Eleventh Circuit: Non-Comparator “Same-Misconduct” Evidence and Shifting Explanations Can Create a Convincing Mosaic (and Pretext) Defeating Summary Judgment

I. Introduction

This appeal concerns race discrimination and retaliation claims brought by DeMarkus Hall and Eddie Hughes (both Black) against their employer, Coal Bed Services, Inc., and (for § 1981 claims) its parent, Pate Holdings, Inc. Hall and Hughes alleged that after they complained to their supervisor, Willie Williams, that another supervisor, James Toxey, was discriminating against them because of race, the company terminated them and then refused to rehire them.

The precipitating termination event was a jobsite drug test: Hall, Hughes, and a White coworker, Brandon Ramsey, all refused to take the test and left the jobsite, understanding they were terminated. The company soon rehired Ramsey—without requiring an immediate drug test—paid him for a full shift that day, but paid Hall and Hughes only for four hours, and then replaced Hall and Hughes with White hires (including a worker previously fired for failing a drug test).

The key appellate issues were whether, viewing the evidence in plaintiffs’ favor, (1) the record contained sufficient circumstantial evidence—under the “convincing mosaic” approach—to allow a jury to infer discriminatory intent despite contested comparator status under McDonnell Douglas; and (2) whether the retaliation claim failed because alleged intervening misconduct (the drug-test refusal and a purported report of marijuana use) “broke” causation and defeated pretext at summary judgment.

II. Summary of the Opinion

The Eleventh Circuit (Carnes, J.) reversed. It held that the record, taken in the light most favorable to Hall and Hughes, presented a genuine dispute of material fact on both discrimination and retaliation under Title VII and § 1981.

• Discrimination: The court found a “convincing mosaic” supporting an inference of intentional race discrimination, emphasizing disparate treatment in rehiring and pay, replacement by White workers, and evidence of shifting explanations surrounding whether the drug test was “random.”
• Retaliation: The court held the plaintiffs’ complaint about Toxey was “objectively reasonable,” and that temporal proximity (about four weeks) plus evidence undermining the employer’s asserted justification could allow a jury to find retaliation. The employer’s theory that intervening misconduct “broke the causal chain” did not warrant summary judgment on this record.
• § 1981: Applying Comcast Corp. v. Nat'l Assoc. of Afr.-Am. Owned Media but-for causation, the court concluded the same disputed facts sufficed to defeat summary judgment on § 1981 discrimination and retaliation.

III. Analysis

A. Precedents Cited (and How They Shaped the Decision)

• Cottrell v. Caldwell — Used to frame summary judgment methodology: “facts” at summary judgment are those viewed in the nonmovant’s favor. This principle underwrote the court’s treatment of disputes about the alleged marijuana report and the Pate/Hall rehiring conversation.
• Sierra Club v. Tenn. Valley Auth. — Cited for de novo review of summary judgment, reinforcing that the appellate court independently assesses whether genuine issues exist.
• McCreight v. AuburnBank — Central to the discrimination analysis. The court relied on McCreight to describe McDonnell Douglas and the “convincing mosaic” as “two paths to the same destination,” and to explain that the mosaic approach remains available even where strict comparator evidence is lacking.
• Poer v. Jefferson Cnty. Comm'n — Provided illustrative categories of evidence that can form a convincing mosaic: suspicious timing/ambiguous statements, systematically better treatment of similarly situated employees, and pretext.
• Phillips v. Legacy Cabinets — Cited repeatedly for (i) viewing evidence in the nonmovant’s favor and (ii) the relationship between Title VII single-motive and § 1981 but-for causation standards; also cited for the non-binding nature of unpublished opinions.
• Tynes v. Fla. Dep't of Juv. Just. — Used to clarify that a “convincing mosaic” is simply sufficient circumstantial evidence for a reasonable factfinder to infer intentional discrimination; importantly, the court rejected the employer’s suggestion that “actual” evidence requires direct evidence.
• Comcast Corp. v. Nat'l Assoc. of Afr.-Am. Owned Media — Anchored the § 1981 requirement that race discrimination be a but-for cause of the contractual injury; the court treated the record as sufficient under that standard, aligning it with the Title VII single-motive framework referenced in Phillips.
• Jenkins v. Nell — Cited as a comparative reference supporting denial of summary judgment where material disputes and circumstantial evidence can establish discriminatory causation.
• Yelling v. St. Vincent's Health Sys. — Provided the retaliation framework: the court “primarily” uses McDonnell Douglas for circumstantial retaliation evidence and articulated the prima facie elements and pretext step.
• Berry v. Crestwood Healthcare LP — Used to confirm that retaliation can be proven without McDonnell Douglas and can rest on any circumstantial evidence permitting a reasonable inference of retaliatory intent.
• Clover v. Total Sys. Servs., Inc. and Furcron v. Mail Ctrs. Plus, LLC — These cases governed the “objectively reasonable” belief requirement for protected opposition. The court applied them to hold the plaintiffs’ report of perceived discrimination by Toxey was close enough to be objectively reasonable even if unlawfulness was disputed.
• Farley v. Nationwide Mut. Ins. Co. and Thomas v. Cooper Lighting, Inc. — Cited for temporal-proximity causation principles. Farley (seven weeks) supported the conclusion that four weeks can establish a prima facie “causal nexus” where the decisionmaker knew of the complaint.
• Jefferson v. Sewon Am., Inc. — Reinforced that, when circumstances could support either inference, the ultimate retaliation question is for the jury—supporting reversal of summary judgment.
• Gogel v. Kia Motors Mfg. of Ga., Inc. — Confirmed that § 1981 retaliation is cognizable and analyzed under the same framework as Title VII retaliation, allowing the court to carry its Title VII analysis over to § 1981.

B. Legal Reasoning

1. Discrimination: The “Convincing Mosaic” and the Role of Non-Comparator Evidence

The court’s discrimination holding turned on the idea that the absence (or assumed absence) of a valid McDonnell Douglas comparator does not end the inquiry. Even assuming Ramsey was not “similarly situated in all material respects” due to experience, tenure, and job title, the court treated Ramsey’s “same day, same incident, same refusal” differential treatment as highly probative in the mosaic analysis.

The court identified a multi-part mosaic that, collectively, could allow a jury to infer discriminatory intent:

• Ramsey (White) was rehired essentially immediately after refusing the same drug test; Hall and Hughes (Black) were “stonewalled” despite seeking rehiring on the same conditions.
• Pay disparity for the same partial day: Ramsey was paid a full shift; Hall and Hughes were paid four hours.
• Replacement by White workers, including rehiring a White worker previously fired for failing a drug test.
• A potentially shifting rationale: the company told the EEOC the drug test was “random,” later asserting in deposition it was prompted by a report that Hall and Hughes were seen smoking marijuana.
• Background evidence of disparate work assignments and development opportunities; and evidence of Toxey’s demeaning and racialized language and manner (“boys,” “y’all,” “them,” “more black friends”).

The court rejected the employer’s suggestion that plaintiffs needed “actual evidence of a racial motive” understood as direct evidence, reiterating that circumstantial evidence is “actual” evidence and can suffice to reach a jury.

2. Retaliation: Objective Reasonableness, Temporal Proximity, and Pretext Despite Alleged Intervening Misconduct

On protected activity, the court held the complaint to Williams about perceived race discrimination by Toxey was objectively reasonable under Clover v. Total Sys. Servs., Inc. and Furcron v. Mail Ctrs. Plus, LLC. The court focused on whether the conduct was “close enough” to support a reasonable belief of unlawfulness, not whether it ultimately was unlawful.

On causation, the court relied on temporal proximity and knowledge: about four weeks elapsed between complaint and termination/non-rehire, which the court treated as within the range recognized in Farley v. Nationwide Mut. Ins. Co. and consistent with Thomas v. Cooper Lighting, Inc.

The company’s main causation defense—intervening misconduct (drug-test refusal and alleged marijuana use)—did not justify summary judgment because the underlying facts and inferences were disputed:

• Hall and Hughes denied drug use; the “report” was undocumented.
• The company had characterized the test as “random” to the EEOC, creating an inference of shifting explanations.
• Ramsey’s refusal was not treated as a bar to rehire, undermining the claim that refusal necessarily “broke the chain.”

Taking these together, the court held a jury could find the stated reason was pretextual and could find retaliation—making summary judgment improper.

3. § 1981: But-For Causation at the Summary Judgment Stage

Applying Comcast Corp. v. Nat'l Assoc. of Afr.-Am. Owned Media, the court recognized that § 1981 requires but-for causation. It nonetheless concluded that the same evidentiary record—rehire and pay disparities, replacement evidence, and shifting explanations—created a genuine dispute that race (and retaliation for complaining about race discrimination) could be found to be a but-for cause of the adverse contractual outcomes.

C. Impact

• Comparator limits do not cap discrimination proof: The decision strengthens the practical force of the “convincing mosaic” approach in the Eleventh Circuit by emphasizing that evidence of sharply different treatment of a White coworker involved in the same incident can be powerful even if that coworker would not qualify as a McDonnell Douglas comparator.
• Shifting explanations remain potent pretext evidence: A divergence between an employer’s EEOC position statement (drug test “random”) and later litigation testimony (drug test prompted by reported marijuana use) can support competing inferences that must be resolved by a jury.
• Retaliation causation is not automatically defeated by asserted intervening misconduct: Where the misconduct rationale is disputed, inconsistently described, or unevenly enforced (as with Ramsey), a jury may still infer retaliatory motive notwithstanding an intervening event.
• Rehire decisions and partial-day pay can be adverse actions and probative facts: The opinion treats the refusal to rehire and differential pay as meaningful components of discrimination/retaliation proof, not peripheral employment details.

IV. Complex Concepts Simplified

• Summary judgment (Fed. R. Civ. P. 56(a)): The judge ends the case before trial only if no genuine dispute of material fact exists and the moving party is entitled to judgment as a matter of law. All reasonable inferences go to the nonmoving party.
• McDonnell Douglas framework: A common method for proving discrimination/retaliation with circumstantial evidence using burden-shifting: prima facie case → employer’s legitimate reason → plaintiff shows pretext.
• Comparator: Under McDonnell Douglas, a plaintiff often compares treatment to a similarly situated employee. This opinion stresses that comparator defects do not eliminate other circumstantial evidence under a “convincing mosaic.”
• Convincing mosaic: A holistic look at multiple pieces of circumstantial evidence that, together, could let a reasonable jury infer discriminatory intent.
• Pretext: Evidence that the employer’s stated reason is not the true reason—often shown through inconsistencies, shifting explanations, or uneven enforcement.
• Protected activity / objective reasonableness: Complaining about discrimination is protected if the employee reasonably (objectively) believes the conduct opposed is unlawful or close enough to be unlawful.
• Temporal proximity: A short time between a complaint and an adverse action can support an inference of causation in retaliation claims, especially if decisionmakers knew of the complaint.
• But-for causation (§ 1981): The plaintiff must be able to prove the harm would not have happened “but for” race (or retaliation). At summary judgment, the question is whether a reasonable jury could so find on the evidence.

V. Conclusion

The Eleventh Circuit’s reversal underscores that employment cases often turn on how a jury interprets circumstantial evidence—especially where an employer’s rationale appears inconsistent and where a White coworker involved in the same incident receives markedly better treatment in rehiring and pay. The opinion reinforces that plaintiffs are not required to produce direct evidence of racial motive, and that even when strict comparator requirements are not met under McDonnell Douglas, a “convincing mosaic” can still carry the case to trial. On retaliation, the court confirmed that objectively reasonable complaints are protected and that alleged intervening misconduct does not automatically sever causation when the surrounding facts suggest selective enforcement or pretext.