No fewer than nine times over the past decade, Western powers have deployed noble rhetoric, soldiers and taxpayer dollars in the service of nation-building. And no fewer than nine times, they have, to one degree or another, failed to build stable, self-sustaining nations.

The litany consists of Somalia, Haiti, Bosnia, Kosovo, Sierra Leone, East Timor, Liberia, Afghanistan and Iraq. The best one could say is that they are works in progress. The worst: Too many of them still can't function on their own and continue to pose threats to their own citizens as well as U.S. national interests. While genuine good -- both humanitarian and security-related -- has come of these efforts, the results have fallen far short of our professed objectives, consumed enormous resources and political capital, and left uncertainty about the U.S. and international commitment.

It is hard to avoid the conclusion that our interest in taking over "problem nations" has far outpaced our ability or willingness to solve those nations' problems.

Iraq, the biggest and most troubled of these undertakings, has focused attention on the future of nation-building as an element of U.S. foreign policy as never before. The subject has become a cottage industry. Some policy experts are proposing a cabinet-level agency for reconstruction, a civilian equivalent of the Pentagon's Central Command, or a coordinating agency like the Federal Emergency Management Agency. Others want a standby budget or a reorganization of the State Department.

The domestic U.S. politics of nation-building have been scrambled by Iraq, too. This election year no major candidate will assert, as George W. Bush did in 2000, that "I don't think our troops ought to be used for what's called nation-building." Yet, neither Bush nor Democratic Sen. John F. Kerry are going to be suggesting any new projects of this sort, either. With American soldiers and civilians dying in Iraq, domestic support for such ventures is waning. And American politicians seem less interested in perfecting nation-building than in redefining success and pulling out or passing off responsibility more quickly to the very nations we deemed incapable of helping themselves.

That's because nation-building has come to mostly mean the comprehensive occupation of collapsed or defeated states, the remaking of entire societies and sky-high, endless costs. Another approach, however, would feature peacekeeping, economic aid, technical assistance and support for elections that might, in some instances, make costly and frustrating military intervention less likely.

Since the end of the Cold War, the international community can claim noteworthy achievements in countries in crisis: ending years of conflict in Bosnia and Liberia; saving hundreds of thousands of lives in Somalia; halting the ethnic cleansing of Kosovo; arranging elections in Cambodia, East Timor and Afghanistan; and contributing to the ouster of Saddam Hussein, Charles Taylor, Slobodan Milosevic and Mullah Omar.

But more often, the results of nation-building efforts have been disheartening. We have repeat offenders, such as Haiti and Liberia. We have failures, such as Somalia, where the state's complete collapse benefited al Qaeda. Even East Timor, which the world first allowed to be utterly destroyed and now calls a nation-building success story, is decades from running its own affairs without pervasive U.N. involvement.

Bosnia and Kosovo are often advanced as paragons of international nation-building. But we have spent billions of dollars keeping people alive without producing much of a nation in either case. After almost 10 years, Bosnia remains a quasi-state. The central government depends heavily on international handouts and possesses little power. The nationalists who hold sway in the regional governments have only recently begun to feel their writ diminish under the crypto-imperial leadership of international representative Paddy Ashdown. Kosovo's fundamental dilemma -- final status inside Serbia or outside it -- is no closer to resolution than it was the day U.S. bombing ended five years ago. Instead Kosovo is a failed non-state, and the international community's willingness to export this dilemma to the future has stalled economic reconstruction, limited progress toward reconciliation, and generated uncertainty in Kosovo and throughout the Balkans.

In Kosovo and Bosnia we have imposed order while trying to build nations atop unresolved political conflicts. In Afghanistan and Iraq, we have resolutely refused to put in enough forces to even establish order.

The effort to build a decent Afghan state in a place 12 times the size of Bosnia, decimated by 25 years of war, and beset by ethnic rivalries is a remarkable undertaking. A representative government has been established and unprecedented nationwide elections are scheduled less than three years after the fall of the Taliban. However, Afghanistan has also resumed its place as the world's leading heroin exporter; its women remain mistreated and disenfranchised; and warlords and Taliban remnants alike continue to menace the nation. We chose a minimalist approach because of the sense that the place was too big, too complex and potentially too contentious to take over and because of military requirements in Iraq and elsewhere. Are we and our NATO allies really committed to the huge cost and lengthy time needed to reach the larger goals that we profess to seek? No one has the slightest idea.

In Iraq, the U.S. failure to prepare for the end of major combat or deploy enough forces to deal with resisters left us an extraordinarily difficult task -- fighting an insurgency that impedes reconstruction while trying to build a political process whose legitimacy depends on successful reconstruction. Although we have put the Iraqis front and center now, we are still essential to an outcome that is far from preordained and perhaps years away.

Yes, nation-building is an incredibly difficult undertaking -- but much of the difficulty stems from the problems we make for ourselves:

* The nature of peace agreements. When Western countries decide to intervene in conflicts, they have focused -- understandably -- on ending the fighting. Limiting casualties has been the name of the game, an objective that is often at odds with the goals of establishing self-sustaining states. In Bosnia, peace meant leaving the country divided into three ethnic mini-states. In Kosovo, our fear of using ground troops led to a U.N. resolution leaving Milosevic with a legal role in Kosovo. These choices bedevil Bosnia and Kosovo to this day. In Afghanistan, we opted to leave the warlords in charge of their fiefdoms, and in both Iraq and Afghanistan we left hostile elements on the field with access to arms -- a factor that may yet erase our gains in both places.

* The nature of international mandates. If international problem-solvers cannot agree on how a problem should be solved, trouble almost certainly follows. This lack of consensus has undermined efforts in both Afghanistan and Iraq, yet we might have learned its pitfalls from Kosovo.

Michael Steiner, a former head of the U.N. mission in Kosovo, put the problem bluntly: His function was not nation-building, because the U.N. Security Council had not mandated him to give Kosovo a state. The result is that Kosovo, indeed, is not a state but a colonial U.N.-run entity -- and a poorly functioning one -- that the international community refuses to alter despite the depressing economic conditions and the growing threat of internal violence.

* The problem of legitimacy. International legitimacy for military intervention and occupation derives from a U.N. mandate. The Bush administration hoped that the operation in Iraq could be legitimate without U.N. approval. But continuing violence has further undercut the intervention's validity and has made it more difficult to prolong involvement, share burdens and sustain domestic support. Mandates come with their own burdens. During negotiations to gather support, compromises often end up, as in Bosnia and Kosovo, limiting the military operation and making post-conflict administration more difficult.

But international legitimacy -- however important -- is no substitute for local legitimacy. Without local acceptance and participation, nation-building fails. Not enough of that involvement has happened yet in Kosovo. And when the United States has sought local involvement in Iraq, it has seemed to be mainly in response to violence.

When it comes to the comprehensive remake of collapsed and defeated states, it would be best to say straightforwardly that we do not know how to build nations in faraway places. And it is hard to foresee any enthusiasm for the next candidates for takeover and large-scale nation-building -- Iran, North Korea and the Congo, although military action against the first two cannot be ruled out.

Yet, despite the sobering record, the United States and the international community do not have the luxury of ignoring unsavory or crumbling states and the threats they can spawn for us or their own people. Disregarding Afghanistan during the 1990s has shown the peril of that approach.

This should put a premium on "preventive action" -- and not simply the sort aimed at enemies. This kind of preventive action should be aimed at preventing war, not relying on it. It needs to come earlier and discourage hostilities between states, try to prevent others from descending into chaos, and steer leaders away from self-destructive, repressive practices. There are many candidates now -- particularly in central Asia, the Caucasus and Africa -- where diplomacy, economic pressure and incentives might do some good. Unfortunately, leaders of democracies have shown little dedication to preventing conflict, the disintegration of states or genocide despite all their genuflections and apologies. That is because this sort of preventive action can exact short-term political costs in exchange for vague long-term goals and unmeasurable progress.

There isn't much choice, though. We can go on voicing ambitious objectives and watching them founder, or we can be realistic about our constraints and limitations and thus escape being immobilized by them.

Morton Abramowitz, former assistant secretary of state for intelligence and research and former president of the Carnegie Endowment for International Peace, is a senior fellow at the Century Foundation. Heather Hurlburt, a Michigan-based writer, wrote speeches about foreign policy for the Clinton administration and was deputy director of the International Crisis Group's Washington office.

Nation-building without end? In March, five years after the Kosovo war ended, U.N. peacekeepers were still needed to end clashes between ethnic Albanians and Serbs in the northern part of the province.