The Third Circuit held oral argument today in the case on the legal standard for historical cell-site information. I blogged about this important case here last week, and the oral argument audio from this morning has been posted here at the Third Circuit’s website. It was a very unusual and free-ranging argument that went for almost an hour-and-a-half, more than twice the scheduled 40 minutes. I wanted to blog my thoughts on the argument as I listened to it, as well as provide some of the key exchanges based on my best effort to transcribe them.
To set the scene, the presiding judge is Dolores Sloviter, a former Chief Judge of the Third Circuit, together with Judge Wallace Tashima, a visiting judge from the Ninth Circuit. Judge Jane Roth was also assigned to the panel, but she was absent from the hearing. Judge Sloviter is a Carter appointee; Judge Tashima a Clinton appointee; and Roth a Bush 41 appointee. The lawyer for the United States is Mark Eckenwiler, DOJ’s guru on electronic surveillance. Kevin Bankston from EFF and U of San Francisco lawprof Susan Freiwald argued as amici. This is a very strong group of advocates: All three work in this area regularly and know their way around these issues.
Here are my thoughts about the argument as I listen to it:
Around 5 minutes in, Judge Sloviter has lots of general questions about the Stored Communications Act. Her questions are somewhat surprising, as they concern sections of the statute that are not actually implicated by this case: Judge Sloviter is asking Eckenwiler to opine on why Congress drew a distinction between 2703(a) and 2703(b), while this case is about 2703(c). (I answer Judge Sloviter’s questions in my article A User’s Guide to the Stored Communications Act, if you’re curious.)
At the 15-minute mark, Judge Sloviter contends that the technology is advancing so that cell phones are giving more and more accurate location information. Eckenwiler contends that this isn’t accurate, and takes the court through the history of the FCC regulations on Enhanced 911 and the resolution of cell-site tracking. The oral argument is actually really technically interesting; there’s a lot of interesting stuff about the government’s ability to conduct cell cite surveillance. (As an aside, Eckenwiler really is a guru on this stuff, both as to the law and the technology: He was my neighbor in the Computer Crime Section in my first year at DOJ, and I spent a ton of time in his office annoying him with questions about ECPA, the Fourth Amendment, and the like.)
At the 23-minute mark, Judge Tashima asks if these sorts of records are routinely kept, and thus available to be disclosed. Eckenwiler says that they are not. Judge Sloviter comments that such records are kept for at least 180 days, which reflects a misunderstanding of the statute: The 180-day distinction is about legal standards for compelling contents if kept, and Judge Sloviter seems to think it is about a requirement to keep non-content records.
The most interesting exchange happens at the 28-minute mark. Judge Sloviter has been asking lots of questions about the technology, and she explains that she wants to say why she is asking so many technology questions. Judge Sloviter:
Let me tell you. As I work on this, I listen to the news. And you know that there are governments in the world that would like to know where some of their people are or have been. For example, have been at a what may be happening today in Iran — have been at a protest. Or at a meeting, a political meeting. Now can the government assure us that (1) it will never try to find out that information and (2) whether that informaton would not be covered by [18 U.S.C. 2703](d)?
Eckenwiler responds that he can’t speak to future hypotheticals like that. Judge Sloviter responds:
But don’t we have to be concerned about that? If the statute would permit the government — not this government right now, but a government — to get information as to where, and it doesn’t have a GPS, but this could be instead of a GPS, wouldn’t the government — a government — find it useful, if it could get that information without probable cause?
Eckenwiler then responds by discussing Supreme Court Fourth Amendment caselaw which indicates that this information isn’t protected by the Fourth Amendment. Judge Sloviter responds that she is looking to express her concerns through a reading of statutory law:
But I’ve set aside the Constitution. My concern — and I can’t talk for Judge Tashima — if we can decide this on a statutory basis, aren’t we obliged to do that rather than hypothesize what the decision would be under the Fourth Amendment? . . . And the Congress knew how to say when you have to come up with a warrant, and that’s a constitutional requirement. And it did that in [2703](a), when it was looking at content. It didn’t say a warrant in [2703](d). And that isn’t that what concerned Magistrate Judge Lenihan?
I’m not sure where Judge Sloviter is going here. She seems to want to read the statute as requiring a warrant to address her concerns with the possibility that a future government might monitor political protesters. But the statute is clear that no warrant is required, and besides, the government couldn’t monitor political protesters under 2703(d)’s Terry order standard anyway.
Around the 36-minute mark, it’s interesting that the panel is still so focused on the technology. I don’t think think they have asked Eckenwiler a legal question about the terms of the relevant section of the statute or the relevant Fourth Amendment law yet.
At the 38-minute mark, Judge Sloviter explains that she has been on the Third Circuit for more than 30 years, and she thinks that magistrate judges “are not very grudging” about issuing warrants. She seems to think that judges will sign applications for warrants, so the government should just get warrants for these records. She asks, “Why do you want to make the point that you don’t have to show probable cause?” Eckenwiler makes the point that the statute requires a lower threshold — that this is what the statute says. Judge Sloviter responds, “okay, let’s go to the statute.” Interestingly, I think this is the first time the actual part of the statue at issue here has been raised, and we’re 40 minutes in to the argument.
At the 40-minute mark, Judge Sloviter suggests that judges have discretion to decline applications under 18 U.S.C. 2703(d). Here’s the language of the statute:
A court order for disclosure under subsection (b) or (c) may be issued by any court that is a court of competent jurisdiction and shall issue only if the governmental entity offers specific and articulable facts . . . .
Judge Sloviter asks, doesn’t the use of the word “may” suggest that judges have the discretion to decline applications if the judge, at his or her discretion, wants the government to get a warrant? I think Judge Sloviter’s suggestion is wrong here. The use of “may” reflects the fact that judges need authorization to issue specific kinds of orders. 18 U.S.C. 2703(d) orders were created in 1994 for the first time; The section confers the power to issue these orders on any court of “competent jurisdiction,” a term defined elsewhere in the statute. The use of “may” creates the power to issue the order; it doesn’t signify that the judge has the option to do it at the judge’s discretion.
At the 46-minute mark, Eckenwiler makes a critical admission: He has never seen the HBO series The Wire. Judge Sloviter is surprised. (I am, too. It may be the best show on TV ever.) Eckenwiler declines to mention that at least he watches a lot of Jeopardy!; he even appeared on the show twice.
Kevin Bankston starts at the 48-minute mark. He argues that 2703(d) orders are optional, and that courts should construe the statute to avoid the Fourth Amendment problem by allowing judges to say a warrant is required. The discussion then turns again to the technical question of how accurate cell-site information actually is. Bankston is arguing that cell phone cell site data is protected under Karo; if you make a call from a home, the cell site information could reveal the phone is inside the home, information about the home. I don’t think this works under Smith v. Maryland, though: In response to the claim that numbers dialed from a home revealed information from inside the home, Smith responded, “Regardless of his location, petitioner had to convey that number to the telephone company in precisely the same way if he wished to complete his call. The fact that he dialed the number on his home phone rather than on some other phone could make no conceivable difference, nor could any subscriber rationally think that it would.”
At the 60-minute mark, Judge Tashima says he finds it hard to read the statute as vesting discretion in the magistrate judge as to what the standard is to issue the order. Bankston responds that the discretion is bounded, as there are only two standards — the Terry equivalent or the warrant equivalent. Bankston argues that the plain language of 2703(d) vests that discretion in the msgiatrate by saying that the language “may issue.”
Susan Freiwald goes at the 66-minute mark (she is an amicus together with EFF). She is arguing that the Fourth Amendment requires a warrant for this information. Judge Tashima asks if Freiwald agrees with Eckenwiler’s assessment of how accurate the cell site data is; Freiwald responds that it’s complicated, and it’s not really clear how accurate such data can be. Freiwald also argues that cell cite information is distinguishable from Smith and Miller because the information is generated without the knowledge and control of the user: the user knows he is transferring numbers to the phone company to place the call, but he doesn’t know that his cell phone is transferring location information to the company to keep in touch. She also argues that the information revealed via cell cite data is more intimate than the numbers dialed in Smith, and that the government should be able to satisfy the minimal requirement of a warrant.
Eckenwiler is back at the 75-minute mark for his rebuttal. Judge Sloviter has questions about the Fourth Amendment issues, and whether and when the government needs a warrant to use a tracking device. Eckenwiler explains Karo and Knotts, but I think that this case is all about Smith v. Maryland. People who have cell phones know they are there; that’s pretty different from a tracking device or GPS case, I think. That’s why in my view the discussion of the resolution of cell-site information is beside the point; That’s the standard that applies when the person is unaware of the device that is sending the signals, whereas Smith is the case on point when the user is aware of the device.
All in all, I found the argument puzzling, and the likely path of the decision quite uncertain. Judge Sloviter was obviously quite skeptical of the government’s case, but I wasn’t sure what hook she might try to find (if any) to address her concerns that a government could abuse this power if it had to get reasonable suspicion orders instead of probable cause orders. Judge Tashima didn’t show his cards as much, but I thought he seemed to lean the government’s way. If Tashima and Sloviter disagree, the key vote becomes Judge Roth: Judge Roth wasn’t present at the argument, so it’s hard to know where the case might be going.