Earlier this month, the Supreme Court issued its decision in Bowe v. United States, a case about the Antiterrorism and Effective Death Penalty Act (AEDPA). While observers who had hoped for a ruling in the tariff cases were left disappointed, the opinions in Bowe are noteworthy in their own right. In particular, the Justices’ discussion of clear-statement rules deserves some attention.

The status of the various clear-statement rules (and substantive canons more generally) has been a perennial subject of debate among academics, and several Justices have individually joined the conversation in recent years. But I cannot recall another case where the entire Court engaged with this question as explicitly as in Bowe. Briefly, the merits issue in Bowe was whether federal prisoners are covered by AEDPA’s “old-claim bar,” which requires dismissal of second or successive challenges to convictions that raise the same claim as a prior postconviction challenge. In an opinion by Justice Sotomayor, the majority held that only state prisoners are subject to that bar. In dissent, Justice Gorsuch took the contrary view. For present purposes, though, the most important issue was a threshold jurisdictional question the Court addressed: whether AEDPA stripped the Supreme Court of certiorari jurisdiction over Bowe and other cases like it.

That issue arose from the interaction between two sections of AEDPA. Under 28 U.S.C. § 2255, a prisoner in federal custody may in certain limited circumstances “move the court which imposed the sentence to vacate, set aside or correct the sentence.” § 2255(a). If a prisoner seeks to bring a second or successive motion, the motion “must be certified as provided in section 2244 by a panel of the appropriate court of appeals to contain” either “newly discovered evidence” or “a new rule of constitutional law.” § 2255(h). That language—specifically, “certified as provided in section 2244”—directs us to the second relevant section of AEDPA, 28 U.S.C. § 2244. That section, which sets forth the procedures for state prisoners who seek authorization to file a second or successive habeas petition in federal court, includes the following provision: “The grant or denial of an authorization by a court of appeals to file a second or successive application shall not be appealable and shall not be the subject of a petition for rehearing or for a writ of certiorari.” § 2244(b)(3)(E).

Thus, the question in Bowe was whether § 2255’s cross reference to § 2244 meant that federal prisoners could never get Supreme Court review of an appellate court’s decision not to certify a second or successive motion. Justice Sotomayor’s opinion for the Court (joined by Chief Justice Roberts, Justice Kavanaugh, Justice Kagan, and Justice Jackson) determined that AEDPA did not have this jurisdiction-limiting effect. In so holding, the majority relied heavily on the principle that “Congress must speak clearly if it seeks to impose exceptions” to the Supreme Court’s certiorari jurisdiction. The majority never said whether its reading of the statute was the best one; it was content simply to show that its reading was plausible, which sufficed because “[a]mbiguous language … cannot provide the required clear indication.” In dissent, Justice Gorsuch—joined by Justices Thomas, Alito, and Barrett^[1]—protested that “the majority’s bespoke certiorari-for-federal-prisoners-clear-statement rule lacks any grounding in law.” In the dissenters’ view, the best reading of AEDPA was that it did strip jurisdiction over this case, and there were no proper grounds to depart from that best reading.

For now, I’ll largely put aside the debate over the best reading of the relevant provisions of AEDPA. It is worth noting, however, a path that the majority decided to forgo. Justice Sotomayor presumably could have avoided reliance on the clear-statement rule by simply arguing that the most natural reading of the statute supported the Court’s exercise of jurisdiction. One could probably make a decent elephants-in-mouseholes sort of argument here, given that the allegedly jurisdiction-stripping provision in this case made no direct reference to jurisdiction, with the government’s argument instead relying on an inference from a cross-reference. Nevertheless, the majority decided to stake its opinion on a newly minted clear-statement rule instead. (Of course, this would hardly be the first time that the Court has invoked substantive canons when ordinary tools of construction at least arguably could have done the trick.)

Notably, the majority expressly declined to rely on constitutional avoidance to ground its clear-statement rule. Several notable scholars had argued in an amicus brief that “construing 28 U.S.C. § 2244(b)(3)(E) to apply to this case” would likely “exceed Congress’s Exceptions Clause power” by preventing the Supreme Court from ensuring uniformity in federal law, so “the canon of constitutional avoidance weighs especially strongly in favor of construing § 2244(b)(3)(E) not to apply to this case.” To be sure, it is not at all obvious that this argument is correct. For many decades after the Constitution was ratified, lower federal courts had the final word in a wide swath of criminal cases. As Akhil Amar has noted, “Until 1891, the Supreme Court had no general appellate jurisdiction over criminal cases decided by federal circuit courts—even when such cases posed the most important of constitutional issues.” Moreover, Justice Gorsuch’s dissent argued that there was no constitutional problem because “for much of our Nation’s history, there was no collateral review of the vast majority of convictions, much less a right to certiorari in such cases.” But in any event, the majority did not purport to reach any constitutional issues.

Instead, the majority sought to ground its clear-statement rule in several existing presumptions. Initially, and in my view most persuasively, the Court presented its rule as effectively just an application of the presumption against implied repeals. In the Court’s view—although it did not phrase the matter in quite these terms—the broad grant of certiorari jurisdiction in 28 U.S.C. § 1254(1) creates a kind of default rule. Congress is therefore presumed to adhere to that baseline and “must speak clearly if it seeks to impose exceptions to that jurisdiction.”^[2] There is considerable power to this argument, and it seems to have at least partly motivated Felker v. Turpin and Hertz Corp. v. Friend (both of which the Court here cited). But the dissent’s argument was that this was an express limitation, not an implied repeal.^[3] So, Justice Gorsuch contended, the majority needed something stronger than “the (unremarkable) principle” that “we have historically declined to infer limitations on our certiorari jurisdiction based on statutory silence.”

The majority also cited language from the 2003 case Castro v. United States expressing a desire to avoid “clos[ing] our doors to a class of habeas petitioners.” The majority did not elaborate extensively on why this concern justifies creation of a clear-statement rule. The majority did point to the longstanding principle that “we read limitations on our jurisdiction to review narrowly.” It is worth noting, though, that many of the cases applying that presumption are expressly based at least in part on constitutional avoidance. See, e.g., Webster v. Doe, 486 U.S. 592, 603 (1988) (“We require this heightened showing in part to avoid the ‘serious constitutional question’ that would arise if a federal statute were construed to deny any judicial forum for a colorable constitutional claim.”); Bowen v. Michigan Academy of Family Physicians, 476 U.S. 667, 681 n.12 (1986) (noting “serious constitutional question”); Johnson v. Robison, 415 U.S. 361, 366-67 (1974) (leading off analysis with constitutional avoidance); see also PDR Network, LLC v. Carlton & Harris Chiropractic, Inc., 588 U.S. 1, 19 (2019) (Kavanaugh, J., concurring in judgment) (suggesting there would be due process concerns if Congress “preclude[d] judicial review of agency interpretations of statutes in enforcement actions”).^[4]

So if the majority in Bowe really did mean to rely on that line of cases, it is possible that the majority’s analysis actually is constitutionally grounded after all. On this reading, Justice Sotomayor may have been engaged in a kind of “stealth constitutional avoidance,” indirectly relying on constitutional avoidance by citing to cases that in turn rely on constitutional avoidance. That explanation is not entirely satisfying, however, because it is not obvious that the kind of constitutional concern present in those cases is the same as the one some saw in Bowe. After all, as Justice Gorsuch observed, the presumption-of-reviewability cases were generally about whether “federal agency action is amenable to some form of judicial review,” which is a different question from whether a litigant who has “already enjoyed extensive judicial review” should get an additional chance to seek certiorari.

In any event, Justice Sotomayor also pointed to Boechler v. Commissioner, a unanimous 2022 decision in which the Court explained that “we treat a procedural requirement as jurisdictional only if Congress ‘clearly states’ that it is.” Sotomayor acknowledged that Boechler did not “control[] this case,” but asserted that it “demonstrates that this Court requires clarity in some contexts even absent constitutional concerns.” But reliance on that precedent is somewhat dicey. True, some version of this clear-statement requirement for distinguishing ordinary claims-processing rules from jurisdictional limits traces back to at least Arbaugh v. Y & H Corp. in 2006. But as recently as two years ago, there seemed to be uncertainty among the Justices about the nature of this canon and where it came from. At oral argument in Harrow v. Department of Defense, multiple Justices inquired about the basis or contours of the jurisdictional clear-statement rule, with Justice Gorsuch leading the way. As Eli Nachmany noted on this blog at the time, counsel for both parties described the rule as reflecting “fidelity to congressional intent” and being based on “assumptions about Congress.” Perhaps that justification had some purchase in Harrow itself; as petitioner’s counsel noted, “[i]n terms of the timing of the amending of [the provision at issue], the most recent amendment was 2012. And in 2012, the clear statement rule was already established…” But for provisions that have not been amended since 2006 (when Arbaugh was decided), the congressional intent argument ought to be a much tougher sell, since Congress could not even arguably have been aware of a rule that had yet to be invented. (Notably, the statutory provisions at issue in Bowe have not been substantively amended since AEDPA’s enactment in 1996, a full decade before Arbaugh.^[5])

Although Justice Gorsuch ended up joining a unanimous opinion in Harrow that matter-of-factly applied the clear-statement rule, one gets the sense from his dissent in Bowe that he (and the other Justices who joined his dissent) may still be uncomfortable with the principle’s foundations. Here is the key paragraph from the Bowe dissent:

Two years ago, in an obscure administrative law case, this Court expressly refused to treat Boechler as a permission slip for creating any manner of new interpretive presumptions that may happen to suit our tastes but that do not “approximat[e] rea[l]” statutory meaning. Compare Loper Bright Enterprises v. Raimondo, 603 U. S. 369, 455, n. 1 (2024) (KAGAN, J., dissenting) (arguing that, in light of cases like Boechler, we may as well accept other interpretive fictions like Chevron), with 603 U. S., at 399 (majority opinion) (holding otherwise). Yet that is exactly how the majority treats Boechler today.

This paragraph appears to suggest that the clear-statement rule applied in Boechler (and Harrow) is an “interpretive fiction[]” like Chevron deference. Given the dissenters’ attitude toward Chevron, that does not seem like a compliment. And the word “fiction” suggests a rejection of the approach to justifying the jurisdictional clear-statement rule promoted by both parties at the Harrow oral argument—that is, the argument that the rule reflects a presumption about congressional intent. Recall that, in Loper Bright, the Court rejected as “fictional” the notion that Congress intends ambiguity to delegate interpretive authority to agencies. Justice Kagan’s dissent responded that Boechler and other cases applying presumptions and clear-statement rules could similarly be called “fictions.” After citing a laundry list of such presumptions, Kagan stated that “[t]hey all are best guesses—and usually quite good guesses—by courts about congressional intent.” (Textualists reluctant to invoke “intent” can still make a similar kind of move by seeking to root canons in the context that informs an ordinary reader’s interpretation of a statute, as Justice Barrett has prominently tried to do with the major questions doctrine.)

But if, as Gorsuch suggests in Bowe, Boechler’s presumption (like Chevron’s) in fact fails to “‘approximat[e] rea[l]’ statutory meaning,” on what ground can it be justified? Certainly, no Justice in Bowe claimed it was grounded in any sort of constitutional value. Instead, both the majority and the dissent (the latter perhaps more half-heartedly) pointed to “the harsh consequences of finding a procedural requirement jurisdictional,” which can lead to “the waste of judicial resources and may unfairly prejudice litigants.” But that formulation, faithful though it is to what the precedents say, may raise more questions than it answers. Most strikingly, this sort of nakedly consequentialist justification seems in considerable tension with the Court’s other jurisprudential commitments. It is one thing to say that a substantive canon that does not reflect congressional intent or ordinary meaning may nevertheless be rooted in constitutional values (though even that has been controversial). But can such a canon be rooted in other, free-floating policy concerns that happen to strike the Court as important? If so, then it is hard to see what limits exist on the Court’s ability to create new substantive canons.

Although it is impossible to say for certain, it seems plausible that concerns like these drove some or all of the dissenters in Bowe (one of whom was Justice Barrett, who is well-known as having written at length about substantive canons when she was a law professor). Justice Gorsuch’s opinion, to be sure, did not repudiate all non-constitutional clear-statement rules, and indeed he seemed to accept the legitimacy of Boechler (while confining it to its “particular context”). But the opinion seems laden with a general skepticism of their pedigree, and a desire at least to put a stop to their growth.

I do not have anything especially insightful to say about clear-statement rules, or substantive canons more generally; many others have written extensively on those issues. But I hope this post has at least shown that Bowe represents a striking case study in how the Justices explain and defend the creation of new substantive canons, in an era when the propriety of many such canons has been increasingly called into question.

The fact that the clear-statement rule announced here was somewhat novel does not, of course, necessarily make it wrong. It is common in legal reasoning to support a proposition by reference to an array of somewhat analogous cases, no one of which is exactly on point. And indeed, as Will Baude noted back in 2022, that is precisely how many substantive canons have been created: “there are a lot of putative substantive canons that are not directly rules of positive law, but that are derived from other canons” by “extrapolat[ing] from one substantive principle to another.” Despite the concerns from multiple Justices that he noted at the time, it seems from Bowe that a majority of the Court has decided it is comfortable with that kind of extrapolation. Regardless of one’s opinion on the specific outcome the Court reached in Bowe, the case lays bare yet again the difficulty in articulating a consistent and theoretically sound approach to the creation of substantive canons.

Will Foster is a 2L at Yale Law School.

---

^[1] Justice Barrett only joined Part I of the dissent (the jurisdictional analysis), perhaps because she felt it was unnecessary to reach the merits given her view that the Court lacked jurisdiction. Importantly, though, Part I contained all of the discussion of clear-statement rules.

^[2] A related constitutional argument was suggested by Akhil Amar in a 2014 conversation with Richard Re. Under Amar’s argument, as summarized by Re, Article III creates “a constitutional default rule in favor of Supreme Court appellate review over the listed categories” of cases. Accordingly, when Congress seeks to use its Exceptions Clause power to “overcome the constitutional default in favor of appellate jurisdiction … a clear statement rule should apply.” Thus, though the majority in Bowe didn’t make this argument, it may be that the default rule is not merely the creature of statute but of the Constitution itself.

^[3] One might also question whether a narrow exception to a broad grant of power should really be considered a “repeal” at all. Cf. Savage Servs. Corp. v. United States, 25 F.4th 925, 946 (11th Cir. 2022) (“[W]hen a highly specific statute narrowly displaces a general one, it’s not at all clear that the presumption against implied repeal applies … [I]t’s not at all clear that the legislature is repealing anything.”).

^[4] The presumption in favor of judicial review is sometimes also justified based on the Administrative Procedure Act. See Department of Commerce v. New York, 588 U.S. 752, 771 (2019) (“The Administrative Procedure Act embodies a ‘basic presumption of judicial review’” (quoting Abbott Laboratories v. Gardner, 387 U.S. 136, 140 (1967)). That justification is presumably of no help here because Bowe did not involve an APA claim.

^[5] § 2244 has not been amended at all since 1996, and § 2255 underwent only one technical amendment in 2008 (to “designat[e] the 8 undesignated paragraphs as subsections (a) through (h)”).