Real and imagined facts in Rwandan history

Jos van Oijen writes that Michela Wrong in her new book moves around real and imagined facts and witnesses, revives a double genocide theory based on inflated casualty numbers, re-labels victims, discredits bona fide genocide experts and promotes layman’s opinions as irrefutable evidence, while revising the history of the genocide against the Tutsi. Van Oijen makes an appeal for properly corroborated and verified research.

By Jos van Oijen

My critique of Michela Wrong’s Do Not Disturb in September was by and large a technical discussion to explain how the book allocates undue credibility to fringe theories and recycled myths by neglecting better evidence from forensic investigations, judicial inquiries and academic research. Without this flaw, which affects about two-fifths of Do Not Disturb, it could have been a useful book of 250 pages. Instead, it’s a problematic one that mixes facts and fabrications.

Several commenters – university professors among them – nevertheless manage to politicize my review, using an “us” versus “them” dichotomy. According to their binary logic, the “us” category is reserved for fierce critics of the Rwandan government, like Michela Wrong. In contrast, others who don’t fall in line and highlight major flaws in her work are labeled pro-government and filed under “them”.

Judging by the emails addressed to the editors of the ROAPE website, “us” means good and “them” means bad. Let’s keep in mind, however, that unless fabricated or manipulated, facts have no side. They are what they are, whether they confirm a popular theory or expose it as a hoax. That’s reality, not politics.

In this follow-up post I provide additional examples of myths, recycled in the book, to illustrate my observation that the intuitive method employed by the author is prone to error. Fact checking the information obtained from informants and other informal sources is an essential component of investigative journalism. How else are we to know if a story is based in reality or on a fiction? Intuition isn’t a useful tool to determine this. Feelings and impressions are not facts, they can’t substitute critical thinking.

However, my central argument – that the book would have benefited from applying professional principles and guidelines – does not exclusively concern one author or a single book. Hidden behind the errors there can be myths created decades ago. Their appeals to emotion or to a suggested logic have convinced other authors before Michela Wrong, sufficiently to evade scrutiny for years.

The Hourigan Affair

The manner in which hundreds of scholars and journalists have discussed the story of International Criminal Tribunal for Rwanda (ICTR) investigator Michael Hourigan, perpetuates an international conspiracy theory, which, we will see, is not even remotely plausible. The case revolves around a memorandum written by Hourigan in 1997. It was intended for ICTR chief prosecutor Louise Arbour to inform her about intelligence collected from alleged whistle blowers in the Rwandan Patriotic Front (RPF).

Do Not Disturb summarizes the Hourigan Affair as follows: “… when ICTR investigators Jim Lyons and Michael Hourigan, a former FBI staffer and an Australian lawyer, respectively, were approached in 1997 by three former RPF fighters claiming direct knowledge of [Paul] Kagame’s responsibility [for assassinating President Habyarimana], they excitedly called Louise Arbour to say they had compiled a dossier outlining grounds for prosecution.”

Arbour invited Hourigan to her office in The Hague to provide her with the details. A couple of weeks later a meeting was arranged where Hourigan delivered his memo conveying the essence of the witness statements. Unexpectedly, the story goes, Arbour reacted with hostility. She dismissed the information and ended the investigation. “Shocked investigators,” Wrong writes, meaning Lyons, “speculated that Kigali’s various Western “friends” – most likely, the US government – had applied pressure. What prompted her about-turn remains unclear.”

Wrong’s source for the cover-up is a brief interview with Lyons in the controversial BBC documentary Rwanda’s Untold Story. It’s risky to rely on a TV program, however, as they’re rarely adequate sources and should be treated with caution. This one omits the fact that Lyons’ contribution was almost entirely hearsay. When Hourigan met Arbour in The Hague, Lyons was in New York as his one-year contract had expired. He has never met the informants and didn’t discuss the incident with Arbour to get her side of the story.

The “outside pressure” element was debunked in 2010 by Hourigan when he was interviewed about the incident at a conference in Brussels. About his meeting with Arbour in The Hague he said: “I presented her there a memorandum … about informants’ information. She read that, but thereafter it was completely different.” If Arbour’s attitude changed after she finished reading the memo, any clues to the cause of her “about-turn” must be in the text. The section of the memo relevant for this discussion is reproduced below, in Fig. 1.

The many journalists and scholars who have cited the memo tend to highlight statements in the text that appear to confirm a consensus view of fifteen years ago: that Kagame was responsible and that a missile was fired from Masaka hill. The other information has been largely ignored, especially in the academic literature. This is odd because it exposes the affair as a hoax.

Luc Reydams has dedicated an entire journal article to the Hourigan Affair in 2018, titled Politics or Pragmatism. It quotes large portions of an affidavit Hourigan submitted in 2007, but the memo – which was attached to it – and what it claims (that FAR [government] soldiers were responsible and that Camp Kanombe, where the Presidential Guard was stationed, served as mission control), are ignored.

French professor André Guichaoua, who is widely regarded as a leading expert on Rwanda, has published the memo on his website as an annex to his book From War to Genocide. However, the places where it mentions Camp Kanombe are covered with black bars (see Fig, 1). Why his readers aren’t supposed to take note of that information remains unclear. Guichaoua didn’t respond to questions.

Figure 1: The original text is quoted on the left; Guichaoua’s Annexe 49 is on the right

We can imagine Louise Arbour’s astonishment when she read the memo. If the suggestion that the assassination was a joint venture of the RPF and FAR was farfetched, there were more glaring errors: Gasogi Hill as a shooting location, for instance. Even in 1997 it was understood that the missiles were fired from the left side of the plane. Gasogi was on its right. Not helping matters was Hourigan telling Arbour that, like Jim Lyons, he had neither met the witnesses, nor verified their information.

The team member who handled the informants was Amadou Deme. This is a notable aspect because in his 2014 memoir Deme describes his close friendships with genocidaires like Georges Rutaganda and Aloys Ntabakuze that go back to the pre-genocide era. This raises questions about the integrity of the investigations he had been involved in. In the memoir, Deme still refers to his genocidal buddies as “falsely accused heroes”.

Unsettling as this knowledge may be, it doesn’t seem to bother Filip Reyntjens, another leading expert according to some. In his 2020 Working Paper on the Habyarimana assassination, Reyntjens still evokes the obsolete information of Hourigan and Deme as credible evidence to support his own conclusions.

When I contacted Louise Arbour about this case, she explained that she didn’t know Hourigan very well when he contacted her. She wasn’t encouraged by what she learned when she inquired about him. “I felt at that time that our capacity to investigate thoroughly that event was seriously compromised: our resources (human and otherwise) were not adequate to the task and our operating from Kigali made it even more problematic.”

Arbour wasn’t surprised at the negative reactions to her decision to shelve the case: “The lens of a journalist, or that of a historian, is obviously not the same as those of a prosecutor, constrained by rules of legal relevance, admissibility of evidence and a very high standard of proof, all necessary to engage personal criminal responsibility. I did not think at the time that we could meet those standards in that case.”

An analysis of the relevant jurisprudence by ICTR defense lawyers Peter Robinson and Golriz Ghahraman supports Arbour’s decision. It  concluded, “… the inconclusive determination of whether the attack constituted perfidy or treachery, or instead a permissible ruse of war, makes it more prudent not to bring such a prosecution, and to leave the debate to scholars and historians.”

Whether it is wise to leave it up to scholars and historians in this polarised academic field is contestable as well. The tendency to resist new information that threatens one’s belief, is a major factor on both sides of the debate. The examples in this and my previous blogpost demonstrate that influential scholars may have great difficulty adjusting to evidence which emerges after they’ve already formed their opinions and published their analyses.

Michela Wrong, primed by a man on TV and unaware of the factual and historical facets of her information, displays the same dismissive attitude towards research that invalidates the stories she believes to be true and has staked her reputation on.

Re-labeling genocide victims

The next example is one that I mentioned briefly in my previous blogpost. But due to its nature – the recycling of a hate radio message from May and June 1994 – it merits a more detailed description. The flaw in Wrong’s method is quite similar to the one in the first example. She apparently found the story in a fringe report, didn’t verify it, and took it as credible evidence to support her argument.

On 25 April 1994, fishermen from the village of Kasensero in Uganda noticed the remains of Rwandan genocide victims floating in the Kagera river, close to where it flowed into Lake Victoria. Over the next few weeks, tens of thousands of bodies would follow. In the first weeks of May they were reported from Butare in southwestern Rwanda, 900 kilometers upstream from Kasensero, all the way to Musoma, 250 kilometers across the lake in Tanzania.

A clean-up operation retrieved 11,000 bodies from along the shore, but the total number of victims in the lake was estimated to be in the range of 25,000 to 50,000. Because the operation was sponsored by Western aid organizations, international journalists flocked to Uganda to report the story. Most victims were women and children, they wrote, murdered in the most horrific manners. Here was tangible evidence of systematic killing on such a massive scale that the regular deflections and excuses of the Hutu Power government in Rwanda no longer sufficed.

This PR problem was resolved quickly by mirroring foreign news reports on the Rwandan radio stations with the difference that the perpetrator-victim roles were reversed. The alternative version broadcast by the hate radio towards the end of May and throughout June blamed the RPF for the catastrophe and re-labeled the genocide victims in Lake Victoria as innocent Hutus killed in RPF territory.

This version was kept in circulation after the genocide was over, first by the ousted regime and army leaders, followed by a motley crew of genocide deniers and lawyers of genocide suspects. The latter group finally managed to push it into the mainstream media by persuading journalists it was one of the examples of long suppressed evidence of RPF atrocities.

The persuasive element in the story was the fact that after the RPF had captured the town of Rusumo on 30 April 1994, a large stretch of the Akagera river was under their control, although most of it was in a game reserve. Wrong’s colleague Judi Rever rehashed the story in her book In Praise of Blood: “Near the end of April, the most southern prefecture of Kibungo was securely held by the RPF. The refugees escaping to Tanzania were therefore not Tutsis, but Hutus being chased and killed by the RPF. The corpses dumped in the Kagera from late April onward were Hutus.”

This explanation sounds reasonable until one takes a look at the map (see Fig. 2 below). The Akagera isn’t the only major river in Rwanda. Upstream from Rusumo lie the Nyabarongo and the Akanyaru. These rivers meander through regions where in April 1994 most Tutsis were concentrated before their combined currents discharge into the Akagera.

The troops movements as depicted in Alison Des Forges’ Leave None to Tell the Story and other sources suggest that on 1 May 1994, around 90% of the riverbanks along the main rivers in inhabited areas were in government territory. A few reports incriminate RPF troops suggesting they’re accountable for at least some of the victims. However, the genocide in the rest of the country did not abate after the RPF took Rusumo, which implies that the vast majority of bodies reaching Lake Victoria towards the end of May and early June would still have been genocide victims. For a general idea of the situation, I’ve combined the data in Fig. 2.

Figure 2: Map of rivers in the region.

The version in Do Not Disturb is slightly different from the Radio Rwanda/RTLM/Judi Rever version. Wrong writes: “Reporters would later recall, with retrospective unease, how eerily quiet the first areas captured by the RPF had always seemed. When bodies with their hands tied kandoya-style behind their backs surfaced in Lake Victoria, brought by the Kagera and Nyabarongo Rivers, many observers wondered how fresh genocide victims could be washing down from areas the RPF had long cleared of interahamwe.”

The “fresh genocide victims” element caught my attention. The distances by river between Rwanda and Lake Victoria seemed too great for a dead body to arrive in Lake Victoria and still look‘fresh’, given the tropical conditions. The stretch from Butare and Gikongoro to Kasensero is more than 900 kilometres. From Rusumo it’s still 525 kms. But I always try to check such extraordinary claims, just to be sure. In this case I received help from an engineer at Delft University of Technology and two other specialized scientists. The exercise led me to conclude that if fresh bodies indeed surfaced in Lake Victoria, they would have been thrown in locally, not in Rwanda.

So where did Michela Wrong get those ideas? The only reports I could find which mention fresh bodies in the water are from the post-genocide period and they do not refer to Lake Victoria. The scale is incomparable as well. In July 1994 a UNHCR spokesperson told journalists: “We are seeing between 10 and 20 bodies a day floating down the river past Ngara [in Tanzania, opposite Rusumo], “some of them are very, very fresh bodies, so the killing continues.”

Did Wrong mix up the reports? There is a more likely explanation. When I was reviewing publications authored by Paul Rusesabagina a couple of weeks ago, I came across a version of the ‘lake bodies story’ that’s almost identical to Wrong’s. In his report ‘Compendium of RPF crimes‘ from 2006, the details and opinion are all the same, which suggests that the mistake originates with him. Wrong could have read it, paraphrased it in her book, and forgot to credit it to Rusesabagina.

Moral choice

In Do Not Disturb, Michela Wrong moves around real and imagined facts and witnesses, revives a double genocide theory based on inflated casualty numbers, re-labels victims, discredits bona fide genocide experts and promotes layman’s opinions as irrefutable evidence, while she casually revises the history of the genocide against the Tutsi as if she’s redecorating her living room. This can’t all be explained in detail without turning this blog into another book. Most omissions and distortions have layers of facts, or a history attached to them, as demonstrated by the examples above. Unfortunately, people generally remember false facts and alternative histories better because they’re simple and catchy compared to the refutations that are often more complex and require some effort to understand.

Other reviewers have highlighted how Wrong portrays her informants as Prince Charmings, and their former comrades as scary monsters. She even defends an informant who is in prison for embezzling a few hundred thousand pounds from a charity. Proximity matters. So does cognitive dissonance. Two of these Prince Charmings who play a prominent role in the book, fell out with each other as well, one now accusing the other of genocide. And the dissidents who claim to be witnesses of the plane attack have refuted each other’s testimonies before the investigative judges. Such bubble-bursting details are rationalised in passing and remain largely obfuscated in the book.

Journalists and scholars who write about Rwanda and the genocide are from a wide spectrum of disciplines and backgrounds: historians, cultural anthropologists, sociologists, psychologists, criminologists, forensic scientists, political scientists, law professors, human rights activists, linguists, economists, medical doctors, engineers, and so forth. Besides these we have of course the genocide survivors and other eyewitnesses, priests, attorneys, diplomats, aid workers all contributing to the literature. The advantage of this diversity is that together they produce a multifaceted perspective providing opportunities for reaching a comprehensive analysis of the genocide, its history and aftermath, and to learn from it.

But it also causes friction, misunderstandings and personal resentments that, while they endure, hinder rather than advance our understanding. Everything will stand or fall on the quality of the data and the ability to identify independently established facts from opinions. Theories and judgments should follow the evidence, not the other way around. If a claim is made that according to science, the laws of nature, or the logistics involved is simply not possible, it must be discarded, and the theories should be adjusted accordingly. Unfortunately, not everyone is prepared to do that.

Let me conclude this blogpost by reminding the reader that my critique of Do Not Disturb has addressed methodological flaws that affect about two-fifths of the book. The other 250 pages are largely accurate and worth reading. As long as the author checks her facts and makes clear distinctions between personal feelings, beliefs, assumptions, opinions, facts, and evidence, there’s no crime in letting people have their say, even if what they say is sometimes offensive.

Jos van Oijen is an independent researcher from The Netherlands who publishes on genocide-related issues in various online and print media. His first blogpost on Michela Wrong’s book Do Not Disturb: The Story of a Political Murder and an African Regime Gone Bad on roape.net can be read here.

Featured Photograph: The International Criminal Tribunal for Rwanda (ICTR) in Arusha, Tanzania.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.