Daniel Ellsberg, Edward Snowden, and the Modern Whistle-Blower

From their backgrounds to their motivations, the two men have some striking differences.
Illustration by Javier Jaén

In the summer of 1967, Secretary of Defense Robert McNamara commissioned a group of thirty-six scholars to write a secret history of the Vietnam War. The project took a year and a half, ran to seven thousand pages, and filled forty-seven volumes. Only a handful of copies were made, and most were kept under lock and key in and around the Beltway. One set, however, ended up at the RAND Corporation, in Santa Monica, where it was read, from start to finish, by a young analyst there named Daniel Ellsberg.

Ellsberg was dismayed by what he learned. For a generation, the U.S. government had been lying to the American people about the Vietnam War. He put the first of the volumes in his briefcase, praying that the security guards at RAND would not stop him, and made his way to a small advertising agency in West Hollywood, where a friend told him there was a Xerox machine he could use.

“It was a big one, advanced for its time, but very slow by today’s standards,” Ellsberg writes in his 2002 autobiography, “Secrets: A Memoir of Vietnam and the Pentagon Papers”:

It could do only one page at a time, and it took several seconds to do each page. I tried pressing the book down on the glass to do two pages at a time, but the middle section was faint and uneven. Fortunately the books were bound with metal tapes through holes so they could be taken apart. . . . The machine didn’t collate, and the bar had to come back and travel just as slowly for each copy.

Night after night, Ellsberg repeated the process: copying until dawn, rushing home, hiding the pages, grabbing a few hours’ sleep, then stumbling bleary-eyed into the office. Eventually, he gave copies of what came to be known as the Pentagon Papers to the Times and then the Washington Post, and passed into history as the archetypal modern whistle-blower.

Ellsberg was handsome and charismatic. He had served in the Marine Corps as a company commander in Korea. He did his undergraduate and graduate studies at Harvard, where he wrote a Ph.D. dissertation on game theory and collaborated with Thomas Schelling, who went on to win a Nobel Prize. He took a senior post in McNamara’s Defense Department, represented the State Department in Vietnam, and had two stints as a senior intelligence analyst at the RAND Corporation. Ellsberg knew about the Pentagon Papers because he was a member of the select team that wrote them, working on the section dealing with the very early nineteen-sixties. Before he approached the Times, he went to the Senate, where he tried to get someone to release the documents formally and hold public hearings. He walked the halls and dropped in on people he knew. “I had Senator Mathias in mind, and Senator Mike Gravel,” who, he notes, had “written me a letter congratulating me on my New York Review of Books article,” about the bombings in Laos. (It seems safe to say that the subject, verb, and object here—“Senator,” “written,” “New York Review of Books article”—may never again appear together in a sentence.)

In another passage in “Secrets,” Ellsberg relates a conversation he had in 1968 with Henry Kissinger, a Harvard acquaintance. Ellsberg was trying to prepare Kissinger for the White House. For the first time in your life, Ellsberg tells him, you will have a security clearance and gain access to a steady flow of government secrets. He goes on:

You will feel like a fool for having studied, written, talked about these subjects, criticized and analyzed decisions made by presidents for years without having known of the existence of all this information, which presidents and others had and you didn’t, and which must have influenced their decisions in ways you couldn’t even guess. That feeling of foolishness, Ellsberg continues, will last two weeks, at which point you will become so enthralled by your access to “whole libraries of hidden information” that

it will have become very hard for you to learn from anybody who doesn’t have these clearances. Because you’ll be thinking as you listen to them: “What would this man be telling me if he knew what I know? Would he be giving me the same advice, or would it totally change his predictions and recommendations?” And that mental exercise is so torturous that after a while you give it up and just stop listening. . . . The danger is, you’ll become something like a moron.

Ellsberg says that Kissinger listened carefully, but “I knew it was too soon for him to appreciate fully what I was saying. He didn’t have the clearances yet.”

What’s striking about the account isn’t just its wry brilliance. It is that the man who defined the modern category of whistle-blower was the sort who gave career advice to Henry Kissinger. Ellsberg was an insider—and that fact puts him in stark contrast with the man who has come to be seen as his heir, Edward Snowden.

Snowden left his job as a National Security Agency contractor in Hawaii three years ago, with thousands of the U.S. government’s most closely held secrets in his possession. In the course of his work, Snowden had learned things that dismayed him, and many of those secrets soon found their way onto the front pages of the world’s newspapers, earning him his reputation as the post-9/11 Ellsberg.

But Snowden did not study under a Nobel Prize winner, or give career advice to the likes of Henry Kissinger. He was a community-college dropout, a member of the murky hacking counterculture. He enlisted in the Army Reserves, and washed out after twenty weeks. He worked at the C.I.A. for a few years and left under a cloud. He learned about the innermost secrets of American intelligence-gathering and policy not because he was personally involved with that intelligence-gathering or policymaking but because he was a technician who helped service the computer systems that managed these things. The élites, Snowden once said, “know everything about us and we know nothing about them—because they are secret, they are privileged, and they are a separate class.” Had Snowden been a whistle-blower in 1967, at the launch of the Pentagon Papers, he would have blown the whistle on Daniel Ellsberg. The whistle-blower as insider has become the whistle-blower as outsider. That is a curious fact, and, as we come to terms with the consequences of Snowden’s actions, it may be an underappreciated one.

In “The Leaky Leviathan,” a study published three years ago in the Harvard Law Review, David Pozen attempts to understand a puzzle. Strict laws prohibit government officials from disclosing secrets, yet leaking has been a constant feature of American political life. Since the passage of the Espionage Act, in 1917, the federal government has prosecuted only about a dozen cases concerning media leaks of state secrets. That’s an astonishingly small number. Pozen, a Columbia law professor, cites one estimate that, between 1949 and 1969, 2.3 per cent of the front-page stories in the Times and the Washington Post were based on government leaks. Another study looked at just the first six months of 1986 and found that a hundred and forty-seven stories in the country’s eight major newspapers were based on leaks. The entire career of Bob Woodward, perhaps the best-selling political writer of his generation, is based on leaks. And yet, with a few symbolic exceptions, nothing is done.

“For a crime that Presidents describe as a major threat to national security and good government, the degree of ‘underenforcement’ is stunning,” Pozen writes. “Even if we were to limit the denominator to classified information leaks that the Intelligence Community (IC) is known to have otherwise documented publicly—which may be a small fraction of the universe of potentially prosecutable offenses—the historic indictment rate for leak-law violators would be below 0.3%. The actual rate is probably far closer to zero.” Even the recent uptick in leak prosecutions during the Obama Administration, Pozen argues, does not alter the fundamental pattern. In Washington, giving away secrets to the press is a crime largely without consequences.

Pozen easily dispenses with the idea that Administrations don’t prosecute leakers because they can’t find them. They can: information—particularly sensitive information—has a pedigree. When I worked on the science desk at the Washington Post, my colleagues and I would read a front-page story by our counterparts at the Times and invariably know where the leak on which the story was based came from. The first order of business was typically to call the leaker and complain that he or she was playing favorites.

Pozen argues that governments look the other way when it comes to leaks because it is in their interest to do so. He cites a story that ran in the Wall Street Journal and the Washington Post in 2012 about how the C.I.A., with the coöperation of Yemeni authorities, was using drone strikes against Yemen-based Al Qaeda militants. The drone program was classified: that story didn’t come from a press conference. Pozen says the story was clearly a “plant”—that is, a leak made with the full authorization of the White House. Letting the facts slip out served a purpose for the Obama Administration. A plant like that, Pozen writes, “keeps the American people minimally informed of its pursuits, characterizes them in a manner designed to build support, and signals its respect for international law.”

But if you want to reserve your right to plant an authorized leak, Pozen argues, you have to allow unauthorized leaks as well:

For a strategy of planting to work, it is critical that relevant audiences not immediately assume that every unattributed disclosure they encounter reflects a concerted White House effort to manipulate the information environment. The practice of planting requires some amount of constructive ambiguity as to its prevalence and operation.

“Hey! Guess what?”

In a world where every stealthy disclosure is a plant, the journalist is a stooge, the Administration’s motives are transparent, and the government of Yemen is exposed. But, when the origin of the disclosure is uncertain, all parties save face. “Plants need to be watered with leaks,” Pozen writes. He continues, “It is loosely analogous to what a game theorist would call mixed-strategy equilibrium: an approach that generates sufficient randomness (or apparent randomness) across government sources as to degrade the ability of outsiders to predict the nature and origin of any given disclosure.”

Leaks are also a form of governmental “self-binding,” Pozen argues. The executive branch, he says, has a persistent problem in the American democratic system. It needs to justify to the public the extraordinary power it wields. One of the ways it does this is to allow periodic disclosures that let the voters and the Congress see behind the curtain, reassuring them that most of the troubling things the executive branch engages in will eventually come to light. The White House allows leaks—even if those leaks hamper and embarrass it in the short run—because they help it maintain its power in the long run. The public needs regular and convincing reassurance that the strong President will be caught before he becomes a bad President. “A mechanism that never made the President look bad would quickly lose its capacity to signal credibility; the whole point is that the power-enhancing second- and third-order effects of these arrangements ultimately come to swamp the power-reducing first-order effects,” Pozen writes. “No pain, no gain.”

In short, the relationship between the government and the press—between the source of leaks and the beneficiary of leaks—is symbiotic. Governments may make a fuss about how much leaks are harming them. But they need leaks as much as the press does. The legitimacy of government requires sunshine, and the practice of governance sometimes requires darkness—and, in the face of that contradiction, leaks are a kind of informal workaround.

But, crucially, the workaround depends on a common understanding among the participants, a degree of discretion and judgment. You need to have laws against leaking—and occasionally enforce them—so that leakers think twice about what they’re leaking. You can’t over-enforce those laws, though, because then you’ve killed planting. The press, meanwhile, has to preserve the ambiguity of plants, in order to preserve its access to leaks. And the press, too, has to think twice about what it will do: if you discredit the institution of leaking, by airing disclosures too damaging to the national interest, you’ll make it much more difficult for the next leaker to come forward. Leaking is not a mechanical exchange of information. It is a ritual that obliges its participants to play by certain rules.

The Pentagon Papers documented a history of policy failure. Throughout the Vietnam conflict, people inside the national-security apparatus—senior, thoughtful people with an intimate knowledge of the arena—had repeatedly come to the same conclusion. The war in Southeast Asia was costly and pointless. Nonetheless, the White House continued to tell the American people that it was going swimmingly. For someone like Ellsberg, an expert in the decision sciences, the failure of the policy establishment to make its conclusions heard was galling.

In August of 1970, Ellsberg went to see Henry Kissinger, then Nixon’s national-security adviser, and asked him if he knew about the Pentagon’s secret Vietnam history. Kissinger replied that he had a copy in the White House. In his autobiography, Ellsberg recounts:

I was encouraged to hear it. I asked, “Have you read it?”

“No, should I?”

I said I thought strongly that he should. . . .

“But do we really have anything to learn from this study?”

My heart sank. I thought: My God! He’s in the same state of mind as the rest of them all along. They each thought that history started with his administration and that they had nothing to learn from earlier ones.

When Ellsberg went to the Senate, one of his stops was William Fulbright, who was squarely in the antiwar camp. But he, too, needed convincing:

He asked if I would give him an example of a revelation that would make a big splash.

I said that it was not any individual page or revelation, or even a small set of them, that was very important. It was the overall detailed documentation of our involvement over the years and the repetitive patterns of internal pessimism and of desperate escalation and deception of the public in the face of what was, realistically, a hopeless stalemate. It was the total lack of a good reason for what we were doing anywhere in the whole story. You had to read a lot of it, perhaps a thousand pages, covering a number of periods, to get the full effect.

Ellsberg was disillusioned. He had exhausted all formal remedies. So he decided to leak. But he didn’t leak all forty-seven volumes of the Pentagon Papers. The last four volumes of the study dealt with ongoing, and highly sensitive, diplomatic efforts to negotiate a peace and the release of P.O.W.s. Ellsberg held those pages back because, as he said later, “I didn’t want to get in the way of the diplomacy. . . . I wanted to get in the way of the bombing and killing.”

And whom does he entrust with those forty-three volumes? The Times, which rents a suite at the Hilton, posts security guards outside, and assigns a team to spend the next three months reading through the collected documents. To recap: a studious political-science Ph.D. wishes to instruct the upper echelon of American leadership in the benefits of reading several thousand pages of history, and, after his initial efforts prove unavailing, assigns a carefully curated set of course materials to the most august institution in American journalism. This isn’t the behavior of a dissident. It’s the behavior of a graduate student, who realizes, with great disappointment, that his thesis adviser has not kept up with the literature.

Ellsberg’s behavior brings to mind what the sociologist Beryl Bellman refers to as “the paradox of secrecy.” When a secret is disclosed, Bellman observes, the holders of secrets say “Don’t tell anyone” even as they are, of course, telling someone: “The respondent is instructed to disregard the telling as an exposing and to treat the occasion as an exception to the rule of not talking secrets.” As Bellman explains, a secret isn’t invalidated by its disclosure; it’s defined by its disclosure. What makes a secret a secret is simply the set of operating instructions that accompany its movement from one person to the next.

The same logic applies to leaks. Leaks differ from secrets only in the specifics of their operating instructions: “Don’t tell anyone” becomes “Don’t tell anyone I told you.” Leaks do not invalidate government secrecy; they depend upon it. When Ellsberg was serving in the Pentagon during the Johnson Administration, for example, he became distressed by the fact that, whenever the generals secretly asked Johnson for additional troops in Vietnam, the President would comply. Johnson, Ellsberg writes, feared “that if he didn’t meet their requests, they would leak their demands to hawks in Congress and cause him domestic trouble.” So Ellsberg resolved to strike back: “What I had in mind was very simple: a leak a day of a closely held secret, something that showed high-level access. The content was much less important than the leak itself.” The real message would be clear to the President: the next time he decided to grant the military’s request, it would become known—and receive the public scrutiny and resistance it deserved.

Ellsberg’s best answer to a governmental entity using leaks to push the White House in a given direction was to counter with a leak designed to push the White House in the opposite direction. When he says that “the content was much less important than the leak itself,” he is making Bellman’s point. His power came from his status as the holder of secrets. Would the Times have won a Pulitzer for publishing the Pentagon Papers if the study had been unclassified? Not a chance. This was a seven-thousand-page government report that, even Ellsberg admitted, didn’t start to make sense until you got a thousand pages in. (“There is an endless amount of garbage in there, but nothing really important,” Ellsberg’s RAND colleague Konrad Kellen once said. “It was impossible to read them because they were so long.”) If governments need leaks to validate their secrecy, leakers need secrecy to validate their leaks.

Even the farcical dénouement of the Ellsberg case upholds the Pozenian principle that the good leak is the leak that perpetuates the practice of leaking. The first thought of some Washington insiders was that the leak must have come from the Nixon White House. After all, the study stopped at the end of the Johnson Administration. It discredited the previous generation of Democratic Presidents. That sounds like a plant, doesn’t it?

Kissinger, who knew a thing or two about game theory, called Nixon delightedly the day the story broke. “In public opinion, it actually, if anything, will help us a little bit, because this is a gold mine of showing how the previous administration got us in there. . . . What this massively proves is that, if it’s anybody’s war, it’s Kennedy’s and Johnson’s.” Nixon, being Nixon, immediately thought to take the exercise one step further. Since they were under the protective umbrella of mixed-strategy equilibrium, why not join in? “Stuff on Kennedy I’m gonna leak,” Nixon responded. “We’ll just leak it out. . . . Now that it’s being leaked, we’ll leak out the parts we want.” The leak that looked like a plant opened the door for a plant that would look like a leak. The Nixon Administration then tried to prosecute Ellsberg, only to have the case thrown out because Nixon’s operatives broke into the office of Ellsberg’s psychiatrist looking for Ellsberg’s patient records. And why did they want the records? To leak them, of course.

Imagine a young man called Daniel Snowberg. He has a doctorate in international relations, and once spent a summer interning at the Electronic Frontier Foundation, the think tank specializing in digital freedom. He gets a job as an analyst at the National Security Agency, and while there he runs across a copy of the infamous Foreign Intelligence Surveillance Court (FISC) authorization, under Section 215 of the Patriot Act. This is the order that led Verizon to hand over its telephone records to the N.S.A.

“Of course I’m pissed! I just found out we’re getting these things for our Christmas bonus!”

The order troubles him. The Patriot Act allows the N.S.A. to obtain phone records and the like if it provides “a statement of facts showing that . . . the tangible things sought are relevant to an authorized investigation.” But this isn’t a search specific to an investigation. It appears to be a fishing expedition. He surreptitiously copies the authorization: it’s not that long. He sends it to his old colleagues at the Frontier Foundation. They share his alarm: the legal opinion in the FISC order looks unconstitutional to them. They set up a meeting with Ron Wyden, the ranking Democrat on the Senate Intelligence Committee. Wyden, too, is troubled. He encourages them to leak it to the Washington Post. And they do. That is the leaker as insider.

Edward Snowden took a different path. He used a Web crawler (a search engine preprogrammed with key words) to roam through the N.S.A. files, “touching” as many as 1.7 million of them. Among those files was the FISC order. But Snowden also accessed, and ultimately passed on to journalists, thousands of files concerning activities that had nothing to do with domestic surveillance.

Daniel Snowberg, the insider, would have sparked a national debate that focussed on the question of what access the N.S.A. should have to the private data of ordinary American citizens. And when, in May, 2015, a federal court ruled that the N.S.A.’s telephone-records collection violated the intent of the Patriot Act, Snowberg would have stood as someone who restored the legitimacy of the national-intelligence apparatus: who, in the spirit of Pozen’s notion of self-binding, embarrassed the executive branch in the short term in order to preserve the prerogatives of the executive branch in the long term.

Snowden does not belong in the same category as Julian Assange’s WikiLeaks, which indiscriminately makes stolen material available to all comers. Snowden went through intermediaries; he expected his journalistic outlets to curate the material he gave them. Nonetheless, Snowden didn’t leak, in the traditional sense. He flooded, and in that difference of degree is a difference in kind. Edward Jay Epstein’s upcoming book, “How America Lost Its Secrets: Edward Snowden, the Man and the Theft,” presents the hard-line intelligence community’s case against Snowden, and what is remarkable about it is how little time Epstein spends on the propriety of the N.S.A.’s domestic surveillance. Instead, he dwells on everything else that Snowden took from the N.S.A. files: details of N.S.A. methodologies, overseas operations, foreign sources—revelations that, national-security officials maintain, gravely compromised our foreign intelligence-gathering.

In light of the sensitivity of this other material, Epstein makes the claim that Snowden, now living under asylum in Russia, may have been the dupe of a foreign-intelligence service. Epstein could never have made an argument like that about Snowberg. No matter how well intentioned Snowden’s actions—and there are many who see him as a hero—by violating the norms of insider disclosure, he suffered a self-inflicted wound. A much needed national conversation about the N.S.A.’s encroachment on civil liberties became sidetracked by debates about his own motivations.

So why didn’t Snowden release just the FISC order? Because he’s not fundamentally a leaker. He’s a hacker. Leakers are concerned with Bellman’s operating instructions and Pozen’s mixed-strategy equilibrium, because they are interested in using and exploiting secrecy: they believe that secrecy, by its preservation and strategic violation, serves an essential purpose. The hacker, on the other hand, is a skeptic of secrecy. The anthropologist Gabriella Coleman has described hacking as the “aesthetic disposition” of craftiness and guile. “When lecturing to my class one security researcher described the mentality: ‘You have to, like, have an innate understanding that [a security measure is] arbitrary, it’s an arbitrary mechanism that does something that’s unnatural and therefore can be circumvented in all likelihood,’ ” she writes.

That’s Snowden. According to Eric Schmitt, of the Times, while Snowden was at the C.I.A. his supervisor issued a “derog,” suspecting him of breaking into classified files. According to a former N.S.A. director, Mike McConnell, Snowden once applied to the N.S.A., and broke into its files so that he could get the answers for the application test. He “touched” those 1.7 million files to show that he could: that the N.S.A.’s pretensions to secrecy were an arbitrary mechanism doing something unnatural and therefore could be circumvented. Snowden says that he left a digital trail, designed to show the N.S.A. what he had and hadn’t taken, but that the agency’s auditors missed the signs. Because the N.S.A. cannot determine precisely what Snowden compromised, it says that it now must spend many years and many billions of dollars rebuilding its operations. The leaker would be remorseful about that fact. The hacker just rolls his eyes. “I figured they would have a hard time,” Snowden told Wired. “I didn’t figure they would be completely incapable.”

The hacker’s world view is centered on technological systems, and radicalized hackers like Snowden usefully remind us how vulnerable we are to the digitized state. Snowden spoke eloquently of this threat during a live-stream interview at Fusion’s recent Real Future Fair, in Oakland: “If you want to build a better future, you are going to have to do it yourself. Politics will only take us so far. And if history is any guide they are the least reliable means of achieving the effective change that we really want to see.” That means getting our technological systems in order, so that we can enforce “a digital right to privacy.” He went on:

Technology works differently than the law. Technology knows no jurisdiction if properly implemented. When you change the protocols, the technologies and the services that are comprising this fabric that controls how the cell phone in your pocket operates, how your communications get from your computer to the servers on the Internet that provide you the services that you are looking for, that’s all happening through science, through math, technology, engineering. And if we change that, if we get that right . . . There is wisdom in these words. But Snowden went further. Politics wasn’t just challenged by technology. It had been rendered obsolete by technology:

When we talk about saving lives, when we are talking about fighting cancer, treating AIDS, ameliorating poverty, these solutions typically are not coming from government. . . . While law is important . . . at the end of the day law is simply letters on a page.

This is a perplexing statement. The AIDS epidemic was not confronted through a wiki. The war against H.I.V. and diseases like cancer is bought and paid for by taxpayers, through the services of the National Institutes of Health. Acts of Congress, in the form of Medicaid, Medicare, and Social Security, did an enormous amount to alleviate systemic poverty. All these measures are guided by policy, authorized by law, and enacted by institutions of government: they are examples of things the government does really well. Perhaps it is to be expected that Snowden has little patience with or understanding of this process: for the hacker, the digital world is sometimes more real than the physical one. The irony of Snowden’s position, however, ought not to escape notice. He challenged the N.S.A. because it reduced a complicated social question (what is the role of surveillance in a democratic society?) to a technical one: how much data can we acquire and analyze? Yet Snowden himself reduced the complicated social question of when and how the disclosure of sensitive information strengthens democratic government to a technical one: how many classified files can I touch with my Web crawler? Snowden did not repudiate the N.S.A.’s error. He replicated it.

“Things That Can and Cannot Be Said,” a new book by the writer Arundhati Roy and the actor John Cusack, is a short account of a conversation that took place not long ago in Russia between Daniel Ellsberg and Edward Snowden. The meeting was Cusack’s idea. He knew both men from his work with the Freedom of the Press Foundation. Calls were made, tickets were bought, and the four met in Room 1001 of the Moscow Ritz-Carlton.

The result is a strange and ultimately disappointing volume. Roy explains, mysteriously, “The jokes, the humor, and repartee that took place in Room 1001 cannot be reproduced. The Un-Summit cannot be written about in the detail that it deserves.” Snowden is present, yet absent. Roy asks him “whether the N.S.A. was just feigning annoyance at his revelations but might actually be secretly pleased at being known as the All Seeing, All Knowing Agency—because that would help keep people fearful, off balance.” We aren’t told Snowden’s response to this or to any number of other questions. By the end of the book, Snowden has all but disappeared. In his place we get Daniel Ellsberg, still a compelling figure at the age of eighty-five. Ellsberg strides through Russian immigration—the same country, in his Pentagon years, he worked to destroy—flashing a peace sign. He buys himself a Cossack hat. He tells war stories from the Pentagon under Robert McNamara. He ends the day “exhausted and blissful,” quoting from “The Man Without a Country,” by Edward Everett Hale, and begins to weep.

“Oddly, when I think back on the meeting in the Moscow Ritz, the memory that flashes up first in my mind is an image of Daniel Ellsberg,” Roy writes. “Dan, after all those hours of talking, lying back on John’s bed, Christ-like, with his arms flung open, weeping for what the United States has turned into.” Oddly? Surely one obvious lesson of the Snowden affair is that Whistle-blower 2.0 does not really belong in the same room as Whistle-blower 1.0.

Ellsberg says to Snowden, “You’ve seen ‘Dr. Strangelove’?” The leaker tries, as best he can, to find common cultural ground with the hacker.

Snowden: “No.”

Cusack: “Oh my God!”

Snowden: “Before my time, man.” ♦