Article III, Appellate Review, and the Leahy Bill: A Response to Orin Kerr

Orin’s post from yesterday afternoon wonders whether the Leahy bill’s provision for “certification” of decisions by the FISA Court to the FISA Court of Review (and from there to the Supreme Court) violates Article III’s case-or-controversy requirement. In effect, Orin’s charge is that, at the point of certification, the FISA Court is (or, at least, could be) effectively seeking an advisory opinion from the FISA Court of Review. To be clear, there's no definitive case law upholding against an Article III challenge the statutory certification procedures on which the Leahy bill is modeled. But, while I don’t have any objection to Orin’s proposed “fixes,” I explain in the following post why I don’t think there’s any constitutional problem in the first place—and why the California Medical Ass’n decision that Orin cites isn’t to the contrary. I.  Certification and Article III To start, it’s worth stressing how unusual statutory certification procedures are in general. Although a few specialized statutes (like the FECA provision at issue in California Medical) provide for certification, the three most common examples of certifications by federal courts are: (1) certifications of questions of state law to state courts of last resort (which is irrelevant for present purposes, since Article III doesn't bind state courts); (2) certifications of otherwise unappealable interlocutory questions under 28 U.S.C. § 1292(b); and certifications by the Courts of Appeals to the Supreme Court under 28 U.S.C. § 1254(2) (although the Court last accepted such a certificate in the 1981 Marschalk case). There is vanishingly little case law suggesting that certification under (2) or (3) raises distinct questions under the case-or-controversy requirement, and for good reason. In both sets of cases, there is necessarily a case-or-controversy in the certifying court at the time of certification (or else the dispute would no longer be justiciable even in the certifying court), and the existence of such a case-or-controversy presumably follows the certificate to the appellate court. Usually, the case-or-controversy requirement shows up in appellate review by requiring that the appellant have Article III standing to appeal--the very defect that proved fatal to the Prop. 8 proponents in the Supreme Court's 2013 decision in Hollingsworth v. Perry. But certificates aren't really "appeals" in at least two key respects: their substantive scope (which, at least initially, is usually confined to the specific certified order); and the triggering entity (the lower court judge--who certainly doesn't have "standing"--not the nominally adverse party who might). To that end, I’m unaware of any case law suggesting that the validity of a certificate turns on a party’s standing to appeal—nor should it. Consider in that regard the Supreme Court’s Rule 19, which not only provides wholly different procedures for cases in which a question has been certified by a Court of Appeals, but also stresses that “When a question is certified by a United States court of appeals, this Court, on its own motion or that of a party, may consider and decide the entire matter in controversy.” In other words, without any action by the "parties," the Justices can not only answer the certified question, but they can resolve the whole underlying dispute. (If it seems weird that the Justices could go further on an interlocutory appeal than an intermediate appellate court could in the same case, that's already the law under 28 U.S.C. § 1254(1), as this amicus brief explains.) Finally, Rule 19 further underscores the absence of an appealing party by providing that the opening brief is to be filed by “the party who is the appellant or petitioner below," even though there is no reason to believe that they desire or are entitled to pursue such appellate review. I don't mean to belabor the point; the key for present purposes is the conclusion that certificates don't of themselves raise Article III issues so long as the court that issued the certificate did so in the context of an ongoing case-or-controversy. That doesn't mean appellate courts are then free to pontificate on any matter they see fit; ordinary rules of judicial review would still apply. But whereas Orin's post suggests that the appellate courts would lack the authority to resolve questions not directly central to the order putatively being reviewed through certification, my own view is that such statements would simply be dicta insofar as they don't bear on the decision below--but that the appellate court's jurisdiction follows from the jurisdiction of the certifying court. II.  Cases or Controversies in the FISA Court Of course, that only begs the question of whether the FISA Court would be overseeing an ongoing case or controversy at the time it certifies a question for appellate review under the Leahy bill (and thus had jurisdiction to issue the certificate). In Orin's view, the Leahy bill "envisions that after the FISC has issued an order, when there is no application actually pending, the FISC can act sua sponte to certify 'any question of law' to the Court of Review. And after the Court of Review answers the certified question, it can then in turn certify 'a question of law' for the Supreme Court." The key is Orin's claim that "there is no application actually pending." Respectfully, I think that's the wrong point of emphasis. FISA Court orders are best viewed as prospective, not retrospective. That is to say, they authorize government action going forward (often for a specific period of time) that is subject to compliance with various procedural rules imposed (and administered) by the FISA Court. Thus, if the government's application suffices to create a case or controversy for Article III purposes (more on that in a moment), that case or controversy does not cease to exist once the application is granted; to the contrary, it exists for so long as the government is acting under the relevant application. In other words, even if no application is pending, there is a FISA Court order in place at the moment of certification. The harder question, in my view, is whether there is an Article III case or controversy in the first place when the government makes applications to the FISA Court. In our post from last November, Marty Lederman and I flagged the problem--and noted how "classic" FISA was initially predicated on the analogy to search warrants in criminal cases, which arguably don't violate Article III because they are ancillary to subsequent judicial proceedings. This, at least, was the view of the Office of Legal Counsel at the time FISA was enacted, and it's the ground on which FISA Court orders were upheld against Article III challenges prior to September 11. But applications under non-individualized authorities like sections 215 and 702 are quite another matter, since the warrant analogy completely breaks down in those contexts. And so it's worth stressing that it's not at all clear how there is ever an Article III case or controversy when the government files an application in the FISA Court under sections 215 and 702. But again, the key for present purposes is that, if the government's application suffices to create an Article III case or controversy, that case or controversy necessarily persists for the duration of the authorities that the FISA Court's granting of the application provides. That's why, for example, there is no Article III problem with having a "special advocate" participate in the FISA Court itself, even after the initial application has been approved. And if there is a case or controversy in the FISA Court, it should follow that the FISA Court has the power to certify relevant questions of law to the FISA Court of Review (and it, in turn, to the Supreme Court). III.  The California Medical Footnote The one contrary authority to which Orin cites in his post is footnote 14 of Justice Marshall's opinion in California Medical, which, in relevant part, provided as follows:

The unusual procedures embodied in this section are, at the very least, circumscribed by the constitutional limitations on the jurisdiction of the federal courts. A party seeking to invoke § 437h must have standing to raise the constitutional claim. Furthermore, § 437h cannot properly be used to compel federal courts to decide constitutional challenges in cases where the resolution of unsettled questions of statutory interpretation may remove the need for constitutional adjudication. Moreover, we do not construe § 437h to require certification of constitutional claims that are frivolous, or that involve purely hypothetical applications of the statute. Finally, as a practical matter, immediate adjudication of constitutional claims through a § 437h proceeding would be improper in cases where the resolution of such questions required a fully developed factual record.

Orin cites this passage for the proposition that "[f]ederal courts lack the power to decide cases unless the justiciability requirements of Article III have been met. Using a certification procedure doesn’t change that requirement." That's true, of course, but only so far as it goes--and it doesn't support Orin's conclusion. Under § 437h (the statute at issue), a party had to have standing "to invoke" that provision because a party had to have standing to bring a § 437h action in the first place, after which point the statute provided for certification of constitutional questions. Nothing suggests that a party must have separately had standing to appeal in order for the district court to sua sponte certify "all questions of constitutionality of this Act to the United States court of appeals for the circuit involved." And as Justice Marshall explained, courts receiving certificates will apply ordinary doctrines to avoid unnecessarily resolving constitutional questions, including those that are frivolous, hypothetical, or can be avoided through alternative statutory construction. There's no reason to believe the same wouldn't apply to the Leahy bill--the certification procedures of which are far narrower than those the Court endorsed in California Medical.

*                                                 *                                                 *

To be sure, the certification provisions in the Leahy bill are still a strange way to run a railroad. As should hopefully be clear from my prior posts, I'd far prefer a more concrete set of procedures that provide not only that the special advocate is allowed to participate as of right in at least some FISA cases, but that he is given a sufficient stake in the proceedings to allow him to directly appeal adverse decisions by the FISA Court. The Leahy bill, as should be clear to all, is a compromise--one designed to at once maximize the potential for meaningful judicial review of FISA Court decisions while minimizing the intrusiveness of outside parties in such review. I'm not convinced that this is the best way forward as a policy matter--and certainly agree with Orin that there are ways to make the bill less awkward even with respect to its current approach to certification. But there's a pretty big difference between an unusual certification procedure and an unconstitutional one. In my view, at least, this is definitely not the latter.