THE BLOODIEST century in history has closed with a new idea: that lives can be saved if foreign troops are willing to shoot human rights violators before they begin, or at least before they complete, their tasks--and that if lives can be saved this way, morality requires that we not shrink from acting, wherever it takes us.

The theory of humanitarian intervention is very much the product of the twentieth century. It is, first of all, a response to decade after decade of massive human rights violations, including genocide. Perhaps it is only survivors' guilt that motivates democratic counties to wonder if they could have done more--for the Jews or the Cambodians or the Kosovars or Rwandans or Sudanese or Timorese. But humanitarian intervention depends on timely knowledge, itself a product of modern technology that allows news to travel everywhere fast, and on the ability to respond, a product of modern military technology: Today we can watch the horrors of mass killing on television and airlift soldiers in by nightfall. But the question remains: When may, or must, the United States and its allies intervene in the affairs of other sovereign states in order to "stop the killing"?

Prior to the recent rise of humanitarian intervention, state sovereignty was the organizing principle of world affairs--not least in the United Nations, which is after all a collection of states and not of peoples. In a twist on the old saw that in America "all politics is local", in world politics all crimes, including human rights crimes, were local too. There was no appeal beyond the national capital. Indeed, the UN Charter well-nigh prohibits intervention across state borders, allowing only two exceptions: self-defense in cases of "armed attack" (Article 51) and actions taken with Security Council approval as matters of collective security (Chapter VII). Humanitarian intervention without Security Council approval is, under the Charter, illegal. The British Foreign Office summed up the traditional view in a 1986 statement:

The overwhelming majority of contemporary legal opinion comes down against the existence of a right of humanitarian intervention, for three main reasons: first, the UN Charter and the corpus of modern international law do not seem to specifically incorporate such a right; secondly, State practice in the past two centuries, and especially since 1945, at best provides only a handful of genuine cases of humanitarian intervention, and, on most assessments, none at all; and finally, on prudential grounds, that the scope for abusing such a right argues strongly against its creation. ... In essence, therefore, the case against making humanitarian intervention an exception to the principle of non-intervention is that its doubtful benefits would be heavily outweighed by its costs in terms of respect for international law.¹

But the British were fighting a rearguard action. At both the state and individual levels, change has been coming for years. The Genocide Convention permits, and may even be said to require, intervention. Time after time in the last half century, nations crossed borders and named humanitarian motives as their key concerns: India in East Pakistan in 1971, Vietnam in Cambodia in 1978, Tanzania in Uganda in 1979, as well as the more recent cases of the United States in Somalia or Haiti, and NATO (with the British Foreign Office in the lead) in Kosovo. If some of these claims of humanitarian motivation were laughable, they fit La Rochefoucauld's definition of hypocrisy: the tribute vice pays to virtue. A century ago, a country need do no more than invoke its national interest to justify military action; more recently, it is usually necessary at least to mention humanitarian motives.

The turning point was, of course, the end of the Cold War. Today, intervention in local disputes and internal abuses will not escalate into a superpower conflict and risk a major war. At the same time, whatever restraints the superpowers placed on their client states during the Cold War are gone as well, so that both potential abusers and potential interveners are freer to act. The resulting change in the international system is evident: humanitarian intervention is a fact of international life as much as are trade treaties. Labor standards, environmental regulations, regional bodies such as the European Union or Mercosur, and international human rights conventions now abound, adding layers of new rules to the older set that long governed international commerce. But while treaties continue to work fairly well and will no doubt remain in force at the end of the twenty-first century, humanitarian intervention, a far newer phenomenon, may not. For so far, it works badly in practice and not at all in theory. The question is whether its flaws are intrinsic and unavoidable, or can be fixed.

Krauthammer's Law

THE MOST serious pragmatic problem posed by humanitarian intervention is that the willingness of the intervening country to place lives and resources at risk is determined by the degree of national interest involved. Charles Krauthammer has described the resulting problem in an article in these pages, in which he calls the notion of humanitarian war "an idea whose time has come, and gone." He elaborates: "The central contradiction--the Iron Law of Humanitarian War--is this: Humanitarian war requires means that are inherently inadequate to its ends." The first reason that humanitarian war has no future, in his view, is that it involves "a contradiction of means: bloodless war."²

Krauthammer's point is well taken, as is his observation that the very effort to avoid bloodshed (your own or the opponent's) may prolong a war. When the motive is purely humanitarian, national interests vague or nonexistent, and the risks high, humanitarian intervention will not, ironically, be an attractive option.

There are, however, cases where conditions are more favorable. Sometimes the risks seem low due to a huge disproportion of power between the interveners and the target state; sometimes a measure of national interest is involved. For example, when the United States intervened in Haiti for "humanitarian" reasons, did we not also want a decent government in place there so that we could in good conscience send back Haitian "boat people"?

In fact, absolutely pure humanitarian interventions may be the exception rather than the rule. To diminish Krauthammer's gap between ends and means, it is necessary only to reduce the risks a bit and raise the national interest rewards above zero, and at some point a balance may be found. Indeed, considering the cases in which the world either did not intervene at all or did so without much energy--as in the cases of Rwanda and Tibet--Krauthammer's 's problem solves itself: when ends and means are too far out of whack, there will be no intervention. The decision to intervene is, after all, still being made most often by political leaders whose pursuit of humanitarian ends will continue to be tempered by their need to win elections.

Krauthammer's suggestion that means will be limited by ends--that humanitarian goals presage a weak-kneed effort that will fall short--may be wrong as well. Acting with absolute certainty of our moral rightness, we may be willing to pound the malefactors into the ground. In Kosovo, that is what happened: while the NATO effort began weakly, it later became so intense that some (including myself) complained of the civilian casualties being caused in Serbia. The moral arithmetic here became perverse: we were unwilling to risk many American casualties in air or ground combat but, given the rightness of our cause, we felt justified in a high-altitude bombing campaign that inevitably killed civilians. Instead of Krauthammer's limited war dilemma, we may in the future find ourselves confronted with a much different problem if we choose to ignore the "just war" doctrine of proportionality: we may just blast the bastards to smithereens--and feel better about it than we have any business feeling, especially in the hum anitarian context.

These are practical problems: how hard do you fight, how much do you risk, how many do you kill when direct national interests are quite limited? To these questions one more must be added: how likely is success? To impose hardship and to take lives when there is little chance of solving anything is not only feckless but immoral. Krauthammer argues powerfully that few cases will provide any reasonable chance of a positive outcome because the social rifts deep enough to produce massive human rights crimes cannot be solved by a brief bit of international policing. We may stop the bloodshed while we are there, only to see it take off again when we leave.

In many cases, we will indeed be too ignorant about the society in question to do much good except keep the human rights violators apart from their intended victims for a while. Or we will be willing to intervene briefly but unwilling to meet the costs of the prolonged stay that real change requires. On other occasions our self-interested intervention will be wrongly labeled humanitarian, and we will advance our own national goals even if the locals benefit little (as in the case of Haiti). In all these situations, humanitarian intervention may achieve little long-term improvement in the lives of the people it claims to help.

But not always. The practical difficulties of successful humanitarian intervention remind us that our goal is to do good rather than feel good, but they should not block action when there is something useful to be done. Sometimes the problem may be an individual dictator: Panama, for example, is clearly better off with General Noriega in jail (whether one calls the U.S. intervention there humanitarian or imperialist) and has become a working democracy. Sometimes even deep social divisions can exist without sectarian violence or mass murder, and a period of foreign intervention may open a new phase in the political life of a nation. The "realist" school provides important warnings against allowing moralizing to replace judgment, but prudence does not always dictate inaction or indifference.

A Theoretical Vacuum

AS DIFFICULT as these practical issues are, the doctrinal problems involved in humanitarian intervention may be worse. It is a practice in search of a theory.

The traditional defense of humanitarian intervention under international law and in the United Nations was that it is meant to solve a threat to international peace. This argument is both non-humanitarian and simply unpersuasive. It helps when you are trying to shoe-horn things into UN categories, but the argument doesn't hold up in the real world: while all human rights crimes may not be local, they are not all international either. Sometimes borders can be sealed. The "threat to peace" argument must be reserved for appropriate cases; in others the "forfeiture of sovereignty" argument is better. This argument posits the following: sovereignty rests in the people. There cannot be popular consent to severe human rights violations, hence the state is acting against the true sovereign (the people) when it commits human rights crimes. The intervention is then undertaken on behalf of the "true" sovereign and does not represent a breach of sovereignty.

For Americans, weaned on "We the People", this formulation is attractive. In fact, the forfeiture of sovereignty doctrine can be applied to many cases of human rights abuse. But before applying this doctrine too widely, we should consider its implications. What about crimes committed willingly by a sovereign majority against a small minority, up to and including genocide? Here the "general will" may be to slaughter a particular sect or religious or racial group, and the sovereignty argument fails. The interveners must instead simply impose their morality on the local majority, insisting that there is a global standard of human rights law that no one may breach. The Genocide Convention and the Convention Against Torture are codified examples of such laws, and it may be that in this new century their narrow confines will be widened. Courts in England have ruled in the Pinochet case that under international law, torturers are like pirates: they may be captured anywhere because there is a universal interest in s topping torture. One can imagine broader and broader categories of human rights violations being brought under international oversight, providing justifications not only for court actions against individual abusers but for some forms of intervention as well. Why only genocide? Why large-scale atrocities and not smaller scale ones? Why not mass murder, or political murder, or violent repression?

Such questions lead to other practical ones, such as who would have the stomach for constant intervention in the dozens of countries that fall short of peaceful and democratic politics, but again there are also doctrinal issues to consider. Who has the right to decide that a human rights crime is being committed, or that it rises to the level of offense covered in some treaty or convention? Who has the right to organize an intervention force? If the moral prerequisites for intervention fall (allowing or demanding it at ever lower levels of human rights crimes), the procedures for intervention must be improved. Since almost every intervener claims some moral basis for his action, expanding the opportunities for intervention expands those for mischief as well. In recent years, the intellectual battle has been fought mostly by human rights activists favoring intervention. They have in the main strung together arguments justifying moves by the great democracies to rescue people in reasonably backward places. In the not-so-distant future, they may wish to direct their attention to the other side of the coin: standards and procedures aimed at preventing misbegotten military ventures whose goals are flimsily clothed in the language of humanitarianism. Stopping a bogus humanitarian intervention may come to be as pressing a challenge as launching a genuine one.

A Growing Problem

WHAT WILL THE coming decades bring? There is reason to believe that humanitarian intervention will be needed more often, rather than less. The last decades of the twentieth century suggest that ethnic and religious identities, long expected to decline in the face of "modernity", are instead rising in salience to millions of people, not just in the former Yugoslavia or in East Africa but around the globe.³ Sectarian strife may well become more prevalent, especially when combined with renewed nationalism. For the collapse of the Soviet empire and the end of the Cold War have unquestionably unleashed (or, to use a more neutral term, liberated) and strengthened desires to have a state coterminous with the nation and representing its true aspirations--whether in Eritrea, East Timor, Kurdistan or Chechnya. Moreover, increased religious devotion has in some areas led to human rights violations against nonbelievers, and in others has combined with nationalism in an explosive brew that threatens minority rights. I t seems that as the world economy becomes more integrated, the desire for, indeed the intensity of feelings about, narrower religious and national identities is strengthened. And when identity politics becomes violent, the preconditions for large-scale human rights abuses and thus for humanitarian intervention are in place.

It is probably true as well that the "creative destruction" we associate with capitalism will add to these trends. Globalization of trade and finance can often disrupt society in a beneficial manner and hasten the arrival of democracy; it did so in South Korea, Taiwan and Chile, to take three familiar examples. But modernization may also be disruptive to manners and morals, thereby producing a counter-reaction: a search for stronger traditional identities, the reassurance and security of group membership, and the comforting exclusion of those who are different. One sees this phenomenon at its most benign when the government in London cedes sovereignty "up" to Brussels--or "down" to the Scottish and Welsh assemblies. It will appear in a less benign guise elsewhere.

The concept of sovereignty, already under pressure, will be weakened further if this scenario is accurate. To use a corporate legal analogy, we have already "pierced the veil" of inviolable state frontiers, and there will be no turning back. Practical objections to particular armed interventions will prevent some, but the cards are stacked against indifference. In many Third World situations the victims are Christians whom the leading Western powers will find it politically difficult to ignore. Meanwhile, ignoring victims who are very poor and non-white will offend the world's pretensions to believe in the equal dignity of all mankind. Humanitarian nightmares that are the product of nationalism are very likely to involve border wars or border changes, and therefore legitimately to involve other countries and the international community. Samuel Huntington's vision in The Clash of Civilizations is proving to be prescient.

Finally, the so-called "revolution in military affairs" may play a role as well. If the greatest pragmatic limit on intervention is the unwillingness of potential interveners to bear great risks and to take many casualties, new weaponry may lighten their load. Pilotless planes, robot vehicles, smart bombs, information warfare and the like can reduce the risks to a politically acceptable level. The war in Kosovo, like the Gulf War intervention against Iraq, may be a harbinger of a huge military imbalance that would allow leading powers to intervene without much risk of sacrificing soldiers. Who would not want to rescue thousands of desperate people if the only risk were that a few robots would have to die for their country?

Use It or Lose It

CURIOUSLY, even the efforts to establish rules and limits for intervention will erode traditional views of sovereignty. It is difficult to imagine what practical standards can be written into international law--when should a large-scale human rights violation be handled with speeches or boycotts, and when is armed force justified? The words international lawyers choose will never provide much guidance or much restraint. It is far more likely that a requirement of collective action will be imposed; that is, armed intervention is justified under international law when some group of nations says so: the Security Council, the General Assembly, NATO or ad hoc combinations that may be assembled in a particular case. The requirement for group action revives the ideals of collective security that we associate with the League of Nations and United Nations and appeals to the hope that the burdens of intervention will be widely shared. But its main effect will be to disallow any single capital from deciding to use arme d force, and this is a large interference with a right once thought to be at the very heart of sovereignty.

This requirement of collective action is inevitable, and on the whole a positive development. Humanitarian intervention is defined not by means but by ends. The tools of humanitarian intervention are military, just as they are in cases of blatant aggression, so we can only distinguish the moral from the immoral intervention by making a judgment about motives. It is unlikely today that an armed action aimed at repression or characterized by aggressive intent will capture much international support. It is equally unlikely that an action aimed at saving lives will elicit no such support. To return to realpolitik, it is unlikely in the foreseeable future that a powerful American diplomatic effort will not be able to drum up support for any action that we feel we must take. Moreover, our military dominance makes intervention without our own approval and participation difficult and often impossible.

For Americans, this most idealistic of discussions cannot escape the debate over the uses of our extraordinary power. Those who fear that the United States leans always to imperial adventures and to the support of repressive regimes will have no trouble in urging great restraint, but this camp has shrunk a good deal since the end of the Cold War. As we saw in Kosovo, many who attacked American interventionism repeatedly during the 1960s, 1970s and 1980s--even many who voted against U.S. intervention in the Gulf War--were loudly calling for more of it in the 1990s. Today the more consequential debate is between those who believe power is bolstered when it is husbanded, and those in the "use it or lose it" camp. The former group argues that the United States is uniquely able to mount an operation like Operation Desert Storm or to counterbalance growing Chinese power in East Asia, and should save its powder for those requirements. To use our resources in interventions that could be undertaken by lesser powers is to squander them--and with them the political will to intervene. The interventionists counter that it is our moral leadership that will be squandered if we stand aloof when massive human rights abuses are being committed and our allies are considering action. Successful interventions, they argue, breed confidence in our will and our ability, engendering more political support at home and abroad; they may also deter tomorrows human rights abuser.

The outcome of this debate over the use of American power will tell us more about the future of humanitarian intervention than any other single factor. Such is the power and reputation of the United States as this century begins, so remarkable its cultural and military dominance, that humanitarian intervention has neither "come and gone" nor is it about to take over world politics. Like almost everything else in global affairs, its future depends largely on where the Americans come out.

THERE will be no clear and final answer to that question. The coming decades will bring circumstances where violent human rights abuses are frequent, the temptation and opportunity for intervention appears, and there are no insurmountable obstacles to action. We will then be faced with a myriad of practical and legal objections, for our moral standards have gotten ahead of international law and practice, and far ahead even of our ability to change the world. The situation unavoidably will be messy: there will often be no clarity as to when we fall on the side of the line where armed action is permissible, desirable or required, and when we are about to do something stupid. We will return again and again to the same debate, amongst ourselves and with other democratic nations that share our desire to deter, diminish, stop or punish human rights violations. If we will sometimes find a case where action or inaction seems an easy conclusion, we will face more cases where it is hard to figure out whether anything truly useful can be done. At the back of our minds we will worry about the costs we must pay, the impact on our reputation and self-esteem of action or inaction, and whether in the end we would be better off in a world where national sovereignty has all the legitimacy of the divine right of kings.

Yet this is exactly the debate we should be having, and these are exactly the doubts we should be entertaining. If, as I started by saying, the theory of humanitarian intervention is a product of the twentieth century, the practice of it is the product of the unique circumstances at that century's end and at the beginning of this one. The current international debate over humanitarian intervention reflects some advance in moral standards, but it reflects more fundamentally a single fact: that for the first time in human history a democratic society is the dominant world power. We Americans are all engaged in a novel enterprise: judging whether and how our nation's power, combined with that of like-minded friends and allies, can be used to eliminate or at least reduce abuses of the rights of man. In the course of the last century, those rights were more and more carefully elaborated in international conventions that were, usually, not worth the paper they were printed on. They were signed by the likes of Stal in and coexisted with a vast empire of repression that took in much of the world's population. In a few short years we can now envision, indeed we are actively engaged in, making those words real--not for all mankind, but for an ever increasing portion of it.

It is to be expected that this effort, tangled as it is with the great debate over the uses of American power, confuses and divides us as it does friends of human rights everywhere. We must get used to it, for this effort, and this debate, will characterize international politics and much of our own in the coming decades.

1 UK Foreign Office Policy Document No. 148, The British Yearbook of International Law 1986, p. 614.

2 Krauthammer, "The Short, Unhappy Life of Humanitarian War", The National Interest (Fall 1999), pp. 6-8. [emphasis in original]

3 See Peter L. Berger, ed., The Desecularization of the World: Resurgent Religion and World Politics (Grand Rapids, MI: Win. B. Ferdmans, 1999).

Elliott Abrams is president of the Ethics and Public Policy Center. From 1981 to 1989 he served as assistant secretary of state for International Organization Affairs, Human Rights and Humanitarian Affairs, and Inter-American Affairs. A version of this article will appear this April in Freedom House's Freedom in the World.