Polarising the debate over Rwanda
Simplistic arguments on both sides of the 'genocide debate' just make it harder to learn the truth about the 1994 killings
René Lemarchand recently summed up the problem very well: “Few events in history are more subject to controversy than the mass killings commonly designated as genocide. This is hardly surprising considering the lack of anything like a consensus of scholarly opinion about the precise meaning of the term, the different interpretations of the phenomenon offered by social scientists, and the enormous emotional charge it carries. Rwanda is no exception. There are few parallels for the sheer depth of the discords and disagreement the 1994 genocide has generated among observers, survivors and perpetrators”.1
Discords and disagreement, in one sense, should of course be welcomed. As Scott Straus recently put it, “the Rwandan genocide is a very difficult, even impossible, story to tell”.2 It is so difficult precisely because there is no one story to be told: the enormous volume of literature now available on the genocide allows for many different interpretations and narratives. Attempting to tell 'the story' of the Rwandan genocide has two detrimental effects. Firstly, it invites macro-level generalisations that fail to do justice to genocide as a composite of thousands of individual acts of violence, each with its own specificity and context.3 Secondly, it feeds into the singular focus on the genocide at the expense of other causes of excess mortality in Rwanda throughout the early and mid-1990s. Both these phenomena have unfortunately affected a large proportion of research into the events of 1990s Rwanda. As Lemarchand notes, there is a clear “one-sidedness [to] the early accounts of the tragedy” that lead those accounts to “reflect the official version of the facts projected by the [Rwandan Patriotic Front]”, the rebel army made up mostly of Tutsi immigrants from Uganda that took control of Rwanda in the genocide's aftermath.4
As a result of this earlier one-sidedness, numerous critiques of earlier research have now emerged. Many of these critiques are quite reasonable and justifiable in light of the identified deficiencies of early accounts. However, the revisionist church in this 'genocide debate', if it can be called that, is a broad one. Indeed, early interpretations of the events of 1990s Rwanda have provided space for accounts that place themselves at the opposite end of the spectrum. I'm talking about interpretations that deny a genocide occurred in Rwanda in 1994. The majority of these accounts largely suffer from the same deficiencies as the earlier accounts that were biased in the opposite direction; yet these revisionist accounts are increasingly gaining legitimacy.
I want to briefly explore, through a recent example, the way these revisionist accounts are leading to a polarisation of the debate over the events in Rwanda – where one is accused of being either an unwitting stooge of RPF (or worse, United States) propaganda, or a despicable genocide denier. This polarisation threatens to make it even harder to develop a more nuanced picture of the different forms of violence that took place – a task which surely we owe to the victims of that violence to attempt.
Raging against the “mainstream camp”
This April, Monthly Review Press published a new book by Edward Herman and David Peterson, The Politics of Genocide. The book is presented as an examination of the politicised nature of the label 'genocide' - “that in the United States it is used by the government, journalists, and academics to brand as evil those nations and political movements that in one way or another interfere with the imperial interests of U.S. capitalism”. Herman and Peterson's book comes complete with praise from political commentators and journalists whose opinions I'm sure most radical and progressive activists would value highly. It also comes with a forward by Noam Chomsky. The book is consequently bound to receive widespread attention from the Western Left. It's likely many who read the book will be approaching the topic of 1990s Rwanda for the first time, perhaps having previously only heard journalistic accounts that present wildly misleading narratives of good versus evil or ethnic hatreds. Those of us who are critical of mainstream media will therefore be on the lookout for narratives that critique these views, and for narratives that scrutinise the role of Western governments. It's important to recognise that this is the frame of mind through which many will approach Herman and Peterson's book.
Herman and Peterson's discussion of Rwanda, which makes up the middle section of their book and which is available online (http://www.monthlyreview.org/100501herman-peterson.php – all quotes which follow are from this text), can be summarised in their words as follows: “To a remarkable degree, all major sectors of the Western establishment [have] swallowed a propaganda line on Rwanda that turn[s] perpetrator and victim upside-down”. This “official propaganda line” is one of “a Hutu conspiracy to commit genocide”, an “established narrative” of “800,000 or more largely Tutsi deaths resulting from a “preprogrammed genocide” committed by “Hutu Power”” which according to Herman and Peterson “appears to have no basis in any facts, beyond the early claims by Kagame's RPF and its politically motivated Western sponsors and propagandists”. This “successful framing” of events has served to mask reality, which according to Herman and Peterson is that “the mass killings were not directed against the Tutsi population” and that “the chief responsibility for Rwandan political violence belonged to the RPF, and not to the ousted coalition government, the FAR, or any Hutu-related group”. This violence was supported by the United States, among other reasons, as a way for “Washington [to gain] a strong military presence in Central Africa” in the form of Paul Kagame's government.
Readers unfamiliar with the literature on 1990s Rwanda may not fully appreciate the significance of this argument: it runs counter to the overwhelming majority of research, itself made up of literally hundreds of pieces of evidence, a vast number of eyewitness accounts, a number of UN-sponsored investigations and dozens upon dozens of academic studies, on the mass killings of 1994 which argue that a genocide was carried out aimed at Rwanda's Tutsi population. That such a body of work exists does not of course invalidate Herman and Peterson's claims; it does, however, place a heavy burden of proof upon Herman and Peterson to both provide strong evidence for their claims and to show where others who argue a genocide did occur have gone wrong in their research.
As I see it, Herman and Peterson fail to satisfy that burden of proof. Their chapter on Rwanda suffers from a number of problems regarding evidence and interpretations which for reasons of space I won't cover too thoroughly here; the Canadian academic Gerald Caplan does a good job of this himself in his review article 'The politics of denialism' (about which more later). I want to focus on the ways in which Herman and Peterson encourage a polarisation of the debate over the 1994 killings.
First, despite numerous references to “the establishment narrative”, “the mainstream camp” and “Western circles”, there is barely any engagement with the vast literature on 1990s Rwanda. One need only skim through Lemarchand's review article 'Rwanda: The State of Research' to get a sense of the depth of research that has been conducted into the mass killings; since his article was written, further important research has also been conducted. If one compares the bibliography of Lemarchand's article with that of Herman and Peterson's chapter on Rwanda, it emerges that the latter make barely any reference to the dozens of academics, human rights campaigners, journalists and eyewitnesses whose work anyone wanting to write on the events of 1994 Rwanda is expected to be familiar with. No attempt is made by Herman and Peterson to explain how the majority of research on 1990s Rwanda has managed to get it so wrong in its collection of evidence to support a genocide thesis. The one person whose work they do discuss, Alison Des Forges, is dismissed on account of her having “worked for the U.S. Department of State and National Security Council” prior to carrying out her investigation for Human Rights Watch into the 1994 killings. Herman and Peterson conclude that Des Forges' work “helped provide cover for the U.S. takeover”. In making this assertion, Herman and Peterson do not engage at all with the actual contents of Des Forges' work or indeed demonstrate that her work on Rwanda is incorrect in its conclusions.
Secondly, Herman and Peterson repeatedly claim that important evidence which supports their thesis has been 'suppressed' or 'ignored'. One of the most significant issues in this context is that of the shooting-down on April 6 1994 of the plane carrying the Presidents of both Rwanda and Burundi. It is widely acknowledged that this event precipitated the mass killings. Herman and Peterson claim that that “the established perpetrator-victim line requires suppression of the crucial fact that the April 6 shooting-down of the government jet... was carried out by RPF commandos”. In fact, far from being suppressed, there has been open discussion of the shooting-down of the plane and the possibility of RPF involvement for a number of years. Lemarchand explicitly states that “there is a growing body of evidence pointing to [the RPF's]... alleged participation in the military operation that brought down Habyarimana's plane”.5 Both the prospects of trying President Kagame for Habyarimana's death and the indictments of RPF officials by French and Spanish judges have also been discussed, for example in the Journal for International Criminal Justice.6 It should also be noted that RPF responsibility for the shooting-down of the plane is by no means certain: for instance in January 2010 an extensive Rwandan special commission headed by Jean Mutsinzi concluded that extremist Hutus in the then-government were responsible for shooting down the plane (http://mutsinzireport.com).
More generally, Herman and Peterson claim that “[t]o accept the standard model of “The Genocide”, one must ignore the large-scale killing and ethnic cleansing of Hutus by the RPF”. It would seem, however, that a key person held up by Herman and Peterson as exemplifying the “mainstream camp”, Alison Des Forges, manages to accept the 'standard model' and pay attention to RPF crimes. In fact, a large proportion of evidence related to RPF crimes that Herman and Peterson discuss in their book is discussed in Des Forges' investigation of the killings, Leave None to Tell the Story. Furthermore, many of the Rwanda scholars that Herman and Peterson do not discuss at all – Lemarchand, Scott Straus, Mahmood Mamdani, Colette Braeckman, Alan Kuperman, Filip Reyntjens – both accept that a genocide occurred against the Tutsi and highlight documented killings committed by the RPF.
The final crucial way in which Herman and Peterson contribute to polarisation is in their interpretation of evidence, or perhaps more accurately the implicit notion that their interpretation is the only one that works. This is best understood through another claim of 'suppressed evidence'. Herman and Peterson cite a “key verdict” of the International Criminal Tribunal for Rwanda (ICTR) in December 2008, that four former high-ranking Hutu members of the Rwanda military were acquitted of the charge of participation in a conspiracy to commit genocide. As Patrick Karuretwa notes, for those who want to deny a genocide occurred, “the concept is quite simple: no planning = no intent = no genocide”.7 Indeed, this gets to the core of Herman and Peterson's argument and the resultant polarisation in the debate over the 1994 killings. The crux of their argument is that, while the Hutu members of the Rwandan government were in no real position to plan a genocide against the Tutsi, “the RPF was the only well-organised killing force within Rwanda in 1994, and the only one that planned a major military offensive”.
Herman and Peterson's conclusion, however, is not by any means the only one that can be drawn from the evidence they present. Once again, academic discussion of this point in the “mainstream camp” is not nearly as closed-minded as Herman and Peterson suggest. Lemarchand cites the work of Michael Mann and James Gasana, who separately argue that the plan envisaged by Hutu extremist elites was most likely that of selective killings or politicide aimed at the political opposition; then, as the regime lost cohesion and both it and ordinary Hutu civilians felt threatened by RPF actions, a wider plan was improvised by opportunistic elites – from limited to total genocide.8 This conclusion is supported by Scott Straus, whose micro-level field research suggests that rather than involving elite planning and machine-like state orchestration, the mass killings of 1994 were the outcome of complex and uneven local power struggles between moderate and extremist Hutu, struggles whose impetus lay in destabilisation caused by the RPF offensive.9 As Lemarchand notes, Straus “refutes (or seriously qualifies) the notion of a planned total genocide”.10 Yet he argues there was a genocide nonetheless.
The implications for discussion of 1990s Rwanda can be put as follows: by refusing to engage with the vast majority of research into the mass killings in Rwanda while at the same time claiming evidence has been suppressed or not discussed by that research, and by providing selective interpretations of such evidence that do not do justice to the interpretations presented by other research, Herman and Peterson imply a homogeneous “mainstream camp” that denies everything they themselves assert. From this, Herman and Peterson seem to assume that acknowledging the role of the Rwandan Patriotic Front and the Western governments that either passively or actively supported its actions requires that one deny that a genocide occurred – either you open your eyes to 'what really happened' or you're an unwitting defender of the United States' imperial designs for the Great Lakes region. If as I've suggested The Politics of Genocide receives a wide readership among the radical left, and if Herman and Peterson's conclusions are taken at face value by those who don't know much about either the region or previous research into it, discussion of Rwanda will become polarised, as readers accept the parameters of the debate provided by this book.
The politics of name-calling
While Herman and Peterson promote polarisation on one 'side' of the 'genocide debate', reaction so far to The Politics of Genocide has worked to promote it from the other. The most high-profile review of the book has been from Gerald Caplan, the Canadian academic and founder of the Remembering Rwanda movement. Caplan's review, 'The politics of denialism', does a good job as I've said of pointing out the lack of rigour in Herman and Peterson's collection of evidence to support their argument. Caplan's framing of debate over the 1994 killings, however, is in many ways as problematic as those he criticises.
The problem with Caplan's framing is its characterisation of both 'sides' of the debate. On one side are “the overwhelming number of those who have ever written about the genocide”. Significantly, though to a much lesser degree than Herman and Peterson, Caplan portrays this overwhelming number as holding a more unified view of the 1994 killings than actually exists. For example, Caplan refuses the notion put forward by Herman and Peterson that the RPF “were perceived as serving U.S. interests”; Caplan writes that “[n]o other historian of the genocide of whom I'm aware makes this claim and no evidence for it exists”. While it's true that no other major writer understands the RPF solely as a tool of U.S. imperialism, much has been written of the diplomatic support given to the RPF by the United States. Barrie Collins documents the role of the U.S. in pressuring Habyarimana, often with the threat of withdrawal of international funds, throughout the 1990s to treat the RPF as a legitimate opposition movement and to concede to some of its demands.11 Crucially, as Alan Kuperman argues, the 1993 Arusha accords, which favoured the RPF, were signed by Habyarimana under international pressure and despite the acknowledgement of U.S. officials that hard-liners among the Hutu elite would never accept the measures.12 There was indeed a perception, as Collins notes, in the Rwandan government “that the RPF enjoyed discreet American approval”.13
Linked to this issue of U.S. support is Caplan's assertion that “[a]lmost every well-known writer on the genocide condemns the international community, led by the U.S., for refusing to intervene to stop the massacres of the Tutsi”. This image of U.S. non-intervention in Rwanda has been forcefully criticised by Mahmood Mamdani, who writes: “the U.S. did intervene in Rwanda, through a proxy. That proxy was the RPF... Instead of using its resources and influence to bring about a political solution to the civil war, and then strengthen it, the U.S. signalled to one of the parties that it could pursue victory with impunity. This unilateralism was part of what led to the disaster”.14
Arguably more significant in the context of a polarisation of the Rwanda debate is Caplan's characterisation of the other side of the debate. Herman and Peterson are placed alongside a number of other writers branded “genocide deniers”, whose evidence for their arguments comprise “fabrications, distortions, innuendo and gross ignorance”. These deniers are described as surviving by simply citing each other's work, in Caplan's words “gleefully drink[ing] each other's putrid bath water”.
While Caplan's choice of language is arguably unhelpful, it is his treatment of specific writers that is most damaging to rational debate. Caplan includes among these genocide deniers Christian Davenport and Allan Stam. Stam and Davenport, the latter a well-respected Professor of Peace Studies at the University of Notre Dame, have compiled data on killings from April-June 1994 from a wide range of sources, including Rwandan ministries, survivors associations and human rights organisations. Applying a Bayesian latent variable model to the data across 164 administrative communes, they estimate approximately 890,000 victims of killings in the zone under the jurisdiction of the FAR, 77,000 killings in the zone under RPF jurisdiction, and 93,000 in areas contested by the FAR & RPF.15
I contend that by labelling Davenport and Stam genocide deniers, Caplan does them and their research a great disservice. Caplan is right to note controversy around their work – Davenport, Stam and members of their research team have been interrogated and threatened by the Rwandan government. Branding them genocide deniers, however, is quite inaccurate. To quote their research paper: “[c]learly evidence of genocide is evident within this analysis but it is also clear that a variety of other activities exist as well, which merit discussion and consideration within journalistic, scholarly, legal and political circles”.16 As the two have noted elsewhere, “we have never denied that a genocide took place; we just noted that genocide was only one among several forms of violence that occurred at the time”.17
Caplan does have a more specific source of outrage at Davenport and Stam's work, namely their “sensational estimate” that “the majority of victims [were] likely Hutu and not Tutsi”. Caplan says the methodology used to arrive at this “Orwellian assertion” has been “totally discredited”. Unfortunately no source is provided for this 'total discrediting'. It is true that one could level criticisms at aspects of Davenport and Stam's methodology. Their claim that the majority of victims were likely Hutu is based on what they admit is a simple method: subtracting the estimated number of Tutsi survivors (300,000) from 1991 census data on the number of Tutsi in Rwanda (600,000). This number is then subtracted from the number of those estimated to have been killed overall – 800,000 to 1 million. It's likely, however, that in the 1991 census the number of Tutsi was under-reported, perhaps by as much as 40%, by both the Habyarimana regime, in order to keep school and public employment quotas of Tutsi low, and by Tutsi themselves, in order to avoid discrimination.18 The census data also obviously does not take population growth from 1991-4 into account. Nevertheless, in light of the fact that previous scholarly estimates have placed the numbers of Hutu and Tutsi deaths much closer to eachother than the current Rwandan government's estimates,19 to call Davenport and Stam's estimate “Orwellian” is unhelpful. In sum, Caplan's name-calling is just as distorting, and ultimately polarising, as the claims of Herman and Peterson.
'Hacks' and 'facts'
A war of words is already reverberating thanks to The Politics of Genocide and written reviews of it. Those looking at Caplan's review online will see a response to it from international criminal lawyer Christopher Black, who is himself currently Lead Counsel at the ICTR for Augustin Ndindiliyimana, former head of Rwanda's Gendarmerie. Black accuses Caplan of being “in the pay of the RPF military junta” and describes Caplan's list of scholars who support the genocide thesis as “a long list of other RPF hacks”. Yet aside from Black's continuation of the name-calling-game, much of what he says regarding the RPF – that is was committed to gaining state power rather than protecting the Tutsi population, that it chose a military takeover rather than making concessions in negotiations, and that it refused cease-fire offers even as the genocide began – is supported by academic research.20 The problem with Black's argument, as with Herman and Peterson's, is the implicit notion that in order to hold these opinions of the RPF's influence on the mass killings of 1994, one must deny that a genocide was carried out aimed at Rwanda's Tutsi population.
Lemarchand's regret that early accounts of the 1994 killings too often reflected the official position of those who in the end won control of the Rwandan state unfortunately continues to have relevance. In a recent research piece, Marijke Verpoorten argued that “[t]he lack of an open debate on all forms of violence makes it difficult to collect equally accurate information on the different types of violence and assess their relative impact on the population”.21 This lack of an open debate is perpetuated by the Rwandan government under President Kagame, who himself stands accused of war crimes. As many observers, including Rwandan genocide survivors, recognise, the Rwandan government continues to use a one-sided view of the 1994 killings to shield itself from accusations of political repression and authoritarianism at home and crimes against humanity in the Democratic Republic of Congo.22
Because of this political reality, the direction of debate over the events of 1994 has real ramifications for the viability of holding Kagame and members of the RPF to account for their actions over the last two decades. A polarisation of this debate, whereby you're either a RPF stooge or a genocide denier, will surely erode that viability. One can only hope that as increasing amounts of evidence are collected and collated, a more nuanced and complex view of the violence in Rwanda is allowed to develop.
1 Rene Lemarchand, “Rwanda: The State of Research” Online Encyclopedia of Mass Violence, 4 November 2007, http://www.massviolence.org/Article?id_article=51: 2.
2 Scott Straus, Intimate Enemy: Images and Voices of the Rwandan Genocide (Zone Books, 2006): 14-15.
3 Straus, Intimate Enemy: 14-15; Scott Straus, “Origins and aftermaths: The dynamics of genocide in Rwanda and their post-crime implications” in Pouligny, Chesterman & Schnabel (eds.), After Mass Crime: Rebuilding states and communities (United Nations University Press, 2007): 122-4.
4 Lemarchand, “Rwanda: The State of Research”: 7.
5 Ibid: 12-13. See also Rene Lemarchand, “Review Essay: Controversy Within the Cataclysm” African Studies Review 50(1), April 2007: 140-144.
6 See the articles by Peter Robinson & Golriz Ghahraman, Vanessa Thalmann and an anonymous commentator in Journal of International Criminal Justice 6(5), November 2008: 981-1011.
7 Patrick Karuretwa, “Release of Rwanda's mastermind of death promotes genocide denial” The Harvard Law Record, 4 December 2009, http://www.hlrecord.org/opinion/release-of-rwanda-s-mastermind-of-death-promotes-genocide-denial-1.951557.
8 Lemarchand, “Rwanda: The State of Research”: 16-17.
9 Scott Straus, “Origins and aftermaths”.
10 Lemarchand, “Rwanda: The State of Research”: 19.
11 Barrie Collins, “New Wars and Old Wars? The Lessons of Rwanda”, in David Chandler (ed.), Rethinking Human Rights: Critical Approaches to International Politics (Palgrave Macmillan, 2003): 157-175.
12 Alan J. Kuperman, “Provoking genocide: a revised history of the Rwandan Patriotic Front” Journal of Genocide Research 6(1), 2004: 61-84, 75.
13 Collins, “New Wars and Old Wars?”: 163.
14 Mahmood Mamdani, “The Politics of Naming: Genocide, Civil War, Insurgency” London Review of Books 29(5), 8 March 2007: 5-8.
15 Christian Davenport & Allan Stam, “Rwandan Political Violence in Space and Time” Discussion paper, 2009, http://web.mac.com/christiandavenport/iWeb/Site%2040/Publications_files/rwanda031708c.pdf.
16 Ibid: 36.
17 Christian Davenport & Allan Stam, “What Really Happened in Rwanda?” Miller-McCune Research Essay, 6 October 2009, http://www.miller-mccune.com/politics/what-really-happened-in-rwanda-3432/http://www.miller-mccune.com/politics/what-really-happened-in-rwanda-3432.
18 Marijke Verpoorten, “The death toll of the Rwandan genocide: a detailed analysis for Gilkongoro Province” Population 60(4), 2005: 331-368.
19 See the discussion of estimates by Des Forges and Reyntjens in: Lemarchand, “Rwanda: The State of Research”: 12.
20 See: Kuperman, “Provoking genocide”.
21 Marijke Verpoorten, “Detecting Hidden Violence: The Spatial Distribution of Excess Mortality in Rwanda” Discussion Paper 254/2010, LICOS Centre for Institutions and Economic Performance, May 2010: 3.
22 Alice Gatebuke, “Deadly silence: Rwanda's never again, once again?” Pambazuka News 487, 24 June 2010, http://www.pambazuka.org/en/category/features/65430.