The State Secret Protection Act: Unnecessary, Unconstitutional, and Undemocratic

Testimony before
Subcommittee on the Constitution, Civil Rights,
and Civil Liberties,
Committee on the Judiciary,
United States House of Representatives

June 4, 2009

The misnamed State Secret Protection Act of 2009 (H.R. 984) is dangerous, in terms of both its effect on national security and the violence it would do to the constitutional separation of powers. Congress should be aware of the following key points:

• The state secrets privilege has a 200 year history in the United States and has existed in essentially its present form for 135 years. It has been used by every president since Lyndon Johnson, up to and including President Barack Obama.
• There is absolutely no evidence of abuse of the state secrets privilege. Data from 1954 through 2008 show that its use is rare. In reported opinions, the privilege was asserted just seven times in 2007, and three times in 2008.
• There is no evidence that the state secrets privilege is being used more frequently than in the past or in cases where it is not needed. There is no evidence that it is being used to stifle cases on political grounds. There is no evidence that judges are unduly deferential to the executive when it is asserted; the trend is actually in the opposite direction.
• The State Secret Protection Act would force the government to admit highly classified secrets, such as the identities of spies, in the course of litigation, putting national security at risk.
• The State Secret Protection Act would give activists a "heckler's veto" over many national security programs created by the democratic branches of government.
• The State Secret Protection Act attempts to transfer powers clearly assigned to the President to judges, in violation of the Constitution. It is unconstitutional.
• The State Secret Protection Act is a cynical attempt by Congress to duck tough decisions in the national security arena--where bad decisions can have catastrophic consequences--by passing the buck to the courts.
• The state secrets privilege is only one of several "immunities" that can bar litigation altogether. For example, courts have cited the Speech or Debate Clause to dismiss suits against Members of Congress and other legislators involving invasion of privacy, defamation, wiretapping, incitements to violence, age, race, and sex discrimination, retaliation for reporting sexual discrimination, and larceny and fraud.
• The modern application of the privilege was defined in a 1953 case, U.S. v. Reynolds. The Reynolds framework carefully balance the sometimes harsh results of the state secrets privilege--the exclusion of relevant evidence or dismissal of a claim altogether--with the genuine needs of U.S. national security.
• Seven separate requirements, including Department of Justice review and "personal consideration" by high-ranking federal officials, ensure that the state secrets privilege is used only when necessary to protect national security.

My name is Andrew Grossman. I am Senior Legal Policy Analyst at The Heritage Foundation. The views I express in this testimony are my own, and should not be construed as representing any official position of The Heritage Foundation.

My testimony this afternoon concerns the misnamed State Secret Protection Act of 2009 (H.R. 984, "SSPA"), which would regulate, and in some cases prohibit, the federal government's invocation of the state secrets privilege to prevent the disclosure of sensitive national security information and programs in civil litigation. The SSPA is dangerous, in terms of both its direct effect on national security and the violence it would do to the constitutional separation of powers, and I thank the Subcommittee for holding this hearing and considering my testimony on the consequences of this legislation.

As I will explain, Members of Congress should be wary of the SSPA for three reasons. First, it is unnecessary because there is no evidence of abuse of the state secrets privilege. Second, it raises serious constitutional concerns, particularly as regards the Article II duties assigned to the President. Third, the legislation can be seen as a cynical attempt by Congress to evade its constitutional duty to make tough decisions about our national security, and this abdication puts the nation's safety at risk. For these reasons, Congress should resist succumbing to pressure from political partisans and activists to force the disclosure of closely held national security information in civil lawsuits.

I. No Evidence of Abuse

On the terms of the justifications offered by its supporters, the SSPA is unnecessary. Contrary to repeated claims by civil liberties groups and others, recent use of the state secrets privilege is not different in kind or quantity than in the past. Despite more attention paid to the privilege in recent years--largely as a result of political opposition to the policies of the George W. Bush Administration and their embrace by the Obama Administration--the strong accountability mechanisms built into it continue to guarantee that it is not overused or otherwise abused. To understand this point requires some understanding of the privilege's historical pedigree.

Though usually discussion of the state secrets privilege begins with the Supreme Court's 1953 decision in United States v. Reynolds[1], that approach presents a pinched view of the privilege's history and scope--and perhaps this is deliberate. The privilege's first acknowledgement in the law of the United States--or at least the first in written reports uncovered by modern scholars--is typically accredited to Chief Justice John Marshall, who referred obliquely to executive privilege in Marbury v. Madison and, while riding circuit, to an intelligence-based privilege in the trial of Aaron Burr for treason. In the former case, Marshall allowed that while Attorney General Levi Lincoln could not be "obligated" to disclose "any thing [that] was communicated to him in confidence," the fact whether the disputed commissions had been found in the office of the Secretary of State the disposition of the commissions "could not be a confidential fact," thereby relegating Marshall's brief description of the privilege to dicta.[2]

Marshall elucidated that privilege's application to secret communications and intelligence while presiding over the treason trial of Aaron Burr. Burr sought to admit a letter from General James Wilkinson, an essential witness against him, to President Jefferson, over the government's objection that the letter "contains matter which ought not to be disclosed."[3] The balancing of a party's need with the necessity of government secrecy in certain matters "present a delicate question," explained Marshall--one "which, it is hoped, will never be rendered necessary in this country."[4] Yet again, Marshall sidestepped the need for such balancing, because "certainly nothing before the court which shows that the letter in question contains any matter the disclosure of which would endanger the public safety."[5] But "If it does contain any matter which it would be imprudent to disclose, which it is not the wish of the executive to disclose, such matter...will, of course, be suppressed."[6] In such a case, said Marshall, "much reliance must be placed on the declaration of the president."[7] And, Marshall made clear, "The propriety of withholding it must be decided by [the President], not by another for him."[8] Though the issue was made moot when Jefferson, pressed to make a decision, consented to admission of the letter "excepting such parts as he deemed he ought not to permit to be made public,"[9] this formulation, as well as its rationale, would greatly influence the Reynolds court.

The issue would next arise in U.S. courts in the matter of Totten v. United States, the Supreme Court's 1875 decision which, though brief, merits careful consideration. Totten, heir to one William Lloyd, brought an action against the United States claiming that Lloyd had entered into a contract with President Lincoln to ascertain troop placements in the South and "other information as might be beneficial" to the North during the Civil War.[10] Such a contract would ordinarily be binding, Justice Field explained for the Court, but not in the circumstances presented by this particular case: "The service stipulated by the contract was a secret service; the information sought was to be obtained clandestinely, and was to be communicated privately; the employment and the service were to be equally concealed."[11]

On the basis of these facts, the Court propounded two rules, quite intertwined, one narrow and one broad. The narrow rule, that offered in most discussions of the case,[12] is simply that no suit may require disclosure of a spy's employment by the government. This was not framed as a privilege but as an absolute bar to litigation. Without such a bar, "whenever an agent should deem himself entitled to greater or different compensation than that awarded to him, the whole service in any case, and the manner of its discharge, with the details of dealings with individuals and officers, might be exposed, to the serious detriment of the public."[13] As the Court explained, "A secret service, with liability to publicity in this way, would be impossible."[14] It is thus an implied term of such contracts that the very act of suing for compensation is a breach of contract that defeats recovery.

Yet it is the broader rule, though less discussed by academics, which has proven more influential to the development of the law. Put simply:

It may be stated as a general principle that public policy forbids the maintenance of any suit in a court of justice, the trial of which would inevitably lead to the disclosure of matters which the law itself regards as confidential, and respecting which it will not allow the confidence to be violated.[15]

Among these confidential matters are those typically covered by the various privileges, such as between a husband and wife or patient and physician. But, said the Court, "Much greater reason exists for the application of the principle to cases of contract for secret services with the government, as the existence of a contract of that kind is itself a fact not to be disclosed."[16] Thus, in some circumstances relating to national security, lawsuits that would inevitably disclose state secrets, such as a spy relationship, are simply barred, because their very existence "is itself a fact not to be disclosed" and disclosure would be "a detriment to the public."[17] The Court would affirm this rule's vitality in Reynolds[18] and subsequently reaffirm it in a 2005 case.[19]

By 1875, then, the basic contours of the law were set. In cases where parties sought to subpoena or otherwise introduce state secrets, the courts would exclude materials that the executive determined to be "imprudent" to disclose, in effect giving the executive a privilege to protect certain information from disclosure. And lawsuits that, at their core, concern secret government relationships and activities, such as spy contracts, would simply be barred as non-justiciable. Even at this early date, the state secrets "privilege" was not strictly a privilege in every case; sometimes it would be a "threshold question" that could defeat a claim at the outset of the case.

Understood in this historical context, Reynolds was less a revolution than a refinement, one that began the task of regularizing invocation and application of the privilege with respect to modern civil procedure. The case was brought under the Federal Tort Claims Act (FTCA) by the widows of three civilians killed in the crash of an Air Force bomber testing "secret electronic equipment."[20] The government refused to disclose its post-accident report, arguing that disclosure would, according to an affidavit of the Judge Advocate General of the Air Force, hamper "national security, flying safety, and the development of highly technical and secret military equipment."[21] The district court resolved the case in the plaintiffs' favor after the government declined to present the report for ex parte, in camera inspection.[22] The Third Circuit affirmed, holding that the FTCA had waived any privilege that the government might have had.[23]

In an opinion by Justice Vinson, the Supreme Court expounded a new framework for invocation of the privilege, drawing freely from legal precedent:

The privilege belongs to the Government and must be asserted by it; it can neither be claimed nor waived by a private party. It is not to be lightly invoked. There must be formal claim of privilege, lodged by the head of the department which has control over the matter, after actual personal consideration by that officer. The court itself must determine whether the circumstances are appropriate for the claim of privilege, and yet do so without forcing a disclosure of the very thing the privilege is designed to protect.[24]

Of this final step, the Court provided some elucidation by analogy to the privilege against self-incrimination, as described by Justice Marshall during the Burr trial. A court should consider "all circumstances of the case" in determining whether "there is a reasonable danger that compulsion of the evidence will expose military matters which, in the interest of national security, should not be disclosed."[25] But once the court has reached that determination, "the occasion for the privilege is appropriate, and the court should not jeopardize the security which the privilege is meant to protect by insisting upon an examination of the evidence, even by the judge alone, in chambers."[26] Courts thus have significant flexibility and discretion in determining whether the government's use of the privilege is appropriate, but their inquiry is limited, with great focus, to that question alone. It also recognizes that in some instances, particular evidence will pose such a significant and obvious danger to national security that even in camera review is inappropriate. (After all, among other concerns, very few judges review evidence in secured rooms, encased in reinforced concrete and with doors that seal, designed to prevent eavesdropping or outright theft.[27])

The necessity of the evidence to the party seeking to admit it is also a relevant consideration. "[T]he showing of necessity which is made will determine how far the court should probe in satisfying itself that the occasion for invoking the privilege is appropriate."[28] The district court, explained the Supreme Court, should have been satisfied with the government's assertion alone, for the plaintiffs' need was tempered by the availability of alternative evidence on the same factual allegations. This is not, however, a balancing test, weighing necessity against risk. The Court was careful to explain that no showing of necessity, no matter how great, may overcome a determination that the privileges was properly asserted.[29] At most, great necessity may prompt a judge to scrutinize the basis of the assertion more closely.

It is worth, at this point, a brief historical detour. Some have argued, in recent years, that the declassified accident report proves that the privilege was asserted unnecessarily and improperly in Reynolds--in other words, that there was no risk at all that giving the report to the plaintiffs or court would have risked disclosing national security secrets. Courts that have examined this issue directly, however, reject that claim. After finding the declassified report, heirs of those killed in the crash brought suit against the United States in 2003, alleging that the Air Force had misrepresented the nature of the information contained in the report and thereby committed fraud on the court by improperly asserting the privilege.[30] A district court and the Third Circuit directly considered the issue of whether the government officers asserting the privilege had committed perjury; both courts rejected the accusation.[31] The courts found that the report contained extensive technical information about the B-29 bomber, as well as details about the electronic equipment (a classified experimental radar system) that was being tested.[32] As the district court explained, "Details of flight mechanics, B-29 glitches, and technical remedies in the hands of the wrong party could surely compromise national security."[33] In short, in this much-assailed case, history has confirmed that the assertion of the privilege was appropriate.

Since Reynolds, the courts have done little more than flesh out its approach. One notable development was judicial embrace of the analogy of foreign intelligence gathering to the construction of a mosaic: "Thousands of bits and pieces of seemingly innocuous information can be analyzed and fitted into place to reveal with startling clarity how the unseen whole must operate."[34] This view counsels strong deference to the executive's assertion of the privilege because "What may seem trivial to the uninformed [e.g., a judge], may appear of great moment to one who has a broad view of the scene and may put the questioned item of information in its proper context."[35] Another notable development was the taxonomy of possible dispositions of a case in which the privilege has been asserted[36]; these are incorporated into the discussion below. Most importantly, the courts built up a body of case law that would provide guidance in evaluating assertions of the privilege.

The Reynolds framework can be seen as a deliberate effort to balance the harsh reality of the state secrets privilege--the exclusion of relevant, and perhaps determinative, evidence or dismissal of a claim altogether--with the genuine needs of U.S. national security. It ensures that assertion of the privilege comports with procedural due process as it is practiced today and, to the extent it intrudes on substance, provides a check against abusive assertions. This dual nature--protecting procedural and substantive rights--is evident in the long list of requirements and protections that, post-Reynolds, must be satisfied to ensure that the privilege is not "lightly invoked."

First of all, the privilege may be invoked only by the United States, and not by a private litigant. This requirement alone greatly circumscribes the potential for abuse, as relatively few civil cases touch upon national security or classified matters.

Second, the privilege may not be asserted by a line attorney or even supervising attorney but only by the head of the department that has control over the matter, usually an agency head. This requirement ensures that the decision to assert the privilege will be subject to more extensive review, by more individuals and at higher levels of responsibility. It is analogous to the similar requirements in the Foreign Intelligence Surveillance Act (FISA) that high-ranking officials, such as the Attorney General and National Security Advisor, certify that applications made to the FISA Court meet the exacting requirements of the law.[37] Indeed, recognizing the value of independent certifications made by high-ranking officials, paired with precise judicial review, Congress greatly increased the FISA's reliance on this mechanism in the FISA Amendments Act of 2008.[38]

Third, it must be formally invoked. This requires a separate determination of the propriety of invoking the privilege by the Department of Justice, which is charged with conducting litigation for the United States and supervising litigation carried out by the government.[39] An agency head, acting alone, generally cannot assert the privilege without the concurrence of the Department of Justice. This ensures not only additional levels of review and accountability, but also that the proposed assertion of the privilege will be evaluated by legal and security specialists who will ensure that the United States uses the privilege in a consistent fashion that promotes national security over any agency's parochial interests. Department of Justice lawyers are also especially attentive to identifying and rejecting weak claims that might ultimately undermine the privilege, providing another check against overuse of the privilege and its assertion in cases in which it is not necessary.

Fourth, the department head asserting the privilege must undertake "actual personal consideration" of the matter, just as Justice Marshall ruled was required of the President in the Burr trial. For a high-ranking official, typically carrying great responsibility, this is a significant and potentially burdensome requirement, demanding that he or she personally review the evidence or matter at issue and produce a declaration (or several in cases where classified and unclassified declarations are required) explaining, to the satisfaction of the court, why disclosing the evidence at issue would endanger national security. Typically, both requests and declarations will be reviewed by agency counsel and mid-level officials.[40] Only then are declarations signed and filed--under the penalty of perjury.

Fifth, after many levels of executive-branch review, the "court itself must determine whether the circumstances are appropriate for the claim of privilege"--that is, whether the government has demonstrated that there is a "reasonable danger" disclosure would harm national security.[41] Reynolds counsels that the privilege is not to be "lightly accepted" and that the showing of necessity of the party seeking to compel the evidence "will determine how far the court should probe" in determining whether the privilege is appropriate.[42] This inquiry may even include examination of the evidence at issue in camera.[43] Only when a claim is not supported by necessity will assertion of the privilege, with nothing more, suffice to invoke it.

That, however, is a rare occurrence because, as explained by Carl Nichols, former Deputy Assistant Attorney General in the Civil Division of the Department of Justice, the government's disclosures to the court are typically extensive:

In making its determination, moreover, a court often reviews not just the public declarations of the Executive officials explaining the basis for the privilege, but also classified declarations providing further detail for the court's in camera, ex parte review. One misperception about the state secrets privilege is that the underlying classified information at issue is not shared with the courts, and that the courts instead are simply asked to dismiss cases based on trust and non-specific claims of national security. Instead, in every case of which I am aware, out of respect for the Judiciary's role the Executive Branch has made available to the courts both unclassified and classified declarations that justify, often in considerable detail, the bases for the privilege assertions.[44]

In a recent opinion, in a case that aroused no little controversy due to the government's assertion of the state secrets privilege, the Ninth Circuit professed itself "satisfied that the basis for the privilege is exceptionally well documented."[45] Among the evidence filed by the government for in camera review was the complete document which it sought to protect from disclosure.[46] In reaching its decision to affirm the exclusion of the document, the court relied on "[d]etailed statements," including classified information, that "underscore that disclosure of information concerning the Sealed Document and the means, sources and methods of intelligence gathering in the context of this case would undermine the government's intelligence capabilities and compromise national security."[47]

Sixth, the court, if it upholds assertion of the privilege, must decide what effect that decision has on the case before it. Assertion of the state secrets privilege does not, in theory and in fact, necessarily result in the dismissal of a case. As in Reynolds, the case may be able to proceed, just without the privileged evidence. In others cases, where the evidence is crucial, it will not.

This is no different than the application of any other privilege that results in the exclusion of evidence. For example, the attorney-client privilege protects communications between criminal defendants and their lawyers that would be extremely to government prosecutors; in some instances, without this evidence, prosecutors are unable to bring charges. Another example is the Speech or Debate Clause, which grants Members of Congress a testimonial privilege under which they "may not be made to answer questions" no matter the gravity of the claim involved.[48] It should not be controversial, then, when cases are not allowed to proceed for the same reasons that apply in other contexts. This generally occurs when, once the privileged evidence has been excluded, the plaintiff is simply unable to establish a prima facie case. This is the same as summary judgment following invocation of the doctor-patient or attorney-client privilege.

But the state secrets "privilege" is, as described above, sometimes more than a privilege, because it protects against the disclosure of secret facts, rather than just the use of certain evidence in court. For example, it is no violation of the doctor-patient privilege to prove the factual matters confessed to a psychiatrist by other means--e.g., an invoice or an email; under the state secrets "privilege," however, the very facts themselves may be off-limits. This would include, for example, lawsuits based on covert espionage agreements.[49] In the Reynolds court's words, cases are simply non-justiciable when "the very subject of the action...was a matter of state secret."[50] Or as formulated in Totten: "[P]ublic policy forbids the maintenance of any suit in a court of justice, the trial of which would inevitably lead to the disclosure of matters which the law itself regards as confidential."[51]

This is, as in Totten and Tenet v. Doe, a complete bar on litigation. It is also, as the Supreme Court explained in Tenet, a "threshold question," like abstention, that a court may resolve before it even addresses jurisdiction.[52]

Yet this seemingly harsh result is not unusual in the law. For example, the Speech and Debate Clause, another of the Constitution's means to affect the separation of powers[53], renders Members of Congress, as well as their staff and invited witnesses, completely "immune from suit" for a wide variety of conduct that is "within the sphere of legitimate legislative activity."[54] On this ground, the courts have dismissed claims of invasion of privacy,[55] slander and libel,[56] civil rights violations,[57] wiretapping,[58] incitements to violence,[59] violations of First Amendment rights,[60] age discrimination,[61] racial discrimination,[62] sexual discrimination,[63] retaliation for reporting sexual discrimination,[64] larceny and fraud,[65] and McCarthyism.[66] Qualified immunity, as well, has a similar effect, shielding government officials from immunity for violations of civil rights and ending cases before the plaintiff has had an opportunity to conduct discovery.[67] As the Supreme Court has explained, this is so because "broad-ranging discovery and the deposing of numerous persons...can be peculiarly disruptive of effective government."[68]

In addition, dismissal or summary judgment may be mandated when the assertion of the privilege denies the defendant a complete defense to the claim.[69] This remedy may be available only when the court, through its review of affidavits and other materials to resolve the privilege claim, is also satisfied that the defense is availing.[70] Any other result "would be a mockery of justice," observed one court.[71]

Thus, outside the core of the state secrets privilege--that is, lawsuits specifically targeted at national security secrets--a judge exercises his or her usual discretion in determining whether a case will proceed, providing yet another procedural check on assertion of the privilege. Only when a lawsuit moves from the periphery to the core of clandestine operations is this discretion limited--for example, wholesale challenges to government intelligence programs. This is as it should be, considering the purpose of the privilege. The objects of these suits should usually be pursued, if at all, through the political process.

Seventh and finally, as with other privileges, assertions of the state secrets privilege are appealable and are usually reviewed de novo by the courts of appeal.[72] This means that the appellate court accords the trial court's application of the standard no deference whatsoever and considers the issue anew. Aggrieved appellants thus have the opportunity for a second bite at the apple, to correct any legal errors, such as undue deference to the government's assertion, the trial court may have made. Further, in some cases, appellate courts have taken the unusual step of reconsidering factual determinations made at the trial court level[73]; when this occurs, the appellant is essentially afforded a second trial--and a second opportunity to defeat aspects of the factual basis underlying the assertion of the privilege.

As this exercise makes clear, the requirements of the Reynolds framework provide extensive protections against abusive or improper assertions of the state secrets privilege and afford adverse parties significant opportunity to challenge both its invocation and its effect. The evidence supports this conclusion.

One source of evidence is quantitative analysis of cases. It is difficult, of course, to provide an exact count of the number of cases in which the privilege has been at issue, because not all cases result in published opinions. It is possible, however, to catalogue all published opinions adjudicating assertions of the privilege. Robert Chesney did this in a 2007 article, providing an appendix listing all such opinions since Reynolds through 2006.[74] The data collected in that useful article disproves many of the claims made about the state secrets privilege, particularly those concerning its use during the George W. Bush Administration.

As should be expected, given the procedural hurdles and checks, assertion of the state secrets privilege is rare. From 1954 through 2006, the privilege was adjudicated in 89 cases.[75] Most of these cases concern intelligence operations; a few concern each of military technology, military contracts, and diplomatic communications. In most, but not all, the assertion of the privilege was upheld. This demonstrates that the government uses the privilege only sparingly, when necessary, and that courts are willing to push back when they doubt its application.

A few trends in the usage of the privilege are visible, though the paucity of cases prevents confident analysis. Assertion of the privilege was rare until the early 1970s, when cases became more frequent, reaching a peak in 1982. This rise coincided somewhat with popular concern over the government's domestic intelligence activities--the subject of many of these cases--and dissipated as reforms engineered by the executive and Congress branch reined in excesses. There was another surge in the early 1990s, and then one beginning in 2004, about two years into the war on terrorism and around the time that the media began to report on classified programs. The final year of Chesney's study, 2006, witnessed seven assertions of the privilege in published opinions--a seemingly low number but, in fact, a new high.

Despite claims to the contrary, the privilege was not claimed more frequently by the Bush Administration. At least through 2006, Chesney concludes, the data "does not support the conclusion that the Bush administration chooses to resort to the privilege with greater frequency than prior administrations"--that is, the rate of assertion of the privilege relative to the amount of litigation implicating classified national security programsis little changed.[76] This is a more appropriate measure than just counting the number of cases because, unlike with prosecutions, the government does not control the number of civil cases filed that implicate state secrets. Indeed, the more irresponsible and obviously barred suits that are filed, the more the government will be forced to assert the privilege.

I attempted to replicate Chesney's methodology to provide data for the years 2007 and 2008. Federal courts, I found, issued seven reported opinions adjudicating the state secrets doctrine in 2007, and just three in 2008--for ten in total.[77] By comparison, the government contractor defense was adjudicated in more than twice as many published opinions over the same period. One reason for the decrease between 2007 and 2008 may be the aggregation of several lawsuits challenging National Security Agency programs in one court, perhaps resulting in fewer total opinions.

This result and Chesney's data are strong evidence that the privilege is asserted only rarely and that it is rarely, if ever, misused. After all, cases demonstrating misuse or inappropriately harsh results, such as dismissal based on a peripheral connection to national security, are those more likely to be contested and appealed, whether by the government or the party against whom the privilege has been asserted. Such cases, then, are disproportionately likely to result in published opinions. Similarly, activist litigation intended to alter government policies or strike down government programs are of some public interest, receive significant coverage in the media, and are often aggressively litigated. These too are more likely to result in appeals and published opinions.

There is also no evidence that the privilege is being asserted with respect to different kinds of subject matter than it was in the past. Chesney's data show that surveillance programs were the subject of extensive litigation in the 1970s and 1980s, resulting in some assertions of the privilege.[78] The other regular subjects of cases in which the privilege was asserted during that period were employment and contractual disputes within the military and the intelligence agencies and cases risking the disclosure of purely technical information, such as the operation of stealth aircraft technology.[79] The data, concludes Chesney, "does not support the conclusion that the Bush administration [was] breaking new ground with the state secrets privilege" in terms of subject matter.[80]

The 2007 and 2008 data do not alter that conclusion. Five of the opinions were in cases challenging NSA intelligence programs.[81] Two concerned "extraordinary rendition."[82] Two concerned the Valerie Plame affair and her attempt to collect damages from the federal government, as well as other defendants, for the disclosure of her identity.[83] (Neither of the Plame opinions directly adjudicated an assertion of the privilege but both considered it relevance to a Bivens inquiry.) The remaining case concerned allegations that a CIA agent stationed in Rangoon, Burma, had tapped the phone of a Drug Enforcement Administration agent also stationed there.[84]

There is also no evidence that the government has sought harsher remedies, such as dismissal, more often than in the past. Indeed, the government sought, and received, dismissal in the first state secrets case decided after Reynolds in 1954 and has sought dismissal regularly since the early 1970s.[85] Roughly, the Bush Administration sought dismissal or summary judgment in 70 percent of the cases in which it asserted the privilege through 2006. The Clinton Administration sought dismissal or summary judgment in 55 percent. Both were more likely to seek summary disposition in cases relating to intelligence policy and employment disputes involving classified programs. The Bush Administration simply faced a higher proportion of these suits, leading it to seek summary disposition in a slightly higher proportion of cases.

There is also no evidence that the courts have accorded inappropriate deference to executive assertions of the privilege in recent years. In cases with reported opinions, courts granted the government's requested relief 83 percent of the time during the Clinton Administration. Through 2006, courts granted the Bush Administration its requested relief 65 percent of the time; many of the rejections were in cases alleging warrantless domestic surveillance. Including the data from 2007 and 2008 reduces the Bush Administration's "win rate" to just 60 percent. And in its first few months, the Obama Administration has racked up a single loss, in the Ninth Circuit[86], and a looming loss (several procedural issues are disputed) in district court.[87] If anything, the courts have become less deferential to executive assertions of the privilege.

Finally, the privilege has been embraced by the Obama Administration as a necessary tool to protect national security. This should come as no surprise, despite the charged rhetoric of the 2008 presidential campaign; the privilege has been used by every presidential administration since the Johnson Administration asserted it in 1967 to block discovery concerning warrantless surveillance by the FBI. (That assertion was rejected by the court.) The new Administration declined to change course in Mohamed, a suit challenging the CIA's "extraordinary rendition" program; the government lawyer responding to insistent questions from a Ninth Circuit judge stated that the position had been "thoroughly vetted with the appropriate officials within the new administration" and that "these are the authorized positions."[88] In al-Haramain, a suit by an Islamic charity accused of funding terrorism challenging an intelligence program, the Obama Administration stated, in a motion challenging the court's refusal to sustain its assertion of the privilege, that the "disclosure of classified information ... would create intolerable risks to national security."[89] The Administration has stated that, if the court orders it to disclose details about the program to the charity, it will appeal swiftly.[90] It is significant that President Barack Obama, who as a candidate was so critical of the Bush Administration's use of the privilege[91], has come to agree that, in some cases, its use is necessary and legitimate.

To summarize this review of cases applying the privilege, there is no evidence that the state secrets privilege--quite separate from the underlying legal doctrines that implicate the merits of any case--has been abused or misused during the Bush Administration or, more broadly, at all. There is no evidence that the privilege is being used frequently or in cases where it is not needed, no evidence that it is being used to stifle cases on political grounds, and no evidence that judges are unduly deferential to the executive when it is invoked. The fact that some cases concerning government policies have been bounced out of court by the privilege is both unexceptional (assertion of any privilege may, in some cases, defeat a claim) and appropriate--many of these cases, had they been allowed to proceed, would quite obviously have exposed secrets that would put U.S. national security at risk. Finally, the privilege has been employed sparingly by all administrations since Lyndon Johnson was in office, including the current Administration, demonstrating that protecting state secrets from disclosure is not, and should not be, a partisan or ideological issue.

This would seem to defeat any argument in favor of substantively limiting or procedurally hobbling the state secrets privilege, such as the SSPA would do. The Act would radically alter the privilege, placing a much higher--and at times, insurmountable--burden on the government to protect national security information that, in other contexts, is protected by strict laws and regulations carrying heavy criminal penalties for their violation.[92]

In general, the Act would require the government to disclose all evidence it claims is privileged to the court and then prove that public disclosure of each piece of evidence "would be reasonably likely to cause significant harm to the national defense or diplomatic relations of the United States"--a higher standard than that articulated in Reynolds. Gone would be Reynold's sliding scale approach based on necessity, replaced with a mandate that the court personally review every bit of evidence, no matter the obviousness of the consequences of disclosure, the lack of necessity, or the risk of interception during proceedings, as acknowledged by the Court in Reynolds. Counsel for all parties would be presumptively authorized to participate in proceedings concerning the privilege--at least one hearing would be required. The court would accord government officials and experts no deference at all on national security matters, thereby requiring judges to determine weighty matters of national security policy and classification.

Further, in every case, whether or not the privilege is sustained, dismissal or summary judgment would be forbidden until the party against whom the privilege has been upheld "has had a full opportunity" to complete discovery and litigate the issue or claim to which the privileged material is relevant.[93] This would essentially overturn Totten, forcing the government to admit highly classified secrets, such as the identities of spies, in the course of litigation. It would also force the government to submit to "broad-ranging discovery" that itself would be "disruptive of effective government," particularly national defense.[94]

It is difficult to see how these changes could cut down on abuse or misuse of the state secrets privilege, because, as described above, none has been documented. It is clear, however, that the Act would cause the government to lose more often on the privilege issue and to expend greater effort, and disclose more information, even when it is able to prevail. The effect would be particularly harsh in Totten-style cases, in which the government would face the unattractive choice of being uncooperative and losing (by default or on summary judgment) or actually litigating, which would itself confirm the existence of secret relationships and programs. One would expect the government to settle most cases or lose on default judgment, and this strategy would encourage a flood of litigants, some of them with frivolous claims that the government could not challenge lest it disclose state secrets merely by doing do. This would not be, in any way, an improvement over current law--quite the opposite.

The most significant effect, however, may be in activist lawsuits challenging government programs, which would be difficult or impossible to settle without shutting down large portions of our national security infrastructure. The only choice, then, would be to litigate, at considerable expense in terms of dollars and distraction. Because the government would lose more often--as a result of the Act's heightened standard and inordinately complex procedural requirements--the courts would play a major role in making national security policy by disclosing details about, and effectively ending, programs that have been authorized by the President and Congress. This would give activists a "heckler's veto" over many national security programs created by the democratic branches of government, to which such powers are textually committed in the Constitution. This consequence is discussed further below.

Finally, it is likely that the Act's procedures would result in the inadvertent disclosure of closely held national security information. This too is discussed below.

Far from being necessary, the SSPA would endanger national security. It offers no apparent benefits, other than the possibility, attractive to some, that activists with unpopular ideas could use it to achieve an end-run around the democratic process on issues relating to national security.

II. Serious Constitutional Concerns

The SSPA raises serious constitutional concerns by altering a privilege that has a constitutional dimension. Unlike most other privileges, which are supported solely by the common law or statutory law, the state secrets privilege is grounded in the powers committed to the President in Article II of the Constitution. Congress's undisputed power to codify or even abrogate common-law privileges by statute cannot extend to altering to the Constitution's assignments of authority and responsibility. Because it would radically restrict the authority of the President to safeguard military and diplomatic secrets and intelligence, the Act is likely unconstitutional.

As the Supreme Court explained in C & S Airlines v. Waterman S.S. Corp., courts simply lack the constitutional authority and the expertise to make certain types of decisions that are assigned to the executive:

The President, both as Commander-in-Chief and as the Nation's organ for foreign affairs, has available intelligence services whose reports neither are nor ought to be published to the world. It would be intolerable that courts, without the relevant information, should review and perhaps nullify actions of the Executive taken on information properly held secret. Nor can courts sit in camera in order to be taken into executive confidences. But even if courts could require full disclosure, the very nature of executive decisions as to foreign policy is political, not judicial. Such decisions are wholly confided by our Constitution to the political departments of the government, Executive and Legislative. They are delicate, complex, and involve large elements of prophecy. They are and should be undertaken only by those directly responsible to the people whose welfare they advance or imperil. They are decisions of a kind for which the Judiciary has neither aptitude, facilities nor responsibility and have long been held to belong in the domain of political power not subject to judicial intrusion or inquiry.[95]

This is not merely a prudential limitation on judicial power, but a bar to its exercise altogether. The courts both should not and "could not" second guess such decisions.[96]

Justice Potter Stewart provides a compelling explanation for the Constitution's investiture of this narrow but absolute band of power in the executive, and concomitant narrow and absolute bar on judicial discretion:

[I]t is elementary that the successful conduct of international diplomacy and the maintenance of an effective national defense require both confidentiality and secrecy. Other nations can hardly deal with this Nation in an atmosphere of mutual trust unless they can be assured that their confidences will be kept. And within our own executive departments, the development of considered and intelligent international policies would be impossible if those charged with their formulation could not communicate with each other freely, frankly, and in confidence. In the area of basic national defense the frequent need for absolute secrecy is, of course, self-evident.[97]

The Court confirmed and elucidated the C & S Airlines rule in U.S. v. Nixon, in which it rejected the President's claim of executive privilege.[98] Nixon's assertion of the privilege fell short, explained the Court, because "[h]e does not place his claim of privilege on the ground they [the materials sought] are military or diplomatic secrets."[99] "As to those areas of Art. II duties," the opinion continues, the courts have "shown the utmost deference to Presidential responsibilities."[100] Put plainly, "to the extent this interest [a President's interest in confidentiality] relates to the effect discharge of a President's powers, it is constitutionally based."[101] In such cases, said the Court, the rule in Reynolds applies: "the court should not jeopardize the security which the privilege is meant to protect by insisting upon an examination of the evidence, even by the judge alone, in chambers."[102]

The SSPA seemingly ignores this clear jurisprudence and constitutional imperative, running roughshod over the separation of powers. Whereas Reynolds required a court to determine only whether the executive had properly asserted the privilege--i.e., that it had complied with the requisite procedures and that the assertion concerned a matter assigned to the executive, as determined by the deferential "reasonable power" standard--the Act would require courts to "determine whether the privilege claim is valid" by reviewing, for itself, all of the evidence asserted to be privileged and then determining "whether the harm identified by the Government...is reasonably likely to occur should the privilege not be upheld."[103] In this inquiry, the court "shall weigh testimony from Government experts in the same manner as it does, and along with, any other expert testimony," including that from other parties' experts or experts appointed by the court.[104]

In this way, the Act attempts to transfer a power clearly assigned to the executive to the courts. Under the Act, even when a matter falls clearly within the executive's constitutional purview, and clearly outside of the judiciary's, the executive's assertion of the need for confidentiality would be afforded no deference at all, nullifying the executive's power to maintain secrecy in state affairs. This is directly contrary to C & S Airlines, Nixon, and Dept. of the Navy v. Egan, in which the Court explained that the President's "authority to classify and control access to information bearing on national security...flows primarily from this constitutional investment of power [Art. II, § 2] in the President and exists quite apart from any explicit congressional grant."[105] Because the state secrets privilege is raised only by the executive and only (rather axiomatically) in cases that the executive determines threaten to reveal state secrets, a determination which is assigned to the executive, the Act would be "unconstitutional in all of its applications," easily satisfying the most stringent test for facial invalidity.[106]

Further, the SSPA may also impermissibly intrude on the judicial authority conferred in Article III of the Constitution. As the Supreme Court observed in City of Boerne v. Flores, Congress lacks "the power to establish the meaning of constitutional provisions."[107] This limitation incorporates the Supreme Court's constitutional precedents:

When the Court has interpreted the Constitution, it has acted within the province of the Judicial Branch, which embraces the duty to say what the law is. When the political branches of the Government act against the background of a judicial interpretation of the Constitution already issued, it must be understood that in later cases and controversies the Court will treat its precedents with the respect due them under settled principles, including stare decisis, and contrary expectations must be disappointed.[108]

On this basis, the Court struck down the Religious Freedom Restoration Act ("RFRA") as an impermissible intrusion on its power to interpret the Constitution, explaining "it is this Court's precedent, not RFRA, which must control."[109]

The SSPA is similar to RFRA in that it purports to define the Constitution. Specifically, it would impose a rule of decision on the courts requiring them to adopt a narrow construction of presidential power in cases where the state secrets privilege is asserted. But as in Boerne, Congress does not legislate on a blank slate. As discussed above, the Court has held clearly and repeatedly that the state secrets privilege is grounded in Article II of the Constitution. By ignoring this precedent, the SSPA would usurp the power of the judicial branch "to say what the law is."[110]

Finally, several of the procedures specified in the Act also impinge on the executive prerogative described in Egan. First, the court may, at its discretion, order the executive to submit "all of the information that the Government asserts is privileged" for review by the court.[111] Second, the court may order the executive to produce "an adequate substitute [for protected information], such as a redacted version, summary of the information, or stipulation regarding the relevant facts, if the court deems such a substitute feasible."[112] Third, the court may order the executive "to provide a manageable index of the information that the Government asserts is subject to the privilege."[113] Fourth, the court may order the executive to conduct a "prompt" review of any party or counsel to determine whether to provide that individual with a security clearance.[114] Fifth, the court may require that protected information be disclosed to counsel at the hearings required by the Act.[115] Sixth, the court may order the executive to disclose protected information after it determines that the privilege claim is not "valid."[116] Each of these procedures would, in some or all instances, violate the executive's constitutional authority "to classify and control access to information bearing on national security" and to "to determine whether an individual is sufficiently trustworthy to...give that person access to such information."[117]

In short, the no fewer than seven provisions of the SSPA, including its core operative provision, attempt to alter the constitutional separation of powers by reassigning powers from the executive to the judicial branch. In addition, the core provision may also impermissibly intrude the judicial power. Outside of the constitutional amendment process specified in Article V of the Constitution, Congress lacks the power to affect such changes.

III. Weakening Congress and National Security

By altering the structural relationship between the branches, the SSPA would also allow the courts to usurp Congress's power and responsibility--a result that the most cynical Members of Congress may welcome for its political benefits. This, in turn, threatens to undermine the effectiveness of national security policy, putting Americans at risk.

The constitutional separation of powers is no mere legal nicety but an essential bulwark against both tyranny and impotence. The Framers had experience with each of these ills, the former under British rule and the latter as citizens of states weakly bound by the Articles of Confederation. Thus they created an executive energetic in foreign affairs and national security but comparatively weak in domestic policy, recognizing that diplomacy and defense have aspects inimical to drawn-out deliberation, particularly in public. John Jay described one such need in Federalist No. 64:

It seldom happens in the negotiation of treaties, of whatever nature, but that perfect secrecy and immediate despatch are sometimes requisite. These are cases where the most useful intelligence may be obtained, if the persons possessing it can be relieved from apprehensions of discovery. Those apprehensions will operate on those persons whether they are actuated by mercenary or friendly motives; and there doubtless are many of both descriptions, who would rely on the secrecy of the President, but who would not confide in that of the Senate, and still less in that of a large popular Assembly.

Further, as Alexander Hamilton concluded in Federalist No. 74, "Of all the cares or concerns of government, the direction of war most peculiarly demands those qualities which distinguish the exercise of power by a single hand."

Thus, proposals which make exercise of this kind of executive power contingent on the approval of another branch sacrifice these advantages, as well as the accountability of the President for foreign affairs and national security.

Not all matters relating to foreign entanglements and defense are, of course, strictly executive affairs. In many, Congress plays an essential role in legislating programs that draw on its powers for the executive to carry out and in appropriating funds to these programs. And it is Congress, of course, that is responsible for declaring war, defining the laws of war, and raising and supporting armed forces.[118] This raises a question: what is sacrificed when this legislative power and responsibility is transferred to another branch, in this case the judiciary?

Perversely, that would be one of the consequences of the SSPA. This is because the state secrets privilege enforces the separation of powers not just between the executive and the judiciary, but also between Congress and the judiciary. By limiting judicial discretion in certain fields, it protects congressional policymaking in those fields. Broadly speaking, this is but one example of a textual commitment made to the legislative branch, the exercise of which should be and usually is met with the utmost judicial deference. In many instances, these powers are shared, in whole or in part depending on context, with the executive.[119] Thus, the courts should not, and usually will not, adjudicate matters that are "political questions," a subset of the larger class of matters committed to the legislature, the executive, or both of the political branches.

By impinging on the executive's ability to carry out programs that demand stealth and secrecy, the SSPA (if not struck down as unconstitutional) would allow the courts to intrude on matters that would otherwise be outside of the powers committed to the judiciary. Courts would have the power to expose and effectively end or at least hinder all manner of intelligence and national security programs approved by Congress that rely, in any measure, on stealth or secrecy. Even when the court ultimately rules that an assertion of the state secrets privilege was "valid," the damage of exposure will already have been done. In this way, the Act would empower courts, and private parties bringing cases before them, to make policy that had previously been the exclusive domain of the political branches.

This proposition is not far-fetched. As described above, many of the lawsuits in which the government asserts the state secrets privilege concern intelligence and national security programs, some of which have been specifically authorized by Congress and some of which have proceeded under more general legislative authority with Congress's acquiescence.[120] Several of these lawsuits, launched and litigated by activist groups that have failed to convince Congress to adopt their agendas, are intended to end or significantly restrict these programs. The Act would facilitate these efforts, enabling activists to make an end-run around Congress's legislative process.

A Totten-style case presents a simple example. Assume that the most recent intelligence authorization bill, passed by Congress and signed by the President, permits the Central Intelligence Agency to conduct human intelligence activities in foreign states. Assume, as well, that Congress has also funded this program in its most recent appropriations bill. A single overseas agent or informant would be empowered, under the Act, to extract "graymail" from the Agency by threatening to reveal aspects of its human intelligence program through litigation.[121] The Agency would face a choice: it could either pay (through a settlement or default judgment) and thereby make itself a target for identical threats from scores of sources and employees, or it could fight the lawsuit, effectively acknowledging aspects of its program, including the existence of an intelligence relationship. In either case, potential informants are likely to regard the CIA with great wariness, for fear that their identities could be disclosed in litigation or by association with the disclosure of another. The human intelligence program, authorized and funded by Congress and carried out by the executive, would be effectively shut down by judicial interference.

As this simple hypothetical demonstrates, the direct consequences to national security of disclosing state secrets could be immense. But they are also straightforward and so need not be dwelled upon.

Less obvious, but no less pernicious, are the indirect consequences. The benefit of Congress's deliberative process, as concerns any number of intelligence and national security programs, would be undermined, as the courts upset carefully crafted balances hashed out in congressional committee and on the floor before being wrought into law. If such programs are to be ended or scaled back, it should be Congress, which legislates over a far broader canvas than any court hearing a particular case, that should do it, relying on its understanding of the nation's needs and the appropriate means of satisfying them. The courts, considering the law case by case, simply lack the institutional expertise and resources to make policy. The result, in all likelihood, would be worse policy that does not strike the appropriate balance between national security, individual rights, expense, efficacy, and all the other factors that Congress considers in writing legislation.

The other indirect effect, premised on cynicism, could do far more damage to Congress and our representative democracy. In recent years, Members of Congress have been accused for their unwillingness to intervene in controversial actions carried out by the executive branch. Some of these activities, such as certain aspects of foreign intelligence collection, may be beyond Congress's power to affect. Others, however, are not. This latter group includes the Bush and Obama Administration's use of funds earmarked for financial institutions to bail out and then purchase General Motors and Chrysler; the Federal Reserve's bank bailouts; the AIG bailout; the CIA's use of "enhanced interrogation techniques" including waterboarding; surveillance programs that intercept some communications that are arguably domestic in nature; and the use of National Security Letters. In each case, Congress held multiple hearings and many Members of Congress expressed their criticism, often in harsh, accusatory tones. In none, however, did Congress pass legislation significantly curtailing the executive's discretion or rescinding the statutory authority upon which the executive relied.

The SSPA takes this cynicism to a new level. It would allow Congress to duck tough decisions in the national security arena--where bad decisions can have catastrophic consequences--by passing the buck to the courts. These are the same courts that have already come under criticism from the majority party in Congress for upholding state secrets claims and thereby declining to invalidate programs that Congress itself could eliminate with a single bill. The Act would take the pressure off of Congress to check executive overreaching, while giving Members still more targets to criticize in overheated floor statements. This result, pushing contentious matters out of the realm of debate, would be politically safe--which no doubt explains its attraction to some Members--but absolutely poisonous to the American politic.

Rather than attempt to alter the constitutional separation of powers so as to evade responsibility for government actions and omissions, Congress should confront these issues directly and forthrightly.

IV. The Greater Public Good

"Dismissal is a harsh sanction," the Fifth Circuit observed in one state secrets case. "But the results are harsh in either direction and the state secret doctrine finds the greater public good--ultimately the less harsh remedy--to be dismissal."[122] Congress should not sacrifice this greater good to ameliorate the unfortunate plight of the very few who suffer a harsh remedy under the law.

The SSPA would have that effect, putting the nation and its citizens at risk to aid the undemocratic efforts of activists who have been unable to sway Congress to adopt their risky policies. The Act, however, offers an indirect approach--shifting controversial and contentious issues to the courts--thereby promising to shield Congress from deserved opprobrium for allowing our nation's security to be placed at risk.

This legislation is cynical. It is also unconstitutional and completely unnecessary to remedy any genuine ill. Congress should look past the parochial interests of those who would use the courts to make policy, as well as political expediency, and focus on the greater public good.