The controversy is Washington over majority leader Harry Reid's statement that the war is lost is an interesting case study in the logic of war. Reid's statement has been countered by

• General David A. Petreaus' claims of barely perceptable improvement in the situation on the ground
• President Bush's statement to the effect that the new surge of violence in Iraq is the U.S. surge going according to plan - the expected reaction of the insurgency to the surge
• and former secretary of defense Donald Rumsfeld's comment on Larry King to the effect that this is a "contest of wills."

Meanwhile, some strongly critisize the Democratic plan of setting a deadline for U.S. disengagement from the conflict.

My question is this: How can a war like this ever end if it is lost, but arguments like those of Petreaus (which might be right, but sound a lot like what we were hearing in Vietnam with all the stuff about pacified areas and body counts) and Rumsfeld are taken seriously? Of course it is a test of wills. To follow that logic and not give because it is a test of wills means never-ending war. If we assume for a moment that

• we are not fanatics and do not want eternal war
• and acknowledge that defeat is possible
• and note the fact that in a war like this the war cannot end in enemy occupation of U.S. territory

it becomes a war of pure choice where we need some kind of way out in the event of defeat. The central questions then becomes What are the indicators that the war is lost? If we go the "test of wills" route, there is always reason to keep fighting. If increased enemy violence is an indicator of the plan succeeding, defeat becomes almost absurdly an indicator of victory. How many more weeks of record-breaking slaughter will be "the plan"? When does it turn into an indicator of the opposite - failure?

I can't answer these questions, but it seems to me that we must find a way to avoid the deadly, murderous logic of these kinds of arguments while, at the same time, not slipping into automatic defeatism every time something goes wrong. (Which might set a poor precident for situations where the war is indeed necessary, unlike it was in this case.)

At the very, very least, the answer must include a situation where someone can at least make a public statement to the effect that the war is lost without being accused of Wehrkraftzersetzung. And an exit plan with a calendar date must be an option. Otherwise, if even presenting such ideas is treason, it will be a never-ending game of "emperor's new clothes" and never-ending slaughter. In a life-or-death struggle on our own territory, this kind of totalitarian political discourse would serve a purpose that just about anyone other than a total pacifist might accept, depending on the situation. In a situation of a war of choice on the other side of the globe, however, it stifles the very discussion needed to eventually end the war in the event that it really is a lost cause. Where defeat is not total, the mechanism for accepting it involves discourse that runs counter to the military logic of mobilizing the entire population's total will and sentiment. If that makes winning harder in the "test of wills," then that is part of fighting wars of choice that those who choose them must accept.