United States v. Christopher Mitchell

Opinion

PUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 23-4291

UNITED STATES OF AMERICA, Plaintiff - Appellee,

v. CHRISTOPHER LADARISS MITCHELL, Defendant - Appellant.

Appeal from the United States District Court for the Middle District of North Carolina, at Greensboro. William L. Osteen, Jr., District Judge. (1:21–cr–00227–WO–3)

Argued: September 10, 2024 Decided: November 7, 2024

Before DIAZ, Chief Judge, and WYNN and THACKER, Circuit Judges.

Affirmed by published opinion. Judge Wynn wrote the opinion, in which Chief Judge Diaz and Judge Thacker joined.

ARGUED: Mark A. Jones, BELL, DAVIS & PITT, PA, Winston-Salem, North Carolina, for Appellant. Stephen Thomas Inman, OFFICE OF THE UNITED STATES ATTORNEY, Greensboro, North Carolina, for Appellee. ON BRIEF: Sandra J. Hairston, United States Attorney, Angela H. Miller, Assistant United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Greensboro, North Carolina, for Appellee.

2

WYNN, Circuit Judge:

Defendant Christopher Ladariss Mitchell appeals his sentence, which the district court imposed after Mitchell pleaded guilty to one count of conspiracy to commit Hobbs Act robbery and one substantive count of Hobbs Act robbery. On appeal, he raises a single challenge: that the district court erred when it included conspiracies to commit four separate robberies in the calculation of his total offense level. Instead, he argues, it should have considered only two robbery conspiracies. We affirm.

I.

In June 2021, a grand jury returned an eight-count indictment against Mitchell and eight others. The indictment opened with “General Allegations,” which stated in relevant part that, “[a]t all times material to th[e] Indictment,” “Company A” and “Company B” were “wireless communication retailer[s]” with stores located at several identified addresses, “among other locations within the Middle District of North Carolina, and elsewhere.” J.A. 6–7. 1 “Company A” is Boost Mobile, and “Company B” is MetroPCS. However, for ease of reference, we will continue to refer to them as Company A and Company B.

Count One charged all nine defendants with conspiracy to commit Hobbs Act robbery. Specifically, it alleged that “[f]rom on or about August 16, 2020, continuing up to and including on or about September 15, 2020,” Mitchell and his codefendants conspired to rob “employees of Company A and Company B” of “property, including United States

1

Citations to the “J.A.” and “S.J.A.” refer, respectively, to the Joint Appendix and Sealed Joint Appendix filed by the parties in this appeal.

3

currency and wireless devices.” J.A. 7–8. Counts Two through Eight each brought a substantive charge of Hobbs Act robbery, alleging that particular defendants had robbed a particular location on a particular date. For example, Counts Two and Three alleged that Mitchell and others had robbed Company A at, respectively, 121 National Highway in Thomasville, North Carolina, on August 16, 2020, and 4411 West Gate City Boulevard in Greensboro, North Carolina, on August 19, 2020. Counts Four through Eight also specified robberies by date and location but did not implicate Mitchell. 2 In January 2023, Mitchell pleaded guilty to Counts One and Two pursuant to a written plea agreement. The parties did not stipulate to a factual basis at the time Mitchell entered his guilty plea.

The Probation Office prepared a draft presentence investigation report (“PSR”). In a section titled “The Offense Conduct,” the PSR listed five robberies or attempted robberies in which it stated Mitchell had directly participated: (1) a robbery of a Circle K store in Salisbury, North Carolina, on August 14, 2020; (2) the Thomasville robbery of Company A described in Count Two of the indictment; (3) the Greensboro robbery of Company A described in Count Three; (4) a robbery of a Company B store in Kannapolis, North Carolina, on August 21, 2020; and (5) an attempted robbery of a Company B store in Sumter, South Carolina, on September 14, 2020. A table summarizing these robberies follows:

2

To be more precise, Count Six named Mitchell as a participant in the robbery described in that count. However, “subsequent investigation indicate[d] there [was] no evidence he participated in th[at] robbery.” S.J.A. 122.

4

Date (2020) Location Company Indictment August 14 Salisbury Circle K (not listed)

August 16 Thomasville Company A Count 2 August 19 Greensboro Company A Count 3 August 21 Kannapolis Company B (not listed)

September 14 Sumter Company B (not listed)

The PSR applied the 2021 United States Sentencing Guidelines Manual (“Guidelines”) to calculate Mitchell’s Guidelines range. First, the PSR noted that, under § 1B1.2(d) of the Guidelines, “[a] conviction on a count charging a conspiracy to commit more than one offense shall be treated as if the defendant had been convicted on a separate count of conspiracy for each offense that the defendant conspired to commit.” S.J.A. 118 (quoting U.S. Sent’g Guidelines Manual § 1B1.2(d) (U.S. Sent’g Comm’n 2021)). The PSR understood this Guideline to mean that Count One should be treated as if Mitchell was convicted on a separate conspiracy count for each of the five aforementioned robberies. It then noted that a multiple-count adjustment applied pursuant to § 3D1.4 of the Guidelines. Specifically, the Guidelines instruct that, “[w]hen a defendant has been convicted of more than one count, the court shall” group closely related counts, determine the offense level for each group, and then consult the table in § 3D1.4 to deduce the combined offense level. U.S.S.G. § 3D1.1(a).

Applying these rules, Mitchell’s offense level was 26 plus an increase in offense level based on the number of units (which in this case was five because the robberies were not grouped together). Five units led to a four-level increase, bringing Mitchell’s offense level to 30. Id. § 3D1.4. A three-level decrease for acceptance of responsibility reduced the

5

total offense level to 27. Based on this offense level and Mitchell’s criminal history category of V, the draft PSR calculated his Guidelines range as 120 to 150 months in prison.

Mitchell objected to the inclusion of the August 14, 2020, Circle K robbery as outside the scope of the charged conspiracy. He further took the position that only those robberies “identified in Count One as an object of the conspiracy offense alleged therein” could be counted pursuant to § 1B1.2(d) and its commentary. S.J.A. 151. And he argued that “[b]ecause Count One does not allege any specific robberies, the objects of that offense should be limited to the two objects the Government did elect to allege: a robbery of Company A and a robbery of Company B.” S.J.A. 156. He contended that the two robberies that should count were the Greensboro and Kannapolis robberies. See S.J.A. 158. With only those two robberies considered, the increase pursuant to § 3D1.4 would be two, rather than four, levels, resulting in a total offense level of 25 rather than 27.

The Probation Office agreed that the Circle K robbery should not be included and removed it from the revised PSR, but concluded that the other four robberies were properly included. Four units, just like five units, produces a four-level increase in offense level under § 3D1.4, so the revised PSR continued to calculate Mitchell’s total offense level as 27.

In his sentencing memorandum and during the sentencing hearing, Mitchell reiterated his argument that § 1B1.2(d) and its commentary allowed for an offense-level increase based only on “the two general offense/objects [the Government] did elect to mention [in the conspiracy count, Count One]: a robbery of Company A and a robbery of Company B.” S.J.A. 197. At times, however, he seemed to imply that the appropriate

6

robberies to consider were the Thomasville and Greensboro robberies—even though both were of Company A—as those robberies (unlike the Kannapolis and Sumter robberies of Company B) were specifically identified in the indictment. E.g., S.J.A. 198 (appearing to object to the inclusion of the Kannapolis and Sumter robberies).

At sentencing, the district court confirmed with defense counsel and Mitchell himself that he made no factual objection regarding his participation in all four robberies. That is, Mitchell explicitly agreed that the court could appropriately find “that the Government [could] prove beyond a reasonable doubt [his] participation in the commission of the four robberies of the wireless stores.” J.A. 45–46. After hearing arguments related to § 1B1.2(d), the district court rejected Mitchell’s view of that Guideline and adopted the revised PSR, with its inclusion of all four robberies.

Although Mitchell’s Guidelines range was 120 to 150 months, the Government argued for a downward variance to 108 months. The district court agreed and sentenced Mitchell to 108 months on each count, to run concurrently. 3 Mitchell timely appealed. 4

3

The court later granted Mitchell’s motion for a sentence reduction pursuant to 18 U.S.C. § 3582(c)(2) and reduced his sentence to 100 months.

4

We ordered supplemental briefing on the level of deference owed to the Guidelines and their commentary and whether any error was harmless.

7

II.

Mitchell raises a single challenge on appeal: that the district court erred when it increased his offense level by four (rather than two) levels pursuant to Guideline § 1B1.2(d). 5 “‘In assessing whether a sentencing court has properly applied the Guidelines, we review factual findings for clear error and legal conclusions de novo.’ Because this appeal involves a purely legal question—interpretation of the Guidelines—we review de novo.” United States v. Thompson, 874 F.3d 412, 414 (4th Cir. 2017) (citation omitted) (quoting United States v. Llamas, 599 F.3d 381, 387 (4th Cir. 2010)).

III.

A.

In the present appeal, the parties dispute the meaning of the applicable Guideline and its accompanying commentary. We thus begin by setting forth the principles that shape our interpretation of the Guidelines.

The Sentencing Commission promulgates the Sentencing Guidelines Manual, which “contains text of three varieties”: the Guidelines themselves; policy statements; and commentary on “both guidelines and policy statements.” Stinson v. United States, 508 U.S. 36, 41 (1993).

5

In a footnote in his Opening Brief, Mitchell also “submits that his sentence is substantively unreasonable.” Opening Br. at 14 n.1. This “passing shot at the issue” is insufficient to avoid forfeiture of the argument, so we do not consider it further. United States v. Smith, 75 F.4th 459, 468 (4th Cir. 2023) (quoting Mowery v. Nat’l Geospatial- Intel. Agency, 42 F.4th 428, 433 n.5 (4th Cir. 2022)).

8

In 1993, the Supreme Court held in Stinson v. United States that “commentary in the Guidelines Manual that interprets or explains a guideline is authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that guideline.” Id. at 38. Thus, under Stinson, the interpretive weight owed to the commentary did not turn on whether the Guideline was ambiguous. See id. at 44 (“[C]ommentary explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice.”).

In so holding, Stinson analogized the commentary to “an agency’s interpretation of its own legislative rules.” Id. at 45. Under so-called Seminole Rock (or Auer) deference, such an agency interpretation was—at the time of Stinson—“given ‘controlling weight unless it [was] plainly erroneous or inconsistent with the regulation,’” or unless it “violate[d] the Constitution or a federal statute.” Id. (quoting Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 414 (1945)); see Auer v. Robbins, 519 U.S. 452, 461 (1997).

In its 2019 decision in Kisor v. Wilkie, however, the Supreme Court reconsidered the scope of Auer deference. While “uphold[ing]” the Auer doctrine, the Court “reinforce[d] its limits.” Kisor v. Wilkie, 588 U.S. 558, 563 (2019). In particular, the Court emphasized that “the possibility of [Auer] deference can arise only if a regulation is genuinely ambiguous . . . . after a court has resorted to all the standard tools of interpretation.” Id. at 573. Further, even if the regulation is genuinely ambiguous, “the agency’s reading must still be ‘reasonable.’” Id. at 575 (quoting Thomas Jefferson Univ. v. Shalala, 512 U.S. 504, 515 (1994)). And even then, “not every reasonable agency reading of a genuinely ambiguous rule should receive Auer deference”; instead, courts “must make

9

an independent inquiry into whether the character and context of the agency interpretation entitles it to controlling weight.” Id. at 576; see id. at 576–79 (setting forth aspects of the agency’s interpretation to consider in undertaking this inquiry).

The Supreme Court’s decision in Kisor left lower courts in limbo regarding how to understand Stinson. True, Kisor did not explicitly mention the Guidelines context of Stinson, and Stinson had relied on Auer deference as an “analogy” that was “not precise.” Stinson, 508 U.S. at 44. Nevertheless, Stinson also stated that the Court thought “the Government [was] correct in suggesting that the [Guidelines] commentary be treated as an agency’s interpretation of its own legislative rule,” id.; Stinson explicitly quoted Seminole Rock, id. at 45; and Kisor cited Stinson as one of the Court’s pre-Auer “decisions applying Seminole Rock deference,” Kisor, 588 U.S. at 568 n.3. So, while “it is th[e Supreme] Court’s prerogative alone to overrule one of its precedents,” State Oil Co. v. Khan, 522 U.S. 3, 20 (1997), Kisor’s alteration of the standard cited in Stinson raises the question of whether it modified the test applicable to the Guidelines and its commentary.

The Circuits are deeply divided on the answer to this question. The Third, Sixth, Ninth, and Eleventh Circuits have held that Kisor limits the deference originally required by Stinson, meaning courts must find a Guideline ambiguous before deferring to its commentary. 6 The Second, Fifth, Seventh, Eighth, and Tenth Circuits, by contrast, have

6

See United States v. Nasir, 17 F.4th 459, 471 (3d Cir. 2021) (en banc); United States v. Riccardi, 989 F.3d 476, 485 (6th Cir. 2021); United States v. Castillo, 69 F.4th 648, 655 (9th Cir. 2023); United States v. Dupree, 57 F.4th 1269, 1275 (11th Cir. 2023) (en banc).

10

concluded that the standard in Stinson continues to govern, unaltered by Kisor. 7 We are not aware of published authority from the First or D.C. Circuits explicitly taking a position in this dispute. 8 In United States v. Campbell, we adopted the first approach: reading Stinson through the lens of Kisor. See United States v. Campbell, 22 F.4th 438, 445 & n.3 (4th Cir. 2022) (holding that “Kisor limited when courts will afford Seminole Rock/Auer deference” and “ma[de] clear that [its] modifications to Seminole Rock/Auer deference apply equally to judicial interpretations of the Sentencing Commission’s commentary”); United States v. Boler, 115 F.4th 316, 322 (4th Cir. 2024) (noting that “Stinson’s analysis was altered by Kisor” (citing Campbell, 22 F.4th at 445)).

Mitchell nevertheless argues that this Court has in fact adopted the approach of the second group of circuits noted above: that Kisor does not apply to the Guidelines and their commentary. He relies on United States v. Moses, which issued a short while after

7

See United States v. Rainford, 110 F.4th 455, 475 n.5 (2d Cir. 2024); United States v. Vargas, 74 F.4th 673, 678 & n.3 (5th Cir. 2023) (en banc), cert. denied, 144 S. Ct. 828 (2024) (mem.); United States v. White, 97 F.4th 532, 535 (7th Cir. 2024), cert. denied, --- S. Ct. ---, 2024 WL 4427289 (Oct. 7, 2024) (mem.); United States v. Donath, 107 F.4th 830, 838 (8th Cir. 2024); United States v. Maloid, 71 F.4th 795, 809 (10th Cir. 2023), cert. denied, 144 S. Ct. 1035 (2024) (mem.).

8

See United States v. Gadson, 77 F.4th 16, 20 (1st Cir. 2023) (noting for purposes of plain-error review that “even assuming that Kisor abrogated Stinson,” any related error by the district court “was not ‘clear or obvious’”), cert. denied, 144 S. Ct. 823 (2024) (mem.); United States v. Lewis, 963 F.3d 16, 24 (1st Cir. 2020) (relying on pre-Kisor circuit precedent to interpret a Guideline because nothing in Kisor would have altered that precedent). Compare United States v. Jenkins, 50 F.4th 1185, 1197 (D.C. Cir. 2022) (citing both Stinson and Kisor in the sentencing context), with United States v. Sargent, 103 F.4th 820, 822, 825 (D.C. Cir. 2024) (suggesting the question is open in the D.C. Circuit).

11

Campbell and stated that “Stinson continues to apply unaltered by Kisor.” United States v. Moses, 23 F.4th 347, 349 (4th Cir. 2022), cert. denied, 143 S. Ct. 640 (2023) (mem.).

To the extent there may be a conflict between Campbell and Moses, in this Circuit, we follow the bright-line rule that the first-decided case controls. McMellon v. United States, 387 F.3d 329, 333 (4th Cir. 2004) (en banc) (“When published panel opinions are in direct conflict on a given issue, the earliest opinion controls, unless the prior opinion has been overruled by an intervening opinion from this court sitting en banc or the Supreme Court.”); accord Moses, 23 F.4th at 359–60 (King, J., dissenting in part and concurring in the judgment).

But Mitchell contends that there is no direct conflict between Campbell and Moses because Campbell’s application of Kisor was mere “dicta,” such that Moses must govern even though Campbell was the first-decided case. Supp. Opening Br. at 3. That is simply incorrect.

In Campbell, this Court initially analyzed the Guidelines question pursuant to Stinson, and then made it clear that “if there were any doubt” of the result “under Stinson[,] . . . the Supreme Court’s recent decision in Kisor v. Wilkie renders this conclusion indisputable.” Campbell, 22 F.4th at 444 (citation omitted). And across “nearly four pages” of its opinion, Campbell discussed “the impact of Kisor on the question at issue.” United States v. Moses, No. 21-4067, 2022 U.S. App. LEXIS 7694, at *9 (4th Cir. Mar. 23, 2022) (Wynn, J., voting to grant rehearing en banc). In doing so, it “expressly relie[d] on Kisor to hammer home its conclusion,” making its “repeated citations to Kisor . . . key analytical building blocks that support its overall conclusion.” Id. at *9–10. Thus, Campbell

12

alternatively held that Kisor applies to the Guidelines. That’s “not dicta.” Gestamp S.C., L.L.C. v. Nat’l Lab. Rels. Bd., 769 F.3d 254, 262 n.4 (4th Cir. 2014) (citing MacDonald, Sommer & Frates v. County of Yolo, 477 U.S. 340, 346 n.4 (1986)); accord Moses, 2022 U.S. App. LEXIS 7694, at *14–15 (Motz, J., dissenting from denial of rehearing en banc and voting to grant rehearing en banc). Instead, that holding in Campbell is binding precedent. E.g., United States v. Ford, 703 F.3d 708, 711 n.2 (4th Cir. 2013) (“Where a court makes alternative holdings to support its decision, each holding is binding precedent.”); Est. of Van Emburgh ex rel. Van Emburgh v. United States, 95 F.4th 795, 804 (4th Cir. 2024) (“We may, of course, sometimes issue alternative holdings that have precedential effect.”).

In this Circuit, then, the opinion in Campbell controls on the question of whether Kisor modified Stinson, unless and until the question is revisited by the Supreme Court or this Court sitting en banc. 9 Accordingly, our first stop in analyzing the meaning of the relevant Guideline is whether it is genuinely ambiguous. Boler, 115 F.4th at 323. If it is, “we next consider whether the commentary’s [relevant] definition . . . falls within the ‘zone of ambiguity’

9

The Supreme Court has, to date, declined to address this issue, despite the deep circuit split and numerous requests from litigants, amici, and lower courts for guidance. E.g., Ratzloff v. United States, 144 S. Ct. 554 (2024) (mem.) (denying petition for writ of certiorari regarding the impact of Kisor on Stinson); Vargas, 144 S. Ct. 828 (same); Maloid, 144 S. Ct. 1035 (same); Gadson, 144 S. Ct. 823 (same); United States v. Rivera, 144 S. Ct. 861 (2024) (mem.) (same); Moses, 143 S. Ct. 640 (same); Moses, 2022 U.S. App. LEXIS 7694, at *6 (Niemeyer, J., supporting the denial of rehearing en banc) (noting that “the Supreme Court’s advice on whether Stinson or Kisor controls the enforceability of and weight to be given Guidelines commentary” would be “welcome,” as it is “an issue that could have far-reaching results”).

13

such that it should be given deference.” Id. at 327 (quoting Kisor, 588 U.S. at 576, and United States v. You, 74 F.4th 378, 398 (6th Cir. 2023)). And, if so, “we independently inquire as to whether the commentary’s ‘character and context’ entitles it to ‘controlling weight.’” Id. at 328 (quoting Kisor, 588 U.S. at 576). An ambiguous Guideline is also subject to the rule of lenity. Campbell, 22 F.4th at 446.

If the Guideline is not genuinely ambiguous, however, we cannot defer to the commentary. “If uncertainty does not exist, there is no plausible reason for deference. The [Guideline] then just means what it means—and the court must give it effect, as the court would any law.” Kisor, 588 U.S. at 574–75. For that reason, “when commentary is inconsistent with an unambiguous guideline”—for example, if the commentary would expand the application of a Guideline beyond its plain meaning—“‘the Sentencing Reform Act itself commands compliance with the guideline.’” Campbell, 22 F.4th at 447 (emphasis added) (quoting Stinson, 508 U.S. at 43).

The fact that commentary is not owed deference where the Guideline is unambiguous does not, however, mean that courts must entirely ignore the commentary. “Rather, commentary explains the guidelines and provides concrete guidance as to how even unambiguous guidelines are to be applied in practice.” Stinson, 508 U.S. at 44. Put another way, while an unambiguous Guideline “means what it means,” Kisor, 588 U.S. at 575, such that “we have no need to consider” commentary when a Guideline is unambiguous, United States v. Dupree, 57 F.4th 1269, 1279 (11th Cir. 2023) (en banc) (emphasis added), and cannot defer to it, that does not mean “concrete” examples of how that unambiguous Guideline applies are unhelpful, Stinson, 508 U.S. at 44.

14

B.

Having set forth the principles that are to guide our analysis, we turn to “[t]he ‘first step in interpreting’ [the] guideline,” which “‘is to determine whether the language at issue has a plain and unambiguous meaning with regard to the particular dispute in the case.’” United States v. Sargent, 103 F.4th 820, 826 (D.C. Cir. 2024) (quoting Robinson v. Shell Oil Co., 519 U.S. 337, 340 (1997)). In evaluating ambiguity, we must consider a term’s “ordinary meaning” and employ “the various canons of statutory interpretation,” including considering “‘the specific context in which the language is used, and the broader context of the’ regulation as a whole.” Boler, 115 F.4th at 323, 325 (quoting Hurlburt v. Black, 925 F.3d 154, 158 (4th Cir. 2019) (en banc)).

Under Guidelines § 1B1.2(d), “[a] conviction on a count charging a conspiracy to commit more than one offense shall be treated as if the defendant had been convicted on a separate count of conspiracy for each offense that the defendant conspired to commit.” To apply this Guideline, we must first understand the meaning of “more than one offense.” Count One indisputably charged Mitchell with conspiring to violate a single statute (18 U.S.C. § 1951(a)) by means of multiple instances of the same type of criminal act (robbery). If “more than one offense” requires conspiracy to violate multiple statutes, or to commit multiple types of criminal acts, then the Guideline does not apply at all in Mitchell’s case. If it refers merely to conspiracy to commit more than one criminal act, however, the Guideline applies, and we must then determine how to apply it.

We conclude that, at least as to robbery, the Guideline unambiguously carries the latter definition: § 1B1.2(d) applies where a defendant is convicted on a count charging a

15

conspiracy to commit more than one robbery, even if those underlying robberies would constitute violations of the same statute. 10 Dictionary definitions of the term “offense” are of little help, as they could point to either of the meanings noted above. 11 The same is true of Mitchell’s acknowledgment that the term “unambiguously means a specific violation of federal criminal law.” Supp. Opening Br. at 5. The context provided by the Guidelines as a whole, however, resolves any ambiguity in this case. Chapter Two of the Guidelines, “Offense Conduct,” identifies various “Offenses,” including “Basic Economic Offenses” such as “Robbery.” U.S.S.G. § 2B3.1. The inclusion of “Robbery” within Chapter Two indicates that the robbery itself is the offense. Reviewing the Guidelines section on “Robbery” confirms this understanding: it cites several statutory provisions under which a defendant might be charged for that offense (demonstrating that the focus of the “offense” is on the act of robbery, not the particular statute) and provides “Specific Offense Characteristics” that relate to the facts of particular robberies (demonstrating that the focus of the “offense” is on each individual robbery, not the class of robberies).

The commentary provides a relevant “concrete” example of how this “unambiguous” Guideline, § 1B1.2(d), is “to be applied in practice” to robbery

10

We need not consider what constitutes a single “robbery” for purposes of this Guideline—for example, whether a defendant who robs multiple victims within the same store has committed one “offense” or several—because the robberies at issue in this case were indisputably distinct, occurring on different dates at different locations.

11

E.g., Offense, Black’s Law Dictionary (12th ed. 2024) (defining “offense” as “[a]

violation of the law; a crime, often a minor one”); Offense, Merriam-Webster.com, https://www.merriam-webster.com/dictionary/offense (last visited Nov. 3, 2024) (defining “offense” in relevant part as “an infraction of law”) [https://perma.cc/PS7S-VXLD].

16

conspiracies. Stinson, 508 U.S. at 44. Application Note 3 states that, “where a conviction on a single count of conspiracy establishes that the defendant conspired to commit three robberies, the guidelines are to be applied as if the defendant had been convicted on one count of conspiracy to commit the first robbery, one count of conspiracy to commit the second robbery, and one count of conspiracy to commit the third robbery.” U.S.S.G. § 1B1.2 cmt. n.3.

Our opinion in United States v. Gutierrez, 963 F.3d 320 (4th Cir. 2020), is not to the contrary. That case involved a “conspiracy to engage in a pattern of racketeering activities, in violation of the Racketeer Influenced and Corrupt Organizations Act (‘RICO’).” Id. at 328 (citing 18 U.S.C. § 1962(d)). Although “[t]he racketeering activities described in the indictment were numerous,” we concluded that § 1B1.2(d) did not apply. Id.; see id. at 343. But that holding does not suggest that “offense” in § 1B1.2(d) should be understood to refer to the statute, rather than the underlying criminal acts. Rather, our decision in Gutierrez rested on the unique situation posed by RICO conspiracies, which by their nature “are of the single-object variety, with the object being to engage in racketeering.” Id. at 343 (quoting United States v. Garcia, 754 F.3d 460, 482 (7th Cir. 2014)). Further, RICO conspiracies are “specifically govern[ed]” by a different Guideline. Id. (citing U.S.S.G. § 2E1.1(a)(2)). Accordingly, our treatment of RICO conspiracies in Gutierrez has no bearing on how to understand § 1B1.2(d).

Turning back to the text of § 1B1.2(d), then, Mitchell has “[a] conviction on a count charging a conspiracy to commit more than one offense”: he pleaded guilty to Count One, which charged a conspiracy to commit more than one robbery (at least one against each

17

company, which had different locations). So we must next determine the meaning of the remainder of the Guideline: such a conviction “shall be treated as if the defendant had been convicted on a separate count of conspiracy for each offense that the defendant conspired to commit.” U.S.S.G. § 1B1.2(d).

The question becomes what to make of the phrase “each offense that the defendant conspired to commit.” “Offense” necessarily takes the same meaning as it does earlier in the sentence: it refers to individual robberies. And “conspired” necessarily refers to the “conspiracy” noted earlier in the sentence, thus limiting our consideration to “offenses” falling within that “conspiracy.” For this reason, the Circle K robbery that the draft PSR attributed to Mitchell for § 1B1.2(d) purposes was appropriately excluded from the final PSR, as it did not fall within the conspiracy charged in Count One.

Still, the phrase “each offense that the defendant conspired to commit” raises at least two questions: Must “each” of these offenses have been identified with particularity in the “count charging a conspiracy”? And, if the plea or verdict does not plainly establish which of multiple offenses a defendant “conspired to commit”—for example, because the conspiracy charge did not identify the conspired-to offenses with particularity, or because the conspiracy charge set forth a number of particular offenses allegedly conspired to, and the plea or verdict did not indicate which offense or offenses were in fact implicated— under what standard should a court determine “each offense that the defendant conspired to commit”?

Regarding the first question, we conclude the answer is unambiguously “no.” It is black-letter law that a “count” can “charg[e] a conspiracy to commit more than one

18

offense”—understood here as more than one robbery—without specifically identifying each of the robberies, e.g., by date, location, and victim. Indeed, under certain circumstances, the particular acts underlying a conspiracy charge need not be specified in the indictment at all. See United States v. Janati, 374 F.3d 263, 270–71 (4th Cir. 2004) (reversing district court’s ruling that evidence allowed to be presented on conspiracy count was limited to overt acts described in other substantive charges in indictment); accord United States v. Coleman, 349 F.3d 1077, 1088 (8th Cir. 2003) (applying this principle in a § 1B1.2(d) case). And while the word “each” turns our focus to the particular offenses (robberies), it does so with the qualifier “that the defendant conspired to commit”—not, for example, “that the defendant was specifically alleged to have conspired to commit.” Accordingly, an “offense that the defendant conspired to commit” can only mean one that the defendant did, in fact, conspire to commit.

This view finds support in other Guideline provisions, too. First, § 1B1.2(a) defines “the offense of conviction” as “the offense conduct charged in the count of the indictment or information of which the defendant was convicted.” U.S.S.G. § 1B1.2(a). The fact that the phrase “offense of conviction” refers to the specific acts alleged in the indictment suggests that the more general “offense” used in § 1B1.2(d) refers to a broader range of conduct. Second, § 1B1.2(c) requires a court sentencing a defendant who pleaded guilty to treat the defendant as having been convicted for both any offenses of conviction and “additional offense(s).” U.S.S.G. § 1B1.2(c). This provision, which comes just before § 1B1.2(d), contemplates holding a defendant accountable for unindicted conduct, and so supports a reading of “offense” in § 1B1.2(d) that also includes conduct not specifically

19

alleged. Third, § 1B1.3(a)(1)(A) provides that, “[u]nless otherwise specified,” base offense levels, specific offense characteristics, and adjustments to the offense level—including the multiple-counts provision of § 1B1.2(d)—must be calculated based on “all acts and omissions committed, aided, abetted, counseled, commanded, induced, procured, or willfully caused by the defendant.” U.S.S.G. § 1B1.3(a)(1)(A) (emphasis added). This language is not limited to conduct identified in the indictment.

These provisions suggest that “each offense that the defendant conspired to commit,” as used within § 1B1.2(d), is not limited to offenses specifically alleged in the indictment but refers to all offenses the defendant conspired to commit that fall within the conspiracy that is alleged in the indictment. The Guideline thus instructs courts to determine which offenses falling within the overall charge of conspiracy a defendant in fact conspired to commit, and to treat each of those offenses as a separate conviction for sentencing purposes.

This conclusion leads to the second question noted above: how to determine the conspired-to offenses when the conviction itself (e.g., the plea or verdict) does not explicitly establish which of various possible offenses should be considered. We have no need to answer that question in this case, however, because the district court concluded that the evidence supported Mitchell’s involvement in the four robberies under any standard, including beyond a reasonable doubt, and Mitchell does not challenge that conclusion on appeal. Cf. United States v. Dickerson, 27 F. App’x 236, 246 (4th Cir. 2001) (per curiam) (concluding, based on the commentary and Appendix C of the Guidelines, that “when

20

exercising its authority under § 1B1.2(d), the district court’s decision is governed by a reasonable doubt standard”).

In sum, under the circumstances of this case, § 1B1.2(d) is unambiguous. Mitchell has “[a] conviction on a count” (Count One) “charging a conspiracy to commit more than one offense” (robberies of Company A or Company B that occurred from on or about August 16, 2020, to on or about September 15, 2020). So, that conviction is to “be treated as if” he was “convicted on a separate count of conspiracy for each” of the four robberies he undisputably “conspired to commit”: (1) the August 16 robbery of a Company A store in Thomasville; (2) the August 19 robbery of a Company A store in Greensboro; (3) the August 21 robbery of a Company B store in Kannapolis; and (4) the September 14 attempted robbery of a Company B store in Sumter. Accordingly, for purposes of sentencing, § 1B1.2(d) applies, and Mitchell’s conviction on Count One is to be treated as if he were convicted of four robbery conspiracies. We therefore affirm the district court’s calculation of Mitchell’s Guidelines range.

C.

Mitchell raises several arguments as to why only two robbery conspiracies, rather than four, should be scored for purposes of § 1B1.2(d). None are persuasive.

Mitchell’s primary contention, relying on Application Note 4, is that under § 1B1.2(d), only those offenses “alleged in the conspiracy count” may be counted. Opening Br. at 11 (quoting U.S.S.G. § 1B1.2 cmt. n.4). And he contends that Count One charges a conspiracy to commit only two robberies: “one robbery of Company A and one

21

robbery of Company B.” Id. at 19; see id. at 26 (specifying that the two robberies to be counted are the Greensboro and Kannapolis robberies).

This argument falls flat for several reasons. Because we have determined that the Guideline is unambiguous on the relevant question, we do not owe deference to the commentary on which Mitchell relies. Further, that reliance may be misplaced. The commentary states that, where the “verdict or plea does not establish which offense(s) was the object of the conspiracy,” § 1B1.2(d) “should only be applied with respect to an object offense alleged in the conspiracy count if the court, were it sitting as a trier of fact, would convict the defendant of conspiring to commit that object offense.” U.S.S.G. § 1B1.2 cmt. n.4. Mitchell would like to read this commentary to mean, affirmatively, that the Guideline can “only be applied with respect to an object offense alleged in the conspiracy count.” But the statement is conditional: the Guideline can “only be applied with respect to an object offense alleged in the conspiracy count if” certain evidentiary requirements are met. Id. (emphasis added); see United States v. Robles, 562 F.3d 451, 454–55 (2d Cir. 2009) (per curiam) (noting that the commentary in question “places its emphasis not on the specificity of the conspiracy charge but on the standard of proof that must be satisfied,” and upholding the district court’s application of § 1B1.2(d) to count two robbery offenses that “were charged as substantive offenses” but “were not specifically designated as the objects of the conspiracy ‘in the conspiracy count’”); accord United States v. Ford, 761 F.3d 641, 659– 60 (6th Cir. 2014) (similarly rejecting Mitchell’s understanding of the commentary).

Even giving Mitchell the benefit of his premise, however, nothing in Count One limits the conspiracy to a single robbery each of Company A and Company B. To the

22

contrary, under Count One’s plain terms, any robbery of any location of either company within the specified time frame would qualify. See J.A. 6–8 (indictment alleging that, “[f]rom on or about August 16, 2020, continuing up to and including on or about September 15, 2020,” Mitchell and others conspired to rob “employees of Company A and Company B,” and incorporating “General Allegations” stating that the companies were “wireless communication retailer[s] with store locations” at six identified addresses, “among other locations within the Middle District of North Carolina, and elsewhere”).

Changing tack, Mitchell contends that this breadth of the indictment is itself a problem. He argues that only those robberies identified with particularity (i.e., by date and location) somewhere in the indictment—rather than all those falling within the broader parameters set by Count One—may be counted. E.g., Opening Br. at 17 (suggesting that only “robberies . . . identified in the Indictment” should be considered); id. at 20–21 (distinguishing Mitchell’s case from others in which the indictment somewhere listed the offenses with particularity); id. at 25 (suggesting that due process requires that “evidence of ‘object offenses’” must be “reflected by the allegations set forth in the indictment” to be used “for sentencing purposes”). Under this approach, it is the two Company A robberies (Thomasville and Greensboro) that should be counted, rather than one robbery of Company A and one robbery of Company B (e.g., Greensboro and Kannapolis).

This second argument rests on concerns about notice and due process: that Mitchell was not told up front, in the indictment, precisely which stores he was accused of conspiring to rob, and when. See id. at 25. But Mitchell never challenged the indictment as unconstitutionally vague; never moved for a bill of particulars under Federal Rule of

23

Criminal Procedure 7(f); 12 never sought to withdraw his plea; and, indeed, has explicitly disclaimed on appeal any argument that the indictment was flawed or that his plea was unknowing. See Oral Arg. at 11:19–12:44, 38:55–39:44, 40:44–41:24, https://www.ca4.uscourts.gov/OAarchive/mp3/23-4291-20240910.mp3. And the indictment did provide Mitchell with at least “broad” notice, Robles, 562 F.3d at 456: he was informed that the Government was charging him with a conspiracy to commit a certain crime (robbery) of certain materials (currency and wireless devices) from certain victims (employees of Company A and Company B) between certain dates (on or about August 16 to on or about September 15, 2020) at locations that included several identified in the indictment.

Further, the Supreme Court has made clear that while “[a]n indictment must set forth each element of the crime that it charges . . . . [,] it need not set forth factors relevant only to the sentencing of an offender found guilty of the charged crime.” Almendarez-Torres v. United States, 523 U.S. 224, 228 (1998). Rather, this Court has held that the “notice . . . necessary to give the defendant ‘an opportunity to contest the validity or applicability of . . . [a] sentencing enhancement’” is “typically” provided through the PSR. United States v. Hodge, 902 F.3d 420, 427 (4th Cir. 2018) (quoting United States v. Moore, 208 F.3d 411, 414 (2d Cir. 2000) (per curiam)); cf. United States v. Jinwright, 683 F.3d 471, 487 (4th Cir. 2012) (upholding enhancement where “the PSR provided notice that” the

12

See United States v. Am. Waste Fibers Co., 809 F.2d 1044, 1047 (4th Cir. 1987)

(per curiam) (rejecting the argument that the indictment was too vague, and noting that if the defendant needed more “information to enable it to prepare its defense,” the “proper course [was] to seek a bill of particulars”).

24

enhancement “was a factor relevant to sentencing and supplied a non-exhaustive list of” facts “that amply supported application of the enhancement in and of themselves”).

We therefore conclude that where, as here, a defendant is convicted “on a count charging a conspiracy to commit more than one” robbery, that conviction must “be treated” at sentencing “as if the defendant had been convicted on a separate count of conspiracy for each” robbery falling within that conspiracy, regardless of whether the robberies were specifically identified by date and location in the indictment. U.S.S.G. § 1B1.2(d). In so holding, we join the other circuits to have explicitly addressed this question in published authority (albeit before Kisor). See Coleman, 349 F.3d at 1088; Ford, 761 F.3d at 658–59.

IV.

For the foregoing reasons, we affirm the district court’s calculation of Mitchell’s Guidelines range and affirm Mitchell’s sentence.

AFFIRMED