The Price of the Surge

In January 2007, President George W. Bush announced a new approach to the war in Iraq. At the time, sectarian and insurgent violence appeared to be spiraling out of control, and Democrats in Washington -- newly in control of both houses of Congress -- were demanding that the administration start winding down the war. Bush knew he needed to change course, but he refused to, as he put it, "give up the goal of winning." So rather than acquiesce to calls for withdrawal, he decided to ramp up U.S. efforts. With a "surge" in troops, a new emphasis on counterinsurgency strategy, and new commanders overseeing that strategy, Bush declared, the deteriorating situation could be turned around.

More than a year on, a growing conventional wisdom holds that the surge has paid off handsomely. U.S. casualties are down significantly from their peak in mid-2007, the level of violence in Iraq is lower than at any point since 2005, and Baghdad seems the safest it has been since the fall of Saddam Hussein's regime five years ago. Some backers of the surge even argue that the Iraqi civil war is over and that victory on Washington's terms is in sight -- so long as the United States has the will to see its current efforts through to their conclusion.

Unfortunately, such claims misconstrue the causes of the recent fall in violence and, more important, ignore a fatal flaw in the strategy. The surge has changed the situation not by itself but only in conjunction with several other developments: the grim successes of ethnic cleansing, the tactical quiescence of the Shiite militias, and a series of deals between U.S. forces and Sunni tribes that constitute a new bottom-up approach to pacifying Iraq. The problem is that this strategy to reduce violence is not linked to any sustainable plan for building a viable Iraqi state. If anything, it has made such an outcome less likely, by stoking the revanchist fantasies of Sunni Arab tribes and pitting them against the central government and against one another. In other words, the recent short-term gains have come at the expense of the long-term goal of a stable, unitary Iraq.

Despite the current lull in violence, Washington needs to shift from a unilateral bottom-up surge strategy to a policy that promotes, rather than undermines, Iraq's cohesion. That means establishing an effective multilateral process to spur top-down political reconciliation among the major Iraqi factions. And that, in turn, means stating firmly and clearly that most U.S. forces will be withdrawn from Iraq within two or three years. Otherwise, a strategy adopted for near-term advantage by a frustrated administration will only increase the likelihood of long-term debacle.

THE SURGE'S FALSE START

After the February 2006 bombing of the Askariya shrine in Samarra, the White House started to become increasingly concerned that there were too few U.S. troops in Iraq. A network of retired army officers led by Jack Keane, a former vice chief of staff of the U.S. Army, had been pushing from the outside for an increase in forces, and Senators John McCain (R-Ariz.) and Lindsey Graham (R-S.C.) kept up a drumbeat of criticism of what they saw as a lackluster military effort. The November 2006 congressional elections, which handed the House and the Senate to the Democrats, added to the sense that a new strategy was needed. In a December 2006 memo, Bush's national security adviser, Stephen Hadley, somewhat gingerly noted that the United States might "need to fill the current four-brigade gap in Baghdad with coalition forces if reliable Iraqi forces are not identified."

On December 13, 2006, Bush met with the Joint Chiefs of Staff at the Pentagon to persuade them to allocate more troops to Iraq. It was not an easy sell. U.S. ground forces are not configured to fight such a long war, and the repeated deployment of the same active-duty and Reserve units had taken a toll. The reenlistment rate of young captains, for example, had fallen to an unprecedented low; about half of the West Point classes of 2000 and 2001 had decided against an army career. The pace of unit rotations and the tempo of operations had also taken their toll on equipment, which was wearing out at nine times the normal rate, faster than it could be replaced. The chairman of the Joint Chiefs of Staff made clear his concern about the army being stretched too thin. A shortfall of 10,000 company-grade officers meant that the Reserve units would have to rob both people and materiel from other units. Meanwhile, the mounting expense of the war was crowding out the procurement of new combat systems for the navy and the air force, and there was a growing risk that the military might find itself without the capacity to meet other strategic challenges, whether from Afghanistan, Iran, or elsewhere.

Bush tried to allay these worries, pledging to, among other things, increase the size of the U.S. Army and the Marine Corps and boost defense spending. But the Joint Chiefs also conditioned their reluctant support of the surge on a promise from the president to hold Iraqi Prime Minister Nouri al-Maliki's feet to the fire on political reconciliation. So when Bush unveiled his surge strategy in January 2007 (the deployment of an additional 21,500 troops, through September, with the initial military objective of restoring order to Baghdad), the stated purpose was to ensure that "the [Iraqi] government will have the breathing space it needs to make progress in other critical areas. Most of Iraq's Sunni and Shia want to live together in peace -- and reducing the violence in Baghdad will help make reconciliation possible." Bush quoted Maliki's promise that the Baghdad security plan would "not provide a safe haven for any outlaws, regardless of their sectarian or political affiliation."

Even then, however, the administration was already starting to doubt Maliki's competence and willingness to pursue reconciliation, the principal determinant of long-term stability in Iraq. Two months earlier, Hadley had visited Iraq to assess the prospects for a cross-sectarian political rapprochement and come away unsure of Maliki's stance. "Do we and Prime Minister Maliki," Hadley had wondered in his December 2006 memo, "share the same vision for Iraq? If so, is he able to curb those who seek Shia hegemony or the reassertion of Sunni power? The answers to these questions are key in determining whether we have the right strategy in Iraq." Hadley proposed several ways to test Maliki's intentions and bolster his resolve, including initiatives to rejigger parliamentary support to free Maliki from his Shiite base linked to Muqtada al-Sadr and enable him to take conciliatory steps toward the Sunnis. The United States, however, lacked the influence necessary to put this approach into practice. Before long, events in Iraq revealed the answers to Hadley's questions: in both cases, a resounding no.

The deployment of the five new brigades proceeded more or less as planned, but from the start there was little headway made toward the broader goals of the surge, particularly reconciliation, as measured by the Iraqi government's inability to meet key benchmarks. The Constitutional Review Committee, which was charged with redressing Sunni grievances, made little progress, and there was no progress on de-Baathification reform, amnesty, provincial elections, or the implementation of oil legislation. The Sunni Iraqi Accordance Front had walked away from Maliki's cabinet, and Bush's reportedly regular calls to Maliki urging him to mobilize his government were ineffective. The Iraqi committees created to support the Baghdad security plan were left unfilled, and the three Iraqi brigades needed to help implement it arrived late and understrength. Diplomatic efforts to get Iraq's neighbors involved fizzled.

FROM TOP DOWN TO BOTTOM UP

The president's hopes for the top-down political efforts that were supposed to accompany the surge quickly faded. As a substitute, however, a new bottom-up strategy was embraced. Bush had observed in his January surge speech that the Sunnis were challenging al Qaeda's presence in Iraq, and a February 2007 National Intelligence Estimate on Iraq recommended "deputizing, resourcing, and working more directly with neighborhood watch groups and establishing grievance committees -- to help mend frayed relationships between tribal and religious groups, which have been mobilized into communal warfare over the past three years." A few months later, the president signaled a formal shift in strategy in a speech at the Naval War College: "To evaluate how life is improving for the Iraqis, we cannot look at the country only from the top down. We need to go beyond the Green Zone and look at Iraq from bottom up. This is where political reconciliation matters the most, because it is where ordinary Iraqis are deciding whether to support new Iraq or to sit on the fence, uncertain about the country's future." What the president was proposing was a shift in the U.S. approach to counterinsurgency. Now, the United States would work to exploit a grass-roots anti-al Qaeda movement already under way by taking the pressure off the insurgents who had begun to point their weapons at the jihadists and funneling money to tribal leaders. In theory, this would help dismantle the jihadist infrastructure and create islands of stability that would eventually join up like "oil spots."

After the U.S. invasion, the Sunni groups that would go on to make up the insurgency arrived at a marriage of convenience with the foreign and local jihadists who made up al Qaeda in Iraq. The two shared a common goal: to reverse the triumph of the Shiites and restore the Sunnis to their lost position of power. For the Sunni insurgents, the presence of foreign jihadists also helped divert the attention of U.S. forces. Up to a point, therefore, al Qaeda's excesses -- such as its attempt to impose strict Wahhabi-style rule by banning music and satellite dishes and compelling women to cover themselves entirely -- were to be tolerated.

But for al Qaeda, the link with the insurgents was supposed to serve two additional purposes that went well beyond the shared goal of chipping away at Shiite predominance -- and ultimately went against the interests of the Iraqi Sunnis themselves. The first was to establish an al Qaeda-dominated ministate as a base for carrying out jihad against enemies outside of Iraq. (The November 2005 attack against three Western tourist hotels in Amman, Jordan, allegedly ordered by Abu Musab al-Zarqawi, then the leader of al Qaeda in Iraq, was a harbinger of this wider strategy.) The second was to seize a leading position within the insurgency and thereby block a power-sharing arrangement between Baghdad and the Sunni nationalists, an arrangement that would entail the selling out of al Qaeda by the Sunnis.

The Iraqi Sunnis' enthusiasm for the alliance waned as al Qaeda increasingly attempted to assert its leadership. In October 2006, al Qaeda declared the formation of an Islamic state in Iraq, demanding that Sunni insurgent leaders pledge allegiance to the new (and many believed fictional) jihadist commander Abu Omar al-Baghdadi, whose name was supposed to signify an authentically Iraqi origin. To the nationalist insurgents, accepting the declaration of a separate state and ceding leadership to al Qaeda made little sense. Doing so would have fueled the process of decentralization, emboldened those Kurds and Shiites who sought their own fiefdoms, and, crucially, further distanced the Sunnis from eventual access to Iraq's potentially massive oil revenues. Moreover, despite the spectacular successes that had been attributed to al Qaeda, it was the nationalist Sunnis who provided the backbone of the insurgency and had done most of the killing and dying.

Some tribes had also grown increasingly resentful of al Qaeda's efforts to seize control of resources. The Albu Risha tribe, for example, had lost control over portions of the road from Baghdad to Amman, undermining its ability to raise revenue by taxing or extorting traders and travelers. When the Albu Rishas' leaders protested, the chieftain, Sheik Bazi al-Rishawi, was killed along with one of his sons, and two more of his sons were abducted. In response, Rishawi's fourth son, Sheik Abdul Sattar, assembled a small group of tribal figures (with the help of funds from the local U.S. military commander) under the banner of the Anbar Salvation Council to roll back al Qaeda's influence. The bodies of al Qaeda personnel soon began turning up in alleyways.

This strategic schism might have been papered over had the jihadists not overreacted to the opposition of other insurgent groups. In 2007, there was a wave of sensational killings of Sunni leaders by al Qaeda, including Abdul Sattar (who had met with President Bush two weeks before his death). The assassinations of Sunni leaders warranted retaliation under the prevailing tribal code, opening the door to more systematic cooperation between the tribes and U.S. forces. In the wake of Abdul Sattar's death, a Sunni leader complained that al Qaeda's assassinations had "left resistance groups with two options: either to fight al Qaeda and negotiate with the Americans or fight the Americans and join the Islamic State of Iraq, which divides Iraq. Both options are bitter." After their defeat in the battle of Baghdad -- thanks to the entrenched power of Sadr's Shiite Mahdi Army and the arrival of additional U.S. troops -- the Iraqi Sunnis went decisively with the first option, marking the start of the Sunni Awakening groups. The United States, for its part, had its own incentive to cooperate with the insurgents: June 2007, with 126 troop deaths, was the second-worst month for the U.S. military in Iraq, and General David Petraeus, the U.S. ground commander, was facing pressure to reduce casualties quickly. The most efficient way to do so was to strike deals with the newly pliable insurgents.

The deals were mediated by tribal leaders and consisted of payments of $360 per month per combatant in exchange for allegiance and cooperation. Initially referred to by the United States as "concerned local citizens," the former insurgents are now known as the Sons of Iraq. The total number across Iraq is estimated at over 90,000. Although the insurgents turned allies generally come well armed, at least one unit leader, Abu al-Abd, commander of the Islamic Army in Iraq, who controls Sunni neighborhoods in Baghdad, has said that he receives weapons as well as logistical support from U.S. units. His arrangement is probably typical. In November 2007, he agreed to a three-month pact, open to extension.

This strategy has combined with other developments -- especially the fact that so much ethnic cleansing has already occurred and that violence in civil wars tends to ebb and flow, as the contending sides work to consolidate gains and replenish losses -- to bring about the current drop in violence. The Sunni sheiks, meanwhile, are getting rich from the surge. The United States has budgeted $150 million to pay Sunni tribal groups this year, and the sheiks take as much as 20 percent of every payment to a former insurgent -- which means that commanding 200 fighters can be worth well over a hundred thousand dollars a year for a tribal chief. Although Washington hopes that Baghdad will eventually integrate most former insurgents into the Iraqi state security services, there are reasons to worry that the Sunni chiefs will not willingly give up what has become an extremely lucrative arrangement.

TRIBAL REALITIES

The surge may have brought transitory successes -- although if the spate of attacks in February is any indication, the decrease in violence may already be over -- but it has done so by stoking the three forces that have traditionally threatened the stability of Middle Eastern states: tribalism, warlordism, and sectarianism. States that have failed to control these forces have ultimately become ungovernable, and this is the fate for which the surge is preparing Iraq. A strategy intended to reduce casualties in the short term will ineluctably weaken the prospects for Iraq's cohesion over the long run.

Since the mid-nineteenth century, ruling powers in the Middle East have slowly and haltingly labored to bring tribal populations into the fold, with mixed success. Where tribes and tribalism have remained powerful, the state has remained weak. The Ottomans attempted forced sedentarization of the tribes, weakening tribal authorities by disrupting settlement patterns and replacing tribal sheiks with smaller cadres of favored leaders who became conduits for patronage. The colonial powers after World War I faced a different problem: the threat of nationalist urban elites opposed to foreign rule. In an effort to counter defiant urban leaders, they empowered rural tribes on the periphery. In Iraq, the British armed the tribes so that the sheiks could maintain order in the countryside and balance the capabilities of the nominal local governments operating under League of Nations mandates. Thus, the tribal system that Ottoman rule sought to dismantle was revitalized by British imperial policy, and the power of the nominal Iraqi government was systematically vitiated. In 1933, Iraq's King Faisal lamented, "In this kingdom, there are more than 100,000 rifles, whereas the government has only 15,000."

The tribes lost some power over the subsequent decades. This was in part a result of increasing direct British involvement in activities such as law enforcement, land tenure, and water distribution and in part a result of urbanization: as Iraqis moved from the country to the city, their affiliations shifted from the tribe to urban institutions -- principally the trade union and the mosque -- even as they held on to tribal symbols. When the Baathists took power in 1968, they explicitly rejected "religious sectarianism, racism, and tribalism ... the remnants of colonialism." The tribes, in their minds, were inevitable rivals of a centralizing state. But after taking control in a coup in 1979, Saddam leaned on his own Sunni tribal networks to staff his security services, army leadership, and bureaucracy, while suppressing other tribal life. He tried to rein in tribes by dispersing Baathist apparatchiks throughout the hinterland, but he nonetheless came to rely on the tribal system as a whole to make up for the shortcomings of the state as times became harder.

During the Iran-Iraq War, Saddam used Shiite tribes to defend regions near the Iranian border, and elsewhere tribal leaders regained some of their traditional authority as the war forced the redeployment of Baathist officials to the front. Amid the hardships created by the conflict, the flow of resources from the center shrank, leading to greater self-reliance in tribal areas and the renewed importance of tribal leaders. The Gulf War, and the grinding international sanctions that followed, accelerated these trends. In 1996, a high council of tribal chiefs was established and was granted political privilege, weapons, and land. Selected tribal leaders were allowed to enrich themselves by any means, fair or foul, and in return they were expected to defend the regime. Saddam, in effect, fostered a process of retribalization in Iraq.

Iraq's Arab neighbors, particularly Jordan and Saudi Arabia, provide a counterexample. They won enduring stability by corralling the tribes through a combination of reward and punishment. In Transjordan, King Abdullah I and the British -- helped by famine and the effects of the Great Depression -- confronted recalcitrant tribes militarily and then secured their allegiance with a steady flow of resources from the emerging state. More recently, Jordan's Hashemite monarchy has preserved the tribes' loyalty by guaranteeing them prestigious positions in the government and the military and by playing them off against the Palestinians. In Saudi Arabia, the al Saud dynasty consolidated its state by subduing the tribal challenge of rebellious Ikhwan and then endowing them with status and a military role. Strategic marriages between the al Saud family and the tribes cemented these ties. Although such efforts occasionally faltered, the thrust of the policy was always clear: to subordinate the tribes to the state.

Now, U.S. strategy is violating this principle by fostering the retribalization of Iraq all over again. In other countries in the region, such as Yemen, the result of allowing tribes to contest state authority is clear: a dysfunctional country prone to bouts of serious internecine violence. Such violence can also cross borders, especially if neighboring states are willing to use the tribes as their own agents. Pakistan provides a particularly ominous example of this dysfunctionality: its failure to absorb its Pashtun population has threatened the viability of the Pakistani state. The continued nurturing of tribalism in Iraq, in a way that sustains tribes in opposition to the central government rather than folding them into it, will bring about an Iraqi state that suffers from the same instability and violence as Yemen and Pakistan.

U.S. officials in Iraq have taken note of how the current U.S. approach has exacerbated the dangers of tribalism. Last month, a senior U.S. military adviser conceded, "We're not thinking through the impact of abetting further corruption and perpetuating tribal power." In December, a U.S. diplomat warned, "The absence of government in a lot of areas has allowed others to move in, whether militias or others." The net effect has been a splintering of the country rather than the creation of a unified nationalist Sunni front that, having regained its confidence, would be prepared to deal constructively with Baghdad.

THE CRUMBLING CENTER

The growth of warlordism is another consequence of the surge. By empowering the tribes and other networks without regulating their relationship to the state, the United States has enabled them to compete with one another for local control and what is mostly criminal revenue. It is worth noting that warlordism is not just a creeping Sunni phenomenon. Kurdish and Shiite criminals have been equally adept at exploiting the current security situation to their advantage. Indeed, warlordism appears even to be altering the sectarian divide. In Najaf, where gang warfare has erupted on more than one occasion, supporters of Sadr's Mahdi Army are engaged in street battles with members of the Badr Organization, even though both are Shiite groups.

Last December, a committee of British MPs charged with examining the security situation in Basra as British forces began to draw down concluded that warlords and criminal gangs had all but taken over the city. "Although the reduction in attacks on UK forces can only be welcome," the committee's report noted, "this alone cannot be a measure of success. The initial goal of UK forces in South Eastern Iraq was to establish the security necessary for the development of representative political institutions and for economic reconstruction. . . . This goal remains unfulfilled."

The United States' bottom-up strategy is also worsening sectarianism. For many Sunnis, reconciliation means restoration -- not inclusion in power-sharing arrangements but regaining control of the state. Instead of discouraging this mindset, the evolution of the surge into a bottom-up operation has validated it, fostering the impression that Washington has at last recognized that its strategic interests lie with the Sunnis. As the Sunnis see it, the current U.S. strategy is a policy of organizing, arming, and training them to challenge Shiite supremacy.

The Shiites and the Kurds naturally have sharply different notions of what reconciliation means. For the Kurds, reconciliation means respect for their claims to autonomy as well as for their potential territorial gains. The Shiites have tended to emphasize the need for justice before reconciliation, which, as they see it, requires that they be compensated for their suffering under previous regimes (not only Saddam's). This, in their mind, necessitates the subordination of Iraq's Sunni population to the Shiite community. Some Shiite leaders have defied such thinking -- Ayatollah Ali al-Sistani most prominently -- but Sadr has made clear that he will use violence to secure Shiite hegemony, and Maliki's government has shown no willingness to be pried away from Sadr and like-minded Shiites. Indeed, in postconflict situations, reconciliation often founders on the unwillingness of victims to surrender their claims to justice.

Some Sunnis have started to recognize that the United States has no intention of restoring their supremacy. The realization that civilian jobs and vocational training is all that is in store for the 80 percent of the former insurgents who are blocked from membership in the Iraqi army (Shiite leaders want to dominate the army in order to use it as their own instrument of control) has eroded Sunni cooperation with U.S. forces. As one volunteer told a reporter, "The Sunnis were always the leaders of the country. Is it reasonable that they are turned into service workers and garbage collectors? . . . We had not anticipated this from the American forces. Of course we will not accept that." One response has been to head back to al Qaeda. An Awakening commander in the Diyala provincial capital of Baqubah, which has never been fully pacified, said in February, "Now there is no cooperation with the Americans. . . . We have stopped fighting al Qaeda." This was doubtless an exaggeration, but one that pointed to the hard truth that for many Sunnis, Shiite rule remains unacceptable. When former Sunni insurgents no longer believe that Washington will restore them to dominance, their current U.S. paymasters will once again be their targets.

Given the current trajectory, significant Sunni segments of the postsurge Iraqi state will continue to be funded by the United States, but they will remain beyond the control of either Baghdad or Washington. They will also be in a position to establish ties with neighboring countries. All of this may well accelerate the centrifugal forces unleashed by the bottom-up strategy. When it withdraws from Iraq, the United States will be leaving a country more divided than the one it invaded -- thanks to a strategy that has systematically nourished domestic rivalries in order to maintain an illusory short-term stability.

This could mean that Iraq will remain essentially unreconstructed. The authority of the state would plummet, and the United States' ability to influence events, already limited, would become even weaker. Iraq would become a running sore, and successive crises within the country and on its borders would distract Washington from other priorities and sap its ability to normalize relations with Iran. For the Iraqis, safety, security, and economic advancement would remain uncertain. Those who could leave would. Stability would become an ever-receding prospect.

One plausible consequence of this turmoil would be the emergence of a U.S.-trained and U.S.-equipped Iraqi army, increasingly open to former officers of Saddam's military, as a powerful force in Iraqi politics. The professionalism and esprit de corps of the army is already on the rise. Officers who see themselves as having to navigate a maelstrom of unregulated militias, weak and irresponsible government officials, tribes emboldened and then embittered by their U.S. connections, and overbearing but uneven U.S. assertions of control could turn inward, as they did under the British and under Saddam. They might adopt a posture of superiority to politicians, impatience with upstart tribal leaders, and passive-aggressiveness toward their U.S. patrons and then sideline the civilian government and take control of the state. This result might be less disastrous than complete long-term breakdown: to the degree that Iraq needs a mediating military presence to sustain a fragile peace, this role might ultimately be better served by a military with its own corporate identity rather than by U.S. troops. But still, the United States would be confronted by a strong, centralized state ruled by a military junta that would resemble the Baathist regime Washington overthrew in 2003. Rather than an anarchic situation, the United States would face potentially aggressive nationalism and a regime unsympathetic to U.S. regional priorities.

RESPONSIBLE RETREAT

At this stage, the United States has no good option in Iraq. But the drawbacks and dangers of the current bottom-up approach demand a change of course. The only alternative is a return to a top-down strategy. To be more effective this time around, Washington must return to the kind of diplomacy that the Bush administration has largely neglected. Even with 160,000 troops in Iraq, Washington lacks the leverage on its own to push the Maliki government to take meaningful steps to accommodate Sunni concerns and thereby empower Sunni moderates. (The legislative package and the de-Baathification reform law passed earlier this year were seriously flawed and did more to spur the Sunnis' anxieties than redress their grievances.) What the United States could not do unilaterally, it must try to do with others, including neighboring countries, European allies, and the United Nations (UN).

In order to attain that kind of cooperation, Washington must make a public commitment to a phased withdrawal. Cooperation from surrounding countries and European partners is unlikely to be forthcoming without a corresponding U.S. readiness to cede a degree of the dubious control it now has over events in Iraq. Currently, the dominant U.S. presence in Iraq allows the rest of the world to avoid responsibility for stability in and around Iraq even as everyone realizes the stakes involved. A plan to draw down U.S. forces would therefore contribute to the success of a larger diplomatic strategy, prompting Middle Eastern states, European governments, and the UN to be more constructive and proactive in working to salvage stability in the Persian Gulf.

The point, therefore, is not to focus on the precise speed and choreography of a troop withdrawal. Rather, what is necessary is to make clear that the United States intends to withdraw. Should the Bush administration suspend the currently programmed withdrawals of the surge force, it would send precisely the opposite message. President Bush, Secretary of Defense Robert Gates, and General Petraeus have all signaled their interest in halting any further drawdowns after the last surge brigade has come home this summer. Petraeus, who has already begun to lay out his case in interviews, argues that "the key is to hang on to what you've got." The president has suggested that he is unwilling to withdraw additional troops until after the Iraqi provincial elections -- which, although originally scheduled for October, could very well be delayed. It is therefore possible that the next U.S. president will have to decide what to do with approximately 140,000 troops, a considerably larger number than most observers assumed would still be on the ground in Iraq at the end of 2008. (Some consideration will also have to be given to the problem of removing 56,000 contractors and facilitating the departure of a segment of the 30,000-50,000 Iraqi and foreign workers supporting the U.S. presence.)

Given that the laws of physics are as relevant to troop redeployments as are the laws of strategy and politics, the higher baseline bequeathed by Bush would mean a longer timeline for withdrawal. As of last summer, there were 1,900 tanks and other armored vehicles, 43,000 trucks, and 700 aircraft in Iraq. Equipment is scattered over 70 bases throughout the country, along with 38 major supply depots, 18 fuel-storage centers, and 10 ammunition dumps. According to the conservative rule of thumb used by military logisticians, the U.S. Army and the Marine Corps could move a brigade per month from the Iraqi theater. Moving the 15 brigades likely to be in Iraq in January 2009 would require up to 10,000 truck trips through potentially hostile zones within Iraq.

Although fixating on an exact timetable for withdrawal might be unhelpful at this juncture, a new administration should begin to draw down deliberately and in phases as soon as its internal deliberations are complete and the process has been coordinated with Baghdad. These steps could take months, as the new team conducts its policy-review process; military planners plot safe and efficient withdrawal routes; congressional consultations are carried out; conclusions are reached about where the forces being drawn down should be redeployed; planners determine the size, roles, and missions of the residual force; and the numerous dependencies created by the occupation and the surge are gradually shed. Once under way, however, a drawdown of most of the troops now in Iraq could be completed within two years. The redeployment might proceed more quickly if U.S. public support for the war collapsed, the Iraqi government demanded a swifter withdrawal, or the political situation in Iraq settled down; alternatively, the process might take more time if U.S. forces were under attack, an atrocity claiming the lives of many Americans occurred, or a responsible, reconciliation-minded Iraqi government and a concerned international community sought a slower drawdown.

RECONCILIATION FROM ABOVE

Announcing a withdrawal will entail certain risks. Aware that U.S. forces will finally be departing, Iraqi factions might begin to prepare for a new round of fighting. The Sunnis, aware of their vulnerabilities to attack by militant Shiite forces without the United States to protect them, might resuscitate their alliance with al Qaeda. The government in Baghdad might be concerned about its own exposure to attack in the absence of a U.S. shield and proceed to forge tighter links with Tehran or encourage greater activism by the Mahdi Army. It is all the more vital, therefore, that the drawdown take place as part of a comprehensive diplomatic strategy designed to limit these risks. The interval between a decision to withdraw and the removal of the bulk of U.S. forces should provide the space in which the UN can convene a multilateral organization to foster a reconciliation process in Iraq.

There is much that can be done to revitalize a top-down approach to reconciliation if it is under UN auspices and led by a credible special envoy. First, the international community should be energized to help Iraq move forward on provincial elections, which would test the popularity of the new Sunni leaders who have emerged during the surge and lash them up to Baghdad. This would have the added benefit of isolating the radical federalists from the majority of Shiites, who would prefer to live in a united Iraq. A UN envoy would have a better chance of brokering a deal on the distribution of provincial and federal powers, the issue that led to the veto of the provincial election law, than would Washington. In a multilateral setting that is not conspicuously stage-managed by the United States, regional states, including Iran and Saudi Arabia, could play a pivotal role in this process. Although Tehran's cooperation is inevitably hostage to its broader relations with Washington, UN sponsorship of this effort might provide the leaders of Iran with the cover they need to act in their own interest. The Saudis, for their part, would like to see the UN involved and are prepared to use their influence and money to impel the parties in Iraq toward reconciliation.

Second, an institutionalized multilateral group of concerned states should mobilize the broader international community to assist with the care, feeding, and permanent housing of the millions of refugees and internally displaced Iraqis who have not been able to get to the United States or Europe. This is essential, since refugee camps and squatter settlements are incubators of radicalism and radiate violence. The longer these populations remain unmoored and cut off from education, employment, and access to adequate social services and health care, the harder it will be to resettle them permanently, whether in Iraq or elsewhere.

Third, before a new and more intense phase of the civil war begins, there should be a multilateral process put in place to prod Saudi Arabia and other Persian Gulf states to finance investment projects that provide real employment in Iraq. Furthermore, Iraq's neighbors, including Iran, should be pressing the Iraqi government to bring far more Sunni Awakening volunteers into the regular Iraqi army and, crucially, into the provincial police forces funded by the central government. The latter step would reinforce the positive effects of the provincial elections and the emergence of politically legitimate local leaders. The current commitment to enlist 20 percent of the Awakening's members is far too small to have an impact.

Finally, the tribes feeding off the surge must be weaned from U.S. assistance and linked firmly to Baghdad as their source of support. Intertwining the tribes with Baghdad in this way, as the Iraq specialist Charles Tripp has noted, would yield something very much like the imperial protectorates in the Middle East of the first half of the twentieth century. The "club of patrons" in the capital would dole out goods to tribes through favored conduits. At this juncture, the U.S. military is performing the role of the patrons -- creating an unhealthy dependency and driving a dangerous wedge between the tribes and the state. Through coordinated action by the UN sponsors of the multilateral process, the government in Baghdad, and U.S. commanders on the ground, payment responsibilities will have to be transferred from the U.S. military to Iraqi government representatives.

There is no guarantee that the old way of giving tribes a taste of the lash followed by a dollop of state largess -- the model that successfully integrated tribes in Jordan and Saudi Arabia in the twentieth century -- can be successfully applied to a divided Iraq today. Iraq is heterogeneous, unlike Jordan or Saudi Arabia, where the state and the tribes shared a religious heritage. Furthermore, overestimating Iranian or Saudi influence on Iraqi politics and the willingness of the UN Security Council to plunge into the existing morass is all too easy. In any event, it will be a slow and hazardous undertaking. Many things have to happen more or less simultaneously in a carefully coordinated chain of actions. Washington has to announce that it will begin withdrawing the bulk of its forces. The UN secretary-general, with the backing of the Security Council, must select a special envoy. A contact group of key states must be formed under UN sponsorship. Priorities and milestones will need to be set for the distribution of resources within Iraq, the recruitment of Sunnis to the army, provincial elections, foreign investment, dealing with refugees, and development assistance. Crucially, the phasing of the troop drawdown will have to mesh with this diplomatic process but not hinge on its ultimate success. This course is risky and possibly futile. Yet it is still a better bet than a fashionable, short-term fix divorced from any larger political vision for Iraq and the Middle East.

Listen to this essay: Download

This article is part of the Foreign Affairs Iraq Retrospective.