United States v. Steve Pritchard

964 F.3d 513

• Court: Court of Appeals for the Sixth Circuit
• Decided: July 7, 2020
• Docket: 18-6210
• Status: Published

Opinion

RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 20a0203p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

UNITED STATES OF AMERICA, ┐ Plaintiff-Appellee, │ │ > No. 18-6210 v. │ │ │ STEVE ALLEN PRITCHARD, │ Defendant-Appellant. │ │ ┘

Appeal from the United States District Court for the Western District of Kentucky at Bowling Green. No. 1:16-cr-00028-1—Gregory N. Stivers, District Judge.

Argued: October 23, 2019

Decided and Filed: July 7, 2020

Before: CLAY, THAPAR, and NALBANDIAN, Circuit Judges. _________________

COUNSEL

ARGUED: Frank W. Heft, Jr., OFFICE OF THE FEDERAL DEFENDER, Louisville, Kentucky, for Appellant. Terry M. Cushing, UNITED STATES ATTORNEY’S OFFICE, Louisville, Kentucky, for Appellee. ON BRIEF: Frank W. Heft, Jr., Donald J. Meier, OFFICE OF THE FEDERAL DEFENDER, Louisville, Kentucky, for Appellant. Monica Wheatley, UNITED STATES ATTORNEY’S OFFICE, Louisville, Kentucky, for Appellee.

NALBANDIAN, J., delivered the opinion of the court in which CLAY and THAPAR, JJ., joined. CLAY, J. (pp. 22–24), delivered a separate concurring opinion. No. 18-6210 United States v. Pritchard Page 2

_________________

OPINION _________________

NALBANDIAN, Circuit Judge. Some men just want to watch the world burn. Others start fires to collect insurance money. Steve Pritchard is the latter. But after playing with fire several times, Pritchard’s penchant for profiting from arson took a deadly turn. Instead of only damaging property, a fire started by Pritchard in June 2011 led to firefighter Charles Sparks’s death. While responding to the fire, Sparks suffered a fatal heart attack. Although Pritchard could have supposed that a firefighter would respond to the blaze, he had no reason to suspect that Charles Sparks would arrive on the scene, bringing with him a history of coronary disease and spotty use of his prescription medication. At issue is whether Pritchard caused Sparks’s death within the meaning of the federal arson statute, 18 U.S.C. § 844(i).

Pritchard’s appeal turns on first principles of causation. The common law typically permits liability only when the perpetrator acts as both the but-for and the legal cause of the harm. And that distinction often matters. Every fledgling law student knows that “a kingdom might be lost ‘all for the want of a horseshoe nail,’” see Massachusetts v. EPA, 549 U.S. 497, 546 (2007) (Roberts, C.J., dissenting), but we still don’t hold the blacksmith responsible for the defeat despite his faulty craftmanship’s role as the but-for cause. For that reason, laws that invoke proximate causation generally impose liability when the harm was foreseeable. Under 18 U.S.C. § 844(i), Sparks’s death need only be “a direct or proximate result of [Pritchard’s] conduct.” Because the government introduced testimony showing how firefighting can trigger a heart attack, we find that the jury had sufficient evidence to hold Pritchard responsible for Sparks’s death. Pritchard also alleges that the lower court committed various reversible errors, including wrongly admitting propensity evidence, denying his motion to suppress cell phone data seized by the government, and applying an unwarranted sentencing enhancement. Those arguments also lack merit. Thus, we AFFIRM. No. 18-6210 United States v. Pritchard Page 3

I.

This case involves a fire, a firefighter, and a fraudster. First, the fire. At 3:05 AM on June 30, 2011, a 911 caller reported a fire at the Pritchard residence. Eleven minutes later, firefighters, including Assistant Chief Charles Sparks, arrived on the scene. These firefighters found the house engulfed in “[a] lot of fire, a lot of flames.” (R. 145, Trial Tr. Vol. 2 at PageID# 1055.) During the firefighting, Sparks lost consciousness. Personnel at the scene administered CPR to Sparks and called for an ambulance. Although the emergency workers tried to revive Sparks, who was not breathing and did not have a pulse, he never regained consciousness. Eight days later, Sparks died after being taken off life support.

Before fighting the fire on June 30, 2011, Sparks suffered from various medical ailments. In 2005, Sparks had a heart attack that required inserting a stint in his coronary artery. Sparks also suffered from coronary disease, hypertension, and diabetes—all of which are risk factors for a heart attack. Around the time of his death, Sparks had not been taking his prescribed heart medication or insulin to treat his diabetes.

Brandi Pritchard, Steve Pritchard’s wife, clocked into work at 3:34 AM on June 30, 2011. She worked about forty-five minutes away from her house, meaning she left for work around 2:45 AM. Around 7:00 AM that morning, Brandi received a call informing her that her home was on fire.

Now for the fraud. To Brandi’s purported surprise, Pritchard decided to drive her to work that morning, telling her that it “would be a good morning to go ahead and start . . . this fire in this house[.]” (R. 146, Trial Tr. Vol. 3 at PageID# 1352.) On the way to work, Pritchard proclaimed, “I did it[,]” referring to the fire. (Id. at 1353.) Pritchard had arranged for Brandi’s children and his dog to be out of the house that morning. Later, Brandi’s children would testify that Pritchard showed them photographs of the fire he took from his phone and that Pritchard implied he started the fire. And Dale Clark, a longtime acquaintance of Pritchard and Brandi, also testified that Pritchard insinuated that he started the fire with Brandi’s help. Finally, Sherry Waggoner, a friend of the Pritchards, recounted overhearing the couple argue about the fire. She testified that Pritchard claimed to be “the genius in the project [and] . . . the one that master No. 18-6210 United States v. Pritchard Page 4

planned it.” (R. 145, Trial Tr. Vol. 2 at PageID# 1143.) Sherry’s husband, Robbie Waggoner, confirmed this conversation and that Pritchard took credit for the arson.

Later that day, Pritchard picked Brandi up from work and instructed her on “what [she] needed to say when [they] got [home].” (R. 146, Trial Tr. Vol. 3 at PageID# 1356.) Pritchard wanted Brandi to tell investigators that he had spent the night in Louisville. And that’s the same story he gave to the police. Yet Pritchard’s alibi did not withstand scrutiny. Cell tower records revealed that Pritchard had not been in Louisville on the night of the fire. The government obtained Pritchard’s cell-site location information (CSLI) without a warrant in July 2015, relying on the warrant exception in the Stored Communications Act (SCA).

June 30 wasn’t the first time Brandi heard about Pritchard’s plot. Brandi bought a $50,000 insurance policy for the house six days before the fire. So when Pritchard learned about the insurance policy, he remarked about “how easy it would be to burn the house down and get the money for it.” (Id. at 1335.) At first, Brandi balked at Pritchard’s plan, finding it “ludicrous[.]” (Id. at 1343.) To persuade Brandi, Pritchard recounted previous fires that he started to collect insurance money, including burning down a house owned by Tena Gowen, the couple’s close friend. He also mentioned setting a car belonging to Whitney Clark, his niece, on fire to collect insurance money. And Pritchard discussed setting his own car on fire, as well as the home of David Newcomb.

After the fire, Brandi collected on her insurance policy. In the ensuing investigation of the fire, carried out by Kentucky State Police and the FBI, Pritchard pressured Brandi to lie to cover up the arson. When he became worried about Brandi’s resolve, Pritchard resorted to threats of violence to coerce Brandi not to confess. In response, Brandi obtained multiple protective orders against Pritchard. Pritchard’s threats continued for years; after the FBI interviewed Pritchard in 2014, he broke into Brandi’s house and threatened to harm her and her children if she told the truth about the arson. Brandi eventually confessed to her role in the arson, the insurance fraud, and the cover up.

Following the investigation, a federal grand jury charged Pritchard and Brandi with Malicious Destruction of Property by Fire, a violation of 18 U.S.C. § 844(i) that permits No. 18-6210 United States v. Pritchard Page 5

punishment for arson causing death, and Mail Fraud under 18 U.S.C. § 1341. Brandi pleaded guilty and Pritchard went to trial.

At Pritchard’s trial, Dr. Thomas Hales, a physician for the National Institute for Occupational Safety and Health, testified about risk factors relevant to Sparks’s death. Hales concluded that firefighting “triggered” Sparks’s fatal heart attack. (Id. at 1313.) He also testified that firefighting requires “a lot of heavy physical exertion” and recounted how studies show a link between strenuous physical activity and cardiac malfunction. (Id. at 1258–64.) But he could not conclude, given Sparks’s history, that Sparks would not have had a heart attack independent of the fire. Dr. Divyesh Bhakta, who treated Sparks after his heart attack, testified that Sparks’s untreated diabetes and failure to take his heart medicine produced “a number of risk factors[.]” (R. 134, Trial Tr. Vol. 5 at PageID# 728–40.) He also could not conclude whether the fire, and not these risk factors, caused Sparks’s heart attack. After hearing this evidence, the jury found Pritchard guilty of arson causing death.

Pritchard’s counsel made many objections before, during, and after trial. Before the trial, he filed an unsuccessful motion to exclude evidence about Pritchard’s previous arsons and the Emergency Protective Orders (EPOs) obtained by Brandi. During trial, he objected to the court’s refusal to give his proposed jury instructions about proximate cause. He also argued that the district court should not have admitted CSLI seized by the government because the officers lacked a valid warrant for the search. So Pritchard moved to suppress, contending that admitting the cell phone records violated the exclusionary rule. In addition, Pritchard’s counsel challenged the relevancy of testimony given by the government’s expert witness, David West, about the characteristics and motives shared by arsonists. And during sentencing, Pritchard’s counsel argued against applying a two-level leadership enhancement to Pritchard’s sentence. The lower court overruled these challenges.

In sum, Pritchard received a sentence of 360 months for arson causing death and a concurrent 240-month term for mail fraud. Pritchard now appeals his conviction and sentence. No. 18-6210 United States v. Pritchard Page 6

II.

An arsonist started a fire to collect insurance money. Then a firefighter lost his life putting that fire out. Before us is whether the firefighter’s death was the “direct or proximate result” of the arsonist’s actions, the predicate necessary for guilt under 18 U.S.C. § 844(i). While ordinarily a straightforward question, this case poses a wrinkle: The firefighter had a history of cardiac disease and passed away from a heart attack suffered during the fire. Pritchard, who started the fire, tells us that he isn’t responsible for the firefighter’s existing heart condition. But the government argues that Pritchard’s arson set in motion foreseeable events where a firefighter could lose his life. In short, the parties ask us to decide the appropriate causation standard for arson causing death under 18 U.S.C. § 844(i).

As we must, we begin with the statute’s text. See Binno v. Am. Bar Ass’n, 826 F.3d 338, 346 (6th Cir. 2016). Section 844(i) permits punishment when “death results to any person, including any public safety officer performing duties as a direct or proximate result of conduct prohibited by this subsection[.]” 18 U.S.C. § 844(i) (emphasis added). Because the statute covers unlawfully destroying a building or other property with fire, § 844(i) covers injuries caused by Pritchard’s arson. See id. So this case turns on when, exactly, death is “a direct or proximate result” of arson.

Courts traditionally answer questions of causation by looking to the common law. See United States v. Hayes, 589 F.2d 811, 821 (5th Cir. 1979) (“We adhere to the accepted practice among federal courts in construing a federal criminal statute where specific terms are left undefined. We give those terms their common law meaning.” (citing United States v. Turley, 352 U.S. 407, 430 (1957))). Addressing common law causation principles, the Supreme Court remarked: “The law has long considered causation a hybrid concept, consisting of two constituent parts: actual cause and legal cause.” Burrage v. United States, 571 U.S. 204, 210 (2014) (citing H. Hart & A. Honore, Causation in the Law 104 (1959)). As with § 844(i), “[w]hen a crime requires ‘not merely conduct but also a specified result of conduct,’ a defendant generally may not be convicted unless his conduct is ‘both (1) the actual cause, and (2) the “legal” cause (often called the “proximate cause”) of the result.’” Id. (quoting 1 W. LaFave, Substantive Criminal Law § 6.4(a), pp. 464–66 (2d ed. 2003)). Just like the statute in Burrage, No. 18-6210 United States v. Pritchard Page 7

§ 844(i) “does not define” its terms describing causation. See id. So following the framework set out by the Supreme Court, it likely makes sense that the statute contemplates actual cause when invoking direct cause. That’s because pairing actual cause and proximate cause would reflect common law causation principles. But we need not address the exact meaning of “direct cause” under § 844(i) because this case can be resolved on proximate causation grounds.1

While proximate cause lacks a “precise definition,” proximate cause language generally “inject[s] a foreseeability element into the statute.” Babbitt v. Sweet Home Chapter of Cmtys. for a Great Or., 515 U.S. 687, 713 (1995) (O’Connor, J., concurring). And this Circuit understands proximate cause as permitting liability even when a defendant’s “acts were not the immediate cause of the victim’s death or injury.” United States v. Martinez, 588 F.3d 301, 319 (6th Cir. 2009) (quoting United States v. Guillette, 547 F.2d 743, 749 (2d Cir. 1976)). Typically, proximate cause presents a narrower range of liability than an actual, or direct, cause. That’s because it acts as a:

limitation which the courts have placed upon the actor’s responsibility for the consequences of the actor’s conduct . . . [because] the causes of an event go back to the dawn of human events and beyond. But any attempt to impose

1To avoid redundancy, the text’s use of both types of causation in the alternative implies that they each have a different meaning or add something unique to the legal analysis. See Rimini St., Inc. v. Oracle USA, Inc., 139 S. Ct. 873, 881 (2019) (“If one possible interpretation of a statute would cause some redundancy and another interpretation would avoid redundancy, that difference in the two interpretations can supply a clue as to the better interpretation of a statute.”). Although the rule against surplusage is not always “a silver bullet[,]” the text here, notably its use of the word “or,” provides two distinct types of causation. Id. That drafting decision suggests that we should permit liability under § 844(i) for either direct or proximate causation. But the government does not separate the concepts of direct and proximate causation in its briefing, choosing instead to contend generally that Pritchard’s actions were the “direct or proximate cause” of Sparks’s death. To be fair, some formulations of proximate cause include a concept of directness within them. For instance, Black’s Law Dictionary provides this secondary definition of proximate cause: “A cause that directly produces an event and without which the event would not have occurred.” Cause, Black’s Law Dictionary (11th ed. 2019) (emphasis added). And to further add to the confusion, the Restatement eschews the usage of the term proximate cause entirely and finds the phrase inadequate to explain the general concept of limitations on scope of liability as a result of remoteness or lack of foreseeability. See Restatement (Third) of Torts: Liability for Physical and Emotional Harm ch. 6 Special Note on Proximate Cause (“Although the term ‘proximate cause’ has been in widespread use in judicial opinions, treatises, casebooks, and scholarship, the term is not generally employed in this Chapter because it is an especially poor one to describe the idea to which it is connected.”); see also id. § 29 cmt. b (“[T]he term ‘proximate cause’ is a poor one to describe limits on the scope of liability. It is also an unfortunate term to employ for factual cause or the combination of factual cause and scope of liability.”). In any event, our analysis focuses on what the Supreme Court and our Circuit have generally said about proximate cause. No. 18-6210 United States v. Pritchard Page 8

responsibility upon such a basis would result in infinite liability for all wrongful acts.

W. Page Keeton et al., Prosser and Keeton on the Law of Torts § 41, at 264 (5th ed. 1984).

Setting a necessary limit for imposing liability, this Circuit finds proximate causation when a person suffers harm from events that “are foreseeable and naturally result from [the defendant]’s criminal conduct[.]” Martinez, 588 F.3d at 319 (alteration in original) (quoting Guillette, 547 F.2d at 749). In sum:

A fundamental principle of criminal law is that a person is held responsible for all consequences proximately caused by his criminal conduct. Thus, where events are foreseeable and naturally result from one’s criminal conduct, the chain of legal causation is considered unbroken and the perpetrator is held criminally responsible for the resulting harm.

United States v. Wiegand, No. 93-1735, 1994 WL 714347, at *2–3 (6th Cir. Dec. 22, 1994) (quoting Hayes, 589 F.2d at 821).

Applying that proximate causation principle, this Circuit held that “[i]njury to a firefighter is a foreseeable result of arson[.]” Id. at *3. But that applied to burns suffered during firefighting. Id. at *1. Pritchard rightly observes that the facts here are different. Rather than perishing from typical injuries sustained during firefighting, such as burns and trauma from falling debris, Charles Sparks died from a heart attack. And evidence at trial showed that Sparks had a pre-existing heart condition, along with other maladies, that he left untreated. That means, according to Pritchard, Sparks’s failure to treat his condition and decision to keep fighting fires with a diseased heart broke the chain of causation between Pritchard’s fire and Sparks’s death.

Drawing on that observation, Pritchard contends that insufficient evidence supports his conviction because the government did not show enough causation between Pritchard’s arson and Sparks’s heart attack. First, he contends that his actions needed to be “the cause of death,” and not simply “a cause of death.” (Appellant’s Br. at 20.) But Pritchard’s theory that there can only be one cause of death sufficient for imposing liability flouts our precedent. See Hausrath v. N.Y. Cent. R.R. Co., 401 F.2d 634, 637 (6th Cir. 1968) (“There may, of course, be more than one proximate cause.”). Still, Pritchard argues that Sparks’s “heart attack could have occurred at home due to diabetes or coronary artery disease” and the arson only acted as a “contributing No. 18-6210 United States v. Pritchard Page 9

factor” that didn’t set the events of Sparks’s death in motion. (Appellant’s Br. at 22–23.) Rephrased, Pritchard tells us that a firefighter dying from a heart attack wasn’t foreseeable when he started the fire and there is not an unbroken causal link between starting the fire and Sparks’s death. And we agree that this case isn’t as clear cut as Wiegand, where we held an arsonist liable for causing the burns suffered by a firefighter because that injury was foreseeable and the causal chain between the arson and the burns appeared unbroken. 1994 WL 714347, at *1–3.

Still, a defendant raising a sufficiency of the evidence claim faces a high burden. That’s because a court must determine “whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” United States v. Porter, 886 F.3d 562, 565 (6th Cir. 2018) (quoting Jackson v. Virginia, 443 U.S. 307, 319 (1979)). It’s true that Sparks suffered from serious ailments, including diabetes and blocked arteries, and evidence showed that Sparks had not been taking his prescribed heart medication or insulin for some time. So Pritchard contends that the presence of these “other factors that could have been the cause” of Sparks’s heart attack means the fire could not have proximately caused Sparks’s death. (Appellant’s Br. at 22.) But in Martinez, we upheld a finding of proximate cause when a doctor’s improper treatment contributed to a patient’s fatal overdose. 588 F.3d at 319–21. There, we denied a similar insufficiency of the evidence claim. The jury found prescribing over the counter drugs and performing nerve-block injections proximately caused the patient’s fatal overdose, even though the patient had used narcotic drugs. Id. at 321–22. And we found those facts sufficient to affirm a conviction.

Martinez’s reasoning applies here. Even though Pritchard offered evidence that Sparks’s pre-existing medical condition led to his death, the jury found an unbroken chain of causation between the fire and Sparks’s heart attack. Just as we did not disturb a jury’s finding that the patient’s “actions as an addict cannot be said to break the chain of proximate causation[,]” we do not disturb the jury’s conclusion that Sparks’s medical ailments did not sever the causal link between the fire and his death. Id. at 322. After all, the government entered evidence showing that firefighting stresses the heart, meaning that heart attacks can be caused by responding to arsons just like severe burns. And to show proximate cause, the government only needed to No. 18-6210 United States v. Pritchard Page 10

enter sufficient evidence that Sparks’s death was a foreseeable and natural result of Pritchard’s actions. Even though a heart attack might not be the most obvious injury for a firefighter to suffer, the government entered evidence showing that firefighting strains the heart enough to possibly trigger a heart attack. So Pritchard isn’t really saying the jury lacked evidence supporting its conviction; his actual complaint is that the jury sided with the government. In short, Pritchard’s sufficiency of the evidence claim asks us to substitute our view for the jury’s, which we refrain from doing. See Ratliff v. Wellington Exempted Vill. Schs. Bd. of Educ., 820 F.2d 792, 795 (6th Cir. 1987).

Despite listing proximate and direct causes as distinct terms, § 844(i) permits liability for an arson that either proximately or directly causes a firefighter’s death. Because the jury relied on evidence showing Pritchard proximately caused Sparks’s death, a direct causation analysis is unnecessary. We affirm Pritchard’s conviction for arson causing death.

III.

Pritchard not only challenges his conviction by offering a causation theory, but also applies that principle to the jury instructions given over § 844(i). This argument repackages his complaint that the court below relied on an overly broad understanding of causation. That’s because Pritchard believes that he could only be guilty if he was the cause of the death, and not a cause of the death. Just like he argued that the trial evidence didn’t support a finding that he acted as more than a contributing factor in Sparks’s death, Pritchard complains that the district court provided jury instructions permitting a conviction if Pritchard’s actions were a cause of death and not the cause of death. He made that distinction in his proposed jury instructions, which the district court did not adopt. We review the legal accuracy of jury instructions de novo. United States v. Roth, 628 F.3d 827, 833 (6th Cir. 2011). But a district court’s refusal to give an instruction requested by the defendant must amount to abuse of discretion in order for us to vacate a judgment. Id.

At trial, the judge gave this instruction to the jury:

The death of Charles Sparks resulted from the Defendant’s conduct if the Defendant’s conduct was a direct and proximate cause of Charles Sparks’ death. The Defendant’s conduct was a direct and proximate cause of Charles Sparks’ No. 18-6210 United States v. Pritchard Page 11

death if it was a substantial factor in causing him to die, and he would not have died except for the Defendant’s conduct.

(R. 101, Jury Instr. at PageID# 364.) Pritchard alleges that the trial court erred by rejecting his proposed instructions under which he needed to be “the direct and proximate cause” of Sparks’s death. (Appellant’s Br. at 25.) So, according to Pritchard, the jury convicted him under an overly permissive causation standard. That’s doubly true, Pritchard tells us, because the instruction permitted guilt for being a substantial factor in Sparks’s death. This alleged mistake made it harder for Pritchard to argue that guilt could only be found under § 844(i) if Sparks’s firefighting, and not his medical condition, triggered the heart attack. In short, Pritchard contends that the instruction’s wording wrongly permitted guilt if Pritchard was a cause of death, and not the cause of death. But his proposed jury instruction didn’t track the statutory language, which unambiguously covers injuries that are “a direct or proximate result of conduct prohibited by this subsection[.]” 18 U.S.C. § 844(i) (emphasis added).

Because Pritchard challenges the district court’s “failure to give the requested instruction[,]” we review for abuse of discretion. (Appellant’s Br. at 25–26.) And we only find that a district court abused its discretion over jury instructions “if the instructions, viewed as a whole, were confusing, misleading and prejudicial.” Williams v. Eau Claire Pub. Schs., 397 F.3d 441, 445 (6th Cir. 2005) (quoting Kitchen v. Chippewa Valley Schs., 825 F.2d 1004, 1011 (6th Cir. 1987)). What’s more, a proposed jury instruction must be “a correct statement of law[.]” Id. In short, Pritchard’s proposed jury instruction departed from the language of § 844(i), yet now he complains about the district court’s decision to issue an instruction that more closely mirrored the statute.2 We cannot conclude the district court abused its discretion in

2We note that the challenged jury instruction isn’t without warts. To start, § 844(i) creates criminal liability for “a direct or proximate” cause and not “direct and proximate” causation, as stated by the district court’s instructions. Unsurprisingly, Pritchard didn’t make that observation because the district court’s language implies a more stringent causal requirement. But we also note that the district court relied on “substantial factor” and but-for causation concepts without expanding on those terms. So the jury arguably did not receive proper instructions on the meaning of “direct or proximate” cause as discussed in Section II above. We believe that proper instruction on this issue touches on foreseeability and breaking the causal chain. A good example comes from Martinez. There, the judge informed the jury that: Proximate or direct cause exists where the acts of the Defendant . . . in a natural and continuous sequence directly produces the deaths and without which they would not have occurred. The Defendant is not responsible . . . if his alleged [criminal act] is a remote cause and No. 18-6210 United States v. Pritchard Page 12

denying Pritchard’s proposed instruction in favor of language more faithful to the statute. So Pritchard’s challenge to the jury instruction falls short.

IV.

During trial, the district court admitted evidence that (1) Pritchard set four previous fires; (2) his wife, Brandi, obtained Emergency Protective Orders (EPOs) against him; and (3) opined about what motivates arsonists. Pritchard believes these admissions violated the Federal Rules of Evidence. So he requests a new trial where the jury would not be exposed to this allegedly inadmissible and prejudicial evidence.

We typically review challenges to evidentiary admissibility for abuse of discretion. See United States v. Mack, 258 F.3d 548, 553 n.1 (6th Cir. 2001) (“[W]e conclude that the appropriate standard of review for the admissibility of evidence under Rule 404(b) is ‘abuse of discretion.’”); United States v. Collins, 799 F.3d 554, 572 (6th Cir. 2015) (“We review a district court’s admission or exclusion of evidence for abuse of discretion.”). For challenges to the introduction of evidence under Federal Rule of Evidence 404(b), we have also used a three-step process:

[W]e first review for clear error the district court’s factual determination that the “other . . . acts” occurred. Second, we examine de novo the district court’s legal determination that the evidence was admissible for a legitimate purpose. Finally, we review for abuse of discretion the district court’s determination that the probative value of the other acts evidence is not substantially outweighed by its unfairly prejudicial effect.

not a proximate or direct cause of the deaths. A cause or condition is remote when the result could not have been reasonably foreseen or anticipated as being the likely cause of the deaths. The causal connection is broken when another’s act, which could not have been reasonably foreseen and is fully independent of the Defendant’s alleged [criminal act], intervenes and completely breaks the effect of Defendant’s conduct and such other acts become the proximate cause of the deaths. J.A. Volume II at 807–08, Martinez, 588 F.3d 301 (Nos. 06-3882, 06-4206). Although this jury instruction applies to a statute with different language than § 844(i), it touches on key causation concepts absent from the instruction used in Pritchard’s trial. That said, Pritchard doesn’t complain about any error in the instructions aside from the court’s decision to use a instead of the. Even though appellate courts are not “hidebound by the precise arguments of counsel” made below, we cannot stage a “takeover of the appeal.” United States v. Sineneng-Smith, 140 S. Ct. 1575, 1581 (2020). So we decline to consider whether any unchallenged flaws in the jury instructions impact Pritchard’s appeal. No. 18-6210 United States v. Pritchard Page 13

United States v. Merriweather, 78 F.3d 1070, 1074 (6th Cir. 1996) (second alteration in original). Although previous rulings from this Court suggest “an intra-circuit split” over the standard of review for Rule 404(b) evidence, using the three-step test and reviewing for abuse of discretion are not necessarily inconsistent. See United States v. Mandoka, 869 F.3d 448, 456–57 (6th Cir. 2017). And both standards produce the same result for Pritchard, so we need not delve into that ambiguity. Finally, we are mindful that harmless evidentiary errors do not permit us to vacate a conviction. United States v. Hardy, 643 F.3d 143, 153 (6th Cir. 2011).

A.

Pritchard argues that the district court erred by admitting evidence about previous fires he started because that evidence only shows a propensity for criminality. And that legal proposition is true: The government cannot introduce past crimes as evidence of the defendant’s propensity to commit a crime at issue. United States v. Clay, 667 F.3d 689, 693–94 (6th Cir. 2012). That’s because showing propensity to commit a crime, either as to a specific offense or in general, unfairly prejudices the jury over the defendant’s guilt. See Old Chief v. United States, 519 U.S. 172, 180–81 (1997) (“[I]mproper grounds [for admitting evidence] certainly include . . . generalizing a defendant's earlier bad act into bad character and taking that as raising the odds that he did the later bad act now charged (or, worse, as calling for preventive conviction even if he should happen to be innocent momentarily).”).

That said, Federal Rule of Evidence 404(b) provides exceptions for introducing a defendant’s prior crimes or bad acts for non-propensity purposes, including: “proving motive, opportunity, intent, preparation, plan, knowledge, identity, absence of mistake, or lack of accident.” Just because evidence, such as a past conviction, might not be admissible to show propensity doesn’t automatically mean the district court erred by admitting it. In short, the question becomes whether the fires Pritchard previously set fall under a recognized propensity evidence exception.

Pritchard’s previous arsons played a role in his plan to set fire to his wife’s house. To convince Brandi to join his plot, he told her about his previous arsons. So Pritchard used his prior bad acts to influence Brandi. This scenario is much like United States v. Holley, 57 F. No. 18-6210 United States v. Pritchard Page 14

App’x 639 (6th Cir. 2003) (order), where the government introduced evidence of the defendant’s past arson in proving a second arson. Just like Pritchard, the defendant in Holley bragged about his previous arson to “recruit” a co-arsonist. Id. at 641. For that reason, we found that evidence of a past arson fits into the Rule 404(b) “preparation and planning” exceptions. Id. Although Pritchard’s past arsons might look like propensity evidence at face value, his reliance on his previous bad acts to attract a co-arsonist and develop his insurance fraud plan falls under a Rule 404(b) exception.

That gives us enough to perform the Merriweather test. First, we review the district court’s conclusion that the acts occurred for clear error. Plenty of testimony supports Pritchard committing the previous arsons and the record lacks evidence to the contrary. So there’s no clear error in finding that Pritchard’s prior bad acts took place. Next, we consider de novo whether the evidence should have been admitted under Federal Rule of Evidence 404(b). As discussed above, the previous arsons fall under a recognized Rule 404(b) exception to the prohibition on propensity evidence. Finally, we ask if the prejudicial effect of introducing the previous arsons substantially outweighed the probative value of doing so. Like the district court, we recognize evidence that Pritchard committed arson in the past might prejudice a jury. But we cannot deny the probative value of showing Pritchard’s thought process in plotting the arson and his attempt to recruit Brandi as his aide. Because the prejudicial effect of this admission must substantially outweigh the probative value under Merriweather, Pritchard loses under this prong too. All considered, we cannot conclude that the district court erred by admitting Pritchard’s past arsons as evidence.

B.

As for the three EPOs obtained by Brandi, Pritchard contends that they are irrelevant and thus inadmissible under Federal Rule of Evidence 401. In short, he argues the EPOs unfairly portrayed him as a “an abuser,” which has nothing to do with his guilt for the arson that caused Sparks’s death. (Appellant’s Br. 35.) Pritchard tells us that Brandi obtained: (1) the September 2011 EPO because of a drunken threat; (2) the February 2013 EPO over a threat at knife point; and (3) the 2014 EPO because Pritchard broke into her house and threatened her. (Id.) From that perspective, none of those EPOs relates to the fire. It would follow that they didn’t make No. 18-6210 United States v. Pritchard Page 15

any fact in consequence more or less probable at trial, rendering the EPOs irrelevant (and thus inadmissible) under Federal Rule of Evidence 401.

Yet this argument ignores why Brandi might have sought the EPOs. Because Brandi knew the truth about the fire, Pritchard allegedly threatened her throughout the investigation to prevent her from confessing. And all of the EPOs occurred during a period where Pritchard had been threatening Brandi, as well as her children, to prevent her confession. The prosecution introduced the EPOs to show that Brandi was afraid of Pritchard and to support her testimony that Pritchard used threats of violence to coerce her into lying to investigators. So the EPOs make it more likely that Pritchard committed the crime because they corroborate his coverup efforts. Considering “[t]his Circuit applies an ‘extremely liberal’ standard for relevancy[,]” the EPOs are relevant because they support facts pointing to Pritchard’s guilt. United States v. Ramer, 883 F.3d 659, 681 (6th Cir. 2018) (quoting United States v. Collins, 799 F.3d 554, 578 (6th Cir. 2015)). Thus, we find that admitting the EPOs didn’t violate Federal Rule of Evidence 401. What’s more, the EPOs are probative of Pritchard’s guilt by showing how he coerced Brandi to cover up the crime. So the EPOs’ prejudicial effect of labeling Pritchard as an abuser does not render them inadmissible. As a result, we uphold the district court’s decision to admit the EPOs into evidence.

C.

Finally, Pritchard’s argues that the government improperly introduced David West’s expert testimony at trial about what motivates arsonists. West testified that arsonists derive excitement and gratification from setting fires and often watch the fire or take pictures of the fire’s aftermath as a memento. He also suggested that arsonists generally have financial motives.

Again, Pritchard claims the district court violated Federal Rules of Evidence 401 and 403 by admitting irrelevant and prejudicial evidence. Prichard argues that giving general evidence about the habits and traits of arsonists doesn’t make his guilt any more or less probable. On top of that, Pritchard contends that offering general motivations for arson and attributing them to him created prejudice because it made the jury associate him with arson. Alleging that the evidence lacked probative value because the government gave no supporting evidence about Pritchard’s No. 18-6210 United States v. Pritchard Page 16

motivation, Pritchard concludes that the prejudicial value of the expert testimony substantially outweighed its probative value.

The government responds that this Circuit already found expert testimony about arsonists’ general behavior admissible in United States v. Rayborn, 495 F.3d 328, 334 (6th Cir. 2007). But that’s not the entire truth. In Rayborn, the expert testified about how arsonists tend to set fires—for example they “commonly use paper as an accelerant” and “typically” pour flammable liquid in certain patterns. Id. So Rayborn does not settle whether testimony about the intrinsic motivation of arsonists is relevant under Rule 401. Although we do not permit any conceivably mitigating or harmful evidence as relevant, this Court has noted that the “Federal Rules are extremely permissive as to what evidence is relevant.” Wood v. Wal-Mart Stores E., LP, 576 F. App’x 470, 473 (6th Cir. 2014).

West’s testimony here meets that minimal standard. He explained that, based on his expert knowledge, taking pictures of a fire fits the profile of an arsonist. So Pritchard photographing the fire and showing those images off to Brandi’s children made it more likely that he started the fire. This distinguishes Pritchard’s case from United States v. Taylor, 814 F.3d 340, 362 (6th Cir. 2016), abrogated on other grounds by United States v. Richardson, 948 F.3d 733 (6th Cir. 2020), where we found that statistics and “generalized facts that could be said to apply to every . . . offender” were not relevant to the specific defendant’s case. Because West’s testimony ties to specific facts in the record supporting Pritchard’s guilt, that testimony is both relevant and probative. Thus, the district court committed no error by admitting West’s testimony.

One last point: Even if admitting the above evidence violated the Federal Rules of Evidence, Pritchard would still need to show the district court’s error wasn’t harmless. Hardy, 643 F.3d at 153. But when the “record evidence of guilt is overwhelming” independent of the allegedly offending evidence, admitting that evidence is harmless and does not permit vacating a conviction. Id. Given Brandi’s testimony, testimony from neutral parties, and Pritchard’s faulty alibi, the jury had enough evidence to convict Pritchard without relying on his prior bad acts, the EPOs, or West’s expert testimony. So any error in admitting that evidence would be harmless. No. 18-6210 United States v. Pritchard Page 17

V.

Relying on Carpenter v. United States, 138 S. Ct. 2206 (2018), Pritchard argues that the district court should have suppressed evidence of his cell-site location between June 1, 2011 and July 31, 2011. The government, without a warrant, obtained CSLI showing the whereabouts of Pritchard’s cell phone after receiving a court order in 2015. This information disproved Pritchard’s alibi of being in Louisville when the fire started. Fourteen days before Pritchard’s trial began, the Supreme Court decided Carpenter. There, the Court found that warrantless seizure of CSLI under the SCA violated the Fourth Amendment. Id. at 2221. So Pritchard asserts that his allegedly improperly seized CLSI should have been barred at trial by the exclusionary rule. Although it’s true that Carpenter prohibits warrantlessly seizing CSLI, this case is not about what the government can do post-Carpenter. Instead, we ask if the government acted in good faith considering the status of the law when it seized Pritchard’s CSLI. Illinois v. Krull, 480 U.S. 340, 349–50 (1987) (confirming that the exclusionary rule generally does not apply to evidence obtained in good-faith reliance on a statute before it was declared unconstitutional).

It’s black letter law that good-faith reliance on a statutory warrant exception can excuse an officer’s failure to obtain an otherwise-necessary warrant. See United States v. Parrish, 942 F.3d 289, 293 (6th Cir. 2019) (“[C]ourts will not exclude evidence from trial that was seized ‘by officers reasonably relying on a warrant issued by a detached and neutral magistrate.’” (quoting United States v. Leon, 468 U.S. 897, 913 (1984))). That means as long as an officer relies on a statute in good faith to perform a warrantless search, the fruits of that search are admissible at trial. This is true even if that statute is later found unconstitutional. United States v. Buford, 632 F.3d 264, 271 (6th Cir. 2011) (“[T]he Supreme Court has held suppression is not an available remedy when police officers conducted a search in good faith reliance on some higher authority, such as a warrant or a statute, even if the warrant or statute were later held invalid or unconstitutional (the ‘good faith exception’).” (quoting United States v. Gonzalez, 598 F.3d 1095, 1101 (9th Cir. 2010) (Bea, J. dissenting from the denial of rehearing en banc))). But this standard does not preclude all challenges to warrantless searches—the exclusionary rule bars evidence seized when an officer did not rely on a statute in good faith to execute the offending No. 18-6210 United States v. Pritchard Page 18

search or seizure. Indeed, officers do not rely on a statute in good faith “if [the statute’s] provisions are such that a reasonable officer should have known that the statute was unconstitutional.” Krull, 480 U.S. at 355.

Pritchard claims that our precedent would not have permitted an officer to conduct warrantless CSLI searches under the SCA when the government seized his CSLI in 2015. He relies on United States v. Warshak, 631 F.3d 266, 288 (6th Cir. 2010), where this Circuit found that a warrantless seizure of private emails under the SCA violated the Fourth Amendment. In that case, we applied the good-faith exception, but explained that “after today’s decision,” our “good-faith calculus has changed, and a reasonable officer may no longer assume that the Constitution permits warrantless searches of private emails.” Id. at 289 n.17. Pritchard argues that this holding means that the government could not have relied on the SCA in good faith to collect his CSLI data post-Warshak.

The government disagrees that Warshak prohibited good-faith reliance on the SCA’s warrant exception, at least for CSLI. It counters Pritchard’s reasoning by citing United States v. Pembrook, 876 F.3d 812, 823 (6th Cir. 2017), vacated on other grounds by Calhoun v. United States, 139 S. Ct. 137 (2018), where this Circuit denied the defendants’ motion to suppress CSLI seized in 2014 without a warrant. Rejecting the defendants’ Fourth Amendment claim, we found that no precedent, binding or otherwise, established the need for a warrant to seize CSLI under the SCA. Id. So, as of 2014, courts in our jurisdiction had “no constitutional reason to suppress . . . cell-tower location evidence.” Id.

That takes us back to the controlling issue: How did the law appear to a reasonable officer when the government seized Pritchard’s CSLI in 2015? In Pembrook, this Court scoured caselaw both within and outside our Circuit and determined that, as of 2014, criminal defendants had no Fourth Amendment protections against warrantless seizures of CSLI. Id. Absent evidence that our Fourth Amendment precedent changed until the 2018 Carpenter decision, it would be strange to locate such a protection in 2015. Surely if no caselaw or statutory language existed to negate an officer’s good-faith belief that warrantless CSLI seizures were permissible in 2014, then a Fourth Amendment sea change impacting a 2015 CSLI seizure would not have gone undetected or unannounced. No. 18-6210 United States v. Pritchard Page 19

Proving this point, Pritchard doesn’t argue that any opinion or legislation altered Pembrook. Instead, he claims that Warshak preceded Pembrook and gave a contradictory holding. Thus, he argues that Pembrook was never good law because it violated this Circuit’s precedent when issued. But this ignores the major difference between Warshak and Pembrook: Warshak concerned the seizure of private emails while Pembrook spoke directly about CSLI.3 Private written communications have long enjoyed protection from unreasonable searches and seizures under the Fourth Amendment. See Carpenter, 138 S. Ct. at 2264 (Gorsuch, J., dissenting) (“Under its plain terms, the [Fourth] Amendment grants you the right to invoke its guarantees whenever one of your protected things (your person, your house, your papers, or your effects) is unreasonably searched or seized. Period.”). But until Carpenter, courts didn’t routinely grant similar protection to data, held by a third party, revealing a person’s location, such as CSLI. So in a post-Pembrook, pre-Carpenter world, reasonable officers had every reason to believe in the constitutionality of collecting CSLI without a warrant under the SCA.

What’s more, this Circuit already found that warrantless CSLI searches under the SCA performed before Carpenter stemmed from good-faith reliance on the SCA. Carpenter, 926 F.3d at 318 (“The Government’s acquisition of Carpenter’s CSLI violated the Fourth Amendment. The district court nevertheless properly denied suppression because the FBI agents relied in good faith on the SCA when they obtained the data.” (emphasis added)). This ruling concerns warrantless CSLI collection post-Warshak, meaning it covers the period at issue in Pritchard’s case. Id. at 318. As a result, pre-Carpenter reliance on the SCA for warrantless CSLI collection doesn’t permit the sort of claim Pritchard raises.

Thus, a reasonable officer, when the government seized Pritchard’s CSLI in 2015, could have relied on the SCA in good faith to obtain that data without a warrant. So we affirm the

3This Circuit alluded to this distinction when hearing Carpenter on remand. United States v. Carpenter, 926 F.3d 313, 318 n.1 (6th Cir. 2019) (“Although Warshak announced a prospective rule barring the warrantless search of a suspect’s private emails under [the SCA], the court did not address any other circumstances where reliance on [the SCA] might be unreasonable. The decision in Warshak therefore would not have alerted the agents in Carpenter’s case to the unconstitutionality of seeking the CSLI at issue here.”), reh’g granted on other grounds, 788 F. App’x 364 (6th Cir. 2019); see also In re United States for Historical Cell Site Data, 724 F.3d 600, 611–12 (5th Cir. 2013) (discussing the difference between collecting CSLI and searching email servers under the SCA before the Supreme Court’s decision in Carpenter), abrogation recognized by United States v. Beverly, 943 F.3d 225 (5th Cir. 2019). No. 18-6210 United States v. Pritchard Page 20

district court’s denial of Pritchard’s motion to suppress CSLI evidence showing his location during the arson.

VI.

Apart from his conviction, Pritchard challenges the two-level sentencing enhancement he received under U.S.S.G. § 3B1.1(c). The language of U.S.S.G. § 3B1.1(c) reads: “if the defendant was an organizer, leader, manager, or supervisor in any criminal activity” that involved four or fewer participants and was not otherwise extensive, “increase by 2 levels.” And comments to § 3B1.1 provide factors for applying the leadership enhancement. These include:

the exercise of decision making authority, the nature of participation in the commission of the offense, the recruitment of accomplices, the claimed right to a larger share of the fruits of the crime, the degree of participation in planning or organizing the offense, the nature and scope of the illegal activity, and the degree of control and authority exercised over others.

U.S.S.G. § 3B1.1 cmt. n.4. Pritchard claims the district court erred by applying § 3B1.1 for leading Brandi in the arson and the coverup, thereby imposing a procedurally unreasonable sentence. So he asks us to vacate his sentence and remand to correct this allegedly erroneous sentencing calculation. “This court reviews sentencing enhancements under Section 3B1.1 with deference.” United States v. Rodriguez, No. 18-2396, 2019 WL 7187346, at *3 (6th Cir. Dec. 26, 2019) (citing United States v. Washington, 715 F.3d 975, 982–83 (6th Cir. 2013)).

Disputing the § 3B1.1 leadership enhancement, Pritchard argues both that Brandi acted as an “equal participant” in the criminal offenses and that “she refused to go along with the scheme every time it was mentioned.” (Appellant’s Br. at 52.) From Pritchard’s account, Brandi helped start the fire and Brandi herself filed the fraudulent insurance claim. This shows, according to Pritchard, that he lacked “managerial control over other [criminal] participants.” United States v. Wright, 747 F.3d 399, 412 (6th Cir. 2014) (quoting United States v. Vandeberg, 201 F.3d 805, 811 (6th Cir. 2000)). That allegedly shows a sentencing error because only control over others in the criminal activity, not mere suggestion to commit a crime, generally warrants a § 3B1.1(c) leadership enhancement. U.S.S.G. § 3B1.1 cmt. n.4. No. 18-6210 United States v. Pritchard Page 21

But as much as Pritchard wants to paint Brandi as a co-equal arsonist or an independent actor outside his control, that simply isn’t what the record shows. Pritchard came up with the idea for arson, presented it to Brandi for recruitment purposes, tried to sway her when she resisted, bragged about being a “genius” at arson-based insurance fraud, planned the logistics of the fire, directed the coverup, and threatened domestic violence to execute the coverup. Thus, Brandi acted under Pritchard’s control while (1) planning the fire, (2) committing the coverup, and (3) submitting a fraudulent insurance claim. So we uphold the district court’s decision to apply a § 3B1.1 sentencing enhancement.

VII.

For the above reasons, we AFFIRM Pritchard’s conviction and sentence. No. 18-6210 United States v. Pritchard Page 22

CONCURRING _________________

CLAY, Circuit Judge, concurring. I concur in the majority’s judgment but write separately to note my disagreement with a small, but important, piece of its analysis. Specifically, the majority unnecessarily and wrongly analyzes the meaning of the phrase “a direct . . . result.” See 18 U.S.C. § 844(i) (establishing that individuals may be subject to any term of imprisonment or to the death penalty “if death results to any person, including any public safety officer performing duties as a direct or proximate result of conduct prohibited by this subsection”) (emphasis added). The majority suggests that Congress contemplated “actual”— sometimes referred to as “but-for”—causation when it used this phrase, thus allowing a defendant to be convicted under § 844(i) based on a showing of either actual or proximate causation. I disagree, but nevertheless concur because, as the majority acknowledges, this analysis is unnecessary to this case’s result and therefore constitutes dicta.

At bottom, the majority’s interpretation of “direct . . . result” is simply incorrect. “Direct cause” is not another term for “actual cause,” but is instead a synonym for “proximate cause.” See Cause, Black’s Law Dictionary (11th ed. 2019). While the majority invokes common law to find otherwise, the Supreme Court has clearly explained that that “among the many shapes [that proximate cause] took at common law was a demand for some direct relation between the injury asserted and the injurious conduct alleged.” Holmes v. Sec. Inv’r Prot. Corp., 503 U.S. 258, 268 (1992) (citation omitted); accord, e.g., Paroline v. United States, 572 U.S. 434, 444 (2014); Hemi Grp., LLC v. City of New York, 559 U.S. 1, 9 (2010); Bridge v. Phoenix Bond & Indem. Co., 553 U.S. 639, 654 (2008); see also Babbitt v. Sweet Home Chapter of Cmtys. for a Greater Or., 515 U.S. 687, 732–33 (Scalia, J., dissenting) (“In fact ‘proximate’ causation simply means ‘direct’ causation.”). We have interpreted “proximate cause” similarly. See, e.g., Lopez Sosa v. Barr, 780 F. App’x 307, 309 (6th Cir. 2019) (interpreting asylum statute and concluding that, “to use the tools of proximate causation, the statute requires a ‘direct relationship’ between the social group and the harm” (quoting Hemi Grp., LLC, 559 U.S. at 10)); Crosby v. Twitter, Inc., 921 F.3d 617, 624 (6th Cir. 2019) (agreeing that “proximate cause requires some ‘directness’ No. 18-6210 United States v. Pritchard Page 23

between the tortious conduct and the injury”). The legislative history of § 844(i) confirms this meaning. According to a Senate Committee on the Judiciary report, “[t]he Committee intends that a death or injury is a direct or proximate result of conduct proscribed in subsection 844 . . . (i) if it is reasonably foreseeable.” S. Rpt. 98-225, at 359 (1983). Foreseeability is, of course, a critical component of the proximate cause inquiry.

The majority acknowledges that “some formulations of proximate cause include a concept of directness within them,” but reasons that Congress must have meant actual causation because proximate and actual causation typically go hand in hand and because the rule against surplusage suggests that “direct . . . result” must mean something different than “proximate result.” Neither point justifies its interpretation.

First, far from ensuring that actual and proximate causation remain paired, the majority’s proposed interpretation does just the opposite. In the central case cited by the majority, Burrage v. United States, 571 U.S. 204 (2014), the Supreme Court explained that “[w]hen a crime requires ‘not merely conduct but also a specified result of conduct,’ a defendant generally may not be convicted unless his conduct is ‘both (1) the actual cause, and (2) the “legal” cause (often called the “proximate cause”) of the result.’” Id. at 210 (emphasis added) (quoting 1 W. LaFave, Substantive Criminal Law § 6.4(a), pp. 464–466 (2d ed. 2003)). Interpreting “direct . . . result” as referring to proximate cause comports with this rule because a proximate cause of an event is generally also an actual cause of it. Paroline, 572 U.S. at 444 (explaining that proximate cause means both that an event actually caused another event and that the two events are sufficiently connected). An actual cause of an event, on the other hand, is frequently not also a proximate cause. Cf. Burrage, 571 U.S. at 211 (recognizing that actual causation “represents ‘the minimum requirement for a finding of causation when a crime is defined in terms of conduct causing a particular result’” (alteration in original) (quoting Model Penal Code § 2.03(1)(a) explanatory note)). By interpreting § 844(i) as allowing conviction based on a showing of actual causation alone, the majority thus dramatically expands the scope of conduct covered by the statute. This runs counter to the purpose of pairing the two types of causation in the first place. See Paroline, 572 U.S. at 448 (explaining that requiring proximate cause ensures that a law does not apply in No. 18-6210 United States v. Pritchard Page 24

“situations where the causal link between conduct and result is so attenuated that the so-called consequence is more akin to mere fortuity”).

Second, while courts should interpret statutes to avoid surplusage, that rule does not compel the majority’s conclusion here. As the majority acknowledges, “proximate cause” is an imprecise concept. Accordingly, I understand the addition of “direct . . . result” as simply clarifying what “proximate result” means by focusing courts’ inquiry on the issue of remoteness. Where a phrase is otherwise unclear, adding an explanatory phrase is not redundant.

The majority’s analysis is not simply unjustified—it runs counter to binding case law. “Especially in the interpretation of a criminal statute subject to the rule of lenity, we cannot give the text a meaning that is different from its ordinary, accepted meaning, and that disfavors the defendant.” Burrage, 571 U.S. at 216 (citation omitted). As discussed, the most common meaning of “direct cause” is similar to that of “proximate cause.” To my eyes, § 844(i) thus unambiguously requires a showing of proximate causation, and “direct . . . result” simply confirms this point. However, even if that phrase could also be interpreted to refer to actual causation, it would be at most ambiguous and the rule of lenity would require this Court to interpret it in defendants’ favor. See United States v. Bass, 404 U.S. 336, 347 (1971) (“In various ways over the years, we have stated that ‘when choice has to be made between two readings of what conduct Congress has made a crime, it is appropriate, before we choose the harsher alternative, to require that Congress should have spoken in language that is clear and definite.’” (quoting United States v. Universal C.I.T. Credit Corp., 334 U.S. 218, 221–22 (1952))).

Finally, and perhaps most importantly, the majority’s interpretation of this portion of the statute is not necessary to our ultimate resolution of this appeal. Nor did the parties present this matter for our consideration or provide any briefing to inform our analysis. Under any possible interpretation of § 844(i), a defendant may properly be convicted based on a showing of proximate cause, and the panel agrees that Pritchard proximately caused Sparks’s death. The majority recognizes as much, but presses its interpretation anyway. Fortunately, the majority’s analysis regarding the meaning of “direct . . . result” is dicta and, as such, will not bind any future panel to be confronted with this issue. I therefore concur in the judgment.