In this pathbreaking book, Amy Zegart provides the first scholarly examination of the intelligence failures that preceded September 11. Until now, those failures have been attributed largely to individual mistakes. But Zegart shows how and why the intelligence system itself left us vulnerable.
Zegart argues that after the Cold War ended, the CIA and FBI failed to adapt to the rise of terrorism. She makes the case by conducting painstaking analysis of more than three hundred intelligence reform recommendations and tracing the history of CIA and FBI counterterrorism efforts from 1991 to 2001, drawing extensively from declassified government documents and interviews with more than seventy high-ranking government officials. She finds that political leaders were well aware of the emerging terrorist danger and the urgent need for intelligence reform, but failed to achieve the changes they sought. The same forces that have stymied intelligence reform for decades are to blame: resistance inside U.S. intelligence agencies, the rational interests of politicians and career bureaucrats, and core aspects of our democracy such as the fragmented structure of the federal government. Ultimately failures of adaptation led to failures of performance. Zegart reveals how longstanding organizational weaknesses left unaddressed during the 1990s prevented the CIA and FBI from capitalizing on twenty-three opportunities to disrupt the September 11 plot.
Spying Blind is a sobering account of why two of America's most important intelligence agencies failed to adjust to new threats after the Cold War, and why they are unlikely to adapt in the future.
There are no customer reviews available at this time. Would you like to write a review?
Princeton University Press
March 08, 2009
Number of Print Pages*
Adobe DRM EPUB
* Number of eBook pages may differ. Click here for more information.
Excerpt from Spying Blind by Amy B. Zegart
An Organizational View of 9/11
I was not surprised. I was horrified.
--General Brent Scowcroft, former national security advisor1
IN JANUARY 2000, al Qaeda operatives from around the world gathered secretly in Malaysia for a planning meeting. The Central Intelligence Agency (CIA) was watching. Among the participants was a man named Khalid al-Mihdhar, one of the September 11 hijackers who would later help to crash American Airlines Flight 77 into the Pentagon. By the time the meeting disbanded, the CIA had taken a photograph of al-Mihdhar, learned his full name, obtained his passport number, and uncovered one other critical piece of information: al-Mihdhar held a multiple-entry visa to the United States.2 It was twenty months before the September 11, 2001, terrorist attacks on the World Trade Center and Pentagon. George Tenet, the director of central intelligence (DCI), later admitted that the CIA should have immediately placed al-Mihdhar on the State Department's watch list denying him entry into the United States, and it should have notified other government agencies such as the FBI.3 But the CIA did not do so until August 23, 2001, just nineteen days before the attacks and months after al-Mihdhar had entered the country, obtained a California motor vehicle photo identification card--using his real name--and started taking flying lessons.
The case of Khalid al-Mihdhar provides a chilling example of the subtle yet powerful effects of organization--that is, the cultures, incentives, and structures that critically influence what government agencies do and how well they do it. Why did the CIA take so long to put this suspected al Qaeda operative on the State Department's watch list, especially given Director Tenet's earlier declaration that the United States was "at war" with al Qaeda,4 his clear public warnings to Congress--for three consecutive years--that Osama bin Laden was determined to strike major blows against American targets,5 and when intelligence chatter about preparations for a "spectacular" attack was spiking in the spring and summer of 2001?6
The simplest answer is that the agency had never been in the habit of watch listing al Qaeda operatives before. For more than forty years, the Central Intelligence Agency and the twelve other agencies of the U.S. Intelligence Community (IC) had operated with Cold War procedures, priorities, and thinking, all of which had little need for making sure foreign terrorists stayed out of the United States.7 Before September 11, there was no formal training program, no well-honed process, and no sustained level of attention given to ensuring that intelligence officers would identify dangerous terrorists and warn other U.S. government agencies about them before they reached the United States.8 As one CIA employee told congressional investigators after the September 11 attacks, he believed it was "not incumbent" even on the CIA's special Osama bin Laden unit to place people such as al-Mihdhar on the State Department's watch list.9
No one will ever know whether the World Trade Center and Pentagon attacks could have been prevented. Evidence suggests, however, that the right information did not get to the right places at the right time. Many of the agonizing missteps and missed clues have been widely publicized. There is the star Phoenix FBI agent who warned in a July 2001 memo that Osama bin Laden could be training terrorists in U.S. flight schools, a warning that never made it to the top of the FBI or a single other intelligence agency. There is the refusal by FBI headquarters to seek a search warrant for the computer files of Zacarias Moussaoui, a foreign flight school student who Minneapolis field agents were convinced was plotting a terrorist attack with a large aircraft and who later became the only person convicted in the United States for his connection to the 9/11 attacks.10 And there is the president's August 6, 2001, CIA briefing entitled "Bin Laden Determined to Strike in U.S." which gave the impression the FBI had the threat covered, erroneously suggested that Yemeni tourists taking photographs were terrorists casing federal buildings in New York, and made no mention of crucial pieces of information that should have been pursued aggressively. These included the Phoenix memo, the al Qaeda summit in Malaysia, al-Mihdhar's U.S. visa, and the CIA's discovery that a second September 11 hijacker who had attended the summit, Nawaf al-Hazmi, had also entered the United States.11 Thanks to the extraordinary work of the 9/11 Commission and the House and Senate Intelligence Committees' Joint Inquiry into the attacks, most Americans have a good idea of what went wrong in the weeks and months before September 11.12 The challenge now is to explain why it went wrong.
This book argues that the answer lies in organizations, more specifically, in the deeply rooted organizational weaknesses that have afflicted U.S. intelligence agencies for decades and in the enduring impediments to fixing them. The single most important reason the United States remained so vulnerable on September 11 was not the McDonald's wages paid to airport security workers, the Clinton administration's inability to capture or kill Osama bin Laden, or the Bush administration's failure to place terrorism higher on its priority list. It was the stunning inability of U.S. intelligence agencies to adapt to the end of the Cold War.
During the 1990s, for example, intelligence officials repeatedly warned of a grave and growing terrorist threat even while they continued old funding patterns that favored electronic surveillance--ideal for counting Soviet warheads--over human intelligence efforts better suited for penetrating terrorist groups. Although details about U.S. intelligence spending are classified, conservative estimates based on the declassified 1997 intelligence budget put annual human intelligence spending at $1.6 billion, a little more than the cost of building and launching a single spy satellite.13 The amount of money spent directly to support human intelligence operations in the field was even less. As one official with access to the CIA's human intelligence budget put it, once pensions, salaries, and other expenses were paid, the "The James Bond fund that people think we're doing came down to $500 million," or less than 2 percent of the annual intelligence budget at the time.14
Counterterrorism efforts were as scattered as they were underfunded, split among forty-six different agencies without a central strategy, budget, or coordinating mechanism. Director of Central Intelligence George Tenet declared war on Osama bin Laden in a December 1998 memo and urged that "no resources or people [be] spared" to fight him, but proved unable to mass his troops in the right places.15 Although Tenet tried to increase dramatically the size of the Counterterrorist Center, he failed,16 leaving only five analysts assigned to Osama bin Laden on September 11.17
The CIA was not alone. The FBI formally declared terrorism its number one priority as early as 1998.18 Yet on September 11, 2001, only 6 percent of FBI personnel were working on counterterrorism issues,19 new agents still received more time for vacation than counterterrorism training,20 and the vast majority of the FBI's intelligence analysts--precisely the people who were charged with connecting the dots across different FBI cases-- were found to be unqualified to perform their jobs.21 Steeped in an eighty-year-old culture that prized searching houses more than searching databases, the agency lacked basic computer capabilities to see whether the words "flight training school" showed up in any of its case files and even the FBI Director, Louis Freeh, ordered the computer removed from his office because he never used it.22 In the words of one FBI official, the prevailing attitude was, "real men don't type. The only thing a real agent needs is a notebook, a pen, and a gun, and with those three things you can conquer the world."23 Just weeks before the attacks, a highly classified internal review of the bureau's counterterrorism capabilities gave failing grades to every one of the FBI's fifty-six U.S. field offices.24
These problems were not isolated mistakes, failures of foresight, or the result of poor decisions by individuals asleep at the switch. Instead, they were symptoms of three deeper and more intractable organizational deficiencies: (1) cultural pathologies that led intelligence agencies to resist new technologies, ideas, and tasks; (2) perverse promotion incentives that rewarded intelligence officials for all of the wrong things; and (3) structural weaknesses dating back decades that hindered the operation of the CIA and FBI and prevented the U.S. Intelligence Community from working as a coherent whole. It was these core weaknesses that caused U.S. intelligence agencies to blow key operational opportunities--such as watchlisting al-Mihdhar or searching Zacarias Moussaoui's computer files--that might have disrupted the September 11 plot. And it was these core weaknesses that kept U.S. intelligence agencies from getting more chances to defeat al Qaeda in the first place. With FBI agents keeping case files in shoe boxes rather than putting them into computers, with CIA operatives clinging to old systems designed for recruiting Soviet officials at cocktail parties rather than Jihadists in caves, with career incentives that rewarded intelligence officials for staying cloistered in their own agencies rather than working across agency lines, and with a forty-year-old intelligence structure that gave no person the power to match resources against priorities and knock bureaucratic heads together, the U.S. Intelligence Community did not have a fighting chance against al Qaeda.
The existence of these organizational deficiencies, and the urgent need to fix them, was no secret in Washington before the September 11 attacks. Between 1991 and 2001, intelligence problems and counterterrorism challenges were the subject of at least six classified reports25 and a dozen major unclassified studies.26 The unclassified studies alone issued more than 500 recommendations for reform across the U.S. government. Two-thirds of these recommendations, or 340 in total, targeted the CIA, FBI, and the rest of the U.S. Intelligence Community. Yet only 35 of these 340 intelligence recommendations were successfully implemented before September 11, and most--268 to be exact--resulted in no action whatsoever.27 In January 2001, nine months before the attacks, the bipartisan blue-ribbon Hart-Rudman Commission offered the most comprehensive assessment of U.S. national security challenges and deficiencies since World War II. The commission issued stark conclusions: "the dramatic changes in the world since the end of the Cold War," it noted, "have not been accompanied by any major institutional changes in the Executive branch of the U.S. government."28 The commission presciently predicted that institutional deficiencies left the United States homeland exceptionally vulnerable to catastrophic terrorist attack.
No system is failure-proof. As Richard Betts wrote in Foreign Affairs shortly after September 11, "The awful truth is that even the best intelligence systems will have big failures."29 Evidence suggests, however, that U.S. intelligence agencies were nowhere close to being the best before 9/11, and that they could have been better. When the Soviet Union fell in 1991 and the principal threat to U.S. national security changed, the Intelligence Community was slow to change with it.
Why? What is it that prevented the CIA, the FBI, and other agencies from adapting to the rising terrorist threat during the 1990s? To date, no one has provided satisfying answers. Academics have avoided the subject, concentrating instead on research topics that have more readily available data, fit more squarely into existing theories, and do not require delving into the controversial business of spying. At the same time, politicians and journalists have preferred to point fingers, focusing on who failed to do what and when. The result is a prevailing wisdom that mistakenly attributes the failures of September 11 to individuals.
THE FINGER POINTING FALLACY
Everyone has someone to blame for 9/11. Democrats such as former Clinton National Security Advisor Samuel Berger and Secretary of State Madeleine Albright have faulted President Bush and his administration for giving terrorism short shrift compared to missile defense and other foreign policy issues.30 Republicans, including Secretary of State Condoleezza Rice and Vice President Richard Cheney, have charged the Clinton administration with failing to develop an effective counterterrorism strategy and emboldening bin Laden by responding weakly to earlier terrorist attacks.31 Some, such as former Senate Intelligence Committee Chairman Richard Shelby (R-AL), have laid responsibility squarely on the shoulders of George Tenet, who served as director of central intelligence from 1997 to 2004.32 The most blistering criticism came in the spring of 2004, when Richard Clarke, the White House counterterrorism czar under both Presidents Clinton and George W. Bush, accused Bush and his top aides of dropping the ball on terrorism.33 Although different accusers have found different culprits, their point is the same: individual leadership failures are the root cause of the World Trade Center and Pentagon attacks.
Attributing failure to individuals is both understandable and dangerous. Casting blame after moments of great tragedy is a natural human response. It also makes for good politics and great journalism. No one should be shocked that politicians from both parties rushed to accuse and defend, and the press rushed to cover them. Nor is it surprising that nearly all of the 9/11 books penned by journalists since the attacks have focused on the human causes of tragedy, dissecting the power plays and personality clashes between various intelligence officers in the field and policymakers in Washington.34 It is the nature of the business: journalists usually place individuals at the heart of the story rather than examining the forces that transcend them and tend to rely on anecdotal evidence to meet tight deadlines rather than studying a single problem in systematic and gory detail over a number of years. The best of this genre--such as Steve Coll's Ghost Wars and Lawrence Wright's The Looming Tower-- have much to offer, yet nevertheless emphasize individual failures more than systemic ones. The worst in the genre suffer from what Malcolm Gladwell calls "creeping determinism," a tendency to view pre-9/11 warnings through post-9/11 lenses.35 In hindsight, of course, smoking guns are everywhere--a 1995 report sent to the CIA from Philippine authorities noting that a captured terrorist had plans to fly an airplane into CIA headquarters,36 or an al Qaeda telephone call intercepted during the summer of 2001 that mentions a "terrifying" attack using an airplane.37 But proving that intelligence officials could and should have seen these signals as ominous warnings beforehand is quite another matter.
As the above discussion indicates, highlighting the role of individuals is also dangerous because it suggests the wrong causes of failure and the wrong remedies to address them. We are left to think that if only the right people had been heard, if only a few important officials had connected a few obvious dots, if only more leaders inside the corridors of power had had their hair on fire, tragedy could have been averted. As Bob Woodward, the dean of journalist nonfiction, once wrote, "Decision making at the highest levels of national government is a complex human interaction. . . . This human story is the core."38
Actually, the human story is the problem. What is missing from these accounts is a sense of context, the underlying constraints and forces that make it likely talented people will make poor decisions. It is easy, for example, to blame intelligence officials for overlooking warnings about a terrorist attack in an intercepted telephone conversation. It is much harder when one considers that several million such conversations are intercepted by intelligence officials every day of every week of every year.39 Journalists, the old saying goes, write the first draft of history. In the case of September 11, however, journalists have provided the only draft of history. The fault is not theirs, but ours in the academy: political scientists have devoted almost no attention to studying U.S. intelligence since 9/11.40 The result is that the role of individuals in September 11 has been grossly overstated, while the organizational causes of failure have gone largely unexamined.
Consider the strongest case that has been made for the importance of individual leadership: the Clinton administration's successful prevention of terrorist attacks during the Millennium celebration. The 9/11 Commission found that from December 1999 to January 2000, things were clearly different. Told that there would be between five and fifteen terrorist attacks on domestic targets, senior Clinton officials were at battle stations.41 The national security advisor, the FBI director, the director of central intelligence, the attorney general, the secretary of state, Pentagon officials, the NSC staff--all of the federal government's senior most officials were focused as never before on one goal: preventing a terrorist attack inside the United States. National Security Advisor Samuel Berger held White House meetings with top officials every day for a month.42 Information was flowing, both across agencies and up the chains of command. The commission concluded that the weeks surrounding the Millennium constituted the "one period in which the government as a whole seemed to be acting in concert to deal with terrorism."43 During the summer of 2001, by comparison, the commission found no corresponding senior level push inside the Bush administration to focus and shake up the bureaucracy.44
In the eyes of many, this sense of urgency and attention at the senior levels made all the difference: The presence of high-level leadership led to success in thwarting the Millennium attacks, while its absence in the summer of 2001 led to failure on September 11. As Berger put it, "Things happen when the number one person is in the room. . . . When the principal spends an hour a day at the White House or more, he goes back or she goes back to his agency or her agency and he or she shakes that agency for whatever it has."45 Berger told the 9/11 Commission that he was "convinced that our sustained attention . . . prevented significant losses of life."46 Former counterterrorism czar Richard Clarke was more blunt, suggesting that if only the Bush administration had urged senior officials to "shake the trees" in the summer of 2001, as Clinton officials had done during the Millennium period, they would have discovered what lower-level agents in the FBI already knew: that al Qaeda operatives were in the United States.47 "If Condi Rice had been doing her job and holding those daily meetings, the way Sandy Berger did, if she had a hands-on attitude to being national security adviser, when she had information that there was a threat against the United States, that kind of information was shaken out in December 1999, it would have been shaken out in the summer of 2001," Clarke declared on CNN's Larry King Live.48
The problem here is not facts, but logic. It seems clear that senior Clinton officials devoted more attention to thwarting imminent terrorist attacks during the Millennium period than their Bush administration counterparts did in the summer of 2001. The question is, did that attention matter? It could be that individual leadership made all the difference, forcing the system to move information to the right places and preventing disaster. Or maybe not. When it comes to cause and effect relationships, looks are often deceiving. In public policy 101, this is called the "correlation versus causation" problem: two trends can coincide without strong causal connections between them. If I said, for example, that my hair length historically corresponded to the rise and fall of the U.S. stock market, surely no reader would start adjusting his investments according to my hair styles. Leadership is obviously more connected to counterterrorism than hair length is to the Dow. But the point is the same: causal relationships cannot be assumed. Instead, they must be determined through careful analysis of facts.
In this case, closer examination suggests the individual leadership argument is not so convincing. The Clinton administration's own after-action report of the Millennium attacks concluded that the crucial break stemmed from luck and the experience of a low-level customs agent, not planning and the leadership of high-level officials.49 On December 14, 1999, an alert customs inspector named Diana Dean noticed a fidgety passenger driving his rental car off a ferry traveling from Canada to Port Angeles, Washington. Dean decided to detain the passenger for a secondary inspection. When she began patting him down, the passenger panicked and tried to flee. He turned out to be Ahmed Ressam, an Algerian Jihadist who was planning to blow up Los Angeles International Airport (LAX). When inspectors examined Ressam's car, they found hidden explosives and timing devices.50 Counterterrorism Chief Richard Clarke wrote that "strings from Ressam" led to a sleeper cell of Algerian mujahadeen in Montreal, which in turn led officials to what looked like sleeper cells in Boston and New York.51
Dean testified at Ressam's trial that it was her "training and experience" that led her to notice and stop Ressam.52 "I don't recall any specific threats," she later told reporters. "I don't recall anybody saying watch for terrorists."53 Her recollections turned out to be accurate. U.S. customs records reveal that the agency was not under any heightened state of alert.54 Senior officials in Washington, DC may have been at battle stations to stop a terrorist attack when Ressam's ferry crossed the Canadian-U.S. border, but Diana Dean and every other customs agent in the field were not. The 9/11 Commission concluded, "It appears that the heightened sense of alert at the national level played no role in Ressam's detention."55 In this particular case, the leadership of senior officials contributed much less to the outcome than most people believe.
My point is not that individual leadership never matters, but that the harder-to-see aspects of organizational life--such as training, procedures, cultures, and agency structures--often matter more. This is important, both for our understanding of the past and our expectations of the future. Indeed, if individual leadership determined counterterrorism success and failure, then fixing U.S. intelligence agencies would be easy. One need only identify the few bad apples and toss, or vote, them out. The reality is much worse. Yes, individuals made mistakes, but it was the system that failed us.