Is it really constitutional for the National Security Agency (NSA) to collect “metadata” on Americans' phone usage?
You've probably seen talking heads shouting back and forth about it, and wondered whether someone could just please give you a straightforward, accurate, thousand-word primer on the underlying constitutional doctrine.
Well, here's your primer. I hope it's helpful, at least as a starter for cocktail party conversations.
The Fourth Amendment forbids “unreasonable searches.” In my last Viewpoint essay, I described how “reasonableness” has been applied to various types of searches.
The NSA program implicates the concept of a “search.”
Is it a “search,” under the Fourth Amendment, for the government to monitor and collect “metadata”—namely, records of what phone number called what phone number, and for how long?
Here's the answer: no, it's not a search—not under current case law.
But the governing case is 35 years old, and if the courts take seriously the implications of the 2012 case United States v. Jones (a case about GPS tracking) the law on metadata analysis may be ripe for change.
First, why on earth is it not a “search” under current law to keep a record of every phone number everyone dials? Because of a case called Smith v. Maryland, decided in 1979. The Court had previously held, in Katz v. United States (a 1967 case about wiretapping of phone booths), that what makes a government action a “search” or not is whether the government intrudes on a “reasonable expectation of privacy.”
Listening in on a phone conversation, said the Court, is a search.
Smith raised the question of “pen registers,” or calling histories, generated by the phone company for a particular number: if it's a “search” when the government listens in on your calls, is it also a search when the government gets a record of the numbers you've called?
No, said the Court, it isn't, for the simple reason that phone users reasonably understand that in order to avail themselves of the service provided by the phone company—connecting them to another phone—they have to tell the phone company what number to connect them to.
Perhaps the Justices were remembering their own experience with phones, which, for most of their lives, would have involved speaking to a human operator and requesting a connection to a particular number. Whether the process is human or automated, you still have to tell the phone company what number to connect you to.
Now wait, you say: the fact that you reveal the information to the phone company doesn't mean you want it handed over to the government. That's a perfectly natural response, shared by most Fourth Amendment commentators and a number of state courts—but not by the Supreme Court.
The cases hold that it is not a “search” for the government to obtain information that you have voluntarily shared with a third party, and that very much includes calling metadata.
Since obtaining a call history is not a search, the familiar Fourth Amendment requirements of a warrant, probable cause and so on, are not implicated when the government collects the data.
Now wait, you say again—remember 1979? Computers as we know them barely existed. There was no Internet, hardly anyone had a home computer, and not even the government had enough processing and storage power available to record and analyze everyone's phone metadata. The Court in 1979 did not and could not have anticipated the NSA of 2013.
That's true, but unfortunately, it's all too often how it goes in a system of constitutional law in which court decisions create rules that stand for decades. Technological change is fast; the wheels of justice grind slow.
And in this country, the basic questions about privacy rights in the face of governmental surveillance technology are decided by nine elderly people doing their best to assimilate a world of gadgetry they probably understand only dimly. (This is not necessarily a bad thing, by the way—would you be happier if those cool techy kids at Facebook were writing the rules on privacy? I wouldn't.)
So your cocktail-party takeaway is that constitutional rules like the “third party” doctrine can be overtaken by technological change. But the Court can adapt, and the evolution of the case law on GPS trackers gives us reason to hope that Smith may be due for a rewrite.
For 40 years, the case law governing the use of tracking devices to monitor vehicle movements was driven by the same intuition as the case law on call histories. Whatever you do in “plain view” (for example, drive on a public road) is fair game for anyone, including the government, to observe.
For decades, all the cases held that monitoring a subject's movements with a tracker was not a “search,” because when a person drives on a public roadway, anyone can see where she's going. Trackers were pure catnip for law enforcement.
And they got better and better, as the core technologies advanced.
The batteries got better, so you didn't have to wire them into the car's electrical system. The transmitters got better, so you didn't have to have a chase car hovering to maintain a sightline. The storage capacity got better, so you could store as much travel data as you wanted. And the processing power got better, so you could make a detailed model of multiple subjects' movements, and really connect the dots.
Then, finally, in 2010, something remarkable happened. A few judges looked at the new technical capabilities of the government's tracker technology, and began rethinking their Fourth Amendment analyses of trackers.
Eventually, the issue reached the Supreme Court and, in January 2012, the Court overturned decades of precedent and held, unanimously, that installing a tracker to monitor a car's movements is indeed a “search.”
Readers of my previous column will recall that the majority's rationale in Jones was decidedly retro—the five-justice majority held that installing a tracker on a car is a search because it requires the physical appropriation of the subject's property, which is a “trespass” under 18th century common law.
But what's exciting about Jones for the question of phone-call metadata is not the majority opinion, but rather the concurring opinion that also got five justices (Justice Sotomayor joined both opinions) and proposed a different reason for why tracker monitoring is a search.
The concurring justices wrote that the monitoring and analysis technology has improved to the point at which it is possible to create a mosaic of someone's life through myriad individual data points. Each individual point might be “public,” but when the government uses technology to knit them all together into a portrait, the result can be a violation of your reasonable expectation of privacy.
The concurring justices didn't say exactly how detailed that portrait had to be to constitute a search, but they were prepared to say that the amalgamation of data from a month-long tracker investigation would be enough to constitute a search.
Intimacy may be in the eye of the beholder, but I bet most readers will agree with me that the NSA could stitch together a much more intimate portrait of your life from your communications metadata (which is not just phone data, but that's another essay) than from a month of tracker data from your car.
So the question is whether those five justices—Justices Breyer, Ginsburg, Kagan, Alito, and Sotomayor—would be prepared to apply the “mosaic theory” of tracker data to the much richer mosaic of NSA metadata.
I think they probably are.
And lower courts may also be emboldened by the Jones concurrence to start setting some “mosaic” parameters for metadata analysis. Maybe a day in the life will be OK, but a month will be too intrusive.(Intuitions are going to be all over the map here.)
But we'll have to wait for actual cases.
Courts don't decide abstract principles; they need parties, and concrete facts. So we'll have to see what the government does with all this data.
Maybe it won't do anything with it. But if there is any sharing with the Department of Justice—and if you're a prosecutor looking for a way to prove up a tough case, wouldn't you be asking the NSA to share?—then the issue will get into the courts.
If it does, I think the mosaic-theory analysis may prove to have legs—and if it does, judges will become a lot more willing to police metadata collection than they are now.
Caleb Mason is a regular contributor to The Crime Report. He is an attorney at Miller Barondess LLP in Los Angeles, and a former federal prosecutor. He welcomes comments from readers.