United States Senate Committee on the Judiciary
Reporters' Privilege Legislation: Preserving Effective Federal Law Enforcement
September 20, 2006

Mr. Steven Clymer
Professor , Cornell Law School

Testimony of Steven D. Clymer
Professor of Law, Cornell Law School
On the Proposed Journalists' Privilege Legislation
Before the United States Senate Committee on the Judiciary
September 20, 2006

Mr. Chairman and Members of the Committee: Thank you for inviting me to appear before you today to testify about a proposed statutory "journalists' privilege" in federal trials and proceedings. I am a Professor of Law at Cornell Law School, where I teach courses on Evidence, Criminal Procedure, Terrorism and the Law, and Criminal Law. I also have been an Assistant District Attorney in Philadelphia and a federal prosecutor in both California and New York. Most recently, during a leave of absence from Cornell, I served as the Chief of the Criminal Division of the United States Attorney's Office in Los Angeles. I left full-time service with the Department of Justice
[DOJ] on June 17, 2005. I now work part-time as a federal prosecutor, on a single case. The opinions that I express are my own and should not be attributed to DOJ.

The "Free Flow of Information Act of 2006" attempts to strike a balance between two interests that are fundamental in a free society: (a) the dissemination of accurate information, including information about government operations, to the public through the news media; and (b) the truthseeking function of federal grand juries, federal courts, and federal administrative proceedings.
Although that objective is a worthy one, the proposed legislation fails to accomplish its stated goal of "guarantee[ing] the free flow of information to the public through a free and active press." Unfortunately, it nonetheless imposes considerable costs on "effective law enforcement and the fair administration of justice."

My observations fall into two categories. First, I offer the general view that the proposed legislation is unlikely to have its desired effect of persuading reluctant sources to divulge newsworthy information to journalists. The legislation sets out a privilege qualified by at least eight exceptions, some of which apply only upon satisfaction of multiple requirements and involve the application of a subjective and unstructured balancing test. Although, if enacted, this legislation would offer lawyers, judges, and law professors much to ponder and discuss, for reluctant sources deciding whether to leak information to the news media, it would provide more confusion than clarity or assurance. Confusion would not be the only byproduct. The legislation also would result in the loss of reliable, probative evidence in criminal, civil, and administrative proceedings; delay investigations; and encumber courts with additional litigation.

Second, I identify several concerns about specific features of the proposed legislation.
1. It is Unlikely That the Proposed "Free Flow of Information Act of 2006" WillAppreciably Increase the Flow of Information to the Public

When assessing the proposed legislation, it is critical to understand that the benefit, if any, of a journalists' privilege accrues at a different time than the costs. The desired benefit -- the increased flow of information to the news media and the public -- occurs, if at all, when a source who desires anonymity seeks an assurance from a journalist that a court will not or cannot compel the journalist to reveal the source's identity. For the privilege to result in an increased flow of information, it must enable journalists to honestly provide guarantees of anonymity that are sufficient to persuade a significant number of otherwise unwilling sources to leak information.

The costs come later, when federal grand juries, federal courts, federal prosecutors, criminal defendants, civil litigants, and administrative bodies are denied access to reliable and probative evidence that may have a bearing on indictment, guilt, innocence, sentencing, or liability. This loss of evidence is a direct result of the restrictions imposed on federal courts' ability to compel journalists to provide such information. Significantly, even if the proposed legislation fails to
encourage more disclosures, it still will impose costs by suppressing the truth.

The view that the proposed legislation will increase the flow of information to the news media is premised on two claims. First, there is the claim that the law in its present state -- in which some federal courts and many states recognize a "journalists' privilege" of some sort -- deters sources from providing information because they fear that a federal court later will compel the journalist to whom they made disclosure (or the journalist's employer) to identify them. Second, there is the claim that the proposed legislation, if it were to become law, would remove that deterrent and prompt a significant number of reluctant sources to make disclosures.

Both of these claims merit scrutiny. As to the first claim, it bears mention that even without a federal statutory journalists' privilege, many sources provide information, including classified information, to journalists. For example, there have been recent and prominent leaks to the news media on sensitive topics such as the National Security Agency's [NSA] warrantless wiretapping program and the Central Intelligence Agency's [CIA] overseas detention and interrogation of al Qaeda operatives. Significantly, these leaks occurred in the face of widespread news coverage of the jailing of former New York Times reporter Judith Miller, coverage that made clear that there is little or no federal protection of the anonymity of news sources. Those who were involved in the NSA and CIA leaks had to have been aware that federal law usually does not permit journalists to conceal their sources from federal grand juries and courts, and that the journalists to whom they leaked the classified information could be compelled to identify their sources. Despite that, they leaked the information.

Information flows to journalists despite the absence of a federal statutory journalists' privilege because sources correctly perceive that there is only a very small risk that the government or a private litigant will seek and a court will compel disclosure of their identity, or because they have reasons for making disclosure that outweigh perceived risks. In addition, some journalists may promise sources that they will disobey a court order compelling them to testify and go to jail rather than disclose a source. If a prospective source believes such an assurance, the existence or nonexistence of a federal shield law may not affect the decision to leak information.

To be sure, however, there likely are journalists who are unwilling to disobey a court order and sources who are unwilling to disclose information without a guarantee of confidentiality. A true measure of the proposed legislation is its effect when a source who requires such a guarantee contacts a journalist who will provide it only if the law safeguards him from being compelled to disclosure his source.

This leads to examination of the second claim: that the proposed legislation will result in an increase in the flow of information to the public through the news media. That claim does not withstand scrutiny. As things now stand, without a federal statutory journalists' privilege, a journalist (who is unwilling to disobey a court order) cannot assure confidentiality. He can, however, tell a potential source that: (a) it is not common for journalists to receive subpoenas for source information; (b) if a DOJ attorney seeks to learn the source's identity, the request will be subject to rigorous internal DOJ scrutiny detailed in federal regulations, scrutiny that rarely results in approval of a subpoena for source information; (c) if the journalist is subpoenaed, there sometimes are ways to resolve subpoenas without revealing sources; (d) if the journalist is subpoenaed, the subpoena cannot be resolved, and the litigation is in state court, the journalist may have a privilege to refuse to disclose the source's identity, depending on the specifics of the law that applies in the state in which the litigation occurs; and (e) if the litigation is in federal court, the journalist may or may not be compelled to disclose the source's identity, depending on the specifics of the law that applies in the federal circuit in which the litigation occurs and whether the subpoena was issued in a criminal or a civil case. In the event that the source asks for additional details, and if the journalist knows details, the journalist can explain the potentially applicable state shield law and federal decisional law. Ultimately, however, with so many variables at issue, the journalist cannot truthfully provide the source with a clear-cut assurance that a court will not compel disclosure or a meaningful estimate of the probability of court-ordered disclosure of the source's identity.

The proposed legislation does nothing to eliminate that uncertainty. Instead, it creates a privilege subject to multiple possible exceptions. Different rules apply when government attorneys seek disclosure in criminal investigations and proceedings [Section 4(b)]; when criminal defendants seek disclosure [Section 5(b)]; when litigants in civil or administrative actions seek disclosure [Section 6(b)]; when journalists participate in criminal or tortious conduct [Section 7(A)]; when journalists witness criminal conduct [Section 7(B)]; when disclosure is "reasonably necessary" to prevent death, kidnaping, or serious bodily injury [Section 8]; when disclosure "would assist in preventing" an act of terrorism or specified harm to national security [Section 9(a)(1)]; and when the source reveals "properly classified information" to which he had "authorized access" [Section 9(a)(2)]. The exceptions apply only upon satisfaction of various requirements and, in some cases, only if a court determines that "nondisclosure of the information [about the source's identity] would be contrary
to the public interest, taking into account both the public interest in compelling disclosure and the public interest in newsgathering and maintaining the free flow of information to citizens."

At the moment of truth -- when the source seeks an assurance of anonymity -- it will not be clear which exception or exceptions, if any, may ultimately apply. For example, in the case of a source deciding whether to leak classified information, it is possible, depending on circumstances, that a court later will apply one or more of the exceptions contained in Sections 4(b), 5(b), 6(b), 9(a)(1), and 9(a)(2),1 all of which contain different requirements. It will be impossible to predict in advance what legal test will apply if the journalist is subpoenaed, much less whether a court will order disclosure. If the proposed legislation were to become law, no journalist could honestly guarantee confidentiality unless he was willing to disobey a court order to disclose a source's identity and go to jail.2 At most, the legislation would enable journalists to tell reluctant sources that if a subpoena is issued, and if the subpoena involves a federal matter, and if the subpoena is not resolved without litigation, the new federal law makes it marginally less likely that the reporter will be compelled to testify than it was before the new law. The bottom line is this: If journalists understand the proposed legislation and are honest with their sources about it, they cannot offer the sort of robust guarantees of confidentiality that likely are necessary to cause an appreciable increase in disclosures of newsworthy information.

Although benefits from the legislation in the form of greater information flow are speculative and unlikely, the costs are tangible and unavoidable. The statutory privilege will deny federal grand juries access to probative evidence that bears on decisions whether to indict; jeopardize criminal prosecutions by foreclosing prosecutors from discovering and presenting in court crucial incriminating evidence; impair criminal defendants from gaining access to evidence that may exculpate them; and deny civil litigants information that they otherwise could discover and prove. Even in cases in which those seeking access to the truth are able to satisfy a court that one of the
exceptions to the proposed privilege applies, that effort will require litigation, delay, and the expenditure of considerable resources.

Unfortunately, any effort to increase the certainty of journalists' guarantees to sources would require the elimination of exceptions, thus resulting in the loss of even more probative evidence and increasing the costs of the proposed legislation. Ultimately, the price that must be paid for an effective privilege -- meaning one that has few or no qualifications and thus will provide real assurance to reluctant sources when they decide whether to leak information -- likely is too high for the federal criminal and civil justice systems to bear.

2. The Proposed "Free Flow of Information Act of 2006" Imposes Requirements on Federal Prosecutors in Criminal Matters That Do Not Apply to Civil Litigants or Criminal Defendants
A comparison of the various exceptions to the proposed privilege reveals that Section 4(b), which applies to government attorneys seeking source information "in any criminal investigation or prosecution," imposes more rigorous requirements than Section 5(b),which applies to requests by criminal defendants, and Section 6(b), which applies to requests in civil or administrative actions. For example, only government attorneys in criminal matters must demonstrate, among other things, that the information sought "is critical to the investigation or prosecution, particularly with respect to directly establishing guilt or innocence." Criminal defendants need show only that the information sought "is relevant [not critical] to the question of guilt or innocence." Civil litigants need establish only that the evidence sought "is critical to the successful completion of the civil action," without the added hurdle of "particularly with respect to directly establishing guilt or innocence." Similarly, under the proposed legislation, federal prosecutors, but not criminal defendants, must, to the extent possible, limit subpoenas to journalists to "verification of published information" and "surrounding circumstances relating to the accuracy of the published information." One apparent feature of the higher burden on criminal prosecutors is that they, unlike other litigants, are not entitled to judicial compulsion to obtain source information if it constitutes or will reveal only circumstantial evidence, specifically, evidence that does not "directly establish[ ] guilt or innocence."3

One can imagine a situation in which a journalist has one or more sources of information about events that later result in both criminal and civil litigation. Under the proposed legislation, it is possible that a court would compel the journalist to reveal source information to a criminal defendant who needs it to exonerate himself and to a civil litigant who needs it to prove his case, but, at the same time, refuse to compel the same journalist to reveal the same source information to a grand jury or prosecutor, despite their need for it as part of a criminal investigation or prosecution.

Although there may be unique constitutional concerns with limiting criminal defendants' access to source information, it is difficult to justify legislation making it more difficult for federal prosecutors to obtain information than criminal defendants and civil litigants. If anything, federal prosecutors should be subject to fewer, if any, constraints on their efforts to obtain source information. First, federal prosecutors seek such information as part of their obligation to enforce federal criminal law, an objective more important to public welfare than a civil litigant's personal lawsuit. Second, unlike criminal defendants and private civil litigants, federal prosecutors must comply with rigorous internal Department of Justice regulations set out in 28 C.F.R. 50.10 before seeking access to source information from journalists. Third, DOJ has a powerful institutional interest in carefully limiting its issuance of subpoenas to journalists because abuses could trigger legislation restricting DOJ's discretion in that area. In contrast, neither criminal defendants nor private civil litigants are subject to the regulations or any centralized oversight of their decisions to seek source information from journalists. Nor are they "repeat players" who are constrained by concerns about the long-term effects of overly aggressive use of subpoenas to members of the news media.

3. The Section 4(b) Exception Requires Disclosure of Sensitive Investigative Informationand Involves Courts in Prosecutorial Decision-Making

The proposed exception for source information by government attorneys in criminal investigations and prosecutions in Section 4(b) requires federal courts to make determinations that "the information sought is critical to the investigation or prosecution"; the government "has exhausted reasonable alternative sources of information"; and that "nondisclosure of the information would be contrary to the public interest." The exception further provides that the journalist (or communications service provider) be given "notice and an opportunity to be heard" on the applicability of the exception. In addition, by implication, the proposed legislation requires that prosecutors turn over to subpoenaed journalists the information necessary to persuade courts that the above-described requirements are satisfied.4

The Section 4(b) exception thus would force the government to reveal to both the court and the subpoenaed journalist or media outlet evidence demonstrating the significance of the source information to the overall investigative strategy or prosecution, the nature and extent of the prosecution's other evidence, the investigative steps that it has taken relative to the desired source information, and the overall scope and importance of its prosecution. Such disclosures of law enforcement sensitive information could jeopardize ongoing investigations and prosecutions and are contrary to federal laws requiring the secrecy of investigative information such as matters occurring before grand juries, non-consensual interceptions of communications, and tax-related information.

In addition, Section 4(b) empowers federal courts to scrutinize government investigative strategy and second-guess prosecutorial judgments about the need for source information or the role of such information in the overall investigation. It is by no means clear that courts have the competence or constitutional authority to engage in those endeavors.

4. The Balancing Tests in Sections 4(b)(5), 5(b)(4), 6(b)(5), and 9(a)(2) Are Subjective and Beyond the Expertise of Federal Courts

Sections 4(b)(5), 5(b)(4), 6(b)(5), and 9(a)(2) of the proposed legislation all require that courts conduct so-called "balancing tests" when deciding whether an exception to the statutory privilege applies. In Sections 4, 5 and 6, the test requires that a court determine whether "nondisclosure of the [source] information would be contrary to the public interest, taking into account both the public interest in compelling disclosure and the public interest in newsgathering and maintaining a free flow of information to citizens."5

By mandating judicial consideration of "the public interest in newsgathering and maintaining a free flow of information to citizens," this test appears to require that federal courts assess the general effect that an order requiring a journalist to disclosure a source's identity will have on future newsgathering. It is not clear how a federal judge would have access to sufficient information to intelligently make such an assessment, at least absent full hearings involving, among others, media experts and various confidential media sources, all of whom could provide their views on the likely impact of a single federal court's disclosure order on future source and journalist behavior. Although federal courts certainly are competent to assess the effect in the case before them of the admission or exclusion of evidence, they have no special insight into or expertise about the impact that a disclosure order will have on future newsgathering. In short, this test calls for a policy determination -- one that is within the expertise of Congress -- not a judicial determination.

Similarly, in cases involving disclosure of classified information, Section 9(a)(2) requires that a federal court determine whether the "unauthorized disclosure has significantly harmed national security in a way that is clear and articulable and the harm caused by the unauthorized disclosure of such information outweighs the value to the public of the disclosed information." Such a determination may require the government to disclose to the court large amounts of related classified information so that the court will have a full understanding of all of the national security implications and concerns triggered by the unauthorized leak of classified information. Even if a
federal court were fully informed in this way, it is by no means clear that they have the expertise or constitutional authority to balance threats to national security against the benefits of disclosure.Such national security matters are the responsibility of the Executive Branch .

Supporters of legislation creating a journalists' privilege often make a collateral argument -- that a federal journalists' privilege statute will promote more uniform treatment across the federal system. Although federal Courts of Appeal have reached differing conclusions about the existence and scope of a journalists' privilege, it is unlikely that the proposed legislation will promote uniformity. Most notably, the balancing tests provides no real guidance to federal courts and invite idiosyncratic decisions based on the subjective predilections of individual federal judges.

5. The "Foreign Power" and "Agent of a Foreign Power" Carve-Out From the Definitionof "Journalist" is Problematic

In an apparent response to DOJ concerns that media outlets related to terrorist organizations would be privileged under the proposed legislation, an amendment provides that the term journalist "shall not include any person who is a 'foreign power' or 'agent of a foreign power'" under the Foreign
Intelligence Surveillance Act of 1978 [FISA]. Although well intentioned, this carve-out is problematic.

When the government seeks to prove that a person or entity is a "foreign power" or an "agent of a foreign power," it typically does so using classified evidence as part of an application to the FISA court for an order for FISA electronic surveillance or physical searches. If the government wants to employ such surveillance and searches, it strives to keep secret from the target of the FISA coverage that it is under investigation. If, as the amendment provides, the government were required to allege and prove "foreign power" or "agent of a foreign power" status to a court in order to avoid application of the journalists' privilege, it may be forced to reveal classified information or tip its hand to a FISA target.

A better approach might be to tie the carve-out to the designation of a journalist or media outlet as a "Specially Designated Global Terrorist" [SDGT] under Executive Order 13224, and perhaps include non-designated media outlets that are associated with designated SDGTs. The President, the Secretary of the Treasury, the Secretary of State, and the Attorney General are involved in the process by which persons and entities are designated as SDGTs based on their status as or affiliations with terrorist organizations and terrorists. Because SDGT designations are made public, and can be based on undisclosed classified evidence, use of SDGT status or association with an SDGT as the trigger for the carve-out will not jeopardize classified information or otherwise imperil national security.

6. The Section 10 Requirement of a Promise or Agreement of Confidentiality is Problematic

Section 10 conditions application of the journalists' privilege on "a promise or agreement of confidentiality made by a journalist." It is unclear what this condition requires. Under the proposed legislation, a journalist cannot honestly provide a blanket "promise or agreement of confidentiality" to a source because of the numerous exceptions to the legislation (unless, of course, the journalist is prepared to ignore a court's order to disclosure source information and go to jail). Thus, Section
10 cannot mean that application of the journalist's privilege is dependent on the making of an unqualified promise or agreement. A more plausible interpretation is that the privilege is triggered only by a promise or agreement of confidentiality absent a court order requiring disclosure. If so, the legislation should state this explicitly.

An additional difficulty arises because Section 10 is silent on the issue of waiver. In several recent high profile cases, including the Judith Miller incident, despite journalists' claims to the contrary, sources of information claimed that they either never requested confidentiality or later waived it.6
Those cases reveal that it should not be left to journalists to decide whether the privilege applies or whether a source has waived confidentiality. As is the case with other privileges, such as the attorney-client privilege, a court, not a journalist, should determine the validity of an assertion of the privilege and whether a source has waived confidentiality. Any legislation should so provide.

1 Indeed, Section 7 of the Act provides that Sections 1 through 6 and 8 through 10 apply to illegal disclosures of documents and information.

2 Section 10 of the proposed legislation renders the statutory privilege inapplicable unless there is a "promise or agreement of confidentiality made by the journalist." This section apparently is intended to avoid conferring a privilege when there no assurance of confidentiality has been given. As noted in the text, however, if the legislation were passed, a journalist who was honest with a source could at best offer only a qualified promise or agreement of confidentiality. (This issue is discussed later in the text.) Because, under the proposed legislation , it would cost the journalist nothing to make such a promise, journalists likely would provide such assurances even if the source might be persuaded to provide information without a promise.

3 If the words "particularly with respect to directly establishing guilt or innocence," which appear only in the Section 4(b) exception, are meant to limit federal prosecutors and grand juries to direct evidence of guilt or innocence, as opposed to circumstantial evidence, the limitation is contrary to judicial understanding of the two forms of evidence and present federal practice. Federal courts do not distinguish between the probative value of direct and circumstantial evidence. See, e.g., United States v. Ramirez-Rodriquez, 552 F.2d 883, 884 (9th Cir. 1977). Indeed, some federal courts instruct jurors that the two forms of evidence merit equal treatment during deliberations. See, e.g., Ninth Circuit Model Jury Instruction 1.6 ("The law permits you to give equal weight to both [direct and circumstantial evidence], but it is for you to decide how much weight to give to any evidence.").

4 Section 9 provides for ex parte and in camera judicial review only in connection with the Section 9(a) exception involving leaks that threaten national security interests, thereby suggesting that no similar procedures are available under Section 4(b).

5 It is worth noting application of the balancing test in Section 5(b) could, in some cases, result in violations
of criminal defendants' due process rights to exculpatory and mitigating evidence. The Section 5(b) exception appears to permit a federal court to deny a criminal defendant access to such evidence, no matter how relevant and probative, if the court decides that the public interest in newsgathering outweighs the public interest in disclosure of the source

6 As I described in earlier testimony:
In Providence, Rhode Island, despite a court order, WJAR-TV reporter Jim Taricani refused to disclose the identity of a source who had given him an FBI videotape showing a government official accepting a bribe. After Taricani had been convicted of criminal contempt, his source came forward and claimed that he never had asked Taricani to keep his identity secret. Taricani disputes that claim.
. . . New York Times reporter Judith Miller refused to comply with a court order requiring her to testify in a federal grand jury about a source. After she had been held in contempt and spent 85 days in federal custody, she claimed that her source finally had given her permission to reveal his identity.
But, both the source and his lawyer provided a different version of events, claiming that they had communicated such approval to her attorney a year earlier.