It is depressingly familiar: as in the case of America’s China policy after 1945, as in Vietnam, ambitious but fuzzy aims are being sought by inadequate means. If the purpose of NATO’s bombing of Serbia has been to deter the Serbs from inflicting more harm on the Kosovar Albanians, the policy has failed and we now have to confront the fact that it is, to use the language of strategists, far more difficult to compel the Serbs to get out of Kosovo so that the refugees can return than to deter them—and most unlikely that we will be able to compel them by air warfare alone.

As so many analysts keep telling us, we have stepped into an extraordinarily complicated and messy tragedy. The peaceful citizens of democratic countries, even when they are stirred by compassion and shocked by atrocities, are reluctant to wage war “for others”—self-defense is another matter. And the Balkans seem like the archetype of a quagmire. The essence of the issue is simple: at the end of the twentieth century, crimes against humanity are being committed on a scale and with a brutality that one hasn’t seen in Europe since Nazi Germany (although, as Mark Danner has reminded us, Bosnia was a prelude and a warning that the NATO countries, and indeed the UN, ignored for four years1 ).

Such crimes are doubly intolerable. They are intolerable on moral grounds, of course: no tyrant should be allowed to treat human beings in such a way, on such a scale. But also on grounds more familiar or acceptable to “realists,” for whom, as for Dean Acheson, there are two kinds of issues: moral ones, and real ones. To tolerate this ferocious combination of ethnic cleansing and quasi genocide is to give other potential ethnic cleansers a green light in a world of fragile states and ethnic conflicts, and to invite world disorder on a grand scale. For there can be no order, no “stable structure of peace,” to use Kissingerian language, without a modicum of moderation, especially in areas that are powder kegs.

This is what matters. The fact that NATO has “violated Yugoslav sovereignty” doesn’t: it was Milosevic who violated, by abolishing it, Kosovo’s autonomy, and who later explicitly accepted that Kosovo was a legitimate subject of “international” concern. What the Serbs are doing is not a police operation against political dissenters or ordinary criminals. It is the destruction of a movement of national liberation from extremely repressive rule, the crushing of a drive for self-determination. The fact that the US and its allies have failed to respond to cases of genocide or ethnic cleansing all over the world—in the Sudan, or in Cambodia, or in Rwanda, or in the Krajina when the Croats took it back from the Serbs in 1995—is not a reason for passivity in Kosovo; it is a reason for remorse or at least self-criticism. It was a mistake for the NATO countries not to seek authorization from the Security Council of the UN. But if, as is likely, the Russians, thinking of Chechnya, and the Chinese, mindful of Taiwan and Tibet, had used their vetoes, there would still have been a duty to act.

NATO’s intervention is undoubtedly a just cause. But in order to judge the morality of a war, one also has to take into account two other considerations: Are the values destroyed in the war less important than the values defended? Is there, as the Christian just-war doctrine also requires, a reasonable chance of success? In order to evaluate the diplomatic and strategic justification of a war, we must consider the consequences of a Serbian victory. These would include the destabilization of weak states like Albania and Macedonia, which would be obliged to cope with the continuing influx of refugees, and the danger to the stability and peace of the region that would result from the predictable guerrilla actions Albanians would mount in Kosovo and from acts of hot pursuit that the Serbs might launch against the Albanians’ external sanctuaries. Would these consequences be worse than the destabilization that a major NATO ground and air war would entail, and the return to a kind of cold war between the US and Russia that it might provoke?

This is where the difficult choices lie. One strategy that has been suggested is little better than a cop-out: arming the Kosovar Albanians just as the US armed the Afghans against the Soviets. The trouble is, first, that the Serb blitz in Kosovo has pushed the battered KLA into remote areas that would be difficult to reach, and, second, that these guerrillas and armed Kosovars, after they were infiltrated from Albania or Macedonia, would have to operate, not amid a friendly and supportive population, but in a ravaged country largely emptied of its inhabitants. Arming the Kosovars can be an ingredient of a strategy, a supplement, but it can’t be the main part of it.


Only two options seem workable. The first is the kind of ground operation William Pfaff has called for in these pages2—a bold combined action, à la Inchon, by soldiers and paratroopers aimed at both liberating Kosovo and allowing the refugees to return and rebuild their burned villages with massive Western aid. So far, foolishly, the US and its allies have reassured, and encouraged, Milosevic by repeating that they have no plans and no intention to proceed along these lines. If it is true that there are no such plans, this strategy would be especially hard to pursue. It would take time both to prepare it and to bring the necessary forces into a base as unsuitable as Albania or as reluctant, even hostile, a base as Macedonia.3 (Invading Serbia from Hungary would be wrong, not only because it is Kosovo that needs to be freed, but because of the repercussions on the West’s relations with Russia.) Will the public in the US and in Western Europe remain supportive of NATO for all these weeks of buildup if the images of fleeing refugees stop filling the TV screens once most of the Kosovar Albanians have been thrown out; if, in Germany or France, anti-Americanism fed by NATO bombings and the inevitable “collateral damage” it produces rises; and if, in the US, in a preelection year, doubts about American involvement multiply?

Once the operation is launched—against an adversary who would have had the time to move its tanks to the borders of Kosovo, to dig in, to use whatever Albanians remain as shields, and also to prepare for guerrilla warfare as in World War II should Serb forces be pushed back—would the US “stay the course” if casualties are high, and analogies (however misleading) to Vietnam begin to feed clamors for an “exit strategy” and for what I once called American combatant immunity? The “chance of success” would depend at least as much on the morale and determination of the citizens of NATO countries, and on the skills of their statesmen, as on the hazards of the battlefield—a battlefield familiar to the Serbs and quite unfamiliar to, and difficult for, NATO forces.

Thus this option, aimed at allowing the Kosovar Albanians to return and at giving them a country of their own (whether it is fully independent or part of a loose Yugoslav Federation is secondary), would require the kind of bold leadership that none of the NATO countries (including, in this instance, Tony Blair) has been able to muster. Maybe the high cost of a NATO failure to reach the goals so solemnly set in recent meetings and statements will push our leaders in the direction of decisiveness and daring; after all, the US moved into Haiti and Bosnia once the alternatives appeared worse than the costs of action. But all the talk about a “diplomatic solution” or compromise (which could not but ratify Milosevic’s execrable triumph on the ground) is not encouraging.

If NATO’s leaders decide that their publics aren’t ready to “die for Kosovo,” that they don’t themselves have the energy or desire to try to rally their people, and that the chances of victory at an acceptable price are low, then only one option makes sense. It will, inevitably, look like a NATO defeat, and it will be one, given the ambitious (and legitimate) goals that have been set. This should be clearly recognized, and the reasons for not following the ground-war option candidly spelled out. It would entail a scaling down of our aims to a political version of the goal so often stated by the military (both in this case and in that of Iraq), the “degrading” of the adversary’s forces. It would mean confessing that although Milosevic’s methods are Hitlerian, Serbia’s power and the threat it represents aren’t comparable to those of Nazi Germany in the 1930s or “Communism” in the 1950s, and that therefore the Alliance isn’t going to wage the kind of war to defeat aggression that the US waged in Korea. It would mean admitting that the absence of oil explains the difference between Kuwait and Kosovo (even though Milosevic’s behavior has been far worse than Saddam Hussein’s in Kuwait).

This would signify a painful moral retreat and shrinking of earlier ambitions for a “new world order.” But it would not be a total defeat, because the goal of NATO action would still be to make it clear to the Serbs and to other potential ethnic cleansers that they cannot achieve their goals without having to pay the very high price that bombings inflict—the disruption of the daily lives, and in some cases the death, of largely acquiescing citizens, the destruction of strategic resources, etc. This would mean both that the bombing campaign would continue (but would no longer be presented as aiming at forcing the Serbs to get out of Kosovo) and that it would be limited in time, because as it goes on the likelihood of collateral damage and of selecting targets of increasingly less obvious military significance multiplies, and with it the risk of a backlash of popular revulsion in NATO countries. As in the case of Iraq, we would then be left with an inglorious but not necessarily unworkable policy of containment, aimed at preventing any more Serbian aggressions and annexations. Unlike in the case of Iraq, we would also be left with hundreds of thousands of refugees nobody wants, except—but for how long?—poor Albania.


Thus it is a choice between a genuine war for Kosovo’s liberation, with the high perils and huge costs I have described, and a limited air war for punishment that obliges Milosevic and his people to pay a high price for a ghastly fait accompli we no longer refuse to accept. It is not an easy choice. In either case, we should cease considering Milosevic as a potential negotiating partner. If we choose the first option, we need to defeat him in Kosovo and to leave in place enough forces (whether they are NATO’s, OSCE’s, or the UN’s doesn’t matter so long as they are armed and have a clear mandate) to protect returning Kosovars and to discourage any new aggression.4 If we choose the second option, treating him as a legitimate interlocutor would add insult to injury, although we should encourage the Russian prime minister and the secretary-general of the UN to try to get Milosevic to agree to withdraw from Kosovo and accept an international force not exclusively composed of NATO contingents. In either case, international criminal justice should at least build a case for his indictment as a war criminal.

Those who, like former Secretary of Defense Caspar Weinberger, have exhorted NATO to “win” owe it to the citizens and soldiers of NATO countries to be clear and explicit about the difficulties, dangers, and costs—to reduce the possibility of a formidable backlash, à la Vietnam, as a result of those costs. But one has to be equally candid about the moral and political implications of the more modest (in ends and means) strategy of option two. If one believes that the heart of the issue is the toleration—or not—of genocide and of murderous forms of ethnic cleansing, and the establishment of a clear norm of international law against such acts committed either across or within borders, the choice for the first option, demanding and dangerous as it is, seems to me the right one. —April 22, 1999

This Issue

May 20, 1999