The Foreign Intelligence Surveillance Court: Is Reform Needed?

With the advent of the Edward Snowden leaks commencing in June 2013, much has been written about Snowden and the United States intelligence community. This short blog post examines one of the only proposals to emerge that would constitute systemic procedural change, namely the creation of a special advocate or institutional amicus system before the Foreign Intelligence Surveillance Court – the FISC (hereafter referred to as a “special advocate” reform, for ease of reference). Such a system would be beneficial for both substantive and procedural reasons. A recently passed House bill, which merely keeps the status quo by permitting the court to appoint an amicus — a power it has now — falls short of what is needed.

As a backdrop to this structural proposal, let’s step back and consider the framework of our national security system. Justice Souter presciently wrote in his concurrence in the Supreme Court’s groundbreaking Hamdan decision, that because the executive branch has the responsibility to protect the nation it may understandably give primacy to security concerns, at the expense of privacy. Indeed, the executive branch, faced with even a small chance of a horrendous terrorism event occurring on our shores, will feel the obligation to take whatever lawful steps it can to minimize the risk of the event occurring. This is what I call the “one percent” problem: faced with even a one percent chance of another 9/11 or Boston bombing, the executive will do what it legally can to reduce that risk even further. And so it should.

It is thus largely left to the legislature and judiciary, to the extent permitted, to balance that interest with our concern for civil liberties and transparency. Here I examine one aspect of that “check and balance” function, which is the FISC, to see whether reform would be beneficial.

The FISC

Let’s start by briefly going over what the FISC is. Some of the following may be familiar to Just Security readers, but I also highlight other features of the system that are particularly important in considering reforms.

  • The court was established by Congress in 1978, which also created the authority to obtain various legal processes for national security investigations (e.g., FISA warrants).
  • It now comprises eleven federal district judges. All are regularly confirmed Article III judges. They are not specifically selected as judges solely to serve on the FISC. Instead, they all have normal civil and criminal duties as district court judges, which they retain even if they are selected to be on the FISC.
  • The Chief Justice of the United States Supreme Court selects the judges who will serve on the FISC, from among all the Article III district court judges. The selection is final — i.e. it does not need to be confirmed by any other party.
  • There are eleven federal circuits and judges from at least seven circuits must be represented on the court.
  • The selected judges serve a term of seven years.
  • Three of the eleven judges must be from courts within 20 miles of Washington, D.C.
  • Importantly, a FISC judge sits for only a single week every 11 weeks (a week for each judge). Thus, there may well be less continuity — the sense of the court operating as a unit — than exists in the regular Article III setting. There may also be less judicial expertise as a result of the part-time nature of the assignment – a concern in an area where the law is complex, at times quite novel, and the consequences far reaching. That potential gap is filled, some would argue, by permanent court clerks, who work for the FISC; but critics would note that clerks are not judges and such a system may lead to overreliance on these non-judicial actors. Indeed, given the week-to-week schedule, the reliance on the FISC clerks may be extensive: the then presiding FISC Judge noted in his July 2013 letter to Congress that there is routine, informal interaction between its staff and DOJ. This is atypical in Article III courts, but is the admitted norm in the FISC.
  • In 1978 Congress also created a FISC court of review, which can hear appeals from FISC decisions. It has three judges, who are also selected in the same manner as FISC judges, but from the judges on the federal courts of appeals or the district court. The appeals court has sat only two times since 1978 to issue rulings.

The FISC routinely hears solely from the government, as is the case in a typical Article III criminal proceedings in which warrant applications and grand jury matters are being adjudicated. At times, in both settings, there may not be a particular person being targeted who could receive notice in order to be heard. But even where there is such a target, notifying that person would of course not be conducive to successful law enforcement and intelligence collection.

Post the Snowden leaks, the issue has arisen whether an institutionalized amicus or special advocate could play a useful function in the FISC, setting forth arguments that may not be evident to the FISC judges or its permanent staff. This outside voice would not represent a particular target, for the reasons noted. Such a system thus would not lead to the tipping off a target; and if structured in a manner akin to the Senator Blumenthal FISC reform bill, would not risk delay where speed is necessary (since the FISC would retain wide latitude, e.g. it could decline to await the advocate’s submission or hear it after the fact – though the need for speed hardly arises when considering the bulk collection authorities that have come to light, since those are re-presented to the FISC like clockwork every 90 days).

The FISC and Article III Courts

A principal argument made by advocates for the status quo – including by an eminent former FISC presiding judge – is that the process is working as evidenced by the statistics. These statistics were laid out for first time by then-Judge Walton in his 7/29/13 letter to Congress. In footnote 6, he noted a slightly lower rate of approval of FISA warrants than for Article III criminal warrants, although both are over 99%. A high rate of approval in both settings is understandable — the government seeks to present that which will succeed, whether it is a search warrant, Title III wiretap, or FISA electronic warrant. If you are in the government, you know the standard you must meet and you do your best to meet it and you do not present the matter until you believe you have the goods.

But is it entirely fair to view the two tracks–national security and criminal–as comparable? I don’t think so.

First, the nature of the issues under consideration in national security matters makes having another voice at the table of greater need. In the national security context, the legal issues presented in a government application can be far more complicated and have far greater consequences than in a routine criminal matter. The typical criminal warrant presents a quotidian factual determination as to whether probable cause has been established; while similar factual issues arise in the national security context in connection with individual warrant applications, complex, novel, and broad legal issues are more likely to present themselves in the FISC (see, e.g., the 215 NSA metadata program). And given the stakes, the government may be more willing in the national security context to push the envelope legally and get closer to the line – have chalk on its cleats – thus presenting closer issues for decision.

Second, although the FISC is composed of excellent Article III judges, they are trained civil and criminal lawyers, and do not come to the FISC as national security experts. Indeed, given the part-time nature of the judges’ FISC duties, there is simply not the same level of on-the-job training as exists in the Article III context.

Let’s use the NSA 215 program as an illustration. It is now public that prior to the leaks, it was reauthorized by the FISC over 30 times, by over a dozen different FISC judges (as the program came up for renewal every 90 days since its inception). No judge rejected the government’s applications. Prior to the leaks, there was no written opinion on the legality of the program. There was, however, one thoughtful and detailed decision on a different program that raised similar issues.

Post leaks, we have two written FISC decisions upholding the program, but a split among Article III judges – who had the benefit of adversarial process — as to its legality. Additionally, the President’s Privacy and Civil Liberties Oversight Board (PCLOB) — an independent executive agency, comprising serious and thoughtful people — has found, in a detailed report endorsed by a majority of its five members, that the NSA program was not statutorily authorized. It too had the benefit of input from myriad sources prior to issuing its findings. Of note, no FISC judge has yet to address the principal arguments made by the PCLOB as to why it finds the program not consonant with the 215 statute. That fact alone would suggest the need for an institutionalized advocate or amicus in the FISC, so that the court for both substantive and procedural reasons can address the arguments by the PCLOB, even if they are ultimately not winning arguments.

This lack of written decisions prior to the leaks, and lack of decisions even after the leaks addressing all the best counter-arguments, points to a third difference between the FISC and article III criminal process. The former is rarely if ever subject to the adversarial process. Prior to the leaks, for instance, not a single FISA application has been required to be disclosed to opposing counsel. By contrast, criminal rules require disclosure of warrant applications to “aggrieved” parties. Thus, although not all criminal matters end up in litigation, the specter of judicial review is palpable in a criminal investigation, with its attendant adversarial review. The opposite is generally true in a national security investigation.

Without such adversarial input, it is unrealistic to expect the FISC to come up on its own with all the arguments as to why a government application may be legally wrong. Even if you reject, as I do, the notion that the FISC is a rubber stamp – or that the selection process for FISC judges accounts for its substantive decisions — one can still be in support of another voice being heard in the FISC. That voice would not just give the public greater confidence in the process. That it would undoubtedly do and is no small advantage in favor of such a reform. The reform would also be a means to better decision-making, in an area where getting it right is of critical importance.

As another illustration, after the Snowden leaks, the government has filed under seal repeated applications to reauthorize the 215 program, and the court has also approved those applications, also under seal. The government’s applications and the court’s decisions have been unsealed only after the fact. But given the government’s public acknowledgement of the 215 program, there is no reason at all that those applications — and the court decisions – were sealed in their entirety and only belatedly unsealed. The process can and should be more transparent, but right now there is no institutional player to raise those issues with the FISC.

In sum, to seek a special advocate is not to question the integrity or acumen of the members of the FISC or the DOJ, but to provide the court with views that can better effectuate our dual concerns for security and civil liberties. It is easy to get lost in the minutiae of the Snowden “traitor or hero” debate, or even the value of the NSA 215 program, and miss the opportunity for meaningful change. A structural reform to the FISC — and a modest one at that, since it imposes no costs on our national security mission — is called for. The time is ripe for an institutionalized special advocate. 

Filed under:
About the Author(s)

Andrew Weissmann

Teaches National Security Law at NYU School of Law, Member of the Centers on Law and Security, Member of the Administration of Criminal Law, Former General Counsel for the Federal Bureau of Investigation