The FISA Court and Article III: A Surreply to Orin

As I suspected it would, the exchange between my friend Orin Kerr and me on the constitutionality of the appellate review provisions in the Senate version of the USA FREEDOM Act has morphed into a broader conversation about whether the FISA Court itself is consistent with Article III. Consider the three major points of Orin's reply to me: (1) courts don't have continuing jurisdiction under Article III to supervise prospective orders for so long as they remain in place; (2) even if they did, the existence of jurisdiction doesn't give courts the authority to address any question they might like to resolve; and (3) it's not at all clear that reviewing warrant applications even is an exercise of Article III power in the first place. I suspect we're already quite far into the weeds of this discussion, but for the sake of closing circles, I thought I'd offer brief reactions to each of these three points--and endeavor to do so below the fold. I.  Continuing Jurisdiction The crux of the dispute between Orin and me is whether an Article III case or controversy expires upon the issuance of a warrant or other surveillance authorization by the FISA Court, or whether the case or controversy remains live for the temporal duration of that authorization. This point is critical because, in Orin's view, by the time the FISA Court might certify a question to the FISA Court of Review under the Senate bill, the government's application has been granted, and so the case may well be over. (And, as I wrote in my response, I agree with the underlying premise--that a case or controversy must still exist in the certifying court at the time of certification.) The reason why I think I have the better of this argument is because of the language of FISA itself. Thus, FISA mandates that all applications have a durational limit--either one proposed by the government, or statutory default rules that otherwise govern. Then, provisions like 50 U.S.C. § 1805(d)(3) provide that:

At or before the end of the period of time for which electronic surveillance is approved by an order or an extension, the judge may assess compliance with the minimization procedures by reviewing the circumstances under which information concerning United States persons was acquired, retained, or disseminated.

In other words, FISA itself contemplates a continuing case or controversy by authorizing the issuing judge to sua sponte supervise the government's compliance with the underlying order. If Orin is right, then any judicial action taken pursuant to 50 U.S.C. § 1805(d)(3) would violate Article III. That doesn't prove Orin is wrong, of course. But it does underscore the deeper point, i.e., that the nature of the relief the FISA Court is providing to the government is more akin to prospective authorizations for future conduct than it is to retrospective relief for past conduct. And in more ordinary contexts, it is black-letter law that an injunction presents a justiciable controversy so long as it (1) remains in place; and (2) continues to produce adverse effects. Thus, as I wrote in my response, if there is ever an Article III case or controversy when the government applies to the FISA Court (and, again, that's a big if, as discussed in (III), below), I think it is clear that such a case or controversy persists for the statutorily circumscribed duration of the granted application. Just to be clear, if I'm right on this point, then most of Orin's objections to the Senate bill go by the board--because even Orin would concede, I suspect, that the FISA Court of Review could then at least answer certified questions of law upon the answers to which the validity of the underlying order necessarily rests. II.  Dicta vs. Jurisdiction-Less Pontificating This brings us to Orin's second point, which is that jurisdiction over a dispute doesn't give judges authority to say whatever the heck they want. Just to be clear, I don't disagree with Orin on this point in practice; the source of our disagreement is why this is so--whether the reason why judges can't say whatever they want in such contexts comes from Article III itself, or from common law doctrines that disfavor unnecessary judicial pontificating. Orin's point, if I understand it correctly, is that Article III courts may only answer those questions that are necessary to the underlying dispute--and that anything else is not just dicta, but is lawless. My contrary view is that, once Article III courts have jurisdiction over the underlying subject-matter, they have jurisdiction to say whatever the heck they want in the course of resolving the dispute; they just are precluded from doing so by any number of doctrines circumscribing appropriate judicial decisionmaking (e.g., constitutional avoidance; waiver; abstention; etc.), and if they nevertheless so pontificate, we are trained to usually dismiss such statements as "dicta." Consider qualified immunity appeals. In such cases, courts usually face two distinct questions: (1) Did the defendant's conduct violate the plaintiff's rights; and (2) were those rights "clearly established" at the time of the underlying conduct. The first question goes to the legality of the officer's actions; the second goes to whether he can be held liable even if the answer to (1) is yes (if the answer to (1) is no, the case stops there). As should hopefully be clear, in cases in which the answer to (2) is no (that is, where the officer cannot be held liable), the answer to (1) is almost always irrelevant, and so any judicial discussion thereof is dicta, at best. And yet, courts routinely answer the first question even when ruling for the officer on the second (indeed, from 2001 to 2009, lower courts were required to do so by the Supreme Court.) Under Orin's theory, unless the resolution of (1) is necessary to resolution of the underlying dispute, then courts lack jurisdiction to reach it. Under my view, courts have jurisdiction to reach (1); it's just a question of whether existing common law doctrines of judicial decisionmaking militate in favor of (or against) such resolution. (In this regard, qualified immunity doctrine is a rare instance in which these doctrines at least sometimes favor resolution of questions not directly necessary to the case.) And I think it's safe to say that the courts have gravitated toward my view of this subject, not Orin's. To be sure, Orin may find these doctrines an unconvincing (or insufficient) constraint on the power of federal judges to pontificate on irrelevant subjects, but, if anything, there are even stronger doctrinal constraints on courts using certification (the method prescribed by the Senate bill) to reach unnecessary questions. As then-Justice Hughes explained in 1914,

It is a familiar rule that this court cannot be required through a certificate . . . to pass upon questions of fact, or mixed questions of law and fact; or to accept a transfer of the whole case; or to answer questions of objectionable generality, which, instead of presenting distinct propositions of law, cover unstated matters ‘lurking in the record,’ or questions that are hypothetical and speculative.

So if, per Orin's hypothetical, the FISA Court certified three questions, one of which was utterly irrelevant to the underlying dispute, the FISA Court of Review would have jurisdiction to answer all three; they just would be bound by any number of judicial restraint doctrines to answer only those questions that go to the validity of the underlying surveillance authorization. Orin and I end up in the same place; we just have different reasons for why the court receiving the certificate won't indulge irrelevant or otherwise unnecessary questions. [As a nerdy footnote, the matter gets slightly more complicated--albeit not in a manner relevant to this discussion--when we get to certification to the Supreme Court, because of the indefeasible constitutional constraint on that Court's original jurisdiction.] III.  FISA and Article III, More Generally Orin's last point is, as I wrote in my response (and as Marty Lederman and I suggested back in November) the true elephant in the room: The possibility that FISA judges aren't even exercising Article III power in the first place when they approve surveillance applications. Before turning to the substance of this issue, let me make one (hopefully obvious) observation: Insofar as Orin is correct, then it would be more than a little odd to apply Article III's case-or-controversy requirement to actions that fall entirely outside Article III. That is to say, if surveillance authorizations fall outside Article III, then wouldn't it follow that subsequent oversight of them does, as well--and so there's no formal need for such review to comply with Article III? In this sense, Orin's third objection runs rather sharply into his first two... As for the substance, it's worth clarifying what is settled and what isn't. With regard to individualized warrant applications, I think there's far more authority than just the one district court case I cited (Megahey) for the proposition that federal judges are exercising Article III authority when they issue warrants--as OLC explained back when FISA was enacted in 1978. That conclusion is premised on a fiction, to be sure (that such warrants are ancillary to subsequent judicial proceedings). But plenty of federal courts doctrine rests on far less plausible fictions. The harder (open) question is whether such exercises of Article III authority in the context of authorizations for non-individualized surveillance (under, e.g., section 215 of the USA PATRIOT Act or section 702 of FISA as added by the FISA Amendments Act) satisfy Article III. And the ACLU, at least, has argued that they do not. As I've written before, though, the Article III question is exactly why I think Congress included provisions for adversarial review in both of these authorities--not to placate civil libertarians, but to ensure the existence of at least a nominal adversary in cases arising under these provisions. I'm not necessarily convinced that the hypothetical possibility of adversarial involvement ameliorates the Article III concerns, but, just to reiterate, this last point is something of a red herring. For if there's an Article III problem with how the FISA Court exercises authority, it exists wholly without regard to the provisions of the Senate bill.