90 F.4th 425

5th Cir.

2024

Brian McNeal, Plaintiff—Appellee, versus James LeBlanc, Defendant—Appellant.

No. 22-30180

United States Court of Appeals for the Fifth Circuit

January 5, 2024

Before JONES, STEWART, and DUNCAN, Circuit Judges.

Appeal from the United States District Court for the Middle District of Louisiana, USDC No. 3:18-CV-736

PER CURIAM:

We face another suit against a Louisiana official for overdetention, “now a euphemism for prisoners illegally incarcerated beyond the terms of their sentence.” Hicks v. LeBlanc, 81 F.4th 497, 500 (5th Cir. 2023). Pursuant to 42 U.S.C. § 1983 and state law, Brian McNeal sued Louisiana Department of Public Safety and Corrections (“DPSC”) Secretary James LeBlanc in his individual capacity for wrongfully detaining McNeal 41 days after his sentence expired. The district court denied LeBlanc’s motion to dismiss premised on qualified immunity. Following our precedent, we AFFIRM.

I.

A.

McNeal pled guilty of possessing cocaine and drug paraphernalia in 2015.¹ The Orleans Parish Criminal District Court sentenced him to a five-year suspended sentence, with five years of probation. In 2017, McNeal was arrested for violating probation. His probation was revoked, and on August 3, 2017, he was sentenced to serve 90 days at the Steve Hoyle Program (“Hoyle”)—an in-patient substance abuse program at the Bossier Parish Correctional Center. A few days later, the DPSC generated a release letter that set McNeal’s release date as November 1, 2017. A DPSC employee, anticipating his eventual transfer, sent the release letter to Hoyle.

Rather than transfer him directly to Hoyle, however, DPSC directed the Orleans Sheriff to transfer McNeal to the Elayn Hunt Correctional Center (“Hunt”), where McNeal would be classified before enrolling in Hoyle. While at Hunt, authorities determined McNeal was unfit for Hoyle due to a mental impairment. So, McNeal remained at Hunt and was never transferred to Hoyle. But his release letter remained at Hoyle, and no DPSC official took steps to ensure the letter made its way to Hunt.

McNeal was not released on November 1, 2017. Sometime thereafter, his girlfriend called the New Orleans Probation and Parole Office, asking why he had not been released. On November 15, 2017, McNeal wrote the Hunt warden, complaining he had not yet been taken to court and released. The warden’s office responded: “If your presence was required in court, the proper documents would have been sent for you to be transported.”

On December 6, 2017, after his girlfriend’s further inquiry, McNeal’s probation officer and lawyer investigated the situation. Realizing McNeal’s release date had passed, his probation officer notified DPSC. On December 11, 2017, a DPSC employee emailed McNeal’s release letter to Hunt, stating, “This is your authority to release the offender on 11/01/2017, as having completed said sentence that was imposed at the time of revocation.” The email explained that DPSC thought McNeal was at a different facility. McNeal was released on December 12, 2017, 41 days after his proper release date.

B.

In 2018, McNeal sued LeBlanc and other Louisiana officials in state court. The defendants removed the case to federal court. In 2020, the federal district court granted McNeal’s motion for partial summary judgment but denied the defendants’ summary judgment motion. In that order, the court ruled that McNeal’s overdetention claims were not barred by Heck v. Humphrey, 512 U.S. 477 (1994).

A month later, the district court granted McNeal’s motion to amend his complaint. McNeal’s first amended complaint, the district court found, “failed to state a viable claim of supervisor liability against LeBlanc.” But the court again granted McNeal leave to amend, and McNeal filed a second amended complaint (the operative complaint) against LeBlanc and others. In this complaint, McNeal sued LeBlanc for false imprisonment, negligence, violation of his Fourteenth Amendment due process rights, violations of the Louisiana Constitution, violations of Monell v. Department of Social Services, 436 U.S. 658 (1978), and failure to train or supervise. McNeal sought declaratory relief, compensatory and punitive damages, and attorney fees.

McNeal alleges that, in 2012, DPSC performed an internal review called the “Lean Six Sigma,” examining how long it took to calculate prisoner release dates. The review was “champion[ed]” by LeBlanc, who had been DPSC Secretary since 2008. Lean Six Sigma “found a widespread pattern of people being held past their legal release date,” with 83% of DPSC prisoners being overdetained. The review determined that, on average, inmates were held 71.69 days past their release dates. After learning of the issue, LeBlanc set the goal to detain “450 persons per year, for an average of 31 days per person.” LeBlanc’s changes reduced the number of overdetained persons from “2,252 per year to 1,612, and the average number of overdue days was reduced from 71.7 to 60.52 days.” Despite these efforts, LeBlanc conceded that “the ‘functional processes’ around the transmission of documents” at the DPSC “remain as antiquated as they were in 1996.”

Even after the Lean Six Sigma review, overdetention issues persisted. For example, four DPSC employees testified in a 2015 state case, Chowns v. LeBlanc, La. 37th JDC 26-932, that DPSC employees were well aware of prisoner overdetention. In 2017, a report by the Louisiana Legislative Auditor documented DPSC’s problems calculating and processing prisoners’ release dates. DPSC itself conducted an internal review, which “confirmed that the pattern of overdetention it learned about in 2012” from Lean Six Sigma “was ongoing.” In a grant application to the U.S. Department of Justice, DPSC disclosed that in 2017 it “had an average of 200 cases per month considered an ‘immediate release’ due to [processing] deficiencies,” and the prisoners in these cases “were held an ‘average of 49 days past the end of their sentences.’” In 2018, then-Louisiana Attorney General Jeff Landry wrote an op-ed with U.S. Senator John Kennedy, stating there was “a layer of incompetence so deep that the Corrections Department doesn’t know where a prisoner is on any given day of the week or when he should actually be released from prison.”

McNeal further alleges that LeBlanc knew overdetention issues still plagued the DPSC as of November 2017. Specifically, after Lean Six Sigma, LeBlanc learned that thousands of people in the custody of DPSC “were being held past their release date.” LeBlanc also admitted that, even after the changes he instituted, the DPSC “still had ‘people being held an average of about two months past their release date.’” Yet, LeBlanc never fired, demoted, penalized, or reprimanded anyone for holding inmates past their release dates. Between 2012 and 2017, multiple officials reached out to LeBlanc about overdetained prisoners. LeBlanc was also personally involved in “the back-and forth with the auditor,” which eventually led to the 2017 Louisiana Legislative Audit.

LeBlanc moved to dismiss McNeal’s second amended complaint for failure to state a claim, arguing he enjoyed qualified immunity. The district court denied LeBlanc’s motion to dismiss. LeBlanc now appeals, arguing that (1) Heck bars McNeal’s claims and (2) the district court erred by denying him qualified immunity.

II.

We have jurisdiction to review by interlocutory appeal the denial of a motion to dismiss based on qualified immunity. Ramirez v. Escajeda, 921 F.3d 497, 500 (5th Cir. 2019). We review such denials de novo, “accepting all well-pleaded facts as true and viewing them in the light most favorable to the plaintiff.” Brown v. Miller, 519 F.3d 231, 236 (5th Cir. 2008). On interlocutory appeal, “our review is restricted to determinations of questions of law and legal issues.” Ramirez, 921 F.3d at 500 (citation and internal quotation marks omitted). In this posture, we review only “whether the facts pleaded establish a violation of clearly-established law.” Id. at 501 (quoting Ashcroft, 556 U.S. at 673) (cleaned up).

III.

A.

We first address whether McNeal’s claims are barred under Heck. LeBlanc frames McNeal’s challenge as to both the validity and the duration of his sentence. McNeal counters that Heck’s bar does not apply because he merely challenges his overdetention, not the underlying conviction or sentence. Following our recent caselaw, we are bound to agree with McNeal.

In Hicks, we held that Heck does not bar claims by an overdetained prisoner who “does not challenge the validity of his sentence, [but] merely the execution of his release.” Hicks, 81 F.4th at 506; see also Crittindon v. LeBlanc, 37 F.4th 177, 190 (5th Cir. 2022) (“The Heck defense ‘is not . . . implicated by a prisoner’s challenge that threatens no consequence for his conviction or the duration of his sentence.’” (quoting Muhammad v. Close, 540 U.S. 749, 751 (2004))), cert. denied, 144 S. Ct. 90 (2023). Thus, “Heck is no bar” where success on a § 1983 claim is based on the period a prisoner “was held beyond his original sentence [because] it would not invalidate the conviction or its attendant sentence.” Hicks, 81 F.4th at 506. Such is the case here, where McNeal does not challenge his conviction or attendant sentence, but rather the 41 days he was imprisoned beyond his release date. Accordingly, Heck raises no bar to McNeal’s claims against LeBlanc based on his alleged overdetention.

B.

LeBlanc also raises a qualified immunity argument similar to those we have rejected in prior DPSC overdetention cases. LeBlanc argues he enjoys qualified immunity because McNeal fails to allege a pattern of similar overdetentions at DPSC. We have already addressed this argument based on almost identical allegations made in Parker v. LeBlanc, 73 F.4th 400 (5th Cir. 2023). We held there, as we are bound to hold here, that the overdetained prisoner alleged a pattern of similar violations at the DPSC sufficient to deny LeBlanc qualified immunity at the motion to dismiss stage. See id. at 406.

Qualified immunity protects public officials from liability if “their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.” Pearson v. Callahan, 555 U.S. 223, 231 (2009) (citation omitted). To overcome qualified immunity, the plaintiff must allege facts showing (1) “a violation of a constitutional right,” and (2) that “the right at issue was ‘clearly established’ at the time of defendant’s alleged misconduct.” Id. at 232. A right is clearly established if “it would be clear to a reasonable official that his conduct was unlawful in the situation he confronted.” Porter v. Epps, 659 F.3d 440, 445 (5th Cir. 2011) (alteration adopted) (citation omitted).

1.

At prong one, we ask whether McNeal has alleged facts showing a Fourteenth Amendment violation. The Fourteenth Amendment guarantees that no state may “deprive any person of life, liberty, or property, without due process of law.” U.S. Const. amend. XIV, § 1. Our precedent establishes, accordingly, “that a jailer has a duty to ensure that inmates are timely released from prison.” Porter, 659 F.3d at 445. Relevant to LeBlanc’s liability, we recently “held that ‘it is without question that holding without legal notice a prisoner for a month beyond the expiration of his sentence constitutes a denial of due process.’” Parker, 73 F.4th at 404 (quoting Crittindon, 37 F.4th at 188); see also Douthit v. Jones, 619 F.2d 527, 532 (5th Cir. 1980) (“Detention of a prisoner thirty days beyond the expiration of his sentence in the absence of a facially valid court order or warrant constitutes a deprivation of due process.”). McNeal alleges that LeBlanc, as DPSC Secretary, violated his Fourteenth Amendment rights.

It is black-letter law, however, that § 1983 does not create respondeat superior liability. Jason v. Tanner, 938 F.3d 191, 197 (5th Cir. 2019) (citing Monell, 436 U.S. at 694–95). That being said, a supervisory official may, in limited circumstances, be liable under § 1983 for failure to train or to adopt policies if a plaintiff shows (1) the supervisor “failed to train the officers involved,” (2) “that failure to train . . . caused the violation of the plaintiff’s rights,” and (3) “the failure to train . . . constituted deliberate indifference.” Id. at 196.

Supervisory officials are deliberately indifferent if they retain a program for which they are on “notice that a particular omission in their training program causes [their] employees to violate citizens’ constitutional rights.” Connick v. Thompson, 563 U.S. 51, 61 (2011). Ordinarily, to show deliberate indifference, a plaintiff must point to “[a] pattern of similar constitutional violations by untrained employees.” Id. at 62. Constitutional violations must be “very similar” to “jointly form a pattern.” Jason, 938 F.3d at 198 (citing Connick, 563 U.S. at 62).

In Parker, we held that LeBlanc’s knowledge of three facts put him on notice of “a pattern of similar constitutional violations by untrained employees.” Parker, 73 F.4th at 405. In Parker, the plaintiff alleged that, in September 2017, a DPSC employee improperly labeled him a sex offender. Id. at 402. This error caused the plaintiff, who should have been released on October 9, 2017, to remain incarcerated until September 10, 2018. Id. at 402–03. The plaintiff alleged LeBlanc knew about the following:

an October 2017 legislative audit report on the Louisiana DPSC entitled “CFE Management of Offender Data: Processes for Ensuring Accuracy Department of Corrections”;
a 2018 editorial by Senator John Kennedy and Attorney General Landry entitled, “Criminal Justice Reform Actually Hurting Public Safety,” published in the newspaper “The Advocate”; and
testimony by DPSC employees admitting to rampant over-detention in a similar suit in Louisiana state court, Chowns v. LeBlanc, La. 37th JDC 26-932.

Id. at 405. Relying on Crittindon, we held these allegations demonstrated LeBlanc’s notice of a pattern of similar overdetentions to survive prong one of qualified immunity at the motion to dismiss stage. Ibid.

We are faced with two of the same relevant factual allegations made in Parker, and more.² Like the plaintiff there, McNeal alleges that LeBlanc knew about the October 2017 legislative audit and testimony by DPSC employees in Chowns v. LeBlanc, La. 37th JDC 26-932, admitting to rampant overdetention. McNeal also alleges LeBlanc had intimate knowledge about the results from the Lean Six Sigma report before McNeal’s overdetention occurred. Finally, McNeal alleges that before 2017, multiple public officials reached out to LeBlanc regarding overdetained prisoners. The Parker panel found fewer allegations sufficient to establish LeBlanc’s knowledge of a pattern of overdetention. We are thus bound under the rule of orderliness to find McNeal’s more numerous allegations sufficient to show deliberate indifference and to survive prong one of qualified immunity. See ibid.; see also United States v. Traxler, 764 F.3d 486, 489 (5th Cir. 2014).

2.

At prong two, we ask whether McNeal’s right was clearly established at the time of LeBlanc’s alleged deliberate indifference. We determine “what constitutes clearly established law” by “first look[ing] to Supreme Court precedent and then to our own.” Shumpert v. City of Tupelo, 905 F.3d 310, 320 (5th Cir. 2018). The touchstone at this prong is “fair warning” to the official “that his conduct deprived his victim of a constitutional right.” Hope v. Pelzer, 536 U.S. 730, 740 (2002). In other words, the law can be clearly established “despite notable factual distinctions between the precedents relied on and the cases then before the Court, so long as the prior decisions gave reasonable warning that the conduct then at issue violated constitutional rights.” Ibid. We must determine, accordingly, “whether it would be clear to a reasonable official that his conduct was unlawful in the situation he confronted.” Porter, 659 F.3d at 445 (alteration adopted) (citation omitted).

We held in Parker that by the fall of 2017 our law was “clear that a jailer like LeBlanc ha[d] a duty to ensure inmates [were] timely released from prison.” Id. at 407 (citation omitted). At the time of the overdetention in Parker, there was “sufficient clearly established law regarding the constitutional right to a timely release from prison.” Id. at 408. LeBlanc therefore had “fair warning that his failure to address” overdetention at the DPSC “would deny prisoners . . . their immediate or near-immediate release upon conviction.” Ibid. (cleaned up) (quoting Crittindon, 37 F.4th at 188). LeBlanc’s alleged deliberate indifference thus violated the prisoner’s clearly established right to a timely release. Ibid.

Parker constrains us to reach the same result in this case. McNeal’s alleged overdetention occurred in the fall of 2017, the same period the overdetention occurred in Parker. As in Parker, LeBlanc at that point had “fair warning that his failure to address” rampant overdetention in the DPSC “would deny prisoners like [McNeal] their immediate or near-immediate release upon conviction.” Ibid. (cleaned up) (quoting Crittindon, 37 F.4th at 188). Therefore, we are bound to follow that panel’s decision at prong two of qualified immunity. See Traxler, 764 F.3d at 489.

Accordingly, under our precedents, the district court did not err in denying LeBlanc qualified immunity at this stage.

IV.

The district court’s judgment is AFFIRMED.

EDITH H. JONES, Circuit Judge, concurring:

I concur that our precedent currently requires that Secretary LeBlanc be denied qualified immunity. See Parker v. LeBlanc, 73 F.4th 400 (5th Cir. 2023); Crittindon v. LeBlanc, 37 F.4th 177 (5th Cir. 2022), cert. denied, --- S. Ct. --- (2023). I further agree with Judge Duncan’s special concurrence advocating en banc review of this “mistaken” precedent, which “makes LeBlanc answerable for the errors of subordinates, creating vicarious liability in contravention of Monell v. Dep’t of Soc. Servs., 436 U.S. 658, 98 S. Ct. 2018 (1978), and Connick v. Thompson, 563 U.S. 51, 131 S. Ct. 1350 (2011).” In this case, for instance, there is nothing at all to connect LeBlanc with the events that resulted in McNeal’s overdetention.

But I also write separately because McNeal’s claims fail for an additional reason: they are barred by Heck v. Humphrey, 512 U.S. 477, 114 S. Ct. 2394 (1994). Judge Oldham explained in depth the reasons for Heck bar in Crittindon v. LeBlanc, as he stated “[b]oth the federal habeas statute, 28 U.S.C. § 2241, and the Civil Rights Act of 1871, 42 U.S.C. § 1983, create causes of action for prisoners with constitutional claims. But the remedies offered by those two statutes—and Congress’s limitations on them—differ radically.” 37 F.4th 177, 192 (5th Cir. 2022) (Oldham, J., dissenting). Specifically, “the habeas statute offers a singular equitable remedy: release from custody. But § 1983 goes further and also offers money damages and attorney’s fees.” Id. at 193 (citations omitted). Moreover, § 1983 “comes with none of” the “numerous severe limitations” that Congress has placed on federal habeas. Id. at 193, 192 (citing The Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”), Pub. L. No. 104–132, 110 Stat. 1214).

In Heck, the U.S. Supreme Court “recognized this ‘potential overlap between’ habeas and § 1983, and it cut off access to the latter in cases where the prisoner’s claim sounds in the former.” Id. at 193 (quoting Heck, 512 U.S. at 481, 114 S. Ct. at 2369–70). “The upshot is that, where a prisoner can obtain relief through habeas, he cannot sue under § 1983.” Id. Heck built on the rule announced twenty years earlier in Preiser v. Rodriguez: “Congress has determined that habeas corpus is the appropriate remedy for state prisoners attacking the validity of the fact or length of their confinement, and that specific determination must override the terms of § 1983.” 411 U.S. 475, 490–91, 93 S. Ct. 1827, 1836 (1973). The U.S. Supreme Court has repeatedly reaffirmed this principle, stating that “we have insisted that § 1983 contains an ‘implicit exception’ for actions that lie ‘within the core of habeas corpus.’” Nance v. Ward, 597 U.S. 159, 167, 142 S. Ct. 2214, 2221 (2022) (quoting Wilkinson v. Dotson, 544 U.S. 74, 79, 12 S. Ct. 1242, 1246 (2005)).

In this case, McNeal was released from custody 41 days late because the Louisiana Department of Public Safety and Corrections sent his release paperwork to the wrong facility. Thus, his case concerns a challenge to “the fact and length of his confinement.” Id. “That means [his] only remedy lies in habeas. And the Heck doctrine plainly bars [him] from ignoring the specific terms of the habeas statute, which ‘must override the general terms of § 1983.’” Crittindon, 37 F.4th at 193 (quoting Preiser, 411 U.S. at 491, 93 S. Ct. at 1836).

But McNeal made no good faith efforts to seek state habeas relief. Unlike some of the Crittindon plaintiffs, who at least filed petitions for writs of habeas corpus in Louisiana state court, 37 F.4th at 194, McNeal never made any such filings. At most, he alleges that he “wrote a letter” to the warden of the facility where he was detained and spoke with some of the officers. These actions fall well short of a good faith effort to seek state habeas relief. Under Heck, only after McNeal successfully obtained such relief via a valid state court order declaring the confinement “invalid” could he state a claim under § 1983. 512 U.S. at 477, 114 S. Ct. at 2372.

Allowing McNeal and other “overdetention” plaintiffs to obtain § 1983 relief without requiring them to make even a good faith effort to obtain state habeas relief not only violates the unambiguous language of Heck and Preiser—as well as centuries of habeas jurisprudence—, it yields perverse incentives for litigants as well. It is well established in civil cases that litigants have a duty to mitigate their damages after an injury occurs. See, e.g., Energy Intel. Grp., Inc. v. Kayne Anderson Capital, 948 F.3d 261, 274 (5th Cir. 2020). In the criminal context, Heck imposes an analogous duty to mitigate such that any overdetained prisoner must seek habeas relief at the earliest possible opportunity.¹ State habeas, after all, comprises the most traditional and effective tool to obtain the equitable relief a prisoner ultimately seeks: release from custody. This court should not enable overdetained prisoners to neglect their obligation to seek habeas relief and instead bypass that remedy in order to pursue Section 1983 damages by filing for the wrong type of relief in the wrong court at the wrong time.

Louisiana has serially defaulted in its obligation to release prisoners on time. It is beyond this panel’s purview to analyze, much less solve this critical problem. However, we should have demonstrated confidence in the state courts’ ability, through habeas corpus, to resolve individual cases by remitting individuals like McNeal to the state court system for exhaustion of remedies. This seems to me, as to Judge Oldham, a classic situation that Heck intended to address. We should revisit Crittindon en banc and overrule it.

STUART KYLE DUNCAN, Circuit Judge, concurring:

I concur in denying Secretary LeBlanc qualified immunity but only because our precedent requires that result. See Parker v. LeBlanc, 73 F.4th 400 (5th Cir. 2023); Crittindon v. LeBlanc, 37 F.4th 177 (5th Cir. 2022), cert. denied, 144 S. Ct. 90 (2023). Our precedent is mistaken, however. It makes LeBlanc answerable for the errors of subordinates, creating vicarious liability in contravention of Monell v. Dep’t of Soc. Servs., 436 U.S. 658 (1978), and Connick v. Thompson, 563 U.S. 51 (2011). To repair that far-reaching error, our court should rehear this case en banc.

We have had several overdetention cases involving DPSC and LeBlanc.¹ We will likely have many more. See Hicks v. LeBlanc, 81 F.4th 497, 510 (5th Cir. 2023) (“[O]ur Court remains plagued by claims arising from inexplicable and illegal over-detention in Louisiana prisons[.]”). The question is not whether overdetention is a serious problem (it is) nor whether it should be fixed (it should). The question, instead, is about the proper remedy: whether LeBlanc, the head of a large and complex state agency, can be held personally liable under § 1983 for causing a prisoner’s overdetention.

To answer that question, our circuit borrows the standard for finding a municipality liable under § 1983. See, e.g., Southard v. Tex. Bd. of Crim. Just., 114 F.3d 539, 551 (5th Cir. 1997) (noting “the close relationship between the elements of municipal liability and an individual supervisor’s liability,” and holding “the same standards of fault and causation should govern” (quotation omitted)). Under that framework, LeBlanc cannot be vicariously liable for an overdetention caused by a subordinate’s error. See Thompson v. Upshur County, 245 F.3d 447, 459 (5th Cir. 2001). LeBlanc must have caused the overdetention himself. To establish that, one must show that LeBlanc’s “deliberately indifferent” failure to train subordinates caused the overdetention. See Jason v. Tanner, 938 F.3d 191, 196 (5th Cir. 2019) (holding warden could be liable only if his deliberately indifferent failure to train “caused the violation of the plaintiff’s rights”).² And the usual way of establishing failure-to-train liability is to show “[a] pattern of similar constitutional violations by untrained employees.” Connick, 563 U.S. at 62.³

That’s where the problem begins. The pattern requirement is critical because it keeps failure-to-train from collapsing into respondeat superior. See Jason, 938 F.3d at 197–98 (holding a pattern of “very similar” violations is required to avoid “pure respondeat superior liability under § 1983”); cf. Connick, 563 U.S. at 60 (warning against holding municipal governments “vicariously liable under § 1983 for their employees’ actions”). The prior violations must be closely similar to the present one. Otherwise, a supervisor would not be on notice of a flaw in the agency’s training program, nor would he have any idea how to change the training to fix the problem. See Connick, 563 U.S. at 62 (“Without notice that a course of training is deficient in a particular respect, decisionmakers can hardly be said to have deliberately chosen a training program that will cause violations of constitutional rights.”). In other words, absent an alarm bell rung by a pattern, the supervisor’s liability would depend solely on his subordinate’s error, not on anything the supervisor himself failed to do. That is respondeat superior liability, and it is excluded in § 1983 claims. See Monell, 436 U.S. at 694–95.

The paradigm illustration of the pattern requirement comes from Connick, 563 U.S. 51. A line prosecutor violated Brady v. Maryland, 373 U.S. 83 (1963), by failing to disclose exculpatory evidence. Could the district attorney’s office be liable for that misdeed because the office failed to properly train prosecutors on Brady? The answer turned on whether the office had a pattern of previous violations. Critically, though, it wasn’t enough to say, “The office had past Brady violations.” Cf. Connick, 563 U.S. at 62–63. The Supreme Court demanded more. Prior violations had to be specific enough to “put [the district attorney] on notice that the office’s Brady training was inadequate with respect to the sort of Brady violation at issue here.” Id. at 62 (emphasis added); see also Jason, 938 F.3d at 198 (holding no deliberate indifference where the specific kind of complained-of prison attack had not previously occurred).

Respectfully, our circuit has not correctly applied Connick’s pattern requirement in DPSC overdetention cases. Overdetentions occur for many reasons, as our cases show. A department employee may misclassify a prisoner. See Parker, 73 F.4th at 402. Or an employee may fail to apply time-served credits. See Hicks, 81 F.4th at 501. Or an employee may misapply the law for calculating time-served credits. See Taylor v. LeBlanc, 68 F.4th 223, 225 (5th Cir. 2023), withdrawn by 2023 WL 4155921 (5th Cir. June 23, 2023). Or local jails may fail to timely transmit pre-classification paperwork to DPSC. See Crittindon, 37 F.4th at 183. To make LeBlanc liable for any one of these overdetentions, Connick requires evidence of a pattern of closely similar violations sufficient to notify LeBlanc that his department’s flawed training caused the particular violation. See, e.g., Jason, 938 F.3d at 198 (“[T]he Supreme Court in Connick required that only very similar violations could jointly form a pattern.” (citing Connick, 563 U.S. at 62)). Our cases, however, have watered down this pattern requirement.

Parker is a good example. Parker was overdetained because a DPSC employee allegedly “misclassif[ied] him as a sex offender.” 73 F.4th at 402. Although reciting Connick’s pattern requirement, id. at 405, the panel rejected the argument that Parker failed to allege a pattern of similar misclassifications. See id. at 406 (rejecting LeBlanc’s “distinction” between overdetention due to “misclassification” and overdetention due to other causes). Instead, the panel held that Parker satisfied Connick by pointing to “three pieces of evidence”: (1) an audit of DPSC; (2) the Kennedy-Landry op-ed; and (3) the Chowns testimony. Id. at 405; see also Op. at 8–9 (discussing this evidence). This evidence, the panel explained, supported a pattern of “similar” overdetentions by showing “LeBlanc was aware of the deficiencies of implemented policies that routinely led to errors like the one that violated [Parker’s] constitutional rights.” Ibid.

That reasoning misapplies Connick. “[O]nly very similar violations could jointly form a pattern [under Connick],” as our court has correctly explained. Jason, 938 F.3d at 198. So, what is “very similar” about the pattern of prior overdetentions in Parker and the overdetention at issue there? The pattern evidence, the panel claimed, showed LeBlanc knew about “deficiencies in the way . . . DPSC calculated and implemented release dates.” Id. at 403. That is plainly insufficient. The violation in Parker wasn’t caused by generic errors in “calculating and implementing release dates.” It was caused by an employee’s misinterpreting Parker’s criminal history to include an offense requiring sex-offender registration. See id. at 403–04. Parker did not explain how its pattern evidence had the slightest thing to do with that classification mistake.⁴

Without the pattern evidence required by Connick, a failure-to-train claim against LeBlanc collapses into vicarious liability. That is, under our precedent, LeBlanc can be liable as a supervisory official based merely on the fact that overdetentions and delays in processing release dates, writ large, have occurred within DPSC. If that were enough to prove deliberate indifference, though, Connick would have come out the other way. The district attorney would have been liable for failing to train prosecutors merely because previous Brady violations had occurred in his office. Of course, that is not what Connick held. The required pattern had to show, instead, that “the office’s Brady training was inadequate with respect to the sort of Brady violation at issue [t]here.” Connick, 563 U.S. at 62 (emphasis added). That stringent evidentiary foundation is missing here because our precedent, in contravention of Connick, rejects it. Our en banc court should correct that far-reaching error.

One final note. Our cases speak in the same breath of a supervisor’s liability for “failure to train” and for “failure to adopt policies.” See Parker, 73 F.4th at 404–05 (discussing a supervisor’s liability for “failure to adopt policies if that failure causally results in a constitutional injury” (quoting Crittindon, 37 F.4th at 186)); see also Porter v. Epps, 659 F.3d 440, 446 (5th Cir. 2011) (discussing “[l]iability for failure to promulgate policy and failure to train or supervise”). It is unclear to me whether those are meant to be different theories of supervisory liability or different articulations of the same theory. See, e.g., Parker 73 F.4th at 405–06 (toggling between LeBlanc’s liability for “failure to train” and “failure to adopt policies”). If the former, then I seriously doubt that a “failure to adopt policy” theory has any basis in the Supreme Court’s case law. Connick is quite specific that it was addressing liability for a “failure to train.” See Connick, 563 U.S. at 61–63. It said nothing about a generic “failure to adopt or promulgate policies.”

Imposing liability because a supervisor “fails to adopt policies” opens a much broader vista of supervisory liability than for “failing to train” subordinates. Indeed, Connick explained that the “most tenuous” type of deliberate-indifference liability was “failure to train.” Id. at 61; see also Oklahoma City v. Tuttle, 471 U.S. 808, 822–23 (1985) (plurality opinion) (holding inadequate training is “far more nebulous, and a good deal further removed from the constitutional violation, than was the policy in Monell”). That is because the theory is based not on a municipality’s action but its omission. Premising liability on failing to “adopt policies,” however, is even more tenuous. At least a failure to train is focused on a supervisor’s omissions with respect to a particular duty (training employees) and in response to a problem that training could solve (a pattern of prior employee violations linked to inadequate training). A “failure to adopt policies,” by contrast, appears to open supervisors to liability merely for failing to be clairvoyant. That cannot meet the stringent standard of deliberate indifference. See Connick, 563 U.S. at 62 (“Without notice that a course of training is deficient in a particular respect, decisionmakers can hardly be said to have deliberately chosen a training program that will cause violations of constitutional rights.”).

That problem aside, however, our precedent is clear that a supervisor’s liability in a case like this must be grounded on a pattern of prior, similar violations. See Jason, 938 F.3d at 198 (citing Connick, 563 U.S. at 62). That requirement has not been met here.

We should rehear this case to fix the problem. As noted, we will likely have many more overdetention cases against LeBlanc and others. We need to clarify when officials can be liable for overdetaining prisoners. If we fail to do that, we risk turning § 1983 into a source of vicarious liability for the heads of State agencies. In addition to violating Supreme Court precedent, such a misguided project would be futile. The overdetention problem is obviously a serious one. But if evidence does not connect the problem to something LeBlanc himself has done or failed to do, then making him personally liable for overdetentions will solve nothing.

I urge our court to rehear this pressing issue en banc.

Notes

The facts are taken from the allegations in McNeal’s complaint, which we accept as true. See . Federal habeas relief is textually available to any federal or state prisoner who is “in custody” in violation of the Constitution or laws or treaties of the United States. 28 U.S.C. § 2241(c)(3). The definition of custody does not textually depend on the length or nature of the confinement. I assume state habeas statutes are similarly expansive. The Great Writ, after all, began with Magna Carta to limit the king’s ability to imprison people. See, e.g., ; ; ; (per curiam) (unreported) (vacating denial of qualified immunity where plaintiff alleged his release date was miscalculated); (per curiam) (unreported) (vacating denial of summary judgment and remanding for further proceedings in light of where plaintiff alleged DPSC and Orleans Parish improperly calculated time served credits); (per curiam) (unreported) (granting qualified immunity to LeBlanc where plaintiff alleged DPSC improperly placed a parole hold on him).

While McNeal also alleges LeBlanc knew about the 2018 Kennedy-Landry editorial, that knowledge is not relevant to LeBlanc’s liability in this case. According to McNeal’s complaint, the op-ed was not published until 2018, after McNeal’s November 2017 overdetention and December 2017 release. True, there are other ways to make a supervisory official liable. One could show the violation was caused by an actual policy promulgated by the official or by the official’s own policymaking decision. See ; ; . But neither ground is alleged here. So, the only way to find LeBlanc liable is through a failure-to-train theory.
Here, McNeal has sued LeBlanc under both failure-to-train and failure-to supervise theories. I assume for present purposes those are distinct theories. But like failure-to-train, a failure-to-supervise claim must point to a “causal link” between a supervisor’s failure to supervise, amounting to “deliberate indifference,” and a subordinate’s acts that “cause[d] plaintiff’s constitutional injury.” (citation omitted). McNeal does not allege facts showing LeBlanc’s individual failure to supervise led to this particular delayed release. I therefore only discuss whether McNeal sufficiently alleged a claim for failure to train.

Yes, there’s a rare alternative where a single incident is enough to show that a supervisor was deliberately indifferent. See (discussing (O’Connor, J., concurring in part and dissenting in part)). But no one argues that unusual exception applies here.

Our decision in similarly misapplies the pattern requirement. involved an overdetention caused by a communication breakdown between DPSC and parish prisons. . Focusing on the Lean Six Sigma review, we found LeBlanc on notice merely that “DPSC prisoners were annually held past their release date” due “to delays in determining prisoners’ release dates.” . But the panel did not ask, as requires, whether this particular cause of the delays had cropped up before, much less whether a pattern of similar occurrences should have put LeBlanc on notice that this was an issue. Instead, we merely addressed overdetention writ large and found the evidence sufficient to show a “pattern of delays” in determining release dates. See (holding “[a] reasonable factfinder could conclude that [LeBlanc’s] awareness of this pattern of delays [and unlawful detentions of prisoners] and [his] conscious decision not to address it rises to the level of deliberate indifference”).