Directorate S

by Steve Coll

  “It was a cable station,” Riedel added skeptically.

  The newcomers made no effort to disguise their contempt for Bush’s close cultivation of Karzai and the mutual dependency they believed it had created.

  Holbrooke’s phone rang again. “It’s Zardari,” he confided. He walked out of the meeting, returned, and said he’d told the Pakistani president that he would call him back.

  The conversation turned to Obama’s dilemmas in managing the Pentagon and Petraeus. “This is not 1965,” Holbrooke pronounced. “I would know. I was there.”

  He was referring to President Lyndon B. Johnson’s escalation of the Vietnam War, based on ad hoc tactical decisions. Instead, Obama had commissioned the Riedel review to identify vital interests and goals.

  Yet the Riedel review they were celebrating was merely a light blueprint, Holbrooke acknowledged, not a full-blown military or diplomatic or intelligence plan. He had met Petraeus the night before, he said, to start to “operationalize” the review’s principles.25

  In fact, the review was an unusual, deadline-pressured exercise. It was odd for a new president to commission an outside consultant to study an inherited war. Riedel was a think tank scholar who had advised the Obama campaign but had no desire to enter the administration. Obama valued him and knew Riedel better than he knew James Jones, the tall former Marine general fluent in French whom Obama had appointed as national security adviser. Jones was not interested in running the Afghan review, he told Obama’s aides, because he had to oversee other foreign policy reviews, such as about Iraq.26

  Riedel was a well-informed, direct-speaking analyst, particularly on the subject of Pakistan. He was an ardent I.S.I. skeptic. He wrote well. All of the new team had read Lute’s longer, researched top secret review from the expiring Bush administration. That document made clear that the war wasn’t going well. The general policy direction of escalation in Afghanistan and renewed attention to Pakistan and I.S.I. seemed clear. Riedel offered a convenient solution to Jones’s deflection of responsibility.

  A week after Obama’s inauguration, Riedel had chaired a meeting in Room 445 of E.E.O.B., the same secure conference room where Lute had run the final Bush review, to announce his study process. “We’ll have a draft in a week,” he said. That seemed insanely fast. But Obama had declared that he wanted the review finished before he attended his first N.A.T.O. summit in early April. Another context for the work was that the White House had already defined “the big challenge” as “how to keep Holbrooke out” of the strategy formation, as a participant put it. This had not to do with policy differences but reflected a view that he was disruptive and hard to control. “The whole thing was weird and awkward and ugly, unsavory,” the participant said. Lute and others at the White House were pleased with Riedel’s work, but Robert Gates, who had stayed on as secretary of defense, thought the draft had “no new ideas.”27

  The review team sneakily arranged to brief the report to President Obama on a domestic Air Force One flight to Los Angeles when they knew Holbrooke would not be present. It was already becoming common for the White House to schedule meetings on Afghanistan on dates they knew Holbrooke could not make. The president made no fuss about this maneuver; his staff understood that Holbrooke’s theatrics put Obama off, and the more contact the two men had, the more alienated the president seemed to become.

  The Riedel review mattered because its findings would be distributed as binding presidential guidance to every agency involved in the Afghan war and policy toward Pakistan. Riedel’s final draft found that the United States had only one truly “vital” interest in the region: to defeat Al Qaeda. The highly classified annex identified another high priority, to prevent extremists from acquiring Pakistani nuclear weapons, but this was not something Obama’s advisers wanted him to highlight in public, because it was sensitive and might sound provocative or alarmist. America had other interests in the war, such as stability in South Asia and the reduction of heroin trafficking, but Al Qaeda trumped all others.

  On Al Qaeda, Riedel had proposed that America seek to “disrupt, dismantle and destroy” the organization “and its affiliates” in Pakistan and Afghanistan. The classified paper used that language. On the morning Obama was to deliver his speech announcing the findings, Mike Mullen and Bob Gates from the Pentagon joined him. Twenty minutes before the president took the lectern, Mullen took him aside. Speaking for Gates as well, he said, “We don’t like the term ‘destroy.’ It’s a really high bar.” Obama agreed to change the word to “defeat,” a somewhat looser benchmark. Riedel, who was standing behind the president as this drive-by editing of his work concluded, thought to himself, “The U.S. military doesn’t destroy things anymore?”

  The hard focus on Al Qaeda and its allies raised another question: Did Al Qaeda’s “affiliates” include the Afghan Taliban? This was undefined. From the beginning, Obama felt very strongly that the United States should not set an objective of defeating the Taliban, and he authorized background briefings to the press during Riedel’s study emphasizing that his Afghan war would focus on Al Qaeda. That frustrated some at the Pentagon who wanted to take on the Taliban fully. Riedel himself saw no daylight between the Taliban and Al Qaeda. Indeed, Riedel’s final draft endorsed a “fully resourced counterinsurgency strategy” at least in southern Afghanistan. Yet this did not necessarily equate to a war aim of fully defeating the Taliban; it might be enough, for example, to keep the Taliban at bay in the countryside while eradicating Al Qaeda and its transnational terrorist allies. Because there was disagreement on the point the paper left it vague, deliberately.28

  What did “fully resourced counterinsurgency” war mean? This was also undefined. Defense strategists often cited a ratio of 20 soldiers and police to every 1,000 local inhabitants in estimating the requirements of full counterinsurgency war. In that case Afghanistan might require as many as 600,000 troops and police, an unrealistic number anytime soon. The review elided this problem by defining the Afghan population base to be secured and subdued—in Kandahar, Helmand, and areas of the east—more narrowly. For example, it was not necessary to dispatch counterinsurgency forces to the Panjshir Valley, which was calm and whose population and leaders seemed firmly allied with N.A.T.O.

  The vagueness suited Petraeus. Just before the president’s public speech, the Principals Committee approved a decision to send 4,000 more troops to Afghanistan, in addition to the 17,000 approved in February. Only Biden opposed this escalation, arguing that it was politically unsustainable. Obama’s top aides at the White House and Gates at the Pentagon did not anticipate at the time they announced the Riedel review that Petraeus and Mullen would recommend sending yet tens of thousands more troops in just a matter of months. They assumed the commanders in the field would work with what they had. They were wrong.29

  —

  Gates and Mullen decided they needed to replace General Dave McKiernan, the American commander in Afghanistan. Obama had now endorsed a counterinsurgency strategy, however undefined. McKiernan was a conventional armored officer who had fought in Desert Storm. Like many of his generation in the Army, he had no experience with implementing counterinsurgency doctrine. He was a reserved man, unaccustomed to interagency meetings, and he had never been to Afghanistan before his assignment to lead the war. Gates flew to Kabul and relieved him of command over dinner.30

  Gates and Mullen had already settled on Stan McChrystal as McKiernan’s successor. McChrystal’s résumé as a hunter-killer in Special Operations left many colleagues doubtful about his ability to lead a strategy centered on winning hearts and minds. Yet McChrystal had served years in Afghanistan and his recent tour in Mullen’s office at the Pentagon had exposed him to the ideas he would be expected to implement. Watching two policy reviews up close, he had found the process “awkward at best” and he “saw little enthusiasm” for what he sensed was going to be needed to succeed in Afghanistan. During a brief meeting at the White Hous
e, Obama offered no specific guidance, but thanked him for taking on the responsibility. On Capitol Hill, the longtime chair of the House Armed Services Committee, Representative Ike Skelton, told him, “All you have to do is win.”

  McChrystal flew to Brussels to meet Ashfaq Kayani. N.A.T.O. had invited Kayani to talk to the alliance’s military leadership about his plans to win back control of Pakistani territory and to support N.A.T.O.’s effort in Afghanistan. Pakistan still provided N.A.T.O.’s principal land supply route to Afghan bases. As more American troops poured into Afghanistan after 2008, more supply ships arrived at Karachi ports carrying weapons, ammunition, fuel, food, and construction equipment. Pakistani trucks carried the goods hundreds of miles by road to U.S. bases.

  Kayani asked the room of European generals, “What should be the measure of success?” He offered an answer: “Are we winning the public opinion or not?”

  He urged N.A.T.O. to “look at the fundamental issues.” If N.A.T.O. sent more troops to escalate the war in Afghanistan, while adhering to assumptions that turned out to be wrong, there would be serious consequences, not least for Pakistan’s stability. What if the Bonn constitution’s approach to centralized government concentrated in Kabul turned out to be wrong, and Afghanistan needed a much more decentralized system? What if the assumption that the Afghan National Army could be built rapidly into an effective fighting force turned out to be wrong? This is not like maneuvering a speedboat, Kayani said. The core assumptions of N.A.T.O. strategy had to hold up over four or five years.31

  “The number of troops that you have—you won’t be able to do it,” Kayani said. “You have to control population centers and the roads that link them. But given the number of troops you have and the time constraints, you won’t be able to do it.”

  Outside of Brussels, Kayani met separately with McChrystal, Mullen, and Petraeus. He wanted McChrystal to redirect the new American forces to the east, along the Pakistan border. “I’m in the F.A.T.A. already,” Kayani said. “I can’t pull out anytime soon. Go to eastern Afghanistan first, then go south.” That would squeeze the militants in Waziristan who were rocking Pakistani cities.

  McChrystal countered that Kandahar and Helmand were the “center of gravity” in the war, the Taliban’s birthplace. “We have limited time. We have to hit the center of gravity.”

  “You don’t identify the center of gravity for the purpose of attacking it,” Kayani said, according to one participant. “You find ways to unbalance it without going straight at it.” He might have been describing I.S.I.’s twenty-year strategy against Kabul. “This will become a revolving door in the south—you’ll go in and out, the Taliban will go in and out.”

  The Americans were in no mood to take military advice from Kayani. Petraeus became aggravated. The last person he wanted advice from about the war in eastern Afghanistan was a general whose refusal to tear down the Taliban leadership in Quetta or to clean their militias out of North Waziristan was itself undermining N.A.T.O. strategy enormously. Pakistani sanctuaries were probably the biggest vulnerability in their military plan. The decision to go big in southern Afghanistan could not be undone, in any event. The Marines were in Helmand and Army task forces were headed to Kandahar. Petraeus made his irritation plain and Kayani went outside to cool off with a smoke.32

  Kayani had to accommodate multiple American policies—Central Command’s ambitious counterinsurgency against the Taliban, the C.I.A.’s lethal targeting of Al Qaeda by drones on Pakistani soil, the S-rap’s improvisational diplomacy, Biden’s blunt talk on behalf of the White House, and a push by the Defense Department’s office in Islamabad to train Pakistani troops. “Ten Wars,” Doug Lute’s damning classified slide deck about N.A.T.O.’s fractured chain of command in Afghanistan, might easily be adapted to describe American policy in Pakistan. Bruce Riedel had written an abstract of unified strategy for Obama. But every American agency still ran its own war on the ground.

  TWENTY-ONE

  Losing Karzai

  By 2009, the Arg Palace in Wazir Akbar Khan, where President Hamid Karzai lived, had recovered some of its grace from the degraded years of Taliban occupation. Palace staff tended the eighty-three-acre grounds, which held gardens, a pond, a mosque, and a parade ground. The staircase up from the main entrance displayed polished calligraphy inscribing “Allah” in gold script, ninety-nine different ways. The complex still contained a private residence for Karzai and his family. It also had to accommodate his official office, other offices for aides, and scores of security personnel. Karzai remained subject to a benign form of house arrest because of the risks of traveling outside. Kabul had evolved into a smoky, militarized city hunkered behind blast walls and razor wire. Tajiks in diverse hats wearing perpetual five o’clock shadows waved cars through checkpoints with the muzzles of their assault rifles. From his gardens Karzai could hear car horns, helicopter rotors, and the occasional distant thud of a suicide bomber.1

  In his private quarters, the greatest change in Karzai’s life was the arrival of his first son, Mirwais, born in 2007 to his wife, Zeenat. She was his first and only wife; Karzai had married her relatively late in life. Mirwais plainly brought joy to Karzai. Yet the president increasingly seemed a man who suffered. He was often ill with colds and sinus ailments. He kept vials of pills and vitamins nearby during his endless meetings. He struggled to eat enough, sometimes ordering cakes that he would finish off eagerly.2

  Karzai had a soft, sentimental side. He read poetry. His British interlocutors catered to his fondness for an imagined England of warm beer and immaculate cricket pitches. The British ambassador once delivered a boxed set of Karzai’s favorite television series, Last of the Summer Wine, a long-running middlebrow comedy about three friends in an English town.3

  To relieve stress, Karzai would spend hours walking in the Arg gardens. It was little wonder that he displayed signs of agitation, given his constricted circumstances. Yet by 2009 Karzai’s mood swings had become so visible and intense that it was no longer plausible to explain them merely as manifestations of cabin fever. The British government circulated reports that Karzai had been treated for psychological issues in Quetta and India earlier in his life, but the State Department could not document these accounts. In any event, there was no need for a medical record to confirm Karzai’s volatility.

  The president’s inner circle by now consisted mainly of Pashtun technocrats dependent upon his patronage, men with no substantial popular constituencies of their own. They included Rangin Spanta, who had fled to Germany during the Soviet war, where he became a professor of political science and an activist in the Green Party before returning after 2001. There was Zalmai Rassoul, another foreign affairs adviser, who had become a medical doctor in France during his years of exile, before joining Karzai. There was Hanif Atmar, a capable administrator who had connections with the Communist regime during the Cold War and later worked at a Norwegian nongovernmental organization. Rahim Wardak, a general who defected to the mujaheddin during the Soviet war, served as Karzai’s minister of defense. Umer Daudzai, a former administrator at the United Nations Development program, was his chief of staff. The Panjshiris who were still in Karzai’s orbit, such as Amrullah Saleh at N.D.S. and Ahmad Zia Massoud, the first vice president of Afghanistan, saw the president less frequently, in formalized cabinet meetings or at national security briefings. The sense among them was that U.S. ambassador Zalmay Khalilzad’s project to consolidate power for Karzai in 2004 had turned out to be a kind of ethnic putsch, reducing Tajik influence and elevating dependent Pashtuns around the president. Among the endless parade of foreign ambassadors and generals seeking Karzai’s time, he often favored the Indians and the Turks, although he had little choice but to accommodate the relentless Americans.

  Karzai raged openly at his aides, even in the presence of foreign visitors. His behavior became so bad that his aides and ministers arranged to discuss the problem at the compound of Kai Eide, the Norwegian di
plomat who was the U.N. representative to Afghanistan. They were concerned that Karzai was sabotaging Afghanistan’s relationship with the United States at a time when the country needed the alliance to survive and advance. In the garden, they asked, “What can we do with the President? We must stop this.”4

  They urged Eide to help. He wasn’t sure he could. Eide counseled Karzai to raise his concerns about American conduct of the war in a less confrontational way, but as Karzai came to fear treachery by the Obama administration during the first half of 2009, he grew angrier and angrier.5

  The president’s aides were sympathetic, up to a point. They agreed that the international media and the United States had mistreated Karzai. As one minister put it, Karzai went “from an Afghan Mandela” in 2002 to “an Afghan Mugabe” in 2007. That was not fair. But during 2009, Karzai’s own conduct crossed new lines.

  “If we can’t run the government, we should bring the Taliban back—to punish both the Americans and the Panjshiris,” Karzai declared one day to this minister, as he recalled it.

  Mirwais happened to be with them. “Do you want this boy to grow up under a Taliban regime?” the minister asked. “I don’t want that for my son.”

  They took a walk in the palace gardens. “Mr. President,” the minister said, “yes, I believe the United States was not fair to you. But they bring some good things. We should take some responsibility, too, for the things we have done wrong.”

  If the Kabul government collapses, he continued, “the U.S. will not be threatened, but we will be wiped out.”6

  In the first half of 2009, Karzai’s principal goal was to be reelected president. The best evidence of his sanity was his tactical skill in service of this ambition. He maneuvered deftly to sideline potential rivals, one by one, without revealing his own designs until it was absolutely necessary. He seemed particularly worried about Gul Agha Sherzai, whom he feared Obama had anointed.