24 hours later, Judge Leon’s 68-page opinion issuing (and then staying) a preliminary injunction against the NSA’s bulk telephony metadata program continues to generate a ton of headlines and increasingly sophisticated analyses. Not surprisingly, a number of the early reviews are critical of Judge Leon’s Fourth Amendment reasoning–especially his conclusion that Smith v. Maryland, the 1979 Supreme Court decision applying the “third-party doctrine” to phone records, can be distinguished in light of the myriad practical and technological differences between such collection today as compared to 34 years ago.
Suffice it to say, I’m a bit more sympathetic to Judge Leon’s reasoning than are, for example, Orin Kerr, Paul Rosenzweig, and Ben Wittes. But rather than push back against their arguments, the post that follows suggests that this may all be irrelevant–that a no-less-vulnerable element of Judge Leon’s decision is his holding that the plaintiffs’ APA claim (challenging the metadata program on statutory grounds) is precluded by section 215 itself. If, as I think it should, the D.C. Circuit reverses Judge Leon’s preclusion analysis, then that will bring the statutory question to the fore–for the Court of Appeals to either decide as a matter of first impression or send back to Judge Leon. And if, as many (including me) believe, the program can’t be reconciled with the statute, then we’ll end up in the same place (the program in its present form will be enjoined), without ever having to answer the thorny and far more far-reaching Fourth Amendment question concerning twenty-first century expectations of privacy in metadata.
I. The Preclusion Argument
At its core, the “preclusion” argument holds that Congress, in section 215 (50 U.S.C. § 1861), already provided a comprehensive mechanism for challenges to “production orders” by authorizing the recipients thereof (think Verizon) to object before the FISA Court, and to appeal an adverse decision to the FISA Court of Review–and, if need be, to the Supreme Court. Thus, the argument goes, such a remedy is “preclusive” insofar as it manifests Congress’s intent to displace relief that might otherwise have been available to other plaintiffs through other statutes. As Judge Leon explained in his opinion,
The text of Section 1861, and the structure and purpose of the FISA statutory scheme, as a whole, do indeed reflect Congress’s preclusive intent. Stated simply, Congress created a closed system of judicial review of the government’s domestic foreign intelligence-gathering, generally, and of Section 1861 production orders, specifically. This closed system includes no role for third parties, such as plaintiffs here, nor courts besides the FISC, such as this District Court. Congress’s preclusive intent is therefore sufficiently clear.
Such preclusion doesn’t foreclose consideration of the plaintiffs’ constitutional claims, Judge Leon continued, because of the various canons of statutory interpretation requiring a clear statement from Congress before cutting off judicial review of such arguments. But the upshot of this analysis is that it required Judge Leon to reach the Fourth Amendment challenge, even though the statutory objections to the metadata program are, by many accounts, far stronger.
II. Why the Preclusion Argument Fails to Persuade
The problem with the preclusion argument is not immediately apparent from the above-quoted passage, but is actually relatively easy to explain: At the core of Judge Leon’s analysis is the conclusion that “Congress did not contemplate the participation of third parties in the statutory scheme at all,” as indicated by, among other things, the focus in section 215 on the recipients of production orders, and not the subjects thereof. Leaving aside the long-settled presumption in favor of APA review (which, to be clear, requires more than just implicit inferences to overcome), this argument runs into two more immediate and fatal shortcomings–both of which demonstrate how it’s just not true.
Start with the cause of action Congress provided in FISA itself for “aggrieved persons.” Codified at 50 U.S.C. § 1810, this section provides an express private cause of action for “a person who is the target of an electronic surveillance or any other person whose communications or activities were subject to electronic surveillance,” “who has been subjected to an electronic surveillance or about whom information obtained by electronic surveillance of such person has been disclosed or used in violation of section 1809 of this title.” To be sure, the statute adopts a less-than-obvious definition of “electronic surveillance,” which may or may not extend to the collection of bulk telephony metadata. But that’s not the point; as section 1810 indicates, Congress certainly has contemplated the ability of individual plaintiffs to challenge aspects of FISA’s statutory scheme–at least in those rare and exceptional cases when those parties have standing, and can demonstrate that they have, in fact, been injured by FISA-authorized surveillance. (Of course, without Snowden, the secret nature of such surveillance will usually be an insurmountable barrier to standing.) Judge Leon not only does not seek to distinguish section 1810; he doesn’t even mention it.
In addition to the private cause of action created by section 1810, there is also the presumptive ability of criminal defendants to litigate via a motion to suppress the legality of the metadata program if and when evidence derived from it is introduced against them at trial. Judge Leon suggests (in footnote 26) that no such motion would be available in an appropriate case because FISA does not expressly provide one–but that’s not really the government’s position, nor would it make sense if it were. The whole point of the exclusionary rule is that such remedies are available to criminal defendants even in the absence of express statutory authorization. Kinda like APA review…
In other words, the argument that Congress has implicitly precluded review of section 215 orders by anyone other than their recipients runs headlong into the two circumstances in which it doesn’t seem to be true. Thus, it’s not really an “implied preclusion” argument; it’s an “implied preclusion except for those other two available remedies” argument. But it should follow that a regime that isn’t actually comprehensive can’t be deemed preclusive, at least absent some indication that Congress so intended. Given that, and given the Supreme Court’s repeated admonition that, as Justice Scalia explained last year, “if the express provision of judicial review in one section of a long and complicated statute were alone enough to overcome the APA’s presumption of reviewability for all final agency action, it would not be much of a presumption at all,” it becomes clear why Judge Leon’s preclusion analysis is so unavailing. (Folks interested in a more detailed version of this argument should check out pp. 18-26 of the ACLU’s brief in opposition to the government’s motion to dismiss in ACLU v. Clapper.)
Finally, even if one isn’t persuaded by the above analysis (even though the burden is supposed to be on the government to overcome the APA presumption, and not the other way around), there’s also the constitutional avoidance canon. If Judge Leon’s interpretation of section 215 as precluding the plaintiffs’ APA claim required him to reach and decide a constitutional question that a contrary statutory interpretation could have avoided, then it should have been avoided unless the text would admit of no other construction. Put another way, even if one thinks Congress may have implicitly precluded APA claims by these plaintiffs, there is no basis whatsoever for concluding that Congress unambiguously did so (nor does Judge Leon, who seems not to have heard of the constitutional avoidance canon, suggest otherwise). As such, Judge Leon should have interpreted section 215 to not preclude review, and to thereby potentially avoid the Fourth Amendment question he ended up answering.
Simply put, there’s insufficient evidence that Congress sought to overcome the presumption in favor of APA review; there are at least two contrary examples of how “third parties” might potentially challenge production orders; and there’s the constitutional avoidance canon lurking in the background to resolve any ambiguities in the statutory text. Given these shortcomings, it seems to me that the preclusion analysis of Judge Leon’s opinion is no less–and perhaps far more–vulnerable than the Fourth Amendment reasoning on which so many commentators have focused thus far.
III. Why the Weakness of Preclusion Matters
As for why the weakness of the preclusion analysis matters, that’s easy: As this case goes to the D.C. Circuit, the continuing uncertainty over the constitutional question (and the potentially massive implications of a Fourth Amendment holding in either direction) might well lead judges of any and all persuasion to look for other grounds on which to resolve the appeal. One possibility is reversing Judge Leon’s conclusion that these plaintiffs have standing, but for a host of reasons, that strikes me as unlikely. As Judge Leon quite pointedly explained, the comprehensiveness of the metadata program is the linchpin to the plaintiffs’ standing, even after Clapper.
A far more attractive possibility is to resolve a suit like this one on the basis of the APA claim, i.e., that the metadata program is inconsistent with its underlying statutory authority, and is therefore unlawful (albeit not necessarily unconstitutional). There’s a lot to say about the statutory challenge to section 215, but see especially Orin‘s and Marty‘s posts on the subject. And for judges who find such reasoning persuasive, invalidating the program on statutory grounds would sidestep the far messier (and more significant) Fourth Amendment question, limiting the holding to the specifics of the metadata program, which Congress could always reenact with modifications.
Indeed, constitutional avoidance would only further justify an interpretation of section 215 as not authorizing the metadata program in its current form; the contrary conclusion, i.e., that Congress fully authorized the metadata program in its present form, directly raises the Fourth Amendment concerns highlighted by Judge Leon. And the only obstacle to such a holding is Judge Leon’s preclusion analysis, which concludes that these plaintiffs can’t maintain such an APA claim. Reversal on that ground would open the door to the merits of the APA/section 215 claim, whether in the D.C. Circuit or, after vacatur and remand, before Judge Leon. And so, while folks continue to focus on the Fourth Amendment analysis and the future of Smith v. Maryland, it seems to me that, at least in the context of this litigation, no less attention (and perhaps far more) should be paid to the preclusion argument–and the far less controversial argument against the legality of the bulk records program.