The Hoover Institution’s Task Force on National Security and Law has been running an essay series on its blog, The Briefing, “Secrecy and Accountability in the Digital Age.” (Introductory post by Peter Berkowitz, Hoover senior fellow and Task Force chair.) Most of the essays (written by Task Force members, of which I’m one) are about surveillance, big data, NSA, the FISA court, etc. My contribution, by contrast, returns the debate over secrecy, accountability, and oversight issues to where it was before Edward Snowden sent it into an argument over data collection – drones and targeted killing. (Added: Reading some of the comments, please bear in mind that the discussion of accountability and oversight below is about accountability and oversight of drone warfare and targeted killing conducted beyond US borders; it is not about accountability or oversight for the NSA or data collection programs or Edward Snowden.)
It’s a little difficult to recall, frankly, but before Snowden, the most heated national and international debate was about procedures, oversight, and accountability in drone warfare and targeted killing. In part, that’s because the process questions were, for many, a stand-in for campaigning against the programs’ substance. Many critics (including many in the anti-war Left and the Pauline wing of the Republican Party), then and now, regarded
accountability and oversight as a procedural stalking horse for delegitimizing drone warfare as such, and particularly drone programs conducted by the CIA. No process of either targeting or transparency will ever be enough; they will happily take whatever gains are offered by way of transparency as grounds to ask for more. The appetite grows with the eating.
Still, irrespective of strategic political goal, critics are not wrong to “question the adequacy of the accountability mechanisms and oversight of these programs.” I don’t mean that in the sense that the existing mechanisms are inadequate as such or inherently illegitimate; quite the contrary. But legitimacy is a dynamic process in a democracy, and it just is delegitimizing, over time, to have programs that are widely known, even widely supported by the public (as drone warfare is), are discussed extensively by government officials – but with respect to which the government response in any official setting, including the courts, is “neither confirm nor deny.” It undermines public trust to have a democratic government say that it’s not lying to the public, not technically lying, merely because because it refuses comment. The combination of every unofficial and near-official statement saying, x, but then officially saying not-x, not-not-x, not-saying-anything, not-not-saying-anything – is a falsehood, even if a falsehood delivered sotto voce.
Knowingly making a bunch of statements that government – and the public – know can’t all be true, even if no one is actually a falsehood, amounts in combination to a falsehood. It undermines public confidence in processes that necessarily can’t be shared in any detail with the public. It’s one thing if the government neither confirms nor denies events that are mostly secret and which the government has tried to keep secret. But, well, a genuinely public position of what amounts to “irony” – wink-and-nod secrecy, “notional” secrecy – won’t work for a democracy. Say what you mean, mean what you say – but that means that if government says no comment, it really hasn’t said anything, and hasn’t been talking out of the side of its mouth.
Post-2012 election, the administration understood this and thought it was getting its arms around the controversy – always a peculiar idea, since ordinary Americans of both parties overwhelmingly support drone warfare. President Obama delivered a major speech (May 23, 2013, National Defense University) on counterterrorism policy into the future, in which his robust defense of drone warfare and the government’s processes was at its center. One important thing the speech did was acknowledge, within the meaning of the covert action statute, the existence and certain policies of the drone programs. (Lawfare editor-in-chief and Brookings Institution senior fellow Benjamin Wittes and I analyze it closely as Chapter 3 of our book, Speaking the Law: The Obama Administration’s Addresses on National Security Law, appearing serially as online chapters before final hardback publication at the Hoover Institution Press; Chapter 3 can be downloaded as a free pdf here at SSRN.)
Within days of President Obama’s speech, however, Snowden’s stolen documents began to appear in the press, and the result was a remarkable shift of public debate away from targeting processes to the NSA and data collection. Despite the shift in debate, and though a firm defender of drone warfare and targeted killing as a pillar of US “counterterrorism-on-offense,” I argue in my Hoover essay that the structure of accountability and oversight for these programs as “covert actions” under US Code Title 50 (which establishes the CIA, defines “covert action,” and sets out the statutory basics of oversight and accountability) needs some important, though politically quite unlikely, statutory reforms:
Clearly, there is room for certain reforms of secrecy, accountability, and oversight structures, but reform should proceed under two fundamental premises. First, there are legitimate government secrets. Second, accountability and oversight are the responsibility of the political branches alone, not of unelected and unappointed individuals such as Edward Snowden. These two premises are, importantly, not merely claims of politics or national security necessity; they are propositions about legitimacy and its terms. Moreover, this is not simply a call for transparency in the form of more speeches and statements by senior officials, though those have been a vital tool of articulating national security law and its evolution, and of defending its legitimacy. It is about reform at the statutory level — and one might begin with Title 50’s definition of “covert action” as a prelude to oversight processes that flow from it.
This might seem a strange starting point to those familiar with the functioning of the covert action statute since it was overhauled following the 1970s-era Church Committee hearings. The statute, at Section 413b(e), defines “covert action” as “activities of the United States Government to influence political, economic, or military conditions abroad, where it is intended that the role of the United States Government will not be apparent or acknowledged publicly” (the definition then lists certain exclusions). Covert actions are subject to official secrecy or at least official non-acknowledgment, on the one hand, but also subject to corresponding official oversight and accountability to Congress on the other. The accountability and oversight process is robust and appears to have worked pretty well over the past 30 years. It ensures that the president has to know and approve the substance of these activities and that senior leadership in Congress will know about them as well. It might not seem obvious that this stands in need of reform, and less still that it ought to be one of the first steps.
But three factors come together to create weaknesses in the covert action accountability regime: precision stand-off weapons; increased strategic reliance on discretely targeted counterterrorism uses-of-force that are enabled by new technologies; and the blurring of different kinds of counterterrorism uses-of-force that do not neatly fit into categories of the statutory definition. Most obviously, the existing categories do not adequately address precision weapon technologies in counterterrorism operations. As everyone knows today, these operations have a range of public knowledge about them that includes ‘truly secret'; ‘mostly secret and unacknowledged'; ‘widely known but still unacknowledged'; ‘everybody knows and the US government talks about it under a shredded fig leaf'; and finally, ‘merely preposterously unacknowledged’.
Even if statutory reform of Title 50 were politically possible, however, it would be made more difficult because it would require a corresponding change in the exemptions to the Freedom of Information Act (FOIA). Why? Because the biggest legal driver of characterizations of drone warfare as “covert action” is the risk that any form of acknowledgment would persuade a judge, somewhere, to deem the statute’s new “levels” of acknowledgment sufficient warrant to require disclosure by the intelligence community in response to a FOIA inquiry. If you’re one of the committed critics of drone warfare, or of government secrecy as such, this is a feature. If you’re most people, however, and certainly if you are the Obama administration or most members of the Congressional intelligence oversight committees, this is a bug, and a very big one. The Obama administration has both declassified certain facts about the existence of the drone programs and has acknowledged parts of this in FOIA litigation; but necessary and legitimate secrecy requires certainty that it will be maintained over time, and that needs to be established in advance by statute, not by evolving and uncertain court rulings.