When dangers increase, liberties shrink. That has been our history, especially in wartime. And today we face dangers without precedent: a mass movement of militant Islamic terrorists who crave martyrdom, hide in shadows, are fanatically bent on slaughtering as many of us as possible and—if they can—using nuclear truck bombs to obliterate New York or Washington or both, without leaving a clue as to the source of the attack.
How can we avert catastrophe and hold down the number of lesser mass murders? Our best hope is to prevent al-Qaida from getting nuclear, biological, or chemical weapons and smuglling them into this country. But we need be unlucky only once to fail in that. Ultimately we can hold down our casualities only by finding and locking up (or killing) as many as possible of the hundreds or thousands of possible al-Qaida terrorists whose strategy is to infiltrate our society and avoid attention until they strike.
The urgency of penetrating secret terrorist cells makes it imperative for Congress—and the nation—to undertake a candid, searching, and systematic reassessment of the civil liberties rules that restrict the government's core investigative and detention powers. Robust national debate and deliberate congressional action should replace what has so far been largely ad hoc presidential improvisation. While the USA-PATRIOT Act—no model of careful deliberation—changed many rules for the better (and some for the worse), it did not touch some others that should be changed.
Carefully crafted new legislation would be good not only for security but also for liberty. Stubborn adherence to the civil liberties status quo would probably damage our most fundamental freedoms far more in the long run than would judicious modifications of rules that are less fundamental. Considered congressional action based on open national debate is more likely to be sensitive to civil liberties and to the Constitution's checks and balances than unilateral expansion of executive power. Courts are more likely to check executive excesses if Congress sets limits for them to enforce. Government agents are more likely to respect civil liberties if freed from rules that create unwarranted obstacles to doing their jobs. And preventing terrorist mass murders is the best way of avoiding a panicky stampede into truly oppressive police statism, in which measures now unthinkable could suddenly become unstoppable.
This is not to advocate truly radical revisions of civil liberties. Nor is it to applaud all the revisions that have already been made, some of which seem unwarranted and even dangerous. But unlike most in-depth commentaries on the liberty-security balance since September 11—which argue (plausibly, on some issues) that we have gone too far in expanding government power—this article contends that in important respects we have not gone far enough. Civil libertarians have underestimated the need for broader investigative powers and exaggerated the dangers to our fundamental liberties. Judicious expansion of the government's powers to find suspected terrorists would be less dangerous to freedom than either risking possibly preventable attacks or resorting to incarceration without due process of law—as the Bush administration has begun to do. We should worry less about being wiretapped or searched or spied upon or interrogated and more about seeing innocent people put behind bars—or about being blown to bits.
Recalibrating the Liberty-Security Balance
The courts, Congress, the president, and the public have from the beginning of this nation's history demarcated the scope of protected rights "by a weighing of competing interests...the public-safety interest and the liberty interest," in the words of Judge Richard A. Posner of the U.S. Court of Appeals for the Seventh Circuit. "The safer the nation feels, the more weight judges will be willing to give to the liberty interest."
During the 1960s and 1970s, the weight on the public safety side of the scales seemed relatively modest. The isolated acts of violence by groups like the Weather Underground and Black Panthers—which had largely run their course by the mid-1970s—were a minor threat compared with our enemies today. Suicide bombers were virtually unheard of. By contrast, the threat to civil liberties posed by broad governmental investigative and detention powers and an imperial presidency had been dramatized by Watergate and by disclosures of such ugly abuses of power as FBI Director J. Edgar Hoover's spying on politicians, his wiretapping and harassment of the Rev. Martin Luther King, Jr., and the government's disruption and harassment of antiwar and radical groups.
To curb such abuses, the Supreme Court, Congress, and the Ford and Carter administrations placed tight limits on law enforcement and intelligence agencies. The Court consolidated and in some ways extended the Warren Court's revolutionary restrictions on government powers to search, seize, wiretap, interrogate, and detain suspected criminals (and terrorists). It also barred warrantless wiretaps and searches of domestic radicals. Congress barred warrantless wiretaps and searches of suspected foreign spies and terrorists—a previously untrammeled presidential power—in the 1978 Foreign Intelligence Surveillance Act. And Edward Levi, President Ford's attorney general, clamped down on domestic surveillance by the FBI.
As a result, today many of the investigative powers that government could use to penetrate al-Qaida cells—surveillance, informants, searches, seizures, wiretaps, arrests, interrogations, detentions—are tightly restricted by a web of laws, judicial precedents, and administrative rules. Stalked in our homeland by the deadliest terrorists in history, we are armed with investigative powers calibrated largely for dealing with drug dealers, bank robbers, burglars, and ordinary murderers. We are also stuck in habits of mind that have not yet fully processed how dangerous our world has become or how ill-prepared our legal regime is to meet the new dangers.
Rethinking Government's Powers
Only a handful of the standard law-enforcement investigative techniques have much chance of penetrating and defanging groups like al-Qaida. The four most promising are: infiltrating them through informants and undercover agents; finding them and learning their plans through surveillance, searches, and wiretapping; detaining them before they can launch terrorist attacks; and interrogating those detained. All but the first (infiltration) are now so tightly restricted by Supreme Court precedents (sometimes by mistaken or debatable readings of them), statutes, and administrative rules as to seriously impede terrorism investigators. Careful new legislation could make these powers more flexible and useful while simultaneously setting boundaries to minimize overuse and abuse.
Searches and Surveillance
The Supreme Court's caselaw involving the Fourth Amendment's ban on "unreasonable searches and seizures" does not distinguish clearly between a routine search for stolen goods or marijuana and a preventive search for a bomb or a vial of anthrax. To search a dwelling, obtain a wiretap, or do a thorough search of a car or truck, the government must generally have "probable cause"—often (if incorrectly) interpreted in the more-probable-than-not sense—to believe that the proposed search will uncover evidence of crime. These rules make little sense when the purpose of the search is to prevent mass murder
Federal agents and local police alike need more specific guidance than the Supreme Court can quickly supply. Congress should provide it, in the form of legislation relaxing for terrorism investigations the restrictions on searching, seizing, and wiretapping, including the undue stringency of the burden of proof to obtain a search warrant in a terrorism investigation.
Search and seizure restrictions were the main (if widely unrecognized) cause of the FBI's famous failure to seek a warrant during the weeks before September 11 to search the computer and other possessions of Zacarias Moussaoui, the alleged "20th hijacker." He had been locked up since August 16, technically for overstaying his visa, based on a tip about his strange behavior at a Minnesota flight school. The FBI had ample reason to suspect that Moussaoui—who has since admitted to being a member of al-Qaida—was a dangerous Islamic militant plotting airline terrorism.
Congressional and journalistic investigations of the Moussaoui episode have focused on the intelligence agencies' failure to put together the Moussaoui evidence with other intelligence reports that should have alerted them that a broad plot to hijack airliners might be afoot. Investigators have virtually ignored the undue stringency of the legal restraints on the government's powers to investigate suspected terrorists. Until these are fixed, they will seriously hobble our intelligence agencies no matter how smart they are.
From the time of FDR until 1978, the government could have searched Moussaoui's possessions without judicial permission, by invoking the president's inherent power to collect intelligence about foreign enemies. But the 1978 Foreign Intelligence Security Act (FISA) barred searches of suspected foreign spies and terrorists unless the attorney general could obtain a warrant from a special national security court (the FISA court). The warrant application has to show not only that the target is a foreign terrorist, but also that he is a member of some international terrorist "group."
Coleen Rowley, a lawyer in the FBI's Minneapolis office, argued passionately in a widely publicized letter last May 21 to FBI Director Robert S. Mueller III that the information about Moussaoui satisfied this FISA requirement. Congressional investigators have said the same. FBI headquarters officials have disagreed, because before September 11 no evidence linked Moussaoui to al-Qaida or any other identifiable terrorist group. Unlike their critics, the FBI headquarters officials were privy to any relevant prior decisions by the FISA court, which cloaks its proceedings and decisions in secrecy. In addition, they were understandably gunshy about going forward with a legally shaky warrant application in the wake of the FISA court's excoriation of an FBI supervisor in the fall of 2000 for perceived improprieties in his warrant applications. In any event, even if the FBI had done everything right, it was and is at least debatable whether its information about Moussaoui was enough to support a FISA warrant.
More important for future cases, it is clear that FISA—even as amended by the USA-PATRIOT Act—would not authorize a warrant in any case in which the FBI cannot tie a suspected foreign terrorist to one or more confederates, whether because his confederates have escaped detection or cannot be identified or because the suspect is a lone wolf.
Congress could strengthen the hand of FBI terrorism investigators by amending FISA to include the commonsense presumption that any foreign terrorist who comes to the United States is probably acting for (or at least inspired by) some international terrorist group. Another option would be to lower the burden of proof from "probable cause" to "reasonable suspicion." A third option—which could be extended to domestic as well as international terrorism investigations—would be to authorize a warrantless "preventive" search or wiretap of anyone the government has reasonable grounds to suspect of preparing or helping others prepare for a terrorist attack. To minimize any temptation for government agents to use this new power in pursuit of ordinary criminal suspects, Congress could prohibit the use in any prosecution unrelated to terrorism of any evidence obtained by such a preventive search or wiretap.
The Supreme Court seems likely to uphold any such statute as consistent with the ban on "unreasonable searches and seizures." While the Fourth Amendment says that "no warrants shall issue, but upon probable cause," warrants are not required for many types of searches, are issued for administrative searches of commercial property without probable cause in the traditional sense, and arguably should never be required. Even in the absence of a warrant or probable cause, the justices have upheld searches based on "reasonable suspicion" of criminal activities, including brief "stop-and-frisk" encounters on the streets and car stops. They have also upheld mandatory drug-testing of certain government employees and transportation workers whose work affects the public safety even when there is no particularized suspicion at all. In the latter two cases, the Court suggested that searches designed to prevent harm to the public safety should be easier to justify than searches seeking evidence for criminal cases.
Exaggerated Fear of Big Brother
Proposals to increase the government's wiretapping powers awaken fears of unleashing Orwellian thought police to spy on, harass, blackmail, and smear political dissenters and others. Libertarians point out that most conversations overheard and e-mails intercepted in the war on terrorism will be innocent and that the tappers and buggers will overhear intimacies and embarrassing disclosures that are none of the government's business.
Such concerns argue for taking care to broaden wiretapping and surveillance powers only as much as seems reasonable to prevent terrorist acts. But broader wiretapping authority is not all bad for civil liberties. It is a more accurate and benign method of penetrating terrorist cells than the main alternative, which is planting and recruiting informers—a dangerous, ugly, and unreliable business in which the government is already free to engage without limit. The narrower the government's surveillance powers, the more it will rely on informants.
Moreover, curbing the government's power to collect information through wiretapping is not the only way to protect against misuse of the information. Numerous other safeguards less damaging to the counterterrorism effort—inspectors general, the Justice Department's Office of Professional Responsibility, congressional investigators, a gaggle of liberal and conservative civil liberties groups, and the news media—have become extremely potent. The FBI has very little incentive to waste time and resources on unwarranted snooping.
To keep the specter of Big Brother in perspective, it's worth recalling that the president had unlimited power to wiretap suspected foreign spies and terrorists until 1978 (when FISA was adopted); if this devastated privacy or liberty, hardly anyone noticed. It's also worth noting that despite the government's already-vast power to comb through computerized records of our banking and commercial transactions and much else that we do in the computer age, the vast majority of the people who have seen their privacy or reputations shredded have not been wronged by rogue officials. They have been wronged by media organizations, which do far greater damage to far more people with far less accountability.
Nineteen years ago, in The Rise of the Computer State, David Burnham wrote: "The question looms before us: Can the United States continue to flourish and grow in an age when the physical movements, individual purchases, conversations and meetings of every citizen are constantly under surveillance by private companies and government agencies?" It can. It has. And now that the computer state has risen indeed, the threat of being watched by Big Brother or smeared by the FBI seems a lot smaller than the threat of being blown to bits or poisoned by terrorists.
The Case for Coercive Interrogation
The same Zacarias Moussaoui whose possessions would have been searched but for FISA's undue stringency also epitomizes another problem: the perverse impact of the rules—or what are widely assumed to be the rules—restricting interrogations of suspected terrorists.
"We were prevented from even attempting to question Moussaoui on the day of the attacks when, in theory, he could have possessed further information about other co-conspirators," Coleen Rowley complained in a little-noticed portion of her May 21 letter to Mueller. The reason was that Moussaoui had requested a lawyer. To the FBI that meant that any further interrogation would violate the Fifth Amendment "Miranda rules" laid down by the Supreme Court in 1966 and subsequent cases.
It's not hard to imagine such rules (or such an interpretation) leading to the loss of countless lives. While interrogating Moussaoui on September 11 might not have yielded any useful information, suppose that he had been part of a team planning a second wave of hijackings later in September and that his resistance could have been cracked. Or suppose that the FBI learns tomorrow, from a wiretap, that another al-Qaida team is planning an imminent attack and arrests an occupant of the wiretapped apartment.
We all know the drill. Before asking any questions, FBI agents (and police) must warn the suspect: "You have a right to remain silent." And if the suspect asks for a lawyer, all interrogation must cease until the lawyer arrives (and tells the suspect to keep quiet). This seems impossible to justify when dealing with people suspected of planning mass murder. But it's the law, isn't it?
Actually, it's not the law, though many judges think it is, along with most lawyers, federal agents, police, and cop-show mavens. You do not have a right to remain silent. The most persuasive interpretation of the Constitution and the Supreme Court's precedents is that agents and police are free to interrogate any suspect without Miranda warnings; to spurn requests for a lawyer; to press hard for answers; and—at least in a terrorism investigation—perhaps even to use hours of interrogation, verbal abuse, isolation, blindfolds, polygraph tests, death-penalty threats, and other forms of psychological coercion short of torture or physical brutality. Maybe even truth serum.
The Fifth Amendment self-incrimination clause says only that no person "shall be compelled in any criminal case to be a witness against himself." The clause prohibits forcing a defendant to testify at his trial and also making him a witness against himself indirectly by using compelled pretrial statements. It does not prohibit compelling a suspect to talk. Miranda held only that in determining whether a defendant's statements (and information derived from them) may be used against him at his trial, courts must treat all interrogations of arrested suspects as inherently coercive unless the warnings are given.
Courts typically ignore this distinction because in almost every litigated case the issue is whether a criminal defendant's incriminating statements should be suppressed at his trial; there is no need to focus on whether the constitutional problem is the conduct of the interrogation, or the use at trial of evidence obtained, or both. And as a matter of verbal shorthand, it's a lot easier to say "the police violated Miranda" than to say "the judge would be violating Miranda if he or she were to admit the defendant's statements into evidence at his trial."
But the war against terrorism has suddenly increased the significance of this previously academic question. In terrorism investigations, it will often be more important to get potentially life-saving information from a suspect than to get incriminating statements for use in court.
Fortunately for terrorism investigators, the Supreme Court said in 1990 that "a constitutional violation [of the Fifth Amendment's self-incrimination clause] occurs only at trial." It cited an earlier ruling that the government can obtain court orders compelling reluctant witnesses to talk and can imprison them for contempt of court if they refuse, if it first guarantees them immunity from prosecution on the basis of their statements or any derivative evidence. These decisions support the conclusion that the self-incrimination clause "does not forbid the forcible extraction of information but only the use of information so extracted as evidence in a criminal case," as a federal appeals court ruled in 1992.
Of course, even when the primary reason for questioning a suspected terrorist is prevention, the government could pay a heavy cost for ignoring Miranda and using coercive interrogation techniques, because it would sometimes find it difficult or impossible to prosecute extremely dangerous terrorists. But terrorism investigators may be able to get their evidence and use it too, if the Court—or Congress, which unlike the Court would not have to wait for a proper case to come along—extends a 1984 precedent creating what the justices called a "public safety" exception to Miranda. That decision allowed use at trial of a defendant's incriminating answer to a policeman's demand (before any Miranda warnings) to know where his gun was hidden.
Those facts are not a perfect parallel for most terrorism investigations, because of the immediate nature of the danger (an accomplice might pick up the gun) and the spontaneity of the officer's question. And as Rowley testified, "In order to give timely advice" about what an agent can legally do, "you've got to run to a computer and pull it up, and I think that many people have kind of forgotten that case, and many courts have actually limited it to its facts."
But when the main purpose of the interrogation is to prevent terrorist attacks, the magnitude of the danger argues for a broader public safety exception, as Rowley implied in her letter.
Congress should neither wait for the justices to clarify the law nor assume that they will reach the right conclusions without prodding. It should make the rules as clear as possible as soon as possible. Officials like Rowley need to know that they are free to interrogate suspected terrorists more aggressively than they suppose. While a law expanding the public safety exception to Miranda would be challenged as unconstitutional, it would contradict no existing Supreme Court precedent and—if carefully calibrated to apply only when the immediate purpose is to save lives—would probably be upheld.
Would investigators routinely ignore Miranda and engage in coercive interrogation—perhaps extorting false confessions—if told that the legal restraints are far looser than has been supposed? The risk would not be significantly greater than it is now. Police would still need to comply with Miranda in almost all cases for fear of jeopardizing any prosecution. While that would not be true in terrorism investigations if the public safety exception were broadened, extreme abuses such as beatings and torture would violate the due process clause of the Fifth Amendment (and of the Fourteenth Amendment as well), which has been construed as barring interrogation techniques that "shock the conscience," and is backed up by administrative penalties and the threat of civil lawsuits.
Bringing Preventive Detention inside the Law
Of all the erosions of civil liberties that must be considered after September 11, preventive detention—incarcerating people because of their perceived dangerousness even when they are neither convicted nor charged with any crime—would represent the sharpest departure from centuries of Anglo-American jurisprudence and come closest to police statism.
But the case for some kind of preventive detention has never been as strong. Al-Qaida's capacity to inflict catastrophic carnage dwarfs any previous domestic security threat. Its "sleeper" agents are trained to avoid criminal activities that might arouse suspicion. So the careful ones cannot be arrested on criminal charges until it is too late. And their lust for martyrdom renders criminal punishment ineffective as a deterrent.
Without preventive detention, the Bush administration would apparently have no solid legal basis for holding the two U.S. citizens in military brigs in this country as suspected "enemy combatants"—or for holding the more than 500 noncitizens at Guantanamo Bay. Nor would it have had a solid legal basis for detaining any of the 19 September 11 hijackers if it had suspected them of links to al-Qaida before they struck. Nor could it legally have detained Moussaoui—who was suspected of terrorist intent but was implicated in no provable crime or conspiracy—had he had not overstayed his visa.
What should the government do when it is convinced of a suspect's terrorist intent but lacks admissible evidence of any crime? Or when a criminal trial would blow vital intelligence secrets? Or when ambiguous evidence makes it a tossup whether a suspect is harmless or an al-Qaidan? What should it do with suspects like Jose Padilla, who was arrested in Chicago and is now in military detention because he is suspected of (but not charged with) plotting a radioactive "dirty-bomb" attack on Washington, D.C.? Or with a (hypothetical) Pakistani graduate student in chemistry, otherwise unremarkable, who has downloaded articles about how terrorists might use small planes to start an anthrax epidemic and shown an intense but unexplained interest in crop-dusters?
Only four options exist. Let such suspects go about their business unmonitored until (perhaps) they commit mass murders; assign agents to tail them until (perhaps) they give the agents the slip; bring prosecutions without solid evidence and risk acquittals; and preventive detention. The latter could theoretically include not only incarceration but milder restraints such as house arrest or restriction to certain areas combined with agreement to carry (or to be implanted with) a device enabling the government to track the suspect's movements at all times.
As an alternative to preventive detention, Congress could seek to facilitate prosecutions of suspected "sleepers" by allowing use of now-inadmissible and secret evidence and stretching the already broad concept of criminal conspiracy so far as to make it almost a thought crime. But that would have a harsher effect on innocent terrorism suspects than would preventive detention and could weaken protections for all criminal defendants.
As Alan Dershowitz notes, "[N]o civilized nation confronting serious danger has ever relied exclusively on criminal convictions for past offenses. Every country has introduced, by one means or another, a system of preventive or administrative detention for persons who are thought to be dangerous but who might not be convictable under the conventional criminal law."
The best argument against preventive detention of suspected international terrorists is history's warning that the system will be abused, could expand inexorably—especially in the panic that might follow future attacks—and has such terrifying potential for infecting the entire criminal justice system and undermining our Bill of Rights that we should never start down that road. What is terrorist intent, and how may it be proved? Through a suspect's advocacy of a terrorist group's cause? Association with its members or sympathizers? If preventive detention is okay for people suspected of (but not charged with) terrorist intent, what about people suspected of homicidal intent, or violent proclivities, or dealing drugs?
These are serious concerns. But the dangers of punishing dissident speech, guilt by association, and overuse of preventive detention could be controlled by careful legislation. This would not be the first exception to the general rule against preventive detention. The others have worked fairly well. They include pretrial detention without bail of criminal defendants found to be dangerous, civil commitment of people found dangerous by reason of mental illness, and medical quarantines, a practice that may once again be necessary in the event of bioterrorism. All in all, the danger that a preventive detention regime for suspected terrorists would take us too far down the slippery slope toward police statism is simply not as bad as the danger of letting would-be mass murderers roam the country.
In any event, we already have a preventive detention regime for suspected international terrorists—three regimes, in fact, all created and controlled by the Bush administration without congressional input. First, two U.S. citizens—Jose Padilla, the suspected would-be dirty bomber arrested in Chicago, and Yaser Esam Hamdi, a Louisiana-born Saudi Arabian captured in Afghanistan and taken first to Guantanamo—have been in military brigs in this country for many months without being charged with any crime or allowed to see any lawyer or any judge. The administration claims that it never has to prove anything to anyone. It says that even U.S. citizens arrested in this country—who may have far stronger grounds than battlefield detainees for denying that they are enemy combatants—are entitled to no due process whatever once the government puts that label on them. This argument is virtually unprecedented, wrong as a matter of law, and indefensible as a matter of policy.
Second, Attorney General John Ashcroft rounded up more than 1,100 mostly Muslim noncitizens in the fall of 2001, which involved preventive detention in many cases although they were charged with immigration violations or crimes (mostly minor) or held under the material witness statute. This when-in-doubt-detain approach effectively reversed the presumption of innocence in the hope of disrupting any planned followup attacks. We may never know whether it succeeded in this vital objective. But the legal and moral bases for holding hundreds of apparently harmless detainees, sometimes without access to legal counsel, in conditions of unprecedented secrecy, seemed less and less plausible as weeks and months went by. Worse, the administration treated many (if not most) of the detainees shabbily and some abusively. (By mid-2002, the vast majority had been deported or released.)
Third, the Pentagon has incarcerated hundreds of Arab and other prisoners captured in Afghanistan at Guantanamo, apparently to avoid the jurisdiction of all courts—and has refused to create a fair, credible process for determining which are in fact enemy combatants and which of those are "unlawful."
These three regimes have been implemented with little regard for the law, for the rights of the many (mostly former) detainees who are probably innocent, or for international opinion. It is time for Congress to step in—to authorize a regime of temporary preventive detention for suspected international terrorists, while circumscribing that regime and specifying strong safeguards against abuse.
Civil Liberties for a New Era
It is senseless to adhere to overly broad restrictions imposed by decades-old civil-liberties rules when confronting the threat of unprecedented carnage at the hands of modern terrorists. In the words of Harvard Law School's Laurence H. Tribe, "The old adage that it is better to free 100 guilty men than to imprison one innocent describes a calculus that our Constitution—which is no suicide pact—does not impose on government when the 100 who are freed belong to terrorist cells that slaughter innocent civilians, and may well have access to chemical, biological, or nuclear weapons." The question is not whether we should increase governmental power to meet such dangers. The question is how much.