45. "WHY IN THE
WORLD DIDN'T WE
KNOW?"

Fred Hitz, the CIA's inspector general, said his job was to walk through the battlefield while the smoke cleared and shoot the wounded. His internal investigations were painstaking and pitiless. He was old-school agency, recruited in his senior year at Princeton after being tapped by the dean of students. As fate would have it, his biggest case concerned his classmate from the CIA's career-training cadre of 1967, an alcoholic burnout from the old Soviet division by the name of Aldrich Hazen Ames.

On Presidents' Day, February 21, 1994, a team of FBI agents hauled Ames out of his Jaguar as he left his suburban home for headquarters, slapped on the handcuffs, and took him away forever. I went to see him in the Alexandria county jail after his arrest. He was a gray man of fifty-three who had been spying for the Soviets for nearly nine years. He would soon be sent to a lifetime of solitary confinement, and he was eager to talk.

Ames was a malcontent and a malingerer who got a job with the agency because his father had once worked there. He spoke passable Russian and wrote readable reports when sober, but his personnel records were a chronicle of drunkenness and ineptitude. He had failed upward for seventeen years. In 1985, he had reached a pinnacle: chief of counterintelligence for the Soviet Union and Eastern Europe. He was known to be an alcoholic malcontent. Yet the agency gave him access to the files of nearly every important spy working for the United States behind the iron curtain.

He had become contemptuous of the CIA. He thought it absurd to say that the Soviet threat to the United States was immense and growing. He decided that he knew better. "I know what the Soviet Union is really all about, and I know what's best for foreign policy and national security," he remembered thinking. "And I'm going to act on that."

Ames obtained permission from his superiors to meet with an officer from the Soviet embassy in Washington, pretending that he could recruit the Russian. In April 1985, in exchange for $50,000, he had handed the Soviet intelligence officer the names of three Soviet citizens who were working with the CIA. Then, a few months later, he named every name he knew. Moscow set $2 million aside for him.

One by one, America's spies inside the Soviet Union were arrested, tried, imprisoned, and executed. As they died, Ames said, "bells and whistles" went off inside the clandestine service. "It was as if neon lights and searchlights lit up all over the Kremlin, shone all the way across the Atlantic Ocean, saying, 'There is a penetration.'" Yet the CIA's leaders refused to believe that one of their own had betrayed them. Using double agents and deception, the KGB skillfully manipulated the CIA's perceptions of the case. It had to be a bug. It could not be a mole.

Ames also gave Moscow the identities of hundreds of his fellow CIA officers and a thorough rundown on their work. "Their names were given to the Soviet intelligence service, as were the details of a number of operations that the United States was engaged in," Hitz said. "This began in 1985, but continued until one or two years before his arrest, and Ames was an avid gatherer of information to supply to his Soviet case officer. So in strict intelligence terms it was a horror."

The agency knew that something had destroyed its Soviet operations. But it took seven years to begin to face the facts. The CIA was unable to investigate itself, and Ames knew it. "You would wind up with people throwing up their hands and saying, 'We can't do it,'" he said with a smirk. "You've got two or three or four thousand people running around doing espionage. You can't monitor it. You can't control it. You can't check it. And that's probably the biggest problem with an espionage service. It has to be small. The minute you get big, you get like the KGB, or you get like us."

"A VIOLATION OF COMMANDMENT NUMBER ONE"

It took Hitz more than a year after the arrest to assess the damage Ames had wrought. In the end, he found that the CIA itself had been part of an elaborate deception.

Among the most highly classified papers that the agency produced during and after the cold war were "blue border" reports, with a blue stripe on the side signifying their importance, assessing the strength of Moscow's missiles, tanks, jets, bombers, strategy, and tactics. They were signed by the director of central intelligence and sent to the president, the secretary of defense, and the secretary of state. "That is what the intelligence community exists to do," Hitz said.

For eight years, from 1986 to 1994, the senior CIA officers responsible for these reports had known that some of their sources were controlled by Russian intelligence. The agency knowingly gave the White House information manipulated by Moscow--and deliberately concealed the fact. To reveal that it had been delivering misinformation and disinformation would have been too embarrassing. Ninety-five of these tainted reports warped American perceptions of the major military and political developments in Moscow. Eleven of the reports went directly to Presidents Reagan, Bush, and Clinton. They distorted and diminished America's ability to understand what was going on in Moscow.

"This was an incredible discovery," Hitz said. The most senior CIA official responsible for these reports insisted--as Ames had done--that he knew best. He knew what was real and what was not. The fact that the reporting had come from agents of deception meant nothing. "He made that decision himself," Hitz said. "Well, that was shocking."

"What came out of this whole episode was a feeling that the agency couldn't be trusted," Hitz said. "In short, it was a violation of Commandment Number One. And that's why it had such a destructive impact." By lying to the White House, the CIA had broken "the sacred trust," Hitz said, "and without that, no espionage agency can do its job."

"THE PLACE JUST NEEDS A TOTAL OVERHAUL"

Woolsey acknowledged that the Ames case revealed an institutional carelessness that bordered on criminal negligence. "One could almost conclude not only that no one was watching, but that no one cared," he said. But he announced that no one would be dismissed or demoted for the "systemic failure" of the CIA in the Ames case. Instead, he sent letters of reprimand to six former senior officers and five still on duty, including the chief of the clandestine service, Ted Price. He defined the failures as sins of omission and blamed them on a flawed culture within the CIA, a tradition of arrogance and denial.

Woolsey presented his decision to the House intelligence committee on the afternoon of September 28, 1994. He made a bad impression. "You have to wonder whether the CIA has become no different from any other bureaucracy," the committee's chairman, Dan Glickman, a Kansas Democrat, said upon emerging from the meeting. "You have to wonder if it has lost the vibrancy of its unique mission."

The Ames case created an attack on the CIA that was unprecedented in its intensity. It came from the right and it came from the left and it came from the dwindling center of American politics. Anger mixed with ridicule--a deadly brew--flowed from the White House and Congress. There was a strong sense that the Ames case was not an isolated aberration but evidence of a structural dry rot. Lieutenant General Bill Odom, who had run the National Security Agency under President Reagan, said the solution was radical surgery.

"I would disembowel the CIA," he said. "It's contaminated. And if you take halfhearted measures it will remain contaminated."

Striving to defend the agency from without and within, Woolsey promised the American people that they had a right to ask where the CIA was headed. But he had lost his ability to chart that course. So on September 30, 1994, Congress created a commission on the future of the CIA and gave it the power to blaze a new path for the agency in the twenty-first century. The Ames case had created a once-in-a-generation chance for change.

"The place just needs a total overhaul," said Senator Arlen Specter of Pennsylvania, a Republican who had served six years on the Senate intelligence committee.

What was needed was a push from the president of the United States, which never came. It took three months to select the seventeen members of the commission, four months to draft an agenda, and five months before the panel held its first formal meeting. The commission was dominated by members of Congress, notably Representative Porter J. Goss of Florida, a conservative Florida Republican. Goss had spent an undistinguished stint with the clandestine service in the 1960s, but he was the only member of Congress who could claim hands-on experience at the agency. The commission's most distinguished outsider was Paul Wolfowitz, who came to the table thinking that the CIA's ability to gather intelligence through espionage had collapsed, and who would be among the most influential members of the next president's inner circle.

The commission was led by Les Aspin, who had lost his job as secretary of defense nine months before, fired for his inability to make decisions. Clinton had named him chairman of the President's Foreign Intelligence Advisory Board. Depressed and disorganized, Aspin asked big questions without clear answers: "What does it all mean now? What are the targets now? What are you trying to do?" When he died suddenly of a stroke at fifty-six a few months later, the commission's staff was despondent and its work went adrift. The commissioners headed in a dozen different directions, unable to decide on a destination.

The staff director, Britt Snider, proclaimed: "Our goal is to sell intelligence." But many of the witnesses were warning that salesmanship was not the issue. It was the product.

The commission finally convened and took testimony. Bob Gates, who had drawn up the long list of 176 threats and targets three years before, now said the agency was overwhelmed by the multiplicity of tasks. Case officers and station chiefs said the clandestine service was drowning under too many requests to do too many small-bore things too far afield. Why was the White House asking the CIA to report on the growth of the evangelical movement in Latin America? Was that really important to the national security of the United States? The agency was only capable of a few major missions. Tell us what you want us to do, the CIA's officers begged.

But nothing focused the commission. Not the March 1995 attack by a religious cult that poured sarin gas into the Tokyo subways, killing 12 people and injuring 3,769, an event that signified the transformation of terrorism from nation-states to the self-anointed. Not the April 1995 bombing of the federal headquarters in Oklahoma City, which killed 169 people, the deadliest attack on American territory since Pearl Harbor. Not the discovery of a plot by Islamic militants to blow up a dozen American airliners over the Pacific and crash a hijacked jet into CIA headquarters. Not the warning from a CIA officer that someday the United States would face "aerial terrorism"--an airplane dive-bombing a target. Not the fact that a total of three people in the American intelligence community had the linguistic ability to understand excited Muslims talking to each other. Not the realization that the ability of the CIA to analyze information was being drowned by the explosion of e-mail, personal computers, cellular telephones, and publicly available encryption for private communications. Not the growing realization that the CIA was in a state of collapse.

The report, seventeen months in the making, had no weight and no impact. "Counterterrorism received little attention," said Loch Johnson, a member of the commission's staff. "The limits of covert action were never defined; the weaknesses in accountability went largely unaddressed." No one who read it bought the anodyne arguments that a little fine-tuning would fix the machine.

As the commission completed its report, a total of twenty-five people were enrolled at the CIA's career training center for young new recruits. The agency's ability to attract talent was at an all-time low. So was its reputation. The Ames case had made the CIA's future a casualty of its history.

The clandestine service was "terribly concerned about what they feel are inadequate numbers of people on the front line," Fred Hitz said at the time. "Getting the right people and getting them in the right place is already a different problem to solve. We've got good people but not enough of them, and not enough of them in the places where we need them. If the president of the United States and the Congress of the United States don't help, then the one thing that will bring us around will bring us around too late. Some horrible event happening somewhere in the world, maybe in our own nation, that makes us all wake up as Pearl Harbor made us wake up and say--why in the world didn't we know?"

46. "WE'RE IN
TROUBLE"

At the end of 1994, Jim Woolsey recorded a farewell address to his troops at the CIA, sent a letter of resignation to the White House by courier, and left town in a hurry. Bill Clinton searched the government for someone willing and able to take the job.

"The president asked me whether I was interested in being the director of central intelligence," said his deputy secretary of defense, John Deutch. "I made it very clear to him that I was not. I saw my friend Jim Woolsey having tremendous difficulties as director. I didn't think that there was any reason for me to think I could do better."

Fine, Clinton replied, find someone who can. Six weeks went by before Deutch managed to press-gang a retired air force general named Mike Carns for the job. Six more weeks passed before the nomination wobbled, plummeted, and crashed.

"The president pressed on me the view that I really had to do it," Deutch said. Thus began a short and bitter lesson in the political science of American intelligence. Deutch had good reason to dread the assignment. He had been in and around national-security circles for three decades, and he knew that no director of central intelligence ever had succeeded in fulfilling his charter--serving simultaneously as the chairman of American intelligence and the chief executive of the CIA. He requested and received cabinet rank, as Bill Casey had, to ensure himself some access to the president. He had hopes that he might become the secretary of defense if Clinton was re-elected in 1996. But he knew that the CIA was in a state of turmoil that could not be repaired in a year or two.

"Plagued by poor leadership, the Agency is adrift," a veteran CIA analyst, John Gentry, wrote during the days that Deutch first came to office. "It has a palpable malaise. The unhappiness level of employees well into management ranks is very high. Senior officers are floundering as well." The agency was led by "a corps of senior officers so devoid of real leadership skills that it is largely incapable of independent creative action." With Clinton apparently content to get his intelligence from CNN, Gentry wrote, the CIA had "no one left to pander to."

As deputy secretary of defense, Deutch had been through a yearlong review of American intelligence with Woolsey, trying for a truce in the endless wars over money and power between the Pentagon and the CIA. They would pick an issue--say, the proliferation of nuclear weapons--and at the end of the day they would conclude that much more had to be done. Counterintelligence? After Ames, definitely more. Support to military operations? Hugely important. Human intelligence? More spies. Better analysis? Absolutely crucial. At the end of the review it was clear that there were an infinite number of needs and a finite amount of money and personnel available to meet them. American intelligence could not be reformed from within, and surely it was not being reformed from without.

Deutch and Woolsey both had the well-known I'm-the-smartest-guy-in-the-room syndrome. The difference was that Deutch often was the smartest. He had been the dean of science and the provost at the Massachusetts Institute of Technology; his field was physical chemistry, the science of the transformation of matter at the molecular, atomic, and subatomic levels. He could explain how a lump of coal becomes a diamond. He set out to transform the CIA under that type of pressure. At his confirmation hearings, he had vowed to change the culture of the CIA's clandestine service, "down to the bare bones," but he had no clear idea how. Like his predecessors, he went to learn at the feet of Richard Helms.

Helms, now eighty-two, carried himself with the bearing of a British peer. Shortly after his skull session with the new director, I had lunch with him at a restaurant two blocks from the White House. Helms sipped a noontime beer, sitting beneath slowly revolving ceiling fans, and confided that Deutch was instinctively drawing away from the clandestine service--"seeing it as nothing but trouble. Nor is he the first to be distancing himself. He's got to do a job convincing them he's on the team."

In May 1995, a few days after Deutch showed up for work at CIA headquarters, the leaders of the clandestine service, always conscious of the need to recruit a new boss, presented him with a glossy brochure titled "A New Direction. A New Future." It was the list of their top ten targets: loose nukes, terrorism, Islamic fundamentalism, support for military operations, macroeconomics, Iran, Iraq, North Korea, Russia, China. The new director and his spies all knew that the White House wanted to use the CIA as a private Internet, a database on everything from tropical rain forests to compact-disc counterfeiters, and that its attention needed a far sharper focus. "The trouble is there's too much to do," Deutch said. "You get requests: What's going to happen in Indonesia? What's going to happen in Sudan? What's going to happen in the Middle East?" The call for global coverage was impossible to fulfill. Let us concentrate on a few hard targets, the spies said. Deutch could not settle the argument.

Instead, he worked for five months trying to get a handle on the clandestine service. He flew off to CIA stations around the world, listening, questioning, and weighing what he had to work with. He said he found "tremendously poor morale." He was shocked by the inability of his spies to solve their own problems. He found them in a state approaching panic.

He compared them to the American military after Vietnam. Back then, as Deutch put it in September 1995, a lot of smart lieutenants and colonels had looked at one another and said: "'We're in trouble. We've got to change. We've got to figure out a way to do this differently. We're either leaving or we're going to change the system.' And the people who stayed did change the system." Deutch wanted the clandestine service to solve its own problems. But he found his people incapable of change. "Compared to uniformed officers," he said of his spies, "they certainly are not as competent, or as understanding of what their relative role is or what their responsibilities are." The clandestine service "was not confident of carrying out its day-to-day activity."

This crisis of confidence took many forms. Some were made manifest in misguided operations that backfired. Others were continuing failures of collection and analysis. Some were breathtaking lapses in judgment.

In Bosnia, on July 13, 1995, as the world's press reported mass killings of Muslims by Serbs, a spy satellite sent back pictures of prisoners being guarded by gunmen in fields outside the town of Srebrenica. No one at the CIA looked at that picture for three weeks. No one had thought that the Serbs would conquer the town. No one anticipated a slaughter. No one paid heed to human-rights groups, the United Nations, or the press. The CIA had no officers and no agents in the field to corroborate what they were reporting. It had no information about any atrocities. It had been ordered to devote itself to supporting military operations in the region, and it had neither time nor talent to spare to check out reports from terrified refugees.

Two weeks after the first press accounts of a slaughter, the CIA sent a U-2 over Srebrenica; the plane recorded images of freshly dug mass graves in the fields where the prisoners had stood. Those photos arrived at the CIA on a regular military courier flight three days later. And three days after that, a CIA photo analyst matched up the location of the first satellite image of the prisoners in the field with the second U-2 image of burial sites. The analysis landed at the White House on August 4, 1995.

Thus did the CIA report, three weeks after the fact, the biggest mass murder of civilians in Europe since Hitler's death camps fifty years before. Eight thousand people were dead, and the agency had missed it.

On the other end of Europe, the CIA's Paris station had run an elaborate operation trying to steal the French negotiating position on trade talks. Locked into the idea that free trade was the guiding force of American foreign policy, the White House had aggravated the CIA's woes by demanding more and more economic intelligence. The Paris station was pursuing secrets of minimal importance to the national security of the United States--such as how many American movies would be shown on French screens. The French interior ministry ran a counterespionage operation that included the seduction of a CIA officer working under nonofficial cover as a businesswoman. There was pillow talk, and secrets were spilled. The government publicly expelled the Paris station chief--Dick Holm, a genuine hero at the clandestine service, who had run field operations in Laos and barely survived a fiery plane crash in the Congo thirty years before, and who was on his last tour before retirement. Four hapless and humiliated CIA officers were kicked out of France with him.

Another blown operation, another public embarrassment for the clandestine service, and "another public example of a situation where its ability to carry out its function as its own standards required came into question," Deutch said. He asked his officers time and again: "What are the professional standards of carrying out your very difficult mission? And are you doing it well all over the world?" His answer to that last question was a resounding no.

"IT WAS CLEARLY MALICE"

The problems at the Paris station were a passing annoyance compared to what went on in the Latin American division of the clandestine service. The division was a world apart at the CIA, dominated by veterans of the war against Fidel Castro, men who had their own set of rules and disciplines. Since 1987, station chiefs in Costa Rica, El Salvador, Peru, Venezuela, and Jamaica had been accused of lying to superiors, sexually harassing coworkers, stealing money, threatening underlings at gunpoint, running a counternarcotics operation in which a ton of cocaine wound up on the streets of Florida, and keeping sloppy accounts involving $1 million in government funds. It was the only division in the clandestine service in which station chiefs were removed from their posts for misconduct on a regular basis. The division's isolation flowed in part from the internal politics of the countries it covered. Throughout the cold war, the CIA had worked with military regimes against left-wing insurgencies in Latin America. The old bonds were hard to break.

In Guatemala, 200,000 civilians had died during forty years of struggle following the agency's 1954 coup against an elected president. Between 90 and 96 percent of those deaths came at the hands of the Guatemalan military. In 1994, the CIA's officers in Guatemala still went to great lengths to conceal the nature of their close relations to the military and to suppress reports that Guatemalan officers on its payroll were murderers, torturers, and thieves. This concealment violated a balancing test that Woolsey had started in 1994. The test, called "agent validation," was supposed to weigh the quality of an agent's information against the perfidiousness of his conduct.

"You don't want to be in a position of dealing with military officials or officials in that government who are known by everyone to have blood on their hands unless there is a legitimate intelligence goal to be served," said Inspector General Fred Hitz. "Unless that person knows that there's a cache in southern Guatemala where biological weapons are being put together and they're going to be sold on the open market and he's your only source for it. If a person is notorious for butchering people, breaking the law, then the fact that the CIA is in contact with that individual has to be balanced against the information that individual is likely to provide. If the information is the keys to the holy mystery, we'll take the chance. But let's do it with our eyes open and not because of inertia or momentum."

This problem boiled over when a Guatemalan colonel on the CIA's payroll was implicated in the cover-up of the murders of an American innkeeper and a Guatemalan guerrilla married to an American lawyer. The outcry over the innkeeper's murder had led the Bush administration to cut off millions of dollars in military aid to Guatemala, though the agency continued its financial support for Guatemalan military intelligence. "The CIA station in Guatemala was about twice the size it needed to be," said Thomas Stroock, the American ambassador to Guatemala from 1989 to 1992, but it could not seem to bring itself to report accurately on the case. The station chief, Fred Brugger, failed to tell Ambassador Stroock that the colonel, a prime suspect, was a CIA agent. "Not only did they not tell me," Ambassador Stroock said, "they did not tell my boss, the secretary of state, or the Congress. That was stupid."

Folly turned to malevolence in 1994, when Dan Donahue became the station chief. While the new American ambassador, Marilyn McAfee, was preaching human rights and justice, the CIA stayed loyal to the murderous Guatemalan intelligence service.

The embassy split in two. "The chief of station came into my office and showed me a piece of intelligence, which came from a Guatemalan source, suggesting that I was having an affair with my secretary, whose name was Carol Murphy," Ambassador McAfee remembered. The Guatemalan military had bugged the ambassador's bedroom and recorded her cooing endearments to Murphy. They spread the word that the ambassador was a lesbian. The CIA station transmitted this piece of intelligence--later known as "the Murphy memo"--to Washington, where it was widely distributed. "The CIA sent this report to the Hill," ambassador McAfee said. "It was clearly malice. The CIA had defamed an ambassador by back channels."

The ambassador was a conservative person from a conservative family, she was married, and she was not sleeping with her secretary. "Murphy" was the name of her two-year-old black standard poodle. The bug in her bedroom had recorded her petting her dog.

The CIA station had shown a stronger affinity for its friends in the Guatemalan military than for the American ambassador. "There was a division between intelligence and policy," Ambassador McAfee said. "That's what scares me."

It scared Deutch too. On September 29, 1995, toward the end of his fifth month in office, Deutch went to the Bubble--the once-futuristic six-hundred-seat amphitheater near the entrance to CIA headquarters--to deliver some bad news to the clandestine service. An internal-review board at the CIA had weighed the evidence in Guatemala and told Deutch that he should dismiss Terry Ward, the Latin American division chief of the clandestine service from 1990 to 1993, then serving as the chief of station in Switzerland. It said he should dismiss the former Guatemala station chief Fred Brugger too, and discipline his successor Dan Donahue severely, making sure he never served as a station chief again.

Deutch said there were "tremendous deficiencies in the way the agency carried out its business" in Guatemala. The problem was lying--or, as he put it, "a lack of candor"--between the chief of station and the American ambassador, the station and the Latin American division, the division and headquarters, and finally between the agency and Congress.

It was rare--very rare--for anyone to be fired from the clandestine service. But Deutch said he was going to do exactly as the review board recommended. The announcement did not go over well at the Bubble. The hundreds of officers gathered there were ferociously angry. Deutch's decision signified to them a suffocating political correctness. The director told them that they had to keep going out into the world and taking risks in the name of national security. A low growl rose from the back of the Bubble, a bitter laugh signifying: Yeah. Sure. That was the moment when the director and the clandestine service washed their hands of one another. It sealed his fate at the CIA.

"WE WANT TO GET THIS RIGHT"

The break was unbridgeable. Deutch decided to hand the portfolio of problems at the clandestine service to his number-two man--George Tenet, the deputy director of central intelligence. Now forty-two years old, Tenet, always the tireless and loyal aide, had spent five years as staff director of the Senate intelligence committee and two years as the National Security Council's point man for intelligence. He had vital insights in managing the CIA's tortured relationships with the Congress and the White House. And he soon came to see the clandestine service differently than Deutch did--not as a problem to be solved but a cause to be championed. Tenet would do his utmost to lead them.

"Let me explain life to you," Tenet said he told the clandestine service chiefs. "Here are the ten or fifteen things, that we cannot tolerate to fail against, to advance the national-security interests of the United States. This is what we want you to devote your money, your people, your language training, and your skills to. We want to get this right."

Terrorism soon rose to the top of Tenet's list. In the fall of 1995, a barrage of threatening reports started coming from the CIA station in the Sudan to the agency headquarters and the White House counterterrorism czar, Richard Clarke. They were based on the word of a single recruited CIA agent. They warned of an imminent attack against the station, the American embassy, and a prominent member of the Clinton administration.

"Dick Clarke came to me and said, 'They're going to blow you up,'" remembered Tony Lake, the president's national security adviser. Who is going to blow me up? Lake asked. Maybe the Iranians, Clarke replied, maybe the Sudanese. "So I went to live in a safe house and drove to work in a bulletproof car," Lake said. "They could never show it was real. I suspect not."

The Sudan was an international clearinghouse for stateless terrorists in those days. Among them was Osama bin Laden. The agency first knew him in the late 1980s as a rich Saudi who supported the same Afghan rebels that the agency armed in their fight against their Soviet oppressors. He was known as a financier of people who had grand visions of attacking the enemies of Islam. The CIA never pulled together its shards and fragments of intelligence on bin Laden and his network into a coherent report for the White House. No formal estimate of the terrorist threat he represented was published until after the entire world knew his name.

Bin Laden had returned to Saudi Arabia to rail against the presence of American troops after the 1991 Gulf War. The Saudi government expelled him, and he settled in the Sudan. The CIA's station chief in the Sudan, Cofer Black, was an old-school operator of considerable courage and cunning who had helped hunt down the burned-out terrorist known as Carlos the Jackal. Black tracked bin Laden's movements and motives in the Sudan as best he could. In January 1996, the CIA created a counterterrorism unit of a dozen people devoted entirely to the Saudi--the bin Laden station. There was a sense that he might start taking aim at American targets abroad.

But in February 1996, the CIA, heeding the warnings of its recruited agent, shut down its operations in the Sudan, blinding itself to fresh intelligence on its new target. The station and the American embassy were shuttered and their personnel moved to Kenya. The decision came over the strongest objections of the American ambassador, Timothy Carney, a man of military discipline along with diplomatic sensibilities. He argued that for the United States to withdraw from the Sudan was a dangerous mistake. He questioned the CIA's warnings about an imminent attack, and he was proved right. The agent who had raised the alarm was later found to be a fabricator, and the CIA formally withdrew roughly one hundred reports based on his information.

Shortly thereafter, bin Laden moved to Afghanistan. The chief of the bin Laden station, Mike Scheuer, saw this as a tremendous opportunity. The CIA had reestablished contacts with a network of Afghan exiles in the tribal northwest territories of Pakistan. The "tribals," as the CIA called them, were helping in the hunt for Mir Amal Kansi, the gunman who had killed two agency officers outside headquarters.

The hope was that they could help kidnap or kill bin Laden someday. But that day would have to wait. The CIA had another man in its crosshairs at that moment.

The chief of the Near East division of the clandestine service, Stephen Richter, had been working for two years on a plan to support a military coup against Saddam Hussein. The order had come from President Clinton, the third such command from the White House to the CIA in five years. In Jordan, a team of CIA officers met with Mohammed Abdullah Shawani, a former commander of Iraqi special forces. In London, the agency conspired with an Iraqi exile named Ayad Alawi, who headed a network of rebellious Iraqi military officers and Ba'ath Party leaders. The CIA backed him with money and guns. In northern Iraq, the CIA gathered the tribal leaders of the stateless Iraqi Kurds, renewing an old and troubled romance.

Despite the CIA's best efforts, none of these disparate and fractious forces came together. The agency invested many millions trying to recruit key members of Saddam's military and political circles, hoping they would rise up. But the plot was penetrated and subverted by Saddam and his spies. On June 26, 1996, Saddam began arresting at least two hundred officers in and around Baghdad. He executed at least eighty of them, including General Shawani's sons.

"The Saddam case was an interesting case," Mark Lowenthal, who had been staff director of the House intelligence committee and a senior CIA analyst, said after the coup collapsed. "All right, so we get rid of Saddam Hussein, good thing. But who do we get after him? Who's our guy in Iraq? Anybody that we put in power in Iraq is likely to have the staying power of a flea. So this was a case where you had policy makers saying do something. This do something urge really expressed their frustration." They failed to see that the CIA "had no way to deal with Saddam Hussein," he said. "The problem with the operation was that there were no reliable Iraqis to deal with. And the reliable Iraqis you're looking at have no access to do what you want to them to do. So the operation was a bust. It wasn't feasible. But it's very hard for an operator to say, 'Mr. President, we can't do that.' So you end up with an operation that probably shouldn't have been started in the first place."

"FAILURE IS INEVITABLE"

Deutch infuriated Clinton by telling Congress that the CIA might never solve the problem of Saddam Hussein. His seventeen-month tenure as director of central intelligence ended in bitterness. In December 1996, after Clinton was re-elected, he dismissed Deutch from the government and turned to his national security adviser, Tony Lake, to take the job so few coveted.

"It would have been a great challenge," Lake mused. "What I had in mind was pushing the analytical side to make intelligence--both its sources and its products--fit in with the world of the mid-1990s. What we got was too often an overnight parsing of the news."

But Lake would not be confirmed. The Republican chairman of the Senate intelligence committee, Richard Shelby of Alabama, decided to make him a whipping boy for everything that conservatives found wrong with the Clinton administration's conduct of foreign policy. The appearance of bipartisanship that the intelligence committees had maintained for the better part of twenty years evaporated. There was also an undercurrent of opposition to Lake from inside the clandestine service. The message was: don't send us any more outsiders.

"To the CIA, everyone is an outsider," Lake observed.

It was not even close to a fair hearing. On March 17, 1997, Lake withdrew in anger, telling the president that he was not going to spend three more months as "a dancing bear in a political circus." So the poisoned chalice was handed to George Tenet--the only choice remaining. Tenet was already running the agency as the acting chief. He would become the fifth director of central intelligence in six years.

"It is impossible to overstate the turbulence and disruption that that much change at the top caused," the CIA's Fred Hitz said. "Its impact on morale is hard to overstate, in terms of its destructiveness. You have the feeling--who's in charge here? Can't anybody up there play this game? Don't they understand what we're about? Don't they realize what our mission is?"

Tenet knew what the mission was: save the CIA. But the agency approached the end of the American Century burdened by a personnel system invented in the 1880s, an information conveyor belt resembling assembly lines of the 1920s, and a bureaucracy dating to the 1950s. It moved people and money around in ways that summoned up memories of Stalin's Five-Year Plans. Its ability to collect and analyze secrets was falling apart as the information age exploded and the Internet made encryption--the transformation of language into code--a universal tool. The clandestine service had become a place where "great successes are rare and failure is routine," a report by the House intelligence committee noted.

Those failures once again were front-page news. The CIA's capacity for spying had once again been wounded by a traitor from within. Harold J. Nicholson, who had been station chief in Romania, had taken up a two-year posting as a head instructor at the Farm, the CIA's training school outside Williamsburg, Virginia. He had been spying for Moscow since 1994, selling the Russians files on dozens of CIA officers stationed abroad and the identities of every new officer graduated from the Farm in 1994, 1995, and 1996. The CIA told the federal judge who sentenced Nicholson to twenty-three years in prison that it would never be able to calculate the damage he had done to its operations worldwide. The careers of three years' worth of CIA trainees were blighted; once burned, they could never serve overseas.

On June 18, 1997, three weeks before Tenet's swearing-in, a new report by the House intelligence committee erased the remnants of the prideful notion that the CIA served as America's first line of defense. The committee, led by Porter J. Goss, said the agency was filled with inexperienced officers unable to speak the languages or understand the political landscapes of the countries they covered. It said the CIA had a small and dwindling capability to gather intelligence through espionage. It concluded that the CIA lacked the necessary "depth, breadth, and expertise to monitor political, military, and economic developments worldwide."

Later that summer, a career intelligence officer named Russ Travers published a haunting essay in the CIA's in-house journal. He said America's abilities to gather and analyze intelligence were falling apart. For years, he wrote, the leaders of American intelligence had been insisting that they were putting the agency on the right track. This was a myth. "We fine-tune our structures and marginally change our programs...getting the deck chairs on the Titanic nice and neat." But "we are going to begin making more and bigger mistakes more often," he warned. "We have gotten away from basics--the collection and unbiased analysis of facts."

He offered a prophecy for the future leaders of the CIA. "The year is 2001," he wrote. "By the turn of the century, analysis had become dangerously fragmented. The Community could still collect 'facts,' but analysis had long ago been overwhelmed by the volume of available information and were no longer able to distinguish between significant facts and background noise. The quality of analysis had become increasingly suspect.... The data were there, but we had failed to recognize fully their significance.

"From the vantage point of 2001," he wrote, "intelligence failure is inevitable."

47. "THE THREAT
COULD NOT BE MORE
REAL"

George Tenet was sworn in as the eighteenth director of central intelligence on July 11, 1997. He boasted to me back then, knowing his words would appear in The New York Times, that the CIA was far smarter and more skilled than any outsider could know. This was public relations. "We were nearly bankrupt," he confessed seven years later. He had inherited a CIA "whose expertise was ebbing" and a clandestine service "in disarray."

The agency was preparing to mark its fiftieth anniversary that September, and it had put together a list of the fifty greatest CIA officers as part of the celebration. Most were either old and gray or dead and gone. The greatest among the living was Richard Helms. He was not in a celebratory mood. "The only remaining superpower doesn't have enough interest in what's going on in the world to organize and run an espionage service," Helms said to me that month. "We've drifted away from that as a country." His successor, James Schlesinger, felt much the same. "The trust that was reposed in the CIA has faded," he said. "The agency is now so battered that its utility for espionage is subject to question."

Tenet started rebuilding. He called old stars out of retirement, including Jack Downing, who had served as station chief in Moscow and Beijing and agreed to run the clandestine service for a year or two. Tenet also sought a multibillion-dollar cash infusion for the agency. He promised that the CIA could be restored to health in five years' time, by 2002, if the money started flowing immediately. Porter Goss, who held the agency's purse strings in the House, arranged for secret "emergency assistance" of several hundred million dollars, followed by a onetime $1.8 billion shot in the arm. It was the biggest intelligence spending increase in fifteen years, and Goss promised to find more.

"Intelligence isn't just something for the cold war," Goss said back then. "When you think back to Pearl Harbor, you can understand why. Unpleasant surprises are out there."

"CATASTROPHIC SYSTEMIC INTELLIGENCE FAILURE"

Tenet lived in a state of foreboding, awaiting the next snafu. "I will not allow the CIA to become a second-rate organization," he proclaimed at a headquarters pep rally. A few days later, on May 11, 1998, the agency was caught by surprise again when India exploded a nuclear bomb. The test remade the balance of power in the world.

The new Hindu nationalist government had vowed openly to make nuclear weapons part of its arsenal. India's atomic weapons commissioner had said he was ready to test if political leaders gave the go-ahead. Pakistan had fired off new missiles, all but daring New Delhi to respond. So a nuclear blast by the world's largest democracy should not have come as a shock--but it did. The reporting from the CIA's station in New Delhi was lazy. The analysis at headquarters was fuzzy. The warning bell never rang. The test revealed a failure of espionage, a failure to read photographs, a failure to comprehend reports, a failure to think, and a failure to see. It was "a very disturbing event," said Charles Allen, the CIA's longtime chief of warning, whom Tenet recalled from retirement to serve as his assistant director of central intelligence for collection. It was a clear sign of a systemic breakdown at the CIA.

People started having premonitions of a catastrophe. "The likelihood of a cataclysmic warning failure is growing," Tenet's successor at the National Security Council, Mary McCarthy, wrote in an unclassified report shortly after the Indian test. "Disaster looms!"

Tenet had a reason to be looking the other way at the time of the nuclear test. His troops were rehearsing an operation to capture Osama bin Laden. In February 1998, bin Laden had proclaimed that he was on a mission from God to kill Americans. In Afghanistan, he was gathering the shock troops and camp followers of the holy war against the Soviets for a new jihad against the United States. In Pakistan, the CIA's station chief, Gary Schroen, was perfecting a plan to use the agency's old Afghan allies to snatch bin Laden as he traveled to his mud-walled compound in the southern city of Kandahar. On May 20, 1998, they began a final, four-day, full-scale dress rehearsal. But on May 29, Tenet decided to cancel the operation. Success depended on coordination with Pakistan--which had now exploded its own nuclear test in response to India. The Pakistanis were pounding the war drums. The Afghans were unreliable. Failure was not an option--it was a probability. The chances for capturing bin Laden were slim to start, and the world was now too unstable to risk it.

June passed without the promised attack from bin Laden, then July. On August 7, 1998, President Clinton was awakened by a 5:35 a.m. call reporting the bombings of the American embassies in Nairobi, Kenya, and Dar es Salaam, Tanzania. The explosions took place four minutes apart. The damage in Nairobi was horrific; I saw it with my own eyes. Twelve Americans, including a young CIA officer, died in that blast, which killed hundreds and wounded thousands of Kenyans in the streets and office buildings outside the embassy walls.

The next day, George Tenet went to the White House with the news that bin Laden was heading to an encampment outside Khost, Afghanistan, near the border with Pakistan. Tenet and Clinton's national-security aides agreed to hit the camp with cruise missiles. They wanted a second target to even the score, and they chose al Shifa, an industrial plant outside Khartoum, the capital of Sudan. An Egyptian agent of the CIA had delivered a soil sample from outside the plant suggesting the presence of a chemical used to make VX nerve gas.

The evidence was a very slender reed. "We will need much better intelligence on this facility" before bombing it, warned Mary McCarthy at the National Security Council. None ever appeared.

Navy ships in the Arabian Sea fired a barrage of million-dollar cruise missiles at both targets on August 20. They killed perhaps twenty Pakistanis passing through Khost--bin Laden was long gone--and a night watchman in Sudan. Clinton's inner circle claimed that the evidence for attacking Sudan was airtight. First they said al Shifa was a weapons factory working for bin Laden. In fact, it was a pharmaceutical plant, and the link to bin Laden vanished. Then they said it was part of an Iraqi scheme to distribute nerve weapons. But the Iraqis had not weaponized VX, as tests by United Nations inspectors confirmed. The soil sample might have been a VX precursor, and it might as well have been weed killer.

The case was a dozen dots connected by inference and surmise. Nothing ever corroborated the decision to strike al Shifa. "It was a mistake," said Donald Petterson, the American ambassador to Sudan from 1992 to 1995. "The administration failed to produce conclusive evidence that chemical weapons were being made at the pharmaceutical factory. The administration had grounds for suspicion, but to commit an act of war, which the missile attack was, the evidence should have been iron-clad." His successor, Ambassador Tim Carney, said with deliberate restraint: "The decision to target al Shifa continues a tradition of operating on inadequate intelligence about Sudan." The Clinton administration's counterterrorist attack went off half-cocked.

Three weeks later, Tenet met with the rest of the leaders of the American intelligence community. They agreed that they had to make "substantial and sweeping changes" in the way the nation collected, analyzed, and produced information. If they did not, they said, the result would be "a catastrophic systemic intelligence failure." The date was September 11, 1998.

"WE WILL CONTINUE TO BE SURPRISED"

If the CIA did not reinvent itself, and soon, "in ten years we won't be relevant," Tenet told me in October, his first on-the-record interview as director of central intelligence. "Unless you develop the expertise, we won't achieve what we want to achieve."

Since 1991, the agency had lost more than three thousand of its best people--about 20 percent of its senior spies, analysts, scientists, and technology experts. Roughly 7 percent of the clandestine service was headed out the door annually. That added up to a loss of about a thousand experienced spies, and it left not many more than a thousand in place. Tenet knew he could not guard against the future with so weak a force out on the front lines.

"There will always be days where we have to race to catch up to events we did not foresee, not because somebody is asleep at the switch, but because what's going on is so complicated," he said. "There is an expectation that we have built a no-fault intelligence system, that intelligence is expected not only to tell you about the trends and to tell you about events and give you insight, but in each and every case has a responsibility to tell you when the date, time, and place of an event is." The CIA itself had created that hope and expectation long ago. It was an illusion. "We will continue to be surprised," Tenet said.

He began to organize a nationwide talent hunt, painfully aware that his battle to rebuild the CIA would take many years, many billions of dollars, and many thousands of new officers. It was a desperate struggle against time. It takes five to seven years to turn a novice into a case officer capable of working in the rougher capitals of the world. American-born citizens who were both fluent in foreign cultures and willing and able to work for the CIA were hard to come by. A spy must know how "to use deception, to use manipulation, to use, frankly, dishonesty in the pursuit of his job," said Jeffrey Smith, the CIA's general counsel in the mid-1990s. "The management of the agency must always worry about finding that extraordinarily rare individual who has the talent to deal in this deceptive and manipulative world and keep his or her own moral ballast." Finding, hiring, and keeping such exceptional minds had been a job that never was done.

Over the years, the CIA had become less and less willing to hire "people that are a little different, people who are eccentric, people who don't look good in a suit and tie, people who don't play well in the sandbox with others," Bob Gates said. "The kinds of tests that we make people pass, psychological, and everything else, make it very hard for somebody who may be brilliant or have extraordinary talents and unique capabilities to get into the agency." As a consequence of its cultural myopia, the CIA misread the world. Very few of its officers could read or speak Chinese, Korean, Arabic, Hindi, Urdu, or Farsi--the languages of three billion people, half the planet's population. Far too few had ever haggled in an Arab bazaar or walked through an African village. The agency was unable to dispatch "an Asian-American into North Korea without him being identified as some kid who just walked out of Kansas, or African-Americans to work around the world, or Arab-Americans," Gates said.

In 1992, when Gates was director of central intelligence, he wanted to hire an American citizen raised in Azerbaijan. "He spoke Azeri fluently, but he didn't write English very well," he recalled. "And so he was rejected because he didn't pass our English test. And when I was told this, I just went crazy. I said: 'I've got thousands of people here who can write English, but I don't have anybody here who can speak Azeri. What have you done?'"

The agency started combing the cities and the suburbs of America looking for the children of immigrants and refugees, young men and women who grew up in first-generation Asian and Arab households, reaching out with ads in ethnic newspapers throughout the United States. The harvest was thin. Tenet knew the agency would live or die in years to come by virtue of its ability to project an image of international intrigue and intellectual adventure to smart young people. But new blood was only part of the cure. The recruitment drive never resolved a fundamental question at the agency--could the CIA recruit the kind of person it would need five or ten years down the road? It did not know where the road would lead. It only knew it could not survive in the state into which it had fallen.

"WE'RE GOING TO BOMB THIS"

The enemy was growing stronger as the agency grew weaker. The failed attack on bin Laden supercharged his status and attracted thousands of new recruits to his cause. The urgency of the CIA's campaign against al Qaeda escalated in concert with his popularity.

Tenet revived the plans to use Afghan proxies to capture him. In September and October 1998, the Afghans claimed they had mounted four failed ambushes against bin Laden--which the CIA strongly doubted. But they convinced the agency's field officers that they could track him as he traveled from camp to camp inside Afghanistan. They reported on December 18 that bin Laden was heading back to Kandahar, and that he would spend the night of December 20 at a house inside the governor's compound there. Station chief Gary Schroen sent word from Pakistan: strike tonight--there may never be a better chance. The cruise missiles were spinning in their chambers and locked on the target. But the intelligence was one man's word and hundreds of people were sleeping in the compound that night. Tenet's desire to do away with bin Laden was overcome by his doubts. The word from on high was no go. Courage gave way to caution and gung-ho became go slow.

From the fall of 1998 onward, "the United States had the capability to remove Osama bin Laden from Afghanistan or to kill him," but it quailed when it came time to pull the trigger, said John MacGaffin, the number-two officer in the clandestine service earlier in the Clinton years. "The CIA knew bin Laden's location almost every day--sometimes within fifty miles, sometimes within fifty feet." At least fifteen American special-forces soldiers were killed or injured in training missions for the anticipated assault. Commanders in the Pentagon and civilian leaders in the White House continually backed down from the political gamble of a military mission against bin Laden.

They left the job to the CIA. And the agency could not execute it.

The Afghans reported in the first weeks of 1999 that bin Laden was headed for a hunting camp south of Kandahar favored by wealthy falconers. A spy satellite looked down on the camp on February 8 and fixed its location. A government aircraft from the United Arab Emirates--an American ally--was parked there. The lives of the emirs could not be sacrificed to kill bin Laden and the cruise missiles stayed in their launchers.

The Afghans kept tracking bin Laden's travels in and out of Kandahar throughout April 1999. They locked onto him for thirty-six straight hours in May. Gary Schroen's agents delivered detailed reports on his whereabouts. The intelligence would never be better, said Tenet's deputy director of central intelligence, General John Gordon.

Three times the chance came to strike with cruise missiles. Three times Tenet said no. His confidence in the CIA's ability to pick its targets had been badly shaken days before.

A NATO bombing campaign against Serbia had been launched with the intent of forcing President Slobodan Milosevic to withdraw his troops from Kosovo. The CIA had been invited to pick targets for American warplanes. The task was assigned to the agency's counterproliferation division, the group that analyzed intelligence on the spread of weapons of mass destruction. The analysts identified their best target as the Yugoslav Federal Directorate of Supply and Procurement, at 2 Umetnosti Boulevard in Belgrade. They used tourist maps to help them fix the location. The targeting flowed up through the CIA's machinery to the Pentagon, and the coordinates were loaded onto a B-2 stealth bomber's circuitry.

The target was destroyed. But the CIA had misread its maps. The building was not Milosevic's military depot. It was the Chinese embassy.

"The bombing of the Chinese embassy in Belgrade was an acutely unpleasant experience for me," said Vice Admiral Thomas R. Wilson, who became director of the Defense Intelligence Agency in July 1999. "I was the one who showed the picture of the Chinese embassy to the President of the United States--among 900 other pictures I showed him--and said, 'We're going to bomb this, because it's the Yugoslav department of military procurement.'" He had gotten that picture from the CIA.

This mistake cut deeper than anyone could know. It would be a long time before the White House and the Pentagon would trust the agency to put anything--or anyone--in the crosshairs of an American missile.

"YOU AMERICANS ARE CRAZY"

The military and intelligence services of the United States were still set up to work against armies and nations--hard to kill, but easy to find on the map of the world. The new enemy was a man--easy to kill, but hard to find. He was a wraith moving around Afghanistan at night in a Land Cruiser.

President Clinton signed secret orders that he thought gave the CIA the power to kill bin Laden. In the throes of his impeachment, he daydreamed aloud about American ninjas rappelling out of helicopters to grab the Saudi. He made Tenet the commander of a war against one man.

Tenet fought his own doubts about the CIA's intelligence and its covert-action capabilities. But he had to devise a new plan of attack before bin Laden struck again. With his new counterterrorism chief, Cofer Black, he laid out a new strategy at the end of the summer of 1999. The agency would work with old friends and old foes around the world to kill bin Laden and his allies. Black deepened his ties to foreign military, intelligence, and security services in places such as Uzbekistan and Tajikistan, on the Afghan border. The hope was that they would help CIA officers put their boots on the ground inside Afghanistan.

The goal was to link up with the Afghan warrior Ahmed Shah Massoud at the redoubt he had held for nearly twenty years, since the early days of the Soviet occupation, deep in a mountain valley northeast of Kabul. Massoud, a noble and courageous fighter who wanted to be the king of Afghanistan, proposed a grand alliance to his old contacts from the agency. He offered to attack bin Laden's strongholds--and, with the help of the CIA and American arms, to overthrow the Taliban, the rabble of peasants, mullahs, and jihad veterans that ruled in Kabul. He could help the agency establish a base that would let it get bin Laden on its own. Cofer Black was all for it. His deputies were ready to go.

But the chances of failure were too high for Tenet. Once again, he said no--getting in and out was too risky. Reporters and foreign aid workers took those risks all the time in Afghanistan. The CIA at headquarters would not.

Massoud laughed when he heard that. "You Americans are crazy," he said. "You guys never change."

As the millennium approached, Jordan's intelligence service, created and long supported by the CIA, arrested sixteen men whom it believed were prepared to blow up hotels and tourist sites during Christmas. The agency thought that this plot prefigured a global attack by al Qaeda timed for the new year. Tenet went into overdrive, contacting twenty foreign intelligence chiefs in Europe, the Middle East, and Asia, telling them to arrest anyone associated with bin Laden. He sent an urgent message to all CIA officers overseas. "The threat could not be more real," it said. "Do whatever is necessary." The millennium passed without a catastrophic attack.

The president was briefed on the CIA's covert-action plans against bin Laden in February and March 2000, and he said the United States surely could do better. Tenet and Jim Pavitt, the new chief of the clandestine service, said they would need millions in new funds to do the job. The White House counterterrorism czar, Richard Clarke, thought that the CIA's will, not its wallet, was too thin; he said the agency had been given "a lot of money to do it and a long time to do it, and I didn't want to put more good money after bad."

The political season brought the return of the tradition inaugurated by President Truman: the intelligence briefing for the opposition. The acting deputy director of central intelligence, John McLaughlin, and the deputy chief of the counterterrorist center, Ben Bonk, went down to Crawford, Texas, and led a four-hour seminar for Governor George W. Bush over Labor Day in September. It was Bonk's unhappy duty to tell the Republican candidate that Americans would die at the hands of foreign terrorists at some time in the coming four years.

The first deaths came five weeks later. On October 12, in the harbor of Aden, the capital of Yemen, two men in a speedboat stood and bowed as they approached an American warship, the USS Cole. The explosion killed seventeen, wounded forty, and ripped a $250 million hole in one of the American navy's most sophisticated vessels.

Al Qaeda was the obvious suspect.

The CIA set up a satellite office in Crawford to keep Bush posted on that attack and other world events during the long struggle over the 2000 election. In December, after the Supreme Court declared Bush the victor, Tenet personally briefed the president-elect on bin Laden. Bush remembered specifically asking Tenet if the CIA could kill the guy; Tenet replied that killing him would not end the threat he represented. Bush then met alone with Clinton for two hours to talk about national security.

Clinton remembers telling him: "Your biggest threat is bin Laden." Bush swore that he never heard those words.

48. "THE DARK SIDE"

"American intelligence is in trouble," James Monnier Simon, Jr., the assistant director of central intelligence for administration, warned shortly after Bush took office in January 2001. The CIA "has had its centrality compromised," he said. It lacked the power to collect and analyze the intelligence needed to protect the nation.

"The United States in 2001 is faced with a growing, almost dizzying disparity between its diminished capabilities and the burgeoning requirements of national security," Simon said. "The disconnect between what we are planning for and the likelihood of what the United States will face has never been so stark." The time would come when the president and Congress would have to explain "why a foreseeable disaster went unforeseen."

American intelligence was almost as divided and diffused as it had been in 1941. Eighteen consecutive directors of central intelligence had failed in their duty to unify it. Now the agency was about to fail as an institution of American government.

The CIA stood at seventeen thousand people, about the size of an army division, but the great majority of them were desk jockeys. Roughly one thousand people worked abroad in the clandestine service. Most officers lived comfortably in suburban cul-de-sacs and townhouses in the orbit of the Washington beltway. They were unused to drinking dirty water and sleeping on mud floors. They were unsuited for lives of sacrifice.

Two hundred officers had joined the CIA's clandestine service as charter members in September 1947. Perhaps two hundred were capable and courageous enough to tough it out in hardship posts in January 2001. The full complement of CIA personnel focused on al Qaeda amounted to perhaps twice that number. Most of them were staring at computers at headquarters, cut off from the realities of the outside world by their antiquated information technologies. To expect them to protect the United States from attack was at best a misplaced faith.

"A HOLLOW SHELL OF WORDS WITHOUT DEEDS"

Tenet was in the good graces at the White House, having formally renamed the CIA's headquarters the Bush Center for Intelligence after the president's father, and the new commander in chief liked Tenet's tough-guy attitude. But the agency received the barest support from President Bush during his first nine months in office. He gave the Pentagon an immediate 7 percent budget increase. The CIA and the rest of the intelligence community received a boost of three-hundredths of one percent. The difference was set in meetings at Donald Rumsfeld's Pentagon, where not a single representative of the intelligence community was present. Rumsfeld and Vice President Dick Cheney, partners in the politics of national security since the days of Nixon and Ford, held enormous power in the new administration. They shared an abiding distrust in the capabilities of the CIA.

Bush and Tenet met at the White House almost every morning at eight. But nothing Tenet said about bin Laden fully captured the president's attention. Morning after morning at the eight o'clock briefing, Tenet told the president, Cheney, and national security adviser Condoleezza Rice about portents of al Qaeda's plot to strike America. But Bush was interested in other things--missile defense, Mexico, the Middle East. He was struck by no sense of emergency.

During the Reagan administration, when the president was hard of hearing and the director of central intelligence mumbled unintelligibly, aides used to joke that there was no telling what went on between them. Bush and Tenet had no such infirmities. The problem lay in a lack of clarity at the CIA and a lack of focus at the White House. It's not enough to ring the bell, Richard Helms used to say. You've got to make sure the other guy hears it.

The noise--the volume and frequency of fragmentary and uncorroborated information about a coming terrorist attack--was deafening. Tenet could not convey a coherent signal to the president. As the klaxon sounded louder and louder in the spring and summer of 2001, every nerve and sinew of the agency strained to see and hear the threat clearly. Warnings were pouring in from Saudi Arabia and the Gulf states, Jordan and Israel, all over Europe. The CIA's frayed circuits were dangerously overloaded. Tips kept coming in. They're going to hit Boston. They're going to hit London. They're going to hit New York. "When these attacks occur, as they likely will," Clarke e-mailed Rice on May 29, "we will wonder what more we could have done to stop them."

The agency feared an onslaught overseas during the July 4 holiday, when American embassies across the world traditionally let down their defenses and open their doors to celebrate the American Revolution. In the weeks before the holiday, Tenet had called upon the chiefs of foreign intelligence services in Amman, Cairo, Islamabad, Rome, and Ankara to try to destroy known and suspected cells of al Qaeda and its affiliates throughout the world. The CIA would supply the intelligence, and the foreign services would make the arrests. A handful of suspected terrorists were jailed in the Gulf States and in Italy. Maybe the arrests had disrupted plans of attack against two or three American embassies, Tenet told the White House. Maybe not. Impossible to tell.

Tenet now had to make a life-and-death decisions unlike any that had ever confronted a director of central intelligence. A year before, after a seven-year struggle between the CIA and the Pentagon, a small pilotless aircraft equipped with video cameras and spy sensors called the Predator had been declared ready to be deployed over Afghanistan. The first flight had come on September 7, 2000. Now the agency and the air force had figured out how to put antitank missiles on the Predator. In theory, for an investment of a few million dollars a CIA officer at headquarters would soon be able to hunt and kill bin Laden with a video screen and a joystick. But what was the chain of command? Tenet wondered. Who gives the go-ahead? Who pulls the trigger? Tenet thought he had no license to kill. The idea of the CIA launching a remote-control assassination on its own authority appalled him. The agency had made too many mistakes picking targets in the past.

On August 1, 2001, the Deputies Committee--the second-echelon national-security team--decided that it would be legal for the CIA to kill bin Laden with the Predator, an act of national self-defense. But the agency came back with more questions. Who would pay for it? Who would arm the aircraft? Who would be the air traffic controller? Who would play the roles of pilot and missile man? The hand-wringing drove counterterrorism czar Clarke crazy. "Either al Qaeda is a threat worth acting against or it is not," he fumed. "CIA leadership has to decide which it is and cease these bi-polar mood swings."

The agency never had answered a question put to it by President Bush: could an attack come in the United States? Now was the time: on August 6, the president's daily brief began with the headline "Bin Ladin Determined to Strike in US." The warning beneath the headline was a very weak piece of reporting. The freshest intelligence in it dated from 1999. It was a work of history, not a news bulletin. The president continued his vacation, chopping brush in Crawford, unwinding for five weeks.

The long White House holiday ended on Tuesday, September 4, when Bush's first-string national security team, the Principals Committee, sat down together for the first meeting it ever held on the threat of bin Laden and al Qaeda. Clarke sent an agonized note that morning to Condoleezza Rice, begging the national security adviser to envision hundreds of Americans lying dead from the next attack. He said the agency had become "a hollow shell of words without deeds," relying on foreign governments to stop bin Laden, leaving the United States "waiting for the big attack." He implored her to move the CIA to action that day.

"WE'RE AT WAR"

Intelligence fails because it is human, no stronger than the power of one mind to understand another. Garrett Jones, the CIA station chief during the disastrous American expedition in Somalia, put it plainly: "There are going to be screw-ups, mistakes, confusion, and missteps," he said. "One hopes they won't be fatal."

September 11 was the catastrophic failure Tenet had predicted three years before. It was a systemic failure of American government--the White House, the National Security Council, the FBI, the Federal Aviation Administration, the Immigration and Naturalization Service, the congressional intelligence committees. It was a failure of policy and diplomacy. It was a failure of the reporters who covered the government to understand and convey its disarray to their readers. But above all it was a failure to know the enemy. It was the Pearl Harbor that the CIA had been created to prevent.

Tenet and his counterterror chief, Cofer Black, were at Camp David on Saturday, September 15, laying out a plan to send CIA officers into Afghanistan to work with local warlords against al Qaeda. The director returned to headquarters late Sunday and issued a proclamation to his troops: "We're at war."

The agency, as Cheney said that morning, went over to "the dark side." On Monday, September 17, President Bush issued a fourteen-page top secret directive to Tenet and the CIA, ordering the agency to hunt, capture, imprison, and interrogate suspects around the world. It set no limits on what the agency could do. It was the foundation for a system of secret prisons where CIA officers and contractors used techniques that included torture. One CIA contractor was convicted of beating an Afghan prisoner to death. This was not the role of a civilian intelligence service in a democratic society. But it is clearly what the White House wanted the CIA to do.

The CIA had run secret interrogation centers before--beginning in 1950, in Germany, Japan, and Panama. It had participated in the torture of captured enemy combatants before--beginning in 1967, under the Phoenix program in Vietnam. It had kidnapped suspected terrorists and assassins before--most famously in 1997, in the case of Mir Amal Kansi, the killer of two CIA officers. But Bush gave the agency a new and extraordinary authority: to turn kidnapped suspects over to foreign security services for interrogation and torture, and to rely on the confessions they extracted. As I wrote in The New York Times on October 7, 2001: "American intelligence may have to rely on its liaisons with the world's toughest foreign services, men who can look and think and act like terrorists. If someone is going to interrogate a man in a basement in Cairo or Quetta, it will be an Egyptian or a Pakistani officer. American intelligence will take the information without asking a lot of lawyerly questions."

Under Bush's order, the CIA began to function as a global military police, throwing hundreds of suspects into secret jails in Afghanistan, Thailand, Poland, and inside the American military prison in Guantanamo, Cuba. It handed hundreds more prisoners off to the intelligence services in Egypt, Pakistan, Jordan, and Syria for interrogations. The gloves were off. "Our war on terror begins with al Qaeda, but it does not end there," Bush told the nation in an address to a joint session of Congress on September 20. "It will not end until every terrorist group of global reach has been found, stopped, and defeated."

"I COULD NOT NOT DO THIS"

There was a war at home as well, and the CIA was part of it. After 9/11, James Monnier Simon, Jr., the assistant director of central intelligence, was placed in charge of homeland security for the intelligence community. He went to a meeting at the White House with Attorney General John Ashcroft. The subject was the creation of national identity cards for Americans. "What would it have? Well, a thumbprint," Simon said. "Blood type would be useful, as would a retinal scan. We would want your picture taken a special way so that we could pick your face out of a crowd even if you were wearing a disguise. We would want your voice print, because the technology is coming up that will pick your voice out of every other voice in all the cell phones on earth, and your voice is unique. In fact, we would like to have a bit of your DNA in there, so if something ever happens to you we can identify the body. By the way, we would want the chip to tell us where this card is, so that if we needed to find you we could. Then it dawned on us that if we did that, you could set the card down. So we would put the chip in your bloodstream."

Where would this drive for security end? Simon wondered. The names of Stalin's and Hitler's intelligence services sprang to his mind. "It could in fact end up being the KGB, NKVD, Gestapo," he said. "We, the people, need to watch and be involved." Precisely how the American people were supposed to watch was a problematic question. What a representative of the director of central intelligence was doing at the White House discussing the implantation of microchips in American citizens was another. The national identity card never materialized. But Congress did give the CIA new legal powers to spy on people in the United States. The agency was now permitted to read secret grand jury testimony, without a judge's prior approval, and obtain private records of institutions and corporations. The agency used the authority to request and receive banking and credit data on American citizens and companies from financial corporations. The CIA had never had the formal power to spy inside the borders of the United States before. It did now.

Tenet talked to General Michael Hayden, the director of the National Security Agency, shortly after the attacks. "Is there anything more you can do?" he asked. "Not within my current authorities," Hayden answered. Tenet then "invited me to come down and talk to the administration about what more could be done." Hayden came up with a plan to eavesdrop on the communications of suspected terrorists within the United States without judicial warrants. It was arguably illegal but arguably justified on a theory of "hot pursuit"--chasing suspects beyond the borders of the map and outside the limits of the law. President Bush ordered him to execute the plan on October 4, 2001. It had to be done, Hayden said: "I could not not do this." The NSA once again began to spy within the United States.

Cofer Black ordered his counterterror corps to bring him bin Laden's head in a box. The Counterterrorist Center, born fifteen years before as a small freestanding unit of the clandestine service, still working in the basement at headquarters, was now the heart of the CIA. Retired officers returned to duty and new recruits joined the agency's tiny cadre of paramilitary commandos. They flew into Afghanistan to make war. The agency's men handed out millions of dollars to marshal the loyalties of Afghan tribal leaders. They served nobly for a few months as advance troops for the American occupation of Afghanistan.

By the third week of November 2001, the American military knocked out the political leadership of the Taliban, leaving behind the rank and file, but paving the way for a new government in Kabul. It left tens of thousands of Taliban loyalists unscathed. They trimmed their beards and melted into the villages; they would return when the Americans began to tire of their war in Afghanistan. They would live to fight again.

It took eleven weeks to organize the hunt for Osama bin Laden. When that hunt began in earnest, I was in eastern Afghanistan, in and around Jalalabad, where I had traveled on five trips over the years. An old acquaintance named Haji Abdul Qadir had just reclaimed his post as the provincial governor, two days after the fall of the Taliban. Haji Qadir was an exemplar of Afghan democracy. A well-educated and highly cultured Pathan tribal leader in his early sixties, a wealthy dealer in opium and weapons and other basic staples of the Afghan economy, he had been a CIA-supported commander in the fight against the Soviet occupation, the governor of his province from 1992 to 1996, and a close associate of the Taliban in their time. He personally welcomed Osama bin Laden to Afghanistan and helped him establish a compound outside Jalalabad. Now he welcomed the American occupation. Haji Qadir was a good host. We walked in the gardens of the governor's palace, through swayback palms and feathery tamarisks. He was expecting a visit from his American friends any day now, and he was looking forward to the renewal of old ties and the ritual exchange of cash for information.

Haji Qadir had gathered the village elders of his province at the governor's palace. On November 24, they reported that bin Laden and the Arab fighters of al Qaeda were holed up in an isolated mountain hideout thirty-five miles south-by-southwest of the city, near the village of Tora Bora.

On November 28, at about five in the morning, as the first call to prayer sounded, a small plane landed at the rocket-pocked runway of the Jalalabad airport with a delegation of CIA and Special Forces officers aboard. They were carrying bales of $100 bills. They met with Haji Zaman, the newly appointed Jalalabad commander of the self-proclaimed government. He told the Americans he was "90 percent sure" that bin Laden was in Tora Bora. The dusty road south from Jalalabad to Tora Bora ended in a rough mountain trail impassable to all but men and mules. The trailhead connected with a network of smugglers' routes leading to mountain passes into Pakistan. Those routes had been a supply line for the Afghan rebels, and Tora Bora had been a place of great renown in the fight against the Soviets. A cave complex dug deep into the mountainside had been built, with the CIA's assistance, to meet NATO military standards. An American commander with orders to destroy Tora Bora would have been well advised to use a tactical nuclear weapon. A CIA officer with orders to capture bin Laden would have needed to requisition the Tenth Mountain Division.

On December 5, as American B-52 bombers pounded away at that stony redoubt, I watched the attack from a few miles' distance. I wanted to see bin Laden's head on a pike myself. He was within the agency's reach, but beyond its grasp. He could only be taken by siege, and the CIA could not mount one. Those who went after al Qaeda in Afghanistan were the best the agency had, but they were too few. They had come armed with lots of cash, but too little intelligence. The futility of hunting bin Laden with dumb bombs was soon made manifest. Moving from camp to camp in the Afghan borderlands, bin Laden was protected by a phalanx of hundreds of battle-hardened Afghan fighters and thousands of Pathan tribesmen who would sooner die than betray him. He had the CIA outnumbered and outmaneuvered in Afghanistan, and he escaped.

Tenet was red-eyed, furious, gnawing on cigar stubs, near the limits of his endurance. His counterterror troops were pushed beyond their capacity. Alongside American special-operations soldiers, they were hunting, capturing, and killing bin Laden's lieutenants and foot soldiers in Afghanistan, Pakistan, Saudi Arabia, Yemen, and Indonesia. But they started hitting the wrong targets again. Armed Predator attacks killed at least twenty-four innocent Afghans in January and February 2002; the CIA handed out $1,000 in reparations to each of their families. Fanning out across Europe, Africa, and Asia, working with every friendly foreign intelligence service on earth, CIA officers snatched and grabbed more than three thousand people in more than one hundred countries in the year after 9/11, Tenet said. "Not everyone arrested was a terrorist," he cautioned. "Some have been released. But this worldwide rousting of al Qaeda definitely disrupted its operations." That was inarguable. But the fact remained that as few as fourteen men among the three thousand seized were high-ranking authority figures within al Qaeda and its affiliates. Along with them, the agency jailed hundreds of nobodies. They became ghost prisoners in the war on terror.

The focus and intensity of the mission to kill or capture bin Laden began to ebb in March 2002, after the failed assault on Tora Bora. The CIA had been commanded by the White House to turn its attention to Iraq. The agency responded with a fiasco far more fatal to its fortunes than the 9/11 attacks.

49. "A GRAVE
MISTAKE"

"There is no doubt that Saddam Hussein now has weapons of mass destruction," Vice President Dick Cheney said on August 26, 2002. "There is no doubt he is amassing them to use against our friends, against our allies, and against us." Secretary of Defense Don Rumsfeld said the same: "We know they have weapons of mass destruction," he said. "There isn't any debate about it."

Tenet provided his own grim warnings in a secret hearing before the Senate intelligence committee on September 17: "Iraq provided al Qaeda with various kinds of training--combat, bomb-making, and chemical, biological, radiological and nuclear." He based that statement on the confessions of a single source--Ibn al-Shakh al-Libi, a fringe player who had been beaten, stuffed in a two-foot-square box for seventeen hours, and threatened with prolonged torture. The prisoner had recanted after the threat of torture receded. Tenet did not correct the record.

On October 7, on the eve of congressional debate over whether to go to war with Iraq, President Bush said that Iraq "possesses and produces chemical and biological weapons." He went on to warn that "Iraq could decide on any given day to provide a biological or chemical weapon to a terrorist group or individual terrorist." This created a dilemma for Tenet. Days before, his deputy, John McLaughlin, had contradicted the president in testimony to the Senate intelligence committee. On orders from the White House, Tenet issued a statement saying, "There is no inconsistency between our view of Saddam's growing threat and the view as expressed by the President in his speech."

It was the last thing he should have said, and he knew it. "It was the wrong thing to do," Tenet testified almost four years later. Throughout his years of public service, Tenet had been a fundamentally decent man. But under the enormous pressures he faced after 9/11, his one flaw, his all-consuming desire to please his superiors, became a fault line. Tenet's character cracked, and the CIA did too. Under his leadership, the agency produced the worst body of work in its long history: a special national intelligence estimate titled "Iraq's Continuing Programs for Weapons of Mass Destruction."

A national estimate is the best judgment of the American intelligence community, produced and directed by the CIA, and distributed with the authority and imprimatur of the director of central intelligence. It is his word.

The estimate was commissioned by members of the Senate intelligence committee, which thought it might be wise to review the evidence before going to war. At their request, the CIA's analysts spent three weeks gathering and reviewing everything the agency knew from spy satellites; from foreign intelligence services; and from recruited Iraqi agents, defectors, and volunteers. The CIA reported in October 2002 that the threat was incalculable. "Baghdad has chemical and biological weapons," the top secret estimate said. Saddam had bolstered his missile technology, bulked up his deadly stockpiles, and restarted his nuclear weapons program. "If Baghdad acquires sufficient fissile material from abroad," said the estimate, "it could make a nuclear weapon within several months." And most terrifying of all, the CIA warned that Iraq could conduct chemical and biological attacks inside the United States.

The CIA confirmed everything the White House was saying. But the agency was saying far more than it knew. "We did not have many Iraqi sources," Jim Pavitt, the chief of the clandestine service, admitted two years later. "We had less than a handful." The agency produced a ton of analysis from an ounce of intelligence. That might have worked if the ounce been solid gold and not pure dross.

The CIA as an institution was betting that American soldiers or spies would find the evidence after the invasion of Iraq. It was a hell of a gamble. It would have appalled Richard Helms, who died at age eighty-nine on October 22, 2002, after the estimate was completed. In tribute to his legacy, the CIA reprinted parts of a speech he had made years before. The full text was buried in the agency's archives, but its power had not dimmed. "It is sometimes difficult for us to understand the intensity of our public critics," Helms had said. "Criticism of our efficiency is one thing, criticism of our responsibility quite another. I believe that we are, as an important arm of government, a legitimate object of public concern.... I find it most painful, however, when public debate lessens our usefulness to the nation by casting doubt on our integrity and objectivity. If we are not believed, we have no purpose."

"WE HAD NO ANSWERS"

To understand how the CIA was able to say that Iraq's weapons of mass destruction existed, go back to 1991 and the end of the first Gulf war. After the war came seven years of intensive international scrutiny, led by United Nations inspectors looking for evidence that Saddam had a hidden arsenal. They combed the country and captured what they could.

In the mid-1990s, Saddam feared international economic sanctions more than another attack from the United States. He destroyed his weapons of mass destruction in compliance with the commands of the United Nations. But he kept his weapons-production facilities, he lied about it, and the United States and the United Nations knew he was lying. That legacy of lying caused the inspectors and the CIA to distrust everything Iraq did.

In 1995, General Hussein Kamal, Saddam's son-in-law, defected along with a few of his aides. Kamal confirmed that Saddam had destroyed the weapons. The CIA disregarded what he said, judging it as deception. The fact that Kamal went back to Iraq and was assassinated by his father-in-law did not alter the agency's belief.

His aides told the CIA about Iraq's National Monitoring Directorate, which aimed to conceal Saddam's military intentions and capabilities from the world. The CIA wanted to pierce that system of concealment, and a stroke of good fortune made it possible. Rolf Ekeus, the chairman of the United Nations inspection team, was Swedish. So was Ericsson, the telecommunications giant that made the walkie-talkies used by the National Monitoring Directorate. The CIA, the NSA, Ekeus, and Ericsson devised a way to tap the Iraqis' telecommunications. In March 1998, a CIA officer in the guise of a United Nations weapons inspector went to Baghdad and installed the eavesdropping system. Intercepted conversations were beamed to a computer in Bahrain, which searched for key words such as missile and chemical. A sterling operation, with one exception: the CIA learned nothing about the existence of any weapons of mass destruction in Iraq.

That spring, the weapons inspectors found what they thought were remnants of VX nerve gas in Iraqi missile warheads. Their report was leaked to The Washington Post. Baghdad called it an American lie. Charles Duelfer, who had led some of the inspection teams in the 1990s and returned to Iraq as Tenet's lead weapons hunter in 2004, concluded, "Ultimately, I think, the Iraqis were right. They did not weaponize VX."

The confrontation over the VX report was a turning point. Iraq no longer trusted the inspectors, who had never trusted Iraq. In December 1998, the United Nations pulled its inspectors out and the United States once again started bombing Baghdad. The information the CIA had gleaned from its Ericsson taps was used to target American missiles on the people and institutions it penetrated--including the home of the man who ran the National Monitoring Directorate.

Iraq declared to the United Nations that it had rid itself of weapons of mass destruction. The declarations were essentially accurate; the substantial violations were minor. But Saddam had been deliberately ambiguous about his arsenal, fearing he would stand naked before his enemies if they believed he did not have the capability to produce the weapons. He wanted the United States, his foes in Israel and Iran, his internal enemies, and above all his own troops to believe that he still had the weapons. Illusion was his best deterrent and his last defense against attack.

This was the state of affairs that confronted the CIA after 9/11. Its last reliable reports from inside Iraq were very old news. "We were bereft of any human intelligence--zero, nada, in terms of agents on the ground," said David Kay, who also had led the United Nations team and preceded Duelfer as the CIA's chief weapons hunter in Iraq. The White House wanted answers. "We had no answers," Kay said.

Then, in 2002, "suddenly, what looked like a golden source of human intelligence stepped forward: defectors," he said. "These defectors coming out of Saddam's regime told us about his weapons programs and weapons progress. Not all of them came out to the United States; many came out to the intelligence services of France, Germany, Britain, and other countries. The information seemed to be unbelievably good." One of the most riveting stories was the one about mobile biological weapons laboratories. The source was an Iraqi in the hands of the German intelligence service, code-named Curveball.

"The Iraqi defectors understood two things: one, we shared a mutual interest in regime replacement; and, two, the U.S. was very concerned about weapons of mass destruction in Iraq," Kay said. "So they told us about weapons in order to get us to move against Saddam. It was basic Newtonian physics: give me a big enough lever and a fulcrum, and I can move the world."

Only one thing was worse than having no sources, and that was to be seduced by sources telling lies.

The clandestine service had produced little information on Iraq. The analysts accepted whatever supported the case for war. They swallowed secondhand and thirdhand hearsay that conformed to the president's plans. Absence of evidence was not evidence of absence for the agency. Saddam once had the weapons. The defectors said he still had them. Therefore he had them. The CIA as an institution desperately sought the White House's attention and approval. It did so by telling the president what he wanted to hear.

"FACTS AND CONCLUSIONS BASED ON SOLID INTELLIGENCE"

President Bush presented the CIA's case and more in his State of the Union speech on January 28, 2003: Saddam Hussein had biological weapons sufficient to kill millions, chemical weapons to kill countless thousands, mobile biological weapons labs designed to produce germ-warfare agents. "Saddam Hussein recently sought significant quantities of uranium from Africa," he said. "Our intelligence sources tell us that he has attempted to purchase high-strength aluminum tubes suitable for nuclear weapons production."

All of this was terrifying. None of it was true.

On the eve of the war, on February 5, 2003, Secretary of State Colin Powell, whose international stature was unmatched in the Bush administration, went to the United Nations. With George Tenet at his shoulder, ever the loyal aide, his presence a silent affirmation--and the American ambassador to the United Nations, the future director of national intelligence, John Negroponte, at his side--the secretary of state began: "Every statement I make today is backed up by sources, solid sources. These are not assertions. What we are giving you are facts and conclusions based on solid intelligence."

Powell said: "There can be no doubt that Saddam Hussein has biological weapons and he has the ability to dispense these lethal poisons and diseases in ways that can cause massive death and destruction." He again warned of Iraq's mobile biological-weapons laboratories, how they could park in a shed, make their poison, and move on undetected. He said Saddam had enough lethal chemical weaponry to fill sixteen thousand battlefield rockets. And perhaps worst of all, there was the threat of the "much more sinister nexus between Iraq and the al Qaeda terrorist network."

This was not a selective use of intelligence. It was not "cherry-picking." It was not fixing the facts to fit the war plans. It was what the intelligence said, the best intelligence the agency had to offer. Powell had spent days and nights with Tenet, checking and rechecking the CIA's reporting. Tenet looked him in the eye and told him it was rock solid.

On March 20, 2003, the war began ahead of schedule with a bad tip from the CIA. Tenet ran to the White House with a flash report that Saddam Hussein was hiding at a compound south of Baghdad called Doura Farms. The president ordered the Pentagon to destroy the compound. The bunker-busting bombs and the cruise missiles rained down. Vice President Cheney said, "I think we did get Saddam Hussein. He was seen being dug out of the rubble and wasn't able to breathe." It was a false report: Saddam was nowhere to be found. The first targeting failure of the war was not the last. On April 7, 2003, the CIA reported that Saddam and his sons were meeting at a house next to the Saa Restaurant in the Mansur district of Baghdad. The air force dropped four one-ton bombs on the house. Saddam was not there either. Eighteen innocent civilians were killed.

The agency had predicted that thousands of Iraqi soldiers and their commanders would surrender all along the route of the attack once it was launched over the border from Kuwait. But the American invading force had to fight its way through every town of any size on the road to Baghdad. The CIA envisioned the wholesale capitulation of Iraqi military units, and its intelligence was specific: the Iraqi division based at An Nasiriya would lay down its arms. The first American troops into the city were ambushed; eighteen marines were killed, some by friendly fire, in the first major combat of the war. American forces were told they would be welcomed by cheering Iraqis waving American flags--the clandestine service would provide the flags--and showered with candy and flowers on the streets of Baghdad. In time, they were met with bullets and bombs.

The CIA compiled a list of 946 suspect sites where Saddam's arsenals of mass destruction might be found. American soldiers were wounded and killed hunting for the weapons that never were. The agency missed the threat posed by the assault rifles and rocket-propelled grenades stockpiled by the Fedayeen, the irregular forces led by Saddam's son Uday Hussein. The failure led to the first major series of combat deaths of American soldiers. "The Fedayeen and other paramilitary forces proved more of a threat than anyone had expected," wrote the authors of On Point, the U.S. Army's official history of the invasion of Iraq. "The intelligence and operations communities had never anticipated how ferocious, tenacious, and fanatical they would be."

The CIA organized a paramilitary squad of Iraqis called the Scorpions to conduct sabotage before and during the war. During the occupation, the Scorpions distinguished themselves by beating an Iraqi general to death. Major General Abed Hamed Mowhoush, who was suspected of directing insurgent attacks but had turned himself in voluntarily to American forces, was clubbed senseless with sledgehammer handles by the Scorpions in the presence of a CIA officer who led them, a retired Special Forces officer who had signed on with the agency for the war. Mowhoush died of his injuries two days later, on November 26, 2003. Earlier that month, an Iraqi prisoner named Manadal al-Jamadi was tortured to death at the Abu Ghraib prison while in the custody of a CIA officer. The brutal interrogations were part of what the White House had called upon the agency to do when the gloves came off.

As the CIA concluded three years after the invasion, the American occupation of Iraq became "the cause celebre for jihadists, breeding a deep resentment of U.S. involvement in the Muslim world and cultivating supporters for the global jihadist movement." The assessment came far too late to be much use to American forces. "Every Army of liberation has a half-life beyond which it turns into an Army of occupation," wrote Lieutenant General David H. Petraeus, who commanded the 101st Airborne Division in the first year of the war, oversaw the effort to train the Iraqi army on a second tour, and returned as the commander of American forces in 2007.

"Intelligence is the key to success," he said. Without it, military operations fall into "a catastrophic downward spiral."

"JUST GUESSING"

The agency went pouring into Baghdad when the war was over. "As Iraq transitions from tyranny to self-determination, Baghdad is home to the largest CIA station since the Vietnam war," proclaimed Jim Pavitt, the chief of the clandestine service. "I am extremely proud of our performance in Iraq, and of our role in liberating its people from decades of repression." The officers at the Baghdad station worked with Special Forces soldiers, trying to create a new political climate in Iraq, selecting local leaders, paying off politicians, trying to rebuild society at the grass roots. They tried to work with their British counterparts to create a new Iraqi intelligence service. But very little came of all that. When the Iraqi insurgency began to rise up against the American occupation, those projects started to fall apart and the leadership at the CIA's Baghdad station began breaking down.

As the occupation spun out of control, the CIA's officers found themselves pinned down at the American embassy compound in the capital, unable to escape the protection of the high walls and razor wire. They became prisoners of the Green Zone, powerless to comprehend the Iraqi insurgency, spending far too many hours drinking at the Babylon bar, run by the Baghdad station. Many would not accept a rotation of more than one to three months, barely enough time to get their bearings in Baghdad.

The station, whose ranks approached five hundred officers, ran through three chiefs in the course of a year. The CIA simply could not find a replacement for the first station chief in 2003. "They had grave, grave difficulty finding a competent individual to go out," said Larry Crandall, a veteran Foreign Service officer who had worked closely with the CIA during the Afghan jihad and served as the number-two manager of the $18 billion American reconstruction program in Iraq. The agency had no one from the clandestine service willing or able to serve. It finally selected an analyst with next to no experience in running operations. He lasted for a matter of months. It was an extraordinary failure of leadership in a time of war.

The CIA sent the best of the American inspectors who had hunted down Saddam's arsenal in the 1990s back to Iraq. David Kay led a team of 1,400 specialists, the Iraq Survey Group, working directly for the director of central intelligence. Tenet continued to stand by the CIA's reporting, rejecting the growing criticism as "misinformed, misleading, and just plain wrong." But the survey group scoured Iraq and found nothing. When Kay returned to report that, Tenet put him in purgatory. Kay nonetheless went before the Senate Armed Services Committee on January 28, 2004, and spoke the truth.

"We were almost all wrong," he said.

When it became certain that the agency had only imagined Iraq's doomsday arsenal, a moral exhaustion began to settle over the CIA. A dark bitter anger overtook the fiery spirit that had gripped it after 9/11. It was evident that it no longer mattered much to the White House or the Pentagon or the State Department what the agency had to say.

President Bush disdained the CIA's increasingly dire reports on the course of the occupation. The agency was "just guessing," he said.

This was a death knell. If we are not believed, we have no purpose.

"THE EVIDENCE WAS COMPLETELY FRAIL"

"We're at war," said Judge Laurence Silberman, whom President Bush appointed on February 6, 2004, to lead an investigation into the ways in which the CIA had conjured Saddam's weaponry. "If the American Army had made a mistake anywhere near as bad as our intelligence community, we would expect generals to be cashiered."

He continued: "It would have been eminently justifiable to have told the President and the Congress that it was likely that Saddam had weapons of mass destruction based on his past use, insufficient indications of destruction, and his deceptive behavior." But the CIA had made "a grave, grave mistake in concluding that there was a ninety percent degree of certainty that he had weapons of mass destruction. And it was a grave mistake not based on hindsight," he said. "The evidence was completely frail, some quite faulty, and their tradecraft wasn't good. Moreover, there was such an abysmal lack of internal communication within the intelligence community that often the left hand didn't know what the right hand was doing."

The CIA had reached its conclusions on Iraqi chemical weapons solely on the basis of misinterpreted pictures of Iraqi tanker trucks. The CIA had based its conclusions on Iraqi biological weapons on one source--Curveball. The CIA had based its conclusions on Iraqi nuclear weapons almost entirely on Saddam's importation of aluminum tubes intended for conventional rocketry. "It's almost shockingly wrong to conclude that those aluminum tubes were appropriate or designed for centrifuges for nuclear weapons," Judge Silberman said.

"What was such a disaster," he said, "was for Colin Powell to have gone to the United Nations and set forth that absolutely unmistakable certain case which was based on really bad, bad stuff."

Judge Silberman and his presidential commission received unprecedented permission to read every article on Iraq's weapons of mass destruction from the president's daily brief. They found that the CIA's reports for the president's eyes were no different from the rest of its work, including the infamous estimate--except in one regard. They were "even more misleading," the commission found. They were, "if anything, more alarmist and less nuanced." The president's daily briefs, "with their attention-grabbing headlines and drumbeat of repetition, left an impression of many corroborating reports where in fact there were very few sources.... In ways both subtle and not so subtle, the daily reports seemed to be 'selling' intelligence--in order to keep its customers, or at least the First Customer, interested."

"WE DIDN'T GET THE JOB DONE"

George Tenet saw that his time was over. He had done his best to revive and renew the agency. Yet he would always be remembered for one thing: reassuring the president that the CIA had "slam dunk" evidence on Iraq's weapons of mass destruction. "Those were the two dumbest words I ever said," Tenet reflected. No matter how long he lived, no matter what good deeds he might do in years to come, they would feature in the first paragraph of his obituary.

Tenet, to his credit, asked Richard Kerr, the former deputy director of central intelligence, to investigate what had gone wrong with the Iraq estimate. The study was classified upon completion in July 2004 and stayed that way for nearly two years thereafter. When it was unsealed, it was clear why the agency had kept it under wraps. It was an epitaph. It said the CIA had all but ceased to be when the cold war came to a close; the fall of the Soviet Union had an impact on the agency "analogous to the effect of the meteor strikes on the dinosaurs."

In the case of Iraq, and in many other cases as well, analysts were routinely forced to "rely on reporting whose sourcing was misleading and even unreliable." In the infamous case of Curveball, CIA officers had fair warning that the man was a liar. This warning went unheeded. That was not dereliction of duty, but it came close.

The clandestine service routinely "used different descriptions for the same source," so that the readers of its reports believed they had three corroborating sources of information when they had one. This was not fraud, but it came close.

The CIA had been working on the questions of the Iraqi arsenal for more than a decade, and yet Tenet had gone to George Bush and Colin Powell on the eve of war brandishing falsehoods cloaked as hard truths. That was not a crime, but it came close.

Tragically, this was Tenet's legacy. He finally conceded that the CIA was wrong--not for "political reasons or a craven desire to lead the country to war" but because of its incompetence. "We didn't get the job done," he said.

The meaning of that failure was left for the CIA's chief weapons inspector, David Kay, to explain in full. "We think intelligence is important to win wars," he said. "Wars are not won by intelligence. They're won by the blood, treasure, courage of the young men and women that we put in the field.... What intelligence really does when it is working well is to help avoid wars." That, in the end, was the ultimate intelligence failure.

50. "THE BURIAL
CEREMONY"

On July 8, 2004, seven years after he took office, George Tenet resigned. In his farewell at CIA headquarters, he summoned up the words of Teddy Roosevelt: It is not the critic who counts, nor the man who points out how the strong man stumbles, or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood. Richard Nixon had quoted the same speech the day before he left the White House in disgrace.

Tenet retreated to write a painful personal memoir of his time at the CIA. It was a prideful, bitter book. He justly boasted of the CIA's success--with invaluable help from British intelligence--at dismantling secret weapons programs in Pakistan and Libya. He maintained that he had transformed the agency from a shambles into a dynamo. But the machine had broken down under intolerable pressure. Tenet could not strike at al Qaeda before 9/11: "In the absence of hard intelligence," he wrote, "covert action is a fool's game." And ever since the attacks, he had been swamped by tidal waves of threats that never materialized. Every day he conveyed the newest fears to the White House, and "you could drive yourself crazy believing all or even half" of what he reported. He nearly had. Gripped by uncertainty, he and the CIA had convinced themselves that Iraq's arsenal existed. "We were prisoners of our own history," he wrote, for the only hard facts they had were four years old. He confessed error, but it was a condemned man's plea for absolution. Tenet came to believe that the White House wanted to blame the decision to go to war on him. It was too great a weight to bear.

And now the critic took his turn as the man in the arena.

Porter Goss had never been a great success at the CIA. Recruited in his junior year at Yale in 1959, he joined the clandestine service and served under Allen Dulles, John McCone, and Richard Helms. He had worked in the Latin American division for a decade, focused on Cuba, Haiti, the Dominican Republic, and Mexico. The highlight of his time in the Miami station was running Cuban agents on and off the island in small boats under the cover of the night in the fall of 1962.

Nine years later, Goss was serving at the London station when a bacterial infection seized his heart and lungs and nearly killed him. He retired, recovered, bought a small newspaper in Florida, and parlayed the paper into a seat in Congress in 1988. He had a net worth of $14 million, a gentleman's farm in Virginia, an estate on Long Island Sound, and a viceroy's dominion over the CIA as chairman of the House intelligence committee.

He was modest about his accomplishments at the agency. "I couldn't get a job with CIA today," he said in 2003. "I am not qualified." He was right about that. But he had decided that he and he alone should be the next director of central intelligence. He had taken aim at Tenet with a vicious fusillade. His weapons were the words of the intelligence committee's annual report on the agency.

"IT WILL TAKE US ANOTHER FIVE YEARS"

Published on June 21, 2004, three weeks before Tenet stepped down, the Goss report warned that the clandestine service was becoming "a stilted bureaucracy incapable of even the slightest bit of success." Though 138,000 Americans had applied to work at the CIA in the previous year, few made the grade as spies. Tenet had just testified that "it will take us another five years of work to have the kind of clandestine service our country needs."

Goss seized on that sad truth: "We are now in the eighth year of rebuild, and still we are more than five years away from being healthy. This is tragic."

Goss then turned his fire on the CIA's intelligence directorate for producing spot news of scant value instead of the long-range strategic intelligence that had been the initial reason for creating the agency. Goss was right about that, too--and everyone in the intelligence world knew it. "We haven't done strategic intelligence for so long that most of our analysts don't know how to do it anymore," said Carl W. Ford, Jr., the assistant secretary of state for intelligence and research from May 2001 to October 2003, and a former CIA officer.

"As long as we rate intelligence more for its volume than its quality, we will continue to turn out the $40 billion pile of crap we have become famous for," Ford said. He was incensed that the agency, while transfixed by the chimerical arsenal of Saddam Hussein, had learned nothing about the nuclear weapons programs of the rest of the president's axis of evil. "We probably knew a hundred times more about Iraq's nuclear program than Iran's and a thousand times more than Korea's," Ford said. North Korea was a blank, as it had always been. The CIA had tried to rebuild a network of agents in Iran but failed. Now Iran was a blank, too; the agency actually knew less about those nuclear programs than it had known five or ten years before.

The CIA was in ruins, Ford said: "It's broken. It's so broken that nobody wants to believe it." The Goss report made that clear. "There is a dysfunctional denial of any need for corrective action," it said. "CIA continues down a road leading over a proverbial cliff."

Goss was sure he had the answers. He knew that the CIA had been fooling itself and others about the quality of its work. He knew that most of the clandestine service had spent four decades of the cold war waiting and hoping for Soviets to volunteer their services as spies. He knew that its overseas officers in the war on terror spent days and nights waiting and hoping for their counterparts in Pakistan and Jordan and Indonesia and the Philippines to sell them information. He knew that the solution was to overhaul the agency.

The 9/11 Commission created by Congress was about to issue its final report. The commission did a splendid job reconstructing the events leading up to the attacks. It did not chart a clear path forward. Nor had Congress done much to fix the agency since 9/11, other than giving it billions of dollars and plenty of free advice. The commission correctly described congressional oversight of intelligence as "dysfunctional"--the same epithet Goss hurled at the agency. For years, there had been next to no engagement on the life-and-death issues that confronted the CIA by the House and Senate intelligence committees. The House committee under Goss had produced its last substantive report on the conduct of the CIA in 1998. A quarter of a century of congressional oversight of the agency had produced little of lasting value. The intelligence committees and their staffs had applied an occasional public whipping and a patchwork of quick fixes for ever-present problems.

It was known that the 9/11 Commission would recommend the creation of a new national intelligence director. The idea had been kicked around since Allen Dulles's heyday. It offered no real solution to the crisis at the CIA. Rearranging the boxes on the flowchart of the government would not make it easier to run the CIA.

"It is an organization that thrives through deception," said John Hamre, a former deputy secretary of defense and the president of the Center for Strategic and International Studies in Washington. "How do you manage an organization like that?"

It was one of many questions the CIA and Congress never had answered. How do you run a secret intelligence service in an open democracy? How do you serve the truth by lying? How do you spread democracy by deceit?

"IN THE END, THEY WON'T STAY"

The myth about the CIA dated back to the Bay of Pigs: that all its successes were secret, that only its failures were trumpeted. The truth was that the CIA could not succeed without recruiting and sustaining skilled and daring officers and foreign agents. The agency failed daily at that mission, and to pretend otherwise was a delusion.

To succeed, the CIA needed to find men and women with the discipline and self-sacrifice of the nation's best military officers, the cultural awareness and historical knowledge of the nation's best diplomats, and the sense of curiosity and adventure possessed by the nation's best foreign correspondents. It would help if those recruits were able to pass for Palestinians, Pakistanis, or Pathans. Americans like that were hard to find.

"Can CIA meet the ongoing threat? The answer at this moment is no--absolutely not," said Howard Hart, who had put his life on the line to run agents in Iran, smuggled guns to the Afghan rebels, and led the agency's paramilitary officers. Hart said he took offense when Goss called the CIA "a bunch of dysfunctional jerks" and "a pack of idiots." But he conceded that "CIA's clandestine service can be criticized for not having done as well as it should have. That is a fair statement. Because we have people who just don't pull their weight. And the reason that most of them are there is that we have no way to replace them."

President Bush pledged to increase the agency's ranks by 50 percent. But quality, not quantity, was the crisis at hand. "What we don't need is more money and people, at least not for now," Carl Ford said. "Fifty percent more operators and fifty percent more analysts equals fifty percent more hot air." The personnel problem was the same one Walter Bedell Smith had faced as director of central intelligence while the Korean War was raging: "We can't get qualified people. They just simply don't exist."

The CIA could not find enough talented Americans to serve as spies on a government salary. Hundreds had resigned at headquarters and in the field during 2004, infuriated and humiliated at the collapse of credibility and authority at the agency. Recruiting, hiring, training, and retaining young officers still remained the most difficult task at the agency.

Goss vowed to find them. He went to his Senate confirmation hearings with a swagger on September 14, 2004, saying he could fix the CIA once and for all. "I don't want to give aid and comfort to the enemy by telling you how bad I think the problem is," he said before the cameras, but the problem would be solved. Upon his confirmation by a 77-to-17 vote from the full Senate, Goss went straight to CIA headquarters in a state of exhilaration.

"I never in my wildest dreams expected I'd be back here," he told the men and women he had roundly condemned three months before. "But here I am." He proclaimed that his powers would be "enhanced by executive orders" from the president: he would be Bush's intelligence briefer, the head of the CIA, director of central intelligence, national intelligence director, and chief of a new national counterterrorism center. He would not wear two hats, like his predecessors. He would wear five.

On his first day of work, Goss began a purge more swift and sweeping than any in the history of the Central Intelligence Agency. He forced almost every one of the CIA's most senior officers out the door. He created a bitterness that had not been seen at headquarters in nearly thirty years. The resentment over the expulsion of Stephen Kappes as chief of the clandestine service was ferocious. Kappes, an ex-marine and former station chief in Moscow, represented the very best of the CIA. In partnership with the British intelligence service, he had only recently played a leading role in a triumph of intelligence and diplomacy by persuading Libya to abandon its long-running program to develop weapons of mass destruction. When he questioned Goss's judgment, he was shown the door.

The new director surrounded himself with a team of political hacks he had imported from Capitol Hill. They believed they were on a mission from the White House--or some higher power--to rid the CIA of left-wing subversives. It was the perception at headquarters that Goss and his staff, the "Gosslings," prized loyalty to the president and his policies above all, that they did not want the agency athwart of the White House, and that those who challenged them would pay the price. The scourging of the CIA was rightly a question of competence. It wrongly became a question of ideology.

The director issued orders against dissent from the president's policies. His message was clear: get with the program or get out. The latter choice looked more and more attractive for the talented tenth of the CIA's personnel. A huge homeland security industry was growing at the outer edge of the beltway, selling its services to a government outsourcing expertise. The best of the agency's people sold out. The CIA had been top-heavy with aging cold warriors fifteen years before. Now it was bottom-heavy with beginners. By 2005, half of the CIA's workforce--operators and analysts alike--had five years' experience or less.

The president's offhand proclamation that the agency was "just guessing" about Iraq ignited a smoldering anger that burned throughout the ranks of those professionals who remained. The CIA's officers in Baghdad and in Washington tried to warn that the path the president was pursuing in Iraq was disastrous. They said the United States could not run a country it did not understand. Their words carried no weight at the White House. They were heresy in an administration whose policies were based on faith.

Four former chiefs of the clandestine service tried to contact Goss to advise him to go slow, lest he destroy what remained of the CIA. He would not take their calls. One of them went public: "Goss and his minions can do a great deal of damage in short order," Tom Twetten wrote in an opinion piece published in the Los Angeles Times on November 23,2004. "If the professional employees in the agency don't believe the agency's leadership is on their side, they won't take risks for it and, in the end, they won't stay." The next day, John McLaughlin, who had held the agency together as acting director after Tenet's resignation, delivered another riposte. The CIA was not "a 'dysfunctional' and 'rogue' agency," he wrote in The Washington Post. "The CIA was not institutionally plotting against the president." Haviland Smith, who had retired as chief of the counterterrorism staff, weighed in. "Porter Goss and his troops from the Hill are wreaking havoc," he wrote. "Purging the CIA at this unfortunate moment, when we need to be dealing with real issues of terrorism, is cutting off our nose to spite our face." In all the years that the agency had been battered in the press, never had the director been attacked in print, on the record, by the most senior veterans of American intelligence.

The facade was falling. The CIA was tearing itself apart.

"Here is one of the most peculiar types of operation any government can have," President Eisenhower had said fifty years before. "It probably takes a strange kind of genius to run it." Nineteen men had served as director of central intelligence. Not one met the high standard Eisenhower had set. The agency's founders had been defeated by their ignorance in Korea and Vietnam and undone by their arrogance in Washington. Their successors were set adrift when the Soviet Union died and caught unaware when terror struck at the heart of American power. Their attempts to make sense of the world had generated heat but little light. As it was in the beginning, the warriors of the Pentagon and the diplomats of State held them in disdain. For more than half a century, presidents had been frustrated or furious when they turned to the directors for insight and for knowledge.

The job, having proved impossible to fill, now would be abolished.

In December 2004, with the upheaval at the agency in full force, Congress passed and the president signed a new law establishing the director of national intelligence, as the 9/11 Commission had urged. Hastily drafted, hurriedly debated, the law did nothing to ease the chronic and congenital problems that had plagued the agency since birth. It was continuity masquerading as change.

Goss thought the president would choose him. The call never came. On Feburary 17, 2005, Bush announced that he was nominating the American ambassador to Iraq, John D. Negroponte. A diplomat of rigorously conservative stripe, suave, subtle, a skilled infighter, he had never worked a day in the world of intelligence, and he would not long serve in it.

As in 1947, the new czar was given responsibility without commensurate authority. The Pentagon still controlled the great bulk of the na tional security budget, now approaching $500 billion a year, of which the CIA's share was roughly 1 percent. The new order served only as a formal recognition that the old order had failed.

"FAILURE CANNOT BE EXPLAINED"

The CIA was gravely wounded. In accordance with the laws of the jungle and the ways of Washington, stronger beasts fed upon it. The president gave great power over espionage, covert action, eavesdropping, and reconnaissance to the Pentagon's undersecretary for intelligence, and elevated that job to the number-three position at the Department of Defense. "That sent seismic shudders through the intelligence community," said Joan Dempsey, who was a deputy director of central intelligence and executive director of the foreign intelligence advisory board under Bush. "That's much more of a Kremlin approach."

The Pentagon moved stealthily and steadily into the fields of overseas covert operations, usurping the traditional roles, responsibilities, authorities, and missions of the clandestine service. It recruited the most promising young paramilitary officers and retained the most experienced ones. The militarization of intelligence accelerated as the nation's civilian intelligence service eroded.

Negroponte's new chief analyst, Thomas Fingar, had run the State Department's small but first-rate office of intelligence and research. He surveyed the state of the agency's directorate of intelligence and quickly determined that "nobody had any idea of who was doing what where." He moved to pull the functioning remains of the CIA's analytic machinery under his aegis. The best and the brightest thinkers left at the agency signed on with him.

The agency as constituted was vanishing. The building was still there, and there would always be an institution in it. But on March 30, 2005, a wrecking ball struck what remained of the spirit of the CIA. It came in the form of the six-hundred-page report of Judge Silberman's presidential commission. The judge was as rigorous a thinker as could be found in the capital. His intellectual badge was as strong as his intensely conservative credentials. He had twice come close to being named director of central intelligence. In fifteen years as a federal appeals court judge in Washington, he had consistently supported the means and ends of national security, even when they encroached on the ideals of liberty. His staff, unlike the 9/11 Commission's, was deeply experienced in intelligence operations and analysis.

Their judgment was brutal and final. The realm of the director of central intelligence was "a closed world" with "an almost perfect record" of resisting change. The director had presided over a "fragmented, loosely managed, and poorly coordinated" patchwork of intelligence collection and analysis. The agency was "often unable to gather intelligence on the very things we care the most about" and its analysts "do not always tell decision-makers just how limited their knowledge really is." The CIA was "increasingly irrelevant to the new challenges presented by weapons of mass destruction." Its overriding flaw was "poor human intelligence"--an inability to conduct espionage.

"We recognize that espionage is always chancy at best; fifty years of pounding away at the Soviet Union resulted in only a handful of truly important human sources," the commission said. "Still, we have no choice but to do better." The CIA "needs fundamental change if it is to successfully confront the threats of the 21st century." That was "a goal that would be difficult to meet even in the best of all possible worlds. And we do not live in the best of worlds."

On April 21, 2005, the office of director of central intelligence disappeared into history. Goss called Negroponte's swearing-in "the burial ceremony" for the agency of old. On that day, the new boss received an odd blessing: "I hope the spirit of Wild Bill Donovan guides and inspires his efforts," said Senator John Warner of Virginia, chairman of the Armed Services Committee.

A bronze statue of Donovan stands guard at the entrance to CIA headquarters, where every living former director of central intelligence gathered at Goss's invitation on August 21, 2005, to receive medallions commemorating their service and to mark the end of their long line. George H. W. Bush was there, at the center that bore his name. So were Jim Schlesinger and Stan Turner, so bitterly resented as outsiders; Bill Webster and Bob Gates, failed reformers and restorers; Jim Woolsey, John Deutch, and George Tenet, who had fought to right a ship that had lost its bearings. Some of these men cheerfully despised one another; others shared deep bonds of trust. It was a pleasant enough wake, with a touch of pomp. There was a luncheon and a lecture on the vanished office from the CIA's chief historian, David S. Robarge. Goss sat in the front row, writhing inside. He had spent weeks agonizing over an inspector general's report he himself had demanded while still chairman of the House intelligence committee. It was a scathing look at the flaws that contributed to the 9/11 attacks, a knife in the agency's heart, a surgical examination of its inability to wage anything resembling a war against the nation's enemies. In the tradition of Allen Dulles, Goss had decided to bury it. The agency would never account for its failure to protect the United States. But in truth the reckoning had come to pass.

The CIA's historian recalled President Eisenhower's words when he came to lay the cornerstone for the agency's headquarters on November 3, 1959:

America's fundamental aspiration is the preservation of peace. To this end we seek to develop policies and arrangements to make the peace both permanent and just. This can be done only on the basis of comprehensive and appropriate information.

In war nothing is more important to a commander than the facts concerning the strength, dispositions, and intentions of his opponent, and the proper interpretation of those facts. In peacetime the necessary facts...and their correct interpretation are essential to the development of policy to further our long-term national security and best interests.... No task couldbe more important. Upon the quality of your work depends in large measure the success of our effort to further the nation's position in the international scene.... This agency demands of its members the highest order of dedication, ability, trustworthiness, and selflessness--to say nothing of the finest type of courage, whenever needed. Success cannot be advertised: failure cannot be explained. In the work of intelligence, heroes are undecorated and unsung.

"On this spot will rise a beautiful and useful structure," the president had concluded. "May it long endure, to serve the cause of America and of peace."

As Americans died in battle for want of the facts, the directors of central intelligence arose, shook hands, walked out into the heat of the summer afternoon, and went on with their lives. As the old soldier feared long ago, they had left a legacy of ashes.

"ADMIT NOTHING, DENY EVERYTHING"

On May 5, 2006, President Bush fired Porter Goss after nineteen months of ceaseless backstabbing at the CIA. The fall of the last director of central intelligence was swift and inglorious, and the bequest he left was bitter.

The next day, Goss got on a plane and delivered the commencement address at Tiffin University, ninety miles west of Cleveland, Ohio. "If this were a graduating class of CIA case officers, my advice would be short and to the point," he said. "Admit nothing, deny everything, and make counteraccusations." With those words, he disappeared from view, leaving behind the weakest cadre of spies and analysts in the history of the CIA.

One week after Goss resigned, a team of FBI agents raided CIA headquarters. They seized control of the office of Dusty Foggo, who had just stepped down as executive director, the third-highest job at the agency. This was the man whom Goss had inexplicably put in charge of running the CIA from day to day. In his previous post, Foggo had been quartermaster for the clandestine service. Based in Frankfurt, he kept CIA officers from Amman to Afghanistan supplied with everything from bottled water to body armor. Among his tasks was ensuring that his own accountants and cargo-kickers complied with the CIA's rules and regulations. "Having been the 'Ethic's Guy,'" he wrote to a fellow officer, "I wish you the best with this annual exercise." Foggo evidently had trouble with the word ethics.

The indictment in United States of America v. Kyle Dustin Foggo was painfully specific in its particulars. Unsealed on February 13, 2007, it charged Foggo with fraud, conspiracy, and money laundering. It said Foggo had fixed million-dollar contracts for a close friend who had wined and dined him in high style, treated him to extravagant trips to Scotland and Hawaii, and promised him a lucrative job--old-fashioned palm greasing. There had never been a case remotely like it in the history of the CIA. At this writing, Foggo has entered a plea of not guilty. He faces twenty years in prison if convicted.

On the same day that Foggo was indicted, a federal judge in North Carolina sentenced a CIA contract worker named David Passaro to eight years and four months in prison for beating a man to death in Afghanistan. Passaro served with a CIA paramilitary team based in Asadabad, the capital of Kunar Province, a few miles west of the border with Pakistan. The agency had hired Passaro despite his history of criminal violence; he had been fired from the police force in Hartford, Connecticut, after he was arrested for beating a man in a brawl.

The man who died at his hands was Abdul Wali, a well-known local farmer who had fought the Soviets in the 1980s. Wali had heard that he was wanted for questioning after a series of rocket attacks near the American base. He came to the Americans of his own free will, and he said he was innocent. Passaro doubted his word and threw him in a cell. He beat Wali so badly that the prisoner pleaded to be shot to end his pain; he died of his injuries two days later. Passaro was indicted and convicted under a provision of the Patriot Act allowing American citizens to be tried for crimes committed on territory claimed by the United States overseas. The judge noted that the absence of an autopsy had shielded him from a charge of murder.

The court received a letter from the former governor of Kunar, who said that Wali's death had done grievous damage to the American cause in Afghanistan and served as powerful propaganda for the resurgent forces of al Qaeda and the Taliban. "The distrust of the Americans increased, the security and reconstruction efforts of Afghanistan were dealt a blow, and the only people to gain from Dave Passaro's actions were al Qaeda and their partners," the governor wrote.

Three days after Passaro was sentenced, a judge in Italy ordered the indictment of the CIA's Rome station chief, the Milan base chief, and two dozen more officers in the abduction of a radical cleric who spent years under brutal interrogation in Egypt. A court in Germany charged thirteen CIA officers for the wrongful kidnapping and imprisonment of a Lebanese-born German citizen. The government of Canada formally apologized and paid a $10 million settlement to one of its citizens, Maher Arar, who had been seized by the CIA while changing planes in New York after a family vacation, transported to Syria, and subjected to the cruelest interrogation for ten months.

By then, the CIA's prison system had been condemned. It could not long survive when it was no longer secret. Americans were asked to take it on faith that the kidnapping, imprisonment, and torture of innocent people had been part of a program essential to preventing another attack on the United States. It may be so, but the evidence is scant. It is unlikely that we will ever know.

Porter Goss was succeeded at the CIA by General Michael Hayden, the deputy director of national intelligence, the former chief of the National Security Agency, the executor of President Bush's orders to train electronic eavesdropping on American targets, the first man to hold the diminished title of director of the Central Intelligence Agency, and the first active-duty military officer to run the CIA since Walter Bedell Smith left in 1953. General Hayden declared at his Senate confirmation that "amateur hour" was over at the CIA. But it was not.

By the CIA's own standards, roughly half its work force were still trainees. Few were ready and able to produce results. But there was nothing to be done about it; the CIA had no choice but to promote them beyond their levels of ability. As youngsters in their twenties replaced people in their forties and fifties, the result was an abridgment of intelligence. The clandestine service began to abandon the techniques of the past--political warfare, propaganda, and covert action--because it lacked the skills to conduct them. The agency remained a place where few people spoke Arabic or Persian, Korean or Chinese. It still denied employment to patriotic Arab Americans on security grounds if they had relatives living in the Middle East--as most did. The information revolution had left officers and analysts no more capable of comprehending the terrorist threat than they had understood the Soviet Union. And as the agency's reporting was overtaken by catastrophe in Iraq, the fifth Baghdad station chief in less than four years packed his bags for the closed world of the Green Zone.

The CIA was at a nadir. It no longer had the president's ear, and American leaders were looking elsewhere for intelligence--to the Pentagon and private industry.

"THE DISASTROUS RISE OF MISPLACED POWER"

Bob Gates took over the Pentagon on December 18, 2006--the only entry-level analyst ever to run the CIA and the only director ever to become secretary of defense. Two weeks later, John Negroponte, the new national intelligence czar, resigned after nineteen months to become the number-two man at the State Department. He was replaced by a retired admiral, Mike McConnell, who had run the National Security Agency during its first great collapse at the dawn of the digital age and who had spent the past decade making money as a military contractor at Booz Allen Hamilton.

When Gates settled in at the Pentagon, he looked around at the American intelligence establishment and he saw stars: a general was running the CIA, a general was the undersecretary of defense for intelligence, a general was in charge of State Department's counterterrorism programs, a lieutenant general was the Pentagon's deputy undersecretary for intelligence, and a major general was running spies at the CIA. Every one of these jobs had been held by civilians, going back many years. Gates saw a world in which the Pentagon had crushed the CIA, just as it had vowed to do sixty years before. He wanted to close the military prison at Guantanamo Bay, bring the suspected terrorists from Cuba to the United States, and either convict them or recruit them. He wanted to contain the Defense Department's dominance over intelligence. He longed to reverse the decline in the CIA's central role in American government. But there was very little he could do.

The decline was part of a slow rot undermining the pillars of American national security. After four years of war in Iraq, the military was exhausted, bled by leaders who had invested far more in futuristic weapons than in uniformed soldiers. After five years of defending a foreign policy based on born-again faith, the State Department was adrift, unable to give voice to the values of democracy. And after six years of willful ignorance imposed by know-nothing politicians, congressional oversight of the agency had collapsed. The 9/11 Commission had said that of all the tasks facing American intelligence, strengthening congressional oversight might be the most difficult, and the most important. In 2005 and 2006, Congress responded by failing to pass the annual authorization bill for the CIA, the basic law governing the agency, its policies, and its spending. The roadblock was a single Republican senator who obstructed the bill because it ordered the White House to file a classified report on the CIA's secret prisons.

The failure of authority made the congressional intelligence committees irrelevant. Not since the 1960s had there been so little congressional control over the agency. Now a far different force gained great influence over intelligence: corporate America.

At the end of Dwight Eisenhower's years as president, a few days after he lamented the legacy of intelligence failures he would pass on to his successors, he gave his farewell address to the nation and famously warned: "We must guard against the acquisition of unwarranted influence, whether sought or unsought, by the military-industrial complex. The potential for the disastrous rise of misplaced power exists and will persist." Little more than half a century later, the surge of secret spending on national security after 9/11 had created a booming intelligence-industrial complex.

Corporate clones of the CIA started sprouting all over the suburbs of Washington and beyond. Patriotism for profit became a $50-billion-a-year business, by some estimates--a sum about the size of the American intelligence budget itself. This phenomenon traced back fifteen years. After the cold war, the agency began contracting out thousands of jobs to fill the perceived void created by the budget cuts that began in 1992. A CIA officer could file his retirement papers, turn in his blue identification badge, go to work for a much better salary at a military contractor such as Lockheed Martin or Booz Allen Hamilton, then return to the CIA the next day, wearing a green badge. After September 2001, the outsourcing went out of control. Green-badge bosses started openly recruiting in the CIA's cafeteria.

Great chunks of the clandestine service became wholly dependent on contractors who looked like they were in the CIA's chain of command, but who worked for their corporate masters. In effect, the agency had two workforces--and the private one was paid far better. By 2006 something on the order of half the officers at the Baghdad station and the new National Counterterrorism Center were contract employees, and Lockheed Martin, the nation's biggest military contractor, was posting help-wanted ads for "counterterrorism analysts" to interrogate suspected terrorists at the Guantanamo prison.

Fortunes could be made in the intelligence industry. The money was a powerful attractor, and the result was an ever-accelerating brain drain--the last thing the CIA could afford--and the creation of companies like Total Intelligence Solutions. Founded in February 2007, Total Intel was run by Cofer Black--the chief of the CIA's counterterrorist center on 9/11. His partners were Robert Richer, who had been the number-two man at the clandestine service, and Enrique Prado, Black's chief of counterterror operations. All three had decamped from the Bush administration's war on terror in 2005 to join Blackwater USA, the politically wired private security company that served, among many other things, as the Praetorian Guard for Americans in Baghdad. They learned the tricks of the government-contracting trade at Blackwater, and within little more than a year Black and company were running Total Intel. These were among the best of the CIA's officers. But the spectacle of jumping ship in the middle of a war to make a killing was unremarkable in twenty-first-century Washington. Legions of CIA veterans quit their posts to sell their services to the agency by writing analyses, creating cover for overseas officers, setting up communications networks, and running clandestine operations. Following their example, new CIA hires adopted their own five-year plan: get in, get out, and get paid. A top secret security clearance and a green badge were golden tickets for a new breed of Beltway bandits. The outsourcing of intelligence was a clear sign that the CIA could not perform many of its basic missions unaided after 9/11.

Above all, it could not help the army impose democracy at gunpoint in Iraq. Action without knowledge was a dangerous business, as Americans found to their sorrow.

"TO ORGANIZE AND RUN AN ESPIONAGE SERVICE"

In the cold war, the CIA was condemned by the American left for what it did. In the war on terror, the CIA was attacked by the American right for what it could not do. The charge was incompetence, leveled by such men as Dick Cheney and Don Rumsfeld. Say what one may about their leadership, they knew from long experience what the reader now knows: the CIA was unable to fulfill its role as America's intelligence service.

The fictional CIA, the one that lives in novels and movies, is omnipotent. The myth of a golden age was of the CIA's own making, the product of the publicity and the political propaganda Allen Dulles manufactured in the 1950s. It held that the agency could change the world, and it helps explain why the CIA is so impervious to change. The legend was perpetuated in the 1980s by Bill Casey, who tried to revive the devil-may-care spirits of Dulles and Wild Bill Donovan. Now the agency has revived the fable that it is America's best defense. With orders to train and retain thousands of new officers, it needs to project an image of success to survive.

In truth, there haven't been many halcyon days. But there have been a few. When Richard Helms was in charge, the agency spoke the truth to Lyndon Johnson and Robert McNamara about the war in Vietnam, and they listened. There was another such fleeting moment when Bob Gates ran the CIA; he kept calm and carried on as the Soviet Union crumbled. But fifteen years have passed since then, and the glory is gone. The CIA found itself unable to see the way forward in a battle where information and ideas were the most powerful weapons.

For sixty years tens of thousands of clandestine service officers have gathered only the barest threads of truly important intelligence--and that is the CIA's deepest secret. Their mission is extraordinarily hard. But we Americans still do not understand the people and the political forces we seek to contain and control. The CIA has yet to become what its creators hoped it would be.

"The only remaining superpower doesn't have enough interest in what's going on in the world to organize and run an espionage service," Richard Helms said a decade ago. Perhaps a decade from now the agency will rise from the ashes, infused with many billions of dollars, inspired by new leadership, invigorated by a new generation. Analysts may see the world clearly. American spies may become capable of espionage against America's enemies. The CIA someday may serve as its founders intended. We must depend on it. For the war in which we are now engaged may last as long as the cold war, and we will win or lose by virtue of our intelligence.

ACKNOWLEDGMENTS

I am fortunate to have spent part of the past twenty years talking with CIA directors and officers whose professional lives spanned the course of six decades. I am particularly grateful to Richard Helms, William Colby, Stansfield Turner, William Webster, Bob Gates, John Deutch, George Tenet, John McMahon, Tom Twetten, Milt Bearden, Tom Polgar, Peter Sichel, Frank Lindsay, Sam Halpern, Don Gregg, Jim Lilley, Steve Tanner, Gerry Gossens, Clyde McAvoy, Walter Pforzheimer, Haviland Smith, Fred Hitz, and Mark Lowenthal. A tip of the hat goes to the men and women of the CIA's history staff, who do their part for the cause of openness in the face of fierce resistance from the clandestine service, and to present and former members of the agency's public affairs staff.

I am deeply in debt to the work of Charles Stuart Kennedy, a retired Foreign Service officer and the founder and director of the Foreign Affairs Oral History Program. The library he has created is a unique and invaluable resource. The State Department's historians, who produce The Foreign Relations of the United States, the official record of American diplomacy, published since 1861, have done more in the past decade to unseal secret documents than any other arm of government. They, along with the staffs of the presidential libraries, deserve the thanks of a grateful nation.

A reporter is lucky to have one great editor in a lifetime. I have had more than my share, and over the years they have given me time to think and freedom to write. Gene Roberts gave me my start at the Philadelphia Inquirer. Bill Keller, Jill Abramson, Andy Rosenthal, and Jon Landman help make The New York Times a daily miracle. They are keepers of a public trust.

Three tireless researchers helped create this book. Matt Malinowski transcribed interview tapes, Zoe Chace dug into the diplomatic history and the National Security Council files, and Cora Currier did groundbreaking research at the National Archives. I am grateful to my high school chum Lavinia Currier for introducing me to her fiercely intelligent daughter. Zoe is the daughter of the late James Chace and the sister of Beka Chace, two friends whose spirits sustain me.

I want to salute the journalists who have covered the CIA, the struggles in Iraq and Afghanistan, and the agonies of American national security since 9/11. Among them are John Burns, Dexter Filkins, Matt Purdy, Doug Jehl, Scott Shane, Carlotta Gall, John Kifner, and Steve Crowley of The New York Times; Dana Priest, Walter Pincus, and Pam Constable of The Washington Post; Vernon Loeb, Bob Drogin, and Megan Stack of the Los Angeles Times; and Andy Maykuth of The Philadelphia Inquirer. We remember our brothers and sisters who gave their lives to get the news, among them Elizabeth Neuffer, Mark Fineman, Michael Kelly, Harry Burton, Azizullah Haidari, Maria Grazia Cutuli, and Julio Fuentes.

My gratitude goes to Phyllis Grann, who had the good grace to edit and publish this book, and to Kathy Robbins, the world's most brilliant literary agent.

Legacy of Ashes took shape at Yaddo, the retreat for artists and writers in Saratoga Springs, New York. For two months, the good people of Yaddo housed and fed me while thousands of words a day went pouring into my ThinkPad. I was honored to be the first recipient of the Nora Sayre Endowed Residency for Nonfiction, created in her memory to support her literary legacy. A thousand thanks to the poet Jean Valentine for introducing me to Yaddo; to Elaina Richardson, president of the Corporation of Yaddo; and to the trustees, supporters, and employees of this magnificent refuge.

The book grew longer and stronger at the house of my in-laws, Susanna and Boker Doyle, who supported me with their great good nature.

My will to write began when I first saw my mother, Professor Dora B. Weiner, working on a book in the basement of our home in the quiet before dawn. Forty-five years later, she is still writing and teaching and inspiring her students and her sons. All of us wish my father were here to hold this book in his hands.

Legacy of Ashes ends as it began, with a dedication to the love of my life, Kate Doyle; to our daughters, Emma and Ruby; and to the rest of our lives together.