This post updates readers on the current status of the mosaic theory of the Fourth Amendment. As regular readers know, that’s the novel approach to the Fourth Amendment introduced by the DC Circuit in United States v. Maynard — and then suggested by the concurring opinions in United States v. Jones — by which an aggregation of non-searches and subsequent analysis of the collected data at some point becomes a Fourth Amendment search.
There has been a lot of litigation on the mosaic theory recently. I wanted to flag three recent developments: an oral argument before the Eleventh Circuit, a decision by the Foreign Intelligence Surveillance Court, and a novel procedural step by Magistrate Judge Facciola.
1) On Friday, the Eleventh Circuit heard oral argument in a case that directly raises the mosaic theory for historical cell-site data, United States v. Quartavious Davis. ACLU attorney Nathan Wessler presented argument, presumably drawing from the joint civil liberties group amicus brief. The groups filed a very similar brief before the Fourth Circuit in the Graham case. Davis came before Graham only because Graham is on hold pre-argument pending the Supreme Court’s decisions in Wurie and Riley.
As an aside, it’s interesting to me that the civil liberties group amicus brief opts not to use the phrase “mosaic theory” to describe its position. That’s an understandable litigation strategy for the pro-mosaic side. If you’re trying to get judges to adopt a fundamentally new approach to an important part of the Constitution, it’s probably better to downplay the newness of the approach — and its many difficult implications — by not labeling it as something new.
I don’t know much at this point about how the panel reacted to the mosaic arguments in Davis. Eleventh Circuit oral arguments aren’t available online. Plus, the AP story doesn’t give much in the way of clues. The story reports that Judge Sentelle of the DC Circuit was on the panel sitting by designation. Judge Sentelle asked the government a critical question: “Why isn’t that at least as much an invasion of privacy as a GPS driving down the highway?” That might mean something, but it’s hard to know what to make of one question by one judge. Plus, we know that Judge Sentelle was highly skeptical of the mosaic approach of Maynard before the Jones decision. In dissenting from denial of rehearing en banc, Judge Sentelle described the mosaic theory as “novel aggregation approach” inconsistent with Supreme Court precedent. He also rejected the basic thinking behind the mosaic theory with a blunt response: “The sum of an infinite number of zero-value parts is also zero.” Perhaps the concurring opinions of Jones persuaded Judge Sentelle to look at the matter differently. But if not, I would think he is a hard vote for the defense in Davis. Anyway, If I learn more about the argument, I’ll add an update.
2) As my co-blogger Stewart Baker noted earlier, Judge Collyer authored a recently-declassified opinion for the Foreign Intelligence Surveillance Court disagreeing with Judge Leon’s view for the DC District Court that the telephony metadata program violates the Fourth Amendment. Notably, Judge Collyer’s opinion directly rejects the mosaic theory around pages 24-30.
3) I’ve written several posts about Magistrate Judge Facciola’s string of warrant application denials raising novel theories of the Fourth Amendment. In light of those posts, I should point out that Judge Facciola has recently suggested that he may adopt — or, in his case, re-adopt, after being previously reversed for adopting — the mosaic theory of the Fourth Amendment for historical cell-site records. Here’s some context. Soon after Maynard was decided by the DC Circuit, Judge Facciola denied an application for historical cell-site records on the ground that the mosaic theory required a warrant for such records. DOJ appealed, and then-Chief Judge Lamberth reversed Magistrate Judge Facciola. According to then-Chief Judge Lamberth, the mosaic theory does not extend to access to historical cell-site records.
In his most recent order, Judge Facciola is looking into the issue again based on a routine DOJ application for historical cell site data. This time, Judge Facciola is reinvestigating the prior bases for a possible warrant requirement through a novel procedural step: He is planning to hold a mini-trial to conduct fact-finding that could effectively challenge at least some parts of the reasoning from Judge Lamberth’s opinion that reversed Magistrate Judge Facciola last time. In his new opinion, Judge Facciola asks EFF to represent the interests of the unknown subject of the records. He then describes his plan to to find facts — and have a trial if necessary — about several questions that he feels he must have answered to establish how the surveillance statutes and Fourth Amendment will apply in light of the state of caselaw. That caselaw includes what Judge Facciola revealingly describes as “the shadow majority opinion in United States v. Jones, 132 S.Ct. 945 (2012)”. (Pro-mosaic advocates sometimes call the two concurring opinions in Jones, viewed collectively, as a “shadow majority” opinion.)
As I see it, Judge Facciola is trying to overcome the fact that there is no factual record — and therefore no constitutionally ripe dispute — by having a trial on what the facts might turn out to be in some typical case. If I am reading the opinion correctly, Magistrate Judge Facciola then plans to apply the Fourth Amendment and statutes to the expected facts, presumably on the assumption that the application before him raises that typical set of facts. Here are sample factual questions that Judge Facciola plans to answer, as listed in his latest order:
How frequently is CSLI generated and how closely can it track a person’s location? Is it only generated when a phone call is made? If the government asks for only a narrow subset of CSLI, what does the provider actually give the government? What do the terms of service say about tracking a person’s location? Does any location information appear on a person’s bill? . . . .
• How precise is CSLI in an urban setting like Washington, D.C.?
• If the phone is in range of multiple towers when it makes a call, does it generate CSLI with each tower in range or only the tower that has the strongest signal?
• Is CSLI generated when the data functions are used? E.g., e-mail, Twitter, Facebook, web surfing, etc.
What kind of CSLI records are stored by the provider?
• For what length of time is CSLI stored?
• Does the provider have the capability to sort CSLI for any given phone number so that it provides CSLI that is made only for incoming/outgoing phone calls?
• When a phone “updates” with the network—independently of specific use by the user—is CSLI generated?
• How often does the average phone “update” with the network and connect to a cell tower?
• Is the collection of CSLI at all affected by the network that is used? E.g., LTE, WIMAX, 3G, 2G
• Are all phones certified for E-911 Phase II regulations per 47 C.F.R. § 20.18(h)(1)(I), (ii)?
• When a request for CSLI is received, what exactly is given to the government?
I don’t personally don’t think that judges can establish a ripe dispute by appointing an amicus and holding a mini-trial to establish future “facts” based on a hypothetical typical case. On the other hand, I suppose it’s better than Magistrate Judge Stephen Smith’s decision in very similar circumstances to just announce “facts” on his own based on his own independent research. (Incidentally, for a fascinating look at how Magistrate Judge Smith and Magistrate Judge Facciola consult each other and other judges on these questions, see this Washington Post story from last week.)
Anyway, we’ll see what happens. This should certainly be interesting to watch. Some have wondered why Magistrate Judge Facciola would try this approach after being reversed on the issue last time. I don’t know, but here are at least three possibilities. First, appeals from magistrate judges in the DC District Court go to the Chief Judge, and the DC District Court has a new Chief Judge as of July 2013. The appeals now go to Chief Judge Richard Roberts, not former Chief Judge Royce Lamberth, and it’s possible that Chief Judge Roberts will look at the issues differently. Second, at least some of Lamberth’s reasoning was about how invasive historical cell-site monitoring can be, and I gather that Judge Facciola’s focus on that question is designed to establish a factual record to help on appeal assuming that DOJ appeals Judge Facciola’s decision. Third, in ex parte litigation, the case ends when the government first wins. The only way to get an ex parte case to the DC Circuit that invented the mosaic theory in Maynard is to present it in a way that persuades the district court and then prompts another DOJ appeal to the circuit court.
For Fourth Amendment nerds, we certainly live in interesting times.