As forces fighting Libyan leader Muammar al-Qaddafi consolidated control of Tripoli in the last days of August 2011, many pundits began speaking of a victory not just for the rebels but also for the idea of humanitarian intervention. In Libya, advocates of intervention argued, U.S. President Barack Obama had found the formula for success: broad regional and international support, genuine burden sharing with allies, and a capable local fighting force to wage the war on the ground. Some even heralded the intervention as a sign of an emerging Obama doctrine.

It is clearly too soon for this kind of triumphalism, since the final balance of the Libyan intervention has yet to be tallied. The country could still fall into civil war, and the new Libyan government could turn out to be little better than the last. As of this writing, troubling signs of infighting among the rebel ranks had begun to emerge, along with credible reports of serious human rights abuses by rebel forces.

Yet even if the intervention does ultimately give birth to a stable and prosperous democracy, this outcome will not prove that intervention was the right choice in Libya or that similar interventions should be attempted elsewhere. To establish that requires comparing the full costs of intervention with its benefits and asking whether those benefits could be achieved at a lower cost. The evidence from the last two decades is not promising on this score. Although humanitarian intervention has undoubtedly saved lives, Americans have seriously underappreciated the moral, political, and economic price involved.

This does not mean that the United States should stop trying to promote its values abroad, even when its national security is not at risk. It just needs a different strategy. Washington should replace its focus on military intervention with a humanitarian foreign policy centered on saving lives by funding public health programs in the developing world, aiding victims of natural disasters, and assisting refugees fleeing violent conflict. Abandoning humanitarian intervention in most cases would not mean leaving victims of genocide and repression to their fate. Indeed, such a strategy could actually save far more people, at a far lower price.


As the Cold War ended, many foreign policy analysts predicted that the United States would return to isolationism. Without the need to counter the Soviet Union, it was argued, Americans would naturally turn inward. It hardly needs saying that these predictions have not been borne out. Throughout the 1990s, the United States continued to play the leading role in global affairs, maintaining military bases around the world and regularly intervening with military force. The 9/11 attacks only reinforced this pattern. Politicians from both parties today regard the deployment of military forces as a routine part of international relations.

It was not always this way. Although isolationism among conservatives went virtually extinct in the 1950s, during the Cold War, and especially after Vietnam, liberals almost always opposed the use of military force, even for humanitarian purposes. But after the Soviet Union collapsed, many on the left began to embrace the idea that the vast military capabilities assembled to check its influence could now be used to save lives rather than destroy them. The evaporation of Soviet power also made it easier to use those forces, lifting one of the most important constraints on the deployment of U.S. troops abroad. The astonishing success of the U.S. military in the Persian Gulf War of 1990-91, meanwhile, convinced many people that Americans had finally lost their aversion to intervention abroad, kicking the "Vietnam syndrome" once and for all. The costs of using force appeared to have fallen dramatically.

The end of the Cold War also touched off a bloody civil war in Yugoslavia, the first major conflict in Europe in almost 50 years. Although the United States had few national security interests at stake there, the brutal nature of the fighting prompted many calls for intervention, mostly from the left. These calls did not move President George H. W. Bush to intervene in the Balkans, but his decision to send forces to Somalia in 1992 was partially an effort to demonstrate that he was willing to use the military for humanitarian missions if the conditions were favorable. Under President Bill Clinton, the United States went further, undertaking major humanitarian interventions in Bosnia, Haiti, and Kosovo. A surprising number of opinion-makers on the left, including Peter Beinart, Thomas Friedman, Christopher Hitchens, Michael Ignatieff, and Anne-Marie Slaughter, later lent their support to the 2003 invasion of Iraq out of the conviction that it would end decades of human rights abuses by Saddam Hussein.

Prominent Democrats also called on the United States to use military force to end the mass killings in Darfur, Sudan. In 2007, then Senator Joe Biden told the Senate Foreign Relations Committee, "I would use American force now. . . . I think it's not only time not to take force off the table. I think it's time to put force on the table and use it." During the 2008 Democratic presidential primary race, Hillary Clinton repeatedly called for the imposition of a no-fly zone in Sudan. Most recently, in March, Obama defended the intervention in Libya, saying, "There will be times . . . when our safety is not directly threatened, but our interests and values are. . . . In such cases, we should not be afraid to act." The public agreed: a poll conducted days after NATO began air strikes against Libya found that, even with two other ongoing wars, majorities of both Democrats and Republicans supported the military action. Only self-described independents were more likely to disapprove than approve.


Proponents of such interventions usually make their case in terms of the United States' moral responsibilities. Yet perhaps the most important costs incurred by military interventions have been moral ones. On the ground, the ethical clarity that advocates of human rights have associated with such actions -- saving innocent lives -- has almost always been blurred by a much more complicated reality.

To begin with, aiding defenseless civilians has usually meant empowering armed factions claiming to represent these victims, groups that are frequently responsible for major human rights abuses of their own. Although advocates of humanitarian intervention in the 1990s frequently compared the atrocities of that period to the Holocaust, the moral calculus of intervening in these conflicts was inevitably more problematic. The Tutsi victims of Hutu génocidaires in Rwanda and the Bosnian Muslim and Kosovar Albanian victims of Serbian paramilitaries in the former Yugoslavia were just as innocent as the Jewish victims of the Nazis during World War II. But the choice to aid these groups also entailed supporting the less than upstanding armed factions on their side.

In Bosnia, for example, the United States eventually backed Croatian and Bosnian Muslim forces in an effort to block further aggression by Serbian President Slobodan Milosevic. These forces were far less brutal than the Serbian forces, but they were nevertheless implicated in a number of large-scale atrocities. In August 1995, for example, Croatian forces drove more than 100,000 Serbs in the Krajina region of Croatia from their homes, killing hundreds of civilians in what The New York Times described as "the largest single 'ethnic cleansing' of the war." It was later revealed that the U.S. State Department had allowed private U.S. military consultants to train the Croatian army in preparation for the offensive. In April of this year, two Croatian military leaders in charge of the campaign were convicted of crimes against humanity at The Hague.

Similarly, after the NATO bombing campaign in 1999 helped evict Serbian forces from Kosovo, the Kosovo Liberation Army turned on the Serbian civilians remaining in the province and in neighboring Macedonia, killing hundreds and forcing thousands to flee. Since the end of the war, human rights groups and the Council of Europe have repeatedly called for investigations of high-ranking KLA officials suspected of engaging in executions, abductions, beatings, and even human organ trafficking.

Another set of moral costs stems not from the unsavory behavior of the groups being protected but from the unavoidable consequences of military intervention. Even if the ends of such actions could be unambiguously humanitarian, the means never are. Using force to save lives usually involves taking lives, including innocent ones. The most advanced precision-guided weapons still have not eliminated collateral damage altogether. Many Americans remember the 18 U.S. soldiers who died in Somalia in 1993 in the "Black Hawk down" incident. Far fewer know that U.S. and UN troops killed at least 500 Somalis on that day and as many as 1,500 during the rest of the mission -- more than half of them women and children.

In Kosovo, in addition to between 700 and several thousand Serbian military deaths, Human Rights Watch estimates that NATO air strikes killed more than 500 civilians. NATO pilots, ordered to fly above 10,000 feet to limit their own losses, found it difficult to distinguish between friend and foe on the ground. Sixteen civilians were also killed when NATO bombed a Serbian television station that it accused of spreading pro-government propaganda. These and other incidents led Human Rights Watch to conclude that NATO had violated international humanitarian law in its conduct of the war. Amnesty International accused NATO of war crimes.

Although military interventions are calculated to increase the costs of human rights abuses for those who commit them, perhaps interventions' most perverse consequence has been the way they have sometimes actually done the opposite. If perpetrators simply blame the victims for the setbacks and suffering inflicted by the intervention, the incentives to retaliate against victim groups, and possibly even popular support for such retaliation, may rise. Foreign military interventions can change victims from being viewed as a nuisance into being seen as powerful and traitorous enemies, potentially capable of exacting revenge, seizing power, or breaking away from the state. Under these conditions, even moderates are more likely to support harsh measures to meet such threats. And with most humanitarian missions relying on airpower to avoid casualties, potential victims have little protection from retaliation.

In Kosovo, for example, the NATO bombing campaign hardened Serbian opinion against the Kosovar Albanians and rallied public support behind Milosevic, at least initially. Many Serbs donned T-shirts with a bull's-eye and attended anti-NATO rock concerts to express their solidarity against the West and for Milosevic's regime. One Serb told a reporter, "When Milosevic thought he could do whatever he wanted with us, I was against him. Now I am against NATO because they are strong and we are weak." Still worse, the bombing may have actually provoked a major upsurge in the violence, or at least given Milosevic the excuse he needed to implement a long-held plan to ethnically cleanse the region. Either way, when Serbian attacks on Kosovars escalated, NATO planes were flying too high and too fast to protect civilians on the ground.

The prospect of foreign military intervention also may encourage victims to rise up -- a perilous course of action if the intervening forces are not equipped to protect them or if the intervention arrives too late or not at all. Perhaps the most clear-cut example of this perverse dynamic occurred in the aftermath of the Persian Gulf War. During the war, Bush said the Iraqis should "take matters into their own hands and force Saddam Hussein, the dictator, to step aside." Many Iraqi Kurds and Shiites responded to this call, believing that the United States would send military forces to assist them or at least protect them from retaliation by Saddam. It was not to be. Wishing to avoid a quagmire, Bush decided to end the war just 100 hours after the ground invasion had begun. Saddam responded to the domestic uprisings with extreme brutality, killing perhaps 20,000 Kurds and 30,000-60,000 Shiites, many of them civilians.

Another set of costs associated with humanitarian interventions are political. The United States' humanitarian interventions have won the country few new friends and worsened its relations with several powerful nations. The United States' long-term security depends on good relations with China and Russia, perhaps more than any other countries, but U.S.-sponsored interventions have led to increasing distrust between Washington and these nations. Both countries face serious secessionist threats and strongly opposed U.S. intervention in Bosnia and Kosovo out of fear of setting an unwelcome precedent. The accidental bombing of the Chinese embassy in Belgrade in 1999, which killed three Chinese citizens, resulted in major demonstrations outside the U.S. embassy in Beijing and an acute deterioration of relations between the two countries that lasted almost a year. Conflict with Russia over Kosovo continues to this day. 

The political strains have not been limited to relations with potential U.S. adversaries. Brazil and India, two of the United States' most important democratic allies in the developing world, also opposed the intervention in Kosovo and have refused to recognize its independence. More recently, both countries sided with China and Russia and condemned the intervention in Libya, arguing that NATO's actions significantly exceeded what the UN Security Council had authorized.

A less tangible political cost of these interventions has been their corrosive effect on the authority of international organizations such as the UN. In regard to Kosovo, the threat that China and Russia would veto a resolution to intervene in the UN Security Council forced proponents of intervention to insist that the mission did not require UN authorization. A few years later, however, many of these one-time advocates found themselves arguing against U.S. intervention in Iraq, at least in part on the grounds that Washington had failed to obtain UN approval. Having ignored the UN when it came to Kosovo and Iraq, it will be more difficult for the United States to condemn the use of force by other states that fail to obtain UN approval.


Perhaps the most frequently ignored costs of humanitarian interventions, however, have been what economists call opportunity costs -- the forgone opportunities to which the resources for a military mission might have been put. These costs are considerable, since military intervention is a particularly expensive way to save lives.

Each of the more than 220 Tomahawk missiles fired by the U.S. military into Libya, for example, cost around $1.4 million. In Somalia, a country of about 8.5 million people, the final bill for the U.S. intervention totaled more than $7 billion. Scholars have estimated that the military mission there probably saved between 10,000 and 25,000 lives. To put it in the crudest possible terms, this meant that Washington spent between $280,000 and $700,000 for each Somali it spared. As for Bosnia, if one assumes that without military action a quarter of the two million Muslims living there would have been killed (a highly unrealistic figure), the intervention cost $120,000 per life saved. Judging the 2003 Iraq war -- now a multitrillion-dollar adventure -- primarily on humanitarian grounds, the costs would be orders of magnitude higher. 

The lesson that many human rights advocates have drawn from these calculations is not that intervention is too costly but that it is no substitute for prevention. A careful study commissioned by the Carnegie Corporation of New York, for example, concluded that early but robust efforts at conflict prevention were almost always more cost-effective than reactive interventions. If only the math were so simple: this argument seriously underestimates the full costs of preventive efforts by assuming that the international community will correctly identify catastrophes long before they occur and intervene only in those cases. In reality, predicting which hot spots will turn violent is extremely difficult. As then UN Secretary-General Boutros Boutros-Ghali told reporters in Sarajevo in 1992, although the situation there seemed dire, his job was to think about all the conflicts around the world that might benefit from intervention. "I understand your frustration," he said, "but you have a situation that is better than ten other places in the world. . . . I can give you a list." Thus, although the costs of prevention in any given conflict would surely be much lower than the costs of a purely reactive intervention, these costs must be multiplied many times over because forces would end up intervening in crises that were never going to rise to a level that would have justified military intervention.

What is more, the record of low-cost preventive missions has been at least as bad as the record of interventions reacting to atrocities. One of the most tragic aspects of the genocides in Bosnia, Rwanda, and Darfur was that international peacekeepers were present during some of the worst episodes of violence, such as the slaughter of some 8,000 Bosnian Muslims in Srebrenica in 1995, which was witnessed by 400 UN peacekeepers. The problem in these cases was not that no one was sent to prevent the violence; it was that the forces that were deployed were not given the resources or the mandates to stop the violence breaking out around them. In some cases, they could not even protect themselves. More robust preventive deployments might have been more effective, but they would not have been cheap.


To be sure, $120,000 or even $700,000 does not seem like an unreasonably high price to pay to save a life; developed countries routinely value the lives of their citizens much more highly. Although these costs may seem low in absolute terms, in comparison to the other ways the United States' scarce resources might have been spent to save lives abroad, humanitarian intervention begins to look almost extravagant. Three strategies offer the prospect of helping more people with a much lower moral, political, and economic cost: investing in international public health initiatives, sending relief aid to victims of natural disasters and famines, and assisting refugees fleeing violent conflict. Millions more lives could be saved if the billions of dollars spent on humanitarian interventions were instead spent on these efforts.

International public health programs are almost certainly the most cost-effective way to save lives abroad. The World Health Organization estimates that every year at least two million people die from vaccine-preventable diseases alone (millions more die from other easily treatable infectious diseases, such as malaria or infectious diarrhea). This is an annual toll more than twice as large as the Rwandan genocide and more than 200 times the number of civilians who died in Kosovo. Measles alone killed more than 160,000 people in 2008, almost all of them children. It costs less than $1 to immunize a child against measles, and since not every unvaccinated child would have died from measles, the cost per life saved comes out to an estimated $224. Even using the exceedingly generous estimates above of the number of lives saved by military intervention, this means that on a per-life basis, measles vaccination would be 3,000 times as cost-effective as the military intervention in Somalia and more than 500 times as cost-effective as the intervention in Bosnia. The provision of antimalarial bed nets may be more efficient still -- costing only between $100 and $200 per life saved. The final bill may be even lower, since preventive public health expenditures such as these often more than pay for themselves in averted medical costs and increased productivity.

The lifesaving potential of such public health programs is enormous. Indeed, because of intensive vaccination initiatives, measles deaths have dropped by almost 80 percent since 2000, probably saving well over four million lives in the last ten years. And of course, vaccinating children for measles did not require killing anyone, violating international laws, or damaging important relationships with powerful countries.

A second way that the United States can save lives without the use of force is through disaster-relief efforts. The International Red Cross estimates that more than one million people were killed between 2000 and 2009 in natural disasters such as earthquakes, floods, and hurricanes. It is difficult to estimate how many lives were saved by international relief efforts in these disasters or how many more might have been saved had even greater resources been devoted to disaster preparedness and response. Disaster-relief programs are almost certainly less economically efficient in saving lives than the most effective public health programs, but like public health efforts, they avoid many of the moral and political costs of military intervention. Few forms of intervention are more deeply appreciated by recipients. After the U.S. military sent rescue and medical teams and emergency supplies to Indonesia in the wake of the devastating 2004 tsunami, the proportion of Indonesians who held favorable views of the United States, which had plummeted following the invasion of Iraq, more than doubled -- an important gain in the world's largest Muslim country.

A third set of strategies focuses on aiding potential victims of violent conflict and repression, including genocide and mass killing. Although using military forces to halt perpetrators and protect victims on the ground is usually very expensive, it is possible to assist victims of violent conflict at much lower cost by helping them escape to safer areas. Large refugee flows are rightly seen a humanitarian emergency in themselves, but refugees of violence are also survivors of violence. In practice, measures designed to help victims reach safety across international borders and to care for refugee populations once they arrive have probably saved more lives from conflict than any other form of international intervention.

History provides numerous examples that illustrate the potential of providing safe havens for refugees. Although the Nazis clamped down on emigration after World War II began, between 1933 and 1939 Germany actively encouraged it, a process that ultimately resulted in the exodus of approximately 70 percent of Germany's Jews. Had Western nations put up fewer barriers to Jewish immigration or actively sought to assist Jewish emigration, they would surely have saved many more lives. The ability of potential victims to escape likely played an even greater role in limiting the toll from repressive governments during the Cold War. Following the communist takeover in North Korea, for example, more than one million people, around ten percent of the population living above the 38th parallel, made their way to the South between 1945 and 1947. Had they been unable to flee, many would surely have been labeled enemies of the state and executed or sent to the North Korean gulag. Similarly, roughly 3.5 million Chinese refugees, mostly supporters of Chiang Kai-shek who would have been prime targets of Mao Zedong's subsequent campaigns against political enemies, escaped to Taiwan and Hong Kong following the communist victory in the Chinese Civil War in 1949. Today, many of the 250,000 Sudanese refugees surviving in camps in eastern Chad likely would have joined the 300,000-400,000 victims of the mass killing in Darfur had they not fled the fighting.

The first order of business, then, should be for outside powers to keep their borders open to victims fleeing violence. The large numbers of refugees who managed to escape the bloodshed in North Korea, China, and Kosovo were able to do so only because they could flee across open borders into neighboring states. Many victims are not so fortunate. For example, Iraqi Kurdish refugees attempting to flee the crackdown following the Gulf War initially faced closed borders as they tried to go to Iran and Turkey. Diplomatic pressure and economic assistance from the United States and NATO, however, ultimately prompted these countries to open their borders, at least temporarily.

Even when neighboring states are willing to open their doors, perpetrators often try to block victims' escape. Such was the case in Rwanda, where Hutu génocidaires set up roadblocks to prevent Tutsis from crossing into Burundi, Congo, Tanzania, and Uganda. In cases like these, the use of limited military force may make sense. In Rwanda, a relatively small military intervention, perhaps with airpower alone, could have destroyed roadblocks and secured key escape routes, helping tens of thousands reach safety. By one estimate, this strategy might have saved 75,000 lives.

The international community should also ensure the survival of refugees once they reach their destinations. The conditions awaiting most refugees of mass violence seldom provide much better odds of survival than do those faced by victims who remain behind. Not only are food, water, and shelter in short supply, but refugees are also frequently subject to violence and thievery at the hands of other refugees or local populations. Few refugees would survive for long without substantial external assistance. As a result, when the options for potential refugees are unattractive, many will prefer to stay and fight, even when their chances of success are slim. When refugees can expect more hospitable conditions across the border, however, more will choose to flee and more will survive when they arrive.


Proponents of humanitarian intervention may object that the calculus laid out here understates its effectiveness by neglecting the other U.S. interests that these military missions serve. Even the most ardent advocates of intervention in such places as Kosovo, Sudan, or Libya, however, usually concede that the United States' safety was never directly threatened by the crises there. At the same time, helping refugees and saving lives through public health programs and disaster relief also serve a variety of secondary U.S. interests -- improving relations with other countries, promoting economic development, and increasing regional stability. A full accounting cannot neglect these benefits, either.

Some may also protest that the United States cannot give up on humanitarian intervention since it is the only country with the capability to project power around the globe. This may be true, but it would be a relevant concern only if other countries or nongovernmental organizations were already devoting sufficient resources to nonmilitary forms of humanitarian aid. The millions of easily preventable deaths that still occur every year are evidence that much more is needed. Still others may assert that the United States has a special responsibility to oppose governments that are engaged in massive human rights violations, even at much greater cost, because doing so sends a message that the world will not tolerate crimes against humanity and despotism. But that message need not be sent with bombs. A stronger message, in fact, should be sent to governments that fail to provide even inexpensive health care or essential services to save the lives of their own citizens. Finally, some will argue that the United States does not need to choose between military intervention and humanitarian aid since it can afford both. This is correct, but given the number of people who could benefit from increased humanitarian aid, the country will have to vaccinate many more children and assist many more refugees before military intervention begins to look affordable in comparison.

The strategies suggested here are not without their own dilemmas, of course. Large refugee populations can foster instability if the refugees attempt to fight their way home or fall into conflict with local populations. And humanitarians have learned the hard way that relief aid and medical supplies can be hijacked by corrupt governments or violent rebel groups. Fortunately, these problems are less severe than the problems of military intervention, and there are ways to mitigate them, even if they cannot be eliminated altogether. The provision of humanitarian aid should be more closely monitored, the aid should be linked to other forms of aid that recipients desire, and the aid should be targeted to those countries and local groups that demonstrate that they can use it most effectively. Strategies to assist refugees must be combined with diplomatic coercion and tough economic sanctions designed to end the conflicts that forced the refugees out in the first place. With defenseless victims out of harm's way, international pressure on perpetrators would be much less likely to provoke further crackdowns.

As with most of the choices in international relations, these strategies are simply the best of a poor set of alternatives. Even so, a foreign policy based on them would not mean simply standing by in the face of atrocity and injustice. Indeed, efforts such as helping refugees could save thousands of lives even when a major military intervention is out of the question. Equally important, these strategies would do much to allow Americans to wholeheartedly embrace a less militarized foreign policy, restoring the United States' image as a force for good in the world and providing Americans with an alternative perspective on the use of force, something that has been absent from U.S. foreign policy debates. U.S. foreign policy has always sought to promote the values of its citizens, as well as protect their material and security interests abroad. The country should not abandon that noble impulse now. It simply needs a better way to act on it.


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