What We Owe Whistleblowers
Jim Frazier

What We Owe Whistleblowers

Jameel Jaffer argues that their disclosures since 9/11 have been vital, and that we should protect them better than we do

Occasional Papers

An essay series tackling pressing issues at the intersection of speech, privacy, and technology

A few weeks ago, not long before the United States carried out a drone strike in Kabul that killed a humanitarian worker and nine members of his family, a federal court in Virginia sentenced Daniel Hale, a former intelligence analyst, to 45 months in prison for sharing classified information about the drone program with a reporter. Hale’s indictment and conviction under the Espionage Act received little attention, probably because the prosecution of national security whistleblowers has now become routine. During the 20th century, only one person was convicted under the Espionage Act for having shared information with the press, but, since 9/11, both Republican and Democratic administrations have used the Espionage Act aggressively against journalists’ sources.

This isn’t generally regarded as a scandal, but it should be. It’s morally indefensible that we imprison journalists’ sources without even considering the value to the public of the information they’ve disclosed, and without considering the justifications they offer for their actions. It’s also deeply dangerous to our democracy. Twenty years after 9/11, the American public is more reliant on national security whistleblowers than ever before. It’s worth reflecting on the debt we owe them, and considering the ways in which Congress and courts could better protect them.


Let’s begin by acknowledging what can’t reasonably be disputed: In the years since 9/11, the United States has paid a staggering price for excessive secrecy. Time and again, national security policies crafted behind closed doors and shielded from public scrutiny have proved to be deeply flawed, with far-reaching consequences for life, liberty, and security. Mistakes, excesses, and abuses that might have been avoided or quickly corrected had they been subjected to prompt outside review have led to ill-considered war, the gross infringement of human rights, the alienation of our allies , the fortification of our enemies, the misdirection of colossal amounts of money, the erosion of the United States’ influence around the world, and the sapping of Americans’ confidence in the competence of their government.

And yet the price we’ve paid for excessive secrecy would have been greater still had it not been for government insiders who disclosed official secrets in order to inform the public about official decisions they believed to be wrong. Americans learned about the abuses at Abu Ghraib only because a soldier, in defiance of orders from superiors, supplied photographs to The New Yorker and “60 Minutes.” When we learned about the CIA’s network of overseas black sites, it was because intelligence officials uneasy about the legality and morality of the CIA’s activities revealed their apprehensions to The Washington Post. It was Edward Snowden, a government contractor, who exposed the National Security Agency’s dragnet collection of Americans’ telephone records and other programs of mass surveillance. Documents and videos furnished to Wikileaks by Chelsea Manning undermined official narratives about the wars in Afghanistan and Iraq. And when American media organizations reported that official statements about American drone strikes had dramatically understated the number of bystander casualties, and grossly mischaracterized the military’s rules of engagement, they relied in part on information supplied by Hale; on accounts from drone operators haunted by abuses they had witnessed; and on disclosures by other military personnel and public servants, most of them still anonymous, who shared official secrets without authorization.

One needn’t believe that these people acted with pure motives, that all of their disclosures were justified, or that none of their disclosures were harmful, to recognize the pivotal role that they played. Informed public debate about national security policy would have been impossible had these insiders not revealed what the government would have preferred to suppress. And it’s not simply that the American public would have remained unaware of some of the government’s most consequential mistakes and abuses, or that we would have learned of them much later than we did. It’s also that these mistakes and abuses would have gone uncorrected for much longer than they did. In the wake of insiders’ unauthorized disclosures, the government made significant adjustments to policies relating to interrogation, detention, surveillance, and extrajudicial killing. In some instances, unauthorized disclosures provoked executive branch officials to course-correct almost immediately. But even when they didn’t spur immediate adjustments to government policy, these disclosures fueled investigative journalism and civil litigation, and sparked investigations by congressional committees, inspectors general, and the military—investigations and processes that in many instances led eventually to significant reform.

Government officials habitually call for the imposition of harsh penalties on insiders who share official secrets without authorization. The rhetoric has become markedly more heated even as the prosecution of leakers has become more frequent, and even as the sentences imposed on leakers have become more severe. In the absence of a course correction of some kind, the government’s use of the Espionage Act against journalists is likely to become even more common. Whatever norms once kept the government from using the Espionage Act against journalists’ sources have now been abandoned. And new surveillance technologies, and new legal authorities, allow the government to identify leakers without incurring the political costs that have historically been associated with serving subpoenas on journalists.

None of this would be troubling if unauthorized disclosures were always socially harmful. But a few features of our system make the unauthorized disclosure of official secrets not simply inevitable but vital to the proper functioning of our democracy.

First, Executive Order 13526, which establishes the classification system, allows the government to classify information without regard to whether and to what extent disclosure would aid public deliberation. Under that executive order, an “original classifying authority” may classify government information if the information falls within one of a number of specified categories—e.g., “military plans,” “intelligence activities,” or “foreign activities of the United States”—and if disclosure “reasonably could be expected to cause damage to the national security that the original classification authority is able to identify or describe.” If an original classifying authority determines that these criteria are satisfied, information can be classified even if it is plain that the benefits of disclosure would outweigh the harms. In practice, it is common for the government to classify the kinds of documents that could be expected to be especially important to the public’s ability to understand, evaluate, and influence national security policy—including documents that describe government policy, documents that explain the government’s understanding of its legal authorities or obligations, and documents that describe conduct that is unlawful. The executive order gives decisive weight to the security interest, and no weight at all to the interest in informed public deliberation.

Second, despite the limiting language in the executive order, the government routinely classifies information whose disclosure could not reasonably be expected to cause damage to the national security, as many government studies have found. J. William Leonard, then director of the Information Security Oversight Office, testified in 2004 that “half of all classified information is overclassified.” Former New Jersey Governor and 9/11 Commission Chairman Thomas Kean said that “three quarters” of the classified material he had reviewed in connection with the 9/11 Commission “should not have been classified in the first place.” The executive overclassifies for many different reasons—among them, that officials are rarely sanctioned for overclassifying information; that classifying information can afford the classifier bureaucratic advantage; and that classifying information can shield controversial decisions from scrutiny both inside and outside the government. For these reasons and others, “we overclassify very badly,” as former CIA Director Porter Goss once observed, and the mere fact of classification is not a reliable indicator that disclosure could reasonably be expected to cause harm.

Third, the courts have rejected the proposition that the First Amendment guarantees the public or even the press any general right of access to information in the hands of government. Acknowledging the “structural” role that the First Amendment plays in “securing and fostering our republican system of government,” the Supreme Court has held that the First Amendment protects the public’s right of access to criminal proceedings and certain documents filed in connection with them, and lower courts have held that this right extends to civil proceedings. The courts have declined, however, to recognize a more general right of access to government information. Moreover, even in the narrow circumstances in which a right of access has been recognized to attach, some courts have held that the mere fact of classification is sufficient to overcome the constitutional right.

Fourth, Congress has not afforded the public or the press a meaningful statutory right of access to information relating to national security. While the FOIA allows ordinary citizens to request records from the government on any topic, it does not permit courts to order the disclosure of information that is properly classified, and the courts almost always defer to the government’s assertion that the information is properly classified. (Indeed, courts have overturned the executive’s classification decisions only a handful of times.) Because the FOIA does not include a public interest “override,” courts affirm the government’s withholding of classified information under the act without reference to whether the benefits of disclosure are likely to outweigh the harms. Indeed, courts affirm the government’s withholding of classified records even when they believe that the records describe or authorize government conduct that is unlawful.

Fifth, the courts have operationalized the FOIA in a way that accommodates selective disclosure. Congress enacted the FOIA in large part out of concern that the government had distorted debate about the war in Southeast Asia by cherry-picking the information it shared with the public. The way the courts have implemented the FOIA, however, facilitates the very practice that the statute was meant to foreclose. As interpreted by the courts, the FOIA allows the government to disclose classified information unofficially—through unattributed leaks to the media—without waiving its right to withhold that same information in response to FOIA requests. As it has been interpreted by the courts, the statute also permits the government to disclose classified information selectively without waiving its right to withhold closely related information.

Sixth, statutory protections for insiders who use internal channels to report official malfeasance are weak in the national security sphere. The whistleblower protections available to other government employees are not available to employees of the intelligence agencies. Recent legislation protects intelligence community employees from reprisal for certain disclosures to Congress provided they file a complaint first with the intelligence community’s inspector general. This channel, however, is more likely to be useful to employees who are concerned about isolated abuses than to employees concerned about failures that are systemic—ones that relate, for example, to the lawfulness of executive branch policy, or the failure of institutional oversight mechanisms. And yet these are the complaints that are most vital to ensuring that national security policy does not escape popular control.

No doubt, it would be possible to design a system in which unauthorized disclosures of classified information would be less necessary. The executive branch could narrow the categories of classifiable information, incorporate consideration of the public interest into the classification process, or subject classifiers’ decisions to additional internal review and audit within the executive branch. Congress could expand protection for intelligence community whistleblowers who submit complaints through internal channels, narrow FOIA’s national security exemptions, require more robust judicial review in certain classes of FOIA cases, or modify FOIA doctrine in ways that might disincentivize selective disclosure. The courts could construe the First Amendment to encompass a general constitutional right to information in the hands of the government, and clarify that the mere fact of classification cannot be sufficient to overcome this constitutional right.

In the system we have now, however, unauthorized “disclosure of some national security secrets is not only inevitable but also essential for the proper functioning of our government,” as Mary-Rose Papandrea has written. Without unauthorized disclosures, the public’s ability to understand and evaluate national security policy would be almost entirely dependent on the goodwill of executive officials. Public debate about national security policy would take place in an information environment controlled almost entirely by the officials most responsible for developing, authorizing, and overseeing that policy. The most significant mistakes and abuses—those that result in unlawful policies (rather than isolated abuses), or that reflect failures of the oversight system—would be insulated from public scrutiny. And public policy would become unmoored from the democratic consent that gives it its legitimacy.


Unauthorized disclosures of classified information have always been crucial to our democracy, but developments over the past two decades have made whistleblowing more necessary as well as more perilous. Post-9/11 developments have deepened our dependence on whistleblowers for multiple reasons. The military and the intelligence agencies have grown dramatically in their budgets, personnel, prestige, and influence. They are relying heavily on potent new technologies whose implications for individual freedom and human rights are poorly understood even by their developers. And the increased reliance by the military on remotely piloted aircraft and covert operations has rendered war, and many of its implications, less visible to most Americans.

The record of the past two decades reveals the extent of our new dependence on whistleblowers. Consider:


The abuses at Abu Ghraib came to public attention because a soldier, defying orders from superiors, supplied photographs to The New Yorker and “60 Minutes.” The public learned of the CIA’s network of overseas prisons because officials who were concerned about the lawfulness and morality of the CIA’s policies supplied classified information to The Washington Post. When the public learned, several years later, that the CIA had destroyed videotapes of prisoners being waterboarded, this, too, was the result of unauthorized disclosures.

These disclosures were inflection points in the public debate about the Bush administration’s interrogation policies and about the “war on terror” more generally, and they led, sometimes circuitously but often directly, to institutional reflection and legal reform. At least in part in response to information made public by whistleblowers, the military conducted multiple investigations relating to the abuse of prisoners; the Senate Armed Services Committee undertook a major investigation and held public hearings; and the Justice Department launched criminal investigations. Some of these investigations and processes were whitewashes or failures but others resulted in significant changes to the law and to the practices of the military and intelligence agencies. Congress clarified and strengthened the prohibition against cruel, inhuman, and degrading treatment. The Defense Department narrowed military interrogators’ authority to use techniques that had led to abuse. President Bush transferred prisoners out of the CIA’s black sites. The Office of Legal Counsel withdrew a memo that purported to supply a legal basis for so-called “enhanced interrogation techniques,” including waterboarding, a technique that the United States had previously prosecuted as a war crime. When he took office in 2009, President Obama shut down the black sites, rescinded related OLC memos, and declared that the United States would disavow torture. Later, the Senate Intelligence Committee undertook a yearslong investigation and wrote a 6,000-page report meant to ensure that the “system of detention and interrogation described in [the] report is never repeated.” These reforms and processes came about for many reasons, but it is inconceivable that they would have come about if government insiders had not disclosed classified information without authorization.


The public learned of the NSA’s mass surveillance activities because Snowden, an NSA contractor, leaked classified documents to The Guardian and The Washington Post. In the months before Snowden did this, government officials who spoke publicly about the government’s surveillance practices described them in terms that obscured the scope and implications of those practices; the director of national intelligence testified inaccurately to Congress that the NSA was not collecting information about large numbers of Americans; the Supreme Court dismissed a constitutional challenge to the NSA’s mass surveillance activities on the grounds that the plaintiffs lacked proof that the government was actually engaged in those activities; and the Justice Department systematically failed to notify criminal defendants of its use of evidence derived from the NSA’s mass-surveillance activities, though the law required it to provide such notice. Public debate about the NSA’s activities was not simply impoverished but profoundly distorted.

Snowden’s disclosures alerted the public to the scope of the NSA’s surveillance activities, the implications of those activities for the rights of people living in the United States, and significant failures on the part of the congressional intelligence committees and the specialized courts tasked with overseeing some of the NSA’s surveillance practices. Like the unauthorized disclosures about interrogation, Snowden’s disclosures about surveillance led to institutional reflection and legal reform. The Foreign Intelligence Surveillance Court gave serious consideration for the first time to whether the NSA’s call-records program complied with the Fourth Amendment. A federal appeals court declared the program to be unlawful. The Justice Department began to comply with its obligation to notify criminal defendants of the provenance of evidence derived from the NSA’s mass surveillance activities, which had the effect of permitting criminal defendants to challenge the constitutionality of those activities in court. Congress passed legislation that dramatically curtailed the call-records program, empowered the FISA court to hear from court-appointed amici in certain contexts, and obliged the director of national intelligence to declassify and release FISA court opinions of broad significance. Obama instructed the director of national intelligence to establish a “review group” to assess the government’s use of communications and surveillance technologies, and the review group made a series of recommendations that the executive adopted, at least in part. None of this could have taken place if Snowden had not disclosed classified information without authorization.

“Targeted” killing

Like the public debate about the NSA’s surveillance practices, public debate about the government’s use of armed drones overseas was characterized by selective disclosure and official misdirection. Even as it normalized the practice of extrajudicial killing, the Obama administration refused to disclose statistics on civilian casualties, the legal memos purporting to supply the basis for the government’s policies, and the standards and procedures by which suspected enemies, including American citizens, were added to government “kill lists.” The ACLU and The New York Times brought litigation under the FOIA, but the courts proved unwilling to compel the government to publish more than it wanted to, repeatedly deferring to government’s assertion that additional transparency would cause grave damage to national security. At the same time, military and intelligence officials routinely supplied cherry-picked facts to the media under cover of anonymity. Through a veritable flood of officially sanctioned leaks, they painted a picture of a program that was closely supervised, legally sound, and narrowly focused on terrorists believed to be planning imminent attacks.

The officially sanctioned disclosures did not decrease the necessity of unauthorized ones. To the contrary, unauthorized disclosures became especially valuable because they served not merely to inform the public but to correct disinformation as well. Documents supplied by Hale allowed The Intercept to report that many of the government’s targets were not terrorists, that most of its strikes were predicated on evidence that was thin and unreliable, and that its “body counts” were based on the indefensible assumption that bystanders were combatants. The accounts of drone operators who were distraught by abuses they had witnessed further complicated the picture of a program that was supposedly discriminating and, in Obama’s description, “kept on a very tight leash.” On the whole, unauthorized disclosures about the drone program were probably less consequential than the ones that shaped public debate about interrogation and surveillance policy, but they nonetheless contributed in important ways to public discussion of the legality, morality, and effectiveness of the program, and helped spur significant reforms, including ones that brought U.S. policy into closer alignment with human rights law and almost certainly reduced civilian casualties.

Thus, in each of the areas discussed above—interrogation, surveillance, and extrajudicial killing—whistleblowers played an essential role. Indeed, consider what public debate about the “war on terror” would have looked like if the government had been able to conceal, for longer than it did, what these whistleblowers disclosed to us. The decisions to hold prisoners in secret black sites, to subject them to cruel treatment and torture, to deploy the NSA’s awesome surveillance powers against billions of people, to authorize extrajudicial killings that led to the deaths of hundreds of innocents in half a dozen countries—these decisions were surely among the most morally treacherous and politically consequential our government made in the two decades after 9/11. Would a democracy in which decisions of such moment had been concealed from the people indefinitely be worthy of the label? The question would be easy to answer even if one were to stipulate (against the evidence) that the decisions the government made in secret were ones the public would have endorsed. But the question is even easier to answer once it is recognized that, in each of the areas addressed above, government policy had escaped democratic control. We needed whistleblowers to narrow the distance between what the public had consented to and what the government was actually doing. Whistleblowers’ disclosures enabled more informed public debate, and ordinary citizens used what whistleblowers disclosed in order to press for reforms that were in many cases adopted. It was whistleblowers’ disclosures, in other words, that allowed the democratic process to work.

The essential role that whistleblowers have played over the past two decades is reason enough to doubt the defensibility of our current legal regime. But there is another reason, too: Even as post-9/11 developments have made whistleblowers more necessary, whistleblowing has become more difficult and more dangerous.

One reason for this is that new technology has made it easier for the government to identify leakers. Because virtually all government files are now stored electronically, the government usually has a precise record of who accessed the files and when. It also has detailed records of government employees’ communications and activities. The government determined that it was Reality Winner who leaked files relating to Russian cyberattacks because the government’s digital records showed that Winner was one of six people who had printed the records, and that Winner was the only one of the six who had communicated with The Intercept, the media organization that published the files. It is easier to leak information today than it was 50 years ago, but it is more difficult to preserve anonymity. Snowden assumed the government would be able to identify him quickly, and it did.

New surveillance authorities have also reduced the political costs associated with surveillance of journalists. Through mass surveillance programs, the government collects journalists’ communications metadata and (sometimes) content without targeting journalists specifically; the collected data is stored in databases that can be queried secretly. In addition, the USA Patriot Act expanded the government’s authority to use “national security letters” to obtain the call records of individuals not themselves suspected of engaging in criminality or espionage. The letters are served on communications providers without prior judicial authorization and typically accompanied by nondisclosure orders that foreclose the providers from alerting the targets of the government’s interest in their communications. The letters give the government the ability to identify reporters’ sources without paying the high political price sometimes associated with serving subpoenas on journalists and media organizations. During the Obama administration, Attorney General Eric Holder adopted guidelines that imposed new restrictions of the Justice Department’s use of subpoenas to identify sources, but the new restrictions did not apply to national security letters. Nor did Attorney General Merrick Garland’s more recent amendments to the guidelines restrict the use of these letters.

Perhaps most significantly, insiders who share official secrets with the press face a far more hostile legal landscape than they did 20 years ago. To be sure, it has long been the case that government insiders who consider supplying classified information to the public or press must contemplate the possibility of losing their public service employment, being stripped of their security clearances, and being prosecuted under a range of unforgiving criminal statutes, including, most significantly, the Espionage Act, which criminalizes the unlawful retention and communication of “national defense information.” The Espionage Act subjects insiders who disclose classified secrets to the possibility of harsh sanctions regardless of the value to the public of the information they disclose. Like the classification system, the act gives decisive weight to the government’s asserted interest in secrecy and no weight at all to the public’s interest in informed deliberation about national security policy.

But although the Espionage Act was enacted a century ago, only in recent years has the government begun using it against government insiders who disclose information to the press. During the 20th century, Samuel Loring Morison was the only person convicted under the Espionage Act for sharing classified secrets with the press, and President Clinton pardoned him in 2001. In the years since, however, the government’s use of the Espionage Act against journalists’ sources has become routine. The Obama administration indicted eight people under the act for leaking secrets to the press; the Trump administration indicted another five. In a sign of the times, James Comey, then the FBI director, told President Trump in 2017 that he “was eager to find leakers and would like to nail one to the door as a message,” and Jeff Sessions, then the attorney general, later told the media that the Justice Department was engaged in 27 investigations into classified leaks—a major escalation over previous years. Thanks to decisions by administrations of both major political parties over a period of two decades, whatever norms once dissuaded the government from using the Espionage Act against journalists’ sources seem to have evaporated.


Legal scholars, press freedom advocates, and others have argued that insiders should be protected in some circumstances for disclosing official secrets without authorization. One proposition common to many of these proposals is that leakers shouldn’t be held liable under the Espionage Act without a court first considering the public value of the information they disclosed.

In the remainder of this essay, I want to make a case for one version of this proposition—specifically, for allowing Espionage Act defendants to argue, as a defense to liability, that any harms caused by their disclosures were outweighed by the value of the disclosures to the public. At the risk of disappointing the reader, I’m not going to endorse or even describe any particular implementation of this public value defense here, though such a defense could be implemented in many different ways. What I want to do here is make a case, at a more conceptual level, for recognizing a public value defense in some form. The arguments for recognizing such a defense, as I’ll explain, are more numerous, diverse, and substantial than commonly appreciated.

The principal argument usually made for affording Espionage Act defendants a public value defense is that doing so would lessen the deterrent to socially beneficial leaks. This is a powerful argument, and perhaps reason enough to make such a defense available. In general, of course, we want government insiders to respect their nondisclosure agreements. It wouldn’t serve the public interest if security-cleared government employees and contractors subjected every classification decision to their own de novo review. We want government insiders to hesitate, and to think carefully, before disclosing official secrets without authorization. At the same time, we need government insiders to inform the public when they learn of significant abuses or systemic failures.

A public value defense would allow us to treat different kinds of leaks in different ways. It would allow us to distinguish socially beneficial leaks from socially harmful ones. While the defense would be unavailable as to most leaks (because the leaker would not be able to demonstrate that the social benefits of disclosure outweighed the harms), it would, by design, change the calculus for insiders with access to information of most value to the public. Under the new regime, insiders would consider not only the possibility that they would be charged under the Espionage Act but also the likelihood that they would be able to satisfy the public value defense. The Espionage Act would continue to serve as a strong disincentive to most leaks, but the possibility of satisfying the public value defense would reduce the disincentive for the leaks that are most important to the public. An appropriately calibrated defense would accommodate the government’s legitimate interest in protecting (most) classified information but also give appropriate weight to the public interest in maintaining democratic control over national security policy.

It might be argued that the current system doesn’t in fact prevent the public from learning about controversial national security programs the government would prefer to keep secret. The same evidence I marshal above to establish the indispensability of whistleblowers could be deployed instead to support the argument that the threat of heavy penalties does not dissuade insiders from bringing crucial information to public attention, at least when it really matters.

But this argument is ultimately unpersuasive. First, we don’t know what we don’t know; we know that some abuses were disclosed, but we have no idea what abuses might have been disclosed if we had had something other than a strict liability regime for leaks. In addition, we don’t know how the availability of a public value defense might have altered the timing of leaks. If a public value defense had been available, perhaps an insider might have leaked the infamous torture memos in 2002, when they were written; perhaps an insider might have disclosed the NSA’s call records program in 2003 rather than a decade later; perhaps the public might have learned of drone strikes’ civilian toll early enough to compel changes to government policy in 2010 or 2011 rather than 2013 or 2015. More fundamentally, there is simply no reason to believe that the logic of incentives and disincentives that operates in other spheres would not operate in this one. It seems reasonable to assume that the availability of a public value defense for leaks would have resulted in more leaks and a more informed public.

Another argument sometimes offered in support of the status quo is that a public value defense would over-incentivize leaks, perhaps in part because it would erode respect for the classification system by encouraging insiders to treat the government’s secrecy decisions as something other than final and binding. It is doubtful, however, that insiders would systematically overestimate the social benefits of disclosure, or systematically underestimate the social harms—especially because the disincentive to that kind of miscalculation would be severe, since a leaker who miscalculated in one of these ways would suffer the same consequences she would suffer under the current regime. Indeed, it seems quite likely that whistleblowing would be undersupplied even if we made a public value defense available, because even an insider who is justified in her belief that the social benefits of disclosure would exceed the harms would have to take into account the possibility that a jury might disagree—as well as the financial costs of mounting a defense to a possible prosecution. For this reason, it is unlikely that the recognition of a public value defense would result in a flood of leaks, or even an oversupply of them.

Nor is it likely that the recognition of a public value defense would erode respect for the classification system in some more fundamental sense. Consider other contexts in which we recognize analogous “necessity” defenses to general criminal prohibitions. The availability of a self-defense argument is not usually thought to erode respect for the prohibition against deliberate killing. The requirement that military service members refuse orders that are unlawful is not usually thought to erode respect for military discipline. To the contrary, it’s frequently suggested that these exceptions are vital to the legitimacy of the general rules.

Thus, the usual argument for extending public value defense to whistleblowers is compelling. Certain kinds of leaks—the ones that are most crucial to our democracy—are almost certainly undersupplied under our current legal regime, and affording leakers a public value defense could go some way towards correcting this.

But affording leakers a public value defense would have other important benefits, too, including some that are perhaps less obvious. For example, recognizing a public value defense could help regularize the treatment of leakers. Even in a broader landscape that is famously “disorderly,” our treatment of leakers stands out as especially irregular. Defendants charged under the Espionage Act often argue that their actions were justified, but they direct those arguments to the executive, which has the power to determine whether to offer a plea bargain, how severe a sentence to recommend, what the conditions of confinement should be, and whether to offer a pardon or another form of clemency. Because the executive is less concerned about precedent and consistency than the courts, and more exposed to political pressure, and also because the executive branch is poorly positioned to assess the value of disclosures relating to misconduct by executive branch officials, these decisions are unprincipled.

Allowing a public value defense would make our treatment of whistleblowers more predictable—that is, more orderly. While the executive would of course still control prosecutorial decisions, and the president would of course still control the pardon power, the courtroom would become the principal forum for consideration of the social benefit of the defendants’ disclosures, and the question whether the public value defense was satisfied in any given case would be answered within a consistent analytical framework by actors insulated to some extent from the politics of the moment.

Allowing a public value defense could also incentivize responsible leaking, as Yochai Benkler has observed. If the defense entailed a balancing of costs and benefits, the possible availability of the defense would give prospective leakers reason to consider the costs carefully. It would also give them reason to mitigate costs to the extent possible—for example, by limiting the scope of the information disclosed, redacting some information, or disclosing the information to a news organization likely to exercise independent editorial discretion in determining what to publish. With a public value defense, the law would recognize and give weight to these efforts at mitigation, instead of ignoring them, as it does now. It would encourage leakers “to be careful what they leak.”

Relatedly, allowing a public value defense would more closely align our regulatory framework with widely shared intuitions about moral responsibility. Most of us share the view that individuals sometimes have a moral duty to break promises, disobey orders from superiors, or violate the law. Many of us believe there are circumstances in which government insiders who have agreed to guard the government’s confidences are morally obliged to betray those confidences instead—though of course we may disagree about the circumstances in which this obligation should attach, and about whether the obligation attached in a given case. Important social institutions frequently honor whistleblowers and the journalists and media organizations that publish the information they provide. Allowing a public value defense would better align our legal system with the widely shared view that whistleblowing is justified and socially beneficial in some circumstances.

And if allowing a public value defense would legitimize some whistleblowing, it would also have the effect of lending legitimacy to Espionage Act convictions. Because the current regime denies defendants the opportunity to argue that their actions were justified, it allows those convicted under the Espionage Act to claim the mantle of whistleblowers even if the requirements of a public value defense would not have been satisfied in their cases. An insider who discloses official secrets of little value to the public, or with little regard for the harms likely to be caused by the disclosures, can claim the same mantle as one who acted more responsibly. As a result, the unavailability of a public value defense has the effect of tainting even those Espionage Act prosecutions in which the defense would not have been available.

Recognizing a public value defense could have still other benefits. For example, the executive branch might disclose more information of its own accord if government insiders could sometimes expect to receive legal protection for responsibly disclosing official secrets of especial value to the public. Under the existing regime, executive branch officials have almost total control over what information the public receives about executive branch policy. Empowering whistleblowers would have the effect of subjecting the executive branch’s classification decisions to a kind of public interest override in some narrow contexts. The possibility that a whistleblower might disclose information would incentivize government officials to consider disclosing the information preemptively. An appropriately calibrated public value defense would change the calculus of the executive branch with respect to the withholding and selective disclosure of information of unusual relevance to the public’s ability to understand, evaluate, and influence national security policy.

Allowing a public value defense could also help resolve one of the fundamental tensions of the post-Pentagon Papers legal framework. As others have observed, there is something paradoxical, troubling, and even unseemly about our system’s disparate treatment of whistleblowers and journalists. Insiders who disclose classified secrets routinely receive the harshest punishments; the journalists who report the same secrets are more likely to receive Pulitzer Prizes. Allowing a public value defense would close this gap to some extent. There would still be (many) circumstances in which the insiders who disclosed official secrets could be punished but the journalists who published the secrets could not, because the public value defense would apply only to some classes of disclosures, and only where the harms caused by the disclosures were outweighed by their social benefits. (I assume for present purposes that the First Amendment would preclude the government from prosecuting a journalist or publisher under the Espionage Act except in truly extraordinary circumstances.) But with respect to some class of disclosures of especial significance to the public’s ability to maintain democratic control of national security policy, the new regime would treat leakers and journalists the same way.

Finally, allowing a public value defense would also provide the courts an appropriately central role in protecting a channel of information that is crucial to the functioning of our democracy. Given the essential role that whistleblowers play in protecting democratic control over national security policy, the treatment of whistleblowers should not be left entirely to the executive branch. The courts are better positioned than the executive to assess the social benefits of whistleblowers’ disclosures, because they are relatively insulated from immediate political pressures and accustomed to giving legal effect to democratic values. Moreover, even if the executive branch is better positioned than the courts to assess the risks associated with disclosure of classified information, as it is sometimes suggested to be, it is not better positioned to assess the risks associated with disclosures relating to malfeasance by the executive branch, and this is the category of information we are most concerned with here.


For the past 20 years, whistleblowers’ disclosures have been vital to our ability to understand national security policy and to hold government officials accountable for their decisions. If national security policy can still be said to reflect democratic consent, it’s in large part because whistleblowers shared information with the press and public without authorization. As we reflect on the two decades that have passed since 9/11, we should recognize the immense debt we owe to whistleblowers. Extending new legal protections to them should be regarded as a moral and democratic imperative.


This piece is adapted from an essay first published in National Security, Leaks & Freedom of the Press (Lee C. Bollinger & Geoffrey R. Stone, eds. 2021). For helpful comments on drafts, I would like to thank, in addition to Lee Bollinger and Geoffrey Stone: Alex Abdo, Carrie DeCell, Ashley Gorski, Brett Kaufman, Meenakshi Krishnan, Alice Ristroph, Larry Siems, Patrick Toomey, and Ben Wizner. Thanks also to Meenakshi Krishnan for invaluable research support and to Helen Zhong for help with cite-checking and proofreading.


Printable PDF


© 2021, Jameel Jaffer.


Cite as: Jameel Jaffer, What We Owe Whistleblowers, 21-08 Knight First Amend. Inst. (Sept. 9, 2021), https://knightcolumbia.org/content/what-we-owe-whistleblowers [https://perma.cc/GH2F-S3DY].

Matthieu Aikins and Najim Rahim, Afghan Family Says Errant U.S. Missile Killed 10, Including 7 Children, New York Times (Aug. 30, 2021), https://www.nytimes.com/2021/08/30/world/asia/afghanistan-drone-attack-ISIS.html.

Ryan Devereaux & Murtaza Hussain, Daniel Hale Sentences to 45 Months in Prison for Drone Leak, The Intercept (July 27, 2021), https://theintercept.com/2021/07/27/daniel-hale-drone-leak-sentencing/.

Senate Select Committee on Intelligence, Report on the U.S. Intelligence Community’s Prewar Intelligence Assessments on Iraq (2004), https://perma.cc/AZE2-W29B (concluding that excessive secrecy impeded information sharing within the intelligence community, interfered with external oversight, and made it more difficult to detect analytical errors and policymakers’ exaggerations after they had occurred).

Peter Finn & Julie Tate, Justice Department To Investigate Deaths of Two Detainees in CIA Custody, Wash. Post (June 30, 2011), https://perma.cc/V9SN-4SFM; Human Rights Clinic at Columbia University and Center for Civilians in Conflict, The Civilian Impact of Drones: Unexamined Costs, Unanswered Questions (2012), https://perma.cc/EBM6-XKNG.

Senate Select Committee on Intelligence, Report (2014), https://perma.cc/X9XC-BRP9 (stating that CIA’s detention and interrogation program created tensions with U.S. partners and allies and complicated bilateral relationships).

Douglas A. Johnson, Alberto Mora, & Averell Schmidt, The Strategic Costs of Torture, Foreign Affairs (Sept. 2016), https://perma.cc/UKQ4-6NB3.

See, e.g., Senate Select Committee on Intelligence, supra note 3 (stating that CIA contractors who helped develop “enhanced interrogation techniques” were paid $81m despite lacking specialized knowledge of al Qaeda, a background in counterterrorism, or any relevant cultural or linguistic experience).

Tom Malinowski, Restoring Moral Authority: Ending Torture, Secret Detention, and the Prison at Guantanamo Bay, 618 Annals Am. Acad. Pol. & Soc. Sci. 148 (2008), https://perma.cc/GK9T-CFFF; Senate Select Committee on Intelligence, supra note 3 (“the program caused immeasurable damage to the United States' public standing”).

Pew Research Center, Public Trust in Government: 1958-2019 (Apr. 11, 2019), https://perma.cc/E9AL-R84L.

Exec. Order No. 13526, 75 F.R. 705 (2009), https://perma.cc/Z3CA-3D57.

See, e.g., ACLU v. Dep’t of Justice, 681 F.3d 61, 73-75 (2d. Cir. 2012) (holding that CIA could withhold records relating to use of unlawful interrogation method); N.Y. Times v. Department of Justice, 872 F. Supp. 2d 309, 317 (S.D.N.Y. 2012) (holding that Justice Department could withhold surveillance-related records even if the records described or contained the agency’s procedural or substantive law).

Brief of Amici Curiae Scholars of Constitutional Law, First Amendment Law, and Media Law at 7 n.9, United States v. Albury, No. 18 Crim. 00067 (D. Minn. Oct. 4, 2018), https://perma.cc/3T3K-3V46.

See Heidi Kitrosser, Reclaiming Accountability: Transparency, Executive Power, and the U.S. Constitution 63 (2015).

Ron Wyden, Statement on the Intelligence Declassification Provision in the Intelligence Reform Bill (Dec. 8, 2004), https://perma.cc/VV2F-22LM.

Executive Order 13526 contemplates the possibility of sanctions, Section 5.5, but in practice sanctions are almost never imposed, Exec. Order No. 13526, 75 F.R. 705 (2009), https://perma.cc/NZ3X-UTYJ; J. William Leonard, When Secrecy Gets Out of Hand, L.A. Times (Aug. 10, 2011), https://perma.cc/8KFU-Y7M2.

Daniel P. Moynihan, The Culture of Secrecy, National Affairs (1997) at 67.

Erwin N. Griswold, Secrets Not Worth Keeping, Wash. Post (Feb. 15, 1989), https://perma.cc/TD9W-ZW3R (“It quickly becomes apparent to any person who has considerable experience with classified material ... that the principal concern of the classifiers is not with national security, but rather with governmental embarrassment of one sort or another.”).

National Commission on Terrorist Attacks Upon the United States, Federation of American Scientists (May 22, 2003), https://perma.cc/TD9W-ZW3R (quoting former CIA Director Porter Goss).

President Obama acknowledged this himself when he was asked about Secretary Hilary Clinton’s use of a private email server to store emails that may have contained classified information. See Obama On Clinton’s Emails: ‘There’s Classified, And Then There’s Classified’, CSBN (Apr. 10, 2016), https://perma.cc/ES7C-XNXX (“What I also know, because I handle a lot of classified information, is that there are — there’s classified, and then there’s classified,” Obama told Fox News. “There’s stuff that is really top-secret, top-secret, and there’s stuff that is being presented to the president or the secretary of state, that you might not want on the transom, or going out over the wire, but is basically stuff that you could get in open-source.”); see also Geoffrey Stone, Top Secret (2007) at 192 (classification is a ““highly imperfect guide to the need for confidentiality“).

Press-Enter. Co. v. Superior Ct., 478 U.S. 1, 13 (1986) (recognizing First Amendment right of access to preliminary hearings); Press-Enter. Co. v. Superior Ct., 464 U.S. 501 (1984) (recognizing First Amendment right of access to voir dire); Globe Newspaper Co. v. Superior Ct., 457 U.S. 596 (1982) (recognizing First Amendment right of access to criminal trials); Richmond Newspapers v. Virginia, 448 U.S. 555, 576 (1980) (recognizing First Amendment right of access to criminal trials).

See, e.g., Rushford v. New Yorker Magazine, Inc., 846 F.2d 249, 253 (4th Cir. 1988); Westmoreland v. Columbia Broad. Sys., Inc., 752 F.2d 16, 23 (2d Cir. 1984).

Houchins v. KQED, 438 U.S. 1, 9 (1978) (“This court has never intimated a First Amendment guarantee of a right of access to all sources of information within government control.”).

Ctr. for National Security Studies v. Dep’t of Justice, 331 F.3d 918 (D.C. Cir. 2003) (stating that the Constitution “is not a Freedom of Information Act” and observing that, in our system, “disclosure of government information generally is generally left to the political forces”); United States v. Hasbajrami, Nos. 15 Civ. 2684, 17 Civ. 2669, 2019 WL 6888567 (2d Cir. 2019) But see In re Washington Post Co., 807 F.2d 383, 390 (4th Cir. 1986); Doe v. Gonzales, 386 F.Supp.2d 66 (D. Conn. 2005), appeal dismissed as moot, 449 F. 3d 415 (2d. Cir. 2006).

ACLU v. CIA, 710 F. 3d 422 (D.C. Cir. 2013); N.Y. Times v. Dep’t of Defense, 499 F. Supp. 2d 501 (S.D.N.Y. 2007); Ctr. for Int’l Environmental Law v. Office of U.S. Trade Representative, 777 F. Supp. 2d 77 (D.D.C. 2011); Cable News Network, Inc. v. FBI, 384 F. Supp. 3d 19 (D.D.C. 2019).

See, e.g., N.Y. Times v. Dep’ t of Justice, 915 F. Supp. 2d 508, 515–16 (S.D.N.Y. 2013), https://perma.cc/J8DQ-FJ7W (“I can find no way around the thicket of laws and precedents that effectively allow the [government] to proclaim as perfectly lawful certain actions that seem on their face incompatible with our Constitution and laws, while keeping the reasons for its conclusion a secret.”). Executive Order 13526 bars the government from classifying information “in order to … conceal violations of law,” but this narrow prohibition turns on the reasons why the records are being withheld, not on the content of the records, and the courts almost never question the government’s proffered reasons.

Jameel Jaffer, Selective Disclosure About Targeted Killing, Just Security (Oct. 7, 2013), https://perma.cc/7HRC-6V4Y. Thus, the Republican Policy Committee’s statement in support of the proposed legislation declared that, “[i]n this period of selective disclosures, managed news, half-truths, and admitted distortions, the need for this legislation is abundantly clear.”


Jameel Jaffer, Known Unknowns, 48 Harv. C.R.-C.L. L. Rev. 457 (2013), https://perma.cc/FXG4-B6FM.

See National Security Whistleblowing: A Gap in the Law, Brennan Ctr. for Justice (Aug. 21, 2013), https://perma.cc/6ZRY-73UW.

Benkler calls these “accountability leaks”—“those that expose substantial instances of illegality or gross incompetence or error in certain classes of particularly important matters associated with the activities of the national security system.” Yochai Benkler, A Public Accountability Defense for National Security Leakers and Whistleblowers, 8 Harv. L. & Pol'y Rev. 281 (2014) at 303.

For thoughtful attempts to imagine such a system, see, e.g., David A. Strauss, Keeping Secrets, in The Free Speech Century (Geoffrey R. Stone & Lee C. Bollinger, eds. 2018); David McCraw & Stephen Gikow, The End to an Unspoken Bargain? National Security and Leaks in a Post-Pentagon Papers World, 48 Harv. C.R.-C.L. L. Rev. 473 (2013), https://perma.cc/8739-TCRY; Lee Bollinger, Uninhibited, Robust, and Wide-Open 121 (2010).

Elizabeth Goitein & J. William Leonard, America’s Unnecessary Secrets, N.Y. Times (Nov. 7, 2011), https://perma.cc/NH2W-KACJ.

McCraw & Gikow, supra note 31, at 498–501.

Bollinger, supra note 31 at 120-26.

Mary-Rose Papandrea, National Security Information Disclosures and the Role of Intent, 56 Wm. & Mary L. Rev. 1381 (2015) at 1435.

Deposition of Max Frankel, 328 F. Supp. 324.

See Benkler, supra note 30, at 303.

Daniel P. Moynihan, The Culture of Secrecy, National Affairs (1997).

Dana Priest & William Arkin, Top Secret America (2011); Jack Shafer, The Spies Who Came in to the TV Studio, Politico (Feb. 6, 2018), https://perma.cc/E8G9-HSAZ.

Dana Priest & William M. Arkin, ‘Top Secret America’: A Look at the Military’s Joint Special Operations Command, Wash. Post (Sept. 2, 2011), https://perma.cc/98NG-LSK3.

Seymour Hersh, Torture at Abu Ghraib, New Yorker (Apr. 30, 2004), https://perma.cc/Z6SU-8CX9 (noting that the photographs were first broadcast on 60 Minutes a week prior).

Dana Priest, CIA Holds Terror Suspects in Secret Prisons, Wash. Post (Nov. 2, 2005), https://perma.cc/29KT-93T5.

Mark Mazetti, C.I.A. Destroyed 2 Tapes Showing Interrogations, N.Y. Times (Dec. 7, 2007), https://perma.cc/J59N-Z5PS.

U.S. Military Medicine in War on Terror Prisons, Univ. of Minn. Human Rts. Lib., https://perma.cc/3LZS-FDFC (listing some of the investigations).

Inquiry into the Treatment of Detainees in U.S. Custody, U.S. Senate Committee on Armed Services, S. Prt. No. 110-54 (Nov. 20, 2008), https://perma.cc/33CT-WN9C.

Josh Meyer & Greg Miller, Holder Opens Inquiry of CIA Interrogations, L.A. Times (Aug. 25, 2009), https://perma.cc/Y8SB-ETZL.

Michael John Garcia, Interrogation of Detainees: Requirements of the Detainee Treatment Act, Cong. Res. Serv. (Aug. 26, 2009), https://perma.cc/Q4UD-2R2X.

David Stout, Interrogation of Detainees: Requirements of the Detainee Treatment Act, N.Y. Times (Sept. 6, 2006), https://perma.cc/T7PD-7JLM.

Jack Goldsmith, The Terror Presidency 152–60 (2007).

Obama Orders CIA Prisons, Guantanamo Shut, NBC (Jan. 22, 2009), https://perma.cc/N2S7-JB5H.

Senate Select Committee on Intelligence, supra note 3 at i.

Jameel Jaffer & Brett Max Kaufman, How to Decode the True Meaning of What NSA Officials Say, Slate (July 12, 2013), https://perma.cc/D8DH-6G2A.

DNI James Clapper says that the NSA does not collect data on millions of Americans, YouTube (Jun. 6, 2013), https://perma.cc/4XRU-745S. Clapper later apologized for the statement, acknowledging that it was “clearly erroneous.” Bill Chappell, Clapper Apologizes For Answer On NSA's Data Collection, NPR (July 2, 2013), https://perma.cc/93KC-6VHP.

Clapper v. Amnesty Intern. USA, 568 U.S. 398 (2013).

Jameel Jaffer & Patrick C. Toomey, The Solicitor General Should Correct the Record in Clapper, Just Security (Oct. 18, 2013), https://perma.cc/F2XG-RSBB.

See Jameel Jaffer, Obama Is Cancelling the NSA Dragnet. So Why Did All Three Branches Sign Off?, Guardian (Mar. 25, 2014), https://perma.cc/DRP6-S5NF.

ACLU v. Clapper, 785 F.3d 787 (2015).

Charlie Savage, Warrantless Surveillance Challenged by Defendant, N.Y. Times (Jan. 29, 2014), https://perma.cc/L2WM-ENL7.

USA FREEDOM Act, 129 Stat. 268 (2015).

Statement by the Press Secretary on the Review Group on Intelligence and Communications Technology, White House Archives (Aug. 27, 2013), https://perma.cc/4WXG-5KQ8.

Geoffrey R. Stone, The View from Inside the NSA Review Group, 63 Drake L. Rev. 1033 (2015), https://perma.cc/ZYG7-Y886.

See generally Jameel Jaffer, The Drone Memos 24–36 (2010).

Id. at 25.

Two appeals courts eventually ordered “disclosures” but only after concluding that the government had already “officially acknowledged” the information that the litigants had sought. N.Y. Times v. Dep’t of Justice, 756 F.3d 100 (2014) (requiring publication of OLC memo after concluding that Justice Department had officially acknowledged its contents); ACLU v. CIA, 710 F.3d 422 (D.C. Cir 2013) (rejecting CIA’s “Glomar” response after concluding that CIA had officially acknowledged an intelligence interest in targeted killings).

Because the leaks were authorized or tolerated by the administration and designed to marshal public support for the government’s policies, these disclosures are perhaps more accurately labeled “plants” or “pleaks.” See David Pozen, The Leaky Leviathan, 127 Harv. L. Rev. 512, 567 (2013).

As Jack Goldsmith later wrote, “the global picture [was] one of a concerted and indeed official effort by the [U.S. government] to talk publicly about and explain the CIA drone program—almost always in a light favorable to the administration, or at least to the person or interest[s] of the person who [was] speaking to the reporter.” Jack Goldsmith, Drone Stories, the Secrecy System, and Public Accountability, Lawfare (May 31, 2012), https://perma.cc/423C-HELS.

Cora Currier, The Drone Papers, Intercept (Oct. 15, 2015), https://perma.cc/E8WU-RKJ3; see also Jameel Jaffer, The Espionage Act and a Growing Threat to Press Freedom, New Yorker (June 25, 2019), https://perma.cc/9RWH-ZG9Z.

Mark Landler, Civilian Deaths Due to Drones Are Not Many, Obama Says, N.Y. Times (Jan. 30, 2012), https://perma.cc/98WU-7734.

Press Release, President Obama Outlines Counterterrorism Approach Rooted in Human Rights, Human Right First (Dec. 6, 2016), https://perma.cc/F6YR-XWRT.

Affidavit in Support of Application for Arrest Warrant, United States v. Winner, No. 17-024 (S.D. Ga. 2017) https://perma.cc/93AS-ALQS.

Benkler, supra note 30; McCraw & Gikow, supra note 31.

Edward Snowden, Permanent Record 262–63 (2019).

See Charlie Savage, Fight Brews Over Push To Shield Americans in Warrantless Surveillance, N.Y. Times (May 6, 2017), https://perma.cc/B4DB-M47U.

See, e.g., 18 U.S.C. § 2709(c).

McCraw & Gikow, supra note 31.

Charlie Savage, Holder Tightens Rules on Getting Reporters’ Data, N.Y. Times (July 12, 2013), https://perma.cc/Y2YQ-DQYA.

Anna Diakun & Jameel Jaffer, The Justice Department’s New Media Protections Are (Mostly) a Promise, Not Yet a Reality, Knight Institute (July 29, 2021), https://knightcolumbia.org/blog/the-justice-departments-new-media-protections-are-mostly-a-promise-not-yet-a-reality.

18 U.S.C. § 793(d) (2018).

Following the Fourth Circuit in Morison, federal courts have limited the scope of “national defense information” in two ways, requiring prosecutors to show, first, that the information in question was “closely held” and, second, that its disclosure could “potentially” damage the United States or be valuable to its enemies. United States v. Morison, 844 F. 2d 1057 (4th Cir. 1988) at 1071-72. The courts have viewed classification as “strong evidence” that information was closely held. See Stephen P. Mulligan & Jennifer K. Elsea, Criminal Prohibitions on Leaks and Other Disclosures of Classified Defense Information, Cong. Res. Serv. 16 (Mar. 7, 2017).

Steven Aftergood, When the President Pardoned a Leaker, Fed. of American Scientists (Sept. 21, 2016), https://perma.cc/AVY7-PR87.

Gabe Rottman, A Typology of Federal News Media “Leak” Cases, 93 Tulane L. Rev. 1147, 1181–86 (2019); Alex Emmons, The Espionage Act is Again Deployed Against a Government Official Leaking to the Media, Intercept (Oct. 9, 2019) (describing the October 2019 indictment of counterterrorism analyst Henry Frese); Adam Goldman, Ex-Intelligence Analyst Charged with Leaking Information to a Reporter, N.Y. Times (May 9, 2019) (describing the May 2019 indictment of former intelligence analyst Daniel Hale).

Nicholas Fandos, Comey Memos Provide Intimate Look Into Trump Presidency, N.Y. Times (Apr. 19, 2018), https://perma.cc/R86J-NMXQ.

Betsy Swan, Leak Investigations Rise 800% Under Jeff Sessions, Daily Beast (Nov. 14, 2017), https://perma.cc/X7K9-55AJ.

For a discussion of why these norms have evaporated, see Mary-Rose Papandrea, Leaker Traitor Whistleblower Spy: National Security Leaks and the First Amendment, 94 B.U. L. Rev. 449, 454-64 (2014).

See, e.g., Brief of Amici Curiae Scholars of Constitutional Law, First Amendment Law, and Media Law, supra note 12 at 29–35 (arguing that sentences imposed on leakers convicted under the Espionage Act should take into account the “public value” of the defendant’s disclosures); Benkler, supra note 30, at 286 (arguing that Congress should extend a general criminal law defense to leakers who disclose government secrets “on the reasonable belief that by doing so they will expose to public scrutiny substantial violations of law or substantial systemic error, incompetence, or malfeasance”); Papandrea, supra note 84, at 539 (arguing that the First Amendment should be understood to bar the government from imposing criminal liability on leakers who intend to inform the public (as opposed to the enemy) except where the government can demonstrate that the leaks pose a “direct and grave danger to the nation’s security that is not outweighed by the public’s interest in the information”); Heidi Kitrosser, Leak Prosecutions and the First Amendment: New Developments and a Closer Look at the Feasibility of Protecting Leakers, 56 Wm. & Mary L. Rev. 1221, 1264 (2015) (arguing that the First Amendment should be understood to preclude the government from imposing criminal or other severe sanctions on leakers except where the government can show that the leaker lacked an “objectively reasonable basis” to believe that the public interest in disclosure outweighed the identifiable national security harms”).

There are many other ways, of course, that our legal regime could account for the public value of leaks. For example, we could preclude liability under the Espionage Act except where the defendant intended to aid the enemy rather than inform the public; or we could afford defendants the opportunity to point to the public value of their disclosures at the sentencing stage. But the idea of a public value defense can be used here as a stand-in for the broader category of reforms that would, in one way or another, give weight to the public value of the information disclosed.

Strauss, supra note 31, at 135 (“It would be a mistake to suggest that a government employee or contractor, no matter how knowledgeable or expert, should routinely appoint himself or herself the ultimate judge about what government secrets should be disclosed.”).

See, e.g., Christopher Fonzone, What the Law of Military Obedience Can (and Can’t) Do—What Happens if a President’s Orders are Unlawful, Just Security (May 4, 2018), https://perma.cc/J5D9-RCYM.

Alexander Bickel, The Morality of Consent 80 (1977).

Lee Bollinger, supra note 31 at 123 (“an orderly system is preferable”).

Benkler, supra note 30.

Strauss, supra note 31, at 135.

John F. Kennedy Profile in Courage Award: Joe Darby, John F. Kennedy Presidential Library and Museum, https://perma.cc/LL7D-MM9B (last visited Jan. 8, 2020); The Ridenhour Prize for Truth-Telling 2014: Edward Snowden, The Ridenhour Prizes, https://perma.cc/M93C-CAQ2 (last visited Jan. 8, 2020); 2006 Pulitzer Prize in Beat Reporting, The Pulitzer Prizes, https://perma.cc/QT86-SA4H (last visited Jan. 8, 2020); Ed Pilkington, Guardian and Washington Post win Pulitzer prize for NSA revelations, The Guardian (Apr. 14, 2014), https://perma.cc/K3K2-LDBS (NSA reporting at the Guardian was led by Glenn Greenwald).

Bollinger, supra note 31.

The government’s indictment of Julian Assange may test this proposition. Charlie Savage, Assange Indicted Under Espionage Act, Raising First Amendment Issues, N.Y. Times (May 23, 2019), https://perma.cc/C3T7-N8AL.

Deborah Pearlstein, After Deference, 159 U. Pa. L. Rev. 783 (2011).

Stone, supra note 19 at 193 n.27.

Jameel Jaffer is executive director of the Knight Institute.