When not obsessing about the election over the past 48 hours I have, for book purposes, been revisiting the Iyad Allawi Era in Iraq, from July 2004 to February 2005. This has gone down, I think, as something of a "lost period" in America's war effort for a couple of reasons. For one thing, during pretty much his entire term in office -- and certainly during his non-lame-duck months -- the American press was dominated by the 2004 presidential election. Part of the election dynamic was for the debate over Iraq to become very ossified with Kerry and Bush just trading the same barbs over and over again in a way that discouraged people from paying attention to changes on the ground. The other thing, however, is that attention to the Allawi Era simply doesn't do much to advance the political arguments of either of our main teams in Iraq.

The thing about this period is that, though neither Democrats nor Republicans are eager to admit it, what Bush was doing during this time was essentially exactly what "mainstream" "moderate" administration critics say the administration should have been doing. Allawi -- a secular Iraqi Arab of Shiite ancestry who used to be a Baath Party member -- was a decent candidate to try and bridge the sectarian divide in Iraq. His policies, moreover, were aimed at just that. His administration tried to roll back de-Baathification, and during the Allawi era you had Sunni Arabs serving in the government security forces operating in Sunni-populated areas. While attempting these measures of Sunni conciliation, Allawi also fought the Sunni insurgency, but he also fought Muqtada al-Sadr's Shiite militia.

Moderate administration critics put a lot of emphasis on Bush's failure to attempt such policies during Year One of the war, but tend to neglect the fact that these policies actually were tried during Year Two.

And they didn't work, at all. Allawi came to be seen not as a unifying, reconciling figure, but as an object of universal disdain -- an American puppet. Sunni Arabs were happy to grab the hand he extended to them and join the security forces, but only in order to infiltrate the Iraqi government and work on behalf of the insurgency against the collaborationist government. The fight against Muqtada was perfectly successful at killing Mahdi Army members in large numbers, but simply denuded the government of popular support in the Shiite community. Heading into the elections, Allawi had essentially no popularity and his party was overwhelmingly rejected at the polls despite American efforts to aid his campaign in favor of more sectarian formations.

Since his time in office, clearly, things have, in a sense, only gotten worse and worse. But it was during the Allawi Era that the fundamental futility of the American mission became clear. The problem in Iraq wasn't that there was "an insurgency" or "a militia" that needed to be defeated by American forces. The problem in Iraq was that Iraqi society had nothing resembling a political consensus about the legitimate terms of the state. Given such a consensus, American troops would be unnecessary. Absent such a consensus, American troops were useless. And, clearly, forming such a consensus was going to be an extremely difficult task.

Obviously, the argument can be made that if these strategies had been implemented sooner they might have worked better. Be that as it may, it's notable that nowadays plans for Iraq coming from the right (Fred Kagan, Ralph Peters) tend to amount to an espousal of a return to Allawi-ism and that there's no reason to think that trying this again will work out any better.

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