Stoney LESTER, Petitioner-Appellant, v. UNITED STATES of America, Respondent-Appellee.
A petition for rehearing having been filed and a member of this Court in active service having requested a poll on whether this case should be reheard by the Court sitting en banc, and a majority of the judges in active service on this Court having voted against granting a rehearing en banc, it is ORDERED that this case will not be reheard en banc.
A majority of the Court has voted not to rehear this appeal en banc to reconsider whether a subset of federal prisoners classified as career offenders under the United States Sentencing Guidelines can seek resentencing on collateral review based on the void-for-vagueness doctrine explicated in Johnson v. United States, ––– U.S. ––––, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015), which invalidated the residual clause of the Armed Career Criminal Act. As background, we held in United States v. Matchett, 802 F.3d 1185 (11th Cir. 2015), and the Supreme Court later confirmed in Beckles v. United States, ––– U.S. ––––, 137 S.Ct. 886, 197 L.Ed.2d 145 (2017), that prisoners who were sentenced as career offenders after the Supreme Court’s decision in United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), cannot benefit from Johnson because—at least after Booker—the void-for-vagueness doctrine does not apply to the Sentencing Guidelines. But neither Matchett nor Beckles resolved whether those sentenced as career offenders before Booker can collaterally attack their sentences.
Our published order In re Griffin, 823 F.3d 1350 (11th Cir. 2016), answered that question in the negative for two independent reasons. First, Griffin applied the reasoning of Matchett to hold that the Guidelines are not susceptible to void-for-vagueness challenges no matter whether a prisoner was sentenced before or after Booker. See id. at 1354–55. Second, although the Supreme Court has held that the new legal rule of Johnson applies retroactively to cases on collateral review because it is “substantive,” see Welch v. United States, ––– U.S. ––––, 136 S.Ct. 1257, 194 L.Ed.2d 387 (2016), Griffin held that an analogous new rule applying the void-for-vagueness doctrine to pre-Booker career-offender sentences under the Guidelines would be not be substantive and, as a result, could not be applied retroactively. See 823 F.3d at 1355–56.
Griffin and Matchett speak for themselves as to the first holding, but I write in response to Judge Martin’s statement respecting the denial of rehearing en banc to explain that Griffin’s second holding must be correct. In brief, Johnson was a substantive decision that entitled prisoners sentenced under the residual clause of the Armed Career Criminal Act to relief because it made clear that the law had never authorized their convictions and sentences. By contrast, a new rule extending Johnson to set aside pre-Booker career-offender sentences—all of which fall within the substantive statutory ranges prescribed by Congress—would require only the procedural formality of a resentencing in which the district court would have the power to impose exactly the same sentence as before. Such a rule, as we concluded in Griffin, would not be substantive. See id. at 1355–56. To my mind, this nutshell summary of Griffin’s second holding says all that needs to be said. But, because some of my colleagues see the matter differently, this statement explains at greater length why a new rule extending Johnson’s void-for-vagueness holding to invalidate pre-Booker career-offender sentences would not be substantive.
I divide my discussion in three parts. First, I explain that the rationale underlying the retroactivity of substantive rules is that courts lack the authority to enter a judgment of conviction or impose a sentence the substance of which the law does not authorize, and I explain that the new rule that Judge Martin’s statement advocates would not be substantive in this sense because the sentences it would invalidate were imposed within the statutory ranges established by Congress. Second, I explain that Judge Martin’s counterargument cannot be squared with Booker. Third, I respond briefly to Judge Rosenbaum’s statement.
A. The Retroactivity of Substantive Rules.
In Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), the Supreme Court “adopt[ed] [the] approach to retroactivity for cases on collateral review” that the second Justice Harlan had developed in two influential opinions. Id. at 292, 109 S.Ct. 1060; Mackey v. United States, 401 U.S. 667, 675–702, 91 S.Ct. 1160, 28 L.Ed.2d 404 (1971) (Harlan, J., concurring in the judgment); Desist v. United States, 394 U.S. 244, 256–69, 89 S.Ct. 1048, 22 L.Ed.2d 248 (1969) (Harlan, J., dissenting). Under this familiar framework, “new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” Teague, 489 U.S. at 310, 109 S.Ct. 1060 (O’Connor, J., plurality opinion). Although the Court has maintained that there exists an exception for “a small set of ‘watershed rules of criminal procedure implicating the fundamental fairness and accuracy of the criminal proceeding,’ ” it has also stressed that this exception is “extremely narrow,” and it has never held that any rule falls within the exception. Schriro v. Summerlin, 542 U.S. 348, 352, 124 S.Ct. 2519, 159 L.Ed.2d 442 (2004) (some internal quotation marks omitted) (quoting Saffle v. Parks, 494 U.S. 484, 495, 110 S.Ct. 1257, 108 L.Ed.2d 415 (1990)).
In contrast to new procedural rules, “[n]ew substantive rules generally apply retroactively.” Id. at 351, 124 S.Ct. 2519. This category “includes decisions that narrow the scope of a criminal statute by interpreting its terms, as well as constitutional determinations that place particular conduct or persons covered by the statute beyond the State’s power to punish.” Id. at 351–52, 124 S.Ct. 2519 (citation omitted). It also includes “rules prohibiting a certain category of punishment for a class of defendants because of their status or offense.” Penry v. Lynaugh, 492 U.S. 302, 330, 109 S.Ct. 2934, 106 L.Ed.2d 256 (1989), overruled on other grounds by Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002).
Although it is common for courts to “refer[ ] to rules of this ․ type as falling under an exception to Teague’s bar on retroactive application of procedural rules,” “they are more accurately characterized as substantive rules not subject to the bar.” Schriro, 542 U.S. at 352 n.4, 124 S.Ct. 2519. After all, “Teague by its terms applies only to procedural rules.” Bousley v. United States, 523 U.S. 614, 620, 118 S.Ct. 1604, 140 L.Ed.2d 828 (1998). To be sure, there is no practical difference between a rule that falls within “an exception to Teague’s bar” and a rule “not subject to the bar.” Schriro, 542 U.S. at 352 n.4, 124 S.Ct. 2519. But the doctrinal basis of the retroactivity of substantive rules is independent of Teague and indeed long predates it.
More than a century before Teague, the Supreme Court held that prisoners could use the writ of habeas corpus to challenge the substantive validity of their convictions and sentences. Traditionally, of course, the writ was unavailable to challenge mere error by the sentencing court. The habeas jurisdiction of the federal courts was subject to “limitations ․ arising from the nature and objects of the writ itself, as defined by the common law.” Ex parte Siebold, 100 U.S. 371, 375, 25 L.Ed. 717 (1879). And “the general rule [was] that a conviction and sentence by a court of competent jurisdiction [was] lawful cause of imprisonment,” from which “no relief [could] be given by habeas corpus.” Id. “The only ground” for relief “to a prisoner under conviction and sentence of another court [was] the want of jurisdiction in such court over the person or the cause, or some other matter rendering its proceedings void.” Id.
In Siebold, the best known of a number of late-nineteenth-century decisions with similar rationales, the Court held that a prisoner’s argument that his statute of conviction was unconstitutional did not exceed the limited scope of the writ. The Court reasoned that an unconstitutional statute “affects the foundations of the whole proceedings.” Id. at 376. After all, “[a]n unconstitutional law is void, and is as no law.” Id. So, the Court deduced, “[a] conviction under it is not merely erroneous, but is illegal and void.” Id. And because the trial court’s “authority to indict and try the petitioners arose solely upon the[ ] laws” they challenged, “if the laws [were] unconstitutional,” it followed that “the [trial court] acquired no jurisdiction of the causes.” Id. at 377. In other decisions, the Court employed the same jurisdictional logic to hold that habeas corpus was available to correct a sentence imposed without legal authority. See, e.g., In re Bonner, 151 U.S. 242, 255, 14 S.Ct. 323, 38 L.Ed. 149 (1894); In re Mills, 135 U.S. 263, 270, 10 S.Ct. 762, 34 L.Ed. 107 (1890); Ex parte Lange, 85 U.S. 18 Wall. 163, 176, 21 L.Ed. 872 (1873). The Court treated each of these cases “not [as] a case of mere error, but [as] one in which the court below transcended its powers”—that is, exceeded its jurisdiction. Mills, 135 U.S. at 270, 10 S.Ct. 762.
To be sure, the Court’s reasoning in Siebold and in related decisions has had its doubters. Some jurists have argued that these decisions expanded the traditional concept of jurisdiction, see, e.g., Wright v. West, 505 U.S. 277, 285, 112 S.Ct. 2482, 120 L.Ed.2d 225 (1992) (opinion of Thomas, J.); Fay v. Noia, 372 U.S. 391, 450–51, 454–55, 83 S.Ct. 822, 9 L.Ed.2d 837 (1963) (Harlan, J., dissenting), maybe even to the point of unintelligibility, see Paul M. Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 Harv. L. Rev. 441, 470–71 & n.66 (1963). And Supreme Court justices have disagreed about how to interpret Siebold and about how far its logic extends. Compare Montgomery v. Louisiana, ––– U.S. ––––, 136 S.Ct. 718, 730–32, 193 L.Ed.2d 599 (2016), with id. at 740–41 (Scalia, J., dissenting), and id. at 748 (Thomas, J., dissenting).
But whatever doubt there may be about the reasoning in Siebold and similar decisions, there should be no doubt that, as a historical matter, it supplied the rationale for what we now know as the substantive “exception.” When Justice Harlan explained that courts should not bar the retroactive application of new rules “that place ․ certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe,” he added—with specific reference to Siebold and kindred decisions—that “the writ has historically been available for attacking convictions on such grounds.” Mackey, 401 U.S. at 692–93 & n.8, 91 S.Ct. 1160 (Harlan, J., concurring in the judgment); see also Desist, 394 U.S. at 261 n.2, 89 S.Ct. 1048 (Harlan, J., dissenting) (discussing Siebold’s jurisdictional rationale); Fay, 372 U.S. at 454–55, 83 S.Ct. 822 (Harlan, J., dissenting) (same). As the Court started to embrace Justice Harlan’s retroactivity framework in decisions leading up to Teague, it too connected retroactivity with the sentencing court’s lack of jurisdiction. See United States v. Johnson, 457 U.S. 537, 550, 102 S.Ct. 2579, 73 L.Ed.2d 202 (1982) (“recogniz[ing] [that] full retroactivity [is] a necessary adjunct to a ruling that a trial court lacked authority to convict or punish a criminal defendant in the first place,” making its “judgments or sentences ․ void ab initio”); see also Griffith v. Kentucky, 479 U.S. 314, 324, 107 S.Ct. 708, 93 L.Ed.2d 649 (1987) (paraphrasing Johnson as having explained that retroactivity attaches “when the new ruling was that a trial court lacked authority to convict a criminal defendant in the first place”).
Nor can there be any doubt that the substantive exception continues to rest on this jurisdictional rationale. In Montgomery v. Louisiana, the Court explained that “[b]y holding that new substantive rules are ․ retroactive, Teague continued a long tradition.” 136 S.Ct. at 730. The Montgomery Court illustrated that “tradition” with a detailed discussion of Siebold, which, it explained, had “addressed why substantive rules must have retroactive effect.” Id. The Court explained that “the same logic”—that is, Siebold’s jurisdictional rationale—“governs a challenge to a punishment” that the states lack “authority to impose.” Id. at 731. Montgomery reaffirmed Siebold’s rule that “[a] conviction or sentence imposed in violation of a substantive rule is not just erroneous but contrary to law and, as a result, void.” Id. And, the Court reasoned, “[i]t follows ․ that a court has no authority to leave in place a conviction or sentence that violates a substantive rule, regardless of [when] the conviction or sentence became final.” Id. The Court’s explanation in Welch of why Johnson is substantive—that because the residual clause of the Armed Career Criminal Act “is invalid,” it cannot “mandate or authorize any sentence,” 136 S.Ct. at 1265—also would have been at home in any of the decisions in the Siebold family. Cf., e.g., Mills, 135 U.S. at 270, 10 S.Ct. 762 (“The court below was without jurisdiction to pass any such sentences ․”).
The doctrinal pedigree and rationale of the substantive exception logically dictate its scope. The exception is based on the idea that a court has no authority to enter a judgment of conviction under a void statute or to impose a sentence not authorized by law. Necessarily, then, when a prisoner’s conviction is supported by a valid statute and his sentence is authorized by law, the substantive exception cannot apply. Cf. Lockhart v. Fretwell, 506 U.S. 364, 373, 113 S.Ct. 838, 122 L.Ed.2d 180 (1993) (“Cessante ratione legis, cessat et ipsa lex.”).
From this understanding of the substantive exception, it follows that the exception does not apply to pre-Booker sentences under the Guidelines. Every sentence that has ever been imposed under the Guidelines—either before or after Booker—falls within the statutory range of punishment prescribed by Congress for the defendant’s offense. See Mistretta v. United States, 488 U.S. 361, 374–75, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989) (explaining that when Congress empowered the Sentencing Commission to promulgate the Guidelines, it “instructed the Commission that these sentencing ranges must be consistent with pertinent provisions of Title 18 of the United States Code and could not include sentences in excess of the statutory maxima”). And every court that has ever imposed a sentence under the Guidelines had the authority to impose a sentence within the statutory range.
Take Stoney Lester as an example. Lester pleaded guilty in 2003 to distributing more than five grams of crack cocaine, an offense with a statutory range of five to 40 years of imprisonment. See 21 U.S.C. § 841(a)(1), (b)(1)(B)(iii). The sentence he received under the Guidelines—21 years and 10 months—was squarely within the statutory range and, as such, within the authority granted to the district court by Congress. So even if the career-offender provision that was used to calculate Lester’s sentence under the Guidelines were somehow unconstitutionally vague, the substantive exception would not extend to Lester’s sentence.
To look at the same idea from a different angle, consider the result if Lester or a similarly situated prisoner prevailed in a vagueness attack on the career-offender sentencing Guideline. A successful vagueness challenge would establish only that the district court improperly relied on an unacceptably unclear directive in arriving at the prisoner’s sentence. But it would not in the slightest establish that the district court lacked the authority to impose the substantive sentence at which it arrived. On the contrary, as Griffin observed, the district court could impose exactly the same sentence on remand after the appellate court corrected its error. 823 F.3d at 1355. In this respect, erroneous reliance on an unconstitutionally vague sentencing Guideline, if such a thing were to exist, resembles “procedural” sentencing errors—errors like “failing to calculate (or improperly calculating) the Guideline range, treating the Guidelines as mandatory, ․ or failing to adequately explain the chosen sentence,” Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007)—for which the remedy is a remand with corrective instructions but with no guarantee of a different outcome. And, as we explained in Griffin and as the Solicitor General argued in Beckles, this fact means that a new rule extending Johnson to the career-offender sentencing Guideline would also be “procedural,” not substantive, for purposes of retroactivity. See Griffin, 823 F.3d at 1355–56; Brief for the United States at 11–33, Beckles v. United States, ––– U.S. ––––, 137 S.Ct. 886, 197 L.Ed.2d 145, (No. 15-8544).
B. The “Mandatory Guidelines” and Booker.
Judge Martin’s statement responds that this reasoning “fundamentally misunderstands how the mandatory Guidelines worked” because it ignores that, before Booker, “the shortest sentence the District Court was legally allowed to give [Lester]” “[o]nce categorized as a career-offender” was greater than “the longest sentence he could have received if not deemed a career offender.” Statement of Martin, J., at 48. The argument is not expressly framed in terms of whether the district court had the substantive authority to impose Lester’s sentence, but to translate it into those terms is easy enough. In this view, it does not matter to retroactivity that the district court could sentence a pre-Booker career offender like Lester to the same term of imprisonment today. See id. at 49. Instead, the critical point is that—without the career-offender provision—it lacked the substantive authority to impose the sentence then, in the pre-Booker era when the Guidelines dictated what sentences were “legally allowed.” Id. at 48; see also id. at 38–40.
This response contains an important internal tension. In Judge Martin’s account, the putative unconstitutionality of the career-offender Guideline, which her statement contends was established by the Supreme Court’s 2015 decision in Johnson, must “relate back” to the time of Lester’s sentencing in 2003; otherwise, there would be no ground to argue that Lester’s sentence was substantively invalid. But the advisory nature of the Guidelines, which was established by the Supreme Court’s 2005 decision in Booker, must not relate back to Lester’s sentencing. If Booker’s construction of the Guidelines’ legal status was the correct view of the law at the time of Lester’s sentencing, then the “mandatory Guidelines” did not exist in a legal sense, did not limit the district court’s discretion to choose an appropriate sentence within the statutory range, and did not render his actual sentence substantively invalid.
The premise that the Sentencing Guidelines were in the strictest sense legally binding before Booker pervades Judge Martin’s statement. As her statement puts it, “[u]nder the mandatory Guidelines, federal district courts were bound by law to impose sentences within the guideline range.” Id. at 38, 128 S.Ct. 586. Her statement explains, “the mandatory Guidelines were laws that increased punishment”—laws that, “[i]n effect, ․ set minimum and maximum sentences.” Id. at 40, 48. And it insists that the Guidelines, “[b]y law,” dictated the sentences that the district court “was legally allowed to give” Lester. Id. at 48. In this view, “the sentencing judge had no legal authority to sentence [Lester] to other than a sentence mandated by the Guidelines then in effect.” Id. at 38–39.
This manner of speaking, of course, is by no means peculiar to Judge Martin. Virtually all of us are in the habit of distinguishing, in one way or another, between the “mandatory Guidelines” that operated before Booker and the “advisory Guidelines” that have operated since. See, e.g., Matchett, 802 F.3d at 1196 (referring to “when the guidelines were still mandatory”); United States v. Martin, 455 F.3d 1227, 1236 (11th Cir. 2006) (“As we all now know, Booker made the guidelines advisory.”).
Purely as a description of judicial doctrine and practice, this language makes sense. Cf. Oliver Wendell Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 457–61 (1897). Before Booker, district courts believed they were bound to apply the Guidelines as mandatory, and the courts of appeals would reverse district courts that treated them as merely advisory. After Booker, district courts recognized that they were bound not to apply the Guidelines as mandatory, and the courts of appeals will now reverse district courts that fail to treat them as advisory. This change in court behavior makes it natural to say that Booker “made the Guidelines advisory.”
But even if our idioms adequately describe judicial practice, they may be erroneous or inadequate in other respects. And we must be careful not to let the imprecision of our language control our reasoning. “[W]ords are wise men’s counters, they do but reckon by them; but they are the money of fools.” Thomas Hobbes, Leviathan 19 (Edwin Curley ed., Hackett 1994  ).
In particular, we should be mindful of the difference between a change in judicial doctrine and a change in law. This distinction, although sometimes easy to overlook, is fundamental. Without distinguishing between judges’ understanding of the law and the law itself, we could not meaningfully say, for example, that a change in judicial doctrine “proved Mr. Lester’s objection right and Eleventh Circuit case law wrong,” Statement of Martin, J., at 33, nor could the Supreme Court meaningfully describe a past decision of its own as “wrong the day it was decided,” Planned Parenthood of Se. Penn. v. Casey, 505 U.S. 833, 863, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992) (joint opinion of O’Connor, Kennedy, and Souter, JJ.); see also Stephen E. Sachs, Finding Law, 107 Calif. L. Rev. 527, 662-63 (2019). And we would be hard-pressed to explain why Johnson required the retroactive invalidation of final convictions and sentences based on the residual clause of the Armed Career Criminal Act, see Welch, 136 S.Ct. at 1265, when an act of Congress that merely repealed the clause would not do so. The answer, of course, is that Johnson did not strictly speaking change the law; “rather, it held and therefore established that the prior decisions [treating the residual clause as valid] were incorrect.” Rivers v. Roadway Express, Inc., 511 U.S. 298, 312, 114 S.Ct. 1510, 128 L.Ed.2d 274 (1994); see also Harper v. Va. Dep’t of Taxation, 509 U.S. 86, 97, 113 S.Ct. 2510, 125 L.Ed.2d 74 (1993). In the light of Johnson, the residual clause was, as a matter of law, always invalid.
So it can be misleading to say, as we often do, that “the Sentencing Guidelines were mandatory,” Statement of Martin, J., at 31, and that the Supreme Court “made” them advisory in Booker. These idioms suggest that the Guidelines were once really mandatory as a matter of law and that they became advisory as a matter of law only after—and only because of—Booker. But that cannot be right. If Booker’s effect was to change the essential legal status of the Guidelines from mandatory to advisory, then it must have been one of two things: either a legislative amendment that removed the textual command that courts apply the Guidelines as mandatory, see 18 U.S.C. § 3553(b)(1), from the Sentencing Reform Act and replaced it with a provision that the Guidelines be treated as advisory or, alternatively, a constitutional amendment that newly prohibited the mandatory application of the Guidelines. But Booker—a judicial decision—was neither of those things.
The Supreme Court wields judicial power, not legislative power. See U.S. Const. art. I, § 1; id. art. III, § 1. It has no power to amend the text or meaning of any statute; it has the authority only “to say what a statute means,” which means “what the statute meant before as well as after the decision of the case giving rise to that construction,” in the course of deciding a case or controversy. Rivers, 511 U.S. at 312–13, 114 S.Ct. 1510. Nor does the Supreme Court have the authority to amend the Constitution. See, e.g., Bell v. Maryland, 378 U.S. 226, 288, 84 S.Ct. 1814, 12 L.Ed.2d 822 (1964) (In constitutional cases, the Court’s “duty is to construe, not to rewrite or amend, the Constitution.” (internal quotation marks omitted)); South Carolina v. United States, 199 U.S. 437, 448, 26 S.Ct. 110, 50 L.Ed. 261 (1905) (“The Constitution is a written instrument. As such its meaning does not alter. That which it meant when adopted, it means now.”). So, instead of a legislative act or constitutional amendment that changed the law, Booker was a judicial decision that held that the statutory requirement that courts apply the Guidelines as mandatory is without legal effect.
Booker established that the Guidelines are not mandatory through two holdings. First, the Court held that the statutory requirement of mandatory application of the Guidelines would violate the Sixth Amendment as applied whenever it resulted in a sentence greater than the maximum sentence that could be imposed based only on the facts reflected in a jury’s guilty verdict or admitted by the defendant. See Booker, 543 U.S. at 233–37, 125 S.Ct. 738 (Stevens, J., opinion of the Court). Second, as a matter of severability, the Court held that the Guidelines could not be applied as mandatory in any cases, even when their mandatory application would not violate the Sixth Amendment, because the resulting system would be structurally unsound and contrary to the intent of Congress. See id. at 249–59, 265–67, 125 S.Ct. 738 (Breyer, J., opinion of the Court).
Although Booker radically changed judicial doctrine with respect to the Guidelines, neither of its holdings was the equivalent of a legislative or constitutional amendment to the substance of the law. As far as the constitutional holding is concerned, this point should be obvious: “Constitutional invalidity of federal statutes ․ is produced by the Constitution itself, not by the order of a court.” John Harrison, Severability, Remedies, and Constitutional Adjudication, 83 Geo. Wash. L. Rev. 56, 87 (2014); see also Marbury v. Madison, 5 U.S. 1 Cranch 137, 177–78, 2 L.Ed. 60 (1803). The Sixth Amendment did not mean one thing the day before Booker and another thing the day Booker was decided. The Court in Booker did not make the statutory provision requiring the mandatory application of the Guidelines unconstitutional. Instead, the Court recognized that the provision was unconstitutional by determining the meaning of the Sixth Amendment—that is, by interpreting it—and that its meaning was incompatible with that of the statutory provision in certain circumstances. In those applications, the statutory provision was always unconstitutional; it was always “void, and ․ as no law.” Siebold, 100 U.S. at 376; accord Marbury, 5 U.S. at 177.
Booker’s severability conclusion is no different. As the Supreme Court and commentators have long acknowledged, severability is fundamentally “an exercise in statutory interpretation.” Murphy v. Nat’l Collegiate Athletic Ass’n, ––– U.S. ––––, 138 S.Ct. 1461, 1486, 200 L.Ed.2d 854 (2018) (Thomas, J., concurring); see also, e.g., Dorchy v. Kansas, 264 U.S. 286, 290, 44 S.Ct. 323, 68 L.Ed. 686 (1924) (Severability is “a question of interpretation and of legislative intent.”); Michael C. Dorf, Facial Challenges to State and Federal Statutes, 46 Stan. L. Rev. 235, 289 (1994) (“Severability presents a question of statutory interpretation ․”). Specifically, issues of severability pose the interpretive question of how to understand a legal text as a whole in the light of a newly acknowledged fact—that part of the text cannot be given legal effect. See Murphy, 138 S.Ct. at 1486 (Thomas, J., concurring) (“[T]he severability doctrine has courts decide how a statute operates once they conclude that part of it cannot be constitutionally enforced.”); Harrison, supra, 83 Geo. Wash. L. Rev. at 88; see also Booker, 543 U.S. at 247, 125 S.Ct. 738 (Breyer, J., opinion of the Court) (observing that “severability questions ․ can arise when a legislatively unforeseen constitutional problem requires modification of a statutory provision as applied in a significant number of instances”); id. at 265, 125 S.Ct. 738 (“[W]e have examined the statute in depth to determine Congress’ likely intent in light of today’s holding.”); Dorchy, 264 U.S. at 290, 44 S.Ct. 323 (posing the question whether one section of a state statute was “so interwoven with the [other sections] held invalid that the section cannot stand alone”). It is because severability issues are interpretive questions that courts answer them by using the same kinds of tools that are applied to other issues of statutory construction, see, e.g., Alaska Airlines, Inc. v. Brock, 480 U.S. 678, 687, 107 S.Ct. 1476, 94 L.Ed.2d 661 (1987) (finding “abundant indication of a clear congressional intent of severability both in the language and structure of the Act and in its legislative history”), and that they produce the same kinds of disagreements about interpretive methods as do other interpretive problems, see, e.g., Booker, 543 U.S. at 313, 125 S.Ct. 738 (Thomas, J., dissenting in part) (“disagree[ing] with [Justice Stevens’s] restatement of severability principles and reliance on legislative history”).
The interpretive character of severability holdings—Booker’s included—also means that they cannot be understood as if they changed the law in the manner of a legislative amendment. As Justice Scalia once observed, even if it sometimes right to say that “judges in a real sense ‘make’ law,” we must not forget that “they make it as judges make it, which is to say as though they were ‘finding’ it—discerning what the law is, rather than decreeing what it is today changed to, or what it will tomorrow be.” James B. Beam Distilling Co. v. Georgia, 501 U.S. 529, 549, 111 S.Ct. 2439, 115 L.Ed.2d 481 (1991) (Scalia, J., concurring in the judgment). No less than a decision in any other interpretive context, a judicial construction of the Sentencing Reform Act in the light of its conflict with the Sixth Amendment—a conflict that always existed, even if the Court did not formally acknowledge it until Booker—“is an authoritative statement of what the statute meant before as well as after the decision of the case giving rise to that construction.” Rivers, 511 U.S. at 312–13, 114 S.Ct. 1510. Were it otherwise, Booker and other severability decisions would “entail quintessentially legislative work,” which is exactly what the Supreme Court has said they do not entail. Ayotte v. Planned Parenthood of N. New England, 546 U.S. 320, 329, 126 S.Ct. 961, 163 L.Ed.2d 812 (2006); accord Whole Woman’s Health v. Hellerstedt, ––– U.S. ––––, 136 S.Ct. 2292, 2319, 195 L.Ed.2d 665 (2016).
So, our usual figures of speech notwithstanding, it follows from the holdings in Booker that the Guidelines were never really mandatory—not as a matter of law. To be sure, an act of Congress instructed that the sentencing court “shall impose a sentence of the kind, and within the range,” defined in the Guidelines, 18 U.S.C. § 3553(b)(1), and district courts thought that this mandatory instruction was legally binding. The courts of appeals thought so too. But, if Booker is right, we were all wrong. The law—not judges’ understanding of the law, but the actual law—was otherwise. The Sixth Amendment forbade the mandatory application of the Guidelines whenever they prescribed a sentence greater than the law allowed based only on the facts reflected in a guilty verdict or plea. And, as interpreted by Booker in the light of that constitutional prohibition, the Sentencing Reform Act did not prescribe the mandatory application of the Guidelines in any other circumstances either.
The district court did err when it sentenced Lester, but the error was not imposing an illegal sentence under a mandatory, but unconstitutionally vague, sentencing Guideline. Instead, its error—an error that also happened to be shared by every other district court at the time—was treating the career-offender Guideline as mandatory in the first place. To be sure, that was an error of “constitutional significance,” as Booker confirmed. Statement of Martin, J., at 49. But it was a procedural constitutional error, not a substantive one. See Varela v. United States, 400 F.3d 864, 868 (11th Cir. 2005) (holding “that Booker’s constitutional rule falls squarely under the category of new rules of criminal procedure that do not apply retroactively”). And, although Judge Martin’s statement suggests that my acknowledgment of that error should convince me to “agree to fix the sentences mistakenly imposed as mandatory,” the whole point of Teague is that merely procedural “mistake[s]” do not warrant the extraordinary “fix” of collateral relief. Statement of Martin, J., at 39.
That job is reserved for substantive rules, and, in the light of Booker, Johnson cannot possibly have established that the district court lacked the substantive authority to impose Lester’s sentence. If the sentencing court had known what Booker later clarified to be the correct understanding of the law, it would not have recognized that, “[b]y law, the longest sentence [Lester] could ․ receive[ ] ․ was 151 months.” Id. at 48 (emphasis omitted). Instead, Johnson or no Johnson, it would have recognized that it was “legally allowed,” id., to impose any sentence within the statutory range of five to 40 years—just as it could if it sentenced Lester today.
Of course, we do not know what sentence Lester would have received if the district court had been aware of what Booker later held. But all we need to know is whether his sentence was within the substantive authority of the district court at the time it was imposed. And it clearly was. So Griffin was absolutely right to hold that a new rule extending Johnson to set aside pre-Booker sentences based on the career-offender Guideline would not be substantive and would not apply retroactively to cases on collateral review.
C. Judge Rosenbaum’s Statement.
Judge Rosenbaum’s statement responds to my argument in three parts. The first and second parts fail to engage with my argument. And the third part propounds a doctrine of judicial lawmaking that contradicts Marbury’s rationale for judicial review.
The first part of Judge Rosenbaum’s statement quarrels with my account of the substantive exception. I am not sure Judge Rosenbaum’s belief that the exception concerns equally ultra vires action by any branch, not just the judiciary, can be squared with basic habeas principles or with the precedents—all of which stress the sentencing “court[’s] lack[ ] [of] power to exact a penalty that has not been authorized by any valid criminal statute,” Welch, 136 S.Ct. at 1268; accord Siebold, 100 U.S. at 377 (“We think [habeas lies], because, if the laws are unconstitutional and void, the Circuit Court acquired no jurisdiction of the causes.” (emphases added)), or the habeas court’s lack of “authority to leave in place a conviction or sentence that violates a substantive rule,” Montgomery, 136 S.Ct. at 731—but I also doubt that it matters. After all, my argument is that Lester’s sentencing court did not exceed its powers because the statutory provision that purported to limit its sentencing discretion was unconstitutional and therefore void. Far from “gloss[ing] over th[e] constitutionally significant problem” of the mandatory application of the Guidelines, Statement of Rosenbaum, J., at 55, I argue that Booker’s holding that their mandatory application was unconstitutional controls the question we are discussing. And it establishes that the only rule the district court violated when it sentenced Lester was procedural, not substantive.
The second part of Judge Rosenbaum’s statement identifies the heart of my argument, but it offers no meaningful response. I have said that statements like “Booker made the Guidelines advisory” are ubiquitous but not precisely accurate. Judge Rosenbaum’s statement responds only by confirming that they are ubiquitous but makes no effort to refute my point that they are imprecise. See id. at 58–60. I have said that courts used to treat the Guidelines as mandatory but that, as Booker held, they committed legal error by doing so. Judge Rosenbaum’s statement responds only by insisting that courts used to treat the Guidelines as mandatory. See id. at 60. Her statement’s flotilla of quotations from the United States Reports, see id. at 58–59, ignores, first, that Booker held that the literal sense of those statements is false and, second, that courts routinely describe the terms and intended effects of statutes as if they were valid even as they hold the opposite. See, e.g., Murphy, 138 S.Ct. at 1483 (stating that the unconstitutional Professional and Amateur Sports Protection Act “banned the authorization of sports gambling in casinos” and “prohibited the spread of state-run lotteries”); Marbury, 5 U.S. at 176 (“The authority ․ given to the supreme court ․ appears not to be warranted by the [C]onstitution ․” (emphasis added)). The second part’s only direct response to my argument—that “the Booker Court did not make the Guidelines advisory because they were always advisory, since the Sixth Amendment never allowed them to be mandatory”—is that it “is certainly interesting on a metaphysical level.” Statement of Rosenbaum, J., at 60. I appreciate the compliment.
Only the third part of Judge Rosenbaum’s statement is responsive, but unfortunately it fumbles the basic tenets—one might even say the metaphysics—of judicial review. Before Booker, it tells us, “18 U.S.C. § 3553(b)(1) was the law of the land.” Id. at 61. Wrong. As Chief Justice Marshall explained in Marbury, “[t]he question, whether an act, repugnant to the [C]onstitution, can become the law of the land” is to be answered resoundingly in the negative. Marbury, 5 U.S. at 176; see also U.S. Const. art. VI, cl. 2 (“This Constitution, and the Laws of the United States ․ in Pursuance thereof ․ shall be the supreme Law of the Land ․” (emphasis added)); Mackey, 401 U.S. at 678, 91 S.Ct. 1160 (Harlan, J., concurring in the judgment) (observing that Marbury provides “from that day to this the sole continuing rationale for the exercise of” judicial review).
Judge Rosenbaum’s statement continues: “[T]he Court’s change effectively excised the Guidelines’ mandatory language from the United States Code.” Statement of Rosenbaum, J., at 62. The word “effectively” bears the whole weight of this sentence. Section 3553(b)(1)’s mandatory language is still there in the Code. See 18 U.S.C. § 3553(b)(1); see also generally Jonathan F. Mitchell, The Writ-of-Erasure Fallacy, 104 Va. L. Rev. 933 (2018). The only thing that has changed is that now, thanks to Booker, we know that it is legally inoperative because the Sixth Amendment—adopted in 1791—requires and has always required that result.
It follows that the correct application of the Guidelines has always been advisory—even for offenders, like Stoney Lester, convicted before Booker. And it follows from that that the substantive exception would not apply to a new rule extending Johnson to set aside Lester’s sentence and others like it. Understanding the law on this point requires no familiarity with “the laws of physics” or “the space-time continuum.” Statement of Rosenbaum, J., at 60. It instead requires only the humility that comes with appreciating the modest role that the federal judiciary performs in our constitutional system: “to say what the law is,” not what it should be. Marbury, 5 U.S. at 177.
In 2004, Stoney Lester was sentenced in the Middle District of Georgia to almost 22 years in prison for possessing more than five grams of cocaine. This long sentence was required by law at the time because the sentencing court had to treat Mr. Lester as a “career offender” under the then-mandatory United States Sentencing Guidelines. The Eleventh Circuit’s binding precedent required the sentencing judge to count Mr. Lester’s past conviction for walkaway escape as a “crime of violence” under a provision of the career-offender guideline known as the “residual clause.” U.S.S.G. § 4B1.2(a) (2003). If Mr. Lester had not been sentenced as a “career offender,” his term of incarceration would have been at least nine years shorter.
About eleven years after he was sentenced, the Supreme Court invalidated the residual clause of the Armed Career Criminal Act. Johnson v. United States, 576 U.S. ––––, 135 S. Ct. 2551, 192 L.Ed.2d 569 (2015). Since the residual clause under which he was sentenced was identical to that stricken by the Supreme Court, Mr. Lester filed a motion in Federal Court in Georgia under 28 U.S.C. § 2255, seeking to correct his sentence. At that time, he already had a pending petition under 28 U.S.C. § 2241 in the Eastern District of Virginia, where the Bureau of Prisons assigned him at some point. His petition in Virginia asked that court to correct his sentence based on Chambers v. United States, 555 U.S. 122, 129 S.Ct. 687, 172 L.Ed.2d 484 (2009), which overruled the Eleventh Circuit precedent that characterized walkaway escape as a crime of violence. Lester v. Wilson, Case No. 1:12cv681-LO-JFA, 2019 WL 982853 (E.D. Va. Feb. 27, 2019). Mr. Lester’s § 2241 petition would have failed in the Eleventh Circuit because our precedent foreclosed such a claim based on procedural grounds. See McCarthan v. Dir. of Goodwill Indus.-Suncoast, Inc., 851 F.3d 1076, 1090 (11th Cir. 2017) (en banc).
So too with Mr. Lester’s § 2255 petition. Our Court denied him relief based on our precedent in In re Griffin, 823 F.3d 1350 (11th Cir. 2016) (per curiam), which held that inmates like Mr. Lester cannot bring Johnson-based challenges to mandatory Guidelines sentences. Happily for Mr. Lester, things turned out better for him in the Fourth Circuit. That court said Mr. Lester’s sentence contained a “fundamental defect.” Specifically, the Fourth Circuit said Mr. Lester’s longer sentence based on his walkaway escape conviction was fundamentally defective since the Supreme Court had made plain that this should not count as a crime of violence. Lester v. Flournoy, 909 F.3d 708, 714 (4th Cir. 2018). While this Court was debating whether to consider Mr. Lester’s case en banc, the Fourth Circuit granted him relief.
I understand that Mr. Lester is a free man today. Dkt. No. 44, Lester v. Wilson, Case No. 1:12cv681-LO-JFA, 2019 WL 982853 (E.D. Va. Feb. 27, 2019). He served 15 years and seven months in prison. This was more than two-thirds of the sentence originally imposed on him, and three years longer than he would have served if his walkaway escape conviction had not counted in the first place.
I asked the Court to withhold the issuance of the mandate in Mr. Lester’s case, in hopes that we could correct what I see as two errors in In re Griffin. First, I believe In re Griffin erroneously held that the mandatory Guidelines are not subject to vagueness challenges. Second, I think In re Griffin was wrong to say that Johnson does not apply retroactively in mandatory Guidelines cases. This Court declined to take up either issue. In making that decision, the Court also denied Mr. Lester all relief.
I wrote a dissent from the denial of en banc rehearing to express my disagreement with Griffin. But since Mr. Lester has won his freedom, there is no longer a need for en banc review of his case. Still, I am mindful that Mr. Lester is breathing free air solely because of a happenstance administrative decision to move him to a prison in the Eastern District of Virginia, together with his savviness in filing a § 2241 petition while he was there. Failing either one of these things, Mr. Lester would have been required to serve six more years in prison. His case is a testament to the arbitrariness of contemporary habeas law, where liberty can depend as much on geography as anything else. It also illustrates how limited is relief for the thousands of inmates sentenced within the Eleventh Circuit and whose sentence Johnson affected.
Not all inmates will be moved out of this Circuit. Also, Griffin continues to stand as binding precedent in this Circuit. For these reasons I take this opportunity to review why I believe the opinion in In re Griffin is mistaken.
In 2004, Mr. Lester pled guilty to possession with intent to distribute more than five grams of crack cocaine. At that time, the Sentencing Guidelines were mandatory, see United States v. Booker, 543 U.S. 220, 233–34, 125 S.Ct. 738, 750, 160 L.Ed.2d 621 (2005), so the calculation of Mr. Lester’s guideline range set his prison sentence. The District Court calculated Mr. Lester’s guideline range as 262–327 months and, because it had no statutory basis to depart, see 18 U.S.C. § 3553(b)(1), it imposed a 262-month sentence.
Mr. Lester’s long sentence resulted from the District Court’s finding that he was a “career offender” as that term was then defined in the Guidelines. See U.S.S.G. § 4B1.1 (2003). The career-offender provision required longer sentences for defendants sentenced for a “crime of violence” or a “controlled substance offense,” and who in addition have two past convictions that are either a “crime of violence” or “controlled substance offense.”1 Id. § 4B1.1(a). Earlier in his life, Mr. Lester had been convicted in Georgia once for sale of marijuana and once for walkaway escape. The District Court counted the marijuana conviction as a “controlled substance offense.” It also counted Mr. Lester’s walkaway escape conviction as a “crime of violence” under the “residual clause” of the Guidelines. That “residual clause” defined a crime of violence as an offense punishable by more than a year in prison that “involves conduct that presents a serious potential risk of physical injury to another.” Id. § 4B1.2(a). Mr. Lester objected to counting his walkaway escape conviction as a crime of violence under the residual clause, but our Circuit precedent foreclosed his objection. See United States v. Gay, 251 F.3d 950, 953–55 (11th Cir. 2001) (per curiam), abrogation recognized by McCarthan, 851 F.3d at 1080.
In light of his career-offender status, Mr. Lester was assigned an offense level of 34 and a criminal history category of VI. This yielded a guideline range of 262 to 327 months. U.S.S.G. chap. 5, pt. A. At the time, the District Court had no discretion to sentence Mr. Lester outside of this range. If (as we now know should have been the case) Mr. Lester had not been designated a career-offender, he would have had an offense level of 30 and a criminal history category of III, which together yield a guideline range of 121–151 months. Id. Because of the residual clause, Mr. Lester faced at least nine years more in prison than he would have without it.
Over the years after his sentence was imposed, developments in the law proved Mr. Lester’s objection right and Eleventh Circuit case law wrong. In 2009, the Supreme Court overturned the Eleventh Circuit rule that walkaway escape counts as a crime of violence. Chambers, 555 U.S. at 130, 129 S.Ct. at 693, abrogated on other grounds by Johnson, 135 S.Ct. at 2563. If Mr. Lester were sentenced today, his conviction for walkaway escape could not count as a “crime of violence,” and as a result, he would not be sentenced as a career offender. See McCarthan, 851 F.3d at 1080. If he had been sentenced under Eleventh Circuit law as it exists today, Mr. Lester would have been a free man in 2016 at the latest.
Then in 2015, the Supreme Court changed the law in a second way that made clear Mr. Lester’s sentence resulted from not only a mistaken interpretation of the Guidelines, but an unconstitutionally vague Guideline provision as well. Johnson, 135 S.Ct. at 2563. In Johnson, the Supreme Court struck down a provision of the Armed Career Criminal Act (ACCA) that is identical to the career-offender residual clause in the Guidelines. Id. ACCA requires longer sentences for defendants with certain types of earlier convictions who are convicted of possessing a firearm as a convicted felon in violation of 18 U.S.C. § 922(g). The maximum sentence for possessing a firearm as a convicted felon is ordinarily ten years. 18 U.S.C. § 924(a)(2). But ACCA mandates a sentence of no less than fifteen years for a person who has three earlier convictions for either a “violent felony” or a “serious drug offense.” 18 U.S.C. § 924(e). ACCA defines “violent felony” with the same words that the career-offender guideline used when Mr. Lester was sentenced. Specifically, ACCA defines a “violent felony” as an offense punishable by more than a year in prison that either “has as an element the use, attempted use, or threatened use of physical force against the person of another” or “is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.” Compare id. § 924(e)(2)(B), with U.S.S.G. § 4B1.2 (defining “crime of violence” as the term is used in U.S.S.G. § 4B1.1). After years of litigation over how to interpret the residual clause, see Johnson, 135 S.Ct. at 2558–60, in Johnson, the Supreme Court struck it down as unconstitutionally vague, id. at 2563.
The constitutional problem with the residual clause is apparent in the application of what has become known as the categorical approach. When deciding whether a person’s earlier criminal conviction counts as a violent felony under the ACCA residual clause, the Supreme Court instructed us not to look at what the defendant actually did. See Johnson, 135 S.Ct. at 2557. For that matter, this is also true for the residual clause in the Guidelines. See United States v. Lockley, 632 F.3d 1238, 1243 n.5 (11th Cir. 2011). Rather than reconstruct the conduct that resulted in the earlier conviction, courts were instructed to “picture the kind of conduct that the crime involves in the ordinary case” and “judge whether that abstraction presents a serious potential risk of physical injury.” Johnson, 135 S.Ct. at 2557 (quotation marks omitted). But the residual clause gave no guidance “about how to estimate the risk posed by a crime,” because it gave no way for us to decide “what kind of conduct the ‘ordinary case’ of a crime involves.” Id. In a string of cases over the years, the Supreme Court used different “ad hoc test[s] to guide [the] inquiry” before abandoning the endeavor in 2015. Id. at 2558–59. Among other things, the Supreme Court was concerned that the residual clause left “uncertainty about how much risk it takes for a crime to qualify as a violent felony.” Id. at 2558. Ultimately the Supreme Court held that these features of the residual clause “produce[d] more unpredictability and arbitrariness than the Due Process Clause tolerates.” Id.
Johnson “called into question the sentences of thousands of federal prisoners sentenced under ACCA and other, similarly worded statutes,” including the career-offender guideline. Ovalles v. United States, 905 F.3d 1231, 1263 (11th Cir. 2018) (en banc) (Martin, J., dissenting). Johnson told us it is unconstitutional to require a longer sentence based on a judge imagining whether a defendant’s past convictions involve a “serious potential risk of physical injury.” Compare 18 U.S.C. § 924(e), with U.S.S.G. § 4B1.2(a). This is precisely how Mr. Lester got his longer sentence.
To correct an unlawful sentence, a prisoner may file a motion under 28 U.S.C. § 2255. But the Antiterrorism and Effective Death Penalty Act limits a prisoner to one such motion unless the prisoner comes to a Court of Appeals and applies for permission to file a second or successive § 2255 motion. Id. § 2244(b)(3)(A). This Court can, in turn, grant permission only if the prisoner makes a prima facie case that his second or successive motion will rely on newly discovered evidence or a new rule of constitutional law made retroactively applicable on collateral review by the Supreme Court. Id. §§ 2244(b), 2255(h). Once the Supreme Court told us that Johnson applies retroactively (to those sentenced before Johnson was decided) in Welch v. United States, 587 U.S. ––––, 136 S.Ct. 1257, 1268 (2016), this Circuit faced a flood of requests to file a second or successive motion, including one from Mr. Lester. The Court granted him permission to file a second § 2255 motion in District Court. In re Lester, No. 16-11730, slip op. at 9 (11th Cir. May 19, 2016).
I think Johnson entitled Mr. Lester to relief. Yet in the Eleventh Circuit he gets none. Just as Mr. Lester was about to have his sentence reviewed, his efforts were halted by a decision this Court published a few days after he got permission to file his second or successive § 2255 motion. In re Griffin ruled that the mandatory Guidelines are not subject to vagueness challenges and, alternatively, that Johnson is not retroactive as applied in this context. 823 F.3d at 1353–56. Griffin foreclosed Mr. Lester’s claim, so the District Court denied Mr. Lester’s § 2255 motion. Mr. Lester appealed, and a panel summarily affirmed, again relying on Griffin. Lester v. United States, No. 18-10523, slip op. at 3 (11th Cir. June 15, 2018).
But Griffin was wrong in both its rulings, and it should not have stood in the way of Mr. Lester having his sentence corrected.2 If it were not for the Fourth Circuit’s intervention, Mr. Lester would still be in prison. Because Griffin will continue to bar relief to others like Mr. Lester, I will elaborate on how it erred.
First, Griffin held that the mandatory Guidelines are not subject to vagueness challenges. 823 F.3d at 1353–56. But this cannot be accurate. The mandatory Guidelines are indeed subject to vagueness challenges because they “ha[d] the force and effect of laws.” Booker, 543 U.S. at 234, 125 S.Ct. at 750. Like any law, to comply with constitutional due process requirements the mandatory Guidelines had to “give ordinary people fair notice of the conduct [they] punishe[d]” and could not be “so standardless” as to “invite[ ] arbitrary enforcement.” Johnson, 135 S.Ct. at 2556.
Supreme Court precedent amply establishes that the Guidelines were equivalent to law. The Supreme Court characterized the mandatory Guidelines as “the equivalent of legislative rules adopted by federal agencies.” Stinson v. United States, 508 U.S. 36, 45, 113 S.Ct. 1913, 1919, 123 L.Ed.2d 598 (1993); see also Mistretta v. United States, 488 U.S. 361, 371–379, 109 S.Ct. 647, 654–59, 102 L.Ed.2d 714 (upholding the Sentencing Commission’s mandate to set penalties for crimes as a valid delegation of Congress’s legislative power). Under the mandatory Guidelines, federal district judges were bound by law to impose sentences within the guideline range calculated for any given defendant. See 18 U.S.C. § 3553(b) (1984); Booker, 543 U.S. at 234, 125 S.Ct. at 750 (“[S]ubsection (b) of [§ 3553] directs that the court ‘shall impose a sentence of the kind, and within the range’ established by the Guidelines, subject to departures in specific, limited cases.” (quoting 18 U.S.C. § 3553(b))). In Judge William Pryor’s statement respecting the denial of rehearing, he says that once Booker invalidated the mandatory guideline system, “it follows from the holdings in Booker that the Guidelines were never really mandatory – not as a matter of law.” Pryor Statement, p. 22. However, no one can dispute that when Mr. Lester was in the courtroom facing his sentence in 2004, the sentencing judge had no legal authority to sentence him to other than a sentence mandated by the Guidelines then in effect. But if it is true that Judge Pryor and I agree that Mr. Lester’s sentence never should have been imposed as though it were mandatory – I say we should agree to fix the sentences mistakenly imposed as mandatory on Mr. Lester and others like him.
We know that in the mine-run case, if any given Guidelines provision—say, the career-offender provision—required a longer sentence, courts were legally bound to impose that longer sentence, just as they would be bound to follow the statutory provisions of ACCA. For that reason, the mandatory Guidelines must be subject to vagueness challenges in the same way regulations are. E.g., FCC v. Fox Television Stations, Inc., 567 U.S. 239, 253, 132 S.Ct. 2307, 2317, 183 L.Ed.2d 234 (2012) (“Th[e] requirement of clarity in regulation is essential to the protections provided by the Due Process Clause of the Fifth Amendment. It requires the invalidation of laws that are impermissibly vague.” (citation omitted)). And five Courts of Appeal, this one included, entertained vagueness challenges to the mandatory Guidelines. See United States v. Savin, 349 F.3d 27, 38–39 (2d Cir. 2003); United States v. Rutherford, 175 F.3d 899, 906 (11th Cir. 1999); United States v. Johnson, 130 F.3d 1352, 1354 (9th Cir. 1997); United States v. Jones, 979 F.2d 317, 318–19 (3d Cir. 1992); United States v. Moore, No. 95-5586, 107 F.3d 868, 1997 WL 71707, at *1 (4th Cir. Feb. 20, 1997) (per curiam) (unpublished). But see United States v. Brierton, 165 F.3d 1133, 1139 (7th Cir. 1999) (concluding the mandatory Guidelines were not subject to vagueness challenges), abrogation recognized by Cross v. United States, 892 F.3d 288, 296 (7th Cir. 2018); United States v. Smith, 73 F.3d 1414, 1417–18 (6th Cir. 1996); United States v. Pearson, 910 F.2d 221, 223 (5th Cir. 1990); United States v. Wivell, 893 F.2d 156, 160 (8th Cir. 1990).
Mr. Lester’s case well illustrates why the mandatory Guidelines must give fair notice of the conduct they punish. If he had not been designated a career-offender, the law would not have permitted a sentence of higher than 151 months. Once he was designated a career-offender, however, the law did not permit the District Judge to impose a sentence of less than 262 months. The mandatory Guidelines established sentences in the same way ACCA does. Under ACCA, three prior qualifying convictions require a minimum fifteen-year sentence instead of a maximum ten-year sentence. 28 U.S.C. § 924(e). Under the career-offender guideline, two prior qualifying convictions required a minimum 262-month sentence for Mr. Lester instead of a maximum 151-month sentence. U.S.S.G. § 4B1.2(a) (2003). Just like ACCA, the mandatory Guidelines were laws that increased punishment. So in the same way as ACCA, they must be subject to vagueness challenges. See Johnson v. United States, 135 S.Ct. at 2557; Cross, 892 F.3d at 304–06 (holding the mandatory Guidelines are subject to vagueness challenges); In re Sapp, 827 F.3d 1334, 1338–39 (11th Cir. 2016) (Jordan, Rosenbaum, and Jill Pryor, JJ., concurring).
The statute that made the Guidelines mandatory did allow departures in limited circumstances. But this is not enough to insulate them from vagueness challenges. For one thing, “departures [were] not available in every case, and in fact [were] unavailable in most.” Booker, 543 U.S. at 243, 125 S.Ct. at 750; see Cross v. United States, 892 F.3d at 305–06 (concluding that Booker “addressed, and rejected, the argument that the possibility of departures from the mandatory guideline range was enough to make it advisory”). For another, the Supreme Court suggested that a defendant had a due process right to notice when a court was going to sentence him above the guideline range. Specifically, in Burns v. United States, 501 U.S. 129, 111 S.Ct. 2182, 115 L.Ed.2d 123 (1991), abrogated on other grounds as recognized by Dillon v. United States, 560 U.S. 817, 130 S.Ct. 2683, 177 L.Ed.2d 271 (2010), the Supreme Court read Federal Rule of Criminal Procedure 32 to require notice to a defendant before a district court sua sponte departed upward from the mandatory range. Id. at 135, 111 S.Ct. at 2186. Burns interpreted Rule 32 this way in part because of the due process problems that would result if defendants were not given notice. Id. at 138, 111 S.Ct. at 2187. I say that if due process demanded notice of a departure above the mandatory Guidelines, certainly due process would likewise demand notice that certain conduct would require imposition of a longer sentence on account of one’s career offender status. At any rate, we know departures were available from ACCA-enhanced sentences, see 18 U.S.C. § 3553(e), but this feature did not stop the Supreme Court from ruling the ACCA residual clause unconstitutionally vague in Johnson.
This Supreme Court precedent makes plain that the Griffin panel erred when it held the mandatory Guidelines are not subject to vagueness challenges. See 823 F.3d at 1354–56. The Griffin panel said that the “logic and principles” of United States v. Matchett, 802 F.3d 1185, 1193–96 (11th Cir. 2015), which held the advisory Guidelines are not subject to vagueness challenges, extended to the mandatory Guidelines, thus insulating them from vagueness challenges as well. See Griffin, 823 F.3d at 1354. The Supreme Court has endorsed the rationale of Matchett,3 recognizing the advisory Guidelines neither define a crime nor fix a sentence, but rather merely guide a judge’s sentencing discretion within the applicable statutory range. Beckles v. United States, 580 U.S. ––––, 137 S.Ct. 886, 892, 197 L.Ed.2d 145 (2017).4 Griffin missed the mark when it said that the mandatory Guidelines likewise only “cabin discretion.” 823 F.3d at 1355. The Supreme Court has consistently held that the mandatory Guidelines were like sentencing laws. They did not guide discretion—they fixed punishments. For that reason, they should be subject to vagueness challenges.
The Griffin panel also wrongly held that the rule established in Johnson does not apply retroactively to cases like Mr. Lester’s for the reason that he was sentenced under the mandatory Guidelines, as opposed to ACCA. 823 F.3d at 1355–56. Griffin’s refusal to give retroactive application to Johnson flies in the face of the Supreme Court’s instruction to us that it should be applied retroactively. Welch, 136 S.Ct. at 1268. Again, Johnson pronounced it unconstitutional to impose longer sentences on people who were earlier convicted under statutes that required judges to decide whether, in the ordinary case, the prior conviction posed a “serious potential risk of physical injury to another.” 135 S.Ct. at 2563; see also 18 U.S.C. § 924(e)(2)(B). The Supreme Court has told us that this rule applies retroactively, and Mr. Lester should have been able to benefit from it.
Griffin’s recognition of the retroactive application of the Johnson rule for one statute, but not for others, is a novel approach I have not found elsewhere. Notably, Griffin cited no legal precedent that would allow us to parcel out retroactivity in this way. See In re Sapp, 827 F.3d at 1339–40 (Jordan, Rosenbaum, and Jill Pryor, JJ., concurring) (noting that Griffin had no authority for its assumed premise “that a substantive rule of constitutional law expressly made retroactive by the Supreme Court can later be made only partially retroactive by a circuit court”). Neither have my independent efforts located support for this concept. To my knowledge, the Supreme Court has never sanctioned giving intermittent retroactive application to a decision since Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), established the current law on retroactivity. One of our sister circuits seems to have assumed, as Griffin did, that Johnson might not always apply retroactively. See Cross, 892 F.3d at 306–07 (observing that the Supreme Court’s reasoning in Welch “justifies treating Johnson as substantive, and therefore retroactive, when applied to the mandatory guidelines”). But like the Griffin opinion, Cross did not explain its retroactivity principle either. Also, Cross is in some tension with other Seventh Circuit precedent. See Price v. United States, 795 F.3d 731, 732 (7th Cir. 2015) (“We now conclude, consistently with the government’s position, that Johnson announces a new substantive rule of constitutional law that the Supreme Court has categorically made retroactive to final convictions.” (emphasis added)). Other circuits appear to have rejected intermittent retroactivity. See In re Hubbard, 825 F.3d 225, 234 (4th Cir. 2016) (“[T]he government has cited no case to support the proposition that a rule can be substantive in one context but procedural in another.”); see also United States v. Doe, 810 F.3d 132, 154 & n.13 (3d Cir. 2015) (accepting the government’s concession that Begay v. United States, 553 U.S. 137, 128 S.Ct. 1581, 170 L.Ed.2d 490 (2008), an ACCA case, applies retroactively on collateral review of a sentence and observing that “[u]nder Teague, either a rule is retroactive or it is not”).
The Supreme Court does not ask in each case whether a new rule, as applied, would meet Teague’s retroactivity standard. Instead, it analyzes whether the rule itself meets the retroactivity standard and then determines whether the petitioner can benefit from the rule. See Danforth v. Minnesota, 552 U.S. 264, 266, 128 S.Ct. 1029, 1032, 169 L.Ed.2d 859 (2008) (“New constitutional rules announced by this Court that place certain kinds of primary individual conduct beyond the power of the state to proscribe ․ must be applied in ․ all federal habeas corpus proceedings.”) (emphasis added)).
Two cases make the point. In Sawyer v. Smith, 497 U.S. 227, 110 S.Ct. 2822, 111 L.Ed.2d 193 (1990), the Supreme Court considered whether Caldwell v. Mississippi, 472 U.S. 320, 105 S.Ct. 2633, 86 L.Ed.2d 231 (1985), which prohibited death sentences imposed by juries who were led to believe the responsibility for the sentence rested elsewhere, applied retroactively. In conducting the retroactivity analysis, the Sawyer Court left to one side “the significant questions concerning the merits of petitioner’s Caldwell claim ․ , or the question whether application of Caldwell to the facts presented ․ would itself involve a new rule of law.” Sawyer, 497 U.S. at 234, 110 S.Ct. at 2827. It spoke only to “whether Caldwell is available to petitioner as a ground upon which he may seek relief”—that is, whether Caldwell announced a retroactively applicable rule, separate and apart from the merits of the petitioner’s Caldwell claim. Id.
For another example, in O’Dell v. Netherland, 521 U.S. 151, 117 S.Ct. 1969, 138 L.Ed.2d 351 (1997), the Supreme Court addressed the retroactivity of Simmons v. South Carolina, 512 U.S. 154, 114 S.Ct. 2187, 129 L.Ed.2d 133 (1994), which requires a capital defendant be permitted to inform a jury deciding whether to impose a death sentence that he is not eligible for parole when the prosecution argues he presents a future danger. Before considering whether the petitioner’s claim came within the scope of the Simmons rule, the Court explained it had to “determine whether the rule of Simmons was new for Teague purposes, and, if so, whether that rule falls within one of the two exceptions to Teague’s bar.” O’Dell, 521 U.S. at 159, 117 S.Ct. at 1974.
This Court’s own precedent (Griffin aside) likewise treats rules as either retroactive or not. E.g., In re Hill, 777 F.3d 1214, 1223–24 (11th Cir. 2015) (per curiam); In re Holladay, 331 F.3d 1169, 1173 (11th Cir. 2003). In In re Hill, for example, the panel first ruled that Hall v. Florida, 572 U.S. 701, 134 S.Ct. 1986, 188 L.Ed.2d 1007 (2014), did not apply retroactively on collateral review and only then ruled in the alternative that the petitioner did not come within the Hall rule. 777 F.3d at 1223–24. Our sister circuits seem to follow a similar practice. See, e.g., Hubbard, 825 F.3d at 234; Price, 795 F.3d at 734. But see Cross, 892 F.3d at 306–07.
It is worth noting that this across-the-board approach to retroactivity emerged from a backlash to pre-Teague jurisprudence. Before Teague, the Supreme Court decided whether a new rule applied retroactively in any given collateral review case by weighing the purpose of the new rule, the reliance of the States on prior law, and the effect on the administration of justice of a retroactive application of the new rule. See Linkletter v. Walker, 381 U.S. 618, 636–39, 85 S.Ct. 1731, 1741–43, 14 L.Ed.2d 601 (1965). The Linkletter standard resulted in an “unfortunate disparity in the treatment of similarly situated defendants on collateral review,” and the Supreme Court abandoned it for that reason. Teague, 489 U.S. at 305, 109 S.Ct. at 1072. The Griffin holding that substantive rules may apply retroactively in some cases but not others likewise invites disparity between habeas petitioners. Its holding is thus inconsistent with the purpose that drove the Teague decision: that new substantive rules should apply to all petitioners on collateral review. Id. at 316, 109 S.Ct. at 1078.
Perhaps worst of all, Griffin gave no explanation for its novel retroactivity holding. If this Court intends to intermittently apply rules retroactively, it should give the many habeas petitioners who come to us seeking relief an understanding of when a rule will apply retroactively to their case. This is particularly true where, as here, this Court does not intend to apply the rule after the Supreme Court has told us it applies retroactively. Will we do it case by case? Statute by statute? Griffin gives no answer to these questions.
Griffin also indicates that Johnson should be treated differently for sentencing Guidelines, because in that context Johnson would merely “produce changes in how the sentencing procedural process is to be conducted.” Griffin, 823 F.3d at 1355. The suggestion is that Johnson, applied to a mandatory Guidelines sentence, would merely change how a sentence is calculated and not the sentence itself. This fundamentally misunderstands how the mandatory Guidelines worked.
Mr. Lester’s case is illustrative. By law, the longest sentence he could have received if not deemed a career offender was 151 months. Once categorized as a career-offender, however, the shortest sentence District Court was legally allowed to give him was 262 months. In effect, the mandatory Guidelines set minimum and maximum sentences. Sapp, 827 F.3d at 1340 (Jordan, Rosenbaum, and Jill Pryor, JJ., concurring) (citing United States v. Wright, 607 F.3d 708, 718 (11th Cir. 2010) (William Pryor, J., concurring) (“Because Congress required sentencing courts to apply the [mandatory] Sentencing Guidelines and impose a sentence within the applicable guideline range, it was reasonable to view the Guidelines as effectively setting minimum and maximum penalties that varied based on the circumstances of the offense and the characteristics of the offender.”)).
Neither is it an answer to say, as Griffin does, that Johnson’s rule is simply procedural because Mr. Lester might receive the same sentence if he were resentenced now under the advisory Guidelines. Judge William Pryor makes this argument as well. Pryor Statement, pp. 11–12. On remand a district court would resentence Mr. Lester under the advisory Guidelines, as opposed to the mandatory Guidelines, and this is a difference with constitutional significance. See Booker, 543 U.S. at 233–34, 125 S.Ct. at 750; see also Beckles, 137 S.Ct. at 892 (noting that the advisory Guidelines “merely guide the exercise of a court’s discretion in choosing an appropriate sentence within the statutory range”). There is no constitutional doubt about the authority of a district judge to “exercise broad discretion in imposing a sentence within a statutory range,” including taking a defendant’s criminal history into account. Booker, 543 U.S. at 233, 125 S.Ct. at 750. But Johnson makes clear that Congress had no power to require a longer sentence based on a prior conviction because it “involves conduct that presents a serious potential risk of physical injury to another.” U.S.S.G. § 4B1.2(a)(2) (2003). In that sense, Johnson “alter[ed] the range of conduct ․ that the law punishes.” Schriro v. Summerlin, 542 U.S. 348, 353, 124 S.Ct. 2519, 2523, 159 L.Ed.2d 442 (2004).
What’s more, the Supreme Court has taught us in other contexts that a rule is not procedural just because a court might impose the same sentence after resentencing. In Montgomery v. Louisiana, 577 U.S. ––––, 136 S.Ct. 718, 193 L.Ed.2d 599 (2016), the Supreme Court held that the right of all but permanently incorrigible juveniles not to be sentenced to life without parole was a substantive rule. Id. at 732–34. Montgomery addressed a class of juvenile offenders who were being punished in an unconstitutional way. The opinion recognized that a subset of the juvenile offender class might be permanently incorrigible, and therefore entitled to no relief. However, this did not change the substantive right afforded to all juveniles. The same is true here. On remand, a district court could have decided that the sentence it imposed under the mandatory regime remains appropriate in light of the § 3553(a) factors. But this possibility does not change Johnson’s pronouncement of a substantive right not to have a longer sentence mandated under an unconstitutionally vague law. See Sapp, 827 F.3d at 1341 (Jordan, Rosenbaum, and Jill Pryor, JJ., concurring).
Mr. Lester should have been able to correct his 262-month career-offender sentence by a § 2255 motion in this Circuit, where he was sentenced. The residual clause in the mandatory Guidelines is unconstitutional for the same reasons it is unconstitutional in ACCA. There is simply no meaningful distinction between the two. See United States v. Brown, 868 F.3d 297, 311 (4th Cir. 2017) (Gregory, C.J., dissenting). This Court treats precedent interpreting the two as interchangeable. See Gilbert v. United States, 640 F.3d 1293, 1309 n.16 (11th Cir. 2011) (en banc). If a sentence imposed under the ACCA residual clause “does not comport with the Constitution’s guarantee of due process,” it must be true that relying on an identical “shapeless ․ provision” to impose a mandatory Guideline sentence doesn’t either. Johnson, 135 S.Ct. at 2560.
The relief Mr. Lester got from the Fourth Circuit highlights the limits this Court has placed on the Great Writ. His sentence was doubly infirm. He should not have been sentenced as a career offender. And his sentence was the product of an unconstitutionally vague Guidelines provision. The Fourth Circuit corrected the former error. If Mr. Lester had been convicted in the Seventh Circuit, that court would have corrected the latter. Eleventh Circuit precedent holds we have no power to do either.
I hoped our en banc Court would give Mr. Lester relief. Fortunately, the Court’s decision not to do so did not stand in the way of his freedom. But it bears notice that our unwillingness to reconsider Mr. Lester’s sentence will impact scores of inmates sentenced under the mandatory Guidelines in Alabama, Georgia, and Florida who have not been incarcerated outside of the Eleventh Circuit.
I agree with Judge Martin’s well-reasoned Statement respecting the denial of rehearing en banc. I write separately only to add a few points in response to Judge William Pryor’s Statement respecting the denial of rehearing en banc (“Pryor Statement”).
Section I of this dissent takes issue with the Pryor Statement’s argument that prisoners sentenced under the pre-Booker 1 Guidelines can never state a successful Johnson 2 claim because Johnson’s rule is not retroactively applicable to them. As I explain, Johnson’s rule, in fact, is retroactively applicable to such prisoners since pre-Booker, prisoners wrongly sentenced under the career-offender guideline were subjected to higher mandatory-minimum sentences than they otherwise would have been. Section II explores the Pryor Statement’s theory that the Sentencing Guidelines “were never really mandatory,” Pryor Statement at 22, even though individuals sentenced during the two decades when courts applied the pre-Booker Guidelines—including the career-offender provision—remain in prison because of a sentencing regime that supposedly never was. And Section III shows how the Pryor Statement’s theory that judicial opinions never change the law cannot be conceptually squared with the real effects of judicial opinions and the courts’ role in our constitutional system.
The Pryor Statement asserts that Lester could not state a viable Johnson claim because Johnson is not retroactively applicable to cases where the career-offender guideline was applied pre-Booker. Pryor Statement at 22-23. It arrives at this incorrect conclusion by oversimplifying the “doctrinal pedigree and rationale of the substantive exception” that requires retroactivity. That oversimplification incorrectly causes the Pryor Statement to overlook the constitutional significance of wrongly applied mandatory-minimum sentences. Id. at 10.
In fact, as I have explained in McCarthan v. Director of Goodwill Industries—Suncoast, Inc., 851 F.3d 1076, 1121-41 (11th Cir. 2017) (en banc) (Rosenbaum, J., dissenting), retroactivity applies when a previous rule imposed detention in violation of the separation of powers or of the government’s limited powers—that is, when the Congress, the Executive, or the Judiciary exercised powers it does not have in relation to causing the detention of a person. See, e.g., Mackey v. United States, 401 U.S. 667, 692, 91 S.Ct. 1160, 28 L.Ed.2d 404 (1971) (Harlan, J., concurring) (defining substantive rules that warrant retroactivity as “those that place, as a matter of constitutional interpretation, certain kinds of primary, private individual conduct beyond the power of the criminal lawmaking authority to proscribe”) (emphasis added).
In Johnson, for example, Congress exceeded its authority by enacting an unconstitutionally vague law—the residual clause of the Armed Career Criminal Act (“ACCA”)—“that fix[ed] the permissible sentences for criminal offenses.” Beckles v. United States, ––– U.S. ––––, 137 S.Ct. 886, 892, 197 L.Ed.2d 145 (2017) (emphasis omitted). The Judiciary also exceeded its authority when it applied this unconstitutional law to, among other things, increase mandatory-minimum sentences. See id. So not surprisingly, the Supreme Court held Johnson’s rule retroactively applicable. Welch v. United States, ––– U.S. ––––, 136 S.Ct. 1257, 194 L.Ed.2d 387 (2016).
The residual clause of the career-offender guideline, as employed under the pre-Booker regime, caused the same problem as the residual clause of the ACCA. First, the career-offender guideline mirrored the language of the ACCA’s residual clause that the Supreme Court ruled unconstitutionally vague. And second, as Judge William Pryor has previously explained, when “Congress required sentencing courts to apply the Sentencing Guidelines and impose a sentence within the applicable guidelines range, it was reasonable to view the Guidelines as effectively setting minimum and maximum penalties that varied based on the circumstances of the offense and the characteristics of the offender.” United States v. Wright, 607 F.3d 708, 717–18 (11th Cir. 2010) (Pryor, J.); see also Booker, 543 U.S. at 233–34, 125 S.Ct. 738 (opinion of Stevens, J.) (explaining how the pre-Booker Guidelines fixed sentences).
So a defendant wrongly found to qualify for the career-offender guideline was subjected to a higher mandatory-minimum sentence than he would have been under the mandatory Guidelines regime, had the career-offender guideline not been applied. And that is a constitutional problem. See Alleyne v. United States, 570 U.S. 99, 112, 133 S.Ct. 2151, 186 L.Ed.2d 314 (“It is impossible to dissociate the floor of a sentencing range from the penalty affixed to the crime.”), 113 (“Elevating the low-end of a sentencing range heightens the loss of liberty associated with the crime: the defendant’s expected punishment has increased as a result of the narrowed range, and the prosecution is empowered, by invoking the mandatory minimum, to require the judge to impose a higher punishment than he might wish.”) (cleaned up) (2013). In other words, by requiring courts to apply the unconstitutionally vague residual clause of the career-offender guideline in a mandatory way, Congress exceeded its constitutional power, since it prescribed mandatory-minimum sentences based on solely an unconstitutionally vague guideline it gave the force of binding law. And we, in applying these mandatory-minimum sentences based on solely an unconstitutionally vague guideline, also acted beyond our jurisdiction.
The Pryor Statement conveniently glosses over this constitutionally significant problem, diverting attention by focusing on the fact that the pre-Booker Guidelines did not cause defendants to be sentenced to higher maximum sentences than they would have been under the post-Booker Guidelines. But that defendants were not exposed to higher maximum penalties does not somehow render constitutionally insignificant the separate constitutional problem of “fix[ed]” higher mandatory-minimum sentences for defendants than were permissible under the pre-Booker regime. See Beckles v. United States, ––– U.S. ––––, 137 S.Ct. 886, 892, 197 L.Ed.2d 145 (2017) (citing Alleyne, 570 U.S. at 112-13, 133 S.Ct. 2151, as “describing the legally prescribed range of available sentences [which includes the mandatory minimum to the maximum] as the penalty fixed to a crime”).
The Pryor Statement has no answer to this problem. Instead, it conclusorily insists all that matters is whether the sentence that was actually imposed could have been applied under the substantive criminal statute for which the prisoner was convicted. But that ignores the fact that when a court applied the career-offender guideline during the mandatory Guidelines regime, the “legally prescribed range of available sentences” differed from what it would have been in the absence of the career-offender guideline during the mandatory Guidelines regime. In other words, it disregards the constitutional pickle created when Congress exceeded its powers in requiring, through 18 U.S.C. § 3553(b)(1)’s provision making the Guidelines mandatory, courts to apply higher mandatory-minimum sentences than would have otherwise applied in the absence of the unconstitutionally vague residual clause of the career-offender guideline, and courts followed that mandate. And it pays no attention to this problem even though the Supreme Court has expressly identified the constitutional significance attached to mandatory minimums. See Beckles, 137 S.Ct. at 892 (citing Alleyne, 570 U.S. at 112-13, 133 S.Ct. 2151).
The Pryor Statement divorces the fact that Congress made all the Guidelines mandatory through 18 U.S.C. § 3553(b)(1), from Congress’s exceeding of its powers by making the unconstitutionally vague residual clause of the career-offender guideline mandatory. By making this move, the Pryor Statement picks its real fight with Booker, not Johnson. See Pryor Statement at 25 (“I argue that Booker’s holding that their mandatory application was unconstitutional controls the question we are discussing.”). In doing so, the Pryor Statement misses that Congress’s Johnson error under the pre-Booker Guidelines was substantive in nature, since Congress exceeded its powers in requiring courts to apply higher mandatory-minimum sentences based solely on the unconstitutionally vague residual clause of the career-offender guideline, and courts followed that mandate.
Perhaps for this reason, the Pryor Statement takes a second tack to argue prisoners incorrectly sentenced as career offenders pre-Booker have no cognizable § 2255 claim. In an unusual move, the Pryor Statement denies the reality that these prisoners were actually sentenced under a mandatory regime. It reasons that since the Supreme Court in Booker found that the mandatory Guidelines violated the Sixth Amendment, they “were never really mandatory,” even though courts applied them that way for two decades. Pryor Statement at 22 (emphasis in original).
I doubt the perhaps 1,000-plus inmates 3 who sit in prison right now because a court sentenced them using a mandatory version of the Guidelines with an indisputably unconstitutionally vague career-offender clause would agree.4
Nor does the Pryor Statement’s conclusion that the Guidelines “were never really mandatory” comport with what the Supreme Court has had to say on the subject. On the contrary, the Court has bluntly explained that the pre-Booker Guidelines “b[ou]nd judges and courts in the exercise of their uncontested responsibility to pass sentence in criminal cases,” Mistretta v. United States, 488 U.S. 361, 391, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989), and “ha[d] the force and effect of laws, prescribing the sentences criminal defendants are to receive,” id. at 413, 109 S.Ct. 647 (Scalia, J., dissenting); see also Stinson v. United States, 508 U.S. 36, 42, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993) (“The principle that the Guidelines Manual is binding on federal courts applies as well to policy statements”). And just before excising the statutory provision that made the Guidelines mandatory, the Supreme Court stated that the Guidelines “are not advisory; they are mandatory and binding on all judges” because § 3553(b) “directs that the court ‘shall impose a sentence of the kind, and within the range’ established by the Guidelines.” Booker, 543 U.S. at 233–34, 125 S.Ct. 738 (opinion of Stevens, J.).
Indeed, the Pryor Statement’s conclusion today that the Guidelines “were never really mandatory” conflicts even with Judge William Pryor’s own prior statements. Judge Pryor has previously described the Sentencing Reform Act of 1984 as “ma[king] the Guidelines mandatory.” Wright, 607 F.3d at 717–18. In fact, as I have noted, Judge Pryor found that “[b]ecause Congress required sentencing courts to apply the Sentencing Guidelines and impose a sentence within the applicable guidelines range, it was reasonable to view the Guidelines as effectively setting minimum and maximum penalties that varied based on the circumstances of the offense and the characteristics of the offender.” Id. (emphasis added). And he explained that “the Court in ․ Booker ․ made the Guidelines advisory.” Id.
Today, though, the Pryor Statement chalks these remarks up to a failure of linguistic precision and seeks to rewrite history. See Pryor Statement at 17. According to the Pryor Statement, the Booker Court did not make the Guidelines advisory because they were always advisory, since the Sixth Amendment never allowed them to be mandatory. Id. at 19. That is certainly interesting on a metaphysical level.
But it ignores reality. Back here on Earth, the laws of physics still apply. And the Supreme Court’s invalidation of a law does not alter the space-time continuum. Indeed, there can be no dispute that from when the Guidelines were adopted in 1984 to when the Supreme Court handed down Booker in 2005, courts mandatorily applied them, as § 3553(b) required, to scores of criminal defendants—including many who still sit in prison because of them.
It’s also a particularly mindboggling bit of judicial fiction to, in one breath, conclude that the Guidelines were always advisory, and in the next, withhold relief from individuals in Lester’s circumstances by noting the advisory Guidelines do not apply retroactively because Booker is a procedural rule, even though, according to the Pryor Statement, the Guidelines always were advisory. Under the Pryor Statement’s reasoning, the Guidelines were never mandatory, but to inmates like Lester, they will always be mandatory, since these prisoners remain subject to their punishment. This heads-I-win-tails-you-lose logic cannot withstand scrutiny. Either the Guidelines were never mandatory, in which case, Lester and inmates like Lester would not have been sentenced under the mandatory regime or at least would not remain in prison because of the mandatory regime (a circumstance that is clearly not the case), or they were mandatory until Booker ruled they weren’t, and inmates like Lester can mount Johnson challenges.
The last portion of the Pryor Statement that warrants a response is its theory that when Booker invalidated the mandatory nature of the Guidelines, that was really just a change in “judicial doctrine,” but not “a change in law.” Pryor Statement at 16. According to the Pryor Statement, the Supreme Court could not have made the Sentencing Guidelines advisory because the “Supreme Court wields judicial power, not legislative power” and therefore “has no power to amend the text or meaning of any statute ․” Id. at 17.
To be sure, the “Supreme Court wields judicial power, not legislative power” and therefore “has no power to amend the text or meaning of any statute.” But that does not mean the Supreme Court’s actions do not, in a very real sense, change the law. In my view, there are at least two problems with the Pryor Statement’s line of thinking to the contrary: it ignores the reality of what happened in Booker, and it undermines the courts’ constitutional power of judicial review.
First, I address what happened in Booker. When the United States went to sleep on January 11, 2005, 18 U.S.C. § 3553(b)(1) was the law of the land, and sentencing courts were required to “impose ․ sentence[s] of the kind, and within the range” established by the Guidelines. But when the United States went to sleep on January 12, 2005—hours after the Court issued Booker—§ 3553(b)(1) no longer had the force of law. And that is so solely because the Court eliminated § 3553(b)(1) “through severance and excision”5 in order to “make the Guidelines system advisory.” Booker, 543 U.S. at 245, 125 S.Ct. 738 (opinion of Breyer, J.). In striking § 3553(b)(1), the Court recognized that it was “alter[ing] the system that Congress designed.” Id.
Despite the Supreme Court’s express words, the Pryor Statement opines that though the Court may have “radically changed judicial doctrine,” that change was not “equivalent” to a change in “the substance of the law.” Pryor Statement at 19. Call it a change in doctrine, law, policy—pick the noun—but the reality will always be the same: the Court’s change effectively excised the Guidelines’ mandatory language from the United States Code, rendering what were once statutorily imposed mandatory Guidelines advisory. Importantly, the Court had a duty under the Constitution to make this change to our law because “[i]t is emphatically the province and duty of the judicial department to say what the law is.” Marbury v. Madison, 5 U.S. 1 Cranch 137, 177, 2 L.Ed. 60 (1803). And here, § 3553(b)(1) violated the Constitution. That leads to my second point. Our duty to “say what the law is” likewise makes it our job, when a statute is unlawful, to render it unenforceable by saying what the Constitution requires of the statute. See id. (“If two laws conflict with each other, the courts must decide on the operation of each.”); see also The Federalist No. 78. So while it is true that the “Supreme Court ․ has [only] the authority to say what a statute means,” Pryor Statement at 17, sometimes those statements necessarily force a change in the law. And under our Constitution, those changes are every bit as valid and must be followed every bit as much as if Congress itself had enacted them.
Booker is a good example; one day the Guidelines were mandatory, and the next day they were not, solely because of the Supreme Court’s opinion striking down § 3553(b)(1). So while we can pretend that the Supreme Court’s decisions do not “change[ ] the law,” Pryor Statement at 17, the reality is that in our constitutional system, sometimes they can and they do. And as I have noted, when that happens, those changes are just as valid under our constitutional system as if Congress had statutorily enacted them.
Finally, I must respond to Judge Pryor’s charge that this Statement does not “say what the law is,” but rather what I supposedly think “it should be.” See Pryor Statement at 27 (quoting Marbury, 5 U.S. at 177) (quotation marks omitted). I trust that any honest reading of this Statement soundly rebuts that accusation. Nor does daring to disagree with the Pryor Statement’s analysis make my analysis either what I think the law “should be” or immodest. And I will let the irony of the Pryor Statement’s suggestion to the contrary speak for itself.
In short, Johnson’s rule applies retroactively when an unconstitutionally vague law results in the application of a fixed sentencing range that is higher than it otherwise would have been. Pre-Booker, when courts applied the career-offender guideline, that resulted in a higher mandatory minimum than otherwise would have pertained to a defendant under the mandatorily applied Guidelines. So Johnson’s rule applies to pre-Booker sentences that imposed the career-offender guideline based on at least one predicate conviction that qualified under only the unconstitutionally vague residual clause. Nor does the judicial fiction that the Guidelines “were never really mandatory” somehow erase the fact that they were applied in a mandatory way for more than 20 years. So if because of the residual clause, a prisoner was incorrectly sentenced under the career-offender guideline in the pre-Booker sentencing regime, he can establish a viable Johnson claim. And our Circuit precedent to the contrary is simply wrong.
1. The Guidelines contain a career-offender provision in their current iteration, see U.S.S.G. § 4B1.1 (2018), but the provision and our interpretation of it have changed since Mr. Lester was sentenced. My discussion here is about the career-offender provision that was operative when Mr. Lester was sentenced.
2. Griffin ruled on the merits of this issue of first impression, even though the only question before the panel was whether the inmate there had made a prima facie case that his motion would contain a new rule of constitutional law made retroactive by the Supreme Court. This Court has had a long-running debate about this practice, and some of my colleagues and I have criticized it on various grounds. See Ovalles, 905 F.3d at 1266–68 (Martin, J., dissenting); In re Williams, 898 F.3d 1098, 1101–02 (11th Cir. 2018) (Wilson, J., specially concurring).
3. In the interest of full disclosure, I thought that Matchett’s ruling that the advisory Guidelines are not subject to vagueness challenges was wrong. In re Clayton, 829 F.3d 1254, 1258–61 (11th Cir. 2016) (Martin, J., concurring in the result). Of course, I stand corrected by the Supreme Court’s ruling in Beckles.
4. Beckles did not decide anything about the mandatory Guidelines. 137 S.Ct. at 890 (“Because we hold that the advisory Guidelines are not subject to vagueness challenges under the Due Process Clause, we reject petitioner’s argument.” (emphasis added)); id. at 903 n.4 (Sotomayor, J., concurring) (noting that Beckles leaves open the question of vagueness challenges to the mandatory Guidelines).
1. United States v. Booker, 543 U.S. 220, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005).
2. Johnson v. United States, ––– U.S. ––––, 135 S.Ct. 2551, 192 L.Ed.2d 569 (2015).
3. Brown v. United States, ––– U.S. ––––, 139 S.Ct. 14, 16 n.4, 202 L.Ed.2d 302 (2018) (noting that there are likely at least 1,187 Johnson challenges to sentences under the pre-Booker career-offender clause).
4. The Pryor Statement expresses wonderment about whether the career-offender provision of could “somehow [be] unconstitutionally vague ․” Pryor Statement at 12. He need not wonder about the career-offender’s vagueness though because it employs the same language as the ACCA, and the Sentencing Commission expressly modeled the career-offender Guideline after the ACCA. U.S. Sentencing Guidelines Manual app. C, amend. 268 (U.S. Sentencing Comm’n 1989) (“The definition of crime of violence used in this amendment is derived from 18 U.S.C. § 924(e).”).
5. The Pryor Statement takes issue with my characterization of what the Supreme Court did in Booker as “excis[ion].” See Pryor Statement at 27. Of course, that is precisely the word the Supreme Court used. See Booker, 543 U.S. at 245, 125 S.Ct. 738 (opinion of Breyer, J.). And it used that word because in the real world, that was the legal effect of what the Supreme Court did in Booker to the mandatory nature of the Guidelines, whether the Pryor Statement wishes to acknowledge that fact or not.
BY THE COURT: