Access to National Security Information under the US Freedom of Information Act.

by Stephen J. SCHULHOFER, Robert B. McKay Professor of Lawm New York University, New York, USA.

Introduction

A genuine democracy is incompatible with secrecy. Meaningful citizen participation in policy formation and oversight presupposes access to relevant information. Information is the fulcrum on which every form of accountability turns.

In no domain is such access more important than in matters involving “national security” - the government’s responses to perceive external and internal threats to public safety and the territorial integrity of the State. Yet democratic societies typically give executive officials (who have multiple motives to opt for unjustified secrecy) the unilateral power to conceal any information to which they choose to attach the “national security” label. This common practice thus disconnects part of the essential machinery of democracy. In an era of transnational terrorism and ever expanding conceptions of what “national security” means, secrecy’s potential for eroding democratic values is growing, even at a time when societies seek more than ever to promote openness in government.

The US Freedom of Information Act (FOIA) gives courts the power (and responsibility) to counteract these tendencies. Although FOIA’s “Exemption 1” excludes from obligatory disclosure any “properly classified” information,1 FOIA instructs U.S. courts to review whether any secrecy classification was “proper” and thus sufficient to trigger Exemption 1.

This paper examines US judicial decision-making in FOIA Exemption 1 cases. It finds that American judges, with perhaps surprising frequency, have successfully compelled the release of important national security information. It also finds, however, that US FOIA has not always lived up to its potential. US courts often give a strong presumption of validity to executive secrecy claims, largely because they see national security as a domain of unique executive expertise. Such deference, and the assumption of executive expertise on which it rests, go to the heart of FOIA’s promise and limitations in the national security context. I argue, however, that judicial reverence for executive expertise and the converse assumption of outsider incompetence are both misplaced.

Though executive officers often have undeniable expertise in military matters and foreign affairs, they face powerful incentives to impose secrecy for reasons tied only to bureaucratic self-interest. Equally important, agencies in the executive branch are not the only ones that have national security expertise. Of course, individual judges typically do not have this kind of knowledge and experience. But courts as institutions routinely deploy procedures to insure that relevant expertise can be brought to bear. Moreover, appropriate judgments about whether to disclose information depend on striking a balance between the relative benefits of both secrecy and transparency. Judges may lack personal expertise in national security, but they are well placed to assess the value of transparency, and they are society’s quintessential specialists in crafting fine-grained compromises that duly accommodate conflicting interests. National security officials, in contrast, instinctively resist transparency as a matter of principle, and typically they have little appetite for the fine-grained, fact-sensitive judgments that can permit limited disclosure without posing any genuine risk to national security.

The paper explores the possibilities for using US FOIA (and analogous freedom-of-information authorities in other nations) to afford a more effective check on executive self-dealing and self-protection, in order to assure greater transparency in national security matters, while also maintaining robust protection for truly sensitive secrets.

§ 1 – The US Freedom of Information Act (FOIA)

Under FOIA, all US government records must be publicly available, except when a specific statutory exemption applies.2 Several exemptions potentially apply in national security matters, most centrally Exemption 1, which protects all information that “in the interest of national defense or foreign policy [is] properly classified.”3

This FOIA exemption is much narrower that the “state secrets privilege” available in other litigation. When a litigant seeks damages or an injunction against government action, state secrets privilege shields any information that a high official certifies as too sensitive to disclose.4 And if the litigation cannot fairly proceed without such information (for example, when “the very subject matter” involves a state secret or when a state secret is allegedly relevant to a possible government defense), state secrets privilege requires courts to dismiss the case.5 In contrast, in a FOIA case (one in which the applicant seeks only disclosure of specified documents), Exemption 1 procedures require much closer judicial oversight -- in six distinct ways:

1) FOIA requires agencies to identify classified material page-by-page and section-by-section;6 they cannot, as in a state secrets case, claim protection for an entire document simply because the subject matter involves national security.

2) FOIA imposes a filtering process, requiring that “[a]ny reasonably segregable portion of a record shall be provided … after deletion of the portions which are exempt.”7

3) Although state secrets precedent discourages judges from looking at the classified material in camera,8 FOIA carries no similar presumption against in camera inspection.9

4) In considering FOIA exemption 1 issues, the court must “determine the matter de novo, … and the burden is on the agency to sustain its action.”10

5) The FOIA mandate to segregate non-exempt material is bolstered by “a detailed and strictly enforced set of procedural requirements.”11 Exemption claims must be explained in a so-called Vaughn index, a meticulous log of requested documents, with a specific, non-boilerplate rationale for exemption in each case.12

6) Ex parte determinations are disfavored. FOIA courts insist that the Vaughn index provide “as complete a public record as is possible”13 and reject the index where, due to lack of specificity, it “does not permit effective advocacy.”14 In contrast, courts uphold state secrets privilege on grounds never disclosed to the plaintiff, occasionally acknowledging that this procedure “is no doubt frustrating” to the plaintiff.15

Although FOIA can thus be a formidable weapon for transparency in national security matters, its procedures have softened regarding both in camera inspection and de novo review.

In Camera Inspection

In 1973, in EPA v. Mink,16 the Supreme Court held that legislation then in effect gave the executive exclusive authority to determine when secrecy classification was appropriate. As interpreted in Mink, FOIA did not permit judicial review of government claims that FOIA disclosure was precluded by Exemption 1. The decision was widely condemned in Congress, which at the time was intensely resisting the claims of executive prerogative asserted by the Nixon Administration. As a result, Congress promptly overruled Mink; in 1974 Congress amended FOIA to require courts to decide whether records “are in fact properly classified.”17

The amendments encouraged judges to determine the propriety of classification by examining the targeted documents in camera,18 but from the start Congress indicated ambivalence on this score. The legislative history is replete with statements describing in camera inspection as crucial for “the accountability necessary for Government to function smoothly,”19 but the Conference Committee also stated that “[b]efore the court orders in camera inspection, the Government should be given the opportunity to establish by means of testimony or detailed affidavits that the documents are clearly exempt.”20

In the first decade following the 1974 amendments, courts frequently examined material in camera,21 but then many courts began deploying a presumption against it.22 Under one test, if “agency affidavits … demonstrate that material withheld is logically within the domain of the exemption claimed, and … are neither controverted by contrary record evidence nor impugned by bad faith on the part of the agency,” then in camera inspection is unnecessary, and summary judgment should be granted for the government.23 This standard almost returns the law to its pre-1974 bar on judicial oversight, because a petitioner has little hope of demonstrating that documents held by the intelligence community are not “logically within the domain” of national security and has no way to provide “record evidence” contradicting the agency affidavit. Some courts have made the petitioner’s burden even more difficult by accepting Vaughn indexes and agency affidavits under seal, so that the judge never sees the allegedly sensitive material and the petitioner never even sees the government’s description of it.24

Compounding these obstacles to in camera review is the “Glomar response,” in which an agency refuses to confirm or deny the existence of the documents sought, on the theory that merely acknowledging their existence could disclose protected information.25 Courts have permitted a Glomar response even when they find it hard to credit the factual basis for the claim.26 When upheld, of course, the Glomar response moots any possibility of in camera inspection, because there are no longer any known documents to inspect.

Despite these developments, in camera inspection has by no means disappeared.27 But prevailing conceptions of expertise and appropriate deference often seem to make independent judicial inspection in camera a superfluous formality. As a result in camera inspection occurs far less often than a system of assertive oversight would seem to require.

De Novo Review

As in the case of in camera inspection, Congress’s commitment to de novo review was ambivalent. The 1974 amendments, as originally proposed, limited Exemption 1 review to determining whether classification had a “reasonable basis.” But on the Senate floor, that language drew strong opposition and was stricken, with the effect of extending to Exemption 1 the same de novo standard applicable to review of other FOIA exemption claims.28 After that version passed both houses, President Ford (who had succeeded Nixon after the latter’s resignation under fire) vetoed it, calling judicial oversight of classification an unconstitutional infringement on executive authority. He also proposed in the alternative that courts “would have to uphold the classification if there is a reasonable basis to support it.”29 Congress ignored Ford’s compromise proposal and overrode the veto, enacting the bill with the de novo review standard intact.30 As the US Court of Appeals for the DC Circuit explained shortly after the enactment:31

[T]he Administration … argued that de novo responsibility … could not properly be assigned to judges, in part because of logistical problems, and in part because of their lack of relevant experience and meaningful appreciation of the implications of the material involved. Those who prevailed in the legislature … [rejecting that view,] stressed the need for an objective, independent judicial determination, and insisted that judges could be trusted to approach the national security determinations with common sense, and without jeopardy to national security.

But while Congress rejected “reasonable basis” review, it left room for some shading of the de novo standard. The Conference Committee report noted that:32

the Executive departments … have unique insights into what adverse effects might occur as a result of public disclosure of a particular classified record. Accordingly the conferees expect that Federal courts, in making de novo determinations … will accord substantial weight to an agency’s affidavit concerning the … classified status of the disputed record.

Early decisions applying the new standard hewed closely to the spirit of de novo review.33 But most courts quickly became more deferential, and deference has remained the dominant pattern,34 especially after a 1985 Supreme Court decision declared that CIA assertions concerning national security impact “are worthy of great deference given the magnitude of the national security interests and potential risks at stake.”35 In effect, the “reasonable basis” standard rejected by Congress in 1974 has become the test that most courts apply in practice.36 Deference is sometimes carried even further. Courts have declared that agency affidavits are entitled to “utmost deference,”37 that a FOIA judge must accept the affidavit if it is “not called into question by contradictory evidence in the record,”38 or that deference is required whenever the documents sought “arguably” or “logically” fall within Exemption 1.39

If these readily invoked arguments should happen to be unavailable, the government can still support an Exemption 1 claim by resorting to the so called “mosaic” theory. In Halperin v. CIA, the plaintiffs filed a FOIA demand for disclosure of amounts that the CIA had paid for legal fees. Acknowledging that there was no apparent way this information could endanger national security, the DC Circuit nonetheless found Exemption 1 applicable: “[M]uch like a piece of a jigsaw puzzle,” the court said, this bit of information “[w]hen combined with other small leads, could well prove useful for identifying a covert transaction.”40 When this argument is pressed, courts are left with little role to play; even “seemingly innocuous”41 information can qualify as “properly classified” and therefore protected from FOIA disclosure.

After Halperin, mosaic arguments proliferated,42 and courts usually have acquiesced, even in cases involving exceptionally far-fetched claims.43 The mosaic theory is so elastic that it can encompass dangers the experts themselves do not perceive. A Seventh Circuit decision upheld the CIA position that all of its information from public sources was exempt from disclosure, on the ground that if the agency were required to identify and release non-sensitive items, “whoever makes the decision on the behalf of the CIA may miss some clue that foreign intelligence services would catch and thus may inadvertently reveal secrets.”44 The theory of unseen but perilous mosaic inferences thus allows secrecy not only for alleged dangers a judge cannot appreciate but also for imagined dangers a trained intelligence expert might overlook. When taken to that extreme, mosaic theory in effect precludes any possibility of judicial review. Mosaic arguments, because they can’t be falsified, in effect force courts to abdicate their reviewing function altogether.45

The reasons for judicial resistance to de novo review are not mysterious. The 1974 Congress concluded that federal courts, after giving “substantial weight” to the agency’s view, can then make a sound independent assessment. But many judges believe they are not competent to disagree with national security experts.46 Compounding that concern is anxiety about the magnitude of the harm if they should err in discounting the dangers of disclosure.47

In giving primacy to such concerns, however, most courts have simply ignored the congressional judgments that motivated the 1974 legislation. Congress found that decisions of “experts” were often guided by bureaucratic self-interest, not by their expertise. It found that insufficient oversight had led to abuses and to a weakening of national security itself. And Congress expressed a foundational political value, that just as too much disclosure can be dangerous, too much secrecy can harm the nation as well. Federal judges confronting Exemption 1 claims typically give little or no weight to these concerns.48 They posit that agency decisions are motivated solely by concern for national security harm, even when circumstances strongly suggest the contrary.49 And they assume that any risk of imprudent disclosure is intolerable, while imprudent secrecy is assumed to impose little or no social costs.

Assessing FOIA’s Impact

Although FOIA oversight is far from robust in the national security arena, the statute remains a significant vehicle for transparency. In camera inspection, if less frequent than Congress expected, is by no means rare.50 Likewise, review de novo is often an empty fiction, but many judges bring skeptical scrutiny to Exemption 1 claims.51 Moreover, the evidence suggests that the prospect of meaningful oversight -- through in camera inspection and non-deferential review -- has provided a significant check on extravagant secrecy claims, even when in the end the court has merely confirmed the agency’s position.52 Finally, FOIA cases create a more subtle dynamic of transparency, because litigation triggers de-classification review within the agency, often followed by partial disclosures under settlement agreements with FOIA claimants. Thus, beyond the visible record of court-ordered disclosures, the statute has prompted an impressive array of ostensibly “voluntary” national security disclosures that almost certainly would not have occurred in the absence of indirect pressure attributable to FOIA.53

Impressions of the FOIA process are often dominated by cases in which disclosure carries the potential for the page-one political bombshell. In such cases, the agency and the White House are usually determined to preserve secrecy at all costs, and FOIA litigation will seldom overcome such resistance. But high-visibility, high-stakes cases are the exception. Thousands of noteworthy classified documents have faced FOIA scrutiny and thousands have been released, with important revelations concerning U.S. diplomacy, Justice Department and CIA legal opinions, FBI surveillance of public officials, and countless other matters of legitimate public concern.54 Dozens of books and articles have been written on the basis of this “invaluable” material.55

Recently, moreover, judicial oversight has become increasingly vigorous. Many FOIA decisions signal discomfort with executive demands for deference, mirroring the uneasiness emerging in Congress and in public opinion. Deference to intelligence community assessments remains common, and departures from that pattern can be hesitant and unpredictable. Where such departures occur, moreover, the executive branch under President Obama continues to push back tenaciously.56 But growing judicial assertiveness is unmistakable. Courts often insist on in camera inspection,57 and orders to release classified documents are no longer unusual.58 In a striking but not isolated example, the Second Circuit recently mandated the release of photographs that vividly document the abuse of Abu Ghraib detainees, rejecting impassioned administration claims that the disclosure would endanger the lives of American troops.59

In sum, although assertive judicial review has been episodic and often disappointing, independent oversight has unquestionably enhanced governmental transparency, even in the highly sensitive area of national security secrets.

§ 2 – FOIA’s Place in the Wider Information Access Environment

In the US -- as no doubt in other Western democracies -- statutes like FOIA do not provide the only avenue for public access to information. Whistleblowers and the press, for example, have made substantial contributions, as the Snowden and Manning revelations make apparent. Yet even those voluminous leaks have involved only a small portion of the national security information that remains unjustifiably classified.60 And in any case, a democracy under the rule of law can hardly stake its survival on the willingness of whistleblowers to commit serious crimes in order to expose matters of justified public concern.

American reporters enjoy a degree of constitutional protection from prosecution that the press cannot claim in many other countries, as the strictures of the UK Official Secrets Act make clear. But even in the U.S., where reporters are somewhat shielded, their sources are not. Manning was court-martialed for violating the US Espionage Act and is currently serving a sentence of 35 years’ imprisonment.61 Snowden remains a fugitive facing prosecution for harshly punished crimes. A CIA officer who leaked information to NY Times reporter James Risen was recently convicted of nine serious offenses and faces a potential sentence of many decades in prison.62 Under those conditions, uncommon courage on the part of public-spirited officials cannot obviate the need for formal legal mechanisms to grant information access to reporters and other interested citizens.63

The US Congress has potent, legally legitimate information-access tools, but it usually has little appetite for using them constructively.64 As a result, oversight in the American context often has been weak or misdirected, even when control of Congress and the Executive Branch is politically divided. And at best, Congress’s tools operate only in one direction. They allow Congress to get national security information for itself but give Congress almost no authority to release classified documents and testimony to the public.65 And this inability to share classified information with the press and the general public undermines the access privileges that Congress itself enjoys, because Congress’s motivation to seek information and to act on it depends on political incentives, which are likely to remain dormant in the absence of public awareness.

Formal information access rights under FOIA therefore assume crucial importance. FOIA creates legal authority for courts to force disclosure when they conclude that alleged needs for secrecy are overblown. In addition, FOIA judges, unlike those assessing state secrets privilege, can make fine-grained judgments, filtering out sensitive material and releasing the remainder. In that process, FOIA judges can also encourage semi-voluntary disclosure and restrain government tendencies to conjure far-fetched national security fears. FOIA, of course, is concerned only with access to information, not with formal remedies for executive abuse. But if FOIA makes the facts available, the press, the public, and then Congress can do the rest, at least on matters that arouse majoritarian concern.

Nonetheless, FOIA practice has exposed practical limitations that do much to offset its promise. Many courts, reluctant to honor the statutory mandate for de novo review, grant the executive a degree of latitude that often seems equivalent to carte blanche. Overall, the statute, as currently applied, the statute affords offers only a weak, somewhat unpredictable weapon for challenging executive control of national security information.

§ 3 – Consequences

Exclusive executive control over national security secrets is harmful to all the institutions concerned. Without access to relevant information, Congress cannot wisely decide whether or how to legislate, courts cannot accurately adjudicate disputes or enforce the rule of law, and citizens cannot intelligently participate in public affairs.

The unchecked monopoly is a mixed blessing even for the executive branch. Over-classification generates a staggering volume of paper and electronic files in need of protection. One survey found that government archives contained over 1.5 billion pages of classified material that was more than 25 years old.66 In 1997 alone, more than 6.6 million new secrets were created,67 and by 2007 the pace of classification had tripled.68 Yet most assessments suggest that “roughly nine-tenths” of classified material does not need to be so treated.69

Over-classification hampers the intelligence community itself. Out-of-pocket expenditures alone are enormous.70 Worse are the intangible costs. Elaborate secrecy and the need-to-know principle lead to counter-productive “stove-piping” – information is so highly compartmentalized that decision-makers are partially blinded, seeing only part of the relevant data.71 Constrained vision has been a notorious cause of major blunders, not least the intelligence failures leading up to September 11, 2001.72 Moreover, secrecy dilutes oversight within the Executive Branch. Absent accountability for bad choices and incompetence, incentives atrophy and quality control withers.73

The implications for democratic processes are more fundamental. That secrecy impedes political deliberation is self-evident. Moreover, this consequence, often, is not a side-effect of secrecy but its very purpose: “The classification system … is used too often to deny the public an understanding of the policymaking process, rather than for the necessary protection of … highly sensitive matters.”74 As Arthur Schlesinger warned, secrecy gives the executive three strong weapons: to withhold, to leak and to lie.75 By withholding information, the executive prevents outside participation in policy decisions. By leaking, the executive builds support on an artificial basis – disclosing only those facts that serve its purposes. Lying allows the executive to pursue policies without having to account for them at all.

Many acknowledge these problems, at least in general terms, but argue that in times of national danger, unilateral executive control of sensitive information is justified by the unique expertise of executive officials in matters pertaining to national security.

§ 4 – Which Body Can Best Make the Required Decisions about Transparency?

Opponents of judicial oversight in national security matters believe that this controversial function lies outside the proper mission of the courts. But in a society governed by the rule of law, the core mission of the courts -- with their relative insulation from distorting personal and political incentives -- is precisely to resolve cases and controversies between individuals and government, and to provide a remedy when officials exceed their lawful authority.

A more subtle attribute of courts, less well-known but equally crucial, is their attention span. If judges were simply self-governing overseers, free to intervene (or not) as they saw fit, their ability to check executive overreach would inevitably be episodic and undependable. But courts are a very different kind of institution. They operate under a mandate to receive claims from aggrieved citizens, and then to assess and resolve those claims through the detailed steps prescribed for adversarial litigation.

A legislature, of course, lacks any comparable obligation. And legislators face political incentives that in practice tend to weaken any commitment they may have to diligent oversight.76 As public concern and attention shift, as they inevitably will, legislative bodies all too often lose interest in a problem and move on to more newsworthy matters. Courts have no such prerogative; they offer concerned citizens an assured focus and a staying power that legislative checks rarely duplicate.77 Importantly, therefore, oversight functions assigned to the courts stand a far greater chance of actually being performed.

National security matters pose a distinctive difficulty because of the sensitivity of the material that judges are called upon to review; opponents of judicial oversight routinely cite the concern that courts are prone to leaks. Yet in criminal cases, American judges conducting trials under the Classified Information Procedures Act (CIPA)78 have developed experience dealing with vast quantities of classified material. In CIPA cases, U.S. courts have tested and fine-tuned the necessary document-protection measures, without a single recorded mishap.79

That said, CIPA provides only a partial test of concerns about courthouse security, because in criminal cases the prosecution can use its charging discretion to insure that the most sensitive matters are never exposed to CIPA processes. FOIA has broader scope, and most importantly, the choice of what classified material to subject to oversight is not under government control from the outset. Thus, FOIA provides a window into judges’ capacity to handle any classified material safely and constructively.

In that regard US FOIA has been a success on many fronts. It has exposed self-interested secrecy and made available mountains of improperly classified material. At the same time, there is no known national security damage attributable to such disclosures, and no record of unintended leaks attributable to FOIA review. Yet because judges typically use their FOIA powers with restraint, past experience under FOIA leaves open the question whether more active judicial oversight would simply produce greater benefits, or would instead simply produce greater national security risk.

FOIA litigation, however, has shifted discussion about national security expertise away from abstract rhetoric and focused it on concrete disputes over specific portions of specific documents. That experience demonstrates that debates about expertise are distorted by a profound misconception. Commentators who are skeptical of judicial review invariably stress that judges almost always lack any specialized knowledge or experience in intelligence matters.80 But judges also lack training in many other subjects that they routinely decide. They usually know little or nothing about the level of care that is called for when a psychiatrist diagnoses mental illness, when doctor performs surgery or when architects and engineers design buildings, roads and tunnels. But judges are specialists in assessing such matters independently and critically assessing the self-interested claims of opposing litigants.

A sound decision about whether the public should see sensitive secrets requires the ability to evaluate conflicting arguments impartially and to strike careful compromises, getting assistance when appropriate from independent experts, a special master, a special advocate or (ideally) security cleared counsel -- the latter being the preferred approach in US criminal prosecutions that implicate classified information.81 Sound judgment requires above all a generalist’s perspective and the capacity to arbitrate between rival claims. In those domains, judges are the experts. National security officials, in contrast, usually have no particular skill, experience or disposition to reach well-balanced conclusions regarding the relative values of secrecy and transparency. To the contrary, they have strong incentives to seek unjustified secrecy, for reasons of self-interest or bureaucratic and political self-protection. Courts, therefore, are far less likely to undervalue the public interest in disclosure, and this important strength has even greater relevance on matters that concern politically weak minorities.

§ 5 – Solutions

In US freedom of information law, one of the major obstacles to effective judicial oversight is simply psychological. FOIA courts already have a legal mandate to carry out de novo review of the question whether national security material is properly classified. But most judges simply refuse to exercise their existing powers in that regard. If the judiciary is determined to be passive, what can be done?

Legislative insistence on the importance of judicial oversight (in the US case, emphatic re-affirmation of this principle) would represent one important first step. Such a change, of course, would make FOIA even more time-consuming and expensive than it already is. Critics of transparency litigation cite FOIA’s cost ($400 million annually) as a big strike against it.82 But that figure, the cost of FOIA demands across the entire government, is only a fraction of the counterpart expenditures for maintaining secrecy just within the publicly reporting national security agencies ($10 billion annually),83 and little more than a rounding error in the budget of the entire enterprise (the U.S. government) that FOIA monitors (almost $4 trillion annually).84

With respect to national security material, the US FOIA process already generates substantial transparency benefits, but it remains disappointing and imperfect. Steps to make it more effective will, in the long run, depend less on doctrinal and statutory reform than on efforts to raise judicial awareness and a persistent focus in public conversation on the need for an independent checking mechanism. Most American judges focus only the importance of national security expertise and tend to discount their own strength in other equally essential skills. This psychological dimension of the problem is not easy to fix. But more prominent public discussion of the relevant judicial expertise could undoubtedly have significant impact.

Conclusion

Citizen participation in public affairs requires access to information. Yet in national security matters, we give the executive largely unchecked power to control whatever information it chooses to consider sensitive. We have done so unwisely and unnecessarily, because citizens, lawyers and judges alike have held a myopic view of the required expertise and of the issues at stake in judgments about secrecy.

Although societies have long accepted executive monopoly over control of information in the national security area, new developments permit the scope of executive control to expand exponentially. Smaller, cheaper weapons of mass destruction, less complex delivery systems, and less deterrable adversaries create an environment of heightened, never-ending danger. Supposedly vital national assets can now be attacked from any cyber café in the world. National security secrecy is nothing new, but its potential for eroding the structures of democracy is growing.

The executive’s claim to exclusive control over national security secrets is buttressed by the widely accepted assumption of a unique executive expertise and the corresponding assumption of outsider incompetence with regard to military matters and intelligence. Yet even in these distinctively specialized areas, many capabilities necessary for striking the right balance between secrecy and transparency are found not in the executive branch but only in independent oversight bodies like the courts. Judicial decision-making expertise can subject large areas of national security activity to accountability and the rule of law, without exposing a nation to greater danger. Indeed, robust checks and balances are essential for safeguarding any nation’s security effectively over the long-term.