Orin Kerr is the Fred C. Stevenson Research Professor of Law at The George Washington University Law School.
One of the most basic ideas in Fourth Amendment law is what you might call the eyewitness rule: The government can always talk to eyewitnesses. If the police find out a bank was robbed, they can go to the bank and interview those who saw the crime occur. They can talk to the bank clerk about what he observed. They can talk to the security guard about what she experienced. They can talk to bank customers about what happened. These interviews, whether voluntary or compelled, don’t trigger the Fourth Amendment. There’s just no Fourth Amendment right to prevent people from talking about what they saw you do.
If the Supreme Court sticks with the eyewitness rule, Carpenter v. United States is an easy case. Cellphone companies are eyewitnesses. Customers use their services and hire the companies to place calls for them. The companies generate records of what they did for their customers. If the police find out a phone was used in the commission of a crime, they can go to the phone company and get its records about how the phone was used. The phone company can be forced to tell the government what it did for its users. Because there’s no Fourth Amendment right to prevent people from telling the police about what they saw you do, getting those records does not trigger the customer’s Fourth Amendment rights.
The challenge for Timothy Carpenter’s lawyers is to get the Supreme Court to carve out an exception to the eyewitness rule. They need to persuade the court that sometimes the law shouldn’t let people talk about what they saw you do without a warrant. The strategy will be to say that the eyewitness rule shouldn’t apply because Carpenter had a “reasonable expectation of privacy” in his location. Because he had a reasonable expectation of privacy, they’ll say, the government should need a warrant to find out what the phone company knows about him.
But this framing is based on a conceptual error. Properly understood, this case has nothing to do with reasonable expectations of privacy. To see why, we need to understand the origins of the reasonable-expectation-of-privacy test.
In 1967, when Katz v. United States was decided, two kinds of cases dominated Fourth Amendment law defining what is a search. The first kind of case identified the spaces that merited Fourth Amendment protection. Homes received protection, but open fields didn’t. Katz was one of these cases, because it asked whether public phone booths were protected, like homes, or unprotected, like open fields.
The second kind of case considered when disclosure from inside a protected space eliminated privacy. The Supreme Court decided a long string of those cases in the decade before Katz. And it consistently adopted a simple rule: A person who knowingly exposed private information from a private space to outside observation waived Fourth Amendment protection. Hoffa v. United States, decided the year before Katz, is a helpful example. The Supreme Court held that James Hoffa had no Fourth Amendment rights in what he told his friend Edward Partin even though the conversation occurred inside the protected space of Hoffa’s hotel room.
This background is necessary because it explains Justice John Marshall Harlan’s two-part Katz test that the Supreme Court later adopted. Harlan explained the test as his “understanding of the rule that has emerged from prior decisions,” so he was simply trying to summarize what the cases had held. And Harlan’s summary nicely (if briefly) encapsulated the two lines of cases. First, a space had to be protected, which was the case if society was willing to recognize an expectation of privacy as reasonable there. Second, the person had to “exhibit” an actual expectation of privacy, an intention “to keep” the protected space “to himself,” by not exposing the space “to the plain view of outsiders.”
In other words, the Katz two-part test just restated the two requirements of then-existing Fourth Amendment doctrine. To establish Fourth Amendment protection, a person needed to have a place that society would recognize as justifying privacy and had to take steps to shield that space from outside observation.
Why does this history matter for Carpenter? It matters because this case, like all eyewitness cases, involves the second aspect of Fourth Amendment protection: the requirement that a person must shield his information from observation to get Fourth Amendment protection. Even assuming a cellphone user’s location should count as a protected space, the user has revealed his location to the phone company. Whether Carpenter had a “reasonable expectation of privacy” should be irrelevant. Carpenter shouldn’t win because he didn’t shield his location from his phone provider. Carpenter’s relationship with the phone company is like Hoffa’s relationship with Partin. Carpenter can’t both share his information with the phone company and demand a warrant before the phone company gives that information to the government.
This point is hard to see because the Supreme Court veered off-course in the 1970s and 1980s with cases involving what it has called the “third party doctrine.” Those cases are normatively correct. But they’re in the wrong doctrinal box. As I detailed in a recent article, the court mistakenly moved the content of the subjective-expectation-of-privacy test over to the reasonable-expectation-of-privacy test and relabeled it the third-party doctrine. Students of the Fourth Amendment have been confused ever since. Isn’t it sometimes reasonable, they ask, to expect privacy in information that a person knowingly disclosed? But that’s not the right question. The right question is, should you have a right to stop others from telling the government about what they saw you do?
This understanding explains the maddening difficulty Carpenter’s side has articulating the limits of its argument. The Fourth Amendment calls for clear rules. The government needs to know what is a search and what isn’t. But Carpenter’s side always struggles to explain when the Fourth Amendment should offer protection against government access to business records. If cell-site records are protected, how about credit-card records? Telephone records? Bank records? Should the amount of time covered by the government’s request matter, and if so, how? Advocates for Fourth Amendment protection in cases like Carpenter generally decline to say where the lines should be.
The reason for this reluctance, I think, is that there is no obvious line to draw for when you should have a right to stop others from telling the government what they know about you. Carving out an exception to the eyewitness rule creates a puzzle: There are no pre-existing principles that explain which eyewitnesses can be forced to talk to the government and which ones can’t. Legislatures can just draw arbitrary lines. But courts have no traditional tools to use to decide when a warrant is needed to make an eyewitness speak.