What's at Stake in Chatrie v. United States

Audio of this conversation is available via your favorite podcast service.

At the end of last month, the Supreme Court heard oral argument in Chatrie v. United States. The case involves the use of a geofence warrant, which police use to demand information on all cellphones within a certain area and period of time. The outcome of the case, which revolves around Fourth Amendment questions, could have profound implications for location tracking and privacy in the digital age.

To learn more, Tech Policy Press fellow Jake Laperruque, who is monitoring the case, spoke to Michael Price, who serves as litigation director for the Fourth Amendment Center at National Association of Criminal Defense Lawyers (NACDL), one of the lawyers representing the plaintiff.

What follows is a lightly edited transcript of the discussion.

Jake Laperruque:

I'm Jake Laperruque. I'm a Tech Policy Press fellow for this year and also the Deputy Director of the Security and Surveillance Project at the Center for Democracy and Technology. I'm here today talking with Mike Price, who's an attorney working on what's been a really interesting and exciting privacy and surveillance issue this year, which is geofence warrants, a topic that recently was discussed before the Supreme Court. Mike, really thank you for joining us. Could you introduce yourself to our audience, who you are and what kind of work you do?

Michael Price:

My pleasure. Thanks for having me, Jake. I'm Mike Price from the Fourth Amendment Center at the National Association of Criminal Defense Lawyers. We work on cases just like this. We provide litigation support to defense lawyers around the country dealing with issues of new technology and privacy.

Jake Laperruque:

Well, we're really happy to have you here today be talking about this issue. So the case I mentioned that the Supreme Court recently heard is one called Chatrie. And this case deals with a surveillance technology, geofence warrants, that have recently in the last decade emerged, but then become a huge widely used surveillance tool. So to start us off, could you just briefly explain what geofence warrants are, why they're so sensitive and important, and what the biggest privacy dangers are from them?

Michael Price:

Let's take one step back and say geofence warrants are a type of what we call reverse warrants. And to understand that, imagine what you would think of as a regular warrant where the police identify a suspect and they establish probable cause to go and search their data and they get a judge to sign off on that. The police go in with a warrant. Now imagine that process in reverse. The police have no suspect, they don't know who did it, but instead they want to go and search everybody's data and come up with a suspect later. It's like the process in reverse. And to do this, in this case, we're talking about something called Google location history. That's a feature that Google has. About a third of Google users have it opted in on their phones, about 592 million people. And it's a journal of where you've been. It's stored in your Google account just like your documents and your photos and your email.

And what happens is the government goes to Google and they say, "Well, tell me everybody who was around the scene of a crime at this particular time and this location." And they search all 592 million accounts to come up with that data. That's step one.

Then they decide who they think they want to have more information on and they lift the geographic restrictions. They get extra time. They can see where people are going to and coming from and then they get more information and they say, "Well, tell me who these people are. Give me their subscriber information." That's the reveal and step three.

And so what we're arguing here is that that location history data is private because the government here says that it isn't. Of course, that implicates other types of data that we store in the cloud. And we're also saying that this is a general warrant. It's a digital dragnet. The government shouldn't be able to search everybody all at once. They should have to identify particular accounts they want to search, develop probable cause just like normal. So there's a lot at stake here, there's a number of different issues, but one of the big ones is can the government just sort of troll through all of your data that you have stored on the cloud? If they can do it with location history data, what about email? Can they search everybody who emailed about a certain phrase? So those are the potential privacy stakes here.

Jake Laperruque:

So before we dive in a bit more to the history and particulars of this case, I do want to ask a little more about geofence warrants. How many people can they end up sweeping up in this sort of reverse search process?

Michael Price:

So at the beginning of the process, it searches everybody. Everybody who has location history enabled, which Google has told us is about 592 million people, about a third of all Google users. And that runs against everybody. It doesn't matter whether you were there or not. Google has no way of determining that ahead of time and they have to go in and take a peek at everybody's account just to check. So at the very first step, everybody is implicated.

At the second step, that number drops and it's approximately about a third of the cell phones that were in the area at the time. And that number can range from the dozens to the hundreds depending on the area and the length of time. Here in the Chatrie case, it was 19 people who had their information returned to the government after that initial sweep.

And then from there, the government actually asked for information on all 19 of them and Google had to shout them down and say, "No, be more reasonable." And so they came back and they said, "How about these nine?" And then Google said, "Okay." And they got more information about those nine and then they decided there were three more that they were really interested in. And so those were their finalists. Obviously, only one of them was involved in this case.

Jake Laperruque:

And so in this case, it involved a bank robbery. Were all those people being scooped up by this geofence warrant, people in the bank or at that specific building where this crime happened?

Michael Price:

Yeah, that's a general idea is that it's trying to identify literally everybody in the area at the time. I mean, it's limited by who has location history enabled, but the basic idea is to get information on everybody who was in the vicinity of that crime, whether they were involved or not.

Jake Laperruque:

In this case, sort of vicinity ended up being a little broader. I mean, we weren't talking about just one building.

Michael Price:

Right. So in this case, this case involved a bank robbery, but the geofence here didn't just cover the bank. It covered the entirety of the Journey Christian Church next door along with its entire parking lot. And the data returned as a result of this stretched much further, even outside of the range where the government said that it was going to reach. And we're talking about people in hotels, hospitals, homes all over the place, in sensitive locations that had absolutely nothing to do with the robbery and couldn't have.

Jake Laperruque:

I was really struck by the amicus brief that Google submitted for this case where they outlined some other geofence warrants they've received that along the lines of this case, but sometimes even more severely, didn't just go to the spot where a crime occurred or the building, but entire neighborhoods, entire sections of towns. It seems like the parameters of how these geofence warrants, how much they can sweep in broad sets of people and the location data is really potentially daunting.

Michael Price:

That's absolutely true. This is a relatively early geofence warrant in the scheme of things. And after we had some initial wins, the government started changing tactics a little bit, narrowing the time or the area covered, but also often adding additional locations, trying to cross-reference numbers that might appear in one location or two locations or three locations and actually asking Google to do that work for them, putting them in the position of a deputy.

Jake Laperruque:

So as far as the work that Google does in these cases, in this case, Chatrie, the geofence warrant is sent to Google, Google's the main known recipient of geofence warrants. But in the last couple of years, they've actually changed their data retention policies so that they no longer have the location data stored in the company that responds to them. All that information is now just on users' own devices. So in light of this, well, why is this case still important? Some folks might ask, "Well, if Google's the one who receives all these geofence warrants and they now say that they can't respond to them, doesn't that kind of moot the issue or mean that this is no longer relevant?"

Michael Price:

So I should start by saying that what Google did was really significant I think both practically and legally. Practically because it really does make future geofence warrants impossible, at least with Google. And if we haven't explained it, Google changed its policy to say, "You know what? We're not going to store all this location history data, which is user data on our central servers anymore. We are going to from now on store it on people's phones directly if they choose to keep it and we're not going to have a copy of it anymore." And so it's therefore impossible for Google to respond to a geofence warrant anymore. Google has changed the way that they store copies of this data such that if the government wanted to search everybody's location history data now they would have to go and get a warrant for everybody's cell phone with location history data on it.

That is significant, obviously, from a practical perspective, but legally because we're having this fight over whether location history data is private, whether it's somebody's digital property. And of course, the Fourth Circuit here said that there is no privacy interest in location history data because you have it stored with this third party company to which Google replied here, "Hold my beer." Days later after the Fourth Circuit argument where the court seemed to indicate where they were going, Google demonstrated this by deed, transferring location history data or announcing that it was going to transfer location history data from its servers to people's phones directly. So it's saying it's showing this is user data, it's private data, it's not our data, it's not a business record. And so I think that is significant, but we're still having these fights. There's a time lag. Criminal cases are still pending. Mr. Chatrie's case is still pending and that was one of the first ones. And I think the underlying questions here are not going away because of Google's policy change.

Jake Laperruque:

I think we could all definitely hope that more companies take a hold my beer and protect user privacy approach across issues. Do you think there's other companies and services that would also be getting geofence warrants in the future potentially that might not be taking up that sort of same proactive policy Google did?

Michael Price:

So in the past, I mean, we know that companies like Uber and Lyft and Snapchat have all received geofence warrants. So has Apple. But those companies have either not been able to provide, I guess, granular or comprehensive enough location data, or they just in Apple's case, for example, don't keep it. That does not mean that we are not seeing these types of warrants still.

After Google's policy change in particular, we saw a huge uptick in something I call Frankenwarrants or what the government calls area dumps or area searches, which are more or less the same sort of thing issued to a cell phone service provider. So AT&T or T-Mobile for some reason, not Verizon yet, but it's using a different type of cell phone data than your sort of regular cell phone records, something called timing advance, which is maybe another podcast, but that's not optional. And this is not an opt-in or opt-out thing. Everybody that has a cell phone is transmitting this data and the government is using it in a geofence-like way. So I don't think we're done with this issue just because Google changed its policy, but it is a welcome change.

Jake Laperruque:

So Mike, you mentioned also that the Fourth Circuit, the court below that considered this before the Supreme Court, that some of the justices there said there wasn't even a privacy right in location data. But the Supreme Court back in 2018, they issued a rule in another case, Carpenter, that specifically took on location tracking for cell phones and said that there was a Fourth Amendment right there and that you need a warrant for continuous tracking of people's cell phones. So what sort of was the reasons why the lower court distinguished that and what would be the ramifications if the Supreme Court said that location tracking in a geofence warrant for one of those reasons wasn't Fourth Amendment protected?

Michael Price:

Right. So the Fourth Circuit in this case wasn't actually able to reach a determination one way or the other on really any of the issues in this case. They issued a one sentence per curium decision affirming the district court's judgment and that was it aside from another 120 pages of 9 concurrences and dissents. And there weren't eight votes there for any of the major issues in this case, including whether location history data is private. And I think that is one of the most critical issues in this case because without that finding, as the government argues, no warrant at all is needed to search this data.

Jake Laperruque:

So it wouldn't even be a geofence warrant in that case. It would just be a geofence demand that gets handed over without court approval, with some very low standard maybe even.

Michael Price:

Right. I think the government might've even said at some point that the reason they're seeking a warrant at all is just because, well, Google asked them to and wasn't going to comply otherwise. That may speak to another set of power dynamics, but Google was saying, "We're just not going to comply unless you get a warrant." And so the government did, but in court they're saying, "Actually, we never needed to get one," because this data, they believe isn't private at all and should just be available to them.

And of course the problem there is that we're talking about location history data here, but Google doesn't treat that any differently than other types of user data. It stores it in the same way in your account, in your password protected account, in your virtual digital locker in the same way that it stores your emails and your photographs and your documents and your other messages. And so if there's no privacy in location history just because it's on a third party's servers, then what about all of this other data? And I think those implications have been echoing throughout this case and certainly did at argument.

Jake Laperruque:

So yeah, I mean, this kind of brings up the third party issue and something called the third-party doctrine. Could you explain a little about what that is and how that plays into this question of whether this data is private or not?

Michael Price:

Right. So in the 1970s, the Supreme Court issued a pair of decisions about bank records and telephone numbers dialed saying that because that information was voluntarily shared with a third party, I.E. the bank or the phone company, that you didn't have a privacy interest in it and the government didn't need a warrant to go and search it. And that became known as something called the third-party doctrine, which basically holds, although the court has never really extended it, that information that you share knowingly and voluntarily with a third party loses its Fourth Amendment privacy interests. And obviously that doctrine taken to its logical conclusion today with the amount of information that we all store digitally on the cloud has tremendous implications. And so that is one of the central issues in this case. Is this data private? Let's say it's a journal, just a written journal of where you went and you write this down. Is the fact that you are storing it in a virtual private locker with Google, does that take away your privacy interest in it? Obviously we are arguing no.

We're also making an important other argument here about who owns that data and saying that, well, we believe and Google believes, and pretty much everybody else believes that that kind of data that you create yourself and you store with Google or a company like them is your data. And Google says that too in their terms of service and their privacy policy and in the briefs that they filed in this case, they say, "This is not a business record. This is not our bookkeeping. This is your data that we're holding onto for you." And in that sense, even a little intrusion, no matter whether we're talking about a day or two or seven or a couple of months of location history data, any intrusion is a Fourth Amendment violation because that is your data and the government can't search it without a warrant.

Jake Laperruque:

I mean, the justices did seem to understand the magnitude of this problem. They mentioned a lot of different digital records and cloud stored items that have become really common to life, email, online calendars, online photos that you store, online documents. It seems pretty clear at this point, it's just that sort of third-party doctrine is just not really compatible with a digital world. I mean, do you think we're headed in the direction away from that and maybe towards some sort of new standard along the lines of what you're saying of is the company using this as a business record versus is it acting sort of as a steward and just holding something that's really your record? Would that be a better way to do this?

Michael Price:

So we're not actually asking the court to overrule the third-party doctrine, but I think it is important to recognize that the court has never actually applied it outside of the 1970s and certainly hasn't applied it to digital records. And so just as the court declined to apply it in Carpenter to cell site location data, we think that this is similarly private enough to fall outside of that exception, and of course that it is user data and any trespass on that should also trigger a warrant requirement. So that's where we are.

Jake Laperruque:

So I want to dive into some of the arguments and discussion that we had from the court last month, but just one final question about the merits of the case first. You mentioned that this is part of a class of reverse warrants and Frankenwarrants. So what other other similar surveillance technologies and techniques exist in that broader category beyond geofence warrants? What are they and how do they tie into this case even if they're not geofence warrants themselves?

Michael Price:

There are lots of other types of location-based searches that exist out there. Geofence warrants are just one type. We mentioned area dumps a second ago. Tower dumps are another. They didn't come up with an argument, but that was fin by us. And I think there are some important similarities and differences between the different types of location tracking technologies. And I imagine we're going to continue to see litigation on that front regardless of how the court rules in Chatrie. Geofense warrants in this context have some really particular attributes and whether the court focuses on those or on the issues more broadly, I think will determine the impact on other types of surveillance technologies. But technology is advancing very, very rapidly, more rapidly than the courts are able to keep up with the law. So any decision that the courts make here is unlikely to capture the full extent of what we're likely to se next. So I think some guidance is definitely necessary, but I don't know that we're going to see a ruling that solves or answers all of the questions out there.

Jake Laperruque:

The justices mentioned a couple of times cell tower dumps as sort of another type of reverse warrant and another location tracking technology, but not much else in terms of these other reverse warrants. Did that surprise you or did that seem to fit with how the case is being analyzed?

Michael Price:

Like I said, I think geofence warrants have some similarities and some differences with tower dumps, but those are significant and I understand why the court wouldn't want to tackle both at the same time.

Jake Laperruque:

In general, did the arguments go the way you expected or were there things that surprised you about the way that the oral arguments played out?

Michael Price:

I don't know that oral argument ever goes the way you expect, but I guess at least in my opinion, there weren't any big surprises in the questions. Remember, this is an issue that split circuits and state supreme courts. The Fourth Circuit couldn't come to an agreement on any of the issues. So we had some tough questions and some helpful ones, but nothing that we weren't prepared for.

Jake Laperruque:

So when you're preparing for arguments like this and putting together the briefs and planning out what are going to be the main points, federal arguments, and how you're going to respond to different questions, how much are you and the team you worked with shaping your arguments around specific justices versus just trying to lay out what you think is the right case overall and convince them as a group?

Michael Price:

It's interesting. You have to remember that a lot of the cake was baked back when we started this case in 2019. So the Fourth Amendment Center along with the federal defenders in the Eastern District of Virginia have been representing Mr. Chatrie since the initial motion to suppress back in 2019. And that was just a year after the Carpenter decision and a year after Justice Gorsuch filed his dissent in that case, which I sometimes characterize as a concurrence because his gripe wasn't that the outcome was wrong, but that the lawyers below didn't argue that he had a property interest in his cell phone location data, only a privacy interest.

And one of the things that we were very careful to do here was to make that record in the trial court establishing that Mr. Chatrie had a property right in his data. And we subpoenaed Google, we had them on the stand, they filed briefs in the case and said the same thing, that it is user data, it's not a business record, it doesn't belong to Google. And so we made our best attempt at a good record thinking that this might be one of the cases that reaches him.

Jake Laperruque:

So since that Carpenter case back in 2018, which was the last big Fourth Amendment case and issue of sort of surveillance and technology, an entire third of the court has turned over. We have three new justices, Justice Kavanaugh, Justice Barrett and Justice Jackson, who this is really the first time that we've seen them take on a Fourth Amendment and technology case. How do you plan arguments directed to those justices who you haven't seen talk about this issue or haven't written as a Supreme Court justice on these types of issues before? And what did you think of the way they were reacting and the questions they were asking in arguments?

Michael Price:

I think one of the most important things to do in the Supreme Court or in district court is to have the judges realize that this is a privacy issue that could affect them too. And we kind of saw this in the Jones case back in 2012 when Chief Justice Roberts asked the Solicitor General, "Are you saying you could put a GPS tracker on my car in the Supreme Court parking lot right now?" And the government kind of sheepishly answered, "Yeah." And I think that was the moment they lost that case. It happened in the district court in this case too. I remember after we had Google on the stand after testimony in district court and we took a recess, the court clerk came up to me and said, "Can you look at my phone and make sure that this stuff is off?" And that made me think that what we had done had resonated. And so here I thought it was interesting when Justice Barrett said that she had to go and check the location history settings on her phone, that to me signaled that we had conveyed our message.

Jake Laperruque:

Yeah, that's really interesting because it's a common sort of concern both for courts when you're talking to judges and also with lawmakers of how much do they understand new technology. But like you said, it did seem like a lot of this was resonating of this isn't some sort of complex, convoluted new tech. This is the ins and outs of daily life, these types of tools. It does feel like that is resonating and it's interesting to hear that the thought of keeping the stakes personal is important even at some level here where the stakes are so high. So two of the justices, Justice Alito and Thomas, seemed like they were not really eager to even have this case in front of them. They said that basically you guys were asking for an advisory opinion because changes in the Google policy that we talked about and also something called the good faith doctrine rendered this case moot. So could you talk a little bit about what the good faith doctrine is and why you think it shouldn't, presuming you guys don't think it should, block off a ruling here?

Michael Price:

Right. Yeah, this came up at the very beginning of the argument. Justice Alito, I agree, did not seem pleased that the court was hearing this case, but as we argued Google's policy change happened back in 2023. The court was well aware of it before granting cert and it certainly doesn't change the facts of Mr. Chatrie's case or anyone else caught up in other geofences.

The court also specifically declined to grant cert on the good faith question, and the good faith doctrine just briefly, it's the freebie rule for police. It pretty much says that law enforcement should get a pass on the first unconstitutional warrant if they didn't know it was unconstitutional at the time. And I guess importantly, it allows courts to avoid reaching difficult Fourth Amendment questions and developing the law, but that may be a different podcast.

Here, the district court and the Fourth Circuit ultimately determined this case on the good faith doctrine and said that even if this warrant was unconstitutional or even though this warrant was unconstitutional, we're not going to suppress the data in this case because of the good faith doctrine and the court declined to hear argument on that particular issue. I think it is not moot, certainly if we had a win here, Mr. Chatrie would be raising this issue on remand and the court would have to make a determination about whether it applies or not under these circumstances and given the Supreme Court's ruling. So we of course did disagree and do disagree that it is moot and hope that the court does provide some guidance on these issues.

Jake Laperruque:

It does seem like this notion of, well, we don't need to actually say what the rule should be here because this freebie exception is probably going to apply, would create a very dangerous catch-22 because police are always going to be acting in good faith and they're not going to know what the law is on these novel tech questions until the courts step in and say when a surveillance technology is so powerful and what exactly the rules are for. And I don't see if that's how the law was applied, how we ever get out of that cycle.

Michael Price:

Exactly. By pulling a good faith rip cord, judges can avoid making rules, but those rules would guide law enforcement for the next time and they don't have that guidance so the process repeats. And because technology advances more quickly than the law, any ruling now, say even on geofence warrants, may not be directly applicable to the next thing, say Frankenwarrants. And so we end up with good faith again and we end up without any development in the law, even as technology is changing as rapidly as it is.

Jake Laperruque:

In general, this does seem to be a perpetual problem with the law not keeping up with tech. I mean, you mentioned this case spans all the way back to 2019. The Carpenter case that talked about long-term cell phone tracking, that was 2018, that took years to go through and that was a tracking technique that was already underway when the Jones ruling on GPS trackers you mentioned was decided in 2012. I mean, do you see this as sort of a serious problem we need to fix or is this sort of the system slowly working its way out but on delay? Because it seems like we're always a bit behind the ball.

Michael Price:

I think the good faith doctrine is one of the main reasons we're always behind the ball because it provides this easy out to avoid deciding the substantive questions, which can be difficult. And it involves new types of technology every time around and maybe a novel application of those to existing law and without additional guidance from the court, without timely guidance from courts, I find it difficult to see how we get to a modern Fourth Amendment.

Jake Laperruque:

So do you think a revised view on the good faith doctrine is the solution on that to make things work a little more quickly or are there other sort of-

Michael Price:

Absolutely. And NACDL has filed amicus briefs on that point in particular warning about the application of the good faith doctrine and the way that it stifles development of Fourth Amendment law and privacy rights.

Jake Laperruque:

Justice Alito was one of the main justices raising this point about the good faith doctrine and seemed less interested in the merits of this case. He's been a bit enigmatic on privacy issues because back in 2012 in that Jones case on GPS trackers, he issued a concurrence that was a really strong opinion supporting privacy rights and broad protection of privacy rights. And that seemed to shape a lot of the court's future rulings on this, but by the time we get to 2018 in Carpenter, he actually dissented in that case and took a very narrow view on location tracking and now here several years later, he seems like he doesn't want to be talking about the merits one way or the other.

I'm kind of curious both for that shift and in general, do you think justices' views on issues that don't really fall into a neat sort of liberal conservative doctrine divide like privacy issues and also issues where there's these types of new tech and new tools that keep changing so much so rapidly, do you think that makes things more prone to ideological shifts between justices than they would normally have? Or do you think this was just kind of a standard thing that could happen?

Michael Price:

I can't say what Justice Alito was going to do here.

Jake Laperruque:

In general, do you think that new technologies like this make things more prone to there being less consistency for how judges and justices will rule over and over?

Michael Price:

Well, I think it is really important to recognize that the landmark digital privacy cases that we've had over the last 20 years, Jones, Riley, Carpenter haven't split along traditional ideological lines. In fact, they didn't split at all some of them. Jones was 9-0, although they disagreed about why. Same with Riley. Carpenter had some dissents, but as we talked about, Gorsuch's was kind of in name only. He wanted that property argument to be raised. And I think more generally, these issues aren't political in the same way that maybe some of the other courts' cases are. And we have seen justices who disagree on most things who come together over privacy issues. I think anytime you get people to recognize that this can affect them, those political concerns become secondary and the political wins may change direction, but nobody wants a government that can invade their privacy without strong checks.

Jake Laperruque:

So we don't know how this ruling's going to go. We'll see when the opinion comes out at some point this spring or summer, but what do you see as the next big issues on the horizon? What are going to be the most important and urgent Fourth Amendment and tech issues that courts are going to be focusing on in the near future?

Michael Price:

I think that regardless of how the court rules here, there is going to be lots of analysis and litigation over what it means and not just for geofence warrants, but for other similar types of searches. But if things go sideways from my perspective, I think it's important to remember that things like ECPA, like the Electronic Communications Protection Act from 1986, the Right to Financial Privacy Act, RFPA were constitutional responses or congressional responses to the Supreme Court's constitutional decisions in those third-party doctrine cases. So they recognized that those were bad decisions from a privacy perspective and passed laws to establish their privacy and regulations for government access to them. But I would say lawmakers do not need to wait. It has been 40 years since ECPA in 1986. That was I think four years before the Worldwide Web existed. Our privacy laws are in desperate need of an update and I know that there is bipartisan support for it in Congress and there has been for years.

Also along those lines is the Fourth Amendment Is Not For Sale Act. I think that is incredibly important even if we get a good win in Chatrie because what the government can do right now through data brokers and buying this information from private companies is the same kind of thing just for a fee from a portal at police headquarters. So I think legislation is desperately needed here and it shouldn't be that we have to go to the Supreme Court to get a case every 10 years.

Jake Laperruque:

Well, hopefully it won't be another 10 years before the next case because as you're describing, there's a lot of really important and really fascinating issues. We'll be keeping our eye on those and we'll be keeping our eye on to see how the court rules on the Chatrie case. Mike, thanks so much for taking the time to chat with us about it and thanks for all your work.

Michael Price:

Thanks, Jake.