« Conference Announcement: New England Regional Junior Faculty Scholarship Workshop | Main | Student-athlete speech »

Friday, January 08, 2016

Vehicle Problems vs. Unusual Vehicles: The Supreme Court's Bizarre Cert. Grant in Welch

This afternoon, the Supreme Court granted certiorari in Welch v. United States, a case that raises a question I've written about previously ad nauseam--whether the Court's June 2015 decision in Johnson v. United States may be retroactively enforced by federal prisoners in both original and second-or-successive petitions for post-conviction relief. The headline is straightforward; Welch means the Court will almost certainly resolve the 6-2-1 circuit split on that issue before AEDPA's one-year statute of limitations runs in June, which is, in my view, a Very Good Thing. And given that the Solicitor General agrees with Welch on the merits (it recommended a GVR at the cert. stage), the odds of the Court siding with the six circuits that have held that Johnson is substantive, and thus retroactively enforceable, seem high.

All that said, the fact that the Justices chose Welch as the vehicle through which to reach this question is bizarre in at least two respects:

First, Welch comes to the Court in a strange (and to my knowledge, unique) procedural posture: The Eleventh Circuit denied his application for a certificate of appealability _before_ the Supreme Court decided Johnson. That doesn't mean the Eleventh Circuit was right in denying his COA (it sure could've seen Johnson coming); it just means that a Supreme Court decision now holding that Johnson can be enforced retroactively going forward wouldn't actually reverse the Eleventh Circuit's holding in Welch; it would merely provide a new ground on which he may pursue a certificate of appealability. (The Eleventh Circuit denied a petition for rehearing after Johnson, but that's not usually enough to justify a cert. grant.)

Second, and related, Welch is presumably such an attractive vehicle for the Justices because it's a regular petition for certiorari after judgment, over which the Court only has jurisdiction because his is a first (and not second-or-successive) motion for post-conviction relief. For second-or-successive petitioners, relief doesn't depend just upon whether a new rule of constitutional law articulated by the Supreme Court (like Johnson) is substantive, but whether it has been "made retroactive" by the Supreme Court. In other words, it's at least theoretically possible that the Court could confirm that Johnson is substantive (which is all Welch needs in order to win), without taking the next step of holding that such a conclusion necessarily "makes" it retroactive, such that the decision wouldn't actually settle the circuit split (or allow second-or-successive petitioners in the Fifth, Tenth, or Eleventh Circuits to prevail). 

Neither of these problems are fatal; the strange posture has no bearing on the Court's jurisdiction, and I have to think that the Court will make it clear, if it holds that Johnson is substantive, that it can also therefore be enforced retroactively in second-or-successive petitions as well as in cases like Welch. But it does highlight why Welch is an especially strange vehicle through which to take up the Johnson retroactivity circuit split, in contrast to the petition for certiorari before judgment in Harrimon, and especially the petitions for original writs of habeas corpus in Sharp and Williams (both of which are second-or-successive cases). Indeed, perhaps for those reasons, in the government's various briefs in those cases arguing why the Court should wait for a more "appropriate" vehicle, Welch was never mentioned...

As for why this matters going forward, it suggests, if nothing else, that when confronted with a choice between cases with vehicle problems, and cases presenting unusual vehicles (like original habeas or cert. before judgment), the Court is far more inclined to rely upon the former. I leave it to others whether, in the long term, that's a healthy practice (since, among other things, it makes the vehicle problems seem completely discretionary). It certainly seems in this context to perpetuate the effective unavailability of extraordinary relief even in cases in which the Court's criteria appear to be satisfied--and to therefore raise the constitutional questions the Court ducked in Felker v. Turpin.

Posted by Steve Vladeck on January 8, 2016 at 04:22 PM in Steve Vladeck | Permalink


Are you certain that if the court answers Souter's serious constitutional question from Felker the answer will be one you like? It seems quite plausible that the current court would say that there is no constitutional problem with cutting off review of the Court of Appeals gatekeeping function or having a wholly illusory original habeas jurisdiction. The four most conservative members almost certainly think it's okay and the court as a while seems to have become substantially less friendly towards habeas petitioners in general.

Posted by: bill | Jan 8, 2016 4:45:10 PM

Bill -- I'm not certain (or even confident) by any stretch. But (1) if anything gets Kennedy's dander up, it's jurisdiction-stripping; and (2) in any event, if original habeas really _is_ illusory, and not just elusive, then I'd rather have the Court answer the question even if I don't like the answer. Otherwise, original habeas is a fiction, and Congress already _has_ effectively taken away the Court's appellate jurisdiction in these cases.

Posted by: Steve Vladeck | Jan 8, 2016 4:58:21 PM

What are the chances the Court will hold that Teague doesn't apply on 2255 motions (it only governs on 2254 petitions), such that Johnson is per se retroactive under 2255, thus leaving second and successive petitioners out in the cold? The Court ducked that issue in Chaidez.

Posted by: Da Man | Jan 8, 2016 5:47:58 PM

Da Man -- Seems like a long shot, since Johnson ought to be retroactive _here_ on either theory, no? So why resolve that issue here (especially given footnote 16 in Chaidez), as opposed to a case in which it would make a difference?

Posted by: Steve Vladeck | Jan 8, 2016 6:15:08 PM

Well I disagree that Johnson ought to retroactive under the first Teague exception. I've not found a single SCOTUS case applying retroactively a decision invalidating a part of a statute on vagueness grounds.

Posted by: Da Man | Jan 9, 2016 10:04:41 AM

It is not as if SCOTUS said Congress lacks the power to enact a sentence-enhancer for prior convictions.

Posted by: Da Man | Jan 9, 2016 10:11:00 AM

How could the Court "confirm that Johnson is substantive . . . without taking the next step of holding that such a conclusion necessarily 'makes' it retroactive"? Isn't a holding, by the Supreme Court, that a case is a new, substantive rule of constitutional law, all that's required to "make" it retroactive for second-or-successive petitioners? The fight in the Circuits which have found Johnson not applicable to second-or-successive petitions seems to me to be over whether it's truly substantive, in the sense contemplated by Teague. If the Supreme Court says it is, doesn't that settle the question?

Posted by: KK | Jan 9, 2016 1:57:21 PM

Of course it would settle it. The BIG question is whether Johnson is a new substantive rule. It clearly is a new rule. But is it of a piece with decisions holding, for example, that the First Amendment forbids prosecution of those who produce animal crush videos? I think not.

Posted by: Da man | Jan 9, 2016 5:41:08 PM

Da Man -- Of course, it doesn't have to be "of a piece" with decisions like Stevens. Summerlin held that "A rule is substantive rather than procedural if it alters the range of conduct or the class of persons that the law punishes. In contrast, rules that regulate only the manner of determining the defendant's culpability are procedural." As the Seventh Circuit held in Price, Johnson falls on the substantive side of this line because "A defendant who was sentenced under the residual clause necessarily bears a significant risk of facing 'a punishment that the law cannot impose upon him.'"

Yes, a vagueness challenge is different because Congress _can_ clarify the law going forward, but the Ex Post Facto Clause would prohibit it from applying that new definition to prior cases. So I don't think it's a remotely close question...

KK -- I certainly _think_ so, but that requires assuming that lower courts will follow the O'Connor concurrence from Tyler v. Cain, rather than Justice Thomas's more rigid majority opinion.

Posted by: Steve Vladeck | Jan 9, 2016 5:55:24 PM

A couple things:

First, I'm very glad the Court is hearing this in the first motion context. It appears to me, given dicta in Felker, and the text of the gatekeeper provision itself, that had the Court taken an original petition (which followed a prior first petition), the Court was going to itself be limited to deciding whether it had *previously* made Johnson retroactive. It would be very strange if the bar for second or successive original petitions was lower than ordinary second and successive petitions. And, while I agree with your last comment where you say that this is clearly substantive given Summerlin's general description of what's substantive and what's procedural, the courts below all seemed to want to fit Johnson into one of the specific sorts of rules that Summerlin and prior cases recognize as substantive - rules about primary conduct, rules narrowing a statute, rules that say that certain people can't receive a certain penalty on the basis of their status. If those pockets of substantive rules are all that's clearly settled, and the rest of Summerlin is just gloss, I'm afraid that Johnson hasn't been made retroactive, because it doesn't fit into any of those pockets. (It's definitely not a rule interpreting a statute, and it definitely isn't, contra to the Seventh Circuit in Price, a status-based rule about punishment, like a rule about juveniles and the death penalty is.) I think that would be the wrong way to read Summerlin, as what's doing the work in the holding in Summerlin is actually its general description of substantive rules, not the list of kinds of substantive rules. But some people seem to read it as a non-exclusive list without an organizing principle.

Second, to Da Man's points, the reason why you're wrong is that a rule is substantive if it changes what the substantive criminal law is, regardless of whether the *reasons* for the rule sound in procedural due process, which Johnson's reasoning does. The hallmark of a substantive rule is that it affects each person convicted or sentenced under a statute, regardless of the procedural bugs of their individual trials. If the remedy for the problems that Johnson identifies were more granular and stopped short of invalidating the residual clause, Johnson would be procedural. For example, Johnson at one point suggests that the residual clause was so vague that it invited judges to apply it arbitrarily; you could imagine some kind of heightened sufficiency review being the remedy for that, instead of invalidation.

Third, I'm completely unworried about whether a holding in Welch that Johnson is retroactive on first motions "makes" it retroactive for second-or-successive purposes. Even if you ignore O'Connor, a flat holding in Case Y that Case X is retroactive makes it retroactive, by the lights of the majority in Tyler. That's the gold standard, in their view, for retroactivity-making. (After all, what more could a lower court ask?) And even if the Court only said in Welch that Johnson is substantive - which would never happen, as the whole point of the case is to address retroactivity, not to engage in a theoretical exercise of categorizing the rule - that would still be enough for the majority in Tyler, which acknowledges in passing that a combination of cases - here, one case holding that substantive rules are retroactive, another holding that a rule is substantive - can make a case retroactive. What's a little laxer about O'Connor's approach is that she invites lower courts to decide for themselves whether a case so clearly fits one of the Teague exceptions that the Court has made that case retroactive, without ever categorizing it as substantive in some subsequent case.

Posted by: Asher Steinberg | Jan 9, 2016 11:43:19 PM

I have a more basic question, then, about Welch: how does the Supreme Court have jurisdiction over Question No. 1 when the statute governing certificates of appealability says that such a certificate shall issue only "the applicant has made a substantial showing of the denial of a constitutional right." 28 U.S.C. § 2253(c)(2). The question whether Welch's robbery conviction qualifies as a prior under the "elements" clause of ACCA seems to present a pure question of statutory construction. So how could Welch make in the Supreme Court the showing required by § 2253c)(2). Obviously, the court has jurisdiction to decide question 2 assuming that Welch's priors qualified under the residual clause, but Question 1 puzzles me, unless there's some sort of pendent appellate jurisdiction whereby the Court must decide whether the prior qualifies under some valid aspect of the statute before it will tackle the retroactivity question.

Posted by: Da Man | Jan 11, 2016 9:40:09 AM

Is there case law applying the COA requirement to SCOTUS? Section 2253(c)(1) seems to require a COA only for appeals to the Courts of Appeals: "Unless a circuit justice or judge issues a certificate of appealability, an appeal may not be taken *to the court of appeals* . . . ."

Posted by: KK | Jan 11, 2016 10:27:47 AM

KK -- I think that, in its current posture, the Welch cert grant means that the SCOTUS is deciding the question whether the court of appeals should have granted a COA. The SCOTUS has jurisdiction to decide its own jurisdiction, but to answer the questiona affirmatively, it will have to hold that the issues presented meet the standard set forth in § 2253(c)(2).

Posted by: Da Man | Jan 11, 2016 11:19:34 AM

Having done some further digging, it appears that Welch on direct appeal claimed that his robbery conviction was not an ACCA predicate. The 11th Circuit ducked the question whether the robery statute fit under the "elements" clause, concluding that it fit under the residual clause. In his 2255, Welch claimed that his robbery conviction could not qualify as a predicate under ACCA's residual clause based on Descamps (not a constitutional claim), but that argument is now irrelevant because Johnson invalidates the residual clause.

So, does the fact that the 11th Circuit never decided whether the Florida robbery statute fits the "elements" clause give the SCOTUS the jurisdiction to consider that question under the elements clause when the case is up on an application for a certificate of appealability and the main question is whether Johnson applies retroactively? I think Prof. Vladeck is onto something when he's complaining about the SCOTUS's choice of vehicles to resolve that question.

Posted by: Da Man | Jan 11, 2016 1:10:13 PM

Post a comment