It has been five years since Syria descended into brutal warfare, and whatever happens with the current talks on its political transition, it is difficult to imagine the country returning to its prewar borders as a unified state. Indeed, negotiators are beginning to think about something radically different: partition.
Speaking before the Senate Foreign Relations Committee last month, Secretary of State John Kerry implied that if the present cease-fire and political negotiations on Syria failed, partition could be Plan B. The Russians have openly proposed a federal solution, and, according to a United Nations Security Council diplomat, the idea of a “very loose center with a lot of autonomy for different regions” is gaining traction among major Western powers.
It is by no means the ideal solution: It would mean acquiescing to President Bashar al-Assad’s savagery over strident opposition objections, require still more internal displacement by way of sectarian relocation, and perhaps concede territory to the Islamic State.
Partition, or a loose confederation, might not even end the civil war and the accompanying humanitarian catastrophe. It would notionally leave intact an Assad-run statelet against which the Islamic State could continue to rally support, and a Sunni-run statelet that Alawis might seek to undermine.
More broadly, partition resulting from war has a deservedly dubious reputation: In Korea and Germany, for instance, it has meant the brutal sequestration of large populations and laid tripwires for future major wars. In India, it produced a huge demographic rupture and seemingly permanent enmity. The partition of Vietnam in 1954 proved unsustainably vulnerable.
In exceptional circumstances, partition can produce relative long-term stability and security in small and isolated populations. Cyprus has been partitioned for over 40 years, with United Nations troops patrolling the Green Line separating the Greek and Turkish Cypriot portions of the island pending theoretical unification. Similar stability has held in Northern Ireland, where the power-sharing Belfast Agreement has helped smooth over residual tensions. The formal partition of Bosnia under the Dayton Accords satisfactorily ended a terrible war among the Serbs, Bosnian Muslims, and Croats. But in all three cases partition was eased by Europe’s overall peace and stability, the presence of a powerful supranational body in the European Union that diluted the significance of putative national sovereignty and, most important, the relative absence of external provocation.
None of these conditions apply to Syria. The Syrian civil war has occurred in an extraordinarily tough neighborhood. The Arab League has been unable to ameliorate it. It has been stoked by outside parties — notably Iran and Russia on the regime’s side, and Saudi Arabia on the opposition’s — inclined to treat it as a proxy war. And the conflict has been bloodier and its combatants more plentiful and heavily armed than those in Cyprus or Ireland.
The only way to make partitioning work in Syria is for the outside powers to get involved — to agree to an armistice that would freeze the conflict and isolate the Islamic State, and to use force under United Nations auspices to maintain that armistice, facilitate partition and discourage regional actors from geopolitical meddling.
That, in turn, would require a large multinational peace enforcement effort, probably with an American/NATO component alongside Arab League troops and possibly Russian elements. Such a force would also have to oversee the relocation of Syrians who felt unsafe in their current situations.
Forging that kind of arrangement is no easy task: The different powers would have to figure out how to work together, agree to binding personnel and funding commitments, and deal with the inevitable and potentially substantial casualties. And it is unlikely that the United States, other NATO countries or Russia, let alone Saudi Arabia or Turkey, would support a large-scale combat deployment to enforce Syrian partitioning if that meant actively engaging the Islamic State — which may not otherwise be containable in a rogue statelet — and other jihadists on the ground. Even if an agreed force could avoid major ground operations against ISIS and gradually shift to a less muscular peacekeeping role, it would become increasingly brittle.
Other knotty challenges would remain, including contentious negotiations between the regime and the moderate opposition over who got control of which cities. The regime would insist on keeping Damascus, which would likely mean the mass relocation of Sunnis. (The drawback to partitioning Iraq 10 years ago was that Baghdad’s sectarian realignment would have uprooted populations even more traumatically than a dysfunctional unitary state has done organically.)
Meanwhile, a Kurd-dominated entity would be unacceptable to Turkey and would practically have to be scotched to preserve a NATO consensus supporting partition — a catch-22, and an unfair one, given how hard the Kurds have fought.
Finally, Syria’s partition, at least if it were permanent, would contravene the 1923 Lausanne Treaty and therefore would itself have to be legally effected by treaty, which is a tougher haul than a mere ad hoc agreement. Partitioning Syria could also spur destabilizing partitionist impulses elsewhere in the Middle East.
The argument for partition or loose confederation rests on the idea that there’s no better alternative: that it would take diplomatic alchemy to achieve a unitary Syria with a power-sharing government. But diplomacy has never been a science, and the thinking behind partitioning is scarcely less magical. Given that a unitary state is by far the neater and more desirable alternative, Washington should continue to pursue it within the existing framework. Just because it isn’t yet imagineable doesn’t mean it isn’t possible.
Jonathan Stevenson, a senior fellow at the International Institute for Strategic Studies, was the director for political-military affairs for the Middle East and North Africa at the National Security Council from 2011 to 2013.