by Steve Coll
It did not require an expeditionary battalion of anthropologists and political scientists to discover much of what was at issue. Daily reading of The New York Times, The Washington Post, and The Wall Street Journal alone would provide any Obama cabinet member a directionally reliable sense of rising Afghan anger over civilian casualties and intrusive night raids, at least in Pashtun areas of the country; popular disgust at the predatory Afghan government and police; and declining faith that American-led troops could defeat the Taliban.
If policy makers discounted journalism, they could read raw field reporting from the State Department, C.I.A., and D.I.A. that documented the same attitudes. In Zabul, Special Forces raids and casualties brought so many protesters into the streets at one point that the local governor feared public outrage might tip the balance and hand the Pashtun-dominated province to the Taliban. N.A.T.O. convoys following safety protocols continued to flee the scene when they ran over Afghan pedestrians or crashed into civilian taxis. Their lack of accountability provoked angry Afghan mobs to gather and sack the next foreign vehicle through.3
The same files documented the growing influence of the Taliban’s shadow governments around Afghanistan. Across the south, Taliban commanders ordered commercial Afghan cell phone providers backed by N.A.T.O. to shut down certain cell towers so that the guerrillas could infiltrate for attacks without being tracked by their phone signals. The Taliban destroyed switching stations if the carriers disobeyed.4
That summer in Helmand, as the Marines settled in, a State Department political officer touring the province described in a classified cable the residue of British occupation in New Zad, once Helmand’s second-largest city, now “Fallujah-like,” with packs of dogs roaming the streets, dead trees in the fields, and buildings abandoned “amid piles of rubble.” The officer met with elders and mullahs on the town’s outskirts and recorded the local Afghans’ advice. They wanted security, strongly preferred Afghan forces to American Marines, noted the absence of any Afghan government presence, and reported on the closure of all local schools, as well as the resilient influence of the Taliban. “We are like rocks here,” one man said. “You kick us, the Taliban kick us, no one listens to us.”5
From Tampa, Petraeus commissioned Marine Major General Douglas Stone to study Afghan prisons. In July, as Flynn and McChrystal settled in, Stone produced a seven-hundred-page classified report that shocked some of those in the Obama administration who read it. It found that Taliban commanders had effectively taken control of major Afghan prisons and were running sections of the war by cell phones from inside. (Hundreds of Taliban broke out of Sarposa Prison in Kandahar in the summer of 2008, the first of two mass escapes from that facility.)6
In July 2009, the National Security Council also reviewed intelligence reporting on “threat finance,” meaning the Taliban’s budget. The latest reporting listed the Taliban’s key financial sources as fund-raising in the Persian Gulf emirates, including Saudi Arabia and the United Arab Emirates, which brought in an estimated $100 million annually, and then drug dealing, protection rackets, local taxation, extortion, and kidnapping. Doug Lute noted to an interagency meeting that the Taliban appeared to be succeeding with very lean operating funds: “We spend $60 billion a year,” Lute remarked. “They need $60 million a year.”7
Gates had formally tasked McChrystal to conduct an assessment of the war by the end of August. Mike Mullen, the chairman of the Joint Chiefs, asked Jeff Eggers, a Navy SEAL who had completed graduate studies at Oxford University, to join the command team as his strategic adviser. McChrystal also recruited Christopher Kolenda, a West Pointer who had graduate degrees in national security studies, to his Kabul headquarters. They flew around Afghanistan for three weeks of interviews and fact-finding. Stone’s alarming prison study folded into their research. Following a model Petraeus pioneered in Iraq, McChrystal also invited sympathetic think tank specialists to Kabul to kibbitz about the study. It was a way to take outside advice but also to implant McChrystal’s ideas with influential op-ed writers whose work might shape American public opinion.
McChrystal’s effort constituted the third review of American strategy in the war in less than twelve months. As with the previous two, it suffered from a lack of clarity about the plausibility of counterinsurgency doctrine in Afghanistan. McChrystal was not necessarily advocating for a big war in Afghanistan. He was assessing the mission he had been given, in the form of the Riedel review. McChrystal knew he could not “defeat” the Taliban with the troops available, although it was not clear at this point whether that was truly America’s objective. Taking into account the typical requirements of counterinsurgency, it would be impossible to provide enough capable troops—international and Afghan—to secure the entire Afghan population anytime soon. McChrystal solved this by identifying eighty “key terrain districts” out of the four hundred administrative districts in Afghanistan. These included many with urban populations in the south and east. There were an additional forty-one “Areas of Interest.” Most of the territory in question lay in Helmand and Kandahar, as well as around Kabul and along the Pakistan border. The districts constituted a map of “ink spots” that might be secured and then linked together gradually through governance, aid, and security. Within this map the United States would now resource a new counterinsurgency campaign of a classical clear-hold-build-transfer type.
The eighty “key terrain districts” formulation, once accepted by the Pentagon and the White House, soon became an overdetermined engineering diagram. The plan birthed a jargon-filled language of acronyms and “District Stability Frameworks.” Pentagon briefers would exclaim slogans such as “It’s great to make sure the population is attached to the government.” The plan also suffered from the decision made before McChrystal’s arrival to deploy the Marines to Helmand. The entire province held just 4 percent of Afghanistan’s population yet it contained half a dozen key terrain districts that would absorb a large share of the new American military resources. The Marines were in Helmand by order of McChrystal’s predecessor and largely because their generals had demanded independence, or, as one of McChrystal’s aides put it, “We need our own science project.” The Marines were not even under McChrystal’s command at this point; they reported directly to Marine leadership at Central Command in Tampa, Florida. The problem of fractured command identified in the last Bush administration review remained almost a year later.8
At least the key terrain districts formulation solved the counterinsurgency math problem—by reducing the geographical scope of the campaign, enough American, N.A.T.O., and Afghan troops might be identified to carry it out, in rough line with traditional troop-to-population ratios. What the plan could not solve was the absence of an Afghan government capable of receiving the “transfer” of security and governance once N.A.T.O. cleared, held, and built.
“Afghan capacity is an illusion,” Sherard Cowper-Coles, the British envoy with two years of hard experience in Kabul, told Richard Holbrooke bluntly during a July 2009 meeting. The entire sequence of hold-build-transfer was “based on wishful thinking.”9
“Mr. President,” Lute advised Obama during this same period, “you can send a battalion of U.S. Marines, not only anywhere in Afghanistan, but literally anywhere in the world, and they will clear an area. Anywhere in South-Central Asia, a battalion of Marines is going to be so tactically dominant that they can clear that area. And as long as you are willing to keep them there, they can hold it. . . . The problem is handing the cleared area to the Afghans and doing something with it.”10
McChrystal and Flynn and their allies had two ideas about how to prove the doubters wrong. One was to attack corruption in Afghanistan and promote good governance. Their other idea was less conventional: McChrystal would listen to Hamid Karzai’s advice and make a concerted effort to reduce civilian Afghan casualties in the war, to try to make the American military presence more welcome and sustainable. “What is it that we don’t understand?�
� McChrystal asked colleagues at I.S.A.F. that summer. “We’re going to lose this fucking war if we don’t stop killing civilians.”11
As a career door kicker, McChrystal had the credibility to promote what he called “courageous restraint” in Afghanistan, even if it occasionally meant walking away from a firefight with the enemy. He hosted dinners with Afghan human rights advocates and European civil society groups to listen to their ideas about how to find a better balance between the use of force and the protection of civilians. Yet McChrystal endured grumbling and skepticism from the start within his own I.S.A.F. command, particularly from the Marines. Their doctrine emphasized merciless combat; they worried that the Taliban would exploit any new tactical rules promoting battlefield restraint. It was doubtful that there was a form of just-violent-enough warfare carried out by foreign armies and air forces that could inspire confidence among beleaguered, rural Afghan Pashtuns.
On August 30, McChrystal submitted his “Commander’s Initial Assessment” to Gates. An unclassified version of the report ran sixty-six pages. “Many indicators suggest the overall situation is deteriorating,” the commander acknowledged. “We face not only a resilient and growing insurgency; there is also a crisis of confidence among Afghans.” A “new strategy” was now required, one that would rapidly train Afghan forces to take the lead from N.A.T.O. while foreign forces, during the transition, carried out “classic counterinsurgency operations” in which “our objective must be the population.”
“Success is achievable,” McChrystal assured his superiors. It would require recognizing civilian casualties as a strategic issue, addressing Flynn’s insights about the failures of intelligence, and refusing to accept any longer “abuse of power, corruption or marginalization.”12
Karzai warmed to McChrystal that summer, even as he raged at Holbrooke over the envoy’s scheming to dump him. Yet as McChrystal sent his analysis to Washington, he discovered that Hamid Karzai’s vision of the Afghan war was fundamentally at odds with the precepts of his campaign plan.
—
During the last week of August, McChrystal, Greg Vogle, Ambassador Karl Eikenberry, and a few of their aides arrived at the Arg Palace to brief Karzai on McChrystal’s completed assessment of the war. They assembled around a table in Karzai’s small conference room next to the main cabinet room. McChrystal had distilled his study into about twenty PowerPoint slides. He presented them for about ninety minutes.
Eikenberry watched Karzai. At first, the president’s facial expression suggested that he might be thinking, “Why don’t I have a staff that puts up slides that look this good?” The president took out a notebook and started scribbling.
“Mr. President, do you have any questions?” McChrystal asked at last.
“Two, general. You say the situation is bad and getting worse. I disagree. You say the situation is ‘dire.’ Are you going to make the briefing public?”
McChrystal said he was, in summary form.
“We have to think about this,” Karzai continued. “We don’t want the Afghan public to panic. Is there another word for ‘dire’?” This inaugurated a search for euphemisms.
Karzai went on to a more fundamental observation. “You call this an insurgency,” he said. “This is not an insurgency. An insurgency, as I understand the meaning, suggests there are citizens of a country who are fighting against their government because they think the government is illegitimate. Now, we are a conservative, simple Muslim people. If they are fighting against an illegitimate government, then who are you, the United States? You are propping up an illegitimate government. No. There is no insurgency. There is a problem of international terrorism. We are allies in a battle against international terrorism.”13
Karzai believed, like many other Afghans, that the true story of the war—the essential problem—was not his own legitimacy but the mysterious unwillingness of the United States to challenge I.S.I. and Pakistan. Karzai did not seem to see himself as Afghanistan’s commander in chief. He eschewed the martial symbols of a wartime head of state. He didn’t seem sure that the Taliban and other fellow Afghans his security forces fought were truly enemies; they were merely the misguided hired hands of Pakistan.
Perhaps the boldest of McChrystal’s promises in his Commander’s Initial Assessment concerned Pakistan. The report’s key sentence was convoluted: “While the existence of safe havens in Pakistan does not guarantee I.S.A.F. failure, Afghanistan does require Pakistani cooperation.” Essentially, McChrystal asserted that the United States could achieve its goals even if the I.S.I. continued to provide the Taliban sanctuary.
Karzai believed that Pakistan should be the main effort of the American war. As Eikenberry once put it to him, “If you had a choice about where to deploy thirty thousand new American troops, you would put five thousand into training Afghan forces, five thousand along the border with Pakistan, and twenty thousand in the Federally Administered Tribal Areas,” inside Pakistan.
“That’s exactly the point,” Karzai answered. “You’re fighting a second-best strategy. You’re fighting Taliban foot soldiers in Afghanistan and destabilizing the country. You can’t play the game of saying Pakistan is your ally and telling me in private that they’re not.”14
For his part, Eikenberry doubted that McChrystal’s strategy could succeed. Along with Doug Lute and Joe Biden at the White House, he was a well-placed and increasingly vocal skeptic of the premises of counterinsurgency in Afghanistan. Among other things, the ambassador agreed with the C.I.A.’s Vogle and Wood that the Taliban insurgency could not be defeated unless the movement’s sanctuary in Pakistan was eliminated or at least badly disrupted.
Immediately after the meeting with Karzai in late August, Eikenberry wrote a series of three highly classified cables laying out his concerns. They were pursuing a war strategy that Karzai did not endorse. (Unlike a later cable Eikenberry wrote that fall containing similar analysis, this one did not leak.) McChrystal’s war strategy was premised on Karzai’s reliability, but the ambassador worried that Karzai’s growing waywardness did not seem to be registering with Obama’s cabinet and White House advisers. He tried to address this by inviting note takers to his meetings with the Afghan president. Eikenberry instructed the aides to write down Karzai’s extemporaneous remarks verbatim. Then he ensured that the classified cables contained long paragraphs of Karzai’s raw speech. He hoped it would shock Washington into recognition. He certainly got McChrystal’s attention. The commanding general reacted furiously to Eikenberry’s analysis and took the criticism personally. But Eikenberry and colleagues at the embassy felt the military wasn’t taking their views seriously. Karzai was playing the Americans by always scheduling separate one-on-one meetings with the ambassador, Greg Vogle, and McChrystal. One view at the highest levels of the U.S. embassy in Kabul by summer’s end was that Karzai “was a very clever madman—just because he was insane doesn’t mean he was stupid.”15
—
Every Monday at 5:00 p.m., Holbrooke assembled five dozen or so aides and outsiders at a “shura” meeting in Hillary Clinton’s principal conference room at the State Department. The purpose was to drive policy execution about Afghanistan and Pakistan across the government. On other days, Holbrooke ran smaller shuras to discuss sensitive subjects. Senator John Kerry attended one Wednesday that summer and asked Frank Archibald, the C.I.A. liaison to Holbrooke’s office, about the “‘humint’ situation in Afghanistan,” meaning human intelligence collection. Archibald said Kabul Station was “good” on counterterrorism, palace and electoral politics, and was “working on” Iran’s activity in the country. An “increased number” of analysts and officers were now “looking at the Taliban,” Archibald added.16
Mike Flynn wanted 360-degree intelligence collection about Afghanistan’s society and economy, from subtribal loyalties to graft networks that alienated the public. Holbrooke was curious about all that too but he was more interested in finding deeper in
sights into the Taliban’s leadership. Like Flynn, Holbrooke could see that battlefield intelligence produced reams of information about Taliban commanders and tactics but hardly any insight about the movement’s Pakistan-based leaders or their political and military strategy. Pakistan was no help on this score. Its army officers continued to inform American counterparts that autumn “that the Quetta Shura is a fiction,” as one Pakistani brigadier put it, nothing more than an “unsubstantiated fabrication.” Unfortunately, the United States had “fallen victim to rumors” that the Taliban’s leaders had a base in Pakistan, the brigadier said. Holbrooke chuckled at the denials but noted that, “on analysis, what they really deny is the word ‘Quetta.’” That was a fair point because many Taliban leaders appeared to live in Karachi, where they were even harder to trace. Pakistani generals had been lying to American counterparts about their support for the Taliban since the movement’s birth in 1994. It was evidently a hard habit to break.17
—
It was not until they received McChrystal’s dire assessment of the war that Lute and other aides to Obama understood that their field commander and Petraeus intended to recommend that tens of thousands more American troops be dispatched to Afghanistan to carry out the “key terrain districts” counterinsurgency strategy. Until then, the White House had thought the troops Obama ordered forward in February and March were all that would be required. At the time of the Riedel review in April, neither Gates nor Petraeus had said they would be looking for more soldiers. Petraeus, however, had been careful to make no promises. Informally, McChrystal now let the White House know that he might request up to eighty thousand more American troops. To Biden, Lute, and other skeptics in the White House, the news was stunning. Robert Gates, too, at the Pentagon, was surprised by the request.18
