When not to start a war

Gen. David Petraeus's report of some "surge"-driven success on the ground does not change the reality that invading Iraq has proved disastrous, and that nothing plausibly salvageable there is worth thousands more GI lives, or hundreds of billions more dollars. So our soldiers should come home.

But President Bush is right to remind the nation that humanitarian horrors followed closely on the heels of America's withdrawal from Vietnam, and that a horrific scenario could unfold in Iraq if US forces leave abruptly.

Unconscionable as it may have been to stand by during the Rwandan genocide in 1994, consider how much greater the moral culpability when the slaughter would flow from a sequence of events that America's own war of choice set in motion.

We're trapped in Iraq. But just as Mr. Bush now has the foresight to envision the consequences of a departure, he and his national security advisers ought to have foreseen the trap.

Consider the five major wars US forces have fought since World War II: Korea, Vietnam, the Gulf War of 1990-91, and the ongoing Afghan and Iraqi wars. True, no genocide followed the Korean War. But tens of thousands of US troops have never been able to withdraw from the Korean peninsula, and to this day their presence has only bought a highly flawed peace that has tolerated vast human suffering and a nuclear-armed North.

Bush has now reminded us that the Vietnam conflict begat, among other horrors, the Khmer Rouge in Cambodia. And while a US departure after the Gulf War – celebrated as an unequivocal US victory, unlike Korea and Vietnam – did not lead to genocide, it did precipitate an uptick in brutality that Saddam Hussein soon visited on Kurds and Shiites.

Given the extraordinary strategic freedom Bush has enjoyed in conducting the current Iraq war – almost wholly unconstrained by Congress, public opinion, an expensive social agenda, or the prospect of great power intervention against US forces (all of which presidents Johnson and Nixon had to contend with during Vietnam) – it is shameful that Bush still finds himself able to warn us (accurately!) that something even worse than what Iraqis have today would probably follow a "premature" US departure.

This 60-year track record suggests that it would be reckless for any US president to expect to leave the scene of a major war without risking carnage.

Indeed, we have every reason to take seriously the prospect that, once having gone to war, a president will eventually face the choice of staying in perpetuity to put a lid on violence, or leaving and looking the other way while something approaching genocide is perpetrated.

There is little evidence that the Bush administration evaluated the prospect of getting ensnared in this type of trap before committing the US to war in Iraq. Indeed, insider accounts have suggested remarkably little cost-benefit analysis or contingency planning.

But the first MBA president should have considered how good business leaders approach major resource allocation decisions. First, they consider alternative scenarios. And second, having made a decision to invest, they remain prepared to "cut and run" from investments that (contrary to optimistic projections) prove to be value-destroying. They do so not out of cowardice nor lack of will, but rather because they recognize that the only investment the business can't afford to lose on is its portfolio as a whole.

Regrettably, US national security could legitimately require fighting another war in the near future. It is essential, then, that US leaders improve drastically on the judgment and decisionmaking that trapped the US military in Iraq. First, they need to recognize before initiating a war that experience suggests a high probability of finding themselves in a trap. If the risk of this trap can't be reduced to manageable levels, the war shouldn't be started.

Second, US leaders need to be prepared, psychologically and politically, to withdraw forces from conflicts where their continued presence, weighing both strategic and humanitarian concerns, no longer serves the national interest – even though critics would label that "losing."

When it comes to a war of choice, America's CEO can't afford to start one without anticipating traps; and then only if he or she is willing to "lose" to keep from jeopardizing even larger stakes.

• Andy Zelleke is a lecturer in public policy at Harvard University's Kennedy School of Government, and codirector of its Center for Public Leadership.