International Law

A Reply to Ryan Goodman

By Benjamin Wittes
Tuesday, March 25, 2014, 9:58 AM

Over at Just Security, the estimable Ryan Goodman of NYU Law School has responded to my post of Saturday taking on Glenn Greenwald over press eagerness to blow lawful classified programs. Goodman advances the remarkable claim that the legality of what he and Greenwald call "mass surveillance" is actually "an open legal question" under international law, and therefore that the distinction I have drawn between legality and illegality in NSA programs "is much weaker, if not lost." Greenwald triumphantly tweeted Goodman's post as follows:  

Goodman acknowledges that the international legal prohibitions against mass spying are subtle. Indeed, he says, "I don’t fault Ben for making th[e] mistake" of assuming "that there are no legal proscriptions on intelligence surveillance of foreign populations." The mistake, he assures me, "is a common one."

So what is Goodman's evidence that there is, in fact, international law restricting collection against foreigners overseas? He cites three pieces of what he terms "solid support for the proposition that states recognize privacy constraints on foreign surveillance as a matter of international human rights." The first is this UN General Assembly Resolution, passed late last year. The second is this very recent resolution by the European Parliament. The third is something that hasn't even happened yet: "forthcoming concluding observations by the UN Human Rights Committee. During the recent US report to the UN body, at least some members of the Committee drew attention to NSA extraterritorial intelligence practices in strong terms," writes Goodman.

Color me unimpressed.

For one thing, the General Assembly resolution, which Goodman cites as evidence that states accept a legal distinction between mass surveillance and targeted surveillance doesn't quite do that. It objects to "arbitrary or unlawful interference with . . . privacy, family, home or correspondence," and it insists on "the right to the protection of the law against such interference," but it never defines what "unlawful" surveillance is. And I suspect that the US would argue that laws like Section 215, Section 702, and Executive Order 12333 offer "the protection of the law against such interference." Yes, the resolution expresses deep concern about "the collection of personal data, in particular when carried out on a mass scale," but it does not purport to ban such collection. It only requests a report on the subject and reaffirms whatever existing privacy rights may govern the international space. What's more, to the extent it considers "mass surveillance" inconsistent with the ICCPR, the US does not accept the ICCPR's extraterritorial application anyway.

Perhaps this is why the United States did not oppose this resolution, which was passed without a recorded vote. It's hard to me for imagine that the US delegation would not have spoken up had it thought it was acknowledging the illegality of major US intelligence programs.

As to the European Parliament resolution, yes, I acknowledge that this resolution shows that many states consider mass surveillance conducted by the US to be highly offensive. But I hardly think a resolution by the European Parliament passed this month establishes anything like state practice, much less state practice undertaken as a matter of legal obligation. (For a discussion of this resolution, listen to last week's Steptoe Cyberlaw Podcast.) The EU Parliament resolution may say a great deal about what conduct European states think the US should refrain from out of sense of legal obligation, but it doesn't say much about what conduct the intelligence services of the EU's own states refrain from out of a sense of legal obligation. Nor do actions by the Human Rights Committee, particularly actions that have not happened yet.

Norms, to be sure, may change, but the evidence Goodman cites that there is some existing norm under which states acknowledge the illegality of mass surveillance is exceptionally weak. Perhaps the "mistake" of assuming that no such norm exists "is a common one" because it's not a mistake.