3 COUNTERTERRORISM EVOLVES

In chapter 2, we described the growth of a new kind of terrorism, and a new terrorist organization-especially from 1988 to 1998, when Usama Bin Ladin declared war and organized the bombing of two U.S. embassies. In this chapter, we trace the parallel evolution of government efforts to counter terrorism by Islamic extremists against the United States.

We mention many personalities in this report. As in any study of the U.S. government, some of the most important characters are institutions. We will introduce various agencies, and how they adapted to a new kind of terrorism.

3.1 FROM THE OLD TERRORISM TO THE NEW: THE FIRST WORLD TRADE CENTER BOMBING

At 18 minutes after noon on February 26, 1993, a huge bomb went off beneath the two towers of the World Trade Center. This was not a suicide attack. The terrorists parked a truck bomb with a timing device on Level B-2 of the underground garage, then departed. The ensuing explosion opened a hole seven stories up. Six people died. More than a thousand were injured. An FBI agent at the scene described the relatively low number of fatalities as a miracle.¹

President Bill Clinton ordered his National Security Council to coordinate the response. Government agencies swung into action to find the culprits. The Counterterrorist Center located at the CIA combed its files and queried sources around the world. The National Security Agency (NSA), the huge Defense Department signals collection agency, ramped up its communications intercept network and searched its databases for clues.² The New York Field Office of the FBI took control of the local investigation and, in the end, set a pattern for future management of terrorist incidents.

Four features of this episode have significance for the story of 9/11.

First, the bombing signaled a new terrorist challenge, one whose rage and malice had no limit. Ramzi Yousef, the Sunni extremist who planted the bomb, said later that he had hoped to kill 250,000 people.³

Second, the FBI and the Justice Department did excellent work investigating the bombing. Within days, the FBI identified a truck remnant as part of a Ryder rental van reported stolen in Jersey City the day before the bombing.⁴

Mohammed Salameh, who had rented the truck and reported it stolen, kept calling the rental office to get back his $400 deposit. The FBI arrested him there on March 4, 1993. In short order, the Bureau had several plotters in custody, including Nidal Ayyad, an engineer who had acquired chemicals for the bomb, and Mahmoud Abouhalima, who had helped mix the chemicals.⁵

The FBI identified another conspirator, Ahmad Ajaj, who had been arrested by immigration authorities at John F. Kennedy International Airport in September 1992 and charged with document fraud. His traveling companion was Ramzi Yousef, who had also entered with fraudulent documents but claimed political asylum and was admitted. It quickly became clear that Yousef had been a central player in the attack. He had fled to Pakistan immediately after the bombing and would remain at large for nearly two years.⁶

The arrests of Salameh, Abouhalima, and Ayyad led the FBI to the Farouq mosque in Brooklyn, where a central figure was Sheikh Omar Abdel Rahman, an extremist Sunni Muslim cleric who had moved to the United States from Egypt in 1990. In speeches and writings, the sightless Rahman, often called the "Blind Sheikh," preached the message of Sayyid Qutb's Milestones, characterizing the United States as the oppressor of Muslims worldwide and asserting that it was their religious duty to fight against God's enemies. An FBI informant learned of a plan to bomb major New York landmarks, including the Holland and Lincoln tunnels. Disrupting this "landmarks plot," the FBI in June 1993 arrested Rahman and various confederates.⁷

As a result of the investigations and arrests, the U.S. Attorney for the Southern District of New York prosecuted and convicted multiple individuals, including Ajaj, Salameh, Ayyad, Abouhalima, the Blind Sheikh, and Ramzi Yousef, for crimes related to the World Trade Center bombing and other plots.

An unfortunate consequence of this superb investigative and prosecutorial effort was that it created an impression that the law enforcement system was well-equipped to cope with terrorism. Neither President Clinton, his principal advisers, the Congress, nor the news media felt prompted, until later, to press the question of whether the procedures that put the Blind Sheikh and Ramzi Yousef behind bars would really protect Americans against the new virus of which these individuals were just the first symptoms.⁸

Third, the successful use of the legal system to address the first World Trade Center bombing had the side effect of obscuring the need to examine the character and extent of the new threat facing the United States. The trials did not bring the Bin Ladin network to the attention of the public and policymakers.

The FBI assembled, and the U.S. Attorney's office put forward, some evidence showing that the men in the dock were not the only plotters. Materials taken from Ajaj indicated that the plot or plots were hatched at or near the Khaldan camp, a terrorist training camp on the Afghanistan-Pakistan border. Ajaj had left Texas in April 1992 to go there to learn how to construct bombs. He had met Ramzi Yousef in Pakistan, where they discussed bombing targets in the United States and assembled a "terrorist kit" that included bomb-making manuals, operations guidance, videotapes advocating terrorist action against the United States, and false identification documents.⁹

Yousef was captured in Pakistan following the discovery by police in the Philippines in January 1995 of the Manila air plot, which envisioned placing bombs on board a dozen trans-Pacific airliners and setting them off simultaneously. Khalid Sheikh Mohammed-Yousef's uncle, then located in Qatar-was a fellow plotter of Yousef's in the Manila air plot and had also wired him some money prior to the Trade Center bombing. The U.S. Attorney obtained an indictment against KSM in January 1996, but an official in the government of Qatar probably warned him about it. Khalid Sheikh Mohammed evaded capture (and stayed at large to play a central part in the 9/11 attacks).¹⁰

The law enforcement process is concerned with proving the guilt of persons apprehended and charged. Investigators and prosecutors could not present all the evidence of possible involvement of individuals other than those charged, although they continued to pursue such investigations, planning or hoping for later prosecutions. The process was meant, by its nature, to mark for the public the events as finished-case solved, justice done. It was not designed to ask if the events might be harbingers of worse to come. Nor did it allow for aggregating and analyzing facts to see if they could provide clues to terrorist tactics more generally-methods of entry and finance, and mode of operation inside the United States.

Fourth, although the bombing heightened awareness of a new terrorist danger, successful prosecutions contributed to widespread underestimation of the threat. The government's attorneys stressed the seriousness of the crimes, and put forward evidence of Yousef's technical ingenuity. Yet the public image that persisted was not of clever Yousef but of stupid Salameh going back again and again to reclaim his $400 truck rental deposit.

3.2 ADAPTATION-AND NONADAPTATION-IN THE LAW ENFORCEMENT COMMUNITY

Legal processes were the primary method for responding to these early manifestations of a new type of terrorism. Our overview of U.S. capabilities for dealing with it thus begins with the nation's vast complex of law enforcement agencies.

The Justice Department and the FBI
At the federal level, much law enforcement activity is concentrated in the Department of Justice. For countering terrorism, the dominant agency under Justice is the Federal Bureau of Investigation. The FBI does not have a general grant of authority but instead works under specific statutory authorizations. Most of its work is done in local offices called field offices. There are 56 of them, each covering a specified geographic area, and each quite separate from all others. Prior to 9/11, the special agent in charge was in general free to set his or her office's priorities and assign personnel accordingly.¹¹

The office's priorities were driven by two primary concerns. First, performance in the Bureau was generally measured against statistics such as numbers of arrests, indictments, prosecutions, and convictions. Counterterrorism and counterintelligence work, often involving lengthy intelligence investigations that might never have positive or quantifiable results, was not career-enhancing. Most agents who reached management ranks had little counterterrorism experience. Second, priorities were driven at the local level by the field offices, whose concerns centered on traditional crimes such as white-collar offenses and those pertaining to drugs and gangs. Individual field offices made choices to serve local priorities, not national priorities.¹²

The Bureau also operates under an "office of origin" system. To avoid duplication and possible conflicts, the FBI designates a single office to be in charge of an entire investigation. Because the New York Field Office indicted Bin Ladin prior to the East Africa bombings, it became the office of origin for all Bin Ladin cases, including the East Africa bombings and later the attack on the USS Cole. Most of the FBI's institutional knowledge on Bin Ladin and al Qaeda resided there. This office worked closely with the U.S. Attorney for the Southern District of New York to identify, arrest, prosecute, and convict many of the perpetrators of the attacks and plots. Field offices other than the specified office of origin were often reluctant to spend much energy on matters over which they had no control and for which they received no credit.¹³

The FBI's domestic intelligence gathering dates from the 1930s.With World War II looming, President Franklin D. Roosevelt ordered FBI Director J. Edgar Hoover to investigate foreign and foreign-inspired subversion-Communist, Nazi, and Japanese. Hoover added investigation of possible espionage, sabotage, or subversion to the duties of field offices. After the war, foreign intelligence duties were assigned to the newly established Central Intelligence Agency. Hoover jealously guarded the FBI's domestic portfolio against all rivals. Hoover felt he was accountable only to the president, and the FBI's domestic intelligence activities kept growing. In the 1960s, the FBI was receiving significant assistance within the United States from the CIA and from Army Intel-ligence. The legal basis for some of this assistance was dubious.

Decades of encouragement to perform as a domestic intelligence agency abruptly ended in the 1970s.Two years after Hoover's death in 1972, congressional and news media investigations of the Watergate scandals of the Nixon administration expanded into general investigations of foreign and domestic intelligence by the Church and Pike committees.¹⁴ They disclosed domestic intelligence efforts, which included a covert action program that operated from 1956 to 1971 against domestic organizations and, eventually, domestic dissidents. The FBI had spied on a wide range of political figures, especially individuals whom Hoover wanted to discredit (notably the Reverend Martin Luther King, Jr.), and had authorized unlawful wiretaps and surveillance. The shock registered in public opinion polls, where the percentage of Americans declaring a "highly favorable" view of the FBI dropped from 84 percent to 37 percent. The FBI's Domestic Intelligence Division was dissolved.¹⁵

In 1976, Attorney General Edward Levi adopted domestic security guidelines to regulate intelligence collection in the United States and to deflect calls for even stronger regulation. In 1983,Attorney General William French Smith revised the Levi guidelines to encourage closer investigation of potential terrorism. He also loosened the rules governing authorization for investigations and their duration. Still, his guidelines, like Levi's, took account of the reality that suspicion of "terrorism," like suspicion of "subversion," could lead to making individuals targets for investigation more because of their beliefs than because of their acts. Smith's guidelines also took account of the reality that potential terrorists were often members of extremist religious organizations and that investigation of terrorism could cross the line separating state and church.¹⁶

In 1986, Congress authorized the FBI to investigate terrorist attacks against Americans that occur outside the United States. Three years later, it added authority for the FBI to make arrests abroad without consent from the host country. Meanwhile, a task force headed by Vice President George H.W. Bush had endorsed a concept already urged by Director of Central Intelligence William Casey-a Counterterrorist Center, where the FBI, the CIA, and other organizations could work together on international terrorism. While it was distinctly a CIA entity, the FBI detailed officials to work at the Center and obtained leads that helped in the capture of persons wanted for trial in the United States.

The strengths that the FBI brought to counterterrorism were nowhere more brilliantly on display than in the case of Pan American Flight 103, bound from London to New York, which blew up over Lockerbie, Scotland, in December 1988, killing 270 people. Initial evidence pointed to the government of Syria and, later, Iran. The Counterterrorist Center reserved judgment on the perpetrators of the attack. Meanwhile, FBI technicians, working with U.K. security services, gathered and analyzed the widely scattered fragments of the airliner. In 1991, with the help of the Counterterrorist Center, they identified one small fragment as part of a timing device-to the technicians, as distinctive as DNA. It was a Libyan device. Together with other evidence, the FBI put together a case pointing conclusively to the Libyan government. Eventually Libya acknowledged its responsibility.¹⁷ Pan Am 103 became a cautionary tale against rushing to judgment in attributing responsibility for a terrorist act. It also showed again how-given a case to solve-the FBI remained capable of extraordinary investigative success.

FBI Organization and Priorities
In 1993, President Clinton chose Louis Freeh as the Director of the Bureau. Freeh, who would remain Director until June 2001, believed that the FBI's work should be done primarily by the field offices. To emphasize this view he cut headquarters staff and decentralized operations. The special agents in charge gained power, influence, and independence.¹⁸

Freeh recognized terrorism as a major threat. He increased the number of legal attaché offices abroad, focusing in particular on the Middle East. He also urged agents not to wait for terrorist acts to occur before taking action. In his first budget request to Congress after the 1993 World Trade Center bombing, he stated that "merely solving this type of crime is not enough; it is equally important that the FBI thwart terrorism before such acts can be perpetrated." Within headquarters, he created a Counterterrorism Division that would complement the Counterterrorist Center at the CIA and arranged for exchanges of senior FBI and CIA counterterrorism officials. He pressed for more cooperation between legal attachés and CIA stations abroad.¹⁹

Freeh's efforts did not, however, translate into a significant shift of resources to counterterrorism. FBI, Justice, and Office of Management and Budget officials said that FBI leadership seemed unwilling to shift resources to terrorism from other areas such as violent crime and drug enforcement; other FBI officials blamed Congress and the OMB for a lack of political will and failure to understand the FBI's counterterrorism resource needs. In addition, Freeh did not impose his views on the field offices. With a few notable exceptions, the field offices did not apply significant resources to terrorism and often reprogrammed funds for other priorities.²⁰

In 1998, the FBI issued a five-year strategic plan led by its deputy director, Robert "Bear" Bryant. For the first time, the FBI designated national and economic security, including counterterrorism, as its top priority. Dale Watson, who would later become the head of the new Counterterrorism Division, said that after the East Africa bombings, "the light came on" that cultural change had to occur within the FBI. The plan mandated a stronger intelligence collection effort. It called for a nationwide automated system to facilitate information collection, analysis, and dissemination. It envisioned the creation of a professional intelligence cadre of experienced and trained agents and analysts. If successfully implemented, this would have been a major step toward addressing terrorism systematically, rather than as individual unrelated cases. But the plan did not succeed.²¹

First, the plan did not obtain the necessary human resources. Despite designating "national and economic security" as its top priority in 1998, the FBI did not shift human resources accordingly. Although the FBI's counterterrorism budget tripled during the mid-1990s, FBI counterterrorism spending remained fairly constant between fiscal years 1998 and 2001. In 2000, there were still twice as many agents devoted to drug enforcement as to counterterrorism.²²

Second, the new division intended to strengthen the FBI's strategic analysis capability faltered. It received insufficient resources and faced resistance from senior managers in the FBI's operational divisions. The new division was supposed to identify trends in terrorist activity, determine what the FBI did not know, and ultimately drive collection efforts. However, the FBI had little appreciation for the role of analysis. Analysts continued to be used primarily in a tactical fashion-providing support for existing cases. Compounding the problem was the FBI's tradition of hiring analysts from within instead of recruiting individuals with the relevant educational background and expertise.²³

Moreover, analysts had difficulty getting access to the FBI and intelligence community information they were expected to analyze. The poor state of the FBI's information systems meant that such access depended in large part on an analyst's personal relationships with individuals in the operational units or squads where the information resided. For all of these reasons, prior to 9/11 relatively few strategic analytic reports about counterterrorism had been completed. Indeed, the FBI had never completed an assessment of the overall terrorist threat to the U.S. homeland.²⁴

Third, the FBI did not have an effective intelligence collection effort. Collection of intelligence from human sources was limited, and agents were inadequately trained. Only three days of a 16-week agents' course were devoted to counterintelligence and counterterrorism, and most subsequent training was received on the job. The FBI did not have an adequate mechanism for validating source reporting, nor did it have a system for adequately tracking and sharing source reporting, either internally or externally. The FBI did not dedicate sufficient resources to the surveillance and translation needs of counterterrorism agents. It lacked sufficient translators proficient in Arabic and other key languages, resulting in a significant backlog of untranslated intercepts.²⁵

Finally, the FBI's information systems were woefully inadequate. The FBI lacked the ability to know what it knew: there was no effective mechanism for capturing or sharing its institutional knowledge. FBI agents did create records of interviews and other investigative efforts, but there were no reports officers to condense the information into meaningful intelligence that could be retrieved and disseminated.²⁶

In 1999, the FBI created separate Counterterrorism and Counterintelligence divisions. Dale Watson, the first head of the new Counterterrorism Division, recognized the urgent need to increase the FBI's counterterrorism capability. His plan, called MAXCAP 05, was unveiled in 2000: it set the goal of bringing the Bureau to its "maximum feasible capacity" in counterterrorism by 2005. Field executives told Watson that they did not have the analysts, linguists, or technically trained experts to carry out the strategy. In a report provided to Director Robert Mueller in September 2001, one year after Watson presented his plan to field executives, almost every FBI field office was assessed to be operating below "maximum capacity." The report stated that "the goal to 'prevent terrorism' requires a dramatic shift in emphasis from a reactive capability to highly functioning intelligence capability which provides not only leads and operational support, but clear strategic analysis and direction."²⁷

Legal Constraints on the FBI and "the Wall"
The FBI had different tools for law enforcement and intelligence.²⁸ For criminal matters, it could apply for and use traditional criminal warrants. For intelligence matters involving international terrorism, however, the rules were different. For many years the attorney general could authorize surveillance of foreign powers and agents of foreign powers without any court review, but in 1978 Congress passed the Foreign Intelligence Surveillance Act.²⁹This law regulated intelligence collection directed at foreign powers and agents of foreign powers in the United States. In addition to requiring court review of proposed surveillance (and later, physical searches), the 1978 act was interpreted by the courts to require that a search be approved only if its "primary purpose" was to obtain foreign intelligence information. In other words, the authorities of the FISA law could not be used to circumvent traditional criminal warrant requirements. The Justice Department interpreted these rulings as saying that criminal prosecutors could be briefed on FISA information but could not direct or control its collection.³⁰

Throughout the 1980s and early 1990s, Justice prosecutors had informal arrangements for obtaining information gathered in the FISA process, the understanding being that they would not improperly exploit that process for their criminal cases. Whether the FBI shared with prosecutors information pertinent to possible criminal investigations was left solely to the judgment of the FBI.³¹

But the prosecution of Aldrich Ames for espionage in 1994 revived concerns about the prosecutors' role in intelligence investigations. The Department of Justice's Office of Intelligence Policy and Review (OIPR) is responsible for reviewing and presenting all FISA applications to the FISA Court. It worried that because of the numerous prior consultations between FBI agents and prosecutors, the judge might rule that the FISA warrants had been misused. If that had happened, Ames might have escaped conviction. Richard Scruggs, the acting head of OIPR, complained to Attorney General Janet Reno about the lack of information-sharing controls. On his own, he began imposing information-sharing procedures for FISA material. The Office of Intelligence Policy and Review became the gatekeeper for the flow of FISA information to criminal prosecutors.³²

In July 1995, Attorney General Reno issued formal procedures aimed at managing information sharing between Justice Department prosecutors and the FBI. They were developed in a working group led by the Justice Department's Executive Office of National Security, overseen by Deputy Attorney General Jamie Gorelick.³³ These procedures-while requiring the sharing of intelligence information with prosecutors-regulated the manner in which such information could be shared from the intelligence side of the house to the criminal side.

These procedures were almost immediately misunderstood and misapplied. As a result, there was far less information sharing and coordination between the FBI and the Criminal Division in practice than was allowed under the department's procedures. Over time the procedures came to be referred to as "the wall." The term "the wall" is misleading, however, because several factors led to a series of barriers to information sharing that developed.³⁴

The Office of Intelligence Policy and Review became the sole gatekeeper for passing information to the Criminal Division. Though Attorney General Reno's procedures did not include such a provision, the Office assumed the role anyway, arguing that its position reflected the concerns of Judge Royce Lamberth, then chief judge of the Foreign Intelligence Surveillance Court. The Office threatened that if it could not regulate the flow of information to criminal prosecutors, it would no longer present the FBI's warrant requests to the FISA Court. The information flow withered.³⁵

The 1995 procedures dealt only with sharing between agents and criminal prosecutors, not between two kinds of FBI agents, those working on intelligence matters and those working on criminal matters. But pressure from the Office of Intelligence Policy Review, FBI leadership, and the FISA Court built barriers between agents-even agents serving on the same squads. FBI Deputy Director Bryant reinforced the Office's caution by informing agents that too much information sharing could be a career stopper. Agents in the field began to believe-incorrectly-that no FISA information could be shared with agents working on criminal investigations.³⁶

This perception evolved into the still more exaggerated belief that the FBI could not share any intelligence information with criminal investigators, even if no FISA procedures had been used. Thus, relevant information from the National Security Agency and the CIA often failed to make its way to criminal investigators. Separate reviews in 1999, 2000, and 2001 concluded independently that information sharing was not occurring, and that the intent of the 1995 procedures was ignored routinely.³⁷ We will describe some of the unfortunate consequences of these accumulated institutional beliefs and practices in chapter 8.

There were other legal limitations. Both prosecutors and FBI agents argued that they were barred by court rules from sharing grand jury information, even though the prohibition applied only to that small fraction that had been presented to a grand jury, and even that prohibition had exceptions. But as interpreted by FBI field offices, this prohibition could conceivably apply to much of the information unearthed in an investigation. There were also restrictions, arising from executive order, on the commingling of domestic information with foreign intelligence. Finally the NSA began putting caveats on its Bin Ladin-related reports that required prior approval before sharing their contents with criminal investigators and prosecutors. These developments further blocked the arteries of information sharing.³⁸

Other Law Enforcement Agencies
The Justice Department is much more than the FBI. It also has a U.S. Marshals Service, almost 4,000 strong on 9/11 and especially expert in tracking fugitives, with much local police knowledge. The department's Drug Enforcement Administration had, as of 2001, more than 4,500 agents.³⁹ There were a number of occasions when DEA agents were able to introduce sources to the FBI or CIA for counterterrorism use.

The Immigration and Naturalization Service (INS), with its 9,000 Border Patrol agents, 4,500 inspectors, and 2,000 immigration special agents, had perhaps the greatest potential to develop an expanded role in counterterrorism. However, the INS was focused on the formidable challenges posed by illegal entry over the southwest border, criminal aliens, and a growing backlog in the applications for naturalizing immigrants. The White House, the Justice Department, and above all the Congress reinforced these concerns. In addition, when Doris Meissner became INS Commissioner in 1993, she found an agency seriously hampered by outdated technology and insufficient human resources. Border Patrol agents were still using manual typewriters; inspectors at ports of entry were using a paper watchlist; the asylum and other benefits systems did not effectively deter fraudulent applicants.⁴⁰

Commissioner Meissner responded in 1993 to the World Trade Center bombing by providing seed money to the State Department's Consular Affairs Bureau to automate its terrorist watchlist, used by consular officers and border inspectors. The INS assigned an individual in a new "lookout" unit to work with the State Department in watchlisting suspected terrorists and with the intelligence community and the FBI in determining how to deal with them when they appeared at ports of entry. By 1998, 97 suspected terrorists had been denied admission at U.S. ports of entry because of the watchlist.⁴¹

How to conduct deportation cases against aliens who were suspected terrorists caused significant debate. The INS had immigration law expertise and authority to bring the cases, but the FBI possessed the classified information sometimes needed as evidence, and information-sharing conflicts resulted. New laws in 1996 authorized the use of classified evidence in removal hearings, but the INS removed only a handful of the aliens with links to terrorist activity (none identified as associated with al Qaeda) using classified evidence.⁴²

Midlevel INS employees proposed comprehensive counterterrorism proposals to management in 1986, 1995, and 1997. No action was taken on them. In 1997, a National Security Unit was set up to handle alerts, track potential terrorist cases for possible immigration enforcement action, and work with the rest of the Justice Department. It focused on the FBI's priorities of Hezbollah and Hamas, and began to examine how immigration laws could be brought to bear on terrorism. For instance, it sought unsuccessfully to require that CIA security checks be completed before naturalization applications were approved.⁴³ Policy questions, such as whether resident alien status should be revoked upon the person's conviction of a terrorist crime, were not addressed.

Congress, with the support of the Clinton administration, doubled the number of Border Patrol agents required along the border with Mexico to one agent every quarter mile by 1999. It rejected efforts to bring additional resources to bear in the north. The border with Canada had one agent for every 13.25 miles. Despite examples of terrorists entering from Canada, awareness of terrorist activity in Canada and its more lenient immigration laws, and an inspector general's report recommending that the Border Patrol develop a northern border strategy, the only positive step was that the number of Border Patrol agents was not cut any further.⁴⁴

Inspectors at the ports of entry were not asked to focus on terrorists. Inspectors told us they were not even aware that when they checked the names of incoming passengers against the automated watchlist, they were checking in part for terrorists. In general, border inspectors also did not have the information they needed to make fact-based determinations of admissibility. The INS initiated but failed to bring to completion two efforts that would have provided inspectors with information relevant to counterterrorism-a proposed system to track foreign student visa compliance and a program to establish a way of tracking travelers' entry to and exit from the United States.⁴⁵

In 1996, a new law enabled the INS to enter into agreements with state and local law enforcement agencies through which the INS provided training and the local agencies exercised immigration enforcement authority. Terrorist watchlists were not available to them. Mayors in cities with large immigrant populations sometimes imposed limits on city employee cooperation with federal immigration agents.A large population lives outside the legal framework. Fraudulent documents could be easily obtained. Congress kept the number of INS agents static in the face of the overwhelming problem.⁴⁶

The chief vehicle for INS and for state and local participation in law enforcement was the Joint Terrorism Task Force (JTTF), first tried out in New York City in 1980 in response to a spate of incidents involving domestic terrorist organizations. This task force was managed by the New York Field Office of the FBI, and its existence provided an opportunity to exchange information and, as happened after the first World Trade Center bombing, to enlist local officers, as well as other agency representatives, as partners in the FBI investigation. The FBI expanded the number of JTTFs throughout the 1990s, and by 9/11 there were 34.While useful, the JTTFs had limitations. They set priorities in accordance with regional and field office concerns, and most were not fully staffed. Many state and local entities believed they had little to gain from having a full-time representative on a JTTF.⁴⁷

Other federal law enforcement resources, also not seriously enlisted for counterterrorism, were to be found in the Treasury Department.

Treasury housed the Secret Service, the Customs Service, and the Bureau of Alcohol, Tobacco, and Firearms. Given the Secret Service's mission to protect the president and other high officials, its agents did become involved with those of the FBI whenever terrorist assassination plots were rumored.

The Customs Service deployed agents at all points of entry into the United States. Its agents worked alongside INS agents, and the two groups sometimes cooperated. In the winter of 1999-2000, as will be detailed in chapter 6, questioning by an especially alert Customs inspector led to the arrest of an al Qaeda terrorist whose apparent mission was to bomb Los Angeles International Airport.

The Bureau of Alcohol, Tobacco, and Firearms was used on occasion by the FBI as a resource. The ATF's laboratories and analysis were critical to the investigation of the February 1993 bombing of the World Trade Center and the April 1995 bombing of the Alfred P. Murrah Federal Building in Oklahoma City.⁴⁸

Before 9/11, with the exception of one portion of the FBI, very little of the sprawling U.S. law enforcement community was engaged in countering terrorism. Moreover, law enforcement could be effective only after specific individuals were identified, a plot had formed, or an attack had already occurred. Responsible individuals had to be located, apprehended, and transported back to a U.S. court for prosecution. As FBI agents emphasized to us, the FBI and the Justice Department do not have cruise missiles. They declare war by indicting someone. They took on the lead role in addressing terrorism because they were asked to do so.⁴⁹

3.3 . . .AND IN THE FEDERAL AVIATION ADMINISTRATION

The Federal Aviation Administration (FAA) within the Department of Transportation had been vested by Congress with the sometimes conflicting mandate of regulating the safety and security of U.S. civil aviation while also promoting the civil aviation industry. The FAA had a security mission to protect the users of commercial air transportation against terrorism and other criminal acts. In the years before 9/11, the FAA perceived sabotage as a greater threat to aviation than hijacking. First, no domestic hijacking had occurred in a decade. Second, the commercial aviation system was perceived as more vulnerable to explosives than to weapons such as firearms. Finally, explosives were perceived as deadlier than hijacking and therefore of greater consequence. In 1996, a presidential commission on aviation safety and security chaired by Vice President Al Gore reinforced the prevailing concern about sabotage and explosives on aircraft. The Gore Commission also flagged, as a new danger, the possibility of attack by surface-to-air missiles. Its 1997 final report did not discuss the possibility of suicide hijackings.⁵⁰

The FAA set and enforced aviation security rules, which airlines and airports were required to implement. The rules were supposed to produce a "layered" system of defense. This meant that the failure of any one layer of security would not be fatal, because additional layers would provide backup security. But each layer relevant to hijackings-intelligence, passenger prescreening, checkpoint screening, and onboard security-was seriously flawed prior to 9/11.Taken together, they did not stop any of the 9/11 hijackers from getting on board four different aircraft at three different airports.⁵¹

The FAA's policy was to use intelligence to identify both specific plots and general threats to civil aviation security, so that the agency could develop and deploy appropriate countermeasures. The FAA's 40-person intelligence unit was supposed to receive a broad range of intelligence data from the FBI, CIA, and other agencies so that it could make assessments about the threat to aviation. But the large volume of data contained little pertaining to the presence and activities of terrorists in the United States. For example, information on the FBI's effort in 1998 to assess the potential use of flight training by terrorists and the Phoenix electronic communication of 2001 warning of radical Middle Easterners attending flight school were not passed to FAA headquarters. Several top FAA intelligence officials called the domestic threat picture a serious blind spot.⁵²

Moreover, the FAA's intelligence unit did not receive much attention from the agency's leadership. Neither Administrator Jane Garvey nor her deputy routinely reviewed daily intelligence, and what they did see was screened for them. She was unaware of a great amount of hijacking threat information from her own intelligence unit, which, in turn, was not deeply involved in the agency's policymaking process. Historically, decisive security action took place only after a disaster had occurred or a specific plot had been discovered.⁵³

The next aviation security layer was passenger prescreening. The FAA directed air carriers not to fly individuals known to pose a "direct" threat to civil aviation. But as of 9/11, the FAA's "no-fly" list contained the names of just 12 terrorist suspects (including 9/11 mastermind Khalid Sheikh Mohammed), even though government watchlists contained the names of many thousands of known and suspected terrorists. This astonishing mismatch existed despite the Gore Commission's having called on the FBI and CIA four years earlier to provide terrorist watchlists to improve prescreening. The longtime chief of the FAA's civil aviation security division testified that he was not even aware of the State Department's TIPOFF list of known and suspected terrorists (some 60,000 before 9/11) until he heard it mentioned during the Commission's January 26, 2004, public hearing. The FAA had access to some TIPOFF data, but apparently found it too difficult to use.⁵⁴

The second part of prescreening called on the air carriers to implement an FAA-approved computerized algorithm (known as CAPPS, for Computer Assisted Passenger Prescreening System) designed to identify passengers whose profile suggested they might pose more than a minimal risk to aircraft. Although the algorithm included hijacker profile data, at that time only passengers checking bags were eligible to be selected by CAPPS for additional scrutiny. Selection entailed only having one's checked baggage screened for explosives or held off the airplane until one had boarded. Primarily because of concern regarding potential discrimination and the impact on passenger throughput, "selectees" were no longer required to undergo extraordinary screening of their carry-on baggage as had been the case before the system was computerized in 1997.⁵⁵ This policy change also reflected the perception that nonsuicide sabotage was the primary threat to civil aviation.

Checkpoint screening was considered the most important and obvious layer of security. Walk-through metal detectors and X-ray machines operated by trained screeners were employed to stop prohibited items. Numerous government reports indicated that checkpoints performed poorly, often failing to detect even obvious FAA test items. Many deadly and dangerous items did not set off metal detectors, or were hard to distinguish in an X-ray machine from innocent everyday items.⁵⁶

While FAA rules did not expressly prohibit knives with blades under 4 inches long, the airlines' checkpoint operations guide (which was developed in cooperation with the FAA), explicitly permitted them. The FAA's basis for this policy was (1) the agency did not consider such items to be menacing, (2) most local laws did not prohibit individuals from carrying such knives, and (3) such knives would have been difficult to detect unless the sensitivity of metal detectors had been greatly increased. A proposal to ban knives altogether in 1993 had been rejected because small cutting implements were difficult to detect and the number of innocent "alarms" would have increased significantly, exacerbating congestion problems at checkpoints.⁵⁷

Several years prior to 9/11, an FAA requirement for screeners to conduct "continuous" and "random" hand searches of carry-on luggage at checkpoints had been replaced by explosive trace detection or had simply become ignored by the air carriers. Therefore, secondary screening of individuals and their carry-on bags to identify weapons (other than bombs) was nonexistent, except for passengers who triggered the metal detectors. Even when small knives were detected by secondary screening, they were usually returned to the traveler. Reportedly, the 9/11 hijackers were instructed to use items that would be undetectable by airport checkpoints.⁵⁸

In the pre-9/11 security system, the air carriers played a major role. As the Inspector General of the Department of Transportation told us, there were great pressures from the air carriers to control security costs and to "limit the impact of security requirements on aviation operations, so that the industry could concentrate on its primary mission of moving passengers and aircraft.. . . [T]hose counterpressures in turn manifested themselves as significant weaknesses in security." A longtime FAA security official described the air carriers' approach to security regulation as "decry, deny and delay" and told us that while "the air carriers had seen the enlightened hand of self-interest with respect to safety, they hadn't seen it in the security arena."⁵⁹

The final layer, security on board commercial aircraft, was not designed to counter suicide hijackings. The FAA-approved "Common Strategy" had been elaborated over decades of experience with scores of hijackings, beginning in the 1960s. It taught flight crews that the best way to deal with hijackers was to accommodate their demands, get the plane to land safely, and then let law enforcement or the military handle the situation. According to the FAA, the record had shown that the longer a hijacking persisted, the more likely it was to end peacefully. The strategy operated on the fundamental assumption that hijackers issue negotiable demands (most often for asylum or the release of prisoners) and that, as one FAA official put it, "suicide wasn't in the game plan" of hijackers. FAA training material provided no guidance for flight crews should violence occur.⁶⁰

This prevailing Common Strategy of cooperation and nonconfrontation meant that even a hardened cockpit door would have made little difference in a hijacking. As the chairman of the Security Committee of the Air Line Pilots Association observed when proposals were made in early 2001 to install reinforced cockpit doors in commercial aircraft, "Even if you make a vault out of the door, if they have a noose around my flight attendant's neck, I'm going to open the door." Prior to 9/11, FAA regulations mandated that cockpit doors permit ready access into and out of the cockpit in the event of an emergency. Even so, rules implemented in the 1960s required air crews to keep the cockpit door closed and locked in flight. This requirement was not always observed or vigorously enforced.⁶¹

As for law enforcement, there were only 33 armed and trained federal air marshals as of 9/11.They were not deployed on U.S. domestic flights, except when in transit to provide security on international departures. This policy reflected the FAA's view that domestic hijacking was in check-a view held confidently as no terrorist had hijacked a U.S. commercial aircraft anywhere in the world since 1986.⁶²

In the absence of any recent aviation security incident and without "specific and credible" evidence of a plot directed at civil aviation, the FAA's leadership focused elsewhere, including on operational concerns and the ever-present issue of safety. FAA Administrator Garvey recalled that "every day in 2001 was like the day before Thanksgiving." Heeding calls for improved air service, Congress concentrated its efforts on a "passenger bill of rights," to improve capacity, efficiency, and customer satisfaction in the aviation system. There was no focus on terrorism.⁶³

3.4 .. . AND IN THE INTELLIGENCE COMMUNITY

The National Security Act of 1947 created the position of Director of Central Intelligence (DCI). Independent from the departments of Defense, State, Justice, and other policy departments, the DCI heads the U.S. intelligence community and provides intelligence to federal entities.

The sole element of the intelligence community independent from a cabinet agency is the CIA. As an independent agency, it collects, analyzes, and disseminates intelligence from all sources. The CIA's number one customer is the president of the United States, who also has the authority to direct it to conduct covert operations.⁶⁴ Although covert actions represent a very small fraction of the Agency's entire budget, these operations have at times been controversial and over time have dominated the public's perception of the CIA.

The DCI is confirmed by the Senate but is not technically a member of the president's cabinet. The director's power under federal law over the loose, confederated "intelligence community" is limited.⁶⁵ He or she states the community's priorities and coordinates development of intelligence agency budget requests for submission to Congress.

This responsibility gives many the false impression that the DCI has line authority over the heads of these agencies and has the power to shift resources within these budgets as the need arises. Neither is true. In fact, the DCI's real authority has been directly proportional to his personal closeness to the president, which has waxed and waned over the years, and to others in government, especially the secretary of defense.

Intelligence agencies under the Department of Defense account for approximately 80 percent of all U.S. spending for intelligence, including some that supports a national customer base and some that supports specific Defense Department or military service needs.⁶⁶ As they are housed in the Defense Department, these agencies are keenly attentive to the military's strategic and tactical requirements.

One of the intelligence agencies in Defense with a national customer base is the National Security Agency, which intercepts and analyzes foreign communications and breaks codes. The NSA also creates codes and ciphers to protect government information. Another is the recently renamed National Geospatial-Intelligence Agency (NGA), which provides and analyzes imagery and produces a wide array of products, including maps, navigation tools, and surveillance intelligence. A third such agency in Defense is the National Reconnaissance Office. It develops, procures, launches, and maintains in orbit information-gathering satellites that serve other government agencies.

The Defense Intelligence Agency supports the secretary of defense, Joint Chiefs of Staff, and military field commanders. It does some collection through human sources as well as some technical intelligence collection. The Army, Navy, Air Force, and Marine Corps have their own intelligence components that collect information, help them decide what weapons to acquire, and serve the tactical intelligence needs of their respective services.

In addition to those from the Department of Defense, other elements in the intelligence community include the national security parts of the FBI; the Bureau of Intelligence and Research in the State Department; the intelligence component of the Treasury Department; the Energy Department's Office of Intelligence and Counterintelligence, the former of which, through leveraging the expertise of the national laboratory system, has special competence in nuclear weapons; the Office of Intelligence of the Coast Guard; and, today, the Directorate of Intelligence Analysis and Infrastructure Protection in the Department of Homeland Security.

The National Security Agency
The National Security Agency's intercepts of terrorist communications often set off alarms elsewhere in the government. Often, too, its intercepts are conclusive elements in the analyst's jigsaw puzzle. NSA engineers build technical systems to break ciphers and to make sense of today's complex signals environment. Its analysts listen to conversations between foreigners not meant for them. They also perform "traffic analysis"-studying technical communications systems and codes as well as foreign organizational structures, including those of terrorist organizations.

Cold War adversaries used very hierarchical, familiar, and predictable military command and control methods. With globalization and the telecommunications revolution, and with loosely affiliated but networked adversaries using commercial devices and encryption, the technical impediments to signals collection grew at a geometric rate. At the same time, the end of the Cold War and the resultant cuts in national security funding forced intelligence agencies to cut systems and seek economies of scale. Modern adversaries are skilled users of communications technologies. The NSA's challenges, and its opportunities, increased exponentially in "volume, variety, and velocity."⁶⁷

The law requires the NSA to not deliberately collect data on U.S. citizens or on persons in the United States without a warrant based on foreign intelligence requirements. Also, the NSA was supposed to let the FBI know of any indication of crime, espionage, or "terrorist enterprise" so that the FBI could obtain the appropriate warrant. Later in this story, we will learn that while the NSA had the technical capability to report on communications with suspected terrorist facilities in the Middle East, the NSA did not seek FISA Court warrants to collect communications between individuals in the United States and foreign countries, because it believed that this was an FBI role. It also did not want to be viewed as targeting persons in the United States and possibly violating laws that governed NSA's collection of foreign intelligence.⁶⁸

An almost obsessive protection of sources and methods by the NSA, and its focus on foreign intelligence, and its avoidance of anything domestic would, as will be seen, be important elements in the story of 9/11.

Technology as an Intelligence Asset and Liability
The application of newly developed scientific technology to the mission of U.S. war fighters and national security decisionmakers is one of the great success stories of the twentieth century. It did not happen by accident. Recent wars have been waged and won decisively by brave men and women using advanced technology that was developed, authorized, and paid for by conscientious and diligent executive and legislative branch leaders many years earlier.

The challenge of technology, however, is a daunting one. It is expensive, sometimes fails, and often can create problems as well as solve them. Some of the advanced technologies that gave us insight into the closed-off territories of the Soviet Union during the Cold War are of limited use in identifying and tracking individual terrorists.

Terrorists, in turn, have benefited from this same rapid development of communication technologies. They simply could buy off the shelf and harvest the products of a $3 trillion a year telecommunications industry. They could acquire without great expense communication devices that were varied, global, instantaneous, complex, and encrypted.

The emergence of the World Wide Web has given terrorists a much easier means of acquiring information and exercising command and control over their operations. The operational leader of the 9/11 conspiracy, Mohamed Atta, went online from Hamburg, Germany, to research U.S. flight schools. Targets of intelligence collection have become more sophisticated. These changes have made surveillance and threat warning more difficult.

Despite the problems that technology creates, Americans' love affair with it leads them to also regard it as the solution. But technology produces its best results when an organization has the doctrine, structure, and incentives to exploit it. For example, even the best information technology will not improve information sharing so long as the intelligence agencies' personnel and security systems reward protecting information rather than disseminating it.

The CIA
The CIA is a descendant of the Office of Strategic Services (OSS), which President Roosevelt created early in World War II after having first thought the FBI might take that role. The father of the OSS was William J. "Wild Bill" Donovan, a Wall Street lawyer. He recruited into the OSS others like himself-well traveled, well connected, well-to-do professional men and women.⁶⁹

An innovation of Donovan's, whose legacy remains part of U.S. intelligence today, was the establishment of a Research and Analysis Branch. There large numbers of scholars from U.S. universities pored over accounts from spies, communications intercepted by the armed forces, transcripts of radio broadcasts, and publications of all types, and prepared reports on economic, political, and social conditions in foreign theaters of operation.

At the end of World War II, to Donovan's disappointment, President Harry Truman dissolved the Office of Strategic Services. Four months later, the President directed that "all Federal foreign intelligence activities be planned, developed and coordinated so as to assure the most effective accomplishment of the intelligence mission related to the national security," under a National Intelligence Authority consisting of the secretaries of State, War, and the Navy, and a personal representative of the president. This body was to be assisted by a Central Intelligence Group, made up of persons detailed from the departments of each of the members and headed by a Director of Central Intelligence.⁷⁰

Subsequently, President Truman agreed to the National Security Act of 1947, which, among other things, established the Central Intelligence Agency, under the Director of Central Intelligence. Lobbying by the FBI, combined with fears of creating a U.S. Gestapo,⁷¹ led to the FBI's being assigned responsibility for internal security functions and counterespionage. The CIA was specifically accorded "no police, subpoena, or law enforcement powers or internal security functions."⁷² This structure built in tensions between the CIA and the Defense Department's intelligence agencies, and between the CIA and the FBI.

Clandestine and Covert Action. With this history, the CIA brought to the era of 9/11 many attributes of an elite organization, viewing itself as serving on the nation's front lines to engage America's enemies. Officers in its Clandestine Service, under what became the Directorate of Operations, fanned out into stations abroad. Each chief of station was a very important person in the organization, given the additional title of the DCI's representative in that country. He (occasionally she) was governed by an operating directive that listed operational priorities issued by the relevant regional division of the Directorate, constrained by centrally determined allocations of resources.

Because the conduct of espionage was a high-risk activity, decisions on the clandestine targeting, recruitment, handling, and termination of secret sources and the dissemination of collected information required Washington's approval and action. But in this decentralized system, analogous in some ways to the culture of the FBI field offices in the United States, everyone in the Directorate of Operations presumed that it was the job of headquarters to support the field, rather than manage field activities.

In the 1960s, the CIA suffered exposure of its botched effort to land Cuban exiles at the Bay of Pigs. The Vietnam War brought on more criticism. A prominent feature of the Watergate era was investigations of the CIA by committees headed by Frank Church in the Senate and Otis Pike in the House. They published evidence that the CIA had secretly planned to assassinate Fidel Castro and other foreign leaders. The President had not taken plain responsibility for these judgments. CIA officials had taken most of the blame, saying they had done so in order to preserve the President's "plausible deniability."⁷³

After the Watergate era, Congress established oversight committees to ensure that the CIA did not undertake covert action contrary to basic American law. Case officers in the CIA's Clandestine Service interpreted legislation, such as the Hughes-Ryan Amendment requiring that the president approve and report to Congress any covert action, as sending a message to them that covert action often leads to trouble and can severely damage one's career. Controversies surrounding Central American covert action programs in the mid-1980s led to the indictment of several senior officers of the Clandestine Service. During the 1990s, tension sometimes arose, as it did in the effort against al Qaeda, between policymakers who wanted the CIA to undertake more aggressive covert action and wary CIA leaders who counseled prudence and making sure that the legal basis and presidential authorization for their actions were undeniably clear.

The Clandestine Service felt the impact of the post-Cold War peace dividend, with cuts beginning in 1992. As the number of officers declined and overseas facilities were closed, the DCI and his managers responded to developing crises in the Balkans or in Africa by "surging," or taking officers from across the service to use on the immediate problem. In many cases the surge officers had little familiarity with the new issues. Inevitably, some parts of the world and some collection targets were not fully covered, or not covered at all. This strategy also placed great emphasis on close relations with foreign liaison services, whose help was needed to gain information that the United States itself did not have the capacity to collect.

The nadir for the Clandestine Service was in 1995, when only 25 trainees became new officers.⁷⁴ In 1998, the DCI was able to persuade the administration and the Congress to endorse a long-range rebuilding program. It takes five to seven years of training, language study, and experience to bring a recruit up to full performance.⁷⁵

Analysis. The CIA's Directorate of Intelligence retained some of its original character of a university gone to war. Its men and women tended to judge one another by the quantity and quality of their publications (in this case, classified publications). Apart from their own peers, they looked for approval and guidance to policymakers. During the 1990s and today, particular value is attached to having a contribution included in one of the classified daily "newspapers"- the Senior Executive Intelligence Brief-or, better still, selected for inclusion in the President's Daily Brief.⁷⁶

The CIA had been created to wage the Cold War. Its steady focus on one or two primary adversaries, decade after decade, had at least one positive effect: it created an environment in which managers and analysts could safely invest time and resources in basic research, detailed and reflective. Payoffs might not be immediate. But when they wrote their estimates, even in brief papers, they could draw on a deep base of knowledge.

When the Cold War ended, those investments could not easily be reallocated to new enemies. The cultural effects ran even deeper. In a more fluid international environment with uncertain, changing goals and interests, intelligence managers no longer felt they could afford such a patient, strategic approach to long-term accumulation of intellectual capital. A university culture with its versions of books and articles was giving way to the culture of the newsroom.

During the 1990s, the rise of round-the-clock news shows and the Internet reinforced pressure on analysts to pass along fresh reports to policymakers at an ever-faster pace, trying to add context or supplement what their customers were receiving from the media. Weaknesses in all-source and strategic analysis were highlighted by a panel, chaired by Admiral David Jeremiah, that critiqued the intelligence community's failure to foresee the nuclear weapons tests by India and Pakistan in 1998, as well as by a 1999 panel, chaired by Donald Rumsfeld, that discussed the community's limited ability to assess the ballistic missile threat to the United States. Both reports called attention to the dispersal of effort on too many priorities, the declining attention to the craft of strategic analysis, and security rules that prevented adequate sharing of information. Another Cold War craft had been an elaborate set of methods for warning against surprise attack, but that too had faded in analyzing new dangers like terrorism.⁷⁷

Security. Another set of experiences that would affect the capacity of the CIA to cope with the new terrorism traced back to the early Cold War, when the Agency developed a concern, bordering on paranoia, about penetration by the Soviet KGB. James Jesus Angleton, who headed counterintelligence in the CIA until the early 1970s, became obsessed with the belief that the Agency harbored one or more Soviet "moles." Although the pendulum swung back after Angle-ton's forced retirement, it did not go very far. Instances of actual Soviet penetration kept apprehensions high.⁷⁸ Then, in the early 1990s, came the Aldrich Ames espionage case, which intensely embarrassed the CIA. Though obviously unreliable, Ames had been protected and promoted by fellow officers while he paid his bills by selling to the Soviet Union the names of U.S. operatives and agents, a number of whom died as a result.

The concern about security vastly complicated information sharing. Information was compartmented in order to protect it against exposure to skilled and technologically sophisticated adversaries. There were therefore numerous restrictions on handling information and a deep suspicion about sending information over newfangled electronic systems, like email, to other agencies of the U.S. government.⁷⁹

Security concerns also increased the difficulty of recruiting officers qualified for counterterrorism. Very few American colleges or universities offered programs in Middle Eastern languages or Islamic studies. The total number of undergraduate degrees granted in Arabic in all U.S. colleges and universities in 2002 was six.⁸⁰ Many who had traveled much outside the United States could expect a very long wait for initial clearance. Anyone who was foreign-born or had numerous relatives abroad was well-advised not even to apply. With budgets for the CIA shrinking after the end of the Cold War, it was not surprising that, with some notable exceptions, new hires in the Clandestine Service tended to have qualifications similar to those of serving officers: that is, they were suited for traditional agent recruitment or for exploiting liaison relationships with foreign services but were not equipped to seek or use assets inside the terrorist network.

Early Counterterrorism Efforts
In the 1970s and 1980s, terrorism had been tied to regional conflicts, mainly in the Middle East. The majority of terrorist groups either were sponsored by governments or, like the Palestine Liberation Organization, were militants trying to create governments.

In the mid-1980s, on the basis of a report from a task force headed by Vice President George Bush and after terrorist attacks at airports in Rome and Athens, the DCI created a Counterterrorist Center to unify activities across the Directorate of Operations and the Directorate of Intelligence. The Counterterrorist Center had representation from the FBI and other agencies. In the formal table of organization it reported to the DCI, but in fact most of the Center's chiefs belonged to the Clandestine Service and usually looked for guidance to the head of the Directorate of Operations.⁸¹

The Center stimulated and coordinated collection of information by CIA stations, compiled the results, and passed selected reports to appropriate stations, the Directorate of Intelligence analysts, other parts of the intelligence community, or to policymakers. The Center protected its bureaucratic turf. The Director of Central Intelligence had once had a national intelligence officer for terrorism to coordinate analysis; that office was abolished in the late 1980s and its duties absorbed in part by the Counterterrorist Center. Though analysts assigned to the Center produced a large number of papers, the focus was support to operations. A CIA inspector general's report in 1994 criticized the Center's capacity to provide warning of terrorist attacks.⁸²

Subsequent chapters will raise the issue of whether, despite tremendous talent, energy, and dedication, the intelligence community failed to do enough in coping with the challenge from Bin Ladin and al Qaeda. Confronted with such questions, managers in the intelligence community often responded that they had meager resources with which to work.⁸³

Cuts in national security expenditures at the end of the Cold War led to budget cuts in the national foreign intelligence program from fiscal years 1990 to 1996 and essentially flat budgets from fiscal years 1996 to 2000 (except for the so-called Gingrich supplemental to the FY1999 budget and two later, smaller supplementals).These cuts compounded the difficulties of the intelligence agencies. Policymakers were asking them to move into the digitized future to fight against computer-to-computer communications and modern communication systems, while maintaining capability against older systems, such as high-frequency radios and ultra-high- and very-high-frequency (line of sight) systems that work like old-style television antennas. Also, demand for imagery increased dramatically following the success of the 1991 Gulf War. Both these developments, in turn, placed a premium on planning the next generation of satellite systems, the cost of which put great pressure on the rest of the intelligence budget. As a result, intelligence agencies experienced staff reductions, affecting both operators and analysts.⁸⁴

Yet at least for the CIA, part of the burden in tackling terrorism arose from the background we have described: an organization capable of attracting extraordinarily motivated people but institutionally averse to risk, with its capacity for covert action atrophied, predisposed to restrict the distribution of information, having difficulty assimilating new types of personnel, and accustomed to presenting descriptive reportage of the latest intelligence. The CIA, to put it another way, needed significant change in order to get maximum effect in counterterrorism. President Clinton appointed George Tenet as DCI in 1997, and by all accounts terrorism was a priority for him. But Tenet's own assessment, when questioned by the Commission, was that in 2004, the CIA's clandestine service was still at least five years away from being fully ready to play its counterterrorism role.⁸⁵ And while Tenet was clearly the leader of the CIA, the intelligence community's confederated structure left open the question of who really was in charge of the entire U.S. intelligence effort.

3.5 . . .AND IN THE STATE DEPARTMENT AND THE DEFENSE DEPARTMENT

The State Department
The Commission asked Deputy Secretary of State Richard Armitage in 2004 why the State Department had so long pursued what seemed, and ultimately proved, to be a hopeless effort to persuade the Taliban regime in Afghanistan to deport Bin Ladin. Armitage replied: "We do what the State Department does, we don't go out and fly bombers, we don't do things like that[;] . . . we do our part in these things."⁸⁶

Fifty years earlier, the person in Armitage's position would not have spoken of the Department of State as having such a limited role. Until the late 1950s, the department dominated the processes of advising the president and Congress on U.S. relations with the rest of the world. The National Security Council was created in 1947 largely as a result of lobbying from the Pentagon for a forum where the military could object if they thought the State Department was setting national objectives that the United States did not have the wherewithal to pursue.

The State Department retained primacy until the 1960s, when the Kennedy and Johnson administrations turned instead to Robert McNamara's Defense Department, where a mini-state department was created to analyze foreign policy issues. President Richard Nixon then concentrated policy planning and policy coordination in a powerful National Security Council staff, overseen by Henry Kissinger.

In later years, individual secretaries of state were important figures, but the department's role continued to erode. State came into the 1990s overmatched by the resources of other departments and with little support for its budget either in the Congress or in the president's Office of Management and Budget.

Like the FBI and the CIA's Directorate of Operations, the State Department had a tradition of emphasizing service in the field over service in Washington. Even ambassadors, however, often found host governments not only making connections with the U.S. government through their own missions in Washington, but working through the CIA station or a Defense attaché. Increasingly, the embassies themselves were overshadowed by powerful regional commanders in chief reporting to the Pentagon.⁸⁷

Counterterrorism
In the 1960s and 1970s, the State Department managed counterterrorism policy. It was the official channel for communication with the governments presumed to be behind the terrorists. Moreover, since terrorist incidents of this period usually ended in negotiations, an ambassador or other embassy official was the logical person to represent U.S. interests.

Keeping U.S. diplomatic efforts against terrorism coherent was a recurring challenge. In 1976, at the direction of Congress, the department elevated its coordinator for combating terrorism to the rank equivalent to an assistant secretary of state. As an "ambassador at large," this official sought to increase the visibility of counterterrorism matters within the department and to help integrate U.S. policy implementation among government agencies. The prolonged crisis of 1979-1981, when 53 Americans were held hostage at the U.S. embassy in Tehran, ended the State Department leadership in counterterrorism. President Carter's assertive national security advisor, Zbigniew Brzezinski, took charge, and the coordination function remained thereafter in the White House.

President Reagan's second secretary of state, George Shultz, advocated active

U.S. efforts to combat terrorism, often recommending the use of military force. Secretary of Defense Caspar Weinberger opposed Shultz, who made little headway against Weinberger, or even within his own department. Though Shultz elevated the status and visibility of counterterrorism coordination by appointing as coordinator first L. Paul Bremer and then Robert Oakley, both senior career ambassadors of high standing in the Foreign Service, the department continued to be dominated by regional bureaus for which terrorism was not a first-order concern.

Secretaries of state after Shultz took less personal interest in the problem. Only congressional opposition prevented President Clinton's first secretary of state, Warren Christopher, from merging terrorism into a new bureau that would have also dealt with narcotics and crime. The coordinator under Secretary Madeleine Albright told the Commission that his job was seen as a minor one within the department.⁸⁸ Although the description of his status has been disputed, and Secretary Albright strongly supported the August 1998 strikes against Bin Ladin, the role played by the Department of State in counterterrorism was often cautionary before 9/11. This was a reflection of the reality that counterterrorism priorities nested within broader foreign policy aims of the U.S. government.

State Department consular officers around the world, it should not be forgotten, were constantly challenged by the problem of terrorism, for they handled visas for travel to the United States. After it was discovered that Abdel Rahman, the Blind Sheikh, had come and gone almost at will, State initiated significant reforms to its watchlist and visa-processing policies. In 1993, Congress passed legislation allowing State to retain visa-processing fees for border security; those fees were then used by the department to fully automate the terrorist watchlist. By the late 1990s, State had created a worldwide, real-time electronic database of visa, law enforcement, and watchlist information, the core of the post-9/11 border screening systems. Still, as will be seen later, the system had many holes.⁸⁹

The Department of Defense
The Department of Defense is the behemoth among federal agencies. With an annual budget larger than the gross domestic product of Russia, it is an empire. The Defense Department is part civilian, part military. The civilian secretary of defense has ultimate control, under the president. Among the uniformed military, the top official is the chairman of the Joint Chiefs of Staff, who is supported by a Joint Staff divided into standard military staff compartments-J-2 (intelligence), J-3 (operations), and so on.

Because of the necessary and demanding focus on the differing mission of each service, and their long and proud traditions, the Army, Navy, Air Force, and Marine Corps have often fought ferociously over roles and missions in war fighting and over budgets and posts of leadership. Two developments diminished this competition.

The first was the passage by Congress in 1986 of the Goldwater-Nichols Act, which, among other things, mandated that promotion to high rank required some period of duty with a different service or with a joint (i.e., multiservice) command. This had strong and immediate effects, loosening the loyalties of senior officers to their separate services and causing them to think more broadly about the military establishment as a whole.⁹⁰ However, it also may have lessened the diversity of military advice and options presented to the president. The Goldwater-Nichols example is seen by some as having lessons applicable to lessening competition and increasing cooperation in other parts of the federal bureaucracy, particularly the law enforcement and intelligence communities.

The second, related development was a significant transfer of planning and command responsibilities from the service chiefs and their staffs to the joint and unified commands outside of Washington, especially those for Strategic Forces and for four regions: Europe, the Pacific, the Center, and the South. Posts in these commands became prized assignments for ambitious officers, and the voices of their five commanders in chief became as influential as those of the service chiefs.

Counterterrorism
The Pentagon first became concerned about terrorism as a result of hostage taking in the 1970s. In June 1976, Palestinian terrorists seized an Air France plane and landed it at Entebbe in Uganda, holding 105 Israelis and other Jews as hostages. A special Israeli commando force stormed the plane, killed all the terrorists, and rescued all but one of the hostages. In October 1977, a West German special force dealt similarly with a Lufthansa plane sitting on a tarmac in Mogadishu: every terrorist was killed, and every hostage brought back safely. The White House, members of Congress, and the news media asked the Pentagon whether the United States was prepared for similar action. The answer was no. The Army immediately set about creating the Delta Force, one of whose missions was hostage rescue.

The first test for the new force did not go well. It came in April 1980 during the Iranian hostage crisis, when Navy helicopters with Marine pilots flew to a site known as Desert One, some 200 miles southeast of Tehran, to rendezvous with Air Force planes carrying Delta Force commandos and fresh fuel. Mild sandstorms disabled three of the helicopters, and the commander ordered the mission aborted. But foul-ups on the ground resulted in the loss of eight aircraft, five airmen, and three marines. Remembered as "Desert One," this failure remained vivid for members of the armed forces. It also contributed to the later Goldwater-Nichols reforms.

In 1983 came Hezbollah's massacre of the Marines in Beirut. President Reagan quickly withdrew U.S. forces from Lebanon-a reversal later routinely cited by jihadists as evidence of U.S. weakness. A detailed investigation produced a list of new procedures that would become customary for forces deployed abroad. They involved a number of defensive measures, including caution not only about strange cars and trucks but also about unknown aircraft overhead. "Force protection" became a significant claim on the time and resources of the Department of Defense.

A decade later, the military establishment had another experience that evoked both Desert One and the withdrawal from Beirut. The first President Bush had authorized the use of U.S. military forces to ensure humanitarian relief in war-torn Somalia. Tribal factions interfered with the supply missions. By the autumn of 1993, U.S. commanders concluded that the main source of trouble was a warlord, Mohammed Farrah Aidid. An Army special force launched a raid on Mogadishu to capture him. In the course of a long night, two Black Hawk helicopters were shot down, 73 Americans were wounded, 18 were killed, and the world's television screens showed images of an American corpse dragged through the streets by exultant Somalis. Under pressure from Congress, President Clinton soon ordered the withdrawal of U.S. forces. "Black Hawk down" joined "Desert One" as a symbol among Americans in uniform, code phrases used to evoke the risks of daring exploits without maximum preparation, overwhelming force, and a well-defined mission.

In 1995-1996, the Defense Department began to invest effort in planning how to handle the possibility of a domestic terrorist incident involving weapons of mass destruction (WMD). The idea of a domestic command for homeland defense began to be discussed in 1997, and in 1999 the Joint Chiefs developed a concept for the establishment of a domestic Unified Command. Congress killed the idea. Instead, the Department established the Joint Forces Command, located at Norfolk, Virginia, making it responsible for military response to domestic emergencies, both natural and man-made.⁹¹

Pursuant to the Nunn-Lugar-Domenici Domestic Preparedness Program, the Defense Department began in 1997 to train first responders in 120 of the nation's largest cities. As a key part of its efforts, Defense created National Guard WMD Civil Support Teams to respond in the event of a WMD terrorist incident. A total of 32 such National Guard teams were authorized by fiscal year 2001. Under the command of state governors, they provided support to civilian agencies to assess the nature of the attack, offer medical and technical advice, and coordinate state and local responses.⁹²

The Department of Defense, like the Department of State, had a coordinator who represented the department on the interagency committee concerned with counterterrorism. By the end of President Clinton's first term, this official had become the assistant secretary of defense for special operations and low-intensity conflict.⁹³

The experience of the 1980s had suggested to the military establishment that if it were to have a role in counterterrorism, it would be a traditional military role-to act against state sponsors of terrorism. And the military had what seemed an excellent example of how to do it. In 1986, a bomb went off at a disco in Berlin, killing two American soldiers. Intelligence clearly linked the bombing to Libya's Colonel Muammar Qadhafi. President Reagan ordered air strikes against Libya. The operation was not cost free: the United States lost two planes. Evidence accumulated later, including the 1988 bombing of Pan Am 103, clearly showed that the operation did not curb Qadhafi's interest in terrorism. However, it was seen at the time as a success. The lesson then taken from Libya was that terrorism could be stopped by the use of U.S. air power that inflicted pain on the authors or sponsors of terrorist acts.

This lesson was applied, using Tomahawk missiles, early in the Clinton administration. George H.W. Bush was scheduled to visit Kuwait to be honored for his rescue of that country in the Gulf War of 1991. Kuwaiti security services warned Washington that Iraqi agents were planning to assassinate the former president. President Clinton not only ordered precautions to protect Bush but asked about options for a reprisal against Iraq. The Pentagon proposed 12 targets for Tomahawk missiles. Debate in the White House and at the CIA about possible collateral damage pared the list down to three, then to one- Iraqi intelligence headquarters in central Baghdad. The attack was made at night, to minimize civilian casualties. Twenty-three missiles were fired. Other than one civilian casualty, the operation seemed completely successful: the intelligence headquarters was demolished. No further intelligence came in about terrorist acts planned by Iraq.⁹⁴

The 1986 attack in Libya and the 1993 attack on Iraq symbolized for the military establishment effective use of military power for counterterrorism- limited retaliation with air power, aimed at deterrence. What remained was the hard question of how deterrence could be effective when the adversary was a loose transnational network.

3.6 .. . AND IN THE WHITE HOUSE

Because coping with terrorism was not (and is not) the sole province of any component of the U.S. government, some coordinating mechanism is necessary. When terrorism was not a prominent issue, the State Department could perform this role. When the Iranian hostage crisis developed, this procedure went by the board: National Security Advisor Zbigniew Brzezinski took charge of crisis management.

The Reagan administration continued and formalized the practice of having presidential staff coordinate counterterrorism. After the killing of the marines in Beirut, President Reagan signed National Security Directive 138, calling for a "shift . . . from passive to active defense measures" and reprogramming or adding new resources to effect the shift. It directed the State Department "to intensify efforts to achieve cooperation of other governments" and the CIA to "intensify use of liaison and other intelligence capabilities and also to develop plans and capability to preempt groups and individuals planning strikes against U.S. interests."⁹⁵

Speaking to the American Bar Association in July 1985, the President characterized terrorism as "an act of war" and declared: "There can be no place on earth left where it is safe for these monsters to rest, to train, or practice their cruel and deadly skills. We must act together, or unilaterally, if necessary to ensure that terrorists have no sanctuary-anywhere."⁹⁶ The air strikes against Libya were one manifestation of this strategy.

Through most of President Reagan's second term, the coordination of counterterrorism was overseen by a high-level interagency committee chaired by the deputy national security adviser. But the Reagan administration closed with a major scandal that cast a cloud over the notion that the White House should guide counterterrorism.

President Reagan was concerned because Hezbollah was taking Americans hostage and periodically killing them. He was also constrained by a bill he signed into law that made it illegal to ship military aid to anticommunist Contra guerrillas in Nicaragua, whom he strongly supported. His national security adviser, Robert McFarlane, and McFarlane's deputy, Admiral John Poindexter, thought the hostage problem might be solved and the U.S. position in the Middle East improved if the United States quietly negotiated with Iran about exchanging hostages for modest quantities of arms. Shultz and Weinberger, united for once, opposed McFarlane and Poindexter.

A staffer for McFarlane and Poindexter, Marine Lieutenant Colonel Oliver North, developed a scheme to trade U.S. arms for hostages and divert the proceeds to the Contras to get around U.S. law. He may have had encouragement from Director of Central Intelligence William Casey.⁹⁷

When the facts were revealed in 1986 and 1987, it appeared to be the 1970s all over again: a massive abuse of covert action. Now, instead of stories about poisoned cigars and Mafia hit men, Americans heard testimony about a secret visit to Tehran by McFarlane, using an assumed name and bearing a chocolate cake decorated with icing depicting a key. An investigation by a special counsel resulted in the indictment of McFarlane, Poindexter, North, and ten others, including several high-ranking officers from the CIA's Clandestine Service. The investigations spotlighted the importance of accountability and official responsibility for faithful execution of laws. For the story of 9/11, the significance of the Iran-Contra affair was that it made parts of the bureaucracy reflexively skeptical about any operating directive from the White House.⁹⁸

As the national security advisor's function expanded, the procedures and structure of the advisor's staff, conventionally called the National Security Council staff, became more formal. The advisor developed recommendations for presidential directives, differently labeled by each president. For President Clinton, they were to be Presidential Decision Directives; for President George

W. Bush, National Security Policy Directives. These documents and many others requiring approval by the president worked their way through interagency committees usually composed of departmental representatives at the assistant secretary level or just below it. The NSC staff had senior directors who would sit on these interagency committees, often as chair, to facilitate agreement and to represent the wider interests of the national security advisor.

When President Clinton took office, he decided right away to coordinate counterterrorism from the White House. On January 25, 1993, Mir Amal Kansi, an Islamic extremist from Pakistan, shot and killed two CIA employees at the main highway entrance to CIA headquarters in Virginia. (Kansi drove away and was captured abroad much later.) Only a month afterward came the World Trade Center bombing and, a few weeks after that, the Iraqi plot against former President Bush.

President Clinton's first national security advisor, Anthony Lake, had retained from the Bush administration the staffer who dealt with crime, narcotics, and terrorism (a portfolio often known as "drugs and thugs"), the veteran civil servant Richard Clarke. President Clinton and Lake turned to Clarke to do the staff work for them in coordinating counterterrorism. Before long, he would chair a midlevel interagency committee eventually titled the Counterterrorism Security Group (CSG).We will later tell of Clarke's evolution as adviser on and, in time, manager of the U.S. counterterrorist effort.

When explaining the missile strike against Iraq provoked by the plot to kill President Bush, President Clinton stated: "From the first days of our Revolution, America's security has depended on the clarity of the message: Don't tread on us. A firm and commensurate response was essential to protect our sovereignty, to send a message to those who engage in state-sponsored terrorism, to deter further violence against our people, and to affirm the expectation of civilized behavior among nations."⁹⁹

In his State of the Union message in January 1995, President Clinton promised "comprehensive legislation to strengthen our hand in combating terrorists, whether they strike at home or abroad." In February, he sent Congress proposals to extend federal criminal jurisdiction, to make it easier to deport terrorists, and to act against terrorist fund-raising. In early May, he submitted a bundle of strong amendments. The interval had seen the news from Tokyo in March that a doomsday cult, Aum Shinrikyo, had released sarin nerve gas in a subway, killing 12 and injuring thousands. The sect had extensive properties and laboratories in Japan and offices worldwide, including one in New York. Neither the FBI nor the CIA had ever heard of it. In April had come the bombing of the Murrah federal building in Oklahoma City; immediate suspicions that it had been the work of Islamists turned out to be wrong, and the bombers proved to be American antigovernment extremists named Timothy McVeigh and Terry Nichols. President Clinton proposed to amend his earlier proposals by increasing wiretap and electronic surveillance authority for the FBI, requiring that explosives carry traceable taggants, and providing substantial new money not only for the FBI and CIA but also for local police.¹⁰⁰

President Clinton issued a classified directive in June 1995, Presidential Decision Directive 39, which said that the United States should "deter, defeat and respond vigorously to all terrorist attacks on our territory and against our citizens." The directive called terrorism both a matter of national security and a crime, and it assigned responsibilities to various agencies. Alarmed by the incident in Tokyo, President Clinton made it the very highest priority for his own staff and for all agencies to prepare to detect and respond to terrorism that involved chemical, biological, or nuclear weapons.¹⁰¹

During 1995 and 1996, President Clinton devoted considerable time to seeking cooperation from other nations in denying sanctuary to terrorists. He proposed significantly larger budgets for the FBI, with much of the increase designated for counterterrorism. For the CIA, he essentially stopped cutting allocations and supported requests for supplemental funds for counterterrorism.¹⁰²

When announcing his new national security team after being reelected in 1996, President Clinton mentioned terrorism first in a list of several challenges facing the country.¹⁰³ In 1998, after Bin Ladin's fatwa and other alarms, President Clinton accepted a proposal from his national security advisor, Samuel "Sandy" Berger, and gave Clarke a new position as national coordinator for security, infrastructure protection, and counterterrorism. He issued two Presidential Decision Directives, numbers 62 and 63, that built on the assignments to agencies that had been made in Presidential Decision Directive 39; laid out ten program areas for counterterrorism; and enhanced, at least on paper, Clarke's authority to police these assignments. Because of concerns especially on the part of Attorney General Reno, this new authority was defined in precise and limiting language. Clarke was only to "provide advice" regarding budgets and to "coordinate the development of interagency agreed guidelines" for action.¹⁰⁴

Clarke also was awarded a seat on the cabinet-level Principals Committee when it met on his issues-a highly unusual step for a White House staffer. His interagency body, the CSG, ordinarily reported to the Deputies Committee of subcabinet officials, unless Berger asked them to report directly to the principals. The complementary directive, number 63, defined the elements of the nation's critical infrastructure and considered ways to protect it. Taken together, the two directives basically left the Justice Department and the FBI in charge at home and left terrorism abroad to the CIA, the State Department, and other agencies, under Clarke's and Berger's coordinating hands.

Explaining the new arrangement and his concerns in another commencement speech, this time at the Naval Academy, in May 1998, the President said:

First, we will use our new integrated approach to intensify the fight against all forms of terrorism: to capture terrorists, no matter where they hide; to work with other nations to eliminate terrorist sanctuaries overseas; to respond rapidly and effectively to protect Americans from terrorism at home and abroad. Second, we will launch a comprehensive plan to detect, deter, and defend against attacks on our critical infrastructures, our power systems, water supplies, police, fire, and medical services, air traffic control, financial services, telephone systems, and computer networks.. . . Third, we will undertake a concerted effort to prevent the spread and use of biological weapons and to protect our people in the event these terrible weapons are ever unleashed by a rogue state, a terrorist group, or an international criminal organization.. . . Finally, we must do more to protect our civilian population from biological weapons.¹⁰⁵

Clearly, the President's concern about terrorism had steadily risen. That heightened worry would become even more obvious early in 1999, when he addressed the National Academy of Sciences and presented his most somber account yet of what could happen if the United States were hit, unprepared, by terrorists wielding either weapons of mass destruction or potent cyberweapons.

3.7 .. . AND IN THE CONGRESS

Since the beginning of the Republic, few debates have been as hotly contested as the one over executive versus legislative powers. At the Constitutional Convention, the founders sought to create a strong executive but check its powers. They left those powers sufficiently ambiguous so that room was left for Congress and the president to struggle over the direction of the nation's security and foreign policies.

The most serious question has centered on whether or not the president needs congressional authorization to wage war. The current status of that debate seems to have settled into a recognition that a president can deploy military forces for small and limited operations, but needs at least congressional support if not explicit authorization for large and more open-ended military operations.

This calculus becomes important in this story as both President Clinton and President Bush chose not to seek a declaration of war on Bin Ladin after he had declared and begun to wage war on us, a declaration that they did not acknowledge publicly. Not until after 9/11 was a congressional authorization sought.

The most substantial change in national security oversight in Congress took place following World War II. The Congressional Reorganization Act of 1946 created the modern Armed Services committees that have become so powerful today. One especially noteworthy innovation was the creation of the Joint House-Senate Atomic Energy Committee, which is credited by many with the development of our nuclear deterrent capability and was also criticized for wielding too much power relative to the executive branch.

Ironically, this committee was eliminated in the 1970s as Congress was undertaking the next most important reform of oversight in response to the Church and Pike investigations into abuses of power. In 1977, the House and Senate created select committees to exercise oversight of the executive branch's conduct of intelligence operations.

The Intelligence Committees
The House and Senate select committees on intelligence share some important characteristics. They have limited authorities. They do not have exclusive authority over intelligence agencies. Appropriations are ultimately determined by the Appropriations committees. The Armed Services committees exercise jurisdiction over the intelligence agencies within the Department of Defense (and, in the case of the Senate, over the Central Intelligence Agency). One consequence is that the rise and fall of intelligence budgets are tied directly to trends in defense spending.

The president is required by law to ensure the congressional Intelligence committees are kept fully and currently informed of the intelligence activities of the United States. The committees allow the CIA to some extent to withhold information in order to protect sources, methods, and operations. The CIA must bring presidentially authorized covert action Findings and Memoranda of Notification to the Intelligence committees, and it must detail its failures. The committees conduct their most important work in closed hearings or briefings in which security over classified material can be maintained.

Members of the Intelligence committees serve for a limited time, a restriction imposed by each chamber. Many members believe these limits prevent committee members from developing the necessary expertise to conduct effective oversight.

Secrecy, while necessary, can also harm oversight. The overall budget of the intelligence community is classified, as are most of its activities. Thus, the Intelligence committees cannot take advantage of democracy's best oversight mechanism: public disclosure. This makes them significantly different from other congressional oversight committees, which are often spurred into action by the work of investigative journalists and watchdog organizations.

Adjusting to the Post-Cold War Era
The unexpected and rapid end of the Cold War in 1991 created trauma in the foreign policy and national security community both in and out of government. While some criticized the intelligence community for failing to forecast the collapse of the Soviet Union (and used this argument to propose drastic cuts in intelligence agencies), most recognized that the good news of being relieved of the substantial burden of maintaining a security structure to meet the Soviet challenge was accompanied by the bad news of increased insecurity. In many directions, the community faced threats and intelligence challenges that it was largely unprepared to meet.

So did the intelligence oversight committees. New digitized technologies, and the demand for imagery and continued capability against older systems, meant the need to spend more on satellite systems at the expense of human efforts. In addition, denial and deception became more effective as targets learned from public sources what our intelligence agencies were doing. There were comprehensive reform proposals of the intelligence community, such as those offered by Senators Boren and McCurdy. That said, Congress still took too little action to address institutional weaknesses.¹⁰⁶

With the Cold War over, and the intelligence community roiled by the Ames spy scandal, a presidential commission chaired first by former secretary of defense Les Aspin and later by former secretary of defense Harold Brown examined the intelligence community's future. After it issued recommendations addressing the DCI's lack of personnel and budget authority over the intelligence community, the Intelligence committees in 1996 introduced implementing legislation to remedy these problems.

The Department of Defense and its congressional authorizing committees rose in opposition to the proposed changes. The President and DCI did not actively support these changes. Relatively small changes made in 1996 gave the DCI consultative authority and created a new deputy for management and assistant DCIs for collection and analysis. These reforms occurred only after the Senate Select Committee on Intelligence took the unprecedented step of threatening to bring down the defense authorization bill. Indeed, rather than increasing the DCI's authorities over national intelligence, the 1990s witnessed movement in the opposite direction through, for example, the transfer of the CIA's imaging analysis capability to the new imagery and mapping agency created within the Department of Defense.

Congress Adjusts
Congress as a whole, like the executive branch, adjusted slowly to the rise of transnational terrorism as a threat to national security. In particular, the growing threat and capabilities of Bin Ladin were not understood in Congress. As the most representative branch of the federal government, Congress closely tracks trends in what public opinion and the electorate identify as key issues. In the years before September 11, terrorism seldom registered as important. To the extent that terrorism did break through and engage the attention of the Congress as a whole, it would briefly command attention after a specific incident, and then return to a lower rung on the public policy agenda.

Several points about Congress are worth noting. First, Congress always has a strong orientation toward domestic affairs. It usually takes on foreign policy and national security issues after threats are identified and articulated by the administration. In the absence of such a detailed-and repeated-articulation, national security tends not to rise very high on the list of congressional priorities. Presidents are selective in their use of political capital for international issues.

In the decade before 9/11, presidential discussion of and congressional and public attention to foreign affairs and national security were dominated by other issues-among them, Haiti, Bosnia, Russia, China, Somalia, Kosovo, NATO enlargement, the Middle East peace process, missile defense, and glob-alization. Terrorism infrequently took center stage; and when it did, the context was often terrorists' tactics-a chemical, biological, nuclear, or computer threat-not terrorist organizations.¹⁰⁷

Second, Congress tends to follow the overall lead of the president on budget issues with respect to national security matters. There are often sharp arguments about individual programs and internal priorities, but by and large the overall funding authorized and appropriated by the Congress comes out close to the president's request. This tendency was certainly illustrated by the downward trends in spending on defense, intelligence, and foreign affairs in the first part of the 1990s. The White House, to be sure, read the political signals coming from Capitol Hill, but the Congress largely acceded to the executive branch's funding requests. In the second half of the decade, Congress appropriated some 98 percent of what the administration requested for intelligence programs. Apart from the Gingrich supplemental of $1.5 billion for overall intelligence programs in fiscal year 1999, the key decisions on overall allocation of resources for national security issues in the decade before 9/11-including counterterrorism funding-were made in the president's Office of Management and Bud-get.¹⁰⁸

Third, Congress did not reorganize itself after the end of the Cold War to address new threats. Recommendations by the Joint Committee on the Organization of Congress were implemented, in part, in the House of Representatives after the 1994 elections, but there was no reorganization of national security functions. The Senate undertook no appreciable changes. Traditional issues-foreign policy, defense, intelligence-continued to be handled by committees whose structure remained largely unaltered, while issues such as transnational terrorism fell between the cracks. Terrorism came under the jurisdiction of at least 14 different committees in the House alone, and budget and oversight functions in the House and Senate concerning terrorism were also splintered badly among committees. Little effort was made to consider an integrated policy toward terrorism, which might range from identifying the threat to addressing vulnerabilities in critical infrastructure; and the piecemeal approach in the Congress contributed to the problems of the executive branch in formulating such a policy.¹⁰⁹

Fourth, the oversight function of Congress has diminished over time. In recent years, traditional review of the administration of programs and the implementation of laws has been replaced by "a focus on personal investigations, possible scandals, and issues designed to generate media attention." The unglamorous but essential work of oversight has been neglected, and few members past or present believe it is performed well. DCI Tenet told us: "We ran from threat to threat to threat.. . . [T]here was not a system in place to say, 'You got to go back and do this and this and this.'" Not just the DCI but the entire executive branch needed help from Congress in addressing the questions of counterterrorism strategy and policy, looking past day-to-day concerns. Members of Congress, however, also found their time spent on such everyday matters, or in looking back to investigate mistakes, and often missed the big questions-as did the executive branch. Staff tended as well to focus on parochial considerations, seeking to add or cut funding for individual (often small) programs, instead of emphasizing comprehensive oversight projects.¹¹⁰

Fifth, on certain issues, other priorities pointed Congress in a direction that was unhelpful in meeting the threats that were emerging in the months leading up to 9/11. Committees with oversight responsibility for aviation focused overwhelmingly on airport congestion and the economic health of the airlines, not aviation security. Committees with responsibility for the INS focused on the Southwest border, not on terrorists. Justice Department officials told us that committees with responsibility for the FBI tightly restricted appropriations for improvements in information technology, in part because of concerns about the FBI's ability to manage such projects. Committees responsible for South Asia spent the decade of the 1990s imposing sanctions on Pakistan, leaving presidents with little leverage to alter Pakistan's policies before 9/11. Committees with responsibility for the Defense Department paid little heed to developing military responses to terrorism and stymied intelligence reform. All committees found themselves swamped in the minutiae of the budget process, with little time for consideration of longer-term questions, or what many members past and present told us was the proper conduct of oversight.¹¹¹

Each of these trends contributed to what can only be described as Congress's slowness and inadequacy in treating the issue of terrorism in the years before 9/11.The legislative branch adjusted little and did not restructure itself to address changing threats.¹¹² Its attention to terrorism was episodic and splintered across several committees. Congress gave little guidance to executive branch agencies, did not reform them in any significant way, and did not systematically perform oversight to identify, address, and attempt to resolve the many problems in national security and domestic agencies that became apparent in the aftermath of 9/11.

Although individual representatives and senators took significant steps, the overall level of attention in the Congress to the terrorist threat was low. We examined the number of hearings on terrorism from January 1998 to September 2001.The Senate Armed Services Committee held nine-four related to the attack on the USS Cole. The House Armed Services Committee also held nine, six of them by a special oversight panel on terrorism. The Senate Foreign Relations Committee and its House counterpart both held four. The Senate Select Committee on Intelligence, in addition to its annual worldwide threat hearing, held eight; its House counterpart held perhaps two exclusively devoted to counterterrorism, plus the briefings by its terrorist working group. The Senate and House intelligence panels did not raise public and congressional attention on Bin Ladin and al Qaeda prior to the joint inquiry into the attacks of September 11, perhaps in part because of the classified nature of their work. Yet in the context of committees that each hold scores of hearings every year on issues in their jurisdiction, this list is not impressive. Terrorism was a second- or third-order priority within the committees of Congress responsible for national security.¹¹³

In fact, Congress had a distinct tendency to push questions of emerging national security threats off its own plate, leaving them for others to consider. Congress asked outside commissions to do the work that arguably was at the heart of its own oversight responsibilities.¹¹⁴ Beginning in 1999, the reports of these commissions made scores of recommendations to address terrorism and homeland security but drew little attention from Congress. Most of their impact came after 9/11.