The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Ninth Circuit hears NSA program challenge


On Monday, the Ninth Circuit held oral argument in Smith v. Obama, a Fourth Amendment challenge to the Section 215 telephony metadata program. You can watch a video of the argument here. The panel consisted of Judges Hawkins, McKeown, and Tallman. This was the third argument by a federal circuit involving a challenge to the telephony metadata program. The others are the Second Circuit and the DC Circuit, neither of which has handed down a ruling yet.

To win the case before the Ninth Circuit, the plaintiff needs to win on three basic questions: 1) did the plaintiff have standing; 2) did a search occur; and 3) was the search constitutionally unreasonable. There was significant questioning on standing and a lot on what is a search, but very little on reasonableness. On the whole, I think that emphasis is probably a good sign for the government. With that said, I'm not sure which way the case will come out. Judge Tallman seemed pretty likely to vote for the government on either or both of the first two questions. I had less sense where Judges Hawkins and McKeown might come out.

I want to focus on an interesting question that Judge McKeown asked Thomas Byron, counsel for the government defendant: When applying the reasonable expectation of privacy test, how do we know what society expects when it comes to a big surveillance program like Section 215's program? Byron gave what I think is the correct doctrinal answer: Smith v. Maryland tells us as a matter of law that people have no reasonable expectation of privacy in pen register data from their phones, and there is nothing in Smith that suggests that scale or aggregation can make any difference. Although that's a good doctrinal answer, I think there's a historical explanation that is more satisfying. I cover the historical explanation in this forthcoming article, starting at page 11, and I thought I would give a basic outline here.

As I explain in the article, the original design of Justice Harlan's two-part Katz test was to summarize the two basic requirements of establishing Fourth Amendment rights: first, that the intrusion was into the kinds of spaces that the Fourth Amendment protects, like homes or cars or (in Katz) telephone booths; and second, that the person had not openly exposed his protected space to outside observation. The first part was labeled the "objective" test, as it rested on whether the space invaded was the kind of space that merited Fourth Amendment protection in a general sense; and the second test was labeled the "subjective" test, as it hinged on whether the person took steps to hide the space from outside observation, manifesting an intent to keep it private in those specific circumstances.

The rule that you can't have any Fourth Amendment protection in what you share with a third-party was originally part of the subjective test. By revealing your information to a third party, the thinking ran, you no longer manifested an intent to keep your information private and gave up your rights in that information. The government therefore could get it from the third party without implicating your rights. Although that was the original design of Harlan's test, later opinions of the Court misunderstood this point. Later decisions, including Smith v. Maryland, simply assumed that the test labeled "subjective" was an inquiry into what a person actually expected to happen. Smith had to fit the traditional rule that a person has no Fourth Amendment rights in what they disclose to third-parties somewhere, however, so the rule ended up being squeezed artificially into the reasonable expectation of privacy test instead. The result has confused law students for decades: We get the strange-seeming announcement that there's no reasonable expectation of privacy in circumstances when a lot of people probably think it feels pretty reasonable to them.

Why does all this matter? I think it matters because it shows that Judge McKeown's question is based on the Smith court's accidental rephrasing of the third-party rule. Understanding the accident explains why the question that seems really difficult at first blush ends up having a simple answer. The reason the government does not violate the Fourth Amendment rights of a telephone user when collecting metadata from the phone company should have nothing to do with what "society" expects. Instead, the reason is that if you knowingly disclose information to a third party, you are not manifesting your subjective expectation of privacy in that information and don't have any way to control it under the Fourth Amendment as a matter of law. From that perspective, we don't need to know what "society" expects, or what privacy policies say, to apply the third-party doctrine. What "society" expects is irrelevant, as the doctrine was originally and properly rooted in the subjective test instead of the objective test.

To be clear, the doctrinally correct result Byron offered brings you to the same result in the end. If you say that Smith v. Maryland requires lower courts to say that there is no reasonable expectation of privacy in pen register information as a matter of law, you get to the same result that no search has occurred. But I hope it's at least a little bit illuminating to see how we got here, and in particular to see why applying Smith does not call for courts to make first-principles inquiries into societal expectations.