United States v. Muhammad, No. 15-60300 (5th Cir. 2021)

Annotate this Case
Justia Opinion Summary

After defendant was convicted of violating both the Controlled Substances Act and the Controlled Analogue Enforcement Act, the district court sentenced him to 120 years in prison. On appeal, the parties agree that the jury instructions regarding the Analogue Act were erroneous because they omitted an element of the offense. Defendant claimed that the error was structural.

The Fifth Circuit concluded that there is more than enough to demonstrate defendant's knowledge of the chemical structure of the drugs he sold and the controlled substances they mimicked. The court rejected defendant's contention that United States v. Stanford, 823 F.3d 814 (5th Cir. 2016), permits the court to bypass the "thorough examination" of the "whole record" that is required in Neder v. United States, 527 U.S. 1 (1999). Rather, the court applied the Neder harmless-error standard on the same "functional, case-by-case" basis as usual, and concluded that a jury could not rationally find that the Government failed to prove the knowledge requirement in McFadden v. United States, 576 U.S. 186, 188–89 (2015). Therefore, the court concluded that the omitted element in defendant's jury instruction was harmless. The court rejected defendant's remaining arguments on appeal.

Download PDF
Case: 15-60300 Document: 00516017191 Page: 1 Date Filed: 09/16/2021 United States Court of Appeals for the Fifth Circuit United States Court of Appeals Fifth Circuit FILED September 16, 2021 Lyle W. Cayce Clerk No. 15-60300 United States of America, Plaintiff—Appellee, versus Rasheed Ali Muhammad, Defendant—Appellant. Appeal from the United States District Court for the Southern District of Mississippi USDC No. 1:14-cr-36-2 Before Ho, Oldham, and Wilson, Circuit Judges. Andrew S. Oldham, Circuit Judge: This is a case about harmless error. The United States convicted Rasheed Ali Muhammad of violating both the Controlled Substances Act and the Controlled Analogue Enforcement Act. The district court sentenced him to 120 years in prison. On appeal, everyone agrees the jury instructions regarding the Analogue Act were erroneous because they omitted an element of the offense. Muhammad invites us to treat that omission as if it were, for all intents and purposes, structural error. The Supreme Court forbids that approach. We affirm. Case: 15-60300 Document: 00516017191 Page: 2 Date Filed: 09/16/2021 No. 15-60300 I. A. The Controlled Substances Act (“CSA”) makes unlawful the knowing manufacture, distribution, or possession with intent to distribute substances on the federal controlled-substance schedules. 21 U.S.C. § 841(a)(1). The Controlled Analogue Enforcement Act (“Analogue Act”) directs courts to treat certain substances not on the schedules as if they were—including for purposes of the CSA. Id. §§ 802(32)(A), 813. The Analogue Act covers every substance: (i) the chemical structure of which is substantially similar to the chemical structure of a controlled substance in schedule I or II; (ii) which has a stimulant, depressant, or hallucinogenic effect on the central nervous system that is substantially similar to or greater than the stimulant, depressant, or hallucinogenic effect on the central nervous system of a controlled substance in schedule I or II; or (iii) with respect to a particular person, which such person represents or intends to have a stimulant, depressant, or hallucinogenic effect on the central nervous system that is substantially similar to or greater than the stimulant, depressant, or hallucinogenic effect on the central nervous system of a controlled substance in schedule I or II. Id. § 802(32)(A). So the Analogue Act is an antidote to statutory evasion: It expands the CSA’s coverage to include substances that, while technically not on the schedules, mimic those that are. B. Rasheed Ali Muhammad was a drug dealer. Muhammad’s basic plan was to sell substances that (a) could get people high but (b) were not yet on the federal drug schedules. His business was well underway by 2012, and he 2 Case: 15-60300 Document: 00516017191 Page: 3 Date Filed: 09/16/2021 No. 15-60300 actively pitched his products as a legal way to get high. Muhammad set up multiple email accounts for his business and created several websites to ply his wares. Muhammad enlisted his cousin, Michael Young, to help him run the business. And Roslyn Chapman helped Muhammad distribute drugs. Muhammad used chemistry in his efforts to evade the reach of the ever-expanding drug schedules. He watched YouTube videos to research the chemical structures of substances he was interested in selling. He then compared those structures to those of on-schedule substances to see how similar they were. On several occasions, Muhammad shared his chemistry insights with coconspirators. Those communications suggest Muhammad had a knack for combining off-schedule substances to mimic the effects of onschedule substances. Based in part on testimony from Task Force Agent Adam Gibbons (a police officer for the City of Gulfport, Mississippi, who was assigned to a DEA task force), a federal grand jury indicted Muhammad for conspiracy, violation of the CSA, and violation of the Analogue Act. The charges, which covered activities ranging from December 2012 through March 2013, were as follows: (1) Conspiracy with intent to distribute controlled substances and controlled-substance analogues in violation of 21 U.S.C. §§ 802, 813, and 846 and 18 U.S.C. § 2 (“Count One”) (multiple substances); (2) Possession with intent to distribute a controlled-substance analogue in violation of 21 U.S.C. §§ 802, 813, and 841 and 18 U.S.C. § 2 (“Count Two”) (for substance XLR11); (3) Possession with intent to distribute a controlled substance in violation of 21 U.S.C. § 841 and 18 U.S.C. § 2 (“Count Three”) (for substance AM2201); 3 Case: 15-60300 Document: 00516017191 Page: 4 Date Filed: 09/16/2021 No. 15-60300 (4) Possession with intent to distribute a controlled-substance analogue in violation of 21 U.S.C. §§ 802, 813, and 841 and 18 U.S.C. § 2 (“Count Four”) (substance a-PVP); (5) Possession with intent to distribute a controlled substance in violation of 21 U.S.C. § 841 and 18 U.S.C. § 2 (“Count Five”) (a separate count for another instance involving substance AM2201); (6) Possession with intent to distribute a controlled-substance analogue in violation of 21 U.S.C. §§ 802, 813, and 841 and 18 U.S.C. § 2 (“Count Six”) (substance 4-MEC); (7) Possession with intent to distribute a controlled-substance analogue in violation of 21 U.S.C. §§ 802, 813, and 841 and 18 U.S.C. § 2 (“Count Seven”) (a separate count for another instance involving substance a-PVP); (8) Possession with intent to distribute a controlled-substance analogue in violation of 21 U.S.C. §§ 802, 813, and 841 and 18 U.S.C. § 2 (“Count Eight”) (a separate count for another instance involving substance XLR11). Muhammad stood trial before a jury. The district court instructed the jury that, for each count alleging a violation of the Analogue Act, the Government had to prove (1) the alleged analogue’s chemical structure was substantially similar to that of a controlled substance, and (2) the alleged analogue had a stimulant, depressant, or hallucinogenic effect on the nervous system similar to a controlled substance’s effect—or that Muhammad represented it to have such an effect. See 21 U.S.C. § 802(32)(A). The district court added that the Government also needed to prove Muhammad knew the substance was intended for human consumption and what the substance was. But, the court instructed, the Government need not prove Muhammad knew the substance was an analogue. It was enough to prove that Muhammad both knew what the substance was and that he possessed (or conspired to possess) it with the intent to distribute. 4 Case: 15-60300 Document: 00516017191 Page: 5 Date Filed: 09/16/2021 No. 15-60300 The jury convicted Muhammad of Counts One, Three, Four, Five, Six, and Seven. It acquitted him of Counts Two and Eight. The Guidelines recommended a sentence of 1,440 months. See U.S.S.G. § 5G1.2(b). The district court imposed that sentence. After some procedural acrobatics not relevant to the issues before us, Muhammad filed this appeal. II. Muhammad’s principal contention on appeal is that his conviction should be vacated due to a defect in the jury instructions. Five months after Muhammad’s jury found him guilty, the Supreme Court held that the CSA “requires the Government to establish that the defendant knew he was dealing with a controlled substance.” McFadden v. United States, 576 U.S. 186, 188–89 (2015) (quotation omitted). When the controlled substance is an analogue, the Government can satisfy the knowledge requirement in one of two ways. First, it can show that “the defendant knew that the substance was controlled under the . . . Analogue Act.” Id. at 189. Second, it can show that “the defendant knew the specific features of the substance that make it a controlled substance analogue.” Id. at 189, 194 (quotation omitted). This second route requires showing that the defendant knew the substance has a substantially similar structure and produces (or is represented to produce) a substantially similar high to “a controlled substance in schedule I or II.” Ibid. The Government concedes that the jury instructions in this case omitted McFadden’s knowledge requirement. Because Muhammad gets the benefit of McFadden on direct review, see Whorton v. Bockting, 549 U.S. 406, 416 (2007), his Analogue Act convictions cannot stand unless that omission was harmless, see Fed. R. Crim. P. 52(a); McFadden, 576 U.S. at 197. We first apply the harmless-error principles the Supreme Court has told us to apply. Then we address Muhammad’s argument that our precedent requires us to do something else. 5 Case: 15-60300 Document: 00516017191 Page: 6 Date Filed: 09/16/2021 No. 15-60300 A. 1. The Supreme Court has long held that “the omission of an element from a jury charge is subject to harmless-error analysis.” McFadden, 576 U.S. at 197. Its principal decision on this point is Neder v. United States, 527 U.S. 1 (1999). The defendant in Neder was prosecuted for lying on his federal income tax return. Id. at 6. To secure a conviction, the Government had to “prove that the defendant filed a tax return ‘which he d[id] not believe to be true and correct as to every material matter.’” Id. at 16 (quoting 26 U.S.C. § 7206(1)). But the trial court erroneously refused to submit the element of materiality to the jury. Id. at 8. That prompted two questions. First, is the failure to instruct the jury as to an element a structural error that “def[ies] analysis by harmless error standards”? Id. at 7 (quotation omitted). Second (and if not), what is the standard for harmless error in omitted-elements cases? Id. at 15. Neder answered the first question “no.” It explained that trial errors are structural “only in a very limited class of cases” where a constitutional defect “necessarily” infects “the entire trial process” and “render[s] a trial fundamentally unfair.” Id. at 8–9 (emphasis added) (quotation omitted). Incomplete jury instructions do not satisfy those stringent requirements because they do not automatically convert a trial into “an unreliable vehicle for determining guilt or innocence.” Id. at 9. Rather, in cases where the “overwhelming and uncontroverted evidence support[s]” the omitted element, the omission of the element will not “render a trial unfair.” Ibid. (quotation omitted). Turning to the second question, Neder held that “the harmless-error inquiry” is whether it is “clear beyond a reasonable doubt that a rational jury would have found the defendant guilty absent the error.” Id. at 18. The Court 6 Case: 15-60300 Document: 00516017191 Page: 7 Date Filed: 09/16/2021 No. 15-60300 took care to emphasize that this inquiry is not limited to the “evidence of guilt the jury . . . actually consider[ed].” Id. at 17 (emphasis omitted). Our task instead is to review the record “in typical appellate-court fashion” and determine whether a rational jury could find the Government failed to prove the omitted element. 2. The omitted element in this case is McFadden’s knowledge requirement. Based on the district court’s instructions, the jury found that Muhammad knew he was dealing a-PVP and 4-MEC. But the jury did not find that Muhammad knew those substances were in fact analogues. The question, then, is whether the record contains “uncontroverted evidence” establishing beyond a reasonable doubt that he did. Neder, 527 U.S. at 18. The answer to that question turns on whether Muhammad knew that a-PVP and 4-MEC were “controlled under the . . . Analogue Act” or, alternatively, whether he “knew the specific features” of a-PVP and 4-MEC that make them controlled-substance analogues. McFadden, 576 U.S. at 189. We conclude it is clear beyond a reasonable doubt that a rational jury would’ve found Muhammad knew the features that make a-PVP and 4-MEC controlled-substance analogues. Muhammad knew a-PVP and 4-MEC had substantially similar structures to controlled substances and that they produced a substantially similar high. See 21 U.S.C. § 802(32)(A). Muhammad researched and understood the chemical structures of the substances he sold. According to his trial testimony, he watched YouTube videos to find “the chemical structure of everything in . . . 3D form, so that when it’s in 3D form, you can see, like, animation . . . and you can see how it rotates and how it goes around and everything.” Muhammad would then compare the 3D images of the drugs he wanted to sell with images from known controlled substances to see “the difference between what is 7 Case: 15-60300 Document: 00516017191 Page: 8 Date Filed: 09/16/2021 No. 15-60300 considered substantially similar.” He even provided details regarding how this process helped him learn about specific chemicals listed in his indictment: For instance, in my case we have a thing called MAM2201 and AM2201. MAM2201 will put you out of business because it’s just no good. Like it’s only one ring difference. Right? It’s like one ring. And that one ring difference is what makes it . . . get on the analogue list. ... The web site talks about things being chemicals, and then it tells you that once you add a certain ring to it, the rings—and I don’t know what the rings are, but, like, let’s say you got AM2201. Then you got MAM2201. The M stands for an additional ring. That is more than enough to demonstrate Muhammad’s knowledge of the chemical structure of the drugs he sold and the controlled substances they mimicked. Other evidence adds to the heap. For example, Muhammad created a website to tell prospective customers that “[a]ll of our products have been tested for chemicals by independent labs” and that Muhammad would furnish copies of lab results upon request to confirm that none of the chemicals appeared in “[t]he new DEA law that was voted on March 1, 2011.” The same website lists a host of substances controlled under federal and state law and includes their chemical composition. A police officer, a coconspirator, and even Muhammad himself testified that Muhammad set up that website. Further, communications between Muhammad and a coconspirator reveal a comprehensive understanding of how a-PVP and 4MEC are structured. In sum, no rational jury could doubt Muhammad’s knowledge of the chemical structures of a-PVP, 4-MEC, and the similar 8 Case: 15-60300 Document: 00516017191 Page: 9 Date Filed: 09/16/2021 No. 15-60300 controlled substances he researched. See United States v. McFadden, 823 F.3d 217, 227–28 (4th Cir. 2016) (holding on remand from the Supreme Court’s McFadden decision that the error in the jury instructions was harmless because the record demonstrated the defendant’s “thorough and detailed knowledge of chemicals identified” in the indictment). Nor could a jury rationally doubt that Muhammad knew the “effect on the central nervous system” produced by a-PVP, 4-MEC, and the prohibited drugs he wanted to replicate. 21 U.S.C. § 802(32)(A)(ii); McFadden, 576 U.S. at 194. Muhammad admitted at trial that the purpose of his entire operation was “to get around the drug laws of the United States.” He testified that he would “start with a drug that is not banned” and move on when the drug schedules caught up. He did all of this to ensure that his customers—including a target market of “raves” and “colleges”—could continue getting high while “think[ing] they are doing it legally.” In keeping with the theme, one of his email accounts was called alley.legal.high@gmail.com. The only reasonable inference to draw from Muhammad’s testimony is that he was familiar with the effects of the drugs he sold and the substantially similar drugs he abandoned when they appeared on the controlled-substance schedules. And once again, chemistry-intensive communications with both customers and coconspirators back up Muhammad’s own testimony. No rational jury could look at this evidence and conclude that Muhammad lacked the requisite knowledge that his drugs had a “stimulant, depressant, or hallucinogenic effect on the central nervous system.” 21 U.S.C. § 802(32)(A)(ii); McFadden, 576 U.S. at 194. So the error in the district court’s jury instructions was harmless. 9 Case: 15-60300 Document: 00516017191 Page: 10 Date Filed: 09/16/2021 No. 15-60300 B. Muhammad does not meaningfully engage with any of this evidence. He contends instead that our decision in United States v. Stanford, 823 F.3d 814 (5th Cir. 2016), permits us to bypass the “thorough examination” of the “whole record” that Neder requires. 527 U.S. at 16, 19 (quotation omitted). We first review our decision in Stanford. Then we explain why it does not apply. 1. Stanford was an Analogue Act case much like this one. Daniel Stanford faced charges of conspiracy to distribute and to possess with intent to distribute a controlled-substance analogue. 823 F.3d at 822. Stanford contended before trial that the Government had to prove he knew the substance he trafficked was an analogue. Id. at 826. The district court disagreed and ultimately issued an instruction that omitted Stanford’s requested element. Id. at 826–27. Nevertheless, the court permitted Stanford to put on evidence regarding his lack of knowledge to preserve the issue in light of a circuit split. Ibid. It also submitted a special interrogatory to the jury that asked whether Stanford “knew that [the relevant substance] was a controlled substance analogue.” Id. at 827–28. The interrogatory said its “sole purpose” was to “assist the Court” and did not instruct the jury that its finding needed to be beyond a reasonable doubt. Id. at 828. The interrogatory did specify, however, that the jury’s answer “must be unanimous just as it is on the other questions on the verdict form.” Ibid. The jury convicted Stanford and answered “yes” to the question on the special interrogatory. Id. at 827. Then McFadden came down and clarified that a defendant’s knowledge that a drug is a controlled-substance analogue is indeed an element of an Analogue Act offense. That meant the trial court’s jury instructions were erroneous and that Stanford’s conviction could stand 10 Case: 15-60300 Document: 00516017191 Page: 11 Date Filed: 09/16/2021 No. 15-60300 only if the incomplete instructions were harmless. The Government contended they were, relying heavily on the jury’s affirmative answer to the special interrogatory. Id. at 828. In the Government’s view, “there was no need to guess as to whether a rational jury would have found [Stanford] guilty if the proper instructions were given because a rational jury did find that he met the additional element of the statute—that is, that he knew” the drug he trafficked was an analogue. Ibid. (quotation omitted). We disagreed and vacated Stanford’s conviction. Id. at 835. We started with Neder’s general observation that “[e]rroneous jury instructions are harmless if a court, after a thorough examination of the record, is able to conclude beyond a reasonable doubt that the jury verdict would have been the same absent the error.” Id. at 828 (quotation omitted); see Neder, 527 U.S. at 19. Then we held that the special interrogatory left room for reasonable doubt because (1) the interrogatory’s failure to state the standard of proof made it plausible that “the jury hastily answered the extra question without considering any degree of certainty,” id. at 832, and (2) the knowledge McFadden requires “was [not] inherent in the other elements that the jury actually found,” id. at 834. We also held the error was not harmless because the district court’s pretrial rejection of Stanford’s knowledge argument prevented him from presenting a complete defense. See id. at 835–38. 2. Muhammad derives two general principles from Stanford. First, he claims Stanford requires us to focus only on what the jury actually found and to disregard what a rational jury would have found had it been properly instructed. He says this principle helps him because the knowledge element was not inherent in the other elements the jury did find, so it’s possible the jury convicted him without focusing on it. Second, Muhammad suggests Stanford requires us to vacate a conviction whenever a defendant asserts that 11 Case: 15-60300 Document: 00516017191 Page: 12 Date Filed: 09/16/2021 No. 15-60300 he would’ve adopted a different defense strategy but for the trial court’s instructional error. He believes this principle helps him because the error foreclosed defense strategies that otherwise would have been available to him. He also contends the error influenced his choice of arguments to make and evidence to present at trial. Taken out of context, there is some language in Stanford that supports Muhammad’s arguments. See, e.g., id. at 835 (“[I]t is one thing for the government to look back now that the Court has provided the proper framework and pick out evidence that fits into that framework; it is another to assume that the jury focused on the same evidence, without the benefit of that framework . . . .”); id. at 838 (“We are left with [the] claim that [Stanford] would have structured his defense differently if he were aware that knowledge was a required element. That is not an unreasonable assertion.”). But Muhammad’s broad reading of Stanford squarely conflicts with Supreme Court precedent. And there are contextualized readings of Stanford that make much more sense. We therefore decline to adopt Muhammad’s interpretation. See United States v. Arizaga-Acosta, 436 F.3d 506, 508 (5th Cir. 2006) (per curiam) (“This court must follow [Supreme Court] precedent . . . unless and until the Supreme Court itself determines to overrule it.” (quotation omitted)). Start with the idea that Stanford limits our review of the record to evidence the jury actually considered. The defendant in Neder argued for precisely that standard. See 527 U.S. at 17 (recounting Neder’s argument that “[t]o rely on overwhelming record evidence of guilt the jury did not actually consider . . . would be to dispense with trial by jury and allow judges to direct a guilty verdict on an element of the offense”). But the Supreme Court refused to adopt it. See ibid. The Court held instead that we must “conduct a thorough examination of the record” to determine whether “a rational jury would have found the defendant guilty absent the error”—not whether the 12 Case: 15-60300 Document: 00516017191 Page: 13 Date Filed: 09/16/2021 No. 15-60300 actual jury did so find. Id. at 18–19 (emphases added). And for good reason. An actual-jury standard would convert all cases with potentially harmless instruction defects into structural-error cases that require automatic reversal. See id. at 17–18 (“[I]n the case of an omitted element, as the present one, the jury’s instructions preclude any consideration of evidence relevant to the omitted element, and thus there could be no harmless-error analysis.”). Since Neder “concluded that harmless-error analysis is appropriate” in this context, an actual-jury standard cannot be “the proper mode of analysis.” Id. at 18. 1 Far from rejecting Neder, Stanford embraced it. The latter began by recognizing that Neder “rejected the formal, [structural] approach in favor of a functional, case-by-case determination regarding whether an instruction error can be considered harmless.” Stanford, 823 F.3d at 831. Given the special interrogatory at the heart of the Stanford case, our “functional, caseby-case” inquiry obviously focused on that specific question and what the jury actually found in answering it. See id. at 828–29 (“[T]he government here contends that because the jury made a finding on the missing element of knowledge, any error in failing to include it as one of the elements . . . was harmless.”); id. at 835 (“[I]t is one thing for the government to look back now that the Court has provided the proper framework and pick out evidence that fits into that framework; it is another to assume that the jury focused on the same evidence, without the benefit of that framework, when it answered 1 Nor can Muhammad’s proposed inherent-in-the-proof-of-conviction standard be the proper mode of analysis. See Neder, 527 U.S. at 13 (rejecting a proposal to limit harmless-error analysis to cases “where other facts necessarily found by the jury are the functional equivalent of the omitted . . . element” (quotation omitted)); id. at 16–17 & n.1 (holding an omitted element was harmless due to “overwhelming evidence” on the issue without finding that “the jury verdict necessarily included a finding on that issue” (quotation omitted)). 13 Case: 15-60300 Document: 00516017191 Page: 14 Date Filed: 09/16/2021 No. 15-60300 the special interrogatory.” (emphasis added)); ibid. (“We do not know whether, in answering [the interrogatory], the jury credited the testimony to which the government directs us.” (emphasis added)). But it should be equally obvious that Neder prohibits us from focusing exclusively on the jury’s actual findings outside Stanford’s unique context. Muhammad’s reading of Stanford’s complete-defense discussion is similarly flawed. He seems to suggest that Stanford permits a defendant to escape uncontroverted evidence of guilt simply by asserting on appeal that he would’ve litigated things differently had the jury been instructed on the missing element. That’s far afield from the complete-defense right that the Supreme Court has articulated—a fact that Stanford itself recognized. See id. at 836 (“Cases involving a claim that the defendant was denied the right to present an adequate defense typically involve the court’s excluding certain testimony or evidence rather than a contention that the defendant would have changed his trial strategy if he had known a particular element was required.”). Further, if Muhammad is right, then once again Stanford becomes a case about structural error. Every defendant could always claim that he wants a do-over, and no omission will qualify as harmless error. This result would obviously contravene Supreme Court precedent. See Neder, 527 U.S. at 9 (holding that “an instruction that omits an element of the offense does not necessarily render a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence”); see also Stanford, 823 F.3d at 828, 831 (acknowledging that some instruction errors do not require reversal). In reality, three unusual features combined in Stanford to create the defense-based harm our court recognized. First, the district court ruled before trial that the Government didn’t need to prove the missing element. See 823 F.3d at 826. This enabled the defendant to make a colorable argument that the court’s ruling affected his trial strategy. An erroneous pretrial 14 Case: 15-60300 Document: 00516017191 Page: 15 Date Filed: 09/16/2021 No. 15-60300 instruction, after all, is especially likely to hamper a defendant’s ability to mount his defense. That’s because such rulings not only get the law wrong; by deciding the relevant issue at the outset, they also tend to steer the defense in the wrong direction from the start. See id. at 837 (“Stanford argues that the court’s ruling that knowledge was not an element of the drug conspiracy required him to adopt a different trial strategy.” (emphasis added) (quotation omitted)). 2 Second, the defendant in Stanford identified specific evidence that the pretrial ruling prevented him from introducing, and he explained how that evidence would have helped him. See ibid. This brought his argument closer to the heartland of the evidence-based complete-defense cases described above. And third, the defendant lacked notice of the governing legal standard because McFadden postdated his trial, and “he could not have been aware of the two Supreme Court-sanctioned methods for proving [the requisite] knowledge.” Id. at 838. This supported his argument that “he would have structured his defense differently if he were aware that knowledge was a required element.” Ibid. Muhammad’s complete-defense argument is a far cry from the one we embraced in Stanford. First, unlike in Stanford, the court did not make its jury-instruction ruling in this case until after trial. To the contrary, Muhammad prepared his case on the assumption that the Government would be bound by language in the indictment alleging that he knew the drugs he 2 That’s not to say an error is reversible on complete-defense grounds only if the court ruled on the relevant issue before trial. Our court’s complete-defense cases do not impose such a requirement. See, e.g., United States v. Ortiz-Mendez, 634 F.3d 837, 839 (5th Cir. 2011) (“A court commits a reversible error in failing to give an instruction proposed by the defense where (1) the requested instruction is substantially correct; (2) the requested issue is not substantially covered in the charge; and (3) the instruction concerns an important point in the trial so that the failure to give it seriously impaired the defendant’s ability to effectively present a given defense.” (quotation omitted)). But the pretrial timing of the ruling in Stanford is undeniably part of what made it a special case. 15 Case: 15-60300 Document: 00516017191 Page: 16 Date Filed: 09/16/2021 No. 15-60300 marketed were analogues. Muhammad even devotes several paragraphs of his briefing to a summary of how he contested that knowledge issue. In sharp contrast with Stanford, there is no reason to suspect the error had any prospective effect on the way Muhammad defended his case. Second, although Muhammad asserts that the McFadden error foreclosed otherwiseavailable defense strategies and influenced his other trial decisions, he offers nary a detail to explain how or why. Nor can we imagine those details based on our own review of the record. Finally, it’s of course true that Muhammad’s case was—like Stanford—decided before McFadden. That is precisely why the jury instructions omitted McFadden’s knowledge requirement. But this alone can’t suffice for harm: If it did, then the mere fact of an omitted element would be structural error. That would, yet again, contradict Neder. See 527 U.S. at 17–18 (holding the harmless-error rule applies “in the case of an omitted element,” even though “the jury’s instructions preclude any consideration of evidence relevant to [that] element”). Muhammad thus asks us to adopt a reading of Stanford that is both unnecessarily broad and flatly inconsistent with Supreme Court precedent. We decline to do so. Instead, we apply the Neder harmless-error standard on the same “functional, case-by-case” basis as usual. See Stanford, 823 F.3d at 831. On the basis of the record as a whole—and beyond a reasonable doubt— a jury could not rationally find that the Government failed to prove McFadden’s knowledge requirement. We therefore conclude that the omitted element in Muhammad’s jury instructions was harmless. 16 Case: 15-60300 Document: 00516017191 Page: 17 Date Filed: 09/16/2021 No. 15-60300 III. Muhammad raises four other arguments on appeal. None has merit. 3 First, Muhammad argues Task Force Agent Gibbons violated his constitutional rights by presenting false testimony to the grand jury. Even if grand jury testimony is false, the indictment should be dismissed only if (1) the Government knowingly sponsored the false testimony and (2) the false testimony was material—that is, “capable of influencing the grand jury’s decision.” United States v. Cessa, 861 F.3d 121, 142 (5th Cir. 2017) (quotation omitted). And Muhammad makes this argument for the first time on appeal, so our review is for plain error. That means Muhammad must show “(1) that the district court committed an error (2) that is plain and (3) affects his substantial rights and (4) that failure to correct the error would ‘seriously affect the fairness, integrity or public reputation of judicial proceedings.’” United States v. Sanchez-Hernandez, 931 F.3d 408, 410 (5th Cir. 2019) (quoting Johnson v. United States, 520 U.S. 461, 466–67 (1997)). Muhammad points to six statements, all made by Agent Gibbons before the grand jury. 4 Five of those statements were either true or made without knowledge of their falsity—so Muhammad’s arguments about those fail under Cessa. See 861 F.3d at 142. As for the sixth: Even if the statement 3 Muhammad’s reply brief might be read to raise some additional arguments for the first time. If that is correct, we will not consider them: Litigants may not raise arguments for the first time in a reply brief. See United States v. Jackson, 426 F.3d 301, 304 n.2 (5th Cir. 2005). 4 Most of Gibbons’s grand jury testimony is in the record on appeal, though in scattered form. Some of Gibbons’s testimony is missing, however. And other parts are included, but with the bottoms of the pages missing. This court generally does not consider evidence outside the record on appeal. See, e.g., McIntosh v. Partridge, 540 F.3d 315, 327 (5th Cir. 2008). For the sake of argument, we take Muhammad’s representations of Gibbons’s testimony at face value from his briefs. Even on that assumption, all of Muhammad’s arguments fail. 17 Case: 15-60300 Document: 00516017191 Page: 18 Date Filed: 09/16/2021 No. 15-60300 was material, false, and made with knowledge of its falsity, Muhammad hasn’t shown the statement affected his substantial rights. So that argument fails plain-error muster. Statement one: Agent Gibbons testified before the grand jury that Muhammad lived in Bridgeport, Connecticut. The record doesn’t show this statement was false—much less that Gibbons knew of its falsity. Instead, the record shows Muhammad was associated with several addresses in Bridgeport and nearby Shelton, Connecticut. Statements two through five are birds of a feather. Agent Gibbons testified before the grand jury that Muhammad and codefendant Roslyn Demetrius Chapman conspired to ship drugs to various States and to Canada. He testified that Muhammad was the source of various drugs distributed or possessed by Chapman. He testified that Chapman continued to receive drugs from Muhammad until April 22, 2014. And he testified that Muhammad was distributing methylone from December 2012 until the date of the indictment. Muhammad objects that each of these statements was based on a false premise: that Muhammad was operating an email account entitled mohammed1pooser@gmail.com. As Gibbons admitted at trial, that premise turned out to be false. Even so, Muhammad hasn’t shown any of Gibbons’s statements themselves were false—even if they were partially based on a false premise. To the contrary, the record provides abundant independent support for Gibbons’s statements. Further, even if the statements were false, the record doesn’t establish Gibbons knew that when he testified. It’s at best unclear when Gibbons learned Muhammad wasn’t mohammed1pooser. Because Muhammad can’t establish falsity—much less knowing falsity—these arguments don’t clear the Cessa hurdle. Statement six: Gibbons told the grand jury there was a detectable amount of methylone in the AM2201 that had been recovered from 18 Case: 15-60300 Document: 00516017191 Page: 19 Date Filed: 09/16/2021 No. 15-60300 Chapman’s car during a search. Lab reports later showed this statement was false. Yet even if Gibbons knew the truth all along, Muhammad can’t show plain error. The third plain-error prong, “in the ordinary case,” requires a showing of “a reasonable probability that, but for the error, the outcome of the proceeding would have been different.” Molina-Martinez v. United States, 136 S. Ct. 1338, 1343 (2016) (quotation omitted). Connecticut State Police found 487.2 grams of methylone in Muhammad’s bedroom closet on August 7, 2013. Given the strong evidence Muhammad had a large amount of methylone, there’s no reason to think Gibbons’s false testimony about a small amount of methylone affected the grand jury’s decision. So there’s no plain error. Second—and likewise raised for the first time on appeal— Muhammad argues a search warrant for the mohammed1pooser@gmail.com address was invalid, and that the resulting search and arrest therefore violated his constitutional rights. The nub of the argument: The warrant application was granted on the assumption that Muhammad was mohammed1pooser. But as discussed above, that wasn’t true. When it comes to search warrants, we grant relief only if the goodfaith exception to the exclusionary rule doesn’t apply. See United States v. Allen, 625 F.3d 830, 835 (5th Cir. 2010) (laying out the full two-step inquiry). The good-faith exception provides that “evidence obtained in objectively reasonable reliance on a subsequently invalidated search warrant typically should not be excluded.” United States v. Contreras, 905 F.3d 853, 857 (5th Cir. 2018) (quotation omitted). Relevant here, the exception is inapplicable “when the issuing magistrate was misled by information in an affidavit that the affiant knew or reasonably should have known was false.” Id. at 858 (quotation omitted) (listing three other triggers for the exception’s inapplicability). 19 Case: 15-60300 Document: 00516017191 Page: 20 Date Filed: 09/16/2021 No. 15-60300 The good-faith exception applies here because Gibbons reasonably believed Muhammad was mohammed1pooser when he applied for the warrant. At trial, Gibbons testified about drug activity associated with the mohammed1pooser address and explained that, at the time, he believed the address was Muhammad’s “because of the similar name, and one of the IP addresses hit in Bridgeport, Connecticut. I later learned . . . through the investigation that that was in fact false.” Further, recall that Muhammad didn’t contest this warrant until appeal. So even if the district court erred by failing to sua sponte invalidate the warrant, we would have to find that failure plainly erroneous before we could do anything about it. See SanchezHernandez, 931 F.3d at 410. Muhammad offers nothing to support such a finding. So this one isn’t even close. Muhammad’s third argument, again made for the first time on appeal, faults the court’s jury instructions about AM2201 and methylone. The judge instructed that, “at all times relevant to the charges in the indictment,” AM2201 and methylone were each Schedule I controlled substances. And therefore (said the judge) the Government did not need to prove those substances were on the schedule during the relevant timeframe. Muhammad points out that methylone wasn’t added to the schedule until October 21, 2011, and AM2201 wasn’t added until July 9, 2012. See Synthetic Drug Abuse Prevention Act of 2012, Pub. L. No. 112-144, 126 Stat. 1130, 1131 (2012) (adding AM2201 to Schedule I). But at trial, some of the Government’s evidence covered Muhammad’s dealings with these drugs before those dates. Therefore, Muhammad argues the judge should have instructed that, for the pre-scheduling period, methylone and AM2201 were controlled substances only to the extent they were analogues. As with the first two arguments, our review is for plain error. See Sanchez-Hernandez, 931 F.3d at 410. Muhammad’s argument fails because he hasn’t shown any error at all. See ibid. The earliest time listed in any count 20 Case: 15-60300 Document: 00516017191 Page: 21 Date Filed: 09/16/2021 No. 15-60300 of the superseding indictment is December 2012. And by that time, both methylone and AM2201 were on Schedule I. So the judge’s instruction— that the drugs were scheduled “at all times relevant to the charges in the indictment”—was a correct statement of the law. See United States v. Berrojo, 628 F.2d 368, 369–70 (5th Cir. 1980) (holding it is not plainly erroneous for a judge to correctly instruct a jury that a substance is controlled as a matter of law). The mere fact that some of the Government’s evidence involved prior activities does nothing to change that. Muhammad’s final argument, which he did raise below, is that the district court miscalculated his offense level under the Sentencing Guidelines. The district court relied on the Pre-Sentence Report (“PSR”) to arrive at a base offense level of 34. The PSR derived that offense level by finding Muhammad responsible for the equivalent of 10,634.11 kilograms of marijuana. See U.S.S.G. § 2D1.1(c)(3) (base offense level of 34 for “[a]t least 10,000 KG but less than 30,000 KG of Marihuana”), U.S.S.G. § 2D1.1, cmt. n.6 (“In the case of a controlled substance that is not specifically referenced in this guideline, determine the base offense level using the converted drug weight of the most closely related controlled substance referenced in this guideline.”). From there, the court found 12 levels of total enhancement, and then capped Muhammad’s offense level at the maximum level of 43. That score, combined with Muhammad’s criminal history, resulted in a Guidelines imprisonment range of 1,440 months, which is the sentence the court ultimately imposed. See U.S.S.G. § 5G1.2(b). Because Muhammad preserved his challenge to the Guidelines calculation, we review the “district court’s interpretation of the Sentencing Guidelines de novo and its factual determinations for clear error.” United States v. Garza, 587 F.3d 304, 308 (5th Cir. 2009). We find no error. Muhammad says the court erred by partially basing its sentence on behavior that occurred before the date listed in the 21 Case: 15-60300 Document: 00516017191 Page: 22 Date Filed: 09/16/2021 No. 15-60300 indictment. Our precedent says otherwise. See United States v. McCaskey, 9 F.3d 368, 375 (5th Cir. 1993) (“[D]rug transactions occurring before the precise time frame of the conspiracy for which a defendant is convicted may be considered for sentencing purposes if those transactions otherwise satisfy the criteria for relevant conduct prescribed by the guidelines.”). Muhammad also says the district court wrongly held him responsible for drugs sold via an account called legal.high@yahoo.com. But the district court based its factual findings partly on the PSR, which contained an in-depth discussion of the account’s activities. See United States v. Harris, 702 F.3d 226, 230 (5th Cir. 2012) (“Generally, a PSR bears sufficient indicia of reliability to be considered as evidence by the sentencing judge in making factual determinations.” (quotation omitted)). And the court specifically found that Muhammad’s efforts to contradict the PSR were not credible. We find no error in the district court’s decision. AFFIRMED. 22
Primary Holding
The Fifth Circuit concluded that there is more than enough to demonstrate defendant's knowledge of the chemical structure of the drugs he sold and the controlled substances they mimicked; the omitted element in defendant's jury instruction was harmless.

Disclaimer: Justia Annotations is a forum for attorneys to summarize, comment on, and analyze case law published on our site. Justia makes no guarantees or warranties that the annotations are accurate or reflect the current state of law, and no annotation is intended to be, nor should it be construed as, legal advice. Contacting Justia or any attorney through this site, via web form, email, or otherwise, does not create an attorney-client relationship.