In the year since the United States Marines pulled down Saddam Hussein’s statue in Baghdad’s Firdos Square, things have gone very badly for the United States in Iraq and for its ambition of creating a model democracy that might transform the Middle East. As of today the United States military appears committed to an open-ended stay in a country where, with the exception of the Kurdish north, patience with the foreign occupation is running out, and violent opposition is spreading. Civil war and the breakup of Iraq are more likely outcomes than a successful transition to a pluralistic Western-style democracy.

Much of what went wrong was avoidable. Focused on winning the political battle to start a war, the Bush administration failed to anticipate the postwar chaos in Iraq. Administration strategy seems to have been based on a hope that Iraq’s bureaucrats and police would simply transfer their loyalty to the new authorities, and the country’s administration would continue to function. All experience in Iraq suggested that the collapse of civil authority was the most likely outcome, but there was no credible planning for this contingency. In fact, the US effort to remake Iraq never recovered from its confused start when it failed to prevent the looting of Baghdad in the early days of the occupation.

Americans like to think that every problem has a solution, but that may no longer be true in Iraq. Before dealing at considerable length with what has gone wrong, I should also say what has gone right.

Iraq is free from Saddam Hussein and the Baath Party. Along with Cambodia’s Pol Pot, Saddam Hussein’s regime was one of the two most cruel and inhumane regimes in the second half of the twentieth century. Using the definition of genocide specified in the 1948 Genocide Convention, Iraq’s Baath regime can be charged with planning and executing two genocides—one against the Kurdish population in the late 1980s and another against the Marsh Arabs in the 1990s. In the 1980s, the Iraqi armed forces and security services systematically destroyed more than four thousand Kurdish villages and several small cities, attacked over two hundred Kurdish villages and towns with chemical weapons in 1987 and 1988, and organized the deportation and execution of up to 182,000 Kurdish civilians.

In the 1990s the Saddam Hussein regime drained the marshes of southern Iraq, displacing 500,000 people, half of whom fled to Iran, and killing some 40,000. In addition to destroying the five-thousand-year-old Marsh Arab civilization, draining the marshes did vast ecological damage to one of the most important wetlands systems on the planet. Genocide is only part of Saddam Hussein’s murderous legacy. Tens of thousands perished in purges from 1979 on, and as many as 300,000 Shiites were killed in the six months following the collapse of the March 1991 Shiite uprising. One mass grave near Hilla may contain as many as 30,000 bodies.

In a more lawful world, the United Nations, or a coalition of willing states, would have removed this regime from power long before 2003. However, at precisely the time that some of the most horrendous crimes were being committed, in the late 1980s, the Reagan and Bush administrations strongly opposed any action to punish Iraq for its genocidal campaign against the Kurds or to deter Iraq from using chemical weapons against the Kurdish civilians.

On August 20, 1988, the Iran–Iraq War ended. Five days later, the Iraqi military initiated a series of chemical weapons attacks on at least forty-nine Kurdish villages in the Dihok Governorate (or province) near the Syrian and Turkish borders. As a staff member of the Senate Foreign Relations Committee, I (along with Chris Van Hollen, now a Maryland congressman) interviewed hundreds of survivors in the high mountains on the Turkish border. Our report, which established conclusively that Iraq had used nerve and mustard agents on tens of thousands of civilians, coincided with the Senate’s passage of the Prevention of Genocide Act of 1988, which imposed comprehensive economic sanctions on Iraq for crimes against the Kurds. The Reagan administration opposed the legislation, in a position orchestrated by the then national security adviser, Colin Powell, calling such sanctions “premature.”

Except for a relatively small number of Saddam Hussein’s fellow Sunni Arabs who worked for his regime, the peoples of Iraq are much better off today than they were under Saddam Hussein. The problems that threaten to tear Iraq apart—Kurdish aspirations for independence, Shiite dreams of dominance, Sunni Arab nostalgia for lost power—are not of America’s making (although the failure to act sooner against Saddam made them less solvable). Rather, they are inherent in an artificial state held together for eighty years primarily by brute force.


American liberation—and liberation it was—ended the brute force. Iraqis celebrated the dictatorship’s overthrow, and in Baghdad last April ordinary citizens thrust flowers into my hands. Since then, however:


• Hostile action has killed twice as many American troops as died in the war itself, while thousands of Iraqis have also died.

• Terrorists have killed the head of the United Nations Mission, Sergio Vieira de Mello; Iraq’s most prominent Shiite politician, the Ayatollah Baqir al-Hakim; and the deputy prime minister of the Kurdistan Regional Government, Sami Abdul Rahman, along with hundreds of others.

• Looting has caused billions of dollars of damage, most of which will have to be repaired at the expense of the US taxpayer.

• $150 billion has already been spent on Iraq, an amount equal to 25 percent of the non-defense discretionary federal budget. (By contrast, the first Gulf War earned a small profit for the US government, owing to the contributions of other nations.)

• Discontent with the US-led occupation boiled over into an uprising in the Shiite areas of Iraq on the first anniversary of liberation and a persistent insurgency in the Sunni Triangle degenerated into a full-scale battle in Fallujah. Many on the US-installed Iraqi Governing Council strongly opposed the US military response, and the US-created security institutions—the new Iraqi police and the paramilitary Iraqi Civil Defense Corps—refused to fight, or in some cases, joined the rebels.

• US credibility abroad has been undermined by the failure to find weapons of mass destruction. Spain’s elections, Tony Blair’s sinking poll results, and the prospective defeat of Australia’s Howard government underscore the political risk of too close an association with the United States.

• Relations with France and Germany have been badly hurt, in some cases by the gratuitous comments made by senior US officials.

• The United States does not now have the military or diplomatic resources to deal with far more serious threats to our national security. President Bush rightly identified the peril posed by the nexus between weapons of mass destruction and rogue states. The greatest danger comes from rogue states that acquire and disseminate nuclear weapons technology. At the beginning of 2003 Iraq posed no such danger. As a result of the Iraq war the United States has neither the resources nor the international support to cope effectively with the very serious nuclear threats that come from North Korea, Iran, and, most dangerous of all, our newly designated “major non-NATO ally,” Pakistan.

With fewer than one hundred days to the handover of power to a sovereign Iraq on June 30, there is no clear plan—and no decision—about how Iraq will be run on July 1, 2004. Earlier this month, the Bush administration praised itself generously for the signing of an interim constitution for Iraq—a constitution with human rights provisions it described as unprecedented for the Middle East. Three weeks later, as I write, the interim constitution is already falling apart.

As is true of so much of the US administration of postwar Iraq, the damage here is self-inflicted. While telling Iraqis it wanted to defer constitutional issues to an elected Iraqi body, the US-led Coalition Provisional Authority could not resist trying to settle fundamental constitutional issues in the interim constitution. The US government lawyers who wrote the interim constitution, known formally as the Transitional Administrative Law, made no effort to disguise their authorship. All deliberations on the law were done in secret and probably fewer than one hundred Iraqis saw a copy of the constitution before it was promulgated. To write a major law in any democracy—much less a constitution—without public discussion should be unthinkable. Now that Iraqis are discovering for the first time the contents of the constitution, it should come as no surprise that many object to provisions they never knew were being considered.

Iraq’s Shiite leaders say that the National Assembly due to be elected in January 2005 should not be constrained by a document prepared by US government lawyers, deliberated in secret, and signed by twenty-five Iraqis selected by Ambassador Bremer. In particular, the Shiites object to a provision in the interim constitution that allows three of Iraq’s eighteen governorates (or provinces) to veto ratification of a permanent constitution. This, in effect, allows either the Kurds or the Sunni Arabs, each of whom make up between one fifth and one sixth of Iraq’s population, to block a constitution they don’t like. (It is a wise provision. Imposing a constitution on reluctant Kurds or Sunni Arabs will provoke a new cycle of resistance and conflict.) The Shiite position makes the Kurds, who are well armed, reluctant to surrender powers to a central government that may be Shiite-dominated.

At the moment the Sunni Arabs have few identifiable leaders. The Kurds, however, are well organized. They have an elected parliament and two regional governments, their own court system, and a 100,000 strong military force, known as the Peshmerga. The Peshmerga, whose members were principal American allies in the 2003 war, are better armed, better trained, and more disciplined than the minuscule Iraqi army the United States is now trying to reconstitute.


Early in 2005, Iraq will likely see a clash between an elected Shiite-dominated central government trying to override the interim constitution in order to impose its will on the entire country, and a Kurdistan government insistent on preserving the de facto independent status Kurdistan has enjoyed for thirteen years. Complicating the political struggle is a bitter territorial dispute over the oil-rich province of Kirkuk involving Kurds, Sunni Arabs, Shiite Arabs, Sunni Turkmen, and Shiite Turkmen.

It is a formula for civil war.


How did we arrive at this state of affairs?

I arrived in Baghdad on April 13, 2003, as part of an ABC news team. It was apparent to me that things were already going catastrophically wrong. When the United States entered Baghdad on April 9 last year, it found a city largely undamaged by a carefully executed military campaign. However, in the two months following the US takeover, unchecked looting effectively gutted every important public institution in the city with the notable exception of the Oil Ministry. The physical losses include:

• The National Library, which was looted and burned. Equivalent to our Library of Congress, it held every book published in Iraq, all newspapers from the last century, as well as rare manuscripts. The destruction of the library meant the loss of a historical record going back to Ottoman times.

• The Iraqi National Museum, which was also looted. More than 10,000 objects were stolen or destroyed. The Pentagon has deliberately, and repeatedly, tried to minimize the damage by excluding from its estimates objects stolen from storage as well as displayed treasures that were smashed but not stolen.

• Hospitals and other public health institutions, where looters stole medical equipment, medicines, and even patients’ beds.

• Baghdad and Mosul Universities, which were stripped of computers, office furniture, and books. Academic research that took decades to carry out went up in smoke or was scattered.

• The National Theater, which was set ablaze by looters a full three weeks after US forces entered Baghdad.

Even more surprising, the United States made no apparent effort to secure sites that had been connected with Iraqi WMD programs or buildings alleged to hold important intelligence. As a result, the United States may well have lost valuable information that related to Iraqi WMD procurement, paramilitary resistance, foreign intelligence activities, and possible links to al-Qaeda.

• On April 16, looters attacked the Iraqi equivalent of the US Centers for Disease Control, stealing live HIV and live black fever bacteria. UNMOVIC and UNSCOM had long considered the building suspicious and had repeatedly conducted inspections there. The looting complicates efforts to understand and account for any Iraqi bioweapons research in the past. A Marine lieutenant watched the looting from next door. He told us, “I hope I am not responsible for Armageddon, but no one told me what was in that building.”

• Although US troops moved onto the grounds of Iraq’s sprawling Tuwaitha nuclear complex, they did not secure the warehouse that contained yellowcake and other radiological materials. Looters took materials that terrorists could use for a radiological weapon, although much of that material was eventually recovered. The looted nuclear materials were in a known location, and already had been placed under seal by the International Atomic Energy Commission.

• Ten days after the US took over Baghdad, I went through the unguarded Iraqi Foreign Ministry, going from the cooling unit on the roof to the archives in the basement, and rummaging through the office of the foreign minister. The only other people in the building were looters, who were busy opening safes and carrying out furniture. They were unarmed and helped me look for documents. Foreign Ministry files could have shed light on Iraqis’ overseas intelligence activities, on attempts to procure WMD, and on any connections that may have existed with al-Qaeda. However, we may never know about these things, since looters scattered and burned files during the ten days, or longer, that this building was left unguarded.

The looting demoralized Iraqi professionals, the very people the US looks to in rebuilding the country. University professors, government technocrats, doctors, and researchers all had connections with the looted institutions. Some saw the work of a lifetime quite literally go up in smoke. The looting also exacerbated other problems: the lack of electricity and potable water, the lack of telephones, and the absence of police or other security.

Most importantly, the looting served to undermine Iraqi confidence in, and respect for, the US occupation authorities.


In the parts of Iraq taken over by rebels during the March 1991 uprising, there had been the same kind of looting of public institutions. In 2003, the United States could not have prevented all the looting but it could have prevented much of it. In particular, it could have secured the most important Iraqi government ministries, hospitals, laboratories, and intelligence sites. It could have protected the Iraq National Museum and several other of Iraq’s most important cultural and historical sites.

In the spring of 2003, Thomas Warrick of the State Department’s Future of Iraq Working Group prepared a list of places in Baghdad to be secured. The Iraq National Museum was number two on the list. At the top of the list were the paper records of the previous regime—the very documents I found scattered throughout the Foreign Ministry and in other locations. What happened next is a mystery. My State Department informants tell me the list was sent to Douglas Feith, an undersecretary in the Department of Defense, and never came out of his office. Feith’s partisans insist that uniformed American military failed to take action. In either case, the lack of oversight was culpable.

During the war in Kosovo, the Clinton White House was criticized for insisting on presidential review of proposed targets. President Bush, notorious for his lack of curiosity, seems never to have asked even the most basic question: “What happens when we actually get to Baghdad?”

The failure to answer this question at the start set back US efforts in Iraq in such a way that the US has not recovered and may never do so.

The Bush administration decided that Iraq would be run by a US civilian administrator—initially, Retired General Jay Garner—and American advisers who would serve as the de facto ministers for each of the Iraqi government ministries. All this was based on the expectation that the war would decapitate the top leadership of the Saddam Hussein regime, and the next day everyone else would show up for work.

Predictably, this did not happen. In 1991, all authority disappeared in the areas that fell into rebel hands. But even had things gone as the Bush administration hoped, it was not prepared to run Iraq. As the war began, the Bush administration was still recruiting the American officials who would serve as the de facto Iraqi ministers. The people so recruited had no time to prepare for the assignment, either in learning about Iraq or in mastering the substantive skills needed to run the ministry assigned to them. Many mistakes were made. For example, the US official in charge of prisons decided to work with Ali al-Jabouri, the warden of Abu Ghraib prison, apparently unaware of the prison’s fearsome reputation as the place where tens of thousands perished under Saddam Hussein. The coalition rehabilitated Abu Ghraib and today uses it as a prison. The symbolism may be lost on the US administrators but it is not lost on Iraqis.

In late 2002 and early 2003, I attended meetings with senior US government officials on Kirkuk, the multi-ethnic city that is just west of the line marking the border of the self-governing Kurdish region. When Kirkuk, which is claimed by the Kurds, was held by Saddam Hussein, horrific human rights abuses had taken place there. I had been to Kirkuk in the 1980s, and I was concerned that Kurds brutally expelled in the 1980s and 1990s would return to settle scores with Arabs who had been settled in their homes. The week the war began, I asked the US official responsible for Kirkuk how he planned to deal with this problem. We will rely on the local police, he explained. I asked whether the local police were Kurds or Arabs. He did not know. It remains astonishing to me that US plans for dealing with ethnic conflict in the most volatile city inall of Iraq rested on hopes about the behavior of a police force about which they did not have the most basic information.

The Kirkuk police were, in fact, Arabs, and had assisted in the ethnic cleansing of the city’s Kurds. They were not around when Kurdish forces entered the city on April 10, 2003. Many other Arabs also fled, although this was largely ignored by the international press.

The United States’ political strategies in Iraq have been no less incoherent. General Garner arrived announcing that he would quickly turn power over to a provisional Iraqi government. Within three weeks Ambassador Bremer and a new structure, the Coalition Provisional Authority (CPA), replaced him. US officials indicated that Iraqi participation would be limited to an advisory council and that the United States expected to stay in Iraq for up to three years. The US would write a democratic constitution for the country and then turn power over to an elected government. After a few weeks, Bremer changed course and announced he was sharing power with a representative Iraqi governing council. In November, as Bush’s poll numbers plummeted, Bremer was summoned back to Washington to discuss a new strategy. The United States, it was decided, would turn power over on June 30, 2004, to a sovereign Iraqi government that would be chosen in a complicated system of caucuses held in each of Iraq’s “governorates (or provinces).” By January this plan was put aside (it was widely described as “election by people selected by people selected by Bremer”).

The latest strategy—based on the interim constitution and a takeover of sovereignty on June 30 by an as yet undetermined body—the fifth in a year by my count, is now falling apart in the face of Shiite opposition and mounting violence.

The Bush administration’s strategies in Iraq are failing for many reasons. First, they are being made up as the administration goes along, without benefit of planning, adequate knowledge of the country, or the experience of comparable situations. Second, the administration has been unwilling to sustain a commitment to a particular strategy. But third, the strategies are all based on an idea of an Iraq that does not exist.


The fundamental problem of Iraq is an absence of Iraqis.

In the north the Kurds prefer almost unanimously not to be part of Iraq, for reasons that are very understandable. Kurdistan’s eighty-year association with Iraq has been one of repression and conflict, of which the Saddam Hussein regime was the most brutal phase. Since 1991, Kurdistan has been de facto independent and most Iraqi Kurds see this period as a golden era of democratic self-government and economic progress. In 1992 Kurdistan had the only democratic elections in the history of Iraq, when voters chose members of a newly created Kurdistan National Assembly. During the last twelve years the Kurdistan Regional Government built three thousand schools (as compared to one thousand in the region in 1991), opened two universities, and permitted a free press; there are now scores of Kurdish-language publications, radio stations, and television stations. For the older generation, Iraq is a bad memory, while a younger generation, which largely does not speak Arabic, has no sense of being Iraqi.

The people of Kurdistan almost unanimously prefer independence to being part of Iraq. In just one month, starting on January 25 of this year, Kur- dish nongovernmental organizations collected 1,700,000 signatures on petitions demanding a vote on whether Kurdistan should remain part of Iraq. This is a staggering figure, representing as it does roughly two thirds of Kurdistan’s adults.

In the south, Iraq’s long-repressed Shiites express themselves primarily through their religious identity. In early March I traveled throughout southern Iraq. I saw no evidence of any support for secular parties. If free elections are held in Iraq, I think it likely that the Shiite religious parties—principally the Supreme Council for Islamic Revolution in Iraq (SCIRI) and the Dawa (the Call)—will have among them an absolute majority in the National Assembly.

The wild card is Moqtada al-Sadr, the leader of the Shiite uprising. If he is allowed to compete in elections, he will certainly take a share of the Shiite vote. If he is excluded (or imprisoned or killed), his supporters will likely influence the policies of the mainstream Shiite parties, or conceivably disrupt the elections. None of this is good for hopes of creating a stable, democratic Iraq.

The Shiites are not separatists but many of them believe their majority status entitles them to run all of Iraq, and to impose their version of an Islamic state. They also consider connections with Shiites elsewhere as important as their nationalist feelings about Iraq. Iranian Shiites, such as the Ayatollah al-Sistani and, from the grave, Ayatollah Khomeini, have enormous political and spiritual influence in southern Iraq. Their portraits are ubiquitous. Mainstream Iraqi Arab Shiites, such as SCIRI’s leader Abdel Azziz al-Hakim, often advocate a very pro-Iranian line.

Sunni Arabs have always been the principal Iraqi nationalists, and a part of the anti-US uprising in the Sunni Triangle is a nationalist one. The Sunni Arabs have long been accustomed to seeing the Iraqi state as a part of a larger Arab nation, and this was a central tenet of the Baath Party. As Sunni Arabs face the end of their historic domination of Iraq, they may seek to compensate for their minor- ity status inside Iraq by further identifying themselves with the greater Arab nation. Connections with other Sunni populations may eventually become even more important among the Sunni Arabs than pan-Arabism. As elsewhere in Arab Iraq, the Sunni religious parties appear to be gaining ground in the country’s Sunni center at the expense of the secular parties.

Radical Sunni Islamic groups, including those with recent links to al-Qaeda, appear to have an ever more important part in the uprising in the Sunni Triangle (which explains the increasing use of suicide bombers, not a tactic that appeals to the more worldly Baathists). By attacking Shiite religious leaders and celebrations (for example the deadly bombings this March during the as-Shoura religious holiday in Baghdad and Karbala, and the car bomb assassination of SCIRI leader Baqir al-Hakim), Sunni extremists seek to provoke civil war between Iraq’s two main religious groups.


The United States strategy is to hold Iraq together by establishing a strong central government. So far, all its successes have been on paper. The interim constitution gives the central government a monopoly on military force, control over natural resources, broad fiscal powers, and oversight over the judiciary.

Little of this will come to pass. The Kurdistan National Assembly has put forward a comprehensive proposal to define Kurdistan’s relations with the rest of Iraq. In it the Kurdistan National Assembly retains lawmaking power for the region, preserves its fiscal autonomy, and would eventually own the region’s natural resources. Kurdistan will retain the Peshmerga (which would be converted into an Iraqi Kurdistan National Guard nominally under the overall authority of the Iraq central government) and other Iraqi armed forces could only enter Kurdistan with the consent of the Kurdistan National Assembly. Iraq would be fully bilingual (Arabic and Kurdish) and Kurdistan would remain secular.

This places the Kurds on a collision course with the Shiites and the Sunni Arabs. The Shiite religious parties insist that Islam must be the principal source of law throughout Iraq. Both Shiites and Sunni Arabs object to downgrading Arabic to one of two official languages. Sunni Arab nationalists and Shiite religious leaders object to Kurdistan retaining even a fraction of the autonomy it has today.

There are also acute conflicts between Shiite Arabs and Sunni Arabs. These have to do with the differing interpretations of Islam held by the two groups’ religious parties and conflicts between pro-Iranian Shiites and Arab nationalist Sunnis.

Shiites are now providing moral and material support for the Sunni insurgents in Fallujah. An anti-American alliance of radicals from both confessions will not necessarily lead to political unity, nor will it erase Sunni fears of Shiite domination. That said, the confessional divide between Iraq’s Arabs is far less than the ethnic gulf between Arabs and Kurds. Democracy requires tolerance and a willingness to compromise. Except tactically, none of these traits is apparent in a political culture (except for the north) which has been ruled by absolutists.

In my view, Iraq is not salvageable as a unitary state. From my experience in the Balkans, I feel strongly that it is impossible to preserve the unity of a democratic state where people in a geographically defined region almost unanimously do not want to be part of that state. I have never met an Iraqi Kurd who preferred membership in Iraq if independence were a realistic possibility.

But the problem of Iraq is that a breakup of the country is not a realistic possibility for the present. Turkey, Iran, and Syria, all of which have substantial Kurdish populations, fear the precedent that would be set if Iraqi Kurdistan became independent. Both Sunni and Shiite Arabs oppose the separation of Kurdistan. The Sunni Arabs do not have the resources to support an independent state of their own. (Iraq’s largest oil fields are in the Shiite south or in the disputed territory of Kirkuk.)

Further, as was true in the Balkans, the unresolved territorial issues in Iraq would likely mean violent conflict. Kirkuk is perhaps the most explosive place. The Kurds claim it as part of historic Kurdistan. They demand that the process of Arabization of the region—which some say goes back to the 1950s—should be reversed. The Kurds who were driven out of Kirkuk by policies of successive Iraqi regimes should, they say, return home, while Arab settlers in the region are repatriated to other parts of Iraq. While many Iraqi Arabs concede that the Kurds suffered an injustice, they also say that the human cost of correcting it is too high. Moreover, backed by Turkey, ethnic Turkmen assert that Kirkuk is a Turkmen city and that they should enjoy the same status as the Kurds.

It will be difficult to resolve the status of Kirkuk within a single Iraq; it will be impossible if the country breaks up into two or three units. And while Kirkuk is the most contentious of the territories in dispute, it is only one of many.

The best hope for holding Iraq together—and thereby avoiding civil war—is to let each of its major constituent communities have, to the extent possible, the system each wants. This, too, suggests the only policy that can get American forces out of Iraq.

In the north this means accepting that Kurdistan will continue to govern its own affairs and retain responsibility for its own security. US officials have portrayed a separate Kurdistan defense force as the first step leading to the breakup of Iraq. The Kurds, however, see such a force not as an attribute of a sovereign state but as insurance in case democracy fails in the rest of Iraq. No one in Kurdistan would trust an Iraqi national army (even one in which the Kurds were well represented) since the Iraqi army has always been an agent of repression, and in the 1980s, of genocide. The Kurds also see clearly how ineffective are the new security institutions created by the Americans. In the face of uprisings in the Sunni Triangle and the south, the new Iraqi police and civil defense corps simply vanished.

Efforts to push the Kurds into a more unitary Iraq will fail because there is no force, aside from the US military, that can coerce them. Trying to do so will certainly inflame popular demands for separation of the Kurdish region in advance of January’s elections.

If Kurdistan feels secure, it is in fact more likely to see advantages to cooperation with other parts of Iraq. Iraq’s vast resources and the benefits that would accrue to Kurdistan from revenue sharing provide significant incentives for Kurdistan to remain part of Iraq, provided doing so does not open the way to new repression. (Until now, most Iraqi Kurds have seen Iraq’s oil wealth as a curse that gave Saddam the financial resources to destroy Kurdistan.)

In the south, Iraq’s Shiites want an Islamic state. They are sufficiently confident of public support that they are pushing for early elections. The United States should let them have their elections, and be prepared to accept an Islamic state—but only in the south. In most of the south, Shiite religious leaders already exercise actual power, having established a degree of security, taken over education, and helped to provide municipal services. In the preparation of Iraq’s interim constitution, Shiite leaders asked for (and obtained) the right to form one or two Shiite regions with powers comparable to those of Kurdistan. They also strongly support the idea that petroleum should be owned by the respective regions, which is hardly surprising since Iraq’s largest oil reserves are in the south.

There is, of course, a logical inconsistency between Shiite demands to control a southern region and the desire to impose Islamic rule on all of Iraq. Meeting the first demand affects only the south; accepting the second is an invitation to civil war and must be resisted.

Federalism—or even confederation—would make Kurdistan and the south governable because there are responsible parties there who can take over government functions. It is much more difficult to devise a strategy for the Sunni Triangle—until recently the location of most violent resistance to the American occupation—because there is no Sunni Arab leadership with discernible political support. While it is difficult to assess popular support for the insurrection within the Sunni Triangle, it is crystal clear that few Sunni Arabs in places like Fallujah are willing to risk their lives in opposing the insurgents.

We can hope that if the Sunni Arabs feel more secure about their place in Iraq with respect to the Shiites and the Kurds, they will be relatively more moderate. Autonomy for the Sunni Arab parts of Iraq is a way to provide such security. There is, however, no way to know if it will work.

Since 1992, the Iraqi opposition has supported federalism as the system of government for a post-Saddam Iraq. Iraq’s interim constitution reflects this consensus by defining Iraq as a federal state. There is, however, no agreement among the Iraqi parties on what federalism actually means, and the structures created by the interim constitution seem unlikely to move from paper to reality.

Last November, Les Gelb, the president emeritus of the Council on Foreign Relations, created a stir by proposing, in a New York Times Op-Ed piece, a three-state solution for Iraq, modeled on the constitution of post-Tito Yugoslavia. The Yugoslav model would give each of Iraq’s constituent peoples their own republic.* These republics would be self-governing, financially self-sustaining, and with their own territorial military and police forces. The central government would have a weak presidency rotating among the republics, with responsibilities limited to foreign affairs, monetary policy, and some coordination of defense policy. While resources would be owned by the republics, some sharing of oil revenues would be essential, since an impoverished Sunni region is in no one’s interest.

This model would solve many of the contradictions of modern Iraq. The Shiites could have their Islamic republic, while the Kurds could continue their secular traditions. Alcohol would continue to be a staple of Kurdish picnics while it would be strictly banned in Basra.

The three-state solution would permit the United States to disengage from security duties in most of Iraq. There are today fewer than three hundred coalition troops in Kurdistan, which would, under the proposal being made here, continue to be responsible for its own security. By contrast, introducing an Iraqi army and security institutions into Kurdistan, as the Bush administration says it still wants to do, would require many more coalition troops—because the Iraqi forces are not up to the job and because coalition troops will be needed to reassure a nervous Kurdish population. If the United States wanted to stay militarily in Iraq, Kurdistan is the place; Kurdish leaders have said they would like to see permanent US bases in Kurdistan.

A self-governing Shiite republic could also run its own affairs and provide for its own security. It is not likely to endorse Western values, but if the coalition quickly disengages from the south, this may mean the south would be less overtly anti-American. Staying in the south will play directly into the hands of Moqtada al-Sadr or his successors. Moderate Shiite leaders, including the Ayatollah al-Sistani, counseled patience in response to al-Sadr’s uprising, and helped negotiate the withdrawal of al-Sadr’s supporters from some police stations and government buildings. The scope of the uprising, however, underscores the coalition’s perilous position in the south. The failure of the Iraqi police and the Iraqi Civil Defense Corps to respond highlights the impotence of these American-created security institutions. The sooner power in the south is handed over to people who can exercise it, the better. Delay will only benefit anti-American radicals like al-Sadr.

As for the Sunni Triangle, one hope is for elections to produce a set of leaders who can restore order and end the insurrection. Presumably this is an outcome the Sunni rebels do not want to see happen; they will use violence to prevent a meaningful election in large parts of the Sunni Triangle. In these circumstances, the United States may face the choice of turning power over to weak leaders and living with the resulting chaos, or continuing to try to pacify the Sunni Triangle, which may generate ever more support for the insurrection. There may be no good options for the United States in the Sunni Triangle. Nevertheless the three-state approach could limit US military engagement to a finite area.

Baghdad is a city of five million and home to large numbers of all three of Iraq’s major constituent peoples. With skilled diplomacy, the United States or the United Nations might be able to arrange for a more liberal regime in Baghdad than would exist in the south. Kurdish and Shiite armed forces and police could provide security in their own sections of the capital, as well as work together in Sunni areas (with whatever local cooperation is possible) and in mixed areas. Such an arrangement in Iraq’s capital is far from ideal, but it is better than an open-ended US commitment to being the police force of last resort in Iraq’s capital.

Because of what happened to Yugoslavia in the 1990s, many react with horror to the idea of applying its model to Iraq. Yet Yugoslavia’s breakup was not inevitable. In the 1980s, Slovenia asked for greater control over its own affairs and Milosevic refused. Had Milosevic accepted a looser federation, there is every reason to think that Yugoslavia—and not just Slovenia—would be joining the European Union this May.

Still, a loose federation will have many drawbacks, especially for those who dreamed of a democratic Iraq that would transform the Middle East. The country would remain whole more in name than in reality. Western- style human rights are likely to take hold only in the Kurdish north (and even there not completely). Women’s rights could be set back in the south, and perhaps also in Baghdad.

In administering elections and allowing a federation to emerge, the US would badly need the help of the UN and other international organizations and, if it can get it, of the principal European nations as well. The alternative is an indefinite US occupation of Iraq in which we have fewer and fewer allies. It is an occupation that the US cannot afford. It also prevents the US from addressing more serious threats to its national security.


The American involvement in Iraq will be a defining event for the US role in the world for the coming decades. Will it be seen as validating the Bush administration’s doctrines of preventive war and largely unilateral action?

In my view, Iraq demonstrates all too clearly the folly of the preventive war doctrine and of unilateralism. Of course the United States must reserve the right to act alone when the country is under attack or in imminent danger of attack. But these are also precisely the circumstances when the United States does not need to act alone. After September 11 both NATO and the United Nations Security Council gave unqualified support for US action, including military action, to deal with the threat of international terrorists based in Afghanistan. After the Taliban was defeated, other countries contributed troops—and accepted casualties—in order to help stabilize the country; and they have also contributed billions to Afghanistan’s reconstruction. Because the US so quickly diverted its attention to Iraq, many acute problems remain in Afghanistan, including warlordism and the deprivation of basic rights. International support for helping Afghanistan remains strong, however, and the effort can be revitalized with a new administration.

In Iraq the United States chose to act without the authorization of the Security Council, without the support of NATO, and with only a handful of allies. Aside from the British and the Kurdish Peshmerga, no other ally made any significant contribution to the war effort. The United States is paying practically all the expenses of the Iraq occupation. Even those who supported the unilateral intervention in Iraq seem by now to realize that it cannot be sustained. The Bush administration, having scorned the United Nations, is now desperate to have it back.

It turns out that there are some things that only the United Nations can do—such as run an election that Iraqis will see as credible or give a stamp of legitimacy to a political transition. But the most urgent reason to want United Nations participation is to share the burden. Internationalization is a key element of John Kerry’s program for Iraq. Unfortunately, it is a far from easy policy to achieve. While a less confrontational US administration would certainly be able to win greater international support and contributions, it will be a challenge to persuade the major European countries to have either the United Nations or NATO take over the major responsibilities in Iraq.

The reason is cost. Taking all expenses into account, one year of involvement in Iraq costs between $50 billion and $100 billion. Under the mandatory assessment scale for the United Nations this would cost France and Germany some $5 billion to $10 billion each, and they would face pressure to put their own troops in harm’s way. NATO assessments are similarly costly. While our allies may wish a Kerry administration well, they may not be willing to commit resources on this scale to help the United States get out of Iraq. As a European diplomat told me before last year’s war, “It will be china shop rules in Iraq: you break it, you pay for it.”

I believe United States policy is most successful when it follows international law and works within the United Nations, according to the provisions of the Charter. This is not just a matter of upholding the ideals of the UN; it is also practical. As our war in Iraq demonstrates, we cannot afford any other course.

—April 15, 2004

This Issue

May 13, 2004