Among the documents that Edward Snowden released are reports showing that the NSA had been picking up email and phone conversations by and among foreign leaders. Among the alleged targets were officials from the EU, individual EU member countries, Brazil, and Mexico. While each subject of this reported surveillance has expressed outrage, perhaps no state has been more agitated than Germany. Revelations about NSA activity directed at the EU have posed significant problems for the German government, given East Germany’s history of widespread surveillance of its own citizens by the Stasi. Chancellor Angela Merkel is under political pressure as she runs for re-election, and opposition parties have threatened to delay US-EU trade talks unless and until they obtain greater clarity about these NSA allegations.
One way the United States has addressed Germany’s concern is by agreeing to negotiate an arrangement pursuant to which neither state will spy on the other for governmental or industrial purposes. We might suspect that Germany proposed the idea and the United States acceded to the request, although Germany’s Chancellery Minister Roland Pofalla (in charge of Germany’s secret services and its intelligence cooperation with other states) told the German parliament that the United States had offered to enter into these talks. Negotiations are to begin in September. Merkel’s primary challenger in the upcoming German elections called on her to seek a “binding pledge from the U.S. government” not to spy on Germany, though the United States does not seem to have indicated publicly precisely what kind of “agreement” it is prepared to negotiate.
In view of these pending negotiations, it is worth considering at least two things: (1) the potential impact on international law of an arrangement intended to regulate espionage; and (2) the strategic and practical effects such an arrangement might have on U.S. intelligence in the future.
(1) As to the first issue, there is something inherently odd—as Duncan Hollis noted over at Opinio Juris—about the idea of an international agreement not to do something that states largely decline to acknowledge that they do, and that many states already view as unlawful (at least as a matter of domestic law). But there are at least two ways to think about espionage and international law: you may believe that peacetime espionage violates international law, or you may take the view that international law simply does not purport to regulate espionage, an activity nearly as old as time. If you take the former view, you presumably would invoke customary international law norms such as non-intervention and respect for sovereignty, which the use of secret listening posts and wiretaps by one state in another state would contravene. If you take the latter view, you would argue that ideas such as non-intervention and sovereignty developed against a background understanding that states do and will spy on each other, thus establishing a carve-out within those very concepts that allows—or at least turns a blind eye to—espionage.
Because espionage fits uncomfortably with international law, it is unsurprising that there are few (public) precedents of states agreeing not to spy on each other. The most commonly cited example is the “Five Eyes” agreement among the United States, UK, Canada, Australia, and New Zealand. In a paper submitted by the Canadian executive branch to a Member of Parliament, Canada stated, “Five Eyes allies, in their own national interests as sovereign states, can lawfully collect intelligence in accordance with their own domestic laws while respecting the long-standing convention not to target the communications of one another.” Of course, this sounds like an “understanding” rather than a binding legal arrangement, and there is no way to know the extent to which the Five Eyes states honor such standing arrangements.
In 2010, then-DNI Director Dennis Blair sought a comparable arrangement with France. According to the Telegraph, “Mr Blair proposed an unprecedented written pledge even more binding than the post-war ‘gentlemen’s agreement’ between the US, Britain, Canada, Australia and New Zealand as trusted partners who do not spy on each other. The deal would also have given France access to a highly secure intelligence retrieval and exchange system.” President Obama ultimately scuttled the deal out of concern that the agreement might handcuff the United States if a less U.S.-friendly French government came into power in the future. (Note the underlying assumption that the United States would feel obliged to alter its behavior in the face of such an agreement, even if were not in U.S. interests to do so.) In short, I am unaware of any publicly available bilateral “no spy” agreements involving the United States. However, if the United States and Germany do come to an arrangement, it would illustrate the idea that international law can regulate espionage, however unnatural it may seem.
(2) As to the second issue, what are the potential implications for the United States in entering into such an agreement? In the first place, it depends what the “agreement” looks like. If it is a legally binding arrangement, the United States may find itself torn in the future between violating an international legal commitment and conducting espionage in Germany to pursue, say, reports of an imminent armed attack. If—as seems more likely—it ends up being an arrangement that binds as a political matter but not as a legal one, the United States would retain more leeway to act in ways that don’t strictly comply with whatever the final language is. But even political agreements raise the stakes when violations occur; if the United States were caught spying on Germany in violation of a political arrangement, Germany undoubtedly would be exercised. The specific wording of any such agreement also will be important, of course: a limitation on spying on German officials or industries is different from a limitation on spying in Germany at all (against known terrorist groups, for example).
One potential downside of concluding either a binding or non-binding agreement is that other states (including Brazil and Mexico, for instance) may clamor for comparable arrangements, and express outrage and suspicion if the United States proves unwilling to negotiate such deals with them. Another downside is simply the loss of intelligence if the United States agrees not to spy on Germany—or the loss of access to matters or third parties to which the German government might have unique access. The United States conceivably might be able to glean important intelligence via third parties (such as other Five Eyes states), however. Yet another reason such an arrangement might be undesirable is the reason given by President Obama in the French context: a future German government might prove less friendly to the United States than the current one is. Finally, we might think that the United States has more to lose in such a bilateral arrangement because the United States presumably has a broader capacity to collect intelligence on (and in) Germany than Germany does on the United States. So the quid and quo in the arrangement won’t be equivalent.
In short, such an agreement, whether concluded as a legally binding agreement or only a politically binding one, seems unlikely to eliminate spying between the two countries, but would raise the political and reputational costs of conducting espionage. As a result, each government might attempt to focus only on the most serious threats emanating from the other state, if and to the extent it is technologically possible to limit one’s spying in this way.