Opinion
22-5567
11-22-2024
United States of America, Plaintiff-Appellee, v. Gerald Lynn Campbell, Defendant-Appellant.
Dana C. Hansen Chavis, FEDERAL DEFENDER SERVICES OF EASTERN TENNESSEE, INC., Knoxville, Tennessee, for Appellant. Luke A. McLaurin, UNITED STATES ATTORNEY'S OFFICE, Knoxville, Tennessee, for Appellee.
Appeal from the United States District Court for the Eastern District of Tennessee at Chattanooga. No. 1:21-cr-00005-1-Charles Edward Atchley, Jr., District Judge.
ON BRIEF:
Dana C. Hansen Chavis, FEDERAL DEFENDER SERVICES OF EASTERN TENNESSEE, INC., Knoxville, Tennessee, for Appellant.
Luke A. McLaurin, UNITED STATES ATTORNEY'S OFFICE, Knoxville, Tennessee, for Appellee.
READER, J., delivered the amended opinion of the court in which GIBBONS and DAVIS, JJ., joined. DAVIS, J. (pp. 13-16), delivered a separate concurring opinion.
Before: GIBBONS, READLER, and DAVIS, Circuit Judges.
AMENDED OPINION
CHAD A. READLER, CIRCUIT JUDGE
Gerald Lynn Campbell brandished a pistol and told a group of laborers that he had a bullet for each of them. Following an indictment, Campbell pleaded guilty to being a felon in possession of a firearm. The district court sentenced him to fifteen years' imprisonment, concluding that Campbell's prior convictions for robbery and drug offenses triggered the Armed Career Criminal Act's mandatory minimum. Campbell appealed, challenging his sentence. We affirm.
I.
A group of construction workers were performing road repairs in Chattanooga, Tennessee. Gerald Lynn Campbell approached the group in his vehicle. When he arrived, he asked who had voted for Donald Trump in the recent presidential election. Seemingly unhappy with the responses he received, Campbell brandished a firearm. He told the laborers that he "had a bullet for each one of them." Campbell then drove away, eventually parking his truck not far from the construction site. When responding officers found the vehicle, they also discovered Campbell sitting in his vehicle with a semi-automatic pistol in his lap.
A grand jury indicted Campbell for being a felon in possession of a firearm, in violation of 18 U.S.C. § 922(g)(1). Campbell pleaded guilty. At the time of Campbell's offense, being a felon in possession of a firearm typically carried a maximum penalty of ten years' imprisonment. See 18 U.S.C. § 924(a)(2) (2021). The Armed Career Criminal Act, or ACCA, however, mandates a 15-year minimum sentence for a defendant with "three previous convictions" for "a violent felony or a serious drug offense," each committed on "occasions different from another." 18 U.S.C. § 924(e)(1). With respect to Campbell, the potential predicate offenses the district court considered were: (1) an August 1985 Tennessee robbery; (2) an April 1987 Tennessee aggravated assault; (3) an August 1992 Tennessee drug trafficking offense; (4) a December 1992 Virginia drug trafficking offense; and (5) a March 1993 Virginia drug trafficking offense.
Campbell objected on multiple grounds. Among them, he argued that Tennessee robbery is not a predicate offense, that the drug offenses were not committed on different "occasions," and that the Fifth and Sixth Amendments require that the different-occasions issue be decided by a jury. Campbell also challenged one of the Eastern District of Tennessee's standard supervised release conditions. The condition states that if Campbell's probation officer determined Campbell posed a "risk" to another, the probation officer could require Campbell to notify the person about the risk. E.D. Tenn. L.R. 83.10(c)(12). Campbell objected on the grounds that the provision was unconstitutionally vague.
The district court overruled Campbell's objections. It sentenced him to 180 months' imprisonment, ACCA's mandatory minimum, relying on the conviction for Tennessee robbery and the three drug offenses as predicates. Campbell timely appealed.
II.
A. The Armed Career Criminal Act triggers an enhanced sentence for felonious possession of a firearm when a criminal defendant has three or more prior "violent felony" or "serious drug offense" convictions committed on different "occasions." 18 U.S.C. § 924(e)(1). Congress has defined "violent felony" for us. Under the statute's "elements" clause, a violent felony is a crime punishable by imprisonment for more than one year that "has as an element the use, attempted use, or threatened use of physical force against the person of another." Id. § 924(e)(2)(B)(i). We utilize the so-called categorical approach to determine whether an offense satisfies that clause. United States v. Smith, 70 F.4th 348, 354 (6th Cir. 2023). Under that approach, we ask whether the generic offense "necessarily" involves such force. See id. The offense also must have a mens rea greater than recklessness. United States v. White, 58 F.4th 889, 895 (6th Cir. 2023) (citing Borden v. United States, 141 S.Ct. 1817, 1825 (2021) (plurality opinion)). We review challenges to whether an offense is a violent felony de novo. United States v. Buie, 960 F.3d 767, 770 (6th Cir 2020).
Before the district court, the government's position was that Campbell's aggravated assault conviction was categorically a violent felony, satisfying the ACCA predicate offense requirements. The district court did not reach the issue, however, and the government does not ask us to do so now. That leaves the robbery and drug offenses as potential predicate crimes.
1. We begin with Tennessee robbery. Tennessee defines "robbery" as "the intentional or knowing theft of property from the person of another by violence or putting the person in fear." Tenn. Code Ann. § 39-13-401(a). We have repeatedly considered that offense categorically a violent felony. That is so, we have explained, because Tennessee robbery requires the use or threat of physical force with the requisite knowing or purposeful mens rea. United States v. Mitchell, 743 F.3d 1054, 1060-61 (6th Cir. 2014); see also Southers, 866 F.3d at 367; United States v. Belcher, 40 F.4th 430, 431 (6th Cir. 2022), cert. denied, 143 S.Ct. 606 (2023).
Campbell responds with two intervening Supreme Court decisions he says require a new approach. According to Campbell, those cases-Elonis v. United States, 575 U.S. 723 (2015), and Borden v. United States, 141 S.Ct. 1817 (2021)-instruct that ACCA's definition of violent felony excludes offenses that are triggered when a defendant threatens physical force negligently. And, Campbell adds, Tennessee robbery sweeps in negligent threats of force, citing State v. Witherspoon, 648 S.W.2d 279 (Tenn. Crim. App. 1983), and Sloan v. State, 491 S.W.2d 858 (Tenn. Crim. App. 1972). But here again, our precedent forecloses Campbell's argument. See Belcher, 40 F.4th at 432 (referencing Witherspoon, 648 S.W.2d at 281, and Sloan, 491 S.W.2d at 861). Throughout the "long history" of Tennessee's robbery statute, "not once" has a Tennessee court construed the fear element to encompass instances where a defendant negligently caused the victim fear. Id. at 431-32; see also United States v. Hubbard, No. 216219, 2023 WL 319604, at *2 (6th Cir. Jan. 19, 2023); United States v. Dorsey, 91 F.4th 453, 456 (6th Cir. 2024).
Changing gears, Campbell argues that Tennessee robbery stretches beyond threats of physical force and includes threatening to falsely accuse the victim of sodomy. Although Campbell grounds his argument on a case that pre-dates the Civil War, see Britt v. State, 26 Tenn. 45 (1846), it too is a familiar one. See Hubbard, 2023 WL 319604, at *3. In Hubbard, we described Britt as holding that "fear constituting an element of the crime is a fear of present personal peril from violence offered or impending." Id. (citing Britt, 26 Tenn. at 46). Although Britt mused that robbery could include threatening to prosecute one for sodomy, no subsequent Tennessee case appears to have ever picked up the proposition. See id.
2. Campbell next turns our attention to his drug offenses. He argues that those offenses were not committed on three separate "occasions," meaning they do not amount to three ACCA predicate offenses. Relatedly, Campbell contends that a jury, and not the district court, should have been the one to make findings on the occasions question.
Some background on the evolving legal framework is necessary. The occasions element springs from the statute's text. See 18 U.S.C. § 924(e)(1) (offenses must be committed "on occasions different from one another"). A multi-factor inquiry guides the analysis. United States v. Williams, 39 F.4th 342, 350 (6th Cir. 2022) (citing Wooden v. United States, 595 U.S. 360, 367-70 (2022)). Those factors include the timing, the locations, the character, and the relationship of the offenses. Id. (citing Wooden, 595 U.S. at 369-70). Substantial time and distance gaps will generally indicate different offense occasions. Wooden, 595 U.S. at 369. On the other hand, offenses with a "common scheme" are more likely to be the same occasion. Id. In many instances, however, the analysis can be limited to inquiries like whether the offenses were committed a "day or more apart" or at "significant" distances, factors that can singularly decide the question. See id. at 370 (citations omitted).
Our historical approach to the "occasions" question has now been informed by Erlinger v. United States, 144 S.Ct. 1840 (2024). When Campbell raised his argument at sentencing, circuit precedent allowed the district court to decide whether prior offenses were committed on different occasions. Williams, 39 F.4th at 351; Belcher, 40 F.4th at 432. The Supreme Court has recently said otherwise. Earlier this year, it held that the Fifth and Sixth Amendments require a unanimous jury to determine whether a defendant's past offenses were committed on separate occasions. Erlinger, 144 S.Ct. at 1851-52. In so doing, Erlinger viewed the question as being "nearly on all fours" with a pair of Supreme Court precedents-Apprendi v. New Jersey, 530 U.S. 466 (2000), and Alleyne v. United States, 570 U.S. 99 (2013)-which collectively bar a judge from increasing the "prescribed range of penalties to which a criminal defendant is exposed" under a preponderance of the evidence standard. Erlinger, 144 S.Ct. at 1851-52 (quoting Apprendi, 530 U.S. at 490). With all parties now acknowledging that Erlinger settles whether the district court violated the principles of Apprendi and Alleyne in sentencing Campbell, we turn to the appropriate remedy for that underlying constitutional error.
a. In so doing, we do not write on a blank slate. Start with the understanding that a constitutional error alone does not entitle a defendant to an automatic reversal. Arizona v. Fulminante, 499 U.S. 279, 306 (1991). Indeed, most constitutional errors are subject to a harmless error analysis, leaving us to ask whether the government has made it clear "beyond a reasonable doubt that the outcome would not have been different" without the constitutional violation. United States v. Mack, 729 F.3d 594, 608 (6th Cir. 2013). Errors arising from violations of Apprendi and Alleyne are not exceptions to this general rule. "Failure to submit a sentencing factor to the jury" that enhanced the defendant's final sentence "is not structural error" that "requir[es] automatic reversal." Washington v. Recuenco, 548 U.S. 212, 218, 222 (2006); see also Neder v. United States, 527 U.S. 1, 18 (1999) ("[I]nfringe[ments] upon the jury's factfinding role [that] affect the jury's deliberative process" are "subject to harmless-error analysis."); Mack, 729 F.3d at 609. That reality informs our analysis here. Again, as the constitutional violation identified in Erlinger-a sentencing enhancement resulting from judicial findings concerning ACCA's occasions clause-is part and parcel with the errors in Apprendi and Alleyne, we likewise ask whether the error at issue in Campbell's case was harmless. Erlinger, 144 S.Ct. at 1852; Recuenco, 548 U.S. at 222; see also Erlinger, 144 S.Ct. at 1860-61 (Roberts, C.J., concurring) (recognizing harmless error applies in this setting); id. at 1866-67 (Kavanaugh, J., dissenting) (same).
Campbell, we note, thinks otherwise. As he sees things, Erlinger error is structural in nature, a breakdown so grave that it rendered his underlying proceedings "fundamentally unfair or an unreliable vehicle for determining guilt or innocence." Neder, 527 U.S. at 9. Yet Campbell does not point to anything unique about judicial factfinding on the different-occasions question to distinguish it from any other Apprendi or Alleyne error. Instead, he analogizes a jury failing to consider the factual predicate behind the ACCA enhancement as tantamount to a judge wholly abridging a criminal defendant's right to a jury trial. See Rose v. Clark, 478 U.S. 570, 578 (1986) ("[H]armless-error analysis presumably would not apply if a court directed a verdict for the prosecution in a criminal trial by jury."). But in categorizing what amounts to structural error, the Supreme Court has twice rejected the notion that any missing piece of the legal puzzle before the jury demands a new trial. Instead, it utilizes a harmless error analysis, even when, it bears emphasizing, a jury does not consider an entire element of an offense or a sentencing enhancement altogether. See Recuenco, 548 U.S. at 220-22; Neder, 527 U.S. at 10. So it is at odds with Supreme Court precedent to equate the failure to allow a jury to conduct fact finding as to a particular sentencing enhancement as on par with the wholesale denial of a jury trial right. See Owens v. Parris, 932 F.3d 456, 460 (6th Cir. 2019) (noting that in the habeas context, "the question as to harmlessness is whether the jury would have made the necessary finding had the jury been asked to make it.").
Even so, says Campbell, as he pleaded guilty, engaging in harmless error would just invite speculation about how a hypothetical jury would have considered the ACCA enhancement. Here too, Campbell's contention is at odds with Supreme Court precedent, this time holdings rejecting the premise that harmless error presupposes an "actual verdict of guilty-beyond-a reasonable doubt." Neder, 527 U.S. at 11-13 (emphasis in original) (rejecting "alternative reasoning" from Sullivan v. Louisiana, 508 U.S. 275 (1993), and cataloging cases applying harmless error review where the jury failed to render a "complete verdict"). Following suit, we have recognized that a harmlessness analysis applies to Apprendi errors stemming from "judicial overreach post-verdict, or guilt admission." See United States v. Belcher, 92 F.4th 643, 650 (6th Cir. 2024) (emphasis added); see, e.g., United States v. Climer, 591 Fed.Appx. 403, 410 (6th Cir. 2014) (applying harmlessness analysis to Apprendi error following a guilty plea). So too with our sister circuits, which have held that the failure to submit the occasions clause question to the jury is subject to harmless error review. See United States v. Stowell, 82 F.4th 607, 610 (8th Cir. 2023) (en banc) (applying harmless error following guilty plea); see also United States v. Rodriguez, No. 21-2544, 2022 WL 17883607, at *2 (7th Cir. Dec. 23, 2022) (order) (same); United States v. Man, No. 21-10241, 2022 WL 17260489, at *2 (9th Cir. Nov. 29, 2022) (same). This practice has generally persisted, even in Erlinger's wake. See United States v. Johnson, 114 F.4th 913, 917 (7th Cir. 2024) (reviewing Erlinger error for harmlessness after guilty plea); United States v. Saunders, No. 23-6735, 2024 WL 4533359, at *2 (2d Cir. Oct. 21, 2024) (same); United States v. Robinson, No. 23-3438, 2024 WL 4448849, at *1 (8th Cir. Oct. 8, 2024) (per curiam) (reviewing Erlinger error for harmlessness after jury trial); United States v. Trahan, 111 F.4th 185, 198 (1st Cir. 2024) (reviewing Alleyne challenge for harmlessness while favorably citing calls from four Justices to apply such review to Erlinger errors). But see United States v. Billings, No. 22-4311, 2024 WL 3633571, at *6 (4th Cir. Aug. 2, 2024) (deeming Erlinger error non-harmless without explanation).
Finally, Campbell reads Erlinger as implicitly viewing the underlying error at issue there as structural. But Campbell's arguments-which turn on purported implications from the Erlinger majority's choice of citation and closing valediction as well as the composition of Justices joining the various separate writings in that case-are purely speculative. We are hard pressed to read between the lines in the way Campbell would have us do. Our concern is the precedent itself, and we take heed in Erlinger's statement that it "decide[d] no more than" the underlying Apprendi question. Erlinger, 144 S.Ct. at 1852. Accordingly, we remain bound to view Apprendi-Alleyne errors as subject to harmless error analysis.
b. Turn, then, to the question of harmlessness. We can "confidently say," based on the "whole record," that the government has shown that the constitutional error at issue here was "harmless beyond a reasonable doubt." See United States v. Henderson, 626 F.3d 326, 333 (6th Cir. 2010) (quoting Delaware v. Van Arsdall, 475 U.S. 673, 681 (1986)).
Given Campbell's armed robbery conviction, we start his predicate offense count at one. This means that the Erlinger error was harmless if at least two of Campbell's three past drug-related offenses occurred on different occasions. On that note, consider the undisputed facts that the jury would have heard. The offenses at issue are a Chattanooga drug trafficking offense in August 1992, a Norfolk drug trafficking offense in December 1992, and another Norfolk drug trafficking offense from March 1993. Months separate each of the offenses-far more than the "day" of separation the Supreme Court used as a benchmark in Wooden. 595 U.S. at 370; see also Williams, 39 F.4th at 350 (robberies separated by "at least six days" were separate occasions). And the first two offenses are likewise remote as a matter of proximity: eight hundred miles as the bird flies separated the Tennessee offense from those in Virginia. See Wooden, 595 U.S. at 369. In fact, Campbell's counsel admitted as much at sentencing. There, he recognized that Campbell committed the ACCA drug predicates on three different dates in geographically distinct locations. See Belcher, 40 F.4th at 432 (rejecting defendant's similar argument in part because defendant admitted that at least three of his prior offenses were separated by at least six years). And, by and large, Campbell's predicates were for wholly distinct crimes-possession with intent to distribute crack cocaine and trafficking in powder cocaine. Taking heed of the Supreme Court's instruction on the meaning of the term "occasion," it becomes evident that overwhelming evidence showed Campbell committed his predicate offenses "on occasions different from one another." Wooden, 595 U.S. at 367 (defining "occasion" as a single "event, occurrence, happening, or episode"); see Erlinger, 144 S.Ct. at 1856 (recognizing that typically, a "defendant's past offenses will be different enough and separated by enough time and space that there is little question he committed them on separate occasions"). As a result, the record evidence shows beyond a reasonable doubt that a jury's failure to consider the different-occasions question had no effect on Campbell's sentence.
Campbell offers little in response. He characterizes the offenses as sharing a common "purpose" (trafficking drugs) because the Tennessee federal court considered the Virginia drug trafficking offenses as "relevant conduct" during Campbell's sentencing. But the Guidelines' relevant conduct analysis differs from Wooden's § 924 "occasions" test. Compare U.S. Sent'g Guidelines Manual § 1B1.3 cmt. n.5(B)(ii) (U.S. Sent'g Comm'n 2015) (defining relevant conduct to include "ongoing series of offenses"), with Wooden, 595 U.S. at 367 (defining "occasion" as a single "event, occurrence, happening, or episode"). Campbell's view-that all offenses with the purpose of "trafficking" drugs occur on the same occasion-would fail to honor the statute's text. See 18 U.S.C. § 924(e)(1) (indicating that "three" "drug offense[s]" can be three ACCA predicates).
Campbell also critiques the government's use of Shepard documents to show harmlessness, a measure he views as inherently flawed. As Campbell sees things, Shepard documents can be "prone to error" because they are potentially dated or untested at the time of issuance. True, these flaws are a reason why Erlinger determined that the occasions inquiry must be submitted to a jury. 144 S.Ct. at 1855. But Erlinger did not preclude the use of Shepard documents in reviewing an error for harmlessness. Instead, harmless error review is based on an assessment of all "relevant and reliable information" in the "entire record," which in turn is gauged through "case-by-case adjudication rather than through a categorical bar against" certain categories of evidence. Greer v. United States, 593 U.S. 503, 510-11 (2021) (considering the substantial-rights prong of plain-error review). And consideration of the entire record is not limited to admissible evidence introduced at a trial. It can include, for example, "information contained in a pre-sentence report." Id. at 511; see also United States v. Stewart, 306 F.3d 295, 324 (considering presentence report in harmless error review of Apprendi error after jury trial); United States v. Zidell, 323 F.3d 412, 434 (6th Cir. 2003) (same); United States v. Stubbs, 97 Fed.Appx. 564, 566 (6th Cir. 2004) (considering plea agreement and presentence report in harmless error review of Apprendi error after guilty plea); United States v. Evans, 568 Fed.Appx. 368, 369-70 (6th Cir. 2014) (same for Alleyne error after guilty plea). Turning to that full record, Campbell offers nothing to undermine the reliability of any of the evidence utilized by the government here, including the relevant Shepard documents. Nor could he, it seems. The record evidence as to the different occasions question is powerful, especially considering Campbell admitted at sentencing that his underlying crimes took place at entirely different times and locations.
As a last resort, Campbell points to cases where a jury acquitted a defendant on the different occasions question, using that as proof that he too could have been acquitted on the issue. But this argument, like his broadside attack on Shepard documents, is "simply another form of the argument that [an Erlinger error] is not subject to harmless-error analysis." Neder, 527 U.S. at 17. After all, if only a theoretical possibility of acquittal were sufficient to defeat demonstrated harmless error, harmless error could not exist. Rather, our focus is on Campbell's case and whether any error in his sentencing was harmless. See Greer, 593 U.S. at 510-11. Given the undisputed record here, we can comfortably say that any constitutional error was harmless, a conclusion that likewise settles his substantive challenge to whether his crimes occurred on different occasions.
B.1. That leaves Campbell's challenge to one of the Eastern District of Tennessee's standard supervised release conditions. We review the imposition of the condition for an abuse of discretion. United States v. Widmer, 785 F.3d 200, 203 (6th Cir. 2015).
First, some history on the condition in question. District courts have discretion to impose supervised release conditions pursuant to 18 U.S.C. § 3563(b)(22). The United States
Sentencing Commission promulgates recommended "standard" conditions. See U.S. Sent'g Guidelines Manual § 5D1.3(c) (U.S. Sent'g Comm'n 2016); 28 U.S.C. § 994(a)(2)(B). One is the risk notification condition, which has seen a transformation. The Commission's prior version stated that a probation officer could direct a defendant to notify "third parties of risks that may be occasioned by the defendant's criminal record or personal history or characteristics." See United States v. Thompson, 777 F.3d 368, 379 (7th Cir. 2015). Finding "personal history" and "characteristics" impermissibly vague, a few circuits (including ours) directed district courts to modify the condition. See id. at 379, 382; United States v. Evans, 883 F.3d 1154, 1162-64 (9th Cir. 2018); United States v. Sexton, 719 Fed.Appx. 483, 484-85 (6th Cir. 2017) (per curiam). In response, the Commission updated the language to the current text, which the Eastern District of Tennessee adopted. See U.S. Sent'g Guidelines Manual supp. to app. C, amend. 803 (U.S. Sent'g Comm'n 2016); id. § 5D1.3(c)(12); E.D. Tenn. L.R. 83.10(c)(12).
The risk notification condition now instructs that if a probation officer determines that a defendant poses "a risk" to "another person," the probation officer may require the defendant to notify the person about the risk, and the defendant "shall," U.S. Sent'g Guidelines Manual § 5D1.3(c)(12), or "must," E.D. Tenn. L.R. 83.10(c)(12), comply. This standard condition is not uncommon in the district courts. See United States v. Cruz, 49 F.4th 646, 652-53 (1st Cir. 2022) (collecting cases). According to Campbell, however, the new language is still impermissibly vague such that it violates the Due Process Clause. A prior panel of our Court has upheld the condition's new language, albeit in an unpublished opinion. See United States v. Marcum, No. 20-5280, 2021 WL 1038272, at *4 (6th Cir. Jan. 20, 2021) (per curiam) (order). Other circuits have rejected similar challenges. See Cruz, 49 F.4th at 653; United States v. Janis, 995 F.3d 647, 653 (8th Cir. 2021); United States v. Gibson, 998 F.3d 415, 423-24 (9th Cir. 2021); United States v. Hull, 893 F.3d 1221, 1223 (10th Cir. 2018). But see United States v. Boles, 914 F.3d 95, 111-12 (2d Cir. 2019) (holding that the condition was impermissibly vague). We uphold the language again today.
True, as Campbell notes, the Due Process Clause forbids vague supervised release conditions. United States v. Shultz, 733 F.3d 616, 622 (6th Cir. 2013). At the same time, a condition need not "spell out every one of its applications." Id. Instead, it must "fix an ascertainable standard of guilt." Id. (cleaned up). That is the case here. The prior version of the condition was fatally flawed because the district court provided no clarification regarding what was meant by "risks," "personal history," or "characteristics." Sexton, 719 Fed.Appx. at 484-85. But the latter two categories have been removed from the condition. See E.D. Tenn. L.R. 83.10(c)(12). And the district court clarified the scope of "risk" in the challenged condition; the "risk" referred to Campbell's "past criminal history"-a definition that clears the "ascertainable" bar. See Marcum, 2021 WL 1038272, at *4; see also Gibson, 998 F.3d at 423 (rejecting vagueness challenge because "risk" was cabined to those posed by the "defendant's criminal record" (emphasis omitted)).
2. Campbell has one last salvo. He argues that the condition was an unconstitutional delegation of judicial authority. Because Campbell did not object before the district court, we review the issue for plain error. See Fed. R. Crim. P. 52(b). That requires Campbell to show (1) error, (2) that was obvious or clear, (3) that affected his substantial rights, and (4) affected the fairness, integrity, or public reputation of his judicial proceedings. United States v. Johns, 65 F.4th 891, 893 (6th Cir. 2023). He falters from the start.
Federal law affords probation officers extensive authority. See 18 U.S.C. § 3603; United States v. Logins, 503 Fed.Appx. 345, 349 (6th Cir. 2012). That authority, however, has its limits. Article III draws a line in the sand: imposing punishment upon a defendant convicted of a crime, a core judicial function, cannot be delegated. See Whitehead v. United States, 155 F.2d 460, 462 (6th Cir. 1946); United States v. Henderson, 29 F.4th 273, 276 (5th Cir. 2022); see also Ex parte United States, 242 U.S. 27, 41-42 (1916). Here, the record indicates that the court reserved the "ultimate authority" to determine the condition's contours. See United States v. Thompson, 653 F.3d 688, 693 (8th Cir. 2011); Cruz, 49 F.4th at 654; Janis, 995 F.3d at 653; see also United States v. Kidd, No. 12-5420, 2013 WL 870263, at *2 (6th Cir. Mar. 11, 2013) (per curiam) (citation omitted). In particular, the court explained that if there were "any questions or concerns," the court would "take [the issue] up" to "resolve" it. See Cruz, 49 F.4th at 654 (rejecting a delegation challenge to an identical condition); Janis, 995 F.3d at 653 (same); Henderson, 29 F.4th at 276 (same). That was enough to satisfy any delegation concerns under our precedent.
We affirm the district court's judgment.
CONCURRENCE
DAVIS, Circuit Judge, concurring.
I agree with and join the majority's opinion. I write separately, however, to address the potential conundrum occasioned by the use of Shepard documents as part of the evaluation of the district court's different-occasions inquiry. The Supreme Court, in Erlinger, held that the Fifth and Sixth Amendments require that a jury, not a judge, conduct the ACCA's different-occasions inquiry. Erlinger v. United States, 602 U.S. 821, 835 (2024). And as observed in the majority opinion, Erlinger's logic springs directly from Apprendi and Alleyne, so precedent supports our application of harmless error review to the district court's undertaking of the different-occasions inquiry. See Neder v. United States, 527 U.S. 1, 15 (1999); Washington v. Recueno, 548 U.S. 212, 222 (2006); United States v. Stewart, 306 F.3d 295, 322-23 (6th Cir. 2002); United States v. Mack, 729 F.3d 594, 608 (6th Cir. 2013).
Applying harmless error review here, we ask whether the government has shown that the Erlinger error was harmless beyond a reasonable doubt-i.e., that "a jury's failure to consider the different-occasions question had no effect on Campbell's sentence." Majority Op. at 8, 9. To answer this question, the majority looked to the Court's decision in United States v. Greer, 593 U.S. 503 (2021), which permits reviewing courts to look to the whole record. Majority Op. at 9. Specifically, the Court observed that, when analyzing the substantial rights prong of plain error review, an appellate court "can examine relevant and reliable information from the entire record"; if a defendant believes certain information is irrelevant or unreliable, that contention should be addressed "through case-by-case adjudication rather than through a categorical bar against considering evidence outside the trial record." Greer, 593 U.S. at 511; see also Majority Op. at 9.
We would be wise, however, to view Greer's permissive approach in the light cast by the Erlinger majority's strong admonition against the use of Shepard documents to satisfy the different occasions requirement of the ACCA.
As a refresher, in Shepard v. United States, the Court held that only a limited swath of documents could be used to determine whether a defendant's prior offense of conviction categorically matched a generic offense and could serve as the basis for applying the ACCA's career-offender enhancement. 544 U.S. 13, 26 (2005). Those documents include a defendant's admissions, "judicial records, plea agreements, and colloquies between a judge and the defendant" and have come to be known as Shepard documents. Erlinger, 602 U.S. at 839; see also Shepard, 544 U.S. at 26.
Apprendi and its progeny established and reiterated that a jury must determine beyond a reasonable doubt any fact(s), "other than the fact of a prior conviction," that will increase a defendant's punishment. Apprendi v. New Jersey, 530 U.S. 466, 490 (2000). And in service of this rule, Almendarez-Torres established a limited exception concerning the evidence the government may use to satisfy its burden: a judge may use Shepard documents solely to find "the fact of a prior conviction." Alleyne v. United States, 570 U.S. 99, 111 n.1 (2013) (citing Almendarez-Torres v. United States, 523 U.S. 224 (1998)). This exception is narrow; it prohibits a judge from going "beyond identifying the crime of conviction to explore the manner in which the defendant committed that offense." Mathis v. United States, 579 U.S. 500, 511 (2016).
Against this backdrop and citing reliability concerns associated with Shepard documents, the Erlinger majority expressed grave doubts about the propriety of using such documents to decide the different-occasions inquiry at the trial level. See Erlinger, 602 U.S. at 840-41. Erlinger's majority observed that using Shepard documents to determine that a defendant's previous convictions "occurred on at least three separate occasions" does "more than Almendarez-Torres allows." Erlinger, 602 U.S. at 839. In that regard, the majority cautioned that the details required to make an informed different-occasions inquiry under Wooden v. United States, 595 U.S. 360, 369 (2022), are beyond the limits placed on the use of such documents by Almendarez-Torres. To allow the use of Shepard documents to make that determination "would be to allow a sentencing court to do exactly what the Fifth and Sixth Amendments forbid." Erlinger, 602 U.S. at 840.
The Erlinger majority did not qualify its cautionary guidance concerning the use of potentially unreliable Shepard documents. Id. at 841. By contrast, the Greer majority found a case-by-case approach more appropriate and placed the responsibility on the defendant to raise any concerns about the reliability (or lack thereof) of evidence from the "entire record" during his plain-error substantial rights inquiry. Greer, 593 U.S. at 511. This makes sense in the context of plain error review where the burden is on the defendant. Id. at 508. The plain-error standard "is difficult" by design for a defendant to meet and understandably so; when possible, a defendant should afford the trial court the opportunity to avoid or remedy any error(s) that could affect the outcome of the proceedings. See, e.g., Puckett v. United States, 556 U.S. 129, 134 (2009). Permitting defendants an appellate remedy for every unpreserved objection to such errors could result in defendants "sandbagging" the trial court. Id. (quoting Wainwright v. Sykes, 433 U.S. 72, 89 (1977)). Harmless error review, on the other hand, does not spring from any such defense-related negligence. Indeed, the government bears the burden of showing that the Erlinger error did not contribute to the challenged outcome because such errors implicate a defendant's trial right guarantee. See Reiner v. Woods, 955 F.3d 549, 555 (6th Cir. 2020); Neder, 527 U.S. at 19 (holding that harmless error review of a court's failure to instruct the jury on an element of the offense "safeguard[s] the jury guarantee").
The differences between Greer and this case give me pause in extending its logic here. Use of the whole record could compound the effect of the initial Erlinger error because of the grave reliability problems associated with the Shepard documents often used during a judge-made different-occasions inquiry. To begin, such documents often lack the information necessary to conduct the different-occasions inquiry. Erlinger, 602 U.S. at 840. Additionally, they "can be prone to error" because "adversarial testing" on facts relevant to the different-occasions inquiry may not have been necessary "in the prior proceeding." Id. at 841 (citations omitted). A trial likely would have protected against such reliability problems.
While Erlinger focused on problems with Shepard documents' utility in a jury-trial context, those concerns do not melt away with the appellate gaze. The Erlinger majority's strong warning speaks in contrast to the Greer majority's invitation to review the whole record when conducting the plain-error substantial-rights prong inquiry. Cf. Greer, 593 U.S. at 515 (Sotomayor, J., concurring) ("[T]he Court's analysis in Greer's case does not extend to the distinct context of harmless-error review."). Yet, as the law currently stands, the majority's conclusion that we may use the "entire record" to assess the harmlessness of an Erlinger error appears inescapable. That said, inescapable does not mean inevitable. After all, "[t]here is no efficiency exception to the Fifth and Sixth Amendments." Erlinger, 602 U.S. at 842. And given Erlinger's caution, we should well consider whether the jury right we seek to protect in calling out an Erlinger error is best served through harmless error review reliant on Shepard documents.
Importantly, as discussed in the majority opinion, the district court record contained ample evidence- including Campbell's counsel's acknowledgment at sentencing-that the convictions subject to different-occasions analysis occurred months and (in two of the three drug-trafficking offenses) miles apart. (R. 63, PagelD 413-15).