March 29, 2002 -- Vol.7, No.1
This article refers to:
Disconnecting the Threads:
Yehuda Bauer, Rethinking the Holocaust, p. 39
Because the Holocaust is often regarded as the apotheosis of genocide and is the best known genocide in the western world, it is the paradigmatic genocide for political manipulation and revising the past Comparisons based on either the Holocaust or the Gulag Archipelago as a single archetype which assume that there is one mechanically recurring script are bound to be misleading.
Helen Fein, Genocide: A Sociological Perspective, p. 55, 56.
The Holocaust and the Rwanda genocide are two of the most terrifying and complex catastrophes of the 20th century. Whether measured by the scale of the atrocities committed against Jews and Tutsi, the distinctiveness of their collective identities, and the deliberate, purposeful manner of their annihilation, there are compelling reasons for seeing in the Rwanda carnage a tropical version of the Shoa. Little wonder if time and again the better known of the two has been used as the paradigmatic frame for analyzing the other.
The aim of this discussion is to challenge -- or at least problematize -- this analogy by placing the concept of genocide in comparative discourse. The sense of revulsion inspired by mass murder on such an appalling scale is no reason to gloss over the singularity of each catastrophe. For if the points of convergence between them are undeniable, to treat Rwanda as the carbon copy of the Holocaust is likely to obscure its historical specificity and regional context, and ultimately lead to a misunderstanding of the motivations behind the killings. Not only does it make short shrift of the very different logics at work in each case, one ideological, the other retributive; it also renders the prospects of national reconciliation in Rwanda even more remote. History, as someone said, never repeats itself, but it sometimes rhymes.
Before going any further, a few notes of caution. Although the title of this paper is meant as a rejoinder to Mark Levene’s effort to identify the “common threads” linking Rwanda to the Holocaust, it is by no means intended to settle scores with the author.  There is much in his discussion that I find illuminating and pertinent. In pointing to its shortcomings – his neglect of the regional and historical contexts – my aim is to raise problems of analysis which are of immediate concern to historians of the Holocaust yet seldom appear to cross the minds of Rwanda specialists, namely the relative importance of context and circumstance as distinct from intention or ideology. This where the ongoing debate among historians of the Holocaust – notably the controversies surrounding the intentionalist and functionalist schools – offers a particularly useful vantage point from which to look at the etiology of the Rwanda genocide. 
Contrary to the impression conveyed by most journalistic accounts, the history of Rwanda does not begin in 1994, or even in 1990, when a group of Tutsi refugee warriors invaded the country, setting in motion an extremely bloody civil war. We need to remind ourselves of the pivotal role of the Hutu revolution of 1959-62, culminating with the overthrow of the Tutsi monarchy and the rise to power of politicians claiming to represent the voice of the Hutu majority. If, as Robert Melson has conclusively demonstrated, revolution and civil war were central elements in the background of the Holocaust and Armenian genocide,  his thesis also finds a perfect illustration in the case of Rwanda. Not the least of the merits of his model is that it offers a framework for understanding not just similarities but differences between the cases at hand. There are indeed significant differences between the Nazi revolution and the Hutu revolution, and Hitler’s disastrous invasion of the Soviet Union (code named Barbarossa) in 1941 has little in common with the 1990 invasion of Rwanda by the Rwanda Patriotic Front (RPF). In one case the perpetrator is the one invading his neighbor, in the other it is the perpetrator who is faced by an invasion from a neighboring state.
The distinction drawn by Helen Fein between ideological and retributive genocide is crucial to the argument set forth here.  Whereas the Holocaust is the classic example of an ideological genocide, rooted in the most stridently racist ideology, the Rwanda genocide is better seen as the byproduct of the mortal threats posed to the revolutionary Hutu-dominated state by the RPF. Like all ideal types these categories are analytic tools, and thus do not take into account the full complexity of real life situations. This is not to suggest therefore that racist propaganda did not play a major role in inciting Hutu mobs to kill innocent Tutsi civilians, only to emphasize the extent to which threat perceptions enhanced the receptivity of the killers to the poisonous ideology distilled on the airwaves of the infamous Radio Mille Collines. Nor is this meant to ignore the anxieties inspired by Nazi allegations of a Judeo-Bolshevik plot, only that such fears belonged to the realm of pure fantasy, whereas in Rwanda they were part and parcel of the every day reality of a vicious civil war 
To put it baldly: Jews did not invade Germany with the massive military and logistical support of a neighboring state; nor did they once rule Germany as the political instrument of an absolute monarchy; nor were they identified with a ruling ethnocracy; nor did Jewish elements commit a partial genocide of non-Jews in a neighboring state 22 years before the Holocaust. Again, Jews did not stand accused of murdering the head of state of a neighboring state (as happened in Burundi with the assassination of Melchior Ndadaye in October 1993). And while Jews were insistently accused by the Nazi propaganda mill of working hand in hand with Bolchevism to subvert the state, at no time did their actions, within or outside Germany, lend the slightest credibility to these accusations. Immensely more threatening was the military posture of the RPF on the eve of the Rwanda genocide. 
The Case for Parallelism
Once this is said, there is a sense in which the analogy remains unambiguously compelling: Tutsi and Jews share a sense of victimhood for which here are few other parallels in recent or past history; both have been the target of a “total domestic genocide”, to use Melson’s phrase. It is not a matter of coincidence if the Rwanda genocide makes immediate claims on the collective memories of Jews everywhere, if Jewish commentators are instinctively drawn to identify with the agonies of the Tutsi, and if an exceptionally close relationship has since developed between the state of Israel and post-genocide Rwanda. Referring to the “murderous trauma which they have respectively endured”, William Miles notes that “it is in this vein that contacts between the RPF and the sate of Israel have been close; that cash-strapped Kigali maintains an embassy in Jerusalem; and the Israeli branches of the Simon Wiesenthal Center and Amcha have actively assisted in survivor treatment and national healing programs in Rwanda, as well as advising Rwandan prosecutors on conducting war crime trials. In this sense post-1994 Tutsi are post-Shoa Jews”. 
The analogy applies at another level as well, albeit a more complex one. In a fascinating discussion of the problems involved in “ ‘Judaizing’ the Rwanda Genocide”, Miles draws attention to their collective self-awareness as “chosen peoples”, not to mention the “positive affinities” supposedly inscribed in their biblical “sibling relationship”: “Tutsi are also Jews in the more problematic sense of being – albeit in the African context – a ‘chosen people’, one whose historically privileged status stemmed from colonial favoritism. (According to some Hamitic interpretations, divine patronage also played a part in Tutsi superiority)”.  But as the author would readily concede, it is one thing to be “chosen” by a colonial administration imbued of the racially-inspired notions of 19th century European ethnography, and quite another to be a divinely chosen people (am nichvar) in the biblical sense. Of this crucial distinction Miles is fully aware. Nor is he oblivious of the ideological convergence between Hamitic and Aryan mythologies.  For there is indeed a sense in which Tutsi claims to superiority -- whether induced by early European colonizers and missionaries, or stemming from a culturally ingrained disposition to see themselves as belonging to a higher order of humanity -- remind one of nothing so much as of the way in which pseudo-scientific theories were turned into racist myths at the hands of Nazi ideologues. 
“Who, then, in the moral universe of Holocaust parallelism, are the Tutsi? Are they ‘the Jews’, victims of intended extermination? Or are they ‘the Nazis’, putative embodiment of a superior race?”  The question Miles raises is one that defies simple answers. His position is unambiguous: “To posit that they are, in some sense, both Aryans and Jews is unacceptable”. I am inclined to think otherwise: seen through the prism of history once might conceivably argue that they are both Aryans and Jews, albeit at different moments of their destiny. Their “Arianness” is inscribed in the explicitly racist connotations of the Hamitic hypothesis -- in part also in some of the myths of origins associated with the birth of the Nyginya monarchy – and their Jewishness in their tragic destinies, their shared victimization at the hands of a racist state.
This is only one of the many ambivalent issues raised by “Judaizing” the Rwandan genocide. Another has to do with the parallel relationship between the Nazi and Hutu revolutions, on the one hand, and genocide on the other.
The Historical Nexus: Revolution and War
Seldom is history determined by accident or contingency. Hitler alone does not explain the Holocaust, any more than Habyalima or his entourage, the so-called akazu, are the causes of the Rwandan tragedy. Like any major event in history genocide must be contextualized. No one has done it more effectively than Robert Melson in his remarkable inquest into the roots of the Armenian and Nazi genocides. Rather than looking at any single individual or culture or mentalité for an explanation, he shows how in each case the combination of revolution and war provided the structural opportunities for the systematic extermination of Jews and Armenians.
While providing the condition for the coming to power of “ideological vanguards”, revolutions redefine “the identity of the political community as the ‘people’, the ‘nation’, the ‘class’, the ‘race’”; the occurrence of war heightens of the sense of vulnerability of the new community, and creates strong ties between domestic and external foes: “Those that earlier have been labeled as ‘the enemies of the revolution’ are part of an insidious plot with the regime’s international foes to undo revolution or even to destroy the state and the political community itself”. Neither “expulsion, assimilation or segregation” are viable options in dealing with such threats; systematic extermination is the only solution.  More often than not, genocide is the deliberate, calculated response of the self-appointed custodians of the revolution to the menace posed by counterrevolutionaries at home and abroad.
The Melson thesis brings to light a crucial parallel between the Holocaust and Rwanda. In both instances the roots of genocide are traceable to the same lethal mix of revolutionary fervor and wartime conditions. Closer scrutiny of the evidence, however, shows that there are fundamental differences in the types of revolutionary upheavals experienced by each state, and the character of the war that followed in their wake.
All revolutions involve the drastic and violent restructuring the social order, but not all revolutions stem from the same ideological roots. The Nazi revolution was nothing if not stridently anti-semitic, aiming at the regeneration of state and society in the name of Aryan “purity”. Under the leadership of the Fuhrer the master-race would emerge as the only source of salvation in the face of a world Jewish conspiracy. The Rwanda revolution was an entirely different phenomenon. The aim was not to enthrone a “master race” but to end the hegemony of the Tutsi minority, the nearest equivalent of a master race during much of the colonial period, and in so doing free the Hutu masses of the shackles of the Tutsi-dominated monarchy.
The ethnic underpinnings of the Rwanda revolution (1959-1962) cannot be denied any more than the anti-Tutsi violence that has accompanied the rise to power of the Hutu counter-elites. As many as 20,000 Tutsi may have lost their lives (out of a total of some 350,000); tens of thousands fled the country, most of them to Uganda and Burundi, others to the Congo and Tanzania. Although some Tutsi do not hesitate to view the revolution as the first of the several genocides they have suffered at the hands of the Hutu, neither the scale nor the circumstances of the human losses have anything in common with the 1994 carnage.
Paradoxically, the exclusionary implications of the revolution were the flip side of its populist, egalitarian aspirations. The “emancipation” of the Hutu masses meant recognition of the claims of the humble and downtrodden – “le menu people”, to use the self-description most frequently used by Hutu politicians -- against the age-old domination of a “feudo-hamitic” monarchy.  Though utterly oblivious of the rights of the minority, this revolutionary agenda attracted considerable sympathy and support from Brussels. It is noteworthy that the revolution got underway three years before the advent of independence in 1962; a successful transfer of power to Hutu politicians would have been unthinkable in the absence of the wholehearted support they received from the trusteeship authorities and the Church. The tone of the revolution was populist and anti-feudal. Unlike most other varieties of African nationalism, its rhetoric was anything but anti-Western; its target was not the trusteeship authorities but the Tutsi-dominated Union Nationale Rwandaise (Unar), seen by Belgium as a dangerous radical movement, close to Lumumba’s Mouvement National Congolais (MNC) and strategically allied to Communist China. In sum, to see in the Hutu revolution a tropical replica of Nazi revolutionary anti-semitism makes little sense.
What does make sense is the perversion of the Hutu revolution into an increasingly anti-Tutsi crusade, culminating in the years following the RPF invasion with an outpouring of rabidly racist propaganda. It is tempting – but quite misleading -- to explain the Rwanda genocide by projecting the events of 1994 into the past, and infer therefrom an undiluted commitment to racism on the part of the Habyalimana regime and its predecessor under the presidency of Grégoire Kayibanda. To cite but one example, Peter Uvin refers to “the long-standing and deeply ingrained racism of Rwanda society”, noting that “for decades Rwandan society has been profoundly racist. The image of the Tutsi as inherently evil and exploitative was, and still is, deeply rooted in the psyche of most Rwandans; this image was a founding pillar of the genocide to come”.  This naively ahistorical view of the roots of the genocide makes short shrift the fact that ethnic discrimination was indeed the hallmark of the traditional Tutsi monarchy long before it was appropriated by Hutu ideologues  -- and suggests an obvious parallel with the Goldhagen argument: just as the Holocaust is historically linked to a long tradition of “eliminationist anti-Semitism”, the Rwanda genocide is likewise anchored in a long-standing legacy of anti-Tutsi racism.  This is a gross oversimplification of a far more complex reality. Overt, officially-sanctioned racism, as distinct from “anti-feudal” or “anti-monarchical” propaganda, was largely absent from the political discourse of Hutu revolutionaries in the 1950s. There was no Mein Kampf to provide ideological direction to the revolution, no Fuhrer to instill hatred in the minds of the masses, no lebensraum to justify conquest, no Final Solution to deal with Tutsi threats (at least not until 1993, when the assassination of Melchior Ndadaye in Burundi set in motion a trend towards “cumulative radicalization” best illustrated by the rise of Hutu Power). 
Anti-Tutsi sentiment has never been a constant in Rwanda history. It waxed and waned depending on the historical moment. The key variable was the salience of threat perceptions, a phenomenon closely linked to the political events in the region. Anti-Tutsi violence reached ominous proportions in 1963-4, in the wake of repeated attempts by Tutsi commandos – the so-called inyenzi, or “cockroaches” in Kinyarwanda  -- to fight their way back into the country. The most disastrous of such raids occurred in December 1963, when a group of armed Tutsi refugees from Burundi nearly captured the capital city; in response an estimated 5,000 Tutsi civilians were murdered by Hutu mobs in the Gikongoro prefecture. Ineffective though they were in bringing down the republic, the impact of these incursions on the radicalization of the Kayibanda regime has been profound. Several new elements, all of them harbingers of a future apocalypse, came into focus: (a) the growing identification of the enemies of the revolution with foreign enemies, (b) the conflation into the same subversive frame of exile and resident Tutsi elements, the latter supposedly acting as a spies (ibiyetso) for the former; (c) the radicalization of the domestic arena through the elimination of moderates. 
The following years saw a distinct lowering of the ethnic tension. As cross-border raids came to an end, so did anti-Tutsi violence -- only to resurface immediately after the 1973 genocidal slaughter of Hutu by Tutsi in neighboring Burundi. Scores of Tutsi students were killed by their Hutu schoolmates in secondary schools and on the campus of the National University of Rwanda, in Butare, causing another major exodus of Tutsi civilians to neighboring states. The 1973 pogroms played a major role in the army coup that brought Habyalimana to the presidency of the Second Republic, and in vesting power in the hands of Hutu elites from the north. Possibly to enlist their support against its domestic opponents the new regime at first showed unmistakable signs of sympathy towards the Tutsi minority. Few today seem to recall that in the years following his accession to power in 1973, President Habyalimana went to great lengths to integrate Tutsi elements into society, and publicly stress the need for national reconciliation. In a document titled “Protocole de la Reconcilation Nationale entre les Rwandais”, written in 1976 at the request of Habyalimana by a well-known Hutu politician, Joseph Gitera, a recurrent theme was the need to bring Tutsi, Hutu and Twa in a common unifying ideological framework, a goal in keeping with the stated objective of the Manifesto of the ruling party, the Mouvement Révolutionnaire National pour le Développement (MRND). 
By the early 1990s the government was faced with unprecedented challenges both from within and without. The plummeting of coffee prices sent the economy into a tailspin; the exigencies of structural adjustment placed further strains on the state; famine conditions were reported in several southern prefectures, in turn intensifying the long-standing tensions between north and south. It was at this critical juncture, in a climate of rising tensions, that some 6,000 RPF troops, with considerable logistical and military support from Uganda, marched into northern Rwanda on October 1st 1990. So far from being viewed as a force dedicated to the overthrow of a dictatorship, as they had hoped, the invaders were immediately perceived as Ugandan-supported counter-revolutionaries in league with a Tutsi fifth column inside Rwanda. Predictably, in a matter of weeks anti-Tutsi racism emerged full-blown. The ethnic cleansing of hundreds of Tutsi civilians in 1991 and 1992 were the premonitory signs of the 1994 apocalypse. The huge bloodletting -- precipitated by the shooting down of the presidential plane carrying President Habyalimana and his Burundi counterpart, Cyprien Ntaramyira, on April 6, 1994 -- did not come to an end until a hundred days later, when the RPF finally seized power in Kigali. It is with reason that a retrospective homage is sometimes paid to the RPF for stopping the killings, but this does not detract from the fact that it bears much of the onus of responsibility for the carnage, for without the RPF invasion there would have been no genocide.
The contrast with the wartime conditions ushered by the Nazi invasion of Poland and the Soviet Union could not be clearer. Barbarossa was a war of conquest launched by the perpetrator state in the name of a racist ideology.  The invasion of Rwanda, on the other hand, immediately metamorphosed into a civil war in which the victims were for the most part ethnically identified with the invaders, not the perpetrators. The conflict was between Tutsi invaders anxious to overthrow a dictatorship (and return to their homeland) against Hutu defenders who saw in the invasion the fearsome prospect of a return to servitude.
Tempting though it may be, in the light of these considerations, to argue that in Rwanda genocide was brought onto themselves by the victims, this is far too simplistic as an explanation, in part because the Tutsi invaders were a very different group of people in terms of their history, social backgrounds and generational ties from the victim group, i.e. the resident Tutsi community of Rwanda, and because it ignores the crucial role played by Hutu extremists in the scuttling of the Arusha accords (of which more later). Thus in response to the argument that there would have been no genocide if the RPF had not invaded the country, one might argue with equal plausibility that there would have been no genocide had Hutu extremists not chosen the path of violence. Once this is said, the case of Rwanda comes as close as any other, and probably closer, to giving qualified validation to the provocation thesis  . For a more critical assessment of the provocation thesis, something must be said of the regional parameters of the Rwanda tragedy.
Intentionalist vs. Functionalist Explanations
How much weight should one place on the intention to kill, as against the chain of circumstances leading to the killings? This, in a nutshell, is the central question at the heart of the debates among German historians grappling with the roots of the Holocaust.
For the intentionalists the role of Hitler in orchestrating mass murder is the irreducible, overriding element behind the annihilation of six million Jews; for the functionalists, “circumstance” is the key. While some emphasize the significance of “cumulative radicalization” that stemmed from the incoherence of the Nazi state, or what Christopher Browning describes as “the chaotic decision-making process of a polycratic regime”, others point to the disastrous consequences of Barbarossa.  As Arno Mayer puts it, “the radicalization of the war against the Jews correlated with the radicalization of the war against the Soviet Union”.  An extreme and highly debatable interpretation is set forth by the most controversial of German historians, Ernst Nolte, in his book on Germany and the Cold War. For Nolte the rise of anti-Semitism in its most murderous form is inseparable from criminal record of Bolshevism during and after the Soviet revolution; anti-Semitism and fascism, according to this reasoning, were simply the means through which the German people were effectively mobilized against the external threat of communism.  In Edouard Husson’s words, for Nolte “Hitler was, almost exclusively, an anti-Lenin”. 
Such contrasting interpretations are directly relevant to Rwanda as they bring into focus two critical dimensions of analysis, one focusing on the role of the genocidal state, its ideologues, militias and racist propaganda, the other drawing attention to the domestic and regional contexts in which the killings occurred. While I am in general agreement with Christopher Browning and others who point out that the intentionalist and functionalist arguments are in some ways complementary rather than mutually exclusive, I would give considerably more merit to the former in explaining the Holocaust.
The intentionalist dimension of the Holocaust is nowhere more pithily captured than in the following quote from Ian Kershaw’s magisterial history of the Third Reich:
Replacing the name of Hitler by that of Habyalimana and the Nazi regime by the Hutu regime can only convey a totally misleading view of the roots of the Rwanda bloodbath.
To emphasize Hitler’s pivotal role in the extermination of Jews in no way diminishes the centrality of ideology; they are two sides of the same anti-semitic coin. Yehuda Bauer’s contention that “Nazi racial anti-semitic ideology was the central factor in the development toward the Holocaust” only adds strength to the intentionalist thesis; significantly, however, he goes on to observe that “one major difference between the Holocaust and other forms of genocide is that pragmatic considerations were central with all other genocides, abstract ideological motivations less so”.  This is where the Rwanda genocide deviates substantially from the Holocaust: if by “pragmatic considerations” is meant a conscious attempt to counter the clear and present danger of a Tutsi takeover, these loomed considerably larger than ideological ones. It is not without reason, therefore, that Bauer views Rwanda as “a pragmatically motivated genocide”. 
It is easy to see why some would find incongruous if not downright offensive the use of the term of “pragmatic” to describe the monstrous crimes committed against innocent Tutsi civilians. All the more so when one considers the murderous role played by the Hutu-controlled media in fanning the flames of genocidal murder. Several observers have chronicled the outpouring of scurrilous accusations leveled against the Tutsi community through the airwaves of Radio Mille Collines and in the pages of Kangura.  Racist propaganda is not enough, however, to explain the spiral to murder. To leave out of the picture Hutu perceptions of the multiple threats to their security posed by Tutsi elements within and outside Rwanda is to miss the key factor that made Hutu extremists so receptive to a rabidly racist propaganda.
Fear and hatred -- both contributing, in Helen Fein’s words, to place prospective victims “outside the universe of obligation”  -- were crucial motives behind the Rwanda bloodbath. Behind the paranoid fears raised by the RPF invasion lay the widespread suspicion that Tutsi everywhere in the region were in league with the invaders. This is where the regional context played a major role in magnifying the threat posed by the RPF. Nowhere was the image of the Tutsi as the embodiment of a mortal danger more hauntingly evident than in Burundi, where Tutsi hegemony was achieved at considerable cost in human lives. From the assassination of Burundi’s first Hutu prime minister (Pierre Ngendadumwe), in 1965, by a Tutsi refugee from Rwanda to the assassination of its first Hutu president (Melchior Ndadaye) in 1993, the history of the country is written in blood, mostly Hutu blood. 
The 1972 carnage marks a turning point in the escalation of ethnic violence in Burundi. In response to a local Hutu-sponsored insurgency as many as 200,000 educated Hutu males, including university students and secondary school children, may have died at the hands of the all-Tutsi army. Surprisingly little attention has been to this watershed event in the long chain of circumstances leading to the Rwanda tragedy. Whether referred to as partial genocide or selective genocide, the 1972 killings have been a pervasive presence in the historical memory of Hutu through the entire region.  Nothing has had a more profound effect in crystallizing anti-Tutsi sentiment than the presence of those tens of thousands of Hutu refugees who sought asylum in Rwanda. The same is true of the some 350,000 Hutu who fled Burundi into Rwanda after the assassination of Melchior Ndadaye in 1993. It is no coincidence if these were among the most dedicated murderers of Tutsi during the 1994 bloodbath.
Ethnic memories transcend boundaries. Just as the flow of Tutsi refugees into Burundi during the Rwanda revolution contributed in no small way to polarize Burundi society, the genocide of Hutu in Burundi served as a powerful magnifier of the danger posed by the RPF. What needs to be underscored is the consistency with which history has influenced the receptivity of the Hutu masses to anti-Tutsi propaganda: from the assassination of Ngendadumwe in 1965 to the assassination of Ndadaye in 1993, from the systematic elimination of scores of Hutu politicians in 1965 and 1969 to the 1972 genocide, from the confiscation of the electoral victory of Hutu candidates in 1965 to the 1993 coup, the image of the Tutsi projected by the recent history of Burundi has been consistently threatening, whether as murderers of elected Hutu officials or genocidaires. The incredible fantasies conveyed by the Hutu-controlled media on the eve of the genocide are inseparable from their collective memory of past catastrophes.
To argue with Mark Levene that “there was at least a kernel of truth in the Hutu fear of the Tutsi” is an understatement.  Even more surprising is Yves Ternon’s contention that “there has never been in Rwanda the threat of a seizure of power by the Tutsi, who only represent in 1994 ten percent of the population and have been persecuted for the last thirty years”.  If anything, the deadly cross-border raids of the early 1960s, not to mention the invasion of October 1st, 1990, suggest precisely the opposite. The fact is that by 1993 fear was omnipresent in many sectors of Rwanda society. Not just fear of the Tutsi, but, especially among Hutu affiliated to the ruling MRND, fear of the Hutu opposition. The confrontation between the RPF and the Rwandan army in the north has often overshadowed in public attention the simultaneous outbreaks of intra-Hutu violence in the south.
The introduction of multiparty elections in 1991 was the signal for a violent competition among rival Hutu parties, and as the Arusha talks conferred upon RPF the status of a legitimate opposition party another source of anxiety arose among MRND extremists: the ominous possibility that the Hutu opposition would work out a deal with the RPF to take over the state. All of this gives considerable plausibility to Jaques Semelin’s proposition that “fear, exploited by propaganda, has played a fundamental role in the construction of a criminal project, a project subsequently implemented with method and organization”.  The situation of collective psychosis arising from a political environment saturated with tension and chronic violence is a key element in the background of the 1994 catastrophe.
The comparison with the regional context of the Holocaust is instructive. For if there are ample grounds on which disagree with Nolte that Nazism was in essence a response to the “class genocide” of the Bolcheviks, and the Holocaust a preemptive strike against the communist threat from the east, in Rwanda the perceived threats arising from both the domestic and regional environments were infinitely more substantial. One can only agree with Mark Levene that “the Nazi projection of threat (by contrast with the situation in Rwanda) remained entirely in the realms of fantasy”.  By the same token, there are excellent reasons to question Alain Destexhe’s dismissal of “background circumstances” as irrelevant to an understanding of the Rwanda genocide. “In Rwanda”, writes Destexhe, “some commentators were very quick to explain that the killings were due to background ‘circumstances’: the war, the death of the Hutu president, the ‘excesses of crowds gripped by fear and ancient hatred’, the ‘justifiable anger of the people’, the ‘provocations by the Tutsi’ and their ‘historical domination of the country’, etc. A consequence of this kind of reasoning is that ‘collective guilt leaves us with no one to blame Therefore, so the argument continues, genocides and systematic massacres fall in the same category as volcanic eruptions or earthquakes”.  Not only does attention to “circumstance” not rule out the apportionment of blame, it provides important clues to an understanding of the element of rationality underlying the motivation to kill.
The Rationality of Mass Murder
There is evidently more to genocide than a sudden outburst of murderous irrationality rooted in fear and prejudice. That human beings are capable of committing the most heinous crimes to promote specific political objectives, or for ideological reasons, or to save their own lives, or because they feel they can act with impunity are some of the most disturbing facts brought out by students of genocide.  But what kind of rationality can conceivably explain the systematic extermination of six million Jews? Where is the logic behind the killing of hapless Tutsi civilians, men, women and children, and the cold-blooded murder of one’s nearest kin and neighbors?
The works of Hannah Arendt, Raul Hilberg and Christopher Browning point to the vulnerability of “ordinary men” to the conditioning impact of racist propaganda, the polarizing effects of “race war”, the crushing weight of an all-embracing bureaucratic machinery.  Is there any reason to believe that the same forces that conspired to the extermination of Jews did not also operate in Rwanda? The answer must necessarily remain speculative. Nothing comparable to the sustained, empirically-grounded research conducted by Hilberg and Brown is as yet available for Rwanda. Nonetheless, enough is known of the circumstances of the killings in Rwanda to suggest significant variations in the logic at work in each case.
Perhaps the most obvious refers to greater salience of ideological motives in the planning and orchestration of the Holocaust. Without trying to minimize the impact of anti-Tutsi propaganda in the years following the RPF invasion, the impetus to kill all Tutsi cannot be traced to a long-standing ideological commitment. “What I view as singular about the Holocaust”, Helen Fein writes, “is the length of warning time, its transnational scope and Hitler’s announcement of his intention to eliminate the Jews two decades before their extermination”.  None of these singularities applies to the case of Rwanda.
What, then, is the rationality that impelled tens of thousands of Hutu extremists to engage in mass murder? Taking a leaf from Yehuda Bauer, one might invoke “pragmatic” reasons, but the phrase is too vague to capture the different sets of actors involved at different junctures of the crisis, and the diversity of their motives.
From the October 1990 invasion to April 1994 the overriding objective was to prevent the RPF from seizing power, but throughout this period recourse of violence served different intermediate goals.
The Instrumental Uses of Violence
It is useful to distinguish three phases in the sequence of violence leading to the ultimate carnage, each corresponding to separate but convergent motivations. During the first phase, stretching from October 1990 to the opening of Arusha talks in August 1992, hundreds if not thousands of Tutsi civilians in the north and west of the country were massacred by youth groups (the infamous “reseau zero”) acting under the supervision of communal authorities. The aim was essentially to physically eliminate those Tutsi elements who might join hands with the aggressors while at the same time accelerate the polarization of as yet unmobilized peasant communities. Contrary to an all-too-frequent opinion, the murder of Tutsi civilians was not a spontaneous mass phenomenon. It required the dismantling of inter-ethnic social mechanisms for controlling the use of force, along with the manipulation of ethnic identities against a common enemy. The pattern revealed by the Kilibira, Bugogwe and Bugesera massacres in 1991 is one in which trained activists were spurred on by state officials to engage in random acts of anti-Tutsi violence.
Next came the crucial period of the Arusha negotiations, from August 1992 to August 1993. The worst killings during that time occurred in the Gisenyi and Ruhengeri prefectures in February 1993, taking the lives of an estimated 300 Tutsi. The RPF responded by launching a massive and largely successful attack against government positions in the north, thus causing a temporary suspension of the peace talks. By then multiparty democracy had been formally endorsed by the Habyalimana government, and the civil war had given way to a long drawn-out negotiation among five different parties. The key players, the RPF and the MRND, were now joined by three opposition groups, the multiethnic Liberal Party (PL), the Democratic Republican Movement (MDR), with its roots in the south-central region, and the Social Democratic Party (PSD). Significantly, the most rabidly anti-Tutsi party, the Coalition for the Defense of the Republic (CDR), had been left out of the talks at the request of the FPR, and was thus disqualified from participating in the so-called broadly-based transitional government (BBTG) agreed upon in Arusha. The exclusion of the CDR, together with the rejection by its leadership and other extremist fringe groups of certain key provisions of the Arusha accords, were the decisive factors leading to the killing of Tutsi civilians in 1993. The aim, in short, was the scuttling of Arusha. The point has been conclusively demonstrated by Bruce Jones: referring to CDR extremists and hard-liners within the MRND he notes that “these extremists reacted bitterly to the provisions of the Arusha accords which called for power-sharing and, especially, the integration of the armed forces The genocide was, in the first instance, an attempt by these extremists to maintain their power by destroying the Arusha accords and its supporters, including the moderates within the government parties who were among its first victims”. 
Before the onset of mass murder extremists had another practical purpose in mind: the use of force as a means of seizing power from their immediate Hutu competitors within and outside their respective political formations. “Ukubohoza” – “liberation”: the Kinyarwanda term to designate the campaign of violence directed by members of the opposition against the ruling MRND speaks volumes for the kind of rationality underlying the use of force. “Intimidation” is a better word to describe the criminal activities orchestrated by the MDR in the south to wrest control first from the local MRND cadres, then against the PSD and ultimately against the Tutsi. The process is excellently analyzed by Alison Des Forges in her discussion of how the burgomaster of Nyakizu, Ladislas Ntaganza, managed to use every resource at his disposal, including force, to break the power of the MRND, then to turn against his former ally, the PSD, and ultimately to manipulate ethnic solidarity to enlist the support of Burundi refugees against the Tutsi: “Asked how to define the basis of Ntaganzwa’s power, people said repeatedly and simply: fear”. 
Much the same strategy was used by radical politicians to rid themselves of their moderate rivals within their own parties. As the date for the installation of the BBTG came into view intra-party competition for access to the transitional government became increasingly fierce, putting anti-Tutsi demagoguery at a premium. Nothing seemed to pay higher dividends than to accuse one’s rival of being soft on the FPR. As anti-Tutsi rhetoric picked up momentum, so did violence. Before long every opposition party (except the PSD) was split right down the middle between radicals and moderates. The result, in the words of Vincent Ntezimana, a leading MDR personality, was “l’éclatement de l’opposition pacifique”. 
The phenomenon is directly linked to the emergence of “Hutu Power” (locally referred to as Hutu Pawa), a label that came to designate extremists across a wide spectrum – not just CDR or MRND fanatics but anti-Tutsi zealots within the PL and the MDR. The precipitating factor leading to the birth of Hutu Power was the assassination of President Ndadaye in Burundi. As Alison Des Forges puts it, “The movement known as Hutu Power, the coalition that would make the genocide possible, was built on Ndadaye’s corpse”.  News of Ndadaye’s death had an immediate and devastating impact on Hutu attitudes. Almost overnight many moderates turned into radicals, and radicals into Hutu power extremists, and those who did not were as likely as not to be faced with accusations of disloyalty (ibisiyo). Ndadaye’s death made a chimera of the implementation of the Arusha accords, and with the flood of Hutu refugees from Burundi fleeing into southern Rwanda anti-Tutsi sentiment gathered fresh momentum, driving home what many already suspected: “You simply cannot trust the Tutsi”.
The Security Dilemma
The crash of the presidential plane, on April 6, 1994, took the lives of two Hutu presidents, Habyalimana of Rwanda and Cyprien Ntaramyira of Burundi, bringing to three the total of Hutu presidents killed in six months. Although responsibility for the shooting down of Habyalimana’s plane remains a mystery, few Hutu doubted that the RPF was directly implicated. In an atmosphere already saturated with fear and uncertainty it is easy to see why Hutu extremists quickly yielded to their long-standing plan of using force preemptively. By shooting down Habyalimana’s plane the FPR made dramatically clear that it was about to strike first; rather than wait for Kagame to preempt, for Hutu extremists in the government and the army the exigencies of security -- indeed survival -- called for an immediate response. The decision to apply the full force of genocidal violence against all Tutsi as well as every Hutu suspected of Tutsi sympathies stemmed from straightforward rational choice proposition: either we kill them first, or else we’ll be killed. Thus framed, the logic of the “security dilemma” left no alternative but to annihilate the enemies of the nation.
The argument that crisis situations generate irrational fears that are rationally exploited by perpetrators of mass violence is nowhere more dramatically illustrated than by the renewed outpouring of racist propaganda diffused through Kigali’s hate-radio in the days following the crash. In drawing attention to the critical nexus between irrational fears and the rational manipulation of such fears, Rothchild and Groth give us a clue to an understanding of how the shooting down of Habyalimana’s plane played into extremist hands: “Ethnic psychosis may create rational opportunities this kind of atmosphere, quintessentially irrational, is paradoxically compatible with a perfectly rational exploitation of mass psychosis by communal brokers and entrepreneurs”.  The entrepreneurs of death saw their opportunity and proceeded to exploit it to the full: the global targeting of all Tutsi as the common enemy went hand in hand with the setting in motion of the institutional machinery of murder, of which the key elements were the prefectoral and communal cadres and militias.
The Grass-Roots Killers
Tempting though it is to portray all “grass-roots killers”  as zombies mechanically responding to orders from above, the realities on the ground tell a more complex story. Admittedly, much more research is needed before we get a coherent picture of the full range of motives that led the lower-echelon genocidaires to kill friends and neighbors as well as relatives; there is as yet too little of the kind of fine-grained, local-level investigation conducted by Alison Des Forges in the prefectures of Gikongoro and Butare. Relatively little is known of the dynamics at work in regional arenas. 
What is reasonably clear, however, is that we are here dealing with very different kinds of phenomena from those so carefully analyzed by Christopher Browning in his landmark work on Ordinary Men. For if the men of Reserve Police Battalion 101 killed for reasons that had little to do with the threat of sanctions, and more to do with peer pressure, careerism and self-serving ambition, the same is not true of the grass-roots killers in Rwanda. Many Hutu were driven to kill their Tutsi neighbors because they knew they had no other option; refusal to comply meant that they themselves would be killed the next day.
Again, while there is reason to agree with Browning that the psychological constraints of binding orders were little more than a convenient alibi for the men of Battalion 101, notwithstanding Stanley Milgram’s assertions to the contrary  , in the case of Rwanda the culture of obedience cannot be dismissed out of hand. Several commentators have correctly emphasized the extent to which conformity is a dominant trait of Rwandan political culture. People, whether Hutu or Tutsi, rarely challenge authority. Compliance with orders from above is part and parcel of what Filip Reyntjens describes as “social conformism”: “many Rwandans tend to do what their neighbors do or what a person of authority tells them to do”.  Not every Hutu fit the pattern of Milgramite robots, however; indeed, many took great risks to save the lives of their Tutsi neighbors. As reported by Des Forges, “some Hutu tried to protect their Tutsi neighbors, particularly those to whom they were bound by the ties of marriage, clientage, or long-standing friendships. Other Hutu opposed the killings on the grounds of principle.”  Sometimes, the killing of one Tutsi, unknown to the killer, did not prevent the same individual from going to great lengths to save the life of a Tutsi friend. (Le crime a ses raisons que la raison ne connait pas.) Clearly, the “culture of conformity” argument cannot be accepted without strong qualifications, but the fact remains that it has a far greater explanatory force in Rwanda than in Nazi Germany.
Given the extreme poverty of the country, among the poorest of the poor, it goes without saying that economic motivations loomed infinitely larger in Rwanda than in Nazi Germany. That some Hutu killed for no other reason than to pillage and loot Tutsi property is well established. “The killers pillaged the goods of their victims”, writes Alison Des Forges, “whether Tutsi in flight or local residents. One witness recounts seeing ‘people returning from Nkakwa with bags of beans, clothing, mats One man came by with cushions for a couch. He had six of them. He wanted to sell them to buy beer People were returning with things which they had found free. There was no punishment. It was like a festival.’”  Nor were wealthy Hutu necessarily safe. In those northern communes where land hunger was especially acute scores of land-owning Hutu were killed by their landless kinsmen. In many instances, the dynamics of grassroots murder were closely related to intra-ethnic class differences, with land ownership, rather than ethnicity, being the key determinant of the victim’s identity. Nonetheless, as reported by Maurice Niwese, cases occurred where putting a Tutsi label on a wealthy Hutu served to legitimize the theft of his/her property. 
Finally, and most importantly, many Hutu became killers because of an enduring sense of ethnic hatred born of the sufferings and hardships they experienced at the hands of the RPF. I refer to the hundreds of thousands of internally displaced persons (IDPs) from the north, who were driven from their homes by the advancing RPF army. After the RPF offensive of February 1993 an estimated one million were forced out of their homelands and regrouped into some 40 IDP camps. As James Gasana noted, the IDPs were “explicitly targeted by the FPR rebellion, expelled from their homes and continuously shot at in the camps to force them to move farther into the government-controlled zone”; the result was entirely predictable: “families were separated and scattered heath centers were overwhelmed, mortality increased; suspension of schooling and lack of occupation for the young led to increased delinquency and crime”. Speaking of the “social, environmental and political problems caused by the huge displacement of civilian populations between 1990 and 1994”, the same observer draws attention to “the singular contribution” of the IDP phenomenon to the “combustion of ethnicity”.  Even more decisive was the contribution of the young IDPs to the ranks of the MRND militia, the interahamwe. It is hardly a matter of coincidence if, among the scores of young thugs manning the checkpoints of the capital, the vast majority were recruited among the IDPs of Nyacinga camp, near Kigali.  Seething hatred of every Tutsi in sight, rather than greed or binding orders, is what lay behind the scenes of mayhem in Kigali, Butare and Gikongoro.
What all this adds to is a picture of considerable complexity. The killings cannot be reduced to any single motive. The circumstances that caused Hutu to become killers differed from prefecture to prefecture, sometimes from commune to commune, and while anti-Tutsi propaganda played an important role in driving the genocidaires to murder, its impact varied widely from one sector of Hutu society to another. What remained constant throughout the killings were the sustained efforts made by prefects and burgomasters to mobilize the masses behind the killing machine, and even here there were some notable exceptions. 
Impunity: A Common Denominator
Reflecting on the lessons of Bosnia and Rwanda, Helen Fein brings to light yet another rational dimension behind the horrors of mass murder. “Genocide is preventable”, she writes, “because it is usually a rational act: that is, the perpetrators calculate the likelihood of success, given their values and objectives”.  Certainly, no one familiar with the extent of French complicity during the Rwanda bloodbath, or indeed with the extraordinary indifference of the international community in both Rwanda and Bosnia, can avoid the conclusion that the organizers of the killings entertained few doubts that they could literally get away with murder. The French patron was seen as offering a diplomatic guarantee of impunity as well as the military and financial means with which to prosecute the carnage.
France’s patron role in Rwanda finds a parallel of sorts in the supportive part played by Germany during the Armenian genocide. In his discussion of “German complicity” in the Armenian genocide of 1915-16, Manus Midlarsky writes that “the Berlin-Baghdad railway, the symbol and reality of Germany’s extension of influence into the
Middle East, of course depended on continued Ottoman cooperation, made easier, in the German view, by complicity in the genocide of the Armenians”.  In the same vein, Norman Naimark points out that “The German themselves had played a central role in the Young Turk administration, and a number of Wehrmacht generals had earlier served as advisors to the Ottoman forces during the war. Some German officers may even have played a role in the Armenian genocide itself”. 
But even in the absence of an external patron to assist the genocidaires, the sheer passivity or indifference of the international community can be interpreted as a tacit approval of their plans. Rwanda and Bosnia are cases in point, but so is Germany in the thirties; although signs of Hitler’s genocidal designs could be detected as early as the 1920s and 1930s,  France and Great Britain found in their appeasement policies the justification they needed to refrain from intervening.
If impunity is indeed the rational foundation for genocidaires to become recidivists, public indifference goes a long way to explain impunity. One of the lessons of history that has yet to sink in is that unpunished crimes can provide a precedent for later crimes. In 1939, addressing a group of Nazi leaders and Wehrmacht generals, Hitler is reported to have said, “Who, after all, speaks today about the annihilation of the Armenians?”  Could some Hutu in Rwanda have referred to the 1972 bloodbath in Burundi in similar terms? Few would disagree with the proposition that the first step to prevent the recurrence of genocide is to make sure that the perpetrators are brought to justice and meted out a punishment proportional to their crimes. But as Leo Kuper reminded us many years ago, the sanction of immunity afforded by the concept of national sovereignty is not the least of the obstacles in the way of sanctions. Helen Fein puts it even more graphically: “Abusive powers will continue to abuse as long as it works: the movement to change the taken-for-granted assumption that sovereignty implies indifference to out neighbors’ crimes is still to emerge from gestation in images of mass flight, chaos, blood, and death”. 
One final note: to bring an end to impunity is one thing, just how to calibrate the scale and severity of the punishment is another matter. This is where the Holocaust holds a lesson for the rulers of Rwanda. The men most directly responsible for planning and implementing the Holocaust – some thirty people altogether – were identified, tried and sentenced to die. Although some have argued that the punishment decreed at Nuremberg was disgracefully benign in view of the magnitude of the crime, and that those elements who escaped punishment hardly became pillars of democracy, there is another side to the coin, which needs to be looked at. Can one imagine what the effect would have been on the German people and the future of democracy in Germany had every German involved in the Holocaust, at one level or another, or in one capacity or another, been brought to justice and condemned? Had tens of thousands been sent to the gallows, as some had wished, one wonders whether Germany would have become a flourishing democracy, or would have experienced much success in coming to terms with her past. In Rwanda today, very few of the “brains” behind the genocide have been identified, and none of those currently in detention in Arusha have been dealt a death sentence; meanwhile, scores of mid-level killers have been tried in Kigali and inflicted the death sentence, while some 130,000 Hutu suspected of participating in the killings are still languishing in jail, seven years later. The least that can be said is that the prospects for national reconciliation in such circumstances seem very remote. This is yet another difference with the Holocaust, and perhaps not the least consequential.
 . Mark Levene, “Connecting Threads: Rwanda, the Holocaust and the Pattern of Contemporary Genocide”, in Roger W. Smith ed., Genocide: Essays Toward Understanding, Early-Warning and Prevention (Williamsburg, Va.: 1999), 27-64.
 . On the contrasting interpretations of the Holocaust offered by German historians, see Richard J. Evans, In Hitler’s Shadow: West German Historians and the Attempt to Escape From the Nazi Past (Pantheon Books, New York: 1989, Edouard Husson, Comprendre Hitler et la Shoa: Les historiens de la République fédérale d’Allemagne et l’identité allemande depuis 1949 (Presses Universitaires de France, Paris: 2000), and Christopher Browning, The Path to Genocide (Cambridge University Press, Cambridge : 1992), chap. 5.
 . Robert Melson, Revolution and Genocide: On the Origins of the Armenian Genocide and the Holocaust (The University of Chicago Press, Chicago: 1992)
 . Helen Fein, “Genocide: A Sociological Perspective”, Current Sociology, Vol. 38, no. 1 (Spring 1990), Special Issue, p. 28-9. Especially pertinent in the context of this paper is Fein’s discussion of “the misuses of genocide comparison”, ibid. 55-69.
 Quoting from Norman Cohn’s classic work, Warrant for Genocide, Helen Hintjens argues that “just as the Nazi had done in Germany before the war, extremists in Rwanda ‘started with the fiction of a conspiracy and modeled themselves, more or less consciously after it’ (Cohn 1980, 193)”.
Although there was no dearth of anti-Tutsi conspiracy theories in the Rwanda media, the parallel with Nazi Germany strikes me as profoundly misleading in that allegations of Jewish plots were utterly groundless; in Rwanda as in Burundi, however, there were ample grounds upon which to build such theories, beginning with the assassination of the Hutu Prime Minister Pierre Ngendadumwe at the hands of a Tutsi refugee from Rwanda, in 1965. Elsewhere the author refers to the “supposed genocidal schemes of the (Tutsi) enemy” and the “supposed evil intentions and revanchist designs of the Batutsi enemy”. In the light of the genocide of Hutu by Tutsi in Burundi in 1972, resulting in the massacre of perhaps as many as 200,000, and of the armed invasion of Rwanda by Tutsi refugees in 1990, it is easy to see why many politically conscious Hutu in Rwanda, as well as in Burundi, entertained such “suppositions”! The Hutu extremists did not start with the “fiction of conspiracy”, but with the realities of murder and genocide, ending up with conspiracy theories. Helen Hintjens, “Explaining the 1994 genocide in Rwanda”, The Journal of Modern African Studies, vol. 37, no. 2 (1999), 263.
 Bill Berkeley, one of the few observers to call into question the parallels between the Tutsi’s experience of genocide and that of the Jews, notes in the same vein that “the parallels are not exact: the Jews of Europe were never armed. There was no Jewish conspiracy to dominate Europe, nor had there ever been one. There had been no Jewish tyranny in Germany, as there were Tutsi tyrannies in Rwanda and Burundi, and there had been no Jewish-perpetrated genocides in, say, Austria, as there were Tutsi-perpetrated genocides against Hutus no fewer than three times in a generation in Burundi, just an hour’s drive down the road”. Bill Berkeley, The Graves Are Not Yet Full: Race, Tribe and Power in the Heart of Africa, (Basic Books, New York: 2001), 264-5
 . William F.S. Miles, “Hamites and Hebrews: problems of ‘Judaizing’ the Rwandan genocide”, Journal of Genocide Research, vol. 2, no. 1 (2000), 112.
 . Ibid., 112.
 . The definitive work on Aryan ideology is Leon Poliakov, Le Mythr Aryen (Calmann-Levy, Paris: 1994); on the Hamitic hypethesis, see Edith Saunders, “The Hamitic Hypothesis: Its Origins and Functions in Time Perspective”, Journal of African History, Vol. 10 (1969), 521-532.
 . I have noted elsewhere the more intriguing parallels between the Aryan myth and its Rwandan counterpart: (a) their genealogical roots are traceable to societies located outside the genocidal states, one to ancient Ethiopia, the other to Sanskrit civilization; (b) both were elaborated by 19th century European intellectuals anxious to lend pseudo-scientific respectability to their speculations about the presumed superiority of certain civilizations, including their own; (c) in post-colonial Rwanda as in Nazi Germany these pseudo-scientific were recast into virulently racist myths at the hands of Nazi and Hutu ideologues. See R. Lemarchand, “Where Hamites and Aryans Cross Paths: Myth-Making and Mass Murder” (unpublished ms.)
 . Miles, op. cit., 112.
 . Melson, 18-19.
 . For a fuller discussion of the Hutu revolution, see R. Lemarchand, Rwanda and Burundi (Pall Mall, London: 1970)
 . Peter Uvin, “Tragedy in Rwanda”, Environment (April 1996), p. 13.
 . See Jan Vansina, Le Rwanda ancien (Karthala, Paris: 2001), esp. the last chapter, “L’histoire face au présent”, and R. Lemarchand, Ethnicity as Myth: The View from Central Africa (University of Copenhagen, Center for African Studies, Occasional Paper, May 1999)
 . Daniel J. Goldhagen, Hitler’s Willing Executioners: Ordinary Germans and the Holocaust (Vintage Books, New York: 1997). For critical commentaries, see Ruth Bettina Birn, “Revisiting the Holocaust”, The Historical Journal. Vol. 40, no. 1 (1997), 195-215, and Edouard Husson, Une culpabilité ordinaire? Les enjeux de la controverse Goldhagen (PUF, Paris: 1997). As might have been expected, the Goldhagen thesis has found a highly receptive audience among some Tutsi intellectuals : see, for example, Benjamin Sehene, Le Piège Ethnique (Editions Dagorno, Paris : 1999), 119-121.
 . Significantly, the first signs of an explicitly anti-Tutsi, racist ideology came into view not in Rwanda but among the Tanzania-based Hutu refugees from Burundi, in the aftermath of the 1972 genocide. See R. Lemarchand, Burundi: Ethnic Conflict and Genocide (Cambridge University and Woodrow Wilson Center Press, New York: 1995), 201; see also Liisa Malkki, Purity and Exile: Transformations in Historical-National Consciousness Among Hutu Refugees in Tanzania (University of Chicago Press, Chicago: 1995).
 . The Kinyarwanda term for cockroach, inyenzi, is a deliberate deformation of ingenzi, meaning “brave”. The original label used by Unaristes for designating the Tutsi guerillas was “ingangurarugo ziyemeje kuba ingenzi”, the name of one of king (mwami) Rwabugiri’s armies, meaning roughly “the brave ones in the service of the king’s army”. I am indebted to Emmanuel Hakizimana for this information.
 . For a more detailed discussion of the impact of the abortive inyenzi raids on post-revolutionary Rwanda, see R. Lemarchand, “Refelections on the Rwanda Genocide: A Backward Glance into the Past”, paper presented at the annual meeting of the African Studies Association (ASA), Nashville, November 2000.
 . The full text of the document is reproduced in the newspaper Intego (Kigali), no. 12, June 11, 1996.
 . Arno Mayer describes Barbarossa as “an incarnation of the major tenets of Hitler’s action-ideology. Rooted in racist social Darwinism, the war in the east had the fourfold purpose of conquering lebensraum from Russia, of enslaving the Slavic populations, of crushing the Soviet regime, and of liquidating the alleged nerve-center of international Bolchevism.” Arno Mayer, Why Did the Heavens Not Darken? The Final Solution in History (Pantheon Books, New York: 1989) 459.
 . For an excellent discussion of the glaring misconstructions of the provocation thesis applied to the Holocaust and Armenia, see Melson, op. cit. 10-12.
 . Christopher Browning, The Path to Genocide, op. cit., 86.
 . Arno J. Mayer, op. cit., loc. cit. See also Christopher Browning’s critical commentary of Mayer’s interpretation, Browning. op. cit., 77-85
 . Ernst Nolte, Deutschland und der Kalte Krieg (Stuttgart: 1985)
 . Husson, op. cit., 136. See also Richard Evans’ excellent critique of the Nolte thesis, in In Hitler’s Shadow, op. cit., 24-46.
 . Ian Kershaw, Hitler: 1936-1945, Nemesis (Penguin Books, London: 2000), 841.
 . Yehuda Bauer, Rethinking Genocide (Yale University Press, New Haven and London: 2001), 42, 47.
 . Ibid. 46.
 . See for example, Alison Des Forges, Leave None To Tell the Story (Human Rights Watch and International Federation of Human Rights, New York and Paris: 1999), 65-91, and Jean-Pierre Chrétien, Rwanda. Les media du genocide (Karthala, Paris: 1995).
 . Helen Fein, “Scenarios of Genocide: Models of Genocide and Critical Responses”, in Israel Charny ed., Towards the Understanding and Prevention of Genocide (Westview Press, Boulder: 1984), cited in Melson op. cit., 15.
 . See R. Lemarchand, Burundi: Ethnic Conflict and Genocide (Cambridge University Press and Wilson Center Press, Cambridge: 1995)
 . See R. Lemarchand and David Martin, Selective Genocide in Burundi (Minority Rights Group, London: 1974); for a more detailed (and revised) analysis of the 1972 bloodbath, see R. Lemarchand, Burundi: Ethnic Conflict and Genocide, op. cit., 76-105. It is noteworthy that most analysts of the Rwanda genocide more or less systematically shun the use of the term “genocide” to describe the 1972 bloodbath, as if recognition of the anti-Hutu genocide in Burundi might diminish the horror of the anti-Tutsi genocide in Burundi. For example, in her definitive work on Rwanda, Alison Des Forges refers to the “slaughter of tens of thousands of Hutu by Tutsi in 1972” (p.65) and elsewhere to the “slaughter” of “some 100,000 Hutu” (p. 134). The G word never appears in the Burundi context. Alison Des Forges, op. cit.
 . Levene, op. cit., 37.
 . Yves Ternon, L’innocence des victimes au siécle des génocides (Desclée de Brouwer, Paris : 2001), 47.
 . Jaques Semelin, “Qu’est-ce qu’un crime de masse? Le cas de l’ex-Yougoslavie , Critique Internationale, no. 6 (Winter 2000), 146.
 . Levene, op. cit. 37.
 . Alain Destexhe, Rwanda and Genocide in the Twentieth Century (New York University Press, New York: 1994) 13.
 See Jaques Semelin, “Penser les massacres”, typescript (Paris, 2001); see also “Qu’est-ce qu’un crime de masse?”, op. cit.
 . Hannah Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (Viking Press, New York: 1963), Raul Hilberg, The Destruction of European Jews (Quadrangle, New York: 1967), Christopher Browning, Ordinary Men: Reserve Police Battalion 101 and the Final Solution in Poland (Harper Collins, New York: 1992.
 . Fein, Genocide : A Sociolgical Perspective, op. cit. 55.
 . Bruce Jones, “’Intervention without Borders’: Humanitarian Intervention in Rwanda, 1990-94”, Millenium: Journal of International Studies, Vol. 24, No. 2 (1995), 243. See also, Bruce Jones, “Civil War, the Peace Process and Genocide in Rwanda”, in Taisier M. Ali and Robert O. Matthews eds., Civil Wars in Africa (McGill-Queens University Press, Montreal: 1999) , 53-86.
 . Des Forges, op. cit, 361.
 . Vincent Ntezimana, La Justice Belge Face au Génocide Rwandais (Karthala, Paris: 2000), 42. The author’s phrasing ignores the fact that the MDR’s activities in the Butare prefecture were not exactly “peaceful”, and that he himself has been accused by a Belgian tribunal of having taken part in the genocide.
 . Des Forges, op. cit. 137.
 . Donald Rothchild and Alexander J. Groth, “Pathological Dimensions of Domestic and International Ethnicity”, Political Science Quarterly, Vol. 110, No. 1 (1995), 75.
 . The term is borrowed from Christopher Browning to designate low-level genocidaires as distinct from the regular army, the Presidential Guard and the prefectoral and communal cadres. Christopher Browning, “Ordinary Germans or Ordinary Men”, Address and Response at the Inauguration of the Dorot Chair of Modern Jewish and Holocaust Studies (Emory University, Atlanta: 1994), 9.
 . The research currently undertaken by Scott Straus, a PhD candidate in political science at the University of California at Berkeley, promises to fill some major gaps in the literature on the dynamics of lower-echelon genocidaires.
 For a persuasive critique of the Milgram argument that obedience to orders from above is the key to an understanding of the behavior of Holocaust perpetrators, see David R. Mandel, “ The Obedience Alibi: Milgram’s Account of the Holocaust Reconsidered”, Analyse & Kritik (October 1998), 74-94.
 . Filip Reyntjens, “Rwanda: Genocide and Beyond”, Journal of Refugee Studies, Vol. 9, No. 3 (1996), 6.
 . Alison Des Forges, op. cit., 378.
 . Ibid., 379.
 “Someone could be called Tutsi simply to take away his cow; it would have been illogical to steal his cattle without calling him a Tutsi, and once thus named his fate was sealed”. Maurice Niwese, Le peuple rwandais un pied dans la tombe. Récit d’un réfugié étudiant (L’Harmattan, Paris: 2001), 55.
 . James Gasana, “Natural resource scarcity and violence in Rwanda”, A Report Prepared for the Swiss Organization for Development and Cooperation (Intercooperation), n.p., 2000, available from the author at email@example.com
 . I am indebted to Emmanuel Hakizimana for this information.
 The widespread participation of prefects and communal councillors in the killings does not mean that these are traceable to the presence of a genocidal “strong state”; the dynamics of murder must be seen against the background of highly personalized networks headed by a handful of key “patrons” (such as Theoneste Bagosora, Joseph Nzirorera, Fernand Nahimana, to cite but the most prominent) with multiple ramifications into local administrative cadres and the civil society. The principal vectors of death, i.e. the interahamwe, the parties’ youth wings, the Radio et Television Mille Collines, were all part and parcel of the civil society. As the killings got under way, the state had virtually ceased to exist, and next to the president the most visible embodiment of the state, Prime Minister Uwilingiyimana, along with several other cabinet members, had already been killed.
 . Helen Fein, “Patrons, Prevention and Punishment of Genocide: Observations on Bosnia and Rwanda”, in Helen Fein ed. The Prevention of Genocide: Rwanda and Yugoslavia Reconsidered, A Working Paper of the Institute for the Study of Genocide (New York: 1994), 5.
 . Manus I. Midlarsky, “The Killing Trap: Genocides and Other Mass Murders of the Twentieth Century”, Unpublished ms., 2000, 6.
 . Norman M. Naimark, Fires of Hatred: Ethnic Cleansing in Twentieth Century Europe (Harvard University Press, Cambridge, Mass: 2001) 57.
 . “’Out with the Jews’, insisted Hitler, June 24, 1920, ‘those who poison our people’. In an August 13, 1920 speech entitled ‘Why we are Anti-Semites’ Hitler again made clear that the Jews would have to go ”, Ibid. 63.
 . Ibid., 57.
 . Fein, “Patrons, Prevention and Punshment of Genocide”, op. cit., 12.
This article refers to: