Lawanna Tynes v. Florida Department of Juvenile Justice

88 F.4th 939

• Court: Court of Appeals for the Eleventh Circuit
• Decided: December 12, 2023
• Docket: 21-13245
• Status: Published

Opinion

[PUBLISH]

In the

United States Court of Appeals For the Eleventh Circuit

____________________

No. 21-13245 ____________________

LAWANNA TYNES, Plaintiff-Appellee, versus FLORIDA DEPARTMENT OF JUVENILE JUSTICE,

Defendant-Appellant.

Appeal from the United States District Court for the Southern District of Florida D.C. Docket No. 0:18-cv-62891-WPD ____________________ 2 Opinion of the Court 21-13245

Before JILL PRYOR, NEWSOM, and GRANT, Circuit Judges. GRANT, Circuit Judge: This appeal results from an all-too-common confusion in employment discrimination suits: whether the evidentiary framework set out in McDonnell Douglas is a stand-in for the ultimate question of liability in Title VII discrimination cases. We repeat today what our precedents have already made clear: It is not. Properly understood, McDonnell Douglas is an evidentiary framework that shifts the burden of production between the parties to figure out if the true reason for an adverse employment action was the employee’s race. It is not a set of elements that the employee must prove—either to survive summary judgment or prevail at trial. To be sure, in some cases a lack of success in establishing a prima facie case will also reflect a lack of success in showing employment discrimination. But, as both this Court and the Supreme Court have explained, the ultimate question in a discrimination case is whether there is enough evidence to show that the reason for an adverse employment action was illegal discrimination. The prima facie case in the McDonnell Douglas framework can help answer that question—but it cannot replace it. Here, the Florida Department of Juvenile Justice is distracted by a perceived failure on the part of its former employee, Lawanna Tynes, to meet her initial burden of production at the prima facie stage of McDonnell Douglas. But that distraction comes with a price—a lack of focus on whether Tynes put forward 21-13245 Opinion of the Court 3

enough evidence to show that she was fired because of racial discrimination. The jury thought so, and the Department does not challenge the sufficiency of the evidence for that conclusion. The verdict thus stands. The Department also argues that Tynes did not adequately plead a claim for race discrimination under 42 U.S.C. § 1981, which requires a different standard of causation than Title VII—and, perhaps more importantly for the Department’s purposes here, offers a higher level of potential damages. But again, the Department sets its sights on the wrong target. Though the district court’s order expressly relied on its authority to permit amendments to the pleadings under Rule 15(b)(1) of the Federal Rules of Civil Procedure, the Department does not even cite Rule 15(b)(1) on appeal. That means the challenge is forfeited, so we also affirm the district court’s order denying the Department’s motion for judgment as a matter of law on Tynes’s § 1981 claim. I. Tynes was employed by the Florida Department of Juvenile Justice for sixteen years. At the time of her termination, she was the superintendent of the Broward Regional Juvenile Detention Center. The superintendent’s responsibilities include overseeing the facility’s operations and ensuring that both juvenile detainees and staff are in a safe environment. One Sunday, while Tynes was off for medical leave, an unusually high number of incidents required an officer to call for back up. The assistant secretary of detention services, Dixie Fosler, 4 Opinion of the Court 21-13245

followed up by assembling a technical assistance team to review staffing and personnel issues. After the team’s review was complete—but before its report was issued—Fosler terminated Tynes. Tynes had no prior negative performance review or reprimands. Even so, the Department offered a laundry list of reasons for the termination: poor performance, negligence, inefficiency or inability to perform assigned duties, violation of law or agency rules, conduct unbecoming of a public employee, and misconduct. Tynes sued, alleging race and sex discrimination. Her complaint unambiguously alleged two violations of Title VII of the Civil Rights Act of 1964, which prohibits employers from terminating employees because of their race or sex. 42 U.S.C. § 2000e-2(a)(1). The complaint also stated that it brought “other causes of actions [sic] which can be inferred from the facts herein.” The basis of Tynes’s discrimination case was that similarly situated white and male employees were treated differently and that the Department’s stated reasons for her termination were pretextual. For comparator evidence, Tynes pointed to Joseph Seeber, a white male, and Daryl Wolf, a white female, who were both superintendents of juvenile detention centers with incidents that reflected a lack of control or failure to abide by the Department’s policies. 1 But, unlike Tynes, neither was terminated.

1 At summary judgment, the district court held that Seeber and Wolf were

both appropriate comparators. 21-13245 Opinion of the Court 5

Far from it—they received only oral reprimands, were allowed to transfer to different facilities, and were granted multiple opportunities to comply with various recommendations for improvement. As for pretext, Tynes presented evidence of Fosler’s personal bias against her. Gladys Negron, Tynes’s direct supervisor, testified that she believed Tynes’s termination was based on Fosler’s personal feelings rather than professional concerns. She said that Fosler’s written report “contained several inaccuracies,” and even characterized the technical assistance team’s efforts as a “search- and-kill mission” against Tynes. At trial, Fosler faltered in her testimony; she could not recall the basis for her conclusion that Tynes had engaged in “conduct unbecoming as a public employee,” nor could she point to another employee fired without negative performance reviews or prior reprimands. The jury returned its verdict in favor of Tynes and made specific findings in a special verdict form: (1) “race or sex was a motivating factor”; (2) the Department would not have discharged Tynes if it had not taken into account her race or sex; and (3) Tynes’s race was a but-for cause of her termination. The jury awarded $424,600 in compensatory damages and $500,000 in damages for emotional pain and mental anguish. The district court ordered the Department to reinstate Tynes to a similar position— but not under Fosler’s supervision. The Department filed a renewed motion for judgment as a matter of law or, alternatively, for a new trial. It argued that the 6 Opinion of the Court 21-13245

Department was entitled to judgment on Tynes’s Title VII claims because she did not present comparators who were “similarly situated in all material respects” and therefore failed to satisfy her burden to establish a prima facie case under McDonnell Douglas. The filing also asserted that Tynes had not properly pleaded her § 1981 claim. A § 1981 claim differs in two relevant ways from a Title VII claim—there is no cap on damages and the causation standards are higher. 42 U.S.C. § 1981a(b)(3)–(4); see Comcast Corp. v. Nat’l Ass’n of Afr. Am.-Owned Media, 140 S. Ct. 1009, 1017–19 (2020). The district court denied the motion on both issues. It rejected the Department’s Title VII arguments because “the circumstantial evidence regarding the two comparators was sufficient to establish the discrimination claims,” and “[c]redibility was for the jury to decide.” The court also rejected the § 1981 argument, saying that even if Tynes had not properly pleaded that violation in the first place, Rule 15(b)(1) of the Federal Rules of Civil Procedure gave it “the discretion to allow an amendment” to the complaint during the trial. The Department now appeals the district court’s denial of its renewed motion for judgment as a matter of law. II. Judgment as a matter of law is appropriate when “the facts and inferences point so overwhelmingly in favor of one party that reasonable people could not arrive at a contrary verdict.” Brown v. Alabama Dep’t of Transp., 597 F.3d 1160, 1173 (11th Cir. 2010) 21-13245 Opinion of the Court 7

(alterations adopted and quotation omitted). We review the denial of a motion for judgment as a matter of law de novo. Id. III. A. Title VII of the Civil Rights Act of 1964 outlaws employment discrimination because of “race, color, religion, sex, or national origin.” 42 U.S.C. § 2000e-2(a)(1). Likewise, 42 U.S.C. § 1981 prohibits employers from intentionally discriminating on the basis of race in employment contracts. See Johnson v. Ry. Express Agency, Inc., 421 U.S. 454, 459–60 (1975); Ferrill v. Parker Grp., Inc., 168 F.3d 468, 472 (11th Cir. 1999). To prove a claim under either statute, a plaintiff can use direct evidence, circumstantial evidence, or both. See Jenkins v. Nell, 26 F.4th 1243, 1249 (11th Cir. 2022). Early on, though, it became clear that when only circumstantial evidence was available, figuring out whether the actual reason that an employer fired or disciplined an employee was illegal discrimination was difficult and “elusive.” Texas Dep't of Cmty. Affs. v. Burdine, 450 U.S. 248, 255 n.8 (1981). After all, an employer can generally fire or discipline an employee for “a good reason, a bad reason, a reason based on erroneous facts, or for no reason at all,” so long as that action “is not for a discriminatory reason.’” Flowers v. Troup Cnty. Sch. Dist., 803 F.3d 1327, 1338 (11th Cir. 2015) (quoting Nix v. WLCY Radio/Rahall Commc’ns, 738 F.2d 1181, 1187 (11th Cir. 1984)). To deal with the difficulties encountered by both parties and courts, the Supreme Court in McDonnell Douglas set out a burden 8 Opinion of the Court 21-13245

shifting framework designed to draw out the necessary evidence in employment discrimination cases. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802 (1973). It works like this. Step one is for the plaintiff, who establishes what McDonnell Douglas calls a “prima facie” case of discrimination when she shows that (1) “she belongs to a protected class,” (2) “she was subjected to an adverse employment action,” (3) “she was qualified to perform the job in question,” and (4) “her employer treated ‘similarly situated’ employees outside her class more favorably.” McDonnell Douglas, 411 U.S. at 802; Lewis v. City of Union City, 918 F.3d 1213, 1220–21 (11th Cir. 2019) (en banc). The last requirement is met when the plaintiff presents “evidence of a comparator—someone who is similarly situated in all material respects.” Jenkins, 26 F.4th at 1249 (quotation omitted). The prima facie showing entitles the plaintiff to a rebuttable presumption of intentional discrimination. U.S. Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 714–15 (1983). The defendant then rebuts that presumption (if it can) by offering evidence of a valid, non-discriminatory justification for the adverse employment action. Id. at 714. Once that justification is offered, the presumption of discrimination falls away and the plaintiff tries to show not only that the employer’s justification was pretextual, but that the real reason for the employment action was discrimination. Id. at 714–15; Burdine, 450 U.S. at 256. This final question “merges with the plaintiff’s ultimate burden of persuading the factfinder that she has been the victim of intentional discrimination.” Lewis, 918 F.3d at 1221 (quoting Burdine, 450 U.S at 256 (alterations adopted)). 21-13245 Opinion of the Court 9

McDonnell Douglas, in short, is an evidentiary tool that functions as a “procedural device, designed only to establish an order of proof and production.” St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502, 521 (1993); see also Burdine, 450 U.S. at 255 n.8; Furnco Constr. Corp. v. Waters, 438 U.S. 567, 577 (1978). What McDonnell Douglas is not is an independent standard of liability under either Title VII or § 1981. Nor is its first step, the prima facie case—“establishing the elements of the McDonnell Douglas framework is not, and never was intended to be, the sine qua non for a plaintiff to survive a summary judgment motion.” Smith v. Lockheed-Martin Corp., 644 F.3d 1321, 1328 (11th Cir. 2011); see also Brady v. Off. of the Sergeant at Arms, 520 F.3d 490, 493–94 (D.C. Cir. 2008). Often, however, parties (and sometimes courts) miss this fundamental point and wrongly treat the prima facie case as a substantive standard of liability. To be fair, the McDonnell Douglas court’s terminology likely bears some responsibility for the continuing confusion on this point. When the Supreme Court uses the term “prima facie case” in this context, it does so “in a special sense.” Wells v. Colorado Dep’t of Transp., 325 F.3d 1205, 1223 (10th Cir. 2003) (Hartz, J., writing separately). The Court itself has explained that although that phrase may sometimes “describe the plaintiff’s burden of producing enough evidence to permit the trier of fact to infer the fact at issue,” within the McDonnell Douglas framework the term “prima facie case” has a different meaning—it marks “the establishment of a legally mandatory, rebuttable presumption.” Burdine, 450 U.S. at 254 n.7 (citing 9 J. Wigmore, Evidence § 2494 (3d ed. 1940)). 10 Opinion of the Court 21-13245

So, although in other contexts a prima facie case typically does mean enough evidence for a plaintiff to prevail on a particular claim, here the meaning is different. Under McDonnell Douglas, a plaintiff who establishes a prima facie case is entitled to a “legally mandatory, rebuttable presumption” that the employer intentionally discriminated against her. Id. What that means is that once a plaintiff satisfies her prima facie burden, the defendant “knows that its failure to introduce evidence of a nondiscriminatory reason will cause judgment to go against it.” Hicks, 509 U.S. at 510 n.8. The presumption of discrimination introduced by the prima facie case thus helps narrow things down and “frame the factual issue” by drawing out an explanation that the plaintiff can then seek to demonstrate is pretextual. Burdine, 450 U.S. at 255. In this way, the prima facie showing exerts a sort of “practical coercion” that forces the defendant to “come forward” with evidence explaining its actions. Hicks, 509 U.S. at 510 n.8, 511. It also offers a benefit for the defendant employer, who now has a better idea of what evidence needs to be rebutted. See id. But once the prima facie case has “fulfilled its role of forcing the defendant to come forward with some response,” it no longer has any work to do. Id. at 510–11. Where “the defendant has done everything that would be required of him if the plaintiff had properly made out a prima facie case, whether the plaintiff really did so is no longer relevant.” Aikens, 460 U.S. at 715 (emphasis added). This is so because the “district court has before it all the evidence it needs to decide whether the defendant intentionally discriminated against the plaintiff.” Id. (quotation omitted). So when the 21-13245 Opinion of the Court 11

defendant employer offers evidence of the reason for its actions toward the plaintiff, the presumption of discrimination created by the prima facie case “simply drops out of the picture.” Hicks, 509 U.S. at 511; see also Turnes v. AmSouth Bank, NA, 36 F.3d 1057, 1061 (11th Cir. 1994). That is a far cry from serving as a substitute standard necessary to survive summary judgment. Another reason for the confusion? A failure in the prima facie case often also reflects a failure of the overall evidence. Even though we do not dwell on whether the technical requirements of the prima facie case are met once the defendant has met its burden of production, we keep in mind that the questions the plaintiff must answer to make a prima facie case are relevant to the ultimate question of discrimination. A plaintiff who fails to prove that she was a member of a protected class, for example, or that she suffered an adverse employment action, will be unable to prove that she was unlawfully discriminated against. See Standard v. A.B.E.L. Servs., Inc., 161 F.3d 1318, 1327–28 (11th Cir. 1998); Kidd v. Mando Am. Corp., 731 F.3d 1196, 1202–04 (11th Cir. 2013). We’ll admit that we have at times framed that analysis in terms of whether the plaintiff has established a prima facie case, but the more fundamental problem with such a failure of evidence is that it means the plaintiff cannot prove a necessary element for his employment discrimination case. See, e.g., Kidd, 731 F.3d at 1202–04. This distinction is important because the components of a prima facie case are not necessarily coextensive with the evidence needed to prove an employment discrimination claim. That is why 12 Opinion of the Court 21-13245

a plaintiff need not plead the elements of a prima facie case to survive a motion dismiss. Swierkiewicz v. Sorema N. A., 534 U.S. 506, 515 (2002). And it explains why courts in this Circuit do not instruct juries on the prima facie case or the McDonnell Douglas framework. See Dudley v. Wal-Mart Stores, Inc., 166 F.3d 1317, 1322 (11th Cir. 1999). It is also why “the plaintiff’s failure to produce a comparator does not necessarily doom the plaintiff’s case.” Smith, 644 F.3d at 1328. Indeed, “the plaintiff will always survive summary judgment if he presents circumstantial evidence that creates a triable issue concerning the employer’s discriminatory intent.” Id. at 1328. That is because McDonnell Douglas is “only one method by which the plaintiff can prove discrimination by circumstantial evidence.” Vessels v. Atlanta Indep. Sch. Sys., 408 F.3d 763, 768 n.3 (11th Cir. 2005). A plaintiff who cannot satisfy this framework may still be able to prove her case with what we have sometimes called a “convincing mosaic of circumstantial evidence that would allow a jury to infer intentional discrimination by the decisionmaker.” Smith, 644 F.3d at 1327–28 (footnote and quotation omitted); see also Lewis v. City of Union City, 934 F.3d 1169, 1185 (11th Cir. 2019) (Lewis II). This rearticulation of the summary judgment standard arose in large part because of widespread misunderstandings about the limits of McDonnell Douglas—the same misunderstandings that persist today. A “convincing mosaic” of circumstantial evidence is simply enough evidence for a reasonable factfinder to infer 21-13245 Opinion of the Court 13

intentional discrimination in an employment action—the ultimate inquiry in a discrimination lawsuit.2 Jenkins, 26 F.4th at 1250. This approach to analyzing the evidence treats an employment discrimination suit in same way we would treat any other case— jumping directly to the ultimate question of liability and deciding whether the moving party is entitled to judgment at that stage of the case. It is no different than the standards we ordinarily apply in deciding summary judgment and post-trial motions. “If the plaintiff presents enough circumstantial evidence to raise a reasonable inference of intentional discrimination, her claim will survive summary judgment.” Hamilton v. Southland Christian Sch., 680 F.3d 1316, 1320 (2012). All that to say, in deciding motions for summary judgment or judgment as a matter of law, parties already understand that, when we use what we have called the convincing mosaic standard,

2 A plaintiff proving her case through the convincing mosaic standard may

point to any relevant and admissible evidence. As we have said, “no matter its form, so long as the circumstantial evidence raises a reasonable inference that the employer discriminated against the plaintiff, summary judgment is improper.” Smith, 644 F.3d at 1328. Evidence that is likely to be probative is “evidence that demonstrates, among other things, (1) suspicious timing, ambiguous statements, or other information from which discriminatory intent may be inferred, (2) systematically better treatment of similarly situated employees, and (3) pretext.” Jenkins, 26 F.4th at 1250 (quotation omitted). Given the wide scope of available evidence, the convincing mosaic standard “can be of particular significance when the plaintiff cannot identify a similarly situated comparator,” as the McDonnell Douglas framework requires. Bailey v. Metro Ambulance Servs., Inc., 992 F.3d 1265, 1273 n.1 (11th Cir. 2021). 14 Opinion of the Court 21-13245

we look beyond the prima facie case to consider all relevant evidence in the record to decide the ultimate question of intentional discrimination. But parties do not always understand that we are answering that same question when using the McDonnell Douglas framework. Under McDonnell Douglas, the failure to establish a prima facie case is fatal only where it reflects a failure to put forward enough evidence for a jury to find for the plaintiff on the ultimate question of discrimination. This may mean that there was not enough evidence to infer discrimination. Or it may be that there was no adverse employment action. But the analysis turns on the substantive claims and evidence in the case, not the evidentiary framework. For these reasons, we have repeatedly emphasized that after a trial we “should not revisit whether the plaintiff established a prima facie case.” Cleveland v. Home Shopping Network, Inc., 369 F.3d 1189, 1194 (11th Cir. 2004); see also, e.g., Holland v. Gee, 677 F.3d 1047, 1056 (11th Cir. 2012); Collado v. United Parcel Serv., Co., 419 F.3d 1143, 1150 (11th Cir. 2005); Tidwell v. Carter Prods., 135 F.3d 1422, 1426 n.1 (11th Cir. 1998); Richardson v. Leeds Police Dep’t, 71 F.3d 801, 806 (11th Cir. 1995); Carmichael v. Birmingham Saw Works, 738 F.2d 1126, 1129 (11th Cir. 1984). Instead, we ask only one question: whether there is a sufficient evidentiary basis for the jury to find that the defendant intentionally discriminated against the plaintiff. Cleveland, 369 F.3d at 1194. 21-13245 Opinion of the Court 15

B. That analysis solves this case. The Department’s only argument is that the comparator employees that Tynes offered were not adequate to establish a prima facie case of discrimination under McDonnell Douglas. That may be true; under our precedent a comparator employee must be “similarly situated in all material respects”—a high bar to meet. Lewis, 918 F.3d at 1218. But the jury’s factual inquiry was whether the Department intentionally discriminated against Tynes, and its answer was “yes.” The Department does not contend that the evidence, taken as a whole, could not support the jury’s verdict. By focusing exclusively on Tynes’s comparator evidence, the Department has forfeited any challenge to the ultimate finding of discrimination. Of course, the strength of Tynes’s comparator evidence is relevant to the ultimate question of intentional discrimination. Holland, 677 F.3d at 1056–57. But to the extent that there are material differences between Tynes and her comparators at this stage of the case, it is the jury’s role—not ours—to determine how much weight the comparator evidence should be given. In other words, it is possible that her comparators were insufficient to establish a prima facie case yet still relevant to the ultimate question of intentional discrimination. See Lewis II, 934 F.3d at 1187–88. To win after trial, the Department would have needed to explain why the evidence, taken as a whole, was insufficient to support the jury’s verdict. Because it failed to do so, we affirm the judgment of the district court denying the Department’s renewed motion for judgment as a matter of law on the Title VII claims. 16 Opinion of the Court 21-13245

IV. The Department also challenges the jury’s verdict on Tynes’s § 1981 claim, arguing that her complaint did not adequately plead the § 1981 claim and that she did not prove that race was a “but-for” cause of her termination.3 The Department, however, has forfeited both arguments. The Department is right about one thing—Tynes’s complaint may not have set out a separate claim under § 1981. See Weiland v. Palm Beach Cnty. Sheriff’s Off., 792 F.3d 1313, 1322–23 (11th Cir. 2015) (requiring a complaint to set out a different count for each cause of action or claim for relief ).4 Even so, the district court held that it had discretion to allow an amendment to the pleadings during the trial under Rule 15(b)(1). That rule permits the pleadings to be amended at trial when “a party objects that evidence is not within the issues raised in the pleadings” so long as “doing so will aid in presenting the merits and the objecting party fails to satisfy the court that the evidence would prejudice that

3 In Comcast, the Supreme Court held that but-for causation was required to

prove a § 1981 claim. 140 S. Ct. at 1019. 4 In addition to the Title VII claims, the complaint says it brings “other causes

of actions [sic] which can be inferred from the facts herein.” But it does not set out a § 1981 claim in its own count; instead, it refers to § 1981 in the jurisdictional section of the complaint as a federal question presented in the case. What’s more, each of Tynes’s Title VII counts alleges that she “is a member of a protected class under § 1981,” and the prayer for relief requests that the court “[a]djudge and decree that Defendant has violated 42 U.S.C. § 1981.” 21-13245 Opinion of the Court 17

party’s action or defense on the merits.” Fed. R. Civ. P. 15(b)(1). The district court stated that it found that permitting amendment would not prejudice the Department. The Department does not challenge the district court’s authority under Rule 15. Indeed, at oral argument counsel expressed a lack of familiarity with that rule. And when “an appellant fails to challenge properly on appeal one of the grounds on which the district court based its judgment, he is deemed to have abandoned any challenge of that ground.” Sapuppo v. Allstate Floridian Ins. Co., 739 F.3d 678, 680 (11th Cir. 2014). So while it is not clear whether the district court properly invoked Rule 15(b)(1)—after all, Tynes did not actually move to amend her complaint—any challenge on that ground is forfeited. See Molinos Valle Del Cibao, C. por A. v. Lama, 633 F.3d 1330, 1352 (11th Cir. 2011); Green Country Food Mkt., Inc., v. Bottling Grp., LLC, 371 F.3d 1275, 1281 (10th Cir. 2004). The Department’s second § 1981 argument—that Tynes did not prove that race was a but-for cause of her termination—is also forfeited. In its post-trial motion, the Department argued that because Tynes did not plead a § 1981 claim, her complaint did not allege that race was a but-for cause. But it did not argue that Tynes failed to prove that race was a but-for cause. 5 “It is well-settled that

5 The clear intention of the Department’s Rule 50 motions was to challenge

the adequacy of the pleadings. The Department may contend (though it did not do so directly before this Court) that it preserved a proof-based argument with this statement: “Plaintiff offered no testimony or evidence at trial that her 18 Opinion of the Court 21-13245

we will generally refuse to consider arguments raised for the first time on appeal.” Ramirez v. Sec’y, U.S. Dep’t of Transp., 686 F.3d 1239, 1249 (11th Cir. 2012). The Department cannot now repackage its pleading argument into a claim that Tynes did not prove an essential element at trial. * * * After a full trial on the merits, a defendant cannot successfully challenge the jury’s verdict by arguing only that the plaintiff’s comparators were inadequate or that the prima facie case was otherwise insufficient. Here, the Department was required to demonstrate why the record evidence could not support the jury’s verdict and failed to do so. Because the Department also failed to adequately challenge the grounds upon which the district court denied its motion with respect to Tynes’s § 1981 claim, the district court’s order is AFFIRMED.

race was the ‘but-for’ cause of her termination.” In context, both we and the district court read this as support for the pleading-based argument, but in any event, such a statement is far too conclusory on its own to preserve the issue for appeal. 21-13245 NEWSOM, J., Concurring 1

NEWSOM, Circuit Judge, concurring: Today’s majority opinion offers an important critique of the role that McDonnell Douglas’s burden-shifting analysis has come to play in deciding Title VII cases. In particular, the majority explains that McDonnell Douglas (1) provides only an “evidentiary framework” and (2) was never meant to establish “an independent standard of liability” or specify a “set of elements that the employee must prove—either to survive summary judgment or prevail at trial.” Maj. Op. at 2, 9. Unfortunately, as the majority notes, “parties (and sometimes courts)” often “miss this fundamental point and wrongly treat” McDonnell Douglas, and in particular its initial prima-facie-case step, “as a substantive standard of liability.” Id. at 9. And although this case doesn’t arise on summary judgment, the majority correctly observes that the overreading of—and consequent overemphasis on—McDonnell Douglas has become particularly acute at the Rule 56 stage, where courts have increasingly taken to treating the test’s prima-facie- evidence benchmark “as a substitute standard necessary to survive summary judgment.” Id. at 11; see also id. at 9–12 (detailing the problems with courts’ applications of McDonnell Douglas at summary judgment). Yes, yes, and yes—I completely agree. I’ll confess, though, that I’ve developed an even deeper skepticism of McDonnell Douglas. The majority opinion seeks to put courts back on the right path in their application of McDonnell Douglas; I tend to think we might be better off on an altogether different path. Here’s what I mean: I’d 2 NEWSOM, J., Concurring 21-13245

long taken for granted that McDonnell Douglas’s three-step framework provided the presumptively proper means of deciding Title VII cases at summary judgment. I’ve changed my mind. McDonnell Douglas, it now seems to me, not only lacks any real footing in the text of Rule 56 but, worse, actually obscures the answer to the only question that matters at summary judgment: Has the plaintiff shown a “genuine dispute as to any material fact”—in the typical Title VII case, as to whether her employer engaged in discrimination based on a protected characteristic. Instead of McDonnell Douglas—which, to be clear, neither the Supreme Court nor we have ever said provides the sole mechanism for adjudicating summary-judgment motions—courts should employ something like our oft-maligned “convincing mosaic” standard, which I had always viewed as something of a rogue but which, upon reflection, much more accurately captures and implements the summary-judgment standard. For me, it’s quite the turnabout, so I should explain myself. I Title VII of the landmark Civil Rights Act of 1964 broadly prohibits workplace discrimination. In relevant part, its operative provision states that— It shall be an unlawful employment practice for an employer to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of 21-13245 NEWSOM, J., Concurring 3

such individual’s race, color, religion, sex, or national origin . . . .

42 U.S.C. § 2000e-2(a)(1). Title VII was (and is) an historic piece of legislation that tackled (and continues to tackle) one of the country’s weightiest social problems. Legally speaking, though, it’s just a statute, no different from hundreds of others. And so, as the Supreme Court has repeatedly reminded us, the “ordinary rules” of civil procedure apply to Title VII cases. Swierkiewicz v. Sorema N.A., 534 U.S. 506, 511 (2002) (“[T]he ordinary rules for assessing the sufficiency of a complaint apply.”); see also, e.g., U.S. Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 716 (1983) (“[N]one of this means that trial courts or reviewing courts should treat discrimination differently from other ultimate questions of fact.”). Many, if not most, Title VII cases are decided at summary judgment. The “ordinary rule[]” for evaluating the propriety of summary judgment, of course, is Federal Rule of Civil Procedure 56: The court shall grant summary judgment if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.

Fed. R. Civ. P. 56(a). In the mine-run discrimination case, the key issue is whether the employer engaged in some action, in the statute’s words, “because of ” an employee’s race, sex, religion, or other protected characteristic. Accordingly, the fundamental question at summary judgment is—or should be—whether there is 4 NEWSOM, J., Concurring 21-13245

a genuine dispute of material fact about that all-important causation issue. But not all analytical frameworks hew closely to that question. Briefly, we assess employment-discrimination cases at summary judgment using one or more of three approaches. First, a reviewing court might consider whether the plaintiff has pointed to direct evidence of discrimination. If the case instead turns on circumstantial evidence, the court might ask—second—whether the plaintiff can survive McDonnell Douglas’s burden-shifting analysis or—third—whether she can assemble what we have called a “convincing mosaic” of evidence suggesting discrimination, Smith v. Lockheed-Martin Corp., 644 F.3d 1321, 1328 (11th Cir. 2011). In terms of consistency with Rule 56, the direct-evidence analysis, reserved for cases featuring particularly “blatant” and overtly discriminatory comments or conduct, see Fernandez v. Trees, Inc., 961 F.3d 1148, 1156 (11th Cir. 2020), performs well enough. But direct-evidence cases are increasingly rare, so most Title VII suits these days are turn on circumstantial evidence. Among those, McDonnell Douglas is clearly the dominant framework, with “convincing mosaic” trailing along as something of an afterthought. 1 And until recently, that seemed exactly right to

1 So far as I can tell, we have considered the convincing-mosaic test in only

five published Title VII decisions, three of which involved cursory single- paragraph rejections of a plaintiff’s invocation of it. See, e.g., Smith v. Lockheed- Martin Corp., 644 F.3d 1321, 1328 (11th Cir. 2011); Flowers v. Troup Cnty. Sch. Dist., 803 F.3d 1327, 1335 (11th Cir. 2015); Trask v. Secretary, Dep’t of Veterans Affs., 822 F.3d 1179, 1193 (11th Cir. 2016), abrogated on other grounds by Babb v. 21-13245 NEWSOM, J., Concurring 5

me—I had marinated in McDonnell Douglas and its progeny for so long that I had come to view the convincing-mosaic test as an interloper, a hack contrived to save cases that might otherwise go out on summary judgment. I’ve concluded that I was wrong about that—as in 180° wrong. Upon reflection, it now seems to me that McDonnell Douglas is the interloper—it is the judge-concocted doctrine that obfuscates the critical inquiry. The convincing-mosaic standard, by contrast—despite its misleadingly florid label—is basically just Rule 56 in operation. Quite unlike McDonnell Douglas, it actually asks the key question: Does the “record, viewed in a light most favorable to the plaintiff, present[] a convincing mosaic of circumstantial evidence that would allow a jury to infer intentional discrimination by the decisionmaker”? Smith, 644 F.3d at 1328 (internal quotations and footnote omitted). Strip away the grandiloquence—after all, “convincing mosaic of circumstantial evidence” just means “evidence”—and that is exactly Rule 56’s summary-judgment standard. In the discussion that follows, I’ll explain briefly why I’ve come to believe (1) that McDonnell Douglas is the wrong framework to apply in deciding Title VII cases at summary judgment and (2) that our convincing-mosaic standard—which I’d rebrand slightly—

Wilkie, 140 S. Ct. 1168 (2020); Lewis v. City of Union City, 934 F.3d 1169, 1185 (11th Cir. 2019) (on remand); Bailey v. Metro Ambulance Servs., Inc., 992 F.3d 1265, 1273 n.2 (11th Cir. 2021). 6 NEWSOM, J., Concurring 21-13245

is the right one. I’ll also try to anticipate and respond to a few objections. II To start, why the loss of faith in McDonnell Douglas? In short, I fear that it doesn’t reliably get us to the result that Rule 56 requires. See also Maj. Op. at 11 (noting that “the components of a prima facie case are not necessarily coextensive with the evidence needed to prove an employment discrimination claim”). And in retrospect, that shouldn’t be particularly surprising, because McDonnell Douglas’s reticulated, multi-step framework forces courts to ask and answer a series of questions that only peripherally relate to the one that Rule 56 poses: Has the plaintiff presented “a genuine issue as to any material fact”—in the typical Title VII case, about her employer’s discriminatory intent? Let me unpack my concern, in three parts. First, as a threshold matter, McDonnell Douglas seems (in retrospect) awfully made up. Here’s how the Supreme Court has described its handiwork: In McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973), we set forth the basic allocation of burdens and order of presentation of proof in a Title VII case alleging discriminatory treatment. First, the plaintiff has the burden of proving by the preponderance of the evidence a prima facie case of discrimination. Second, if the plaintiff succeeds in proving the prima facie case, the burden shifts to the defendant “to articulate some legitimate, nondiscriminatory reason 21-13245 NEWSOM, J., Concurring 7

for the employee’s rejection.” Third, should the defendant carry this burden, the plaintiff must then have an opportunity to prove by a preponderance of the evidence that the legitimate reasons offered by the defendant were not its true reasons, but were a pretext for discrimination.

Texas Dep’t of Cmty. Affs. v. Burdine, 450 U.S. 248, 252–53 (1981) (internal citations and footnote omitted). There’s certainly no textual warrant in Title VII or the Federal Rules for so elaborate a scheme, and so far as I know, no one has ever even sought to justify it as rooted in either. Perhaps a product of its time, the whole thing is quite legislative, quite Miranda-esque—“set forth,” to use the Supreme Court’s own words. See also Maj. Op. at 7–8 (observing that McDonnell Douglas “set out” the burden-shifting framework). And for me, the framework’s made-up-ed-ness is a flashing red light—prima facie evidence, if you will, that something is amiss. Cf. Club Madonna Inc. v. City of Miami Beach, 42 F.4th 1231, 1261 (11th Cir. 2022) (Newsom, J., concurring) (“[U]nelected, unaccountable federal judges shouldn’t make stuff up.”). Second, whatever it was that the Supreme Court initially conjured, it seems to have taken on a life of its own. Perhaps most jarringly, McDonnell Douglas’s burden-shifting framework has become the presumptive means of resolving Title VII cases at summary judgment—despite the facts (1) that McDonnell Douglas itself arose not on summary judgment but out of a bench trial, see Green v. McDonnell Douglas Corp., 299 F. Supp. 1100, 1102 (E.D. Mo. 1969), and (2) that, so far as I can tell, the Supreme Court has 8 NEWSOM, J., Concurring 21-13245

specifically addressed McDonnell Douglas’s application to Title VII cases at summary judgment only once, and in that decision held that it didn’t apply, see Trans World Airlines, Inc. v. Thurston, 469 U.S. 111, 118–19 (1985). 2 Even beyond that, despite the Supreme Court’s occasional reminders that McDonnell Douglas’s “procedural device” was intended “only to establish an order of proof and production,” St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502, 521 (1993), 3 lower courts have become progressively obsessed with its minutiae, allowing it to drive substantive outcomes. The framework’s constituent details have grown increasingly intricate and code-like, as courts have taken to forcing a holistic evidentiary question—

2 Ironically, resolving cases at summary judgment seems to be McDonnell

Douglas’s sole remaining office. The Supreme Court has clarified that its burden-shifting analysis is inapplicable both at the pleading stage, see Swierkiewicz, 534 U.S. at 508, and in deciding post-trial motions, see Aikens, 460 U.S. at 715, and most courts of appeals have excised references to McDonnell Douglas’s framework from their pattern jury instructions, see Timothy M. Tymkovich, The Problem with Pretext, 85 Denv. U. L. Rev. 503, 528 & nn.189– 91 (2008) (collecting cases). To be fair, the Court has utilized McDonnell Douglas to evaluate claims under other statutes at summary judgment. None of those decisions, though, has squarely addressed McDonnell Douglas’s consistency (or inconsistency) with Rule 56. See, e.g., Babb v. Wilkie, 140 S. Ct. 1168, 1172 (2020); Young v. United Parcel Serv., Inc., 575 U.S. 206, 231 (2015); Raytheon Co. v. Hernandez, 540 U.S. 44, 51–52 (2003); O’Connor v. Consolidated Coin Caterers Corp., 517 U.S. 308, 312 (1996). 3 See also Burdine, 450 U.S. at 255 n.8 (observing that the McDonnell Douglas

framework was designed merely to help the parties progressively “sharpen the inquiry into the elusive factual question of intentional discrimination”). 21-13245 NEWSOM, J., Concurring 9

whether all the evidence, viewed in the light most favorable to the plaintiff, creates a genuine factual dispute—into a collection of distinct doctrinal pigeonholes. For instance, we have explained— and we’re hardly alone—that McDonnell Douglas’s first stage, the prima facie case, further entails a “four-step test,” one step of which requires the plaintiff to show that she was treated differently from a similarly situated “comparator.” Lewis v. City of Union City, 918 F.3d 1213, 1220–22 (11th Cir. 2019) (en banc). We’ve then treated these requirements as a series of standalone, case-dispositive elements—boxes to be checked—rather than simply asking the controlling question whether the facts give rise to a triable issue of discrimination. In so doing, we’ve mistakenly allowed the tool to eclipse (and displace) the rule. 4 Finally, and perhaps worst of all, it now strikes me that the McDonnell Douglas three-step—particularly as supplemented by the first step’s constituent four-step—obscures the actual Title VII inquiry, especially at summary judgment. I’ll readily confess that others have beaten me to this conclusion, but they make for pretty good company. For instance, while a judge on the D.C. Circuit,

4 See Sandra F. Sperino, Rethinking Discrimination Law, 110 Mich. L. Rev. 69, 71

(2011) (“[T]he key question in modern discrimination cases is often whether the plaintiff can cram his or her facts into a recognized structure and not whether the facts establish discrimination.”); see also Deborah A. Widiss, Proving Discrimination by the Text, 106 Minn. L. Rev. 353, 374–75 (2021) (“In practice, however, the causation standard employed is less important than whether a plaintiff can successfully squeeze the evidence into an arcane and complicated body of judge-made law . . . .”). 10 NEWSOM, J., Concurring 21-13245

Justice Kavanaugh described the fixation on the plaintiff’s prima facie case as “a largely unnecessary sideshow” that “has not benefited employees or employers,” has not “simplified or expedited court proceedings,” and, in fact, “has done exactly the opposite, spawning enormous confusion and wasting litigant and judicial resources.” Brady v. Office of Sergeant at Arms, 520 F.3d 490, 494 (D.C. Cir. 2008). Worse, he explained, the McDonnell Douglas framework isn’t just wasteful, it is potentially misleading in that it entices reviewing courts to focus on non-core issues: At summary judgment, the prima facie case is “almost always irrelevant” and “usually [a] misplaced” inquiry—because once the defendant offers an explanation for its decision, “whether the plaintiff really” made out a prima facie case no longer matters. Id. at 493–94 (quoting Aikens, 460 U.S. at 715). Rather, then-Judge Kavanaugh continued, once the defendant explains itself, “the district court must resolve one central question: Has the employee produced sufficient evidence for a reasonable jury to find that . . . the employer intentionally discriminated against the employee on the basis of race, color, religion, sex, or national origin?” Id. at 494. That, of course, is the Rule 56 question—shorn of all its McDonnell Douglas prophylaxis. 5

5 One clarification: While the prima-facie-case question is undoubtedly “irrelevant” as a formal matter following an employer’s summary-judgment motion—at that point, the employer having explained itself, the focus turns to the ultimate question—that’s not to say that the sort of proof that might inform a plaintiff’s prima facie showing is irrelevant as an evidentiary matter. As the majority opinion observes, “the questions the plaintiff must answer to 21-13245 NEWSOM, J., Concurring 11

To be clear, Justice Kavanaugh is hardly alone. Justice Gorsuch made similar observations during his tenure on the Tenth Circuit. Using the very same descriptor that Justice Kavanaugh had, he explained that McDonnell Douglas’s staged inquiries “sometimes prove a sideshow,” Hinds v. Sprint/United Mgmt. Co., 523 F.3d 1187, 1202 n.12 (10th Cir. 2008), that the framework itself “has proven of limited value,” Walton v. Powell, 821 F.3d 1204, 1210 (10th Cir. 2016), and that courts too often get bogged down “engag[ing] in the business of trying to police the often fine line between” when McDonnell Douglas does and doesn’t apply, id. at 1211. 6

make a prima facie case are relevant to the ultimate question of discrimination”—whether she was a member of a protected class, whether she suffered an adverse employment decision, how her colleagues were treated, etc. Maj. Op. at 11. So it may well be that a plaintiff who lacks the evidence necessary to make out a prima facie case should lose at summary judgment. Importantly, though, she shouldn’t lose because she has failed to dot her Is and cross her Ts under McDonnell Douglas, but rather because she has failed to proffer evidence that gives rise to a genuine issue of material fact concerning whether her employer engaged in unlawful discrimination. Cf. also id. at 11 (“A failure in the prima facie case often also reflects a failure of the overall evidence.”). 6 Others have voiced similar complaints. Judge Easterbrook has described Title

VII summary-judgment cases generally as implicating a “rat’s nest of surplus ‘tests.’” Ortiz v. Werner Enters., Inc., 834 F.3d 760, 766 (7th Cir. 2016). Judge Hartz has observed that the McDonnell Douglas framework, in particular, “only creates confusion and distracts courts from ‘the ultimate question of discrimination.’” Wells v. Colorado Dep’t of Transp., 325 F.3d 1205, 1221 (10th Cir. 2003) (Hartz, J., concurring). Judge Wood has lamented the “snarls and knots that the current methodologies used in discrimination cases of all kinds 12 NEWSOM, J., Concurring 21-13245

* * * So, what’s my takeaway regarding McDonnell Douglas? From a case that didn’t even arise on summary judgment has emerged a purported “procedural device” that, in day-to-day operation, disregards the duly promulgated rules of summary-judgment procedure, that overrides the substance of Title VII, and whose multi-step burden-shifting formula obscures the decisive question: Does the summary-judgment record reveal a genuine dispute of material fact about whether an employer discriminated against its employee “because of ” a protected characteristic? III So, as it turns out, there’s plenty not to like about McDonnell Douglas as a summary-judgment tool. And what of the convincing- mosaic standard, which I’ve confessed to having long dismissed as secondary corollary of sorts or, worse, a manipulable workaround? Turns out there’s a lot to like.

have inflicted on courts and litigants alike” and expressed her view that McDonnell Douglas’s successive inquiries have “lost their utility.” Coleman v. Donahoe, 667 F.3d 835, 863 (7th Cir. 2012) (Wood, J., concurring). And Judge Tymkovich, training his critique on McDonnell Douglas’s third step, has complained that the “focus on pretext has shifted the emphasis of an employment discrimination case away from the ultimate issue of whether the employer discriminated against the complaining employee.” Tymkovich, supra note 2, at 505. 21-13245 NEWSOM, J., Concurring 13

McDonnell Douglas, it now seems to me, leads us away from—or at the very least is orthogonal to—Rule 56’s north star. By contrast, the convincing-mosaic standard points, even if a little clumsily, right at it. Here’s what we said in Smith: [T]he plaintiff will always survive summary judgment if he presents circumstantial evidence that creates a triable issue concerning the employer’s discriminatory intent. A triable issue of fact exists if the record, viewed in a light most favorable to the plaintiff, presents a convincing mosaic of circumstantial evidence that would allow a jury to infer intentional discrimination by the decisionmaker.

644 F.3d at 1328 (internal quotation marks, citations, and footnote omitted). Stripped of the rhetorical flourish—the superfluous “convincing mosaic of ” preface—that is, in essence, just a restatement of Rule 56’s summary-judgment standard. No bells, no whistles—just reasonable inferences and triable facts. What accounts, then, for the convincing-mosaic standard’s failure to launch? Well, inertia for starters. By the time the convincing-mosaic option came along, at least as a stand-alone test, parties, courts, and commentators had been debating and applying McDonnell Douglas for decades. Separately, I think the convincing- mosaic framework suffers from a branding problem of sorts, of which its rhetoric is a big part. The informal moniker— “convincing mosaic”—just sounds contrived, and thus sends formalists like me into a dither. It’s also a little misleading: Satisfying the test requires neither “convincing” a reviewing court 14 NEWSOM, J., Concurring 21-13245

nor presenting enough evidence to compose a “mosaic.” Summary judgment turns on the existence of a genuine factual dispute; courts deciding summary-judgment motions don’t weigh evidence, and they don’t decide (let alone announce) whether they’re convinced. And a mosaic—in its truest sense a collection— isn’t necessary to defeat summary judgment; a single item of evidence can at least theoretically suffice. In any event, as between the two current contestants, it now strikes me that the convincing-mosaic standard—which I’d be inclined to re-brand as, perhaps, just the “Rule 56” standard, to denude it of its unnecessary ornamentation—comes much closer to capturing the essence of summary judgment than does McDonnell Douglas. IV Let me try, in closing, to anticipate and address a few likely objections. A First, does any of this really matter? I think it does. We shouldn’t perpetuate the existing regime by dint of its sheer existence. We should strive to get the cases right according to the governing law. And for present purposes, the “governing law” comprises (1) Title VII’s prohibition on employment discrimination perpetrated “because of ” an employee’s protected characteristics, 42 U.S.C. § 2000e-2(a)(1), and (2) Rule 56’s focus on the existence of a “genuine dispute” about that causation issue, see Fed. R. Civ. P. 56(a). For reasons I’ve tried to explain, McDonnell 21-13245 NEWSOM, J., Concurring 15

Douglas is at best only tangentially directed to those issues; the convincing-mosaic standard—or something like it—is much more immediately so. Moreover, I fear that our increasingly rigid application of McDonnell Douglas may actually be causing us to get cases wrong— in particular, to reject cases at summary judgment that should, under a straightforward application of Rule 56, probably proceed to trial. A plaintiff who can marshal strong circumstantial evidence of discrimination but who, for whatever reason, can’t check all of the McDonnell-Douglas-related doctrinal boxes—for instance, because she can’t quite show that her proffered comparator is sufficiently “similarly situated,” see supra at 9—may well lose at summary judgment, whereas a plaintiff who has a slightly better comparator but little other evidence of discrimination might survive. Especially in light of Rule 56’s plain language—which focuses on the existence of a “genuine dispute as to any material fact,” Fed. R. Civ. P. 56(a)— that seems a little topsy-turvy. B Second, wouldn’t a wholehearted embrace of the convincing-mosaic framework result in more cases going to trial and thereby overburden already busy district courts? Well, maybe. To the extent that McDonnell Douglas’s judge-created elements and sub-elements are currently causing courts to grant summary judgment in cases where, in Rule 56 terms, a genuine dispute exists, then yes, ditching them in favor of something that looks more like 16 NEWSOM, J., Concurring 21-13245

the convincing-mosaic standard would lead to more trials. 7 But inasmuch as that’s a problem, courts shouldn’t manufacture or jerry-rig doctrine to fix it. I’ve never thought that judges should decide cases in an effort to drive good outcomes or avoid bad ones, and now’s not the time to start. For good or ill, the facts are (1) that Title VII gives plaintiffs a right to a jury trial in appropriate circumstances, see 42 U.S.C. § 1981a(c), and (2) that Rule 56 forestalls jury trials only where there is “no genuine dispute as to any material fact”—here, as to the employer’s causal motivation. Some cases will warrant trial under Rule 56’s standard, some won’t. But neither Title VII nor the Federal Rules make an exception for claims that, while legally viable, might prove time- and labor- intensive. C Finally, isn’t the idea of scrapping McDonnell Douglas in favor of something like the convincing-mosaic standard pretty radical? Not particularly. After all, we’ve been using (or at least incanting)

7 Reasonable minds can differ about how many cases are wrongly decided

because of McDonnell Douglas. Many of our early cases doubted whether an employer’s motive is susceptible to summary judgment at all. See Chapman v. AI Transp., 229 F.3d 1012, 1025 (11th Cir. 2000) (en banc) (collecting cases). When we held that it is, we did so on the ground that “the summary judgment rule applies in job discrimination cases just as in other cases” and, thus, that “[n]o thumb is to be placed on either side of the scale.” Id. at 1026. But the questions (1) whether the summary-judgment procedure applies to Title VII cases—of course it does—and (2) how many cases it will weed out are, to my mind, different. 21-13245 NEWSOM, J., Concurring 17

the convincing-mosaic standard as an alternative to McDonnell Douglas for more than a decade now, and other courts have similarly renounced any slavish devotion to McDonnell Douglas’s rigid three-step analysis. Interestingly, we borrowed the phrase “convincing mosaic” from the Seventh Circuit. See Smith, 644 F.3d at 1328 (quoting Silverman v. Board of Educ. of Chi., 637 F.3d 729, 734 (7th Cir. 2011)). That court has since (and wisely) jettisoned the “convincing mosaic” label, but not its substance. Instead, it has adopted what it calls a “direct method”—in effect, a merger of our direct-evidence and convincing-mosaic frameworks—which permits an employee to oppose her employer’s summary-judgment motion using any evidence, whether technically direct or circumstantial, so long as it creates a triable issue of discrimination. See Sylvester v. SOS Child.’s Vills. Ill., Inc., 453 F.3d 900, 902–03 (7th Cir. 2006). The court has described its approach in the following terms, which, to me, sound pretty convincing-mosaic-ish: [The] legal standard . . . is simply whether the evidence would permit a reasonable factfinder to conclude that the plaintiff’s race, ethnicity, sex, religion, or other proscribed factor caused the discharge or other adverse employment action. Evidence must be considered as a whole, rather than asking whether any particular piece of evidence proves the case by itself—or whether just the “direct” evidence does so, or the “indirect” evidence. Evidence is evidence. Relevant evidence must be considered and irrelevant evidence disregarded, but 18 NEWSOM, J., Concurring 21-13245

no evidence should be treated differently from other evidence because it can be labeled “direct” or “indirect.”

Ortiz, 834 F.3d at 765. For its part, the D.C. Circuit has likewise taken steps to reorient McDonnell Douglas toward the ultimate question whether the plaintiff has presented a genuine factual dispute about intentional discrimination. By the time the employer files a summary-judgment motion, that court has explained, it “ordinarily will have asserted a legitimate, non-discriminatory reason for the challenged decision” at step two of McDonnell Douglas’s three-step analysis. Brady, 520 F.3d at 493. At that point, the D.C. Circuit continued, “whether the employee actually made out a prima facie case is ‘no longer relevant’ and thus ‘disappear[s]’ and ‘drops out of the picture.’” Id. (quoting Hicks, 509 U.S. at 510–11, and Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 143 (2000)). Rather, the reviewing court then “has before it all the evidence it needs to decide” the ultimate question—namely, “whether the defendant intentionally discriminated against the plaintiff.” Id. at 494 (quoting Aikens, 460 U.S. at 715). So, to avoid any “lingering uncertainty,” the D.C. Circuit concluded by emphasizing that in the mine-run summary-judgment case, where the employer has offered a non-discriminatory reason for its action, a reviewing court “should not . . . decide whether the plaintiff actually made out a prima facie case” under McDonnell Douglas but, rather, should resolve the “central question” whether the “employee [has] produced sufficient evidence for a reasonable jury to find” that “the 21-13245 NEWSOM, J., Concurring 19

employer intentionally discriminated against the employee on the basis of race, color, religion, sex, or national origin?” Id. All of which is simply to say: It’s not quite as heretical as I once assumed to question whether McDonnell Douglas is the—or even an—appropriate means of deciding Title VII cases at summary judgment. And it wouldn’t be quite as radical as it once seemed to shift the focus away from McDonnell Douglas’s judge- made formulation and toward Rule 56’s plain language.8

8 Bulky footnote alert: At this point, inside baseballers may be asking, “What about the en banc decision in Lewis, which you wrote?” See Lewis v. City of Union City, 918 F.3d 1213 (11th Cir. 2019) (en banc). Fair question. To be clear, though, I needn’t renounce Lewis. For what it set out to do—as we explained there, “to clarify the proper standard for comparator evidence in intentional-discrimination cases” brought under McDonnell Douglas’s burden- shifting regime, id. at 1220—I continue to believe that Lewis gave the right answer. It’s just that I’ve come to doubt that McDonnell Douglas—and our downstream application of it—asks the correct questions. In Lewis, we noted that a Title VII plaintiff can respond to her employer’s summary-judgment motion in “a variety of ways”—“one of which,” we said, “is by navigating the now-familiar three-part burden-shifting framework established by the Supreme Court in McDonnell Douglas,” whose first part, of course, requires the plaintiff to make out a prima facie case of discrimination. Id. at 1217. We further noted the Supreme Court’s repeated directive that one of the ways—seemingly, the presumptive way—that the plaintiff can demonstrate a prima facie case is by satisfying a constituent four- step test, one prong of which requires her to show “that she was treated differently from another ‘similarly situated’ individual—in court-speak, a ‘comparator.’” Id. (quoting Burdine, 450 U.S. at 258–59). Faced with an entrenched intra-circuit split, we granted en banc rehearing to answer a discrete question about the proper implementation of that McDonnell-Douglas- related “comparator” element: “What standard does the phrase ‘similarly 20 NEWSOM, J., Concurring 21-13245

V “Wisdom too often never comes, and so one ought not to reject it merely because it comes late.” Henslee v. Union Planters Nat’l Bank & Tr. Co., 335 U.S. 595, 600 (1949) (Frankfurter, J., dissenting). For a while now, I’ve uncritically accepted the McDonnell Douglas framework as the proper means of resolving Title VII cases on summary judgment, and I’ve long scorned the convincing-mosaic standard as a judge-made bypass. I repent. I had it backwards. Whereas McDonnell Douglas masks and muddles the critical Rule 56 inquiry, “convincing mosaic,” for all intents and purposes, is the critical Rule 56 inquiry. On a going-forward basis,

situated’ impose on the plaintiff: (1) ‘same or similar,’ (2) ‘nearly identical,’ or (3) some other standard?” Id. at 1218. Our response: A Title VII plaintiff must show that her proposed comparators are “similarly situated in all material respects.” Id. at 1224–29. I stand by Lewis’s answer to that operational question—one of the many such questions that lower courts, including ours, have taken to asking in the wake of McDonnell Douglas. I will confess, though, that the question that we confronted and answered in Lewis now strikes me as awfully weedsy— indicative, I worry, of an analysis that (to continue the botanical metaphor) risks missing the forest for the trees. Rather than getting tangled up in prima facie cases, four-step tests, similarly situated comparators, and the like, I’ve come to believe that we’d be better off cutting straight to the Rule 56 chase: Has the plaintiff presented evidence that gives rise to a genuine factual dispute about whether her employer engaged in intentional discrimination? To my surprise, the convincing-mosaic standard—shorn of its frills—does pretty much exactly that. (Interestingly, and perhaps tellingly, on remand from our en banc decision, Lewis won—i.e., survived summary judgment—on convincing-mosaic grounds. See Lewis, 934 F.3d at1186–90 (on remand)). 21-13245 NEWSOM, J., Concurring 21

therefore, I would promote the convincing-mosaic standard to primary status and, to the extent consistent with Supreme Court precedent, relegate McDonnell Douglas to the sidelines.