Ghost Wars

by Steve Coll

  Reno’s disapproval mattered because National Security Adviser Sandy Berger sought a consensus within the Cabinet about the exact wording of the CIA’s instructions. Even though they felt they were on very solid legal ground, the language they were working with month after month, memo after memo, lived in uncomfortable proximity to the long-standing White House ban on assassinations. They did not want Reno to develop dissents to Clinton’s decisions about bin Laden in this area. In the midst of the impeachment mess, none of them wanted to wake up to a newspaper headline that read: ATTORNEY GENERAL OBJECTS TO CLINTON’S TERRORIST ASSASSINATION PLANS. So Jamie Baker’s group drafted and redrafted language to accommodate Reno’s concerns. The resulting consensus formulations, conceded one White House senior official involved, were often convoluted and “Talmudic.”

  More broadly, the president’s covert policy—as fashioned by Sandy Berger, his deputy James Steinberg, Richard Clarke, and the national security cabinet—pursued two different goals at the same time. On the one hand, they ordered cruise missile–equipped submarines to patrol secretly under the Arabian Sea. They hoped to use the submarines to kill bin Laden if they could find him sitting still long enough to strike. On the other hand, they authorized the CIA to carry out operations designed at least on paper to take bin Laden alive. The Small Group debated “whether to consider this a law enforcement matter demanding a judicial response or a military matter in which the use of armed force was justified,” Madeleine Albright recalled. “We decided it was both.” William Cohen argued that debate over war versus law enforcement was a “false choice”; all instruments of American power were required at once.

  The split policy reflected unresolved divisions inside the national security cabinet. Attorney General Janet Reno and FBI director Louis Freeh, along with others at Justice, had invested themselves deeply in the law enforcement approach to terrorism. American counterterrorist policy had since 1986 emphasized bringing terrorists to justice in courtrooms. Even though killing bin Laden would be legal under American law, some at Justice and the White House nonetheless felt uneasy at times about that approach. There might be unintended consequences. They had been willing to endorse the August cruise missile strikes in the immediate aftermath of the embassy bombings. There was a sense of proportion in those attacks. Now, as an ongoing matter, some of them preferred to seek bin Laden’s arrest, not to launch a low-grade war.21

  Clinton himself seemed to lean in both directions. If anything, by his actions and decisions the president seemed to favor lethal force against bin Laden and al-Zawahiri if he could find a way to make an immaculate strike. “Clinton’s desire to kill them was very clear to us early on,” recalled one of his senior aides. But he did not commit himself all the way. The first MON he signed in the summer of 1998 authorized covert action aimed at taking bin Laden and his aides into custody for trial. The ambiguous language might have been crafted to assure Janet Reno’s support, but Clinton etched his own signature on the memo. Yet the president’s second MON explicitly authorized bin Laden’s death in one narrow set of hypothetical circumstances—without overriding the general order in the first memo.

  At one of Richard Clarke’s Counterterrorism Security Group meetings that autumn, they reviewed intelligence about how bin Laden moved around Afghanistan. Sometimes he traveled by road in heavily armed convoys of Land Cruisers. Occasionally, however, he flew in helicopters and aircraft maintained by al Qaeda in conjunction with the Taliban’s small air force. The CIA received occasional reports from its tribal tracking team and other sources about bin Laden’s flights. They wanted to be certain their agents had legal permission to shoot at a helicopter or airplane if they knew that bin Laden was on board. The Pentagon also ordered planning late in 1998 for operations to intercept al Qaeda aircraft. Downing an airplane was not an operation likely to produce an arrest or capture, so it did not seem to be covered by the prevailing MON. Also, such an attack could violate international treaties banning air piracy. This was an area the National Security Council lawyers often worried about: A covert operation might be legal under domestic U.S. law, but it might at the same time violate American treaty commitments abroad. This could lead other countries to abandon their pledges under international treaties. Also, in some cases the United States had passed laws making any treaty violation a domestic crime. A MON might permit crimes abroad but nonetheless place an individual CIA officer in legal jeopardy inside the United States.22

  Jamie Baker’s office presented a new MON for Clinton’s signature. It would authorize the CIA or the Pentagon to shoot down bin Laden’s helicopters or airplanes under certain circumstances. There was no pretense in this MON that bin Laden would be captured for trial. Clinton signed it.

  The president had now authorized the CIA to capture bin Laden for trial and, separately, to kill him. Pentagon planning was equally divided: A December 1998 order sought options for capturing al Qaeda leaders and transporting them from Kandahar, while other plans contemplated stand-off air strikes. Some CIA managers saw their instructions from the White House as legalistic, restrictive, and ambiguous. The drafts of more straightforward proposed instructions they sent over to the White House from Langley came back full of abstract phrases open to multiple interpretations. The CIA received no “written word nor verbal order to conduct a lethal action,” one official involved recalled. “The objective was to render this guy to law enforcement.” Under its written authorities from the White House the CIA had to recruit agents “to grab [bin Laden] and bring him to a secure place where we can turn him over to the FBI.” Some CIA managers saw a big difference between the August 1998 MON language and a pure lethal action. “If they had said ‘lethal action,’ it would have been a whole different kettle of fish and much easier,” the official recalled. Credible planning and supervision of an arrest operation inside Afghanistan, transfer to FBI agents, and extraction to the United States was far more complicated than planning for a lethal strike. The exact language Clinton sent to Langley in his bin Laden-related MONS zigzagged on the issue of lethal force. The first document after the embassy bombings said the CIA’s tribal agents could use lethal force during a capture operation only in self-defense. The TRODPINTS were told they would only be paid if they captured bin Laden, not if they killed him. At the end of 1998 Clinton reversed course and approved paying the tribals either way, as long as they did not execute prisoners or otherwise grossly violate the rules. A new memo during this period also authorized the CIA’s agents to kill bin Laden if capturing him did not look feasible. Yet Clinton later signed at least two other classified memos about operations against bin Laden that reverted to the earlier, less permissive language. The changes demoralized CIA field officers and encouraged them to believe that they and their Afghan allies would be held to account on issues of legal nuance.23

  White House aides saw the same instructions as providing the clearest possible signal that the CIA should get after bin Laden and his leadership group and kill them if necessary. Capture for trial was the stated objective of the August MON, yes, but the White House aides believed they had written the document to provide the CIA with the maximum flexibility to kill bin Laden in the course of an arrest operation. All of them, including the CIA’s managers and lawyers, knew that as a practical matter bin Laden and his bodyguards would resist capture. These were committed jihadists. They would likely martyr themselves long before they were handcuffed. Under the White House’s authorities, as soon as bin Laden’s men shot back, the CIA’s several dozen armed Afghan agents could take them out. Also, as the months passed and new MONs were written, the CIA’s authorizing language, while still ambiguous, was changed to make the use of lethal force more likely. At first the CIA was permitted to use lethal force only in the course of a legitimate attempt to make an arrest of bin Laden or his top aides. Later the key language allowed for a snatch operation or a pure lethal attack if an arrest was not plausible.

  Clinton’s aides thought the CIA’s managers were using the leg
al issues as a dodge. The agency sometimes seemed to believe that under the MON, “unless you find him walking alone, unarmed, with a sign that says ‘I am Osama’ on him, that we weren’t going to attempt the operation,” one White House official involved recalled. “I think we were concerned that there were too many people [at Langley] who will just see the downsides and not enough people motivated to get the job done.” Yet CIA leaders and lawyers alike interpreted their instructions the same way—as orders to capture, not kill, except in certain circumstances.24

  Sandy Berger later recalled his frustration about this hidden debate, confined at the time to only a few dozen officials and lawyers with the proper security clearances: “It was no question, the cruise missiles were not trying to capture him. They were not law enforcement techniques.” Berger said that if “there was ever any confusion, it was never conveyed to me or the president by the DCI or anybody else.”25 What the White House needed most was “actionable intelligence” about bin Laden’s precise location. They depended on the CIA to provide it. The agency had ample authority to put its Afghan agents into action, Berger believed.

  The tension festered. It would not be resolved anytime soon.

  IN THE SAME WEEK that bin Laden’s operatives struck two U.S. embassies in Africa, Mullah Omar’s turbaned Taliban soldiers, their ranks swollen with jihadist volunteers from Pakistan’s madrassas and aided by officers from Pakistani intelligence, finally captured their last major prize in the north of Afghanistan: the sprawling city of Mazar-i-Sharif. “My boys and I are riding into Mazar-i-Sharif,” the longtime ISI Afghan bureau officer Colonel Imam, once a close partner of the CIA, boasted in an intercepted telephone call at the height of the battle.26

  Mazar’s defenders, commanders allied with Ahmed Shah Massoud, succumbed to bribes paid by Pakistani officers, Massoud told his men at a military assembly. The leading local warlord, Abdul Malik, “delivered his city for a fistful of dollars,” Massoud declared.27 Massoud and his militias still controlled the northern town of Taloqan, but increasingly they were being painted into a corner.

  Just weeks after the embassy bombings Massoud wrote a letter to the United States Senate urging that America help him in his war against the Taliban, Pakistani intelligence, and bin Laden. After the expulsion of Soviet troops, Massoud wrote, Afghanistan’s people “were thrust into a whirlwind of foreign intrigue, deception, great-gamesmanship and internal strife… . We Afghans erred, too. Our shortcomings were a result of political innocence, inexperience, vulnerability, victimization, bickering and inflated egos. But by no means does this justify what some of our so-called Cold War allies did to undermine this just victory.” Pakistan and its Arab Islamist allies had fielded twenty-eight thousand paramilitary and military forces in Afghanistan to aid the Taliban’s drive for conquest, Massoud wrote. Afghanistan had been delivered to “fanatics, extremists, terrorists, mercenaries, drug mafias and professional murderers.” America should help him turn them away. Washington should break its long debilitating dependence on Pakistan in shaping its Afghan policies, Massoud urged.28

  But the Clinton administration, especially diplomats at the State Department, remained disdainful of Massoud and his pleas. With the fall of Mazar, the Taliban seemed more than ever an irreversible force inside Afghanistan. Madeleine Albright, Undersecretary Tom Pickering, and regional specialists in State’s South Asia bureau all recommended that the administration continue its policy of diplomatic engagement with the Taliban. They would use pressure and promises of future aid to persuade Omar to break with bin Laden. The U.S. embassy in Islamabad promoted this argument in its cables to Washington. Most State diplomats saw Ahmed Shah Massoud as a spent force tainted by his recent deals to accept arms supplies from Iran and by his reliance on heroin trafficking for income. Some at State, including Inderfurth, said later that they thought it was useful for Massoud to remain viable as a military force in northern Afghanistan because he offered a check on the Taliban’s cross-border Islamist ambitions in Central Asia. But from Albright on down, the State Department certainly was not prepared to join Massoud’s military campaign against the Taliban.29

  State diplomats sought to convince the Taliban’s leaders that America did not see them as the enemy, that the United States was targeting only bin Laden and his Arab lieutenants. The August cruise missile attack “was not directed against Afghanistan or the Taliban,” Assistant Secretary of State Rick Inderfurth explained in October 1998. The Taliban “need to understand that by harboring terrorists, they are becoming increasingly complicit in the acts those terrorists commit.” But there was still time for the Taliban to change its stripes. “We urge the Taliban to respond,” Inderfurth declared. “If it does not, we will have to respond accordingly and adjust our policies.”30

  The underlying premise of this outreach, rarely stated aloud so as to preserve America’s bargaining position, was to trade U.S. diplomatic recognition of the Taliban as Afghanistan’s legitimate government in exchange for custody of Osama bin Laden. Among other things, State’s diplomats hoped Pakistan and its intelligence service would use their presumed leverage over the Taliban to help cut this deal. In effect this was the continuation of an American policy that had long been willing to accept Pakistani hegemony over Afghanistan in the name of regional stability.

  AT THE HEART OF the matter lay an unresolved factual and policy question: Who was the enemy? There was a second question, also unresolved: How dangerous, really, was the threat?

  By the late 1990s, presiding over a historic economic and stock market boom, Bill Clinton had concluded that terrorism—along with ethnic war, diseases such as AIDS, and regressive religious regimes like the Taliban—represented “the dark side” of the “breathtaking increase in global interdependence” witnessed since the collapse of communism. Satellites, air travel, and more recently the World Wide Web had collapsed time and space, suddenly forcing disparate nations and religions and cultures into roiling interaction. America had reaped enormous benefits from these changes. Its rapidly rising wealth flowed directly from “tearing down the walls, collapsing the distances, and spreading the information that we have across the world,” as Clinton put it later. Yet at the same time “you cannot collapse walls, collapse differences, and spread information without making yourself more vulnerable to forces of destruction.” Clinton believed that America’s mission was to accelerate these trends, not resist them. He sought to lead the country and the world from a period of global “interdependence” to one of more complete worldwide “integration.” Terrorist attacks were a “painful and powerful example of the fact that we live in an interdependent world that is not yet an integrated global community,” he believed. Yet Clinton did not want to build walls. He saw the reactionary forces of terrorism, nationalism, and fundamentalism as inevitable; they were intricately connected to the sources of global progress. They were also doomed. In human history, he asserted with questionable accuracy, “no terrorist campaign has ever succeeded.”31

  More specifically, Clinton saw bin Laden and Islamic radicals like him as part of a long historical continuum of “fanatics” who “think they’ve got the truth, and if you share their truth, your life has value. And if you don’t, you’re a legitimate target.” Clinton often described his own Christian faith—shaped in part by his exposure at Georgetown University to the Jesuit tradition—as rooted in a search for God that was constrained by human fallibility. “Most of us believe that no one has the absolute truth,” Clinton said. “As children of God, we are by definition limited in this life, in this body, with our minds.” Life could only be “a journey toward truth,” never fully completed until salvation. The Taliban, bin Laden, and al Qaeda had “very different ideas [than] we have about the nature of truth, the value of life.”32

  Clinton was prepared to “take Mr. Bin Laden out of the picture” if he could, he said later. Yet he defined the broader purpose of his foreign policy as one that would “spread the benefits” of global integration and “reduce the ri
sks” of terrorism by making “more partners and fewer terrorists in the future.” He was inclined to see bin Laden as an isolated fanatic, flailing dangerously but quixotically against the forces of global progress.33

  Most of the Clinton administration’s debates about counterterrorism policy took place far from public view. Some of the most pointed occurred within the Counterterrorism Security Group where virtually every memo was highly classified. Here the CIA’s main representative, Paul Pillar, joined tense, sometimes hostile debates with Richard Clarke and his principal counterterrorism aides, Steven Simon and Daniel Benjamin. Their day-to-day arguments involved some of the most critical strategic issues.

  Their discussions were substantive, intellectual, and visceral. They involved basic questions about modern terrorism, bin Laden’s network, its threats, and American policy. All four men were exceptionally intelligent and well spoken. They were bookish, intense, well read, nervous, and argumentative. Their disagreements had the hyperarticulate character and unyielding passion of ideological disputes among Ivy League faculty. The hours they worked together were long beyond count, and the pay was mediocre. Yet they were debating day to day the most important issues in their country’s clandestine war against bin Laden. The pressure was almost unbearable. There was little reward for being right in these disputes. There was the continual potential of catastrophe for being wrong.

  The four of them agreed about a great deal. Their differences were often subtle, yet they were also substantial. As the longtime deputy director of the Counterterrorist Center, Pillar wielded great influence over the CIA’s terrorism analysis. Along with Clarke, Simon and Benjamin were instrumental in White House counterterrorism policy in the first year after the Africa embassy bombings.