Should We Trust Leakers?
Anonymous Sources and the Problem of Regulation
In chapter 5 I argued that though an official can be justified in making an unauthorized disclosure when she encounters an abuse of state secrecy, she is not obliged to make such a disclosure if doing so will expose her to significant retaliation, and that even when an official is obliged to make such a disclosure, it will take great courage for her to voice her concerns in spite of the prospect of retaliation. The implication of this analysis is that we should expect officials to continue to raise the alarm by leaking classified information rather than by blowing the whistle. This state of affairs is troubling because, as we have seeen, an official can take advantage of anonymity to make selective disclosures that favor sectional interests or partisan causes. Consequently, we need to examine whether the practice of leaking can be justified, and if so, how we might prevent its misuse. But first we must investigate whether it really is possible for leakers to evade detection.
Why Leaks Persist
Given the threat of retaliation, officials who wish to disclose classified information are often advised to leak the incriminating information rather than lodge a formal complaint. In the words of The Art of Anonymous Activism, a prominent how-to guide, anonymous disclosures provide a means of “serving the public while surviving public service.”1 But why should an official who discloses classified information anonymously be more confident of evading retaliation? One might reasonably assume that a president could use the powers granted by the laws that prohibit unauthorized disclosures to hunt down and punish the source of an embarrassing disclosure, and that anonymity can therefore provide officials with little more than temporary cover. But the evidence belies this assumption, as only a handful of officials have ever been prosecuted for leaking classified information. Why have successive administrations allowed leaks of classified information to go unpunished even as they have come down hard on whistleblowers like Richard Barlow, Sibel Edmonds, and Coleen Rowley (whom we met in chapter 5)?
The dearth of prosecutions for leaking is typically attributed to the absence of a blanket statutory prohibition against unauthorized disclosures akin to the United Kingdom’s Official Secrets Act.2 The Espionage Act requires the prosecution to show that an official responsible for “willfully” disclosing “information relating to the national defense” to a person who is “not entitled to receive it” had “reason to believe” that such information “could be used to the injury of the United States or to the advantage of any foreign nation.”3 It is widely accepted that this so-called bad faith requirement of the Espionage Act works to the advantage of defendants because often it is difficult for prosecutors to establish that an official had “reason to believe” that the information revealed by his anonymous disclosure could be used “to the injury of the United States.”4 This is especially true when an unauthorized disclosure reveals a prima facie violation of the law. In such cases, prosecutors are confronted with the very real possibility that a jury will decline to convict an official who contends that the violation gave her “reason to believe” that its disclosure would be in the public interest.
Prosecutors are also hindered by the notoriously imprecise language of the Espionage Act. For example, what does it mean to “willfully” communicate information? How do we determine who is a “person not entitled to receive” information relating to the national defense when the classification system does not “entitle” anyone? And what about diplomatic and intelligence secrets that cannot be easily subsumed under the heading of “information relating to the national defense?”5 Although the courts have made attempts to pin down the meaning of these phrases (or at least to narrow the range of possible interpretations), the absence of settled guidelines has prompted prosecutors to tread carefully, lest an unfavorable interpretation or decision weaken the deterrent effect of the Espionage Act.
The executive has not stood idly by in the face of challenges to its authority over classified information. On the contrary, presidents have periodically urged Congress to revise the Espionage Act and to revisit the Classified Information Procedures Act, a statute that is meant to prevent defendants from engaging in the practice of “graymail” (that is, requesting access to classified information on the grounds that this information is needed to mount a defense).6 However, these calls have had little effect because lawmakers have long been divided over whether and how to address the practice of leaking. These divisions can be traced to differences among lawmakers over the wisdom of strengthening the president’s control over the national security apparatus. Thus, for the foreseeable future at least, the threat of criminal prosecution is likely to remain—in the words of the widely cited report of the Interdepartmental Group on Unauthorized Disclosures of Classified Information (commonly referred to as the Willard Report)—“so illusory as to constitute no real deterrent to the prospective leaker.”7
Besides, even if the president is able to convince Congress to revise the Espionage Act, it is unlikely that this step alone will turn the tide. Consider that successive administrations have failed to enforce a statute like §798(a), which clearly prohibits the disclosure of classified information relating to communications intelligence.8 It is notable that the United States has thus far secured only a single conviction under this statute; in that case, moreover, the official in question, Shamai Leibowitz, an FBI linguist who had disclosed the transcripts of a counterintelligence wiretap to an Internet blogger, conceded his guilt as part of a plea bargain.9 This record suggests that the dearth of prosecutions for leaking owes to factors that extend beyond the shortcomings of the Espionage Act.
What might these factors be? Political constraints constitute one such factor. It can often prove embarrassing or divisive to prosecute public officials, particularly members of Congress. For example, in 2004 the FBI identified Senator Richard Shelby as the person who disclosed to the media that the NSA had intercepted two Arabic-language messages on the eve of September 11, 2001: “The match is about to begin” and “Tomorrow is zero hour.”10 But Senator Shelby was never charged under §798(a) even though his disclosure revealed what President Bush described as “alarmingly specific” information.11
Another important factor contributing to the dearth of prosecutions is, paradoxically, a concern for national security. On numerous occasions the government has chosen not to prosecute an official responsible for leaking classified information for fear that such a move would be perceived as confirming the validity of what has been leaked and might even lead to the revelation of additional sensitive information. A recent example is the case of the CIA official Mary McCarthy, who reportedly disclosed to the Washington Post highly classified information pertaining to the CIA’s use of so-called black sites (or secret prisons) during the war on terror. Although the CIA reportedly uncovered McCarthy’s role after she failed a lie-detector test conducted as part of an effort to trace the source of the Post story, it chose to fire her rather than press criminal charges, because a trial would likely “wind up airing sensitive information as well as policy dissents” over its handling of detainees in the war on terror.12
Of course presidents do not have to rely on criminal laws alone in order to punish disobedient employees. The “more promising approach,” the Willard Report notes, is to utilize administrative sanctions, because “for most government employees, a realistic prospect of being demoted or fired for leaking classified information would serve as a deterrent.”13 These sanctions should certainly not be taken lightly, since presidents have shown a willingness to use them—as seen in the McCarthy case. Nonetheless, the fact that leaks continue to appear on the front pages of the New York Times and the Washington Post suggests that a sizable number of officials are no more deterred by the specter of administrative sanctions than they are by the threat of criminal charges. This outcome can be attributed to two factors. The first is sheer defiance. Officials who decide to leak classified information are sometimes so troubled by the wrongdoing they have discovered that they feel they must share it with the public—even at the risk of being discovered and punished.14 This sort of defiance is most likely to make an appearance when the relevant official is already on the verge of retirement or has a viable career alternative (as was the case with McCarthy, who was identified and then dismissed a few days before she was due to retire to start a new career as a public interest lawyer).
The other, more deep-seated, explanation for the limited deterrent effect of administrative sanctions lies in the fact that investigators are frequently unable to identify the source of an anonymous disclosure. This difficulty was publicly acknowledged by the Willard Report, which observed in 1982 that the FBI had so far “rarely been successful in identifying the sources of such disclosures.”15 Though the Willard Report urged the president to make “better efforts to identify leakers,” little seems to have changed in the intervening decades: the Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction concluded in 2005 that “the greatest barrier” the president faces in punishing leaks still lies “in identifying the leaker.”16
What explains this continuing inability to hunt down the officials responsible for leaking classified information? In part, the answer lies in the size and complexity of the national security bureaucracy, where even highly classified information is often accessed by dozens and sometimes hundreds of individuals.17 This makes it impractical for the FBI to repeatedly undertake the exhaustive investigations needed to identify the individual responsible for a leak of classified information. Indeed, this obstacle is now thought to be so substantial that it has become de rigueur to conclude that “it is generally fruitless to try to discover the source of a leak.”18 The available statistics certainly support this conclusion. For instance, between 2005 and 2009 nearly 200 leaks were referred to the FBI, but “investigators opened only 26 of these cases, identified 14 suspects, and prosecuted none of them.”19
Although such statistics speak volumes about the obstacles that the Justice Department faces in identifying the officials responsible for leaking classified information, we should not assume that these obstacles are wholly insurmountable. It is important to remember that the Justice Department’s poor track record in tracing leakers owes substantially to internal guidelines directing prosecutors to avoid compelling reporters to reveal their sources. Should a future administration discard these guidelines, the odds would almost certainly shift in favor of prosecutors. Nor is the cooperation of reporters always essential to prosecutors. The widely reported successes that the Bush and Obama administrations have recently had in investigating a number of high-profile leaks, including those by McCarthy, Leibowitz, Thomas Tamm, Jeffrey Sterling, Thomas Drake, Bradley Manning, Stephen Kim, and John Kiriakou, provide a sobering reminder that it is not impossible for determined investigators to hunt down the source of a leak even in the absence of cooperation from reporters.20 And, as the Justice Department informed Congress in 2002, there is reason to think that improvements in information technology, especially in the areas of digital content management and forensics, will further improve the FBI’s chances of detecting the officials responsible for leaks.21
These possibilities notwithstanding, there is little reason to believe that future administrations will be able to clamp down on leaks in a comprehensive manner. The evidence suggests that stubbornness and shrewdness greatly constrain how far any administration can police its employees. Consider, for example, the aftermath of Dana Priest’s report in the Washington Post identifying the locations and operations of the CIA’s secret prisons, and revealing dissent within the CIA over the legality and morality of utilizing extreme interrogation methods.22 Priest’s report attracted strong criticism from Porter Goss, then director of the CIA, who testified before Congress that it had caused “very severe” damage to relations with the foreign intelligence agencies that had helped the CIA establish these prisons.23 Yet, in spite of Goss’s assurances, the CIA’s efforts to track down Priest’s sources met with only limited success. While the CIA was able to ascertain that McCarthy had had “unauthorized conversations” with Priest, both McCarthy and the CIA have since claimed that her contact with Priest was not related to the Post story.24 Meanwhile Priest has attributed her report to “multiple current and former intelligence officials on three continents,” a claim suggesting that even if McCarthy was in fact a source, she was only one of a number of sources, the bulk of whom have apparently gone undetected.25
Questionable Uses of Anonymity
We have now established that an official who wants to make an unauthorized disclosure can take refuge in anonymity. We should not, however, be too quick to celebrate this fact, since anonymity can also be utilized to make disclosures that further sectional interests or partisan causes. But how realistic is the threat that leaks will be misused? Is there really a danger that an official will leak classified information—and thereby risk her peace of mind—in order to covertly impose her own values or beliefs on the public? Many commentators presume otherwise; they believe that leakers are invariably motivated by high principles. Bruce Ackerman, for instance, has described leakers as “patriotic” on the grounds that their disclosures “don’t endanger our national security. They promote it, by preserving our constitutional integrity.”26 A closer inspection of the facts, though, shows such enthusiasm to be misplaced, as there appear to be numerous cases where officials have disclosed classified information for reasons unrelated to the merits of the relevant policy. Consider two such instances.
The first concerns the disclosure of the Terrorist Surveillance Program (TSP), a Bush administration initiative that allowed the NSA to eavesdrop on electronic communications without obtaining a warrant from the Foreign Intelligence Surveillance Court (FISC). This program was made public in December 2005 by James Risen and Eric Lichtblau of the New York Times. In their report, Risen and Lichtblau claimed that, “nearly a dozen current and former officials” discussed the program with them “because of their concerns about the operation’s legality and oversight.”27 Though the reporters have never identified their sources, subsequent developments have allowed us to learn a little about two of them.
One source whom Risen and Lichtblau relied upon was Thomas Tamm, a former Justice Department employee. Tamm joined the Justice Department in 1998 and was initially assigned to a unit evaluating death penalty cases. He reportedly became “disaffected” in 2001 when John Ashcroft took over as attorney general, because unlike his predecessor, Janet Reno, Ashcroft began to encourage prosecutors to seek the death penalty in “as many cases as possible.”28 Subsequently, in 2003, Tamm requested a transfer to the Office of Intelligence Policy and Review, the unit charged with securing wiretap warrants.29 It was here that Tamm “stumbled upon” the fact that certain wiretap requests were being sent directly by Attorney General Ashcroft to the chief judge of the Foreign Intelligence Surveillance Court, Colleen Kollar-Kotelly.30 Although Tamm was not involved in this process and was therefore unaware of the details, including the various (albeit unconventional) safeguards that were in place, he decided on the basis of secondhand information that the whole affair did not “smell right” and therefore contacted Lichtblau, thereby kick-starting the Times’s investigation into the NSA program.31 Note that Risen and Lichtblau’s report contained none of these contextual details, least of all that Tamm was motivated in part by his “anger” at unrelated Bush administration policies.32 These details became public knowledge only after Tamm was declared a target of the FBI investigation into the disclosure, which in turn led him to discuss his case with Newsweek in an effort to defend his actions as well-intentioned.33
Another source whom Risen and Lichtblau relied on was Russell Tice, a former NSA employee, who disclosed his involvement to ABC News shortly after the Times report was published.34 Tice told ABC News that he was concerned about inadequate oversight at the NSA and hinted at the existence of surveillance programs even more extensive than what the Times’s report described. But when reporters subsequently reached out to the congressional intelligence committees about these allegations, congressional staffers reacted cautiously, claiming that Tice “comes with baggage.”35 What this meant became clearer after Tice appeared before Congress in February 2006. His testimony revealed that he had been fired from the NSA in May 2005 following a run-in with administrators over his allegations that a colleague might be engaged in espionage.36 Though an internal investigation had dismissed Tice’s claim, he apparently suggested that the colleague be reinvestigated, which in turn led administrators to demand that he undergo psychiatric evaluation.37 After this evaluation concluded that Tice suffered from paranoia, his security clearance was revoked and he was assigned to a low-level position. When Tice subsequently wrote to 132 members of Congress complaining of mistreatment, the NSA fired him.38 Now, it is important not to assume that Tice’s allegations against the NSA were influenced by his treatment at the hands of administrators. Tice has stated that he had raised concerns about “suspected illegal activity” at the NSA before 2005, and that the NSA had him diagnosed as paranoid in order to get rid of him.39 Since the secrecy surrounding the case makes it difficult to ascertain the actual sequence of events, this claim certainly deserves the benefit of the doubt. What is remarkable, though, is that the report in the Times offered not the slightest hint of this background. Had Tice not come forward of his own accord, we would likely never have had the chance to factor his complicated circumstances into our assessment of Risen and Lichtblau’s account.
The second case involves the disclosure of Operation Merlin, a covert CIA operation to route flawed nuclear technology to Iran, which reportedly went awry after the Russian intermediary had an attack of nerves and tipped off the Iranians, thereby potentially allowing them to make good use of the technology. The public became aware of this operation following the publication of James Risen’s State of War in 2006, which discusses Operation Merlin as part of a broader story focused on the mismanagement of the CIA’s operations inside Iran.40 Risen’s source, it now appears, was Jeffrey Sterling, a former employee of the CIA. A little by way of background: after joining the CIA in 1993 and completing training in Farsi, Sterling was sent to Germany in 1995 to recruit spies who could undertake espionage in Iran. In 1997 Sterling reportedly learned from his supervisor that he was not being given leads to develop because a “big black man speaking Farsi” would “draw too much attention” to Iranian agents working for the CIA.41 Sterling then elected to move to the CIA office in New York in 1999. However, following “several disagreements” with managers about his performance, Sterling was eventually fired in 2002.42 He then filed a racial discrimination lawsuit against the CIA, but the case was dismissed in 2005 after the Fourth Circuit accepted the CIA’s invocation of the state secrets privilege.43 Meanwhile the CIA also refused to approve a draft of a memoir that Sterling wanted to publish, on the grounds that it contained classified information. It is now alleged by the government that Sterling, frustrated by these setbacks, decided to retaliate against the CIA by disclosing to Risen what he knew about Operation Merlin.44 Risen’s story, it should be noted, hints at none of this; none of the details outlined above would have become public knowledge had the Justice Department not indicted Sterling for leaking classified information to Risen.
Why should we be troubled by leaks of classified information like the ones made by Tamm, Tice, and Sterling? There are at least two causes for concern. The first is that disclosures made with a view to embarrass decision-makers may well be selective. Bias is especially problematic in the present context because it can prove difficult for decision-makers to correct the public record, as they will usually not want to reveal additional information about the activities that have been disclosed. As a result, the public may end up with a false impression about the motives of elected officials and the efficacy of public institutions.45 This may have happened in the cases at hand. As more detailed accounts of the inner workings of the Justice Department have emerged, it has become clear that Attorney General Ashcroft and his subordinates actually extracted from President Bush modifications to the NSA program that they saw as necessary to ensure its legality.46 Meanwhile the CIA has claimed that Sterling’s account has “falsely characterized certain facts and circumstances” surrounding Operation Merlin, though it has declined to elaborate further in public.47
These leaks are also troubling because the accompanying anonymity makes it very difficult for citizens and lawmakers to guard against being misled. If citizens and lawmakers are aware that the source of an unauthorized disclosure may be a disgruntled employee, then they are put in a position to treat the disclosure with due caution, even when decision-makers are unable to defend themselves. This problem could of course be mitigated if reporters and publishers were to offer their readers some guidance about the motives and circumstances of their sources, but such candor is rare in practice. It may be objected that offering details about the motives and circumstances of a source could make it easier for investigators to identify and punish the official responsible. This is a reasonable concern, but it is not self-evident that every piece of information about the motives and circumstances of a source will tend to expose her identity. Note also that reporters, editors, and publishers do not have obligations to their sources alone—they also have obligations to citizens and lawmakers, whom they ought not to mislead. This is all the more true when journalists cite the public’s right to know as a justification for publishing classified information, since an insufficient degree of candor on their part actually prevents the public from knowing that a given report provides only a partial perspective.48 This is what seems to have happened in the cases described above, where the reporters involved appear to have made little effort to strike a balance between their obligations to their sources and to the public. For example, after the Times published the NSA story (and before Tamm’s interview with Newsweek was published), Lichtblau claimed that while sources can have an “axe to grind,” the original source for the story had “checked out on all counts,” and his “angst seemed sincere.”49 However, we now know from Tamm himself that he was motivated in part least by his anger at Bush administration policies that had nothing to do with the NSA surveillance program.
The Sterling case is even more troubling on this count. What could Risen have hoped to achieve, one might wonder, by his silence with regard to Sterling’s motives? After all, his silence could hardly have thrown investigators off the scent. Consider the circumstances: Sterling was reportedly the only person who had access to some of the details cited in Risen’s account; the memoir that Sterling had submitted to the CIA’s Publications Review Board apparently contained a description of Operation Merlin; and Risen had previously chronicled Sterling’s employment woes in the Times.50 The persons most affected by Risen’s complete silence, it seems, were not investigators from the FBI, but rather citizens and lawmakers, whom he claimed he wanted to educate about the “recklessness” of the Bush administration, but whom he denied the opportunity to evaluate Sterling’s claims in context.51
Can We Justify Anonymity?
We have seen that though the practice of leaking allows officials to raise the alarm without being easily detected, it can also hinder citizens’ and lawmakers’ ability to guard against misleading disclosures. This complication raises the question of whether it is possible to regulate the practice of leaking: that is, can we prevent officials from disclosing classified information with a view to covertly furthering sectional or personal agendas? If we cannot regulate the practice of leaking in this way, then we will have reason to worry that leaks actually increase rather then reduce the ability of officials to abuse their privileged access to state secrets.
In order to regulate the practice of leaking, we need to outline the conditions under which an official can be justified in making an anonymous disclosure and then identify the means that will allow us to ascertain whether a particular disclosure meets these conditions. The former of these tasks poses the less significant challenge. In order for an official to be justified in leaking classified information, the disclosure must (a) concern an abuse of public authority; (b) be based on clear and convincing evidence; (c) not pose a disproportionate threat to public safety; and (d) be limited in scope and scale as far as is possible. If this formulation sounds familiar, it should. Since an official who leaks classified information does much the same thing as an official who blows the whistle—they both disobey laws prohibiting unauthorized disclosures—it is only to be expected that the conditions under which these practices can be justified should be broadly similar. The crucial difference, of course, is that unlike a national security whistleblower, an official who is justified in leaking classified information is not required to disclose her identity. How can this divergence be justified?
Two justifications can be offered. As noted in chapter 5, an official can be justified in making an anonymous disclosure when doing so exposes activities that are so obviously wrongful that her motives, and therefore her identity, are rendered unimportant. This explains why, for example, we should not be troubled by our ignorance of the identity of the official who disclosed to Seymour Hersh details about the military investigation into the mistreatment of detainees at Abu Ghraib prison.52 This disclosure revealed violations so depraved that the government’s response was not to challenge the motives of the source or the appropriateness of the disclosure but rather to take corrective action—to terminate the offending programs and penalize the officers responsible.
Recall, however, that unauthorized disclosures typically expose suspected rather than gross wrongdoing: violations of the law where the costs and benefits are somewhat more evenly matched. Such disclosures usually prove controversial owing to disagreements over the rightful bounds of public authority. For example, rather than provoking formal apologies or corrective action, the disclosures that revealed the NSA’s warrantless surveillance program and the CIA’s secret prison system were greeted by strong criticism from lawmakers—in the former case Congress responded to the disclosures by enhancing the president’s authority to conduct wiretaps without a warrant; in the latter it enacted legislation allowing the CIA to withhold the pensions of employees dismissed for leaking classified information. Needless to say, when a disclosure is formally condemned in this way, the source of a disclosure cannot claim to have unveiled gross wrongdoing. She will have to avail herself of a different justification for maintaining her anonymity in the face of criticism that she ought to submit to the law since she has thrust her values upon an unwilling public. Arguably, such a justification can be derived from the intention rather than the substance of her disclosure.
As discussed in chapter 5, conflicts over the rightful bounds of public authority invariably lead to differences over whether prima facie wrongdoing actually amounts to an abuse of authority. Unfortunately, these differences can be exploited by officials looking to provide a veneer of respectability to unauthorized disclosures that are actually intended to further a narrow or personal agenda. For instance, an official can cite the public’s right to know as her motivation for disclosing a classified study into the pollution caused by the use of depleted uranium in military ordnance, when her real agenda is to embarrass the president into purchasing ordnance from another supplier. I argued in chapter 5 that in order to be able to guard against such self-interested or manipulative disclosures, officials responsible for disclosing evidence about suspected wrongdoing must be willing to identify themselves so that we can investigate their motives.
However, this demand for self-identification has a major drawback. It can be difficult for an official to know ex ante whether citizens and lawmakers will agree that the violation of the law she has discovered actually constitutes an abuse of authority. For instance, suppose an official discloses information about a violation of the law that she believes will be viewed as wrongdoing, and her disclosure actually ends up being condemned by Congress, which believes that the president is acting rightfully (as in the secret prisons case). Evidently, this official has misjudged what citizens and lawmakers consider wrongful activity. She genuinely believed the president has exceeded his brief, but her suspicion has turned out to be unfounded, as citizens and lawmakers either support or at least do not formally condemn the president’s actions. According to the criteria outlined in chapter 5, this official is now obliged to forgo anonymity. This requirement may not initially seem all that troubling. What does an official who has acted in good faith have to fear from publicity, we might wonder? The answer of course is—informal retaliation. As we saw in chapter 5, even an official whose decision to disclose classified information is pardoned by a jury must still contend with colleagues and managers who may not be as forgiving. It seems reasonable to argue, then, that if the official has acted in good faith—if she has exposed activities that she disinterestedly believes constitute wrongdoing—then she should not have to come forward and identify herself when doing so will likely subject her to forms and degrees of retribution, especially harassment and public humiliation that are not commensurate with her actions, but which the law is powerless to curb.53
So, to summarize, it is problematic to require an official who makes an unauthorized disclosure about a prima facie violation of the law to identify herself, because this can lead to the penalization of disclosures based on honest errors of judgment about the severity of the wrongdoing. This outcome is undesirable because we do not want to inhibit officials from making disclosures about violations of law that seem to amount to an abuse of authority, especially when we cannot rely on lawmakers or judges to provide oversight or review for the reasons discussed in chapters 2–3. Remember, we demand self-identification only because we want to know if the source of the disclosure is less than disinterested. Since the requirement of self-identification is likely to deter both bad faith and good faith disclosures, a more reasonable way to address the threat of distortion or manipulation is to devise a more precise means by which we can filter out disclosures made in bad faith.
Having clarified the conditions under which an official can be justified in making an anonymous disclosure, we need to identify the means by which we can ensure that these conditions have been met. It is not difficult to ascertain whether an official is justified in making an anonymous disclosure that reveals gross wrongdoing. If the activity exposed by an anonymous disclosure is formally condemned by lawmakers or judges, then we can conclude that the official responsible for the disclosure is not obliged to reveal her identity. This is why, for example, it was permissible for Deep Throat—Bob Woodward and Carl Bernstein’s legendary Watergate source—to keep his role secret. But how do we ascertain whether the source of a disclosure faces the prospect of suffering unwarranted retaliation when we do not know her identity and therefore lack the most important clue to her motives? It would clearly be unwise to allow officials to be judges in their own cases, but, fortunately, this is not the only alternative. Since leaks of classified information are typically routed via reporters, editors, and publishers, we could potentially requisition the assistance of these intermediaries, as they are the likeliest to know the identity of the source of an anonymous disclosure and are therefore best situated to ascertain her motives. But are reporters, editors, and publishers willing and able to take on the responsibility of acting as filters in the public interest?
There are two reasons to be concerned on this front. First, they themselves may not always know the identity of their sources; these sources may be anonymous “tipsters.” Though officials have so far had relatively little need to act as anonymous tipsters, this method of disclosing classified information will surely become more prevalent if reporters are routinely compelled to disclose the identity of their sources. The practice of anonymously tipping off reporters also appears likely to become more widespread owing to technological advances. Consider the profusion of WikiLeaks-inspired websites like the Wall Street Journal’s “Safe House” and Al-Jazeera’s “Transparency Unit,” both of which allow officials to anonymously upload evidence of wrongdoing.
Troubling though these developments may be, they do not rule out the possibility that making dutiful inquiries will allow journalists to learn at least something about the motives of an anonymous informant. For example, in investigating the claim that Iraq was attempting to purchase uranium from Niger, journalists were able to discover, albeit somewhat belatedly, that Rocco Martino, the mysterious individual who provided reporters with forged documents attesting to a sales agreement between Iraq and Niger, was actually a front man for one or more intelligence agencies.54 There is no reason to think that similar investigative efforts could not be made in other cases involving anonymous informants. Of course this does not mean that journalists will always be able to ascertain the provenance of an anonymous tip. But in such cases their duty is clear: if they take the public’s right to know seriously, then they are obliged either to withhold the relevant information or, if the information can be independently verified and is troubling enough to outweigh concerns about the informant’s motives, then they ought to publish it alongside an account of the reasons to be concerned about her motives.
The second reason to doubt that reporters, editors, and publishers have the ability to act as filters in the public interest draws on the discussion in chapter 4 where I noted that journalists, who do not have access to the big picture, cannot readily claim to know how far the disclosure of a piece of classified information will harm national security. Even if they are able to verify that a source is acting in good faith, they may still end up publishing information that is deeply harmful to national security. But this challenge too is not insurmountable. The fact that reporters, editors, and publishers do not have access to the big picture should be seen as obliging them to approach the executive branch prior to publication in order to allow the president the opportunity to offer reasons against disclosing classified information.55 In addition, it seems reasonable to think that reporters who have built careers on reporting national security affairs can acquire a network of trusted contacts whom they can use to vet the sensitivity of the information that they intend to disclose. These contacts could also help them gauge the cautionary advice offered by the president. There is of course the risk that “old hands” will come to see the world through the lens of the officials whom they are meant to subject to critical scrutiny. We must remain alert to this danger. But the existence of this danger does not preclude journalists from drawing upon the expertise of senior officials in order to act as filters in the public interest.
So reporters, editors, and publishers can serve as filters in the public interest, but are they willing to assume this responsibility? The challenge lies in the prevalence of the belief that “the presumptive duty of the press is to publish, not to guard security or to be concerned with the morals of its sources.”56 This belief has grown out of Alexander Bickel’s epochal observation in his Morality of Consent that the existence of the First Amendment effectively establishes a “contest” between the press and the government, because the freedom afforded to the press by the First Amendment means that though the government is allowed to “guard mightily” against leaks, it is forced to “suffer them if they occur.”57 Bickel commended this “contest” because—writing in the wake of the Pentagon Papers episode—he saw it as one of the only means by which citizens could get a handle on the president’s vast national security apparatus. Hence, not surprisingly, he took the view that “the chief responsibility of the press is to play its role in that contest”: to publish everything “newsworthy” that it can get its hands on.58 And the need to satisfy this “chief responsibility” in turn meant, Bickel argued, that the press ought to behave—and ought to be allowed to behave—as a “morally neutral, even an unconcerned, agent as regards the provenance of newsworthy material that comes to hand.”59
To be clear, Bickel did not believe that the press has no responsibilities other than to publish. On the contrary, he warned that “not everything is fit to print.”60 There ought to be, he wrote, due regard for “probable factual accuracy, for danger to innocent lives, for human decencies, and even, if cautiously, for nonpartisan considerations of the national interest.”61 However—and this is the key point—Bickel insisted that the enforcement of these norms must depend on “self-discipline and self-restraint, and on public opinion, not on law.”62 He disfavored using the law to regulate this “unruly contest” between the executive and the press for the simple reason that “if we should let the government censor as well as withhold, that would be too much dangerous power.”63 This was equally true, Bickel argued, when it came to the use of anonymous sources, because any legal regulation “forcing reporters to divulge such confidences would dam the flow to the press, and through it to the people, of the most valuable sort of information.”64
The fact that Bickel’s view has garnered widespread support among members of the media establishment has meant that the standard-bearers of the profession have been willing to accept that they ought to act as filters in the public interest. For example, when the New York Times was heavily criticized for revealing the existence of the Treasury Department’s Terrorist Financing Tracking Program, the deans of some of America’s leading schools of journalism issued a statement, entitled “When in Doubt, Publish,” defending the New York Times on the grounds that it “is the business—and the responsibility—of the press to reveal secrets.”65 But the same statement also rapped the Washington Post on the knuckles for having allowed Robert Novak to disclose Valerie Plame’s identity, warning that “the public wants the press to keep a sharp lookout, but wants the job performed responsibly.”66 Bill Keller and Dean Baquet, former editors of the New York Times and the Los Angeles Times, respectively, have expressed similar sentiments. They have argued that while the decision on whether or not to publish classified information “is not one we can surrender to the government,” it is also “not a responsibility we take lightly.”67
These Bickelian affirmations of the need to respect “self-disciplined limits” have extended to the use of anonymous sources as well. For instance, following a series of scandals involving the misuse of anonymous sources, the New York Times issued a revised “Confidential News Sources Policy” in 2004 warning its reporters that “when we use such sources, we accept an obligation not only to convince a reader of their reliability but also to convey what we can learn of their motivation,” adding for good measure that “we do not grant anonymity to people who use it as cover for a personal or partisan attack.”68 The Washington Post offered its readers similar assurances. Its editor, Leonard Downie, pledged in 2004 that his reporters were aware that the Post’s reporting guidelines contain the exhortation that “we must strive to tell our readers as much as we can about why our unnamed sources deserve our confidence.”69
Heartening though these proclamations may be, they seem not to have had great effect. If the troubling cases outlined earlier are not proof enough, consider how extensively the public editors and ombudsmen of the Times and the Post have criticized their employers on this count. Since 2004 the Times’s public editors, Daniel Okrent, Byron Calame, and Clark Hoyt, have altogether written nearly a dozen articles about the “integrity issues” raised by the paper’s use of anonymous sources.70 Hoyt, for example, has pointed out that the Times’s use of anonymous sources “to air opinion, not fact” actually increased after it revised its “Confidential News Sources Policy” in 2004, even though “the policy would seem to discourage that,” adding that nearly 80 percent of the anonymous sources used “were not adequately described to readers.”71 The Post has not been far behind on this count. Writing more than five years after Downie’s pledge, the Post’s ombudsman, Andrew Alexander, observed in 2009 that many of the newspaper’s “lofty standards” on the use of anonymous sources are “routinely ignored,” while others “are unevenly applied.” “The Post’s sourcing rules are fine,” he added, “the problem is compliance.”72
Why have the Times and the Post failed to abide by the standards that they themselves have publicly endorsed? If their public editors and ombudsmen are to be believed, the indiscriminate use of anonymous sources should be attributed to laxness. For instance, Alexander, the Post’s ombudsman, has written that “the solution must come in the form of unrelenting enforcement by editors, starting with those at the top.”73 Similarly, Calame, the Times’s public editor, has recommended “commitment to top-level oversight, and to providing sufficient editing attention.”74 But in spite of all the policies and memos that have been drafted and circulated by the editors of the Times and the Post, the indiscriminate use of anonymous sources largely continues unabated. The editors of the Times, for instance, have themselves admitted that they continue to “stray” from their “own guidelines” regarding the use of anonymous sources.75 The Post has been equally helpless. As Alexander has admitted, ombudsmen, himself among them, have complained “for decades” about the Post’s “unwillingness to follow its own lofty standards” concerning the use of anonymous sources—all to little effect.76
The continuing inability of the Times and the Post to abide by their own declared standards indicates that there is something fundamentally wrong with the diagnosis offered by their public editors and ombudsmen. It may be that this diagnosis rests on a false assumption. The belief that we can prevent the indiscriminate use of anonymous sources simply by demanding that editors make more of an effort to police reporters assumes that editors themselves (and the publishers who hire them) have no incentive to permit the indiscriminate use of anonymous sources. This assumption can be traced directly back to Bickel’s view—and it is demonstrably false. Bickel assumed that the reporter’s interest is inevitably aligned with the public interest, because the former’s desire to profit by reporting the news would have the effect of serving the latter’s desire to know. The “professional interest of the reporter,” he declared, serves as a “sentinel over the public rights.”77 Bickel saw little reason to be concerned about the use of anonymous sources. “The reporter’s access is the public’s access,” he wrote.78 But what about the possibility that instead of treating anonymous sources in a “morally neutral” or “unconcerned” manner, reporters (and the editors who supervise them and the publishers who hire them) will use their privileged access to sources to further their careers or their favored causes? It is certainly not difficult to imagine how the parochial interests (the desire for prizes, promotions, and readers) and political affiliations of reporters, editors, and publishers could end up influencing decisions on this count, especially when their use of anonymous sources is, by definition, not open to public scrutiny.79
Indeed, how else, if not by invoking the narrow interests of reporters, editors, and publishers, can one explain some of the most sensational disclosures of the past decade? Consider, for instance, the Times’s decision to disclose the NSA’s warrantless surveillance program. As is now well known, the Times had initially agreed to withhold Risen and Lichtblau’s report after being warned by President Bush that its publication could jeopardize national security. So what prompted the Times’s editors and publisher to change their minds? Though Risen and Lichtblau’s report cited “concerns” about “the operation’s legality and oversight,” this claim is undercut by their own acknowledgment that the NSA program was neither obviously unlawful (as the chief judge of the Foreign Intelligence Surveillance Court had agreed to cooperate on Article II grounds) nor without oversight (as the Bush administration had briefed the congressional leadership). These safeguards were unconventional, but they placed officials from all three branches of government in the position of being able to raise the alarm in the event that they felt that the NSA program was wholly unconstitutional. So why, then, did the Times suddenly feel the need in late 2005 to give precedence to the concerns of officials like Tice and Tamm? Lichtblau has indicated that the editors at the Times revisited their decision to put the report on hold after being confronted with the prospect of being “scooped” by Risen, who, frustrated by the Times’s willingness to defer publication, had decided to publish the report in his forthcoming book, State of War.80
Consider as well the Times’s decision to disclose the Department of the Treasury’s efforts to monitor international money transfers in spite of the pleas of numerous officials, including two Democrats, Lee Hamilton, cochair of the 9/11 Commission, and Representative John Murtha, an otherwise fierce critic of the Bush administration. This report’s publication was met with an outpouring of criticism; among the critics was the Times’s own public editor, Calame, who argued that the disclosure was unwarranted in view of the “apparent legality of the program in the United States, and the absence of any evidence that anyone’s private data had actually been misused.” Keller, the Times’s editor, responded to critics by defending the story as “newsworthy” on the grounds that “some officials who have been involved in these programs have spoken to the Times about their discomfort over the legality of the government’s actions and over the adequacy of oversight.”81 Who were these officials and what were their motives? This Keller did not say. His silence notwithstanding, it is hard to impute disinterested motives to the Times’s sources, for, as the Wall Street Journal observed at the time, “since the Treasury story broke … no one but … a few cranks have even objected to the program, much less claimed illegality.”82 One is hard-pressed here to discount the Journal’s suspicion that the Times’s decision to run the story owed less to its concern for the “newsworthy” and more to its desire to revive its “watchdog” credentials, which had been tarnished by its evident subservience to the White House in the run-up to the Iraq War.
To be clear, the cases highlighted above should not be seen as implying that reporters, editors, and publishers are bound to use anonymous sources without due care. Happily, there have been instances where members of the press have exercised admirable self-restraint in exposing classified information.83 There have reportedly also been numerous instances where members of the press have refused to grant anonymity to sources with questionable motives (albeit unbeknownst to the public, who cannot witness the confabulations that take place within the newsroom, much less between a reporter and his source). But the above cases do show that it would be unwise to assume that reporters, editors, and publishers will always filter out rash or malicious disclosures; on the contrary, there is always the danger that their parochial interests could warp their judgment.
Can Leaks Be Regulated?
If the interests of reporters, editors, and publishers do not always align with those of the public, then arguably we should disregard Bickel’s view and take recourse to the law in the hope that the threat of punishment will prompt journalists to filter out rash or malicious disclosures. Consider what such a law might look like. The misuse of anonymous sources, as we have seen, can occur either when reporters, editors, and publishers neglect to investigate whether the source of a controversial disclosure has acted in good faith, or when they fail to disclose that the source of a controversial disclosure has less than disinterested motives. The law could track the former kind of wrongdoing by establishing procedural tests or criteria, such as “due diligence” requirements that reporters, editors, and publishers must satisfy before publishing a leak of classified information.84 For instance, the law could require them to obtain from sources an explicit declaration of any potential conflict of interest on the condition that if a source misleads the reporter, then the reporter is no longer obliged to protect the source’s identity. The law could also require reporters, editors, and publishers to sign a statement confirming that they have ascertained that a source does not have a conflict of interest (a step that would help fix individual and corporate responsibility and thus preempt pleas of ignorance or efforts to shift blame).
The law could track the latter kind of wrongdoing by subjecting reporters, editors, and publishers to a “bad faith test,” which would examine whether a failure to disclose a source’s conflict of interest could be reasonably attributed to recklessness, malice, or self-interest. It is difficult to exhaustively specify in advance the sorts of actions that would indicate the presence of bad faith, but the law would be interested in ascertaining whether, for example, a member of the press made a reasonable effort to describe a source’s conflict of interest, or if he misrepresented or even obscured a source’s interests in order to reduce the appearance of conflict and thereby enhance the credibility of a given news report.
Finally, in the event that wrongdoing is proven on either of the counts described above, the law could impose a fine and require the publication of a correction where appropriate. The former penalty, which could extend to the confiscation of profits, would bring home the costs of irresponsible behavior, while the latter penalty would correct the public record. Let me stress that under such a law members of the press would be obliged to disclose a source’s identity only to a judge, and only in the event that her disclosure is formally condemned by Congress or becomes the subject of a criminal investigation. This means that the proposed law could never warrant an injunction (since a disclosure has to be published before it can prove controversial).
In light of our concerns about the misuse of anonymity, such a law seems quite attractive. But would it be feasible? One major obstacle is a cluster of legal challenges emanating from the First Amendment. The most immediate of these challenges comes from the protection afforded to speech under the First Amendment. Under prevailing First Amendment doctrine, “content-based regulations” that restrict “high-value speech” are “presumed unconstitutional.”85 The Supreme Court has made clear that “uninhibited, robust, and wide-open” debate on political issues, and “criticism of official conduct” in particular, constitute speech of the highest value; thus any law attempting to regulate disclosures that allege wrongdoing by public officials is likely to run afoul of the First Amendment.86
There may be an escape route, though. The court has accepted that a law that burdens “political speech” can be upheld “if it is narrowly tailored to serve an overriding state interest.”87 The proposed law satisfies both these requirements. It is narrowly tailored, because it targets only leaks by officials with undisclosed conflicts of interest. Furthermore, it serves a compelling interest because it makes it harder for officials to obscure their efforts to manipulate public opinion. What makes this interest particularly compelling is that the interest that the First Amendment would otherwise protect—vigorous public discussion—is less likely to be realized in the present context. The First Amendment, as Judge Learned Hand famously put it, “presupposes that right conclusions are more likely to be gathered out of a multitude of tongues than through any kind of authoritative selection.”88 But how can the “right conclusions” be drawn when a multitude of tongues are rendered silent or discredited by the lack of an equivalent access to classified information? How can the “marketplace of ideas” serve as a “test of truth” when few, if any, competing voices can warn listeners against being hoodwinked?89 Some might respond that the First Amendment calls for forbearance on the grounds that an “uninhibited, robust, and wide-open” public debate will occur as and when contrary evidence forces its way into public sight. But then again, do we want to extend the protections afforded by the First Amendment to anonymous disclosures made by officials with undisclosed conflicts of interest in the hope that subsequent, and possibly quite belated, disclosures will confirm that their claims and motives were prudent and decent?
Unfortunately, there are few, if any, precedents that can help us gauge how the Supreme Court might react to this sort of argument. There have been, to be sure, strong defenses of anonymous political speech, most notably in McIntyre v. Ohio Elections Commission (1995) where the court, having spoken warmly of a “respected tradition of anonymity in the advocacy of political causes,” struck down an Ohio regulation banning the distribution of anonymous leaflets during an election on the grounds that an “interest in providing voters with additional relevant information does not justify a state requirement that a writer make statements or disclosures she would otherwise omit.”90 But the circumstances the court confronted in McIntyre differed markedly from those envisioned by the law proposed here. The Ohio regulation scrutinized in McIntyre preempted publication, did not discriminate between misleading and valuable anonymous speech, and demanded up-front disclosure of the identity of the speaker, whereas the law proposed here does not prevent publication, targets only anonymous speech that is misleading, and requires only the disclosure of conflicts of interest. Furthermore, McIntyre involved the use of anonymity by an author (rather than a source), who was a private citizen (rather than an official), and whose claims were not misleading. It is hardly surprising, then, that the court concluded in McIntyre that knowing “the name and address of the author [would] add little, if anything, to the reader’s ability to evaluate the document’s message.”91
Equally inconclusive are cases where the court has upheld limitations on anonymous political speech in the face of First Amendment challenges. For instance, in Buckley v. Valeo (1976) and McConnell v. Federal Election Commission (2003) the court accepted the claim that laws requiring campaign contributors to disclose their identity aided “voters in evaluating those who seek federal office,” and it reminded those who objected to such disclosure requirements on First Amendment grounds that they must not overlook “the competing First Amendment interests of individual citizens seeking to make informed choices in the political marketplace.”92 But the court also emphasized here that such disclosure requirements might constitute an “unconstitutional burden” if those whose identities were disclosed were thereby rendered vulnerable to retaliation.93 This qualification creates uncertainty as to whether a court influenced by Buckley and McConnell would support the proposed law, as the disclosure of a source’s conflict of interest could lead to the revelation of her identity and thereby expose her to retaliation. Its decision could ultimately turn on whether it thinks the First Amendment permits the chilling of a particular class of political speech, namely, leaks by unnamed officials with undisclosed conflicts of interest.
There is one case that gives some reason for optimism on this count: Cohen v. Cowles (1991). This case arose after Dan Cohen, a member of a gubernatorial campaign, provided reporters from the Minneapolis Star Tribune and the St. Paul Pioneer Press court records pertaining to a rival candidate. Although the reporters had promised Cohen anonymity, their editors decided to disclose his role. This led to Cohen’s losing his job, prompting him to sue the newspapers for breach of confidentiality. The newspapers responded to Cohen’s action by arguing that allowing him to sue for damages would burden speech “because news organizations will have legal incentives not to disclose a confidential source’s identity even when that person’s identity is itself newsworthy.”94 The court, however, disagreed with this logic. A five-member majority led by Justice White argued that no First Amendment interests were at stake because the burden on publication was “self-imposed”: the reporters had voluntarily offered and promised confidentiality.95 However, the four-member minority responded that this view relied on “a conception of First Amendment rights as those of the speaker alone, with a value that may be measured without reference to the importance of the information to public discourse.”96 If we were to take the audience, rather than the speaker, as the bearer of First Amendment rights, Justice Souter argued, then we would conclude that, on balance, Cohen’s claim for damages ought to be denied, because knowing about “Cohen’s identity expanded the universe of information relevant to the choice faced by Minnesota voters,” as it revealed something about “the character of the candidate who had retained him as an adviser.”97 Although the majority was not convinced by this argument, it remains to be seen whether the court might warm to this view in cases involving classified information leaks where both the public interest in the motive of the source and the threat of retaliation faced by the source are greater.
A second legal challenge to the proposed law comes from the protections afforded to the press under the First Amendment. In particular, the provision requiring a news organization to correct the public record would appear to run afoul of the court’s stance in Miami Herald Publishing Company v. Tornillo (1974) that compelling editors or publishers “to publish that which “ ‘reason’ tells them should not be published” is unconstitutional.98 It is difficult to see how we could overcome this challenge. An answer, insofar as it exists, lies in emphasizing the fact that the court’s remarks in Tornillo were addressed to a candidate for office who was seeking a right to reply to critical remarks published by the Miami Herald, whereas the proposed law seeks to correct news reports featuring misleading content. The success of this line of reasoning depends on whether the court can be persuaded that misleading content is at least as troubling as false content (which the press can be ordered to retract under current First Amendment doctrine). Misleading content, at least of the sort envisioned by the proposed law, is deeply troubling, because its flaws cannot be easily discerned or corrected through careful reading and public debate. But the court may respond here with the argument that it is still up to readers to exercise due caution when digesting anonymous disclosures.99 As the court stated in Tornillo, “a responsible press is an undoubtedly desirable goal, but press responsibility is not mandated by the Constitution and like many other virtues it cannot be legislated.”100 If this view holds, it suggests that even if the court were otherwise amenable to the proposed law, it would uphold only a more modest version of it, one that does not require the press to correct the public record.
A further challenge to the proposed law comes in the form of the so-called reporter’s privilege, a legal privilege granting reporters the right to refuse to disclose information about confidential sources. The proposed law potentially conflicts with the reporter’s privilege because it requires members of the press to disclose before a court information pertaining to the conflicts of interests of their confidential sources. It may seem that this conflict does not raise any significant constitutional issue because (as we saw in chapter 4) the court has stated in Branzburg v. Hayes (1972) that reporter’s privilege cannot be used to thwart the administration of justice. Note, however, that the Branzburg court saw itself as addressing the “sole issue” of whether reporters can be compelled to testify in instances where “news sources themselves are implicated in crime or possess information” relevant to a criminal investigation.101 Its opinion emphasized that “no penalty, civil or criminal, related to the content of published material is at issue here,” adding that “use of confidential sources by the press is not forbidden or restricted … [n]o attempt is made to require the press to publish its sources of information or indiscriminately to disclose them on request.”102 This disclaimer leaves open the question of whether the proposed law, which compels news reporters to testify about their confidential sources in order to ascertain whether they themselves should be punished, is compatible with the First Amendment’s guarantee of the freedom of the press.103
Although there are few, if any, precedents that could shed clear light on this question, the case can be made that the proposed law does not undermine what the First Amendment finds valuable in a reporter’s privilege. The reporter’s privilege is founded on the notion that while the First Amendment “guarantees a free press primarily because of the important role it can play as a vital source of public information,” the “newsgathering and reporting activities of the press are inhibited when a reporter cannot assure a confidential source of confidentiality.”104 Now, does the proposed law really challenge this line of reasoning? It is not self-evident that requiring members of the press to provide a court with details pertaining to their confidential sources would make it harder to obtain the cooperation of sources willing to act in good faith, namely, officials willing to leak classified information that they disinterestedly believe constitutes wrongdoing. The confidentiality of such a source would be preserved since the court would have no reason to compel reporters, editors, or publishers to publicly reveal any additional details that might expose such a source to retaliation. Hence it is not clear what valuable news-gathering interest would be put at risk by legislation making the reporter’s privilege dependent on the willingness of reporters, editors, and publishers to faithfully investigate and disclose conflicts of interest.
This argument may be confronted with a point made earlier. It can prove difficult for an official to gauge how a leak of classified information will be received, and this could mean that sources with merely an appearance of a conflict of interest may not come forward with valuable information. These sources might well fear that should their disclosures prove controversial enough to spark a criminal investigation into the leak, a court could demand that further information about their interests be disclosed, at which point their cover might be blown, exposing them to retaliation, even though their decision to leak classified information was not actually motivated by self-interest.
Admittedly, this danger exists. But it is not clear that it will be easily realized in practice. Under the proposed law, it would be up to a judge to decide if and what corrections to the public record are warranted, and such corrections could, presumably, be tailored to forestall retaliation. It is possible that a judge could, after having balanced the relevant interests, still demand corrections that have a tendency to reveal the identity of the source. Hence an element of risk would always remain for sources who are not willing to have their conflicts of interests disclosed in the event that their disclosures result in a criminal investigation. But it is not clear how much this element of risk would actually impinge on the willingness of such sources to cooperate with reporters. There is little empirical evidence to show causation, and so we are, as Lillian BeVier has pointed out, “simply awash in indeterminacy about the impact that recognizing or not recognizing a reporter’s privilege would actually have.”105
But what if it could be proven that the risk of exposure would cause the aforementioned sources to clam up? In that case, the court would need to consider whether the news-gathering exercise that the First Amendment seeks to protect is better or worse off when sources of this kind—who refuse to disclose potential or apparent conflicts of interest—decline to cooperate with reporters. In light of the cases discussed earlier in this chapter, it seems probable that weeding out sources of this kind would actually increase the likelihood of the press’s furthering what Alexander Meiklejohn and Vincent Blasi have respectively described as the self-governing and checking value of the First Amendment.106
We have so far been discussing challenges to the constitutionality of the proposed law. Let us now consider a juridical challenge relating to whether such a law can be applied fairly. Recall that the proposed law would use two different tests to determine whether anonymous sources have been misused. The first would examine whether a member of the press has followed established due diligence standards, while the second would examine whether there is evidence indicating bad faith on the part of the concerned reporter, editor, or publisher. The former of these tests should not raise serious concerns about fairness, since the criteria will be objective (e.g., did the reporter warn government officials about the impending disclosure and allow them the opportunity to explain or at least mitigate the likely harm?). The same cannot be said about the latter test, though. Here we have reason to be concerned about fairness because the basis of this subjective assessment will not be open to review. Requiring a member of the press to disclose before an open court the details needed to assess whether he has acted in bad faith would almost certainly reveal the identity of his source, an outcome that would defeat the purpose of the trial, which is to determine whether the motive (much less the identity) of the source ought to have been disclosed. Consequently, the judges tasked with applying the proposed law will have to utilize in camera, ex parte hearings, and they will also have to issue redacted or truncated opinions. These arrangements will require them to make the substantive assessments called for by the law without the benefit of adversarial proceedings, while also impeding the public’s ability to understand the reasoning behind their decisions. Under these conditions there will be substantial scope for judicial discretion, and it will not be easy to determine whether the law is being applied consistently—that is, whether like cases are being treated alike.
Troubling as the challenge outlined above may be, it is not without an answer. While we cannot alter the fact that closed proceedings will be necessary, we can lessen the possibility that such proceedings will lead to unfair judgments by raising the standard of proof required to establish wrongdoing. Of course raising the standard of proof comes at a cost—it makes it more likely that only the grossest misuses of anonymous sources are likely to be punished. But there is little alternative here; it makes sense to err on the side of caution when asking judges far removed from the hurly-burly of politics to decode the motives of officials, reporters, editors, and publishers whose activities may at times have more to them than meets the eye.
We have seen that a law intended to punish the misuse of anonymous sources confronts significant legal challenges, and that the concessions required to overcome these challenges are likely to diminish the force and the reach, and therefore the utility, of this law. Let us now consider a final cause of concern: the practical challenge of enforcing such a law. As we saw in chapter 4, no reporter, editor, or publisher has ever been prosecuted for publishing classified information, not even when they have violated a statute like §798, which clearly prohibits the publication of classified information relating to communications intelligence. If presidents do not have an incentive to enforce a statute like §798, it is difficult to see why they will have an incentive to enforce the law under discussion here, since the factors that make it unappealing to enforce the law in the former context are operative in the present context as well. One such factor, as we have seen, is the concern that a trial would draw attention to the classified information revealed by an unauthorized disclosure (and might even demand the disclosure of additional classified information). Another factor is that presidents have a strong incentive to avoid confrontations with reporters and publishers, who can easily retaliate by publicizing the political and personal foibles of the president and his aides. The adage that one should never pick a fight with a man who buys ink by the barrel is only too well known in Washington.
This analysis may provoke the response that the business of law enforcement could be made easier if the authority to initiate and adjudicate cases were vested in the hands of an agency or commission designed to withstand the political power of the press and to use ex parte, in camera procedures. Such a move might solve the reverse problem as well—that is, it would also lessen the possibility that the proposed law would be used to harass reporters, editors, and publishers, especially if this agency or commission were composed of distinguished individuals drawn from civil society, including retired members of the media establishment. But even assuming such an arrangement were feasible, it could do little to overcome a separate obstacle to enforcement, namely, the ease with which reporters, editors, and publishers can evade law enforcement.
It has, of course, long been possible to evade legal constraints on the publication of classified information by operating beyond national boundaries. Take, for example, the publication in 1975 of Inside the Company by Philip Agee, a disaffected former CIA official. Agee avoided legal challenges in America by publishing his book, which revealed the identities of covert CIA officers, in England. Consider as well the 1987 publication of Spycatcher, an exposé penned by Peter Wright, a former member of the British secret service. Wright got around a court order banning his book by having it published in Scotland and Australia (from where it was reportedly smuggled into England in large numbers). The emergence of new media, particularly Internet-based communication channels, has, of course, dramatically increased the ease with which reporters, editors, and publishers can evade laws or regulations pertaining to the publication of classified information. We live in a world where leaks of classified information can be instantly transmitted to “information clearinghouses” like WikiLeaks and OpenLeaks and mirrored on websites based around the world. It is not unreasonable, then, to wonder whether there is any point in enacting the law under discussion here.
It is important not to overstate the challenge. The existence of websites like WikiLeaks and OpenLeaks should not blind us to the fact that the American media establishment continues to exercise enormous influence owing to its subscriber base, editorial independence, and professional standards. Indeed, the recent disclosure of diplomatic cables can be viewed as having reaffirmed the importance of the American media establishment, because it is unlikely that WikiLeaks’s “data dump” would have had the public impact that it did, had the New York Times not “curated” the disclosures, drawing the attention of readers to particular facts, providing context, and commissioning supporting editorials.107 We also must not overlook the fact that the vast majority of leaks pertaining to the United States continue to be published by American news organizations. There is a simple reason for this: the protections offered to the press under the First Amendment have thus far made it unnecessary for officials to take recourse to reporters, editors, and publishers based overseas. Thus, at present, a leak of classified information that is published on the Internet is likely to be—and perhaps ought to be—treated with suspicion: one must wonder why the official responsible for the disclosure could not find an American outlet willing to report her story.
The fact that the ongoing media revolution has failed to undermine the American news establishment’s role in channeling leaks of classified information might lead one to conclude that it is not too late to enact a law regulating the use of anonymous sources. How difficult could it be to enforce such a law so long as the New York Times and the Washington Post continue to serve as the primary channels by which leaks of classified information are disseminated? However, this line of thought overlooks an important point. If the law actually began demanding that the Times and the Post justify their use of anonymous sources, then officials who are fearful of being “outed” by the ensuing investigations will in fact develop a strong incentive to channel their disclosures via the Internet (as noted previously, it is not only officials with bad motives who have reason to fear being outed by investigations). This development would then naturally lead the American public to begin according greater credibility to anonymous disclosures made via the Internet. In short, the problem that the emergence of websites like WikiLeaks and OpenLeaks poses for law enforcement is not that these outlets have already displaced (or are about to displace) the American media establishment as the primary channel by which leaks of classified information are made public, but that they could do so if the American media establishment were to come under stricter scrutiny.108
Making the Best of It
It seems unlikely, then, that we can rely on the law to compel reporters, editors, and publishers to filter anonymous disclosures on behalf of the public. For even if the proposed law is able to overcome the First Amendment challenges discussed above, the executive is unlikely to be able to enforce its provisions in the age of the Internet. Thus we have little choice but to rely on the good faith of reporters, editors, and publishers—and perhaps the skepticism of readers. This is not a cause for outright alarm, since there certainly have been cases where reporters, editors, and publishers have chosen to act responsibly. But the evidence on this count is not exactly overwhelming. As a result, we must learn to make our way in a political universe in which leaks of classified information will continue to be used not only to sound the alarm but also to further sectional interests and partisan causes. Let us now consider what we can do to make the best of a difficult situation.