Expanding the Reverse Targeting Prohibition: A Back Door Repeal of 702?

Several bills are now circulating to amend and reauthorize Section 702 of the Foreign Intelligence Surveillance Act (FISA). They all purport to reauthorize Section 702, in recognition of the important role it plays in the security of the United States and its allies. However, two of the bills would expand the existing prohibition on “reverse targeting” in a way that could render Section 702’s authority null. Thus, far from reauthorizing Section 702, these bills may effectively repeal it.

Much of the reauthorization discussion has focused on whether and how the government should be able to query the data obtained under Section 702 using the identifier of a U.S. person (like an email address). The Government Surveillance Reform Act (the Wyden bill), co-sponsored by Sens. Ron Wyden (D-Ore.),  Mike Lee (R-Utah), Warren Davidson (R-Ohio), and Zoe Lofgren (D-Calif.), and the Protect Liberty and End Warrantless Surveillance Act (the HJC bill) would require the government first to obtain a court order based on a finding that there is probable cause to believe that the U.S. person who is the subject of the query is an agent of a foreign power. The other two bills—from the Senate Select Committee on Intelligence (SSCI) and the House Permanent Select Committee on Intelligence (HPSCI)—would impose stricter internal requirements and controls for those queries without requiring such a court order.

While the U.S. person query issue is a vital one, all four bills seek to make other changes as well. We focus on one change common to both the Wyden and HJC bills—reverse targeting—that illustrates the magnitude of the challenge Congress faces in coming to terms with potential ramifications of these bills. 

Section 702 authorizes surveillance only of non-U.S. persons who are reasonably believed to be outside of the U.S. It also prohibits the government from engaging in reverse targeting. That is, the government may not “intentionally target a person reasonably believed to be located outside the United States if the purpose of such acquisition is to target a particular, known person reasonably believed to be in the United States.” This is intended to prevent the government from pretextually targeting a non-U.S. person when the actual purpose is to obtain the communications of a specific individual in the United States. 

The reverse targeting provision draws from long-standing prohibitions in intelligence policy that predate Section 702. When one of the authors, Alex Joel, served as the civil liberties protection officer at the Office of the Director of National Intelligence, he addressed reverse targeting in a letter in 2007 (before Section 702 was enacted in 2008), in response to a request from the ranking member of the HPSCI: 

Questions have been … raised about “reverse targeting”—that is, could an intelligence agency target a person overseas as a pretext for intercepting the communications of individuals inside the United State with whom the foreign person is in contact? The simple answer is that when the agency’s actual purpose is to surveil the person in the United States, it must obtain a court order as required under FISA. This is not a new problem for either the intelligence or law enforcement communities. ...

It is important to recognize that reverse targeting makes little sense as matter of intelligence tradecraft: if intelligence officers are indeed interested in a target inside the United States, they will have a natural incentive to seek a FISA court order in any event so as to obtain all of that person’s communications, rather than the limited subset that would otherwise be acquired through such reverse targeting.

Compliance with the reverse targeting prohibition has not been an issue in reauthorization discussions. Indeed, only a handful of reverse targeting incidents have been identified in the past. For example, the 16th semiannual compliance assessment reported on one such incident: “Reverse targeting is barred by statute and NSA policy and the prevention of reverse targeting is a key component of both the internal and external review of the Section 702 program .... [T]he joint oversight team assesses that the extreme rarity of reverse targeting incidents demonstrates the success of NSA’s training efforts on this issue.” 

Nonetheless, the Wyden and HJC bills would expand the reverse targeting prohibition. They would replace the operative language “if the purpose of such acquisition is to target a particular, known person reasonably believed to be in the United States” with “if a significant purpose of such acquisition is to acquire the information of 1 or more United States persons or persons reasonably believed to be located in the United States at the time of acquisition or communication” (emphasis added). While the changes involve only a few words, they would have an outsized impact on the efficacy of Section 702.

To understand why, it is important to recall a fundamental reason for Section 702—to detect whether foreign terrorists are communicating with people inside the United States. As stated by then Director of National Intelligence Mike McConnell in 2007 in discussing legislative proposals for what became the FISA Amendments Act of 2008 (containing Section 702): 

[O]ne of the most important and useful pieces of intelligence we could obtain is a communication from a foreign terrorist outside the United States to a previously unknown “sleeper” or coconspirator inside the United States. Therefore, we need to have agility, speed and focus in collecting the communications of foreign intelligence targets outside the United States who may communicate with a “sleeper” or coconspirator who is inside the United States. 

Indeed, this was the very reason members of Congress raised in rejecting a similar attempt to expand the reverse targeting prohibition. During the debate over the FISA Amendments Act in 2008, Sen. Russell Feingold (D-Wisc.) sought to have Section 702’s reverse targeting provision refer to “a significant purpose” rather than “the purpose.” Sen. Jay Rockefeller (D-W. Va.), who was then chairman of the Senate Select Committee on Intelligence, strongly opposed the proposal:

The Feingold reverse targeting amendment ... goes too far. ...  [W]e are revising the Foreign Intelligence Surveillance Act today in large measure precisely because we want the intelligence community to have the ability to detect and acquire the communications of terrorists who call into the United States. In other words, in order to detect and prevent terrorist attacks, finding out if a foreign terrorist overseas is in contact with associates in the United States is actually a significant purpose of this legislation, and it will always be a significant purpose of any targeting of a foreign terrorist target overseas by the intelligence community. ... You cannot in good conscience bar the intelligence community from collecting these communications. That is unacceptable. ... The amendment is not needed to achieve its stated goals. It will harm vital intelligence collection.

The intelligence concern with expanding the reverse targeting prohibition extends beyond counterterrorism. If a foreign government is communicating about its espionage activities against the interests of the U.S. or its allies, the communications the U.S. may be most interested in are communications the foreign government has with or about as-yet-unidentified agents operating inside the country. The same goes for foreign persons seeking to penetrate U.S. computer networks, or proliferators of weapons of mass destruction conducting activities inside the country. One would naturally expect that in conducting foreign intelligence targeting non-U.S. persons abroad, a “significant purpose” would be to acquire information of people in the U.S. who might be in communication with those targets. And it is important to note that the language of the amendment is not limited to communications with U.S. persons; even if a significant purpose of the surveillance is to obtain communications between foreign terrorists in which they identify their U.S. accomplices, the surveillance is prohibited. Thus, 702 would be limited to instances where there is no possibility that a U.S. person would be contacted or discussed—which is hardly appropriate for a tool intended to protect the United States.

Indeed, it is out of concern for the privacy of those very persons—those who are communicating with foreign targets abroad—that the Wyden and HJC bills put in place such stringent new requirements on U.S. person queries. How to balance the privacy of people whose communications are incidentally collected with the need for foreign intelligence surveillance is the proper focus of attention. The debate over so-called “back door searches” needs to be resolved without “back door repeals” lurking in the text.