Another in a continuing series of reports on First Amendment legal issues, from the Newseum Institute.
By Michael Barclay
Special to the Newseum Institute
First Amendment jurisprudence emphatically supports the notion that core political speech is the most highly protected category of speech. The justification for the narrow window through which governmental interests can take precedence over an individual’s right to speak freely and openly about the conduct of his or her governors is that an unhindered exchange of ideas is a prerequisite to an efficacious, accountable, and just form of government. Indeed, this notion is reflected in the Court’s embrace of the logic that political debate benefits from a marketplace of ideas. In such a marketplace, public debate cannot take place without an appropriate supply of information. If one wants his or her political ideas and opinions to carry any weight, he or she must be able to draw upon a steady flow of reliably relevant information about governmental conduct.
In this sense, the ability to freely exchange information with the public carries just as much importance as the actual expression of one’s political opinions. As Senior Judge Damon Keith of the Sixth Circuit U.S. Court of Appeals noted in the Detroit Free Press v. Ashcroft majority opinion, “democracies die behind closed doors.” Clearly, the circulation of relevant political information necessary to educate and inform the public is a prerequisite to a democratic society where political debate is to be considered robust in any way. If we are unable to share and distribute information about governmental conduct, there can be no reasonable expectation that the general public will be able to take the necessary actions to remedy misconduct when it occurs. With this perspective in mind, the nature of the executive branch’s system of classification of secret or confidential materials takes on a particular relevance for the maintenance of good governance.
Executive privilege has been a fundamental aspect of the executive branch since the drafting of the Constitution. But the emphasis placed on the importance of an energetic, unitary executive has always been accompanied by an emphasis on a strong separation of powers and an associated accountability on the part of the President. In fact, the entire justification for having a single President at the head of the executive branch was that he or she would be singly and wholly accountable to the opinions of the voting populace in a way that a bureaucracy is not. As Heidi Kitrosser remarks, “while the executive is designed to have energetic capacities—including a capacity for keeping secrets—it is also structured to maximize the chance that abuse of those capacities will be detected and remedied.” It has been assumed that freedom of the press protections afforded by the First Amendment, combined with the structural protections afforded by Congressional legislation, would act as a sufficient check on the power to keep official secrets. If individuals can make discreet and specific requests for classified information and the press is free to publish any relevant information that it encounters, the public should, in theory, be informed enough to wield the power of the vote in a productive manner.
But the sheer scale of excessive classification may well render these checks and balances insufficient to rein in the executive’s ability to unilaterally classify information related to national security. J. William Leonard remarked, in an interview with the New York Times, that he has “seen information classified that [he has] also seen published in third-grade textbooks.” While the volume of information deemed unfit for public consumption is frightening in and of itself, the real danger of overclassification flows from the incentive to use classification as political cover. As such, the temptation to hide politically embarrassing or damaging information “behind closed doors” (as Judge Keith would say) is too significant to allow rampant overclassification to remain unchecked if one is operating under the conviction that the free flow of information is a prerequisite to good governance.
The executive has a necessary right to keep some information secret in the name of legitimate national security concerns. But given the vast potential for political utility associated with the classification system and its suspect operational norms, the necessity of effective checks on the power of the executive to keep secrets is readily clear. Current structural checks on secrecy (namely, the Freedom of Information Act and the Reducing Over-Classification Act) have not proven successful in reclassifying and making public significant amounts of classified data and do nothing to alter a situation of complete deference to executive decision-making regarding classified information.
With Congress and the press unable to reign in the dangers of overclassification, another check on the executive’s misuse of the classification system rests with the willingness of government employees with access to classified information to disclose information to the public that exposes governmental misconduct. The public’s incentive to protect particular disclosures of classified information is significant, as “the vast range of information classified makes it largely inevitable that leaks will provide information about vital public policy issues.” However, current and former government employees and contractors face a daunting legal obstacle to fulfilling their vital role as “gatekeepers to much information of First Amendment value” in the form of Espionage Act indictments., Most recently, Jeffrey Scudder—a long-time CIA employee—had his security clearance revoked, his house raided by the FBI’s version of a SWAT team, and was forced into retirement under the threat of losing his pension. This punitive action was entirely for attempting to file a Freedom of Information Act request for a series of historical documents that should have been declassified years earlier, not for leaking information to the press. Scudder’s Freedom of Information Act request “destroyed [his] entire career”; an individual facing a full-blown Espionage Act indictment realized a life-shattering consequence.
To supplant existing checks on overclassification, it may be necessary to tip the scales towards an emphasis on the protection of the speech rights of classified information “gatekeepers” when balancing these protections with the right of the executive branch to keep information related to national security secret at its unhindered discretion. This paper attempts to consider such a recommendation’s impact on the classification system as a whole.
Section one will show that overclassification is a barrier to good governance on the grounds that stifling the free flow of politically relevant information destroys the accountability of the executive branch.
Section two will explore the ways in which the classification system is used as a political tool, rather than as a means to protect national security interests. This section will demonstrate that classified information can be mobilized, via information planting, to meet political goals at the expense of good governance. The analysis of planting and its relationship to unauthorized leaks of classified information draws heavily on David Pozen’s recent scholarship on the subject.
Section three will analyze checks on the executive’s power to classify information in the status quo. This section will emphasize the ways in which government employees and contractors with access to classified information are uniquely situated to provide a robust check on the problem of overclassification.
Section four will discuss the First Amendment rights of government employees with access to classified information as they currently stand. This section will also attempt to outline some of the legislative and legal barriers, made coherent through the lens of Heidi Kitrosser’s strength equals prerogative paradigm, that stand in the way of robust protections for speech related to classified information. It will not, however, make positive recommendations for what these rights should look like. Instead, it will focus on hindrances to these rights, with an eye to what would need to be dealt with if more potent protections were desired.
Section five will attempt to outline some of the potential effects of increased speech protections for classified information gatekeepers. The relationship between leak-law enforcement levels and authorized plants of classified information by high-level officials in the executive branch will receive particular attention.
1. Overclassification and Good Governance
In a May 2009 memorandum, President Obama ordered a review of classified information policies in an effort to facilitate an “unprecedented level of openness.” The body of the memorandum emphasized that “a democratic government accountable to the people must be as transparent as possible and must not withhold information for self-serving reasons or simply to avoid embarrassment.” The President’s statement hit the nail on the head in terms of addressing the potential dangers of overclassification. His quest for openness affirms the notion that freedom of information is a prerequisite concern for anyone seeking to establish a responsive and just mode of governance. It also suggests that overclassification is inherently in opposition to a properly functioning democratic government.
The ability of citizens to adequately respond to governmental actions and decision-making is the basis of self-governance. Given that the free flow of politically relevant information is integral to the public’s ability to mobilize in the face of government misconduct, the nature of the classification system has an integral role in determining whether the voting populace is being represented in good faith. With this in mind, it is readily clear that excessive government secrecy has “significant consequences for the functioning of the government itself” and that “power can be increased through controls on the flow of information.”
a. Overclassification At a Glance
Startling statistics published by the Information Security Oversight Office (ISOO) in 2013 support the notion that “overclassification has become an epidemic.” To begin with, a large number of individuals have the power to make classification decisions. They are divided into two categories of authority: original and derivative. Original classifiers “are those individuals designated, either by the President, by selected agency heads, or by designated senior agency officials with Top Secret original classification authority, to classify information in the first instance.” As the first line of defense against security-damaging transparency, original classifiers determine the guidelines followed by derivative classifiers. As the ISOO notes, “derivative classification is the act of incorporating, paraphrasing, restating or generating in new form information that is already classified.” In fiscal year 2013, there were 2,269 original classification authorities, a number touted by the ISOO as an improvement from the 2,326 with the same authority in fiscal year 2012. The number of individuals with derivative classification authorities is not a published figure, but the Moynihan Commission estimated that there were roughly three million with some form of classification authority at the time of the publication of its report in 1997. This number has likely grown larger with the expansion of the national security state since 9/11. There were 58,794 original classification decisions made in fiscal year 2013. In addition to the 58,794 original decisions, there were a reported 80,124,389 derivate classification decisions made during that year. Shockingly, this was a sixteen percent decrease from fiscal year 2012.
It is worth noting that the vast majority of these decisions are set in stone, evidenced by the paltry number of after-the-fact changes in classification status. Of the tens of millions of decisions made in fiscal year 2013, only 68 were formally challenged. Of these 68 challenges, only 12 were overturned after a subsequent review. Either classifiers are the most successfully infallible government officials on the planet or there is scant scrutiny applied to classification decisions. Clearly, the former is much less likely than the latter.
Another indicator of the relatively immutable nature of classification decisions is the slow pace of declassification. In fiscal year 2010, “the federal government spent $201 creating and securing old secrets for every tax dollar spent” on declassification efforts. There are three types of declassification review: automatic, systematic, and discretionary. Automatic declassification refers to the fact that classification status is automatically reviewed after 25 years. Of the 52.47 million pages of records reviewed in fiscal year 2013 under the automatic declassification program, 25.77 million pages were declassified—forty-nine percent of the reviewed pages. Any records exempted from automatic declassification are passed on to systematic declassification review, where 1.7 million of the 6.52 million pages that were reviewed were declassified—twenty-six percent. Discretionary review “is conducted when the public interest in disclosure outweighs the need for continued classification.” Accounting for the most limited category of review, only 55,671 of 346,351 pages—sixteen percent—were declassified under this category in fiscal year 2013.
While the presentation of declassification statistics in number of pages makes it appear to be an effective and wide-reaching process, the amount of information declassified relative to the amount that is made secret every year is miniscule. Since each classification decision usually governs more than one page of information, declassification statistics should be taken with a grain of salt and not compared to classification decision statistics in a one-to-one manner. However one views them, the important takeaway is that the rate of classification far outpaces the rate at which information is being declassified.
b. Factors That Contribute to Overclassification
The startling number of classification decisions made every year demands a further inspection of the factors that contribute to the steadily expanding mountain of information deemed unfit for public consumption. Two factors appear to contribute to overclassification.
The first can be thought of as structural, in the sense that it flows from the nature of the classification system rather than from purposeful misuse. The relationship between original and derivative classifiers provides the latter with an incentive to classify information with an exceptionally heavy hand. If the guidelines used to make derivative classification decisions are unclear and based exclusively on prior precedent, derivate classifiers have an incentive to classify anything that could possibly be included under these categories in an effort to avoid blame for accidentally publicizing closely held information. As Scudder notes, “no one ever gets punished for overclassification, but god forbid you let something out.” In addition to the tendency to want to cover personal liability for information that one encounters, many derivative classifiers may not have received the proper training to make these decisions. The 2013 ISOO Report to the President notes that only 54.55% of classifiers and 61.11% of derivative classifiers received training in “proper classification and declassification each year.” A combination of liability-aversion and improper training presents the perfect storm for over-extending classification guidelines.
The second contributing factor involves purposeful misuse of the classification system. Classification stamps are frequently used as a way to seal embarrassing and politically costly information off from public view. As President Obama’s 2009 openness memo suggested, this practice represents a much more damaging aspect of classification practices as far as public debate and freedom of information are concerned. While derivative classifiers might accidentally over-extend the classification system because of structural factors inherent to their position, classification for political cover is an active attempt to constrict and constrain the free flow of politically relevant information. Long lines of government insiders have attested to the widespread use of classification status to hide evidence of governmental misconduct. Senator John Kerry remarked that “more often than not…documents…remained classified or were classified to hide negative political information, not secrets.” The Moynihan Commission’s report echoed this sentiment with its assertion that “the classification system…is used too often to deny the public an understanding of the policymaking process, rather than for the necessary protection of intelligence activities and other highly sensitive matters.”
2. Mobilizing Classified Information for Political Gain
However, the political utility of classified information does not end with the executive’s ability to simply control which information is and is not released to the public. In addition to the problem of excessive classification for political cover, the institutionalization of classified information represents a powerful and commonly used political tool for the executive branch. Aside from the obvious benefits of hiding embarrassing evidence of governmental misconduct behind a classification stamp, state secrets can be deployed in a variety of politically productive, power-maximizing ways. The most perniciously useful manifestation of instrumentalized secrecy is the authorized, selective leaking of classified information for political gain (also known as planting).
a. Planting Defined
Selective leaking of classified information to the press allows the executive branch to obfuscate governmental conduct and reorient political debate while simultaneously achieving positive policy goals. The vast majority of leaked information comes from high-level government officials employing the political utility of information as part of “a rational, power-enhancing strategy.” At face value, a plant is little more than a leak that is authorized and made for a politically utilitarian purpose. But its status in the context of leak-law enforcement (historically, not legalistically, speaking) puts plants in an entirely different category than leaks by whistleblowers. Pozen suggests that a leak must reflect six conditions; by his definition, a leak is “(i) a targeted disclosure (ii) by a government insider (employee, former employee, [or] contractor) (iii) to a member of the media (iv) of confidential information, the divulgence of which is generally proscribed by law, policy, or convention (v) outside of any formal process (vi) with an expectation of anonymity.”
There are a great number of political situations in which a well-crafted leak of classified information is particularly effective, especially in policy areas subject to an increased level of secrecy (such as national security, foreign affairs, and intelligence policy). Plants allow the executive branch to orient or reignite debate about a particular policy; to navigate the complex interrelations of a multi-tiered bureaucratic structure; to communicate with political constituents (both domestic and foreign); to test the public’s response to particular policy decisions; to take credit for successful policies; to respond to time-sensitive issues; and to avoid the political consequences of calling a press conference and going through the normal channels of declassification (a notoriously arduous process).
Plants are the domain of high-level officials and the individuals that work for them directly (such as aides and assistant secretaries). As such, these individuals take it as given that their word speaks for itself, removing the necessity to hand over classified information in any tangible way. This means that they are insulated from enforcement of the laws that govern leaking as a result of political status. Because of who they are, and of the utility-based justification for divulging sensitive information to the media, planters shy away from the general, and often expository, leaks that characterize the actions of whistleblowers. Instead, they have the most to gain from highly calculated but purposefully minimal information that acts as a guide to a reporter’s story, rather than acting as the backbone of the story itself. For example, while Chelsea Manning’s leak of 700,000 military cables and reports unleashed a whirlwind in American politics, White House plants more often aim to change the direction in which the breeze is blowing. The Banality of Planting
Since the Teddy Roosevelt administrations ubiquitous application of the practice, politically utilitarian plants of classified information have become commonplace, with the press simultaneously “serv[ing] as lapdogs, watchdogs, and scapegoats for the executive branch.” Even the Moynihan Commission’s report on secrecy acknowledged the mundane nature of classified information plants and their relationship to leaks when it noted that “the anonymous leak, often at a senior level, has become an important tool of governing and a form of instant declassification (although the information leaked is likely to remain officially classified notwithstanding its publication).” The wide array of uses and the commonality of plants, in and of themselves, does not fully convey the pronounced role that plants have in executive branch political strategy. To gain an appreciation of their role in policymaking, one only needs to consult the legion of executive branch officials who acknowledge the fact “that leaking is a heterogeneous activity that occurs in a repeat-play environment.” Pozen brings these voices together in a concise manner:
“When President Obama’s former Chief of Staff William Daley raised hackles for acknowledging, ‘I’m all for leaking when it’s organized,’ he was simply giving candid expression to a deeply rooted belief…Planting is not an incidental practice of a few craven officials. It is programmatic, a mode of governance. ‘With every rollout of everything significant we do’ in the national security field, one White House official remarked in an interview, ‘there is a structured layout of how we will do the leaking’ (that is, the planting). All recent administrations appear to have done the same.”
3. Checks on Executive Secrecy and the Epistemic Privilege of National Security Gatekeepers
If the existence of an executive privilege to classify information as it relates to national security has necessitated a “structural and historical directive to balance the President’s capacity for secrecy against checking forces,” then the political utility of complete control over classified information means that special attention should be given to ensuring that those checks are robust and efficacious.
a. Checks and Balances in the Status Quo
There are two main categories of balances to overclassification—structural (Congress) and rights-based (the press and the public). Congress can statutorily impose side limits on the reach of classification through the passage of legislation such as the Freedom of Information Act, but it cannot dictate what kinds of information are deemed classifiable. Similarly, the press’s incentive to publish as much information as possible is thought to provide a counterbalance to the entrenched nature of extensive executive secrecy. Between the two, there should theoretically be enough balance to ensure that politically relevant information regarding national security and foreign affairs can reach the public.
What Heidi Kitrosser calls the “strength equals prerogative” paradigm explains the ways in which deference to the executive branch fundamentally disempowers structural and speech-based checks on secrecy. This argument supercharges “the President’s broad constitutional capacity to keep secrets with a constitutional prerogative to keep secrets in the face of counter-forces such as congressional requests for information or classified information leaks.” When individuals make requests to declassify information, or when classified information is at stake in a court case, the executive’s desire to keep that information secret is enough of a justification to do so. Such a tautological justification for the validity of classification decisions—that particular information needs to be classified because it is security-sensitive and it is security sensitive because it is classified—ensures that attempts to increase informational openness do not endanger the political utility of governmental opacity. An extension of this logic can be seen in “the only portion of FOIA that directly addresses national security,” Exemption 1. In order for the government to claim that particular information is unfit for release under Exemption 1, its classification status is the only justification necessary and “courts [can] not view the documents in camera or otherwise inquire whether the documents had been in fact properly classified.”
b. National Security Gatekeepers and the Power of Access
Government employees and contractors with access to classified information represent a potential supplement to the balancing functions of Congress and the press, but their willful acceptance of a position of trust (defined by the non-disclosure clauses of their employment contracts) complicates their right to speak openly about governmental misconduct. As Kitrosser says, these individuals act as “gatekeepers” to a wide range of policy information that is crucial to the kinds of robust and open debate that efficacious constitutional protections of political speech are meant to foster. The scale, pace, and nature of the classification system makes it inevitable that information that should be public will be classified. Regardless of whether a particular classification decision was the result of a culture of extensive secrecy or an act of political cover, classified information gatekeepers occupy a position of epistemic privilege relative to the general public. By this, I mean that they alone possess the contextualized knowledge to determine the legitimacy of classification decisions in a way that balances that information’s relevance to political debate with its impact on national security efforts, while ignoring the political implications of its publication for individual actors.
Clearly, a world in which government employees are combing through classified information intending to leak anything of political relevance is not productive; after all, a great deal of information is legitimately and reasonably classified. However, the edge cases where a derivative classifier oversteps his or her directive—or where White House officials are classifying documents with no relation to national security, but purely for political cover—demand that there be some means of educating the public about that information, especially when it involves the existence of governmental misconduct. Kitrosser spends a great deal of time and effort combating the notion that First Amendment protections for national security gatekeepers would not necessarily turn them “into a law unto themselves.” If high-level executive branch officials are allowed and encouraged to regularly leak classified information without crippling national security efforts, there is no reason why low-level officials cannot exercise good enough judgment to make the same kind of decisions in good faith. It seems as though the only difference is that plants are politically minded while leaks have the potential to be politically ruinous. Unfortunately for the White House, only the latter are crucial to informed public debate about policy and government conduct.
c. Gatekeepers, The Press, and Adversarial Contest
While the potential to leak classified information can be thought of as a check on overclassification, it is little more than a necessary supplement to the ability of the press to act as a useful counterbalance to executive privilege. The power of rights-based checks flows from the ability of the press to participate in an adversarial game with the executive branch. While the executive has an incentive to keep as much information secret as possible, the press has an incentive to publish as much politically relevant information as possible. But the press has to have access to information in the first place if such an adversarial relationship is expected to facilitate public access to the information necessary to understand and respond to national security policy.
In the status quo, the press is essentially playing this game with one hand tied behind its back by the possibility of costly and ruinous Espionage Act trials—both for sources and for the journalists themselves. The Espionage Act’s prohibition on the retention and solicitation of classified information is written so broadly that “the application of theories of inchoate liability…might render members of the media subject to criminal liability for acts of newsgathering wholly separate from publication.” The prospect of criminal liability for soliciting sources during the newsgathering process makes the press a meager adversary for the executive branch’s secret-keeping capacity. This is in large part because First Amendment jurisprudence suggests that only publication is protected by the speech and press clauses. Inchoate liability for newsgathering is even more disturbing if one considers the vast number of relevant issues that deal with classified information and the likelihood of a journalist soliciting such information without even being aware of doing so. It goes without saying that such a possibility could, and most likely does, have a chilling effect on journalistic efforts to educate the public on foreign affairs, national security, and intelligence policy.
When one takes the importance of these policy areas into account—as well as the massive importance of the press as a check on executive secrecy and, more generally, as a watchdog constitutionally empowered to spur political debate—there is an overt need to buttress the press’s ability to facilitate the disclosure of information that is publicly useful and dubiously classified, particularly if the dangers of overclassification are to be deterred in any meaningful manner. If government employees and contractors with access to classified information were protected from criminal liability for disclosures that overwhelmingly aid the public’s understanding of policy more than they damage national security interests, the press would not have to worry about being criminally liable for soliciting information from gatekeepers. In this legal landscape, potential criminal liability would most likely rest on sources based on whether a particular disclosure of classified information satisfies judicially and statutorily prescribed standards of review.
4. Barriers to Gatekeeper First Amendment Protections and the Strength Equals Prerogative Paradigm
a. Gatekeeper Protections in the Status Quo
While the potentially positive uses of the epistemic privilege afforded to national security gatekeepers are readily obvious, the amount by which disclosures of classified information are protected by First Amendment speech protections is definitive for garnering publicly useful outcomes from such a privilege. Currently, there is “absolutely zero protection” for gatekeepers, “even as a last resort, and even when the information reveals illegal government misconduct.” The Court has made it absolutely clear that “while ‘aggressive’ judicial balancing may be called for in some First Amendment contexts, it is not appropriate with respect to national security information leaks.” This view represents the distinction, made fully coherent under the Obama Administration between whistleblower protections afforded to employees in the intelligence and national security community, and protections afforded to the rest of the civil service. The director of public policy at the Project on Government Oversight, Angela Canterbury, remarked that the landscape for potential “non-national security/intelligence community whistleblowers…has dramatically improved…but for anyone making public disclosures about national security/intelligence wrongdoing, it is worse.”
b. Judicial Balancing and Classified Information
It is not illogical to extend an adjusted version of constitutional speech protections to classified information leaks. The Supreme Court’s decision in Bartnicki v. Vopper affirmed “that sanctioning the disclosure of ‘truthful information of public concern’ not only implicates speech, but ‘implicates the core purposes of the First Amendment.’” Even if this decision is not a directly applicable precedent for government employees indicted for leak-law violations, the existence of criminal statutes that ban the unauthorized disclosure of classified information does not preclude the possibility of weighing the value of such speech against applicable governmental interests. In fact, this is exactly how major precedent in First Amendment jurisprudence has come about—a state or federal law banned some form of speech and the Supreme Court decided that civil liberties were more important in some cases while upholding such statutes in others. Eugene Volokh mirrors this claim in his assertion that “the point of modern First Amendment law is that speech is often protected even though it violates a law restricting it…. Such laws…are nonetheless speech restrictions, and courts rightly evaluate them—and often strike them down—under the First Amendment.” In other words, there is no reason why the potential impact of the publication of security-sensitive information should not be rationally weighed against the importance of educating the public about politically relevant issues—even if that balancing is done with the judge’s thumb on the government’s side of the scales. Kitrosser sums up the precarious balance involved with applying First Amendment values to national security gatekeepers in stating,
“On the one hand, government employees serve as functionaries of Article II, subject to Presidential judgments with respect to national security secrets. In this sense, they bear an Article II responsibility that the press and the general public lack. On the other hand, they have special First Amendment value given their access to information within a vast and powerful secret-keeping system. Government employees thus merit a more moderate level of protection than do the press and the public, but a level substantially greater than that reflected by the automatic or presumptive criminalization of classified information leaks.”
c. The Barriers to Gatekeeper Protections: Heidi Kitrosser’s Strength Equals Prerogative Paradigm
The refusal to treat cases involving classified information leaks in the same way as other First Amendment cases brings us back to Kitrosser’s formulation of the strength equals prerogative paradigm. Two iterations of the argument are particularly useful for understanding why First Amendment arguments have been deemed inapplicable to leak-law violation cases by the courts. The first is that, because national security gatekeepers have accepted a position of trust as part of the terms of their employment (which include non-disclosure clauses), the leaking of classified information is nothing more than a breach of conduct. In Snepp v. United States, this breach of trust was so important to the Court that “whether Snepp violated his [position of] trust does not depend upon whether his book actually contained classified information.” Regardless of whether Snepp’s refusal to submit to a pre-publication review of his book did any actual harm to national security interests, the fact that he disobeyed the terms of his employment contract is all that matters. The emphatic reliance on the terms of employment for national security leakers, rather than on the actual damage to legitimate governmental interests caused by particular leaks, represents the exceptional legal status of government employees. The refusal to consider the material ramifications of a contractual violation—as the judiciary would do in a normal First Amendment case—shows the legalistic power wielded by the government’s prosecutorial team every time that leaks are treated as conduct and not speech.
While legal discourse revolving around breaches of conduct figure prominently in every case involving leak-law violations, Snepp also suggests that gatekeepers’ speech rights could legitimately be limited “even in the absence of an express agreement,” despite the fact that such freedoms, “in other contexts, might be protected by the First Amendment.” Regardless of whether the executive can circumvent First Amendment protections of political speech without an employment contract that contains a non-disclosure provision, the breach of trust involved in the unauthorized disclosure of classified information provides the main justification for a punitive response to such actions.
The second relevant aspect of the strength equals prerogative paradigm is that only the executive branch has the expertise and contextual understanding to make determinations about “whether and when information leaks should be punished.” Judge Ellis’ comments during the hearing to reduce Lawrence Franklin’s sentence seems to step back from the Snepp court’s assertion that the existence of a breach of conduct makes the existence (or non-existence) of harm to national security interests irrelevant. A major issue in the case was Ellis’ assertion that a leak has to include “national defense information” for there to be a violation of leak-laws. Despite the potential that such a determination might have for the protection of gatekeepers who leak non-security-sensitive information in good faith, he completely deflates any protective power by going on to add that “it isn’t the judiciary’s task or obligation to determine what should or shouldn’t be classified. That’s…an Executive Branch decision.” Even if a leak does not include classified information—or includes information that is already public, despite remaining classified—a deferral to the executive branch regarding the admissibility of relevant information could leave defendants completely unable to provide the evidence necessary for a reasonable defense.
These two aspects of the strength equals prerogative paradigm are fundamentally intertwined and contribute to a unanimous, unchecked control over the distribution of classified information without questioning the relevance of particular classification decisions to national security interests. Labeling speech as conduct sidesteps the only legitimate justification for unauthorized disclosures of classified information. As Kitrosser notes, such a flippant bypassing of critical constitutional questions “is an analytical nonstarter where speech is regulated for its content.” Completely accepting an expansive theory of executive privilege makes it impossible to take the legitimacy of particular classification decisions into account when evaluating leak-law violations. If the statistics and first-hand accounts provided in Section One are evidence of a bloated and misused classification system, then it seems unproductive to take the wielding of a classification stamp at face value. Furthermore, the political utility of planting suggests that more effort should be made to counter the ways in which secrecy encourages governmental opacity and the proliferation of politicized classification decisions. If authorized plants can exist without bringing about the collapse of the national security state, an increased role for the judiciary in addressing leak-law violations may be able to increase executive transparency in areas that matter while ensuring that legitimate security concerns are fully and appropriately considered.
d. The Prospect of Judicial Balancing
An increased role for the judiciary would not entail a complete reimagining of inter-branch checks and balances. Such power could come from Congress or from self-imposed court precedent specific to disclosures of classified information by government employees. While this paper does not attempt to make specific suggestions in regard to potential standards or statutes, it is obvious that any change in gatekeeper First Amendment status would have to deal with the strength-equals-prerogative paradigm by taking a more nuanced approach to classified information and the material repercussions of any particular disclosure of it. As long as overclassification exists and is politically incentivized, classification decisions and punitive actions to protect them must be considered suspect and be balanced with the importance of a responsive citizenry armed with the necessary knowledge to engage in informed political debate. A judiciary armed with the power of precedent and congressional statute is just one of the possible paths to ensuring a proper balance of executive privilege and the public’s right to engage in informed political debate.
Judge Ellis’ use of the Classified Information Protection Act (CIPA) in Rosen and Weissman is a good example of why judicial balancing represents a feasible means of weighing the importance of national security interests and the power of executive privilege against the constitutional rights of leakers and the public’s right to have access to politically relevant information. As part of their defense, Rosen and Weissman wanted to use the entirety of the documents they divulged “in an effort to show that the information they disclosed was not the basis for the Report’s classification.” Predictably, the government’s prosecution team “sought to redact large portions of the Report, asserting that those portions were not relevant to the defense.” Faced with these two interpretations of the same piece of classified information, the Court was able to solicit the justifications for the admittance of both versions of the Report and “fashion a substitution that would protect the defendants’ rights, while simultaneously preventing the unnecessary disclosure of classified information.” It seems as though Judge Ellis sided more with the defendants than he did with the executive, because the prosecution ended up dropping the charges against Rosen and Weissman on the grounds that the evidence ruling was unfavorable to their interests. While it is impossible to know for sure, it seems likely that the charges were in fact dropped because judicial balancing gave the defendants a more robust defense, rather than because security-sensitive information would have been revealed at trial.
Regardless of the outcome of judicial balancing in this particular instance, the use of CIPA in Rosen and Weissman’s case injects a scintilla of hope into the notion that the court system can adeptly evaluate and resolve tensions between public and governmental interests in the realm of classified information. Congress seems to agree with this optimism. The FOIA Improvement Act of 2014—recently introduced in the Senate by Patrick Leahy (D) and John Cornyn (R)—aims to add a similar type of “public-interest balancing test” to Exemption (b)5. If this provision of the bill becomes law, perhaps it can serve as a kind of litmus test for judicial balancing in the context of classified information.
5. The Aftermath of National Security Gatekeeper Protections
a. Impact on the Number of Leaks and Executive Self-Checking
The first foreseeable consequence of increased protections for whistleblowers via judicial balancing is the potential for an increase in leaks. It follows logically that government employees with the legal grounds to publicize governmental misconduct under the side constraints imposed by judicial balancing would be more likely to do so. However this possibility is, itself, a catalyst for the opposite outcome—an improvement in intra-agency methods for addressing instances of misconduct that are hidden by classification. If the executive branch could prove in court that efficacious and easily accessible internal channels to address misconduct were present, it would make leaks of classified information to the press unnecessary and fully punishable by law. Oddly enough, increasing the likelihood of leaks could provide an incentive for self-checking by the executive branch in a way that more effectively guards security-sensitive information from the public.
The strategy of enhanced self-checking was already applied to non-national security gatekeepers with the passage of the Whistleblower Protection Enhancement Act (WPEA), which drastically increased the efficacy of the Office of Special Counsel for potential whistleblowers outside of the national security and intelligence communities. A reason that there is a difference in legal protection for national security and intelligence community whistleblowers is because the original version of the bill (which did not make such a distinction) would likely have been vetoed by President Obama, even if it had passed the Senate. Amending the WPEA to expand whistleblower protections to national security gatekeepers stands as a feasible option to buttress self-checking in the face of increased protections for leakers in a way that would encourage the utilization of the former over the latter as a means to address governmental misconduct. Alternatively, continuing to provide inadequate means of executive self-checking would give gatekeepers license to disclose classified information to the press as long as those disclosures abide by whatever statutorily or jurisprudentially prescribed standards exist. In such a scenario, leakers would still be held accountable for their actions, but a punitive response would only be justified in instances where a judge determines that a particular leak did more harm than good.
b. Planting and Leak-Law Enforcement: David Pozen’s Theory of Constructive Ambiguity and Manufactured Scarcity
The effect of an exceptional gatekeeper First Amendment status on planting and enforcement levels is much more complicated than the potential ramifications for the number of leaks because it begs the question of the relationship between plants and the enforcement of leak-laws. In order for plants to be successful as a core aspect of the executive branch’s political strategy, that strategy has to keep two characteristics intact—what Pozen calls “constructive ambiguity” and “manufactured scarcity.” The maintenance of these structural characteristics depends primarily on the specific ways in which leak-laws are enforced and the message that that enforcement sends to the general public.
Constructive ambiguity refers to the notion that in order for plants to remain politically efficacious, they have to be able to pass as coming from the same publicly-concerned standpoint as information sourced from a low-level, genuinely concerned whistleblower. If the source of classified information leaks is obviously the executive branch, plants get red-flagged in debate as part of a concerted effort to control the flow of political debate through a manipulation of the media. Manufactured scarcity refers to how the impression that leaks are intolerable, highly illegal, and heavily prosecuted gives authorized plants a much more authentic and salient political effect. As long as the public believes that planted information was only made public through an acknowledged self-sacrifice for the public good, plants are perceived as alarm bells when they are almost always carefully cultivated and politicized tools. Manufactured scarcity disguises power-maximizing politicking as an act of anonymous selflessness in the public’s eyes.
Constructive ambiguity and manufactured scarcity gain force through the relationship between plants and leaks. This relationship is made coherent through a strategy of overall “de minimis criminal enforcement of the laws against leaking” in the highest-profile way possible. This statement may seem contradictory, but the key takeaway is that investigations should be as narrowly scoped and rare as possible while simultaneously signaling a hardline stance against the leaking of classified information. Since an energetic enforcement of leak-laws is likely to indict the actions of the executive branch just as much as it would root out low-level whistleblowers and leakers, the best course of action is a strategy of selective enforcement where leaked information that is damaging to the executive branch’s policy agenda is pursued in an exceptionally punitive and publicized manner. While overall enforcement hovers around a “0.3% indictment rate,” the actions of individuals like Chelsea Manning and Edward Snowden become a chance for a president to repudiate the practice of leaking as a whole while leaving its politically useful cousin (the plant) untouched by any resulting investigation. At this point, it becomes readily obvious that leaks and plants rely on each other in a very specific and complex way. As Pozen notes, “Planting depends upon leaking to give it political and epistemological breathing room; leaking, in turn, depends upon planting to give it legal breathing room. The two are fundamentally symbiotic. Plants need to be watered with leaks.”
To be absolutely clear, violations of leak-laws are not left unprosecuted because it is difficult to win these cases or prove wrongdoing. In fact, the opposite is true: in a world where the majority of our digital communications are recorded in some manner, there is no law applicable to criminal cases that prevents the issuance of subpoenas to journalists in an effort to reveal their sources, and the legal defense available to leakers is “tightly constrained.” Despite these broad punitive powers, such measures are rarely exercised. This is a matter of active willfulness to leave the vast majority of leak-law violators untouched by the critical eye of a full-scale investigation. Of the 25 leak investigations opened by the FBI between 2005 and 2009, the Bureau was able to identify suspects in 14 of those cases. But the majority of these investigations were closed after executive agencies were unwilling to cooperate with investigators. The repeated failure of feasible attempts to enforce leak-laws as much as possible drives home the point that the executive has an interest in selectively allowing disclosures of classified information that are politically beneficial. President Clinton vetoed the fiscal year 2001 Intelligence Authorization Act because it included a provision that would “ensure the prosecution of all unauthorized disclosures of classified information.” The Bush administration routinely “den[ied] security clearance to Justice Department ethics attorneys to investigate leaks…thus shutting down the investigation[s].”
If full enforcement of leak laws is logistically viable, and even quite easy, the question of how to enforce them in a way that leaves constructive ambiguity and manufactured scarcity in place presents itself as the lynchpin to understanding the increase in Espionage Act indictments under President Obama. David Pozen discusses at length that the executive finds itself between a rock and a hard place in its efforts to preserve constructive ambiguity. On the one hand, the DOJ can’t prosecute every single leak of information because most leaks are plants sourced from high-level officials who are functionally immune from such indictments. On the other hand, pursuing every single unauthorized leak would mean that every leak left unprosecuted would blatantly be a product of plants by the executive branch, thus destroying any sense of constructive ambiguity. Without that level of ambiguity, the executive branch is just as well off holding a press conference and openly disclosing classified information. This is obviously an untenable strategy in a world where low-level officials are routinely prosecuted for the same crime—but only when they reveal information that is politically costly and embarrassing. The opportunity cost of enforcement represents a huge threat to a political tool that allows the White House and executive agencies to wield a fair amount of power without risking the consequences of publicly linking oneself to any particular statement.
The consequences of an overly broad investigation are evident in the experience of the second Bush administration and the debacle that resulted from the planting of intelligence reports regarding Iraq. While these plants were politically advantageous in the sense that they publicized information that turned the tide of public opinion about Saddam Hussein and the invasion of Iraq, the fallout from their exposure as plants was as costly as the plants were useful. What had previously come off as a concerned official letting the world know that Saddam Hussein had attempted to purchase uranium from Niger was now seen as a calculated, politicized response to an Op-Ed column by former Ambassador to Gabon Joseph Wilson, “What I Didn’t Find in Africa.” Suddenly, Harry Reid was demanding that President Bush “fully disclose his participation in the selective leaking of classified information,” a death knell to a successful plant campaign. Once it becomes clear that an administration has disclosed any amount of classified information to the press, all leaks become suspect and the administration loses its credibility. The ability to make anonymous statements without the associated risks of doing so is an effective political tool. The example of Bush’s leaking debacle shows us that constructive ambiguity serves as a prerequisite to the use of that tool. Pozen notes that the public reaction to the Obama administration’s zealous use of the Espionage Act further confirms the notion that a particularized level of enforcement is critical to effective planting.
“President Obama is currently learning this lesson—and validating the theory. By increasing criminal enforcement against suspected leakers, his DOJ has “exposed the White House to accusations…that it clamps down on whistleblowing when the disclosures undermine its agenda but eagerly volunteers anonymous ‘senior administration officials’ for interviews when politically expedient.” Critics in the media, civil society, and Congress now scrutinize every such unattributed disclosure for signs of hypocrisy, national security harm, or other grounds for de-legitimization. It took only a handful of prosecutions to erode the Administration’s credibility in the eyes of important constituencies and, with it, the capacity to shape the policy narrative.”
While constructive ambiguity is a highly complex problem for enforcement, ensuring a manufactured scarcity is a matter of straightforward public signaling in an effort to affirm the criminality of leaking as a whole. The weight of the sentencing proscribed by the Espionage Act successfully portrays leaks of classified information as inherently damaging to national security and the public good. In addition to this, signaling is intended to send a clear message that anyone seeking to leak information can potentially face a ruinous array of consequences. As John Kiriakou—one of the eight individuals who have faced Espionage Act charges under President Obama—remarked in an Op-Ed written for the Guardian, his case was meant to send a clear and simple message to potential whistleblowers and leakers: “Challenge us and we will destroy you.” Similarly, Jeff Scudder remarked, “quickly you can walk up the line,” but “if you rock that boat any more, your career is going to be over.” Even when the government’s case against Lawrence Franklin lost momentum after the related charges against Rosen and Weissman were dropped, Judge Ellis made sure that his decision still conveyed the proper public signals.
“So, what I do today—what has happened to Mr. Franklin and what will happen after I rule today—has to stand as a beacon to government officials, because Mr. Hammerstrom is absolutely right, it is important that government officials, more than anyone else, get this message: You cannot engage in disclosure of classified information, certainly not NDI…you are precluded by your agreement with the government and by internal regulations from disclosing classified information…and that if you do so, there are consequences; and that noble motives don’t erase the violation. And it’s important that what I do today reflects that.”
Clearly such a message, regardless of whether it is also applicable to plants—although it likely is not,—drastically raises the stakes of leaking, especially if that leak represents an embarrassment for the executive branch. If perceptions of illegality are the sole determining factor for the existence of a manufactured scarcity, then the impunity of the Espionage Act is certainly a boon to the broader strategy of implementing plants for political purposes. It doesn’t matter how many of these cases get brought up. What seems to be important to Judge Ellis is that there is a clear message sent to the public that leaking classified information is illegal and that individuals who choose to violate leak-laws will essentially have their lives torn apart.
c. Gatekeeper Protections and Plant-Minded Selective Enforcement
With the relationship between planting and leak-law enforcement levels in mind, it becomes readily apparent that affording First Amendment protections to government employees with access to classified information poses a massive threat to manufactured scarcity and a potential boon to constructive ambiguity. Obviously, admitting that there are instances in which the unauthorized disclosure of state secrets can be justified sends a signal entirely contradictory to the notion that leaks are deplorable and criminal in every instance. But the likelihood that judicial review of gatekeeper leaks would be highly sympathetic to the executive branch’s right to keep information secret means that it could potentially validate and legitimize the heavy-handed prosecutions currently employed in the status quo. The punishments won’t change, but the government’s efforts may be deemed more appropriate if the prosecution could prove that a leaker’s actions did not fall within whatever gatekeeper protections were established by the judiciary and/or Congress.
On the other hand, leaving the door open for justifiable leaks would further obscure the source of anonymous disclosures of classified information, potentially improving the power of constructive ambiguity. Even if internal methods of self-checking successfully reduced the overall number of leaks by providing an effective alternative to leaking information to the press, there would be an added layer of legitimacy to plants. This is possible because the public may assume that leaked information was important enough that the source was overwhelmingly confident that it served a public interest and would pass whatever judicial balancing test was used to evaluate a defendant’s actions.
As far as leak-law enforcement levels are concerned, it doesn’t seem that the current strategy of de minimis selective enforcement would need to change in the presence of a specialized gatekeeper First Amendment status. The only difference would be that the prosecutions themselves would require balancing the relevance of leaked information to national security interests more thoroughly with that information’s potential to stimulate public debate about policy. It might be harder for the government to get leak-law indictments to stick, but that does not make it any harder for the executive branch to send a signal to government employees and the public when actionable violations do occur. Actually administering a punishment seems to be less important than the stress that an indictment puts on government employees and contractors—whether it entails raids by militarized FBI teams, suspension of one’s security clearance, administrative retaliation, or the massive costs associated with litigation. Furthermore, if Pozen is correct in stating that the level of leak-law enforcement has little to do with the difficulty of investigating and prosecuting leaks, then making the government’s prosecutorial burden more difficult should have little to no effect on its enforcement strategy in general.
While these predictions about the impact of gatekeeper protections make logical sense, we should remind ourselves not to take them as in any way empirically validated. Pozen points out that “even if we could reliably measure the consequences of the leaks that have occurred, it is impossible to know what outcomes would have followed from a system of slightly more, vastly more, or even paltrier enforcement.” The only means to know how gatekeeper protections might impact enforcement levels, planting, or the overall number of unauthorized disclosures of classified information, is through a process of judicial trial-and-error. Until the courts have established a line of rulings applicable to leak-law violations, the best we can do is understand all of the factors affecting the First Amendment status of national security gatekeepers and pay attention to them if their speech rights are amended in any way.
The importance to the proper functioning of a democracy of facilitating informed public debate cannot be emphasized enough. If the public is unable to hold elected officials accountable as a result of ignorance or misinformation, then we are doing ourselves a great disservice and wholeheartedly deserve an unresponsive and unrepresentative government. In order to ensure that we are educated enough to avoid such an outcome, we have to expect relevant political information (especially when that information points to governmental misconduct) will come to light as it is discovered. The specter of overclassification stands as a threat to such an expectation.
Government employees empowered by specialized First Amendment protections could potentially utilize their position of epistemic authority in relation to classified information to stand as watchdogs in a way that members of the general public cannot. With the actions of these national security gatekeepers closely scrutinized and constrained by the judiciary, we may be able to attain the ideal situation in which security-sensitive information can be kept private—but evidence of governmental misconduct can no longer hide behind a classification stamp. Judicial balancing offers a flexible and nuanced method of addressing overly extensive state secrecy without calling the validity of executive privilege as a whole into question. As individuals in a unique position of access to a wealth of politically relevant information that is off-limits to the public, the First Amendment status of national security gatekeepers is an open question. The answer to that question has the potential to facilitate the jurisprudential momentum necessary to place effective checks on overclassification and governmental misconduct in the realms of national security and intelligence policy.
As Kitrosser expertly notes, “if discussing government policy is core political speech, that status does not magically change whenever the policy is labeled classified.”
Michael Barclay prepared this paper during and after an internship in 2014 at the Newseum Institute, at the Newseum in Washington, D.C. He attends Pitzer College, Claremont, Calif., where he is studying for a degree in International Political Economy.
 Heidi Kitrosser, Classified Information Leaks and Free Speech, 3 U. Ill. L. Rev. 881, 906-08 (2008) [hereinafter, Kitrosser, Classified Information Leaks and Free Speech]; Heidi Kitrosser, Free Speech Aboard the Leaky Ship of State: Calibrating First Amendment Protection for Leakers of Classified Information, 2 Journal of National Security Law & Policy 409, 422 (2012) [hereinafter, Kitrosser, Free Speech Aboard the Leaky Ship of State] (“Protecting speech that helps to facilitate and check self-government indeed is a central purpose of the free speech and free press clauses from the perspectives of constitutional structure and principle more broadly”).
 See Abrams v. U.S., 250 U.S. 616, at 2 (1919) ( “…the ultimate good desired is better reached by free trade in ideas — that the best test of truth is the power of the thought to get itself accepted in the competition of the market, and that truth is the only ground upon which their wishes safely can be carried out.”); see also Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 914.
 Detroit Free Press v. Ashcroft, 303 F.3d 681, 683 (6th Cir. 2002) (quoting Kleindenst v. Mandel, 408 U.S. 753, 773 (1972)).
 see Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 920. See also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 423 (detailing the ways in which free speech and free press clauses work together to place a check on Executive Branch secrecy).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 917-18.
 Id. at 918.
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 422.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 916-26; see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 422.
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 429.
 See David Pozen, The Leaky Leviathan: Why the Government Condemns and Condones Unlawful Disclosures of Information, 127 Harv. L. Rev. 512, 515, 517, 559-65 (2013); see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 411, 419, 423.
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 429.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 923; see also, Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 426.
 See, Kitrosser, Classified Information Leaks and Free Speech, supra note 1 (explaining the legal particularities involved with an increase in First Amendment protections for government employees with access to classified information more in-depth); see also, Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 424 (“The First Amendment’s promise would be empty indeed if its protections did not extend to information that the President wishes to keep secret. This includes information from government insiders, who alone are structurally situated to reveal it.”).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 885.
 Id. at 926.
 See Pozen, supra note 12, at 522-28 (listing all legal statutes applicable to those who mishandle classified information); see Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 415-21 (detailing court cases related to leak-law violation); see also Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 897-904 (detailing the ways in which the strength equals prerogative approach to classified information leaks was applied to New York Times v. United States, United States v. Morison, and United States v. Rosen and Weissman).
 Greg Miller, CIA Employee’s Quest to Release Information ‘Destroyed My Entire Career’, Washington Post, July 4, 2014, http://www.washingtonpost.com/world/national-security/cia-employees-quest-to-release-information-destroyed-my-entire-career/2014/07/04/e95f7802-0209-11e4-8572-4b1b969b6322_story.html.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 926; see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 423, 440-46.
 Memorandum on Classified Information and Controlled Unclassified Information, 74 Fed. Reg. 26277 (May 27, 2009).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 906-08 (detailing the constitutional grounds for considering information sharing as complementary to First Amendment protections of political speech. As Kitrosser notes, “The principle that speech about government is of core First Amendment value extends to information sharing as well as opinion sharing. The right to express viewpoints would mean little if government could stifle the exchange of facts underlying such viewpoints. The Supreme Court has embraced this logic many times.”).
 S. Doc. No. 105-2, at XXI (1997).
 Mary-Rose Papandrea, Lapdogs, Watchdogs, and Scapegoats: The Press and National Security Information, 83 Ind. L.J. 233-306, 243 (2008).
 Info. Sec. Oversight Office, Ann. Rep. 1, 3-5 (2013).
 Id. at 2.
 Ctr. for Dev. of Sec. Excellence, Original Classification Authority Desktop Reference, 1, 1 (2012).
 Info. Sec. Oversight Office, supra note 25, at 5.
 Id. at 1
 S. Doc. No. 105-2, at XXII (1997) (noting that the exact number of individuals with derivative classification authority is not recorded because of its constantly fluctuating nature, but also noting that because of the unusually privileged access to government employees and statistics afforded to this congressional investigation, it determined a rough estimate of this number).
 Faddis, Charles and Martin, Rachel, Has Sept. 11 Changed How Agencies Share Secrets?, NPR. (September 6, 2011) available at http://www.npr.org/2011/09/06/140206911/intelligence-community-still-not-keen-on-sharing-secrets.
 Info. Sec. Oversight Office, supra note 25, at 3.
 Id. at 1.
 Id. at 7.
 National Security Secrecy at the Literal Expense of Openness, OpenTheGovernment.org (Sept. 28, 2011), http://www.openthegovernment.org/node/3235.
 Info. Sec. Oversight Office, supra note 25, at 7.
 Id. at 7.
 Id. at 7.
 Id. at 7.
 Id. at 7.
 Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 Info. Sec. Oversight Office, supra note 25, at 18-9.
 Id. at 19.
 Mark-up of Fiscal Year 1994 Foreign Relations Authorization Act: Hearing Before the Subcomm. on Terrorism, Narcotics and International Operations of the S. Comm. On Foreign Relations, 103rd Cong. 32 (1993).
 S. Doc. No. 105-2, at XXI (1997).
 Pozen, supra note 12, at 515, 517, 559-65; see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 411, 419, 423 (detailing the ways in which classified information is leaked to the public by the Executive Branch for political gain).
 See Pozen, supra note 12, at 559-65 (discussing more in-depth the various politically productive uses of classified information).
 Id. at 515.
 See id. at 559-72 (explaining the ways in which plants differ from leaks).
 Id. at 521.
 Id. at 559.
 Id. at 529 (“Journalists and government insiders have consistently attested that leaking is far more common among those in leadership positions. The ship of state, one often hears, is the only known vessel that leaks from the top80 — starting, that is, from the White House itself”).
 Id. at 533-34 (“High-level officials do not need to turn over official documents to get heard; in the absence of any tangible disclosure, journalists and editors are much more likely to find their statements to be credible and news- worthy. And as the discussion here suggests, high-level officials do not tend to see themselves as whistleblowers on a mission to expose abuse or as dissidents on a large scale”).
 Id. at 534 (“I am unaware of any senior U.S. policymaker who has ever been accused of unlawfully revealing thousands of pages or their equivalent. General leaks are the province of the radically disaffected and the subversive. Top government brass, socialized into and successful in the Washington power culture, are unlikely to be either”).
 Pozen, supra note 12, at 233.
  S. Doc. No. 105-2, at 8 (1997).
 Pozen, supra note 12, at 546.
 Id. at 562.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 904.
 Id. at 916-26; see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 422 (detailing the differences between structural and rights-based checks on executive secrecy).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 921 (“Professor Bickel made this point when he described first amendment doctrine as creating an “adversary game between the press and government” which is analogous to the constitutional system of separation and balance of powers among the institutions of government”).
 Id. at 896.
 Id. at 896.
 Id. at 896-97. (“Another indirect form [of the strength equals prerogative argument] is the argument that only the political branches—particularly the President—are capable of understanding whether and when information leaks should be punished. This argument equates the capacities of the political branches—particularly the President—to designate national security secrets with a singular ability, and hence conclusive authority, to do the same”).
 Papandrea, supra note 24, at 244.
 Id. at 244.
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 410 (“Of the commentary that exists, the common view is that publishers must be strongly protected under the First Amendment, while leakers can be punished with little or no constitutional difficulty. The distinction is said to turn largely on the fact that persons with authorized access to classified information are in special positions of trust and thus have effectively waived protections against prosecution for leaks”); Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 903 (“Citing Morison, as well as cases approving contractually driven prior restraints on publication by former government employees, the court concluded that ‘Congress may constitutionally subject to criminal prosecution anyone who exploits a position of trust to obtain and disclose [national defense information] to one not entitled to receive it’”).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 926.
 Id. at 885 (“Third, the vast range of information classified makes it largely inevitable that leaks often will provide information about vital public policy issues. It is entirely antithetical to First Amendment doctrine and theory for such information exchanges to be made illegal by little more than the wielding of a classification stamp”).
 see Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 440 (detailing the reasons why government employees with access to classified information occupy a position in which they are uniquely suited to provide the public with politically useful information that would otherwise be obscured by overclassification.).
In an in-person interview conducted for the purposes of this article, Jeffrey Scudder noted that he thinks that government employees would have the contextual understanding and capacity necessary to make reasonable decisions about when classified information leaks are justifiable. If national security gatekeepers were to be afforded some degree of First Amendment protection in such instances, Scudder believes that we “wouldn’t see [leaks] often… I don’t think this would happen very often. Mostly, in day-to-day work, it’s not a controversy. But when you have these situations where, like the last one was with the Iraq War, where the government’s making policy and you have analysts saying ‘I’m looking at the data here and this is not true. This is not true.’ If you look at the Pentagon papers, the famous one where it’s like wait a minute, why doesn’t the public know about this? And the bad part is, usually a situation like that, it’s embarrassing to the administration or the government agency and then they’ll move holy hell to keep that from coming out. And that is where you’ve got, you know how do you prevent that from happening, to let the employee—because in hindsight, most of the times when things like this have happened, it was actually probably good that these things had came out. It was controversial at the time, but in hindsight, nobody really holds…that Ellsberg was a traitor.” Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 Mr. Scudder was optimistic about the prospect of judicial balancing when conventional means of addressing overclassification (FOIA Requests, internal reviews, etc.). “You bring up something really interesting—that I, as someone working in the intelligence community, would actually have a venue to say hey courts. That would be something, if that happened, because right now you don’t have that ability to do that. And the only way is to try to force a FOIA request which as you see the government does everything it can to make sure employees know not to do that. That would be interesting. And you look at some of the leaks, its because people have been frustrated. There was a woman who worked for the inspector general’s office who leaked about the renditions and the detainees. You keep hearing the same stories where she saw this, she thought it was wrong, she tried to raise it up and it was ‘okay we’ll look at it. No we’re not gonna do anything about it.’ And finally, she went to the press because she felt like she had no other option” Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 925, 929.
 Id. at 921 (“Professor Bickel made this point when he described first amendment doctrine as creating an ‘adversary game between the press and government’ which is analogous to the constitutional system of separation and balance of powers among the institutions of government”).
 John Kiriakou, Obama’s abuse of the Espionage Act is modern-day McCarthyism, The Guardian, Aug. 6, 2013, http://www.theguardian.com/commentisfree/2013/aug/06/obama-abuse-espionage-act-mccarthyism; See also Vladeck, Stephen I. “Inchoate Liability and the Espionage Act: The Statutory Framework and the Freedom of the Press.” Harvard Law and Policy Review 1, no. 1 (Winter 2007): 236 (“As I suggest below, the application of theories of inchoate liability to the ambiguous language of the Espionage Act might render members of the media subject to criminal liability for acts of newsgathering wholly separate from acts of publication, and therefore less likely to fall within the umbrella of the press protections enmeshed within the First Amendment”).
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 410 (“Of the commentary that exists, the common view is that publishers must be strongly protected under the First Amendment, while leakers can be punished with little or no constitutional difficulty); see also Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 907 (detailing that the Grosjean Court explained that “the predominant purpose of [the free press clause] was to preserve an untrammeled press as a vital source of public information… [I]nformed public opinion is the most potent of all restraints upon misgovernment”).
 Pozen, supra note 12, at 609 (“And as just discussed, the media cannot reliably ascertain which disclosures fall on the unauthorized side of the line; the executive has devised a system that keeps this line vague and discourages reporters from looking into questions of authorization too closely. The resulting ambiguity attenuates the relationship between the crime of leaking national defense information and the practice of researching and writing news stories thereon. In so doing, it renders the Espionage Act overbroad with respect to national security journalism under the logic of cases like Stevens. A law that criminalized only the publication of exceptionally sensitive information received from demonstrably or manifestly unauthorized sources, and that defined these terms, would have a better chance of meeting the “intrinsically related” standard. Because such a law would be contrary to the interests of those who run the government,433 it would also presumably never be enacted”).
 Vladeck, Stephen I. “Inchoate Liability and the Espionage Act: The Statutory Framework and the Freedom of the Press.” Harvard Law and Policy Review 1, no. 1 (Winter 2007): 236 (“For if the press is to be so broadly liable for any newsgathering that might in any way violate the Espionage Act, the chilling effect on speech generally, and on the importance of public debate in a free society specifically, will be manifest”).
 Pozen, supra note 12, at 527.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 900.
 Lilly Maier, Better protection for most, but not the intelligence community, Politifact, Nov. 21, 2013, http://www.politifact.com/truth-o-meter/promises/obameter/promise/426/increase-protections-for-whistleblowers/.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 908.
 Eugene Volokh, Speech as Conduct: Generally Applicable Laws, Illegal Courses of Conduct, “Situation-Altering Utterances,” and the Uncharted Zones, 90 Cornell L. Rev. 1277, 1286, 1311–26, 1336–39, 1347–48 (2005).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 886.
 Id. at 903 (“Citing Morison, as well as cases approving contractually driven prior restraints on publication by former government employees, the court concluded that ‘Congress may constitutionally subject to criminal prosecution anyone who exploits a position of trust to obtain and disclose [national defense information] to one not entitled to receive it’”); see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 410 (“Of the commentary that exists, the common view is that publishers must be strongly protected under the First Amendment, while leakers can be punished with little or no constitutional difficulty.7 The distinction is said to turn largely on the fact that persons with authorized access to classified information are in special positions of trust and thus have effectively waived protections against prosecution for leaks”).
 Snepp v. United States, 444 U.S. 507, at 4 (1980).
 The notion of the transmission of classified information as a breach of conduct (and not as speech) figured prominently in United States v. Morison, United States v. Franklin, and United States v. Rosen and Weissman. see Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 899-904, 905-08; see also Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 419-21 (analyzing case law as it relates to the strength equals prerogative paradigm).
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 896.
 Transcript of Motion Hearing at 33, United States v. Franklin, 171 F. Supp. 2d 1319 (E.D. Va.) (1:05-cr-225, 1:05-cr-421), available at http://fas.org/sgp/jud/aipac/franklin061109.html.  Id. at 34.
 See, e.g., Jon Greenberg, Greenwald: NSA leaker Snowden has no whistleblower protection, Politifact, Jan. 7, 2014 (“[R]ecent trials under the Espionage Act make it clear that Snowden could expect no guarantee that he would be able to tell a jury why his violations might serve the country’s national interests. What he and his lawyers could present might be tightly constrained. ‘The executive branch has asserted the sole authority to determine what remains a secret,’ McClellan said. ‘The way these Espionage Act prosecutions have gone, it has not been fair to the defendants. It has not followed what you would expect in a traditional trial’”).
Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 906.
 see Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 429 (“Given the twin realities of massive over-classification and widespread selective leaks from the top, a broad executive discretion to prosecute classified information leaks is a powerful means for the executive—both through actual prosecutions and through fear of the same—to manipulate information flow. Such discretion can generate a deeply slanted chilling effect. Those who leak information that paints an administration in a bad light have much to fear in an environment where prosecutions occur or are threatened regularly. These realities complement the theoretical problems of any scheme that enables the executive to qualify First Amendment protections for speech about itself. Such a scheme is profoundly in tension with a constitutional structure designed to ensure executive accountability and to do so partly through free speech and a free press”).
 Kitrosser, Free Speech Aboard the Leaky Ship of State, supra note 1, at 441 (suggesting a three-part calibration of leaker protections and punishments: “First, classified leaks should be treated like other government employee speech when employment sanctions such as dismissal or demotion are at issue. Under current doctrine, this means that the Pickering balancing test should apply when employees are disciplined for leaks. Second, the government must meet a considerably higher threshold to impose criminal or civil, rather than employment-based administrative penalties. Third, courts should consider varying the precise nature of the government’s burden with the severity of the penalty sought in the criminal or civil context”).; see id. at 438-445 (discussing in full the legal specificities of potential gatekeeper protections).
 United States v. Rosen, 557 F.3d 192, at 6 (4th Cir. 2009).
 Id. at 6.
 Id. at 6.
 CJ Ciaramella, Cass Sunstein and the Status Quo, Freedom of the Press Foundation, July 11, 2014, https://pressfreedomfoundation.org/blog/2014/07/cass-sunstein-and-status-quo; Patrick Leahy & John Cornyn, FOIA Improvement Act of 2014 Section-By-Section Analysis, https://www.leahy.senate.gov/download/foia-section-by-section-final (noting that the public balancing test included in the FOIA Improvement Act of 2014 “amends subsection (b)(5) of 5 U.S.C. §552—which exempts inter- and intra-agency documents that would be exempt from discovery in civil or criminal litigation—to include a public-interest balancing test. In the case of deliberative process privilege and attorney work-product privilege, the test would mandate disclosure when a public interest in disclosure outweighed the agency’s interest in protecting the records or information. In the case of attorney-client privilege, a compelling public interest would need to outweigh the agency’s interest in nondisclosure. Additionally, this section would sunset the application of Exemption 5 to documents created more than 25 years ago”).
 Jeff Scudder acknowledged the likelihood of an increase in gatekeeper protections leading to an improvement of internal channels to confront misconduct and over-classification when he noted the way in which the courts could act as an advocate for those attempting to challenge the insulated, opaque structure of intelligence and defense agencies. In his opinion, the detached and mission-specific nature of the EEOC made it particularly effective at taking things like discrimination out of the agency’s chain of command when it was necessary to do so. Logically, the court system could serve a similar function in the context of classified information.
Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 See Maier, supra note 81.
 Pozen, supra note 12, at 577, 607.
 Jeff Scudder spoke at length about the ways in which the CIA’s response to his FOIA request was a signal to employees who may wish to challenge the chain of command in any way. “I am a great case for them because they can hold it up and say look at this. Jeff wanted to rock the boat and you know what, one day he wasn’t at work anymore and all this stuff happened to him. So feel free, even though he wasn’t prosecuted, feel free to do this, but it’s a good testament to you don’t want to rock the boat.” He went on to make a connection between making examples out of individuals, like himself, who challenge agency policy and an enforced culture of secrecy at the CIA. Q: “Why would the agency go after you so aggressively if you were technically doing things by the book?” A: “Just the fact that I dared to do it. How dare you do a FOIA request for this. You tried internally and we squashed it and yet you went out and used the law to force us to do this so you know the beast has to rise up. We’ve got to take him out.”’ Q: “Is there a culture of secrecy in the intelligence community?” A: “If you ever say well actually this isn’t secret. There’s nothing—no, its top and bottom, its secret. We’re all raised that way. It’s engrained. Because people like Jeff Scudder said it wasn’t and look what happened to him.”
Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 Pozen, supra note 12, at 562 (“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”).
 Id. at 578 (“Most obviously, lax enforcement allows leaks to be used more freely. A larger volume of administrative samizdat circulates. Less obviously, the backdrop of formal illegality remains relevant, because in depressing the overall amount of leaking it enhances the significance of the disclosures that occur. The Espionage Act casts a shadow even though it is rarely enforced. The government official who discusses classified information with a reporter still assumes some modest amount of legal and professional risk for her actions. That she is willing to do so communicates to her listeners that she or her agency feels she has something notable to say…[T]he paradox here is that leaking must remain illicit or else the speakers’ statements would not necessarily reach their intended audiences or convey the intensity of their preferences, and so would fail to deliver the communicative benefits the executive branch has come to expect from them. On this account, the laws against leaking can be seen as a form of manufactured scarcity which, by raising the cost of leaks, also raises their potential salience and signaling power when they occur. The illegality of leaking enhances its functionality”).
 See supra note 108.
 Id. at 536.
 Id. at 571 (“Formal leak investigations risk exposing top officials’ efforts to manipulate the secrecy rules and marginalize opponents, among other machinations. Criminal cases against quasiauthorized sources risk provoking unfavorable judicial and legislative responses, constraining future flexibility. And all public sanctions risk exacerbating internal conflict as well as underscoring it”); Id. at 572 (“The rare undeniable leak like Chelsea Manning’s can be investigated and punished with only modest externalities for the executive’s broader functioning. Any thoroughgoing effort to separate out the unauthorized from the authorized disclosures, however, would run up against an elaborate web of bureaucratic customs and conventions on which the executive has come to rely — the government by plant, pleak, and leak bequeathed by the New Deal. Because “leaking” is such an ambiguous practice and leak investigations are so difficult to cabin once set in motion, all components of the executive branch that strategically reveal protected information through the press have a shared interest in keeping enforcement levels low”).
 Id. at 536.
 Id. at 565.
 See id. at 548-554 (combating the “constraint-based approach” to an explanation of “de minimis criminal enforcement” of leak-laws by responding to the notion that it is difficult to identify leakers and conduct investigations about their actions and the notion that it is difficult for the government to argue leak-law cases in court without being forced to publicize sensitive information).
Jon Greenberg, Greenwald: NSA leaker Snowden has no whistleblower protection, Politifact, Jan. 7, 2014, http://www.politifact.com/punditfact/statements/2014/jan/07/glenn-greenwald/greenwald-nsa-leaker-snowden-has-no-whistleblower-/.
 Pozen, supra note 12, at 558 (“At a minimum, the idea that the leak laws are inherently or peculiarly incapable of implementation is untenable. Energetic enforcement just comes at a cost — political, practical, personal — that relevant decision makers have been unwilling to accept”).
 Id. at 536.
 Id. at 550-51 (“Nonetheless, an identification rate of more than fifty percent over a five-year period suggests that many leakers can be discovered without extraordinary measures. Furthermore, in a significant portion of the investigations that failed to yield a suspect, other components of the executive branch appear to have borne some fault. Through a 2006 FOIA request, journalist Josh Gerstein obtained FBI records showing that the Bureau had abandoned numerous criminal investigations into classified information leaks because the “victim agency” had failed to cooperate with its work”).
 Id. at 575 (“If their sociological premises are correct, these theories suggest that any enforcement regime that is seen as harsh on leakers—whistleblowers, they will be called—or that has the effect of exposing the administration’s media machinations, represents a profound threat to executive power”); See also id. at 550 (“President Reagan went further than any other President in trying to implement these policies. No recognized constitutional obstacle stood in his way. Yet after journalists protested, senior officials rebelled from within; Secretary of State George Shultz held a press conference to announce that no one at the State Department would submit to a polygraph. The attempted crackdown on leakers lasted all of three weeks before being summarily scrapped.191 Administrations and agencies have experimented to varying degrees with a number of other initiatives over the years, but Reagan’s plan remains the closest the executive has come, at least in the pre-WikiLeaks era, to establishing a comprehensive program for monitoring prospective leakers, ferreting out culprits, or constraining media contacts”).
 Steven Aftergood, A Look Behind President Clinton’s Veto of an Anti-Leak Bill, Federation of American Scientists, July 22, 2014, http://fas.org/blogs/secrecy/2014/07/anti-leak-veto/.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 884.
 Jon Greenberg, CNN’s Tapper: Obama has used Espionage Act more than all previous administrations, Politifact, Jan. 10, 2014, http://www.politifact.com/punditfact/statements/2014/jan/10/jake-tapper/cnns-tapper-obama-has-used-espionage-act-more-all-/; Aubrey Bloomfield, 8 Whistleblowers Charged With Violating the Espionage Act Under Obama, PolicyMic, June 23, 2013, http://www.policymic.com/articles/50459/8-whistleblowers-charged-with-violating-the-espionage-act-under-obama.
 Pozen, supra note 12, at 559-79.
 Id. at 585.
 David Johnston & David E. Sanger, In Court Filings, Cheney Aide Says Bush Approved Leak, N.Y. Times, Apr. 6, 2006, http://www.nytimes.com/2006/04/06/washington/06cnd-leak.html?pagewanted=print&_r=0.
 See Pozen, supra note 12, at 585 (discussing the effect that an exposure of leaks of classified information regarding the invasion of Iraq as such delegitimized the George W. Bush administration’s political goals).
 Id. at 559-79.
 Id. at 585
 John Kiriakou, Obama’s abuse of the Espionage Act is modern-day McCarthyism, The Guardian,Aug. 6, 2013, http://www.theguardian.com/commentisfree/2013/aug/06/obama-abuse-espionage-act-mccarthyism.
 Scudder also noted, in reference to his own case with the CIA, that the agency “just want[ed] to protect the system” and sought to punish him for his FOIA request simply because of “the fact that [he] dared to [submit] it.” He continued, “[h]ow dare you do a FOIA request for this. You tried internally and we squashed it. Yet you went out and used the law to force us to do this, so…the beast has to rise up. We’ve got to take him out.” Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).see also supra note 106.
 Transcript of Motion Hearing at 37, United States v. Franklin, 171 F. Supp. 2d 1319 (E.D. Va.) (1:05-cr-225, 1:05-cr-421), available at http://fas.org/sgp/jud/aipac/franklin061109.html.
 See supra note 106.
 See supra note 106; see also supra note 129 at pincite.
 See supra note 106 (explaining the importance of a culture of secrecy at government agencies).
 In references to the court proceedings related to the FOIA Request that he submitted, Jeff Scudder remarked, “judges will always defer to the integrity of the government, and to their expertise. So the government would make these statements, and I’d read the filings going that is ridiculous… the judge deferred to [the government] saying ‘okay, this must be the case because you’ve said it was so.’ And [the government] lied—That really angered me. I’m not a legal scholar, but I don’t see how you can make statements that are false.” Interview with Jeffrey Scudder, former Program Manager at the CIA and currently a Director at Qlarion, Inc., in Reston, VA (Aug. 10, 2014).
 See supra note 96 (detailing examples of the types of balancing tests that courts could apply to cases involving national security gatekeepers).
 Pozen, supra note 12, at 572 (“The rare undeniable leak like Chelsea Manning’s can be investigated and punished with only modest externalities for the executive’s broader functioning. Any thoroughgoing effort to separate out the unauthorized from the authorized disclosures, however, would run up against an elaborate web of bureaucratic customs and conventions on which the executive has come to rely — the government by plant, pleak, and leak bequeathed by the New Deal. Because “leaking” is such an ambiguous practice and leak investigations are so difficult to cabin once set in motion, all components of the executive branch that strategically reveal protected information through the press have a shared interest in keeping enforcement levels low”).
 See supra note 114, at 548.
 Pozen, supra note 12, at 544.
 Kitrosser, Classified Information Leaks and Free Speech, supra note 1, at 908.