[PUBLISH]
IN THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
________________________
No. 18-11220
________________________
D.C. Docket No. 2:15-cv-00701-GAP-CM
PATRICIA I. ERMINI, a.k.a. Patricia I. Mapes,
Plaintiff–Appellee,
versus
MIKE SCOTT, in his official capacity as Sheriff of Lee County, Florida,
Defendant–Appellant.
________________________
Appeal from the United States District Court for the Middle District of Florida
________________________
(September 10, 2019)
Before WILLIAM PRYOR, NEWSOM, and BRANCH, Circuit Judges. NEWSOM, Circuit Judge:
This case arises out of a routine wellness check that went badly awry. The underlying episode began with three Lee County deputies stopping by to check on 71-year-old Patricia Ermini at the request of her daughter—and ended with the deputies shooting Ermini five times. Ermini—who, incredibly, survived—later sued, bringing a litany of state- and federal-law claims against the deputies and Lee County Sheriff Mike Scott. Only one claim made it past summary judgment—a state-law cause of action against Sheriff Scott in his official capacity, seeking to hold him vicariously liable for the deputies’ negligence in conducting the check. That claim went to trial, the jury ruled in Ermini’s favor, and the district court thereafter denied Scott’s motion for new trial.
Scott now appeals the judgment against him as well as the court’s post-
judgment order refusing his new-trial request. Scott argues (1) that the district court improperly instructed the jury that if it concluded that he proved Florida’s “alcohol defense,” Ermini couldn’t recover; (2) that by introducing evidence regarding certain aspects of the deputies’ conduct during the check, Ermini impermissibly pursued a nonexistent and precluded “negligent-use-of-force” claim; (3) that Ermini’s lawyer made a forbidden “golden-rule” argument when she asked the jurors to “imagine if someone was in [their] house”; and (4) that the trial court abused its discretion by admitting immaterial character evidence concerning two deputies’ post-event (and unrelated) terminations from the Lee County Sheriff’s
3
Office. Because we find no errors that merit a new trial, we affirm the judgment in Ermini’s favor.
I
A
Following a worrisome telephone conversation, Patricia Ermini’s daughter, Maine resident Robin LaCasse, called the Lee County Sheriff’s Office to request a wellness check on her elderly mother. During the call, Ermini had seemed distraught—and possibly suicidal—and LaCasse hadn’t been able to get back in touch with her. LaCasse told the Sheriff’s Office that Ermini might have been drinking wine and that she had a handgun in her home. Shortly after LaCasse’s phone call, Deputies Richard Lisenbee, Robert Hamer, and Charlene Palmese were dispatched to Ermini’s home; they knew that Ermini could be intoxicated and that she owned a gun.
Lisenbee arrived on the scene first, banged on the door, and yelled “Sheriff’s Office,” but got no response. When he opened the unlocked door, Lisenbee found the house dark, quiet, and in disarray, an empty wine bottle on the floor. He retreated out of the house and waited for backup. When Palmese arrived, she and Lisenbee reentered the home and announced themselves, but again got no response. The deputies opted to wait for Hamer before continuing the wellness check. Once all three officers were on the scene, they again announced themselves
4
and entered the dark living room with their flashlights illuminated and their weapons drawn. They made their way to the closed double doors leading to Ermini’s master bedroom.
Lisenbee opened the right door and shined his flashlight into the room, where he saw Ermini lying in bed. Ermini awoke, confused by the strangers in her bedroom—she testified that she remembered asking “who’s there?” and telling the intruders that she had a gun and to get out of her house. She also testified that she recalled the deputies saying that they were with the Sheriff’s Department and her responding that she hadn’t called the Sheriff and that they had better get out of her house. Lisenbee began backing out of the bedroom as Ermini, clothed only in her undergarments, moved toward the door.
The deputies and Ermini recall very differently the critical moments that followed. Hamer testified that Ermini walked toward him with both hands on her gun, which she pointed directly at him. Ermini doesn’t remember grabbing her gun or pointing it at anyone. In any event, Hamer, who was outside the bedroom, fired seven rounds through the partially closed bedroom door, five of which hit Ermini, who collapsed onto the floor. (Further to the parties’ dispute, Ermini’s gun was found on the floor to the left of where she fell after being shot, and a bullet from her weapon was later found lodged in the ceiling.) Hamer began providing emergency medical care to a still-confused Ermini, who (according to
5
the officers and paramedics) repeatedly asked why the deputies were in her home and why they were trying to kill her. Ermini was taken to the hospital for further treatment, and she survived.
B
Ermini sued Deputies Lisenbee, Hamer, and Palmese, as well as Sheriff Scott and William Murphy, an additional officer who hadn’t been on the scene. Among other claims, Ermini alleged excessive force and false arrest under the Fourth Amendment, battery, negligent infliction of emotional distress, and negligence in conducting the wellness check under Florida law. Only Ermini’s vicarious-liability claim against Scott for the allegedly negligent wellness check survived summary judgment. That claim went to trial, and the jury ultimately ruled in Ermini’s favor, awarding her $750,000 in damages.
Several aspects of the pre-trial and trial proceedings are relevant to this appeal. We’ll take them chronologically. First, before trial, Scott submitted a motion in limine under Federal Rules of Evidence 403 and 404(b) to exclude evidence surrounding Lisenbee’s and Hamer’s post-event (and unrelated) terminations from the Sheriff’s Office. The district court held a telephonic hearing and denied the motion, stating that it would allow limited questioning about the timing of and general reasons for the officers’ terminations but that it would exclude additional details and written reports. Second, during closing arguments,
6
Ermini’s lawyer asked the jury, “Can you imagine if someone was in your house that you wouldn’t try to figure out who is that[?]” Scott’s lawyer objected, stating “That’s not right, golden rule.” The district court overruled the objection without further elaboration.
The third item pertains to Scott’s “alcohol defense,” which, under Florida law, prevents a plaintiff from recovering damages if either her “normal faculties were impaired” or she had a blood-alcohol level of 0.08% or higher, and as a result she was more than 50% responsible for her own harm. Fla. Stat. § 768.36(2)(a)– (b) (2019). The district court instructed the jury that if it found that Scott had proved the alcohol defense by a preponderance of the evidence, Ermini couldn’t recover. A similar statement was included on the special verdict form given to the jury, which informed jurors that if they found in Scott’s favor on the defense, they didn’t need to fill out the remainder of the form. Scott objected to the court’s alcohol-defense instruction at trial.
Following the jury’s verdict, Scott moved for a new trial and renewed his motion for judgment as a matter of law. As relevant here, he argued that the evidence about Lisenbee’s and Hamer’s terminations should have been excluded and that Ermini had improperly introduced evidence about the deputies’ use of force, which Scott said had no place in a negligent-wellness-check case. The
7
district court denied both motions. Scott appeals the district court’s judgment and its order denying his motion for new trial. 1
II
Scott’s first argument challenges the district court’s jury instructions and verdict-form entry pertaining to Florida’s alcohol defense, which is codified at Fla. Stat. § 768.36. The district court instructed the jury as follows:
On the first defense, the issue for you to decide is whether Patricia Ermini was under the influence of any alcoholic beverage to the extent that her normal faculties were impaired, or the plaintiff had a blood alcohol level of .08 or higher; and whether as a result of the influence of such alcoholic beverage, Patricia Ermini was more than 50 percent at fault for her own harm.
If you find that the sheriff has proven this defense by a preponderance of the evidence, then plaintiff’s claim is barred and your verdict is for the sheriff.
(emphasis added). Scott objected to the italicized portion of the charge, which wasn’t included in the parties’ proposed instructions. The verdict form similarly informed the jury that if it found that Scott had proved the alcohol defense, it didn’t need to complete the remainder of the form.
Scott contends that he is entitled to a new trial because the district court improperly told the jury about the legal effect of any finding under the alcohol defense—namely, that if proved the defense would bar Ermini from recovering.
1 Although Scott’s notice of appeal also referenced his renewed motion for judgment as a matter of law, he didn’t address that motion in his briefing, so we won’t consider it here.
8
That information, he says, was unnecessary and was likely to evoke sympathy for Ermini. In particular, Scott asserts that just as evidence of a party’s liability insurance is inadmissible at trial, see Fed. R. Evid. 411, information about the legal effect of an alcohol-impairment finding should be verboten since it too has the potential to improperly influence the jurors’ emotions. We disagree.2
A
Before jumping into the merits, we must determine at the outset whether state or federal law governs Scott’s argument. We know, as a general matter, that in a diversity case in federal court, “[t]he substance of jury instructions . . . is governed by the applicable state law, but questions regarding procedural aspects of jury charges are controlled by federal law and federal rules.” Pate v. Seaboard R.R., Inc., 819 F.2d 1074, 1081–82 (11th Cir. 1987) (citations omitted). This case, though, poses a more specific issue, which turns out to be a question of first impression for this Court: What law applies—state or federal—when deciding whether a district court judge properly informed the jury about the legal effect of its finding under state law? More particularly, is informing the jury about the legal effect of a factual finding bound up in the “substance of [the] jury instructions”
2 “We review jury instructions de novo to determine whether they misstate the law or mislead the jury to the prejudice of the objecting party.” Palmer v. Bd. of Regents of the Univ. Sys. of Ga., 208 F.3d 969, 973 (11th Cir. 2000) (citations omitted). So long as the jury “instructions accurately reflect the law,” however, a district court has “wide discretion as to the style and wording employed.” McCormick v. Aderholt, 293 F.3d 1254, 1260 (11th Cir. 2002).
9
within the meaning of Pate, and thus governed by the underlying state law, or is it instead a “procedural aspect[]” of the charge governed by federal law?
We join the Seventh Circuit in holding that this issue is controlled by federal law, as it pertains to procedure, not substance. In Beul v. ASSE International, Inc., 233 F.3d 441 (7th Cir. 2000), the Seventh Circuit faced the question whether state or federal law governed a district court’s handling of a question posed to it by a jury during its deliberations. The question pertained to the special verdict form, which required the jury “to enter separately . . . the amount of damages and the percentage of the plaintiff’s comparative fault and not make the ‘bottom line’ computation, which involve[d] deducting from the amount of damages that amount times the plaintiff’s percentage of comparative fault.” Id. at 449. In particular, the jury asked, “What bearing do the negligence factors”—i.e., the fault percentages assigned to the parties—“have on the amounts we may or may not choose to award?” Id. The judge answered, correctly, that they would indeed have an effect: “[T]he comparison factor, if you find both parties negligent, has a significant impact upon the award that the Court enters.” Id.
In concluding that federal law governed the question whether the judge had acted properly, the Seventh Circuit began with the proposition that “[r]ules of general applicability and purely managerial character governing the jury, such as the form in which a civil jury is instructed, are quintessentially procedural.” Id.
10
From that premise, the court concluded that a district judge’s decision to inform the jury about the legal effect of its findings is an issue of procedure: “We think it follows that whether the federal court should try to keep the jury in the dark about the legal effect of the jury’s answers to the questions posed to it by the special verdict is also a question of federal law.” Id. The court wrote that its classification of the issue as procedural could result either from “an interpretation of [Federal] Rule [of Civil Procedure] 49(a),” which governs special verdicts, or from the application of “a federal common law of special verdicts to supplement the rule.” Id. (The same, we think, could be said with respect to Rule 51, which governs jury instructions. See Fed. R. Civ. P. 51.)
We agree with the Seventh Circuit’s conclusion that federal law controls.
The decision whether to inform the jury about the legal effect of its findings—or, whether instead, in the Seventh Circuit’s words, to “keep the jury in the dark”— has nothing to do with the substance of the applicable legal rule or the merits of the issues before the court. Instead, it has everything to do with the fair and efficient management of the trial. To be sure, a party may take issue with the fairness or efficiency of the district court’s decision; but it must do so by reference to federal law, not state.
11
B
So then, does federal law preclude a district court judge from informing jurors about the legal effects of their factual findings? We hold that it doesn’t.
It’s true, of course, that “[a] judge has no general duty to inform the jury of the legal consequences of its verdict.” Beul, 233 F.3d at 450 (citations omitted). And it may even be the case that “in some circumstances the giving of such information might interfere with the jury’s appraisal of the facts.” Id. But there is most assuredly “no rule against giving the information.” Id. And indeed, we agree with the Seventh Circuit that it is “difficult to conceive” of a scenario in which merely informing jurors of the legal consequences of their findings—i.e., telling them the whole truth—would constitute reversible error. Id. Rule 49 expressly allows judges to use general verdict forms with special interrogatories, which “reveal to the jury the legal consequences of its specific findings,” and “there is no purpose in forbidding [judges] to do the same thing with a special verdict.” Id. And we can think of no reason why a different rule should apply to jury instructions. We hold, therefore, that a district court may inform the jury about the legal effect of its finding so long as it does so accurately. And here no one contends—nor could they—that the court’s instructions were inaccurate.
Scott leans heavily on Harrison v. Gregory, and understandably so, as the court there granted a new trial, in part because the jury was told that if it concluded
12
the defendant had proved Florida’s alcohol defense, the plaintiff couldn’t recover. 221 So. 3d 1273, 1277–78 (Fla. Dist. Ct. App. 2017). Harrison, though, is doubly distinguishable. First, it arose under state law, and as we’ve just explained, federal law governs whether and under what circumstances a federal district court judge may inform the jury about the legal effect of its findings. Second, Harrison is different in an important respect. The problem there wasn’t so much that the jury found out about the legal effect of its finding regarding Florida’s alcohol defense, but rather that the plaintiff’s lawyer had, in closing, attempted to “inflame the minds and passions of the jurors” by harping on the unfairness of the statutory defense: “By the way,” he said, “50 percent or more at fault, there’s no recovery.” Id. at 1277. 3 The situation here, where the district court impassively—and accurately—instructed the jury concerning the legal effect of its alcohol-defense finding, strikes us as entirely different. Walther v. Omaha Public Power District, 412 F.2d 1164 (8th Cir. 1969), on which Scott also relies, is also off-point. There, the Eighth Circuit merely concluded that the legal effect of a jury’s finding needn’t necessarily be included in a jury instruction, not that it can’t be included. See id. at 1169–70.
3 In this connection, it is worth noting that Scott’s own lawyer made a similar comment during his closing argument here: “[B]y the way, if [Ermini’s] own intoxication impairment factored into this incident, precipitated the incident, caused the incident, then this claim is barred.” He went on to tell the jurors that if they found in Scott’s favor with regard to the alcohol defense, “that ends this case.”
13
To summarize: Federal law controls this issue, and federal law doesn’t preclude district court judges from accurately informing jurors of the effects of their findings—in either their instructions or their verdict forms. Accordingly, it wasn’t improper for the judge here to provide an accurate statement of law explaining the legal effect of the jury’s finding under Florida’s alcohol defense.
III
Scott next asserts that he is entitled to a new trial on the ground that Ermini effectively used her negligent-wellness-check claim as a Trojan Horse to reassert her claims about the deputies’ use of force, which the district court had dismissed pretrial. Although his position is difficult to discern, Scott seems to be making either (or both) of two different arguments. First, he contends that Ermini’s introduction of evidence about the deputies’ use of force converted her otherwise cognizable (even if he would say losing) negligent-wellness-check claim into an invalid “negligent-use-of-force” claim that Florida law doesn’t recognize. Second, and separately, because the district court rejected Ermini’s claims surrounding the officers’ use of force at summary judgment, Scott asserts that her claim is precluded. We disagree on both counts.4
4 Our “review of a district court’s denial of a motion for new trial is very limited.” Blu-J, Inc. v. Kemper C.P.A. Grp., 916 F.2d 637, 643 (11th Cir. 1990). “Absent an abuse of discretion, the district court’s disposition of a motion for a new trial will not be disturbed on appeal, especially when that disposition was to deny the motion.” Id.
14
A
We’ll start with Scott’s contention that Ermini used the guise of a negligent-
wellness-check claim to try a non-cognizable negligent-use-of-force claim. And we’ll begin at the beginning, with Ermini’s complaint. Ermini’s complaint expressly alleged that Scott’s deputies negligently performed a wellness check— not that they engaged in a negligent use of force. (To be sure, Ermini asserted that the deputies used force excessively, in violation of federal constitutional standards, but not that they used force negligently, in violation of state law.) In particular, Ermini alleged that “S[cott], acting through his deputies, owed a special duty of care to [her] because of the manner in which his deputies responded to a 911 call for a wellness check,” and that the deputies later “fail[ed] to exercise reasonable care in conducting the wellness check.” And the district court certainly seemed to appreciate the distinction: While the court granted summary judgment in the deputies’ favor on Ermini’s excessive-force claims, it denied them summary judgment on the negligent-wellness-check claim, finding that “triable issues of fact exist[ed] as to whether the deputies exercised reasonable care in carrying out the welfare check, which increased the risk of harm to [Ermini], and whether [Ermini’s] actions were a reasonably foreseeable consequence of the negligent acts or omissions of the deputies.”
15
Even so, Scott insists that as matters unfolded at trial, Ermini’s negligent-
wellness-check claim devolved into a non-cognizable—and thus forbidden— negligent-use-of-force claim. We don’t think so. Ermini didn’t invent and litigate a new cause of action; rather, there is simply an inevitable overlap between the facts underlying her negligent-wellness-check claim and those surrounding the deputies’ use of force. And although Florida doesn’t recognize claims for negligent use of force, it does recognize the commonsense proposition that “distinct act[s] of negligence” can occur in conjunction with alleged excessive- force incidents. City of Miami v. Sanders, 672 So. 2d 46, 48 (Fla. Dist. Ct. App. 1996). Just so here—the wellness check involved distinct acts of alleged negligence during which an allegedly excessive use of force occurred.
To bolster his claim, Scott asserts that “[m]uch of [Ermini’s] counsel’s questions and argument” focused on the officers’ use of force, but he musters only two examples from the trial record: (1) Ermini’s lawyer suggesting that she never pointed her gun at the deputies; and (2) her lawyer referencing the deputies’ failure to issue a warning prior to opening fire. Two instances do not “much” of Ermini’s argument make. Moreover, the facts surrounding the use of force are inextricably intertwined with those surrounding the wellness check—it’s all one seamless story. Accordingly, it seems to us essentially inevitable that Ermini’s lawyer properly would elicit testimony pertaining to the use of force in the course of trying the
16
negligent-wellness-check claim. Indeed, excluding all information about the officers’ force would have made it nearly impossible for the jury to do its job. Could jurors have even possibly determined Ermini’s damages, for instance, if they had been precluded from hearing any information about the specific conduct that caused her injury?
In any event, the fact that Ermini’s lawyer mentioned a few details about the force used by the deputies doesn’t negate or invalidate the trial’s overarching focus on their negligence in performing the wellness check, nor does it convert Ermini’s negligent-wellness-check claim into a nonexistent negligent-use-of-force claim. The complaint, the jury instructions, and the verdict form all properly focused on the vicarious-liability claim against Scott and his statutory defenses. And when Ermini’s lawyer examined her witnesses—particularly the officers—she questioned them extensively about their conduct in connection with the wellness check even before any shots were fired. She asked the deputies, for instance, about their dark uniforms, whether the lights outside Ermini’s home were on, whether the house’s interior was dark and if (and how) they used their flashlights, whether their patrol cars were visible from the road, whether they checked the perimeter of the home before entering, whether they used Ermini’s name when speaking to her, and what, if any, wellness-check-related training or experience they had received prior to the incident. None of those questions had anything to do with the officers’ use
17
of force. The fact that Ermini’s lawyer also occasionally touched on force-related topics doesn’t convert the cause of action. The circumstances surrounding the wellness check and the use of force are bound up in the same unfortunate incident—it’s difficult, if not impossible, to tell the former story without occasionally verging into the latter.
B
Scott separately contends—albeit obliquely—that Ermini’s introduction of evidence pertaining to the deputies’ use of force should have been barred by some form of preclusion doctrine. Again, we disagree.
To the extent that any such doctrine is in play here, we think it is “direct estoppel,” which is a variety of issue preclusion. DuChateau v. Camp, Dresser & McKee, Inc., 713 F.3d 1298, 1301 (11th Cir. 2013). “Direct estoppel, as opposed to collateral estoppel, governs the preclusive effect of a litigated issue in a separate proceeding within a single suit.” Cotton v. Heyman, 63 F.3d 1115, 1118 n.1 (D.C. Cir. 1995); see also DuChateau, 713 F.3d at 1301–03. To the extent Scott means to argue that Ermini was barred from “relitigating” at trial force-related facts that had already been adjudicated against her at summary judgment—in particular, whether Ermini pointed her gun at the deputies and whether they warned her before shooting—his argument fails. A partial summary judgment on some of a plaintiff’s claims cannot directly estop other claims that proceed to trial.
18
We have held that a jury verdict on one claim can directly estop another claim in the same suit. DuChateau, 713 F.3d at 1301–02 (holding that a jury verdict on a claim of employment retaliation directly estopped the plaintiff’s appeal of the district court’s rejection of her claim of pregnancy discrimination because both claims required proof of adverse employment action). But partial summary judgment is materially different from a jury verdict. Unlike a jury verdict, a partial summary judgment is not a “final decision” under 28 U.S.C. § 1291. State Treasurer of Michigan v. Barry, 168 F.3d 8, 11 (11th Cir. 1999).5 Accordingly, the district court’s interlocutory order rejecting some of Ermini’s claims at summary judgment does not preclude her negligent-wellness-check claim. 6
5 The only way for Ermini to obtain appellate review of the district court’s partial summary judgment against her would have been to seek discretionary interlocutory review, 28 U.S.C. § 1292(b), or extraordinary relief such as a writ of mandamus. And a failure to seek permissive interlocutory review does not transform an otherwise nonfinal order into a final decision entitled to preclusive effect. 18 Charles Alan Wright, Arthur R. Miller & Edward H. Cooper, Federal Practice and Procedure § 4418, at 524 (3d ed. 2016) (“[F]ailure to seize opportunities for avowedly interlocutory review should not transform a potentially reviewable order into issue preclusion.”). 6 Moreover, and in any event, both of the issues on which Scott seems to predicate his estoppelbased argument—whether Ermini pointed her gun at the deputies and whether they warned her before shooting—were assumed in Ermini’s favor at summary judgment, not against her; the district court accepted for the sake of argument that Ermini didn’t point her gun at the deputies and that they didn’t warn her before opening fire, but granted summary judgment on her excessive-force claim anyway.
19
IV
Next, the “golden-rule.” Scott objects to the following statement that Ermini’s lawyer made at closing: “The fact of the matter is that . . . actually, the evidence supports Ms. Ermini’s version, which is a logical thing. Can you imagine if someone was in your house that you wouldn’t try to figure out who is that[?]” (emphasis added). Scott asserts that this statement violates the prohibition on “golden-rule” arguments. We disagree. 7
A
A “golden-rule” argument invites jurors to “put themselves in the shoes of the plaintiff and do unto him as they would have him do unto them under similar circumstances.” Ivy v. Sec. Barge Lines, Inc., 585 F.2d 732, 741 (5th Cir. 1978). The risk is that such a statement asks jurors to “decide the case on the basis of sympathy rather than from an objective review of the evidence.” Dempsey v. Mac Towing, Inc., 876 F.2d 1538, 1540 n.1 (11th Cir. 1989). An unremedied golden- rule argument will ordinarily result in reversal. Id.
A golden-rule objection typically arises when a lawyer asks jurors to place themselves in the plaintiff’s position with respect to the calculation of damages. See, e.g., Ivy, 585 F.2d at 741 & n.10. For example, asking the jury, “If you were
7 We review a district court’s ruling regarding allegedly improper statements by counsel for an abuse of discretion. Allstate Ins. Co. v. James, 845 F.2d 315, 318 (11th Cir. 1988) (citation omitted).
20
in the plaintiff’s shoes, how much do you think you would deserve?” would be a clear-cut violation. “The rationale for prohibiting such an argument is that the jury’s sympathy will be unfairly aroused, resulting in a disproportionate award of damages.” Burrage v. Harrell, 537 F.2d 837, 839 (5th Cir. 1976); see also Har- Pen Truck Lines, Inc. v. Mills, 378 F.2d 705, 714–15 (5th Cir. 1967). The question here is whether Ermini’s lawyer’s statement—which pertained not to the appropriate measure of damages but rather to Scott’s liability for the deputies’ actions—violated the prohibition on golden-rule arguments as explicated in our case law. We conclude that it didn’t.
B
Although many courts treat golden-rule arguments as universally improper, 8 we have historically permitted certain golden-rule-ish statements in the liability context. Most notably for present purposes, in McNely v. Ocala Star-Banner Corp., we held that defense counsel didn’t violate the prohibition on golden-rule arguments when he “invit[ed] the jury to put itself in the defendants’ position when
8 See, e.g., Caudle v. District of Columbia, 707 F.3d 354, 360 (D.C. Cir. 2013) (holding that “[i]t is no more appropriate for a jury to decide a defendant’s liability . . . based on an improper consideration than to use the same consideration to determine damages”); Mich. First Credit Union v. Cumis Ins Soc’y, Inc., 641 F.3d 240, 249 (6th Cir. 2011); Ins. Co. of N. Am., Inc. v. U.S. Gypsum Co., 870 F.2d 148, 154 (4th Cir. 1989); Edwards v. City of Philadelphia, 860 F.2d 568, 574–75 (3d Cir. 1988); Joan W. v. City of Chicago, 771 F.2d 1020, 1022 (7th Cir. 1985) (stating that “[t]here is no reason for . . . a distinction [between different types of golden rule statements] because the jury’s departure from its neutral role is equally inappropriate regardless of the issue at stake”).
21
considering [plaintiff’s] alleged work place misconduct and evaluating whether he was terminated because of his disability.” 99 F.3d 1068, 1071 n.3 (11th Cir. 1996). That statement, we held, “was not in any way directed to the question of damages; rather it related only to the reasonableness of appellee’s actions”—an issue that was squarely, and properly, before the jury. Id. (emphasis added) (citing Burrage, 537 F.2d at 839).
Somewhat farther afield—but in the same vein—we have permitted liability-
related golden-rule statements in criminal cases. In Reese v. Secretary, Florida Department of Corrections, for instance, we held that a prosecutor’s comment that a victim experienced “every woman’s wors[t] nightmare,” as well as statements urging the jury to think about “what was going on in [the victim’s] mind,” weren’t improper golden-rule arguments because the government had to show that the crime was “designed to inflict a high degree of pain with utter indifference to . . . the suffering of others” in order to prove an aggravating factor. 675 F.3d 1277, 1292 (11th Cir. 2012) (first alteration in original) (quoting citation omitted). So too, in Davis v. Kemp, we deemed the following statement permissible in a death- penalty case, because the jury could properly consider the defendant’s “future dangerousness”: “Ask yourselves this question, how would you feel living in this community if you looked out of your window one night and saw [the defendant] walking down the street coming up toward your house. If that wouldn’t put a
22
feeling of cold terror in your heart, what would?” 829 F.2d 1522, 1528–29 (11th Cir. 1987).9 The common tie that binds the liability-related golden-rule-ish statements that we have deemed permissible? Their connection to an element or factor genuinely at issue in the case, or their focus on either a party’s behavior or perception of an incident. Here, Ermini’s lawyer’s argument went directly to the reasonableness of Ermini’s response to the deputies’ actions during the wellness check, which bears directly on the foreseeability of her actions—both issues that jurors would necessarily have had to grapple with in deciding Ermini’s negligent- wellness-check claim. The statement seems to us essentially indistinguishable from the reasonableness-related statement that we allowed in McNely.
Admittedly, Ermini’s lawyer could have chosen her words more carefully—
there’s plenty of case law warning attorneys to steer clear of potentially inflammatory closing arguments, see, e.g., United States v. Kopituk, 690 F.2d 1289, 1342–43 (11th Cir. 1982)—but, her lone statement in closing didn’t run afoul of the prohibition on golden-rule statements as we have understood and applied it.
9 See also, e.g., Kennedy v. Dugger, 933 F.2d 905, 913 (11th Cir. 1991) (permitting the following comment about a first-degree-murder victim’s experience: “Can you imagine, in your own living room not bothering a soul on a Saturday afternoon? . . . [The defendant] walked . . . down to your own house, and, a total stranger, because you got in his way, destroys you[?]”).
23
What’s more, even if the statement had crossed the line, “[f]or reversible error to be found in a closing argument, the challenged argument must be ‘plainly unwarranted and clearly injurious.’” Goldsmith v. Bagby Elevator Co., 513 F.3d 1261, 1282 (11th Cir. 2008) (quoting Peterson v. Willie, 81 F.3d 1033, 1036 (11th Cir. 1996)). There is no evidence to suggest that Ermini’s lawyer’s statement was “plainly unwarranted”—again, it seems at the very least to fall within the zone of statements that we have allowed—let alone “clearly injurious” to Scott’s case. Id. Moreover, when evaluating the prejudice resulting from an attorney’s comments, we examine comments in the context of the entire trial and any curative instruction, United States v. Hernandez, 145 F.3d 1433, 1438–39 (11th Cir. 1998), and here the district court judge explicitly instructed the jurors that they “must not be influenced in any way by either sympathy or prejudice for or against either party,” and that they “must follow the law . . . even if [they] do not agree with it.”
V
Finally, Scott challenges the admission of Lisenbee’s and Hamer’s testimony about their post-event terminations from the Sheriff’s Office—which all agree had nothing to do with the incident involving Ermini but which, Scott says, tended to paint the deputies in a bad light. The district court allowed limited questioning about the timing of and general reasons for Lisenbee’s and Hamer’s firings, but it excluded any related documents containing further details. In
24
particular, when examining Lisenbee, Ermini’s lawyer briefly asked whether he had been dismissed from the Sheriff’s Office; questioning revealed only that Lisenbee had been a probationary officer at the time of the Ermini incident and that he hadn’t “me[t his] probationary requirements.” Ermini’s lawyer also confirmed that Hamer had been fired for “[c]onduct unbecoming” after he “violate[d] a policy and fail[ed] to disclose it.” That was the extent of the evidence—no additional details were elicited, no explanatory records or documentation were introduced, and the remainder of the deputies’ examinations focused on the incident at issue.10 Because the issue at trial was whether Scott was vicariously liable for the deputies’ own failure to use reasonable care in conducting the wellness check, Scott contends that allowing Ermini’s lawyer to question Lisenbee and Hamer about their terminations—even to a limited extent—violated Federal Rules of Evidence 404(b) and 403. We discuss the application of those rules in turn.11
10 As a result of the judge’s exclusion of any additional evidence surrounding the deputies’ terminations, jurors didn’t learn (1) that Lisenbee’s failure to meet probationary requirements was partially related to his “inability to control scenes or suspects through verbal commands” and that he “had started remedial training two weeks before this incident”; or (2) that Hamer was dismissed after “an unrelated incident where inappropriate materials were found on his work computer.” 11 We review a district court’s evidentiary rulings for abuse of discretion. United States v. Brown, 665 F.3d 1239, 1247 (11th Cir. 2011).
25
A
In relevant part, Federal Rule of Evidence 404(b) states that “[e]vidence of a crime, wrong, or other act is not admissible to prove a person’s character in order to show that on a particular occasion the person acted in accordance with the character.” Fed. R. Evid. 404(b)(1). The rule’s language is broad—it applies to any “person[,]” not just to plaintiffs, defendants, or parties. Even so, we have held in both the civil and criminal contexts that Rule 404(b) does not—at least of its own force—apply when, as here, the challenged other-bad-acts evidence implicates a witness or another non-party to the litigation. See United States v. Morano, 697 F.2d 923, 926 (11th Cir. 1983) (concluding, in a criminal case, that “Rule 404(b) does not control” in instances where the evidence in question does not relate to acts of the defendant, but rather to “an extraneous offense committed by someone other than the defendant”); Glados, Inc. v. Reliance Ins. Co., 888 F.2d 1309, 1311–12 (11th Cir. 1987) (citing and applying Morano in the context of a civil case).
As we did in United States v. Sellers, we once again “question Morano’s reasoning,” which seems to us to flatly contradict Rule 404(b)’s clear text:
The plain language of Rule 404(b) refers to “persons,” not “defendants,” and Rule 404(a) carves out specific exceptions relating to the “accused,” “victim[s],” and “witness[es].” Where, as in Morano and this case, the non-defendant’s “extraneous” act, if proved, directly supports the guilt of the defendant as to the crime charged, that “extraneous” act should not, it seems, be subject to proof through the improper character-evidence route condemned by Rule
26
404. Indeed, it is difficult to see how the non-defendant’s act could be described as “extraneous” in such circumstances.
906 F.2d 597, 604 n.11 (1990). We recognize, though, as we did in Sellers, that we are “bound by [Morano] as precedent”—especially as extended to the civil- litigation context in Glados. Id. And finally we note—more happily, and again as we did in Sellers—that Rule 404(b)’s non-applicability probably doesn’t make much of a difference, because “Morano held that the factors articulated in Rule 404 ‘should be considered in weighing the balance between the relevancy of this evidence and its prejudice under Rule 403.’” Sellers, 906 F.2d at 604 n.11 (quoting Morano, 697 F.2d at 926). “The functional analysis is thus largely the same”—it just takes place under the auspices of Rule 403 rather than Rule 404(b). Id.
So we conclude—in accord with existing precedent, but dubitante—that Rule 404(b) doesn’t apply here, and we turn to consider whether the admission of evidence pertaining to Lisenbee’s and Hamer’s terminations violated Rule 403 and, if so, whether the error in admitting it was harmless.
B
Rule 403 states that a court “may exclude relevant evidence if its probative value is substantially outweighed by a danger of . . . unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence.” Fed. R. Evid. 403. This Court affords district court judges
27
“the broadest discretion in determining whether evidence should be excluded under Rule 403.” United States v. Costa, 947 F.2d 919, 924 (11th Cir. 1991) (quoting citation omitted).
Taking Rule 404(b) considerations into account—as Morano and Sellers require—we think the challenged testimony here could easily be understood to convey precisely the kind of propensity-based inference that the Rules of Evidence aim to prevent—namely, that Lisenbee and Hamer were bad cops (why else the firings?) with a penchant to act badly, just as Ermini alleges they did the night of the wellness check. Moreover, the circumstances in this case are unique in that while Scott is the defendant, he is only vicariously so—practically speaking, the deputies’ actions are on trial. Accordingly, it seems to us that “the policies underlying Rule 404(b)”—which again, we are instructed to fold into our Rule 403 analysis— are squarely applicable. Morano, 697 F.2d at 926. All things considered, we think it clear that the prejudicial value of this evidence could be thought to substantially outweigh its probative value.
That, though, isn’t the end of the inquiry. Even if a district court abuses its discretion by admitting prejudicial character evidence, evidentiary error “must [still] rise above the threshold of harmless error” to warrant reversal. Perry v. State Farm Fire & Cas. Co., 734 F.2d 1441, 1446 (11th Cir. 1984) (quoting Wallace v. Ener, 521 F.2d 215, 222 (5th Cir. 1975)). The burden of proving that
28
an error was not harmless rests with the party asserting it, who must establish that the error “affect[ed his] . . . substantial rights.” Id. “If . . . the error did not influence the jury, or had but very slight effect, the verdict and the judgment should stand . . . .” Kotteakos v. United States, 328 U.S. 750, 764 (1946).
Although we think the district court may well have abused its discretion in admitting Lisenbee’s and Hamer’s testimony, we conclude that any error was harmless, for several reasons. First, Ermini’s lawyer’s questioning of the witnesses was brief and was largely limited to generalities about their dismissals; the district court properly excluded explanatory records and documents, which would have introduced additional details. Second, Ermini’s lawyer asked all of her witnesses about their employment history and current job status, save for those who appeared in uniform and whose status was therefore obvious—so there wasn’t any undue focus on Lisenbee and Hamer. Third, Lisenbee and Hamer clarified on cross- examination that they were not terminated—or even disciplined or criticized—as a result of anything they did (or didn’t do) in connection with the Ermini incident, and, further, that the Sheriff’s Office never even concluded that they “violated policy” during the wellness check.
Because we conclude that Scott hasn’t shown that the deputies’ testimony about their terminations affected his substantial rights, we find no reversible error.
29
VI
For the foregoing reasons, we hold (1) that it wasn’t improper for the district court to inform the jury about the legal effect of its finding under Florida’s alcohol defense; (2) that Ermini didn’t present a nonexistent negligent-use-of-force claim and that her negligent-wellness-check claim wasn’t precluded; (3) that Ermini’s lawyer didn’t make a forbidden golden-rule argument; and (4) that any error that the district court committed in admitting testimony about Lisenbee’s and Hamer’s dismissals was harmless.12 Accordingly, we affirm the jury’s verdict and the district court’s judgment, as well as that court’s denial of Scott’s motion for new trial. AFFIRMED.
12 Scott also contends that all of these errors, taken together, resulted in cumulative error sufficient to merit a new trial. Because we find no errors, Scott’s cumulative-error argument necessarily fails.