Jill Pryor, Circuit Judge

Peter J. Sholl, Dawn A. Tiffin, U.S. Attorney Service - Middle District of Florida, U.S. Attorney, U.S. Attorney's Office, Tampa, FL, for Plaintiff-Appellee.

Conrad Benjamin Kahn, Joshua Roy Lukman, Rosemary Cakmis, Federal Public Defender's Office, Orlando, FL, for Defendant-Appellant.

Caleb Kruckenberg, Pacific Legal Foundation, Arlington, VA, for Amicus Curiae.

Before William Pryor, Chief Judge, Wilson, Jordan, Rosenbaum, Jill Pryor, Newsom, Branch, Grant, Luck, Lagoa, and Brasher, Circuit Judges.

William Pryor, Chief Judge, filed a concurring opinion.

Grant, Circuit Judge, filed an opinion concurring in the judgment.

Luck, Circuit Judge, filed a dissenting opinion, in which Branch, Circuit Judge, joined.

Jill Pryor, Circuit Judge: This appeal requires us to consider whether an inchoate offense qualifies as a "controlled substance offense" for purposes of the career offender sentencing enhancement under the United States Sentencing Guidelines. U.S. Sent'g Guidelines Manual § 4B1.2(b) (U.S. Sent'g Comm'n 2018). In this case, the district court sentenced Brandon Dupree as a career offender based partly on his conviction for conspiring to possess with intent to distribute a controlled substance in violation of 21 U.S.C. § 846. Dupree appealed his sentence, arguing that his § 846 conspiracy conviction could not serve as a predicate for his career offender enhancement because the Guidelines’ definition of "controlled substance offense" omitted conspiracy and other inchoate crimes.

A panel of this Court affirmed Dupree's sentence, concluding that our decisions in United States v. Weir , 51 F.3d 1031 (11th Cir. 1995), and United States v. Smith , 54 F.3d 690 (11th Cir. 1995), foreclosed his argument. United States v. Dupree , 849 F. App'x 911 (11th Cir. 2021) (unpublished), reh'g en banc granted, opinion vacated 25 F.4th 1341 (11th Cir. 2022). We granted Dupree's petition to rehear the case en banc. After careful consideration, and with the benefit of oral argument, we hold that the definition of "controlled substance offense" in § 4B1.2(b) does not include inchoate offenses. We therefore vacate Dupree's sentence and remand to the district court for resentencing.

I. BACKGROUND

Dupree pled guilty to one count of possession of a firearm after having been convicted of a felony, in violation of 18 U.S.C. §§ 922(g)(1) and 924(a)(2) ; one count of conspiracy to possess with intent to distribute heroin and cocaine, in violation of 21 U.S.C. § 846 ; and one count of carrying a firearm in furtherance of a drug trafficking offense, in violation of 18 U.S.C. § 924(c)(1)(A)(i). Before his sentencing, a probation officer prepared a Presentence Investigation Report ("PSR"). The PSR reported that Dupree had two previous convictions for controlled substance offenses. The PSR considered Dupree's § 846 conspiracy conviction to be his third controlled substance offense. Together, these three offenses qualified Dupree for the career offender enhancement under § 4B1.1(a) of the Sentencing Guidelines.

Applying the enhancement, the PSR assigned Dupree a total offense level of 29 with a criminal history category of VI, for a range of 151 to 188 months of imprisonment under the Guidelines. In addition, Dupree was required to serve a consecutive 60-month term of imprisonment for the carrying a firearm count. See 18 U.S.C. § 924(c). After adding the mandatory minimum consecutive penalty required by § 924(c), Dupree's guidelines range was 211 to 248 months of imprisonment. Without the enhancement, Dupree's guidelines range would have been lower. He would have had an offense level of 23 with a criminal history category of V, resulting in a guideline range of 84 to 105 months’ imprisonment. After adding the mandatory minimum penalty required by § 924(c), his guidelines range would have been 144 to 165 months’ imprisonment.

Dupree objected to the enhancement, arguing that inchoate crimes such as his § 846 conspiracy conviction did not qualify as a controlled substance offense. He argued that without his § 846 conspiracy conviction serving as his third qualifying offense he could not be sentenced as a career offender. The district court overruled Dupree's objection and applied the enhancement. Based on considerations including Dupree's youth and difficult upbringing, as well as his serious medical conditions including paralysis from waist down, the district court varied downward from the guideline range and sentenced Dupree to 106 months’ imprisonment.

On appeal, Dupree renewed his argument that his § 846 conspiracy conviction did not count as a controlled substance offense. He pointed to the plain language of § 4B1.2, which omitted inchoate offenses from the definition of "controlled substance offense." He acknowledged that the commentary to § 4B1.2—specifically Application Note 1—included inchoate crimes in the definition. But he argued that Application Note 1 was unenforceable because it was inconsistent with § 4B1.2's plain text.

The panel rejected Dupree's argument. Dupree , 849 F. App'x at 912. It relied on Weir , in which "we held that conspiracy to possess with intent to distribute marijuana was a controlled substance offense within the meaning of the career offender enhancement." Id. (citing Weir , 51 F.3d at 1031–32 ). The panel further noted that in Smith , decided after Weir , we "held ... that [A]pplication [N]ote 1 to ... § 4B1.2 ‘constitutes a binding interpretation of the term controlled substance offense.’ " Id. (quoting Smith , 54 F.3d at 693 ). This precedent, the panel concluded, bound it to affirm Dupree's sentence. Id.

Dupree petitioned for, and we granted, rehearing en banc to revisit our precedent.

II. STANDARD OF REVIEW

"We review de novo the interpretation and application of the Sentencing Guidelines." United States v. Cingari , 952 F.3d 1301, 1305 (11th Cir. 2020).

III. DISCUSSION

Dupree argues that he was not a career offender under the Guidelines. A defendant is considered a career offender for purposes of sentencing if, among other things, the defendant "has at least two prior felony convictions of either a crime of violence or a controlled substance offense" and his "instant offense of conviction is a felony that is either a crime of violence or a controlled substance offense." U.S. Sent'g Guidelines Manual § 4B1.1(a). Dupree concedes that his prior state drug convictions qualified as controlled substance offenses. But he argues that his § 846 conspiracy conviction does not count as the required third predicate offense because it was not a controlled substance offense. Section 4B1.2 provides:

The term "controlled substance offense" means an offense under federal or state law, punishable by imprisonment for a term exceeding one year, that prohibits the manufacture, import, export, distribution, or dispensing of a controlled substance ... or the possession of a controlled substance ... with intent to manufacture, import, export, distribute, or dispense.

Id. § 4B1.2(b). The commentary in Application Note 1 to § 4B1.2 adds that the term " ‘controlled substance offense’ include[s] the offenses of aiding and abetting, conspiring, and attempting to commit such offenses." Id. § 4B1.2(b) cmt. n.1.

Dupree contends that his § 846 conspiracy conviction is not a controlled substance offense because § 4B1.2(b)’s "controlled substance offense" definition unambiguously excludes inchoate offenses. Because the Guideline is unambiguous, he argues, we must not defer to the commentary's broader definition of controlled substance offense to include inchoate offenses. We agree. We begin with the framework the Supreme Court has established for determining how the Guidelines’ commentary impacts the interpretation of the Guidelines. We then use that framework to interpret § 4B1.2(b). We conclude by applying our ruling to Dupree's case.

A. The Commentary Cannot Expand the Interpretation of Unambiguous Sentencing Guidelines.

The Supreme Court examined whether courts are bound by the commentary's interpretation of the Guidelines in Stinson v. United States , 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993). The Court began by explaining that "[t]he Sentencing Reform Act of 1984 ... created the Sentencing Commission ... and charged it with the task of establishing sentencing policies and practices for the Federal criminal justice system." Id. at 40–41, 113 S.Ct. 1913 (alterations adopted) (internal quotation marks and citations omitted). "The Sentencing Commission promulgate[d] the [G]uidelines by virtue of an express congressional delegation of authority for rulemaking" just as federal administrative agencies promulgate regulations. Id. at 44, 113 S.Ct. 1913. The Court then analogized the Guidelines’ commentary to "an agency's interpretation of its own legislative rules." Id. at 45, 113 S.Ct. 1913 ("[T]he [G]uidelines are the equivalent of legislative rules adopted by federal agencies. ... [T]his type of commentary is akin to an agency's interpretation of its own legislative rules."). Guided by this analogy, the Court determined that the commentary should receive the same level of deference given to an agency's interpretation of its own rules, deference the Court first described in Bowles v. Seminole Rock & Sand Co . Id.

Seminole Rock instructed that when considering how to treat an issuing agency's interpretation of a regulation, a court initially should consider whether "the meaning of the [regulation] is in doubt." Bowles v. Seminole Rock & Sand Co. , 325 U.S. 410, 414, 65 S.Ct. 1215, 89 L.Ed. 1700 (1945). If the meaning is in doubt—if the regulation is ambiguous—the court can then consider the issuing agency's interpretation of the regulation. Id. At that point, the court should afford the agency's construction of its own regulation "controlling weight" unless it is "plainly erroneous or inconsistent with the regulation." Id.

Relying on its decision in Seminole Rock , in Stinson the Supreme Court treated the Sentencing Commission's commentary to the Guidelines like an agency's interpretation of its own regulation: "it must be given ‘controlling weight unless it is plainly erroneous or inconsistent with’ " the Guideline's text. Stinson , 508 U.S. at 45, 113 S.Ct. 1913 (quoting Seminole Rock , 325 U.S. at 414, 65 S.Ct. 1215 ). The Court explained that the Sentencing Commission could resort to the commentary to interpret the Guidelines only "if the [G]uideline which the commentary interprets will bear the construction." Id. at 46, 113 S.Ct. 1913. When the "commentary and the [G]uideline it interprets are inconsistent in that following one will result in violating the dictates of the other, the Sentencing Reform Act itself commands compliance with the [G]uideline." Id. at 43, 113 S.Ct. 1913.

Four years after Stinson , the Supreme Court reaffirmed Seminole Rock in Auer v. Robbins , 519 U.S. 452, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997). In Auer , the Court concluded that the Secretary of Labor's interpretation of a regulation issued by the Department of Labor was "controlling" because it was not "plainly erroneous or inconsistent with the regulation." Id. at 461, 117 S.Ct. 905 (internal quotation marks omitted).

A few years ago, the Supreme Court revisited Auer deference and clarified the proper application of the doctrine. In Kisor v. Wilkie , the Supreme Court examined Auer deference in the context of an administrative agency's (the Department of Veterans Affairs (VA)) interpretation of one of its regulations. ––– U.S. ––––, 139 S. Ct. 2400, 2408–09, 204 L.Ed.2d 841 (2019). The Court reaffirmed Auer ’s "important role in construing agency regulations" while also "reinforc[ing] its limits" and "cabin[ing] ... its scope." Id. at 2408. It explained that Auer was "rooted in a presumption ... that Congress would generally want the agency to play the primary role in resolving regulatory ambiguities." Id. at 2412. Therefore, the Court clarified, "only if a regulation is genuinely ambiguous" should Auer deference be applied. Id . at 2414. To determine whether ambiguity exists, courts first "must exhaust all the ‘traditional tools’ of construction." Id . at 2415 (citing Chevron, U.S.A. Inc. v. Nat. Res. Def. Council, Inc. , 467 U.S. 837, 843 n.9, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) ). "If uncertainty does not exist" after applying these tools, "there is no plausible reason for deference." Id .

The Supreme Court did not overrule Stinson in Kisor , and Kisor did not concern the Sentencing Guidelines. So our task is to figure out how to read Stinson and Kisor together. And we think the only way to harmonize the two cases is to conclude that Kisor ’s gloss on Auer and Seminole Rock applies to Stinson .

In Kisor , the Supreme Court answered the question whether it was overruling Auer or Seminole Rock with a firm "no." Id. at 2408. But the Court nonetheless took to task Seminole Rock ’s "most classic formulation of the [deference] test," which asks "whether an agency's construction is ‘plainly erroneous or inconsistent with the regulation.’ " Id. at 2415 (quoting Seminole Rock , 325 U.S. at 414, 65 S.Ct. 1215 ). The majority criticized this language "in a vacuum," as suggestive of "a caricature of the doctrine, in which deference is reflexive." Id. (internal quotation marks omitted). Instead, the majority said, courts must consider the nature and context of the agency interpretation, and "[f]irst and foremost, a court should not afford Auer deference unless the regulation is genuinely ambiguous." Id. The Court declared, in no uncertain terms, that "[i]f uncertainty does not exist, there is no plausible reason for deference. The regulation then just means what it means—and the court must give it effect, as the court would any law." Id.

We fail to see how this clarification does not apply to the Sentencing Guidelines. Stinson adopted word for word the test the Kisor majority regarded as a "caricature," so the continued mechanical application of that test would conflict directly with Kisor . Even if Stinson ’s application of the test was more considerate of the nature and context of the Sentencing Commission's commentary to the Guidelines than other cases were in applying the test to agency interpretations of regulations, if "uncertainty does not exist" in the Guideline, Kisor says we may not defer.

Kisor did not distinguish between an agency's interpretation of its own regulations and the commentary's interpretation of the Guidelines. This is perhaps unsurprising given that Kisor concerned the VA's interpretation of a VA regulation rather than the Sentencing Commission's commentary to the Guidelines. But Kisor had no need to make this distinction because Stinson had instructed "that the commentary be treated as an agency's interpretation of its own legislative rule." Stinson , 508 U.S. at 44, 113 S.Ct. 1913. Consistent with Stinson ’s instruction about how the Guidelines commentary should be treated, a footnote to the historical background section in Kisor included Stinson alongside administrative agency cases in a list of "(pre– Auer ) decisions applying Seminole Rock deference." Kisor , 139 S. Ct at 2411 n.3.

Stinson adopted Seminole Rock ’s formulation of agency deference. See Stinson , 508 U.S. at 45, 113 S.Ct. 1913 ("[C]ommentary is akin to an agency's interpretation of its own legislative rules. As we have often stated, provided an agency's interpretation of its own regulations does not violate the Constitution or a federal statute, it must be given ‘controlling weight unless it is plainly erroneous or inconsistent with the regulation.’ " (quoting Seminole Rock , 325 U.S. at 414, 65 S.Ct. 1215 )). A few years later, Auer reaffirmed the same test. Auer , 519 U.S. at 461, 117 S.Ct. 905. More recently, Kisor declined to overrule Auer and Seminole Rock , but it "reinforce[d] [the] limits" and "cabined ... [the] scope" of Auer deference, with the primary rule being no deference without ambiguity. Kisor , 139 S. Ct. at 2408. And it took note of Stinson as one of the Court's "legion" Seminole Rock cases that pre-dated Auer. Id. at 2411 n.3, 117 S.Ct. 905. So it follows that Kisor ’s clarification of Auer deference applies to the Guidelines and its commentary. The dissent argues that by applying Kisor ’s clarification to Stinson we have set aside Stinson’ s command that commentary "that interprets or explains a guideline is authoritative," unless it violates the law or is plainly erroneous. Dissent at 1290-91 (quoting Stinson , 508 U.S. at 38, 113 S.Ct. 1913 ). According to our dissenting colleague, we have "essentially overruled Stinson ." Id. at 1290; see Grant Concurrence at 1282, 1285-86. To the contrary, we honor Stinson ’s instruction to "treat[ ]" the commentary "as an agency's interpretation of its own legislative rule." Stinson , 508 U.S. at 44, 113 S.Ct. 1913.

As we have explained, when Stinson was decided, Seminole Rock held that an agency's interpretation of its own rule was "controlling" unless it contravened the rule itself or was plainly erroneous. Seminole Rock , 325 U.S. at 414, 65 S.Ct. 1215. Thus, Stinson ’s conclusion that the commentary is authoritative and entitled to deference is a result of treating the commentary as an agency's interpretation of its own rule. To follow Stinson ’s instruction to treat the commentary like an agency's interpretation of its own rule, we must apply Kisor ’s clarification of Auer deference to Stinson . We have not ignored Stinson or treated it as having been overruled—rather, it directs us to Seminole Rock and Auer , as clarified by Kisor .

We agree, of course, with the dissent that "[t]he Supreme Court doesn't upend decades of precedent through silence." Dissent at 1292. But the Supreme Court has not been silent. It has spoken directly to the issue of whether the Guidelines and its commentary, on the one hand, and an agency's rules and its interpretation of those rules, on the other hand, should be treated differently and concluded they should be treated the same. Our conclusion today flows not from the Supreme Court's silence, but from its affirmation that the commentary should be treated the same as the agencies’ interpretations that were at issue in Seminole Rock , and now Auer and Kisor .

With Kisor ’s refined deference scheme in mind, we turn to whether we should defer to the commentary in Application Note 1, which would supplement § 4B1.2's definition of "controlled substance offense" to include inchoate offenses.

B. Section 4B1.2 Unambiguously Excludes Inchoate Offenses, So the Commentary's Interpretation Is Not Binding.

We begin, as Kisor instructs, with the text of § 4B1.2. After applying our traditional tools of statutory interpretation, we conclude that the plain language definition of "controlled substance offense" in § 4B1.2 unambiguously excludes inchoate offenses.

Section 4B1.2(b), entitled "Definitions of Terms used in Section 4B1.1," provides:

The term "controlled substance offense" means an offense under federal or state law, punishable by imprisonment for a term exceeding one year, that prohibits the manufacture, import, export, distribution, or dispensing of a controlled substance ... or the possession of a controlled substance ... with intent to manufacture, import, export, distribute, or dispense.

U.S. Sent'g Guidelines Manual § 4B1.2(b) (emphasis added). The definition does not mention conspiracy or attempt or any other inchoate crimes. The exclusion of inchoate crimes from the definition of what the term "means" is a strong indicator that the term does not include those offenses. A "definition which declares what a term ‘means’ excludes any meaning that is not stated." Burgess v. United States , 553 U.S. 124, 130, 128 S.Ct. 1572, 170 L.Ed.2d 478 (2008) (alterations adopted) (internal quotation marks omitted); see United States v. Nasir , 17 F.4th 459, 471 (3d Cir. 2021) (en banc) ("The guideline does not even mention inchoate offenses. That alone indicates it does not include them."); United States v. Winstead , 890 F.3d 1082, 1091 (D.C. Cir. 2018) ("Section 4B1.2(b) presents a very detailed ‘definition’ of controlled substance offense that clearly excludes inchoate offenses."). We agree with the en banc Sixth Circuit's observation, in overruling its prior precedent to the contrary, that "[t]o make attempt crimes a part of § 4B1.2(b), the Commission did not interpret a term in the guideline itself—no term in § 4B1.2(b) would bear that construction." United States v. Havis , 927 F.3d 382, 386 (6th Cir. 2019) (en banc). Instead, the Commission purported "to add an offense not listed in the [G]uideline." Id. (emphasis in original).

The lack of any reference to conspiracy or attempt crimes stands in stark contrast to the other definition found in § 4B1.2. In defining "crime of violence," the Sentencing Commission included "any offense ... that has as an element the use, attempted use, or threatened use of physical force against the person of another." U.S. Sent'g Guidelines Manual § 4B1.2(a)(1) (emphasis added). With this definition, the Sentencing Commission demonstrated that it knew how to include attempted conduct in addition to the conduct itself when it meant to do so. A drafting body such as the Sentencing Commission "generally acts intentionally when it uses particular language in one section ... but omits it in another." Dep't of Homeland Sec. v. MacLean , 574 U.S. 383, 391, 135 S.Ct. 913, 190 L.Ed.2d 771 (2015). This "interpretive cannon ... applies with particular force" where the provision that includes specific language is in "close proximity" to the provision that excludes it. Id. at 392, 135 S.Ct. 913. That is true here. The definitions of "crime of violence"—which includes offenses involving the use of physical force as well as the attempted use of physical force—and "controlled substance offense"—which does not include attempt at all—are sister subsections within the same Guideline provision. Because of their proximity, we must infer that the Sentencing Commission intentionally excluded inchoate offenses from the definition of controlled substance offense. See Nasir , 17 F.4th. at 472 (explaining that because § 4B1.2(b) "does not even mention inchoate offenses," but § 4B1.2(a) "explicitly include[s] inchoate crimes" in its definition of "crime of violence," § 4B1.2(b) unambiguously omits inchoate crimes).

The government tries to inject ambiguity into § 4B1.2(b) by pointing to the word "prohibits" in the definition of "controlled substance offense." See U.S. Sent'g Guidelines Manual § 4B1.2(b) ("The term ‘controlled substance offense’ means an offense ... that prohibits ...."). "Prohibit," the government contends, has multiple meanings that make it ambiguous whether controlled substance offenses include inchoate crimes. Because § 4B1.2(b) is ambiguous, the government argues, we must defer to the commentary in Application Note 1 and add inchoate crimes to the definition of "controlled substance offense." The government relies on United States v. Lange , in which we interpreted "prohibit" in § 4B1.2(b). 862 F.3d 1290, 1295 (11th Cir. 2017). Looking to dictionary definitions, we observed that "prohibit" could mean, among other things, "to forbid by a command, statute, law or authority," or "to prevent, hinder." Id. (alterations adopted) (internal quotation marks omitted). Rejecting that the meaning was confined to the narrower definition of "prohibit"—to forbid by law—we concluded that "prohibit" also meant to prevent or hinder because of Application Note 1's inclusion of aiding and abetting, conspiracy, and attempt. Id. ("Because Application Note 1 tells us that an offense prohibits the manufacture of a controlled substance when it prohibits aiding and abetting, conspiring, and attempting that manufacture, ... we must not construe ‘prohibit’ too narrowly." (internal citation omitted)). There are two problems with the government's reliance on Lange .

The first problem, of course, is that in Lange we considered the commentary in Application Note 1 to be "a binding interpretation of the term controlled substance offense" based on Smith and then proceeded to interpret the Guideline light of the commentary. Id. at 1294 (internal quotation marks omitted). So, if Smith is wrong, then Lange is, too.

The second, more fundamental, problem is that "prohibit" cannot mean "to prevent [or] hinder" in the context of § 4B1.2(b) because under that definition there would be no logical limit to the conduct that would come under the controlled substance offense umbrella. Logically, laws criminalizing many different types of conduct, even conduct not involving drugs, prevent or hinder the manufacture or distribution of drugs. We find the Fourth Circuit's articulation of this problem persuasive. In United States v. Campbell , the Fourth Circuit rejected the government's argument that "prohibit" could mean "to prevent or hinder." 22 F.4th 438, 448 (4th Cir. 2022) (alteration adopted) (internal quotation marks omitted). The court explained that "[i]nterpreting ‘prohibits’ to include anything that makes the outlawed conduct more likely to occur would sweep into criminal statutes a vast swath of conduct based on a secondary dictionary definition." Id. And this would result in "controlled substance offense" including such unrelated conduct as "money laundering" and "loitering" because § 4B1.2(b) would have "no logical endpoint." Id. at 448–49. We agree with the Fourth Circuit that the "to prevent or hinder" definition of prohibit does not work in the context of § 4B1.2(b), so it cannot render § 4B1.2(b) ambiguous.

Adopting the broader definition of "prohibit" would lead to results that not only are unworkable, but also would contravene Supreme Court precedent. There can be no doubt that laws forbidding the simple possession of controlled substances prevent or hinder the controlled substance offense of possession with intent to distribute those substances. Under this broader definition of "prohibit," simple possession would meet the definition of a controlled substance offense. But that cannot be right, because the Supreme Court has held the opposite—that a conviction for simple possession does not qualify as controlled substance offense. Salinas v. United States , 547 U.S. 188, 188, 126 S.Ct. 1675, 164 L.Ed.2d 364 (2006). "Prohibit," then, must have the narrower, ordinary meaning it carries in the criminal law context: "to forbid by law."

We conclude that the text of § 4B1.2(b) unambiguously excludes inchoate crimes. Under Kisor , that concludes our analysis, and we have no need to consider, much less defer to, the commentary in Application Note 1. C. Dupree Is Ineligible for the Career Offender Enhancement.

All that remains is to apply our holding to Dupree's case. Dupree concedes that he meets two of § 4B1.1(a)’s requirements for the career offender enhancement, so application of the enhancement turns on whether his "instant offense of conviction" is "a controlled substance offense." U.S. Sent'g Guidelines Manual § 4B1.1(a). Under today's holding, Dupree's conviction for conspiracy to possess with intent to distribute heroin and cocaine in violation of § 846 is not a controlled substance offense because the plain text of § 4B1.2(b) unambiguously excludes inchoate crimes. Dupree must be resentenced without application of the career offender enhancement.

IV. CONCLUSION

The definition of "controlled substance offense" in § 4B1.2(b) of the Sentencing Guidelines does not include inchoate offenses like conspiracy and attempt. To the extent that this holding conflicts with our prior precedent, that precedent is overruled. The district court erred by sentencing Dupree as a career offender because his conspiracy conviction under 21 U.S.C. § 846 was not a controlled substance offense. We vacate Dupree's sentence and remand for resentencing consistent with this opinion.

William Pryor, Chief Judge, concurring:

I join the majority opinion in full because it correctly explains the effect that the decision in Kisor v. Wilkie , ––– U.S. ––––, 139 S. Ct. 2400, 204 L.Ed.2d 841 (2019), has on our precedent United States v. Weir , 51 F.3d 1031 (11th Cir. 1995). I write separately to bring attention to a common misconception about the United States Sentencing Commission that has arisen in debates over the relationship between Kisor and Stinson v. United States , 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993). Courts and commentators tend to justify treating commentary as less authoritative than the guidelines in part on the ground that "[u]nlike the Guidelines themselves, ... commentary to the Guidelines never passes through the gauntlets of congressional review or notice and comment." United States v. Havis , 927 F.3d 382, 386 (6th Cir. 2019) (en banc). But their premise is mistaken. Unlike most agency interpretive rules, Guidelines commentary ordinarily goes through the same notice-and-comment and congressional review procedures as substantive guideline revisions. I agree with the majority that we are bound by Stinson to treat the Guidelines and commentary differently despite this similarity. But in the light of our decision today and others like it, the Commission could shore up the authority of its commentary without substantially modifying its practice by moving what normally goes in the commentary to the main text of the Guidelines in future revisions.

The Stinson Court held that the Sentencing Guidelines commentary should "be treated as an agency's interpretation of its own legislative rule." 508 U.S. at 44, 113 S.Ct. 1913. Administrative agencies usually issue their "legislative rules" through a notice-and-comment procedure but need not use that procedure for issuing interpretive rules. See 5 U.S.C. § 553(b) ; id. § 553(b)(A). Likewise, the Sentencing Commission must follow the notice-and-comment procedure before amending the Guidelines and must present the amendments to Congress for review. See 28 U.S.C. § 994(p), (x). But there is no similar statutory requirement for commentary on the Guidelines: the Commission can modify the commentary without the procedural safeguards Congress requires for guideline changes.

The conventional wisdom reasons that, in the same way that an agency cannot modify a legislative rule without notice and comment by adopting an unreasonable "interpretation" of an unambiguous existing rule, Kisor , 139 S. Ct. at 2415, the Commission cannot dodge the notice-and-comment and congressional review safeguards by creating unreasonable "commentary" on its own unambiguous guidelines. See, e.g. , Havis , 927 F.3d at 386–87. Indeed, concerns about an end-run around notice and comment have led critics to question whether any deference to agency interpretation is appropriate. See, e.g. , Kisor , 139 S. Ct. at 2434 (Gorsuch, J., concurring in the judgment). But the conventional wisdom misses that, in practice, the Commission ordinarily uses the same procedure to revise the commentary as it does to revise the Guidelines.

The application note at issue is an apt example. The commentary that specifies that section 4B1.2 applies to inchoate controlled-substance offenses was present in a different form in the original manual that went through notice and comment and submission to Congress, and its current form was adopted by an amendment that also included guideline revisions and was sent to Congress for review. See United States Sentencing Guidelines Manual § 4B1.2 cmt. n.2 (Oct. 1987); Amendments to the Sentencing Guidelines for United States Courts, 54 Fed. Reg. 21348, 21379 (May 17, 1989).

Indeed, "most amended guideline commentary now undergoes notice and comment and submission to Congress." John S. Acton, Note, The Future of Judicial Deference to the Commentary of the United States Sentencing Guidelines , 45 HARV. J. L. & PUB. POL'Y 349, 357 (2022). To be sure, the Commission's rules of procedure and the underlying statutes do not require that commentary revisions undergo the same process as guideline revisions. See 28 U.S.C. § 994(x) ; U.S. SENT'G COMM'N, RULES OF PRACTICE & PROCEDURE 4.3 (2016). But, except for technical edits, the Commission ordinarily sends its commentary revisions through the same process as its guideline revisions. See Acton, supra , at 358 (citing recent examples).

In the light of this practice, the difference between the Guidelines and the commentary ordinarily boils down to labels and formatting. So, our holding today that Kisor requires courts to give less deference to the Guidelines commentary need not constrain the Commission as much as it may appear. If, for example, the commentary's inclusion of inchoate crimes were added—through the same process the Commission earlier employed—to the definition provided in the main text of section 4B1.2, we would be bound to respect it. In the meantime, we are bound by Stinson ’s holding that they are of different levels of authority.

As an administrative agency within the judicial branch, the United States Sentencing Commission has a "unique composition and responsibilities." Mistretta v. United States , 488 U.S. 361, 384, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989). And unsurprisingly, it does not always operate in the same manner as an executive agency. Federal judges, above all officials, should not assume that it does. Judicial decisions should respect how the Commission, in fact, performs its work lest we create needless uncertainty about the important duty of sentencing criminals.

Grant, Circuit Judge, concurring in the judgment:

The majority is certainly correct that a conspiracy crime cannot qualify as a "controlled substance offense" under § 4B1.2(b). The commentary defining that term to include inchoate offenses is simply inconsistent with the guideline text. But we are still bound by Stinson , and the majority needlessly abandons that decision to get to today's result.

I understand why our Court and others have thought it necessary to at least consider whether Stinson ’s deferential posture to the Guidelines commentary still holds after Kisor . But in answering that question, we should not—cannot—rewrite the precedents to better match our view of first principles or even to create a more coherent body of law. Instead, "we must apply Supreme Court precedent neither narrowly nor liberally—only faithfully." United States v. Johnson , 921 F.3d 991, 1001 (11th Cir. 2019) (en banc). If Stinson is to be overruled, it is up to the Supreme Court to do it. I respectfully concur only in the judgment.

I.

Our duty to faithfully apply precedent continues even when (some of) the reasoning for an old Supreme Court decision has been undermined by a new case. See Rodriguez de Quijas v. Shearson/Am. Express, Inc. , 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989). We leave to that Court "the prerogative of overruling its own decisions." Id. So those decisions remain binding precedent until the Supreme Court sees "fit to reconsider them, regardless of whether subsequent cases have raised doubts about their continuing vitality." Hohn v. United States , 524 U.S. 236, 252–53, 118 S.Ct. 1969, 141 L.Ed.2d 242 (1998). Here, the controlling precedent is Stinson v. United States , 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993). And as a matter of vertical stare decisis, we are bound to follow that case unless and until the Supreme Court instructs us to do otherwise.

Stinson ’s approach to deference was firm. It held, in reversing this Court's more searching approach, that the Guidelines commentary is "authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline." Stinson , 508 U.S. at 38, 113 S.Ct. 1913. The commentary, Stinson informs us, "provides concrete guidance as to how even unambiguous guidelines are to be applied in practice"—even when that application "may not be compelled by the guideline text." Id. at 44, 47, 113 S.Ct. 1913 (emphasis added); see also United States v. Cingari , 952 F.3d 1301, 1308 (11th Cir. 2020).

The majority departs from Stinson . It does so in an attempt to conform that opinion to Kisor v. Wilkie , which either tightened or clarified (depending on whom you ask) the deference required for an administrative agency's interpretation of its own regulation. ––– U.S. ––––, 139 S. Ct. 2400, 204 L.Ed.2d 841 (2019). For decades, the Supreme Court had required courts to give "controlling weight" to an agency's interpretation of its own regulations unless the interpretation was "plainly erroneous or inconsistent with the regulation." Bowles v. Seminole Rock & Sand Co. , 325 U.S. 410, 414, 65 S.Ct. 1215, 89 L.Ed. 1700 (1945) ; Auer v. Robbins , 519 U.S. 452, 461, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997). But Kisor cautioned that this approach might be overly "reflexive" and imposed the threshold requirement that the majority now imports into Stinson : a court may not defer to an agency's interpretation of its own legislative rule unless that rule remains genuinely ambiguous after the court has exhausted all the traditional tools of construction. See 139 S. Ct. at 2415–18. On its own terms, this move has been both celebrated as necessary and denigrated as insufficient. But no matter what one thinks of Kisor , the majority extends rather than follows that precedent by applying it to Stinson and the Sentencing Guidelines.

Two theories could justify this shift. One is that we, as a Court, have always gotten Stinson wrong, and that Kisor opened our eyes to our longstanding misreading of that case. The other is that Kisor partially overruled Stinson , changing the deference owed to the Sentencing Guidelines commentary. Though it is not entirely clear which of these theories the majority adopts, neither stands up to scrutiny.

First, a fair reading of Stinson shows that we have not been misinterpreting the opinion: it does not require—or even allow—an ambiguity analysis before consulting the commentary. To the contrary, Stinson explicitly rejected a search for ambiguity; it found an analogy to Chevron "inapposite" because the commentary, "unlike a legislative rule, is not the product of delegated authority for rulemaking, which of course must yield to the clear meaning of a statute." Stinson , 508 U.S. at 44, 113 S.Ct. 1913 (citing Chevron, U.S.A. Inc. v. Natural Resources Defense Council, Inc. , 467 U.S. 837, 843 n.9, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) ). The role of the Guidelines commentary is different. It "explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice." Id. (emphasis added). So Stinson considered and expressly rejected the key limitation on deference that Kisor requires. That leaves the possibility that Kisor overruled Stinson —or at the very least overruled Stinson ’s broad deference to the commentary. The majority initially acknowledges that this did not happen: "The Supreme Court did not overrule Stinson in Kisor , and Kisor did not concern the Sentencing Guidelines." Maj. Op. at 1275. But as the opinion goes on, it effectively disregards its earlier observation, arguing that Stinson ’s embrace of deference must be understood as a "caricature" of Seminole Rock that cannot survive Kisor . Maj. Op. at 1275 (quoting Kisor , 139 S. Ct. at 2415 ).

The key error that drives this analysis is treating Stinson and Seminole Rock as if they are interchangeable. They are not. Stinson ’s doctrine was unique because the Sentencing Commission is unique. To start, it is not a traditional administrative agency—the Commission is "judicial in nature" while "the role of other federal agencies is typically executive." United States v. Moses , 23 F.4th 347, 355 (4th Cir. 2022). And the Commission promulgates guidelines, not ordinary administrative rules. That distinction is fundamental. Administrative rules exist "not just to inform and guide but also to regulate the broad range of people covered by the particular agency's jurisdiction." Id. The Guidelines, on the other hand, guide courts—they "do not bind or regulate the primary conduct of the public." Mistretta v. United States , 488 U.S. 361, 396, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989).

Stinson respected and even highlighted this distinction. The Supreme Court acknowledged that the text of the Sentencing Reform Act explicitly referenced the commentary to the Guidelines and determined that it was "binding on the courts," just like the Guidelines. Stinson , 508 U.S. at 41–42, 113 S.Ct. 1913 (citing 18 U.S.C. § 3553(b) ). Only after offering those background principles did it consider "[d]ifferent analogies" that had been "suggested as helpful characterizations of the legal force of commentary." Id. at 43, 113 S.Ct. 1913. And only after rejecting analogies to both legislative history and Chevron did the Court cite to Seminole Rock . Id. at 44–45, 113 S.Ct. 1913. Even then, when comparing the commentary to an agency's interpretation of its own regulation, the Court stressed that the analogy was "not precise." Id. at 44, 113 S.Ct. 1913. One reason? Because "Congress has a role in promulgating the guidelines"—specifically a six-month statutory period during which that body can review any changes to the Guidelines before they take effect. Id. ; Id. at 41, 113 S.Ct. 1913 (citing 28 U.S.C. § 994(p) ). And though Stinson did not mention it, the Sentencing Commission has unique appointment requirements, including a minimum number of judges, political diversity, and so on. See 28 U.S.C. § 991. Perhaps in response to the Commission's unique mission and makeup, Stinson went beyond ordinary principles of administrative law. For one, it allowed the Commission to effectively overrule judicial interpretations of the Guidelines: "prior judicial constructions of a particular guideline cannot prevent the Commission from adopting a conflicting interpretation." Stinson , 508 U.S. at 46, 113 S.Ct. 1913. In fact, the Court explicitly endorsed using amendments to the commentary to effect a change in interpretation of the overall Guidelines—the Court reasoned that " ‘Congress necessarily contemplated that the Commission would periodically review the work of the courts, and would make whatever clarifying revisions to the Guidelines conflicting judicial decisions might suggest.’ " Id. (quoting Braxton v. United States , 500 U.S. 344, 348, 111 S.Ct. 1854, 114 L.Ed.2d 385 (1991) ). So Stinson embraced a world in which the Commission (not the judiciary) would authoritatively resolve disputes about the Guidelines and the commentary—and could even reject judicial interpretations of the Guidelines by amending the commentary.

That is not how the Supreme Court treats ordinary administrative agencies. The Court has never held that an agency's reinterpretation of its own legislative rule can override a past judicial construction of that rule. Cf. Kisor , 139 S. Ct. at 2433 (Gorsuch, J., concurring in the judgment) (discussing lower court holdings to that effect). So Stinson went well beyond ordinary principles of administrative law in at least one area. And follow-on cases applying Stinson discuss that case as if it were distinct from Seminole Rock and Auer . See Neal v. United States , 516 U.S. 284, 293, 116 S.Ct. 763, 133 L.Ed.2d 709 (1996) ; United States v. LaBonte , 520 U.S. 751, 757, 117 S.Ct. 1673, 137 L.Ed.2d 1001 (1997) ; see also John S. Acton, Note, The Future of Judicial Deference to the Commentary of the United States Sentencing Guidelines , 45 Harv. J.L. & Pub. Pol'y 349, 377–79 (2022). This further suggests that Stinson hovers outside the world of ordinary administrative law.

The contrary evidence is rather thin. The majority emphasizes that Stinson adopted the government's suggestion that courts should treat the commentary "as an agency's interpretation of its own legislative rule," and claims that today's extension of Kisor honors that command. Maj. Op. at 1275-76, 1275 n.5; Stinson , 508 U.S. at 44, 113 S.Ct. 1913. But that "command" was not without context. The qualifications within the Stinson opinion demonstrate that the Court did not make an unadorned, one-for-one comparison. And even if it had, the Stinson Court could not have anticipated (and would not have adopted, given the other content in the opinion) the pre-conditions for deference that would later follow. Nothing about Stinson suggests that courts should forever index the guidelines to any developments in administrative law, even when such developments contradict Stinson ’s own reasoning.

One last point: the footnote in Kisor that includes Stinson in a list of sixteen pre- Auer cases "applying Seminole Rock " does not shift the analysis. Maj. Op. at 1276; Kisor , 139 S. Ct. at 2411 n.3 (plurality opinion). Any suggestion that it does amounts to an argument by implication from a string cite in a footnote joined by only a plurality of the Court. We can expect more from the Supreme Court if it intends to overrule one of its own cases.

In short—both in Stinson itself and in later cases—the Supreme Court has consistently treated Stinson ’s holding as something distinct from Seminole Rock deference. And because this doctrine is distinct from Seminole Rock , it is not altered by Kisor ’s "gloss." Maj. Op. at 1275.

Where does that leave us? Vertical stare decisis requires that we continue treating the commentary's interpretation of the Guidelines as "authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline." Stinson , 508 U.S. at 38, 113 S.Ct. 1913. Even if the guideline is unambiguous. I respectfully disagree with the majority's choice to either rewrite or overrule Stinson .

II.

My concern is not one of only principle. The shift in this Circuit's approach to the commentary will significantly alter the district courts’ near-constant application of the Sentencing Guidelines. I also fear that it may unsettle much of our caselaw and lead to inappropriate sentencing disparities—frustrating one of the Guidelines’ key purposes.

To be sure, there are serious arguments both from first principles and from policy for extending Kisor to Stinson —or even for overruling Stinson entirely. I do not mean to stake out a position on what the Supreme Court should do or will do if it ultimately revisits the question of Stinson ’s scope. But a change to Stinson deference will be disruptive—perhaps extremely so. And that sort of disruption should be weighed by the Supreme Court as part of its horizontal stare decisis analysis, not invited by our own rejection of vertical stare decisis.

At a bare minimum, courts in this Circuit will now need to perform a Kisor inquiry into whether a guideline's text is ambiguous before considering any commentary that interprets the guideline. Under Stinson , commentary could help ensure that courts agreed about a guideline's meaning. Now, under Dupree , courts in this Circuit will inevitably divide over whether they are allowed to consult the commentary at all.

By way of example, consider Application Note 14(B), commentary to U.S.S.G. § 2K2.1. The guideline applies a four-level enhancement whenever the defendant "used or possessed any firearm or ammunition in connection with another felony offense; or possessed or transferred any firearm or ammunition with knowledge, intent, or reason to believe that it would be used or possessed in connection with another felony offense." U.S.S.G. § 2K2.1(b)(6)(B). The commentary interprets this enhancement to apply for drug trafficking offenses where "a firearm is found in close proximity to drugs, drug-manufacturing materials, or drug paraphernalia." Id. § 2K2.1 cmt. n.14(B). So it explains, or perhaps broadens somewhat, the meaning of "used or possessed." We have considered this commentary and its guideline side-by-side without any concern. See, e.g. , United States v. Martinez , 964 F.3d 1329, 1335 (11th Cir. 2020).

But does it warrant Kisor deference? One of our sister circuits has already splintered on this question. A divided panel in the Third Circuit (which has extended Kisor to the Sentencing Guidelines) decided after several pages of Kisor analysis that deference was appropriate because the guideline was ambiguous. United States v. Perez , 5 F.4th 390, 395–99 (3d Cir. 2021). Another judge disagreed, saying that the commentary went beyond the guideline's "zone of ambiguity" and was "invalid as written" under Kisor . Id. at 404, 402 (Bibas, J., concurring in the judgment) (quotation omitted). Under the majority's new approach, we (and the district courts before us) will now be compelled to wrestle with similar questions for a host of guidelines.

Of course, disagreement between jurists is not an evil in and of itself—sometimes the system invites it. But in enacting the Guidelines, Congress sought to "avoid excessive sentencing disparities." Peugh v. United States , 569 U.S. 530, 536, 133 S.Ct. 2072, 186 L.Ed.2d 84 (2013) (quotation omitted); see also 28 U.S.C. § 991(b)(1)(B). Now, similarly situated defendants may receive substantially different sentences just because courts cannot agree whether a guideline's text is still sufficiently ambiguous after applying the traditional tools of construction. These disagreements will directly undermine the Guidelines’ purposes.

Relatedly, today's holding runs the risk of forcing a full-scale disruption of our Sentencing Guidelines caselaw. Virtually every case that has applied the commentary could be considered presumptively overruled. Thankfully, I do not read the majority to require this approach—though it will be interesting to see whether and how we can avoid it. If the Supreme Court overrules Stinson and tells us to reassess all caselaw applying the commentary, then so be it. But the possibility of such sweeping consequences from today's opinion should prompt us to tread carefully before making that call for ourselves.

In sum, under the majority's new approach, meaningful disagreements about the legitimacy of previously uncontroversial commentary are inevitable. Those disagreements will, in turn, exacerbate the degree of sentencing discrepancies. They will also increase the burden on this Court to provide definitive constructions of the Guidelines. And this whole process will have the end effect of shifting the ultimate responsibility for sentencing policy from the Commission to the courts. That shift undermines Congress's intent, articulated by the Supreme Court, that these disagreements should generally be resolved by the Commission.

III.

Of course, today's case requires more than establishing the standard for interpreting the commentary; we also need to apply that standard. And though Stinson is deferential, that deference is not absolute. If "commentary and the guideline it interprets are inconsistent in that following one will result in violating the dictates of the other," the guideline must prevail. Stinson , 508 U.S. at 43, 113 S.Ct. 1913. After all, "the Sentencing Reform Act itself commands compliance with the guideline." Id.

This is a rare case of true incompatibility between commentary and its underlying guideline. Section 4B1.2 already defines the terms used in U.S.S.G. § 4B1.1. And the definition it provides is comprehensive and clear: " ‘controlled substance offense’ means an offense under federal or state law ... that prohibits the manufacture, import, export, distribution, or dispensing of a controlled substance (or a counterfeit substance) or the possession of a controlled substance (or a counterfeit substance) with intent to manufacture, import, export, distribute, or dispense." U.S.S.G. § 4B1.2(b) (emphasis added).

The commentary expands this definition to include inchoate offenses like attempt and conspiracy. U.S.S.G. § 4B1.2(b) cmt. 1. Though Stinson requires that we give the commentary the benefit of the doubt and seek to reconcile it with the guideline, I do not see how this interpretation can stand. The majority does a good job explaining why this piece of commentary flouts its guideline. Maj. Op. at 1277-79. I agree with that analysis and will not repeat it here.

I will, however, add some additional discussion on the government's failed attempt to save this commentary with creative dictionary use. It offers that inchoate offenses can fit into the guideline's text if we read "prohibit" as if it actually said "hinder." The first level of support is said to come from Black's Law Dictionary which defines "prohibit" to mean: "1. To forbid by law. 2. To prevent, preclude, or severely hinder." Black's Law Dictionary (11th ed. 2019). And according to the government, inchoate offenses like attempt and conspiracy "hinder," even if they do not outright ban, the underlying conduct covered by the guideline.

That is unsupportable. At the very first step, a sleight-of-hand is necessary to transform "prevent, preclude, or severely hinder" into "hinder." "Severely hinder"—particularly when following "prevent" and "preclude"—suggests effectively banning something, not merely impeding it. But the government elides this crucial limitation. What's more, a reader cannot simply pick whatever definition she wishes from a dictionary. "Most common English words have a number of dictionary definitions, some of them quite abstruse and rarely intended. One should assume the contextually appropriate ordinary meaning unless there is reason to think otherwise." Antonin Scalia & Bryan A. Garner, Reading Law 70 (2012). I personally cannot think of any context where "prohibit" naturally means "hinder" (nor does the government provide one). But the fit here is particularly awkward. As every lawyer and citizen knows, criminal law is not suggestive—it either bans conduct or it allows it. In this of all contexts, we should expect the word "prohibit" to have its most ordinary meaning—banning something.

The government's argument is also something of a linguistic puzzle. It does not really argue that inchoate controlled substance offenses themselves "hinder" the underlying conduct. For good reason—can we really say that conspiring to commit a drug crime hinders the commission of that drug crime? Instead, the government seems to argue that the prosecution of these offenses hinders drug possession. To be sure, the government's prosecution of an offense "hinders" its underlying conduct by making that conduct harder to perform in practice. But the very text of a law "prohibits" the underlying conduct. And the guideline doesn't address the secondary effects of prosecution—it only looks at what an offense "prohibits." In effect, the government asks us to add language shifting the guideline's focus to the decision to prosecute (but only for inchoate offenses).

These efforts to harmonize this guideline and its commentary fall short. The government's reading is not just worse than Dupree's—it is plainly erroneous. We have no choice but to hold that the guideline and its commentary are flatly incompatible, and that the enhancement does not apply to Dupree.

* * *

Because I would reach the same result as the majority while continuing to apply Stinson deference, I concur only in the judgment.

Luck, Circuit Judge, joined by Branch, Circuit Judge, dissenting:

Section 4B1.2(b) of the sentencing guidelines defines "[t]he term ‘controlled substance offense’ " as "an offense under federal or state law, punishable by imprisonment for a term exceeding one year, that prohibits the manufacture, import, export, distribution, or dispensing of a controlled substance (or a counterfeit substance) or the possession of a controlled substance (or a counterfeit substance) with intent to manufacture, import, export, distribute, or dispense." U.S.S.G. § 4B1.2(b). The issue in this case is whether conspiring to possess heroin and cocaine with the intent to distribute them is a "controlled substance offense" under guideline section 4B1.2(b).

For thirty years, the answer was yes. See United States v. Weir , 51 F.3d 1031, 1031 (11th Cir. 1995) ("We hold that a conviction of conspiracy to possess with intent to distribute marijuana is a ‘controlled substance offense’ for purposes of career criminal sentence enhancement under section 4B1.1 of the United States Sentencing Guidelines."). The guideline commentary provided that " ‘controlled substance offense’ include[d] the offenses of aiding and abetting, conspiring, and attempting to" possess controlled substances with the intent to distribute them. U.S.S.G. § 4B1.2(b) n.1. And, under Stinson v. United States , 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993), we owed deference to the commentary as an authoritative and " ‘binding interpretation’ of the term ‘controlled substance offense’ " because the commentary neither ran "afoul of the Constitution" or "a federal statute," nor was "it inconsistent with, or a plainly erroneous reading of," the guidelines. United States v. Smith , 54 F.3d 690, 693 (11th Cir. 1995) (applying Stinson to the commentary in guideline section 4B1.2 ).

But, today, the majority opinion answers no. Placing our court with the minority of circuit courts, the majority opinion holds that we must ignore the guideline commentary and finds that conspiring to possess heroin and cocaine with the intent to distribute is not a "controlled substance offense."

The majority opinion reaches this result, and overrules thirty years of precedent, because, it says, Kisor v. Wilkie , ––– U.S. ––––, 139 S. Ct. 2400, 204 L.Ed.2d 841 (2019) clarified Bowles v. Seminole Rock & Sand Co. , 325 U.S. 410, 65 S.Ct. 1215, 89 L.Ed. 1700 (1945) and Auer v. Robbins , 519 U.S. 452, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997) and the Kisor clarification applies to Stinson and the guidelines commentary. Kisor , the majority opinion explains, clarified that commentary is not authoritative, and we do not defer to it, unless the guideline it interprets is genuinely ambiguous.

I respectfully dissent for two reasons. First, despite what the majority opinion says it is doing, it is not really applying Kisor ’s clarification to Stinson . Under the majority opinion's approach, the Kisor clarification applies to Stinson the same way a magnifying glass applies to an ant on a sunny day—total annihilation. The majority opinion is actually applying Kisor to overrule Stinson . But the Supreme Court didn't overrule Stinson and we can't overrule a Supreme Court opinion on our own. Only the Supreme Court can do that. Second, even if the majority opinion isn't overruling Stinson , the Kisor clarification doesn't apply to Stinson .

The Majority Opinion Applies Kisor to Abrogate Stinson

The majority opinion says that Kisor clarified Auer deference and that Kisor ’s clarification applies to Stinson and the guidelines commentary. Maj. Op. at 1275, 1276. But, by applying Kisor to Stinson , the majority opinion has essentially overruled Stinson .

Stinson ’s holding couldn't be clearer. It is in the second sentence of the opinion: "We decide that commentary in the [g]uidelines [m]anual that interprets or explains a guideline is authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline." Stinson , 508 U.S. at 38, 113 S.Ct. 1913. Before Stinson , "various [c]ourts of [a]ppeals ha[d] taken conflicting positions on the authoritative weight to be accorded to the commentary to the [s]entencing [g]uidelines." Id. at 40, 113 S.Ct. 1913. Our court, for example, had held "that commentary to the [g]uidelines, though ‘persuasive,’ [was] of only ‘limited authority’ and not ‘binding’ on the federal courts." Id. at 39, 113 S.Ct. 1913 (quoting United States v. Stinson , 957 F.2d 813, 815 (11th Cir. 1992) ). But the Supreme Court held that our "conclusion that the commentary now being considered [was] not binding on the courts was error." Id. at 42, 113 S.Ct. 1913.

Kisor , by contrast, applies to the deference owed "to agencies’ reasonable readings of genuinely ambiguous regulations." 139 S. Ct. at 2408. There, the Supreme Court "reinforce[d] the limits of Auer deference," explaining that "a court should not afford Auer deference unless the regulation is genuinely ambiguous." Id. at 2415, 2423. "If uncertainty does not exist," the Supreme Court explained, "there is no plausible reason for deference" and "[t]he regulation ... just means what it means." Id. at 2415.

The majority opinion reads Kisor to apply to Stinson and the guidelines commentary. The courts, the majority opinion says, must not afford deference to the commentary unless the guideline it explains or interprets is genuinely ambiguous. If, after applying all of the interpretive tools in the toolbox, uncertainty does not exist, then there is no plausible reason for deference to the commentary.

The majority opinion's new deference standard for guidelines commentary overturns and sets aside Stinson in two important ways. First, the default position under Stinson was that the commentary was authoritative and the presumption was over-come only if the interpretation or explanation violated the law or the guideline would not bear the construction. See Stinson , 508 U.S. at 38, 113 S.Ct. 1913 (holding that a comment "that interprets or explains a guideline is authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline" (emphasis added)). The majority opinion's application of Kisor to the commentary flips the presumption of authoritativeness on its head. Under the majority opinion, the commentary is not authoritative—it is owed no deference—unless the guideline is genuinely ambiguous. Ambiguity must exist "[f]irst and foremost" before the courts even consider whether to defer to the commentary. Kisor , 139 S. Ct. at 2415.

Second, the majority opinion's application of Kisor to the guidelines commentary overrules and sets aside the standard in Stinson for when deference is owed. Under Stinson , the courts defer to the commentary unless the commentary violates the law or its reading of the guidelines is plainly erroneous or inconsistent. 508 U.S. at 38, 113 S.Ct. 1913. In other words, under Stinson , guidelines commentary is binding on the federal courts "if the guideline which the commentary interprets will bear the construction." Id. at 46, 113 S.Ct. 1913.

But the majority opinion's application of Kisor to the commentary sets aside the Stinson inconsistency standard. Under the majority opinion's application of Kisor to the commentary, we no longer ask whether the guideline will bear the commentary's construction. Instead, we must ask if the guideline is genuinely ambiguous after applying the tools in the legal toolbox. If it's not, then, the majority opinion says, we have no need to consider, much less defer to, the commentary.

Critically, the majority opinion adopts the same ambiguity standard for deference to the commentary that the Supreme Court explicitly rejected in Stinson . As the Stinson Court explained, "[d]ifferent analogies have been suggested as helpful characterizations of the legal force of commentary." Id. at 43, 113 S.Ct. 1913. One of the suggested analogies the Supreme Court considered was Chevron deference. "Under Chevron, U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)," the Court continued, "if a statute is unambiguous the statute governs; if, however, Congress’ silence or ambiguity has left a gap for the agency to fill, courts must defer to the agency's interpretation so long as it is a permissible construction of the statute." Stinson , 508 U.S. at 44, 113 S.Ct. 1913 (quotations omitted). The Stinson Court found the analogy to Chevron "inapposite" because "commentary explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice." Id. (emphasis added).

In other words, Stinson deference applies to the commentary even if the guideline it explains is unambiguous. Yet, the majority opinion applies Kisor to adopt the exact opposite deference standard for the commentary—the guideline must be ambiguous before the courts can even consider giving deference. Adopting the standard rejected by Stinson is overturning Stinson .

The majority asserts that " Stinson did not outright reject Chevron deference" but only rejected "the ‘analogy’ between an agency's interpretation of a statute and the commentary's interpretation of a [g]uideline." Maj. Op. at 1277 n.7. But the Stinson Court explicitly noted that, under Chevron , courts must defer to an agency's permissible interpretation of a statute where "Congress’ silence or ambiguity has left a gap for the agency to fill." Stinson , 508 U.S. at 44, 113 S.Ct. 1913 (quotations omitted). The Court rejected this test because commentary (unlike agency regulations interpreting statutes) can "provide[ ] concrete guidance as to how even unambiguous guidelines are to be applied in practice." Id. (emphasis added). The Supreme Court rejected the Chevron test. That's the same test the majority adopts here today. After applying Kisor to the guidelines commentary, the majority opinion has demolished Stinson ’s holding. The majority opinion no longer presumes that the commentary is authoritative. And the majority opinion no longer defers to the commentary if the guideline it interprets will bear the construction. The majority opinion has overturned and set aside Stinson without using the o-word or a-word. But an abrogation by any other name is still an abrogation.

The majority opinion avoids using the o- or a-word because, as an inferior court, we cannot overrule or abrogate a Supreme Court decision. Only the Supreme Court can abrogate or overrule one of its decisions—"it is [that] Court's prerogative alone to overrule one of its precedents." State Oil Co. v. Khan , 522 U.S. 3, 20, 118 S.Ct. 275, 139 L.Ed.2d 199 (1997) ; see also Bosse v. Oklahoma , 580 U.S. 1, 137 S. Ct. 1, 2, 196 L.Ed.2d 1 (2016) (same); United States v. Hatter , 532 U.S. 557, 567, 121 S.Ct. 1782, 149 L.Ed.2d 820 (2001) (same).

But, as the Kisor Court noted, it didn't overrule anything. The "only question presented" in Kisor was "whether" the Supreme Court "should overrule" Auer or Seminole Rock , "discarding the deference they give to agencies." Kisor , 139 S. Ct. at 2408. The Court answered "no." Id. And, as Chief Justice Roberts explained, the Supreme Court "took [ Kisor ] to consider whether to overrule Auer and Seminole Rock . For the reasons the Court discusses ... I agree that overruling those precedents is not warranted." Id. at 2424 (Roberts, C.J., concurring in part).

The majority insists that it is not overturning Stinson but "honor[ing] Stinson ’s instruction to ‘treat[ ]’ the commentary ‘as an agency's interpretation of its own legislative rule.’ " Maj. Op. at 1276 (quoting Stinson , 508 U.S. at 44, 113 S.Ct. 1913 ). But the majority is abandoning the test Stinson required and adopting the test Stinson rejected. We are bound by the Supreme Court's holdings until it tells us otherwise.

Because the Supreme Court didn't overrule Stinson in Kisor , the Stinson deference standard to the commentary remains good law. And so do our cases— Weir and Smith —applying Stinson deference to the definition of "controlled substance offense."

The Kisor Clarification Does Not Apply to Stinson or the Commentary

Even putting aside the improper overruling of Stinson , the majority opinion's Kisor "clarification" does not apply to Stinson and the guidelines commentary. None of the three reasons the majority opinion gives for why the Kisor "clarification" applies to Stinson and the commentary is persuasive.

First, the majority opinion says that the Kisor clarification applies to Stinson because there's no indication that Kisor distinguished between agency rules and the commentary to the guidelines. But, as the majority opinion concedes, " Kisor did not concern the [s]entencing [g]uidelines." Maj. Op. at 1275. And Kisor never mentioned the commentary or the sentencing guidelines. The Supreme Court doesn't upend decades of precedent through silence. Cf. United States v. Florida , 938 F.3d 1221, 1226 n.4 (11th Cir. 2019) ("We do not consider the Supreme Court's silence on an issue that was not presented dispositive.").

Even if we could imply something from silence, the Supreme Court has not been silent about the distinctions between the guidelines commentary and agency rules. In Stinson , the Court explained that the "[c]ommentary ... has a function different from an agency's legislative rule. Commentary, unlike a legislative rule, is not the product of delegated authority for rulemaking, which of course must yield to the clear meaning of a statute." Stinson , 508 U.S. at 44, 113 S.Ct. 1913. "Rather, commentary explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice." Id.

And the Supreme Court has not been silent about the distinctions between the sentencing commission and administrative agencies. "The [s]entencing [c]ommission," the Supreme Court has explained, "unquestionably is a peculiar institution within the framework of our [g]overnment" with "an unusual hybrid in structure and authority." Mistretta v. United States , 488 U.S. 361, 384, 412, 109 S.Ct. 647, 102 L.Ed.2d 714 (1989) ; see also id. at 425, 109 S.Ct. 647 (Scalia, J., dissenting) ("[H]ere we have an anomaly beyond equal: an independent agency exercising governmental power on behalf of a [b]ranch where all governmental power is supposed to be exercised personally by the judges of courts." (footnote omitted)). Unlike with agency rules, "the sentencing function long has been a peculiarly shared responsibility among the [b]ranches of [g]overnment and has never been thought of as the exclusive constitutional province of any one [b]ranch." Id. at 390, 109 S.Ct. 647 (majority opinion). Under the Sentence Reform Act, "Congress’ decision to place the [c]ommission within the [j]udicial [b]ranch reflected Congress’ strong feeling that sentencing has been and should remain primarily a judicial function." Id. (quotation omitted).

Second, the majority opinion says that Kisor recognized that the commentary should be treated as an agency's interpretation of its own legislative rules by "includ[ing] Stinson alongside administrative agency cases in a list of ‘(pre- Auer ) decisions applying Seminole Rock deference.’ " Maj. Op. at 1276. The majority opinion quotes from footnote three in Kisor . But footnote three is in the part of Kisor that failed to get a majority. Chief Justice Roberts—the deciding vote in Kisor —did not join footnote three or anything else in that section of the opinion. And Stinson and the guidelines are not mentioned in the parts of Kisor that Chief Justice Roberts did join. The footnote in Stinson —which couldn't garner a majority—cannot support the majority's position here.

Third, the majority opinion uses the transitive property to find that the Kisor clarification applies to Stinson . " Stinson adopted Seminole Rock ’s formulation of agency deference," the majority opinion explains, "[s]o it follows that Kisor ’s clarification of Auer deference applies to the [g]uidelines and its commentary." Id. at 1276. In other words, because X relied on Y, and Y has been clarified by Z, then X must also have been clarified by Z.

But the Supreme Court has rejected this kind of transitive reasoning to alter its decisions. For example, in Evans v. Gore, 253 U.S. 245, 40 S.Ct. 550, 64 L.Ed. 887 (1920), "Judge Walter Evans challenged Congress’ authority to include sitting federal judges within the scope of a federal income tax law that the Sixteenth Amendment had authorized a few years earlier." Hatter , 532 U.S. at 566, 121 S.Ct. 1782. The Supreme Court agreed with Judge Evans, holding "that the Compensation Clause barred application of the tax to Evans, who had been appointed a judge before Congress enacted the tax." Id.

"A few years later," in Miles v. Graham , 268 U.S. 501, 45 S.Ct. 601, 69 L.Ed. 1067 (1925), "the Court extended Evans , making clear that its rationale covered not only judges appointed before Congress enacted a tax but also judges whose appointments took place after the tax had become law." Hatter , 532 U.S. at 566, 121 S.Ct. 1782. But "[f]ourteen years after deciding Miles, th[e Supreme] Court overruled" it in O'Malley v. Woodrough, 307 U.S. 277, 59 S.Ct. 838, 83 L.Ed. 1289 (1939), although the "Court did not expressly overrule Evans itself." Hatter , 532 U.S. at 566–67, 121 S.Ct. 1782.

The Federal Circuit confronted the continued viability of Evans in Hatter . In 1965, when Medicare was created, Congress "required most American workers (whom Social Security covered) to pay an additional Medicare tax." Id. at 561, 121 S.Ct. 1782. But Congress "did not require [f]ederal [g]overnment employees (whom Social Security did not cover) to pay that tax." Id. "In 1982, Congress ... extended both Medicare eligibility and Medicare taxes to all currently employed federal employees as well as to all newly hired federal employees." Id. The extension "meant that (as of January 1, 1983) all federal judges, like all other federal employees and most other citizens, would have to contribute between 1.30% and 1.45% of their federal salaries to Medicare's hospital insurance system." Id. at 562, 121 S.Ct. 1782.

"[E]ight federal judges, all appointed before 1983, sued the [g]overnment for ‘compensation’ in the United States Claims Court. They argued that the 1983 law, in requiring them to pay Social Security taxes, violated the Compensation Clause." Id. at 564, 121 S.Ct. 1782. The Federal Circuit agreed, holding that the Compensation Clause "prevent[ed] the Government from collecting certain Medicare and Social Security taxes from a small number of federal judges who held office nearly 20 years ago—before Congress extended the taxes to federal employees in the early 1980's." Id. at 560–61, 121 S.Ct. 1782. The Federal Circuit explained that the Supreme Court had not "expressly overrule[d] Evans ," and, "if changes in judicial doctrine had significantly undermined Evans ’ holding, th[e Supreme] Court itself would have overruled the case." Id. at 567, 121 S.Ct. 1782 (quotations omitted). Because the eight judges appointed before the 1983 law were "like Evans (involving judges appointed before enactment of the tax), [and] not like O'Malley (involving judges appointed after enactment of the tax)," the Federal Circuit "held that Evans controlled the outcome." Id.

The Supreme Court affirmed that the Federal Circuit did exactly what it was supposed to do. The Federal Circuit, the Court explained, "was correct in applying Evans to the instant case." Id. Even if the earlier precedent rested on an "increasingly wobbly, moth-eaten foundation[ ]," when it comes to "continuing respect under the doctrine of stare decisis," the "[c]ourt of [a]ppeals was correct in applying that principle" because "it is th[e Supreme] Court's prerogative alone to overrule one of its precedents." Khan , 522 U.S. at 20, 118 S.Ct. 275.

"If a precedent of th[e Supreme] Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the [c]ourt of [a]ppeals should follow the case which directly controls, leaving to th[e Supreme] Court the prerogative of overruling its own decisions." Rodriguez de Quijas v. Shearson/Am. Exp., Inc. , 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989). That is, Supreme Court "decisions remain binding precedent until [it] see[s] fit to reconsider them, regardless of whether subsequent cases have raised doubts about their continuing vitality." Hohn v. United States , 524 U.S. 236, 252–53, 118 S.Ct. 1969, 141 L.Ed.2d 242 (1998). "[W]e are not at liberty to disregard binding case law that is so closely on point and has been only weakened, rather than directly overruled, by the Supreme Court." Fla. League of Pro. Lobbyists, Inc. v. Meggs , 87 F.3d 457, 462 (11th Cir. 1996).

We must do to Stinson what the Federal Circuit did with Evans —apply it even if we have doubts about its continued vitality. Even though Kisor clarified Seminole Rock , and Stinson relied on Seminole Rock , we must follow Stinson because it directly controls the deference we owe to the guidelines commentary, regardless of Stinson resting on an increasingly wobbly, moth-eaten foundation.

There's another problem with the majority opinion's transitive precedent theory. While the Stinson Court used Seminole Rock as an analogy to describe the deference courts owe to the guidelines commentary, Stinson was careful to note that "the analogy [was] not precise because," unlike for agencies, "Congress has a role in promulgating the guidelines." 508 U.S. at 44, 113 S.Ct. 1913. That's why Stinson gave more deference to the guidelines commentary than Seminole Rock gave to an agency's interpretation of its own regulations.

Seminole Rock deference applied to an agency's interpretation of one its regulations "if the meaning of the words used [were] in doubt." 325 U.S. at 413–14, 65 S.Ct. 1215 ("Since this involves an interpretation of an administrative regulation a court must necessarily look to the administrative construction of the regulation if the meaning of the words used is in doubt."). But Stinson went further. The commentary, Stinson said, "is authoritative" and "explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice." Stinson , 508 U.S. at 38, 44, 113 S.Ct. 1913. Unlike in Seminole Rock , the meaning of the guideline didn't have to be in doubt for the commentary to be authoritative.

Also, under Seminole Rock , an agency's interpretation of its regulations was given "controlling weight unless it [was] plainly erroneous or inconsistent with the regulation." 325 U.S. at 414, 65 S.Ct. 1215. Stinson goes further with the guidelines commentary. Under Stinson , the sentencing commission could amend or revise the guidelines by amending the commentary. "Although amendments to guidelines provisions are one method of incorporating revisions, another method open to the [c]ommission is amendment of the commentary, if the guideline which the commentary interprets will bear the construction." Stinson , 508 U.S. at 46, 113 S.Ct. 1913. An agency, by contrast, can't amend its regulations through an interpretation.¹ Conclusion

Court decisions are not like statutes. They are not subject to amendment, and they are not implicitly repealed. If the Supreme Court wants to clarify its decisions, as it did with Seminole Rock and Auer in Kisor , it knows how to say which decisions are clarified. And if it wants to overrule one of its holdings, it knows how to do that too, as it did in Hatter . But, until the Court tells us that it is clarifying Stinson like it did for Seminole Rock and Auer , or that it is overruling Stinson , we are bound to apply Stinson ’s holding to the guidelines commentary.

The majority opinion "think[s] the only way to harmonize" Stinson and Kisor "is to conclude that Kisor ’s gloss on Auer and Seminole Rock applies to Stinson ." Maj. Op. at 1275. That is one way, but it is not the only way, and, more importantly, it is not faithful to how the Supreme Court has told us to read its decisions that have been undermined but not directly overruled by later cases. The other way to harmonize Stinson and Kisor is to apply Stinson to the "commentary in the [g]uidelines [m]anual that interprets or explains a guideline," 508 U.S. at 38, 113 S.Ct. 1913, and Kisor "to agencies’ reasonable readings of genuinely ambiguous regulations," 139 S. Ct. at 2408, until the Supreme Court tells us that Stinson has been overruled or clarified out of existence.

Because the Kisor clarification does not apply to Stinson , Kisor is not "irreconcilable" with Weir and "incongruous" with Smith . Maj. Op. at 1279 n.9. "[A] conviction of conspiracy to possess with intent to distribute" heroin and cocaine remains "a ‘controlled substance offense’ for purposes of career criminal sentence enhancement under section 4B1.1 of the" guidelines. Weir , 51 F.3d at 1031. I would affirm the defendant's sentence.