One of the most complicated questions judges—and all the rest of us—have to grapple with is how to apply the Fourth Amendment to the new order of policing. As we’ve seen, the Fourth Amendment is all about suspicion. The people who wrote and ratified it abhorred “general searches,” which is to say searches aimed at many people, without sufficient cause to suspect any of them. But much of policing today looks like what one might call a general search. In order to deter crime, and prevent terrorist incidents, many policing tactics are aimed at the entire population—without reason to suspect any given individual of anything. That’s true of everything from drunk-driving roadblocks to airport security to bulk data collection.
How can today’s policing be squared up with the Framers’ antipathy of general searches? Answering that question requires us to recall the specific evils that animated the Fourth Amendment: arbitrary, unjustified government intrusions into people’s lives. If we focus on what disturbed the Framers about general searches, rather than on the label itself, it is possible to reconcile their concerns with present-day practice. Not only that, but we can simplify Fourth Amendment law greatly.
IS NO SUSPICION EVER BETTER THAN SOME?
On January 15, 2001, officers from the auto trap division of the Orange County, Florida, Sheriff’s Office visited Wholesale Auto Advantage, Bill Bruce’s auto repair and salvage business. The police say there were eight officers present; employees of Wholesale insist it was closer to twenty. The employees say they were lined up against a fence and patted down, that guns were drawn, that they were kept on the premises against their will for hours. They called it a “raid.” The police deny this. But the police concede that officers from the SWAT squad were present, decked out in SWAT gear, carrying weapons ranging from shotguns to Glock 21s.1
All of which is odd, because what was supposed to be happening that day was an “administrative” search. Administrative searches are humdrum affairs that occur on a regular basis in order to make sure businesses are complying with applicable regulations—like inspections of restaurants or liquor wholesalers. Or home inspections to ensure they meet the building and electrical codes. For an auto salvage place like Wholesale it’s a way of making sure the business is not a “chop shop,” trafficking in stolen vehicles and parts. Typically, a couple of plainclothes inspectors show up, check the records, maybe compare them with the cars on the lot, and unless something is seriously awry, they are soon gone. Judy Bass, the office manager at Wholesale, testified that they’d had exactly that happen in the past, that the officers were “very polite” and stayed for fifteen minutes. This time, though, the first cop into her office was in camouflage, armed, and not identifiable as a police officer; she was “deathly scared.”2
The police spent the entire day taking Wholesale Auto Advantage apart. Literally. Cars were sawed open, drawers and files were searched from top to bottom, then carted off. They hauled away seven pallets of goods, leaving basically nothing behind. Even Judy Bass’s calculator and a Shop-Vac were seized as evidence.3
Bruce had done nothing wrong, as he and Judy Bass tried to demonstrate to the officers both at the time and for days afterward. It’s true that some of the things the police encountered seemed odd. For example, they found a few vehicle identification number tags—the VIN you can see through a car windshield—that allegedly had been unlawfully removed from vehicles. But “seemed” is the operative word here: if the police had not been so hell bent on cracking the big one and listened—or even looked at the paperwork—they would have realized that. Salvage businesses have to take off VINs to do their work. As Bruce explained, “This is for a body shop. They did not raid a library or a car wash.” Florida law plainly allows for this; the cops evidently had insufficient familiarity with the very statutes they claimed to be enforcing.4
There’s no need to take Bruce’s word that he did nothing wrong. The Sheriff’s Office brought a forfeiture action against Bruce, seeking to keep the property it seized, including well over one hundred cars on the lot. The state judge disagreed, ordering almost everything returned to Bruce. The State’s Attorney quickly dismissed all of the charges.5
The cops basically ruined Bruce. The Sheriff’s Office ignored the court order to return all the goods, keeping what they felt they needed to prosecute Bruce—despite the State’s Attorney declining to do so. They held on to all the car titles they had taken, so that Bruce could sell nothing. Eventually he went out of business, busted.6
Bruce sued, and the federal appeals court judges were clearly irate. They believed the “massive show of force” at Wholesale was both inappropriate and unlawful for an administrative search. They sent the case back for a trial on whether the facts were as the employees alleged, and whether the defendants should be held liable.7
But the judges stumbled on the real question in Bruce’s case, which was not the way the police conducted the raid that January day, but whether it should have happened in the first place.
The police showed up at Bruce’s business that day because a few days earlier a young man named Zeeshan Mohammed Shaikh came to them to say that his insurance company had told him the VIN on his car’s dashboard—a 1985 Mustang he bought from Bill Bruce—did not match the “confidential” VIN. A confidential VIN is hidden inside vehicles precisely to deter stealing cars and selling them for parts; the public and confidential VINs are supposed to match.8
As it happened, though, the sale of the Mustang was perfectly legal; it’s a fine story, and had the police just knocked on Bruce’s door and asked instead of tearing his place apart, he would have told them. It turns out some kid had a 1985 Mustang that he totaled. He then stole a 1993 Mustang, and put the VIN from the original car on the stolen car. When he broke up with his girlfriend, she ratted him out, and he went to jail. Bruce bought the car at an auto auction; the car was sold as a “theft recovery” vehicle, with the VIN discrepancy noted on the documents. In fact, Florida had issued a new VIN for the car. This was all clearly indicated on the bill of sale: when Bruce resold the car he even made the buyer and his mom sign that they understood it was a theft recovery. Bruce had the papers (including one signed by a police official) documenting the Mustang’s status as a theft recovery.9
No one—not the cops and not the judges—believed that when the police stormed Bruce’s business that day they had probable cause to conduct a search for criminal violations. When Sheriff’s Deputy Randall Root was asked, “[C]ould [I] just call you up and say, I think I got a stolen vehicle and can you raid Joe Bob’s down the road,” he insisted, “No, no.” In Root’s view, Shaikh was not even “alleging that he had a stolen vehicle.” That’s why the cops never took a formal complaint from Shaikh, and never sought a warrant. And it is why the judges in Bruce’s case concluded that the information the police did have “did not rise to the level of probable cause that would have supported application for a warrant.”10
Rather, the police simply decided, based on Shaikh’s tale about what the insurance company had told him, that they would instead conduct an “administrative search” of Bruce’s business. Unlike a criminal investigation, under Florida law police can conduct administrative searches without any cause whatsoever, and without getting a warrant.11
So here’s the legal conundrum the judges could not solve. If the cops had enough evidence of criminal activity that it equaled probable cause, then they had to get a warrant. If they had no suspicion at all, they could conduct an administrative search without a warrant. But what if—as was the case here—they had a little suspicion, but not enough to equal probable cause? Was a warrantless administrative search acceptable then?
On first glance this question may seem crazy. One of the three judges hearing Bruce’s case, Edward Carnes, thought so. Writing alone, he chastised his other colleagues’ “hand-wringing” over the matter. After all, if no suspicion is fine, how could some suspicion be worse? “[T]his,” Carnes wrote, “is one area covered by Mae West’s observation that: ‘Too much of a good thing is wonderful.’”12
Carnes has a point, right?
THE PROBLEM OF SUSPICIONLESS SEARCHES
Wrong.
Despite the superficial logic of Judge Carnes’s position, things are not nearly so simple. After all—as we know—the police could not have gotten a warrant to search Bill Bruce’s salvage shop. Warrants require probable cause, and no one thought the police had probable cause. As we’ve seen, the Constitution’s requirements of probable cause and warrants are what keep police from doing what they like, when they like. And they prevent mistakes from happening.
If that’s true, should cops be able to get around the probable cause and warrant requirements by just calling what they are doing an “administrative search”?
That’s precisely what worried the other two judges in Bruce’s case. They didn’t believe it really was an administrative search. There was nothing regular about it: the cops came in SWAT gear and tore the place up. If police can call anything they want an administrative search, “the administrative search exception [could] be allowed to swallow whole the Fourth Amendment.” This, the judges felt, would “invest law enforcement with the power to invade the privacy of ordinary citizens.” It was “cause for general alarm.”13
Though the two judges were clearly nonplussed, in light of Supreme Court precedent they didn’t know what to do about the problem in the case before them. The Supreme Court had said administrative searches without any cause at all were fine. Ultimately, like Judge Carnes, they did not understand how having a little cause—though not enough to get a warrant—could eliminate the right to conduct an administrative search.
What all the judges in Bill Bruce’s case failed to consider is precisely why administrative searches are okay without any cause whatsoever. Or, indeed, why any search without cause is ever acceptable. Why it is not prohibited as one of the “general searches” condemned by the Framers of the Constitution. Had they done this, they would have seen the problem with the search of Wholesale that day.
Ponder this puzzle: The government “seizes” millions of people each year and requires them to go through “searches” at airport security if they want to board a plane. The consensus is that airport searches are perfectly constitutional. That they are not, in the words of the Fourth Amendment, “unreasonable searches and seizures.” Yet, why is that exactly? There’s no cause, probable or otherwise, to suspect any of these people of doing anything wrong. At first blush, this seems the very sort of general search that the Fourth Amendment was written to prohibit.
In a world in which policing has shifted from a focus on getting the bad guys through suspicion-based searches, to conducting surveillance of all of us as a way to make society safer, these sorts of suspicion-less searches have become pervasive. That fact forces us to confront the question: Under what conditions are these searches without any suspicion whatsoever permissible? Once we stop caring about cause—about any suspicion of wrongdoing—how can we tell which searches and seizures are barred by the Constitution’s prohibition on “unreasonable searches and seizures” and which are not?
This is one of the most significant problems of our day. If subjecting everyone who hops on a plane to a search without cause is okay, what about drug testing or DNA testing of all of us? Is that also fine? How about the National Security Administration grabbing all our phone records in the name of national security? What of roadblocks of all sorts, from drunk driving to auto safety checks? Or administrative searches of our homes and offices?
On questions such as these, the justices of the Supreme Court are deeply confused. But they are hardly alone. Civil libertarians, scholars, mayors and presidents, spymasters and police chiefs, all have found it difficult to figure out a principled way of distinguishing an “unreasonable” search from one that is reasonable, especially when they are contemplating a search on no suspicion at all. One cost of this confusion certainly has been an enormous loss of civil liberties: many searches are permitted that should never occur. But by the same token, the justices also have banned some valuable forms of policing as “unreasonable” when they plainly should be allowed.14
Things need not be this complicated. If one returns to first principles and simply recalls that the Fourth Amendment exists in large part to prevent arbitrary, unjustified, and discriminatory government searching, it becomes apparent that searches based on some cause, and entirely suspicionless searches, pose two completely different problems. And for that reason, as we will see, they admit of two very different solutions. What matters in the end is having in place protections to make sure that, in the absence of “probable cause,” suspicionless searches are neither arbitrary nor unjustified.
THE DEATH OF “CAUSE”
The justices had billed their 1968 decision in Terry v. Ohio—permitting stop-and-frisk on less than probable cause—as a “narrow” exception to the general rule. Over time, though, the justices allowed other, and larger, exceptions, until cause no longer mattered at all anymore. We’re not talking about lowering the level of suspicion somewhat to allow for things like stop-and-frisk, like we saw in the last chapter. We’re talking about getting rid of cause completely. Each exception on its own may have looked like a “reasonable” little nibble out of the rule requiring probable cause to search or seize people. But like mice with a piece of cheese, nibble by nibble the justices ate up the probable cause rule in its entirety.15
The initial damage was done in cases in the mid-1970s involving automobile stops to enforce the immigration laws. These cases followed Terry’s lead in recognizing that once the courts let go of probable cause, there is “no ready test for determining reasonableness other than by balancing” the government’s need to search against the individual’s interest in avoiding “the invasion which the search [or seizure] entails.” So balance the justices did, and on this basis they approved roadblocks more than fifty miles from the border—based on no cause at all. While “the potential interference with legitimate traffic is minimal,” the justices declared, “[r]outine checkpoint inquiries apprehend many smugglers and illegal aliens who succumb to the lure of such highways.” As we’ve seen time and again, this sort of balancing usually means goodbye to our rights.16
By the mid-1980s, the Court was making strong statements about the unimportance of cause. New Jersey v. TLO, decided in 1985, involved a school disciplinary investigation. (Because she was a juvenile, TLO’s full name was kept out of the records.) Based on a report that TLO was smoking in the bathroom, and faced with her denial, Assistant Vice Principal Theodore Choplick searched TLO’s purse and found marijuana, along with some evidence that the fourteen-year-old TLO was dealing. She ended up in Juvenile Court and challenged the constitutionality of the search. It’s a fair question whether school officials should have to play by precisely the same Fourth Amendment rules that govern policing. But instead of starting from the principle that probable cause and warrants are required, and explaining why school officials are an exception, the Court instead made the bold declaration that probable cause is not “an irreducible requirement of a valid search.” Rather, “the underlying command of the Fourth Amendment” is only “that searches and seizures be reasonable.”17
Then, in 2002, in the case of Board of Education v. Earls, the justices approved the drug testing of all high school students who participated in extracurricular activities, without any evidence to suspect any particular student of using. The justices were quite clear that being made to pee in a bottle with supervision by others, and then having your urine tested for the chemicals it contained, was unequivocally a “search” governed by the Fourth Amendment. But as for the probable cause usually required to justify such a search, the Court stated: “[W]e have long held that ‘the Fourth Amendment imposes no irreducible requirement of [individualized] suspicion.” It depends, one supposes, on what “long” means. It took hundreds of years to move from probable cause to something slightly less, in Terry. Yet, within decades of Terry the justices had become blasé about whether any cause at all was needed before government employees conducted searches.18
Suspicion of any sort has now been done away with in a wide variety of contexts, from roadblocks to administrative searches, to secret collection and searching of our data by the FBI and NSA, to DNA testing of people arrested for crimes as minor as engaging in antiwar protests. Effectively, we’ve all become suspects, whether we’ve given the government any reason to believe we’ve done anything wrong or not. Take Lindsay Earls, the plaintiff in the case challenging the drug testing of students involved in extracurricular activities. Earls was a top student who never took drugs, and was in the show choir, the marching band, the Academic Team, and the National Honor Society. She went to college at Dartmouth, and then on to law school. She was, by her own admission, a “Goody Two-Shoes.” She just didn’t believe government officials could randomly single her out, and humiliate her by making her urinate on command, all to satisfy their own cravings for control over a social problem she had never given any indication of having created or participated in.19
Despite the hand-waving balancing test in these cases, what the justices have never managed to do is explain coherently why some searches are acceptable without any cause whatsoever, while others are not. The justices had a chance to clear things up in a case coming from Indianapolis, Indiana, decided in 2000. Instead, they muffed things badly. And this time, they didn’t limit our rights; instead, they may have made us all less safe.
INDIANAPOLIS ROADBLOCKS
The Near Westside of Indianapolis in the mid-1990s was a tough place. The neighborhoods of Haughville, Hawthorne, and Stringtown were poor, and ridden with crime. The area, which was about 60 percent white and 40 percent black, had a median household income of just under $17,000. Only half of the adults possessed a high school education, and almost a third of the people lived below the poverty line. Drug trafficking was widespread and the gun violence that came with it all too common.20
Residents of the Near Westside were begging their elected officials to do something about the crime problem, and those officials were trying. The police chief and the mayor were looking for creative solutions.
Ultimately, the City of Indianapolis decided to experiment with drug interdiction roadblocks. In 1995, and then on a more widespread basis in 1998, the police set up checkpoints on roads leading into the Near Westside. Motorists were asked for their licenses and registration, while K-9s sniffed for signs of narcotics. If the canines alerted, the cars were pulled to the side for full searches.21
The roadblocks were announced in a press conference at which community leaders wearing hard hats backed up Mayor Stephen Goldsmith. Some 10 percent of people at the roadblocks turned out to be offenders. There were six checkpoints between August and November 1998, at which a bit more than one thousand cars were stopped, and just over one hundred people were arrested. Half of these were for drug crimes, and the rest for something else. (One man had a pit bull locked in his trunk.) Many had small amounts of drugs, but one of the first cars had more than two hundred pounds of marijuana in the trunk.22
Still, the main point of the roadblocks wasn’t catching people with drugs and weapons: it was to forestall the problem. It was precisely the sort of deterrent-based policing that is so popular today. As the mayor and police officials explained, given the publicity attendant on the roadblocks they might not catch many drug traffickers, but they would “deter people” from bringing in drugs in the first place. “It’s not just the statistics,” explained one police official, “it’s the message.” “It made a statement,” said another: “You’re not going to be totally safe driving around with drugs in the car, or with guns.”23
The roadblocks were popular with local residents, but they were challenged in court by the Indiana Civil Liberties Union, representing motorists unhappy that they were stopped. Neighborhood residents stepped up to defend the police, saying they’d asked for aggressive enforcement. “We are the ones who have to lay in bed and listen to the gunshots. We are the ones who can’t use our streets because the drug dealers are there. We like these roadblocks.”24
When the case—City of Indianapolis v. Edmond—got to the Supreme Court, the justices shut down the drug roadblocks. In doing so, one hoped they would offer an intelligible reason why some searches without cause were fine, and others were not. Or even why some roadblocks are fine and others are not. But they didn’t. Instead they handed down a decision that by wide consensus makes no sense at all—yet continues to reverberate through the law, imposing a significant limitation on what government is allowed to do.
THE POLICE CAN’T SEARCH IF THEIR “PRIMARY PURPOSE” IS LAW ENFORCEMENT?
Roadblocks like those used in Indianapolis are common enough in the United States. They are used to address a variety of problems, from unlawful immigration to drunk driving, to unsafe vehicles on the road. Whether one approves or disapproves, the question is whether such vehicle stops, without any cause to believe any driver has done something wrong, are lawful.
Under preexisting law, one would have thought Indianapolis’s roadblocks were perfectly kosher. A full decade before, in a 1990 case called Michigan Department of State Police v. Sitz, the Supreme Court had approved of drunk-driving roadblocks. In doing so, in Sitz, the justices set out a three-part analysis that was really just a variant on the “reasonableness” balancing test. First, did the government have an important problem it was addressing with the roadblock? Second, how much of an intrusion was the roadblock stop? Finally, was the program reasonably effective? In Sitz, the Supreme Court concluded that drunk-driving roadblocks easily met the test. “No one can seriously dispute the magnitude of the drunken driving problem,” wrote the Court, noting the “slight” intrusion the roadblock entailed. And while it was true that only 1.6 percent of motorists traveling through the checkpoint were arrested for driving under the influence, that—the justices said—was good enough.25
Surely Indianapolis’s drug interdiction roadblocks should have been upheld under this test. The justices in Edmond acknowledged that “traffic in illegal narcotics creates social harms of the first magnitude,” not the least of which was the gun violence rampant in the Near Westside. The searches in Indianapolis took a bit longer than in the Michigan roadblock case, but still were only a couple of minutes or so. As for effectiveness, the number of arrests that resulted from the Indianapolis checkpoints was more than twice as high as the Michigan roadblock if narcotics offenses alone were considered, and five times higher if all crimes were taken into account.26
So, what possibly could be wrong?
The failing of the Indiana roadblocks, the justices explained in Edmond, was that they were set up for the “primary purpose” of “crime control.” All the other roadblocks and general searches that they had approved, the justices said, had been used to address “special needs” of law enforcement other than ordinary criminal law enforcement. They had “never approved a checkpoint program whose primary purpose was to detect evidence of ordinary criminal wrongdoing.”27
What does this even mean? It is a very strange thing, to say the least, to condemn the police for conducting a search whose “primary purpose” is “crime control.” Isn’t that their job? And isn’t that also exactly what the police were doing in Michigan with their drunk-driving roadblocks?
How exactly does one distinguish a roadblock (or almost any other search or seizure) for “ordinary crime control” from one furthering “special needs of law enforcement”? After all, the defendants in both Sitz and Edmond were headed to jail for violating the law. And if the argument in Sitz was that the point of drunk-driving roadblocks was to deter drunk driving as much as catch criminals, well, exactly the same was true in Edmond, albeit to deter carrying drugs. Sending people to jail when we catch them is how we deter crime. The Supreme Court’s test is so confused that in a case right after Edmond, the justices got all tangled up in trying to distinguish an “ultimate goal” from an “immediate objective.”28
If you are confused at this point, you are hardly alone. In the wake of Edmond, the lower courts—and all the rest of us—are left completely clueless about what is permissible and what is not. Just look at what’s happened since their decision in 2000. In one case out of Tampa involving a driver’s license checkpoint, a cop explained, “If you get drugs that’s fine too, but basically it’s a driver’s license checkpoint.” The court did not agree. Yet, in another case DEA agents accompanied state officers from the Department of Consumer Protection on an “administrative” search of a pharmacy; this, the courts deemed okay.29
TWO OLD, ODD CASES
There’s more than a little irony in the Supreme Court’s utter confusion about how to tell a valid search without cause from an invalid one. That’s because the secret to unraveling this rampant confusion is found in two of the Supreme Court’s earliest cases involving searches without suspicion: Camara v. Municipal Court (a 1967 decision involving home inspections) and Delaware v. Prouse (a 1979 case involving highway stops). Unlike the Terry stop-and-frisk case, in which the justices used the idea of reasonableness to broaden police power, in Camara and Prouse the Fourth Amendment’s prohibition on “unreasonable searches and seizures” was used to limit what searching the government could do without probable cause—or at least the way in which the government went about conducting suspicionless searches.
At first the reasoning in these two cases is going to seem mighty peculiar. But once we revert to first principles, and remember what the Fourth Amendment was designed to protect against, not only will the reasoning make sense, but it will point the way to clearing up all the confusion that has occurred since, and to how we ought to regulate the new policing.
Camara’s Dilemma
In 1967, the year before Terry was decided, the justices confronted a dilemma involving housing inspectors. Some home owners were unwilling to allow government inspectors to enter their personal castles to check for unsafe wiring and the like without a warrant. But how could the inspectors get a warrant? After all, the Fourth Amendment says “no Warrants shall issue, but upon probable cause,” and in the typical housing inspection there is no probable cause of wrongdoing; usually there is no suspicion at all. Most housing inspections—like most safety or administrative inspections of businesses—simply involve inspectors subjecting owners to periodic inspections. You get inspected when it is your turn, not when someone thinks you’ve done something wrong. Just like Bill Bruce’s assistant Judy Bass testified happened on occasion at Wholesale Auto Supply.
In Camara the justices required warrants for housing inspections, but in doing so they were forced to redefine the notion of what it means to have “probable cause.” After all, the Fourth Amendment says no warrants without probable cause, and in the usual housing inspection there isn’t any. The validity of the search, the justices concluded, was to be decided by balancing the need for the housing inspection against the intrusion of the housing inspector. The former was great; the latter small. In light of this balance, there was “probable cause” (the quotes are the Court’s) whenever the inspectors can show a magistrate the need for a search of the homes in a given area based on some set of criteria such as “the passage of time” since the last inspection, “the nature of the building,” or “the conditions of the entire area.”30
Heretofore, probable cause meant individual suspicion that someone had done something wrong. Now it meant whenever the need outweighed the intrusion. This new definition of probable cause was quite odd, was it not?
Prouse: Misery Loves Company
Equally perplexing was the logic of the 1979 decision in Delaware v. Prouse. In Prouse, a cop stopped a car for a license and registration check. Approaching the car, he smelled pot, and then saw it in plain view. Prouse was arrested for illegal possession. The problem was the police officer had exactly zero reason to stop the car in the first place. “I saw the car in the area and wasn’t answering any complaints, so I decided to pull them off.”31
In deciding that this stop on less than probable cause—indeed no cause at all—was invalid, the Supreme Court used the familiar three-part balancing test. The justices were quick to concede that the state had a “vital interest” in making sure that drivers were properly licensed and cars were registered and safe to drive. But they were not so sanguine about the intrusion that these “random” stops entailed. Such a stop reflects an “unsettling show of authority,” it interferes with “freedom of movement,” and can be inconvenient and time-consuming. It can create “substantial anxiety.”32
What really seemed to tilt the Court toward throwing out Prouse’s conviction, though, was the third part of the balance: the justices doubted the efficacy of such stops. “It seems common sense that the percentage of all drivers on the road who are driving without a license is very small and that the number of licensed drivers who will be stopped in order to find one unlicensed operator will be large indeed.” Wouldn’t it be better, they asked, to stop people whose driving is erratic or whose vehicles seem to have a problem? Such stops, based on probable cause—actual suspicion—“occur countless times each day” and “it must be assumed,” absent evidence to the contrary, that these stops are far more effective in finding unlicensed drivers.33
For this reason, the stop at issue in Prouse was deemed invalid. “In terms of actually discovering unlicensed drivers or deterring them from driving, the spot check does not appear sufficiently productive to qualify as a reasonable law enforcement practice under the Fourth Amendment.”34
So far, so good. But it was just at this moment that the Prouse Court took its own odd turn, holding that its ruling “does not preclude the State of Delaware or other States from developing” alternatives, one of which is the “[q]uestioning of all oncoming traffic at roadblock-type stops.” If the problem with the spot check of motorists like Prouse himself was that it was too intrusive given how unlikely it was to turn up a violation, weren’t these roadblocks only going to be worse? Now even larger numbers of people would be inconvenienced, with the same small chance of finding violations.35
Justice Rehnquist, dissenting from the Court’s ruling, was caustic in pointing out the illogic of Prouse: “Because motorists, apparently like sheep, are much less likely to be ‘frightened’ or ‘annoyed’ when stopped en masse” a patrol officer can “stop all motorists on a particular thoroughfare, but he cannot … stop less than all” of them. (Coincidentally, the case immediately after Prouse in the published Supreme Court decisions is called Leo Sheep Company v. United States.) This, he concluded, “elevates the adage ‘misery loves company’ to a novel role in Fourth Amendment jurisprudence.”36
Rehnquist seems right: taken at face value, Prouse was as strange as Camara. In just the way that the state’s interest in conducting housing inspections hardly seemed to qualify as “probable cause,” how could stopping all these people provide more cause than stopping one?
BACK TO FIRST PRINCIPLES
The peculiarity of Prouse and Camara slips away, though, when one recalls that a primary purpose of the Fourth Amendment is avoiding arbitrary, unjustified police intrusions into our lives. It is about limiting the unbridled discretion of officers—or the government itself—to pick and choose who gets searched. The justices in Prouse reminded readers that “[t]he essential purpose of the proscriptions in the Fourth Amendment is … ‘to safeguard the privacy and security of individuals against arbitrary invasions.’” This could hardly have come as a surprise: the justices said the same thing ten years earlier in Camara—in exactly the same words.37
This concern for arbitrary enforcement goes all the way back to the two eighteenth-century incidents (described in Chapter 5) that led us to have a Fourth Amendment in the first place. Arguing against the Writs of Assistance in 1761, James Otis put his finger on the problem, stating that permitting general searches would allow “[e]very man prompted by revenge, ill-humor, or wantonness to inspect the inside of his neighbor’s house … one arbitrary exertion will provoke another.” Across the pond, condemning the general warrants used to search John Wilkes’s property, Chief Justice Pratt—later Lord Camden—condemned the “discretionary power given to messengers to search wherever their suspicions may chance to fall.” There’s virtually no disagreement that a, if not the, chief purpose of the Fourth Amendment is preventing arbitrary, unjustified searches. Judges and scholars of the Fourth Amendment have said the like, over and over and over, without dissent from the proposition.38
When it comes to figuring out whether a search without cause is permissible, the concern for arbitrary and discriminatory policing is what should matter—and not the Supreme Court’s impenetrable distinction between “ordinary law enforcement” and the “special needs of law enforcement.” As both the Prouse and Camara Courts made clear, the sine qua non of official arbitrariness is allowing officers unfettered “discretion” to invade our lives and our property whenever the whim strikes. The harm of a warrantless home search in Camara was to “leave the occupant subject to the discretion of the officer in the field.” Similarly, the spot check at issue in Prouse represented the “kind of standardless and unconstrained discretion” that is “the evil the Court has discerned when in previous cases it has insisted that the discretion of the official in the field be circumscribed, at least to some extent.”39
In light of the concern to avoid arbitrary enforcement, the seemingly bizarre solutions in Prouse and Camara all of the sudden make perfect sense.
In the ordinary case envisioned by those who wrote the Fourth Amendment, it is the twin requisites of probable cause and a warrant that provide the protection against standardless or arbitrary government action. Probable cause provides the reason for a particular search; it ensures that an officer is not arbitrarily or discriminatorily singling someone out. Then, the warrant ensures that the officer’s judgment as to cause is not biased by crime-fighting zeal (or anything else for that matter).
But because there is no suspicion—and thus no probable cause—in housing inspections and license and registration spot checks, the Supreme Court had to come up with a new means of ensuring that the government was not singling out people for no good reason. The Camara Court solved this problem by requiring housing inspectors to show some regular plan for choosing who got searched and who did not. The “cause” justifying a safety inspection was area-wide rather than individual. If an area was dilapidated, or had not been searched in a long time, or had a particularly tricky set of structures, this would be enough “cause” to justify the search—precisely because it avoided the possibility of the inspectors targeting someone out of thin suspicion, whim, or malevolence. (And, it is worth noting, the justices also required a warrant by a neutral magistrate, in order to confirm this.)40
The Prouse Court’s choice—subject obvious violators to traffic stops, or stop everyone at roadblocks—makes sense for the same reason. Yes, a checkpoint seems to be a greater intrusion for not much higher odds of finding violators. But it also ensures that no one is being arbitrarily, or discriminatorily, treated to the state’s use of force. “At traffic checkpoints the motorist can see that other vehicles are being stopped, he can see visible signs of the officers’ authority.” 41
TWO TYPES OF SEARCHES
Once we see that in Prouse and Camara the justices were putting in place safeguards to prevent arbitrary policing (and its cousins: unjustified and discriminatory policing), we can see that in truth there are two very different kinds of searches, each requiring its own sort of protection: searches that are suspicion-based, and those that are suspicion-less.
Television shapes the typical image of what the police do. In most television shows, the police are after a particular suspect for a particular crime. Police are investigating a violation of the law that has either occurred or is about to occur. The police may know the suspect, but in many cases they are trying to track him (or her) down based on some leads. Think here of Brenda Lee Johnson in The Closer, or Jack Webb in Dragnet, or CSI: Crime Scene Investigation in its many incarnations. In doing their job, the police engage in all kinds of things we might call searches, from looking for information on the Internet or in a police database, to taking fingerprint or DNA samples, to busting down the door of a suspect’s house (hopefully with a warrant).42
The distinguishing characteristic of this sort of investigative search is that it is suspicion-based. The police have cause to believe that some particular person, known or unknown, has committed or is about to commit some specific crime. And the police are trying to discover the facts and put the culprit away.
But in much of the policing we’ve been discussing in this chapter, the police are doing something entirely different. They aren’t investigating to catch a particular suspect; they are for the most part trying to keep there from being a suspect in the first place. That’s what’s going on with the searches in airport security. It is not that we put the TSA officers there to look for a bomb in a suitcase (though assuredly they are—and if they find one the person is surely going to prison). Instead, we spend an untold fortune on such elaborate security in the hope that people will simply leave their agents of destruction at home. The same sort of thing is going on with drunk-driving roadblocks. Yes, if a drunk wanders through then the police are going to get them off the road. But the real reason for the roadblocks—and the publicity that always accompanies them—is to discourage people from getting behind the wheel of a car in the first place. (Studies suggest this strategy is successful.)43
What distinguishes this second sort of search is that there is no cause whatsoever. Because the purpose of the search is deterrence—trying to get people to behave in the first place—there is no suspect and thus no suspicion. A wide swath of society is subjected to policing in order to make people think twice before violating the law.44
Suspicion-less v. suspicion-based searches. It doesn’t matter whether the police are searching for “ordinary law enforcement” or “special needs.” What matters is whether they have suspicion.
TWO TYPES OF PROTECTIONS
Once we see that there are two very different types of searches, it becomes equally apparent we need two different sorts of protections against government arbitrariness. What works for one simply will not work for the other. That’s what the Supreme Court understood when it decided Camara and Prouse, but then seems to have forgotten when it went in a whole new direction with its unworkable approach in the Edmond case involving the Indianapolis roadblocks.
Avoiding Arbitrariness I: Cause-Based Searches
When the police search with suspicion, the protection is obvious. The Fourth Amendment tells us exactly what is needed: probable cause and a warrant. Probable cause ensures there is enough suspicion to single someone out for a search; the warrant confirms the police have it right. And if a search is suspicion-based, but there is inadequate suspicion, then we have the danger of arbitrariness the Fourth Amendment was written to guard against.45
The problem of failing to distinguish suspicion-based v. suspicionless searches is seen in a Supreme Court case that looked remarkably like the one that opened this chapter, only the junkyard owner in this case was named Burger, not Bruce. Burger owned an auto junkyard in New York City. One day the administrative inspectors, who also happened to be police officers, showed up at Burger’s junkyard ostensibly conducting an administrative search. They asked to see his license and the book in which he was supposed to log all the cars and parts on the lot. Burger straight off told the police he had neither: he admitted he was operating illegally. At that point the police searched the junkyard, and Burger ultimately was convicted for dealing in stolen vehicles.46
On appeal, Burger complained (among other things) that the police didn’t have a warrant to search his junkyard. The Supreme Court responded, in a complete non sequitur, that warrantless administrative searches were necessary to achieve an element of surprise. (After all, search warrants are executed all the time by surprise.)47
The justices realized, though, that absent a warrant some “constitutionally adequate substitute” was necessary. They identified three that in their view made everything fine. First, New York law only allowed searches during daylight. Second, it put licensed junkyard owners on notice that they could be searched. And finally, it limited the scope of the search: inspectors could only look at records and vehicles.48
Note how none of those supposed “safeguards” mentioned by the justices is designed in any way to protect against the evil underlying the Fourth Amendment: the police using their unlimited discretion to arbitrarily single out whom to search. Not one of those so-called protections avoids the police picking on someone for the wrong sort of reason or no reason at all.
There’s a telling footnote at the start of the Burger decision that tips readers off to the real problem here. “It was unclear from the record why, on that particular day, Burger’s junkyard was selected for inspection.” That turns out to be a darn good question. The justices simply shrugged it off, saying, “The junkyards designated for inspection apparently were selected from a list of such businesses compiled by New York City police detectives.” But that can’t possibly be correct. After all, recall that when the cops showed up at Burger’s junkyard, he told them right off that he wasn’t licensed. Without a license, he wouldn’t be on any official list of licensees subject to administrative searches. Right?49
Aha. Now we can pretty much guess why the police showed up at Burger’s junkyard that day: they must have gotten a tip of some sort—like Shaikh’s story to the Orange County cops about Bill Bruce and the stolen Mustang—that made them want to search Burger’s junkyard. The search of Burger’s junkyard—just like Bruce’s—was a suspicion-based search. It was not a suspicionless administrative search.
What we don’t know, and will never know, is whether the suspicion that the police had in Burger’s case added up to probable cause. If there was probable cause in Burger’s case, then a warrant would have made all fine. But if there was not probable cause, we are back to our old evil: the danger of arbitrarily picking and choosing whom to search.
You might respond, “So what, they got the bad guy, what could be wrong with that?” But this is precisely the problem we dealt with last chapter. If the police were always this good, then maybe we would decide to live with their supernatural powers of ferreting out crime. Of course, if they were that good, and honest about it, we would not need to worry about unjustified, or discriminatory, or arbitrary searching in the first place.
The problem is that, as we saw, the police are often wrong about their hunches, their searches on insufficient cause. Justice Jackson made precisely this point in the Brinegar case that defines probable cause: “There may be, and I am convinced that there are, many unlawful searches of homes and automobiles of innocent people which turn up nothing incriminating, in which no arrest is made, about which courts do nothing, and about which we never hear.” That’s what we have seen time and again: police searches on less than probable cause are often altogether wrong. And when the police are wrong, then innocent people—like Bill Bruce—pay the price.50
The Constitution is completely clear about the protections for cause-based searching. Warrants and probable cause. The Supreme Court watered these protections down when it allowed stop-and-frisk on “reasonable suspicion.” That alone was problematic—and in the last chapter we explored the fix for that particular problem. But in allowing searches of places like Burger’s junkyard on some undefined suspicion, without any protection against arbitrariness, the justices have gone seriously awry.
Avoiding Arbitrariness II: Suspicionless Searches
What, then, to do about suspicion-less searches? If the search of Burger’s or Bruce’s junkyards had been a true regulatory or deterrent search, it would have made zero sense to require probable cause. There would not have been any. We don’t have suspicion that any given person in line at airport security has a weapon, or that any driver at a drunk-driving roadblock was drinking.
Some people would prohibit suspicionless searches altogether, viewing them the sort of “general searches” the Framers abhorred. But is it really true we can’t have airport security or sobriety roadblocks or bulk data collection, for that matter? That seems to be cutting off our noses to spite our faces.51
The alternative is to have some protection—other than cause—to avoid arbitrary, unjustified, discriminatory searches. And at this point we know exactly what that other protection is.
The answer is you subject everyone to the same treatment. In that way the risk of arbitrary, discriminatory searches disappears. That’s what the Prouse Court was saying about roadblocks, and that’s pretty much what happens (or should happen) at airports.52
Of course, searching everyone can get prohibitively expensive, but there is yet another option available in many circumstances: selecting who gets searched in a truly random way. In Prouse, Justice Harry Blackmun joined in the Court’s decision but wrote separately to point out it was not necessary to stop every car at a checkpoint. It would be perfectly fine to stop every fifth car or every tenth car. If people knew that the police were using roadblocks, this would serve as enough of a deterrent. And from there what matters is only that the police are not picking and choosing based on some possibly illegitimate basis. Random selection does that.53
While randomness can help bring down the cost of a regulatory search, and is often appropriate, it is at least worth making the pitch for searching everyone, at least on some occasions. Very broad searches can help solve the problem of knowing whether a suspicionless search is worth the bother. Think about it. With suspicion-based searches we get some measure of what works by looking at the “hit rate”: the number of times the police come up with the goods. But (and this is a point the Supreme Court has altogether missed) in suspicion-less searches, low hit rates are more likely, and perhaps better, than high ones. That is because the purpose of these suspicionless searches is deterrence. We are trying to keep people from violating the law, so in a successful use of such searches we may have a very low hit rate—i.e., it works. But then, how do we know whether inconveniencing people with this sort of suspicionless search is worth it? It turns out that inconveniencing a lot of people actually helps. Misery does indeed love company.54
First, if the government is forced to search everyone, and bear the cost of doing so, we will know the government truly believes this form of deterrence is worth the cost. Think about airport security, which is unfathomably expensive. The TSA has a huge budget, money that could be spent on many other antiterrorism approaches. Budgets are not unlimited; choices must be made about what deterrent measures are sufficiently efficacious to make them valuable. The public does not always have a basis for knowing what’s worth it or not, but the fact that the government is forced to make these choices provides some guarantee that it has at least considered alternatives and decided this particular intrusion into everyone’s life makes sense.
Second, if enough people are searched, the political process itself will serve as a safeguard of our liberty. The more people who are subjected to police searching of this sort, and put up with it, the more we can be comfortable knowing that people at least believe the effort is appropriate and worthwhile.55
Face it—going through airport security is a pain in the keister. But while people may grumble about the lines, there is no widespread call to stop the searches. The public is persuaded the inconvenience is worth it to avoid airborne terrorism. Note, though, that when the TSA started to use X-ray machines that were too revealing of people’s bodies, there was an immediate outcry and the practice was stopped. The public deemed the intrusion too great given the payoff. With a truly general search, the decision about legitimate government interests and minimal intrusions rests where it should: with the public.56
Indeed, consistent with our notion of democratic policing, these sorts of searches should be approved in advance by the public anyway. And with these suspicionless searches, the government has little argument about the need for keeping them secret. The whole point of the searches is deterrence; people knowing only helps the plan succeed.
So, there you have it. By asking a very simple question, the law governing searches and seizures can be simplified greatly. When examining the constitutionality of police tactics, the first question should be: is it suspicion-based or suspicion-less? If the former, then the protections of sufficient cause (usually probable cause) and warrants are required. If the latter, then we need to ensure that the search is conducted in a way that avoids arbitrariness—usually by making sure everyone is searched.
GETTING IT RIGHT
Now, finally, the answer to the judges’ confusion in Bill Bruce’s case should be obvious. If the police had sufficient suspicion to search Wholesale Auto Salvage, which they did not, then they should have gotten a warrant based on probable cause. On the other hand, if it was truly an administrative search, it should have been random, which it also was not. What the police did at Wholesale Auto that day violated Bill Bruce’s constitutional rights.
Either a search is suspicion-based, and there should be adequate suspicion (plus a warrant when possible). Or a search is suspicionless, and then must be universal, or at least truly random.
The failure to recognize that cause-based and suspicionless searches require very different safeguards against arbitrariness has led the Supreme Court to exacerbate two deeply regrettable problems.
First, allowing suspicion-based searches on something less than probable cause leads to the arbitrary treatment of people like Bill Bruce.
Second, prohibiting suspicionless searches, even for “ordinary law enforcement” (so long as the search is truly general or random), invalidates some searches, like the roadblocks in Indianapolis, that may well be both valuable and appropriate.
The scheme of drug interdiction roadblocks in Edmond was set up to avoid officer arbitrariness and discretion. The affidavits in the case explained that the checkpoint locations were chosen by supervisors “weeks in advance,” “taking into consideration area crime statistics and the ability to locate the checkpoints in a location which will minimize the interference with normal traffic flow.” Cars were stopped in predetermined groups (i.e., five cars at a time), and even if police or fire department vehicles were in that group they were stopped. Most important, the officers were instructed that “every vehicle stopped must be examined in the same manner until particularized suspicion or probable cause develops.” It was emphasized that there was “no discretion given to any officer” in terms of which cars were stopped.57
Whether such roadblocks are a good or bad idea is for each community to decide, but it is hard to see them as unconstitutional.
We’ve lost sight of the reason the Fourth Amendment exists in the first place, and having done so we’ve become confused about what police agencies may and may not do. Yet it is not complicated. Suspicion-based searches should require probable cause. And suspicionless searches need regularity, i.e., subjecting everyone or at least some random set to the search, and avoiding any arbitrariness or discrimination, any whim or caprice.
Two types of searches, two very different protections. In this quite simple way, we can reconcile the Framers’ concerns about general searches with the imperatives of the new deterrent-based policing.
There still is one last, important problem we need to solve. Often suspicionless searches are aimed not at everyone, but at a particular group. The next chapter tackles that problem, by turning to another part of the Constitution, the Equal Protection Clause.