NEW DELHI ― The United States, still mired in a protracted war in Afghanistan that has exacted a staggering cost in blood and treasure, will formally open peace talks with the Taliban, its main battlefield opponent, in the coming days (apparently despite last-minute opposition from Afghan President Hamid Karzai). With the U.S. determined to withdraw its forces after more than a decade of fighting, the talks in Doha, Qatar, are largely intended to allow it to do so “honorably.”

How the end of U.S.-led combat operations shapes Afghanistan’s future will affect the security of countries nearby and beyond. Here the most important question is whether the fate of Afghanistan, which was created as a buffer between Czarist Russia and British India, will be ― or should be ― different from that of Iraq and Libya (two other imperial creations where the U.S. has intervened militarily in recent years).

Foreign military intervention can effect regime change, but it evidently cannot reestablish order based on centralized government. Iraq has been partitioned in all but name into Shia, Sunni, and Kurdish regions, while Libya seems headed toward a similar tripartite, tribal-based territorial arrangement. In Afghanistan, too, an Iraq-style “soft” partition may be the best possible outcome.

Afghanistan’s large ethnic groups already enjoy de facto autonomy, which they secured after their Northern Alliance played a central role in the U.S.-led ouster of the Taliban from power in late 2001. Having enjoyed virtual self-rule since then, they will fiercely resist falling back under the sway of the Pashtuns, who have ruled the country for most of its history.

For their part, the Pashtuns, despite their tribal divisions, will not be content with control of a rump Afghanistan consisting of its current eastern and southeastern provinces. They will eventually seek integration with fellow Pashtuns in Pakistan, across the British-drawn Durand Line ― a border that Afghanistan has never recognized. The demand for a “Greater Pashtunistan” would then challenge the territorial integrity of Pakistan (itself another artificial imperial construct).

The fact that Afghanistan’s ethnic groups are concentrated in distinct geographical zones would simplify partition and make the resulting borders more likely to last, unlike those drawn by colonial officials, who invented countries with no national identity or historical roots. Indeed, both geographically and demographically, Afghanistan’s non-Pashtun groups account for more than half of the country, with Tajiks, Uzbeks, and Hazaras alone making up close to 50 percent of the population.

After waging the longest war in its history, at a cost of tens of thousands of lives and nearly a trillion dollars, the U.S. is combat-weary and financially strapped. But the American effort to cut a deal with the Pashtun-based, Pakistan-backed Taliban is stirring deep unease among the non-Pashtun groups, which suffered greatly under the Taliban and its five-year rule. (The historically persecuted Hazaras, for example, suffered several large-scale massacres.)

While Karzai has been fickle, to say the least, about cooperation with the Americans (indeed, he has since backed away from participation in the Doha talks), the rupture of his political alliance with non-Pashtun leaders has also fueled ethnic polarization. Some non-Pashtun power brokers continue to support Karzai, but many others are now leading the opposition National Front.

These leaders are unlikely to accept any power-sharing arrangement that includes the Taliban. In fact, they suspect that Karzai’s ultimate goal is to restore Pashtun dominance throughout Afghanistan.

Their misgivings have been strengthened by the “Peace Process Roadmap to 2015,” a document prepared by the Karzai-constituted Afghan High Peace Council that sketches several potential concessions to the Taliban and Pakistan, ranging from the Taliban’s recognition as a political party to a role for Pakistan in Afghanistan’s internal affairs. The roadmap even dangles the carrot of cabinet posts and provincial governorships to prominent Taliban figures.

The most serious problem today is that the country’s ethnic tensions and recriminations threaten to undermine the cohesion of the fledgling, multiethnic Afghan Army. Indeed, the splits today resemble those that occurred when Soviet forces withdrew from Afghanistan in 1989, an exit that led to civil war and the Taliban’s eventual capture of the capital, Kabul.

This time, the non-Pashtun communities are better armed and prepared to defend their interests after the U.S. withdrawal. Thus, in seeking to co-opt the Taliban, the U.S. is not only bestowing legitimacy on a thuggish militia; it also risks unwittingly reigniting Afghanistan’s ethnic strife, which would most likely tear the country apart for good.

This raises a fundamental question: Is Afghanistan’s territorial unity really essential for regional or international security?

To be sure, the sanctity of existing borders has become a powerful norm in world politics. Yet this norm has permitted the emergence of ungovernable and unmanageable states, whose internal wars spill across international boundaries, fueling regional tensions and insecurity.

With a war-exhausted U.S. having run out of patience, outside forces are in no position to prevent Afghanistan’s partition along Iraqi (or even post-Yugoslav) lines, with the bloodiest battles expected to rage over control of ethnically mixed strategic areas, including Kabul. In this scenario, Pakistani generals, instead of continuing to sponsor Afghan Pashtun militant groups (like the Taliban and their allies like the Haqqani network), would be compelled to fend off a potentially grave threat to Pakistan’s unity.

A weak, partitioned Afghanistan may not be a desirable outcome; but a “soft” partition now would be far better than a “hard” partition later, after years of chaos and bloodletting ― and infinitely better than the medieval Taliban’s return to power and a fresh reign of terror. Indeed, partition may be the only way to prevent Afghanistan from sliding into large-scale civil war and to thwart transnational terrorists from reestablishing a base of operations in the rubble.

By Brahma Chellaney

Brahma Chellaney is professor of strategic studies at the New Delhi-based Center for Policy Research. ― Ed.

(Project Syndicate)