Document(s) 16 of 37

The birth of Radio-Télévision Libre des Milles Collines (RTLM) in 1993 could not have come at a better time for Rwanda's Hutu elite. Finally, here was a radio station they could use as a mouthpiece for their ideals and a means to propagate their ethno-political war against the Tutsi-dominated Rwandan Patriotic Front (RPF).

A detailed analysis of the content of RTLM broadcasts makes clear that individual broadcasters – not their guests or government officials – were most likely to use the airwaves to disseminate hate. Based on a sample of taped broadcasts, individual journalists dominated the station's programming and accounted for most of the inflammatory statements. In a departure from usual radio programming, monologues by the station's personalities were the most common format of RTLM broadcasts, followed by interviews. News items took up less than 2 per cent of airtime, whereas instructions and directives issued by the station took up less than 3 per cent.

Three-quarters of the time used to make inflammatory statements occurred when there were no guests in the studio; that is, RTLM's on-air personalities were responsible for most of the station's inflammatory statements.

It is important to remember that RTLM broadcasts were not responsible for introducing the language and ideology of hatred into the Rwandan community. Such language and the ideology of ethnic conflict and polarization already existed in Rwanda in the form of a powerful social construct involving ethnic identity. The language had been further developed in 1990–93 by Kangura and other media allied with Hutu Power.

RTLM, coming at the very end of the conflict, drew its commentary, arguments and interpretation of issues from this existing framework. For example, certain words initially used by Kangura over its four years of publication, 1990–94, would become popular and the vocabulary of the ensuing genocide. According to the testimony of language expert Matthias Ruzindana at the ICTR on 21 March 2002, Kangura popularized words like kurimbura and gutsemba (to massacre), gutsembatsemba (to exterminate), kumara (to eat up), gutikiza and gusakumba (to clean), kumarira kw'icumu (to spear), gutema (to cut); others related to 'working', such as gukora (to work), ibikorwa (the work), ibikoresho (the working), abakozi (the workers), umuganda (communal work); and words for nocturnal patrols, irondo, and the people's war, intambara y'abaturage. In those four years Kangura set a trend. By 1994, the media had a language and tools with which to convince the population to undertake violence as a form of 'self-defence'.

The media served the narrow political ends of their owners by playing on the ordinary Hutu's fear of dispossession, violence and displacement. To do this the media relied heavily on half-truths and sometimes-outright lies and threats. The media defined who the enemy was, and why. Later on, they would also explain what was to be done to this enemy.

By 1994, history and reality had been skewed by both politicians and their media to the maximum effect, as the late Archbishop of Rwanda and former journalist André Sibomana outlines, resurrecting patterns of ethnic conflict and making them a norm.

Through a game of repetition, drop by drop, the media build up moral and cultural constructs which eventually become permanent features ... This failure on the part of party officials and media to stem the progress into the abyss contributed significantly to fuelling a climate of intolerance and turned them into agents of destruction of Rwandan society. (Sibomana 1999: 49)

It was into this politically charged environment that RTLM was born in July 1993, a few months before immense international pressure was brought to bear on the parties in the conflict. The result was that the combined weight of the Rwandan Patriotic Army (RPA) and opposition party representation in the new government seemed to effectively marginalize the former ruling party, Mouvement Républicain Nationale pour la Democratie et le Développement (MRND), and its allies.

The composition of the parliament meant that agreement from at least four parties was required to reach a majority. From a government's perspective, it was clear that the RPF would have an easier time mastering the opposition votes than would MRND. (Jones 1999: 139)

In the following months, tensions increased. In particular, the assassination of Burundi's first Hutu president would be a turning point. Melchior Ndandaye was assassinated by soldiers in the almost exclusively Tutsi Burundi army barely four months after winning a landslide election victory. The assassination gave Hutu politicians in Rwanda reason to wonder if the RPA (another largely Tutsi army) would be satisfied with a negotiated solution to power sharing in Rwanda or if it would seek to extend its control. The assassination was also a convenient example that media such as Kangura used to illustrate the 'power hungry nature' of the Tutsi.

What RTLM clearly did was to reach a greater number of people than Kangura or other media had access to. It popularized the Kangura version of reality. Indeed as the judges at the International Criminal Tribunal for Rwanda stated in their judgement against the founders of RTLM and Kangura: 29. ... RTLM broadcasts exploited the history of Tutsi privilege and Hutu disadvantage, and the fear of armed insurrection, to mobilize the population, whipping them into a frenzy of hatred and violence that was directed largely against the Tutsi ethnic group.

99. ... Unlike print media, radio is immediately present and active. The power of the human voice, heard by the Chamber when the broadcast tapes were played in Kinyarwanda, adds a quality and dimension beyond language to the message conveyed. Radio heightened the sense of fear, the sense of danger and the sense of urgency giving rise to the need for action by listeners. The denigration of Tutsi ethnicity was augmented by the visceral scorn coming out of the airwaves – the ridiculing laugh and the nasty sneer. These elements greatly amplified the impact of RTLM broadcasts. (ICTR summary judgement, see Chapter 25)

Through RTLM, the Hutu elite of the time created an environment in which the 'Tutsis of the past' and the 'Tutsis of the present' became the same. The 'Tutsis of the past', who had – according to mistaken beliefs of the past and contemporary histories – ruled Rwanda with an iron fist and exploited the Hutu mercilessly, and the 'Tutsis of the present', who were – allegedly – invading and trying to take over power, were one and the same. They had to be fought and they had to be taught that the Hutu would never submit to them again.

The general picture painted of the Tutsi community by RTLM was that of a treacherous people, people who had hoodwinked the Hutu, living with them in apparent peace while all the while planning an attack.

In the last months before the genocide, RTLM emphasized the similarity between what was happening in 1994 and what had happened in 1959. It was a choice of going back to the slavery that preceded 1959 or rising up and protecting one's freedom, even one's life. For that, one could not trust any Tutsi civilian because it was not possible to know how deeply the RPF had infiltrated such people.

On 23 May 1994, Ananie Nkurunziza, responsible for political analysis in RTLM, made one of these comparisons between the past and the present. He drew on comments made by a politician in one of the parties formed during independence.

Recently, I am always dealing with similarities; those Tutsis killed the president of the Republic. Therefore the masses got up. The population stood up at once. With the attacks of 1966, which are similar to those of 1994, the population has always been vigilant. In addition, Thaddeus Bagaragaza (of APROSOMA) was asked the reason why, during the 1959 revolution, so many Tutsis were killed, whether they all were on the king's side. He well answered that even Tutsis who were not chiefs were all on the king's side. Here we have to see the similarity of things.

Kantano, we've already said it, Tutsis, Rwandan Tutsis, will understand later perhaps that they shouldn't cling on one fixed idea. Clinging to their Tutsi hood is not a sort of loneliness but a constant will to dominate, a feeling of being special, supreme beings. That's why according to Bagaragaza, a so big number died, because even those who weren't chiefs considered themselves as superior to Hutus.¹

Five days later, Kantano Habimana, another RTLM journalist adds:

Wait until we get enough bullets to chase them away, they will regret what brought them here. They said they will kill all Hutus but what can they use to decimate Hutus? One Tutsi may kill one Hutu, but in the end 6 million Hutus can survive, then what ethnic group would be injured the most? (Kantano Habimana, RTLM broadcast)

After 6 April, RTLM criticized and gave instructions and orders to people in the government. RTLM journalists and government officials sent out mutually reinforcing statements targeting sections of the population that were alleged members of RPA or its political wing RPF and anyone believed or suspected to be supporting it. The political elite made sure that people understood that the government supported the message of RTLM. Prime Minister Jean Kambanda called on the station and in a broadcast told the population that RTLM was 'one of the key weapons in the war'.

The station and government officials reinforced each other's call for all able-bodied people in Rwanda to actively participate in the campaign to find the RPF, the RPA and all persons associated with them. Participation was made an imperative. Those who did not participate and wanted to flee the war were mocked by RTLM and sometimes targeted for death. RTLM, as well as political leaders, made it clear that killing 'the enemy' was the duty of every Rwandan.

RTLM's role in the genocide was to reinforce distorted notions of Rwandan history. It used these notions to describe to the ordinary population the 'threat' posed by RPA and an amorphous group of supporters that ranged from innocent Tutsis to the commander of the UNAMIR forces. It used these mythical images of Tutsi cruelty and treachery to spur people into action by presenting a fait accompli; participate or be subjected to slavery and extermination by the blood thirsty RPA. Kill or be killed.

ANALYSIS OF RTLM BROADCASTS

As part of a master's degree in communication psychology, I conducted a content analysis of RTLM broadcasts. Written transcripts of a selection of broadcasts were used; in some instances, the transcripts were compared with audio recordings. I relied on a database created by the International Monitor group, which contained transcripts of 99 tapes, 56 of which were RTLM broadcasts. The recordings varied in length from 60 minutes to several hours. I also had access to a CD-ROM database used by the International Criminal Tribunal for Rwanda (ICTR) prosecution team, which contained more than 100 transcripts of other broadcasts. Of the transcripts obtained from the ICTR, 16 were examined as part of the content analysis. Thus, in total, 72 transcripts of RTLM broadcasts, ranging in length from one to three hours were examined. The tapes had been transcribed into English and French. I am fluent in English and read French with relative ease.

CLASSIFICATION OF DATA AND DEFINITION OF TERMS

The broadcasts were classified according to the following characteristics:

• Format of broadcast;
  
  
• Name of broadcaster;
  
  
• Name of guest or interviewee;
  
  
• Theme or subject of the broadcast;
  
  
• Group targeted for violence or hatred by the particular broadcast;
  
  
• Type of statement that targets a group or person for violence or hatred.

Format of broadcast

The format of RTLM broadcasts included monologues, dialogues, group discussions, interviews, speeches or statements, songs, news, interrogations, instructions.

'Dialogue' is defined as a conversation between two people while 'group discussion' involves three or more people.

An 'interview' is a formal question-and-answer session, making it different from a dialogue, which is an informal exchange of views. RTLM broadcasts used the interview format with guests, but broadcasters also discussed issues with each other. 'Discussion' has, therefore, been used to describe programmes in which two members of the team are talking about an issue while 'interview' applies when questions are put to outsiders.

'Instructions' are directives on what to do and how to act.

Certain broadcasts were difficult to categorize. For example, like other African radio stations, RTLM often broadcast programmes in which the broadcaster read out greetings and messages from the audience, often interspersed with music and the broadcaster's own commentary and witticisms. I have categorized all such broadcasts as 'other' (see Table 9.1).

Name of broadcaster

Only seven RTLM journalists were well known. However, other freelance journalists could be heard from time to time. The staff who transcribed the tapes were sometimes unable to identify the journalist; these are grouped under 'unknown' as opposed to 'other' who were not regular journalists but were recognized. Jean Baptiste Bamwanga is a Radio Rwanda journalist, for example; however, RTLM borrowed Radio Rwanda material and he is included because of the frequency with which he appeared (see Table 9.2).

Name of guest or interviewee

Those interviewed by RTLM are of particular importance to this research as this is where we begin our study of the interaction of the message and the authority figures. The list (Table 9.3) includes a number of known political and military figures and certain persons linked to the launch of RTLM and Kangura newspaper, in addition to representatives of important parties and groups. This is accurate only insofar as these people speak on behalf of the party or militia they claim to represent.

Theme or subject of the broadcast

From study of an initial sample, it was apparent that the classification in this area should indicate the general tone in which the subject matter was discussed. In addition, certain subjects are classified as general, as the subject matter, which was not political, was neither positive nor negative, but purely information, e.g., sports news, news of other countries, etc. (see Table 9.5).

Group targeted for violence or hatred

This category deals with whether the broadcast called for action against a particular group of people and what kind of action was called for. I classified the data into two broad areas: a broadcast that warned people to beware, watch closely for, be vigilant and wary of infiltration by targeted persons; and directives. In the latter, the broadcaster asks for violent action to be taken against the targeted group (beaten, killed, exterminated) or gives instructions for people to go to a certain area and take similar action against a targeted group (see Table 9.6).

I distinguished between use of the words Inyenzi and Inkotanyi. Inyenzi has connotations that go beyond meaning the RPA and could mean a Tutsi or an associate of the RPF. Broadcasts targeting the Inkotanyi are classified as targeting the RPF.

Type of statement that targets a group or person for violence or hatred

Broadcasts including statements used to target people for harm or violence were more difficult to categorize. In deciding what to call a broadcast likely to create hatred or harm, I relied heavily on Article 20 of the International Covenant on Civil and Political Rights (UN 1966). I also considered what would be considered provocative in the Rwandan context.

The definition I have come up with in describing what I was looking for is: a statement that was likely, not only to portray a group, issue or person in a threatening light, but also likely to impute ill or anti-social motives to a group or person, thereby justifying a response for self-defence, or any direct call for negative action, such as killing, and justification for such action.

In some cases, the broadcast may be justified; for instance the negative portrayal of the RPA after a spate of crippling bombing raids by that group cannot be called calls to hatred. However, these have also been identified and classified as they form part of the debate on what role the RTLM played.

I classified insults and allegations of social deviancy separately. Allegations of social deviance imply that a group or person is no longer acceptable in society and is a threat; for instance allegations of cannibalism have very different implications compared with calling a person a dog.

By the end of the research, I found that some of the categories had no entries or that the type of content I had found in a category did not have any relevance to the research, such categories were subsequently removed from the results (see Table 9.7).

RESULTS

Format of broadcast

The most common format of RTLM broadcasts was monologues by the station's own journalists (Table 9.1). This type of broadcast took up 66.29 per cent of instances. Second to monologues, RTLM relied on interviews. It is interesting to note that news items accounted for only 1.52 per cent of all formats, and instructions and directives issued by the station accounted for 2.56 per cent.

Speeches and statements were also more common than news items, taking up 5.84 per cent of segments. The station also aired interrogations of alleged soldiers of the RPA conducted by officers of the Rwandan Armed Forces. These make up 1.28 per cent of the broadcasts, and one of the interrogations featured an alleged RPA member.

RTLM's format is a bit unusual for a news organization. One ordinarily expects to find news, interviews and other forms of programming taking more airtime than monologues by journalists. Standard practice on many broadcast stations is not to have a person talk on air for longer than a few seconds. Even news bites are short and are structured in that manner on the assumption that lengthy monologues lose audience attention.

Table 9.1 Format of broadcast

*Total may not equal 100% due to rounding

Name of broadcaster

In the research sample, Kantano Habimana was the broadcaster on air most frequently (33.51 per cent of instances), almost twice as often as his editor-in-chief, Gaspard Gahigi (Table 9.2). The second most frequent broadcaster was Valérie Bemeriki, followed by Gahigi. RTLM political analyst Ananie Nkurunziza was the broadcaster in 10.65 per cent of instances. These four RTLM journalists account for 75.76 per cent of appearances by broadcasters.

Ruggiu, broadcasting only in French, accounted for 7.97 per cent of appearances, while Noël Hitimana, who left the service in April after being wounded in a mortar explosion, accounted for 5.02 per cent. Unidentified journalists make up 4.16 per cent, and other journalists visiting RTLM take up 4.24 per cent of the total.

Nkomati does not feature much, which can be explained by the allegation that he was sacked in May for failure to appear at work. It is unclear why Rucogoza is not on air more frequently. Mbilizi, a Congolese national, left RTLM in April and attempted to go back to his home country. Ruggiu probably fails to feature prominently because he could not broadcast in Kinyarwanda.

Table 9.2 Name of broadcaster

Name of guest or interviewee

Analysis of the data on interviews indicates that 74.91 per cent of the time, there was no guest at the station. Kambanda accounted for 2.53 per cent of interviews; government officials, 3.34 per cent; unidentified militiamen and women, 2.17 per cent; CDR militia official Stanislas Simbizi and MDR party official Frodould Karamira accounted for 1.53 and 0.99 per cent, respectively; while 1.45 per cent of interviews were with members of the RPA and RPF. Ordinary civilians took up 7.76 per cent (Table 9.3).

To give a clearer picture of how much airtime each interviewee was given, I recalculated the data based on the proportion of interviews of each as a percentage of total interviews (Table 9.4), that is, I excluded the 830 instances reported in Table 9.3 in which there was no interviewee.

Theme or subject of the broadcast

Analysis of the content of the study sample indicates that 16.32 per cent of RTLM broadcasts studied contained statements alleging atrocities carried out by the RPA (Table 9.5), including indiscriminate bombing of civilians, hospitals and orphanages, indiscriminate killings of civilians and extermination of Hutu populations in areas and zones occupied by the RPA. No inflammatory content was found in 13.21 per cent of the broadcasts studied.

Table 9.3 Name of guest or interviewee

Table 9.4 Time spent interviewing various people as a proportion of total interview time

Table 9.5 Inflammatory content of broadcasts

A clear reminder that Tutsi civilians are not necessarily linked to the RPA and its activities and that the two groups should live together peacefully took up only 0.89 per cent of all broadcasts studied. The call on Hutu civilians to fight for their country, look out for infiltrators, avoid re-enslavement and kill all RPA members and their accomplices took up 13.99 per cent. Direct calls for the extermination of members of the RPA and all its supporters, as well as congratulatory messages on the extermination of RPA members and supporters – Inyenzi, Inkotanyi, accomplices – took up 9.16 per cent of the broadcasts.

Allegations that RPA members were not Rwandans and formed part of a regional conspiracy to dominate Tutsis and set up a Tutsi–Hima empire in central Africa took up 7.05 per cent of the broadcasts; allegations that the RPA wanted to take over the country and reinstate the monarchy, subjugating all Hutus accounted for 5.89 per cent.

Other allegations included statements that Tutsi civilians were part of the regional plan by Tutsis to take over power, or that they were actively helping RPA in this cause and stood to benefit from it (3.16 per cent); that Tutsi civilians were actively exterminating Hutus, making such plans, or supporting them (2.17 per cent); that RPA members killed president Habyarimana, precipitating the unrest in the country (2.78 per cent); and that some political parties were helping the RPA (2.50 per cent).

Insults and ethnic slurs directed at the RPA and Tutsis took up 4.89 per cent, and comments depicting the RPA and Tutsis as social deviants – cannibals, sorcerers, vampires – accounted for 2.28 per cent. Reminders about the history of Rwanda, the fall of the monarchy in 1959 during the Hutu revolution and similarities of those events with the present occurred in 3.89 per cent of the broadcasts. Threats of death or harm to the Belgian contingent of UNAMIR and the UNAMIR commander Roméo Dallaire made up 2.00 per cent. Death threats to Hutu supporters of the RPA or those accepting the fact that RPA was winning the war took up 1.33 per cent of the time, while slurs and insults directed at the two groups made up 2.28 per cent.

In the tapes studied, a total of 69.24 per cent of inflammatory messages were linked to either the RPA or Tutsis.

Group targeted for violence or hatred

Calls for the audience to be on the look out for Tutsis accounted for 10.59 per cent of all warnings and directives (Table 9.6). Words that characterized this type of statement were: look out for, tricks, malice, activities of the Tutsi, Tutsi accomplices, Inyenzi-Tutsi and Inyenzi wherever it was not qualified by the addition of Inkotanyi.

Table 9.6 Group targeted for violence or hatred

Directives included targeting of Tutsis for murder or extermination (3.55 per cent); calls to kill or harm RPA soldiers (9.62 per cent); identification of individuals as supporters or accomplices of the RPA (5.75 per cent); instructions to seek out accomplices named on RTLM (2.26 per cent); and instructions to kill such accomplices (0.58 per cent). Warnings to watch the activities of named individuals took up 4.91 per cent of the total, while instructions to kill such individuals accounted for 1.61 per cent.

Broadcasts citing Tutsi women as spies and asking people to watch out for them occurred in 1.36 per cent of all cases, while those calling on Hutu women to actively participate at roadblocks and in fighting the enemy took up 0.26 per cent.

Statements targeting the Belgian UNAMIR contingent took up 0.90 per cent.

The content analysis indicates that 75.14 per cent of all inflammatory statements occurred when there were no interviewees. Consequently, it is accurate to say that RTLM journalists, rather than their interviewees, made most of the inflammatory statements.

Inflammatory comments by civilian interviewees accounted for 6.44 per cent of the total. Out of the total airtime taken up by inflammatory comments, Prime Minister Jean Kambanda accounts for 3.66 per cent; Ferdinand Nahimana, 2.33 per cent; CDR member and militia man Stanley Simbizi, 2.50 per cent; MDR official Frodouald Karamira, 2.00 per cent; unidentified militia, 1.89 per cent; and government officials (mostly burgomasters), 2.33 per cent.

Out of the 7.16 per cent of airtime taken up by comments that RPA members were not Rwandans and formed part of the regional conspiracy to take over central Africa, 5.99 per cent were made in the absence of any interviewees. Similarly, during the 9.32 per cent airtime when direct calls for extermination were made, there were virtually no interviewees present. It is important to note that only 0.06 per cent of the airtime was taken up by Rwanda government officials participating in RTLM interviews to explain that Tutsis were distinct from RPA and not necessarily linked with its activities, and to encourage coexistence.

Out of the total airtime devoted to allegations of RPA atrocities, those made by government officials during interviews account for 0.50 per cent. Out of the calls alleging that Tutsis as a group were planning or committing extermination campaigns against Hutus, Kambanda was responsible for 0.11 per cent.

Of the 9.35 per cent of airtime taken up in direct calls for extermination, RTLM journalists were responsible for 8.10 per cent while Kambanda accounts for 0.11. Allegations that Tutsis and RPA were social deviants, namely cannibals, vampires, or unfit for society took 2.28 per cent of the airtime; RTLM journalists accounted for 2.00 per cent of this airtime while Kambanda accounted for 0.06 per cent.

The call for all Hutus to participate actively in the events accounts for 14.10 per cent of the airtime. Of this, RTLM journalists were responsible for 9.05 per cent, government officials for 0.78 per cent and Kambanda for 0.62 per cent.

Of 3.27 per cent of the time taken up by allegations that the Tutsi community was planning to take over power and subjugate Hutus, 2.16 per cent can be allocated to RTLM journalists, while Kambanda was responsible for 0.06 per cent.

A study of the types of statements used to rouse the Hutu population to fight, defend themselves and actively participate in the events (Table 9.7) showed that 33.33 per cent were accompanied by allegations of RPA atrocities – claims of indiscriminate bombing, massacres and extermination of Hutus. These allegations also included claims that RPA was eviscerating people. In 24.24 per cent of cases, they were accompanied by the allegation that RPA were not Rwandans and were part of a conspiracy to establish a Hima–Tutsi monarchy in the region; 16.16 per cent included allegations that the RPA wanted power and control over the Hutu and might reinstate the monarchy. Reminders of past evils of the monarchy and links with the armed insurgencies of the 1960s accompanied 14.14 per cent of the calls to violence. These also included reminders of how the Hutus had freed themselves from slavery and subjugation.

Table 9.7 Context in which inflammatory statements were made

In 7.07 per cent of the airtime, calls on the Hutu were accompanied by descriptions of Tutsi civilians and RPA as social deviants. This included claims of cannibalism, drinking victims' blood, sorcery, and unredeemable evil natures that did not belong to normal society. In 2.02 per cent, allegations were made that Tutsi civilians were out to subjugate Hutus, were actively helping the RPA in that cause or hoped to benefit from it. In 3.03 per cent, it was announced that Tutsi civilians were killing Hutus, or planning to do so.

CONCLUSIONS

Those who argue that RTLM was set up with the express aim of instigating mass murder right from the start miss the point. They fail to see the frightening process by which a station that was set up merely to air the political views of one group became the megaphone through which people were incited to mass murder.

The content analysis revealed that each time the RPA made advances, the tenor of the broadcasts would change, becoming more extreme, and more blatantly so, as time went on. The level of inflammatory content began climbing in January 1994 – as it became obvious that implementation of the Arusha accords had floundered – and rose steadily until March. With the April assassination of President Habyarimana, the level reached new heights and continued to rise steadily through June as it became evident that the RPA was winning the war.

Until about the beginning of April, RTLM broadcasts tended to focus on anti-RPA, anti-Arusha peace accords propaganda. Broadcasters focused on renditions of the history of Tutsi–Hutu antagonism and allegations that the opposition parties were helping the RPA block installation of the negotiated government. There was a clear change in tone after the assassination on 6 April and even more so after 11 April as the RPA took over parts of Kigali, forcing the provisional government to flee to Gitarama.

During this second phase, which lasted until mid-May, there were clear calls for extermination of the RPA and elimination of Tutsis and all other accomplices. Names of people were broadcast, sometimes along with instructions on what to do to them. The word extermination was used repeatedly during this period. RTLM also broadcast allegations that Tutsi children, women and old men were being armed, making them legitimate targets.

The messages were very clear: the RPA was advancing to take over the country, Tutsis were helping them and they were working together with the Hima and other Tutsis in the region to install a monarchy in the central African region with the intention of subjugating the Hutu. To achieve their end, the RPA and their Tutsi supporters were willing to do anything including exterminating all Hutu.

RTLM told its listeners that the only way to avoid what was coming was for everyone to look out for the 'enemy in their midst' and work with the authorities to ensure the extermination of the RPA and all its accomplices. It was a fait accompli – kill or be killed. Faced with this constant message and the follow-up actions of their leaders, who organized and led militia in attacks against Tutsi, many ordinary civilians believed the propaganda and actively took part in the killing campaign. Those who were not convinced were often coerced and threatened, sometimes even treated as accomplices to 'the enemy'.

Extermination became a project in which lack of involvement or even fleeing into RPA-controlled sectors signified support for the target group.2 Radio had become a tool for directing mass murder.

NOTES

¹ It is likely that RTLM was actually mistaken in the identity of the person they quoted. I have been told that Thaddeus actually belonged to MDR–Parmehutu and not APROSOMA. It is also likely that the spelling is Thadée and not Thaddeus.

² 'I have always told you. All the people who joined the part controlled by the Inyenzi Inkotanyi are Inyenzi themselves. They approve the killings perpetrated by Inyenzi. They are criminals like the Inyenzi Inkotanyi. They are all Inyenzi. When our armed forces will get there, they will get what they deserve. They will not spare anyone since everybody turned Inyenzi. All those who stayed there are all Inyenzi since those who were against Inyenzi have been killed by Inyenzi. Those who suceeded to escape run away to Ngara, Burundi and to the western part of our country. Those who stayed are accomplices and acolytes of the Inyenzi.' (Valérie Bemeriki, RTML broadcast, 28 June 1994)

REFERENCES

Jones, B. 1999. 'The Arusha Peace Process'. In H. Adelman and A. Suhkre (eds). The Path of a Genocide: the Rwanda Crisis from Uganda to Zaire. Transaction Publishers, New Brunswick, NJ, USA.

Sibomana, A. 1999. Hope for Rwanda: Conversations with Laure Guilbert and Harve Deguine. Pluto Press, London, UK, 205 pp.

UN (United Nations). 1966. International Covenant on civil and political rights. General Assembly Resolution 2200A (XXI), 16 December. Available at <www.unhchr.ch/html/menu3/b/a_ccpr.htm> (accessed 6 September 2005).

Document(s) 16 of 37