Options for the New Secretary of Defense

(3 of 5)

Surge forward. Among some active and retired generals, as well as some officers inside Baghdad's Green Zone, there is support for the idea of a temporary surge--boosting U.S. troops levels by 20,000 to 30,000 to stabilize the country. Under this plan, the extra U.S. forces would be deployed to try to quell the sectarian slaughter in Baghdad as well as subdue the jihadists of Anbar province. It is a step that, almost everyone agrees, should have been taken years ago.

The question is whether it is simply too late. There is, for starters, the zero-sum problem. Yes, the more troops we send, the more stability we can buy. But when the troops are withdrawn, instability will return. The second problem is logistics. The U.S. does not have the kinds of reserves that would allow it to beef up its Iraq forces for very long without a further decline in readiness, morale and troop retention. The Army's brigade combat teams are already ragged from fighting two wars; speeding their rotations back into battle would put some units at unacceptable risk. The new Marine Corps commandant said in effect last week he could not maintain the current "operating tempo" without increasing the corps' size--something the Bush Administration has opposed.

Even if the troops were available, there are formidable political barriers to sending them. The Democrats have made it clear that the idea is a nonstarter. While that sentiment could change if Baker and Hamilton support a surge, Abizaid has said he opposes more troops because it would discourage Iraqis from taking responsibility for their own security. "Iraqis will decide Iraq's future," says Major General William Caldwell, the top Army spokesman in Iraq. "Additional coalition troops may produce short-term effects, and we may execute that option, but they are not a long-term solution."

Train and retreat. Quietly gaining favor in military circles, this approach is designed to balance the need for more U.S. forces with a desire to decrease the rate of U.S. casualties. Here's how it would work in theory. First, Washington would boost the number of troops in Iraq and, more specifically, increase the number of G.I.s on military-training teams. The goal is to get the Iraqi army--which is beset by a weak officer corps, weapon shortages and an almost total inability to move around--ready to take over the country as the Americans start to pull out.

If all goes as planned, the U.S. could reduce troops levels as much as 60% within a year. Even if that were possible, "train and retreat" still envisions leaving some 50,000 U.S. troops in Iraq for years. These forces would be dispersed into four or five large garrisons or cantonment zones, far from urban areas. While U.S. troops patrolled the perimeter and guarded against mortar attacks from outside, Iraqi recruits would be trained inside. "You train 'em up and push 'em out," says a well-placed Baker-commission source. "That's what's happening."