Testifying before Congress yesterday, the U.S.'s chief commander in Iraq, Ray Odierno, pointedly refused to say whether he thought U.S. combat troops would be in Iraq past 2011. Whatever the remnant, it will be small. It's a mundane point, noticing that Iraq, which expanded to become an umbrella under which everything soaked by the Bush administration was discredited, is no longer an issue of relevance to the political debate. A quick distinction must be made between an important issue -- one that people care about -- and a relevant issue -- one that people vote about. Afghanistan remains an important issue, and yet there are signs that, by 2010, it will have turned into a relevant issue, even if the contours of the debate stay roughly the same. Relevance is related to salience and immediacy more than import. It's the type of issue that will raise the temperature of the partisan heat map in midterm elections, the sort of issue that independents will use to identify against the party in power.

Or then again, maybe it won't. It's not lost on Obama administration officials that, when Barack Obama ran for president, Iraq was the wrong war, and Afghanistan was -- well, if it wasn't exactly the good war, it was well off the radar screen -- important but not relevant -- and they used it to frame a message about the Bush administration and Republicans that truly resonated: they chose the wrong country and diverted focus off of the real threat. By the time Obama was elected, the Bush administration had basically gifted him a way to fulfill the promise that brought him to prominence -- a withdrawal consensus that was itself a bastard child of poor decisions. No heavy lifting required.

From the time Obama decided to run -- late 2006 -- until the time he was elected -- Iraq, as a political question, firmly resolved itself. And Afghanistan, an issue that began as a consensus, had turned into something completely different. Obama was presented with a request for 30,000 more troops, which he promptly endorsed. It turns out that the consensus -- a narrow focus on getting rid of Al Qaeda's safe haven in Afghanistan -- was entirely misplaced. The Taliban-Al Qaeda nexus had little to do with the geographic boundaries of Afghanistan and a lot more to do with an inextricable link between geopolitics, Afghanistan's natural resources, like opium, and historical forces that even the most powerful country in the world cannot arrest. Politically, Obama is hamstrung by the way he defined Afghanistan during the campaign, a definition that resulted from a way of differentiating that conflict from the conflict in Iraq, a definition that wasn't sui generis. Pretty much the only thing Obama can say to the American people about Afghanistan is that had we properly resourced the war from the start, we wouldn't be where we are today.

All this is a way of saying that the issues one assumes will define an election often turn out to have little to do with what happens after the election; likewise, the type of insight that brought Obama into the consciousness of Democrats -- that Iraq was a bad war -- isn't very helpful in figuring out what to do about the war today. It's quite possible that a variant of the failed strategy in Iraq -- primarily counter-terrorism + bribery + a war of attrition over the resources (substitute poppy for oil) might work better in Afghanistan.

If Obama decides to go this route, and it looks like some influential folks are beginning to favor it, he will have backed away from what one might call a campaign promise to resource Afghanistan properly. That doesn't get you points in politics. For some reason, everyone obsesses about consistency, which this administration believes is just about the worst value to hold when it comes to foreign policy. To them, consistency is dogma. In their view, the U.S. ought to treat situations differently because they are different, because ceteris paribus never applies, because situations are always evolving, because interests are always more pressing than values. The downside of flexibility is that it comes at the cost of legitimacy. But so does dogma, magical thinking, a blind faith in whatever patterns we think we see in the void. I don't know if Brent Scowcroft would call this approach "realism." It's more...ad hoc.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.