Garrett v. Davis: Eighth Amendment Sleep-Deprivation Claims Require Proof of Substantial Risk (Not Causation), No Numeric Sleep Threshold, and a Duration-Sensitive Totality Analysis

1. Introduction

Garrett v. Davis is the Fifth Circuit’s fourth encounter with Michael Garrett’s challenge to Texas Department of Criminal Justice (TDCJ) schedules that allegedly restrict him to roughly three to four hours of nighttime sleep, compounded by interruptions (cell-door operations, lighting, and bed counts). Garrett, a Texas prisoner, sued TDCJ officials under 42 U.S.C. § 1983 seeking injunctive relief requiring at least six hours of nightly sleep, alleging an Eighth Amendment violation.

The central issue is doctrinal: what the Eighth Amendment requires a prisoner to prove—objectively and subjectively—when alleging unconstitutional sleep deprivation. Twice before, the Fifth Circuit remanded because the district court applied the wrong legal standard. In this appeal, the panel again finds the district court demanded proof that Fifth Circuit precedent explicitly does not require: (i) individualized causation of a medical injury, and (ii) a scientifically fixed numerical “threshold” of minimum sleep. The panel also faulted the district court for failing to treat duration as an indispensable part of the objective inquiry.

Parties and posture

• Plaintiff–Appellant: Michael Garrett
• Defendants–Appellees: Director Lorie Davis; Bobby Lumpkin; Eric Guerrero
• Procedural posture: Appeal from a bench-trial/post-remand judgment denying relief on Garrett’s Eighth Amendment sleep-deprivation claim

2. Summary of the Opinion

The Fifth Circuit vacates and remands because the district court, despite prior Fifth Circuit instructions, again evaluated the Eighth Amendment claim under legally erroneous criteria.

Specifically, on the objective prong the district court: (1) discounted expert evidence for showing “association” rather than “cause-and-effect,” effectively requiring causation rather than substantial risk; (2) required proof of a precise “specific number of hours of continuous sleep” needed to avoid harm, treating the absence of a fixed benchmark as fatal; and (3) failed to account adequately for duration—how long Garrett has endured the deprivation—within the totality-of-circumstances analysis.

On the subjective prong, the district court reasoned it could not find deliberate indifference without proof of the “threshold of sleep” required to avoid substantial risk—an approach the panel holds improperly imports the same mistaken premise from the objective inquiry.

The panel declines to render judgment for Garrett because unresolved factual questions remain under the correct standard, and thus the “only one resolution” exception does not apply.

3. Analysis

3.1 Precedents Cited

A. The Garrett trilogy and the governing correction

• Garrett v. Thaler (“Garrett I”), 560 F. App’x 375 (5th Cir. 2014): The Fifth Circuit first held Garrett plausibly alleged both the objective and subjective components of an Eighth Amendment sleep-deprivation claim, reversing a failure-to-state-a-claim dismissal. This established that sleep deprivation can satisfy the Eighth Amendment framework at the pleading stage when sufficiently severe and knowingly imposed.
• Garrett v. Lumpkin (“Garrett II”), 840 F. App’x 807 (5th Cir. 2021): After a bench trial, the district court denied relief based largely on lack of proven cause-and-effect between sleep schedule and medical complaints, and on “legitimate penological interest.” The Fifth Circuit remanded due to intervening transfer and the need to assess conditions at the new unit in the first instance.
• Garrett v. Lumpkin (“Garrett III”), 96 F.4th 896 (5th Cir. 2024): The key doctrinal anchor. The Fifth Circuit clarified that the objective component asks whether conditions pose “a substantial risk of serious harm”—not whether they caused a particular injury—and that penological purpose does not control deliberate indifference. The current opinion is, in substance, an enforcement of Garrett III against repeated district-court reversion to disapproved tests.

B. Standard of review and consequences of legal error

• Barto v. Shore Constr., L.L.C., 801 F.3d 465 (5th Cir. 2015), quoting Becker v. Tidewater, Inc., 586 F.3d 358 (5th Cir. 2009): Bench-trial findings of fact are reviewed for clear error; legal issues de novo.
• Fuji Photo Film Co. v. Shinohara Shoji Kabushiki Kaisha, 754 F.2d 591 (5th Cir. 1985): Clear-error deference does not apply where factual determinations are reached by applying an incorrect legal standard.
• Ball v. LeBlanc, 792 F.3d 584 (5th Cir. 2015): When legal error affects factual findings, remand is proper unless the record permits only one resolution of the factual issue. The panel uses Ball both for remedial posture (remand) and for Eighth Amendment doctrinal statements.
• Harris v. Forklift Sys., Inc., 510 U.S. 17 (1993), and June Med. Servs., L.L.C. v. Phillips, 22 F.4th 512 (5th Cir. 2022): Remand is appropriate where application of the wrong standard may have influenced the ultimate conclusion.

C. Eighth Amendment framework: objective risk and subjective deliberate indifference

• Harper v. Showers, 174 F.3d 716 (5th Cir. 1999): Establishes that sleep is “one of life’s basic needs,” supporting the premise that severe sleep deprivation can implicate the Eighth Amendment.
• Helling v. McKinney, 509 U.S. 25 (1993): Critical to the panel’s rejection of a causation requirement. Helling recognizes Eighth Amendment claims based on exposure to conditions posing future health risks (e.g., secondhand smoke) supported by statistical/scientific evidence, without requiring proof of present disease in the particular inmate.
• Farmer v. Brennan, 511 U.S. 825 (1994): Supplies the canonical articulation: objective “sufficiently serious” deprivation and subjective deliberate indifference (knowledge and disregard of excessive risk).
• Alexander v. Texas Dep’t of Crim. Just., 951 F.3d 236 (5th Cir. 2020): Reinforces the “minimal civilized measure of life’s necessities” framing and the “totality of the record evidence” approach to subjective deliberate indifference.
• McCord v. Maggio, 927 F.2d 844 (5th Cir. 1991): Used to illustrate that unconstitutional conditions (there, unsanitary conditions) need not be proven to have caused illness before they violate the Constitution—risk can suffice.

D. Totality-of-circumstances, no numeric minima, and the special role of duration

• Palmer v. Johnson, 193 F.3d 346 (5th Cir. 1999): Emphasizes totality of circumstances; cited here to reject a requirement of universal quantified minima and to anchor the duration-sensitive evaluation.
• Berry v. Brady, 192 F.3d 504 (5th Cir. 1999), and Talib v. Gilley, 138 F.3d 211 (5th Cir. 1998): Stand for the proposition that whether deprivation falls below constitutional thresholds depends on amount and duration; also support the broader point that the Eighth Amendment does not demand fixed numerical minima (food/calories).
• Ruiz v. Estelle, 679 F.2d 1115 (5th Cir. 1982), amended in part and vacated in part on other grounds, 688 F.2d 266 (5th Cir. 1982): Cited among Fifth Circuit authorities recognizing no fixed minimums (exercise) and reflecting the system-wide, contextual nature of conditions analysis.
• Hutto v. Finney, 437 U.S. 678 (1978), abrogated in part on other grounds as recognized by, Dep’t of Agric. Rural Dev. Rural Hous. Serv. v. Kirtz, 601 U.S. 42 (2024): Quoted for the proposition that the “length of confinement cannot be ignored” and that conditions tolerable for days can become cruel for weeks or months.
• Cooper v. Sheriff, Lubbock Cnty., 929 F.2d 1078 (5th Cir. 1991): Adds the principle that continual use of a facially permissible punishment can become cruel and unusual.

3.2 Legal Reasoning

A. Objective prong: three linked errors and the correct test

The panel frames the objective inquiry as whether Garrett’s “actual conditions pose a substantial risk of serious harm”—a risk-focused assessment that does not require (i) individualized medical causation, (ii) a universally quantified minimum sleep number, or (iii) ignorance of duration.

(1) Risk, not causation

The district court stated Garrett “need not prove actual harm,” but then rejected expert evidence because it showed “only an association” rather than “cause-and-effect.” The Fifth Circuit treats this as a direct reintroduction of an “actual injury/causation” requirement rejected in Garrett III. Drawing on Helling v. McKinney, the panel reiterates that scientific/statistical evidence may establish unconstitutional exposure to serious risk even when no present disease is proven in the individual plaintiff.

(2) No requirement of a numeric societal benchmark

The district court required proof of a “specific number of hours of continuous sleep” and faulted Garrett for not identifying the amount of sleep “a people in a civilized world must actually get” to avoid harm. The Fifth Circuit rejects this as inconsistent with its conditions-of-confinement cases, which routinely refuse to constitutionalize fixed numerical minima (temperature, food, exercise, sanitation). The constitutional question is contextual: whether the conditions, as experienced, deprived the prisoner of “the minimal civilized measure of life’s necessities” and posed a substantial risk of serious harm.

(3) Duration is indispensable

The panel holds that duration is part of the “totality of the specific circumstances presented.” The district court’s analysis failed to integrate how long Garrett has allegedly endured restricted and fragmented nighttime sleep. Quoting Hutto v. Finney, the panel highlights a core Eighth Amendment insight: severity and time interact; what might be tolerable briefly can become unconstitutional when prolonged.

Importantly, the panel does not declare daytime sleep irrelevant. It rejects Garrett’s argument that considering daytime “chunks” exceeded the remand’s scope, explaining the complaint concerns “sleep deprivation,” not strictly “continuous nighttime sleep.” But it insists that any mitigation theory (daytime naps, other opportunities) must be evaluated within the full duration-and-severity context.

B. Subjective prong: deliberate indifference without a “threshold” trap

The district court concluded that without proof of a precise “threshold of sleep” required to avoid substantial risk, it could not find TDCJ knew of the threshold and disregarded it. The Fifth Circuit rejects this logic because it improperly makes the subjective prong depend on the same erroneous premise as the objective prong: that constitutional liability requires a scientifically fixed minimum.

On remand, the district court must reassess deliberate indifference under the correct objective standard, focusing on (i) what TDCJ officials knew about the risks Garrett identified (including through complaints and evidence), and (ii) what responsive measures, if any, were taken—consistent with Farmer v. Brennan, Ball v. LeBlanc, and Alexander v. Texas Dep’t of Crim. Just..

C. Remedy: why the Fifth Circuit remands rather than renders

Garrett sought rendition (judgment in his favor), but the panel applies Ball v. LeBlanc to hold that unresolved factual questions remain once the correct legal standards are applied—especially (1) whether the schedule, considering all meaningful sleep opportunities and the duration of exposure, poses a substantial risk of serious harm, and (2) whether officials knew of and disregarded an excessive risk on the total record. Thus, remand—not rendition—is required.

3.3 Impact

A. Doctrinal impact on Eighth Amendment sleep-deprivation litigation

• Re-centering on “substantial risk”: The opinion reinforces that plaintiffs need not prove individualized medical causation; demonstrating exposure to conditions that pose a substantial risk of serious harm suffices for the objective prong. This materially affects evidentiary framing in prison cases where long-term harms are probabilistic and mediated by multiple factors.
• Rejecting “numeric minimum sleep” requirements: By analogizing to temperature, food, exercise, and sanitation cases, the panel signals that district courts should not demand litigants identify a universally accepted physiological minimum (e.g., “X hours”) as a prerequisite to Eighth Amendment protection.
• Duration as a required analytical dimension: The opinion elevates duration from a generic consideration to an “indispensable” one in assessing sleep deprivation. This encourages courts to scrutinize prolonged exposure, not isolated snapshots, and to avoid treating long-running schedules as constitutionally neutral absent acute injury.
• Subjective prong without scientific “certainty”: The court’s rejection of the “threshold trap” clarifies that deliberate indifference does not require officials to know a precise numerical boundary—only that they know of and disregard an excessive risk under the circumstances.

B. Practical impact on proof and case management

• Expert testimony: The opinion protects the probative value of epidemiological and associational evidence in conditions cases. Experts need not prove deterministic causation; they can support findings of substantial risk.
• Injunctive relief posture: By emphasizing risk and duration, the opinion aligns with forward-looking remedies (altering schedules to reduce risk) rather than backward-looking damages keyed to a proven injury traceable to a single cause.
• Remand discipline: Given the case’s repeated remands, the decision implicitly warns that persistent reapplication of rejected standards is reversible error, increasing the likelihood of vacatur where district courts drift back to causation/benchmark frameworks.

4. Complex Concepts Simplified

• Objective prong (conditions severity): Asks whether the conditions are sufficiently serious—i.e., whether they deprive the prisoner of basic human needs and pose a substantial risk of serious harm. It is not limited to harms that have already occurred.
• Subjective prong (deliberate indifference): Asks whether officials knew of an excessive risk to health or safety and disregarded it. It is more than negligence but does not require certainty or a precise numeric “line” officials consciously crossed.
• Substantial risk vs. causation: “Substantial risk” means a serious likelihood of harm; “causation” would require proof the condition actually caused a specific injury in this plaintiff. The Eighth Amendment permits claims based on serious future risks (e.g., Helling v. McKinney).
• Totality of the circumstances: Courts evaluate the whole picture—severity, frequency of interruptions, lighting/noise, and especially duration—rather than isolating one metric like a universal “minimum sleep hours” number.
• VACATE and REMAND: The appellate court nullifies the judgment (vacate) and sends the case back (remand) for reconsideration under the correct legal rules.

5. Conclusion

Garrett v. Davis underscores a clear rule for Eighth Amendment sleep-deprivation claims in the Fifth Circuit: the objective inquiry is risk-based (substantial risk of serious harm), not causation-based; courts may not require plaintiffs to supply a scientifically fixed numeric minimum sleep benchmark; and duration is indispensable in the totality-of-circumstances analysis. Because the district court’s errors also distorted the deliberate-indifference inquiry, the Fifth Circuit again vacated and remanded for application of the correct standards, leaving the ultimate factual determinations to be resolved on remand.