It's time to think about breaking up Iraq—not into three separate states (Sunni, Shiite, and Kurd) but possibly into semiautonomous federal districts.
I criticized this notion last May after Peter Galbraith proposed such a plan in the New York Review of Books. I complained that it could turn Iraq into a weak state, sire civil war, and lure neighboring countries to intervene, potentially inflaming the entire region. I still fear this possibility. The problem is, Iraq seems headed toward the same nightmare under the status quo. A new political structure might better contain the chaos than blithely muddling through.
Galbraith's idea was based on the premise that Iraq has always been an artifice—the product of lines drawn on a map by British imperialists in the wake of World War I—so why not erase the blunder and revert to traditional ethnic and tribal boundaries?
In his latest online Newsweek column, Christopher Dickey reports that Iraq's national security adviser, Moweffak al-Rubaie, is floating a similar plan, called "democratic regionalism." This would split the country into between four and six districts—the Sunni triangle, the Kurdish territories in the north, at least two Shiite areas in the south, and an administrative center in Baghdad.
Dickey differs from Galbraith in that he recognizes a crucial element in this vision: that the Sunnis must get a piece of Iraq's oil wealth. With a simple tripartite federation, the Kurds would get Kirkuk, the Shiites would have Basra and Nasiriyah—all oil-rich territories—but the Sunnis would be left with nothing. Under Rubaie's plan, the main function of the central government in Baghdad would be to distribute the revenue in some equitable fashion.
To a large extent, this is what the insurgency in the Sunni triangle is about: the Sunnis' sense of impending powerlessness and impoverishment. Sunnis make up just 20 percent of the Iraqi population. In the government to be elected this January (if elections take place), they will be in the minority; the Shiites will have majority rule. This (and not just the recent offensive in Fallujah) is why the leading Sunni parties are threatening to boycott the election.
The Bush administration has made many mistakes in its occupation policy, but one of the biggest was its failure to devise a Sunni strategy—a formula that would give the Sunnis some stake in a new Iraqi order.
The failure had a number of roots. Saddam Hussein was Sunni and lavished favors on his own kind. During the 1991 Gulf War, the first President Bush encouraged Shiites to rebel against Saddam, then allowed them to be slaughtered by Saddam's Sunni soldiers in the war's aftermath. Clearly some payback was in order. But the obligation was pushed beyond reason by the Pentagon's ill-fated backing of Shiite exile Ahmad Chalabi as leader of the new Iraq. Chalabi turned out to be fanatically anti-Sunni and reinforced the Bush administration's leanings in that direction.
This is not to say that the Sunnis now deserve special treatment, but they do need some benefits if Iraq is to avoid spiraling deeper into chaos.
Iraq's current political course holds little hope. If the major Sunni parties boycott the coming election, the new government would have no legitimacy in the least stable part of the country. If the elections do come off and Shiites hold a strong majority in the new government, many clerics will press for the adoption of Islamic law—and there's no way the Kurds will go along. In fact, the Kurds are unlikely to stand for any substantial erosion of the autonomy they've enjoyed for over a decade thanks to U.S. air patrols. In other words, elections or no, Iraq will be subjected to massive centrifugal forces.
These forces might best be reduced by allowing all the country's factions a measure of autonomy, especially in the realm of social customs and religious law.
Possibly neither Galbraith nor Dickey goes far enough in proposing three, four, or six districts. Juan Cole, professor at the University of Michigan and author of a valuable blog on Iraqi affairs, suggests forming a federated order around Iraq's traditional 18 provinces—in part because they already enjoy social legitimacy, in part because hardening Iraq's ethnic divisions may create more problems. Few of these provinces are monolithic, and to act as if they are—to pronounce this one "Kurdish" and that one "Sunni"—could promote ethnic cleansing. (See George Packer's Oct. 4 article in The New Yorker on the explosive tensions between Kurds and Sunnis in Kirkuk.)
Cole boils the problem down to this: "Stop pretending Iraq is not an oil state." Oil states maintain order by paying off their key constituents. If the Iraqi state is to succeed, the Shiites and the Kurds will not be able to keep all their oil revenue. They'll have to pay off—share some of the bounty with—the Sunnis.
Will this end the insurgency? No. Some of the insurgents will reject any order. Some are foreign jihadists who are using the insurgency as a cover for killing Westerners (and their collaborators). But payoffs might placate the merely disgruntled insurgents (even Saddam had to pay off Fallujah's leaders to avoid trouble); and to the extent benefits are felt (a huge infusion of "blood money," to compensate for civilian relatives killed, would be a short-term boost), it could dry up the jihadists' base of support.
One important caveat: This idea—any idea—won't be accepted if it is seen as coming from the West. It has to be Iraqi in conception and implementation. Prime Minister Iyad Allawi hasn't shown any enthusiasm for Rubaie's "democratic regionalism"; he may have to start mustering some.
Finally, none of this means American troops can go home soon. They will be needed to protect Iraq's borders and, to some degree, to quell disorder. But if they are seen strictly as a peacekeeping force, and if the United States appears to be completely out of Iraqi politics, our erstwhile allies might be persuaded to take part—to serve not the occupation but the new Iraq. It's a gamble, but the current course is barely that.