A few weeks back, Ben Wittes wrote a controversial post over at Lawfare on the latest Snowden disclosures, arguing that, “If you’re okay with dumping in the lap of a journalist 160,000 of the most personal conversations a signals intelligence agency can collect, then stop whining to me about ‘bulk’ or ‘mass’ collection.”  As Ben subsequently clarified, his point was not to criticize Snowden for possibly violating the Privacy Act, but to flag what he perceived as the hypocrisy of various media outlets and privacy and civil liberties groups in not criticizing these disclosures—and in thereby appearing to endorse the view that transparency of secret government programs is an unmitigated good. After all, secrecy and privacy are, in many ways, two sides of the same coin—such that those who believe in the virtues of the latter should have a modicum of appreciation for the government’s need for the former in at least some cases.

Indeed, whatever the merits of that specific episode, it illuminates a larger problem that both of us have observed not just in the ever-ongoing dialogue over surveillance reforms after and in light of the Snowden disclosures, but in public discourse over national security law, more generally: That, far too often, proposals to reform government counterterrorism and national security programs are demanding transparency in lieu of accountability—and missing the critical point that the former is just one (of several) means for achieving the (more important) latter. Part of this conflation may come from the different interests of the critics—some of which (e.g., civil liberties groups) may be anti-secrecy and pro-privacy; and some of which (e.g., the media) may be anti-secrecy and anti-privacy.

Regardless of the cause of this trend, by insisting upon greater transparency as a goal unto itself, critics have missed (or wrongly rejected) two separate, but closely related points.  First, recent revelations to the contrary notwithstanding, meaningful accountability of secret government programs is possible even without wide-scale transparency.  Second, there is an array of circumstances in which properly accountable government secrecy is not anathema to civil liberties—and where transparency, as such, might actually compromise individual rights. This is especially true where the government is protecting the confidentiality interests of third parties (e.g., under the Privacy Act), but it may also be true in at least some cases in which the government is protecting its own secrets.

Simply put, comprehensive transparency is neither normatively desirable nor practically achievable in the national security and counterterrorism spheres.  Instead, as we aim to show in this post, true progressive reform in the national security space should be focused first and foremost on measures that will increase accountability, a goal to which increased transparency is only one of a number of potential routes—and, indeed, one with which such transparency is sometimes at odds.  To unpack this argument, we focus on the two most significant external mechanisms for ensuring accountability of government national security programs: judicial review (Part I) and congressional oversight (Part II).  After summarizing the critiques of the status quo, we explain why increased transparency—even if a necessary means of improving accountability—won’t be sufficient, before highlighting what we view as the better way forward.

I.  Accountability Through Secret Judicial Proceedings

A.  Critiques of the Current System

The charge that judicial review provides an inadequate check on government national security and counterterrorism operations has a host of themes, all but one of which we largely agree with.

First, perhaps the most significant hurdle to meaningful judicial review is the virtual absence of meaningful retrospective civil remedies for unlawful government conduct in the name of national security and/or counterterrorism. Steve already documented this trend in detail in The New National Security Canon, but in a nutshell, courts have bent over backwards to (1) not recognize causes of action even for egregious constitutional violations; and (2) use other procedural doctrines to avoid passing, one way or the other, on the legal merits of the programs being challenged. Thus, even in the context of FISA, which provides an express cause of action for damages arising out of violations of the statute, the Ninth Circuit found a clever (if somewhat disingenuous) way to deny relief without passing on whether the government actually violated the statute.

Second, critics of the view that judicial review can provide meaningful accountability in the national security sphere are also quick to point to the review provided by the FISA Court and the FISA Court of Review—which, as we learned thanks to last summer’s Snowden revelations, has often (although not always) been somewhat desultory. Indeed, whether or not the telephone metadata program under section 215 of the USA PATRIOT Act is ultimately upheld by the courts, the one thing that appears clear is that it won’t be on the same grounds upon which the FISC relied in its analyses. If that’s the kind of judicial review that courts have been providing in secret, surely, it can’t be enough.

Third, and related, critics have also highlighted the extent to which courts in national security cases have historically been deferential—perhaps excessively so—to the government.  Unsubstantiated invocations of amorphous national security concerns often tip the scales in favor of rulings that would be unfathomable in more run-of-the-mill challenges to official action.  Indeed, this point appears to be one of the main thrusts of the major new report by Human Rights Watch released earlier this week on how the federal courts have mishandled terrorism prosecutions over the past decade.

Fourth, whereas the above criticisms go to why judicial review of national security or counterterrorism programs have yet to provide meaningful accountability, some have suggested it may be affirmatively inappropriate, insofar as courts arguably lack the competence to resolve such sensitive (and secrecy-laden) questions—and might themselves be the conduit pursuant to which national security secrets are publicly disclosed (whether accidentally or otherwise).

 B.  Why “More Transparency” Isn’t Responsive

Although we don’t share this last critique, we do agree with each of the first three criticisms of the role of courts in national security cases thus far.  But the key for present purposes is that greater transparency won’t meaningfully address any of them.

For starters, more transparency about the government’s conduct of its national security and counterterrorism policies won’t, by itself, have any bearing on the absence of meaningful judicial remedies if and when those policies violate individuals’ rights.  At most, greater public knowledge might help to overcome Article III standing hurdles (as in the litigation over the 215 program), but (1) the availability of standing in that context may be unique; and (2) there’s still the host of other procedural hurdles that better information won’t even begin to allow plaintiffs to surmount.  In this regard, consider the D.C. district court’s decision throwing out the lawsuit arising out of the targeted killing of Anwar al-Aulaqi.  It’s hard to see how more public information about the operation would have had any bearing on the procedural barriers the court relied upon in dismissing the suit.

Related, more transparency isn’t likely, by itself, to improve the quality of litigation before the FISA Court.  The problem there, as has been well documented elsewhere, isn’t the lack of public awareness so much as it is the lack of adversarial participation.  After all, even once the 215 orders became public, there was no way to challenge them directly before the FISA Court or Court of Review, and the Supreme Court refused to step in when EPIC sought such review. We’re not saying that greater transparency would have zero effect on judicial decisionmaking—quite the contrary. Indeed, we would hope that judges might think twice about statutory interpretations that might be subject to vigorous criticism if exposed to the light of day. But compared to the value added of a security-cleared “special advocate” who could present adversarial briefing and argument, the value of greater public scrutiny strikes us as marginal, at best.

In contrast, increased transparency might have a negative effect by incentivizing even greater judicial deference to the government out of a perceived need to protect what remains properly secret.  Thus, in the context of proceedings before the FISA Court, it seems much more likely that the court would be more inclined to push back against new government arguments—and to demand greater showings from the government—when it has confidence that such showings will not compromise the underlying secrecy of the programs being reviewed. In contrast, as the Seventh Circuit’s decision in Daoud illustrates, the regular courts have been reluctant (perhaps excessively so) to require the government to turn over such materials except where absolutely necessary to resolving the underlying legality of the surveillance. To be sure, there may be a greater-than-zero value to the potential outcry that would follow controversial public decisions by the FISA Court. But we’re not convinced that the benefits in this context outweigh the costs.

C.  A Better Way Forward

It’s not that greater transparency in the national security sphere is bad; it’s just that we think it’s not necessarily effective in improving the ability of courts to hold the government accountable in the national security and counterterrorism spheres. Instead, if the real goal of reformers is accountability, we think reforms should focus instead on the following two basic ideas:

  1. Clearer civil remedies. Any discussion of increased judicial accountability must begin with clearer civil remedies for unlawful government action. Whether or not the current state of various judicial doctrines is normatively defensible, it has the practical effect of virtually foreclosing any relief even for the most egregious unlawful conduct by government officers in the conduct of national security and/or counterterrorism policies. As Steve has proposed in the specific context of targeted killing, Congress should generally consider providing an express cause of action for damages against the relevant government officers (or the government itself—if the officer was acting within the scope of his employment), and abrogating the relevant privileges and immunities that would otherwise preclude relief. This should be a no-brainer in any discussion of increased government accountability, but the need for such a reform is especially acute in the national security space.
  2. Meaningful adversarial participation in secret litigation.  If there are contexts in which the opposition to providing such remedies is the need to maintain governmental secrecy, then Congress can and should provide for meaningful adversarial litigation even if many—if not most—of the details must remain secret.  Not only has Congress already so provided in the context of certain kinds of “alien terrorist” removal hearings, but the Guantánamo litigation of the past six years also provides illuminating precedents for how ordinary federal courts can hear cases involving highly sensitive government information while balancing the government’s need for secrecy with the plaintiff’s need for meaningful review.  For those government programs that are properly kept secret, such meaningful adversarial participation is the best way to ensure that courts, at least, are doing as much as they can to hold the government accountable. (And, needless to say, the meaningfulness of the participation will matter quite a lot.)

II.  Accountability Through Secret Congressional Oversight

Of course, courts are neither the primary nor exclusive forum for providing oversight and accountability of government national security and counterterrorism policies. Equally important to this conversation is the role of congressional oversight. And if judicial accountability of national security and counterterrorism policies has left a lot to be desired over the past decade, legislative accountability has perhaps fared even worse.

A.  Critiques of the Current System

Many commentators, and some members of the Senate and House, argue that the congressional oversight of U.S. national security and intelligence apparatus is broken.   There are a number of different lines of criticism, with which we largely agree.

First, observers have been quite critical of coziness between the congressional committees and the Executive Branch agencies within their oversight jurisdiction. Andy has written about some of the institutional incentives within Congress that lead to episodes of “clientitis.”  The recent Stimson Center report on drones also makes some relevant observations about committee behavior.  This problem is especially pronounced in oversight of the intelligence activities due to the intelligence committees’ exclusive jurisdiction over any matter related to sources and methods.  As one article in The Nation argued, the committees’ “members often seem more committed to protecting, rather than scrutinizing, the agencies they are tasked with overseeing.”  Some also focus on overlapping and nonsensical lines of accountability for the U.S. Department of Homeland Security (DHS), whereby it is both drowning in oversight interactions and less accountable to any mission critical priorities at the same time.  Where possible, the Executive Branch engages in forum shopping to find more accommodating committees.  Moreover, Senators and Representatives tend to find it much less politically risky to err in favor of the national security secrecy arguments rather than risk blame in the aftermath of an attack.

Second, Congress suffers from institutional mismatch with a sprawling Executive Branch in terms of size and information disadvantage.  Much like the challenges faced by a board of directors when it comes to management oversight of a Fortune 500 company, Congress is at a serious time, location, and information disadvantage.  Effective congressional oversight requires a mix of constructive interbranch relationships and aggression.

Third, critics charge that oversight access to secret programs is too restrictive within Congress.  Members of Congress from both sides of the aisle (e.g., Reps. Justin Amash (R-Mich.) and Alan Grayson (D-FL)) complain that, as a condition of access to intelligence program information, they are not able to consult with cleared staff, take notes, or even discuss the program at issue with other Members of Congress.  The National Security Act itself restricts disclosure of certain types of secrets to the “Gang of Eight.”

Fourth, and a related point, there has been longstanding criticism of overclassification by the Executive Branch.  Notwithstanding congressional hearings (here), studies (see here, here, ), and requested studies (see here) by the U.S. Government Accountability Office, overclassification is still prevalent and serves to complicate congressional oversight, Freedom of Information Act requests, and other transparency mechanisms.

B.  Why More Transparency Isn’t Responsive

Transparency is incredibly important to a democracy ultimately guided by an informed citizenry. However, the value of public transparency is somewhat more complicated in a republican, representative form of democracy.  Congressional oversight is one of the critical mechanisms for expression of the republican form, and in that frame the premium is on transparency to Congress rather than public transparency as an end in itself.  Public disclosure will necessarily chill conduct; for our purposes, we need to decide what conduct should be chilled, and how cold we should make it.

There are a few important downsides to public transparency in the congressional context beyond the standard concerns—that release of national secrets could burn assets, compromise operational security, damage foreign relations, or encourage illicit weapons proliferation.

Public transparency is a double-edged sword.  On one hand, it disincentivizes misbehavior—immoral, unconstitutional, or otherwise illegal—in absolute terms.  There is real value to the cliché that sunshine is the best disinfectant.  However, as a relative matter, public transparency disincentivizes Executive Branch disclosure to Congress.  Notwithstanding the recent bickering between the Senate Select Committee on Intelligence and Central Intelligence Agency over the handling of classified information, and a few other dust-ups, Congress has generally taken its national security secrecy obligations seriously (as evidenced, for example, in Members’ responsible use of their Speech or Debate Clause immunity).  If the Executive Branch could not have some level of confidence that Congress could maintain secrets, it would likely resist disclosure even more than it does at present.

In addition, public oversight of sensitive national security matters creates unconstructive incentives to derail the fact-finding process with partisan politics.  One need look no further than the congressional investigations of the attack on U.S. facilities in Benghazi, Libya and the problematic law enforcement tactics along the southwest boarder in Operation Fast and Furious to see that grandstanding is the norm rather than the exception.  There, what remain of the Congressional-Executive information access disputes largely go to the politics of administration reaction rather than the underlying conduct under investigation.  As Andy has written previously, there has been more heat than light.

As a general matter, transparency is a good thing, but it must be in the service of other societal ends.  Once we accept that there are valid secret government activities, we have to focus on accountability for those who decide what they are and how they are conducted—and how to provide that accountability without public transparency.

C.  A Better Way Forward

Again, if the real goal is accountability, we think reforms should focus on the following a few basic ideas designed to enhance congressional oversight:

  1. Clear primary and secondary oversight jurisdiction in each chamber.  There needs to be serious committee reform designed to promote committee oversight competition between committees of primary (singular) and secondary (interdisciplinary) jurisdiction.  Therefore, Congress should consider: (a) extending jurisdiction over intelligence sources and methods to a Select Subcommittee on National Security and Intelligence (SSNSI) in each chamber; (b) in the Senate, the SSNSI should be housed under a reconstituted Senate Government Affairs Committee where it would provide secondary competition to SSCI and other primary committees along with interagency jurisdiction; (c) the Senate should establish a freestanding Homeland Security Committee and consolidate the majority of DHS’s primary jurisdiction (as the 9/11 Commission leaders recommended); (d) in the House, the SSNSI should be housed in the Committee on Oversight and Government Reform, where it could likewise provide competition and interagency vision.  We would also consider an appropriate civil liberties oversight role for the House and Senate Judiciary Committees, even in the context of the intelligence community—the activities of which have historically fallen outside their purview.
  2. Meaningful participation by cleared members and congressional staff.  The National Security Act and other authorities need to ensure that members of Congress are able to do their jobs, which include the ability to communicate with one another and a limited number of cleared staff.  Like the adversarial process we have witnessed in the Guantánamo litigation, we can widen the circle of those with access to sensitive information to appropriate Article I and Article III officers without unduly risking national security.
  3. Legal whistleblower access to congressional overseers.  There needs to be a well-publicized safe harbor for whistleblowers to communicate with Congress—and not just those Members most likely to have already signed off on the underlying programs—about matters of national security.  It could perhaps provide some obligations as to progression, such as an obligation to either go up the chain of command, or to the relevant Inspector General, or to the U.S. Department of Justice first, but the congressional oversight process should have institutionalized incentives for national security whistleblowers to go to congressional committees that can handle national secrets.
  4. Classification and overclassification accountability.  The culture of secrecy is strong within the national security establishment, but its excesses defeat transparency and accountability.  This is a delicate balance.  In order to tackle that culture in a responsible way, there need to measured disincentives to overclassification as a function of employees with classifying authority as well as their senior manager.  The Brennan Center has made some sensible recommendations along these lines.

III.  Conclusion – On The Virtues of Democratic Secrecy

Some will no doubt misinterpret the above post as a criticism of calls for transparency-oriented reforms.  Nothing could be further from the truth.  In any number of contexts, we unhesitatingly support calls for greater transparency from the government—including subjects on which we believe the public has a right to know, e.g., who counts as an “associated force” under the AUMF; where we’re using military force; and so on. There is simply no question that the government can and should be more transparent in any number of ways when it comes to national security and counterterrorism policies.

But we don’t join reflexive calls for transparency for its own sake–e.g., arguments that all OLC legal memoranda should be public. We certainly agree that the underlying substance thereof should be publicized in those circumstances in which the government is taking liberties with putative democratic mandates (and thereby acting in an antidemocratic matter). But if one accepts, as we both do, that some lawfully secret (and secretly lawful) government programs are properly kept secret, the question becomes how to ensure that those programs are subject to meaningful checks through secret accountability mechanisms.

To us, too much ink (and blood) is being spilled on transparency as the magic bullet to resolve everything–whether in debates over surveillance reform, targeted killings, or a range of other hot-button topics in national security and counterterrorism law and policy.  In some instances, transparency will not only not solve many of the identified problems, it will be antithetical to optimal protection of privacy and civil liberties.  This is so, we believe, because in some cases calls for greater transparency will provoke the government to resist reforms that might have otherwise led to greater accountability; and in others, greater transparency will directly lead to the revelation of personal information that ought to have remained private.

In short, we believe that national security reform should be a conversation first and foremost about accountability—and that we therefore should be having a conversation about how we improve the existing accountability mechanisms through which we preserve democratic secrecy. (And what, if any, accountability mechanisms should be utilized in cases in which the existing mechanisms are likely to be inadequate or otherwise insufficient.)  If the surveillance reform debate is any guide, we should be very disheartened by what’s happened thus far, to wit: cosmetic changes to the FISA Court and no oversight reform in exchange for all-but cosmetic changes to the underlying substantive authorities.  Neither transparency nor accountability interests are advanced by such reforms.

Ultimately, the more this conversation is about transparency, the less it’s about accountability. And that, to us, is the wrong approach for anyone and everyone who aims to think carefully about the future of U.S. national security and counterterrorism policy.