A glimmer of hope exists in Iraq, but it has not been produced by the surge of extra U.S. forces. Although the surge, prior ethnic cleansing that has separated warring ethno-sectarian groups, and the U.S. military’s paying off Sunni insurgents have reduced violence, the lull is likely to be temporary. The underlying ethno-sectarian fissures in Iraqi society have not been mended and cannot be papered over by centrally passed laws, such as the new statute allowing some Sunni Ba’athists back into the government (in fact, this law bans Ba’athists from serving in the security forces).

Fortunately, some prominent Iraqis realize that centrally imposed institutions, laws, and solutions will not work in this artificial country containing three major ethno-sectarian groups. One of these Iraqis is Mowaffak al-Rubaie, Iraq’s national security adviser. In a Jan. 18, 2008, op-ed in the Washington Post, he proposed a substantially decentralized Iraq in which new autonomous provinces would provide most Iraqi governance through regional executives and parliaments. These regional governments would be funded proportionally out of Iraq’s petroleum proceeds on the basis of the provinces’ populations. The central government would be responsible for defense, foreign policy, interregional affairs, banking, and fiscal and monetary policy.

Rubaie has astutely realized that Iraq ratified a constitution before its ethno-sectarian groups reached consensus on how powerful the central government should be in relation to the powers of the regions. He argues an obvious point that many Iraqis and American policymakers miss: the ethno-sectarian groups are the core of the polyglot Iraqi society, and the central government can only have the legitimacy conferred on it by such groups. He implies that as time passes and the various groups feel more secure, they will confer more legitimacy and power on the central government.

Here Rubaie’s thinking diverges from an even harsher reality. Since Iraq is a make-believe country that has seen many decades of one ethno-sectarian group (the Sunni Arabs) ruling and brutally oppressing the other two (the Kurds and the Shia), such ethno-sectarian identities and suspicions of the other groups will die hard. To ensure long-term stability, Iraq will have to be even more radically decentralized than Rubaie advocates. For example, the ethno-sectarian groups will never allow the central government to control internal security or defense forces. In fact, no one has made a serious attempt to disarm the various ethno-sectarian militias active in Iraq, and the U.S. government’s desperate attempt to reduce violence required training and arming yet a new militia – former Sunni insurgents. Thus, security will also have to be provided at the regional level by essentially deputizing existing militias in the various regions.

Furthermore, the ethno-sectarian groups would be suspicious that one group would control the central government’s oil ministry and divert more revenues to its own region or regions. No doubt, the Sunni Arabs would be most suspicious, because unlike the territory of the Shia and Kurds, their land has many fewer proven oil reserves (at least now). Thus, even oil exploration, production, and sales would have to be decentralized to the regions. To prevent future war among the groups over oil proceeds, the borders of the regions would have to be gerrymandered to give the Sunni region or regions actual oil wells that generate revenues in proportion to their percentage of the total population.

In sum, only foreign policy and interregional commerce would be governed by a very weak central government. With the strong ethno-sectarian identities in Iraq, any stronger central authority would generate fears of recurrent violent oppression and economic exploitation. But Rubaie is on the right track and should be listened to if Iraqis are to avoid a future full-blown civil war.