Welcome to the Berkeley Technology Law [Journal] Podcast.
This is Ben Brokesh, and you’re listening to the Berkeley Technology Law Journal podcast. On August 23, 2020, in Kenosha, Wisconsin, police officers shot 29-year-old Jacob Blake seven times, leading him to be partially paralyzed. This was yet another public display of brutal police violence that had been at the forefront of America’s mind since George Floyd’s murder that May. Protests sparked, and Kenosha became the focal point of these frustrations. The protests were largely peaceful; however, they were interspersed with right-wing militia activity and individual property violations. In an attempt to control these protests, the police used hard tactics like rubber bullets and tear gas. But our story today discusses something else the police used: geofence warrants. First, what are geofence warrants? Unlike traditional warrants related to non-suspects, geofence warrants are used by police when a suspect or suspects cannot be identified. To get the warrant, a law enforcement agency specifies the location and time period. Then, under judicial approval, companies search their databases to give a list of users in a specified location over the specified time. These companies are internet service providers and telecom providers that have collected user information through Wi-Fi, GPS, cellular data, or RFID. In the Kenosha protests, the police used geofence warrants to identify persons in areas where property had been damaged. These geofence warrants would tell the police who was in the vicinity; this vicinity could stretch to a football field away. Since these warrants are not specific to any person of interest, many innocent protesters could have had their information collected without their knowledge. So, are geofence warrants constitutional? They certainly seem to touch on the privacy concerns in the Fourth Amendment. Additionally, for protesters, could unknown data collection about their involvement risk their First Amendment rights? On today’s episode, my colleagues Chris Musachio and Hazim Alwazir here sit down with Berkeley Law’s Dean, Erwin Chemerinsky, and another one of Berkeley Law’s amazing professors, Orin Kerr, who are encouraged to discuss the implications of geofence warrants on First and Fourth Amendment rights. Afterwards, my colleague Hannah Brown will highlight key takeaways from these discussions. This has been Ben Brokesh. Enjoy the conversation.
Interview #1 Erwin Chemerinsky
Based on today’s SCOTUS makeup, do you believe a Fourth Amendment challenge of the use of geofence warrants would be likely to be successful? Why or why not?
[ERWIN CHEMERINSKY] 2:56
I think it’s hard to know whether a challenge to geofence warrants to the Fourth Amendment will be successful in the current Supreme Court. In 2018, the Court decided Carpenter v. United States.1 And there, the Supreme Court said for the government to gather large amounts of cellular location information was a search that requires a warrant based on probable cause.2 And on the basis of Carpenter, I think I could make an argument that a geofence warrant is also gathering information about people, especially if it’s done over a long period of time, and that the principles of Carpenter should apply. For all the geofence warrant is much like what a general warrant is, which the Fourth Amendment, above all, was meant to stop. On the other hand, Carpenter was a five-four decision with Justice Ginsburg in the majority.3 I worry that the Justice who replaced her, Justice Barrett, is much more likely to agree with the conservatives from Carpenter, (Thomas, Alito, Kennedy, Gorsuch) than likely to agree with the more liberal justices such as Ginsburg. Also, the Supreme Court still has the third-party doctrine that says if somebody shares information with a third-party, there’s no expectation of privacy. Everything that’s gathered by a geofence is information gathered through a third party. So, if the court were to follow Carpenter, I think there’s the basis for a successful Fourth Amendment challenge. But I don’t know that the Court’s going to follow Carpenter. And I certainly don’t know that the Court’s going to be willing to extend it.
Do you believe that the standing requirements would limit challenges to those who have actually had their data turned over?
[ERWIN CHEMERINSKY] 4:49
I think it would be very difficult for somebody whose data has not been turned over to law enforcement to bring a challenge. I think that it would be hard for them to show either in terms of Article III standing that they have an injury, or in terms of the usual principles of Fourth Amendment standing, that they can raise an objection. In fact, just this year in a different context, the Supreme Court decided a case, TransUnion LLC v. Ramirez, that involved credit reporting agencies putting on people’s credit reports that they were a terrorist.4 The Supreme Court said those who have reports turned over to others would have standing and injury, but those who just had it on their credit reports wouldn’t have an injury. It’s a different context. But it leads you to believe that somebody who hasn’t had the information turned over would find it very difficult to have standing.
I think we’re going to shift gears a little bit now over to the First Amendment. Are there any other bases upon which a challenge to the use of geofence warrants could be made?
[ERWIN CHEMERINSKY] 6:00
I think the strongest ground for challenging geofence warrants is going to be the Fourth Amendment, that the equivalent to a general warrant that the Fourth Amendment was meant to prohibit. It’s also possible that protesters might be able to bring a challenge under the First Amendment.
What sort of threats do geofence warrants pose to freedom of speech, assembly, freedom of the press?
[ERWIN CHEMERINSKY] 6:28
Geofence warrants will make it possible for the government to learn who’s gathered for a protest It’d give the ability of the government then to have information that otherwise would have required physical surveillance, and perhaps by making it so easy for the government to monitor who’s at a demonstration, chill the speech that people are engaging in.
Would you mind explaining the chilling effects and that doctrine?
[ERWIN CHEMERINSKY] 6:59
There are Supreme Court cases that have said that a claim of chilling of speech is sufficient to show an infringement of speech for First Amendment purposes. One of the initial cases to this was Bantam Books v. Sullivan.5 About 1957, it involved a sheriff who was going to local stores and saying that you’re selling obscene material.6 And the question is, was that so chilling of speech as to violate the First Amendment? And the Supreme Court said yes.7 And there were other cases since when the Supreme Court has said government actions that chill speech violate the First Amendment. On July 1, 2021 in Americans for Prosperity v. Bonta, the Supreme Court said requiring not-for-profits to give to the state the same form they’re giving to the federal government, listing their donors, chilled speech in a way that violated the First Amendment.8 And yet, at other times the Supreme Court has rejected chilling of speech as being sufficient for a First Amendment claim. In Laird v. Tatum, the Supreme Court said the army spying on civilian peaceful protests didn’t raise a First Amendment claim, despite the argument of chill.9 In Clapper v. Amnesty Int’l, about eight years ago, the Supreme Court said the NSA interception of conversations of certain people in the United States to those in foreign countries didn’t raise a constitutional claim.10 There’s no First Amendment violation, because chilling of speech wasn’t sufficient for an infringement. So, the Supreme Court has been very inconsistent about when and whether chilling of speech infringes the First Amendment.
Would the standing issues that we discussed earlier potentially have a place in this chilling effects context?
[ERWIN CHEMERINSKY] 8:51
Whenever anyone wants to sue in federal court, they’re always going to have standing. There’s the question with regard to chilling as whether the Supreme Court would regard chilling of speech, or the likely chilling of speech, as an injury. The cases are flat-out inconsistent about that. Also, if someone is seeking an injunction, City of Los Angeles v. Lyons requires that the person show that he or she is likely to personally be injured again in the future.11
Could the looming threat of being caught in a dragnet geofence warrant fall under the prior restraint for speech?
[ERWIN CHEMERINSKY] 9:35
No, it doesn’t fall under the prior restraint doctrine, assuming the court uses the prior restraint doctrine as it has previously been defined. A prior restraint is where the government prevents speech before it occurs. Traditionally, it’s been things like an injunction that prohibits speech, or the government requiring licensing for speech. And this isn’t a court order stopping speech. This isn’t a license for speech. This might discourage speech. But any law that regulates speech might discourage speech. And then that makes the prior restraint doctrine meaningless. So, I don’t think that this could properly be regarded as a prior restraint.
Would then the distinction between prior restraint and just a regulation on the content of what’s being said, be implicated in this circumstance?
[ERWIN CHEMERINSKY] 10:38
Obviously, with regard to geofences, the question is how it’s used in order to assess whether it’s a content-based restriction on speech. Imagine that there’s a park, and the police obtain the geofence data for all demonstrations in the park for a month, regardless of what the ideology of the demonstration is, regardless of the topic of the demonstration. That would not be a content-based restriction on speech. But imagine that the government only obtains this information for liberal gatherings for speech, then I think you could argue that it’s content restriction. But keep in mind the difficulty: the geofence itself is not a restriction on speech. The claim is that the use of the geofence is chilling speech or restricting speech in a way that violates the First Amendment. That makes it harder to show it’s a content based restriction.
Some have described the First Amendment as being a super-right. Would you agree?
[ERWIN CHEMERINSKY] 11:44
Content-based restrictions on free speech have to meet strict scrutiny. Most other claims of government violation of rights only get rational basis review. Well, relative to rational basis review, that makes free speech a super-right. But there’s a lot of rights that trigger strict scrutiny, and free speech is one among them.
So then to circle back on everything we’ve discussed, given the current makeup of the Supreme Court, the trend of eroding the Fourth Amendment protections, some saying the First Amendment has evolved into this super-right. Do you think that a challenge under the First Amendment would be more or less successful than a challenge under the Fourth?
[ERWIN CHEMERINSKY] 12:34
Obviously, I’m speculating. My instinct is a challenge under the Fourth Amendment has more chance of succeeding than a challenge under the First Amendment. This really is the government gathering information about people without needing to have probable cause or individualized suspicion. That goes against the Fourth Amendment. Carpenter is just three and a half years old, and Carpenter said that under the Fourth Amendment, police need a warrant and probable cause to obtain a long period of stored cellular location information.12 A geofence is analogous to that. So, if I were a litigant and challenging it, I would put more of my money on the Fourth Amendment claim than the First Amendment claim. The difficulty for the First Amendment claim is that a geofence isn’t a restriction on speech. The concern is that a geofence might have the effect of chilling speech. And the Supreme Court has been so inconsistent whether claims of chilling of speech are sufficient for First Amendment violations.
What sort of solutions do you think non-judicially could be used to combat or restrict the use of geofence warrants?
[ERWIN CHEMERINSKY] 13:56
I’d like to see Congress specifically require that geofence use requires a warrant based on probable cause. If the government wants to obtain this kind of information or wants a cellular tower dump and sees everybody’s phone was linked to a cellular tower at a point in time, the government needs a warrant based on probable cause, just like it needs that to get stored cellular location information. I think that would be the easiest solution, the clearest solution to the problem.
So, you wouldn’t advocate for a heightened standard in a similar way to, like, the wiretap standards?
[ERWIN CHEMERINSKY] 14:38
I am arguing that a geofence should require a heightened standard. I would adapt the probable cause standard of the Fourth Amendment, not the standard for, say, pen registers. Wiretapping requires that there be probable cause. Pen registers don’t require probable cause. I’m arguing that the Fourth Amendment standard should be applied to geofences. Maybe to go back to Carpenter, prior to Carpenter to acquire stored cellular location information, the police still had to go before a magistrate judge, but it was a lesser standard than probable cause. I’m arguing that geofences should be treated the same as, say, a large amount of stored cellular location information. Probable cause should be the standard.
And then, what’s the potential for banning it outright? I know in the last year, New York has introduced legislation to ban its use by police. Do you think something like that would be practical on a federal level? Or is it easier to implement on a state level in the circumstance?
[ERWIN CHEMERINSKY] 15:54
To be clear, if a state would have banned the use of geofences, that would apply only to state and local police. A state can’t regulate federal law enforcement. And so, state bans on geofences wouldn’t have any effect on what the federal government could do. I’m skeptical that there will be an outright ban on the use of geofences. I think it’s much more likely that a court would say that probable cause requires that the Fourth Amendment standard has to be met. If you accept that a geofence is a search, then the question is, what’s the standard? And I think probable cause is the one under the Fourth Amendment that should be followed.
I appreciate your time. Thanks for coming.
[ERWIN CHEMERINSKY] 16:41
Oh, my pleasure. Glad we got through the questions. I hope I was able to help.
Oh, so much, really appreciate it. Thank you again.
Interview #2 Orin Kerr
Honestly, it’s an honor to have you here with us. Our first question is, law enforcement authorities have been using warrants to get the identifying information of individuals that are caught within an identified area. And I just wanted to ask you what legal process and checks are required for law enforcement authorities to get these warrants approved as it stands today?
[ORIN KERR] 17:19
Well, we don’t exactly know what legal requirements are required, we actually know only what the providers are demanding. So, this is basically Google, which has all this location information. And Google is demanding that the government get a search warrant. It’s not entirely clear that a warrant is required; a warrant would not be required under the federal privacy law called the Stored Communications Act. But Google is taking the view that to get this information, it requires a warrant either under the Fourth Amendment or for some other reasons. But basically, Google can demand a warrant even if it’s not clear a warrant is required. It’s because basically it just gives the government the burden of challenging whether a warrant is actually required, the government would have to go into court and say Google’s demanding a warrant, we don’t think we should have to get a warrant. And from an investigative standpoint, that’s basically impossible to do or hard to do, because it means putting, however, weeks, months, years hold on the case in order to figure out the legal issue. So at least right now, investigators are going along with Google’s demand for a warrant and the answer is a search warrant is being used.
Interesting. And do you believe that there is kind of this move? That it’s not really practical for investigators to get the search warrant within the context of investigation? But do you think that state authorities would levy a challenge regardless on it? Like do you see this happening in the coming years or coming months in any way?
[ORIN KERR] 18:57
I suspect what will happen is that this will get litigated in the context of a motion to suppress. And one of the government’s arguments will be that no warrant was needed in the first place. So, we shouldn’t be worried about whether this was a defective warrant or not. Because there’s no Fourth Amendment search occurring at all. And therefore, they’ll make that argument in the context of a motion to suppress. They can’t make that argument in the way that this has been litigated so far. So, what’s happened so far is the government will apply—that Google will say we demand a warrant, and then the investigators will go to a federal magistrate judge or state magistrate judge and say, okay, here, we’re applying for a warrant, and then the judge will say, okay, your warrant is or is not sufficient. Whether there is actually a Fourth Amendment search, whether a warrant is required, just doesn’t come up in that setting, because they’re not asking for a ruling on that. You can’t ask a judge for the ruling on that. So, the judges are assuming that a warrant is required and then there have been a couple opinions on what the warrant needs to look like and what would make the warrant valid, all assuming that a warrant is needed. So, we just haven’t had litigation on that first question. And we’ll have that down the road in motions to suppress as they get litigated kind of years, typically months or years, after the warrant is obtained.
Interesting. And when, assuming that the warrants are valid, that the warrants are needed, when executing these warrants for law enforcement to gather the information of all of those individuals within the identified area. And that may include certain individuals that they may lack probable cause as to committing a suspected offense. And so, I just wanted to ask you, assuming that a warrant is required, what do the protections look like and what protections are provided by the Fourth Amendment for unreasonable search of identifying information and location that would be dragged up by a police search of this information?
[ORIN KERR] 21:05
This is a great question. And we don’t really know the answer yet. There are different theories. We’ve had a couple magistrate judge opinions on this, although I don’t think the magistrate judge opinions are particularly strong on the Fourth Amendment questions. It’s really an open question. And here’s what I think the question boils down to: How do you model from a Fourth Amendment standpoint an effort to get to get non-content data about what phone was registered as being located in a certain area, assuming that the information is protected under the Fourth Amendment? And that may or may not be true, but assuming it is? What’s the model for what a warrant is? Is the warrant authorizing kind of the virtual equivalent of going into that space? So, let’s say the government gets a warrant saying Google tell us who was within a 100-yard by 100-yard area on Tuesday from 2 A.M. to 2:15 A.M. Is the warrant akin to a physical search through that place? Is the warrant akin to going to that place and looking around? Is the warrant akin to searching all of the people that are there at that place? Like what is that? We’ve never had anything like that. Because we’ve never . . . Until Carpenter in 2018 the answer would be none of this stuff is protected under the Fourth Amendment.13 So, there’s no search at all. And then Carpenter raises at least the possibility that it is a search, and then you run into the question of how particular is it? So, what’s interesting about this is actually a broader issue raised by Carpenter. This is the case from 2018, saying that it least in the aggregate, historical cell site records over time, if the government collects those, that is a Fourth Amendment search, a warrant is needed for that.14 So then, you run into the question, well, how can a warrant be obtained for that? Like, what’s a particular warrant for those records? Because the whole rationale of Carpenter was that there’s so much data, the government should need a warrant. And then the argument becomes, well, wait a minute, the “so much data” question was usually about the scope of warrants. Do we now just say that there is no scope limitation on warrants? Or if there is a scope limitation, what is it? So, there are a lot of different ways of approaching it. But the short answer is, we really have no idea how to construe the particularity of a search warrant for geolocation information.
In Carpenter, the Supreme Court held that an individual may maintain a legitimate expectation of privacy in the record of their physical movements; in the reasoning, the Supreme Court really looked at basically how pervasive the information is, and how an individual like . . . how it could take up information about a person going to church or a person going about their private life and in private spaces.15 And I was just wondering, so if the Court were to apply this principle into geofencing warrants where that area may be constrained, how do you see that kind of changing the application of Carpenter or the likelihood of Carpenter’s application in this case?
[ORIN KERR] 24:21
Yeah, so this is a way of talking about whether a search occurred at all. So, Carpenter says that at least when the government is collecting seven days or two days, or however, you know, some long term, cell site location records, that you have a reasonable expectation of privacy in the whole of your physical movements.16 It’s not voluntary, because you have to use a phone to participate in daily life. And therefore, when the government collects those records, at least in the long-term sense, it’s a search. It’s not clear that applies to a geofence warrant for a couple of reasons. One is that the geofence warrant is typically limited in time. And so, it’s not obtaining the whole of a person’s physical movements. Second, geofencing basically only works with Google because Google has services that a lot of people opt into or have on their phones. And so, there’s a separate question of whether you voluntarily use Google or using Google is required to participate in daily modern life. And so that’s just a question of whether this implicates the Fourth Amendment at all. Carpenter says nothing, I think relevant to what a warrant for these records would look like. It just said a warrant is required. And so, it’s tricky. What do you do when the legal development takes you from no warrant is required to a warrant is required because it involves a lot of information? And then you get to the step of, well, how limited can the warrant be? Because usually, warrants have to be narrow, because the government can’t just get all this unlimited information. And they know they can’t have an unlimited warrant. So, Carpenter changed the rule of what the Fourth Amendment protects.17 Do you also then say there’s a new rule on how broadly the warrant extends? Do warrants not have a particularity limit anymore? Do they have an extra-strong particularity limit? Is there, you know, like, is there a category of data that can’t be obtained at all? It just kind of, who knows? It’s really something that needs to be explored. And unfortunately, we’ve only seen this come up in litigation in the context of opinions by federal magistrate judges who are getting applications, and then they’re trying to figure out what applications they will grant. And I just don’t think . . . . They’ve had a couple theories of how to do that. And I just kind of think the theories that they’ve grappled with are, you know, I’m not. . . . It’s not obvious to me that any of the Article III judges that review this are going to defer a whole lot to what the magistrate judges have done, because some of them, some of them struck me as slightly off. And I’ll give you an example of one that has got a lot of attention that strikes me as likely wrong. There’s a case from the 1970s, involving a search of a bar, where the bartender is believed to be selling drugs, and they search the bar, and then they search the customers to see if the customers had the drugs that they might have bought from the bartender.18 And the Supreme Court says, no, no, no, when you’re searching the person, when you execute the warrant, you need probable cause in order to search the physical person, the probable cause as to each person in order to search them.19 And so at least one, maybe two magistrate judges have said that applies to geofencing warrants. So, when the government gets a geofencing warrant, there needs to be probable cause as to each phone that is revealed as being present in the location and time that the geofence warrant involves, and I don’t know how that could possibly be right. Like, that’s saying that if it so happens that the warrant reveals that a phone was in the area, then it’s like physically rifling through the person’s pockets, and looking through their wallet, and looking through, you know, taking all their . . . glasses off and taking their watch off, and looking at it strictly, it’s just a terrible analogy. So, I don’t think . . . I’d be really surprised if higher courts agree with that view. But exactly how the Fourth Amendment should apply is actually like, I don’t know. How much space is it? How much time? It’s not obvious to me how much relevance it should be that other people would happen to be there. You know, we don’t say for example, oh, you’re searching an apartment and ten people live in that apartment versus one person lives in that apartment. And so, you can’t search the apartment because a lot of people live there, and they might not be involved in the crime. That there sort of isn’t . . . . You can’t make that argument in Fourth Amendment law generally. So, I’d be a little bit surprised if that argument gets accepted in the higher courts, but we don’t really know. We’re just starting to see decisions on this. So, it’s really kind of too hard, too early to tell like where the law is going to go.
So, in this sense, do you really believe that there shouldn’t be a showing of probable cause as to specific individuals within the geofencing warrant? And how would that affect I would imagine, for example, when law enforcement gets a warrant for let’s say, a house, as you said, like, that’s akin to your example. Or if it was for a protest where there may be tens or even hundreds of individuals in that space?
[ORIN KERR] 29:41
Yeah, so there’s never been a requirement of individual probable cause for searching like in a place. So, if it’s ten people living in a house, one warrant is required, showing probable cause that there’s evidence in the house, and it doesn’t matter whether any of the people are suspects or completely innocent. So that’s the traditional treatment. The rule about individual probable cause for the people at the place where the warrant is executed has only been applied to individuals that are physically searched, like rifling through their pockets and sort of undergoing the physical . . . . This is not a physical search of a person, it’s recovery of a data point about a person. And I’d be surprised if a line of cases creating an exception to a general rule because of the invasiveness of physical searches would be applied to learning a fact about an account. So, in the context of protests, like usually those are just going to be in public and no warrant is going to be required. The government can monitor public protests without any Fourth Amendment oversight. They can take pictures of everyone who’s at the protest, and no Fourth Amendment oversight. It’s all occurring in public. The government can go through the place where the geofence warrant covered, and if the officer was there, no Fourth Amendment oversight. So, the question is, how different is the rule just because a computer is being used to collect the records of it historically? It might be a search; that’s the first sort of uncertain question. But I’m skeptical that the answer is that the government . . . if the government needs probable cause for each person who is found to be there, the answer is you never know if the government’s violating the Fourth Amendment until the data is added. And then, you just hope there happened to have been other people there. Because that adds, that creates, a Fourth Amendment violation. And it strikes me that if you’re engaged in some sort of, you know, if you want to make sure the government can’t track you, just go where other people are. And then whatever the government, the government would be violating the Fourth Amendment to track you. And it’s just that that strikes me as an unlikely principle for courts to adopt, although there are one or two magistrate judges who seem to have at least initially gone that way.
Interesting, and I want to look at specifically the issue of anonymization. So, when law enforcement requests data from people, let’s say from Google, Google only offers a list of anonymized users, along with some relevant coordinates, and a timestamp. Does that affect whether the search is of an individual, since they don’t know which individual it is? Like, at what point does the search occur by law enforcement?
[ORIN KERR] 32:35
Yeah, so assuming there that it is a search to get the records as a whole, the next question would be whether it’s a search of any one person’s data. And then, which is its own sort of separate issue potentially, because, for example, tower dumps, at least so far have been held to not be a search at all. But what Google has been doing by trying to keep the data anonymous has basically been trying to shield information about users who might innocently be in that location. And from a business policy standpoint, that makes a ton of sense, and why should the government find out who was in the area if they don’t have a law enforcement need to find that out and law enforcement doesn’t care because they’re only interested in who they have a law enforcement need to find out? So, that seems to be totally fine. It’s just not, to my mind, it’s not a Fourth Amendment issue. It’s a kind of corporate policy in a context where the provider’s kind of holding the cards, at least for now, but it doesn’t obviously have a Fourth Amendment resonance, it seems to me.
Given that anonymization, you think, doesn’t have a Fourth Amendment element to it, do you believe that there should be, similar to the Wiretap Act, that measures should occur on the front end to protect individuals’ privacy by the government itself, as opposed to requiring the private organization? Or does that not matter? Like in the way that, let’s say, the legislature, as opposed to the courts, chooses to regulate this issue, should they impose those safeguards on the government institution or the private party? And how would the government minimize the data and the information of individuals who may happen to be passing by in a geofenced area?
[ORIN KERR] 34:35
Yeah, so I have a broader view of how the Fourth Amendment should apply to digital searches, which is basically that the government, upon having a warrant to search some protected digital space or digital database, that there should be use restrictions on how the government uses data. Basically, they should be allowed to look through all the data and then use that which is relevant evidence of the crime that they’re investigating. But then everything that’s non-responsive to the warrant should not be usable. It should basically be kind of like a grand jury rule, that the information can’t be revealed. And so, I think the Fourth Amendment should be interpreted as having kind of a similar rule. It’s not obvious to me that it really is implicated in the context of geofence warrants. And I realized, I’m sort of giving pushback and that a lot of your questions are kind of assuming this is like an incredibly severe privacy invasion. And I think this is actually one of the lesser privacy invasions of, of all of the, you know, there’s a lot, the government can do a lot of really invasive things and finding out that a phone was in an area at a particular time, at one specific time, and one specific place, that strikes me as not hugely different from, for example, getting video camera footage from that time and place that showed a particular person there. And then if anything, the video footage, you know, that shows who the person was talking to or walking with and what they were wearing and identifies a person, whereas with the phone, it’s actually just the phone was there, it doesn’t necessarily resolve to a specific person because phones can be passed around and given to different people. So, I guess on the whole, I’m less concerned about this problem in the geofence warrant setting. But I do think that the Fourth Amendment itself should be interpreted to have a use restriction on non-responsive data when the government executes a digital evidence warrant.
I wanted to also ask you about your perspective on Carpenter, in a sense. And also, prior to Carpenter, you published several notes and papers on the third-party doctrine. And I was just wondering, does that affect kind of your theory on why the invasion of privacy isn’t that much in geofencing warrants? As in like, is it because, does that at all affect it in that, is it because individuals share their information or location information with Google? That that kind of influences why they don’t have kind of a large expectation of privacy there?
[ORIN KERR] 37:31
Um, no, my views on Carpenter and on the third-party doctrine generally are actually kind of on a different track. It just seems to me, you know, sort of imagine the government can find out different pieces of information. Imagine . . . . Let’s just lay out a couple options: the government can find out every website you visited in the last month, the government can read through all of your personal emails, or the government can find out that on August 2, 2019, you were within 50 yards of a public park at 2 A.M. Which one of those is kind of the one that makes you feel like, yikes, that is really, really private? And to my mind, the one that is kind of the least, like the fact that a phone was present in a location, that’s private information in some sense. I mean, you don’t want everyone to know where your phone is located, obviously. But that strikes me as a lesser issue than the government knowing every website you visited over a period of weeks or rifling through your emails. And that first example of the government finding out every website you visited, the Seventh Circuit recently said, that’s not a search at all, even under Carpenter.20 And I think that’s wrong. Actually, I think the question of what website you visited should be covered under Carpenter. But that’s at least something at the very least something that’s in flux, and one circuit court has gone the other way, so I think it’s just, you know, basically, you’ve got to think about all the different kinds of data out there. Some pieces of information are going to be really revealing and really private and really sensitive. And other pieces of information are going to be kind of less private, less revealing, less sensitive. And I tend to think the presence of a phone in a location and time is sort of one of the lesser categories of things, as compared to some of them you can have in the digital evidence context where you can have, you know, an incredible amount of deeply private personal information that is none of the government’s business but that the government can access. And so, I think that’s the stuff that I think where the real focus should be from a privacy standpoint. And I’m less concerned with the anonymization of data with which phones were in an area that comes up in a geofence. I mean, if it turns out government doesn’t care, providers want to protect the customers, it may just be there’s no dispute about it. Like no, everyone agrees, the government doesn’t need stuff that has no investigative purpose, and the providers don’t want to give it out. And so, it’s all good. It’s all totally fine. But it strikes me, I’m kind of less worried from a constitutional standpoint about that data, as I am about, you know, the government finding out websites people are visiting or rifling through their emails, and somebody might have 30,000 emails, and the government wants to get one of them. And in the course of that they read the other 29,999. That’s the kind of stuff that I think is where the deepest privacy issues arise in the digital evidence setting.
So, considering all of the things that we’ve talked about, do you think that a challenge to geofencing today in the Supreme Court or the appellate court would be successful? Or would it not?
[ORIN KERR] 40:56
So, I don’t think the issue is whether geofence warrants would be allowed generally. I think the issue is, how broadly can the technology be used? So, it may be that geofence warrants don’t implicate searches at all. In which case, we’re dealing with less a process than a warrant being used to obtain this data. Or it may be that a warrant is required, and then, whatever the appellate courts say the particularity limit is [that limit] is sort of how broadly the warrant can extend. And, you know, in terms of what the U.S. Supreme Court might do, you know, it’s interesting with Carpenter in 2018, it was with a different court. And so, it’s not obvious whether the current Supreme Court would go the same way with Carpenter or would construe it broadly versus construe it narrowly. It’s going to be a really interesting issue once the next digital Fourth Amendment case gets up to the justices. Because, you know, you have Justice Ginsburg passing away, and then Justice Barrett arriving. And I think it’s not entirely clear where Justice Barrett is on these issues. And I think, you know, from an originalist standpoint, I can imagine justices having problems with Carpenter, but being very comfortable with limits on the particularity of warrants, because one of them is very much sort of rooted in the history of the Fourth Amendment. Another is a much more modern path. So, we’re just going to have to wait and see. It’s just, you know, I think we’ve got to watch both pieces, whether there’s a search at all, and if so, how particular does the warrant have to be? But it’s just really early to tell. We just don’t quite know where this is going.
And since we don’t know where this is going, I’m wondering how you think that the legislature should act on this issue?
[ORIN KERR] 42:58
Um, you know, it’s not obvious to me that this is a legislative question. You could always have legislatures address this issue. Usually, the role of the legislature is to enact a privacy rule that the public wants that is more privacy protective than what the Fourth Amendment might [protect]. And so, in order to answer that first usually you’d have a sense of like, well, what might the Fourth Amendment rule be? And that’s going to determine kind of the bookend there where you can’t be more invasive than whatever the Fourth Amendment rule is. But it can be somewhat hard to legislate in an area that is in sufficient flux, before you get some ideas to what the constitutional rule might be. You certainly could. It’s not obvious to me, given kind of at least my sense that the way these warrants are carried out, and the limited information that is revealed, it’s not obvious to me that as a normative policy matter, you’d want a legislative answer that limits these warrants. So, from a policy standpoint, you say, well, how beneficial is this technology? And how risky . . . what are the privacy threats to its use? And it may be that this technology is incredibly useful for solving a lot of really important crimes. And so that is very much in the public interest. And maybe you want that, and maybe the privacy risks are low. It may be that if the government can use this technique, they don’t do other more privacy-invasive techniques. Maybe it’s actually in the net better for privacy to allow geofence warrants so as to not have other forms of surveillance used. So those are the kinds of questions that you’d want to ask if you wanted to approach it from a legislative standpoint. But it’s not obvious to me that this is something that the legislature should be focused on right now. They always can. It’s always up to them. But it’s, this is not something where I think the Fourth Amendment isn’t doing anything. And we need some legislative answer, in particular, in part, because even if the Fourth Amendment doesn’t apply, we still have the Stored Communications Act, which is a legislative answer at the federal level. There are state equivalents. So, there’s always, there’s a statute, it’s just a question of whether you think that statutory standard is too low. That’s something I think people would disagree on.
And we’re just about wrapping up. I wanted to ask you, do you have any kind of final thoughts on this issue, or things that you would ask our listeners or the courts or something to consider when examining the issue of geofence warrants?
[ORIN KERR] 45:51
You know, I would just say, stay tuned for court decisions. I think there’s one case that’s been pending for a really long time, for like a year or something, involving a motion to suppress on this, but even that it’s just a district court case. You know, where you really get the legal action in this area is once you get up to the Court of Appeals, that’s where you start getting, you know, binding precedents. And that’s when the Supreme Court starts to potentially pay attention. So, this is just a really long, slow road until we get some kind of rulings on these questions. And I don’t think we can be confident about what the law is going to turn out to be given how hard this question is, for a while. So yes, be patient is probably my best advice.
Thank you so much for joining us today. It’s been a pleasure.
[ORIN KERR] 46:40
Thanks so much for having me on. Happy to do it.
There has been a dramatic increase in the use of geofence warrants by law enforcement in the U.S. Across all 50 states, geofence requests to Google increased from 941 in 2018 to 11,033 in 2020, accounting for a significant portion of all requests the company receives from law enforcement. These singular warrants can lead to the disclosure of a substantial amount of data for large numbers of people, effectively tracking every person within the parameters of the fence. Privacy advocates have raised objections to the use of such warrants, arguing that they may decrease the feelings of anonymity we rely on when we gather, for example, in protests or public demonstrations. These warrants have raised questions as to their interactions with the First and Fourth Amendments.
The Fourth Amendment was drafted to address the historical concerns with general warrants, allowing for indiscriminate searches and seizures. In our discussion with Dean Chemerinsky, he noted a possible equivalence between geofence warrants and those general warrants due to their limited discretion. Alternatively, Professor Kerr questioned whether these geofencing warrants truly raise a Fourth Amendment concern. This was exemplified in his analogy of a geofence warrant to the search warrant of a home with shared roommates, for which the warrant would not require individual probable cause for every tenant. Additionally, the use of geofencing may have the effect of chilling speech, possibly implicating First Amendment concerns. However, as we heard from Dean Chemerinsky, the Supreme Court has been inconsistent in its decisions as to when a chilling effect reaches the level of a First Amendment violation.
The state of the law relating to geofencing is still in its youth. The first Fourth Amendment challenge against a geofence warrant in United States v. Chatrie is still making its way through the courts.21 Due to the limited number of judicial opinions available on the topic, it is unclear how courts may address the constitutional questions raised by the use of geofencing warrants. Observers may look for guidance from the 2018 Supreme Court decision issued in Carpenter examining the Fourth Amendment’s protections of the reasonable expectations of privacy in the hole of one’s physical movements. The court, in a narrow decision, held that the government generally will need to obtain a warrant to access large amounts of cell site location information.22 However, the makeup of the Supreme Court has changed since Carpenter, and as noted by Dean Chemerinsky, it is unclear how the court will follow its prior decision, let alone if it will expand Carpenter in addressing geofencing warrants.
For now, the control appears to rest with Google to set the rules when law enforcement requests geofencing data. Google has set the standard by requiring warrants, with judges left to figure out what said warrants must address. Resting this control with a private party may be unnerving to those who would prefer to see explicit guidance laid out either by the legislature or the courts. As Professor Kerr explained, the use of geofencing data by law enforcement is expected to continue, and this use will likely push courts to address the standards that law enforcement must conform to in obtaining these warrants.
Thank you for listening. The BTLJ podcast is brought to you by Podcast Editors Seth Bertolucci and Isabel Jones. Our executive producers are BTLJ’s Senior Online Content Editors Karnik Hajjar and Thomas Horn. BTLJ’s Editors-in-Chief are Loc Ho and Natalie Crawford. If you enjoyed our podcast, please support us by subscribing and rating us at Apple Podcasts, Spotify, or wherever you listen to your podcasts. The information presented here does not constitute legal advice. This podcast is intended for academic and entertainment purposes only.