When The Curtain Must Be Drawn

American Experience With Proceedings Involving Information

That, For Reasons of National Security, Cannot Be Disclosed.

by Peter L. STRAUSS1, Betts Professor of Law, Columbia Law School, USA.

If we are to protect our civil rights and civil liberties against [today’s terrorist] threats, the aggressive use of informants, surveillance, wiretaps, searches, inter-rogations, and even group-based profiling must be measured not only against the liberties these practices constrict, but also with respect to the liberties they may protect. ... It is simply not sufficient to say that augmenting [our usual law enforcement] practices with laws and practices more appropriate to counterterrorism is necessarily inconsistent with protecting our civil rights and civil liberties.2

Speaking about counter-terrorism in France today is a fraught subject, given the recent awful events that so dramatically illustrated the tensions between our precious liberties, and the steps that might be required to protect them. The impact of terrorism on liberty is, however, my subject. The Twenty-first Century’s wars against terror, like the Red Scare and Cold War of the Twentieth, have repeatedly put pressure on procedural values ordinarily observed in American adjudications. In criminal, civil, or administrative proceedings, tribunals may be faced with the government’s wish to rely on evidence that, for reasons of national security, cannot be disclosed to the private parties concerned in the matter. There may even be circumstances in which the government is unwilling to reveal to the tribunal itself information about the source of a fact proposition it wishes to assert – for example, the identity of a secret agent who has provided it or the manner in which an intercepted communication has been obtained. Similarly, the government may wish to resist a private demand for information, that ordinarily it would be obliged to supply, because it is a state secret – and, again, it may be unwilling to disclose the information even to the tribunal alone, as might be thought necessary to permit assessing the accuracy of its claim.

In American law, which emphasizes adversarial processes in adjudication, the importance of confrontation (the delivery of testimony in the presence of the party against whose interest it is presented) and cross-examination (rigorous questioning by a party’s attorney of witnesses for the opposing side, once they have presented their testimony) arms resistance to the use of secret evidence and supports claims for its disclosure. In criminal prosecutions, the rights “to be informed of the nature and cause of the accusation; to be confronted with the witnesses against him; to have compulsory process for obtaining witnesses in his favor, and to have the assistance of counsel for his defense” are secured by the Sixth Amendment to the federal Constitution, and place the government’s wishes in shadow. In civil and administrative cases, similar but less explicit counter-pressure arises from one’s Fifth Amendment right not to be deprived of life, liberty or property without “Due Process of law,” a phrase whose meaning derives much from the American commitment to adversarial adjudication.3 When a person’s life, liberty or property has been or might be negatively affected by government action, the process that is due is to be determined by a difficult comparison among the private interest affected, government interests (in efficiency, expedition, national security, etc.), the risk of error created by the procedures actually afforded, and the prospect of reducing that risk by supplying additional procedures.4

The claims to confrontation and cross-examination are overcome in a variety of situations in which statements made or records kept out of court are thought to be reliable – as, for example, records kept in the ordinary course of business rather than created for this particular dispute; or a witness made unavailable by death or departure who has made statements in circumstances suggesting their credibility. Direct confrontation may be thought unacceptably harmful, psychologically, to a minor child or an abused spouse. In these circumstances, however, the person whose interests are at risk will have viewed the records and been able to question the person presenting them; she will know the identity of the witness and the statements she is reported to have given. But in the context of national security, circumstances may suggest that disclosing either or both to the person at risk – even disclosing the fact of or reasons for action – would in itself produce an unacceptable chance of harm.

These issues achieved prominence in American jurisprudence in the wake of World War II, when the development of nuclear weapons, our Cold War with the Soviet Union, and the demagoguery of Joseph McCarthy combined to create high stakes in the possession of military information and an atmosphere of suspicion and fear. Writing this paper brought home that it was in this crucible that my own engagement with administrative law began – in college, through independent study of the Atomic Energy Commission’s security clearance hearings for J Robert Oppenheimer; and, in law school, by an extensive seminar paper on the Supreme Court’s decision in Greene v. MacElroy.5 In both these settings, the latter especially, matters known only to the tribunal, involving reliance on damaging statements by persons neither the tribunal nor the subject of their proceedings ever met, were the driving force producing an adverse decision.

In Greene, an “indispensable” aeronautical engineer and important executive officer for a naval contractor had been stripped of his security clearance, and had for that reason been essentially disqualified from pursuing his profession. This action had been sustained after repeated hearings, reversals, and high-level reconsiderations within the security clearance bureaucracy; what was in the public record revealed it to be, evidently, a case on the margin; and it occurred at the height of the McCarthy era. Greene, his attorney, and supporting witnesses attended those hearings, but the government did not present witnesses. The tribunals, which minutely questioned Greene and his witnesses, evidently relied for its adverse findings on written reports Greene and his attorney were not permitted to examine.6 The tribunals’ questioning did appear in the record available to reviewing courts, and the case eventually reached the United States Supreme Court. The extensive footnotes in Chief Justice Warren’s opinion for that Court make clear how much of the information adverse to Greene came from the written reports of unidentified investigators about “information … from an informant characterized to be of known reliability,”7 reports screened and evaluated for the Board by unidentified “security advisers,” “well-trained people who know how to evaluate reports and evaluate information.”8 The actual holding of the case was that authorization for a program run in this manner must be express, and that express authorization for it was missing; but that holding was driven by constitutional doubts about the permissibility of such departures from American commitments to confrontation and cross-examination in proceedings gravely affecting an individual, and the language expressing those doubts has often been quoted and relied upon in subsequent opinions:

Certain principles have remained relatively immutable in our jurisprudence. One of these is that where governmental action seriously injures an individual, and the reasonableness of the action depends on fact findings, the evidence used to prove the Government’s case must be disclosed to the individual so that he has an opportunity to show that it is untrue. While this is important in the case of documentary evidence, it is even more important where the evidence consists of the testimony of individuals whose memory might be faulty or who, in fact, might be perjurers or persons motivated by malice, vindictiveness, intolerance, prejudice, or jealousy.9

Similar issues recently confronted the United States Court of Appeals for the District of Columbia Circuit, the intermediate federal court having the greatest influence over the development of American administrative law.10 The Ralls Corporation, an American corporation owned by two Chinese nationals associated with a Chinese maker of wind turbines (Sany), purchased four companies with established legal rights to construct wind farms at sites in and around an area of militarily restricted air space; under state law this established Ralls’ right to build the wind farms, which it characterized as a commercial opportunity to demonstrate the quality of its turbines. Other foreign-owned firms with foreign-made wind turbines already operate wind farms in the same area. Ralls negotiated with the Navy a relocation of its one site that actually lay within the restricted space, to reduce airspace conflicts. As required by the Defense Production Act of 1950,11 it filed a 25-page notice with the Committee on Foreign Investment in the United States (CFIUS) detailing its plans and its reasons for believing they posed no security threat to the United States. CFIUS is a high-level executive agency chaired by the Secretary of the Treasury, responsible to review foreign investments that “result in foreign control of any person engaged in interstate commerce”12 for possible effects on national security; if unable to negotiate mitigation of any such effects as it finds, it submits a report to the President detailing its findings and their basis, and he must act on the report within fifteen days, suspending or prohibiting a transaction he finds to present such a threat.13 His actions and findings “shall not be subject to judicial review.”14 During its review, CFIUS asked several questions of Ralls, and entertained a presentation by it, but CFIUS provided Ralls and its attorneys no access to such information as it possessed from other sources. CFIUS then submitted an (undisclosed) recommendation to the President, who promptly found that the Ralls windfarm project created a security threat. He therefore issued orders effectively barring both Ralls and any items made by Sany from the four sites. “[N]either CFIUS or the President gave Ralls notice of the evidence on which they respectively relied nor an opportunity to rebut that evidence.”15

The case presented a number of knotty questions that might have, but did not, prevent the court from addressing Ralls’ constitutional claims to know and have the opportunity to respond to any information adverse to its interests. Just as constitutional doubts about the security clearance procedure’s reliance on undisclosed, unconfronted evidence had led the Greene Court to demand the explicit authorization it found to be lacking for that procedure, due process concerns led the Ralls court to demand more than a “broadly worded statutory bar” to foreclose judicial review of constitutional claims.16 Although the court agreed with the government that review of national security determinations on their merits would be inappropriate for courts,17 it found that resolving Ralls’ claim to certain procedural rights would not necessitate judicial review of those substantive conclusions. The balancing inquiry to determine the requirements of due process, in circumstances in which process is constitutionally “due,” has been stable for almost 40 years.18 The court easily found that the President’s order had deprived Ralls of relationships recognized as “property” under state law, the necessary precondition of Ralls’ constitutional claim.19 What, then, were the minimal procedures having to be employed? Ralls had received neither notice of any of the information on the basis of which the CFIUS and the President acted (and thus no opportunity to challenge or respond to it) nor any explanation of its reasoning in finding a security threat. And the government conceded that the record included unclassified information as well as information the government was obliged to protect for reasons of national security. Invoking both Greene and recent opinions dealing with the designation of organizations as Foreign Terrorist Organizations (FTOs),20 the court concluded that Ralls was entitled to notice of the unclassified information being considered by the CFIUS, and an opportunity to rebut it.

In Greene, the Court had not had to decide whether, in security clearance hearings, security-sensitive material must be revealed to a person whose life could be strongly affected by reliance on it. It was enough to require reversal that reliance on secret information had not explicitly been authorized by President or Congress. But the FTO cases had reached that issue, as the Ralls court remarked:

we made clear – and we iterate today – that due process does not require disclosure of classified information supporting official action. See NCRI (classified information “is within the privilege and prerogative of the executive, and we do not intend to compel a breach in the security which that branch is charged to protect”). We have consistently followed NCRI in subsequent FTO cases...21

Thus, whatever classified information the government might have had and relied on in stripping Ralls of its $6 million investment, Ralls was not entitled to the amenity engineer Greene had in fact enjoyed in Greene v. McElroy, to know what its allegations were. The court did not require that Ralls be informed even of the general gist of the secret information being relied upon, the opportunity Greene had enjoyed, with an opportunity to respond to that description.

Ralls in this way reflects an increased judicial willingness to privilege government claims of secrecy, even if not quite so complete a willingness as the government seeks. It is only one such reflection. Designation as an Foreign Terrorist Organization is another context in which these issues have arisen. Such designations are the current equivalent of the Cold War/McCarthy era “Communist Front organization.” Identification as a CFO was in fact less portentious; it brought community disrepute to the organization and obloquy to its members, but had few formal effects beyond providing evidence useful in security clearance determinations. For FTOs, by contrast, the Anti-Terrorism and Effective Death Penalty Act of 1996 [AEDPA],22 designation has major legal consequences; it results in a criminal-law-enforced blocking of any funds the organization may have on deposit with any financial institution in the United States, and it establishes as a major felony “knowingly providing material support or resources” to the organization.23 The Supreme Court has found that even humanitarian (other than medical or religious) support would permissibly violate this statute.24

CFOs had much more open a chance to challenge their designation then than do FTOs today. During the Cold War, the protests of the Joint Anti-Fascist Refugee Committee against its designation as a Communist Front organization engendered in the Supreme Court an opinion by Justice Frankfurter that later was credibly characterized as “the finest exposition of the need for a hearing”:25

[Designation as] “communist” by the Attorney General of the United States … imposes no legal sanction on these organizations other than that it serves as evidence in ridding the Government of persons reasonably suspected of disloyalty. It would be blindness, however, not to recognize that in the conditions of our time such designation drastically restricts the organizations, if it does not proscribe them. Potential members, contributors or beneficiaries of listed organizations may well be influenced by use of the designation, for instance, as ground for rejection of applications for commissions in the armed forces or for permits for meetings in the auditoriums of public housing projects. Yet, designation has been made without notice, without disclosure of any reasons justifying it, without opportunity to meet the undisclosed evidence or suspicion on which designation may have been based, and without opportunity to establish affirmatively that the aims and acts of the organization are innocent…

… From a great mass of cases, running the full gamut of control over property and liberty, there emerges the principle that statutes should be interpreted, if explicit language does not preclude, so as to observe due process in its basic meaning… The heart of the matter is that democracy implies respect for the elementary rights of men, however suspect or unworthy; a democratic government must therefore practice fairness; and fairness can rarely be obtained by secret, one-sided determination of facts decisive of rights…

… [This principle] should be particularly heeded at times of agitation and anxiety, when fear and suspicion impregnate the air we breathe. “The plea that evidence of guilt must be secret is abhorrent to free men, because it provides a cloak for the malevolent, the misinformed, the meddlesome, and the corrupt to play the role of informer undetected and uncorrected.” United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 551 (dissenting) …

Man being what he is cannot safely be trusted with complete immunity from outward responsibility in depriving others of their rights. At least such is the conviction underlying our Bill of Rights… No better instrument has been devised for arriving at truth than to give a person in jeopardy of serious loss notice of the case against him and opportunity to meet it. Nor has a better way been found for generating the feeling, so important to a popular government, that justice has been done.26

As later in Greene, however, the decisive issue in that case was legislative authorization.

In the AEDPA, Congress created a much more limited procedure for designation of FTO status. Three months before the September 11 attack, which understandably reinforced awareness of terrorist threats and the need to find them in advance of action, the D.C. Circuit described it this way:

The unique feature of this statutory procedure is the dearth of procedural participation and protection afforded the designated entity. At no point in the proceedings establishing the administrative record is the alleged terrorist organization afforded notice of the materials used against it, or a right to comment on such materials or the developing administrative record. Nothing in the statute forbids the use of “third hand accounts, press stories, material on the Internet or other hearsay regarding the organization’s activities. The Secretary may base the findings on classified material, to which the organization has no access at any point during or after the proceeding to designate it as terrorist.

… [Within] thirty days after the publication of the designation in the Federal Register, [a designated organization may obtain judicial review] … based solely upon the administrative record … [U]nder the AEDPA the aggrieved party has had no opportunity to either add to or comment on the contents of that administrative record; and the record can, and in our experience generally does, encompass “classified information used in making the designation,” as to which the alleged terrorist organization never has any access, and which the statute expressly provides the government may submit to the court ex parte and in camera.27

The court indicated that it had indeed reviewed the classified information in the record supplied to it. As would the Ralls court subsequently, it found a due process violation in the government’s failures to provide notice of the pending possible designation as an FTO and the unclassified information it had, with an opportunity to appear and present its side of the story.28 Nothing in the court’s opinion, however, suggested the slightest obligation to provide the kind of information revealed in the Greene record and demanded by Justice Frankfurter, apprising the organization of the gist of the classified, secret information on which the government might rely.

The constitutional right to confrontation and national security secrecy collide most dramatically in criminal prosecutions. Here again developments have been characterized by strong beginnings, congressional action, and judicial retreat. Cold War spy prosecutions brought Ivanov v. United States to the Court in 1967, embedded in what appears to have been an organized crime prosecution, Alderman v. United States.29 At issue was the possibility that wiretap evidence not supported by a search warrant, and thus illegal, had led the government to evidence used in prosecuting Ivanov for espionage. Governing law required a court to exclude from evidence not only material unlawfully seized (which unlawfully overheard conversations would be30) but also any “fruit of the poisonous tree.” The government strenuously argued that this question should be explored in camera and ex parte by the trial judge, to avoid disclosing sensitive information unrelated to the prosecution. The brilliant oral advocacy of Edward Bennett Williams, a preeminent litigator of the time, persuaded Justice White, for the majority, otherwise. Mirroring Williams’ argument, he wrote:

Admittedly, there may be much learned from an electronic surveillance which ultimately contributes nothing to probative evidence. But winnowing this material from those items which might have made a substantial contribution to the case against a petitioner is a task which should not be entrusted wholly to the court in the first instance. It might be otherwise if the trial judge had only to place the transcript or other record of the surveillance alongside the record evidence and compare the two for textual or substantive similarities. Even that assignment would be difficult enough for the trial judge to perform unaided. But a good deal more is involved. An apparently innocent phrase, a chance remark, a reference to what appears to be a neutral person or event, the identity of a caller or the individual on the other end of a telephone, or even the manner of speaking or using words may have special significance to one who knows the more intimate facts of an accused’s life. And yet that information may be wholly colorless and devoid of meaning to one less well acquainted with all relevant circumstances. Unavoidably, this is a matter of judgment, but in our view the task is too complex, and the margin for error too great, to rely wholly on the in camera judgment of the trial court to identify those records which might have contributed to the Government’s case.

… It may be that the prospect of disclosure will compel the Government to dismiss some prosecutions in deference to national security or third-party interests. But this is a choice the Government concededly faces with respect to material which it has obtained illegally and which it admits, or which a judge would find, is arguably relevant to the evidence offered against the defendant.31

And the same choice, even more emphatically, would face the government if it wished to introduce testimony by a secret informant, or from an otherwise admissible document that it was unwilling for reasons of national security to share with the defendant, or if the defendant wished to call as a witness a person who would properly be asked to testify on national security matters. Steps short of giving up the prosecution might be possible – for example, having the witness testify in such a way that her identity would be known only to the defendant, counsel, and court,32 or entering a protective order. Accommodations such as this – even ex parte revelations to a court in the course of determining the need for them – could not be totally secure, with the result that, even so, the government might conclude that it should choose the alternative of foregoing prosecution. For knowledgeable defense attorneys in cases likely involving such information, there grew up a practice of “greymail,” calling witnesses or seeking documents – sometimes late in the course of proceedings – that would create this dilemma for the prosecution.

Congress, driven by its increasing fear of terrorism, subsequently created procedures intended to protect the secrecy of government surveillance and to counter “greymail.” The Foreign Intelligence Surveillance Act [FISA],33 first enacted in 1978, was comprehensively amended in 2008. In its initial form, it created a special court, the Foreign Intelligence Surveillance Court, whose basic operations were secret. Sitting federal judges named to it by the Chief Justice of the United States were authorized to issue surveillance warrants that would be secret, and obtained on the basis of state secret information, but that would in many respects resemble search warrants ordinarily obtained: a specific foreign target must be named, with probable cause shown to believe it a foreign power or its agent, and the government was required to identify the nature and location of each facility or place to be bugged. In addition the government was required in each instance to describe procedures that would be used to minimize the intrusion on the privacy of Americans – those protected by the Fourth Amendment’s warrant requirement. Some combination of the September 11 attacks and its sequellae, and the uncovering of massive warrantless National Security Agency surveillance activities being conducted outside this framework on the claimed inherent constitutional authority of the President, produced the 2008 amendments. While still oriented to surveillance of the communications of non-US persons located abroad, the prior elements of specificity have been withdrawn; submissions to a FISC judge no longer must describe specific targets or facilities; targets need not be a foreign power or its agent; there need only be shown that “a significant purpose of the acquisition is to obtain foreign intelligence information”; and minimization procedures need only be addressed generally, not case-by-case. This legislative loosening of constraint answers the kinds of questions courts have previously asked, and the same fears/reactions to national threat as prompted its enactment may underlie a recent Supreme Court decision making it appear that its constitutionality is unlikely soon to be adjudged by the courts.34

The “greymail” issue was addressed in another act, the Classified Information Procedures Act of 1980.35

CIPA’s purpose is straightforward. The statute establishes procedures meant to protect the government from graymail by balancing the executive branch’s need to protect classified information against a criminal defendant’s rights. It does so in two ways. First, defendants must declare early in the proceedings the “price” [he or she] asserts the government will have to pay if the prosecution continues. CIPA operationalizes this function by imposing notice requirements on defendants intending to submit classified information as evidence36 and requiring pre-trial hearings to determine relevance and admissibility.37 That lets prosecutors calculate ex ante whether the possible national security costs of prosecution are worth its benefits. Second, it authorizes judges to modify discoverable material38 and admissible evidence39 in order to minimize the disclosure of classified information--through the use of unclassified summaries and protective orders, for example--while at the same time vindicating the constitutional and statutory rights of criminal defendants.40

The 1980 Congress, that is, accepted the likely consequence that the government could be put to its choice, and did not attempt to privilege security information over confrontation rights; but it worked to curb the use of this possibility as a litigating weapon. It was a compromise measure, enacted after considerable legislative struggle and grudgingly accepted by the civil rights and defense community; and its implementation by courts has tended to respect the balance struck. “As its opponents recognized, CIPA is, at its core, a pro-government statute. By diminishing defendants’ ability to graymail, it removes an arrow from the defense team’s quiver. A prosecution is a zero-sum game, and one party’s loss is the other’s gain. However, in removing an arrow, CIPA’s procedural framework grants the defendant a partial, patchwork shield against arbitrary or unfounded prosecutorial secrecy claims. Where it favors the government’s interest in secrecy, it provides judges a number of ways to vindicate defendants’ equally important trial rights.”41

Those ways, it might be remarked, include a not unfamiliar model – albeit the one Edward Bennett Williams persuaded the Court to reject in Alderman – in which the tribunal reviews sensitive material ex parte, with government assistance, to determine how much if any of it will be important to the defendant’s case, and whether the defendant’s needs can be satisfied by government concession of matters that might be at issue, or by the provision of an unclassified summary of the gist of the classified material. In Alderman, recall, the Court had reasoned

An apparently innocent phrase, a chance remark, a reference to what appears to be a neutral person or event, the identity of a caller or the individual on the other end of a telephone, or even the manner of speaking or using words may have special significance to one who knows the more intimate facts of an accused’s life.42

But the corresponding argument can be made, and has been made, in the opposite direction – in support of the CIPA procedures and the government interests they protect.

[M] uch of the government’s security interest in the conversation lies not so much in the contents of the conversations, as in the time, place, and nature of the government’s ability to intercept the conversations at all. Things that did not make sense to the District Judge would make all too much sense to a foreign counter-intelligence specialist who could learn much about this nation’s intelligence-gathering capabilities from what these documents revealed about sources and methods. Implicit in the whole concept of an informant-type privilege is the necessity that information-gathering agencies protect from compromise “intelligence sources and methods.” … [These] concerns inform our construction of CIPA and the classified information privilege, and the same concerns must inform analyses by district courts in passing on the discoverability of classified information.43

Defendants’ access, that is, may depend on their capacity to establish the materiality of the information they seek to their defense, without knowing precisely what it is.

Yet in the context of the criminal procedures to which it directly applies, it seems possible that interpretation of CIPA’s procedures will be influenced in defendants’ favor by the requirements of the Sixth Amendment’s confrontation clause. A 2004 opinion of the Supreme Court, Crawford v. Washington,44 emphasized the necessary connection between the right to confrontation and the opportunity for cross-examination in relation to out-of-court statements obtained to prove a criminal case. Historically, “the principal evil at which the Confrontation Clause was directed was the use of ex parte examinations as evidence against the accused … [T]he Framers would not have allowed admission of testimonial statements unless [the declarant] was unavailable to testify, and the defendant had had a prior opportunity for cross-examination.”45 Subsequent decisions have focused on the nature of a “testimonial statement.”46 Thus, in consolidated cases, the transcript of a telephone call made in the midst of an emergency was admissible despite an opportunity to confront and cross-examine the person who had made it,, because the call was made to secure a response to the emergency; but the results of police questioning after responding to a different emergency were not, because made to obtain proof of the prior event.47 Most recently, in Williams v. Illinois,48 the Court failed to agree on a majority opinion in a case challenging one expert’s testimony that reasoned from another expert’s report of test results. For a plurality, it was enough to permit the testimony that the other expert’s report was formally offered to prove only the fact, not the veracity, of it; four Justices were prepared to narrow to “testimonial” only statements “having the primary purpose of accusing a targeted individual of engaging in criminal conduct,”49 and the absent expert’s report did not. Justice Thomas, whose vote was necessary for the result upholding the conviction being challenged, thought this primary purpose test lacked “any grounding in constitutional text, in history, or in logic,”50 and found the use of the report proper because it was not a “solemn declaration or affirmation made for the purpose of establishing or proving some fact.”51 Four dissenters thought use of the absent expert’s report would inevitably be taken to support the truth of the testifying expert’s testimony; “the fact finder can do nothing with it except assess its truth,” and it must thus be recognized as testimonial, requiring confrontation.52 Government offers of secret evidence, even if they do not satisfy the plurality’s “primary purpose” test, often will likely satisfy Justice Thomas and the Williams dissenters as “testimonial” – undercutting the possibility that the CIPA procedures could be interpreted to permit its introduction to establish guilt.53

Outside the context of criminal procedure, any claims to confront witnesses and to know the evidence that may be used to a person’s disadvantage rest exclusively in the Due Process clauses of the Fifth and Fourteenth Amendments, and thus on language whose imprecision and governing interpretation54 belies an irreducible right. As against Congress’ plenary right to exclude non-citizens at the nation’s borders and provide for the expulsion of others unlawfully present, “due process” has offered little security against the government’s use of information known only to it in proceedings leading to exclusion or deportation. Despite citizens’ right to travel, the issuance and validity of passports was for decades controlled by the State Department on the basis of an unexplained conclusion that international travel “would be contrary to the best interests of the United States,” and without recourse;55 only in 1958, in Kent v. Dulles,56 did the Supreme Court find the withholding of a passport unlawful; and here, as in Greene and Joint Anti-Fascists Refugee Committee, its precise ground of decision was want of explicit legislative authorization for constitutionally questionable actions – not ratification of the procedural rights claimed.

In the civil action context, too, the government is able to claim a “state secret” privilege to resist disclosure of information that might be valuable to a plaintiff in litigation pursuing an affirmative claim. Here the foundation was laid by United States v. Reynolds,57 another case with Cold War overtones, in which the widows of civilian observers killed when an Air Force plane caught fire and crashed were denied discovery of the Air Force’s official accident report in connection with their suit for damages under the Federal Tort Claims Act.. The issue, however, was not resolved simply by the executive’s claim of privilege. Plaintiffs’ need for the document had also to be considered.

Regardless of how it is articulated, some … formula of compromise must be applied here. Judicial control over the evidence in a case cannot be abdicated to the caprice of executive officers. Yet we will not go so far as to say that the court may automatically require a complete disclosure to the judge before the claim of privilege will be accepted in any case. It may be possible to satisfy the court, from all the circumstances of the case, that there is a reasonable danger that compulsion of the evidence will expose military matters which, in the interest of national security, should not be divulged. When this is the case, 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.58

In Reynolds, the government had offered to permit surviving crew members to testify about the cause of the crash, which appeared unlikely to involve the experimental electronics present on the plane. This offered compromise weakened any claim that having the report was necessary for the widows’ case, and the action was not dismissed but remanded so that it might go forward on that basis. Seemingly important, too, was that the case arose under a statute that had, to some degree, waived sovereign immunity; that such a waiver of absolute government freedom from financial liability extended to revealing military secrets could be doubted.

Respondents have cited us to those cases in the criminal field, where it has been held that the Government can invoke its evidentiary privileges only at the price of letting the defendant go free. The rationale of the criminal cases is that, since the Government which prosecutes an accused also has the duty to see that justice is done, it is unconscionable to allow it to undertake prosecution and then invoke its governmental privileges to deprive the accused of anything which might be material to his defense. Such rationale has no application in a civil forum where the Government is not the moving party, but is a defendant only on terms to which it has consented.59

Where the government relies on state secrets to deprive an individual of liberty or property, the action that gives right to a claim to due process in doing so, access to that information is not simply an aid to securing affirmative relief. Thus, it would seem the same contrast with the Reynolds circumstances could be drawn. And the case for making that distinction is not weakened by the realization that when, finally, the Air Force report at issue in Reynolds was declassified, the claim to “state secret” privilege appeared to have been hollow, suggesting a judgment influenced by litigation advantage rather than genuine national security need.60

In recent years, however, the “state secret” privilege has been used with some frequency wholly to deny the possibility of suit, and in cases for which government consent is not a requisite. Consider, for example, El-Masri v. United States,61 an action against government officials in their individual capacity seeking damages pursuant to Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics62 for violations of El-Masri’s constitutional rights. These violations were allegedly effected by and under the direction of CIA agents, during brutal interrogations after his “extraordinary rendition.” The government intervened in this otherwise private action, asserting in sealed documents El-Masri and his counsel were not permitted to see, that the law suit could not go forward without exposing state secrets. Although the facts of extraordinary renditions and the brutal interrogation tactics they often entailed were, in general, well known, the government persuaded the Fourth Circuit that the particulars of El-Masri’s case could not be explored at trial without revealing matters entitled to the state secret privilege.

[P] roof of the involvement -- or lack thereof -- of particular CIA officers in a given operation would provide significant information on how the CIA makes its personnel assignments. Similar concerns would attach to evidence produced in defense of the corporate defendants and their unnamed employees … [V]irtually any conceivable response to El-Masri’s allegations would disclose privileged information…

In this matter, the reasons for the United States’ claim of the state secrets privilege and its motion to dismiss were explained largely in the Classified Declaration, which sets forth in detail the nature of the information that the Executive seeks to protect and explains why its disclosure would be detrimental to national security. We have reviewed the Classified Declaration, as did the district court, and the extensive information it contains is crucial to our decision in this matter… It is no doubt frustrating to El-Masri that many of the specific reasons for the dismissal of his Complaint are classified. An inherent feature of the state secrets privilege, however, is that the party against whom it is asserted will often not be privy to the information that the Executive seeks to protect…

As we have observed in the past, the successful interposition of the state secrets privilege imposes a heavy burden on the party against whom the privilege is asserted… That party loses access to evidence that he needs to prosecute his action and, if privileged state secrets are sufficiently central to the matter, may lose his cause of action altogether. Moreover, a plaintiff suffers this reversal not through any fault of his own, but because his personal interest in pursuing his civil claim is subordinated to the collective interest in national security…

… [W]e recognize the gravity of our conclusion that El-Masri must be denied a judicial forum for his Complaint, and reiterate our past observations that dismissal on state secrets grounds is appropriate only in a narrow category of disputes… Nonetheless, we think it plain that the matter before us falls squarely within that narrow class, and we are unable to find merit in El-Masri’s assertion to the contrary.63

The court considered and rejected the possibility that its ruling effectively put in the executive’s hands a tool for avoiding embarrassment and liability – as appeared may have been the case in Reynolds, when the document whose secrecy was so important to protect was finally declassified.64 Courts are responsible to consider carefully the government’s case for invoking the privilege, the court remarked, and any possibilities of partial relief from its effects. Of course they do so in camera and ex parte, and thus without the participation of the individual whose interests may be defeated by the claim.65 And, as has been remarked, they may not do so only in reliance on objective factors – that the executive has a greater capacity to understand the stakes, and that the government interest in national security is a paramount claim. They are human, and may be “afraid. They are afraid of terrorism. They are afraid of what could happen to our security if they rein in government.”66

Scholars have found that courts have required in camera inspection of allegedly privileged documents in fewer than one-third of the reported cases in which the privilege has been invoked, and that this proportion is declining.67 As a result, “even though the Reynolds case held that ‘judicial control over the evidence in a case cannot be abdicated to the caprice of executive officers,’ the practical effect of the decision [has been] to cause precisely that result.”68 When courts fail to scrutinize assertions of the privilege, they leave open the possibility that the privilege will be used to cover up Government wrongdoing, thereby denying justice to litigants and giving the executive branch the ability to violate statutes and constitutional rights with impunity.69

One important difference between the current day and Cold War times has been the development of the Information Age, with its extraordinary capacities for surveillance, record-keeping and data-mining. Since the 9/11 attacks, these capacities have been drawn upon extensively. There now exists a Terrorist Screening Center [TSC] to make coordinated judgments whether a “reasonable suspicion” exists that an “identifiable individual”70 is associated in some way with terrorist activities, an enormous Terrorist Screening Data Base [TSDB]71 it administers as a coordinated data base for holding this information, and a considerable variety of government lists into which that data can then be fed as seems appropriate. The No-Fly List, administered by the Transportation Security Administration [TSA] to prevent identified persons from flying in American air space, is only one of these lists. A person may first become aware of her possible presence on the No-Fly List when she attempts to board a flight, domestic or international, for which she had successfully purchased a ticket and checked in; and that discovery may be attended with interrogations and other police actions. Subsequently, she can submit information to the Department of Homeland Security through its web site, in an effort to persuade it a mistake has been made. However, these procedures will not directly tell her whether she is on the list, provide any information about or characterization of the data that has produced any such designation, or permit her any human contact or hearing, however informal. Her submissions will be processed by the TSA and TSC behind closed doors. Eventually she may receive a letter that, without admitting she was ever on the list, will state formulaically that “after review of any applicable records … [if] it was determined that a correction to records was warranted, these records were modified… This determination constitutes our final agency decision, which is reviewable by the United States Court of Appeals.” Just what the determination was, or on what it was based, is not stated; the administrative record is provided to the reviewing court under seal for its review, ex parte and in camera.72

As with FTO determinations and as in Ralls, litigation has begun somewhat to domesticate these procedures, albeit cautiously, and against concerted government resistance. Three district court actions, each of which defeated strenuous government efforts to preclude judicial review altogether, advance similar analyses of the due process implications of the no-fly list procedure.73 The government need not notify individuals in advance that they have been placed on the TSDB or, more specifically, the no-fly list – a form of notice that could readily compromise anti-terrorism efforts. But once that has become apparent (as by denial of boarding, unusual interrogations, etc.) and the individual has sought relief, it must provide a statement of reasons in as much detail as considerations of national security will permit, access to any unclassified information used in making the determination, an opportunity to submit evidence, and an unclassified explanation, to the extent possible, of any resulting decision with negative consequences, to facilitate participation in judicial review. On judicial review, the court considers the whole administrative record, ex parte and in camera to the extent security considerations so require.

It is striking to notice how significantly these emerging procedures resemble those Congress has created for settings in which the personal claim for access to government information is much weaker. Under the Freedom of Information Act, persons denied information on the basis of one of its nine exemptions may challenge that denial in court, but they and their counsel (if they have one) will have no access to it. Instead they may receive what is known as a Vaughn Index, an attested document drawn up by the government, whose purpose is to describe the basis for the claimed exemption in sufficient detail to permit the reviewing court to decide whether the government has sustained its burden to prove the validity of the exemption claim, but not in so much detail as to reveal the information for which the exemption is claimed. Alternatively, the court or federal magistrate considering the review will examine the document in camera, ex parte, to determine the validity of the claim. In cases in which the Act’s first exemption, for state secrets, is invoked, the reviewing court is to decide whether documents are

(1) (A) specifically authorized under criteria established by an Executive order to be kept secret in the interest of national defense or foreign policy and

(B) are in fact properly classified pursuant to such Executive order

Almost invariably, these determinations are made in camera; and while the second criterion invites independent judicial judgment, they are made with a high degree of deference to classifiers’ determinations. As the Ralls court remarked, in a context in which outsiders had a greater claim of access,74 “[C]lassified information ‘is within the privilege and prerogative of the executive, and we do not intend to compel a breach in the security which that branch is charged to protect.’”75

In one administrative setting where classified information might often have to be used in relation to desired government licenses, a similar approach is also taken, but with greater ostensible effort to minimize the use of that information and to maximize party access to or understanding of it. The procedures of the United States Nuclear Regulatory Commission have long contained a special part to govern the use of classified information in its public procedures.76 Like CIPA, they lean against unnecessary use of classified material in hearings. Parties are to avoid its use unless necessary,77 on advance notice78 which includes “an unclassified statement setting forth the information in the classified matter as accurately and completely as possible,” which the presiding officer is instructed to prefer as the evidence of record to the extent possible;79 classified information is to be admitted only if its “relevance and materiality … are clearly established; and [its] exclusion … would prejudice the interests of a party or the public interest.”80 And any such evidence that may have been admitted is to be expunged from the record unless doing so would “prejudice the interests of a party or the public interest.”81 Hearings are ordinarily to be scheduled to permit the securing of necessary security clearances,82 and where they cannot be obtained, hearing officers are instructed to “tak[e] into consideration any lack of opportunity to rebut or impeach the evidence.”83

When claims to “due process” have been validated in a civilian terrorism context, use of the Mathews v. Eldridge balancing test,84 as in Ralls, has regularly recognized the strength of the private claim. The risk of error, and potential of additional procedures to reduce it, are also plaintiff-favoring, given the blank wall and simple “trust us” claims plaintiffs otherwise face. The government side of the balance, however, framed as its interest in national survival, has constrained the courts that are willing to find process due, to require less than Cold War plaintiffs learned, and either CIPA or the established administrative procedures provide. The terrorist threat, on analogy to ongoing criminal investigations, the seizure of dangerous foodstuffs, or other emergencies, requires no procedure whatever before the deprivation (refusal of boarding; immediately effective notice of an obligation to undo a transaction and remove property associated with it; etc.) has been effected. Post-deprivation, plaintiffs must receive some explanation of the grounds for the deprivation; access to any unclassified information in the record; and some opportunity to provide contrary/rebuttal information and argument. However, no court has yet required an unclassified account of state secret material in the administrative record (such as Greene effectively received through the questioning of him) or the kinds of advance procedures characteristic of CIPA and the established administrative process of the NRC. Strikingly, although oral address to the decisionmaker appears to have been an irreducible element of required administrative due process from the get-go,85 and characterized the Cold War hearings, no court appears yet to have required it in this context.86 The facelessness of the DHS redress procedure, ultimately decided by the TSC, has not yet been disturbed.

Conclusion

A prize-winning student essay at Columbia Law School87 has persuasively chronicled judicial use of the CIPA procedures in assessing procedural claims outside the criminal context for which they were created. Yet the courts have not made full use of the possible analogies. Of the criminal context, recall, the Supreme Court remarked in Reynolds that “since the Government which prosecutes an accused also has the duty to see that justice is done, it is unconscionable to allow it to undertake prosecution and then invoke its governmental privileges to deprive the accused of anything which might be material to his defense.”88 Might one not equally argue “since the Government which deprives a person of liberty or property also has the duty to see that justice is done, it is unconscionable to allow it to effect that deprivation and then invoke its governmental privileges to deprive the person deprived of anything which might be material to his defense”? The Cold War cases regularly required explicit congressional authorization of suspect procedural departures; the no-fly list and its procedures have no such authorization.

The present situation has elicited a range of dark scholarship addressing the impact of emergency on liberties and procedures.89 A colleague has argued in celebrated scholarship that “If we are to protect our civil rights and civil liberties against [today’s terrorist] threats, the aggressive use of informants, surveillance, wiretaps, searches, interrogations, and even group-based profiling must be measured not only against the liberties these practices constrict, but also with respect to the liberties they may protect… It is simply not sufficient to say that augmenting [our usual law enforce-ment] practices with laws and practices more appropriate to counterterrorism is necessarily inconsistent with protecting our civil rights and civil liberties.”90 American courts remain conscious of the rights impinged upon by national security concerns and willing to offer some protection to them; yet in the contrasts between the treatment of procedural claims in the Cold War’s emergencies and today’s one may see a regrettable regression.