After Chatrie, the Real Fight Over Digital Privacy Is Just Beginning

In Chatrie v. United States, the Supreme Court held that when police access a cell phone user’s “location history” they have conducted a “search” that must be constitutionally justified under the Fourth Amendment. This is so, the Court ruled, even when the police obtain that information from a third-party tech company (in this case, Google), rather than from the cell phone user directly. Chatrie has been viewed as a victory for the Constitution, and in many respects it is. But whether Chatrie ushers in, as some predict, an era of greater digital privacy is far from certain. The history of Fourth Amendment jurisprudence gives reason to be concerned that, unless legislatures fill the gaps left by the Chatrie decision, legal and privacy observers may one day look back at Chatrie as a high-water mark and find that Google did a better job than federal courts when it comes to limiting law enforcement access to our digital data.

The Chatrie decision stems from the police investigation of a 2019 bank robbery in Virginia. According to court opinions, police were unable to identify the man seen robbing the bank but could see from surveillance cameras that he was talking on a cell phone just prior to the robbery. Police turned to a tactic that law enforcement was using with increasing frequency: asking Google to produce location history for all users within a specified geographic area (the “geofence”) during a specific time period. Law enforcement obtained a warrant from a judge for this data, not because a court required it, but because Google insisted on it. Google also had a practice of routinely refusing to respond to warrants it unilaterally determined were too broad. Pursuant to a three-step process dictated by Google, Google first searched the accounts of more than 500 million Location History users and determined that 19 accounts appeared within the 150 meter geofence during the requested one-hour time period; Google then produced more information about these accounts over a two-hour time period and ultimately provided subscriber information for three accounts. Two of the accounts belonged to people who apparently had nothing to do with the bank robbery. One of the accounts belonged to 27-year-old Okello Chatrie, who was subsequently prosecuted and convicted for the bank robbery.

Chatrie is a victory for constitutional protection from undue government intrusion because if the Supreme Court had gone the other way—holding that the geofence warrant at issue in Chatrie did not implicate the Fourth Amendment—this and similar “reverse warrants,” in which police demand data from tower dumps, keyword searches, and the like, would not be subject to federal constitutional constraints at all. Relying on tech companies and other businesses to regulate (or not) government access to information about our lives that most would agree with the Supreme Court is “indisputably private” seems untenable. Companies like Google, after all, are in the business of gaining access to private information and then exploiting that access for profit.

As the Supreme Court has reiterated more than once, information obtained through reverse warrants can reveal visits to “the psychiatrist, the plastic surgeon, the abortion clinic, the AIDS treatment center, the strip club, the criminal defense attorney, or the by-the-hour motel.” Particularly combined with tools like automated license plate readers (ALPRs) and ubiquitous CCTV cameras, government actors could use location history to track the comings and goings of millions of Americans for any reason whatsoever, with no constitutional regulation. The Supreme Court has repeatedly acknowledged that this sort of tracking over time provides “an intimate window into a person’s life, revealing not only [one’s] particular movements, but through them, one’s “familial, political, professional, religious, and sexual associations.” The Court’s determination in Chatrie that accessing location history is a “search” thus creates the potential for meaningful judicial constraints on the creation of a government “virtual panopticon with which to scrutinize its citizens’ activities,” that justices across the political spectrum have feared.

Chatrie can also be seen as a victory in that the reasoning underlying the majority’s decision further chipped away at the “third-party doctrine,” which is one of those reality-defying Supreme Court-created rules that essentially says that if you share private information with another party, even for limited, agreed-upon purposes, you have lost any claim to constitutional protection if that third party hands your information over to the police. The Court previously signaled a sea change in its application of the third-party doctrine in the context of cell phone data searches in the 2017 case of Carpenter v. United States, but it went further in Chatrie. Indeed, Justice Neil Gorsuch, concurring with Chatrie’s majority opinion, suggested that Chatrie might be the Court’s attempt to give the third-party doctrine a “quiet burial,” at least as applied to information shared over smartphones.

None of this, however, guarantees meaningful constraints on police use of digital dragnets in the future. It remains to be seen whether the Chatrie decision will result in real constraints on government even in the Chatrie case itself, much less in future cases.

The Supreme Court decided in Chatrie that the government accessing location data is a Fourth Amendment “search,” but that just means the intrusion must be “reasonable” in order to pass constitutional muster. Searches supported by a valid warrant are presumptively reasonable, and the Court sent Chatrie back to the lower court to decide whether the geofence warrant there “satisfied the Fourth Amendment requirements of particularity and probable cause” and was therefore valid. It will be interesting to see how the U.S. Court of Appeals for the Fourth Circuit decides the issue. On the one hand, it’s difficult, as a matter of logic, to see how this warrant, or perhaps any geofence warrant, can meet the probable cause requirement that there is a “fair probability that contraband or evidence of a crime will be found” in the place searched, given the millions of accounts that were accessed, and the increasing likelihood that people robbing banks will turn off their location-tracking apps, or leave their phones at home. The particularity requirement that the warrant provide “cause to believe that the specific things to be searched for will be in the targeted place,” raises similar concerns, given the “dragnet” aspect inherent to reverse warrants. On the other hand, the Supreme Court has often stated that “the touchstone of the Fourth Amendment is reasonableness” and has approved (albeit not in criminal investigations) the widespread use of “administrative warrants” that do not meet the probable cause or particularity standards.

As importantly, reverse warrants and similar tools like ALPRs, facial recognition, and AI tools that detect unusual activity, all use an everything-everywhere, probable-cause-defying approach to identify wrongdoers ranging from kidnappers and child pornographers, to people committing financial crimes, to the Jan. 6 rioters. To be sure, these investigative tools are used far more frequently to investigate far less serious crimes as part of the ongoing “war on drugs,” or the newer “war on immigrants.” Still, it seems highly likely that, when the dust settles, courts will allow police to use, at least to some extent, this set of tools they have found so useful in investigations of serious crimes.

In other words, now is when things get dicey. The Supreme Court decided in Chatrie that geofence warrants trigger Fourth Amendment protection. But it will be hundreds, or even thousands, of judges presiding over criminal cases who shape the contours of that constitutional protection in coming years. And they will do so in a context that structurally discounts the importance of the constitutional protections at stake, perhaps even to the extent that the courts may create a reverse warrant legal regime that is less protective of digital privacy rights than was Google.

The context in which the vast majority of judges will consider reverse warrants differs significantly from the context in which they were considered by Google. When Google set its demands for handing over data to law enforcement, the company was considering whether to, on the one hand, repeatedly disclose to law enforcement digital information about hundreds of millions of its customers who had done nothing wrong, including the identities of some, in order to, on the other hand, provide law enforcement with information that may or may not be important to solve a crime that may or may not be serious. Further, it would have been fair for Google to believe that in many instances the information law enforcement sought was not particularly important: Of 3,000 warrants Google objected to as overbroad, law enforcement did not follow up on 2,500 of them. It was in this context that Google insisted that law enforcement obtain warrants signed by a judge and set up the three-step process described above, despite any clear constitutional requirement for judicial oversight or any constitutional limitations on officer access to the information.

In contrast, after Chatrie, the parameters of the constitutional right to digital privacy will be set almost entirely by judges who are deciding that question at a much later stage of the criminal investigation—the stage where the search was conducted, evidence of criminal conduct was found, and the person being prosecuted for a crime is trying to get that evidence thrown out. These judges will be deciding whether to apply the “exclusionary rule,” another Supreme Court-created doctrine that prohibits (albeit with ample exceptions) the government from admitting evidence against a criminal defendant where that evidence was obtained in violation of the defendant’s constitutional rights. The evidence can be admitted only if the judge decides the search was “reasonable,” which has come to mean essentially that the privacy or property intrusion is outweighed by a valid government interest. And in every case in which the judge is deciding whether the search was reasonable, the government search turned up evidence of a crime. Further, the judge is legally constrained to consider only the privacy interests of the criminal defendant before the court—not the rights of others whose rights may have been infringed by the search. Nor, in deciding the Fourth Amendment reasonableness of the search, is the judge permitted to consider any discriminatory animus on the part of the government actor. Finding a search unreasonable will mean, in the oft-quoted words of then-Judge Benjamin N. Cardozo, that “the criminal is to go free because the constable has blundered.” Of course, as Justice William J. Brennan Jr. wrote, “[S]ome criminals will go free not ... because the constable has blundered but rather because official compliance with the Fourth Amendment requirements makes it more difficult to catch criminals,” but those words do not always land well with (mostly elected) judges deciding criminal cases. 

Judges can readily articulate the dangers of viewing Fourth Amendment protections through the lens of a criminal prosecution, and they of course intellectually understand that the Fourth Amendment protects everyone, including but not only criminal defendants. But hearing search-reasonableness arguments in scores, hundreds, maybe thousands of criminal cases in which the search produced evidence of a crime can make it difficult for a judge to give full weight to the broader impact their Fourth Amendment decisions have, as their worldview collapses into what they see before them every day. Thus it is not uncommon to hear judges refer to the plaintiff in a civil rights suit against police as a “defendant,” as I heard a D.C. Circuit Court judge do repeatedly in a case argued by a Georgetown Law Civil Rights Clinic student this past spring. Nor is it uncommon for judges to frame Fourth Amendment protections as shields used primarily to protect criminal defendants, as did Justice Samuel Alito in Chatrie, when he described the Court’s limitation of the third-party doctrine as giving “criminal suspects a ‘protected Fourth Amendment interest in any sensitive personal information about them that is collected and owned by third parties.’” The geofence warrants at issue in Chatrie underscore how this dynamic can play out even at the early warrant stage in an investigation. Google challenged 3,000 warrants that judges had signed, meaning that some judges were quicker to approve warrants than was Google, even though it is the job of a judge to scrutinize a warrant’s validity to ensure constitutional rights are protected, while Google has no reason to have any interest in protecting privacy beyond what is necessary to further its business model. The geofence warrant experience thus underscores how the context in which judges routinely address questions regarding constitutional rights shapes their views of those rights, and thus the contours of those rights more broadly. 

Sixty-five years of exclusionary rule jurisprudence in a host of Fourth and Fifth Amendment contexts, including surveillance, tells us how this dynamic plays out over time and across jurisdictions: The breadth of the constitutional right will be narrowed, and exceptions to the application of the exclusionary rule—the primary remedy for violations of the constitutional right—will be expanded. Regardless of how the Chatrie warrant fares in that case, these exclusionary-rule-context dynamics make it likely that judges will slowly (or not) but surely whittle down the Fourth Amendment protections afforded to digital privacy in order to facilitate the successful prosecution of an accused murderer here, an alleged insurrectionist there. While some judges will be more protective of the Fourth Amendment than others, judges in general will tend to defer to law enforcement, even on determinations of constitutional law. Reviewing courts, for their part, will tend to defer to the judges who decided the reasonableness of the search in the first instance. It is easy to see how this happens. The urgency of accountability for a serious crime is undeniable, while the slow demise of personal autonomy can be nearly undetectable. 

This is why, in the wake of Chatrie, advocates for digital privacy have renewed, rather than retracted, calls for legislative regulation of geofence warrants and similar law enforcement tools. Those who have been focused on this issue know that legislation at the state and federal levels can be effective at filling the gaps left by broad constitutional pronouncements like the one in Chatrie, allowing for regulation that is both more comprehensive and more nuanced. Specific requirements in statutes not only create an additional avenue of protection—they also anchor and set expectations for how courts should decide the broad and subjective standard of “reasonableness.” Indeed, Congress has in the past responded with legislation to Supreme Court decisions about the impact of technology on Fourth Amendment rights. Title III of the Omnibus Crime Control and Safe Streets Act of 1968 stepped into the breach left by broad Supreme Court opinions regarding the technological issue of the day in 1967, regulating government wiretaps of telephones. The Stored Communications Act similarly was passed in 1986 to address concerns that the Fourth Amendment law was not tailored to the privacy needs raised by new forms of electronic communications. 

The need for updated comprehensive federal legislation to govern reverse warrants and similar tools is clear. But state-level legislation would be useful too, and this is where privacy advocates appear to be putting their efforts. NYU Law School’s Policing Project published information on how and why to regulate geofence warrants and similar data surveillance tools, arguing that legislatures are best positioned to establish detailed rules governing whether and when law enforcement may use data surveillance tools, what safeguards should apply, how data may be retained, and what oversight mechanisms should exist. Regulatory approaches like this would not prohibit all law enforcement use of geofence warrants, but instead seek to establish specific constraints on that use, such as restricting the use of such tools to investigations of serious crimes. Other groups would ban reverse warrants entirely, arguing that their purported benefits are illusory, and their harms impossible to mitigate.

Some groups have drafted model state bills: The Samuelson-Glushko Technology Law & Policy Clinic at the University of Colorado Law School has published a model state statute for regulating geofence warrants. This model would, among other measures, limit the use of geofence searches to investigations of serious crimes where other methods had been unsuccessful or were not feasible. Some states have already passed legislation that addresses the use of reverse warrants to varying degrees. Utah, for example, enacted legislation that generally requires a search warrant for geofence data and restricts keyword warrants. Washington State restricts private entities from establishing geofences around health care facilities.  California restricts the collection and sharing of geolocations related to abortion services. Other states can learn from and build on these and similar efforts.

Recent federal legislation that would regulate data surveillance has been sparse and unsuccessful. Nonetheless, it is worth noting one currently pending bill with bipartisan support that relates directly to questions raised by Chatrie: Senate Bill 4082. Among the provisions relevant to reverse warrants is the act’s Title II, which is the current iteration of the previously introduced Fourth Amendment Is Not For Sale Act. This act would close an enormous loophole in the protection provided by Chatrie by limiting law enforcement’s ability to purchase the same data from commercial vendors that they would otherwise need a warrant to obtain, thereby evading the constitutional floor put in place by Chatrie.

There is plenty in these legislative proposals about which reasonable people can disagree. And legislation will not alleviate the need for courts to fulfill their role in protecting the rights of those who may be overlooked in the democratic process, particularly criminal defendants. But there is little question that relying on courts alone to prevent the erosion of digital privacy is not a winning strategy.

It is worth noting that Chatrie the case will not be a win for Chatrie the man. Instead, Okello Chatrie is likely to share the same fate as the jurisprudentially related defendant, Timothy Carpenter, who was convicted for his involvement in a string of robberies of Radio Shacks and T-Mobile stores. As noted above, the Carpenter case was a historic win for digital privacy, but his conviction and 116-year prison sentence stood, even though the federal government secured Carpenter’s conviction using information the Supreme Court held was obtained in violation of the Constitution. This is because, although the exclusionary rule prohibits the government from admitting evidence obtained in violation of the Constitution, the “good-faith” exception to that rule allows introduction of that evidence if law enforcement was reasonably relying on a warrant that was later declared invalid. In Carpenter, the Court held that the government’s acquisition of data was a search, and that the warrant the government had used to obtain that data was invalid. Nonetheless, Carpenter’s conviction stood because of the good-faith rule. Chatrie is on an even more difficult path, given that the Supreme Court in his case did not hold that the warrant was invalid, and instead remanded for the lower court to determine its legality. Even if Chatrie prevails in his argument that the warrant in his case was invalid, it seems fairly certain that the government will prevail in its good-faith argument, meaning Chatrie’s conviction will stand.

Thus, the Chatrie victory is likely to be pyrrhic for Chatrie. Without legislation to answer the questions left open by the Supreme Court in Chatrie, the victory might turn out to be pyrrhic for everyone.