The Supreme Court will soon decide if Travis Beckles’s prison sentence is illegal. Mr. Beckles was sentenced years ago, and his appeal to the Supreme Court is on post-conviction review. Normally when the Supreme Court invalidates a prison sentence in a post-conviction case, the Court’s holding applies to all other post-conviction cases as well. But the way Mr. Beckles’s lawyers are arguing his case, relief for Mr. Beckles will do nothing for prisoners in certain circuits whose sentences would be illegal for the same reason as Mr. Beckles’s. And if the Supreme Court does not preemptively address these potential circuit splits in Beckles, then it may never have a chance to do so.
Mr. Beckles’s challenge to his sentence is based in part on the Supreme Court’s 2015 decision in Johnson v. United States, which held that the so-called “residual clause” of the Armed Career Criminal Act (ACCA) is unconstitutionally vague. ACCA’s residual clause subjected defendants to longer prison sentences if they had previous convictions for any crime that “involves conduct that presents a serious potential risk of physical injury to another.” Last April the Court made the rule invalidating ACCA’s residual clause retroactive in Welch v. United States. Johnson and Welch were blockbuster decisions that have tied up lower courts in a flurry of litigation that includes thousands of courts of appeals cases. The Court granted certiorari in Beckles to resolve two questions that have split lower courts in the wake of Johnson and Welch: whether an identically worded “residual clause” in the United States Sentencing Guidelines is unconstitutionally void for vagueness, and, if so, whether the rule invalidating the Guideline’s residual clause applies retroactively.
Those are the two questions that Mr. Beckles’s petition for certiorari directly raises. But there are other, equally significant questions that lurk beneath the surface in Beckles. Moreover, the circuits have already split on these other questions, or appear poised to do so. These questions will determine which prisoners would benefit from a favorable decision in Beckles, as well as which prisoners—including ones sentenced under ACCA—will benefit from the rule announced in Johnson. One of these questions is whether the statute of limitations has already expired to raise a challenge that the Guideline’s residual clause is unconstitutionally void for vagueness. Prisoners have one year from the date on which the Supreme Court recognizes a new right to file post-conviction motions asserting that right. Mr. Beckles’s attorneys are arguing that Mr. Beckles is asserting a right that the Court recognized in Johnson. For that reason, they argue that the statute of limitations to challenge Guideline sentences expired in June 2016. But some prisoners may not have challenged their Guideline sentences before that date, and others may need to refile challenges because their previous attempts were denied. If the Court rules as Mr. Beckles’s attorneys are urging, all those prisoners who have similarly illegal sentences may not benefit from a ruling in Mr. Beckles’s favor.
The other question that may prevent prisoners from benefiting from Johnson (or Beckles) is when courts of appeals should allow prisoners to challenge their ACCA sentences or their Guideline sentences based on those decisions. If a prisoner already filed one motion for post-conviction review in the past, the federal habeas statute requires the prisoner to get permission from a court of appeals panel in order to file what is called a “second or successive motion.” Nearly all the prisoners who wish to bring Johnson claims were sentenced years ago, so they already filed their first post-conviction motion. The courts of appeals have been applying divergent standards when deciding whether to authorize second or successive motions in these cases. At least one court of appeals—the Eleventh Circuit—has been denying authorization on the ground that a prisoner’s sentence might still be lawful based on other provisions the defendant was never sentenced under and based on other convictions that were never considered by the court imposing the sentence. One commentator has called the inconsistent treatment of Johnson claims in different circuits (and, in particular, the Eleventh Circuit’s outlier approach) “something very like a travesty of justice.” Yet because of the Antiterrorism and Effective Death Penalty Act’s (AEDPA) restrictions on the Supreme Court’s jurisdiction over second or successive post-conviction cases, the Court may not be able to take another case to address this problem.
The Court should both be aware of these lurking issues and use Beckles as the vehicle to weigh in on them. Doing so may be the only way to ensure that prisoners—particularly those in the Eleventh Circuit—will have a remedy for their unlawful sentences and the only way to ensure that any right announced in Beckles applies uniformly across the country. While the Court typically limits itself to analyzing questions that are directly raised in the petition for certiorari, AEDPA’s restrictions on the Court’s jurisdiction are more than a sufficient reason for the Court to depart from that practice here. Two decades ago, when the Supreme Court upheld AEDPA’s restrictions on post-conviction review, several Justices warned that circuit splits related to successive motions might reopen the constitutionality of AEDPA’s restrictions on the Supreme Court’s jurisdiction. As we show below, the aftermath of Johnson and Welch is precisely what those Justices warned about. Our goal is not to use the post-Johnson developments to reopen the question of AEDPA’s constitutionality. Instead, we aim to show that these developments make real the constitutional concerns that several Justices raised when they initially held that AEDPA was constitutional. And the constitutional concerns that have now materialized (including the troubling state of affairs in which the courts of appeals unreviewably treat identical post-conviction claims in wildly different ways) suffice as reasons why the Court should address questions not directly raised in the petition for certiorari in Beckles.
This short Essay proceeds in four parts. Part I provides some background on how the issues raised by Beckles have divided the courts of appeals. Part II explains how the courts of appeals could come to different conclusions on whether the statute of limitations has already expired for the exact claim that the Supreme Court might grant relief on in Beckles, and how the Court could write its opinion in Beckles to avoid that result. Part III explains how some courts of appeals are prematurely denying prisoners permission to file post-conviction motions raising Johnson claims based on speculation that those motions will fail on the merits. Many of those cases have decided questions of first impression about how and when the rule announced in Johnson invalidates a sentence. AEDPA prevents the Supreme Court from reviewing those cases. Part IV explains why the Court should use Beckles to clarify that courts of appeals should generally not be denying authorizations to file successive motions in this way. This guidance might be the only chance the Supreme Court ever gets to ensure that the lower courts implement Beckles—and Johnson—in a uniform way.
The questions on which the Court granted certiorari in Beckles turn on the differences between ACCA and the Sentencing Guidelines. Johnson held that ACCA’s residual clause is unconstitutionally vague, and Welch applied the rule announced in Johnson retroactively (in other words, to a case where the prisoner’s conviction already became final). Both ACCA and the Sentencing Guidelines subject defendants to higher sentences if they previously committed a crime that “involves conduct that presents a serious potential risk of physical injury to another.” Those thirteen words are called the “residual clause” in both ACCA and the Sentencing Guidelines (specifically, in a provision known as the “career offender guideline”), and the identical language in both provisions has always been interpreted the same way.
Beckles deals with how Johnson affects those identical words in the Sentencing Guidelines. The difference between ACCA and the Guidelines lies in the kinds of penalties they trigger. ACCA subjects defendants to mandatory minimum sentences. When a defendant is sentenced using ACCA’s residual clause, his minimum sentence is fifteen years (with a maximum of life). But without ACCA, the statutory maximum sentence for the same crime is ten years. The Guidelines do not change a defendant’s statutory minimum or maximum sentence, but they require a higher advisory sentencing range. That range has a significant impact on a defendant’s ultimate sentence. Though judges technically have discretion to impose a sentence outside the Guidelines range, the Supreme Court has explained that the “Sentencing Guidelines represent the Federal Government’s authoritative view of the appropriate sentences for specific crimes” and are the “lodestone” of federal sentencing. District courts “must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process.” The sentencing range also serves as the “benchmark” according to which a defendant’s sentence is judged on appeal. The Guidelines’ considerable “force as the framework for sentencing” means that frequently “the judge will use the Guidelines range as the starting point in the analysis and impose a sentence within the range.”
Most defendants are sentenced within the Guidelines range. Just last term, the Supreme Court recognized the “real and pervasive effect the Guidelines have on sentencing. . . . In less than 20% of cases since 2007 have district courts ‘imposed above- or below-Guidelines sentences absent a Government motion.’” The career offender guideline challenged in Beckles has an especially significant pull. Less than 0.57% percent of drug offenders who are sentenced without that Guideline receive sentences longer than the lowest end of the Guidelines range for defendants who were sentenced with that Guideline, even though the defendants were sentenced for the same crimes. Moreover, sentences increase severely as a result of this Guideline. The average sentence imposed on drug offenders classified as career offenders (138.6 months) was over twice as long as the average sentence imposed on drug offenders not classified as career offenders (sixty-two months).
Mr. Beckles’s case illustrates how much the residual clause in the Guidelines can increase a sentence. Without the residual clause, Mr. Beckles would have had a Guidelines sentence of fifteen years in prison. But because he had a single prior conviction for possession of a sawed-off shotgun (ironically, the same exact crime that was used to increase the defendant’s sentence in Johnson), Mr. Beckles was sentenced using the Guideline’s residual clause, and his Guidelines range jumped to thirty years to life in prison. The judge gave Mr. Beckles a sentence at the very bottom of that higher range. If the Supreme Court grants Mr. Beckles relief, his sentence could be halved. This case is not even the most dramatic example. Some prisoners’ current sentences are three or four times higher than what could be lawful after Beckles.
Since the Court decided Johnson and Welch, the courts of appeals have been split on two questions: whether the Guideline’s residual clause is invalid, and whether the Supreme Court has “made” the rule invalidating the Guideline retroactive. On the first question, only the Eleventh Circuit says the Guideline is not unconstitutionally void for vagueness. All eleven other courts of appeals have either held or assumed otherwise. The Eleventh Circuit’s position is especially striking because the United States has been conceding that the Guideline is invalid. It is not difficult to understand why the United States is doing so (and why the Eleventh Circuit’s lone view is likely wrong). If the residual clause in the Sentencing Guidelines is valid, courts must keep trying to interpret it, since a district court’s failure to correctly calculate a Guidelines range is procedural error that requires resentencing. But it makes little sense for courts to continue interpreting the residual clause, given that Johnson described the inquiry as nothing more than “guesswork” and concluded that “trying to derive meaning from the residual clause . . . [is] a failed enterprise.”
It also matters little that the Sentencing Commission deleted the Guideline’s residual clause in a recent amendment because the Commission did not apply that amendment retroactively. Therefore, defendants who were sentenced before that amendment became effective on August 1, 2016 would still be subject to the residual clause, and courts would be forced to determine whether these defendants’ convictions fall within the residual clause’s ambit. But as Johnson set out, that entire enterprise is a farce. And everywhere but in the Eleventh Circuit, that enterprise ended when Johnson was decided.
The more difficult question is whether the Supreme Court has “made” a rule invalidating the residual clause in the Guideline retroactive. Before a prisoner can file a successive § 2255 motion in a district court, AEDPA requires a court of appeals to certify that the motion satisfies certain preconditions, which here means that the motion “contain[s] . . . a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court.” That is, the prisoner needs to show not only that a new rule is retroactive but also that the Supreme Court has made that rule retroactive. Fifteen years ago, the Supreme Court adopted a frighteningly narrow definition of what it means to “make” a rule retroactive, coming close to suggesting that the Supreme Court makes a rule retroactive only where the Court itself applies that rule to a case on collateral review. The courts of appeals initially divided on whether the Supreme Court had made the rule in Johnson—that ACCA’s residual clause is unconstitutionally void for vagueness—retroactive. The Tenth, Fifth, and Eleventh Circuits said it had not; other circuits said it had, at least for purposes of authorizing successive motions. Welch resolved that split by definitively making Johnson retroactively applicable.
Something similar has happened with the rule that the Guideline is unconstitutionally vague. Again, all courts of appeals other than the Eleventh Circuit have held or assumed that Johnson makes the identical Sentencing Guideline language unconstitutionally vague. But of those courts, two have said that the Supreme Court has not “made” that rule retroactive; the rest have said it has. Beckles would resolve this split. If the Court holds the Guideline unconstitutionally void for vagueness, it would then decide whether that rule applies retroactively because Mr. Beckles’s case has already become final. And applying the rule that the Guideline is invalid to a case on collateral review would leave no doubt that the Supreme Court has made that rule retroactive.
Even if the Supreme Court rules that Mr. Beckles’s sentence must be reduced, there would be another potential hurdle to clear for other prisoners who received identical sentences to Mr. Beckles’s: the statute of limitations. The one-year statute of limitations applicable to § 2255 motions runs from “the date on which the right asserted was initially recognized by the Supreme Court.” In Dodd v. United States, the Supreme Court made clear that the statute of limitations runs from the date on which a right asserted was recognized by the Supreme Court, rather than the date on which that right was made retroactive. Mr. Beckles will not be barred by that one-year statute of limitations, since his § 2255 proceeding began long before that deadline. But the same will not be true for others who have sentences identical to Mr. Beckles’s. Depending on how the Court writes the opinion in Beckles, some courts of appeals may say that the time has already expired for other prisoners to challenge their sentences.
Mr. Beckles’s attorneys have argued in their brief to the Supreme Court that prisoners sentenced under the Guidelines are asserting a right that was recognized in Johnson. The implication of this argument—which the attorneys recognized when they urged the Supreme Court to take the case—is that the statute of limitations already expired for all other prisoners to argue that their sentences are unlawful for the same reason as Mr. Beckles’s sentence. Johnson was decided June 26, 2015. That means the deadline for claims based on Johnson expired on June 26, 2016, one day before the Supreme Court granted certiorari in Beckles.
If the Court takes the approach that Mr. Beckles’s attorneys are urging, Mr. Beckles may have his sentence declared illegal, but other prisoners would not because their claims would be foreclosed by the statute of limitations. The statute of limitations does not make a difference for Mr. Beckles, since his § 2255 proceeding began before June 26. But the statute of limitations would pose a bar to other prisoners whose sentences would be illegal for the same reason as Mr. Beckles’s sentence. The statute of limitations would bar prisoners who, like Mr. Beckles, filed an initial motion for post-conviction review and resentencing, if they did not file that motion before June 26. And even for prisoners who filed prior to June 26, their only hope would be for the Supreme Court to either grant, vacate, and remand their cases (assuming they filed petitions for certiorari) or for lower courts to revisit their earlier rulings.
The statute of limitations would pose an even more troubling problem for prisoners who are raising those challenges in successive motions for post-conviction review. Again, before a prisoner can file a successive § 2255 motion based on a new Supreme Court decision, the AEDPA requires that the motion be “certified . . . to contain . . . a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court.” Dodd held that the statute of limitations for successive motions, like the statute of limitations for initial motions, runs from the date on which the right asserted was recognized, rather than the date on which that right was made retroactive. Therefore, prisoners seeking to file a successive motion must show that the Supreme Court has recognized a new right and that the Supreme Court has made that right retroactive within one year. Dodd explained, “an applicant who files a second or successive motion seeking to take advantage of a new rule of constitutional law will be time barred except in the rare case in which this Court announces a new rule of constitutional law and makes it retroactive within one year.”
Two courts of appeals—in addition to the Eleventh Circuit—have held that the Supreme Court has not “made” a rule invalidating the Guideline retroactive. Prisoners therefore cannot challenge their Guideline sentences in these circuits. If the Court writes the opinion in Beckles the way Mr. Beckles’s attorneys are arguing, prisoners in those circuits may not be able to challenge their Guideline sentences after the Court retroactively applied a rule invalidating the Guideline. Before Beckles, the Supreme Court had not “made” a rule invalidating the Guideline retroactive, but after Beckles, it would be too late to challenge a sentence imposed under the Guideline. And unlike for initial motions for post-conviction review, AEDPA provides that the “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.” That means prisoners in those cases may be forever stuck with an incorrect ruling in their cases, even if the Supreme Court corrects this mistake in Mr. Beckles’s case.
Altogether, this could mean that Beckles will benefit the litigant whose case happened to win the certiorari lottery, while arbitrarily leaving others with equally meritorious claims shut out of court simply because they did not file § 2255 motions before June 26 (in other words, because they did not file a Beckles claim before the Court granted certiorari in Beckles or before the Court held the Guideline invalid). Prisoners could also be shut out merely because a court of appeals denied their earlier motion, even though the prisoner filed that motion before June 26.
The Court should prevent that result. There are many reasons why prisoners may not have filed before June 26, 2016 or may need to refile after a favorable decision in Beckles. The Court should therefore clarify that it is recognizing a “new right”—albeit one that represents the best reading of precedent—in Beckles that resets the statute of limitations. An opinion by Judge Martin on the Eleventh Circuit highlighted this possibility shortly before the Supreme Court granted certiorari in Beckles. She explained:
[T]he statute of limitations for § 2255 motions based on Johnson may expire in the next few days. Of course, if the Supreme Court overrules Matchett, that new case could start a new one-year clock. If that happens, the dates of the one-year statute of limitations will turn in part on whether Johnson’s voiding of the identical § 4B1.2(a)(2) language was “apparent to all reasonable jurists.”
Judge Martin therefore argued that a new Supreme Court ruling extending Johnson to the Sentencing Guidelines would announce a new rule and reset the statute of limitations. The Sixth Circuit later suggested the same. That court warned that “it is possible that the [Supreme] Court’s potential invalidation of the residual clause would come too late in our court” and “[p]risoners unaware of the possibility of challenging their Guidelines sentences until after the Supreme Court invalidated the residual clause would be out of luck, at least if the Supreme Court did not also make clear in Beckles that it was announcing a new constitutional rule, distinct from Johnson.”
Recognizing that a favorable decision in Beckles creates a new rule would be consistent with the Supreme Court’s retroactivity doctrine. When the Supreme Court held in Teague v. Lane that “new” constitutional rules of criminal procedure are generally not retroactive, it defined a “new” rule as one that “was not dictated by precedent existing at the time the defendant’s conviction became final.” Commentators have long said that Teague’s definition of “newness” is “far too expansive,” and subsequent cases have held that a rule is dictated by precedent and therefore not new if the rule would have been “apparent to all reasonable jurists.” The Supreme Court has further explained: “That the outcome in [a case] was susceptible to debate among reasonable minds is evidenced . . . by the differing positions taken by the judges of the Courts of Appeals . . . .” Here, the disagreement about whether Johnson applies to the Guidelines suggests that relief for Mr. Beckles would announce a “new rule.” Moreover, describing a rule that invalidates the Guideline as new does not mean that a rule invalidating the Guideline would be an unjustified extension of precedent—far from it. A rule can be new and still represent the best reading of precedent.
But if the Supreme Court decides Beckles in the way Mr. Beckles’s attorneys have urged the Supreme Court to rule (holding that no “new rule” is required to apply Johnson to the Guidelines), there is a risk that the decision will do prisoners no good unless they happened to file a claim before June 26, 2016 and the claim remains pending. Despite the public defenders’ best efforts, there are several reasons why prisoners may not have filed initial or successive § 2255 motions prior to June 26. One, precedent in some circuits squarely foreclosed these motions, so prisoners could not file before Welch was decided on April 18. Even after Welch, Eleventh Circuit precedent barred Johnson claims by prisoners who were sentenced using the residual clause in the Guidelines. Lawyers may not have been able to identify all of the prisoners with Johnson claims in the short period after Welch, and some prisoners may have chosen not to try and file a motion that they knew was barred by circuit precedent. The Eleventh Circuit has even continued to deny authorizations after certiorari was granted in Beckles and has rejected requests to hold cases in abeyance. All those prisoners would need to refile requests for authorization after Beckles. But they might not be able to do that unless Beckles announces a “new rule.”
Two, in the wake of Welch, the Eleventh Circuit ruled that requests for permission to file § 2255 motions are governed by § 2244(b)(1)’s requirement that courts must dismiss any claim that was presented in a prior motion. This interpretation of §§ 2255 and 2244 (which is a minority view) means that prisoners cannot just refile claims that were rejected prior to Beckles. Indeed, the Eleventh Circuit has specified that it will not have jurisdiction over certification motions raising previously presented claims “unless and until the Supreme Court establishes in Beckles or some other future decision ‘a new rule of constitutional law.’” Put another way, the Eleventh Circuit has already said that Beckles will do nothing for prisoners who already filed a § 2255 motion if the case does not announce a “new rule of constitutional law,” even if the Court announces that Mr. Beckles’s sentence is invalid. Mr. Beckles will benefit from that rule, but others might not, even if their sentences became final at the same time.
Three, basic facts about prison litigation mean that some prisoners may not have filed before June 26. A prisoner may be without a lawyer, for example. It is also difficult to identify prisoners who were sentenced in the same manner as Mr. Beckles: the judgment in a case does not indicate which Guidelines a prisoner was sentenced under. Even when this information is recorded elsewhere, it can be hard to uncover. One of the few documents that might list this information is the prisoner’s presentence investigation report (PSR), which the Bureau of Prisons bars prisoners from possessing for security reasons. PSRs are also typically sealed, which means that lawyers other than the counsel of record at sentencing will not have access to them.
In part for these reasons, prisoners in all but one circuit (the First, Third, Fourth, Fifth, Sixth, Seventh, Eighth, Ninth, Tenth, Eleventh, and D.C. Circuits) have requested authorization to raise Johnson-related claims in successive § 2255 motions after June 26, 2016. That is, these prisoners sought permission from the courts of appeals to challenge their sentences after the statute of limitations would have expired if the Court holds that a rule invalidating the Guideline is not a new rule (as Mr. Beckles’s attorneys are arguing). Unless the Supreme Court holds that Beckles creates a “new rule,” all those motions may be denied as untimely.
For all those reasons, the Supreme Court should clarify that a decision invalidating the Guideline’s residual clause is a “new rule.” Doing so avoids the possibility that prisoners would be time-barred from challenging their Guideline sentences before the Supreme Court actually held the Guideline invalid. It also addresses an issue that likely precipitated the certiorari grant in Beckles. If the only circuit split the Court was concerned about was the split on whether the Guideline remains invalid, why not grant certiorari in a case on direct review? A case like that would not raise any questions about retroactivity, since relief on direct review does not depend on whether a rule is “new” or not. Therefore, the Supreme Court likely granted certiorari in Beckles in order to address the circuit split that had emerged over whether the Court had “made” a rule invalidating the Guideline retroactive.
Ordinarily, the Court can address a split on whether it has made a new rule retroactive by retroactively applying that rule to a case that has already become final. Indeed, that is precisely what happened in Welch: the Court retroactively applied the rule announced in Johnson, which allowed prisoners to show beyond any doubt that the Supreme Court had made the rule announced in Johnson retroactive. But the same may not be possible in Beckles if the Court holds that a rule invalidating the Guideline is not a new rule (as Mr. Beckles’s attorneys are arguing): the Court could retroactively apply a rule invalidating the Guideline, but doing so would not allow prisoners to file successive motions based on that rule because the statute of limitations to do so would have already expired. In the Sixth Circuit’s words, “[h]ow strange.” Why permit “successive motions that are barred under the statute of limitations in § 2255(f)(3)”? “It helps no one” to “authorize the filing of successive motions that are routinely barred by the statute of limitations.” The prisoners who could potentially benefit from Beckles include prisoners in the Fifth, Eighth, and Tenth Circuits, which have held that the Supreme Court has not made a rule invalidating the Guideline retroactive. But if the Court holds that a rule invalidating the Guideline is not a new rule (as Mr. Beckles’s attorneys are arguing), Beckles would not benefit those prisoners. For this reason, it makes a good deal of sense for the Court to clarify that a decision invalidating the Guideline resets the statute of limitations to challenge the Guideline.
Of course, the Supreme Court could wait to see whether courts dismiss Beckles claims on statute of limitations grounds and try to pick up another case to address this issue. That is, the Supreme Court could wait to see whether courts of appeals say that the statute of limitations has already expired on Beckles claims and, if they do, review those determinations by way of yet another petition for certiorari. The same possibility does not exist, however, for the second issue lurking beneath Beckles: whether courts of appeals should grant authorization to file a successive motion where it is not clear if a prisoner’s sentence depended on the residual clause. Beckles may be the Supreme Court’s only opportunity to speak on this question, which would affect both Johnson claims and Beckles claims.
How the Court defines the “right” in Beckles also affects whether courts will even allow successive motions to be filed based on that case. The Court should use Beckles—which may present the only such opportunity for the Court to speak on this issue—to clarify how courts of appeals should determine whether to “authorize” successive § 2255 motions in cases where a prisoner claims his or her sentence depends on the residual clause. In this Part, we list some of the problems that arose in lower courts in the aftermath of Johnson and Welch and explain how the Court could prevent these problems from repeating themselves after Beckles. We then argue that the Supreme Court should explain when and why Beckles makes a sentence illegal to ensure that lower courts analyze Beckles claims in a uniform way.
Again, before a prisoner can file a successive § 2255 motion based on a new Supreme Court decision, AEDPA requires that the motion be “certified as provided in section 2244 by a panel of the appropriate court of appeals to contain . . . a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court.” The “section 2244” referred to there is another part of AEDPA, which says that a court of appeals “may authorize the filing of a second or successive application only if it determines that the application makes a prima facie showing that the application satisfies the requirements of this subsection.” Section 2244 also provides that the Supreme Court cannot grant petitions for certiorari to review the “grant or denial of an authorization by a court of appeals to file a second or successive application.”
The lack of review has meant that the lower courts face little accountability in their decisions to deny permission to file second or successive § 2255 motions. After Johnson, the Eleventh Circuit frequently denied authorization to file successive § 2255 motions on the ground that a prisoner will not benefit from the “new rule” recognized in Johnson. This happened in two ways. First, the court ruled that Johnson categorically does not apply to the provision under which the prisoner was sentenced (for example, the residual clause of the Sentencing Guidelines or similar but slightly different language in another penal statute). Second, the court ruled that the prisoner’s criminal history includes prior convictions that could be used to increase the prisoner’s sentence in the same way through a provision that survived Johnson.
Other courts that have denied authorizations have done so primarily on the first ground. The Supreme Court has some ability to weigh in on this kind of reasoning via cases that raise the same issue but were brought by a prisoner who never filed a § 2255 motion in the past and thus did not need to get permission to file a successive motion. This is how the Supreme Court came to hear Welch and also how it will hear Beckles.
But the same is not true for denials of permission to file successive § 2255 motions that are premised on the second kind of reasoning—that a prisoner’s § 2255 motion (which has not even been filed yet) will fail on the merits because the prisoner’s record shows that the prisoner’s prior convictions qualify as criminal history predicates under a provision that remains valid. This kind of reasoning would not be reviewable if the Supreme Court granted certiorari in a case where a § 2255 motion was adjudicated on the merits, because the Court would have no occasion in a case like that to review what “prima facie showing” AEDPA requires or what kinds of evidence may be used to establish a prima facie showing.
The absence of any accountability that might result from the possibility of Supreme Court review in these cases has created a severe problem. The Eleventh Circuit ruled on nearly two thousand requests to certify second or successive § 2255 motions based on Johnson in the three months after the decision in Welch. Those rulings show that the court is both internally divided and likely wrong on at least two questions that come into play only at the authorization stage, and only when a court of appeals denies authorization on the ground that a prisoner’s prior convictions still qualify as criminal history predicates (in other words, only in cases that the Supreme Court cannot review): (1) what is required to make a prima facie showing, and (2) what law applies when assessing whether a prisoner has made a prima facie showing. As we discuss below, the Eleventh Circuit’s outlier view on these two questions has meant prisoners with nearly identical sentences (as in, ACCA sentences that were based on the same state crimes) have fared differently depending on what certification panel they drew. Most of those rulings were made without input from a lawyer, nearly all of them are never reported in a commercial reporter or on the court’s website, and none are subject to further review.
The Eleventh Circuit appears to be internally divided on what amounts to a prima facie showing that a prisoner “falls within the scope of the substantive rule announced” by this Court. For example, some Eleventh Circuit decisions maintain that a prisoner makes a prima facie showing if no “binding precedent” indicates that the prisoner’s prior convictions support an ACCA sentence despite Johnson. Other panels disagree and will determine as a matter of first impression whether a prisoner’s prior convictions can still support an ACCA sentence. Other panels have said prisoners must “make a prima facie showing that they previously were sentenced, at least in part, in reliance on the ACCA’s now-voided residual clause.” Other panels have instead framed a prima facie showing as one where “the record does not refute” the prisoner’s assertion “that the sentencing court relied on the residual clause.” And these are just the divisions that appear in decisions that were published on Westlaw. The court has also issued thousands of rulings that are not available in any commercial reporter.
Also troubling is the Eleventh Circuit’s method of considering whether a prisoner’s prior convictions can support his sentence under one of the definitions of “violent felony” that remain valid. Along with the residual clause definition at issue in Johnson, Welch, and Beckles, both ACCA and the Guidelines also define “violent felony” (or “crime of violence” in the Sentencing Guidelines) as any crime that “has as an element the use, attempted use, or threatened use of physical force against the person of another,” as well as any crime that “is burglary, arson, or extortion, involves use of explosives, or otherwise involves conducts that presents a serious potential risk of physical injury to another.” And Johnson was careful to say that the decision “does not call into question . . . the four enumerated offenses, or the remainder of the . . . definition of a violent felony.” Therefore, the decision in Johnson—and a future decision in Beckles—does not affect sentences that are valid due to the element-of-force or enumerated-offense clauses.
The Eleventh Circuit has held in hundreds of cases that prisoners cannot file § 2255 motions if the convictions listed on their presentence investigation report support a new ACCA sentence even without the residual clause, sometimes even when no judge ever sentenced the prisoner based on those convictions. The Eleventh Circuit has even done the same with prisoners sentenced under the Guidelines. That is, when prisoners sentenced under the Guideline’s residual clause filed challenges based on Johnson, the Eleventh Circuit sometimes gave an alternative ground for denying them permission to file a § 2255 motion. Instead of denying permission to file a § 2255 motion because the Guideline is not unconstitutionally vague (which is the result required by current Eleventh Circuit precedent), the Eleventh Circuit has denied successive § 2255 motions on the ground that even if the Supreme Court later holds that the Guideline is unconstitutional, a prisoner’s other convictions will support a higher sentence anyway so their claim would have to be denied in the future. In doing so, the Eleventh Circuit is preemptively ensuring that these prisoners would not benefit from a favorable decision in Beckles. And the Eleventh Circuit makes this hypothetical determination about how prisoners’ Beckles claims would fare based only on a form filled out by a prisoner and sealed records called up by the court, all without argument or briefing. The form that prisoners fill out does not allow them to attach any materials, including proposed motions, and the Eleventh Circuit issues denials within thirty days of receiving requests for authorization.
The Second and Sixth Circuits have already adopted a different approach to the Eleventh Circuit’s treatment of potential Beckles claims, one which minimizes the risk that prisoners’ claims are judged prematurely (and potentially incorrectly). This approach also minimizes the risk that a prisoner’s claim will get lost in an unnecessary cycle of filing and refiling. When asked to authorize a successive motion by a prisoner sentenced under the residual clause in the Guidelines, the Sixth Circuit has been transferring cases to district courts to hold in abeyance pending a decision in Beckles. The Sixth Circuit explained:
Not only is the standard for assessing [a successive] motion light, but the setting for reviewing it counsels against making more law than necessary. A denial of a motion to authorize a successive petition is unreviewable—not by the en banc court, not by the Supreme Court. By granting such a motion, even many such motions (roughly 1700 Johnson motions have been filed in our circuit), we decide nothing with finality. The habeas statute permits the district court to determine for itself whether the petitioner has met the gatekeeping requirements of § 2255(h). Congress has also asked us to make these decisions quickly, ideally within 30 days of a motion’s filing and often with little if any briefing. All features of this setting considered, it makes sense to leave the district court free to decide [the issue].
The Sixth Circuit further reasoned that “[t]he Supreme Court’s recent decision to review Beckles also supports this approach” because a decision in that case will provide “answers to the pertinent questions.”
The Second Circuit has taken this same approach of holding these cases in abeyance until there is a ruling in Beckles. The Eleventh Circuit, by contrast, has specifically declined requests to hold “application[s] in abeyance due to the grant of certiorari in Beckles.” The Eleventh Circuit was the only court that took that same approach after the Supreme Court granted certiorari in Welch.
Another issue that lurks beneath Beckles is whether Supreme Court decisions like Descamps v. United States and Mathis v. United States play any role in deciding whether prisoners’ prior convictions support their ACCA sentence despite Johnson. Descamps and Mathis set out the proper interpretation of ACCA’s element-of-force and enumerated-crime clauses (and accordingly the Guideline’s identical versions of those clauses). Both cases abrogated many prior court of appeals cases interpreting those other clauses, but the Eleventh Circuit has said Descamps and Mathis can be ignored when deciding whether a prisoner’s prior convictions qualify as criminal history predicates. The Eleventh Circuit has even refused to apply Descamps for Johnson claims filed by prisoners who were sentenced after the decision in Descamps.
The Eleventh Circuit’s reason for refusing to apply the Supreme Court’s interpretation of the element-of-force and enumerated-offense clauses is that the Supreme Court has not “made” decisions like Descamps and Mathis retroactive. But that is not a basis for refusing to apply binding precedent when deciding a Johnson claim. Where a prisoner challenges an ACCA sentence, the prisoner’s new claim is based on Johnson (and where a prisoner challenges a Guideline sentence, the prisoner’s new claim will be based on Beckles). When a court decides a Johnson claim, Descamps and Mathis merely indicate whether a prisoner’s prior convictions serve as predicates under one of the clauses that survived Johnson. Put another way, Descamps and Mathis inform whether any Johnson violation would be harmless because a prisoner’s sentence remains valid despite Johnson. But this does not mean Descamps or Mathis provide the new rule that such a prisoner is seeking relief under. If a prisoner’s sentence was valid up until the moment Johnson (and potentially Beckles) was decided, then Johnson (or Beckles) is the new rule that allows courts of appeals to authorize successive motions.
There are other reasons why courts cannot ignore Mathis and Descamps when deciding whether a claim meets § 2255(h)’s requirements, as Johnson claims do (and as Beckles claims would). First, Mathis and Descamps explain how ACCA’s language (and the Guideline’s language) should be applied, as a matter of statutory interpretation. And the Supreme Court’s “judicial construction of a statute is an authoritative statement of what the statute meant before as well as after the decision.” “[O]nce the Court has spoken, it is the duty of other courts to respect that understanding of the governing rule of law.” Because decisions of statutory interpretation reflect what a statute meant when it was enacted—and, accordingly, when a prisoner was sentenced under it—Mathis and Descamps reflect both the sentence a prisoner could receive when he was initially sentenced and also the sentence a prisoner could receive if the prisoner were resentenced today.
Second, nothing in § 2255 requires courts to apply incorrect interpretations of statutes or Guidelines just because they would have done so at the time of sentencing. AEDPA’s special requirements for successive motions simply say that these motions need to “contain . . . a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court.” After Welch, there is no question that Johnson meets that definition. And so a motion filed by a prisoner whose sentence was valid up until the day Johnson was decided “contain[s]” a Johnson claim. The rest of § 2255 imposes no further restrictions on what cases courts can use to analyze the merits of successive § 2255 motions. To the contrary, § 2255(a) provides for relief where “the sentence was imposed in violation of the Constitution or laws of the United States.” If a prisoner’s prior crimes no longer fall under ACCA’s language after Johnson (or the career offender guideline’s after Beckles), the prisoner is (in the language of § 2255(a)) “in custody under sentence of a court established by Act of Congress claiming the right to be released upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States.”
Third, Mathis and Descamps apply to successive motions under the retroactivity doctrine established by Teague v. Lane. Teague says “new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” Mathis and Descamps are not “constitutional rules of criminal procedure”; they are decisions of statutory interpretation, and the Supreme Court has made clear that “decisions that narrow the scope of a criminal statute by interpreting its terms” are “not subject to the [Teague] bar.” On top of that, even if Descamps and Mathis were subject to the Teague bar, Mathis and Descamps appear to be old rules, rather than new ones. The Supreme Court has made clear that “[u]nder the Teague framework, an old rule applies both on direct and collateral review.” Accordingly, not applying Mathis and Descamps when deciding § 2255 claims violates “the Teague framework.”
AEDPA’s removal of Supreme Court review of denials of permission to file successive § 2255 cases creates a power that is unlike anything else in federal law. Courts of appeals are almost never allowed to act with no possibility of further review. The Eleventh Circuit’s response to Johnson and Welch shows how dangerous this power can be. The Eleventh Circuit’s outlier approach also raises serious constitutional issues. In Felker v. Turpin, the Supreme Court held that AEDPA’s restrictions on review over these cases did not violate the Suspension Clause. Writing separately, Justice Souter, joined by Justices Stevens and Breyer, noted: “[I]f it should later turn out that statutory avenues other than certiorari for reviewing a gatekeeping determination were closed, the question whether the statute exceeded Congress’s Exceptions Clause power would be open. The question could arise if the courts of appeals adopted divergent interpretations of the gatekeeper standard.” Felker was decided less than a month after AEDPA was enacted. Twenty years later, Justice Souter’s concerns have come to pass, along with others he did not anticipate, such as division within a circuit on the proper gatekeeping standard.
Although AEDPA prohibits the Supreme Court from reviewing denials of authorization by way of petitions for certiorari, the Court could use the claim raised by Mr. Beckles to address how courts like the Eleventh Circuit are denying authorization. The Court could, for example, explain the contours of the right announced in Beckles, including whether decisions like Mathis and Descamps inform any determination that a prisoner’s prior convictions qualify as criminal history predicates. The Court could also provide some rare guidance about what is required to make a prima facie showing to obtain authorization to file a successive motion, as well as how courts of appeals should make that determination.
AEDPA and the Eleventh Circuit have combined forces to make it almost impossible for prisoners to get judicial review of potentially unconstitutional sentences. The same prisoners who the Eleventh Circuit has kept out of court would have fared differently in other circuits. When the Supreme Court set out its modern retroactivity doctrine in Teague, it declared that “the harm caused by the failure to treat similarly situated defendants alike cannot be exaggerated: such inequitable treatment ‘hardly comports with the ideal of “administration of justice with an even hand.”’” The aftermath of Johnson and Welch shows that lower courts cannot always be trusted to “treat similarly situated defendants alike” when deciding which prisoners can file successive § 2255 motions based on a new landmark decision. AEDPA insulates those rulings from the review and accountability that exist for nearly everything else courts of appeals do. If the Supreme Court rules in Mr. Beckles’s favor, it should be mindful of those unique restrictions and write its opinion in a way that prevents a repeat of the mess that unfolded after Johnson and Welch.
* Leah Litman is Assistant Professor of Law at the University of California, Irvine School of Law. Shakeer Rahman is a law clerk to Justice Mariano-Florentino Cuéllar on the Supreme Court of California. The authors thank Erwin Chemerinsky, Song Richardson, Eve Brensike Primus, and Kevin Golembiewski for helpful comments, and Ben Levin for making this Essay happen. The views expressed in this Essay are the authors’ alone.
The Johnson opinion was issued on the last day of the Supreme Court’s 2014 term and has been cited in over ten times as many lower court decisions as any other case from that term, according to Westlaw. Welch was decided on April 18, 2016 (an extraordinarily quick nineteen days after the Supreme Court heard oral argument) and has already been cited 824 times by lower courts. And those are just the cases that are reported on Westlaw. Most rulings about whether prisoners can benefit from Johnson and Welch are made in orders that are not on Westlaw. The Eleventh Circuit (which covers Alabama, Florida, and Georgia) has said it ruled on “close to two thousand” of these cases as of August 2, 2016. In re Chance, 831 F.3d 1335, 1342 (11th Cir. 2016) [https://perma.cc/J7H9-LTGL]. The Sixth Circuit (which covers Kentucky, Michigan, Ohio, and Tennessee) said that “roughly 1700 Johnson motions have been filed in our circuit” as of July 29, 2016. In re Embry, 831 F.3d 377, 382 (6th Cir. 2016) [https://perma.cc/L3XE-BB9W].
Noah Feldman, This Is What ‘Travesty of Justice’ Looks Like, Bloomberg View (July 22, 2016, 2:39 PM), http://www.bloomberg.com/view/articles/2016-07-22/appeals-court-fumbles-supreme-court-ruling [https://perma.cc/WE3V-JLP7]. As we show below, the same problem is already unfolding with Beckles claims.
Felker v. Turpin, 518 U.S. 651, 667 (1996) (Souter, J., concurring) (“[I]f it should later turn out that statutory avenues other than certiorari for reviewing a gatekeeping determination were closed, the question whether the statute exceeded Congress’s Exceptions Clause power would be open. The question could arise if the courts of appeals adopted divergent interpretations of the gatekeeper standard.” (footnote omitted)).
18 U.S.C. § 924(e)(2)(B)(ii) (2012); U.S. Sentencing Guidelines Manual § 4B1.2(a)(2) (U.S. Sentencing Comm’n 2015), amended by U.S. Sentencing Guidelines Supplement to the 2015 Manual § 4B1.2(a)(2) (U.S. Sentencing Comm’n 2016). Although this language was removed in the 2016 reiteration of the Sentencing Guidelines, the change benefits no one sentenced prior to August 2016. See infra text accompanying notes 36–37.
See Leah M. Litman, Residual Impact: Resentencing Implications of Johnson’s Potential Ruling on ACCA’s Constitutionality, 115 Colum. L. Rev. Sidebar 55, 64 n.46 (2015) (listing cases) [https://perma.cc/DM44-PLTW].
See Brief of the Federal Public & Community Defenders & the National Ass’n of Federal Defenders as Amici Curiae in Support of Petitioner at 14, Jones v. United States, No. 15-8629 (U.S. Apr. 21, 2016) [https://perma.cc/L7ZG-59M2].
For example, a defendant who is convicted of being a felon in possession of a gun and has two previous felony convictions normally gets a sentencing range of fifteen to twenty-one months. See U.S. Sentencing Guidelines Manual § 2K2.1(a)(7) (U.S. Sentencing Comm’n 2015). If one of those convictions meets the residual clause definition, the range becomes forty-one to fifty-one months. See id. § 2K2.1(a)(4)(A). If both do, it becomes sixty-three to seventy-eight months. See id. § 2K2.1(a)(2). This increase is automatic even if the earlier convictions were punished with just a year in state prison. See id. § 4B1.2(a).
See Brief for Petitioner, supra note 27, at 28 n.7 (listing cases). Since that brief was filed, the en banc Seventh Circuit has also ruled that the residual clause in the Guidelines is unconstitutionally vague. See United States v. Hurlburt, Nos. 14-3611, 15-1686, 2016 WL 4506717, at *7 (7th Cir. Aug. 29, 2016) (en banc) [https://perma.cc/87Q9-5Y5K].
See supra notes 17–21. Indeed, the Eleventh Circuit has expressly told district courts that “[a]lthough Johnson abrogated the previous decisions of the Supreme Court interpreting the residual clause of the Armed Career Criminal Act, sentencing courts interpreting the residual clause of the guidelines must still adhere to the reasoning of cases interpreting the nearly identical language in the Act.” Matchett, 802 F.3d at 1195–96.
Johnson v. United States, 135 S. Ct. 2551, 2560 (2015) (citation omitted). For more colorful explanations in some of Justice Scalia’s opinions, see Litman, supra note 14, at 58 and Derby v. United States, 131 S. Ct. 2858, 2859 (2011) (mem.) (Scalia, J., dissenting from denial of certiorari) (“If it is uncertain how this Court will apply Sykes and the rest of our ACCA cases going forward, it is even more uncertain how our lower-court colleagues will deal with them. Conceivably, they will simply throw the opinions into the air in frustration, and give free rein to their own feelings as to what offenses should be considered crimes of violence . . . .”).
Chief Judge Patti B. Saris, Chair, U.S. Sentencing Comm’n, Remarks for Public Meeting 4 (Jan. 8, 2016), http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-and-meetings/20160108/meeting_minutes.pdf [https://perma.cc/HD3U-4WXG]. One of the sentencing commissioners who voted not to retroactively apply the amendment is Judge William H. Pryor, Jr., who also decided that Johnson does not apply to the Guidelines retroactively as a member of two Eleventh Circuit panels. See Matchett, 802 F.3d at 1195–96 (holding that Johnson does apply to the Sentencing Guidelines); In re Rivero, 797 F.3d 986, 988 (11th Cir. 2015) (holding that a rule that Johnson applies to the Sentencing Guidelines has not been “made retroactive” by the Supreme Court); see also Andrew Hessick, Should Judges Who Sit on the Sentencing Commission Rule on the Legality of Sentencing Guidelines?, Notice & Comment (Oct. 15, 2015), http://yalejreg.com/nc/should-judges-who-sit-on-the-sentencing-commission-rule-on-the-legality-of-sentencing-guidelines/ (noting that Judge Pryor’s “participation in [Matchett and Rivero] raises separation of powers concerns”) [https://perma.cc/F4EH-MPCG].
The Guidelines direct courts to apply the Sentencing Guidelines issued by the Sentencing Commission that are “in effect on the date that the defendant is sentenced” unless doing so would “violate the ex post facto clause of the United States Constitution,” in which case the court is to use the Guidelines Manual “in effect on the date that the offense of conviction was committed.” U.S. Sentencing Guidelines Manual §§ 1B1.11(a), (b)(1) (U.S. Sentencing Comm’n 2015); see 18 U.S.C. § 3553(a)(4)(A)(ii) (2012). Because the Guideline amendment deleting the residual clause imposes less punishment than the prior version of the Guideline, courts could apply the amendment to defendants who committed their offenses prior to the amendment, but were sentenced after the amendment.
Aside from holding the Guideline invalid, one way for the Supreme Court to avoid this result would be to declare that any sentence based on the Guideline is unreasonable. This is the approach that Judge Ikuta on the Ninth Circuit urged in United States v. Lee, 821 F.3d 1124, 1136 (9th Cir. 2016) (Ikuta, J., dissenting) [https://perma.cc/MGN7-LDGU]. It is not clear why this approach would improve on declaring a Guideline unconstitutionally vague: if applying a Guideline whose language is unconstitutionally vague is always unreasonable, why not just hold the Guideline unconstitutionally vague? Judge Ikuta’s proposed rule would also raise difficult retroactivity questions because the nature and source of such a rule would be unclear, given that it is not entirely clear what reasonableness review of sentencing determinations is. Reasonableness review is partially a judicial creation, but also partially statutory, see United States v. Booker, 543 U.S. 220, 245–46 (2005); and reasonableness review has both procedural and substantive components, see Gall v. United States, 552 U.S. 38, 49–50 (2007). It is therefore not clear if Mr. Beckles would benefit from a rule that said that applying the Guideline is unreasonable given that Mr. Beckles’s case is on collateral review. Because this Essay is concerned about the possibility that a rule would benefit Mr. Beckles while leaving certain others out of luck, we do not focus on the possibility that the Supreme Court will hold that sentences based on the residual clause of the Sentencing Guidelines are unreasonable.
Tyler v. Cain, 533 U.S. 656, 663 (2001). For more on Tyler and on the different approaches the courts of appeals have taken to the “made retroactive” requirement, see Leah M. Litman, Resentencing in the Shadow of Johnson v. United States, 28 Fed. Sent’g Rep. 45, 48–49 (2015) [https://perma.cc/7P2P-UGCG].
See In re Arnick, 826 F.3d 787, 788 (5th Cir. 2016) (per curiam) (“[T]he Supreme Court has not addressed whether this arguably new rule of criminal procedure [established in Johnson] applies retroactively to cases on collateral review.”) [https://perma.cc/V84Q-BEJX]; Donnell v. United States, 826 F.3d 1014, 1017 (8th Cir. 2016) (“Donnell’s successive motion seeks to assert a new right that has not been recognized by the Supreme Court or made retroactive on collateral review.”) [https://perma.cc/2WX7-N87S].
See Brief for Petitioner, supra note 27, at 15–17 (“Johnson’s rule is new as to Mr. Beckles because it was announced several years after his conviction became final, and it expressly overruled precedent foreclosing a vagueness challenge.”); id. at 14 (“Johnson has retroactive effect in this collateral proceeding. Johnson announced the following rule of constitutional law: a legal provision is void for vagueness under the Due Process Clause where it ‘requires a court to picture the kind of conduct that the crime involves in “the ordinary case,” and to judge whether that abstraction presents a serious potential risk of physical injury.’” (citation omitted)); id. at 46 (“Johnson has retroactive effect in this collateral case.”).
Petition for Writ of Certiorari, supra note 2, at 5–6 (arguing that the Supreme Court needed to decide the case in its 2015 Term because “the one-year statute of limitations governing collateral Johnson claims will expire on June 26, 2016”). One of the petitioner’s amici also noted this in the brief they filed at the merits stage. See Brief of the Federal Public & Community Defenders & the National Ass’n of Federal Defenders as Amici Curiae, supra note 25, at 2 (explicitly arguing that the statute of limitations has run).
The statute of limitations is subject to equitable tolling, Holland v. Florida, 560 U.S. 631, 649 (2010) [https://perma.cc/S64U-8EFD], and excepts cases of actual innocence, McQuiggin v. Perkins, 133 S. Ct. 1924, 1928 (2013) [https://perma.cc/9U9Q-F8HC]. A favorable decision in Beckles could mean that prisoners sentenced under the career offender guideline are actually innocent of their sentences, but this is by no means certain. See Litman, supra note 14, at 65–73 (discussing how cases are unclear on whether this kind of legal innocence qualifies as actual innocence).
Butler v. McKellar, 494 U.S. 407, 415 (1990). A subsequent case relied on the fact that “[t]wo Federal Courts of Appeals . . . reached conflicting holdings” to conclude that a rule was new. Caspari v. Bohlen, 510 U.S. 383, 395 (1994).
See Chaidez v. United States, 133 S. Ct. 1103, 1110–11 (2013) (“In acknowledging that fact, we do not cast doubt on, or at all denigrate, Padilla. Courts often need to, and do, break new ground; it is the very premise of Teague that a decision can be right and also be novel.”).
See, e.g., In re Anderson, 829 F.3d 1290, 1294 (11th Cir. 2016) [https://perma.cc/LR3D-MBTF]. Other circuits have elected to transfer the motions to district courts and stay them in abeyance of Beckles. See supra note 57 and accompanying text.
Other courts toll the statute of limitations for the disposition of successive motions. See, e.g., Orona v. United States, 826 F.3d 1196, 1198 (9th Cir. 2016) (per curiam) [https://perma.cc/9AL3-PUZC]. The Eleventh Circuit adjudicates requests for certification within thirty days, see In re Clayton, 829 F.3d 1254 (11th Cir. 2016) (Martin & Pryor, JJ., concurring) [https://perma.cc/UAN6-EGHL], and denies them with prejudice, so there may not be much resultant tolling.
See In re Bradford, 830 F.3d 1273, 1277 (11th Cir. 2016) (holding that § 2244 “removes our jurisdiction to consider” a prisoner’s motion to certify a Johnson motion if the prisoner asked for permission to file the same motion in the past). For an explanation of why the Eleventh Circuit’s insistence that the gatekeeping requirements in §§ 2244 and 2255 are jurisdictional is likely incorrect, see Leah M. Litman & Luke C. Beasley, Jurisdiction and Resentencing: How Prosecutorial Waiver Can Offer Remedies Congress Has Denied, 101 Cornell L. Rev. Online 91, 112 (2016) (“Gonzalez makes clear that only a prisoner’s failure to seek or obtain authorization from a court of appeals deprives a district court of jurisdiction over a successive petition. Once the prisoner has filed for and obtained authorization, he has cleared the lone jurisdictional hurdle.”) [https://perma.cc/F77Z-53BH].
Chief Judge Patti B. Saris, Chair, U.S. Sentencing Comm’n, Remarks for Public Meeting 6–7 (Jan. 8, 2016), http://www.ussc.gov/sites/default/files/pdf/amendment-process/public-hearings-and-meetings/20160108/meeting_minutes.pdf (noting that sentencing documentation does not identify which provision led to career offender designation or which criminal history events were counted as predicates).
U.S. Dep’t of Justice, Legal Resource Guide to the Federal Bureau of Prisons 10 (2014), https://www.bop.gov/resources/pdfs/legal_guide.pdf (“The PSR . . . contain[s] sensitive information regarding an inmate’s social contacts and criminal history, and are not permitted to be retained in the possession of the inmate.”) [https://perma.cc/AFA3-GCRT].
See In re Little, No. 16-3023 (3d Cir. filed June 30, 2016) (appears to assert a Beckles claim) (noting the application was received June 30, dated June 25, and postmarked June 27). The application was transferred to the Fourth Circuit because the original conviction was entered in Virginia. Id.
See In re Bradshaw, No. 16-2040 (6th Cir. filed July 20, 2016) (appears to assert a Beckles claim). The docket notes the application was initially filed in the district court, apparently in June 2016. Id.
See In re Toussaint, No. 16-72575 (9th Cir. filed Aug. 2, 2016) (appears to assert a Beckles claim). The docket notes the application was initially filed in the district court on June 27, 2016. Id.
See, e.g., In re Griffin, 823 F.3d 1350, 1354 (11th Cir. 2016) (“Griffin is unable to make a prima facie showing that Johnson applies to him in light of our binding precedent in Matchett that the Sentencing Guidelines cannot be challenged as unconstitutionally vague.”) (citing United States v. Matchett, 802 F.3d 1185, 1193–96 (11th Cir. 2015)); In re Sapp, 827 F.3d 1334, 1336 (11th Cir. 2016) (denying motion based on Griffin).
The other non-Eleventh Circuit cases available on searchable databases are: Dawkins v. United States, 809 F.3d 953, 956 (7th Cir. 2016) (per curiam) [https://perma.cc/3Q5M-RRM3]; Dawkins v. United States, 829 F.3d 549, 549 (7th Cir. 2016) (per curiam) (same case) [https://perma.cc/BUP5-Q8JG]; Hill v. United States, 827 F.3d 560, 562 (7th Cir. 2016) (denying motion) [https://perma.cc/SE8M-GQHR]; and United States v. Bolden, 645 F. App’x 282, 283 (4th Cir. 2016) (Mem.) (per curiam) (relying on prior precedent); cf. United States v. Bell, 622 F. App’x 770, 771 n.1 (10th Cir. 2015) (Mem.) (referring to denial of authorization) [https://perma.cc/QY2V-J6CL].
See In re Leonard, Nos. 16-13528-J, 16-13804-J, 16-13857-J, 2016 WL 3885037, at *3–4 (11th Cir. July 13, 2016) [https://perma.cc/4KY5-ZCJJ]; In re Parker, 827 F.3d 1286, 1288 (11th Cir. 2016) [https://perma.cc/UW2T-NZBC].
See, e.g., In re Smith, 829 F.3d 1276, 1280 n.4 (11th Cir. 2016) (“Although we have binding precedent to support our conclusion, we do not concede that such precedent is required.”); In re Sams, 830 F.3d 1234, 1239 (11th Cir. 2016).
18 U.S.C. § 924(e)(2)(B) (2012); U.S. Sentencing Guidelines Manual § 4B1.2(a)(1) (U.S. Sentencing Comm’n 2015), amended by U.S. Sentencing Guidelines Supplement to the 2015 Manual § 4B1.2(a)(1) (U.S. Sentencing Comm’n 2016).
18 U.S.C. § 924(e)(2)(B); U.S. Sentencing Guidelines Manual § 4B1.2(a)(2) (U.S. Sentencing Comm’n 2015), amended by U.S. Sentencing Guidelines Supplement to the 2015 Manual § 4B1.2(a)(2) (U.S. Sentencing Comm’n 2016).
See, e.g., In re Aiken, No. 16-12847-J (11th Cir. June 23, 2016); In re Alford, 16-12748-J (11th Cir. June 20, 2016); In re Carrasquillo, No. 16-12506-J (11th Cir. June 17, 2016); In re Thompson, No. 16-12595-J (11th Cir. June 17, 2016); In re Branson, 16-12675-J (11th Cir. June 16, 2016); In re Brown, 16-12557-J (11th Cir. June 16, 2016); In re Edwards, No. 16-12693-J (11th Cir. June 16, 2016); In re Townsend, No. 12-12659-J (11th Cir. June 16, 2016); In re Bell, No. 16-12532-J (11th Cir. June 15, 2016); In re Cruz, No. 16-12530-J (11th Cir. June 15, 2016); In re Franks, No. 16-12564-J (11th Cir. June 15, 2016); In re Parrish, No. 16-12652-J (11th Cir. June 15, 2016); In re Venta, No. 16-12698-J (11th Cir. June 15, 2016); In re White, No. 16-12570-J (11th Cir. June 15, 2016); In re Austin, No 16-12699-J (11th Cir. June 14, 2016); In re Creighton, No. 16-12580-J (11th Cir. June 14, 2016); In re Martin, No. 16-12503-J (11th Cir. June 14, 2016); In re Mims, No. 16-12574-J (11th Cir June 11, 2016); In re Sawyer, No. 16-12501-J (11th Cir. June 10, 2016); In re Safeeullah, No. 16-12443 (11th Cir. June 9, 2016); In re Hudson, No. 16-12243-J (11th Cir. June 8, 2016); In re Parks, No. 16-12404-H (11th Cir. June 8, 2016); In re Payne, No. 16-12290 (11th Cir. June 6, 2016); In re Knight, No. 16-12132-J (11th Cir. June 3, 2016); In re Garner, No. 16-12109-J (11th Cir. June 1, 2016); In re Little, No. 16-11979-J (11th Cir. May 27, 2016); In re McKinney, No. 16-11948-J (11th Cir. May 26, 2016); In re Turner, No. 16-11914-A (11th Cir. May 25, 2016); In re Leonard, No. 16-11925-J (11th Cir. May 24, 2016); In re Smith, No. 16-11901-C (11th Cir. May 24, 2016); In re Yawn, 16-12729-J (11th Cir. May 20, 2016); In re Simmons, No. 16-11563-B (11th Cir. May 4, 2016); In re Young, No. 16-11532-A (11th Cir. Apr. 28, 2016). This is a list of some of the split-panel rulings from between April 18 (the date Welch was decided) and June 26 (the one-year statute of limitations deadline on Johnson claims) of which we are aware, all denying motions. Because these orders are not published on the Eleventh Circuit’s website or a searchable database, they are nearly impossible to find other than by serially looking up docket numbers in the Eleventh Circuit’s Case Management/Electronic Case Files system. There were surely many more rulings like this, including ones issued after June 26 and ones issued by unanimous panels.
E.g., In re Burgest, 829 F.3d 1285, 1287 (11th Cir. 2016) (“Even if we were to assume that . . . Johnson also applies to . . . the Guidelines, Burgest would not be entitled to relief.”) [https://perma.cc/X2PC-T5PT]; In re Davis, 829 F.3d 1297, 1302 (11th Cir. 2016) (“[E]ven if we were considering in this current application a Johnson challenge to the district court’s application of the Guidelines’ career offender enhancement, Davis could not make a prima facie showing that Johnson impacted that sentencing decision because he clearly had two qualifying predicate offenses.”) [https://perma.cc/CCN7-RKTA]; In re Sams, 830 F.3d 1234, 1240 (11th Cir. 2016) (“[E]ven if Johnson retroactively applies to the Guidelines, Sams’s claims still fail.”).
For problems with relying on presentence investigation reports at the authorization stage, see In re Leonard, Nos. 16-13528-J, 16-13804-J, 16-13857-J, 2016 WL 3885037, at *8–9 (11th Cir. July 13, 2016) (Martin, J., concurring). Once a § 2255 motion is filed in district court, the additional time, briefing, and potential input from lawyers obviate these concerns.
See Blow v. United States, 829 F.3d 170, 171–72 (2d Cir. 2016) (“[B]ecause the Supreme Court will likely decide in Beckles whether Johnson applies retroactively to the Guidelines, the district court is instructed to hold Blow’s § 2255 motion in abeyance pending the outcome of Beckles.”).
See In re Clayton, 829 F.3d 1254 (11th Cir. 2016) (Martin & Pryor, JJ., concurring) (“[U]nlike all other circuits, the Eleventh Circuit refused to stay applications for successive § 2255 motions pending Welch.” (citation omitted)).
Descamps, 133 S. Ct. at 2281–82 (holding that “sentencing courts may not apply the modified categorical approach when the crime of which the defendant was convicted has a single, indivisible set of elements”).
Mathis, 136 S. Ct. at 2247–48 (A “prior crime qualifies as an ACCA predicate if, but only if, its elements are the same as, or narrower than, those of the generic offense. The question in this case is whether ACCA makes an exception to that rule when a defendant is convicted under a statute that lists multiple, alternative means of satisfying one (or more) of its elements. We decline to find such an exception”).
See, e.g., United States v. Lockett, 810 F.3d 1262, 1266 (11th Cir. 2016) (“Before Descamps, our Court ‘assumed that the modified categorical approach could be applied to all non-generic statutes. . . . The Descamps decision dictates discarding that assumption.’” (quoting United States v. Howard, 742 F.3d 1334, 1343 (11th Cir. 2014)).
See, e.g., In re Cook, No. 16-12745 (11th Cir. June 17, 2016) (unpublished). The Cook ruling helps illustrate how Descamps and Johnson interact. Mr. Cook was sentenced under ACCA because he had a previous Florida burglary conviction. Id. at 4. Seven years before Mr. Cook’s 2014 sentencing, the Supreme Court held that Florida burglary convictions are violent felonies under ACCA’s residual clause even if they are not violent felonies under the enumerated-crimes clause. See James v. United States, 559 U.S. 192, 209–10 (2007). Right before Mr. Cook was sentenced, the Court held in Descamps in 2013 that California burglary convictions can never count as violent felonies under the enumerated-crimes clause. Descamps, 133 S. Ct. at 2293. Descamps essentially confirmed that Florida burglary convictions also can never count as violent felonies under the enumerated-crimes clause. Of course, Descamps did Mr. Cook little good in 2014, since James meant that his burglary conviction still counted under the residual clause. But once Johnson struck ACCA’s residual clause in 2015, the enumerated-crimes clause was the only way this burglary conviction could support an ACCA sentence. But when Mr. Cook asked the Eleventh Circuit for permission to file a § 2255 motion raising a Johnson claim, a split panel denied that request based on pre-Descamps precedent holding that Florida burglary meets ACCA’s “enumerated crimes clause” definition, no matter that Descamps may have overruled that precedent as of the time of Mr. Cook’s sentencing. See Cook, No. 16-12745-J at 5–6.
As it happens, Cook was decided two days after another Eleventh Circuit panel issued a published (and therefore binding) order that granted a different prisoner’s request to challenge his ACCA sentence because the sentence was based on a Florida burglary conviction. See In re Adams, 825 F.3d 1283, 1284 (11th Cir. 2016) (granting motion). This means that even though Mr. Cook and Mr. Adams received mandatory ACCA sentences based on the same exact prior crimes, their Johnson claims had totally different outcomes. Mr. Adams was able to file a § 2255 motion, and the United States (which never gets to weigh in when the Eleventh Circuit denies permission to file a § 2255 motion) agreed that his sentence must be vacated. See United States v. Adams, No. 16-CV-22252, 3 (S.D. Fla. June 30, 2016). Meanwhile, Mr. Cook’s identical § 2255 claim began and ended with the application form he sent to the Eleventh Circuit, and this ruling cannot be reviewed.
Mr. Cook is one of many prisoners in the Eleventh Circuit whose ACCA sentences were based on Florida burglary but who will never get to challenge their sentence, even though others with identical sentences have already won relief. See, e.g., In re Chisholm, Nos. 16-13946-J, 16-14638-J (11th Cir. July 27, 2016); In re Yawn, No. 16-12729-J (11th Cir. June 20, 2016); In re Carrasquillo, No. 16-12506 (11th Cir. June 17, 2016); In re Branson, No. 16-12675-J (11th Cir. June 16, 2016); In re Brown, No. 16-12557-J (11th Cir. June 16, 2016); In re Parrish, No. 16-12652-J (11th Cir. June 15, 2016); In re McKinney, No. 16-11948-J (11th Cir. May 26, 2016); In re Thomas, 823 F.3d 1345, 1349 (11th Cir. 2016); In re Young, No. 16-11532-A (11th Cir. Apr. 28, 2016) (all denying motions for ACCA convictions based on Florida burglary). Again, these are just a few examples we are aware of. Because the majority of these orders are not published or reported, they are difficult to track down.
E.g., In re Hires, 825 F.3d 1297, 1303 (11th Cir. 2016) (holding that “Descamps is retroactive for a first § 2255 motion” but does not apply to successive-motion questions) [https://perma.cc/EPN3-MM6R]. This is another issue on which the Eleventh Circuit is split. See, e.g., Adams, 825 F.3d at 1285–86 (applying Descamps).
To be sure, Descamps discussed additional rationales for the Court’s holding on top of the statutory interpretation rationale. See Descamps, 133 S. Ct. at 2287 (“First, it comports with ACCA’s text and history. Second, it avoids the Sixth Amendment concerns that would arise from sentencing courts’ making findings of fact that properly belong to juries. And third, it averts ‘the practical difficulties and potential unfairness of a factual approach.’” (quoting Taylor v. United States, 495 U.S. 575, 601 (1990))). But these additional rationales do not make Descamps any less of a statutory interpretation decision. The fact that the Court said its reading of ACCA was more convenient and more constitutional than other readings does not make that reading of that statute any less authoritative a reading of the text. To the contrary, “[t]he so-called canon of constitutional avoidance is an interpretive tool,” much like other interpretative tools. FCC v. Fox Television Stations, Inc., 556 U.S. 502, 516 (2009); see also Ezell v. United States, 778 F.3d 762, 763 (9th Cir. 2015) (“We hold that the Supreme Court did not announce a new rule of constitutional law in Descamps. Rather, it clarified—as a matter of statutory interpretation—application of the ACCA in light of existing precedent.”) [https://perma.cc/59B6-97Q8].
Id. at 352 n.4; see also Welch v. United States, 136 S. Ct. 1257, 1267–68 (2016) (describing why decisions of statutory interpretation are not subject to the Teague bar); Bousley v. United States, 523 U.S. 614, 620 (1998) (“[B]ecause Teague by its terms applies only to procedural rules, we think it is inapplicable to the situation in which this Court decides the meaning of a criminal statute . . . .”).
See, e.g., Mathis v. United States, 133 S. Ct. 2243, 2257 (2016) (“Our precedents make this a straightforward case. For more than 25 years, we have repeatedly made clear that application of ACCA involves, and involves only, comparing elements.”); Descamps v. United States, 133 S. Ct. 2276, 2283 (2013) (“Our caselaw explaining the categorical approach and its ‘modified’ counterpart all but resolves this case.”); see also Ezell v. United States, 778 F.3d 762, 763 (9th Cir. 2015) (“[T]he Supreme Court did not announce a new rule of constitutional law in Decamps. Rather, it clarified—as a matter of statutory interpretation—application of the ACCA in light of existing precedent.”); United States v. Davis, 751 F.3d 769, 775 (6th Cir. 2014) (“The Supreme Court in Descamps explained that it was not announcing a new rule, but was simply reaffirming the Taylor/Shepard approach . . . .”) [https://perma.cc/PB8U-NDB5]; United States v. Montes, 570 F. App’x 830, 831 (10th Cir. 2014) (“[T]he Descamps decision did not recognize a new right.”). Though both Descamps and Mathis drew dissenting opinions, “[d]issents have been known to exaggerate the novelty of majority opinions; and ‘the mere existence of a dissent,’ like the existence of conflicting authority in . . . lower federal courts, does not establish that a rule is new.” Chaidez v. United States, 133 S. Ct. 1103, 1110 n.11 (2013) (quoting Beard v. Banks, 542 U.S. 406, 416 n.5 (2004)) [https://perma.cc/5SRP-N57Y].
Copyright 2016 by Leah M. Litman & Shakeer Rahman
Cite as: Leah M. Litman & Shakeer Rahman, What Lurks Below Beckles, 111 Nw. U. L. Rev. Online 69 (2016), http://scholarlycommons.law.northwestern.edu/cgi/viewcontent.cgi?article=1239&context=nulr_online.