One of the most eagerly awaited aspects of the Privacy and Civil Liberties Oversight Board ("PCLOB") report on section 702 surveillance was how the PCLOB would treat human rights issues. In January, 2014, President Obama issued PPD-28, acknowledging that individuals all over the world had privacy interests in data collected by the NSA. Would the PCLOB take a stand on whether human rights agreements like the International Covenant on Civil and Political Rights ("ICCPR") apply extraterritorially, and whether the U.S. complied with that treaty? At first blush, the PCLOB seems to reserve the question, declaring that it will work within the inter-agency PPD-28 process to address the issue. Some, like Amie Stepanovich at Just Security, are disappointed in this absence of definitive conclusions.
However, a couple of observations are in order: 1) the PCLOB recommends measures that will help meet any human rights standard that might plausibly be applied to the U.S.; and, 2) the PCLOB can best further the goal of human rights compliance by working within the PPD-28 process, although this approach also raises the stakes for the Board.
In exploring the first point, assume world bodies will find that ICCPR duties, like the duty to refrain from arbitrary intrusions on privacy, apply extraterritorially; the U.N. Human Rights Committee has suggested this, and the U.N. High Commissioner on Human Rights report, due in August, will likely draw the same conclusion. (See Harold Koh’s memo here, as well as Beth Van Schaack’s excellent paper and Ashley Deeks’s post.) That brings us to the ICCPR’s substantive guidance, found in Article 17, to refrain from arbitrary intrusions on privacy. What is “arbitrary” collection? As my Fordham article on NSA surveillance and human rights argues, a state complies with Article 17 if its foreign surveillance has some relationship to an important governmental objective.
The arbitrariness standard is more deferential than certain readings of the proportionality standard often used in human rights law. The text of a treaty is a key factor in interpretation, under article 31 of the Vienna Convention on the Law of Treaties. Article 17’s use of the term, “arbitrary,” acknowledges that surveillance in law enforcement and national security is inherently overinclusive---as it must be to close gaps that allow criminals, terrorists, and rogue state officials to operate with impunity. The NSA’s foreign surveillance, which is tailored to subjects of foreign intelligence value, meets this standard.
Indeed, collection that is tailored at the collection and/or use stage would meet the more demanding proportionality standard read into Article 8 of the European Convention on Human Rights, which protects privacy but contains broad exceptions for national security, law enforcement, and other legitimate government objectives. In 2006, years before the Snowden revelations, the European Court of Human Rights held in Weber v. Germany that intelligence collection programs involving computer searches of content were consistent with Article 8. Weber contains two key insights relevant to section 702 surveillance. First, public disclosure of criteria that trigger surveillance should not be unduly specific, because knowledge of specific criteria would allow subjects of collection to fly under the radar, thereby defeating the purpose of collection. Second, states have a greater interest in collection regarding persons located abroad, since a state has less power to directly influence such persons’ conduct through its domestic law or acquire information about them through other means. Without the capacity to engage in foreign surveillance, a state must rely for its information on other states that may be indifferent or hostile.
How does the PCLOB report on section 702 fit with human rights norms? Consider its Recommendation (on pp. 134-35) that the NSA, 1) use greater specificity in its targeting procedures assessing the foreign intelligence value of a subject, and, 2) require an explanation in writing of how a particular selector (such as a phone number of email address) is likely to yield foreign intelligence information. The PCLOB notes its assumption that the Foreign Intelligence Surveillance Court (FISC) will review the NSA’s targeting procedures. If some source of adversarial advocacy (either an amicus serving at the FISC’s request or a more institutionalized public advocate per Steve and Marty) weighs in on the revised targeting procedures, the FISC will have the benefit of arguments from both sides. It can then make a truly deliberative decision on whether the procedures comply with statutory and constitutional norms. The attention to tailoring in this process will also highlight U.S. compliance with the ICCPR.
Review by an independent court like the FISC already puts the U.S. ahead of many European countries, which rely on exclusively administrative authorization for their surveillance activities. Moreover, the PCLOB’s proposal would likely cause only minimal disruption in current NSA practice. Of course, any new documentation requirement has costs. However, it would be salutary to ask analysts to provide more than the single sentence that the PCLOB says now serves to explain a given selector’s foreign intelligence value. More precise explanations can yield more efficient targeting. That may well save time in the long run.
The PCLOB’s decision to recommend this course without definitive conclusions on human rights concerns is part of a long game on foreign intelligence collection. Playing the long game raises the stakes for the Board and for the intelligence community as a whole. Will the Board be effective in internal deliberations that are part of the PPD-28 process? If not, its reticence in the section 702 report will seem like a bad bet.
That would be unfortunate, not merely for the PCLOB, but for U.S. interests. The world is looking at U.S. foreign surveillance with more skeptical eyes in the wake of the Snowden revelations. The U.S. will navigate through these turbulent waters if it internalizes constraints that enhance global confidence while preserving effective collection. Failing to adopt modest constraints like those proposed in the PCLOB report may prompt further pressure on our allies to decline to cooperate on intelligence matters. That ominous development might cheer the NSA’s critics, but it would do little to promote the NSA’s mission.