As its new term began this week, the Supreme Court heard two cases involving the boundaries of privacy in an increasingly transparent society; later in the term, it will hear at least one more. The constitutional question in all three cases is whether the government needs to have individualized suspicion before it can conduct searches using relatively unintrusive, extremely effective technologies-- thermal-imaging devices that can detect heat emitting through house walls, for example, or random checkpoints with drug-sniffing dogs. The cases allow the Court to grapple with an injury that is similar, but not identical, to one the Fourth Amendment's Framers meant to prohibit: the indignity of living in a zero- tolerance society, in which there is no place to hide from the all-seeing gaze of the state.
At the time the Constitution was written, the classic examples of unreasonable searches were the writs of assistance and general warrants that had allowed agents of the crown to break into the homes of American colonists and riffle through their private papers. These searches were very intrusive (involving physical trespass into the most private corners of the home), not very effective (requiring the seizure of many innocent papers in the course of discovering suspicious ones), and extremely arbitrary (allowing officers complete discretion in deciding which places should be searched and which persons or things should be seized).
At first glance, the mass searches made possible by twenty-first-century technology avoid these unpleasant side effects. Consider, for example, the virtual speed traps many cities are today setting up. Drivers exceeding the speed limit have their license plates and faces photographed automatically and later receive tickets in the mail. Unlike the eighteenth-century writs of assistance, these searches are not at all arbitrary (all drivers are searched) , completely effective (only the guilty are punished), and generally unintrusive (offenders are ticketed without being physically stopped). Or consider the FBI's unfortunately named Carnivore system, which invisibly searches everyone's e-mail but alerts law enforcement only when it finds the particular messages it has been programmed to look for. This, too, is more effective, and less intrusive and arbitrary, than the eighteenth-century approach of dispatching human beings to rummage through people's desk drawers.
Why, then, should pervasive low-level surveillance raise any constitutional concerns at all? One reason is that citizens in a liberal society have a powerful interest in not being constantly observed by the state. Without " insulation from observability," the sociologist Robert Merton has written, " the pressure to live up to the details of all (and often conflicting) social norms would become literally unbearable." It's arguable that this "insulation from observability," rather than a mere ban on arbitrary searches, was a core value the Fourth Amendment's Framers intended to protect. They believed, for example, that searches of private papers, even with warrants, were unconstitutional, and they accepted some measure of seditious libel because they felt freedom would be threatened if state agents could read citizens' diaries and spy on them in their bedrooms.
Unfortunately, today's courts have been slow to translate this antipathy to pervasive surveillance for a world in which the greatest threats to privacy are virtual rather than physical. On Tuesday, the Supreme Court heard a case that will determine whether the Indianapolis police department violated the Fourth Amendment when, during four months in 1998, it set up roadblocks on city streets and led drug-sniffing dogs around more than 1,000 cars. The Court has already approved the use of random checkpoints to catch drunk drivers and stop illegal immigrants at the border. And lawyers for the city argued that since police have the authority to stop cars to check drivers' licenses and registrations, they should be free to add the drug sniffs. "So long as they have the authority to stop the car, what difference does it make if they have another purpose?" asked Justice Antonin Scalia. But, as Chief Judge Richard A. Posner of the U.S. Court of Appeals in Chicago noted in striking down the program, the main purpose behind it was catching drug offenders, not checking licenses. Furthermore, citizens have a powerful countervailing interest in not being constantly observed on the road unless the police have reason to suspect them of having committed particular offenses. If the Court lets police stop drivers for license checks as an excuse for searching their entire cars, this will inevitably speed the use of less intrusive but more pervasive forms of technological surveillance: automated cameras, for example, that can record and ticket every possible infraction committed in your car, from driving without a seat belt on to driving while talking on a cell phone. These invisible and ruthlessly accurate searches would be less intrusive and more effective than the dog sniffs, but they would be similarly unreasonable. We don't want to live in Bentham's Panopticon.
Unfortunately, the Supreme Court's current approach to privacy may be too wooden to regulate pervasive but unobtrusive forms of surveillance. Consider Kyllo v. United States, a case the Supreme Court will hear later this term. One day in January 1992, at three o'clock in the morning, an officer from the Oregon National Guard sat in an unmarked car and aimed a thermal-imaging device at three houses, including that of Danny Lee Kyllo, who was suspected of growing marijuana with high-intensity lights. With the thermal-imaging device, Kyllo's walls "showed much warmer" than the other two houses, according to an officer; this evidence was used to get a warrant to search Kyllo's house, where officers found the suspected pot. Kyllo challenged the thermal-imaging device as an unreasonable search, but the appellate court ultimately rejected his claim, holding that Kyllo had no legitimate expectation that privacy extended to the heat waves being emitted from his house: He had voluntarily exposed those heat waves to public view, in the same way people choose to put out the trash.
This can't be right. Heat waves, unlike trash, aren't put out; they are emitted without human effort. The question is whether citizens in a civilized society are entitled to expect that their government isn't lurking in front of their homes and peering through their walls without proper cause. If the Supreme Court approves the use of thermal imaging without a warrant to investigate infractions as minor as growing marijuana (as four appellate courts have done), Americans will have no way to be certain that their most intimate movements at home aren't being recorded by hidden sensors. And this uncertainty will inhibit our freedom and autonomy in spaces that should be considered private.
The third privacy case, argued at the Court this Wednesday, asks whether police in Charleston, South Carolina, violated the Fourth Amendment by developing a policy with a city hospital to test the urine of pregnant women suspected of using cocaine and, without their consent, turn the results over to law enforcement officials, who would sometimes arrest them as they left the delivery room. An appellate court upheld the policy, arguing that urinalysis is very reliable and not very intrusive and that its purpose was not to enforce the drug laws but to protect the fetus, which is considered a " person" under South Carolina law. But, as Justice Stephen Breyer noted at the oral argument, the American Medical Association allows doctors to collect intimate medical information and turn it over to the police without consent only in a few special circumstances--when a patient is threatening serious and imminent bodily harm to herself or a third party. In this case, several of the arrests took place after the children were born, which means that any damage had already taken place. Scalia responded that doctors are required by law to report evidence of gunshot wounds, which, similarly, seems more investigative than protective. But a doctor seeing a gunshot wound while examining a patient is like a cop seeing a bloody knife in the backseat while issuing a speeding ticket: It's strong evidence of violent crime, discovered in plain view. South Carolina, by contrast, has enlisted doctors as agents of the police and directed them to conduct intimate medical tests for the sole purpose of protecting third parties. With or without a warrant, state hospitals should not be allowed to test suspected drug users for HIV and then turn the test results over to the police for release to the patient's known sexual partners. Urinalysis, moreover, is already being superseded by more sophisticated forms of medical surveillance; it may soon become routine for hospitals to make genetic maps of DNA samples to determine an individual's predisposition for various diseases--or even for future violence. If the results of these tests could be turned over to state agents without consent, individuals would lose control over the most basic markers of their identity. In the eighteenth and nineteenth centuries, a far slighter invasion--state scrutiny of a diary--was held to violate the Fourth and Fifth Amendments.
There are costs, of course, to prohibiting pervasive, highly effective, and relatively unobtrusive surveillance. If the state is forbidden to search everyone for drugs with dogs and urine tests, it may be more likely to target a few poor, minority suspects with battering rams and street arrests. But although the new technologies of pervasive surveillance might reduce discriminatory law enforcement, they threaten liberty in a subtler but more encroaching way: by making all citizens feel they have no escape from constant observation by the state. Some democracies passively accept these technological encroachments in the name of security: Britain, increasingly, has cameras on street corners. But the roots of our Anglo-American constitutional tradition reflect a horror of this kind of totalizing state control. By rejecting the zero tolerance model, a liberal state treats its citizens as individuals rather than objects. If, by contrast, the Supreme Court abandons the principle that the state needs identifiable suspicion of serious wrongdoing before it can engage in pervasive surveillance, our Constitution in the twenty-first century will protect far less privacy than it did in the eighteenth.
Jeffrey Rosen is legal affairs editor at The New Republic and president and CEO of the National Constitution Center.