Case Study 13.2: Radio Rwanda — RTLM and the 1994 Genocide

"Cut down the tall trees. Cut down the tall trees." — RTLM broadcast, April 6, 1994, using the term "Inkotanyi" (Tutsi RPF fighters) — widely understood to mean: kill the Tutsi

Overview

If the yellow journalism case study demonstrates how commercially motivated print propaganda can help make a war possible, the case of Radio Télévision Libre des Mille Collines (RTLM) demonstrates something incomparably more horrifying: how a single radio station, deploying the specific cognitive affordances of the broadcast medium in a context of severe prior radicalization, contributed directly to the systematic murder of approximately 800,000 people in 100 days.

The RTLM case is the defining modern example of radio as a genocide instrument. It is also the case that most clearly demonstrates the relationship between a propaganda channel's specific properties — intimacy, accessibility, the authority of the spoken word, the ability to deliver real-time operational instructions to a mass audience — and the catastrophic potential of those properties when weaponized by genocidal ideology.

This case study examines four questions: (1) What was RTLM and who built it? (2) What specific propaganda techniques did RTLM deploy, and how did they exploit radio's channel properties? (3) What is the documented relationship between RTLM broadcasts and specific acts of genocide? (4) What did the international criminal tribunal find, and what legal precedents resulted?

A note on approach: This case study deals with mass atrocity in explicit terms. The material is presented with the clinical precision it requires. The purpose is analysis, not sensationalism. Understanding how this propaganda worked is part of understanding how to recognize and resist it.

Rwanda 1994: Essential Context

Understanding RTLM requires understanding the context it operated in. The 1994 genocide did not emerge from radio alone; it emerged from a century of colonial racial categorization, decades of institutionalized discrimination, and years of deliberate radicalization by a political elite facing an existential threat to its power.

Colonial construction of ethnic identity: Rwanda's Tutsi and Hutu populations had historically been social categories with some ethnic dimension but also significant fluidity — a Hutu who acquired cattle could become Tutsi; a Tutsi who lost cattle could become Hutu. Belgian colonial rule, beginning after WWI, hardened these fluid categories into rigid racial identities. Belgian administrators introduced identity cards in 1933 that assigned every Rwandan a fixed "ethnic" category based on cattle ownership (the 10-cow rule: ten or more cows made one Tutsi). The Belgians then administered Rwanda through Tutsi intermediaries, creating a colonial system that assigned superior status to Tutsis and resentment to Hutus — a resentment that Belgian administrators occasionally cultivated as a tool of colonial control.

The 1959 Revolution and its aftermath: In 1959, Hutu political leadership organized a revolution that killed thousands of Tutsis and drove approximately 150,000 into exile. Rwanda's independence in 1962 established a Hutu-dominated government that systematically discriminated against the remaining Tutsi population. Periodic anti-Tutsi violence occurred through the 1960s, 1970s, and 1980s. The children of the 1959 exiles formed the Rwandan Patriotic Front (RPF), primarily Tutsi, which invaded Rwanda from Uganda in October 1990 and began a civil war that continued through 1994.

The Arusha Accords and the hardliners' crisis: International pressure produced the Arusha Peace Accords in August 1993, which provided for a power-sharing arrangement between the Habyarimana government and the RPF. For the Hutu Power hardliners around President Habyarimana — the akazu (inner circle), primarily his wife's family — this was an intolerable prospect. Power sharing with the RPF meant the end of their exclusive access to state resources and, they feared, accountability for past crimes.

Radicalization before RTLM: The groundwork for genocide was laid before RTLM went on the air. The Kangura newspaper, beginning in 1990, published the "Hutu Ten Commandments" — a document that explicitly called for the economic and social exclusion of Tutsis and instructed Hutus not to associate with Tutsis. The term inyenzi ("cockroach") as a derogatory term for Tutsis was established in use before RTLM; RTLM did not invent dehumanization but amplified and extended it. The Interahamwe (Hutu Power militia, literally "those who attack together") were organized and armed beginning in 1992–1993. The genocide had planners, logistics, and organizational infrastructure that radio alone did not create.

What RTLM did was activate this prepared machinery, extending its reach beyond the literate, urban population that read Kangura to the rural, lower-literacy population that listened to radio.

RTLM: The Station

Radio Télévision Libre des Mille Collines (Free Radio and Television of a Thousand Hills — "a thousand hills" being a common description of Rwanda's landscape) began broadcasting on July 8, 1993. It was established by a group of businessmen and political figures closely connected to President Habyarimana and the akazu. Its principal founders included Ferdinand Nahimana, a historian and director of the government's Office Rwandais d'Information; Jean-Bosco Barayagwiza, a senior diplomat and leader of the Coalition pour la Défense de la République (CDR), the most extreme Hutu Power political party; and Hassan Ngeze, editor of Kangura.

RTLM was technically private — it was not officially a government station — and this nominally private status was part of its design. The Rwandan government was under international pressure to maintain press freedom; an officially government station inciting ethnic hatred would have attracted international criticism. A "private" station with the same ownership network could pursue the same agenda while maintaining deniability.

The station was enormously popular from its first broadcasts. Its format was deliberately designed for mass appeal: it played popular music, particularly Congolese soukous and Rwandan folk music, interspersed with news, commentary, and political analysis. The tone was conversational, intimate, sometimes jokey. The presenters — particularly Valérie Bemeriki and Noël Hitimana — cultivated personal, parasocial relationships with listeners through a casual radio-conversation style that was entirely unlike the formal government broadcasts Rwandans were accustomed to.

This format was itself a propaganda technique. By embedding political content within popular entertainment, RTLM normalized its listening as a routine pleasurable activity rather than as ideological instruction. Rwandans who would not have sought out explicit political propaganda tuned in for the music and received the politics as part of the same experience. The entertainment frame lowered critical defenses in exactly the way that propaganda studies predict.

Phase One: Satirical Radicalization (July 1993 — April 1994)

From its launch through the assassination of President Habyarimana on April 6, 1994, RTLM's content can be characterized as systematically radicalizing while maintaining the entertainment format and nominal deniability.

Dehumanization and the "cockroach" terminology: RTLM's broadcasts consistently used the term inyenzi ("cockroach") for Tutsis and, increasingly, for any Hutu who was perceived as insufficiently committed to Hutu Power ideology. The term had been in prior use, but RTLM deployed it with a specific purpose: dehumanization is not merely offensive, it is functionally preparatory for violence. Research in genocide studies (James Waller, Gregory Stanton, and others) has documented consistently that mass atrocity requires prior dehumanization — the psychological preparation that makes killing possible by denying the victim's moral status as a human being. RTLM's repetitive use of animal metaphors for Tutsis was not merely hateful; it was preparatory.

Historical revisionism and threat construction: RTLM broadcasts consistently framed the current moment through the prism of 1959 and the RPF invasion of 1990: the Tutsis had always planned to re-enslave the Hutus; the RPF's military campaign was the execution of a longstanding plan; all Tutsis in Rwanda were potential collaborators with the RPF. This framing served to define the entire Tutsi population — not just RPF combatants — as a threat. It answered the question that genocide perpetrators always have to answer: "Why are we killing civilians?" The answer RTLM provided was: there are no Tutsi civilians; they are all enemies.

Named targets and "radio lists": In what would become the most operationally direct propaganda technique RTLM deployed, the station began broadcasting the names of specific individuals — both Tutsi and "moderate" Hutu — identified as enemies. Addresses, vehicle descriptions, license plate numbers, and locations were broadcast. These broadcasts occurred both before and during the genocide. The named individuals were subsequently killed by Interahamwe militia. The function was unmistakable: RTLM was serving as an operational targeting system for genocidal violence.

Undermining the peace process: RTLM broadcasts from late 1993 through early April 1994 consistently attacked the Arusha Accords, the United Nations peacekeeping force (UNAMIR), and any political figure — Hutu or Tutsi — who supported the power-sharing agreement. Moderate Hutu politicians were named as traitors; UNAMIR's commander, Canadian General Roméo Dallaire, was denounced. The effect was to delegitimize all non-genocidal alternatives in the political environment.

Phase Two: Direct Incitement (April 6 — July 1994)

President Habyarimana's plane was shot down over Kigali on the evening of April 6, 1994, killing him and the president of Burundi. Within hours, the killing had begun. Within 24 hours, RTLM's broadcasts had shifted from what could be described as "merely" radicalizing to explicit incitement.

The April 6 broadcasts and the hours that followed: RTLM immediately attributed Habyarimana's death to the RPF and Tutsi "accomplices" — without evidence and before any investigation. This attribution served a specific function: it provided the trigger narrative. The genocide had been planned; it needed a trigger event and a narrative explanation for that event. RTLM provided both, in real time, to a national audience.

In the days that followed, RTLM broadcasts included:

  • Instructions to set up roadblocks and check identity cards (the Belgian-era identity cards that identified Tutsis)
  • Directions to locations where Tutsi families were sheltering
  • Encouragement to Interahamwe militia who called in from the field
  • Explicit language calling for killing: "Cut down the tall trees" (buza inzitane) was one formulation; there were others more direct
  • Mockery of international observers and appeals for outside intervention, constructing these as interference in a necessary Rwandan "self-defense" operation

The Italian journalist working in Rwanda during the genocide, Alberto Bompani, described RTLM during April 1994 as functioning as a "real-time command and control system" for the genocide: listing roadblocks that needed more personnel, identifying places where Tutsis were hiding, congratulating militia units for their work. General Dallaire, in his memoir Shake Hands with the Devil, describes RTLM as the "voice of the genocide" — the medium through which the killing was coordinated, celebrated, and sustained.

The entertainment format in the killing phase: Even during the genocide, RTLM maintained its music-and-talk format. Killers who took breaks from mass murder could listen to Congolese soukous and then return to work to more RTLM encouragement. The normalization function of the entertainment format continued operating throughout: the killing was framed as necessary, righteous, even joyful work. Presenters expressed satisfaction at reports of large numbers killed. This normalization of atrocity through media is one of the most psychologically disturbing aspects of the RTLM record.

Radio and Genocide: The Causal Evidence

The question of whether RTLM caused the genocide is, like the yellow journalism question, complicated — and for the same structural reasons. The genocide was not caused by radio alone; it was organized by political networks, carried out by armed militia with extensive prior training and coordination, and was possible only because of a century of colonial racial categorization and decades of discrimination that preceded RTLM's existence.

But the research on RTLM's specific contribution is more direct than most media-effect research, for a specific reason: researchers have been able to examine the geographic and temporal relationship between RTLM broadcast reach and killing rates.

David Yanagizawa-Drott's study "Propaganda and Conflict: Evidence from the Rwandan Genocide" (2014, Quarterly Journal of Economics) is the most rigorous quantitative analysis. Yanagizawa-Drott exploited the natural variation in RTLM signal strength across Rwanda's mountainous terrain: some areas had strong RTLM reception, others had weak or no reception. If RTLM affected killing rates, areas with stronger reception should have higher killing rates, all else equal.

The findings were stark. Areas with better RTLM reception had substantially higher participation in violence, even controlling for prior ethnic composition, income levels, and other variables. Yanagizawa-Drott estimated that RTLM incitement accounts for approximately 10 percent of total violence during the genocide — roughly 51,000 deaths attributable specifically to radio incitement. This number is itself almost incomprehensible; but it also means that the overwhelming majority of the genocide's approximately 800,000 deaths cannot be attributed to RTLM alone.

The 10 percent figure should be understood carefully. It represents the marginal effect of RTLM — the additional violence attributable to radio, over and above the violence that occurred in areas without radio access. It does not mean radio caused only 10 percent of the violence; it means radio added approximately 10 percent to the violence that would have occurred anyway. Given that the baseline violence was catastrophic, 10 percent of that baseline represents an enormous causal contribution.

In 1999, the International Criminal Tribunal for Rwanda (ICTR) indicted three individuals for their roles in creating and operating RTLM and Kangura: Ferdinand Nahimana, Jean-Bosco Barayagwiza, and Hassan Ngeze. Their trial, known as the "Media Case" (Prosecutor v. Nahimana, Barayagwiza, and Ngeze), concluded with convictions in December 2003.

The ICTR's Media Case made several findings of historic importance for international law:

The finding on incitement: The tribunal found that RTLM's broadcasts constituted "direct and public incitement to genocide" under Article III(c) of the Genocide Convention, and that this incitement contributed to the commission of genocide. The tribunal was careful to distinguish between speech that expresses hatred (however repugnant) and speech that constitutes incitement to specific violent action. RTLM's broadcasts — naming individuals as targets, providing operational information to killers, encouraging specific acts of violence — crossed the line from the former to the latter.

Nahimana's conviction: Ferdinand Nahimana, who had founded RTLM and directed its editorial policy, was found guilty of genocide, conspiracy to commit genocide, and crimes against humanity. The tribunal found that as RTLM's intellectual architect, he bore command responsibility for its incitement broadcasts. He was sentenced to life imprisonment (later reduced to 30 years on appeal).

The media freedom question: The defendants' primary defense was that their broadcasts were protected speech under international human rights law, and that the prosecution constituted illegitimate state interference with freedom of expression. The tribunal rejected this argument on the grounds that the specific operational incitement that characterized RTLM's April-July 1994 broadcasts — naming targets, directing violence, encouraging killers — was not protected speech under any reasonable interpretation of international human rights standards. The ruling drew an explicit distinction between hate speech (which receives at least some protection under international law) and incitement to genocide (which does not).

The precedent: The ICTR Media Case established that individuals who design, control, and operate media used for genocide incitement can be criminally liable for the genocide itself — not just for the speech. This built on the precedent established at Nuremberg with Julius Streicher's conviction, extending it to radio as well as print. The principle: when a media operation serves as an operational instrument of mass atrocity, its operators bear criminal responsibility for that atrocity.

RTLM and Channel Analysis

The Rwanda case illustrates, with terrible clarity, why channel analysis matters for propaganda studies.

Radio penetration beyond literacy: Rwanda in 1994 had a literacy rate of approximately 45 percent — one of the lower rates in sub-Saharan Africa, a legacy of colonial under-investment in education. Kangura newspaper's genocidal content reached literate, primarily urban audiences. RTLM reached rural areas, less-educated populations, people who had never read Kangura and might not have been reached by print propaganda at all. Radio's bypass of the literacy barrier dramatically expanded the potential audience for genocidal incitement.

Intimacy and authority in a low-trust information environment: In a country with limited media infrastructure, RTLM was experienced as a personal, trusted voice in a way that was probably more powerful than in media-saturated societies with multiple competing information sources. Listeners who tuned in daily for music developed parasocial relationships with presenters that carried real authority. When those presenters said the Tutsis were coming to kill you — that the RPF was at the gates, that your neighbors were collaborators — the message arrived through a channel of established trust.

Simultaneity and collective action: Genocide requires collective action at scale — large numbers of people acting in coordination at roughly the same time. Radio's simultaneity was operationally critical: RTLM could coordinate the timing and location of roadblocks, communicate where people were hiding, and signal to dispersed militia units that the killing was sanctioned, ongoing, and expected. Print could not have achieved this coordination function; radio's real-time broadcasting capacity was the essential feature.

The normalization function of entertainment format: By embedding genocidal content within an entertainment format that Rwandans had chosen to listen to for pleasure, RTLM achieved a normalization effect that straightforward political propaganda could not have reached. The music made the ideology feel like part of daily life; the casual, intimate presentation made the ideology feel like common sense rather than extremism.

Discussion Questions

  1. Yanagizawa-Drott's analysis estimated that RTLM was responsible for approximately 10 percent of the genocide's total violence. Some find this number surprisingly low; others find it impossibly high. How should we think about quantitative estimates of media propaganda effects on mass violence?

  2. The international community had significant advance warning about the genocide and about RTLM's role in preparing for it. UN Special Rapporteur René Degni-Ségui had documented the hate media situation in Rwanda before April 1994. General Dallaire sent his famous "genocide fax" to UN headquarters in January 1994. What, if anything, does the failure of outside intervention tell us about the relationship between propaganda analysis and political will to act?

  3. The RTLM case is sometimes used to argue for strict limits on hate speech in media regulation. What arguments can be made on both sides of this debate, using the Rwanda case? Are there legal and political mechanisms short of speech restriction that might have addressed the RTLM situation?

  4. Comparing RTLM to the yellow journalism case: both used media propaganda that contributed to mass violence (a war vs. a genocide). What are the key similarities and key differences in the propaganda mechanisms, the media structures, and the scale of harm?