The tragic events of September 11, 2001, and the false assessment ofSaddam Hussein's weapons arsenal were terrible reminders that good information isessential to national security. These failures convinced the American public thattheir intelligence system was broken and prompted a radical reorganization ofagencies and personnel, but as Richard K. Betts argues in this book, critics andpoliticians have severely underestimated the obstacles to true reform.One of the nation's foremost political scientists, Betts drawson three decades of work within the U.S. intelligence community to illuminate theparadoxes and problems that frustrate the intelligence process. Unlike America'sefforts to improve its defenses against natural disasters, strengthening itsstrategic assessment capabilities means outwitting crafty enemies who operate beyondU.S. borders. It also requires looking within to the organizational and politicaldynamics of collecting information and determining its implications for policy.Combining academic research with personal experience, Bettsoutlines strategies for better intelligence gathering and assessment. He describeshow fixing one malfunction can create another; in what ways expertise can be both avital tool and a source of error and misjudgment; the pitfalls of always strivingfor accuracy in intelligence, which in some cases can render it worthless; thedanger, though unavoidable, of "politicizing" intelligence; and the issue ofsecrecy& -when it is excessive, when it is insufficient, and howlimiting privacy can in fact protect civil liberties.Bettsargues that when it comes to intelligence, citizens and politicians should focusless on consistent solutions and more on achieving a delicate balance betweenconflicting requirements. He also emphasizes the substantial success of theintelligence community, despite its well-publicized blunders, and highlightselements of the intelligence process that need preservation and protection. Manyreformers are quick to respond to scandals and failures without detailed, historicalknowledge of how the system works. Grounding his arguments in extensive theory andpolicy analysis, Betts takes a comprehensive and realistic look at how knowledge andpower can work together to face the intelligence challenges of the twenty-firstcentury.

Richard K. Betts is director of the Arnold A. Saltzman Institute of War and Peace Studies at Columbia, senior fellow at the Council on Foreign Relations, and author of numerous books on military strategy and foreign policy. He has served on the National Commission on Terrorism, staffs of the Senate Intelligence Committee and National Security Council, and advisory panels for the director of the Central Intelligence Agency.

Preface	p. xi
Twenty-first-Century Intelligence: New Enemies and Old	p. 1
Permanent Enemies: Why Intelligence Failures Are Inevitable	p. 19
Theory Traps: Expertise as an Enemy	p. 53
Incorruptibility or Influence? Costs and Benefits of Politicization	p. 66
Two Faces of Failure: September 11 and Iraq's WMD	p. 104
An Intelligence Reformation? Two Faces of Reorganization	p. 124
Whose Knowledge of Whom? The Conflict of Secrets	p. 159
Enemies at Bay: Successful Intelligence	p. 183
Notes	p. 195
Index	p. 229
Table of Contents provided by Ingram. All Rights Reserved.

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.

(Read this excerpt in pdf format)

Excerpt from the Chapter "Whose Knowledge of Whom"

Intelligence Collection and Civil Liberties

There is a world of difference between the problems of collecting intelligence abroad and at home. All U.S. intelligence operations must conform to U.S. law, but that law leaves operations outside U.S. territory relatively unconstrained. When doing their jobs abroad, American agents may break the laws of the countries in which their operations are undertaken. They may give money to political parties, plant bugs in defense ministries, bribe legislators, tap the phones of diplomats, and do all sorts of things to gather information that the FBI could not normally do within the United States, at least without a court order. More intrusive collection inside the United States would have done the most to boost the chances of averting the September 11 attacks. Great changes in that direction may make Americans fear that the costs exceed the benefits -- indeed, that if civil liberties are compromised, the terrorists will have won.

The unique responsibility of intelligence collectors is to penetrate enemy secrecy, to uncover information that adversaries try to conceal. Since spies or terrorists pose as innocents, maximizing collection of potentially useful information means that intelligence collection becomes a fishing expedition, and that most suspects made subject to surveillance and searches are innocent. To minimize intrusion against innocents, as well as to target intelligence resources efficiently, collection is not maximized, at least in situations short of immediate threats to national survival. Normally, only those who present some substantial cause for suspicion have been subjected to surveillance. How substantial that reason must be -- or whether large numbers of innocent people should be observed to determine which among them are suspects to be scrutinized -- are matters of regular political and legal debate.

Critics of aggressive government intrusion often cite a famous warning by Benjamin Franklin. In congressional debate on the USA Patriot Act, for example, Senator Patrick Leahy cited the gist of what Franklin said as "If we surrender our liberty in the name of security, we shall have neither." As Michael J. Woods points out, however, "Franklin's actual words are more nuanced.... 'Those who give up essential Liberty, to purchase a little temporary safety, deserve neither Liberty nor Safety.'" Whatever Franklin meant, security and liberty can go together, but only if each is compromised to some extent to protect the other.

Priorities Among Liberties

There are two big mistakes one can make when confronted by the trade-off between national security and individual liberty. One is to deny that there is such a trade-off. The other is to embrace it without qualifying which specific aspect of liberty is at issue, instead lumping all liberties together. The first mistake, common only among the most fervent supporters of individual rights, denies the trade-off because it is psychologically unacceptable to admit that good things do not all go together. If forced to admit the trade-off, some zealots insist categorically that civil liberties must always take precedence over security, that if any liberty is sacrificed for the sake of combating terrorists, the terrorists will have won, or that any contraction of liberty is the camel's nose in the tent that portends the collapse of the Constitution. This sort of thinking is a recipe for dangerous constraints on the government's ability to gather intelligence, especially the type that offers the best hope of foiling terrorist plots.

The second mistake is more common: the assumption that the trade-off is between collective national defense and individual civil liberties, period. Thinking in terms of a dichotomy makes it much harder to strike a balance. Civil liberties should not and need not all suffer for the sake of security, because liberties are not all of a piece. The most legitimate trade-off is not between security and liberty in general, but between security and privacy, the one aspect of liberty that inhibits the government's acquisition of information. Limiting the government's knowledge of one's life and activities is an important freedom, but not the most important one. There is no need to compromise the more important elements of civil liberties having to do with freedom of speech, political organization, religion, or especially the right to due process of law -- the freedom from arbitrary arrest and incarceration without the chance to contest one's guilt. Having one's phone tapped without proper cause is not as damaging as being imprisoned for years without trial. I argue that it is more vital to keep within strict interpretations of the limits for the First and Fifth amendments in the Bill of Rights than for the Fourth Amendment. This argument is another that does not have support in the jurisprudence of the Constitution, which recognizes no hierarchy of rights; it is asserted as a practical matter of common sense, not as accepted law.

Zealots at both ends of the spectrum are allied against compromisers. (Consider one set of strange bedfellows: the American Civil Liberties Union has "entered into an alliance, called Patriots to Restore Checks and Balances, with conservative groups such as Grover Norquist's Americans for Tax Reform, Phyllis Schlafly's Eagle Forum and the Citizens Committee for the Right to Keep and Bear Arms.") Libertarian absolutists mimic the National Rifle Association. Like ardent proponents of the right to bear arms, they see even a limited concession as the first step in the unraveling of all privacy rights. Supporters of highly intrusive intelligence collection, on the other hand, too often rest their case on such a breathtaking assertion of presidential powers that any significant improvement in the public's sense of security, or a change of administration, is bound to produce an antiauthoritarian backlash. The compromise argument in this chapter tilts toward liberty on the issue of due process and away from it on the issue of privacy. The latter part of the argument relates to intelligence: It is reasonable to invade the privacy of some citizens in order to gain information that might help to protect the lives of all citizens.

This compromise of one aspect of liberty ultimately serves to protect liberty and the rule of law in general. In the Constitution, life comes before liberty. Life without liberty is possible, but liberty without life is not. Without security, few Americans would be grateful for liberty. This is not readily apparent at most times -- and the proposition strikes some as authoritarian alarmism -- because Americans have habitually taken basic security for granted. They rarely give even a fleeting thought to the tremendous protection afforded the United States by its geography, the two huge ocean moats to the east and west and weak neighbors to the north and south. Indeed, it was the assumption of basic security by most Americans that made the Al Qaeda attacks such a shock to them, even though they caused damage no greater than many counterterrorism specialists had been warning about for a long time.

If those on either side of the debate insist on lumping all civil liberties together in a trade-off with national security, liberty comes out on top in the competition only as long as Americans are spared more attacks on the scale of September 11. Even then, some compromises of liberty remain, as represented by the USA Patriot Act and by congressional acquiescence to warrantless surveillance programs by the National Security Agency (NSA). If September 11 fades into history and terrorists manage only occasional, small provocations, political tides may roll back some of those constraints on freedom. Experience of more dramatic attacks, however, especially ones involving chemical, biological, or radiological weapons, would sweep away most concern for civil liberties in a panic. It is important to differentiate types and priorities of liberties so that they do not get thrown out wholesale in an emergency. It follows that it is also important to do so in order to facilitate intelligence collection that would raise the odds of preventing just the sort of emergency that could produce the panic. In short, reducing a bit of liberty today buttresses a lot of it tomorrow.

Does this exaggerate the risk that vital freedoms could be jettisoned when much greater insecurity puts them under pressure? Not at all. Consider how big a dent was put in privacy and due process just by September 11. The strikes against the World Trade Center and Pentagon were a milder shock than biological weapons attacks that might kill tens or hundreds of thousands would be. If such attacks were attributed to an alienated group of Muslims living in this country, demands for preventive detention of Muslim Americans would hardly remain unthinkable. Recall the internment of Japanese Americans during World War II. Although this event came to be seen, decades after the defeat of Japan, as an aberrant miscarriage of justice rooted in a racist past, it was practically uncontroversial during wartime. More to the point, the Supreme Court decision in the Korematsu case upholding the action has never been overturned. In 1983 Korematsu's individual conviction was reversed because of false information in the government's case. The factual basis was overturned in a coram nobis, "error before us," decision, but the legal point was not readdressed, so the Supreme Court decision technically stands. Before September 11 this was probably among the cases that, although never officially overruled, are universally recognized to be without continuing validity. Yet, if desperate circumstances were to create political demands strong enough, its technical standing could become effective as a precedent, allowing jurists to accede to a repetition. If the Supreme Court decision upholding internment of Japanese Americans can remain standing, it should not be impossible to put the arguments here for less drastic measures to compile information about Americans into law if they are deemed necessary.

One might believe that all this is a moot point, since after September 11 the American system tilted clearly toward the compromises of privacy endorsed here. But the tilt in favor of permissive rules for surveillance and searches is vulnerable to reversal because the Bush administration based its actions in implementing some of them on the overriding of an existing statute, the Foreign Intelligence Surveillance Act (FISA). Stretching the interpretation of legal authority to an extreme is not a solution, irrespective of its damaging effect in subverting the basis of constitutional government. If policy is pursued extralegally because of panic, it is likely to be rejected when alarm subsides. Standards that are to survive for the long haul must be found between the libertarian absolutists and the champions of unlimited presidential authority in wartime.

Although the system tilted against civil liberties in the trade-off with security after September 11 (as it did in several wars), some of that tilt was the wrong kind -- depriving American citizens of due process of law. Abuse of this sort threatens to discredit desirable compromises of privacy rights when times get better. As the war on terror developed, however, public debate reflected the wrong priorities. Aspects of the Patriot Act allowing the scrutiny of library records, for example, generated as much protest as did the imprisonment of American citizens without trial.

The Wrong Tilt

Immediately after September 11, hundreds of aliens were detained and questioned in the United States, and many were deported. These roundups and the conditions of confinement were controversial for critics on the left, but the arrests were temporary and deportations were based on immigration violations. The imprisonments were terminated within months. In the years that followed, dozens of people, many of them U.S. citizens, were held without trial as material witnesses, sometimes for two months or more. Other American citizens who claimed to be businessmen mistakenly suspected of anti-U.S. activity were arrested in Iraq and held in military prisons for long periods. If these were abuses, their duration was limited. The most troubling cases, however, were those of U.S. citizens or legal residents imprisoned as unlawful combatants, without recourse to the courts, for prolonged periods. The most prominent case involved José Padilla, arrested in 2002 on U.S. soil not for committing a violent act but for allegedly having a plan to build and detonate a radiological weapon in a city. Padilla was held for three years without trial, without a chance to contest his guilt in court. (Facing the prospect of a judicial ruling against Padilla's imprisonment without trial, the government eventually changed the accusations, charged him, and put his case into the courts.)

The national security rationale for denying normal rights of due process to citizens in these cases was weak. Attorneys for the executive branch did convince the Fourth Circuit Court of Appeals that giving Padilla the right to contest the accusations against him could materially aid Al Qaeda, but the only reasonable grounds were that a trial could damage intelligence sources. In cases of this sort, the courts may not ultimately rule in favor of the basic right of a defendant to see and contest evidence, but the intelligence problem should not be used as an excuse to imprison a suspect indefinitely. Procedures should be adopted so that secrets can be protected in a legal proceeding that compels the government to prove the suspect's guilt. This is not easily done in any manner that does not compromise the traditional constitutional rights of the accused, but compromised rights to due process are better than none at all. The defense could be allowed to confront secret evidence using procedures similar to those in deportation cases that provide for defense counsels with security clearances to see and challenge evidence. Denying the right of the defendant to see the evidence would have to be conditioned on proving that doing so would reveal a specific and important intelligence source or method that would be lost as a result -- not simply that the defendant would learn something classified. Constitutional issues and legal differences between immigration and criminal cases might keep such solutions in the realm of theory. But either the legal impediments must be overcome or some risk has to be taken with the demonstration of the evidence to strike a reasonable balance in the trade-off between security and liberty.

Experience, as well as principle, precludes justifying a tilt against due process by assuming that the executive's prerogative to imprison suspects indefinitely without charge will not lead to gross injustice. One need look no further than the case of Capt. James Yee, the hapless Muslim army chaplain at Guantanamo who was arrested on suspicion of espionage in 2003. Because Yee was not denied access to judicial procedures, his lawyers were able to probe the evidence and demolish the case against him (although the government did not concede his innocence, even after removing punishment as mild as a reprimand from his record). He was tried not on the original charge of concern to national security, but for adultery. Had authorities been able to hold Yee without trial, he could have been locked away as long as Padilla.

The strongest version of the argument here applies to U.S. citizens but might concern others as well. Philip Heymann notes that the Bush administration consistently maintained "that the location at which a non-citizen is held, the battlefield conditions under which he was seized, the absence of uniform, or a threat to willingly cause civilian deaths will preclude any significant form of judicial review for almost anyone whom the administration suspects of involvement with terrorists." Yet even Israel when facing the intifada had a supreme court that exerted some judicial review over actions against Palestinians. Heymann says that "though our danger is far less than the danger Israel faces, our willingness to abandon the most fundamental judicial protections of personal security has been far greater." (The most difficult issues in this regard were posed by the indefinite imprisonment of suspected enemy combatants at Guantanamo and their trial by military tribunals, but this important set of problems is not considered here.)

Not all customs of due process should be sacrosanct. For example, the demolition of the so-called wall between criminal law enforcement investigations and intelligence collection missions after September 11, allowing the FBI to pass information from the former to the latter in ways not previously permitted, was a good change in policy. But the main point is that the most fundamental right of being presumed innocent until proven guilty need not be abandoned. Protecting American security does not require putting suspects in dungeons and throwing away the keys. Without secure rights to due process in the face of accusations that one is an unlawful combatant, all Americans are vulnerable to the whim of any overzealous president.

Threats to Privacy vs. Threats from Privacy

It is not so vitally important to risk security for the sake of all other civil liberties. The main argument here is in the other direction, in favor of compromising privacy rights for the sake of national security. Personal secrecy is valuable and should be respected when it does not endanger American lives, but it is not as vital as the right to freedom from imprisonment without trial. The government's discovery of embarrassing details about one's personal life may be undesirable, but it does not deprive one of the more essential freedoms.

Unlimited rights to privacy would threaten the ability of the government to gather intelligence that could be important in detecting and blocking enemy plots. Terrorists in particular have one basic advantage: the capacity to conspire -- to mobilize, coordinate, and plan destructive activities under the cover of secrecy. Privacy provides some of the necessary secrecy. If terrorists were able to lower the odds of having communications intercepted by police and intelligence services, and if they were safer from police searches of their property, they could conspire more effectively and the odds that they could successfully execute attacks would improve.

Minimizing limits on intelligence collection would not assuredly prevent disaster, nor can it be proved that reduced intelligence collection would produce more attacks. The easier it is to get information from people who are trying to hide it, however, the better the odds that authorities will get some useful information about plots in time to break them up. If terrorists know that the government has the authority for it, surveillance also provides some deterrent effect and complicates and retards the coordination of plots by inhibiting prompt and efficient communication among plotters. If aggressive invasions of individual privacy for the sake of intelligence collection were to save, say, ten lives in a decade, would that limitation of civil liberties be legitimate? If not, would saving one hundred lives justify such a shift? One life? How many saved lives would warrant major compromises of privacy? The problem, of course, is that it is impossible to know how many lives, if any, will be saved by intrusive intelligence collection. Given that uncertainty, however, it is preferable to err in the direction of saving lives than of maximizing privacy.

Consider the case of Zacarias Moussaoui, who was ultimately convicted and sentenced to life in prison for his involvement in the September 11 plot. The joint congressional investigation concluded that because the FBI had problems applying for FISA warrants during the summer of 2001, there was "a diminished level of coverage of suspected al-Qa'ida operatives in the United States." Some surveillance operations against Al Qaeda suspects stopped because authorizations expired and the FBI did not want to apply for renewal because of uncertainty about the accuracy of their case. "Most of the FISA orders targeting al-Qa'ida that expired after March 2001 were not renewed before September 11," according to the joint congressional investigation. The complications in the FISA process at the time, the "thicket of procedures" and "the wall between intelligence gathering and law enforcement," effectively reduced the chance of interdicting the plot. The FBI did not seek a warrant to target Moussaoui, despite the facts that he was enrolled in flight school and that French intelligence had warned that he was a radical. As a result, his computer was not searched until after September 11. It is by no means certain that intense surveillance of Moussaoui before that day would have revealed the plot -- that possibility was cited by the prosecution in his trial as grounds for requesting the death penalty -- but at the very least it would have connected a few dots and increased the chances of following clues back to the nineteen hijackers.

The hesitation to mount aggressive surveillance and searches in this case was not absolutely required by law. It developed naturally alongside the highly developed set of legal safeguards rooted in the traditional American reverence for privacy and the additional restraints inspired by abuses of domestic intelligence gathering in the first half of the Cold War. Hindsight makes clear that this constraint should have been loosened. (The National Commission on Terrorism realized this more than a year before September 11 when it recommended that the Office of Intelligence Policy and Review stop applying standards more stringent than those required by the FISA statute when approving applications for electronic surveillance. Like the other good recommendations by the commission, it was not adopted before September 11.) If the argument that the restraint of surveillance should have been loosened before September 11 is persuasive, it should remain loosened as long as outside enemies of consequence remain at large. High standards for protecting privacy are like strictures against risking collateral damage in combat. They take precedence more easily when the security interests at stake are not the survival of one's own people but become harder to justify when they are.

Erring on the side of security rather than privacy makes fishing expeditions attractive, although there are legal barriers to promiscuous surveillance. Effective counterterrorism, however, requires fishing. Judge Richard Posner criticized demands for court warrants for eavesdropping on the grounds that they require, in effect, that the government know in advance who the terrorists are. "The challenge is not to track down known terrorists," he said. "It's to find out who the terrorists are." Standards for deciding who is a legitimate target should be based on a combination of indices that are empirically correlated with service to a foreign power or support of terror tactics. To some this raises the specter of "racial profiling." It is profiling, make no mistake, but not racial profiling. It does not mean targeting anyone solely on the basis of ethnicity -- for example, Arab Americans. Instead, proper profiling should rely on a combination of indicators correlated with past cases of espionage or involvement with terrorism. Nationality or religious practice might be included in a set of several actuarially relevant indicators but would not in themselves suffice to trigger investigation.

Any permissive standards for surveillance will worry civil libertarians fearful of the slippery slope toward abuse. For example, controversies developed over reports that the Census Bureau disclosed demographic data on Arab Americans to the Department of Homeland Security and that the FBI and Department of Energy were conducting radiation monitoring of mosques and Muslim businesses to detect possible radiological weapons. The FBI claimed that its actions were based on intelligence leads and patterns associated with Al Qaeda activities, not with Muslims per se. This is where the delicacy of balance is hard to manage, but the answer to the problem is to work on the delicacy, not to abandon either concern.

Until the mid-1970s, effective legal constraints on domestic intelligence collection were weak. Congressional investigation revealed abuses in targeting activists in the civil rights and anti--Vietnam War movements, or other questionable targeting of individuals under programs such as the Huston Plan, COINTELPRO, Operation Chaos, and others. Constraints were tightened, most notably in the Foreign Intelligence Surveillance Act of 1978, which institutionalized the special process for obtaining warrants for surveillance within the United States. There were also so-called attorney general procedures governing legally sensitive aspects of intelligence collection, including "minimization procedures" for handling personal information acquired as a byproduct of legitimate collection. After September 11 the tide moved back in the other direction, diminishing privacy, as recommended here. More intrusive information gathering is controversial, but if it helps avert future attacks it will also avert more severe blows against civil liberties. Americans should remember as well that many solid and humane democracies that have had more permissive rules for collecting information on people than Americans have had seem to live with them without great unease. After the shift toward more intrusion after 2001, the problems became how to keep policy and law in conformity and how to institutionalize safeguards to prevent abuses of power as the government acquired domestic information more freely.

COPYRIGHT NOTICE: Copyright © Columbia University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. For more information, send e-mail to cw270@columbia.edu