Although I've made fun of international relations theorists in the past, there is one IR concept which, when properly applied, has its uses: the realist doctrine of "equilibrium", which can be interpreted in the Machiavellian sense of, "help my weaker enemy against my stronger enemy". This tactic can have several outcomes, all net positive: the stronger enemy could defeat the weaker enemy, but at a greater cost than had the weaker enemy not been helped; the weaker enemy could make use of the help to defeat the stronger enemy, effectively knocking the stronger enemy further down on the "enemies list"; or the conflict could drag on, preoccupying both enemies and thus reducing their opportunities to make trouble elsewhere. The most famous modern example of this strategy in action is the assistance that the US and Britain gave the Soviet Union once it was attacked by Nazi Germany in 1941. Needless to say, that strategy worked brilliantly.
Of course, if the weaker enemy defeats the stronger one, then the weaker enemy can become a problem in itself, and it can seem superficially, in retrospect, as though the strategy has backfired. These days, the most commonly cited example of this "blowback" phenomenon is the assistance that the US provided to Islamic fundamentalists--including, apparently, one Osama bin Laden--in fighting the Soviet occupation of Afghanistan during the 1980s. (For some reason, the Cold War is never described as blowback from "lend-lease".) Now, neocon Michael Rubin is describing Iraq's 1991 invasion of Kuwait as blowback from American aid to Iraq during the Iran-Iraq war.
Both interpretations are sheer nonsense. Global Islamist terrorism is a serious threat, but it can't begin to compare to the threat previously posed by the Soviet Union. Assuming that American assistance to the Afghan rebels in the 1980s actually helped bring about the collapse of the Soviet Union--and its contribution seems in retrospect to have been substantial, if not decisive--the investment paid off spectacularly. In fact, had the Soviet Union not fallen, there's a good chance that Islamist terrorists would be an even greater threat to the US than they are today--thanks to the kind of assistance from the Soviet bloc that anti-Western terrorists received during the 1970s and 1980s.
Likewise, American support for Saddam Hussein during his long war with Iran was hardly a failure. On the contrary, it kept two deadly enemies of the US preoccupied for nearly a decade--a decade during which they made relatively little trouble for the rest of the world. Again, had the US withheld aid, and had Iran managed to defeat Iraq as a result, it would likely have been able to topple Saddam Hussein, and achieve political dominance over its neighbor. Its rebuilding of its military and financial resources--not to mention its nuclear program--would also have started sooner and taken less time. In short, had that assistance not been given to Iraq, Iran would likely be far richer, more powerful and more dangerous than it is today.
Of course, defenders of the first president Bush's decision not to help topple Saddam Hussein in 1991 use exactly the same logic: without Saddam Hussein as a counterweight, Iran would have had a freer hand than it did. The problem with this argument is that by 1991, Saddam Hussein had stopped being an effective counterweight to Iran. On the contrary, he was more of a burden to the US--which had to station tens of thousands of troops to guard the Saudi flank from him, while devoting a portion of its air force to protecting Iraqi Kurdistan--than a hindrance to Iran, which spent the 1990s building and arming Hezbollah, strengthening its ties with Syria, and revving up its nuclear program.
What does this history suggest about America's current situation in Iraq? Well, it would have been nice, of course, if Iraq could have been a peaceful, stable, pro-Western democracy, acting as a bulwark against the various anti-American radicalisms in the region. But given that that unrealistic goal has not been achieved, it seems unlikely that America's worst enemy in the world today is the collection of mutually antagonistic sectarian terrorist militias springing up there. On the contrary, since they all seem to have far more enthusiasm for slaughtering each other (or at least, each other's civilians) than for attacking American troops, it seems far more reasonable for the US to let them have at it than to try to stop them. As Daniel Pipes noted, civil war in Iraq "would be a humanitarian tragedy but not a strategic one". Indeed, it may provide interesting strategic opportunities: by supporting the more anti-Syrian/Iranian groups, the US might be able to draw those countries further into the conflict, forcing them to use more of their resources exerting their influence in Iraq than they would otherwise need to apply.
Now, it has been speculated lately that the Baker-Hamilton commission will recommend instead addressing the Iraq problem by attempting a rapprochement with Iran. Stupidity is never out of the question, of course, but I'm skeptical of that prediction. Whatever else may be said of James Baker--anti-Israel, Arabist, State Department-style schmoozer of dictators--his (and the State Department's) absolute number one favorite anti-Israel Arab dictator to schmooze has always been the reigning monarch of Saudi Arabia. And by siding with Israel, of all countries, against Hezbollah, the Saudis have made it crystal clear that their foremost concern these days is the threat of the Iran-Syria-Hezbollah axis.
Assuming, then, that Baker stays true to his old loyalties, don't be surprised if American policy in Iraq, far from moving towards reconciliation with Iran, takes a sharp turn, as it were, towards the Sunni side of the street.