The Fourth Amendment in 2020
For the Constitution in 2020 conference on The Future of Criminal Justice.
Over the past 200 years, the Fourth Amendment’s guarantees have been construed largely in the context of what might be called “physical searches”—entry into a house or car, a stop and frisk of a person on the street, or rifling through a person’s private papers. But today, with the introduction of devices that can see through walls and clothes, monitor public thoroughfares twenty-four hours a day, and access millions of records in seconds, police are relying much more heavily on what might be called “virtual searches,” investigative techniques that do not require physical access to premises, people, papers or effects and that can often be carried out covertly from far away. This technological revolution is well on its way to drastically altering the way police go about looking for evidence of crime. The Supreme Court’s interpretation of the Fourth Amendment has both failed to anticipate this revolution and continued to ignore it.
Most importantly, the Court’s caselaw has made clear that, outside of interceptions of communications content, a Fourth Amendment “search” usually occurs only when there is a physical intrusion of some sort into the suspect’s property. That means that as a constitutional matter neither a warrant nor any level of suspicion is needed to justify most virtual searches. The one exception to this rule other that electronic surveillance, announced in Kyllo v. United States, is when a virtual device, like a $10,000 thermal imager, is aimed at the home interior. Even then no search occurs if the device is in “general public use” (binoculars, nightscopes, $2000 thermal imagers?), or spies what an officer could see from a lawful vantage point or detects only contraband. Thus, without worrying about the Fourth Amendment, police can watch all of our out-of-home activities and many of in-home ones, and also obtain any records of our financial, communication, and other daily transactions that are maintained by third parties.
The Court’s treatment of government actions that it says are Fourth Amendment searches has also vastly enhanced government’s ability to use technology to discover information about its subjects. In particular, the Court’s expansion of what it calls the “special needs” exception (which does away with the warrant and probable cause requirements for a wide array of regulatory and group-based searches) has opened the door to profligate use of technological surveillance. As long as search and seizure programs are not designed to detect “ordinary criminal wrongdoing,” they need merely be “reasonable” attempts at dealing with a “significant” law enforcement problem. This formula apparently exempts from traditional Fourth Amendment requirements not only most administrative inspections carried out by non-police but also technologically-enhanced anti-terrorism programs like data mining (assuming data mining is a search in the first place, which it probably isn’t).
A Different Fourth Amendment?
If reform of the Fourth Amendment were thought to be important as a means of responding to technological developments, the most obvious first step would be to conform the definition of search to its lay meaning of looking into, over or through something in order to find somebody or something. This move, which Justice Scalia himself contemplated in Kyllo, would immediately encompass virtual searches within the ambit of the Fourth Amendment’s protections. Camera surveillance, tracking, targeting places or people with any type of device (regardless of whether it is in general public use or contraband-specific), and accessing records via computer (or in any other manner) all involve searches under this definition.
Reform could not stop there, however. Current Fourth Amendment law also usually requires probable cause for a search. If police attempts to watch a person walk down the street, follow a car on the public highway, or peruse court records or utility bills all required probable cause, law enforcement would come to a screeching halt. It may have been to avoid such a disaster that most members of the Court, including many of its liberal members, have been willing to declare that these investigative techniques are immune from constitutional review.
But there is a compromise position, suggested by the Fourth Amendment itself. After all, the text of that amendment only requires that searches and seizures be “reasonable.” It does not require probable cause or any other particular quantum of suspicion.
I have argued elsewhere that the Fourth Amendment’s reasonableness inquiry should adhere to a proportionality principle (Privacy at Risk: The New Government Surveillance and the Fourth Amendment (2007), http://ssrn.com/abstract=1026614). The idea of calibrating the justification for an action by reference to its impact on the affected party permeates most other areas of the legal system. For instance, at the adjudication stage the law assigns increasingly heavier burdens of proof (preponderance, clear and convincing, proof beyond a reasonable doubt) as the consequences of the adjudication grow harsher. Equal protection law is famous for its three tiers of scrutiny depending on the interest involved. The proportionality principle has even found its way into the Supreme Court’s Fourth Amendment caselaw. It provides the best explanation, for example, for why arrests and searches of houses require probable cause, while stops and frisks only require reasonable suspicion. As the Court stated in Terry v. Ohio, the case that established this particular hierarchy, “there can be ‘no ready test for determining reasonableness other than by balancing the need to search against the invasion which the search entails.’” Unfortunately, the Court has applied this principle only haphazardly and, when it does apply it, inconsistently.
A more formal adoption of the proportionality principle would state that, for every government action that implicates the Fourth Amendment, government must demonstrate cause to search—meaning a level of certainty that evidence of wrongdoing will be found—roughly proportionate to the intrusiveness of the search. Given the history of the Fourth Amendment, the baseline rule for application of the proportionality principle would be that searches of houses and similarly intrusive actions require probable cause. But less intrusive searches and seizures could be authorized on something less than probable cause. For instance, the Court is clearly correct in its intuition that police viewing of public activities are generally less intrusive than police entries into houses. Camera surveillance and tracking of public movements, use of binoculars to look through a picture window, or perusal of a record of an individual’s food purchases would not require probable cause under proportionality reasoning.
In contrast to the Supreme Court’s jurisprudence, however, only the most minimal intrusions would be exempt from Fourth Amendment regulation. Thus, while randomly surveying the public streets with a camera might be untouched by the Fourth Amendment, using cameras or tracking devices to target an individual would trigger its guarantees (albeit perhaps only in the sense that an articulable reason for the targeting would be required). In further contrast to the Supreme Court’s approach, proportionality reasoning dictates that virtual searches which are as invasive as entry into the home would be permitted only upon a high degree of suspicion. For instance, if empirical research on lay views is to be believed (see Slobogin, Government Data Mining and the Fourth Amendment, 75 U.Chi. L. Rev. 317 (2008), http://ssrn.com/abstract=1001972) society associates bank and phone records with a high degree of privacy. Thus, contrary to Court precedent, police would have to demonstrate reasonable suspicion or perhaps even probable cause before gaining access to such information.
At least two exceptions to the proportionality principle should be recognized, however. When the purpose of a search is to prevent significant, specific, and imminent danger (think: Jack Bauer), society’s interest in protecting itself is sufficiently strong that the justification normally required by the proportionality principle should be relaxed. And when a search or seizure is of large numbers of people—when, in other words, search and seizure programs are involved—the usual Fourth Amendment paradigm focusing on individualized suspicion does not work well. In Government Dragnets (Law & Contemp. Probs., forthcoming; http://ssrn.com/abstract=1640108), I argue that, unless the dragnet involves a full-scale search of homes or mass arrests, this situation should initially be governed by political process theory. As laid out by Richard Worf, political process theory would permit a group search or seizure if it is authorized by legislation that requires neutral or random selection criteria and avoids targeting groups with insufficient access to the political process.
When these requirements are not met (which has often been the case with the group search and seizure cases decided by the Supreme Court) the proportionality principle should be adapted to group settings. Unless the intrusion occasioned by these blunderbuss intrusions is de minimis, proportionality reasoning would require some concrete justification beyond the type of broad pronouncements about “law enforcement problems” on which the Court usually relies. More specifically, proportionality analysis in the group context could require what might be called “generalized suspicion” (to be contrasted with individualized suspicion). Generalized suspicion can be thought of as a measure of a program’s success or “hit rate,” which under proportionality analysis must match its intrusiveness. For instance, in Edmond v. Indianapolis, a roadblock case, police searches produced evidence of drug crime in 5% of the cars stopped. Whether that potential hit rate would be sufficient to justify the intrusion associated with a roadblock would depend on how that intrusion compares to other police actions, such as arrests, that require probable cause (which might require a hit rate of about 50%), and field investigation stops that require reasonable suspicion (which has been quantified at around 30%).
A requirement of generalized suspicion proportionate to the intrusion visited on individuals in the group would force the government to produce concrete justification for its search and seizure programs. Assessment of hit rates might have to be speculative if a particular type of group search or seizure has never been attempted. But presumably a program instituted in good faith is motivated by the perception that a significant crime problem exists. In the absence of such facts (and assuming the danger exception does not apply) courts applying proportionality analysis would be leery of finding that a group investigation is justified.
This is an extremely brief summary of a book and two articles. I’ll be glad to expand on anything that occasions comment.
Christopher Slobogin is Milton Underwood Professor of Law, Vanderbilt University Law School. You can reach him by e-mail at christopher.slobogin at vanderbilt.edu.