In this post, I want to focus on a narrow slice of Charlie Savage's much-anticipated book Power Wars (published today...go ahead, order it now!), one that might not generate as much attention as the material covering more recent national security law episodes. In particular, I want to highlight the book's discussion of an area of NSA surveillance activity sometimes labeled "transit authority." It is a very useful case study of the way in which legal and policy questions may be impacted by technological change (and also a handy illustration of why it is significant that the private sector owns telecommunications infrastructure in the United States).
1. Background: What FISA Did and Did Not Cover In the Beginning
Intercepting "foreign-to-foreign" communications--i.e., sent by a non-US-person abroad to a non-US-person abroad--has always been at the heart of the NSA enterprise. And prior to the late 1980s, it was an area of collection activity that just didn't implicate FISA. This was so by design. The drive to enact FISA in the late 1970s was spurred by leaks and congressional investigations revealing that NSA's activities were not confined to collection of foreign-to-foreign communications, but instead at times involved communications with domestic connections. This was controversial, to put it mildly, and FISA was designed to subject such activity to judicial oversight, FISA was not designed to do the same, in contrast, as to the core foreign-to-foreign surveillance mission.
To strike this balance, the relevant part of FISA (50 USC 1801(f)) made the government seek judicial approval only where one of the following four scenarios applies:
(f) “Electronic surveillance” means—
(1) the acquisition by an electronic, mechanical, or other surveillance device of the contents of any wire or radio communication sent by or intended to be received by a particular, known United States person who is in the United States, if the contents are acquired by intentionally targeting that United States person, under circumstances in which a person has a reasonable expectation of privacy and a warrant would be required for law enforcement purposes;
(2) the acquisition by an electronic, mechanical, or other surveillance device of the contents of any wire communication to or from a person in the United States, without the consent of any party thereto, if such acquisition occurs in the United States, but does not include the acquisition of those communications of computer trespassers that would be permissible under section 2511 (2)(i) of title 18;
(3) the intentional acquisition by an electronic, mechanical, or other surveillance device of the contents of any radio communication, under circumstances in which a person has a reasonable expectation of privacy and a warrant would be required for law enforcement purposes, and if both the sender and all intended recipients are located within the United States; or
(4) the installation or use of an electronic, mechanical, or other surveillance device in the United States for monitoring to acquire information, other than from a wire or radio communication, under circumstances in which a person has a reasonable expectation of privacy and a warrant would be required for law enforcement purposes.
Some things about this approach are clear. Domestic-to-domestic communications can't be surveilled without FISC approval, for example, while no such approval is required for pure foreign-to-foreign collection.
But what about one-end-foreign, one-end-domestic communications (i.e., communications with one end abroad, and other in the US)? Here it gets tricky.
If the target of such collection is a U.S. person, FISA generally applies in such cases thanks to 1801(f)(1) above. But if the target instead is a non-US-person, it gets complicated. Simplifying things a bit, the upshot of the statutory details is that the government remained free to surveil without a court order in such cases except in the specific situation in which the acquisition occurred by tapping a wire from within the United States.
As Charlie relates in Chapter 5, that is not how acquisition typically occurred when the government collected one-end-foreign communications by wire at the time FISA was enacted and for a few years thereafter. In those years, the relevant "wires" were the old copper transoceanic lines, and the government apparently had ample capacity to tap them from international waters. And thus not only foreign-to-foreign but also one-end-foreign communications could be surveilled without need for a FISA order (so long as a U.S. person was not the specific target of the surveillance).
2. The Disruptive Impact of the Fiber-Optic Revolution
The emergence of fiber-optic cables for international communications in the late 1980s presented NSA with a technical challenge that turned out to raise legal questions as well.
As Charlie explains, "it is harder to tap fiber-optic cables from the middle, deep under the waves, than it is to tap coaxial cables that way. To intercept messages from a fiber-optic line, it was easier to collect them at one end or the other...." (p.173) In practical terms, this meant that in order to have the same functionality as it had enjoyed in past years with respect to capturing one-end-foreign communications, NSA had to shift from acquisition in international waters (which FISA did not cover) to acquisition from with telecom facilities on U.S. soil.
Would this change the FISA analysis? Certainly so. Section 1801(f)(2), quoted above, applied in that circumstance, meaning that the government would need to have individualized predication for collection in such cases, as opposed to a vaccuum cleaner approach allowing for post-collection fishing using keywords and other forms of data analysis. Charlie reports that this led the Reagan administration to "secretly began drafting a bill to change FISA." (p. 174) The project did not move forward quickly, however, and Charlie states that the George H.W. Bush administration "decided to shelve it," rather than precipitate a public debate that would be educational for adversaries with respect to our capabilities. (p.174)
3. Acquiring Foreign-to-Foreign Communications from...Inside the U.S.?
There was a silver lining in this situation, from the government's perspective. True, 1801(f)(2) seemed now to apply to the main available method for acquiring one-end-foreign communications. But the same was not true, it turned out, for all the traffic transiting those U.S.-based switches connected to the transoceanic fiber-optic cables. In some instances, in fact, foreign-to-foreign communications hit those switches too. And acquiring them did not trigger 1801(f)(2), since these by definition were not communications to or from a person in the U.S.; they were just foreign-to-foreign communications that happened to transit U.S. infrastructure.
Interestingly, Charlie reports that government lawyers actually struggle with this question of statutory interpretation. Though they ultimately agreed that FISA indeed did not apply, Charlie writes that the lawyers involved "knew what they were doing was edgy." (175) Why "edgy" given that the FISA definition pretty clearly did not apply, and that the Fourth Amendment conferred no other protection on the communications of foreigners? No doubt it was the domestic locus of the activity that caused the concern. It invited criticism of a revived practice of domestic NSA collection without judicial review, notwithstandin that no U.S. persons (in theory) were involved in the communications in question. Indeed, Charlie himself labels the emergence of the "transit authority" scenario as the creation of a "secret new power to employ a form of warrantless surveillance on domestic soil .... " (p.174).
4. The Homefield Advantage Grows: The Internet and Transit Authority
Suffice to say that with the rise of online communications, and especially thanks to packet-switching technology (which meant that communications from all over the world very well might transit U.S. switches), the volume and importance of foreign-to-foreign communications available via transit-authority collection grew by leaps and bounds. (p. 175-77) At least so long as the telecoms were willing to cooperate (remember, there was no FISC order in these cases and thus no legal compulsion to cooperate), NSA had a relatively-convenient way to harvest a considerable amount of data that might be much, much harder to acquire overseas (or undersea, for that matter)--that is, something of a homefield advantage owing to U.S. dominance in the infrastructure of telecommunications bandwidth.
5. How Does this Relate to Events at NSA in September 2001?
This background helps us understand the early post-9/11 developments at NSA. Charlie describes how Director Michael Hayden almost immediately directed NSA to employ its non-US assets (overseas or in space) to ramp up collection on one-end-foreign communications between the U.S. and Afghanistan, so long as no U.S.-person was the specific target. (p.180) But when someone suggested exploiting the homefield advantage by expanding Transit Authority beyond foreign-to-foreign communications in order to also get one-end-foreign communications, NSA's GC (Robert Dietz) correctly concluded that this would require FISC involvement, for the reasons noted above. (p.181)
The question then became much the same one that the Reagan administration lawyers had faced a decade-and-a-half before when fiber-optics first led wire acquisitions to migrate from the ocean's floor to onshore switches: Should the executive branch seek a legislative fix so as to enable one-end-foreign communications to be gathered without individualized, judge-approved predication? Famously, the Bush administration choose not to do so, but unlike the past it went ahead and ordered the acquisition to occur nonetheless, citing wartime authorities. It was not until the Protect America Act in 2007, and the FISA Amendments Act in 2008, that something of a statutory fix would be adopted.