In a post last spring, I discussed the growing use of “Stingrays” by law enforcement officers.  These devices (technically known as International Mobile Subscriber Identity (“IMSI”) catchers) simulate a cell phone tower and thus “trick” a cell phone into sending it data.  These devices are critical for law enforcement surveillance purposes, since criminals will frequently use a specific cell phone for a brief period of time before discarding it and obtaining a new one.  Thus, police officers who know that a suspect is involved in criminal activity and who have have sufficient evidence to obtain a Title III order to wiretap the suspect’s phone are faced with a  dilemma–they have the legal right to listen in on the suspect’s cell phone conversations, but they are unable to do so because there is no way to tell the phone company which phone needs to be tapped.  Armed with a Stingray, the law enforcement officers can fool the suspect’s phone into connecting to the simulated cell phone tower and sharing its identifying information.  The police are able to then take this information and give it to the cell service provider, who can then set up a wiretap for the suspect’s phone.


The problem is that a Stingray will usually gather not just the suspect’s identifying phone information, but also the identifying phone information of everyone else in the area.  Practically, police solve this problem by following the suspect from location to location with the Stingray and determining which cell phone is consistently among the group that is being detected–essentially using process of elimination to determine which piece of identifying data belongs to the suspect’s phone.  But legally, this method raises Fourth Amendment concerns, since the government is gathering information from many innocent people in the process.

I argued earlier that the government should not need a warrant to use Stingrays in this manner, because all they learn about these third parties is the IMSI of the third party’s cell phone–mere “address” information that is already shared with third parties and which does not contain any remotely private information about the user (indeed, the vast majority of cell phone users do not even know their cell phone’s IMSI, nor would they notice or care if other people knew it).  Thus, using a Stingray to obtain a suspect’s IMSI does not implicate the Fourth Amendment in any way.

However, the federal government is taking no chances, and the Department of Justice issued a memo in September of this year that requires its agents to obtain a search warrant prior to using a Stingray device, unless an emergency situation exists.  Consistent with this policy, the government has conceded this point in recent court cases.  This concession makes sense in most situations–if the government already has sufficient information to obtain a Title III wiretap order against an individual, then the government surely has sufficient information to obtain a search warrant,.  Unfortunately, this self-imposed requirement does limit the usefulness of Stingrays in some contexts.  In many cases, the government only has reasonable suspicion that the suspect is committing a crime, and may only want to obtain a pen/trap order to determine whom the suspect is calling and how often–information which is not constitutionally protected but which can be very useful in developing probable cause.  But by requiring probable cause at the outset, the government is denying itself the ability to use this basic investigative tool if the suspect is sophisticated enough to frequently switch cell phones.

More recently, a federal magistrate court imposed even more restrictions on the use of Stingray devices.  In a recent order, Magistrate Judge Iain D. Johnston of the Northern District of Illinois summarized the restrictions that he placed on law enforcement officers when he grants a warrant for using a Stingray.  Essentially Judge Johnston sets three requirements for law enforcement:

(1) Officers must make reasonable efforts to minimize the collection of third party cell phone identification information.

(2) Officers must destroy all third party identifying information that they do collect within 48 hours of its collection.

(3) Any data obtained from the Stingray device can only be used to identify the target’s identifying cell phone information and for no other purpose.

The first two limitations are sensible enough and do not seem to affect the effectiveness of the surveillance.  The use limitation of the third requirement, however, seems unnecessary–there may be a number of legitimate and useful other ways for law enforcement to use the information.  For example, once the police know the target’s IMSI, they can use the Stingray device at a later time to determine the suspect’s general location (This information is not protected under the Fourth Amendment unless it is continuously protected over a long period of time, as in the Jones case).  And some versions of the Stingray device may be able to operate as a pen register, telling the police the numbers the target is calling and the length of those calls.  This information is also unprotected under the Fourth Amendment (there are statutory restrictions on such information, but they are far lower than the probable cause that the police have already developed before they get a warrant).  But for now, the federal government seems content to live with these restrictions and as far as we know, has not challenged them in court.

recent AP article reports on the growing trend of stashing miniature GPS devices inside items that are likely to be stolen and then tracking those items after the theft occurs.  The article describes a string of armed robberies on Long Island in which the perpetrator stole cash from gas stations and convenience stores.  Police investigators then slipped a miniature GPS device into a stack of cash that the perpetrator was likely to steal, and then–once the money was in fact stolen–tracked the perpetrator for a number of days until arresting him.  (The article does not say how the police were able to predict where the robber was going to strike next; presumably they provided these miniature devices to a number of prospective theft victims).  This law enforcement tool has been used in different ways for many years: the article correctly notes that police use a car’s built-in GPS tracking system to trace the car after it has been stolen, and also notes that pharmacists frequently have a pill bottle pre-loaded with a GPS tracker to give to a robber who steals drugs.  In a new twist, police officers who are engaged in a high-speed car chase are now shooting miniature GPS devices out of a special air cannon mounted in the front of the police car.  Once the GPS device sticks to the car, the police can call of the dangerous chase and simply track the criminal to his final destination.


On its face, thes appears to be a valuable law enforcement tool that can help police solve crimes.  Yet the AP news story’s headline delivers a warning: “Hidden GPS devices to track suspects raise legal concerns.”  An ACLU spokesperson argues that “As a baseline, I don’t think people should be tracked with GPS without a warrant,” echoing the theme of the 2012 Supreme Court case United States v. Jones, in which a plurality of the Court held that using a GPS device to track an individual over public roads for four weeks violated that individual’s reasonable expectation of privacy.

These concerns seem alarmist, at best.  Even the concurring Justices in Jones did not imply that any tracking with a GPS required a warrant–they were only concerned with  tracking  over an extended period of time, which would give law enforcement a “mosaic” picture of someone’s life.  And here we are talking about GPS devices which are attached to stolen property–property, in other words, in which the owner has no reasonable expectation of privacy.   This remains true even if the police decide to continue the tracking for a few days or weeks after the theft to see if the stolen property leads them to other contraband or other criminals.  The only feasible argument against this use of GPS tracking would be that the actual criminal might transfer the stolen property–and with it, the GPS device–to an innocent third party, who would then be tracked by the police.  Even then, of course, the GPS device would only show the location of the stolen goods, which the police are entitled to recover even from innocent third parties.

It is certainly important to be vigilant as law enforcement officers develop and employ new technologies in their investigations–we have to ensure that there is the proper balance between liberty and security.  But some uses of new technologies increase security without infringing on our liberty or privacy rights in any significant way.

locked cell phone

This week the New York Times published an op-ed which argued for allowing law enforcement officers with search warrants greater access to the cell phone data of criminal suspects.  The piece was co-written by an impressive set of authors: the District Attorney of Manhattan, the chief prosecutor of Paris, the commissioner of the City of London Police, and the chief prosecutor of the High Court of Spain.  They note that many modern cell phones are password protected, and that Apple and Google (whose operating systems together run about 96% of cell phones) no longer have a copy of that password and therefore police cannot access these cell phones even if they have legal authority to do so.  The piece argues that once law enforcement officers have obtained a warrant (having thus proved to a neutral magistrate that there is probable cause to believe there is incriminating information on the cell phone), there should be no technical barrier (such as password protection) to extracting that information from the digital device.  As they argue:

In the United States, Britain, France, Spain and other democratic societies, the legal system gives local law enforcement agencies access to places where criminals hide evidence, including their homes, car trunks, storage facilities, computers and digital networks.

Carved into the bedrock of each of these laws is a balance between the privacy rights of individuals and the public safety rights of their communities. For our investigators to conduct searches in any of our jurisdictions, a local judge or commissioner must decide whether good cause exists. None of our agencies engage in bulk data collection or other secretive practices. We engage in targeted requests for information, authorized after an impartial, judicial determination of good cause, in which both proportionality and necessity are tested.

It is this workable balance that proscribes the operations of local law enforcement in our cities, and guides our residents in developing their expectations of privacy. But in the absence of laws that keep pace with technology, we have enabled two Silicon Valley technology companies to upset that balance fundamentally.

Judging by the comments posted by the Times, the op-ed was not well-received by the readership: readers argued that encryption protects our data from thieves and hackers as well as from police; that political dissidents and activists rely upon it to communicate safely; and (echoing Riley v. California) that the sheer amount of information on a cell phone means that they need to be protected, even from police officers with search warrants.  The Electronic Frontier Foundation predictably warned that the piece was “nothing more than a blatant attempt to use fear mongering to further their anti-privacy, anti-security, and anti-constitutional agenda.”

It is hard to see what is “unconstitutional” about giving law enforcement access to information once they have obtained a warrant for that information.  Just because we now have the ability to easily password-protect  much of our personal data doesn’t mean that we somehow have greater constitutional rights in that information than we did twenty years ago.  Indeed, if the police had a search warrant for a fine cabinet, they should be able to look inside the file cabinet whether or not the owner has locked it.  The same argument should apply to cell phones–once a court has authorized the search, the police need to (and should be able to) conduct that search.

The real problem–and one that the authors of the op-ed do not really address–is how to go about ensuring that the police do have this ability once a warrant is issued.  The op-ed merely states that “regulators and lawmakers in our nations must now find an appropriate balance between the marginal benefits of full-disk encryption and the need for local law enforcement to solve and prosecute crimes.”  But it is one thing to ask for a “balance” and another to figure out what laws need to be passed to ensure that balance.  One option would be to require the manufacturers of digital devices to provide the government with a “master key” to every cell phone–but the danger of abuse in that context becomes quite obvious.  Another option would be to require the companies that design operating systems to keep a copy of every password (thus making it illegal for Apple or Google to use the operating systems they are currently using)–but this seems like a particularly severe government intrusion into the private sector.  Yet another option would be to allow police to compel the password from the owner of the device, but this raises serious Fifth Amendment questions.  Some courts have held that forcing a suspect to give up his own password is akin to self-incrimination, citing a United States Supreme Court decision which stated that the Fifth Amendment protects a defendant from producing documents which may be incriminating.

In short, the op-ed correctly identified a problem, but was silent on the solution.  Unless and until law enforcement officers develop the tools to break through password-protected phones, this problem will grow more and more severe until one of the more draconian solutions listed above becomes necessary.


This week the Washington Post unveiled a database it has been compiling regarding the fatal police shootings in this country.  This is essential data that, shockingly, nobody had been officially tracking before.  Over the past three years as we have tried to come to terms with the many high-profile police killings across the country, any serious analyses and conclusions were limited by the fact that no organization had been keeping track of this number in any systematic way.  Bravo to the Washington Post for doing what newspapers, at their best, should be doing.  One small critique up front–it seems to me that the database should keep track of all police killings, not just police killings by firearm.  Two of the most controversial police killings over the past year have been the chokehold death in Staten Island, and the death in the police van in Baltimore–neither of which would show up in these numbers.
What do the numbers say?  The raw number is 463 killings in the first six months of 2014.  I am not surprised by the absolute number of killings.  Obviously any killing by a police officer is one too many, but nationally the police engage in over twelve million arrests per year–and certainly if we add in the number of non-arrest police-citizen encounters, that number grows even higher.  Given those numbers, and the high prevalence of gun ownership in this country (see below) some amount of violent encounters between police and civilians is unavoidable.  (The numbers also go both ways–last year 126 police officers were killed in the line of duty).  And of the 463 fatalities so far this year, the suspect was armed in 387 of them, which is at least some evidence that many of those killings were justifiable.
Having said that, it does not take too much digging into the numbers to get to some truly startling and troubling numbers.  The most troubling is the racialized aspect of the statistics.  Although whites made up almost exactly 50% of those who were killed, they made up only 27% of the unarmed civilians being killed–blacks and Latinos made up 75% of that number.  This confirms a lot of what many studies have told us about how implicit bias can change a police officer’s instinctive reaction to a situation depending on the race of the subject.  It also implies that a lot of the killings of unarmed civilians could be avoided–there is no legitimate reason why the numbers for non-whites should be disproportionate to the numbers for whites.
One other point which I can’t resist making.  I am currently in Oxford teaching Comparative Criminal Procedure, and I opened my class today talking about these numbers and the new Washington Post database (it is unrelated to the topics we are discussing in class, but it is a criminal procedure issue and a very important one).  After I went through the numbers, one of the students naturally asked–since we were in a comparative criminal procedure class–how the United States numbers compared to the British numbers.  I looked it up–Britain (with 20% of our population) has had 55 police shootings in the past 24 years–i.e., less than two per year.  Of course one of the big differences (at least to me) is the difference in gun laws between the two countries.  Almost none of the police here in Britain carry guns–because almost none of the civilians are allowed to carry guns.  When a police officer in the United States approaches a suspect, s/he has to always assume the suspect is armed with a gun, which sets a certain dynamic in play before contact even begins.  In Britain, the officer always assumes the opposite.  I don’t think there is the only reason that we have so many more police shootings as a percentage of our population, but it is clearly a significant factor.

In a 5-4 decision yesterday, the Supreme Court held in United States v. Patel that hotel owners cannot be required to turn over their registries to law enforcement without some kind of judicial review.  The case centered on a Los Angeles municipal code provision that required all hotels to keep a registry with information about there guests and to turn the information over to law enforcement upon request.  Law enforcement officers did not need to make any kind of showing of need before making the request, nor did the officers need to get judicial approval.  A hotel owner sued the United States government, mounting a facial challenge to the ordinance on the grounds that it allowed for a suspicionless search that violated its rights.  Five Justices agreed.

There are three interesting aspects to this opinion:

First, eight of the Justices held that plaintiffs can bring facial challenges under the Fourth Amendment.  In my preview of this case, I predicted that the Supreme Court would not allow facial challenges because it would open the door to too many hypothetical cases.  This prediction proved to be completely incorrect.  The Court noted that it allowed facial challenges in many other contexts, and that there was no reason to prohibit them in the Fourth Amendment context.  The Court also noted that a facial challenge to a statute can succeed even if there were some situations in which the search would be authorized by some authority other than the challenged statute:

Similarly, when addressing a facial challenge to a statute authorizing warrantless searches, the proper focus of the constitutional inquiry is searches that the law actually authorizes, not those for which it is irrelevant. If exigency or a warrant justifies an officer’s search, the subject of the search must permit it to proceed irrespective of whether it is authorized by statute. Statutes authorizing warrantless searches also do no work where the subject of a search has consented. Accordingly, the constitutional “applications” that petitioner claims prevent facial relief here are irrelevant to our analysis because they do not involve actual applications of the statute.

Second, this holding will have an important effect on the development of the third party doctrine in Fourth Amendment jurisprudence.  Under the third party doctrine, an individual gives up all of his or her Fourth Amendment rights in an item or a piece of information if he or she hands it over to a third party.  That third party is free to give it to law enforcement, and the original owner of the item or document has no right to object.  Patel does not change any of that.  However, what if the third party–the hotel, the bank, or the internet service provider–does not want to turn the information over to law enforcement upon request?  Before Patel, the third party’s rights to object were relatively fuzzy, but Patel set out clear standards: the government must get some kind of “precompliance review” of its request before the third party can be forced to turn over the information.  This means either an administrative warrant (which is issued by a judge) or–if the government obtained a subpoena and the third party challenged the subpoena–a review of the subpoena by a judge or magistrate.  Granted, the standard for obtaining an administrative warrant or defending a subpoena is relatively low, but the requirement of precompliance review by a neutral judicial official is still a significant one.

Of course, none of this matters if the third party decides to cooperate with law enforcement–as has happened very frequently in the past.  But in the wake of the Snowden revelations and the public outcry that followed, more and more companies are beginning to assert their rights to keep this information from the government–both because the information has intrinsic value to them and because their customers are demanding more privacy.  Some companies, like Google, are challenging subpoenas more and more often.  Thus, for scholars and judges who complain about the third party doctrine, this may perhaps provide some solace–we still may not have the right to protect our information, but the companies we give it to are developing both the incentives and the legal ability to protect it for us.

The final aspect of the holding is one that flew a little bit under the radar, but that nevertheless is important.  The Patel Court assumed, without deciding, that a government search of the hotel’s registry qualified as a special needs search rather than a traditional law enforcement search.  The Court essentially said that there was no need to get into this question into more depth, because even under the lower “special needs” standard, the ordinance was unconstitutional, so it would clearly be unconstitutional under the more stringent law enforcement standard (which requires probable cause and–usually–a warrant).  But the very fact that the Court was willing to even make this assumption for argument’s sake is a bit troubling, and may presage yet another broadening of the already overbroad special needs definition.

Of course, the “special needs” justification has already been broadened way beyond the health and safety inspections of Camera and has been applied to searches which do little more than deter crime.  For example, as I discussed in an earlier article, almost every circuit court has held that the searches at airports are legitimate under the “special needs” doctrine, under the theory that they prevent terrorist action and thus protect public safety.  (The Supreme Court has yet to consider airport searches, but the Court has approved of them in dicta).   So already we have prevention of violent crime as a special need in order to keep society safe.  Likewise, the Supreme Court has approved of drunk driving checkpoints as a special need which keeps other motorists safe by preventing drivers from committing the crime of driving while intoxicated–which in a sense is per se reckless activity.

So we already apply the special needs search doctrine in order to prevent violent or reckless crimes–but at least those applications have some connection to protecting public safety.  Applying the special needs search doctrine to prevent drug use or prostitution in hotel rooms seems to be yet another step–a significant one–towards blurring the line between special needs searches and plain old fashioned law enforcement searches.  This was a line, for example, that the Court was unwilling to cross in Indianapolis v. Edmond.

To be fair, footnote 2 of Patel makes it clear that the Court is not making a decision one way or another on this particular issue.  So perhaps the Court simply wanted to make a broader point about the requirements for special needs searches and so decided to use Patel–even though in the end the search in question might not be an administrative search.  But this nuance may get lost as the months and the years go by, and many lawyers and judges will begin to think of Patel as the “administrative search case” and thus will think of the police accessing hotel records as a special needs administrative search.  Hopefully the Court can clarify this issue before that point.




A few months ago I wrote about (and strongly criticized) the Eleventh Circuit’s decision in United States v. Davis, in which the court held that the government needed to obtain a search warrant before it could access cell tower location information that located the defendant’s cell phone.  Now the Eleventh Circuit, in an en banc decision, has overturned the three-judge panel and held that the third-party doctrine applies to these records; thus, a warrant is not required.


The court began by citing the Fifth Circuit decision which also applied the third party doctrine in deciding this issue.  Then the court applied Smith v. Maryland and found that the Davis case was legally no different from Smith:

For starters, like the bank customer in Miller and the phone customer in Smith, Davis can assert neither ownership nor possession of the third-party’s business records he sought to suppress. Instead, those cell tower records were created by MetroPCS, stored on its own premises, and subject to its control. Cell tower location records do not contain private communications of the subscriber. This type of non-content evidence, lawfully created by a third-party telephone company for legitimate business purposes, does not belong to Davis, even if it concerns him. Like the security camera surveillance images introduced into evidence at his trial, MetroPCS’s cell tower records were not Davis’s to withhold. Those surveillance camera images show Davis’s location at the precise location of the robbery, which is far more than MetroPCS’s cell tower location records show.

The Court not only applies the third party doctrine, it presents a robust defense of the doctrine in this context, harkening back to the Katz test:

As to the subjective expectation of privacy, we agree with the Fifth Circuit that cell users know that they must transmit signals to cell towers within range, that the cell tower functions as the equipment that connects the calls, that users when making or receiving calls are necessarily conveying or exposing to their service provider their general location within that cell tower’s range, and that cell phone companies make records of cell-tower usage. See In re Application (Fifth Circuit), 724 F.3d at 613-14. Users are aware that cell phones do not work when they are outside the range of the provider company’s cell tower network. Id. at 613. Indeed, the fact that Davis registered his cell phone under a fictitious alias tends to demonstrate his understanding that such cell tower location information is collected by MetroPCS and may be used to incriminate him.

Even if Davis had a subjective expectation of privacy, his expectation of privacy, viewed objectively, is not justifiable or reasonable under the particular circumstances of this case. The unreasonableness in society’s eyes dooms Davis’s position under Katz. In Smith, the Supreme Court presumed that phone users knew of uncontroverted and publicly available facts about technologies and practices that the phone company used to connect calls, document charges, and assist in legitimate law-enforcement investigations. See 442 U.S. at 742-43, 99 S. Ct. at 2581. Cell towers and related records are used for all three of those purposes. We find no reason to conclude that cell phone users lack facts about the functions of cell towers or about telephone providers’ recording cell tower usage.

Although the third party doctrine has been routinely criticized, applying it makes sense in this context.  As the court notes, surely every reasonable person knows that the telephone company can track their general location using the person’s cell phone–how else could cell phones function?  And, notwithstanding the famous concurrence in United States v. Jones, a person generally does not have a reasonable expectation of privacy in a public place.

The en banc decision also provides an “alternative” justification for its ruling, which is that even if the third party doctrine did not apply, the search was “reasonable” because the intrusion into privacy was minimal, cell tower location information is routinely used by government investigators, Congress has explicitly endorsed this type of investigation in the Stored Communications Act, and the government’s interest in tracking down criminals is “compelling.”  Professor Orin Kerr had a number of withering critique of this alternative justification in his blog post; one of which was that the “reasonableness” test (as opposed to the warrant requirement) should only be applied in non-criminal cases:

A basic summary of the Supreme Court’s cases might run something like this: When the search involves some kind of non-criminal investigation or purpose, the warrant requirement is often suspended. In that non-criminal context, reasonableness instead becomes a general balancing of interests. The Court has been expanding the general balancing cases, most recently in Maryland v. King. But the Katz rule of a warrant by default is still the Supreme Court’s blackletter law for a traditional criminal investigation search.

In this case, the Eleventh Circuit appears to take a different approach. It begins with the Supreme Court’s non-criminal cases and then applies them to the context of a classic criminal investigation. Instead of the Katz rule of a warrant, the court begins with general balancing. It’s important to catch criminals, the court reasons, and the statute has some good protections given that this wasn’t such an invasive practice. So on the whole the government’s conduct based on reasonable suspicion seems reasonable and therefore constitutional.

This alternative holding is a major development, I think. It’s at odds with the usual rule that a criminal search requires a warrant, and instead replaces it with a totality of the circumstances inquiry into whether the criminal search was the kind of thing that we would generally say is good or would generally say is bad. There’s not only no warrant requirement, there’s no probable cause requirement: It’s just a free-floating reasonableness inquiry.

Professor Kerr has a good point here, but he might be fighting a losing battle.  In reality,  the distinction between “criminal searches” and “non-criminal searches” is becoming blurred almost beyond recognition.  Special needs searches have always been evaluated on a “reasonableness” standard, and many of them are nothing but criminal searches thinly masquerading as non-criminal searches (for example, testing for drugs in schools, stopping cars to check for drunk drivers, and searching passengers before they board an airplane).  Most recently, in Maryland v. King, the Supreme Court applied the reasonableness test to DNA swab of arrestees which was used to determine if the arrestee had committed any other crimes was not a “criminal search.”  Applying the reasonableness test to the obtaining of cell phone location data in a bank robbery investigation definitely pushes the envelope even further, but it continues a trend which has been building for a while.

Of course, this aspect of the Davis en banc decision is merely dicta, so perhaps nothing at all will come of it.  But as far as the holding of the case is concerned, the court has at least brought consistency back to this area of law.


Ohio’s Supreme Court is beginning to get a reputation for aggressively protecting Fourth Amendment rights in the digital age.  Six years ago in State v. Smith it held that police officers may not search a cell phone incident to an arrest, foreshadowing the United States Supreme Court’s Riley decision by five years.  Today the Court released a decision which set out strict requirements for law enforcement officials who are seeking search warrants for computers.  The Court held that the search warrant in question lacked particularity because it essentially authorized a limitless search of the defendant’s computer.

egged cartext








In the case, State v. Castagnola, the defendant was initially being prosecuted for selling alcohol to a minor.  He then took the ill-advised step of egging the prosecutor’s car to show his displeasure with the case.  He then took an even less-advised step and bragged about the egging to a friend, first by text and then in person.  The “friend” ended up being a police informer, who was wearing a wire at the time of the defendant’s verbal confession.  The police then obtained a search warrant and searched the defendant’s home (including his computer) for evidence of the egging incident.  Instead, they found evidence of child pornography.  The defendant was ultimately convicted of the child pornography charge.

The issue at the Ohio Supreme Court was two-fold:

First, the defendant told the police informant that he found the prosecutor’s address by tracing him through a parking ticket the prosecutor had received a few years earlier.  In the detective’s affidavit seeking a warrant, the detective erroneously said that the defendant had mentioned that he found the prosecutor’s address after conducting an “online” search–thus supporting the detective’s request to search the defendant’s computer in order to find evidence of the search.  But the defendant in fact never mentioned how he searched for the prosecutor.  The detective had inferred that the search occurred online, but the Court held that such an inference was not strong enough to rise to the level of probable cause that the defendant had used a computer, noting that “[a]lthough we are in the computer age, records of court activity still exist in paper form and are available to the public in clerk of courts’ offices around the state.”  In other words, even though the defendant had admitted that he had searched through court records for the prosecutor’s address, the likelihood that he did so using a computer does not suffice to establish probable cause.

Second, the warrant’s language, which copied the language from the detective’s affidavit, authorized the police to search:

Records and documents either stored on computers, ledgers, or any other electronic recording device to include hard drives and external portable hard drives, cell phones, printers, storage devices of any kind, printed out copies of text messages or emails, cameras, video recorders or any photo imaging devices and their storage media to include tapes, compact discs, or flash drives.

The Ohio Supreme Court noted that this language violated the Fourth Amendment’s particularity requirement because it did not include any “limitation on what records or documents” were allowed to be searched.  Event though the warrant later specified that any evidence that was recovered was to be used as evidence in a retaliation and criminal damaging case, the warrant was still overly broad.  Thus, the evidence should have been excluded–which almost certainly means that the child pornography case will now be dismissed.

The Court’s first conclusion seems a bit strained–is it really feasible to think that the defendant actually went to the county clerk’s office in person to look up the prosecutor’s old parking ticket?  An inference that the defendant conducted an online rather than a physical search seems not only reasonable (and thus sufficient to support probable cause) but almost certain.

But the Court’s second conclusion is surely correct–courts are already struggling with how to draft warrants in order to properly limit searches of computers, and a broad warrant that permits the police to search through every digital file on every digital device does seem overbroad.  Indeed, the law enforcement technician gave a somewhat weak explanation of why she was looking at picture files when she was allegedly looking for evidence of a search for the prosecutor’s court files.  (She explained that she “went to the images [folder] to find images associated with court websites.”)  On this point, at least, the Ohio Supreme Court’s message is clear: narrow your computer search (and your language in the search warrant) to items that are related to the crime you are investigating.






This week the Court decided Rodriguez v. United States, and held in a 6-3 decision that the police cannot prolong a traffic stop even for a few minutes in order to conduct a drug dog sniff.  More broadly, the Court held that the duration of a traffic stop, like the duration of a Terry stop, must be “reasonable,” and thus may not be lengthened by police conduct which was beyond the “mission” of the traffic stop.

Drug Dog Car Sniff

In my preview of this case, I predicted that the Court would be forced to define what is “reasonable” by essentially legislating from the bench and listing precisely the actions that officers can and cannot do during a routine traffic stop.   I wrote that the Court would define “reasonableness” as “what we are used to seeing the police doing, rather than what is actually reasonable for the police to do.  Thus, in this case, well-established practice will create the constitutional rule, rather than (as it should be) the other way around.”

My prediction was mostly correct, but I give the Rodriguez Court credit for at least trying to create a principled distinction between what is “reasonable” during a traffic stop and what is not.   Specifically, the Court said that the police officer may take a reasonable amount of time to “address the traffic violation that warranted the stop;” to “attend to related safety concerns;” and to conduct “ordinary inquiries incident to [the traffic] stop.”  This last category is included to allow police to “ensure that vehicles on the road are operated safely and responsibly.”  In contrast, the police officer is not allowed to take any extra time to “detect evidence of ordinary criminal wrongdoing.”  The Court used these guidelines to conclude that it is impermissible to take extra time during a traffic stop to use a drug dog.

Unfortunately, the Court’s attempt at making a principled distinction falls short.  The Court helpfully provides a list of what falls under the “ordinary inquiries” category: checking the driver’s license, inspecting the automobile’s registration and proof of insurance, and determining whether there are outstanding warrants against the driver.  But as Justice Thomas points out in his dissent, this last action–checking for warrants–does not really ensure that the vehicle is operated safely and responsibly–it is, plain and simple, an investigation into whether the suspect has been involved in criminal activity.  As such, it seems indistinguishable from the drug dog sniff, at least under the paradigm provided by the Court.

Interestingly, Rodriguez appears to limit another common activity during a traffic stop–questioning the suspect in order to learn if the suspect is engaged in criminal activity and/or requesting to search the car.  The Court does note that the police may “conduct certain unrelated checks during an otherwise lawful traffic stop,” but it states that “[s]he may not do so in a way that prolongs the stop, absent . . . reasonable suspicion.”  Thus, Rodriguez seems to prohibit any investigatory questioning unrelated to the reason behind the stop which prolongs the length of the stop.  It will be interesting to see how lower courts interpret this aspect of the case–does this mean that if the police officer has completed the warrant check and the citation, she is not allowed to ask any further investigatory questions of the driver unless they are somehow related to the offense?  Or maybe lower courts will interpret Rodriguez even more strictly.  As Professor Rory Little notes at SCOTUSblog: “the Court says only that an officer ‘may conduct certain unrelated checks’ during a stop, without discussing what limitations ‘certain’ may imply.  The majority also says that ‘on-scene investigation into other crimes … detours from that mission’ of traffic safety.”  Lower courts will have to interpret this language to determine which questions are or are not permitted to lengthen a traffic stop.

Professor Little also discusses another “loose end” in the Rodriguez decision: he sees the case, when considered alongside the rejection of the front-door dog-sniff in Jardines, as evidence that the Court is backing away from its 1983 decision United States v. Place which stated that drug dog sniffs are not a Fourth Amendment search.  (This would also mean, presumably, that the Court is backing away from its affirmation of Place in the 2005 decision of  Illinois v. Caballes).  He calls Place “one of the more extreme extensions of the ‘reasonable expectation of privacy’ test,” and argues that its rationale is now suspect in light of Jardines, Riley v. California (which prohibited searching cell phones incident to a lawful arrest), and Kyllo v. United States (which prohibited warrantless thermal imaging of a home).   These cases, Little notes, are evidence that the Court has “recognized that privacy has some scope beyond ‘expectations’ that the real world may compel us to abandon,” and that the Court is coming to the realization that drug dogs are like any other investigative tool and should be treated as such.

I am not so sure about this conclusion.  The Court’s rationale that drug dog sniffs do not constitute a Fourth Amendment search is based on a very specific doctrine–the binary search doctrine–and nothing in Kyllo, Jardines, or Riley affects that doctrine in any way.  The binary search doctrine states that government surveillance is not a Fourth Amendment search if it is only able to reveal the absence or presence of illegal activity.  The doctrine is based on the theory (first discussed in Rakas v. Illinois) that the Fourth Amendment only protects legitimate expectations of privacy.  In contrast, the Court’s recent limitations on the use of drug dogs have been based on other doctrines.  In Jardines, the limitation was based on the new  “trespass” Fourth Amendment test found in Jones; and in Rodriguez, the limitation was based on the doctrine that although a drug dog sniff is not a Fourth Amendment search, it is a Fourth Amendment seizure.  All that Rodriguez stands for is that the seizure aspect of the drug dog sniff needs to be justified by reasonable suspicion, and that any unreasonable delay that prolongs a seizure in order to conduct a drug dog sniff is unconstitutional–which is exactly what the Court held in Place thirty-two years ago.

NOTE:  Other commentators have also reacted to this case:  Orin Kerr has stated that he believes the case is correctly decided and that the Court was wise to adopt the “safety-based rationale” for traffic stops, while Steven Schwinn discussed the case on the Constitutional Law Prof Blog.


In the recent Supreme Court case of Hein v. North Carolina, the Supreme Court held that a police officer’s reasonable mistake of law can still validly provide reasonable suspicion for a traffic stop.  Although the case caused some concern that police officers would now be able to ignore (or at least creatively interpret) the law when making a traffic stop and then later argue that they made a “reasonable mistake,” I argued that the holding would end up having a very limited application, since a police officer’s mistake of law would almost never be deemed “reasonable” by a reviewing court.

stop sign changing-lanes

Recently the Hein decision has been applied by two lower courts, one of which found that the mistake by the police officer was reasonable under Hein, and one of which did not.  In People v. Guthrie, the defendant ran a stop sign at the edge of a supermarket parking lot in the Village of Newark in New York.  The police officer pulled the defendant over, noticed that he was intoxicated, and charged him with driving while under the influence.  Upon further investigation, the parties learned that the stop sign had not been officially registered in the Newark Village Code, and therefore the sign was not legally authorized under the state vehicle and traffic law.  Thus, the defendant had no legal duty to stop at the stop sign, and the police officer had no legal right to pull him over after he ran through the stop sign.  The trial court therefore dismissed the evidence, and the appellate court agreed.

On appeal, the New York Court of Appeals reversed, holding that under Hein and the applicable New York state law, the officer’s mistake of law was reasonable, and therefore the stop was valid.  The Guthrie Court did not say much about how to determine whether a mistake of law is reasonable or unreasonable; it spent most of its time explaining why the Hein rule was consistent with New York state law and why it was good policy.  Essentially The Guthrie Court adopted the Hein Court’s theory that there should be no distinction between a police officer’s mistake of law or mistake of fact–as long as the mistake is reasonable, the police officer’s stop is still justified.  In discussing the facts of the Guthrie case, the Court provided this rather unhelpful explanation:

[W]e are not saying that it would have been objectively reasonable for the arresting officer to have claimed ignorance of the requirement in Vehicle and Traffic Law § 1100 (b) that a stop sign in a parking lot be registered to be valid. We are saying that the stop was nonetheless constitutionally justified because the officer was not chargeable with knowing each and every stop sign that was registered under the Newark Village Code.

This explanation seems a bit circular: it doesn’t seem to matter whether the officer is expected to know about the requirement that a stop sign be registered if the officer is not required to know which stop signs are in fact registered.  On the other hand, it does seem reasonable that a police officer would believe it was legal to pull someone over for running a stop sign.  The only other options would be to require the police to memorize all of the stop signs in the precinct which are registered, or to require the officer to always look up the status of a stop sign after seeing a suspect run through it.  Neither of these options seems very feasible.

Meanwhile, the Fifth Circuit was deciding the case of United States v. Alvarado-Zarza, in which the police officer pulled over the defendant for failing to signal 100 feet in advance of making a turn.  The stop lead to a consent search, and the search led to a discovery of cocaine in the car.  In a suppression hearing, the defendant pointed out that he was only making a lane change, not making a turn.  The trail court denied the suppression motion, holding that the defendant made the lane change and then made a turn from the new lane, so that the lane change and subsequent turn were equivalent to “one prolonged turn.”  The Fifth Circuit disagreed, holding not only that the turn signal law did not apply to lane changes, but also that applying the law to lane changes was unreasonable:

In Texas, “[a]n operator intending to turn a vehicle right or left shall signal continuously for not less than the last 100 feet of movement of the vehicle before the turn.” Section 545.104(b) is unambiguous.  Its 100-foot requirement only applies to turns; lane changes are not mentioned.  Further, the statute elsewhere refers to turns and lane changes separately, thereby setting out a distinction between the two.  This distinction is further indicated by the Texas Driver’s Handbook, which defines a lane change as a “lateral maneuver moving the vehicle from one lane to another” and a turn as a “vehicle maneuver to change direction to the left or right.”  See  Mahaffey v. State, 316 S.W. 3d 633, 641 (Tex Crim App. 2010).  The former definition confirms that a lane change, as its name implies, occurs when moving from one lane to another.  A “turn,” on the other hand, involves a change to a vehicle’s direction.  The terms “turn” and “lane change,” therefore, signify distinct actions.   This, Section 545.104(b), by its plain terms, does not apply to lane changes.  

Second, seven months prior to Alvarado-Zarza’s stop, the Texas Court of Criminal Appeals in Mahaffey addressed the distinction between a turn and a lane change.  In that case, a policeman mistakenly concluded that a driver was “turning” by moving out of a lane that was ending.  Although the court did not discuss Section 545.104(b)’s 100-foot requirement, it drew a clear distinction between a turn and other movements, including a lane change.   The court deemed this distinction “plain and unambiguous.”  We interpret Mahaffey to be a rejection of Officer Barrientos’ view.  

Because the caselaw far predates the stop in this case, and because the statute facially gives no support to Officer Barrientos’ interpretation of the 100-foot requirement, we conclude that his mistake of law was not objectively reasonable.

Since the Fifth Circuit was overturning a lower court ruling, it was not merely holding that the police officer’s mistake of law was unreasonable, but also that the district court’s interpretation was unreasonable.  In other words, even when a police officer’s understanding of the law has been confirmed by a trial court judge, it could still be unreasonable.

What lessons can we learn from these two early applications of Hein?   Perhaps the most troubling issue from either of these cases is that the district court in Alvarado-Zarza was willing to construe the turn-signal law so broadly against the criminal defendant, even in the face of all of the arguments to the contrary.  But this is not really a critique of how Hein was applied, since the district court didn’t even think any mistake of law had been made.

In the end, both courts seem to have arrived at the correct result, though the Guthrie Court could have explained its rationale it a bit more detail.  It is hard to argue that the police officer in Guthrie violated the defendant’s rights by pulling him over after he drove through a stop sign–even an “unregistered” one–while the Alvarado-Zarza case refused to give the police office the benefit of the doubt in construing a traffic law, even when a trial court judge agreed with him.  Thus, it seems that so far the Hein decision has not provided police officers with an opportunity to ignore or aggressively interpret the law in their pursuit of traffic stops.