How We'll Know When We've Won

Mahdi Army fighters stand in Basra, Iraq, Thursday, March 27, 2008.
AP Photo
This column was written by Frederick W. Kagan.

The president's nomination of generals David Petraeus and Raymond Odierno to take command of U.S. Central Command and Multinational Force-Iraq, respectively, was obviously the right decision. By experience and temperament and demonstrated success, both men are perfectly suited to these jobs. Given the political climate in Washington, however, their nominations are likely to be attacked with the same tired arguments war critics used to try to drown out reports of progress in Iraq during the recent Petraeus-Crocker hearings. So before the shouting begins again, let us consider in detail one of the most important of these arguments: that no one has offered any clear definition of success in Iraq.

Virtually everyone who wants to win this war agrees: Success will have been achieved when Iraq is a stable, representative state that controls its own territory, is oriented toward the West, and is an ally in the struggle against militant Islamism, whether Sunni or Shia. This has been said over and over. Why won't war critics hear it? Is it because they reject the notion that such success is achievable and therefore see the definition as dishonest or delusional? Is it because George Bush has used versions of it and thus discredited it in the eyes of those who hate him? Or is it because it does not offer easily verifiable benchmarks to tell us whether or not we are succeeding? There could be other reasons - perhaps critics fear that even thinking about success or failure in Iraq will weaken their demand for an immediate "end to the war." Whatever the explanation for this tiresome deafness, here is one more attempt to flesh out what success in Iraq means and how we can evaluate progress toward it.


A stable state. An unstable Iraq is a recipe for continued violence throughout the Middle East. Iraq's internal conflicts could spread to its neighbors or lure them into meddling in its struggles. An unstable Iraq would continue to generate large refugee flows, destabilizing vulnerable nearby states. An unstable Iraq would enormously complicate efforts by the United States or any other state to combat terrorists on Iraqi soil. An unstable Iraq would invite the intervention of opportunist neighbors. The Middle East being an area of vital importance to the United States and its allies, all these developments would harm America's interests.

A representative state. Some war critics (and even some supporters) argue that the goal of "democratizing" Iraq is overoptimistic, even hopeless. So what are the alternatives? Either Iraq can be ruled by a strongman, as it was in the past, or it can be partitioned into several more homogeneous territories, each ruled according to its own desires. Before settling for either of these, we should note that the overwhelming majority of Iraqis continue to manifest their desire for representative government, as evidenced by the 8 million who voted in the last elections, the 90 percent of Sunni Arab Iraqis who tell pollsters they will vote in the upcoming provincial elections, and the sense on the streets that anyone who tries to eliminate representative government will do so at his peril. Beyond that, we must note that neither of the two suggested alternatives is compatible with stability. Nevertheless, let us examine them.

A strongman. Iraq is a multiethnic, multisectarian state just emerging from a sectarian civil war. How could a strongman rule it other than by oppression and violence? Any strongman would have to come from one or another of the ethno-sectarian groups, and he would almost certainly repress the others. Although he might, in time, establish a secure authoritarian regime, the history of such regimes suggests that Iraq would remain violent and unstable for years, perhaps decades, before all opposition was crushed. This option would not sit well with American consciences.

Partition. Partitioning Iraq would generate enormous instability for the foreseeable future. Again, virtually no Arab Iraqis want to see the country partitioned; the Sunni, in particular, are bitterly opposed. But their desires aside, could a partitioned Iraq be stable? The Kurds, after all, already have their region. What would happen if the Shia got all nine provinces south of Baghdad, and the Sunni got Anbar, Salah-ad-Din, and whatever part of Ninewa the Kurds chose to give them? Well, there would be the problem of Baghdad and Diyala, the two mixed provinces, containing mixed cities. Despite the prevailing mythology, Baghdad has not been "cleansed" so as to produce stable sectarian borders. The largely Sunni west contains the Khadimiyah shrine, which the Shia will never abandon, while the largely Shia east contains the stubborn Sunni enclave in Adhamiya. The Sunni in Adhamiya have just gone through many months of hell to hang on to their traditional ground. And there are other enclaves on both sides of the river.

Any "cleansing" of them would involve the death or forced migration of tens or possibly hundreds of thousands. Attempts to divide Diyala and even Ninewa would produce similar results. If ethno-sectarian conflict restarted in Iraq on a large scale, cleansing might make this solution more feasible, but at enormous human cost. In the current context, even to seriously propose it threatens Iraq's stability.

A state that controls its territory. We already have an example of a sovereign, quasi-stable state confronting terrorist foes that is theoretically allied to the United States but has no American troops and does not control all of its own territory. It is Pakistan, whose ungoverned territories in the Federally Administered Tribal Areas and the Northwest Frontier Province have become safe havens for the leaders of the global al Qaeda network. If the United States abandoned Iraq before Iraq could control all of its territory with its own forces, we might make way for similar safe havens in the heart of the Middle East. It is clearly not in America's interests to create a Pakistan on the Euphrates.

A state oriented toward the West. It is also clearly against America's interests for Iraq to become an Iranian puppet. Some in the United States, however, see that development as inevitable; they point to geography and religious ties. Some even say that the United States should not only acquiesce in the inevitable but embrace it, reaching out to the Iranians for their assistance in smoothing our withdrawal as they establish their domination. But why? Iran has not dominated Iraq in centuries. True, the Sunni-Shia divide is profound, but so is the Arab-Persian divide. Iraq's Shia, remember, enthusiastically supported Saddam Hussein's war against their Iranian co-religionists in the 1980s - a sectarian "betrayal" for which the Iranians have never forgiven them. Again, American troops and civilians who live day to day with Iraqis throughout the country report a dramatic rise in anti-Persian sentiment, coincident with a rise in Iraqi Arab nationalism. But back in the United States, the debate over Iraq is scarcely tethered to reality on the ground. In the simple terms suitable to that debate, then, suffice it to say that neither shared Shia faith nor a shared border has historically led to Iranian domination of Iraq. There is no reason to assume it will do so now.