SECRECY, SURVEILLANCE, AND THE SNOWDEN EFFECT - After Snowden: Privacy, Secrecy, and Security in the Information Age (2015)

After Snowden: Privacy, Secrecy, and Security in the Information Age (2015)



THOMAS BLANTON is Director of the National Security Archive at George Washington University in Washington, D.C. In April 2000, the Archive won U.S. journalism’s George Polk Award for “piercing self-serving veils of government secrecy, guiding journalists in search for the truth, and informing us all.” A graduate of Harvard University, he filed his first Freedom of Information Act request in 1976 as a weekly newspaper reporter in Minnesota. His articles have since appeared in The International Herald Tribune, The New York Times, The Washington Post, The Los Angeles Times, The Wall Street Journal, The Boston Globe, Slate, The Wilson Quarterly, and many other publications. He is the author of numerous books, most recently Masterpieces of History: The Peaceful End of the Cold War in Europe, 1989, coauthored with Svetlana Savranskaya and Vladislav Zubok, which won the Arthur S. Link-Warren F. Kuehl Prize for Documentary Editing.

THE EDWARD SNOWDEN leaks starting in June 2013 not only challenged the U.S. government’s surveillance policies but also forced major change in the national security classification system. In fact, only a year after the Snowden documents began hitting the headlines, and directly in response to those disclosures, the U.S. government and particularly its intelligence agencies had actually declassified and published more pages of previously secret surveillance documents (over 3,000 pages) than had the combined media partners holding Snowden materials (2,227 pages as of August 2014).1 At one point in the summer of 2013, the former National Security Agency Director General Michael Hayden—giving yet another interview defending the surveillance programs—complained he could hardly keep up with the official disclosures, so he did not really know what was still a secret and what was not.

Of course, before Snowden, every single page of those new official releases was withheld from the public. The government even claimed—before Snowden—that wiretap court orders and pleadings had to be withheld in full; even the sections of the documents that simply cited and repeated the text of the Fourth Amendment (the Constitution)—were now “top secret.”2 Even after Snowden, the newly declassified pages contained so many redactions, blackouts, and whiteouts—ostensibly on continued national security grounds—that a single Snowden page arguably amounted on substantive grounds to two or even three of the official pages.

The U.S. secrecy system had long privileged at the highest level the communications intercept activities of the National Security Agency, so much so that the standing joke among NSA watchers held that the initials stood for “No Such Agency.” Intelligence products that contained COMINT almost automatically qualified for top secret controls (the definition of “top secret” holds that release of this information would cause “exceptionally grave” damage to national security). From 1950 to 1982 (when the statute protecting CIA agent identities was passed), only one category of “national defense information” was precisely defined in statutes like the Espionage Act, and that was the COMINT Act, Section 798 of Title 18—covering “communications intelligence activities” and “any code, cipher, or cryptographic system” and any classified information obtained from communications intelligence.3 That history of privileged secrecy—presumptive secrecy—is worth a closer look, particularly the lessons it holds for today about the abuses of surveillance power enabled by secrecy, and the potential balance we might strike with public demands for transparency.

This chapter attempts to situate the current debates over secrecy and transparency of surveillance activities in the policy argument as old as this republic, dating back to the revolutionary and necessarily secret activities of the Continental Congress. The narrative here reviews the various and diverse analyses of the national security secrecy system and its incentive structures, and sketches the rise of the NSA itself, as secrecy masked its origins in the Cold War and its vast expansion parallel to the Internet in the 1980s and 1990s. The chapter will describe the renewed secrecy shrouds after the terrorist attacks of 9/11, how this fed the agency’s culture of reflexive secrecy and enabled the aggressive surveillance activities now being partially rolled back—after Snowden. Finally, the chapter will offer the comparisons from more than a dozen examples of the sea change in secrecy from before and after Snowden, and how these belated acknowledgments by the government actually illuminate the credibility gap between what officials claimed and what we now know.

The analysis here, though, cannot offer encouragement that the post-Snowden transformation in the secrecy system amounts to sustainable, much less permanent, change. The credibility of the NSA and its supervisors is shot right now, as the result of demonstrably false statements made to the Congress, the courts, to the public, and to each other—all revealed by the Snowden documents. The new intelligence community transparency is transparently an attempt at public relations rather than systemic reform, although a few institutional arrangements such as the Privacy and Civil Liberties Oversight Board offer some real promise. Right now the trends are in the openness direction, but the same was true of the CIA in the mid-1970s in the midst of congressional investigations and front-page stories, many of them provided by the CIA coming clean itself. Yet within only a few years, the cowboys were back in charge at Langley, and such abuses and fiascos as the Iran-Contra scandal resulted.

Justice Potter Stewart in the famous concurring opinion in the Pentagon Papers case (1971) provided a lasting, useful guide to national security secrecy: “For when everything is classified, then nothing is classified, and the system becomes one to be disregarded by the cynical or the careless, and to be manipulated by those intent on self-protection or self-promotion.… The hallmark of a truly effective internal security system would be the maximum possible disclosure, recognizing that secrecy can best be preserved only when credibility is truly maintained.”

Arguably, the national security classification system is trending right now, thanks to Snowden, toward Justice Stewart’s definition of a truly effective system. But to date, the new disclosures pertain mostly to the surveillance activities that Snowden had access to, with only a few transparency impacts on the rest of the U.S. national security colossus (the most important involve the intelligence budget and drone targeting). The new openness may turn out to be a temporary condition rather than the “zero-based budgeting” that would rebuild the classification system from the ground up. Even with the Snowden revelations—in fact, as proven by Snowden’s ability to leak—we still have a secrecy system that places low fences around vast prairies of classified information, when we actually need very high fences around the tiny graveyards that hold the real secrets that would get someone killed.4

President Obama wrongly defined the surveillance and secrecy challenge in his January 17, 2014, speech as that of “maintain[ing] the trust of the American people, and people around the world.” He was more correct, in the same speech, to say, “Given the unique power of the state, it is not enough for leaders to say: Trust us, we won’t abuse the data we collect. For history has too many examples when that trust has been breached.… [O]ur liberty cannot depend on the good intentions of those in power; it depends on the law to constrain those in power.”5 Building a credible national security secrecy system will require much less secrecy, many more checks and balances built into law, and, yes, much to the discomfort of those in power, more Snowdens.


U.S. government today ranks among the most open in the world, even in matters of national security. Foreigners look in awe at the public and often contentious confirmation process for American spymasters; the frequent debates over specifics on cost overruns in weapons systems; the regular military and foreign policy reporting to Congress, much of which takes place in televised hearings; the wide variety of internal and external audit mechanisms, such as inspectors general; and the often routine openness in other areas of government, which reformers in the rest of the world are trying to emulate. At the same time, however, the U.S. government creates and keeps more secrets than any other state in the world—in fact, if the size of the U.S. national security establishment is any guide, U.S. agencies have more secrets than the next ten largest countries put together. Most open and most secret—such is the paradox of U.S. national security secrecy.

The scholar Daniel Hoffman’s analysis of the Founders of the American republic shows they deliberately meant to overturn monarchical notions of secrets of state, the absolutism found in King Louis’s famous exhortation, “L’état, c’est moi.6 Publicity for congressional proceedings would be the norm, for example—the U.S. Constitution only specifically mentions secrecy a single time, in Article I, Section 5: “Each House shall keep a Journal of its Proceedings, and from time to time publish the same, excepting such Parts as may in their judgment require secrecy; and the Yeas and Nays of the Members of either House on any question shall, at the Desire of one fifth of those Present, be entered on the Journal.”

Conceptually, the Founders’ formulation offers some real guidance for today. An echo may be found in statutes 200 years later, like the Freedom of Information Act, and even in President Obama’s “Day One” commitments to open government—that we need a presumption of openness for government information, with “exceptions” for real secrets as established by some democratic and legal process. Of course, this concept is turned on its head by the almost absolute secrecy around modern communications intelligence matters, before Snowden, where the presumption was secrecy, leaving exceptions to be staked out by leakers or guessed at by the public. This issue—whether a presumption of openness or of secrecy—lies at the heart of the current debate over surveillance transparency.

As on slavery, the Founders’ intent failed to settle the issue. Even at the time, the debate over secrecy raged, and both reformers and securocrats can today find support for their views in what the Founders said and did. Reformers like to cite James Madison’s admonition that “a people that mean to be their own Governors, must arm themselves with the power which Knowledge gives.” However, Madison also defended, against Thomas Jefferson’s criticism, the closed doors of the Constitutional Convention, so that delegates could be protected from outside pressures, change their minds, avoid posturing, and build consensus (the 1787 proceedings were not published in full until 1819, thirty-two years later).7 The Continental Congress, of course, ran the revolution through various secret committees, which had permission to hide even from the rest of Congress the identities of their correspondents; and the first legislative appropriation of the newly independent country gave President George Washington a secret fund to use for ransoming American hostages abroad, or bribing foreign officials—the historic precursor to intelligence budget secrecy today.8

Secrecy figured centrally in the constitutional debates over the leading role of the executive. Alexander Hamilton insisted, in his Federalist Paper number 70, that “[d]ecision, activity, secrecy, and dispatch” were the main qualities required of the executive (he was arguing against those trying to weaken the President’s powers). Yet John Dickinson of Delaware, also arguing for a strong but checked and balanced presidency, pushed back: “Secrecy, vigor & dispatch are not the principle properties reqd. in the Executive. Important as these are, that of responsibility is more so.…”9

These examples illustrate a tension as old as the republic. Perhaps most eloquent of the Founders was Patrick Henry, who wrote, “to cover with the veil of secrecy the common routine of business is an abomination.…” Yet Henry also explained, “Such transactions as relate to military operations or affairs of great consequence, the immediate promulgation of which might defeat the interests of the community, I would not wish to be published, till the end which required their secrecy should have been effected.” However, in the context of an endless war against terror, when would that be? And who would judge?

These questions become most difficult in the context of the modern administrative state. By the 1930s, an alphabet soup of new federal programs joined the regulatory agencies invented by the Progressives, and as government expanded, so did pushback by interest groups and reformers, creating transparency mechanisms like the Federal Register (1936) and rules that would become the notice-and-comment requirement in the Administrative Procedure Act of 1946. During World War II, the U.S. government dramatically grew into what analyst Harold Relyea (of the Congressional Research Service) has described as the “national security state,” with secrecy as “one of its primary characteristics.”10 Writing just before World War I, as corporate and governmental bureaucracies began to proliferate, the German sociologist Max Weber situated the secrecy problem in core bureaucratic practice: “Every bureaucracy seeks to increase the superiority of the professional informed by keeping their knowledge and intentions secret. Bureaucratic administration always tends to be an administration of ‘secret sessions’; in so far as it can, it hides its knowledge and actions from criticism.… The concept of the ‘official secret’ is the specific invention of bureaucracy, and nothing is so fanatically defended by the bureaucracy as this attitude.… In facing a parliament, the bureaucracy, out of a sure power instinct, fights every attempt of the parliament to gain knowledge by means of its own experts or from interest groups.”11

There were of course real reasons for the national security secrecy, to keep enemies in the dark, but mixed motivations characterized the American secrecy system from the start. Take the Manhattan Project, which built the atomic bomb and also gave birth to most of the features of the modern classification system, such as security clearance investigations, compartments, “need to know” restrictions, and document markings. When the Manhattan Project’s commander, the legendary General Leslie Groves, later worked on his memoir with his son, Groves produced a fascinating list of eight justifications for the shroud of secrecy he had placed over all atomic matters, according to his biographer, Robert S. Norris. The Germans and the Japanese provided the top two motivations for secrecy, of course, and the then-allies-future-enemies of the Soviet Union were number three. But “all other nations” including the other allies, like Great Britain, were target number four. Most revealing was number five, “those who would interfere directly with the progress of the work, such as Congress and various executive branch agencies”! Number six limited discussion to a small group of officials; number seven aimed for military “surprise”; and number eight mentioned “need to know” as a management tool, perhaps to control those unruly scientists.12 Max Weber would find much to work with in the Groves list.

Those scientists had been the inventors of nuclear secrecy, not the military, not the security officers. Enrico Fermi credits the Hungarian refuge Leó Szilárd as the one who came up with the idea of self-censorship, of removing physics research related to fission from the scholarly literature after 1939. What Groves and the security officers did in the Manhattan Project was to take the scientists’ peer-review censorship committees and turn them into a pervasive bureaucratic system. Afterward, Szilárd himself complained that in fact the controls on information imposed on the scientists contradicted everything he knew about how scientific progress occurs, and specifically that the compartmented approach—separating scientists into small groups based on “need to know,” as if Groves possessed enough omniscience to decide the need—prolonged the war in the Pacific for more than a year by holding up recognition of how quantities of U-235 could be produced in mass.13 However, the securocrats, not the scientists, were in charge.

Communications intelligence joined nuclear weapons as the main drivers of the new information security system during World War II and after. Breaking the Japanese and German codes ranked among the most sensitive secrets in the U.S. government. When the Chicago Tribune’s front-page coverage of the U.S. victory at the Battle of Midway in 1942 revealed—without explicitly saying so—that the United States was reading Japanese naval messages, security officials wanted to prosecute the newspaper and the Justice Department convened a grand jury to do so. Ultimately the Navy Secretary called off the proceedings because apparently Japan had missed the Tribune story as well as the subsequent controversy (including a Member of Congress actually confirming the fact of the intercepts)—thus the U.S. government feared that any further legal action might bring the matter to Japan’s attention.14 Here, the judgment call in effect concluded that the maximal pursuit of secrecy in the name of security would actually damage that security.

Broken codes also figured prominently in the widely read critiques of secrecy after the Cold War by U.S. Senator Daniel P. Moynihan, especially in the 1997 report of the commission he chaired. The commission’s very title—“On Protecting and Reducing Government Secrecy”—reflected some cognitive dissonance among its congressional sponsors and cochairs (Moynihan the Democrat from New York and Larry Combest, conservative Republican from Texas), although reformers would argue that one better protects the real secrets by reducing the rest. For Moynihan, most striking were the ways in which institutional secrecy enabled the security establishment to get the main problems wrong, and to forestall public knowledge and debate of major Cold War issues, from the true dimensions of Soviet spying in the 1940s to the coming collapse of the Soviet Union in the 1980s. Moynihan focused his argument on “Venona”—the U.S. intelligence community’s decrypted intercepts of 1940s Soviet diplomatic and military cable traffic that proved the existence of multiple Soviet spy rings in the United States. While the U.S. government brought high-profile (and highly controversial) prosecutions against many of these accused spies in the late 1940s, the government refused to declassify even the existence of the intercepts or their use in driving the prosecutions—which Moynihan subsequently believed would have settled the public debate.

Moynihan argued that the excessive secrecy about the “Venona” COMINT left free rein for both McCarthyism and denial, both hysterical overreach and a false conclusion of persecution. All this because of a secrecy system that ironically enabled both understating the real Soviet spy threat of the 1940s and misunderstanding how the United States had largely eliminated that threat by the early 1950s, when the Red Scare mounted into paranoia. For Moynihan, the real cost of national security secrecy included the McCarthyism divisions in society, and the subsequent inflation and overinvestment in military power from overstating the strength of the Communist system by the 1970s and 1980s.15

Moynihan’s subsequent 1998 book, Secrecy, joined the volumes that have been written on the history of Cold War secrecy, and on the periodic scandals that threw back the national security shrouds for significant periods of time. Suffice it here to say that this history is notable for the patterns of enormous growth in classified information since World War II, interrupted in almost every decade by scandal-driven openness, soon followed by retrenchment and more secrecy. The 1970s forced the greatest openness on the U.S. intelligence community, from the Pentagon Papers revelations to the Watergate abuses to the Church and Pike committee investigations of spy agency lawbreaking. For the National Security Agency, that period represents perhaps the only analog to the current post-Snowden period of investigation and exposure (about which much more below).

But during the 1980s, the Iran-Contra scandal showed how limited the 1970s reforms really were, and exposed the return of some of the worst practices of Cold War secrecy and covert operations on and off the books. Investigators later complained that the National Security Agency was the only intelligence agency that stuck to its secrecy claims—forcing unwarranted redactions in the official Iran-Contra reports. Those same reports actually gave NSA some credit for refusing to edit its briefing flow to top officials (intercepts of the arms-for-hostages negotiations, for example) despite White House attempts at cover-up, and those officials’ attempts to seem “out of the loop.”16

The end of the Cold War of course brought large reductions in the military and intelligence budgets, and a parallel decline in national security secrecy during the 1990s. Perhaps most notably, then–Vice President Al Gore’s interest in longitudinal environmental data drove the declassification of decades of spy satellite photos, when less than ten years earlier, a Navy analyst had gone to prison for leaking one of them to Jane’s Defence Weekly. The Clinton administration wrote a new executive order on secrecy, prompting the declassification of hundreds of millions of pages of historic secrets. Also during the 1990s, Congress drove major declassification programs, responding to the hit conspiracy film JFK by establishing a blue-ribbon commission responsible for releasing the classified records around the 1963 Kennedy assassination—including the CIA’s covert operations files on Cuba. Similarly, responding to news reports of Nazi war criminals living in the United States under protection from agencies like the CIA, Congress set up another panel to drive the release of both Nazi and Japanese war crimes records, including the unprecedented release of CIA “name files” on key Nazis like Adolf Eichmann.17

The Moynihan report of 1997 paradoxically represented the high point of the 1990s secrecy reform movement. By 1998 Congress had lost its interest in JFK- and Nazi-style declassification projects, as Chinese spying hysteria swept Capitol Hill (ultimately to collapse with a governmental apology to the poster child, nuclear scientist Wen Ho Lee). The simultaneous Clinton impeachment process devoured the available oxygen in the public policy sphere and undermined the previous progress on open government. The CIA reneged on its declassification program promises on the day the House of Representatives impeached Clinton; and Congress even passed an official secrets act, criminalizing leaks to the media, only to have Clinton veto the bill at the last minute in 2000.

Intelligence officials had persuaded Congress to pass that vetoed official secrets act by presenting classified briefings about cases of leaks that allegedly did severe damage to national security, but some years later, when the FOIA opened those case studies to public view, the stories fell apart.18 Two in particular centered on communications intelligence leaks with relevance to Snowden today. The classified briefings to Congress in 2000 had claimed that a leak to the columnist Jack Anderson in 1971 had cut off U.S. access to the car phones of Soviet Politburo members driving around Moscow, thus depriving the United States of an extraordinary intelligence source. However, Anderson’s biographer Mark Feldstein found Anderson’s notes of phone calls checking with government officials in which Anderson agreed to hold the story. After the CIA Director and the Justice Department confirmed that the Soviets knew about the intercepts, Anderson then wrote about the intelligence coup, but not about the specifics of the car phones: “For obvious security reasons, we can’t give a clue as to how it’s done.”19 The CIA’s James Bruce, the “Vice Chairman of the Director of Central Intelligence Foreign Denial and Deception Committee,” responsible for the classified case studies briefed to Congress, committed his own denial and deception, either advertently or inadvertently missing the entire sequence of events.

The CIA and James Bruce compounded the error by focusing Congress’s attention—in secret sessions—on the purported loss from leaks of U.S. intelligence’s ability to listen to Osama bin Laden’s satellite phone. The Bruce briefing artfully used a quote from White House Press Secretary Ari Fleischer on June 20, 2002 (repeated by President George W. Bush in 2005), blaming a leak and a press report in The Washington Times for Bin Laden’s ceasing to use his satellite phone in 1998, thus thwarting the National Security Agency’s intercepts. The claim also appeared in a best-selling book on terrorism written by former Clinton administration aides, and even in the 9/11 Commission’s final report after Bruce and his team gave classified briefings to the Commission. However, once the Bruce/Bush/Fleischer claim could actually be fact-checked in public, it fell apart, as the 9/11 Commission cochair Lee Hamilton subsequently acknowledged. The purported 1998 leak had been preceded by Bin Laden’s own discussion in a 1997 CNN interview of his satellite phone use, by the 1996 assassination of the Chechen leader Dzhokhar Dudayev by a Russian cruise missile targeted on his satellite phone, and by the August 1998 U.S. cruise missile attack on Bin Laden’s own training camp in Afghanistan, from which he had departed only hours earlier. Like so many alleged secrets, this one turned out to be public information; and The Washington Post concluded the leak story actually ranked as an “urban myth.”20

The backstory of the official secrets act that Congress passed and Clinton vetoed in 2000 shows that secrecy was on the rise even before the terrorist attacks of 9/11. After 9/11, with a heightened sense of threat and even paranoia, government at every level moved its operations under wraps. The massive increases in new security classification decisions might well have happened anyway with the surge in military and intelligence activities, the doubling of those budgets, and the two wars in Afghanistan and Iraq ordered by President George W. Bush, but in addition to the perhaps understandable secrecy related to current war fighting, the securocrats took the opportunity to go back and reverse decisions about historic records, about the documents released in the 1990s, and about whole categories of records having nothing to do with the war on terror that now were pulled back to the vaults. Securocrats at the Department of Energy, the CIA, and the Air Force even began formal but secret reclassification programs (actually full employment programs for their retirees, now working as contract reviewers) to take back the millions of pages released in the 1990s.21

And the new secrecy generated its equal and opposite reaction in the form of leaks. Or more precisely, the overreaching claims of government authority that drove both the new secrecy and even more controversial programs of detention, interrogation, surveillance, and assassination also sparked a pluralistic system’s internal antibodies against authority in the form of whistleblowers. Cases like those of Thomas Drake at the National Security Agency and Thomas Tamm at the Justice Department are discussed elsewhere in this book, as are the advances in electronic surveillance that empowered government prosecutors who started under George W. Bush and then continued under Barack Obama, to bring more leak prosecutions in the last five years than all previous administrations put together.

Thus, even before Snowden, the policy debate over secrecy had reached a new level of ferocity, with members of Congress and conservative analysts calling for prosecution of The New York Times, for example, over various stories based on leaks.22 The media vigorously defended itself with two arguments, first the one about journalistic ethical responsibility for getting comment from any affected parties, especially on national security issues in which reporters must have a dialogue with knowledgeable government officials before publishing classified information. Such a responsibility extended even to the point of censoring themselves (as did The Washington Post on the location of the CIA’s secret prisons in countries like Poland, as did even WikiLeaks on the names of people who talked to U.S. embassies in the Wiki cables). Famously, The New York Timesheld the first version of its story on warrantless wiretapping under President George W. Bush for over a year, until reporting by its Justice Department correspondent (Eric Lichtblau) both confirmed its intelligence correspondent (James Risen) and established that senior officials at Justice believed parts of the program to be illegal.23

The other media argument of course has the virtue of historical consistency. At the core of this argument is the analysis in the famous affidavit in the Pentagon Papers case by New York Times editor Max Frankel:

Presidents make “secret” decisions only to reveal them for the purposes of frightening an adversary nation, wooing a friendly electorate, protecting their reputations. The military services conduct “secret” research in weaponry only to reveal it for the purpose of enhancing their budgets, appearing superior or inferior to a foreign army, gaining the vote of a congressman or the favor of a contractor. The Navy uses secret information to run down the weaponry of the Air Force. The Army passes on secret information to prove its superiority to the Marine Corps. High officials of the Government reveal secrets in the search for support of their policies, or to help sabotage the plans and policies of rival department. Middle-rank officials of government reveal secrets so as to attract the attention of their superiors or to lobby against the orders of those superiors.

Or to quote President John F. Kennedy, the ship of state is the only vessel that leaks from the top.24 In such a context, the media must report and parse the leaks, and does so under the protection of the First Amendment.

Yet the Constitution is not “a suicide pact,” as we are reminded by the securocrats.25 An absolutist First Amendment position remains untenable, according to the federal courts. So in between the securocrats and the First Amenders come the hairsplitters. The most thoughtful of the fencewalkers, Princeton professor Rahul Sagar, situates the problem—the lack of balance between secrecy and openness—in a series of constitutional silences. Sagar argued—before Snowden—that while leaks of national security information like the Times’s 2005 story on warrantless wiretapping are necessary to accountability, they should be maximally discouraged, involving as they do the breaking of oaths, the undercutting of representational democratic channels for decision-making about what information should be public, and mixed motivations by the leakers and the reporters.26

Rebutting Sagar are some interesting apostates. Foremost would be Harvard Law professor Jack Goldsmith, himself no slouch in defending executive privilege as a former head of George W. Bush’s Office of Legal Counsel. Goldsmith wrote some of the still-secret legal opinions justifying warrantless snooping. After Snowden, he became more forthright in his conclusions. Before Snowden, Goldsmith only argued that Bush had made a mistake in seizing executive power unilaterally and secretly, when Congress and the public would have granted almost but not quite all of that power had Bush asked.27 After Snowden, Goldsmith argued against Sagar’s notions by maintaining that secrecy in a democracy has a presumption of illegitimacy. Not because there are no real secrets—of course there are—but because “in a constitutional democracy where the People ultimately rule, secrecy from the People is aberrant (for they cannot govern what they do not know), and it demands special justification (thus the presumption against its legitimacy).”28 Similarly, Suffolk University law professor Alasdair Roberts has argued that the technologies and specifics of intelligence gathering can and likely should be classified, but the policies surrounding their applications cannot be secret and still have legitimacy or accountability. The Snowden revelations have now given even more force to the Goldsmith and Roberts formulations.


The National Security Agency before Snowden enjoyed a presumption of secrecy—an idyllic form of self-protection that came in handy in surviving all kinds of scandal. There was some real basis to the secrecy—cryptological codes and methods and technologies were appropriately classified—but the absolute secrecy covered matters ranging from the pizza boxes produced from the NSA’s shredded papers all to the way up to the policy decisions that allowed, as Snowden’s leaks showed, the collection of the “to-from-date-time-duration” metadata on every single phone call made in the United States on the Verizon phone network, to name just one such company. The contradiction was present at the creation of the NSA—presidential orders pursuant to the National Security Act of 1947 established the NSA in 1952, but not until 1957 was the agency publicly acknowledged in the Government Manual. As mentioned, the initials quickly became a quip, “No Such Agency,” yet the “secret” of its existence actually showed the limits of the secrecy system: Washington-area newspapers had revealed the NSA’s existence in stories as early as 1954 about the building of its sprawling Ford Meade, Maryland, campus.29

That campus would ultimately become what writer James Bamford called “Crypto City”—an enormous collection of thousands of employees and even more thousands of contractors housed in dozens of buildings and spending tens of billions of taxpayer dollars each year. Bamford’s trilogy of books on the NSA argues that the technological opportunities drove much of the expansion of NSA from 1952 onward. In 1929, at the time of Herbert Yardley’s “Black Chamber,” the entire code-breaking apparatus of the U.S. government could fit into a single 25-foot-square room. By the time of the 9/11 attacks, the NSA occupied a whole campus at Ford Meade, with 32 miles of roads, parking lots amounting to 325 acres, and more than 7 million square feet of office space. All covered by a presumption of secrecy.30

For the purposes of a post-Snowden analysis, what is most striking about the NSA’s secret history has been the dramatic ebbs and flows of secrecy around the agency’s operations. The NSA’s secret creation in 1952 took five more years before public acknowledgment in 1957, but actual operational exposure like Snowden’s would come two decades later, in the 1970s. From its earliest days, the NSA picked up a preexisting program, called Project SHAMROCK, that intercepted every single overseas telegram starting in 1945, millions each year, with the secret cooperation of all three cable companies (RCA, Western Union, and ITT). This ultimately produced (as of the early 1970s) some 150,000 messages each month that were selected for NSA’s analysts to review.31 Then, starting in 1962 with some Mafia figures whose names were provided by the FBI and the Justice Department, the NSA cooperated with the CIA on a watch list program, code name MINARET, that ultimately included some 300,000 people in the index, mostly antiwar dissidents and political activists and public figures.32 Here is the precedent for the post-9/11 warrantless wiretapping, but the earlier program at least had specific targets rather than mass collection of data. The targets included the Reverend Martin Luther King Jr., Tom Wicker of The New York Times, even Washington Post humor columnist Art Buchwald! The list also featured two U.S. senators, the Democrat Frank Church from Idaho and Republican Howard Baker of Tennessee.

Senator Church went on to lead the congressional investigations (1975–1976) that exposed the SHAMROCK and MINARET programs—with much help from insider leaks and investigative reporters. Church concluded: “The United States government has perfected a technological capability that enables us to monitor the messages that go through the air.… That capability at any time could be turned around on the American people, and no American would have any privacy left, such is the capability to monitor everything—telephone conversations, telegrams, it doesn’t matter. There would be no place to hide.”33 Those conclusions, along with considerations of immunity for the corporations that had cooperated with wiretapping, and considerations of admissibility of evidence in court, led to the Ford and Carter administrations’ support for what ultimately became the FISA Act of 1978.34

For the NSA, the mid-1970s exposure of their darkest secrets meant a sea change in both operations and self-perception. One senior NSA officer wrote up their new realization in a secret internal newsletter: “We might as well get used to it, used to feeling exposed and unprotected at moments when our accustomed and familiar anonymity seems to be snatched away.” She hoped that their “time of indecent exposure” was coming to an end, but concluded: “Packs of hungry animals of various breeds, having caught the scent, are out there gnawing at the foundations of the storehouses, sniffing and rooting for more beans.”35

Indeed, more “beans” about the SHAMROCK and MINARET programs have emerged in the years since the Church Committee. Through Freedom of Information and Mandatory Review requests, the nongovernmental National Security Archive pried loose the NSA history with the actual names from the watch lists, and a remarkable observation by the agency’s own historian: “The project, which became known officially as Minaret in 1969, employed unusual procedures. NSA distributed reports without the usual serialization. They were designed to look like HUMINT reports rather than SIGINT, and readers could find no originating agency. Years later the NSA lawyer who first looked at the procedural aspects stated that the people involved seemed to understand that the operation was disreputable if not outright illegal.”36

The pendulum swings of NSA secrecy have been perhaps most visible in the agency’s treatment of its outside biographer, journalist James Bamford. Through diligent interviewing of NSA retirees and spelunking in their private archives in the late 1970s, Bamford amassed such documentary gems as a complete series of the NSA employee newsletter and the Justice Department’s internal discussion (as of 1978) of possible prosecution of the NSA over MINARET and SHAMROCK. Before his 1981 book could be published, Bamford faced prosecution threats himself from the government, and the NSA’s security agents repossessed various archival collections used by him from the open shelves of libraries. However, when undertaking his sequel in the 1990s, Bamford enjoyed direct interview time with the NSA Director, a tour of the facilities at Fort Meade, and ultimately even an NSA-hosted book party—so, perhaps not surprisingly, when the book, Body of Secrets, came out in 2001 (before 9/11), there was a noticeably more favorable tone in his depiction of the NSA.37 Likewise, in the 1990s, the NSA actually declassified key directives like the 1993 USSID and the minimization procedures used, for example, to protect First Lady Hillary Clinton’s privacy during foreign travel, along with the NSA’s own 1998 organizational chart. After 9/11, the next version of the organizational chart came out almost completely blacked out, redacted.38

Between 9/11 and the warrantless wiretapping controversy that began with The New York Times story on December 16, 2005, the main secrecy battle at the NSA concerned ancient history—the Vietnam War. An enterprising NSA historian had gone back to the original raw signals intercepts from the controversial Gulf of Tonkin episode in August 1964, in which the Johnson administration claimed two unprovoked attacks by North Vietnamese ships on the U.S. Navy as the basis for bombing the North and escalating the war in the South. The historian’s classified article uncovered the way in which intercepted North Vietnamese naval messages had been mistakenly prioritized, conflated, even merged (and later covered up) in such a way as to reinforce what policy makers wanted to hear, confirming a second attack that never happened but that would justify military response and a ramping up of the Vietnam War. It took years of Freedom of Information pressure, news media coverage, a leak of part of the history, and embarrassment for the NSA before the underlying historical article would be declassified in 2005.39

These ebbs and flows of NSA secrecy illustrate some of secrecy’s costs. Because of the reflexive secrecy, the government fails to understand its own history. The secrecy reinforces that eternal temptation for policy makers to cherry-pick information that suits their preferred policies; suppress dissent and alternative views; and cover for incompetence, inaccuracy, and illegality. In fact, the record shows the 9/11 attacks themselves were enabled by excessive secrecy. Likewise, the dirty little secret of the vastly expanded national security secrecy after 9/11 is that the new shrouds were actually counterproductive. This may seem a controversial finding, but it is one shared by both of the major official investigations of the 9/11 attacks.

For example, the joint congressional investigation of 9/11 found “the U.S. Intelligence Community was involved in fighting a ‘war’ against Bin Ladin largely without the benefit of what some would call its most potent weapon in that effort: an alert and committed American public.” The 9/11 Commission found the problem was not a lack of dots to connect, a lack of haystacks to search, or a lack of intercepts to read, but the active withholding of information inside the government by the CIA from the FBI that might have caught the two future hijackers living in San Diego. According to the 9/11 Commission, the CIA’s interrogations of 9/11 planner Ramzi bin al-Shibh revealed the hijackers would have called off the attacks if there had been any publicity about the arrest of Zacarias Moussaoui, the Minnesota flight school student who only wanted to learn to steer, not take off or land. Most of the ten “operational opportunities” to deter the attacks involved the failure to share information. And after 9/11, first responders complained that the censorship of online maps and diagrams (for example, of Air Force One) blocked effective emergency response. The FBI complained that the haystack approach of vacuuming up phone calls actually generated hundreds of thousands of unproductive leads, overwhelming the ability of agents to focus on real threats.40

By 2005, an emerging consensus began to include overclassification as among the “emerging threats” to U.S. security. A congressional hearing in March 2005 heard estimates of the problem from a number of experts, including the Deputy Undersecretary of Defense for Counterintelligence and Security who answered “50–50” when asked how much overclassification there was. Quoted in the hearing record was 9/11 Commission cochair Tom Kean (former Republican governor of New Jersey). After he finished going through the most recent Osama bin Laden intelligence, he remarked that “[t]hree-quarters of what I read that was classified shouldn’t have been.” Also quoted was President Reagan’s former National Security Council Executive Secretary Rodney McDaniel, who told the Moynihan Commission in 1997 that only 10 percent of classification was for “legitimate protection of secrets.”41 Perhaps it is no coincidence that 2005 also brought the first public notice to the NSA’s programs of warrantless wiretapping.

The secrecy hid all the key policy turning points for surveillance after 9/11. The record post-Snowden shows that the NSA received three new authorizations to go beyond its statutory limits on October 4, 2001, based just on presidential commander-in-chief authority. Another policy turning point secretly occurred in March 2004, with the now-famous hospital room confrontation between the White House and the attorney general (backed up by his deputy and the FBI Director) over the e-mail metadata intercept program that was collecting way too much American citizen information. All in secret, that confrontation shook President Bush to the point that he overruled his own staff and scaled back the program, which ultimately (secretly) ended in 2011 on inefficiency grounds. Just as the secrecy debate was erupting in Washington in 2005, the NSA secretly passed a milestone by collecting all the e-mail and phone calls produced in Iraq that year. The newly proven capacity to vacuum up everything became the template for collections not only abroad but at home, supervised by the aggressive new NSA Director, General Keith Alexander. First Iraq, then the world.42

The hyperactive growth of the intelligence establishment did not go unnoticed. Perhaps the most innovative attempt to grasp the full dimensions of the U.S. government’s spy agencies came with the 2010 Washington Post series and subsequent book by Dana Priest and William Arkin, Top Secret America. Here, counterterrorism is the driver of extraordinary growth especially among government contractors, with the greed motivation front and center. In effect, Priest and Arkin argue, national security secrecy served mainly to protect contractors’ profit margins. The core tools of the ambitious Priest/Arkin mapping process came from contractor job announcements and job descriptions, often containing descriptions of intelligence programs and even code words. Clinching their case was the revolving door created by the first Director of National Intelligence, a centralizing reform of the community pushed by the 9/11 Commission. But former NSA Director Admiral Mike McConnell did not last long in the job, and left the government to take his 1,000 percent raise, as a $2 million a year executive at an intelligence community contractor.43

In all this reporting, there were more than hints of what the NSA was up to. In 2005, the National Security Archive obtained through FOIA the NSA’s transition briefing prepared for the incoming Bush administration in 2000. That briefing straightforwardly announced that the reality of global networks requires that “senior leadership understand that today’s and tomorrow’s mission will demand a powerful, permanent presence on a global telecommunications network that will host the ‘protected’ communications of Americans as well as targeted communications of adversaries.”44 In December 2005, The New York Times broke the warrantless wiretapping story, and in May 2006, USA Today broke the telephone metadata mass collection story—but with much less impact than the Times’s account, at least in part because of the lack of primary source documentation. Edward Snowden would take care of that problem.


Out of the “tens of thousands of documents” (Glenn Greenwald’s phrase) handed over by former NSA contractor Edward Snowden in May and June 2013 to The Guardian columnist Greenwald, the filmmaker Laura Poitras, and The Washington Post reporter Barton Gellman, a handful stand out for their shock value. The very first Snowden stories focused on an April 2013 order from the secret wiretap court, the FISA court, mandating that Verizon turn over to the government “all call detail records” not only for calls between the United States and abroad but also “wholly within the United States, including local telephone calls.” The order cited Section 215 of the Patriot Act as the authority for this mass collection—news to that Act’s authors, who never envisioned such a vacuum cleaner approach. Not only did the document show the government contravening Congress’s intent, but the text showed that the Director of National Intelligence (DNI), James Clapper, had lied to Congress only a month earlier, in March 2013, when he answered a direct question about whether the NSA collected information on millions of Americans by saying, “No. Not wittingly.” After the Snowden revelations, Clapper told NBC News reporter Andrea Mitchell that he had been trying to provide the “least untruthful” answer—an immortal phrase that will be summoned by teenagers facing irate parents for generations to come.45

The Verizon metadata order set the stage for a cascade of contradictions, as the documents and more reporting repeatedly undercut the U.S. government’s claims about its surveillance programs. First up was President Obama himself, on the PBS Charlie Rose show (June 16, 2013), telling a couple of whoppers, that the wiretap court process was “transparent” (until the Snowden leak, not a single wiretap court order had ever seen the light of day), and that “if you are a U.S. person, the NSA cannot listen to your telephone calls and the NSA cannot target your e-mails.” To the contrary, the PRISM documents leaked by Snowden showed the NSA could keep and search for five years any U.S. person’s e-mails and calls “swept up as part of the agency’s court-approved monitoring of a target overseas.” The PRISM documents also showed direct access by the government to the servers of tech companies ranging from Google to Facebook, a revelation that damaged the companies’ credibility and their business model for overseas growth. Subsequently, Google, Yahoo, and Microsoft, among others, announced that they would encrypt their data lines between servers to prevent such direct surveillance.

NSA took another credibility hit with the Snowden leak of the BOUNDLESS INFORMANT slides describing the agency’s capacity to count precisely how many billions of records were being intercepted. Previously, NSA Director General Keith Alexander and DNI Clapper had claimed to Congress that they could not produce any precise numbers on targets or intercepts or messages analyzed. Subsequently, the NSA removed a series of inaccurate “fact sheets” on its surveillance programs from its official Web site.46

The entire administration suffered a hypocrisy hit when the journalists reporting on Snowden’s documents published the classified Presidential Directive signed in November 2012 authorizing offensive cyber operations overseas. During this same period, the U.S. government had repeatedly attempted to focus media attention on hacking activity by the Chinese military and security services. Now, the White House press spokesman had to address the U.S. government’s own behavior and the complicated policy issues that arise from deliberate state-sponsored attacks on computer networks abroad.

Other embarrassing Snowden documents that kept the White House press briefings lively showed the U.S. government surveilling the telephones of allies (like German Chancellor Angela Merkel, who demanded a public cease-and-desist order from President Obama) and friendlies (like Brazilian President Dilma Rousseff, who canceled her planned summit meeting in Washington). The documents detailed the XKEYSCORE program’s capacity from 2007 to today as the “widest-reaching” of NSA’s tools, able to capture “nearly everything a typical user does on the internet,” including the text of e-mails and all searches, even in real time. In April 2013, according to the documents, XKEYSCORE gave the White House the UN Secretary General’s confidential talking points before his meeting with Obama.

The fallout from the Snowden leaks revealed that top officials had lied not only to Congress but also to the wiretap court, to the Supreme Court, and to each other. For example, one wiretap court opinion held that NSA Director Alexander “strains credulity” with his 2009 classified declaration that repeated NSA violations of the court’s limits on searches of stored intercepts arose simply from differences over “terminology.” The lead author of the Patriot Act, Republican Congressman James Sensenbrenner of Wisconsin, contrasted what he learned from the Snowden leaks with the misrepresentations he said government officials had repeatedly made to him about the way they were using his own statute. The solicitor general of the United States, Donald Verrilli Jr., discovered that his own Justice Department lawyers had misled him prior to a Supreme Court argument. Arguing for the Court to throw out a challenge by Amnesty International and the ACLU (among others) to a 2008 wiretap law, Verrilli assured the Justices that defendants would be notified if warrantless wiretap information was used against them, and therefore could in the future challenge the constitutionality of those searches. Only after Snowden did Verrilli find out that prosecutors did not so notify defendants, and that the Supreme Court had dismissed the prior case on false grounds.47

Perhaps most egregiously, NSA Director Alexander’s claims—repeated by many other officials and by members of Congress—that mass collection of call data had prevented fifty-four terrorist plots, soon evaporated under close inspection. Subsequent official investigations found that only thirteen of the fifty-four cases were even connected to the United States, that the bulk telephone metadata program had actually broken no such plots, and only identified a single terrorism supporter, whom the FBI was already tracking. The PRISM targeted collection deserved credit for contributing information to those fifty-four terrorism investigations, but the bulk collection program turned out to be little more than what the retiring deputy director of the NSA called an insurance policy.48

To President Obama’s credit, perhaps realizing that his top intelligence officials were suffering a credibility gap, he ordered his own blue-ribbon review group to look at the surveillance programs, and also encouraged the existing Privacy and Civil Liberties Oversight Board (PCLOB) to investigate. Among other extraordinary findings (such as that the Section 215 metadata vacuum cleaner was illegal, ineffective, and unconstitutional), the PCLOB eviscerated the government’s claims over a decade that congressional action—in the Patriot Act, in the FISA Amendments of 2008, and in the repeated renewals of surveillance authorities—meant that Congress had embraced the executive’s interpretations. Rather, secrecy had completely undermined the constitutional checks and balances:

[W]hen the only means through which legislators can try to understand a prior interpretation of the law is to read a short description of an operational program, prepared by executive branch officials, made available only at certain times and locations, which cannot be discussed with others except in classified briefings conducted by those same executive branch officials, legislators are denied a meaningful opportunity to gauge the legitimacy and implications of the legal interpretation in question. Under such circumstances, it is not a legitimate method of statutory construction to presume that these legislators, when reenacting the statute, intended to adopt a prior interpretation that they had no fair means of evaluating.49

According to the Oversight Board, George Orwell would find familiar the government’s abuse of language and vocabulary:

[S]anctioning the NSA’s [bulk telephony metadata] program under Section 215 requires an impermissible transformation of the statute: Where its text fails to authorize a feature of the program (such as the daily production of new telephone records), such authority must be inferred from silence. Where its text uses limiting words (such as “relevant”), those words must be redefined beyond their traditional meaning. And where its text simply cannot be reconciled with the program (such as its direction that the FBI, not the NSA, receive any items produced), those words must be ignored.

Attorneys for the ACLU took the critique even further, highlighting the fact—revealed by the Snowden leaks—that the NSA was using a completely different dictionary than the rest of us. The words “surveillance,” “collect,” “targeted,” “incidental,” “inadvertent,” “minimize,” and even the word “no” as uttered by James Clapper—all meant something other than the dictionary definition when the NSA employed them.50 Faced with the cascade of Snowden leaks and credibility contradictions, the Office of the Director of National Intelligence (ODNI) started racing to catch up, declassifying all kinds of documents that the NSA and the intelligence community had fought for years to keep secret. By the time of Obama’s speech in January 2014, the President could actually brag that the government had declassified more than forty opinions and orders of the FISA court, as against three that Snowden leaked. James Clapper told reporters that he considered the documents still properly classified but that the damage to national security was outweighed by the public interest, so he was declassifying thousands of pages and putting them all online.51 To David Ignatius of The Washington Post, Clapper argued that the real cost of the leaks was that he had to “throttle back” on collection in order to restore relationships with foreign intelligence partners and corporate partners. However, new documents about Yahoo’s resistance to the government’s orders suggest many of those “partners” no longer see themselves in cahoots.52

Unfortunately, Clapper’s declassifications, as laudable as they were, often simply dug a bigger hole for the defenders of the surveillance programs. For example, the wiretap court had blasted the government in a secret October 2011 opinion for misleading the court. Senator Ron Wyden had hinted at the existence of such an opinion in published letters in the summer of 2012 as Congress was looking at reauthorizing the 2008 FISA Amendments (sunset in 2012). So the nongovernmental Electronic Frontier Foundation went to court with an FOIA request based on the Wyden letters, but the government said the whole opinion, eighty-five pages long, was top secret, so not a word could be released. After Snowden, in August 2013, the government released a redacted version of the opinion, and the public could see that one whole section—previously top secret—consisted of the actual text of the Fourth Amendment!

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

The other result from the ODNI declassifications, especially of the wiretap court submissions and opinions, was the exposure to public view of the government’s almost complete reliance for its legal theory of surveillance and privacy on an obsolete and arguably inapplicable Supreme Court precedent from 1979, before cell phones were even invented. The recent Supreme Court decision in Riley v. California in June 2014 on cell phone privacy suggests the 1979 precedent (Smith v. Maryland) is in jeopardy, and that mere third-party possession of a person’s call data (by the phone company, for example) may not in fact eliminate privacy rights. Such a new standard, which is strongly supported by the PCLOB, might well change the entire legal landscape for the government’s surveillance activities.


The intelligence community in general, and the National Security Agency in particular, remain on the defensive after Snowden, their misrepresentations exposed, their dictionary discredited, and their budgets under pressure. This chastened outlook comes through in the public statements by top officials, now that the Snowden frenzy of summer 2013 has settled down. For example, an interview with the new NSA Director, Admiral Michael Rogers, published in The New York Times in June 2014, appeared under the headline “Sky Isn’t Falling After Snowden, N.S.A. Chief Says: Data Protection Cited: New Leader Calls Leaks Regrettable but Manageable.”53 If true, then we have either undergone a fundamental change in the definition of a secret, or we are witnessing a pragmatic adjustment to a new transparency forced on the NSA (likely the latter). What a contrast to the hyperventilation of Rogers’s predecessor, General Keith Alexander, cited by the Times as calling the Snowden leaks “the greatest damage to our combined nations’ intelligence systems that we have ever suffered.”

No small degree of secrecy remorse is also on view. The ODNI’s general counsel, Robert Litt, told a law school audience, “These leaks have forced the Intelligence Community to rethink our approach to transparency and secrecy.… One lesson that I have drawn from the recent events … is that we would likely have suffered less damage from the leaks had we been more forthcoming about some of our activities, and particularly about the policies and decisions behind those activities.” Litt went on to endorse closer scrutiny of “what truly needs to be classified,” “a mindset of proactively making available as much information as we can”—all because “Greater disclosure to the public is necessary to restore the American people’s trust” that the surveillance was “not only lawful and important” but also “appropriate and proportional in light of the privacy interests at stake.”54

The skeptic should note that this extraordinary reassessment was not driven by congressional oversight, or by judicial restraints, or by internal reformers or external advocacy, but by the public relations hit to the intelligence community’s credibility from Snowden’s leaks. Thus, as secrecy watchdog Steven Aftergood has observed, “leaks emerge as a uniquely powerful tool for shaping intelligence classification policy, while conventional checks and balances appear all but irrelevant by comparison.”55 The hazard is simply that the window is open right now, but it could slam shut again, with a single terrorist attack. There are only a handful of permanent reforms in place ensuring any greater transparency (the PCLOB, mainly, plus some internal civil liberties monitors in the intelligence agencies), so this breakthrough may well be temporary, but the changing mind-sets are certainly more than welcome.

One of the most eloquent voices on the new transparency came from National Security Agency retiring Deputy Director John C. “Chris” Inglis in his interview with NPR’s Morning Edition in January 2014. Asked by host Steve Inskeep whether the NSA should have made the bulk collection program public much earlier, Inglis admitted, “In hindsight yes, in hindsight yes.” Inglis went on, “I think going forward, what I would change is that we need to continue to move in the direction of having greater transparency about the nature of NSA, what its authorities are, how those authorities are brought to bear.”

Even more telling, in the same NPR interview, after Inglis ominously warned that since the terrorists have been alerted to what we can do, they could now take precautions, Inskeep asked if that would really damage our security. Bin Laden had retreated to a house in Abbottabad. He was completely cut off from phone or Internet contact, only communicating via messengers. In effect, the threat of us listening to Bin Laden’s conversations sent him back to the Middle Ages and dramatically reduced his capability. Inglis responded, “Well, at the base of your question I think you’re right.”56

Even President Obama, who criticized Snowden’s actions for months, finally endorsed the public debate over surveillance, at long last, in his January 2014 speech on surveillance: “One thing I’m certain of: This debate will make us stronger.” However, the President acknowledged only in passing the credibility gap that Snowden had exposed—by saying “we must maintain the trust of the American people,” the President certainly implied such trust was lacking at the moment. Inglis put the issue more colorfully during his NPR interview, trying to defend the metadata collection program. When asked about all the programs such as the Merkel intercepts, Inglis came up with an interesting standard for judging the propriety of intelligence activities: “So with respect to the totality of what NSA does, I think that not all of those have withstood the test of the optics, you know, or perhaps you know, the above the fold right side of the newspaper test.”

Another test for the new transparency will be the administration’s approach to intelligence budget secrecy. For decades, open government advocates had argued that keeping the U.S. intelligence budget secret amounted to a “fetish” that distorted rational secrecy decisions down the line. In 1997 and 1998, under pressure from a FOIA lawsuit by the Federation of American Scientists, then–CIA Director George Tenet finally agreed, since he found himself unable to sign an affidavit that release of the budget number would damage national security. Yet in 1999, Tenet resumed old habits, refusing to declassify the budget, and after the 9/11 attacks, further lawsuits seemed likely to fail. But the 9/11 Commission agreed with the reformers as part of the Commission’s overall critique of excessive secrecy, and recommended not only routine declassification of the overall intelligence budget number but also the number for each of the intelligence agencies, so as to enable public policy debate about those agencies’ performances and their claims on taxpayers’ resources. Both President Obama and Congress ultimately acted on the former recommendation but not the latter, writing into law the routine declassification of the overall number.57

The Snowden leaks produced a potential game changer for the budget secrecy debate. Among the documents given to The Washington Post was the top secret fiscal year 2013 budget justification for the intelligence community given to Congress in February 2012. Now a reader can compare and contrast the individual agencies and how they did budget-wise since 9/11. Incongruously, the agency that failed the most on 9/11, by hiding from the FBI its information on two would-be hijackers living in San Diego, profited the most in resource aggrandizement. The Snowden document showed the CIA’s annual budget had soared over the period from 2004 to 2013, from $9 billion to nearly $16 billion. In the same period, the National Security Agency also saw significant increases, from $7 billion to $11 billion. Other data of obvious public policy importance in the budget document included the ominous fact that only about 10 percent of the intelligence community’s personnel qualified for foreign language proficiency payments.58 So far, there’s been no hue and cry about national security damage from the budget revelations; so there comes a test for the Obama administration in the next year: Will the new transparency post-Snowden carry over to fulfilling the recommendation of the 9/11 Commission and routinely declassifying the intelligence budget at the agency level?

Finally, several of the recent declassifications provide tangible evidence of the mind-set change. For example, on June 26, 2014, the Director of National Intelligence released an actual “transparency report,” giving his annual statistics for calendar year 2013 on numbers of court orders, targets, and National Security Letters. One has to read between the lines to understand the data, since a single “target” (say, the country of Afghanistan) could involve millions of people and all of their phone records and all of their e-mail records. But the top secret document was marked for declassification in the year 2039. Instead, the public received it in June 2014, in time for some policy debates and the legislative discussion in Congress about surveillance authorities. That’s the Snowden effect—twenty-five years of secrecy undone.