Published by The Lawfare Institute
in Cooperation With
I. The Origins of the Foreign Intelligence Surveillance Exception
The FIS exception has its real origins in the Supreme Court's 1972 decision in the Keith case. There, the Court expressly disavowed a "domestic security surveillance" exception to the Warrant Clause of the Fourth Amendment, but strongly hinted (without holding) that foreign intelligence surveillance might be different. Thus, in a trio of subsequent circuit court decisions, the Third, Fourth, and Fifth Circuits each recognized a qualified foreign intelligence surveillance exception to the Warrant Clause. As the Fourth Circuit explained in Truong, "this foreign intelligence exception to the Fourth Amendment warrant requirement must be carefully limited to those situations in which the interests of the executive are paramount." In particular, Truong required that the object of the search or surveillance be "a foreign power, its agent or collaborators," and that "the executive should be excused from securing a warrant only when the surveillance is conducted 'primarily' for foreign intelligence reasons." Finally, and this can't be stressed enough, the existence of the FIS exception was squarely tied to the "practical difficulties of obtaining a warrant for foreign intelligence surveillance . . . at the time [the underlying] surveillance was conducted," which the courts described as "particularly acute" prior to the enactment of FISA.
In other words, part of the justification for the FIS exception was the absence of a FISA-like procedure that balanced the need for secrecy with the need to secure ex ante judicial approval before conducting foreign intelligence surveillance. [As I explain below, this is why the 2008 FISA Court of Review decision in the In re Directives case doesn't just follow from these earlier cases.] As Judge Sand explained in 1998 in United States v. Bin Laden, no court had recognized a comparable FIS exception to the Warrant Clause in a case to which FISA applied.
II. The USA PATRIOT Act, the FISA Amendments Act, and the FIS Exception
Of course, it's now familiar that, as part of the USA PATRIOT Act of 2001, Congress eliminated the second prong of Truong--the requirement that the "primary purpose" of the search be the gathering of foreign intelligence surveillance (as opposed to law enforcement evidence). And in its first-ever decision in In re Sealed Case, the FISA Court of Review held that such legislation did not itself violate the Fourth Amendment, at least largely because FISA warrants were probably still effectively "warrants" within the meaning of the Warrant Clause. In other words, Sealed Case upheld the elimination of the primary purpose test without specifically endorsing a categorical foreign intelligence surveillance exception. Given the language of FISA at the time, there was no real need to reach that issue; virtually of the surveillance undertaken pursuant to FISA was with a (FISA) warrant.
This is why the FISA Amendments Act were a game-changer with regard to the existence of a foreign intelligence surveillance exception to the Warrant Clause: Congress in the FAA (building on the Protect America Act of 2007) specifically authorized programmatic warrantless foreign intelligence surveillance in a manner almost guaranteed to sweep up a substantial volume of communications involving U.S. persons. Whereas pre-2007 FISA searches could only really implicate an FIS exception to the Warrant Clause in the rare (emergency) situations in which the statute authorized surveillance without a court order, FAA surveillance and searches routinely implicate the existence of such an exception because, for better or worse, they are conducted pursuant to a "certification" approved by the FISA Court, and not a FISA "warrant."
And so when the FISA Court of Review endorsed a categorical FIS exception in In re Directives in 2008, it was making new law in several respects: First, the Court of Review recognized an FIS exception notwithstanding FISA itself, the existence (and success) of which substantially undermines the justifications the Third, Fourth, and Fifth Circuits had relied upon in recognizing an FIS exception in the pre-FISA cases. In other words, the Court of Review recognized an exception that had previously been justified by the Executive Branch's inability to obtain exactly what FISA created. Second, even if FISA doesn't undermine the case for an FIS exception, the pre-FISA courts recognizing such an exception nevertheless required that the "primary purpose" of the search be to obtain foreign intelligence surveillance. Thanks to the USA PATRIOT Act and the Sealed Case decision, however, the FIS exception recognized by the FISA Court of Review in In re Directives does not include a similar constraint.
In other words, the FISA Court of Review recognized a foreign intelligence surveillance exception to the Warrant Clause that was novel in two critical regards: (1) in existing alongside and notwithstanding FISA; and (2) in applying even to those searches the primary purpose of which is not the gathering of foreign intelligence surveillance. Say what you will about the merits of such an exception to the Warrant Clause; the point I was trying to make yesterday was that such a decision broke important new ground, whether for better or worse.
III. Clapper and the FIS Exception
With all that said, we return to the matter at hand, i.e., the merits in Clapper v. Amnesty International, and whether the FISA Amendments Act violates the Fourth Amendment. To be clearer than I was yesterday, this is in fact two different questions: (1) whether the FAA violates the Warrant Clause; and (2) even if it doesn't, whether it is still unconstitutional to the extent it authorizes "unreasonable" searches. Clearly, the existence of the FIS exception goes only to the first question. And the plaintiffs could still prevail on the merits in the face of a recognized FIS exception if they either show that the surveillance at issue falls outside of such an exception or if they prevail on the reasonableness question. Indeed, given the volume of communications that could be swept up under the FISA Amendments Act, one could easily imagine a court assuming without deciding that there is an FIS exception to the Warrant Clause, but still concluding that the surveillance at issue is unreasonable because of the potential overinclusiveness concerns... To that end, I think it's safe to say I overstated things a bit yesterday when I suggested that "resolving the Fourth Amendment claim necessarily turns on the existence (vel non) of the foreign intelligence surveillance exception recognized by the FISA Court of Review in In re Directives."
At the same time, although the plaintiffs could theoretically prevail in the face of an FIS exception, it's difficult to see how the government could prevail if courts disagree with the FISA Court of Review. Because these searches are conducted within the territorial United States, the government cannot argue that, under Verdugo-Urquidez, the Warrant Clause simply doesn't apply. And I think the ACLU's summary judgment reply brief [see pp. 15-20] is quite compelling as to why the government also cannot take advantage of the body of Fourth Amendment law recognizing that "incidental interception of a person's conversations during an otherwise lawful surveillance is not violative of the Fourth Amendment." Whatever else the FISA Amendments Act authorize, it's hard to view the "vacuum cleaner" as "incidental" in the way that prior courts meant it...
In short, although I think the plaintiffs can win on the merits in Clapper even with an FIS exception, I'm not sure how the government could prevail on the merits without one. That may not have come through yesterday as clearly as I would've liked, but that, to me, is the heart of the matter here--and at least a small part of why I suspect the government is throwing so many eggs into the standing basket.