A Proposal To Reform FISA Court Decisionmaking

Yesterday’s New York Times story on the secret legal opinions of the FISA court prompts a natural question: How should the FISA court reach its decisions, and how do we know it is doing so correctly? That breaks down into two questions. First, what procedures should the FISA court use to reach legal conclusions? And second, when or how should those legal conclusions be made public? The latter has received much more attention than the former. Like a lot of people, I tend to think that it wouldn’t impact national security for the FISA court to release more information about its decisions, at least in those cases when the judges consider abstract legal issues.. Perhaps the court could issue opinions in redacted form; perhaps it could simply release a summary of its legal conclusions and reasoning. Either way, a lot of people have voiced opinions on that issue. In this post, I want to focus on the first question that hasn’t received as much attention: What procedures should the FISA court use to reach its legal conclusions?

The question is hard to answer in part because we don’t know much about the procedures the FISA court presently uses. From what we know, it seems that only the government appears before the FISC judges (putting aside the procedures described in 50 U.S.C. 1881a(i)). We don’t know for sure, but it seems likely that DOJ lawyers apply for an order and the FISC issues an opinion to settle some legal issue raised by the opinion. Presumably they ask for briefing from DOJ, but there’s no evidence they ask for briefing from any other side. The Court then issues an opinion based only on the application and briefing from one side. I’m not sure it works that way, but it seems like a fair guess.

If that’s what happens, though, it is a terrible environment for legal decisionmaking. There are no known facts; the court only gets briefing from one side; and if the government wins, no one can appeal. This is a lousy way to interpret the law. The problem isn’t just that judges who only hear from one side will tend to favor that side. That’s part of the problem. The broader problem is that judges don’t have the adversarial hearing and factual record needed to make a reliable legal decision. Some judges will get lazy and favor the side that briefed the issue; others will feel that they have a duty to be adversarial and favor the other side that was absent. Either way, it’s not a procedure designed to generate reliable legal decisions. (For more details, see my amicus brief before the Fifth Circuit in the still-pending cell cite case on the problems with lawmaking in the context of ex parte court order applications.)

In the criminal law context, we use ex parte procedures like warrant applications as a simple check for a few ex ante thresholds and we leave the real decisionmaking for subsequent ex post litigation in the form of motions to suppress or civil lawsuits. That doesn’t happen in the national security context. The point of watching the bad guys in the national security setting is to find them and watch them, and the surveillance ordinarily will never become public. So it would be good to have sort of procedure to generate more reliable legal interpretations.

Here’s one tentative proposal on what such a procedure could look like. Congress could amend the FISA statute to create a new role for the Oversight Section of the National Security Division at DOJ. The Oversight Section should have a right to file a motion to oppose any application before the FISC. The motion to oppose must be filed within a certain number of days after the application has been made. It will be accompanied by legal briefing explaining why the application should not be granted. Alternatively, if the application has already been granted, the motion should explain why the issued order should be quashed or withdrawn. Under the statute, the FISC should have to entertain the motion to oppose and deny the application if the order would be contrary to statutory or constitutional law. The statute should also have a way to determine the facts. Perhaps the facts in the application should be taken as determinative, subject to review based on a deferential standard such as “clearly erroneous.” Or perhaps the two sides should be ordered to submit their proposed facts and the FISC can choose between them. Either way, there should be a way of settling the relevant facts and some kind of adversarial briefing on the legal issues, along the lines of the procedures currently found in 50 U.S.C. 1881a(i). The FISC can then issue an opinion deciding whether to grant the motion and deny/withdraw the order.

I realize this approach offers a middle ground that may please no one. From the government side, one objection will be that we can’t slow down the process with litigation. From the civil libertarian side, an objection will be that we can’t trust the Oversight Section to oppose the surveillance; instead, the thinking will run, we should have a person or group with civil liberties bona fides take on the role. These are fair concerns on both sides. Given that the FISA process is secret, it’s hard to know what procedures might introduce the needed adversarial review without causing other problems. But with that major caveat made, I very tentatively think that my approach is better. The litigation shouldn’t slow down the monitoring because the court can issue the order and then review the motion to quash/withdraw. And for the system to work, someone needs to have the clearance to review all the FISC applications and see which of them break new ground and should be opposed. It seems to me that it would be easier — and more politically feasible — to give this role to an oversight section within the executive branch rather than to a private group.