In Afghanistan, why does counterinsurgency work in some places but not others?

By Rajiv Chandrasekaran
Sunday, July 25, 2010

MARJA, AFGHANISTAN -- The distance from here to success is only 15 miles.

There, in the community of Nawa, a comprehensive U.S. civilian-military counterinsurgency strategy has achieved what seems to be a miracle cure. Most Taliban fighters have retreated. The district center is so quiescent that U.S. Marines regularly walk around without their body armor and helmets. The local economy is so prosperous, fueled by more than $10 million in American agriculture aid, that the main bazaar has never been busier. Now for sale: shiny, Chinese-made motorcycles and mobile phones. There's even a new ice cream shop.

But here in Marja, the same counterinsurgency strategy has not suppressed the insurgent infection. Dozens of Taliban fighters have stayed in the area, and despite aggressive Marine operations to root them out, they have succeeded in seeding the roads with homemade bombs and sniping at patrols. The insurgent presence has foiled efforts to help and protect the civilian population: Taliban threats -- and a few targeted murders -- have dissuaded many residents from availing themselves of U.S. reconstruction assistance.

In my five trips to the area over the past year, Nawa has felt like progress, while Marja still feels like a war zone. Together, they illustrate the promise and limits of counterinsurgency in Afghanistan and the central challenge facing the new U.S. and NATO commander in Kabul, Gen. David H. Petraeus.

Marja and Nawa have much in common. Both are home to about 80,000 people, almost all of them ethnic Pashtuns. Both are farming communities where opium-producing poppies have been the cash crop of choice. Both are socially conservative southern Afghan backwaters, where tribal chiefs hold sway and women are rarely seen in public, even in head-to-toe burqas.

Both were stricken by the Taliban insurgency four years ago. And over the past year, both have been treated with America's new counterinsurgency formula: Each community has been flooded with U.S. Marines and Afghan security forces, at troop levels that meet or exceed what counterinsurgency theorists prescribe. Each has received a surge of cash and civilian experts in an effort to provide public services, rebuild infrastructure and dole out basic economic assistance. Each has been described as a priority by the central government in Kabul.

So why did all this work in one but not the other?

U.S. military officials contend that Marja needs more time to resemble Nawa. The Nawa operation began last July; efforts in Marja didn't start until February. But when the Nawa campaign was five months old -- where the Marja mission is now -- the district was just as quiet as it is today. The improvements in Nawa occurred quickly, and they seem to have lasted.

By now, Marja was supposed to be a success story as well, demonstrating to a skeptical public in America and Afghanistan that countering the insurgency with more troops, more money and a new strategy could resuscitate a foundering war. Perhaps more important, counterinsurgency proponents in the Pentagon and the State Department hoped to use both towns to make the case to President Obama that counterinsurgency works in Afghanistan and that he should attenuate -- or postpone outright -- his planned drawdown of troops starting next July.

The uneven progress complicates that case. But so, too, does Nawa, for those who maintain that counterinsurgency won't work in Afghanistan because the government is too corrupt, the security forces too inept and Pashtuns too distrustful of foreign troops.

So, which community is the rule? And which is the exception?

It is tempting, and perhaps fair, to view Marja as an outlier with unique tribal and geographic challenges. A patch of desert in Helmand province that was transformed into farmland by canals designed by American engineers in the 1950s, Marja was populated from scratch by the country's late king with settlers from a variety of tribes. The rank and file moved to Marja, but the chiefs didn't. This decades-old experiment in Afghan social engineering has now complicated efforts to find the same sorts of tribal leaders who influence the population in other Afghan communities. They simply don't exist in Marja.

CONTINUED     1           >

© 2010 The Washington Post Company