular and scholarly treatments adopt somewhat different conceptualisations. The UN definition includes attempts to exterminate ethnic groups using means such as preventing births and transferring children, whereas most attention subsequently has been on mass killing. The UN definition excludes mass killing for political reasons whereas many scholars count this as genocide, an example being Cambodia 1975-1979. Indeed, the definition of genocide continues to be contested. I adopt here a definition by genocide scholar Helen Fein: 'Genocide is sustained purposeful action by a perpetrator to physically destroy a collectivity directly or indirectly, through interdiction of the biological and social reproduction of group members, sustained regardless of the surrender or lack of threat offered by the victim'.
Few genocides have received significant popular attention compared to wars with equivalent death tolls, judging for example by the volume of books, films or media commentary on each. The Holocaust is widely known - though there is far more material about war-fighting in World War II - but many other genocides have hardly registered on popular or scholarly consciousness, for example the killing of half a million Indonesians in 1965-1966 or one to three million people in what was East Pakistan in 1971. On the other hand, the term genocide is sometimes applied in conflicts as a way to condemn actions that are far from any common definitions of genocide.
There is a growing body of research in genocide studies, with the topic approached in many different ways, including history, comparative social structures, psychology, classes and state structures, responsibility and prevention. Genocide can be analysed by using frameworks from other fields, as with class analysis. Martin Shaw has brought together the study of genocide and war by conceptualising genocide as a war against civilians. Given the importance of the topic, it seems worthwhile to consider other ways to approach it.
My proposition is that a crucial part of the dynamics of genocide is the management of outrage. Genocide commonly arouses feelings of abhorrence or disgust among many observers. In the face of such reactions, how can perpetrators proceed? The answer is that they need to adopt tactics that reduce the likelihood of outrage.
In many studies of genocide, the assumption seems to be that because genocide is so horrible, what needs to be explained is the lack of response by third parties. In studying the management of outrage, I look as well at active measures, by perpetrators, to inhibit outrage.
In the next section, I introduce a model of outrage management, using the example of the 1991 Dili massacre in East Timor and introducing the relevance of the model to genocide. Then, to show in detail how this model can be applied to genocide, I use the 1994 Rwandan genocide as a case study, examining each of five methods of inhibiting outrage, followed by a discussion of methods for amplifying outrage. In the conclusion I discuss implications for opposing genocide.
Studies of a wide range of injustices - censorship, sexual harassment, unfair dismissal, police beatings, massacres, torture and aggressive war, among others - show that perpetrators use fairly predictable sorts of tactics to reduce outrage, which can be classified into five categories.
Consider, for example, the Indonesian military occupation of East Timor from 1975. Initially, the East Timor independence movement Fretilin waged an armed struggle in resistance. The war and subsequent starvation led to the death of, by some estimates, a third of the East Timorese population, considered by some to constitute genocide.
Fretilin was heavily outnumbered and outgunned. Armed resistance had a further disadvantage: it framed the struggle as a war, which meant the injustice involved in the Indonesian invasion and occupation was less obvious than if the military had attacked an unarmed population. Indonesian forces carried out many massacres, but these were little known to the outside world due to censorship.
In the late 1980s, the East Timorese resistance adopted a new strategy, emphasising peaceful protest in the cities, especially during visits by foreign dignitaries, and using force only in defence. On 12 November 1991, thousands of East Timorese joined a funeral procession in the capital Dili, using it as an occasion to protest against Indonesian rule. As the procession entered Santa Cruz cemetery, Indonesian troops opened fire on the crowd without warning. This assault on peaceful protesters was an obvious injustice. Unlike previous massacres, on this occasion Western journalists were present, witnessing the killings and capturing the events in photos and video footage.
The Indonesian military and government used all the tactics outlined above to reduce outrage.
Although the Indonesian government used all five methods for minimising outrage, on this occasion it was largely unsuccessful. The eyewitness accounts, photos and videotape (smuggled out of the country) cut through cover-up and reinterpretation. Indonesian attempts at devaluation, official channels and intimidation had little salience internationally. The result was a huge expansion of international support for East Timor's independence struggle. The massacre of protesters at Santa Cruz cemetery, intended to subdue East Timorese resistance, instead backfired against the Indonesian occupiers, creating far more support for the resistance.
Quite a number of factors can be used to explain the response to the Dili massacre, including the role of the US, Australian and other governments supporting the Indonesian occupation, the existence of an international East Timor support network and the role of the international media particularly when journalists were attacked. However, my focus here is not on all the factors that explain responses to events but on the tactics of outrage management used by perpetrators and their opponents.
In this example, as in many others, there are two necessary conditions for an action to backfire on perpetrators. First, the action has to be perceived as unfair, excessive or disproportionate, or more generally as a violation of a social norm. Second, information about the action has to be communicated to receptive audiences.
This model of tactics appears to apply quite readily to genocide. Consider some examples from well-known features of the Holocaust.
In practice, there are many complications in the use of these tactics. For example, different levels and types of cover-up might be used in relation to victims, bystanders, killers and planners. Cover-up itself can be covered up: for example, documents about censorship systems may be destroyed. The five-method classification offers a framework on which to build a more detailed analysis.
In this paper, I analyse the 1994 Rwandan genocide as a means of illustrating in detail how this framework can be applied. But first it is worth commenting further on the methods and how this model of tactics relates to other approaches to genocide.
Cover-up can be accomplished by various methods, including secrecy, censorship, hiding correct information in a barrage of misinformation, and operating below the media's threshold of newsworthiness. Different forms and levels of cover-up may apply to different audiences.
Devaluation can occur on the basis of pre-existing prejudice and stereotypes and through active measures such as derogatory labelling, unfavourable media portrayals, and finding or manufacturing damaging information.
Reinterpretation includes any method that justifies or explains an action or event in a way that minimises its seriousness or transfers responsibility for it. Reinterpretation often results from genuine belief, typically a self-serving perspective: perpetrators may not appreciate the consequences of their actions and may see themselves as justified or as victims. It includes claims that consequences were not great, proper procedures were followed and that someone else was to blame. When lying is involved, reinterpretation can shade into cover-up. The distinction, in relation to a specific audience, is that with cover-up there is no awareness that anything happened whereas with reinterpretation audiences know something happened but are given a benign explanation for it.
Official channels, such as grievance procedures, ombudsmen, expert panels, courts and parliaments, are supposed to provide justice. But when the perpetrator is far more powerful than the target, official channels may provide only an appearance of justice, an image without substance. Official channels are typically slow, involve adherence to procedures and rely on experts such as lawyers, and can take up a lot of effort and money. The consequence of lengthy, procedural, complex processes is often that public outrage dies down.
Intimidation and bribery can be used against many targets, most obviously the victims of injustice but also wavering perpetrators and third parties, including journalists, editors, governments and non-government organisations. Intimidation is often used to maintain cover-up: anyone who might reveal what is happening can be threatened with reprisals. Bribery includes incentives for keeping quiet or not intervening, for example opportunities to loot.
These tactics may or may not be used before an injustice. In the beating of Rodney King by Los Angeles police in 1991, there was little done by the police to minimise outrage before or during the beating. On the other hand, in the lead-up to the 2003 invasion of Iraq - perceived by many as an injustice - the US government used all five methods.
There is no time limit to the struggle: tactics of outrage management may be used long after an injustice. Decades after the Holocaust, revisionists continue to attempt to reinterpret the genocide as non-existent or as less serious than it actually was. Nearly a century after the genocide of the Armenians, the Turkish government continues to deny it, using reinterpretation and intimidation.
Tactics can change over time. For example, in the case of massacres such as at Dili or Sharpeville, South Africa in 1960, there were initial efforts to hide what happened; when these efforts failed, other tactics were brought into play, such as official inquiries.
The sorts of tactics used vary from case to case and issue to issue. For example, devaluation is a crucially important technique used to limit outrage over treatment of refugees whereas it is less common in some environmental disasters such as the Exxon Valdez oil spill.
Different audiences may be and often are managed quite differently. Perpetrators may be quite open about their actions with each other but hide them from wider scrutiny. In the case of genocide there are several audiences of significance: perpetrators, potential targets, non-participants within the country, foreign governments and foreign populations, among others. Sometimes an audience may have important subsets: for example, intelligence services may have detailed information about atrocities while government leaders know nothing or have only a superficial understanding of events.
This framework for classifying outrage management tactics is a complement to other approaches to genocide, including history, international relations and other crucially important factors. These factors provide the context in which tactics are used and the resources for using them. For example, state structure, communications infrastructure and norms for public discourse each affect the capacity for cover-up. Policies and practices involving ethnic groups, including beliefs, customs and patterns of employment, housing and intermarriage, affect the possibilities for devaluation. Myths, metaphors and levels of education, as well as the dynamics of the public sphere, affect capacities for reinterpretation. The credibility of the government, police, the military, courts and experts affect the role of official channels. Resources available to perpetrators, including government coordination, police loyalty, funds and technology, affect the capacity for intimidation and bribery. Actual and potential tactics operate within society as it exists, so understanding the society, in full social and historical context, is vital for understanding tactics.
Do perpetrators always use methods to inhibit outrage? Do they always need to? To answer these questions requires an analysis of actual cases. Speaking generally, if no one thinks genocidal actions are disturbing or wrong, then there is no need to manage potential outrage. In practice, individuals and social movements attempt to alter people's perceptions of acceptable behaviour, changing an unnoticed or tacitly accepted practice into something to be condemned. When this occurs, those seen as perpetrators may feel the need to take steps to minimise antagonistic reactions to their actions.
Gregory Stanton has analysed genocide as the end point of a set of stages - not necessarily sequential, and often overlapping - with the focus on precursors to genocide. My approach here, studying tactics to manage outrage over genocide, is complementary to Stanton's study of stages, though there is considerable overlap. Tactics can be used at any stage; indeed, some of the stages, such as dehumanisation, are primarily composed of specific tactics.
From April to June 1994, half a million to a million Tutsi and moderate Hutu were killed in a coordinated extermination operation. My aim here is to outline the tactics used by the perpetrators to minimise outrage from this gross violation of human rights. The Rwandan genocide might seem to be a difficult case, because so much of the killing was carried out in the open, apparently without concern about hiding the atrocities. Yet on closer inspection it is apparent that many types of cover-up were used, as well as the other techniques of inhibiting outrage.
A wider understanding of the Rwandan genocide would include an historical and social analysis of Rwandan society, an examination of the role of international actors, for example in supplying arms and portraying the Rwandan government as benign, and the subsequent geopolitics of central Africa. The 1994 genocide cannot be fully understood in isolation. Nevertheless, for my purposes here, namely illustrating the use of outrage-inhibition techniques, it is sufficient to focus on the April-June 1994 events and immediately related events beforehand and afterwards. This focus is not a dismissal of other atrocities by other groups or at other times; these can be analysed using the same framework.
No new documentary material is introduced here. Rather, material from the rich studies of the genocide is used to show the value of the outrage-inhibition framework. There is a surfeit of examples, due to the many actors involved: killers on the ground; the interim Rwandan government; other governments, notably those of Belgium, France and the United States; the United Nations; non-government organisations; and the mass media. Techniques for inhibiting outrage were used by or in relation to each of these actors. For each technique, I attempt to highlight the principal actors, giving a few illustrations concerning others.
On 6 April 1994, Rwandan president Juvénal Habyarimana was killed when his plane was struck by a missile near Kigali airport and crashed. Massacres of government opponents and of Tutsi began shortly after, coordinated by figures within what became the interim Rwandan government. The leaders of the genocide operation disguised their own role, having set up militias that were trained far from Kigali. Many of the killings, held in remote areas, were denied, telephone wires were cut during massacres, and survivors of massacres were not allowed to leave the area. These were among the methods used to hide information about the killings from outside audiences. As information nonetheless leaked out and international scrutiny increased, the initial large-scale killings were superseded by smaller-scale, less public killings.
Several western governments - especially the Belgian, French and US governments - aided in the cover-up. In the immediate aftermath of the Habyarimana's death and the surge of killing, including 10 Belgians in the UN peacekeeping force, western governments rapidly closed their embassies and withdrew their nationals - thousands of diplomats, aid workers and their families - who otherwise would have been influential witnesses. These governments knew immediately about the massive killing, but did nothing; they had satellite pictures, but refused to supply them to the UN peacekeeping force.
The French government played a particularly important covert role, prior to the genocide having trained and financed the Rwandan Presidential Guard and trained and equipped a massive expansion of the Rwandan army. After the genocide started, 'the French embassy in Kigali was abandoned. Left behind was a huge pile of shredded documents, almost filling a room'. US officials did not seek information, preferring not to know what was happening because this allowed 'plausible denial'.
At the UN Security Council, deliberations were in secret, preventing an examination of government actions. Roméo Dallaire, head of the UN peacekeepers in Rwanda, wrote a report to the UN Department of Peacekeeping Operations, but it was not passed to the Security Council. Meanwhile, Boutros Boutros-Ghali, UN Secretary-General, was detached, perhaps purposely not informed.
The media often play a key role in disseminating information. When the media do not cover an issue - perhaps because it fails to satisfy the usual criteria for newsworthiness - this operates as a form of de facto cover-up. Within Rwanda, media coverage was both limited and biased, allowing much of the killing to continue without being reported. The radio station RTLM - Radio Télévision Libre des Milles Collines - had been set up as private radio, so the government could spread hate without accountability. In the initial weeks of the genocide, RTLM did not reveal atrocities that were occurring. Rural communication and access were limited, so most of the international attention was on Kigali, giving a restricted perspective on the events. Very few foreign journalists remained in Rwanda due to lack of security. However, foreign news media were slow to take interest. For example, Roger Winter, director of the United States Committee for Refugees, wrote an article about what was happening but most US newspapers rejected it.
Although the killings in Rwanda were in public view more than in most genocides, a combination of processes operated to limit awareness in the audiences most likely to take action. Within Rwanda, where there was strong pressure at the provincial and village levels to join in the killings, the hate-promoting radio station RTLM was the dominant media influence. Western governments pulled out their citizens, most journalists left, and western intelligence services did not reveal what they knew about events. As NGOs obtained information about killings and tried to raise awareness about them, the interim Rwanda government moved from open large-scale massacres to more hidden smaller-scale killings. The leaders of the genocidal operation thus seemed to modulate their actions according to the likelihood of exposure. When western governments pulled out their citizens, these leaders judged the risk of intervention minimal and were more brazen; when international scrutiny increased, they encouraged greater discretion in killing. Cover-up thus was not a simple matter of hiding information, but rather a complex process of attempting to keep below the radars of relevant audiences.
Finally, it is worth mentioning cover-up at the individual killer level. Jean Hatzfeld interviewed many genocide survivors and later many killers. After the genocide, the killer's first choice was always silence. It required extraordinary efforts by Hatzfeld to encourage killers to speak freely; this only occurred in prisons, after sentences had been determined, and with a supportive group of other killers. Cover-up can occur at perceptual and psychological levels as well as through organisational processes.
If targets are believed to be dangerous, subversive, worthless or subhuman, then what is done to them does not seem so bad. In Rwanda, there was a long history of derogatory stereotypes and repeated instances of prejudicial policy and killings. A particular Hutu group ran the Rwandan government; some government figures encouraged the expression of anti-Tutsi propaganda to serve their own political ends.
The 'Hutu ten commandments' were a prominent expression of prejudice. Published in the journal Kangura, they included items such as 'we shall consider a traitor any Muhutu who: marries a Tutsi woman; befriends a Tutsi woman; employs a Tutsi woman as a secretary or a concubine' and 'The Bahutu should stop having mercy on the Batutsi'. However, it may be that relatively few rural Hutu ever heard of these 'commandments'.
The Rwandan government sponsored the hate media - magazines and most notably the radio station RTLM - that demonised Tutsi. The radio, for example, warned of the Rwandan Patriotic Front (RPF) - an armed opposition movement mainly composed of Rwandan Tutsi exiles - as devils that did horrible things to their Hutu victims. RTLM identified all Tutsi as the RPF, who were the enemy.
The killers often humiliated and tortured their victims, for example by stripping them naked, raping them, cutting off their body parts, forcing them to kill their own family members, and preventing bodies from being buried. Killing and devaluation reinforced each other.
Hatzfeld, in her interviews, found that killers had different levels of prejudice, but all of them found it hard to speak about. These Hutu killers called Tutsi snakes, zeroes and dogs. As the killers got used to killing, their victims were dehumanised.
Scott Straus, on the basis of his extensive interviews with killers, found that few of them seemed to be motivated by ethnic hatred. A far more significant factor was the identification of Tutsi as the enemy in the context of war.
In western countries, neither prejudice against the Tutsi nor their identification as the enemy had much influence. The role of devaluation, in this case, was largely within Rwanda itself.
Numerous techniques were used that portrayed what was happening in a deceptive way, including self-serving frames, lies, and assignment of responsibility. These techniques were deployed at different times for different audiences.
One of the most pervasive reinterpretations was to conflate the genocide with the war against the RPF. The Rwandan government said massacres were battles, described genocide as interethnic fighting, said Tutsi were accomplices or infiltrators and called Hutu the great mass and the innocent victims of Tutsi aggression. The Prime Minister of the interim government claimed that killings were due to grief over President Habyarimana's death and used the expression 'final war' as a code for genocide. Killing by Hutu was blamed on the RPF or explained as due to spontaneous anger or self-defence. The government made claims about Tutsi infiltration and of Tutsi genocide of the Hutu, in essence attributing their own actions to the Tutsi. Some explanations were bizarre. For example, Tutsi who had been killed were said to have committed suicide and to have caused their own misfortune.
RTLM said the Virgin Mary had appeared to benefit the Hutu and that UN forces had helped the RPF shoot down Habyarimana's plane. RTLM talked about named individuals who had just been killed, claiming no knowledge of where they had gone. This was an aspect of RTLM codes in which, for example, 'war' and 'battle' meant killing and a person 'having ammunition' meant they were targeted for being killed.
The misrepresentations by the Rwandan government and its supporters or agents such as RTLM were aimed at multiple audiences. One audience was those organising the killings. Another was the Hutu population being mobilised to participate in the genocide. Occasionally, lies were aimed at Tutsi, such as that the killing had finished and that Tutsi were to be transported home, when actually they were transported to militias for killing.
Two of the most important audiences were western governments and western media. Rwandan diplomats described what was happening as 'interethnic fighting' or 'tribal violence', gave low figures for the number killed, and falsely claimed the killing was over.
Most other governments were complicit in these interpretations, finding them useful in dampening concern and reducing pressure to intervene. The US and other governments resisted the language of genocide, because it would have suggested greater obligation. The US government used several rationalisations to discourage concern: the events were framed as war, not genocide; the only options presented were either doing nothing or full-scale intervention; and not intervening was said to protect the UN humanitarian mission.
At the UN Security Council, the word 'genocide' was avoided. In the Secretary-General's report of 20 April, the focus was on civil war, with no mention of massacres, reflecting the view of the Rwandan government. The Security Council used a framing like other cases of war-related violence against civilians.
Killers also used techniques of lying, reframing and blaming, reinterpreting their actions, the context and responsibility. Hatzfeld said the killer's second choice, after silence, was lying. Individual killers, in describing their actions, used 'we' rather than 'I'. They avoided the words 'genocide' and 'survivor', preferring instead military language such as 'war with machetes'. The killers tried to shift the blame away from themselves; they blamed God for turning away.
Individual killers, the Rwandan government, and RTLM and other hate media reinterpreted the events using lies, misleading frames and allocation of blame. Western governments and the UN mainly used frames that absolved them of responsibility.
The most important official channels for promoting justice in Rwanda were governments and the United Nations. The interim Rwanda government used its authority to promote the genocide; meanwhile, foreign governments and the UN did little or nothing except for symbolic gestures. Many people look to governments and international bodies to address social problems, but in the Rwandan case such expectations only served to allow the killing to continue.
The interim Rwandan government sought legitimacy itself and, to the extent that it succeeded, gave legitimacy to Hutu power and to the killings. Over a period of several weeks after 6 April, the government gradually imposed its agenda over internal opponents. Given that this government was the key agency behind the genocide, its acquisition of control over formal means for redress over human rights abuses meant that the genocide was legitimated within Rwanda.
Other governments could have taken action, but few did anything. In the days after 6 April, western governments evacuated their nationals, sending a strong message about their priorities to the new Rwandan leaders. The Belgian, French and US governments knew immediately about the killings, but did nothing. Even after NGOs reported on abuses, leading to international awareness, the governments did not act. Members of the US Congress urged their government to act, but to no avail. Instead, US officials 'took solace in the normal operations of the foreign policy bureaucracy, which permitted an illusion of continual deliberation, complex activity, and intense concern'.
It is revealing to take note of actions that might have been taken, but were not, such as expelling Rwandan ambassadors or threatening indictments for genocide. The US and French governments each refused to jam RTLM, offering a range of excuses. When the governments did take action, it was to protect their own interests. After the genocide was nearly over, as a result of the RPF's military victory, the French government launched Operation Turquoise, which in practice protected some Tutsi from death, protected many genocide leaders from capture by the RPF, and served as a public relations exercise within France: as an apparently humanitarian operation, it helped divert attention from the French government's support for the Hutu leadership prior to the genocide and, more covertly, during part of it.
The UN played a key role in Rwanda. It helped promote the 1993 accord between the Rwandan government and the RPF, signed in Arusha, Tanzania. This included a peacekeeping operation, UNAMIR, the UN Assistance Mission for Rwanda, led by Canadian general Roméo Dallaire. But the troops, from Belgium, Ghana and Bangladesh, were inadequate in number and lacked training and equipment. As signs of impending violence increased, Dallaire pleaded for greater support and for authorisation to act more decisively against weapons stockpiles and human rights abuses, but UN officials refused. The presence of UNAMIR raised expectations about what the international community would do, but the UN did not meet these expectations.
Instead, the UN Security Council held meeting after meeting, giving the appearance of concern without much substance. The Rwandan government happened to have a seat on the Security Council, and the Rwandan representative was allowed to continue sitting throughout the genocide. About two weeks into the genocide, the Security Council decided to withdraw most of the UNAMIR troops; this provided a signal for the Rwandan government to press ahead with the genocide. Months later, after publicity made the genocide too difficult to ignore, the Security Council authorised UNAMIR II, but it took so long to be organised - in part due to stalling by the US government - that it played no role in ending the killing.
In Rwanda, the formal apparatus of the international law and expectations for foreign governments and the UN to act against genocide may have made things worse than if they had not existed: they gave the appearance of providing a solution to the killing but not any concrete action. By continually deliberating and stalling, they stymied the translation of public outrage into action against genocide.
Threatening to kill someone is a potent form of intimidation, as is killing neighbours and friends. Intimidation also serves to deter the expression of outrage over killing. It was a key technique used by promoters of genocide.
In the aftermath of 6 April, troops and militias throughout Rwanda began killing opponents of the regime, both Tutsi and Hutu, as well as Tutsi more generally. Some joined in immediately, but others were reluctant. In some parts of the country, officials resisted encouragement to unleash restraints against killing, even going so far as to arrest killers and looters. For the aims of the genocide leaders to be realised, this sort of opposition had to be overcome. Intimidation was a key tool.
Soldiers and police made threats against opponents of the killing, including administrators, and sometimes forced participation in killing. Strong peer pressure - 'intra-Hutu coercion' - operated within groups of Hutu men, with group leaders applying pressure on others. For those who were reluctant to kill, there was the risk of being accused of complicity. Some Hutu women, in the face of attacks on Tutsi, didn't show their compassion for the victims for fear of reprisals.
Administrators also acted against those who refused to kill, for example using police, militia or civilians to burn their houses or physically threaten them. When military officers or local administrators opposed killing, they were bypassed by removing their resources, removing them from their positions and replacing them by compliant officials, or killing them. Many of those likely to oppose the genocide - politicians, lawyers, journalists and human rights activists - were pre-emptively killed.
In parallel with intimidation of resisters was bribery to encourage cooperation. Incentives included food, drink, marijuana, money and opportunities for looting.
Intimidation is central to genocide, but can perform multiple functions in relation to the creation and expression of outrage. Threats, attacks and killing can scare and silence opponents, but on the other hand if these actions are witnessed more widely, they can generate concern and stimulate opposition. In the Rwandan genocide, nearly all the intimidation operated within Rwanda itself, where the génocidaires controlled the government and could operate with virtual impunity. Many of their actions were in public, easily witnessed by observers, serving to terrify potential victims and scare others into joining the attacks. But the killers were careful not to expand their attacks too widely. They killed few western civilians - though many were threatened - and only a few UNAMIR troops, though enough to cause western governments to withdraw their nationals and thus help limit outside awareness of the killings. Too much intimidation of westerners would have risked generating greater outrage and increasing the likelihood of intervention.
The examples presented here are a selection from the extensive evidence that the perpetrators of the Rwandan genocide used all five methods of inhibiting outrage: cover-up, devaluation, reinterpretation, official channels and intimidation. This classification of outrage-inhibition techniques suggests a corresponding set of counter-tactics aimed at amplifying outrage:
Each of these categories of counter-tactics encompasses a wide range of specific actions. The literature on Rwanda provides many examples, though not nearly so many as of perpetrator tactics.
Exposure of the killings to relevant audiences was critical to stopping them. Within Rwanda, witnesses reported on atrocities, and a few were able to send messages out of the country. International human rights NGOs wrote reports and tried to interest the world's media. International public opinion was a powerful force for challenging government indifference and complicity in the genocide, though governments were still able to stall through symbolic gestures.
Within Rwanda, in many cases massacres ended after reporting. Exposure of the massive open killing in early weeks led to international pressure, causing the government to move to more hidden, smaller-scale killing.
Hundreds of people, mostly Tutsi, fled to the Hôtel des Milles Collines, made famous through the film Hotel Rwanda. The hotel twice came under threat of attack, but in both cases the attackers withdrew after international pressure was applied. The regime had failed to cut off an international phone line from the hotel; the hotel's manager used the phone to appeal for assistance.
The presence of witnesses deterred violence. Although most foreign citizens had been evacuated, some remained. Even a few UN soldiers were often enough to deter killing.
Exposure could also undermine the technique of reinterpretation. After the New York Times revealed a US government directive to not use the word genocide, this led to ridicule.
French soldiers in Operation Turquoise had been misled about its purpose. When they witnessed the killing of Tutsi, they realised they had been deceived.
Validation is the counter-tactic to devaluation. People can be validated by asserting their essential humanity and their human rights and by documenting and highlighting their social value, their good works and their social roles, among other methods. This validation of fellow humans occurs through most societies most of the time: systematic devaluation as a prelude to killing is the exception.
Although there is little evidence of validation as an overt technique for amplifying outrage, nonetheless latent and pre-existing validation can be assumed to have played an important role. Outside Rwanda, the ethnic divisions of Hutu and Tutsi had little salience and provided no basis for devaluation: most foreigners assumed that all Rwandans deserved human rights. Even within Rwanda, only a minority of the population was involved in killing, and many individuals were actively opposed to it, as described below. Friendship or family connections motivated some who intervened against killing; general humane instincts motivated others.
To challenge the killers' interpretation of events as spontaneous rage, sporadic violence and limited killing that was now under control, the contrary interpretation of the killings as systematic, extensive and promoted by the Rwandan government was required. Applying the label genocide highlighted the seriousness of the killings. All those who tried to expose and mobilise action against the genocide interpreted the events in these terms, implicitly or explicitly opposing the killers' interpretations.
Official channels - governments and the UN - were for the most part unwilling to act against the genocide. The alternative to appealing directly to governments was mobilising public opinion. This was the approach of those NGOs that promoted their reports to the media.
There are many stories of courageous resistance to intimidation. Some Rwandan military officers and local administrators opposed the killing, even arresting attackers. Many individual Rwandans hid friends, neighbours or total strangers in their homes, or otherwise stood up to the killers. Although participation in the killing was widespread, nevertheless there are sufficient stories of resistance to give hope.
Accounts of the Rwandan genocide offer numerous examples of tactics used by perpetrators to minimise outrage and relatively few examples of tactics used to amplify it. Even so, there is sufficient evidence to suggest that it is fruitful to examine the genocide in terms of a struggle over reactions involving tactics and counter-tactics. Counter-tactics can usefully be classified as exposure, validation, interpretation, mobilisation and resistance.
It may seem natural to assume that genocide, or indeed any killing of civilians, is likely to generate outrage. For genocide to occur or for perpetrators to escape condemnation, this outrage needs to be minimised before, during or after the actions taken. This observation leads to a focus on tactics, including tactics to reduce outrage and tactics to amplify it. It is useful to divide perpetrator tactics into five categories: cover up the action; devalue the target; reinterpret the events or responsibility; use official channels to give an appearance of legitimacy; and intimidate or bribe participants and observers.
The literature on the 1994 Rwandan genocide gives ample evidence of these five methods, including the ways they relate to each other. Although much of the killing was done openly, the Rwandan government remained sensitive to international opinion, distancing itself from the killing through militias and from incitement through nominally independent radio station RTLM, and moving to a policy of 'pacification' - smaller-scale killings out of sight - after large-scale massacres received international attention. Prior to and during the killings, efforts were made, not least through RTLM, to demonise Tutsi, identifying them with the formal enemy, namely the RPF. The Rwandan government lied about the extent of killing and the motivations and responsibility for it; the framing of the killing as due to tribal violence was taken up by much of the world's media. Many people looked to governments and the UN to intervene, but this was misguided: western governments withdrew their citizens, avoided the label genocide, and restrained the UN from any intervention. Finally, the génocidaires systematically threatened opponents, killing many of them, while offering rewards to those who joined the killing and looting.
The Rwandan genocide occurred during a war. The RPF had invaded Rwanda previously and launched another invasion immediately after the death of President Habyarimana and the initiation of killings. This eventually culminated in an RPF victory. Although the RPF ended the genocide through military force, its military threat and invasion were used by the regime to minimise outrage from the genocide.
Outrage from human rights violations is most likely when the injustice is clear-cut: a massacre of peaceful protesters is an obvious injustice, but if even a few protesters use violence, then the engagement is easily framed as violence versus counter-violence. This is why, in nonviolent actions, maintaining commitment to nonviolence is so important: when nonresisting protesters are attacked, the perceived injustice is much greater.
The RPF's invasion was used to help frame the Rwandan situation as one of civil war. Genocidal killing could be reinterpreted as part of the war, and cover-up and devaluation were facilitated by wartime conditions. Rwanda 1994 is just one of many genocides that have taken place during wartime. War facilitates all the methods of inhibiting outrage.
A focus on tactics offers an approach to genocide complementary to the study of history, social structure, political dynamics and belief systems. It throws a critical light on official channels, including laws and governments.
The tactics of amplifying outrage can be used as a checklist for both opposing and preventing genocide. Exposure is central: without documentation and communication to receptive audiences, few people will be concerned. It has been argued that mass starvation is impossible when there is a vibrant public sphere; the same may apply to genocide. Validation of potential targets is vital; advocates of human rights constantly oppose attempts to stigmatise and dehumanise groups of people. Interpretation of actions as human rights abuses is essential: it requires challenging lies and confronting misleading framing of events.
To point to tactics against genocide is of course not to guarantee that these tactics will be successful, any more than understanding military tactics guarantees military victory. Furthermore, it is possible for people to be outraged but for that outrage not to be converted into effective action. The success of anti-genocide efforts depends in part on tactics used and how well they are used but also on resources available to génocidaires, political circumstances, strategic alliances and other factors. Underlying these dynamic factors are structural features such as state sovereignty.
Governments and international organisations are often seen as the solution to the problem of genocide, yet this hope has been dashed for decades. Laws and policies can be helpful if they reflect a mobilised public, but can be deceptive if they are a substitute for action. Finally, resistance is vital. Resistance by those whose lives are at risk is incredibly courageous. For many others, at a distance, resistance is more a matter of will.
I have presented a framework for understanding tactics for minimising or amplifying outrage over genocide and showed how it can be applied to the 1994 Rwandan genocide. Much more can be done to examine outrage-related tactics in genocide, including tactics in other genocides, tactics in periods before and after genocide, and the relation of tactics to key variables such as history, psychology, social and political structure, technology and belief systems.
I thank Ben Kiernan for helpful advice and Michael Barnett, Stephen James, George Myconos, Gregory Stanton and several anonymous referees for useful comments on earlier versions of this paper.
1. Helen Fein, 'Accounting for Genocide after 1945: Theories and Some Findings', International Journal on Group Rights 1, no. 2 (1993): 79-106, at 81.
2. Many authors address the history and politics of the UN definition, for example Adam Jones, Genocide: A Comprehensive Introduction (London: Routledge, 2006), 8-28, who gives a convenient list of definitions by key authors.
3. Helen Fein, Genocide: A Sociological Perspective (London: Sage, 1993), 24. Note that my analysis is not sensitive to the precise definition of genocide so long as people can be outraged by the actions involved.
4. For example, Frank Chalk and Kurt Jonassohn, The History and Sociology of Genocide: Analyses and Case Studies (New Haven, CT: Yale University Press, 1990); Jones, Genocide; Albert J. Jongman, ed., Contemporary Genocides: Causes, Cases, Consequences (Leiden: Projecten Interdisciplinair Onderzoek naar de Oorzaken van Mensenrechtenschendingen, 1996); the journals Journal of Genocide Research, Genocide Studies and Prevention, Holocaust and Genocide Studies and War Crimes, Genocide & Crimes against Humanity.
5. For example, Helen Fein, Accounting for Genocide: National Responses and Jewish Victimization during the Holocaust (New York: Free Press, 1979).
6. For example, Ervin Staub, The Roots of Evil: The Origins of Genocide and other Group Violence (Cambridge: Cambridge University Press, 1989).
7. For example, John G. Heidenrich, How to Prevent Genocide: A Guide for Policymakers, Scholars, and the Concerned Citizen (Westport, CT: Praeger, 2001); Leo Kuper, The Prevention of Genocide (New Haven, CT: Yale University Press, 1985).
8. Martin Shaw, War and Genocide: Organized Killing in Modern Society (Cambridge: Polity, 2003); Martin Shaw, What is Genocide? (Cambridge: Polity, 2007).
9. Samantha Power, 'A Problem from Hell': America and the Age of Genocide (New York: Basic Books, 2002).
10. Brian Martin, Justice Ignited: The Dynamics of Backfire (Lanham, MD: Rowman & Littlefield, 2007). See also Susan Engel and Brian Martin, 'Union Carbide and James Hardie: Lessons in Politics and Power', Global Society: Journal of Interdisciplinary International Relations 20, no. 4 (2006): 475-90; Truda Gray and Brian Martin, 'My Lai: The Struggle over Outrage', Peace & Change 33, no. 1 (January 2008): 90-113; David Hess and Brian Martin, 'Repression, Backfire, and the Theory of Transformative Events', Mobilization 11, no. 1 (2006): 249-67; Sue Curry Jansen and Brian Martin, 'Making Censorship Backfire', Counterpoise 7, no. 3 (July 2003): 5-15.
11. Fein, 'Accounting for Genocide after 1945', 79-106.
12. James Dunn, East Timor: A Rough Passage to Independence, 3d ed. (Sydney: Longueville Books, 2003), 292-3.
13. Chisako M. Fukuda, 'Peace through Nonviolent Action: The East Timorese Resistance Movement's Strategy for Engagement', Pacifica Review 12, no. 1 (February 2000), 17-31.
14. Arnold S. Kohen, From the Place of the Dead: The Epic Struggles of Bishop Belo of East Timor (New York: St. Martin's Press, 1999).
15. Martin, Justice Ignited, 23-33. See also Brian Martin, Wendy Varney and Adrian Vickers, 'Political Jiu-jitsu against Indonesian Repression: Studying Lower-profile Nonviolent Resistance', Pacifica Review 13, No. 2 (June 2001), 143-56.
16. Raul Hilberg, The Destruction of the European Jews (Chicago: Quadrangle, 1961) is a classic treatment.
17. Roy F. Baumeister, Evil: Inside Human Violence and Cruelty (New York: Freeman, 1997); Fred Emil Katz, Ordinary People and Extraordinary Evil: A Report on the Beguilings of Evil (Albany, NY: State University of New York Press, 1993).
18. For more on the failures of official channels in theory and practice, see Martin, Justice Ignited; Brian Martin, 'Illusions of Whistleblower Protection', UTS Law Review 5 (2003): 119-30; Thane Rosenbaum, The Myth of Moral Justice: Why Our Legal System Fails to Do What's Right (New York: HarperCollins, 2004).
19. Brian Martin, 'The Beating of Rodney King: The Dynamics of Backfire', Critical Criminology 13, 3 (2005): 307-26.
20. Brian Martin, 'Iraq Attack Backfire', Economic and Political Weekly 39 (17 April 2004): 1577-83.
21. Donald Bloxham, The Great Game of Genocide: Imperialism, Nationalism, and the Destruction of the Ottoman Armenians (Oxford: Oxford University Press, 2005).
22. Philip Frankel, An Ordinary Atrocity: Sharpeville and its Massacre (New Haven, CT: Yale University Press, 2001).
23. Andrew Herd, 'Amplifying Outrage over Children Overboard', Social Alternatives 25 (Second Quarter 2006): 59-63.
24. Hannah Lendon and Brian Martin, 'Environmental Disasters', in Martin, Justice Ignited, 99-112.
25. An example is 'cultural genocide' that operates through policies prohibiting use of tradition languages and cultural practices. I thank an anonymous referee for this example.
26. Gregory H. Stanton, 'The 8 Stages of Genocide' (New Haven, CT: Yale University Program in Genocide Studies, Working Paper GS01, 1998), http://www.genocidewatch.org/images/8StagesBriefingpaper.pdf (accessed 13 August 2009).
27. See for example Howard Adelman and Astri Suhrke, eds., The Path of a Genocide: The Rwanda Crisis from Uganda to Zaire (New Brunswick, NJ: Transaction, 1999); Nigel Eltringham, Accounting for Horror: Post-Genocide Debates in Rwanda (London: Pluto, 2004); Mahmood Mamdani, When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda (Princeton, NJ: Princeton University Press, 2001); Peter Uvin, Aiding Violence: The Development Enterprise in Rwanda (West Hartford, CT: Kumarian Press, 1998).
28. Among the most useful sources are African Rights, Rwanda: Death, Despair and Defiance (London: African Rights, 1995, rev. ed.); Article 19, Broadcasting Genocide: Censorship, Propaganda and State-Sponsored Violence in Rwanda 1990-1994 (London: Article 19, 1996); Michael Barnett, Eyewitness to a Genocide: The United Nations and Rwanda (Ithaca, NY: Cornell University Press, 2002); Roméo Dallaire with Brent Beardsley, Shake Hands with the Devil: The Failure of Humanity in Rwanda (New York: Carroll & Graf, 2004); Alison Des Forges, 'Leave None to Tell the Story': Genocide in Rwanda (New York: Human Rights Watch, 1999, 2d ed.); Jean Hatzfeld, Machete Season: The Killers in Rwanda Speak (New York: Farrar, Straus and Giroux, 2005); Linda Melvern, A People Betrayed: The Role of the West in Rwanda's Genocide (London: Zed Books, 2000); Linda Melvern, Conspiracy to Murder: The Rwandan Genocide (London: Verso, 2004); Power, 'A Problem from Hell'; Scott Straus, The Order of Genocide: Race, Power, and War in Rwanda (Ithaca, NY: Cornell University Press, 2006).
29. Article 19, Broadcasting Genocide, 16; Des Forges, 'Leave None to Tell the Story', 93.
30. Des Forges, 'Leave None to Tell the Story', 90-1.
31. Article 19, Broadcasting Genocide, 44; African Rights, Rwanda, 236-40.
33. Des Forges, 'Leave None to Tell the Story', 283.
34. Ibid., 595ff.
35. Dallaire, Shake Hands with the Devil, 394: 'Repeated requests to Western nations for aerial photographs and satellite pictures fell on deaf ears'. Dallaire says the Russian government offered to sell satellite pictures, but UN peacekeepers in Rwanda had no budget for this. See also Power, 'A Problem from Hell', 354.
36. Melvern, A People Betrayed, 47-8.
37. Ibid., 48-9.
38. Power, 'A Problem from Hell', 506.
39. Melvern, A People Betrayed, 152.
40. Barnett, Eyewitness to a Genocide, 110.
41. Ibid., 159.
42. Kingsley Chiedu Moghalu, Rwanda's Genocide: The Politics of Global Justice (New York: Palgrave Macmillan, 2005), 87; Article 19, Broadcasting Genocide, 3.
43. Article 19, Broadcasting Genocide, 109.
44. African Rights, Rwanda, 238, 254.
45. Barnett, Eyewitness to a Genocide, 150.
46. Melvern, A People Betrayed, 138.
47. Hatzfeld, Machete Season, 44. Killers had a legal right to remain silent, but this was not the reason for their reticence, because Hatzfeld was interviewing them separately from the legal process.
48. Ibid., 127-8, 137-9; Straus, Order of Genocide, in extensive interviewing took many precautions to elicit valid responses.
49. Sam Keen, Faces of the Enemy: Reflections of the Hostile Imagination (San Francisco: Harper & Row, 1986).
50. On the politics of the Hutu ruling group, see for example Des Forges, 'Leave None to Tell the Story'; Mamdani, When Victims Become Killers.
51. Article 19, Broadcasting Genocide, 67-8; African Rights, Rwanda, 42-3.
52. Straus, Order of Genocide, 131.
53. Melvern, A People Betrayed, 71; Article 19, Broadcasting Genocide, 113; Des Forges, 'Leave None to Tell the Story', 66-71.
54. Melvern, A People Betrayed, 155.
55. Article 19, Broadcasting Genocide, 90.
56. On the role of humiliation in genocide, including in Rwanda, see Evelin Lindner, Making Enemies: Humiliation and International Conflict (Westport, CT: Praeger Security International, 2006). I thank an anonymous referee for recommending Lindner's work in this context.
57. Hatzfeld, Machete Season, 213-5.
58. Ibid., 132.
59. Ibid., 47.
60. Straus, Order of Genocide, 8-9, 96-7, 173.
61. Des Forges, 'Leave None to Tell the Story', 258.
62. Article 19, Broadcasting Genocide, 138, 140.
63. Des Forges, 'Leave None to Tell the Story', 252, 254, 256.
64. Ibid., 74ff.
65. Ibid., 86.
66. Léon D. Saur, 'From Kibeho to Medjugorje: The Catholic Church and Ethno-Nationalist Movements and Regimes', in Genocide in Rwanda: Complicity of the Churches?, ed. Carol Rittner, John K. Roth and Wendy Whitworth (St. Paul, MN: Paragon House, 2004), 220.
67. Article 19, Broadcasting Genocide, 110.
68. Ibid., 111.
69. Ibid., 114-9, 125.
70. Des Forges, 'Leave None to Tell the Story', 259.
71. Ibid., 284-5; African Rights, Rwanda, 250-1.
72. Des Forges, 'Leave None to Tell the Story', 19-20, 641-2.
73. Power, 'A Problem from Hell', 383-4.
74. Barnett, Eyewitness to a Genocide, 133-8; Alain Destexhe, Rwanda and Genocide in the Twentieth Century (New York: New York University Press, 1995), 50; Melvern, A People Betrayed, 177; Power, 'A Problem from Hell', 359-64; Des Forges, 'Leave None to Tell the Story', 641-2.
75. Melvern, A People Betrayed, 171-2.
76. Des Forges, 'Leave None to Tell the Story', 628.
77. Hatzfeld, Machete Season, 44.
78. Ibid., 156.
79. Ibid., 154; Straus, Order of Genocide, 157-60, also reports extensive use of war language.
80. Hatzfeld, Machete Season, 240, 145.
81. African Rights, Rwanda, 240ff; Des Forges, 'Leave None to Tell the Story', 12, 196ff; Straus, Order of Genocide, 173.
82. Mamdani, When Victims Become Killers, 214.
83. Des Forges, 'Leave None to Tell the Story', 595ff.
84. Ibid., 93.
85. Melvern, A People Betrayed, 202-3.
86. Power, 'A Problem from Hell', 508.
87. African Rights, Rwanda, 1134.
88. Article 19, Broadcasting Genocide, 144; Des Forges, 'Leave None to Tell the Story', 641.
89. Gérard Prunier, 'Opération Turquoise: A Humanitarian Escape from a Political Dead End', in Adelman and Suhrke, Path of a Genocide, 281-305.
90. Dallaire, Shake Hands with the Devil.
91. Barnett, Eyewitness to a Genocide, 90.
92. Ibid., 144; Des Forges, 'Leave None to Tell the Story', 25.
03. Melvern, A People Betrayed, 163.
04. On the shortcomings of official channels after the genocide, specifically humanitarian action on behalf of Rwanda refugees in Zaire, among whom former génocidaires reorganised and mounted attacks into Rwanda, see Fiona Terry, Condemned to Repeat? The Paradox of Humanitarian Action (Ithaca, NY: Cornell University Press, 2002), 155-215. I thank an anonymous referee for recommending Terry's book in this context.
95. Des Forges, 'Leave None to Tell the Story', 225, 391; African Rights, Rwanda, 993ff; Straus, Order of Genocide, 135-48.
96. Straus, Order of Genocide, 135-48.
97. Hatzfeld, Machete Season, 119; Straus, Order of Genocide, 141-4.
98. Hatzfeld, Machete Season, 112.
99. Des Forges, 'Leave None to Tell the Story', 236.
100. Ibid., 263-4; Mamdani, When Victims Become Killers, 218-21.
101. African Rights, Rwanda, 177-99.
102. Des Forges, 'Leave None to Tell the Story', 236.
103. Article 19, Broadcasting Genocide, 44.
104. The expression 'des Milles Collines', in English 'of a thousand hills', is another name for Rwanda, so there is no special connection between RTLM (Radio Télévision Libre des Milles Collines) and the Hôtel des Milles Collines besides their location in Rwanda.
105. African Rights, Rwanda, 238, 719ff; Des Forges, 'Leave None to Tell the Story', 633.
106. Power, 'A Problem from Hell', 368.
107. Des Forges, 'Leave None to Tell the Story', 642.
108. African Rights, Rwanda, 1148; Des Forges, 'Leave None to Tell the Story', 681.
109. Des Forges, 'Leave None to Tell the Story', 263.
110. African Rights, Rwanda, 1124ff.
111. Straus, Order of Genocide, 7, 151.
112. Gene Sharp, The Politics of Nonviolent Action (Boston: Porter Sargent, 1973), 207-234, documents the importance of nonviolent discipline.
113. Shaw, War and Genocide.
114. Article 19, Starving in Silence: A Report on Famine and Censorship (London: Article 19, 1990).
115. Leo Kuper, Genocide (Harmondsworth: Penguin, 1981).
Three reviewer reports were provided.
Reviewer #1 stated, among other things, that "Theoretically, the use of tactics as a framework, the author offers nothing new or insightful about genocide prevention beyond what Gregory Stanton provides with his eight stage classification." (I had earlier sent the paper to Gregory Stanton; he was enthusiastic about it.)
Reviewer #2 made several useful comments, but preferred "to see drawing up numerous examples throughout -- Armenian Genocide, Holocaust, Bosnia, Rwanda, Sudan -- in a comparative context".
Review #3 stated "It offers little that is new. Its value consists of the concrete application of behavior patterns to a specific case. ... it smacks to me of an undergraduate paper."
The paper was rejected.
The editors reported that the paper had been studied by the editorial team plus three external reviewers. The editors provided considerable feedback in a combined form (in other words, without separate reviewer reports). Among the comments were the following.
"I feel uneasy with this text. I see a number of problems. I have severe doubts whether it makes sense to think here in terms of categories of mere 'tactics'. This is one of the genocide studies texts with their manichean worldview that can only imagine the evildoers as being conscious of how evil they are and therefore using 'lies', etc. What if (part of) the culprits really think that way, think they are in a legitimate fight, and so on?"
"The author calls for a peaceful overcoming of genociders and blames the RPF in Rwanda of having delivered a pretext for 'reinterpretation' of what he sees as pure genocide, while disecting extermination in Rwanda out of the context of civil war."
The editors said "the reviewers have unanimously advised us not to consider your manuscript for publication in the JGR".
Three helpful referee reports were received. The paper was accepted after revisions.