USA v. Michael Jerome Files, No. 19-11074 (11th Cir. 2023)
Annotate this Case
Justia Opinion Summary
Want to stay in the know about new opinions from the Eleventh Circuit US Court of Appeals?
Sign up for free summaries delivered directly to your inbox. Learn More ›
You already receive new opinion summaries from Eleventh Circuit US Court of Appeals. Did you know we offer summary newsletters for even more practice areas and jurisdictions? Explore them here.
In 2019, Defendant sought a further reduction under the First Step Act. By that time, he had completed four of the sentences—including all three marijuana-only sentences. The district court initially denied relief, holding that even Defendant’s crack-related convictions weren’t “covered offenses.” Defendant appealed, the government confessed error, and the Eleventh Circuit vacated and remanded. The district court then reduced Defendant’s sentences to time served on the eleven crack-related convictions but held that, under Denson’s interpretation of Section 404(b), it lacked authority to modify his sentences on the three non-covered powder-related offenses.
The Eleventh Circuit affirmed. The court held (1) that the Eleventh Circuit’s statement in Denson that a district court “is permitted to reduce a defendant’s sentence” under the First Step Act “only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts that are not ‘covered offenses,’” was a holding; (2) that Concepcion did not abrogate that holding; and (3) that the prior-panel-precedent rule obliges the court to follow it.
This opinion or order relates to an opinion or order originally issued on May 21, 2021.
Download PDF
USCA11 Case: 21-12859 Document: 44-1 Date Filed: 03/24/2023 Page: 1 of 35 [PUBLISH] In the United States Court of Appeals For the Eleventh Circuit ____________________ No. 21-12859 ____________________ UNITED STATES OF AMERICA, Plaintiff-Appellee, versus MICHAEL JEROME FILES, Defendant-Appellant. ____________________ Appeal from the United States District Court for the Southern District of Alabama D.C. Docket No. 2:97-cr-00099-WS-B-10 ____________________ USCA11 Case: 21-12859 2 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 2 of 35 21-12859 Before NEWSOM, LUCK, and TJOFLAT, Circuit Judges. * NEWSOM, Circuit Judge: The First Step Act of 2018 allows federal courts to reduce certain drug-related criminal sentences. In particular, § 404(b) of the Act permits “[a] court that imposed a sentence for a covered offense” to “impose a reduced sentence as if sections 2 and 3 of the Fair Sentencing Act of 2010 . . . were in effect at the time the covered offense was committed.” This case presents the following question: For what offenses may a court “impose a reduced sentence” under § 404(b)—only “covered offenses,” all offenses, or some unspecified middle-ground subset of offenses? Before we can answer that question, though, we have to decide whether this Court has already answered it in a way that binds us. In United States v. Denson, a panel of this Court said that a district court “is permitted to reduce a defendant’s sentence” under § 404(b) “only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts that are not ‘covered offenses.’” 963 F.3d 1080, 1089 (11th Cir. 2020) (citations omitted). The parties here vigorously dispute whether that statement controls our decision. Applying our prior-panel-precedent rule, we must determine whether the Denson panel’s statement was a holding and, if it was, whether the Supreme Court’s intervening * Judge Luck joins the opinion of the Court except for Section II.A.2. USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 3 of 35 3 decision in Concepcion v. United States, 142 S. Ct. 2389 (2022), abrogated it. Although the first question turns out to be somewhat more complicated than at first it may appear, we conclude (1) that Denson’s statement was a holding and (2) that Concepcion did not abrogate it—and, accordingly, that we are obliged to follow it. I A Federal law makes it illegal to sell a “controlled substance” without authorization. 21 U.S.C. § 841(a). The baseline penalties for violations are set forth in § 841(b)(1)(C). Larger quantities of drugs authorize (and sometimes require) higher penalties. Id. § 841(b)(1)(A), (B). But these quantities vary from one drug to another. Before 2010, it took a hundred times more powder cocaine than crack cocaine to trigger the increased penalties. See AntiDrug Abuse Act of 1986, Pub. L. No. 99–570, § 1302, 100 Stat. 3207, 3207-16. Public outcry about that discrepancy—and, in particular, its racially disparate impact—led Congress to pass the Fair Sentencing Act of 2010. See Dorsey v. United States, 567 U.S. 260, 268 (2012). Section 2 of that statute increased the quantity of crack cocaine required to trigger heightened penalties—but it did so only prospectively. See Fair Sentencing Act of 2010, Pub. L. No. 111220, § 2, 124 Stat. 2372. In 2018, Congress adopted the First Step Act to make these changes retroactive. See Pub. L. No. 115-391, § 404, 132 Stat. 5194, 5222 (2018). Section 404 of the Act—at issue here—comprises USCA11 Case: 21-12859 4 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 4 of 35 21-12859 three subsections. The first two are particularly relevant to what we’ll call the “Denson issue.” Section 404(b), the operative provision, allows “[a] court that imposed a sentence for a covered offense” to “impose a reduced sentence as if sections 2 and 3 of the Fair Sentencing Act of 2010 . . . were in effect at the time the covered offense was committed.” Id. § 404(b). Section 404(a), the definitional provision, explains that the term “covered offense” means “a violation of a Federal criminal statute, the statutory penalties for which were modified by section 2 or 3 of the Fair Sentencing Act of 2010.” Id. § 404(a). 1 Because their penalties were “modified by section 2 or 3 of the Fair Sentencing Act,” crack-cocaine offenses are “covered offenses”; powder-cocaine offenses, whose penalties were not changed in the Fair Sentencing Act, are not “covered offenses.” See United States v. Taylor, 982 F.3d 1295, 1299 (11th Cir. 1 The complete text of § 404’s first two subsections: (a) Definition of Covered Offense.—In this section, the term “covered offense” means a violation of a Federal criminal statute, the statutory penalties for which were modified by section 2 or 3 of the Fair Sentencing Act of 2010 (Public Law 111-220; 124 Stat. 2372), that was committed before August 3, 2010. (b) Defendants Previously Sentenced.—A court that imposed a sentence for a covered offense may, on motion of the defendant, the Director of the Bureau of Prisons, the attorney for the Government, or the court, impose a reduced sentence as if sections 2 and 3 of the Fair Sentencing Act of 2010 (Public Law 111-220; 124 Stat. 2372) were in effect at the time the covered offense was committed. USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 5 of 35 5 2020), abrogated on other grounds by Concepcion, 142 S. Ct. at 2396. Separately, § 404(c)—which will become relevant to what we’ll call the “Concepcion issue”—precludes successive motions and clarifies that relief is discretionary: Limitations.—No court shall entertain a motion made under this section to reduce a sentence if the sentence was previously imposed or previously reduced in accordance with the amendments made by sections 2 and 3 of the Fair Sentencing Act of 2010 or if a previous motion made under this section to reduce the sentence was, after the date of enactment of this Act, denied after a complete review of the motion on the merits. Nothing in this section shall be construed to require a court to reduce any sentence pursuant to this section. Id. § 404(c) (citation omitted). B In 1997, a jury convicted Michael Files of eighteen federal drug crimes involving crack cocaine, powder cocaine, and marijuana. A district judge sentenced him to life in prison on eleven counts, forty years on three counts, and twenty years or less on the other four counts—all to run concurrently. In 2017, the judge reduced Files’s sentences to reflect retroactive guidelines changes, trimming the life and forty-year sentences to a bottom-of-the-revised-guidelines-range thirty years. USCA11 Case: 21-12859 6 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 6 of 35 21-12859 In 2019, Files sought a further reduction under the First Step Act. By that time, he had completed four of the sentences—including all three marijuana-only sentences. The district court initially denied relief, holding that even Files’s crack-related convictions weren’t “covered offenses.” Files appealed, the government confessed error, and we vacated and remanded. United States v. Files, 848 F. App’x 412 (11th Cir. 2021). The district court then reduced Files’s sentences to time served on the eleven crack-related convictions but held that, under Denson’s interpretation of § 404(b), it lacked authority to modify his sentences on the three non-covered powder-related offenses. This is Files’s appeal. II To reset briefly, under § 404(b) of the First Step Act, “[a] court that imposed a sentence for a covered offense may . . . impose a reduced sentence as if sections 2 and 3 of the Fair Sentencing Act of 2010 . . . were in effect at the time the covered offense was committed.” Files contends that, read literally, § 404(b) doesn’t limit the convictions for which a federal court “may impose a reduced sentence” to those involving covered offenses—and, accordingly, that § 404(b) gave the district court here authority to reduce even his non-covered powder-related sentences. Files’s argument tees up an interesting question of statutory interpretation. But first, Denson. As we’ve already noted—and as we’ll explain in greater detail—the panel there said that a district USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 7 of 35 7 court “is permitted to reduce a defendant’s sentence” under § 404(b) “only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts that are not ‘covered offenses.’” 963 F.3d at 1089. Needless to say, if we’re bound by that statement, then we must reject Files’s position, whatever its merits. Under our prior-panel-precedent rule, an earlier panel’s holding is controlling “unless and until it is overruled or undermined to the point of abrogation by the Supreme Court or by this court sitting en banc.” United States v. Archer, 531 F.3d 1347, 1352 (11th Cir. 2008) (citations omitted). So, two questions, which we’ll address in turn: (1) Was Denson’s statement—that § 404(b) permits a court to reduce a defendant’s sentence “only on a ‘covered offense’” and not “on counts that are not ‘covered offenses’”—a holding? And if so, has it since been overruled or abrogated? A In order to determine whether Denson’s statement regarding sentencing courts’ authority under § 404(b) was a holding, we need to situate it within the context of the panel’s opinion, which was ultimately addressed to a different issue. That requires some doing. In Denson, a district court had partially denied a defendant’s First Step Act motion—reducing his sentence but by less than he had requested—and had done so “[w]ithout a hearing.” 963 F.3d at 1082. “The issue on appeal,” the panel explained, was “whether the district court [was] required to first hold a hearing at which USCA11 Case: 21-12859 8 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 8 of 35 21-12859 Denson was present.” Id. In answering that question, the panel concluded that neither the First Step Act nor the Due Process Clause required the district court to hold an in-person hearing before ruling on Denson’s sentence-modification motion. Id. Importantly here, though, the panel also said a number of other things along the way—including, again, and most notably for our purposes, that a district court “is permitted to reduce a defendant’s sentence” under § 404(b) “only on a ‘covered offense’” and not “on counts that are not ‘covered offenses.’” Id. at 1089. Our task is to determine exactly how that statement figured into the panel’s ultimate no-hearing conclusion. To that end, we provide the following detailed description of the panel’s multilayered opinion. The Denson panel determined, as an initial matter, that neither the First Step Act itself nor Federal Rule of Criminal Procedure 43 entitles a defendant to an in-person sentence-reduction hearing. Id. at 1086–87. Having cleared away that underbrush, the panel turned its attention to Denson’s “due process claim,” which it rejected on two grounds. Id. First, the panel held that there was “no due process concern” because “the right to be present under Rule 43 is at least as broad as the right under the Due Process Clause”; because Denson had no right to an in-person hearing under the Rule, he necessarily had no such right as an element of due process. Id. at 1087–88 (citations omitted). Second, and separately, the panel addressed Denson’s contention—which he premised on our decision in United States v. Brown, 879 F.3d 1231 (11th Cir. 2018)— that his sentence-reduction hearing was a “critical stage” of his USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 9 of 35 9 criminal proceeding at which due process required his presence. See Denson, 963 F.3d at 1088. In particular, Denson asked the panel to apply “two fact-intensive inquiries” from Brown to hold that his First Step Act motion constituted a “critical stage”: He was entitled to an in-person hearing, he said, because (1) the “errors” that he alleged “undermine[d his] sentence as a whole” and (2) “the sentencing court [had] exercise[d] significant discretion in modifying [his] sentence.” Id. at 1088 (citation omitted). The Denson panel rejected the defendant’s Brown-based argument on two independent bases. First, it held, flatly, that “Brown’s two-part framework,” which had been adopted for 28 U.S.C. § 2255 proceedings, “does not apply to sentencing modifications based on [18 U.S.C.] § 3582(c) motions,” including those brought under the First Step Act. 963 F.3d at 1088–89. Second— and more importantly for our purposes—the Denson panel proceeded to say the following: “Alternatively, and as an independent holding, even assuming arguendo that we should apply Brown’s fact-intensive framework, [the defendant’s] § 3582(c) sentence modification based on the First Step Act is not a critical stage under Brown’s two-part test.” Id. at 1089. In particular, the Denson panel found that neither of Brown’s two necessary preconditions was satisfied. As to Brown’s first prong, the panel concluded that Denson’s First Step Act motion “was not concerned with any ‘errors’ at his original sentencing that may or may not ‘undermine’ his sentence in its entirety.” Id. Then, having concluded that Brown didn’t USCA11 Case: 21-12859 10 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 10 of 35 21-12859 apply to First Step Act motions at all, and that even if it did its first precondition wasn’t satisfied, the Denson panel went on to address Brown’s second prong, the treatment of which is our focus here. Brown’s second factor, again, asks the following question: “‘[W]ill the sentencing court exercise significant discretion in modifying the defendant’s sentence, perhaps on questions the court was not called upon to consider at the original sentence?’” Id. at 1088 (quoting Brown, 879 F.3d at 1239–40). In holding that the court deciding Denson’s First Step Act motion hadn’t exercised “significant discretion” within the meaning of Brown, the panel emphasized that the First Step Act creates only a “limited remedy”—and in explaining why, the panel highlighted several of the First Step Act’s “limit[ations],” which we’ll regroup slightly for the sake of clarity. “[I]n ruling on a defendant’s First Step Act motion,” the Denson panel observed, “the district court”— (1) is “permitted to reduce a defendant’s sentence only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts that are not ‘covered offenses’”; (2) may make reductions “only ‘as if’ sections 2 and 3 of the Fair Sentencing Act were in effect when he committed the covered offense”; (3) may not “change the defendant’s original guidelines calculations that are unaffected by sections 2 and 3”; and USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 11 of 35 11 (4) may not reduce the defendant’s sentence “based on changes in the law beyond those mandated by sections 2 and 3.” Id. at 1089 (emphasis added) (first quoting § 404(b) and then citing United States v. Hegwood, 934 F.3d 414, 418 (5th Cir. 2019)). Based on those four “limit[ations],” the panel stated—as one part of what it called its “[a]lternative[],” “independent holding”— that the court adjudicating Denson’s First Step motion hadn’t exercised “significant discretion” within the meaning of Brown and, therefore, that “a sentencing modification under the First Step Act does not qualify as a ‘critical stage in the proceedings,’” and, therefore—back to the question presented in that case—that Denson was not entitled to an in-person hearing. Id. That is, we understand, a lot to digest. So here’s a recap of the Denson panel’s multiple, cascading due-process holdings, in outline form: Holding 1: Denson has no due-process right to a hearing on his First Step Act motion because the Due Process Clause doesn’t demand more than Rule 43— which doesn’t require a hearing. Holding 2: Denson has no due-process right to a hearing because Brown’s two-part “critical stage” test “does not apply” to First Step Act motions. Holdings 3 and 4: Denson has no due-process right to a hearing because, even assuming that the Brown USCA11 Case: 21-12859 12 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 12 of 35 21-12859 test does apply, it isn’t satisfied for two independent reasons: 3: Brown isn’t satisfied because Denson’s First Step Act motion didn’t allege “errors” that “undermine[d his] sentence.” 4: Brown isn’t satisfied because the court deciding Denson’s First Step Act motion hadn’t exercised “significant discretion.” Importantly here, Denson predicated that fourth and final holding on the ground that the Act grants only a “limited remedy”—which, in turn, it seemed to predicate, at least in part, on the ground that a reviewing court “is permitted to reduce a defendant’s sentence only on a ‘covered offense’” and not “on counts that are not ‘covered offenses.’” The question we must decide is whether that particular statement—that a district court may modify a defendant’s sentence only for a covered offense, and not for a noncovered offense—is a binding determination of law that controls our decision. 2 For the reasons that follow, we conclude that it is. For what it’s worth—not much—our unpublished decisions have divided over that question. Compare United States v. Gee, 843 F. App’x 215, 217–18 (11th Cir. 2021) (holding that Denson’s statement is a holding), with United States v. Hunt, 2022 WL 4115308, at *4–5 (11th Cir. Sept. 9, 2022) (two-judge majority holding that it’s dictum over a single-judge concurrence arguing that it’s a holding). 2 USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 13 of 35 13 1 At the outset, we should dispel two common misconceptions. First, the mere fact that the Denson panel called its statement a “holding” doesn’t make it a holding. As Judge Friendly has explained, “[a] judge . . . cannot transmute dictum into decision by waving a wand and uttering the word ‘hold.’” United States v. Rubin, 609 F.2d 51, 69 n.2 (2d Cir. 1979) (Friendly, J., concurring); see also Bryan A. Garner, et al., The Law of Judicial Precedent 59 (2016) (“[W]hile the court’s statement of [its own] holding is important, it doesn’t necessarily decide the matter.”). Accordingly, “[t]o the extent [an] opinion says one thing but does another, what it does is the holding of the decision.” Ingram v. Comm’r of Soc. Sec. Admin., 496 F.3d 1253, 1265 (11th Cir. 2007) (emphasis added); accord, e.g., Dantzler v. U.S. IRS, 183 F.3d 1247, 1251 (11th Cir. 1999) (“[J]udicial opinions do not make binding precedents; judicial decisions do.”). Second, and by contrast, the mere fact that the Denson panel’s key statement was delivered as part of an alternative holding doesn’t disqualify it from holding status. It is well-established in this Circuit that alternative holdings “are as binding as solitary holdings.” Bravo v. United States, 532 F.3d 1154, 1162 (11th Cir. 2008) (citations omitted). And under our precedent about precedent, the sort of reasoning employed in Denson—that a particular test doesn’t apply but that, even if it does, it isn’t satisfied—constitutes a prototypical alternative holding. Indeed, we have said— albeit (ironically) in dicta—that the alternative-holding rule applies USCA11 Case: 21-12859 14 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 14 of 35 21-12859 in precisely these circumstances. See Hitchcock v. Sec’y, Fla. Dep’t of Corr., 745 F.3d 476, 484 & n.3 (11th Cir. 2014) (issuing “dual holding[s]” that a habeas petitioner’s claim failed under AEDPA deference and, even if AEDPA were inapplicable, on de novo review). 3 Because each of Denson’s four alternative holdings is “as binding” as if it were a “solitary holding[],” Bravo, 532 F.3d at 1162, we can focus exclusively on the one most relevant to us—namely, the panel’s conclusion that Denson had no due-process right to an in-person hearing because the court deciding his First Step Act motion hadn’t exercised “significant discretion.” 2 Remember, the question that we must decide is whether the Denson panel’s key statement—again, that § 404(b) permits a district court to reduce a defendant’s sentence “only on a ‘covered offense’” and not “on counts that are not ‘covered offenses’”—is itself a holding. Files insists that it isn’t, and for support he points to the oft-invoked maxim that portions of a court’s opinion that aren’t “necessary” to its judgment are dicta, see, e.g., United States v. Gillis, 938 F.3d 1181, 1198 (11th Cir. 2019), and contends that the Denson panel’s statement wasn’t strictly necessary to its conclusion that the district court didn’t exercise “significant discretion” in ruling on Denson’s First Step Act motion. Even if that statement was absent, the argument goes—even if that particular limitation didn’t Judge Wilson disagreed that the latter ground constituted a holding. See Hitchcock, 745 F.3d at 490 & n.6 (Wilson, J., concurring). 3 USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 15 of 35 15 exist—the panel might still have concluded that the adjudication of Denson’s motion didn’t entail “significant discretion.” Because nothing in Denson itself indicates that the statement on which we’re now focused was necessary to its no-significant-discretion conclusion, Files reasons, that statement wasn’t a holding. With respect to the application of the necessary-to-the-judgment criterion, we’ll just come right out and say it: We’re in something of a grey area here. We know for certain—and Files would admit—that if all four of the characteristics that the Denson panel identified as features of First Step Act adjudications, see supra at 10–11, were necessary to its no-significant-discretion determination, then all four would constitute holdings. See United States v. Caraballo-Martinez, 866 F.3d 1233, 1244 (11th Cir. 2017). 4 But as Files points out, the Denson panel seemed to cite the four That the four characteristics pertain to what might be called “legal facts” about the First Step Act—rather than legal conclusions relevant specifically to the issue on appeal in Denson—doesn’t change matters. To take just one example from our own precedent, in discerning the elements of federal crimes, we have routinely looked to enumerations of those elements embedded in earlier double-jeopardy decisions applying the Blockburger analysis, which entails a comparison of legal facts—namely, the elements of two different offenses. See, e.g., United States v. Feldman, 931 F.3d 1245, 1257 (11th Cir. 2019) (applying, as binding, the elements of a healthcare-fraud conspiracy—18 U.S.C. §§ 1347, 1349—as specified in the Blockburger analysis conducted in United States v. Gonzalez, 834 F.3d 1206, 1220 (11th Cir. 2016). But cf. Garner, supra, at 84–86 (suggesting—without citation to authority—that some statements regarding legal facts might be mere “[a]ssumptions underlying court decisions” that are not precedential even if necessary). 4 USCA11 Case: 21-12859 16 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 16 of 35 21-12859 characteristics as illustrative or exemplary features, not as strictly necessary conditions. It never suggested, for instance, that the absence of any single characteristic—let alone the first, in particular— would have changed its decision. But—and it’s a big but—our own precedent and common sense both reveal that “necessary” doesn’t mean strictly necessary. The proof? Consider, for example, two common types of determinations—neither of which is strictly necessary to a court’s judgment, but both of which are traditionally accorded holding status. First, one we’ve already explored in some detail: alternative holdings. By definition, an alternative holding isn’t necessary to the court’s judgment—any of two or more alternatives suffices. And yet, as we’ve already explained, our precedent treats alternative holdings “as binding as solitary holdings.” Bravo, 532 F.3d at 1162; see also Garner, supra, at 122–23 (“[A]lternative holdings are still holdings, even though they aren’t logically necessary to the case’s disposition.”). Second, what we’ll call “non-supportive” holdings. Cases often present multiple issues, and courts will more than occasionally decide one question in one party’s favor but then proceed to decide another question—and enter judgment—for the other party. Qualified-immunity cases are illustrative. In those, a court will often conclude, at step one of the familiar two-step analysis, that a government official violated the Constitution but then go on to determine, at step two, that the law wasn’t “clearly established”—and, therefore, that the official is entitled to immunity. USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 17 of 35 17 See, e.g., Camreta v. Greene, 563 U.S. 692, 699–700 (2011). Both conclusions are holdings. Indeed, the Supreme Court has explicitly said that, in that circumstance, a court’s step-one conclusion that the official violated the Constitution is “[n]o mere dictum” but, rather, “creates law that governs the official’s behavior,” even if he prevails at step two. Id. at 708 (allowing an official who prevailed at the second step, and thereby obtained a favorable judgment, to appeal a lower court’s adverse determination that he violated the law). It goes without saying that a non-supportive holding of this sort isn’t necessary to the court’s judgment—because it actually contradicts the court’s judgment. And yet, it is a binding holding nonetheless. See also, e.g., United States v. Steed, 548 F.3d 961, 977 (11th Cir. 2008) (treating as a binding holding an earlier court’s determination that a jury instruction was unlawful despite the earlier court’s ultimate conclusion that the error was harmless). And to be clear, there are still other circumstances in which we depart from a strict-necessity criterion. When, for instance, one of our opinions chooses between two competing legal “tests” in the course of resolving a case, we have characterized our choice of one of them as a holding even when it’s not clear that the case would have turned out differently under the other. 5 For that matter, United States v. Kaley, 579 F.3d 1246 (11th Cir. 2009), is illustrative. There, we had to determine the holding of our earlier decision in United States v. Bissell, 866 F.2d 1343 (11th Cir. 1989). In Bissell, the government had restrained several indicted drug dealers’ bank accounts in order to preserve the funds for eventual seizure under 21 U.S.C. § 853(a). Id. at 1347. The 5 USCA11 Case: 21-12859 18 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 18 of 35 21-12859 statements of a legal rule—whether or not the result of a choice among competing alternatives—are often technically unnecessary to a case’s resolution. In many (if not most) cases, we could simply decide the dispute narrowly on its own particular facts, without separately articulating a test or standard. But no one thinks that when we do state a governing rule—as we typically do—we do so gratuitously and unnecessarily. See Michael Abramowicz & Maxwell Stearns, Defining Dicta, 57 Stan. L. Rev. 953, 984–86 (2005). Given the many exceptions and caveats, it’s easy to see why theories of precedent inextricably tied to strict, logical “necessity”— while “frequently cited” and easy to apply—are “problematic in profound ways” and, among competing theories, “prove[] the easiest to falsify.” Id. at 959, 1056. defendants couldn’t access the money, including for their legal defense, and weren’t given a pre-trial hearing to challenge the restraint. Id. at 1349–50. In determining that the failure to conduct a hearing didn’t violate the defendants’ due-process rights, the Bissell panel applied a test that the Supreme Court had adopted for speedy-trial purposes in Barker v. Wingo, 407 U.S. 514 (1972). See 866 F.2d at 1353–54. Faced with the same issue in Kaley, we concluded—over a concurring judge’s objection—that that the Bissell panel’s adoption of the Barker test was controlling. 579 F.3d at 1253 n.10. We said so because the choice “form[ed] a critical part of the [Bissell] case’s holding” and because the Bissell panel’s conclusion “was driven by, and [could not] be understood apart from[,] [its] application” of the Barker test”—and, notably, despite the fact that Bissell “could have [been] decided . . . on other grounds,” and seemingly without respect to whether that case would come out the same way had the panel adopted a different standard. See id. USCA11 Case: 21-12859 Document: 44-1 21-12859 Date Filed: 03/24/2023 Opinion of the Court Page: 19 of 35 19 But if, as our own precedent reveals, “necessary” doesn’t really mean necessary, what is the true measure of a holding? Does holding status attach, as the Ninth Circuit has posited, to any “issue germane to the eventual resolution of the case [that the court] . . . resolves . . . after reasoned consideration in a published opinion,” United States v. Johnson, 256 F.3d 895, 914 (9th Cir. 2001) (en banc) (plurality opinion), “regardless of whether it was in some technical sense ‘necessary’”? Barapind v. Enomoto, 400 F.3d 744, 751 (9th Cir. 2005) (en banc). See Charles W. Tyler, The Adjudicative Model of Precedent, 87 U. Chi. L. Rev. 1551, 1567–72 (2020) (describing the Ninth Circuit’s view). That approach, we think— while also neat and clean—sweeps too broadly. To be sure, it would comport with some of our caselaw. For example, we’ve treated as dicta—as we think even the Ninth Circuit’s broad rule would—legal conclusions predicated on facts that aren’t actually at issue, 6 as well as aside-like statements about irrelevant legal matters. 7 But the Ninth Circuit’s far-reaching germane-and-considered 6 E.g., Edwards v. Prime, Inc., 602 F.3d 1276, 1298 (11th Cir. 2010) (holding that legal conclusions about hypothetical facts are dicta); Caraballo-Martinez, 866 F.3d at 1244 (similar). Consider, for instance, our decision in United States v. Pickett, 916 F.3d 960, 966 (11th Cir. 2019), that a statement in United States v. Glover, 431 F.3d 744, 749 (11th Cir. 2005), was dictum. In Glover, a panel had considered whether a judge impermissibly found a fact that triggered enhanced penalties. Id. at 749. After concluding that the relevant determination was a question of law rather than fact, the panel had appended an observation that the legal determination was correct. Id. Because that statement wasn’t relevant to the question presented in Glover, we concluded in Pickett that it was dictum. 602 F.3d 7 USCA11 Case: 21-12859 20 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 20 of 35 21-12859 criterion would contradict our precedent in other respects, in that we have relegated to dicta status statements of law that our sister circuit would treat as holdings—for instance, statements regarding a legal framework that the court initially engages but ends up abandoning in favor of an alternative. 8 So far as we can tell, the four relevant-but-not-decisive characteristics of First Step Act proceedings that the Denson panel identified as features of a First Step Act adjudication don’t fit neatly into any of our existing “holding” or “dicta” categories. Without a ready-made, easy-to-apply metric, we are left to determine whether the Denson panel’s statements are more like those that we’ve treated as holdings or those that we’ve deemed dicta. On at 966; see also, e.g., Georgia Ass’n of Latino Elected Offs., Inc. v. Gwinnett Cnty. Bd. of Registration & Elections, 36 F.4th 1100, 1119–20 (11th Cir. 2022) (holding that a previous decision’s statement in an introductory paragraph about a statutory provision’s operation was dictum because it wasn’t applicable to the previous case’s resolution); Fresh Results, LLC v. ASF Holland, B.V., 921 F.3d 1043, 1050 (11th Cir. 2019) (characterizing as dictum a previous decision’s description of a legal standard applicable to a question that it didn’t reach). 8 See, e.g., Pretka v. Kolter City Plaza II, Inc., 608 F.3d 744, 764 (11th Cir. 2010) (treating as dicta a previous decision’s statements about common law where the case was ultimately resolved on statutory grounds); Welch v. United States, 958 F.3d 1093, 1098 (11th Cir. 2020) (treating as dictum a previous decision’s discussion of ACCA’s elements clause where the case was ultimately resolved under the residual clause). USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 21 of 35 21 balance, we think that they are more like the sorts of conclusions that we have accorded holding status. Here’s why: As already noted, if the Denson panel had said that a defendant had to prove all four characteristics in order to demonstrate that the sentencing court hadn’t exercised “significant discretion,” then its conclusions about each would have constituted traditional, necessary-to-the-result holdings. If, at the other end of the spectrum, the panel had said that any one of the four characteristics was sufficient to show an absence of “significant discretion,” then each of its conclusions would, in effect, have constituted a binding alternative holding. Between those two extremes, the same would be true if the panel had said, for instance, that any two were sufficient: In that circumstance, each would be part of an alternative holding—e.g., the first two are present or, in the alternative, the last two are; the first and third are present or, in the alternative, the second and fourth are; etc. (So too if the panel had said that any three would suffice. 9) Accordingly, the only thing separating the Denson panel’s actual statements regarding the four characteristics from binding traditional holdings (on the one hand) or binding alternative holdings (on the other) is that it didn’t specify precisely how many characteristics it took to cross the sufficiency 9 All of this assumes, of course, that the several characteristics of First Step Act adjudications that the Denson panel identified carry equal weight. They seem equally weighty to us, and nothing in Denson suggests otherwise. USCA11 Case: 21-12859 22 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 22 of 35 21-12859 threshold—or precisely which ones. That seems like a pretty thin basis on which to deny them holding status. Bottom line: Our precedent about precedent makes clear that strict necessary-ness is not essential to a statement’s holdingness. And the Denson panel’s statements regarding First Step Act adjudications were clearly significant to its no-significant-discretion conclusion—and are thus fundamentally similar to the other sorts of determinations that we have traditionally accorded holding status. Accordingly, we treat the Denson panel’s statements regarding the scope of sentencing courts’ authority under § 404(b) as holdings. 10 One last thing: Files also asserts that the Denson panel’s statement about the permissibility of reducing sentences for non-covered offenses can’t constitute a holding because it outstripped the facts of the case—because, he says, Denson didn’t have a non-covered offense. Files is right about the law: A “decision can hold nothing beyond the facts of that case.” Edwards v. Prime, Inc., 602 F.3d 1276, 1298 (11th Cir. 2010) (collecting cases). But he’s wrong about the facts—or at least about the relevant facts as the courts in Denson understood them. Files emphasizes that Denson had completed the sentence on his non-covered offense before this Court issued its decision. But because Denson’s Brown-based claim addressed a procedural issue—whether he was entitled to a hearing in the district court—what mattered, as the panel there understood and explained matters, was the fact that Denson had a non-covered offense at the time the district court ruled. See 963 F.3d at 1085; see also Denson, No. 19-11696, Doc. 141 at 5 (district court stating that “[t]he prison sentence on [the noncovered offense] remains 120 months, still to be served concurrently”). See generally United States v. Aguillard, 217 F.3d 1319, 1321 (11th Cir. 2000) (“The holdings of a prior decision can reach only as far as the facts and circumstances presented to the Court in the case which produced 10 USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 23 of 35 23 * * * In sum, then, we conclude that Denson’s determination that a district court “is permitted to reduce a defendant’s sentence” under § 404(b) “only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts that are not ‘covered offenses,’” 963 F.3d at 1089, is indeed a holding. And under our prior-panel-precedent rule, that means it binds us unless it has been overruled or “undermined to the point of abrogation” either by the Supreme Court or by one of our own en banc decisions. Archer, 531 F.3d at 1352. It is to that issue that we turn next. B Files contends that to the extent that Denson held that a district court lacks the authority to reduce a sentence for a non-covered offense, the Supreme Court’s decision in Concepcion abrogated it. We disagree. To be sure, Concepcion abrogated aspects of Denson. In particular, the Supreme Court expressly rejected the fourth of the four considerations that the Denson panel highlighted in its no-significant-discretion analysis: The Court held, contra Denson, that a district court adjudicating a First Step Act motion can consider changes in law unrelated to those specified in the Fair Sentencing Act. Compare Concepcion, 142 S. Ct. at 2396, with Denson, 963 that decision.” (emphasis added) (quoting parenthetically United States v. Hunter, 172 F.3d 1307, 1309 (11th Cir. 1999) (Carnes, J., concurring))). USCA11 Case: 21-12859 24 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 24 of 35 21-12859 F.3d at 1089. But Concepcion certainly didn’t repudiate Denson in toto; indeed, it expressly affirmed the third of the Denson panel’s four considerations—namely, that a court can’t recalculate a defendant’s guidelines ranges for non-Fair-Sentencing-Act reasons. Compare Concepcion, 142 S. Ct. at 2402 n.6, with Denson, 963 F.3d at 1089. Most importantly here, Concepcion didn’t address a sentencing court’s authority to deal with non-covered offenses one way or the other—presumably because the defendant there didn’t have one; he had only a single conviction for a covered crack-related offense. 142 S. Ct. at 2396. Accordingly, we cannot say that Concepcion “abrogated” Denson’s determination that district courts may not reduce defendants’ sentences for non-covered offenses. In arguing otherwise, Files focuses on the Supreme Court’s statements that Congress didn’t “hide any limitations on district courts’ discretion outside of § 404(c)” and that § “404(b) does not erect any additional such limitations.” Id. at 2402. But Files misunderstands the Court’s comments. The Court was referring there only to limits on how district courts exercise their “discretion” in reducing defendants’ sentences—not to their power to do so in the first place. The latter is an issue of authority, not discretion. And with respect to authority, both the Act and the Supreme Court’s opinion contemplate limits outside § 404(c). Take, for example, the very first clause of § 404(b): It clearly states that only the “court that imposed a sentence for a covered offense”—not just any USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Opinion of the Court Page: 25 of 35 25 court—can entertain a sentence-reduction motion under the Act. So too, as just explained, Concepcion itself observed that district courts can’t recalculate defendants’ guidelines ranges based on changes unrelated to those specified in the Fair Sentencing Act. See 142 S. Ct. at 2402 n.6. Those “limitations” remain effective, despite being “outside of § 404(c),” id. at 2402, because they pertain to a court’s ex ante authority to act, not the manner in which it exercises its discretion. The limitation that Denson recognized—i.e., that a court “is permitted to reduce a defendant’s sentence only on a ‘covered offense,’” 963 F.3d at 1089—is similar: It speaks to the whether, not the how. There is thus no fatal inconsistency. A district court adjudicating a First Step Act motion can—without fear of contradiction— apply both Denson’s holding limiting the categories of sentences that can be reduced and Concepcion’s holding empowering courts to exercise broad discretion in imposing reduced sentences for those qualifying offenses. Accordingly, we hold that Concepcion did not abrogate Denson’s holding that a sentencing court “is permitted to reduce a defendant’s sentence” under the First Step Act “only on a ‘covered offense’” and not “on counts that are not ‘covered offenses.’” 963 F.3d at 1089. III For all these reasons, we hold (1) that this Court’s statement in Denson that a district court “is permitted to reduce a defendant’s sentence” under the First Step Act “only on a ‘covered offense’” and “is not free . . . to change the defendant’s sentences on counts USCA11 Case: 21-12859 Document: 44-1 26 Date Filed: 03/24/2023 Opinion of the Court Page: 26 of 35 21-12859 that are not ‘covered offenses,’” 963 F.3d at 1089, was a holding; (2) that Concepcion did not abrogate that holding; and (3) that our prior-panel-precedent rule obliges us to follow it. AFFIRMED. USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 27 of 35 1 NEWSOM, Circuit Judge, joined by TJOFLAT, Circuit Judge, concurring: Reasonable minds can differ, of course, but my own view— fortified by my experience with this case—is that federal appellate courts should issue fewer alternative holdings. I’ll detail my reasons in due course. Let me begin, though, with our decision here. As the majority opinion explains, this appeal tees up an interesting question of statutory interpretation. Section 404(b) of the First Step Act states that “[a] court that imposed a sentence for a covered offense may . . . impose a reduced sentence as if sections 2 and 3 of the Fair Sentencing Act of 2010 . . . were in effect at the time the covered offense was committed.” The question: For what offenses may a court “impose a reduced sentence” within the meaning of § 404(b)? Only for “covered offense[s]”? Or, at the other end of the spectrum, perhaps for any offense, even if non-“covered”—and, for that matter, even if wholly unrelated to the conduct that underlies a covered offense? Or might there be some middle ground, in which courts can reduce sentences for offenses that are (my term) “inextricably intertwined” with a covered offense? The question regarding § 404(b)’s proper scope isn’t just interesting, it’s important. Thousands of federal prisoners were eligible for sentence reductions under the First Step Act, and so determining the extent of courts’ authority under the Act has practical, real-world significance. And, as it turns out, that issue—the meaning of § 404(b)’s “may impose a reduced sentence” clause—is USCA11 Case: 21-12859 2 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 28 of 35 21-12859 squarely presented in this case. The parties thoroughly briefed it. See Br. of Appellant at 23–44; Br. of Appellee at 27–33; Reply Br. of Appellant at 2–9. It was fully vetted at oral argument. See Oral Arg. at 14:18–23:00, 32:18–38:43. And I, for one, think that the answer is pretty clear: Understood in context, and particularly in the light of § 404(b)’s indisputable focus on “covered offense[s],” § 404(b) should be interpreted to empower courts to modify sentences only for those offenses, such that the provision reads, in effect, as follows: “A court that imposed a sentence for a covered offense may . . . impose a reduced sentence for a covered offense as if sections 2 and 3 of the Fair Sentencing Act of 2010 . . . were in effect at the time the covered offense was committed.” That’s the only reading, I submit, that makes any legal or practical sense. The obvious question, then: Why didn’t we say so? Because, as the majority opinion explains, we concluded that we didn’t need to. Having determined that this Court’s earlier decision in United States v. Denson, 963 F.3d 1080, 1089 (11th Cir. 2020), had already resolved the issue in a way that binds us, we found no cause to forge ahead to interpret the statute afresh. See Maj. Op. at 2–3. Even so, one might respond, why not strap on both belt and suspenders? Especially given the difficulty and closeness of the Denson issue—whether its statement regarding § 404(b)’s scope constitutes a binding holding, see Maj. Op. at 14– 23—why not do exactly what the Denson panel itself did and issue an “[a]lternative[] . . . independent holding” that simply “assum[es] arguendo” that Denson doesn’t control, 963 F.3d at 1089, and goes USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 29 of 35 3 on to conclude, in any event, that § 404(b) only narrowly authorizes reviewing courts to modify sentences for covered offenses? What’s the harm? Speaking only for myself, I think the harm can be very real. 1 Let me explain. Some have gone so far as to suggest that alternative holdings are unconstitutional—the premise being, in essence, that once a court has provided a single sufficient basis for resolving the dispute before it, the constitutional “Case[]” has concluded and the “judicial power” has been fully discharged. See U.S. Const. art. III, § 2. Any further statement of law, the argument goes, “present[s] potential advisory-opinion problems.” Phelps v. Alameda, 366 F.3d 722, 729 (9th Cir. 2004) (O’Scannlain, J.). My contention is more modest: At least in appellate courts, issuing alternative holdings is often just a bad idea. I say so for three reasons: (1) my own sense of the judicial role, (2) my desire to facilitate sound judicial decisionmaking, and (3) my concern for impartiality and collegiality on multimember courts. 1. The Judicial Role. Some have a very muscular view of a federal-court authority. They, to use Hart and Wechsler’s wellworn terminology, see federal courts as principally engaged in the And real harm aside, let not the irony be lost that Denson’s cascade of alternative holdings is one of the very things that made it so difficult to determine whether its statement regarding § 404(b)’s scope constituted a holding. See Maj. Op. at 7–12. 1 USCA11 Case: 21-12859 4 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 30 of 35 21-12859 “law declaration” business. Richard H. Fallon, Jr., et al., Hart and Wechsler’s The Federal Courts and The Federal System 73–74 (7th ed. 2015). Courts “have a special function,” they say, “of enforcing the rule of law” by articulating clear, precedential rulings that, taken together, create binding legal doctrine—or, as Hart and Wechsler more soaringly summarize it, “to declare and explicate norms that transcend individual controversies.” Id. at 74. As should be clear, this law-declaration role exists “independent[ly] of” the courts’ “task of resolving concrete disputes.” Id. Not me. I’m in what Hart and Wechsler would call the “dispute resolution” camp. Id. at 73. Federal courts are tasked by the Constitution—and tasked only—with adjudicating the “Cases” and “Controversies” that real parties bring before them. See U.S. Const. art. III, § 2. Without respect to whether the Constitution actually requires it to do so, once a court has fulfilled its obligation—that is, has said enough to resolve the parties’ dispute—it should just stop. It shouldn’t forge ahead, reach out, and declare more law. I realize that mine is a quaint perspective, out of step with the more robust conception of judicial power that has taken hold during the “past half century” or so. Hart & Wechsler, supra, at 74. If that makes me a throwback, so be it. I’ll happily cling to my view—which I’ve expressed elsewhere in lamenting courts’ modern tendency to decide issues unnecessarily—that federal judges should be “reactive, not proactive; passive, not aggressive; modest, USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 31 of 35 5 not bold.” United States v. Campbell, 26 F.4th 860, 895 (11th Cir. 2022) (en banc) (Newsom and Jordan, JJ., dissenting). Alternative holdings—A; and if not A, then B; and if not A or B, then C—rarely reflect “reactive,” “passive,” or “modest” decisionmaking. They’re almost always the opposite—“proactive,” “aggressive,” and “bold.” By “decid[ing] questions that do not matter to the disposition of a case,” courts are “separat[ing] Lady Justice’s scales from her sword.” Friends of Everglades v. S. Fla. Water Mgmt. Dist., 570 F.3d 1210, 1216 (11th Cir. 2009). And wielding this sword verges dangerously close, I fear, to judicial legislation. And that should give us pause. 2 2. Sound Decisonmaking. Of course, in discharging their dispute-resolution function, federal courts will declare some law along the way. “[I]ncidental to [their] responsibility to decide concrete disputes,” that is, courts will necessarily create precedent and doctrine. Hart & Wechsler, supra, at 73. Needless to say, in doing 2 Let me anticipate and attempt to respond here to an objection: “Aren’t your separate concurring opinions—like, oh, say, this one—‘proactive,’ ‘aggressive,’ and ‘bold’ in exactly the same way as the alternative holdings you criticize?” No, actually. A judge’s solo concurrence doesn’t even purport to declare law—it aims only to advance the conversation about a particular topic. In fact, it is my antipathy for alternative holdings that causes me to write separate concurrences in cases where I think they might be of some use. Rather than attempt to wedge my thoughts into a controlling opinion, I offer them, on behalf of myself only, for whatever they may (or may not) be worth. USCA11 Case: 21-12859 6 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 32 of 35 21-12859 so, they should strive to declare the law as accurately as possible. 3 My worry is that there’s likely an inverse relationship between the number of holdings a court purports to issue and the correctness of each. That’s not rocket science—or, to be fair, any kind of science. It’s just a common-sense observation that the more a court bites off, the less time and attention it has to savor and digest each constituent morsel. I’m hardly the first person to express this concern. Judge Leval has observed, for instance, that in his experience, “[c]ourts often give less careful attention to propositions uttered in support of unnecessary alternative holdings.” Pierre N. Leval, Judging Under the Constitution: Dicta About Dicta, 81 N.Y.U. L. Rev. 1249, 1258 n.23 (2006). In particular, he worries that judges joining another’s opinion are “likely to look primarily at whether the opinion fulfills their expectations as to the judgment and the reasoning given in support of it”—and, correlatively, that there is a “high likelihood that . . . alternative explanations[] and dicta will receive scant attention.” Id. at 1262. Judge Kethledge’s warning about dicta applies to alternative holdings, as well: “[J]udges think differently—more carefully, more focused, more likely to think things through—when our words bring real consequences to the parties They should also strive to declare it as clearly as possible. For advocates of the law-declaration model, this should be particularly important. Cluttering up opinion with cascading alternative holdings makes each holding harder to find and discern, see Maj. Op. at 7–12, undermining the very notice and ruleof-law values that the law-declaration model purports to advance. 3 USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 33 of 35 7 before us.” United States v. Burris, 912 F.3d 386, 410 (6th Cir. 2019) (en banc) (Kethledge, J., concurring). Judge Posner has said similar things: Passages that are not an “integral part of the earlier opinion . . . may not have been fully considered.” United States v. Crawley, 837 F.2d 291, 292 (7th Cir. 1988). The unremarkable bottom line: When everyone in the decisionmaking process focuses on a single, necessary ground for resolving a case—when our attention is trained, rather than divided—we’re more likely to arrive at an answer that is well-considered, well-explained, and, most importantly, correct. Let’s go deep, not broad. 3. Impartiality and Collegiality. Two final—and related— problems, both of which are illustrated by a story relayed to a scholar who was investigating different appellate courts’ approaches to precedent. An anonymous Ninth Circuit judge shared the following vignette: “People would ask me, ‘What’s wrong with [the Ninth Circuit’s] assignment system—Judge Reinhardt is on all the big cases?’ [And] I . . . say, . . . ‘There is nothing wrong with [our] assignment system—Judge Reinhardt makes big cases!’” Charles W. Tyler, The Adjudicative Model of Precedent, 87 U. Chi. L. Rev. 1551, 1589–90 (2020) (alterations in original) (emphasis added). That story, I think, contains important lessons about both impartiality and collegiality. First, impartiality. Impartiality is a (perhaps the) cornerstone of our judicial system. As I’ve said before, when a court reaches out to address and decide an issue—here, one unnecessary USCA11 Case: 21-12859 8 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 34 of 35 21-12859 to the resolution of the dispute before it—it increases the risk that outside observers will perceive it (even if mistakenly) to be engaged in “political action” and, accordingly, view its conduct with suspicion and cynicism. See Campbell, 26 F.4th at 896 (Newsom and Jordan, JJ., dissenting). Put simply, when a court decides more than necessary—and even more so when it ranges into tertiary and quaternary holdings—it gives the impression that it wants to make law rather than impartially apply it. Second, collegiality. In their day-to-day work, federal appellate courts sit in panels of three, comprising judges drawn on a random rotational basis to “ensure that all of the [active] judges sit on a representative cross section of the cases heard.” 28 U.S.C. § 46(b). Our prior-panel-precedent rule is designed to afford equal respect to each panel, in that no one collection of three judges can overrule the decision of any other. See United States v. Archer, 531 F.3d 1347, 1352 (11th Cir. 2008). But that rule alone can’t ensure that panels—and the judges who serve on them—share equally in determining circuit precedent. If some judges write narrow, constrained opinions while others reach out to decide as many issues as possible in successive alternative holdings, then our rule systematically preferences the views of the most aggressive. I don’t think anyone wants that. * * * USCA11 Case: 21-12859 21-12859 Document: 44-1 Date Filed: 03/24/2023 Newsom, J., Concurring Page: 35 of 35 9 For all of these reasons, I hope that we will all think, and then think, and then think again before embedding alternative holdings in our opinions. 4 As I said at the outset, I confine my critique to alternative holdings issued by appellate courts. Although any constitutional objection would apply equally to district courts, I’m willing to concede that the practical calculus there is different. First, because their opinions aren’t precedential, see Camreta v. Greene, 563 U.S. 692, 709 n.7 (2011), district courts don’t declare law in the same way that appellate courts do, and thus don’t face the same tension between the dispute-resolution and law-declaration adjudicative models. Second, because district judges don’t sit on multi-member panels, they don’t face the same collegiality issues that can arise from aggressive decisionmaking in the appellate courts. And finally, for a district court, providing redundant decisional grounds can meaningfully increase judicial efficiency by (1) minimizing the risk that a case ping-pongs between it and the appellate court and (2) facilitating the resolution of the entire case in a single appeal. See United States v. Caporale, 701 F.3d 128, 142 n.8 (4th Cir. 2012). 4
Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
