Another Problem With Judge Leon’s NSA Opinion: Absolute vs. Relative Measurements and Fourth Amendment Reasonableness

The more I re-read Judge Leon’s opinion in Klayman v. Obama, the more I am struck by how many parts of the opinion strike me as strange, off-key, or just wrong. This post will focus on an aspect of the opinion that has been ignored so far: Fourth Amendment reasonableness. In particular, I hope to explain why I think Judge Leon’s approach to assessing the reasonableness of the NSA program conflicts with established Supreme Court precedent.

First, some context. According to the Supreme Court, reasonableness requires a cost/benefit analysis: “Whether a search is reasonable is determined by assessing, on the one hand, the degree to which it intrudes upon an individual’s privacy and, on the other, the degree to which it is needed for the promotion of legitimate governmental interests.” Samson v. California, 547 U.S. 843, 848 (2006). That’s the general test, and Judge Leon recognizes it. But I want to focus here on a specific issue: The methodology for analyzing the latter question, the efficacy of the government’s step.

Consider two different approaches. First, a court could make an absolute measurement. That is, the court could measure how much that step advances the government’s interest as compared to no step at all. Alternatively, a court could make a relative measurement. That is, the court could measure how much that step advances the government’s interest as compared to alternative methods that the government could conduct to try to collect the same information.

The Supreme Court cases that I am aware of have endorsed absolute measurement and rejected relative measurement. For example, in United States v. Martinez-Fuerte, 428 U.S. 543 (1976), the Court considered the reasonableness of an immigration checkpoint set up along a highway. The defendants tried to argue that the checkpoint did not advance the government’s interest particularly well because there are other effective ways of finding illegal immigrants. The Court rejected the relevance of the alternatives:

The defendants argue at length that the public interest in maintaining checkpoints is less than is asserted by the Government because the flow of illegal immigrants could be reduced by means other than checkpoint operations. As one alternative they suggest legislation prohibiting the knowing employment of illegal aliens. The logic of such elaborate less-restrictive-alternative arguments could raise insuperable barriers to the exercise of virtually all search-and-seizure powers.

Instead, the Supreme Court tends to look at the effectiveness of the government step taken on its own. Consider the recent decision in Maryland v. King, 133 S.Ct. 1958 (2013), involving the acquisition and analysis of a DNA sample on arrest. In the course of considering the extent to which the DNA test advanced the government’s need to identify suspects taken into custody, the Court compared a world with DNA tests to identify suspects with a world with no tools to identify suspects. Of course, there are lots of ways of identifying suspects, such as fingerprints, photographs, names, Social Security numbers, checking for tattoos, and the like. But the Court treated the existence of alternative ways of identifying suspects as merely proof that the interest was a strong one and that DNA analysis was part of a long tradition: DNA analysis “uses a different form of identification than a name or fingerprint,” the Court assured the reader, “but its function is the same.” The effectiveness of the step was measured on an absolute scale, not relative to the effectiveness of alternatives.

In his NSA opinion, however, Judge Leon takes the opposite approach. Judge Leon only considers the marginal advantage of the telephony metadata program as compared to other investigative techniques the NSA could use to find terrorists. He does this by redefining the government’s interest in a critical way. When the government argues that its interest for the reasonableness balancing was finding terrorists, Leon rejects this absolute interest in favor of a relative one:

A closer examination of the record, however, reveals that the Government’s interest is a bit more nuanced — it is not merely to investigate potential terrorists, but rather, to do so faster than other investigative methods might allow.

(emphasis in original). Judge Leon identifies this interest by closely reading the declarations filed by the government, which he sees as favoring the NSA program not because it finds terrorists but because it is so fast at doing so. After reviewing those declarations, though, Judge Leon announces that he is unpersuaded that the NSA program finds terrorists more quickly than other investigative techniques:

Given the limited record before me at this point in the litigation — most notably, the utter lack of evidence that a terrorist attack has ever been prevented because searching the NSA database was faster than other investigative tactics — I have serious doubts about the efficacy of the metadata collection program as a means of conducting time-sensitive investigations in cases involving imminent threats of terrorism.

If we take Supreme Court precedent seriously, though, that analysis appears to be erroneous. Judge Leon should have measured the efficacy of the governments interest on an absolute scale, not relative to other possible investigative techniques. Put another way, he shouldn’t have required the government to show that the program has been singularly responsible for catching terrorists. Instead, he should have just weighed the extent to which the program advances the government’s interest taken in isolation. Of course, we can guess from the tone of Judge Leon’s opinion that he would have reached the same result regardless of this point. Still, it seems like a significant analytical error in the opinion.

I realize that whether courts should use absolute versus relative measurements is an interesting debate. One problem with absolute measurements is that it allows programs in a constitutional sense that may be undesirable in a policy sense. On the other hand, relative measurements can lead to a recursive loop: If all investigative techniques are equally effective, then they are all unconstitutional because none of them advance the government’s interest relative to the others, but then suddenly there are no legal alternatives, so all of them become constitutional again, but then they’re all unconstitutional again, then constitutional, then unconstitutional, etc. I’ll let readers debate which approach is best in a world with no Supreme Court precedent. My point is just that Supreme Court precedent appears to adopt the absolute approach, so I think Judge Leon was wrong to reject that standard in favor of a relative approach.

Finally, I want to stress that I’m not criticizing Judge Leon for reaching the wrong result in the case. I happen to think that current precedents point to the view that the NSA’s program is constitutional. I also tend to think that result is sensible because the alternative view generally requires either rejecting the third party doctrine (which I think would be a bad idea for reasons I explained here) or else requires adopting mosaic arguments (which I think are problematic for reasons I explained here). But my arguments on these points are intensely pragmatic, based on the difficulty achieving proper equilibrium-adjustment through new line-drawing in the kinds of ways that would seem to be necessary to reach the result that challengers want. If there are ways of addressing those line-drawing concerns, I’m open to rethinking my view. Indeed, I’d be delighted if others would address the line-drawing concerns that I have and persuade me that I’m wrong. But at least so far no one has tried to resolve those problems, so I’m stuck with my pragmatic objections. Either way, my concerns here are not about the ultimate result (constitutional vs. unconstitutional) but rather the legal legitimacy of the arguments used to get there.