How long will the US maintain a large deployment of troops in Iraq? That is now the central question of US President George W. Bush's second term. Until recently, the Bush administration answered with an evasive cliche "as long as it takes and not one day longer." But not anymore.
The ice began to crack on Nov. 17, when Representative John Murtha, a hawkish Democratic congressman and Marine veteran, suggested pulling troops out of Iraq in six months. Soon after, the Republican-controlled Senate voted for "a significant transition to full Iraq sovereignty in 2006." After initial resistance, Bush began to change his rhetoric by suggesting that a troop reduction would occur sooner than previously expected.
The erosion in public support for Bush's Iraq policy is stark. Fifty-four percent of Americans now say that the US erred in sending troops, up from 24 percent at the start of the war in March 2003. In part, this reflects the rising casualty rate, with more than 2,100 US soldiers killed thus far.
But it also reflects a growing belief that the war is failing. As Duke University's Peter Feaver, an expert on public opinion who is now serving as a White House advisor, recently pointed out, Americans will tolerate casualties when they believe that a war is just and has a reasonable prospect of success. But citizens now doubt both these points. The administration is paying the price for overselling the reasons for the war and bungling the post-invasion occupation. Not surprisingly, Bush's new rhetoric stresses that he has a "strategy for victory."
If "victory" remains defined as stable democracy in Iraq, it is unlikely that Bush will have enough time to implement his strategy. In September, General George Casey, the senior US military commander in Iraq, testified to Congress that modern insurgencies last about a decade, and that the Iraqi army had only one battalion capable of fighting without help from US military forces. A month later, the influential International Institute for Strategic Studies in London estimated that US troop withdrawals next year were likely to be small, and that it would take at least five years for Iraq to build the 300,000-strong army needed to fight the insurgency on its own.
But, with Congressional elections next year and a presidential election in 2008, five years is too long. It seems more realistic that the Republican administration will have only 18 months to two years to implement its strategy.
The Democrats, meanwhile, are divided. Some, like Murtha and Representative Nancy Pelosi, the minority leader in the House of Representatives, want a short timetable, while others, like Senator Joseph Biden, resists a firm timetable but predicts a withdrawal of 50,000 US troops next year, with many of the remaining 100,000 to follow in 2007.
Those of us who believe that invading Iraq was a mistake, and that Bush is guilty of hubris in his failure to plan adequately for the aftermath, face a dilemma: if the US withdraws too precipitously, it may compound these mistakes.
Iraq is not like Vietnam, where the US departure was followed by stability imposed by an authoritarian government. In Iraq, the danger is that departure could be followed by civil war and chaos -- ideal conditions for terrorists to maintain havens.
Iraq differs from Vietnam in another way as well. Unlike the North Vietnamese, the Sunni insurgents will have a difficult time taking over a country where they represent only 20 percent of the population. Indeed, with Shia Arabs and Kurds making up 80 percent of the country, the insurgency is focused on only four of Iraq's 18 provinces.
The US' quandary is that it is both part of the problem and part of the solution. So long as a large number of US troops remains as an occupying force, they serve as a recruiting tool for insurgents. As the political scientist Robert Pape has shown in a careful study, resistance to foreign occupation is a prime motivation for suicide bombers.
But, if the US leaves too soon, the elected Iraqi government may be unable to cope with the insurgency, sending Iraq the way of Lebanon in the 1980s or Afghanistan in the 1990s.
Similarly, if Bush sets a short timetable, he may encourage the insurgents to wait him out. But, unless he makes it clear that US troops will leave in the near term, he will reinforce the impression of imperial occupation. The key to resolving this dilemma will be to press for local compromises that involve Sunnis in the political process, and to step up the rate of training of Iraqis to manage their own security. Even then, success is uncertain.
One failure is already clear: that of the neo-conservative dream of creating a military ally that could serve as a long-term base for US troops in the campaign to transform and democratize the Middle East. Three elections have produced some degree of legitimacy for the Shia-dominated Iraqi government, but without a sense of community and effective institutions, elections merely create a tyranny of the majority. That may be better than former president Saddam Hussein's tyranny of the minority, but it is hardly modern democracy.
Bush compares his goal in Iraq to the democratization of Japan after World War II. But Japan was a totally conquered, ethnically homogeneous country with no insurgency, a large middle class and previous experience of political openness. Even then, success took seven years.
Instead, Bush should plan on a two-year window to give the Iraqi government as strong a chance as possible before the Americans leave, while emphasizing that Iraqis will thereafter be responsible for their own security and political salvation.
Joseph Nye is a professor at Harvard's Kennedy School of Government. Copyright: Project Syndicate