Alan J. Kuperman plays word games when he asserts that President Clinton could not have known of the "attempted genocide" of Tutsi in Rwanda until April 20, 1994 -- two weeks into the slaughter -- because the press, nongovernmental organizations (NGOs), and the U.N. did not call it a genocide ("Rwanda in Retrospect," January/February 2000).

Two days, not two weeks, after the slaughter began on April 6, U.S. officials knew that extremists with an avowedly genocidal agenda had murdered legitimate Rwandan authorities and were claiming control of the government. U.S. officials knew that these extremists had used the radio to spew anti-Tutsi propaganda for months and that they had recruited, trained, and armed militias. U.S. officials knew that Rwandans had previously used the highly centralized administrative system to organize massacres of Tutsi. And at least some of these officials knew that an eerily prescient CIA study three months before had foreseen a death toll of half a million if violence began again.

As an April 8 State Department briefing made clear, U.S. officials also knew that Hutu soldiers had been killing Tutsi for two days and that the violence was not limited to the capital. They learned this from U.S. embassy personnel and from the French and the Belgians, who had extensive contacts in Rwanda and with whom Washington was planning a joint evacuation of their citizens. U.S. officials had similar reports from the commander of the U.N. peacekeeping force in Rwanda, as well as from Human Rights Watch and other NGOs, which had been calling the State Department since the preceding day.

During the crucial first weeks, the U.N., at the behest of the United States, ordered the more than 2,000 peacekeepers in Rwanda to do nothing to halt the killing and then withdrew all but a rump force of 400 soldiers. Some 1,000 elite French and Belgian troops (backed by 250 U.S. Marines just across the border) swooped in to rescue foreign nationals (most of them not at risk) and then left, ignoring the slaughter of Rwandan civilians. Clinton and other international leaders said nothing of substance. Seeing the international indifference, Rwandans became convinced that the genocidal government would succeed. Those who hesitated at first now yielded to fear or opportunism and carried the slaughter throughout Rwanda.

U.N. peacekeepers and the evacuation force could have deterred the killings had they acted promptly. Belgian military records show cases in which they did just that when permitted to use their weapons. Firm and coherent international censure could have influenced the organizers of the genocide. On the two occasions when they received outraged telephone calls from foreign governments, the organizers halted attacks on hundreds of Tutsi at the Hotel des Mille Collines in Kigali. Jamming the genocidal radio broadcasts would have kept the organizers from passing orders directly to the population. The military radio, the only other channel accessible to the genocide's organizers, did not broadcast to civilians.

Kuperman scoffs at such measures. He takes seriously only the possibility of dispatching U.S. troops, who, he states, could easily have ended the genocide. Yet he goes on to argue that even a major American force could have saved no more than a quarter of the intended victims. He presents figures, computations, a chart, and a graph that lend apparent solidity to this conclusion.

But this solidity vanishes when his underlying premises are examined. Kuperman's assertion that troops could not have been deployed until several days after April 20 rests on the incorrect assumption that Clinton learned of the genocide only on that date. Kuperman also assumes that the genocide swept through "most" of Rwanda "immediately" after its start and killed half the victims by April 21. This assumption exaggerates the early extent and the speed of the slaughter.

In 1994, the Clinton administration confounded genocide and internal war, and now Kuperman does it again. He slides without explanation from discussing the genocide to speculating about intervention in an "internal war," arguing that helping the "weaker" side may spur it to reject compromise and escalate fighting.

Let us be clear: There was a war in Rwanda, but the weaker party was the genocidal government fighting the militarily stronger Rwandan Patriotic Front. Tutsi civilians were not a party to the conflict. They were a people targeted for extermination. Helping them would not have escalated fighting but would have saved their lives.

Americans must face the truth, as all the other major international actors in Rwanda -- the Belgians, the French, the U.N., and the Organization of African Unity -- have already done by investigating their roles in the genocide. Congress should have the courage to follow this lead and investigate what the U.S. government did do and, even more important, what it could have done during a genocide that slaughtered half a million people.

Alison l. Des Forges is a consultant at Human Rights Watch and author of Leave None to Tell the Story: Genocide in Rwanda.


Alison L. Des Forges' response and William F. Schulz's letter to the editor (March/April 2000) call into question my recent article on Rwandan genocide.

My article relies on three main findings. First, the Rwandan genocide of 1994 was perpetrated extremely quickly; about half the ultimate Tutsi victims were dead before the end of the third week. Second, because of the intentional deception by the perpetrators of the genocide and incomplete reporting by Western sources, President Clinton could not have known that a nationwide genocide against Rwanda's Tutsi was occurring until two weeks into the killing. Third, owing to constraints of strategic airlift and military doctrine, several weeks would have been required for a U.S. intervention to stop the genocide. Thus, my article concludes that even if President Clinton had ordered a maximum intervention as soon as he knew genocide was occurring in Rwanda, at most one-fourth of the ultimate Tutsi victims could have been saved, but the majority would not have survived.

Des Forges takes issue with each of these findings. First, she asserts that my article "exaggerates the early extent and the speed of the slaughter." This is puzzling, given her own recent writing on the subject. My article states that "perhaps 250,000 [Tutsi were killed] in just over two weeks." But her 1999 book for Human Rights Watch, Leave None to Tell the Story, states, "By two weeks into the campaign, they had slain hundreds of thousands of Tutsi." If anything, my estimate is more conservative than her own.

Second, Des Forges claims that President Clinton knew almost immediately that a genocide was under way because of past ethnic violence in Rwanda, an earlier CIA analysis, and news reports during the first two days that extremists had taken control. Each of these arguments falls apart under closer analysis. Des Forges' book estimates that during the approximately three years of civil war prior to the genocide, about 2,000 Tutsi civilians had been killed -- a rate of about 50 per month. Although such killing was extremely troubling, it did not suggest that the death rate would suddenly jump to 250,000 in the following month.

Furthermore, the CIA study was a "desk-level" analysis that contained three possible scenarios, only one of which predicted such mass killings. Intelligence analysts routinely include such a worst-case scenario to cover themselves in case events go awry. Moreover, low-level studies of this sort rarely reach the president. U.S. officials did know, by the second day, about the bloody coup, the large-scale violence in the capital, and the renewed civil war. But they also expected the Tutsi rebels to win quickly. There is no reason to believe that President Clinton or any other U.S. observer knew immediately of the nationwide genocide. Indeed, Des Forges herself, in a Washington Post op-ed published on April 17, 1994 (11 days after the outbreak of violence), failed to raise even the prospect of "genocide." Des Forges' hindsight may be 20-20, but the picture was more muddled at the time.

Third, Des Forges claims that deploying intervention forces throughout Rwanda would not have been necessary to stop the genocide. Either "international censure" or quick military action in the capital by the few European evacuation troops present would have been sufficient, she claims. Although anything is possible, this smacks of wishful thinking. The lightly armed Western evacuators were outnumbered five to one by Hutu forces in the capital, so quick victory would have been unlikely. Meanwhile, large-scale massacres ignited throughout most of the country within the first week and continued for three months -- in the face of repeated international condemnation and sanctions -- until Tutsi rebels and late-arriving French troops ventured out to the countryside to stop them. This strongly suggests that once nationwide genocide began, only large-scale military intervention could have assured its rapid end.

Schulz, by contrast, criticizes three purported claims that are nowhere to be found in my article. First, he claims that I am "at pains to deny ... [that] the U.N.'s inaction ... was caused by an utter failure of will." To the contrary, my article discusses how a lack of will led the United States and the United Kingdom to quash efforts to bolster the peacekeeping force prior to the genocide. Moreover, my article underscores that if the West had possessed sufficient will to launch an intervention immediately upon learning of the attempted genocide, it could have saved up to 125,000 Tutsi.

Second, Schulz claims that my "thesis is that we are under a moral imperative to stop genocide only when we can stop it completely." Although my article debunks previous claims that military intervention in Rwanda after the outbreak of violence could have averted the genocide completely, it never suggests that this "exempts us from making the effort." Rather, it offers detailed recommendations for improving the effectiveness of military intervention. In addition, my article suggests devoting more attention to diplomacy, which offers the best hope of actually averting genocide.

Third, Schulz writes that I "cannot seriously contend" that the innocent victims in ethnic conflict "calculatedly invited their fates." What my article actually states is that the leaders of such groups sometimes provoke retaliation against their own civilians in order to galvanize domestic and, especially, international support. This was a repeated tactic of the Bosnian government in its 1992-95 war, as documented by at least two U.N. commanders on the ground. More recently, this cynical tactic was copied with even greater success by the Kosovo Liberation Army. As long as the West comes to the military assistance of groups being victimized because of their own violent provocations, we risk fostering an escalation of ethnic conflict.

Ultimately, Des Forges, Schulz, and I share the same goal -- to reduce genocide and other massive abuses of human rights. Where we differ is in tactics. Their organizations, Human Rights Watch and Amnesty International USA, respectively, call for coercive pressure to force oppressive rulers to hand over their powers overnight. But when such rulers face the prospect of losing power and suffering retribution, they may opt instead for a final solution of murdering or expelling their opponents -- as in Rwanda, the Balkans, and East Timor. Western military intervention has not and cannot arrive quickly enough to save many innocents. Thus, I offer an alternative prescription: Give less support for violent insurgencies, more incentives for gradual reform, and golden parachutes for departing authoritarian leaders in cases in which forgiving past crimes is the price of preventing future ones.

The best way to stop genocide is not military intervention after the fact but wise diplomacy that prevents genocide from starting in the first place.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now