Partition won’t work

Yesterday Tom Ricks published at foreignpolicy.com a brief piece I wrote for him a couple of weeks ago on why partition of Iraq and Syria is a really bad idea. The basic reason is this:  separation of ethnic or sectarian groups sounds good, but unless they agree on the lines of separation they will sooner or later fight over where to draw them. Agreements about lines of separation are rare. Czechs and Slovaks are the classic case. Far more often, the parties disagree.

Partition proposals don’t prevent war. They cause it.

This is especially true for Iraq and Syria.

In Iraq, there are substantial areas of relative homogeneity:  most people who live in Kurdistan are Kurds, most who live in Anbar and large parts of Ninewa are Sunni, and most who live south of Baghdad are Shia. But that doesn’t mean they would agree on the lines of separation.

For Sunnis, especially for those who support restoration of the caliphate, Baghdad is vital, even though it now has a population that is majority Shia (and partly Kurd). Nor will Sunnis be pleased to see Shia walk off with the lion’s share of Iraq’s massive oil reserves, which lie in the south, or Kurds walk off with much of the rest, which lies in Kurdish-controlled Kirkuk and other so-called disputed territories. Anbar’s natural gas will be little comfort, as it will take years to develop it and build pipelines to ship it out. Partition of Iraq will lead to a war likely to last a decade or more, as Sunnis seek to recover territory they regard as their own. Any guess about which Sunnis, moderates or extremists, will lead that fight?

The situation is even more complicated in Syria, where the same degree of ethnic and sectarian separation does not exist. There are islands of minorities (Kurds, Druze, Christian, Shia and other) spread out in an arc of mostly Sunnis extending from the southern border with Jordan and Israel, through Damascus, Homs and Aleppo to the north and along the Turkish to the Iraqi border. Kurds are not concentrated in one area, and only one of the three areas where they live is contiguous with Iraqi Kurdistan. Alawites are concentrated in the west along the Mediterranean coast, where they are not the majority in many communities, and inland in Damascus, where they are also not the majority.

Division of Syria along ethnic and sectarian lines would therefore mean moving millions of people, in addition to the half of the population that has already been displaced. There is really no way to do that except by force.

Let’s say however that ISIS succeeds in continuing to dominate the Sunni-majority parts of eastern Syria and western Iraq. Should the international community accede to that and hope the jihadists can be contained? Fat chance. They would continue to fight at least for Damascus and Baghdad, which are the historic capitals of the caliphate. And in the meanwhile they would provide safe haven for international terrorists like Khorasan, the Al Qaeda affiliate embedded for now with Jabhat al Nusra, which has just reached an accommodation with the Islamic State.

But, you might ask, aren’t the existing borders artificial? Yes, they are, but it is instructive that they were not established by Sykes and Picot, who are usually cited as the culprits. The map they signed in 1916 had Mosul in the French zone (which is the ancestor of Syria), not the British (which is the ancestor of Iraq):

When ISIS captured Mosul, it was not destroying the Sykes/Picot division but restoring it, in part. The lines we attribute to Sykes and Picot today were drawn in 1923, by the Treaty of Lausanne, though Mosul’s fate was still uncertain (and was supposed to be determined by the League of Nations). But the British were already there and kept it.

The virtue of the existing lines is just that:  they exist. Moving them necessarily creates winners and losers. If the losers are not happy with the result, they will fight. Partition won’t work.