October 21, 2013
Slouching Toward Mordor
Grover Joseph Rees*
I sometimes visit a country whose government pervasively monitors the communications of its citizens, residents, and visitors. I am reluctant to name this country because representatives of its government may also be monitoring this publication, so I’ll call it the Democratic Republic of Mordor.
When I am in Mordor I am careful never to communicate with anyone about politics, religion, or anything else the government might consider subversive. But this is not enough to avoid getting into trouble: before each trip I must also ask all my friends in the United States and elsewhere who might communicate with me by phone or email to avoid any discussion of politics, and particularly of words such as “freedom”, “human rights”, and “democracy.” And if my friends work for organizations that have such words in their names—or for other entities that might be considered suspicious, such as the United States Government—I ask them not to communicate with me at all. The reason I take these steps is that I am reliably informed that while the government of Mordor systematically scans all electronic communications, it is logistically constrained to be far more selective about whose communications it actually reads or listens to. And the way to stay off that list is to avoid red flags such as talking about democracy or receiving a message from someone at Freedom House.
So I had seriously mixed feelings when I learned about the scope of our own government’s collection of communications records. Like most Americans, I consider myself a strong supporter of the war on terror. I believe that Bradley Manning and Edward Snowden committed crimes for which they ought to receive just punishment. As a government official for 25 years I had frequent occasion to complain that we classified way too much information, but it never occurred to me that the remedy was to violate my oath of office by unilaterally declassifying whatever information I thought should be made public. I also agree with my friend Stewart Baker that gathering intelligence about terrorism “isn’t patty cake.” I have therefore been generally supportive of Section 215 of the Patriot Act, which gives the government authority to seize records in the hands of telephone companies and other third parties—but only after proving to a special court that the records in question are relevant to an investigation to obtain foreign intelligence information or to protect against terrorism or espionage.
My support was based in part on my confidence that the court would require the government to present evidence that the records in question were genuinely relevant to a specific investigation. This confidence was bolstered a few months ago when the Director of National Intelligence, testifying before Congress, categorically denied that the NSA had collected “any type of data at all on millions of Americans.”
So I was stunned to learn that the NSA, with the court’s approval, had secretly collected the records of all the calls of all the subscribers of the major telephone companies in the United States and assembled similar collections of email messages, Facebook postings, and other internet communications. Notwithstanding the subsequent reassurances that NSA will never actually listen to my calls or read my emails unless they have reason to believe I have been communicating with terrorists, I find it disturbing that they have collected and are keeping this information.
My reaction stems in part—but only in part—from a visceral distaste for big government. Like many conservatives, I have always opposed the idea of a national identification card, notwithstanding how useful it might be to immigration enforcement and other important objectives. For that matter, many of us are still unhappy that the government broke its promise that Social Security numbers would never be used for anything but keeping track of our Social Security accounts. And most Americas are glad we have drones to spy on Al Qaeda and the Taliban, but we don’t want them spying on us.
The argument that we shouldn’t be worried if we have nothing to hide misses the point. Conservatives and others oppose intrusive government not so much because of specific things we are afraid the government will discover as because they make the government of the United States look and feel too much like the government of Mordor. We know our government’s objectives are mostly noble, while the objectives of dictatorships are frequently ignoble. But the objectives of such governments are not the only thing—indeed, not even the main thing—we find distasteful. The worst thing about them is their methods.
Nor is it clear that these new domestic intelligence gathering powers will not be abused—or, even more likely, expanded even further to include activities that would now be widely regarded as abusive but that will gradually come to be seen as the new normal.
My own experience in government, as an executive branch lawyer and policy-maker and also as a legislative branch employee attempting to exercise oversight, makes me skeptical of the argument that the government will collect and keep this metadata but hardly ever use it. My government experience was not in agencies whose primary function was intelligence gathering. But in parallel situations—in decision making processes on such matters as the official use of firearms by agency employees, the forced repatriation of asylum seekers to dangerous places, and whether the United States should support a United Nations appointment for a foreign government official who was credibly accused of mass murder—the institutional culture of executive branch decision making bodies was to push all the envelopes as far as they could be pushed in order to resolve the crisis at hand. Arguments along the lines of “Americans don’t do things like this” did not resonate at all in these meetings. Related arguments, such as “we have to consider the optics,” were occasionally successful in smoothing some of the roughest edges of proposed policies, but voices of moderation were generally dismissed as “unhelpful” and sometimes the proponents of these arguments were not invited to the next meeting. When employees of NSA or another agency encounter a problem whose solution they believe might be advanced by more ambitious uses of the collected phone and email records—and assuming that the FISA court does not change its self-described practice of pretty much trusting the government to police itself—I can think of nothing that would make the dynamics of the decision making process any less goal-driven than such processes usually are.
One form this expansion could take would be with respect to how many degrees of separation should be required between you and the terrorist before you become a person of interest. If I interpret correctly the statements of those who defend the NSA metadata collection practices, the several hundred cases in which these data have been used to identify and monitor the substance of communications have involved people who received calls or messages directly from suspected terrorists. But surely it might also be useful to know more about the communications of people who regularly call people who are in touch with terrorists. And what about people whose communication links, either direct or attenuated, are not with suspected terrorists but with leakers such as Snowden and Manning? Or with people who are suspected of facilitating the leaks? After only three or four iterations of any of these moves, the NSA could be listening to an awful lot of innocent people. Most of us don’t know much about who calls the people who call us, but few of us would want to be held accountable for the things that are posted by the friends of our Facebook friends.
This is not to say that the current NSA metadata collection practices can never be justified, but only that they need a stronger justification than has been offered so far. In particular, are there less intrusive methods that would serve the program’s objectives as well or almost as well? Although some proponents of the current program argue that the data will disappear if not collected by the government, it appears that the Drug Enforcement Agency has been collecting similar information from AT&T on a case-by-case basis and that the company does maintain the records for long enough to suit DEA’s purposes.
Perhaps there is some reason that seizing communications records only of persons who appear to have received calls or messages from terrorists would not work. If so, this reason should be clearly articulated and thoroughly debated.
Some legal scholars would argue that the proper place for the debate on whose records will be seized is within the NSA and other agencies of the executive branch. According to these scholars, if the President determines that such seizures will be helpful in the conduct of the war on terror, Congress has no constitutional authority to limit executive action pursuant to this determination.
This view of executive power over foreign affairs, and particularly over the conduct of declared and undeclared wars, finds strong support in the Supreme Court’s decision in United States v. Curtiss-Wright Export Corp.,1 which spoke of “the very delicate, plenary, and exclusive power of the President as the sole organ of the federal government in the field of international relations.” The President’s “sole organ” power, according to the Court in Curtiss-Wright, exists apart from the specific executive authorities enumerated in the Constitution. Rather, it consists of certain “attributes of sovereignty” that the Court regarded as implicit in the concept of nationhood. Later defenders of the “sole organ” power have sought to ground it in the text of the Constitution—principally in the designation of the President as Commander-in-Chief of the armed forces and in what they believe to be the plain meaning of the words “the executive power”—and in statements by some of the Framers. But the Curtiss-Wright Court was emphatic that the power was prior to and independent of the Constitution, having been derived from “the external powers passed from the Crown.”
My own view is that the Curtiss-Wright doctrine is essentially anti-constitutional. Its central assumption is that the scope and allocation of foreign affairs powers in the government of the United States cannot have been intended to be fundamentally different than those of other governments. This amounts to a denial of the central theme of our Constitution: the limitation and distribution of sovereignty, often in ways that King George and Metternich would have regarded as unthinkable.
Especially when the executive action in question is directed at activities of United States citizens within the United States, a more appropriate analysis of legislative and executive authority is the one set forth in Justice Jackson’s concurrence in the Steel Seizure2 case: only when the President acts with the “express or implied authorization of Congress . . . may he be said (for what it may be worth) to personify the federal sovereignty.” When, however, the President “takes measures incompatible with the expressed or implied will of Congress, his power is at its lowest ebb . . . .Presidential claim to a power at once so conclusive and preclusive must be scrutinized with caution, for what is at stake is the equilibrium established by our constitutional system.”3
In the current controversy over metadata collection the executive branch, whatever reservations some of its officials may harbor about Congressional authority to limit domestic intelligence gathering, appears to have observed the procedures prescribed by Congress in Section 215 and related statutory provisions and to have relied on the substantive standards set forth in these provisions. It therefore seems appropriate for Congress to consider whether the collection of all telephone, email, and social media records—well in advance of any particular investigation but in anticipation of the possibility of such an investigation—is consistent with the relevancy standard it thought it was enacting in Section 215. If not, Congress should clarify the standard.
Other proposed safeguards, such as declassifying FISA court judgments or making their proceedings adversarial by including a public advocate—an attorney who might or might not be a government official, and who would be charged with defending the rights not of terrorists but of innocent bystanders—should also be considered but would be of limited use if the legal standard for relevancy is as broad as the NSA and the FISA court have construed it to be. Another interesting proposal, for criminal and/or employment-based penalties against federal employees and contractors who abuse collected information, would provide significant protection only if it were not limited to abuse whose motives were extraneous to the employee’s official duties. Although the collection and analysis of secret information on lovers and rivals—a type of surveillance known in the intelligence community as LOVEINT—is indeed abusive, the privacy of most Americans is far more likely to be violated by overzealous federal agents who sincerely but erroneously believe they are doing their jobs.
Most Americans recognize and accept the need to sacrifice some of their privacy in the interest of security, and most will surely acquiesce in the loss of yet another chunk of their privacy if it can be proved to their satisfaction that the collection of all their phone and internet communications records is essential to the prevention of terrorism. Such acquiescence is less likely if the showing is only that the government entities that specialize in the collection of such records have decided it would be more convenient to collect everyone’s records in advance, just in case they should ever come in handy.
1 299 U.S. 304 (1936).
2 Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579 (1952).
3 Id. at 634-55.
*Grover Joseph Rees is a former law professor, judge, and ambassador. Some of the thoughts expressed in this article are based on his essay in The Treaty Power (symposium with A. Sofaer, H. Koh, and J. Nowak), 43 U. Miami L. Rev. 101 (1988).