Johnson v. Baltimore Police Department: Plausible Title VII Comparator Pleading Does Not Require a One-to-One Misconduct Match at Rule 12(b)(6)
1. Introduction
In Wanda Johnson v. Baltimore City, Maryland; Baltimore Police Department, an African American Baltimore police officer (Wanda Johnson) alleged race discrimination and retaliation under Title VII and asserted a municipal-liability claim under Monell v. Department of Social Services of City of New York, 436 U.S. 658 (1978). The case arose from disciplinary proceedings connected to a 2018 nightclub incident and a subsequent Internal Affairs investigation culminating in Johnson’s forced resignation in June 2022.
The appeal presented three pleading-stage questions: (1) whether Johnson plausibly alleged disparate discipline by identifying sufficiently similar non-black comparators; (2) whether she plausibly alleged retaliation causation where the discipline timeline began before her protected activity; and (3) whether she plausibly alleged a “custom or practice” under Monell based on more than her own experience.
2. Summary of the Opinion
Race discrimination (Title VII) — Reversed
The Fourth Circuit held Johnson plausibly alleged disparate treatment because she identified multiple white or non-black officers who allegedly engaged in similar categories of misconduct (assault/use of force and/or false statements) but were not terminated or forced to resign. The district court erred by effectively requiring an identical, “perfect” comparator match at the motion-to-dismiss stage.
Retaliation (Title VII) & Monell — Affirmed
Retaliation failed because the pleaded chain leading to resignation began well before protected activity, and the complaint did not plausibly allege decisionmaker knowledge for the one-day timing theory. The Monell claim failed because Johnson alleged only her own experience plus conclusory statements rather than specific facts showing a persistent, widespread pattern.
3. Analysis
3.1 Precedents Cited
A. Pleading standard at Rule 12(b)(6)
- Ashcroft v. Iqbal, 556 U.S. 662 (2009) and Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007): The court framed all three claims through plausibility pleading—enough factual content to allow a reasonable inference of liability, not “threadbare recitals.”
- Barnett v. Inova Health Care Servs., 125 F.4th 465 (4th Cir. 2025) and Barbour v. Garland, 105 F.4th 579 (4th Cir. 2024): Used for de novo 12(b)(6) review and the requirement to accept well-pleaded facts while rejecting conclusory allegations.
- Wikimedia Found. v. Nat'l Sec. Agency, 857 F.3d 193 (4th Cir. 2017) and Evans v. United States, 105 F.4th 606 (4th Cir. 2024): Cited to reject “naked assertions” (critical to the retaliation knowledge allegation).
B. Title VII discrimination and comparator pleading
- Coleman v. Maryland Court of Appeals, 626 F.3d 187 (4th Cir. 2010) and Bing v. Brivo Sys., LLC, 959 F.3d 605 (4th Cir. 2020): Confirmed that a complaint need not plead a full prima facie case but must plausibly allege a Title VII violation “above a speculative level.”
- McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973); Haynes v. Waste Connections, Inc., 922 F.3d 219 (4th Cir. 2019); Wannamaker-Amos v. Purem Novi Inc., 126 F.4th 244 (4th Cir. 2025): Provided the framework for inference-based discrimination claims and the comparator method of proving discriminatory discipline.
- White v. BFI Waste Servs., LLC, 375 F.3d 288 (4th Cir. 2004): Quoted for the proposition that the inference element can be met by alleging similarly situated employees outside the protected class received more favorable treatment.
- Seabrook v. Driscoll, 148 F.4th 264 (4th Cir. 2025) and Lightner v. City of Wilmington, N.C., 545 F.3d 260 (4th Cir. 2008): Emphasized that comparator misconduct must be comparable in seriousness and similarity must be “meaningful.”
- Spencer v. Virginia State University, 919 F.3d 199 (4th Cir. 2019) and Cowgill v. First Data Techs., Inc., 41 F.4th 370 (4th Cir. 2022): Identified factors relevant to comparator similarity (job, standards, supervisor, qualifications) and clarified “similar in all relevant respects.”
- Cook v. CSX Transp. Corp., 988 F.2d 507 (4th Cir. 1993): Via Haynes, supplied the key comparator principle that comparisons rarely involve “precisely the same set” of offenses and circumstances—central to the majority’s reversal.
- Cosby v. S.C. Prob., Parole & Pardon Servs., 93 F.4th 707 (4th Cir. 2024): Used to rebut the dissent’s “proven vs rumored” framing; the majority distinguished the nature of the conduct at issue in Cosby.
- Smith v. Univ. of North Carolina, 632 F.2d 316 (4th Cir. 1980) and Balderson v. Lincare Inc., 62 F.4th 156 (4th Cir. 2023): Reinforced that courts do not police whether discipline is wise or fair, only whether it is nondiscriminatory.
C. Title VII retaliation causation
- Foster v. Univ. of Md.-Eastern Shore, 787 F.3d 243 (4th Cir. 2015): Set out the prima facie elements (protected activity, adverse action, causal link) and the McDonnell Douglas pathway.
- Barnhill v. Bondi, 138 F.4th 123 (4th Cir. 2024): Critical for two propositions: (1) temporal proximity typically must be within about two months; and (2) the decisionmaker must be actually aware of the protected activity before acting.
- Alberti v. Rector and Visitors of the University of Virginia, 65 F.4th 151 (4th Cir. 2023) and Lettieri v. Equant Inc., 478 F.3d 640 (4th Cir. 2007): Provided the alternative route to causation—evidence of continuing retaliatory conduct/animus bridging temporal gaps.
- Roberts v. Glenn Indus. Grp., Inc., 998 F.3d 111 (4th Cir. 2021): Supported the conclusion that multi-month gaps generally require additional evidence beyond timing.
- Holloway v. Maryland, 32 F.4th 293 (4th Cir. 2022): Illustrated how intervening events (e.g., supervisor comment acknowledging the EEOC complaint) can bridge a longer gap; the majority held Johnson lacked analogous post-activity intervening facts.
D. Municipal liability under Monell
- Monell v. Department of Social Services of City of New York, 436 U.S. 658 (1978): The source of municipal liability doctrine (policy/custom causation).
- Howard v. City of Durham, 68 F.4th 934 (4th Cir. 2023); Starbuck v. Williamsburg James City Cnty. Sch. Bd., 28 F.4th 529 (4th Cir. 2022); Lytle v. Doyle, 326 F.3d 463 (4th Cir. 2003): Organized the recognized pathways to a policy or custom (including widespread practice/condonation).
- Owens v. Baltimore City State's Attorney's Office, 767 F.3d 379 (4th Cir. 2014): Provided the benchmark for adequate “custom” pleading—allegations referencing “reported and unreported cases” and “numerous successful motions” constituted factual content beyond bare conclusions.
- Semple v. City of Moundsville, 195 F.3d 708 (4th Cir. 1999) and Misjuns v. City of Lynchburg, 139 F.4th 378 (4th Cir. 2025): Reinforced that a single incident (or only the plaintiff’s experience) is insufficient to establish a widespread custom.
- Filarsky v. Delia, 566 U.S. 377 (2012): Cited for the general scope of § 1983 liability principles.
3.2 Legal Reasoning
A. Discrimination: comparators at the pleading stage
The central doctrinal move is the majority’s insistence that the district court demanded too much comparator congruence too early. While comparator similarity must be “meaningful” (Lightner v. City of Wilmington, N.C.) and offenses comparable in seriousness (Seabrook v. Driscoll), the majority held that a Rule 12(b)(6) complaint need not identify a comparator who matches the plaintiff’s alleged misconduct “perfect[ly]” across every charged dimension.
In practical terms, the court treated Johnson’s list of thirteen officers as adequate factual content because it described (i) protected-class status (white or non-black), (ii) similar categories of misconduct (false statements and/or assault/use-of-force-type conduct), and (iii) more favorable outcomes (no termination/forced resignation; some promotions). Anchored by Haynes v. Waste Connections, Inc. and its quotation of Cook v. CSX Transp. Corp., the court held that imperfect overlap does not end the analysis when multiple examples—taken together—support an inference of disparate discipline.
The dissent’s comparator theory—“similarly-situated in all respects” (quoting Spencer v. Virginia State University)—would have treated the trial board’s sustained charges as a categorical disqualifier against comparators who were merely charged, suspected, or accused. The majority rejected that approach as too rigid for pleading and reiterated that Fourth Circuit law asks whether comparators are similar “in all relevant respects” (Cowgill v. First Data Techs., Inc.), not identical.
B. Retaliation: causation fails where the discipline path predates protected activity and knowledge is conclusory
On retaliation, the opinion turns on two recurring Fourth Circuit constraints:
- Decisionmaker knowledge is required for timing inferences. Even a one-day temporal proximity (May 31 to June 1) cannot support causation unless the relevant actor was “actually aware” of the protected activity (Barnhill v. Bondi). Johnson’s allegation that BPD “knew” or “should have known” was deemed conclusory without supporting facts.
- Where a disciplinary chain begins before protected activity, later adverse action is less plausibly retaliatory. The court emphasized that the charges leading to the trial board originated in June 2020, while the EEOC charge was filed in February 2021. This chronology made BPD’s later pursuit of discipline consistent with preexisting processes rather than retaliation “because of” protected activity.
Although Lettieri v. Equant Inc. and Alberti v. Rector and Visitors of the University of Virginia allow intervening facts to bridge gaps, the alleged warnings about Internal Affairs “coming for you” predated the protected activity and thus could not function as post-activity retaliatory “intervening events” of the Holloway v. Maryland type.
C. Monell: “custom by condonation” requires specific, external factual content
Johnson framed her municipal-liability theory as a persistent “institutional practice or custom” of covering up misconduct and retaliating against officers who complain. The majority treated this as an attempt to plead a “widespread practice” custom under Lytle v. Doyle and Owens v. Baltimore City State's Attorney's Office.
But the complaint’s supporting facts were essentially limited to Johnson’s own experience plus generalized assertions (“well-known throughout the Department”). Under Semple v. City of Moundsville, Howard v. City of Durham, and Misjuns v. City of Lynchburg, that is not enough; Monell custom requires “numerous particular instances” beyond the plaintiff’s single episode. By contrasting Johnson’s allegations with the more concrete pattern allegations in Owens (e.g., “reported and unreported cases” and “numerous successful motions”), the court held Johnson did not cross the plausibility threshold.
3.3 Impact
- Comparator pleading in disciplinary cases becomes more plaintiff-viable at Rule 12(b)(6). The decision signals that where a complaint supplies multiple comparator examples with job parity (same force/standards) and similar misconduct categories (e.g., false statements and/or assault/use of force), a district court should be cautious about demanding near-identical misconduct matrices at the pleading stage.
- Retaliation claims remain tightly constrained by chronology and knowledge pleading. The opinion reinforces that plaintiffs must plead concrete facts showing awareness by the relevant decisionmaker(s), and that retaliation is difficult to plausibly infer when discipline is a continuation of a pre-protected-activity process.
- Monell custom claims continue to require “pattern” facts, not workplace folklore. The court reaffirmed that generalized “everyone knows” allegations—even if potentially true—must be supported by identifiable external indicia (other incidents, lawsuits, adjudications, statistics, reports, or similar concrete examples) to survive dismissal.
- The dissent highlights an emerging fault line. Judge Wilkinson’s dissent frames police-discipline integrity as a context favoring stricter comparator similarity (including “guilt” outcomes). Future panels may confront how far “relevant respects” extends when disciplinary proceedings produce formal findings.
4. Complex Concepts Simplified
- Rule 12(b)(6) / “plausibility”: At dismissal stage, courts assume well-pleaded facts are true and ask whether those facts make liability a reasonable inference, not merely speculation (Iqbal; Twombly).
- McDonnell Douglas framework: A common method to prove discrimination using circumstantial evidence—often by showing similarly situated employees outside the protected class were treated better for comparable misconduct.
- “Comparator” evidence: Using other employees as reference points. The legal question is not whether their situations are identical, but whether they are similar in the ways that matter to the challenged decision (“in all relevant respects”).
- Retaliation “causation”: The plaintiff must plausibly connect protected activity (like an EEOC charge) to an adverse action (like termination). Close timing can help, but only if the decisionmaker knew about the protected activity; and timing is weaker if discipline began earlier for independent reasons.
- Monell “custom by condonation”: A city/department can be liable under § 1983 only if the constitutional violation was caused by an official policy or a widespread, persistent practice that leaders knew about and failed to correct. One person’s story, without more, is usually insufficient.
5. Conclusion
The Fourth Circuit’s published decision draws a clear pleading-stage line: a Title VII disparate-discipline claim may proceed when the complaint alleges multiple non-black comparators with similar categories of misconduct and more favorable outcomes, without requiring a one-to-one match of every charged act and procedural posture. At the same time, the court tightened familiar constraints on retaliation (knowledge and chronology) and reaffirmed demanding—but settled—requirements for Monell custom pleading. The remand on Count I ensures fuller factual development on the discrimination claim, while Counts III and IV illustrate that conclusory allegations and single-incident narratives will not clear the plausibility bar.
Comments