In his March 17, 2003 speech delivering a final ultimatum to Saddam Hussein, President George W. Bush, after citing the Iraqi dictator's violations of various disarmament requirements and invoking the right of the United States and its allies to preemptive self-defense, also addressed the Iraqi people: "If we must begin a military campaign, it will be directed against the lawless men who rule your country and not against you. As our coalition takes away their power, we will deliver the food and medicine you need. We will tear down the apparatus of terror and we will help you to build a new Iraq that is prosperous and free. In a free Iraq, there will be no more wars of aggression against your neighbors, no more poison factories, no more executions of dissidents, no more torture chambers and rape rooms. The tyrant will soon be gone. The day of your liberation is near."

Whether or not one agrees with the case made for the American-led invasion of Iraq-in retrospect, the reference to "torture chambers and rape rooms" must be extraordinarily uncomfortable of late after the revelations about the goings-on at Abu Ghraib prison-the president's remarks appealed explicitly to an international consensus that would have been unimaginable just a few years ago, an achievement partially obscured amid the heated exchanges between representatives of the U.S. administration, its supporters and opponents in other governments, and the United Nations over the course of the last two years. The object of that consensus was the idea of "humanitarian intervention." In the report it presented to the UN in 2001, the International Commission on Intervention and State Sovereignty contended that where a determined population "is suffering harm, as a result of internal war, insurgency, repression or state failure, and the state in question is unwilling or unable to halt or avert it, the principle of non-intervention yields to the international responsibility to protect." The Commission even suggested that the failure to intervene might, in some cases, be itself a breach of international law. Consequently, in the name of the humanitarian principle, actions-especially the military variety-that would have seemed to the signatories of the original UN Charter clear cases of interference have been justified by citing everything from gross human rights abuses to the threat of terrorism. While critics of the American policy may argue about whether or not the situation in Iraq called for a "humanitarian intervention," few in the international community question the underlying principle of such a use of military force.

In fact, the dust has barely settled from the diplomatic querelle over Iraq when some of most vociferous critics of American actions in the Middle East were demanding interventions elsewhere that would inevitably be either led or supported by the U.S. military (the exclusive possession by the U.S. of the world's only long-haul military cargo airlift capability of any magnitude makes any such operation without American backing almost impossible). Last summer, a 2,300-strong unit of U.S. Marines en route home from the war in Iraq detoured to the coast of Liberia where a small task force led by the amphibious assault ship U.S.S. Iwo Jima had been dispatched to put pressure on the West African country's dictator, Charles Ghankay Taylor, to give up power. Earlier this year, the American government teamed up with that previously "principled" critic of interventions not authorized by the UN, France, to send a military force into Haiti that eased out the island's besieged ruler, Jean-Bertrand Aristide, before obtaining a Security Council resolution. This past week, finally noticing that an estimated one million people in Sudan's western Darfur region had fled their homes after being attacked by the government-backed janjaweed militia (this in addition to the four million already displaced by the long simmering civil war in the country's south), human rights groups and leading editorialists-including the editors of the usually hardnosed Economist newspaper-have called upon the UN Security Council to authorize some kind of intervention.

While armed intervention may sometimes be needed to halt abuses of the vulnerable, each such intervention presents a unique series of complex challenges (as does its aftermath). As the idea of "humanitarian intervention" is here to stay, even with America's military juggernaut preoccupied with the Greater Middle East, it would do well to review the history of recent interventions and ponder some of their lessons about the consequences of such commitments. Among these are the following:

Since most violence is perpetrated more quickly than commonly realized, an intervention will almost inevitably come too late. Even if a consensus about intervention were achieved as soon as news of the humanitarian crisis reaches the international community-an almost impossible task-history teaches us that, sadly, the killers will still be faster than the would-be interveners. In Rwanda, an estimated 500,000 of the 800,000 victims were killed in the first three weeks of the hundred days of the 1995 genocide. In East Timor, Indonesian-backed militias displaced most of the population in one week following the vote for independence in 1999. While this does not mean that the international community should shirk from intervening in the face of grievous abuse, it should be realistic about what it can accomplish even if it mobilized immediately.

Intervention addresses symptoms rather than underlying causes. While a humanitarian intervention might indeed stop human rights abuses, refugee flows, and material insecurity, these symptoms usually manifestations of underlying pathologies-including the failure of the state, the breakdown of the civil society institutions, the arming of militias, the division of society along ethnic or sectarian lines-that do not lend themselves to remedy by armed outsiders, as the fatal 1993 UNOSCOM II debacle in Somalia demonstrated.

Intervention opens the political space to new, often unexpected, actors. Outside intervention, by displacing the old political order, allows new forces to emerge. Some will be old opponents of the former rulers, others will represent entirely new movements. While some of these new actors may be benign, others represent graver threats. As the Coalition Provisional Authority has learned in Iraq, Ambassador L. Paul Bremer's policy of total de-Ba‘athification of state institutions created a vacuum quickly filled by foreign jihadis who took to the Sunni heartland as well as local firebrands like the upstart junior cleric Moqtada al-Sadr who enjoys some support among poorer urban Shi‘ites, in part because of the social services he delivered after the fall of Saddam Hussein. And, as the International Security Assistance Force (ISAF) in Afghanistan has learned from the warlords that the U.S. and other countries initially backed against the Taliban, genies once released do not readily go back into their lamps.

Intervention is the starting point for a complex political process whose eventual end point cannot be predicted. Political conflicts do not come to an end just because outside forces have interposed themselves between combatants. About the most that can be hoped for is that an intervention will channel tensions into the political arena, preventing more violent expressions of communal discord. Iraq's delicate ethnic and religious balance between Arabs, Kurds, Assyrians, and Turkomens and Sunnis, Shi‘ites and Christians will continue to occasion communal tensions when the CPA's relatively clumsy efforts to create a credible government is no more than a dim memory. It would behoove the international community to recall Bertrand de Jouvenel's maxim that political problems "give rise to settlements, not solutions." What form such settlements take and their eventual durability cannot be predetermined, much less imposed.

Economic progress will be difficult if the intervention distorts pre-existing incentive structures. In general, the economies of countries that are the objects of humanitarian interventions are already weakened, if not collapsed altogether. However, an intervening force must tread delicately if it is not to destroy what remains of local markets. Vast provisions of assistance to areas threatened with starvation, for example, may well wipe out remaining farmers, exacerbating the food security situation over the intermediate and long terms, as was the case during the Ethiopian famine of the 1980s. And the presence of large foreign forces can lead to the creation of a state that is unduly dependent upon aid and whose citizens are chiefly employed in servicing their rescuers, as has been the case in Sierra Leone since UNAMSIL helped end that country's civil war two years ago (the budget of the military component of the international intervention, $543.49 million for the current fiscal year, is nearly twice as much as that of the West African country's government and accounts for about one-fourth of its GDP). It will take years to assess the impact on the Iraqi economy of the presence of the U.S.-led coalition's 200,000-plus troops and the various foreign contractors and aid workers.

Intervention can exacerbate, rather than reduce, the humanitarian crisis. In fact, an ill-timed humanitarian military intervention can cause the very tragedies it was supposed to prevent, intensifying the level of violence within a conflict and thus increasing the domestic security threat and spreading regional instability.  As I have argued previously in these pages, the 1990 ECOMOG intervention in the Liberian civil war fell right into this ethical dilemma and contributed to prolonging the conflict by over a decade.^[i] Even non-military interventions, like the 1994 provision of humanitarian materiel to the refugee camps in Eastern Zaire (now the Democratic Republic of Congo) where the Rwandan Hutu génocidaires took refuge, can have disastrous consequences, including the militarization of the camps and the beginning of what would be "Africa's world war."^[ii]