Harvard Law Review Harvard Law Review Harvard Law Review

Civil Rights

The Coherence of Prison Law


In their welcome new article, Justin Driver and Emma Kaufman offer a provocative take on American prison law: that it is “fundamentally incoherent.”1×1. Justin Driver & Emma Kaufman, The Incoherence of Prison Law, 135 Harv. L. Rev. 515, 522 (2021). They base this conclusion on the Supreme Court’s repeated tendency to assert contradictory factual premises about prisoners and prison life.2×2. In this essay, I follow Driver and Kaufman’s practice of at times referring to incarcerated people as “prisoners,” a term that squarely acknowledges the “extraordinary and dehumanizing exercise of state power known as imprisonment,” id. at 525, and foregrounds the experience of being held against one’s will with no power to shape one’s own conditions of life. See Paul Wright, Language Matters: Why We Use the Words We Do, Prison Legal News (Nov. 1, 2021), https://www.prisonlegalnews. org/news/2021/nov/1/language-matters-why-we-use-words-we-do [https://perma.cc/GK52-S45Z] (“[When people are incarcerated, they] are forced into cages at gun point and kept there upon pain of death should they try to leave. What are they if not prisoners? They did not somehow magically appear there and they stay there based on violence and fear of violence . . . .”). In one case, as the authors show, the Court will characterize prisons as violent and in another as “uncomfortable but mundane”;3×3. Driver & Kaufman, supra note 1, at 567. sometimes the Court describes prisoners as illiterate, at other times as strategic and effective litigators;4×4. See id. at 550, 552–53. and so on. If ever one imagined this area of the law to have a stable factual foundation, Driver and Kaufman’s dexterous excavation of the Court’s “selective empiricism”5×5. Id. at 567. puts that notion firmly to rest.6×6. See id. at 567–71.

But viewed through a broader lens, the Court’s prison law jurisprudence proves anything but incoherent. For all the factual switchbacks Driver and Kaufman identify, there is an unmistakable consistency in the overall orientation of the field: it is consistently and predictably pro-state, highly deferential to prison officials’ decisionmaking, and largely insensitive to the harms people experience while incarcerated. These features represent the practical manifestation of the divergent normative inclinations the Supreme Court routinely displays toward the parties in prison law cases. It is hardly a secret that American carceral institutions routinely burden prisoners’ fundamental liberties and fail to provide

even minimally safe and healthy living conditions.7×7. See, e.g., Sharon Dolovich, Prison Conditions, in 4 Reforming Criminal Justice: Punishment, Incarceration, and Release 261, 263–64 (Erik Luna ed., 2017); Sharon Dolovich, The Failed Regulation and Oversight of American Prisons, 5 Ann. Rev. Criminology 153, 156–58 (2022) [hereinafter Dolovich, The Failed Regulation]; Eyal Press, A Fight to Expose the Hidden Human Costs of Incarceration, New Yorker (Aug. 16, 2021), https://www.newyorker.com/magazine/2021/08/23/a-fight-to-expose-the-hidden-human-costs-of-incarceration [https://perma.cc/5TRU-7E58]. Yet with prison law’s moral center of gravity tilting so far in the direction of defendants, plaintiffs bringing constitutional claims in federal court can expect to win only in the most extreme cases, leaving the prison environment largely free of judicial regulation.

In this essay, I explore the mechanisms by which, despite what is known about the reality on the ground in American prisons, courts hearing constitutional challenges brought by prisoners so persistently find in favor of the state. In particular, I zero in on two components of the judicial process in this context: the construction of defendant-friendly doctrinal standards for deciding prisoners’ claims and the deferential posture with which federal courts tend to approach defendants’ assertions in individual cases. As to the doctrine, especially during the Rehnquist Court, the Supreme Court systematically deployed a set of maneuvers — which I have elsewhere termed canons of evasion8×8. See Sharon Dolovich, Canons of Evasion in Constitutional Criminal Law, in The New Criminal Justice Thinking 111, 111 (Sharon Dolovich & Alexandra Natapoff eds., 2017). — to construct doctrinal standards for prison law cases that strongly incline courts to rule in favor of the state. In Part I, by way of illustration, I map the deployment of these various mechanisms in two especially consequential cases, Whitley v. Albers9×9. 475 U.S. 312 (1986). and Turner v. Safley,10×10. 482 U.S. 78 (1987). and show how their use operates to create a doctrinal environment decidedly unfavorable to prisoners’ claims.

Yet skewed doctrinal standards alone cannot explain prison law’s strong pro-state bent. Given the generally noxious character of American prisons, one would still expect incarcerated plaintiffs to prevail more frequently notwithstanding onerous standards. This brings us to the second piece of the puzzle: the way that, in practice, courts hearing prison law cases will often side with defendants even when plaintiffs’ claims are strong on the merits and even when defendants’ proffered arguments strain credulity. To achieve this effect requires a judicial readiness to see the state’s case through an especially sympathetic lens and to exhibit a studied indifference to plaintiffs’ constitutional rights and lived experience. In Part II, I argue that, in prison law cases, judges are primed to approach the parties’ submissions in precisely this way. Examining the Supreme Court’s prison law opinions, I surface the moral psychology they promote, which orients courts to regard prison officials’ arguments favorably while viewing prisoners’ claims with skepticism and even hostility. As a consequence of this posture, which I term dispositional favoritism, federal courts hearing prison cases can wind up favoring defendants in any number of ways hard to square with either the record or the relevant legal rules — including making questionable factual assertions, a phenomenon of a piece with the Court’s use of “selective factual generalizations” that Driver and Kaufman so definitively expose.11×11. Driver & Kaufman, supra note 1, at 522. It is, in other words, the Court’s dispositional favoritism that explains the contradictory factual premises “about the purpose and inhabitants of penal institutions” that Driver and Kaufman track in such revealing detail across the cases and that, as the authors note, “consistent[ly] . . . tend to shift in ways that benefit the government.”12×12. Id. at 568.

Certainly, prisoners who manage to get into federal court13×13. See Dolovich, supra note 8, at 114–16 (canvassing the many obstacles the incarcerated face to getting a hearing on the merits in federal court). will sometimes win. But for incarcerated plaintiffs to prevail on the merits typically requires glaringly indefensible treatment, highly questionable official justifications, dedicated and adept lawyers committed to pressing plaintiffs’ claims, and courts open to taking those claims seriously.14×14. See, e.g., Brown v. Plata, 563 U.S. 493, 493 (2011); Greenhill v. Clarke, 944 F.3d 243, 245–46 (4th Cir. 2019); Gray v. Hardy, 826 F.3d 1000, 1000 (7th Cir. 2016); Prison Legal News v. Cook, 238 F.3d 1145, 1153 (9th Cir. 2001); Coleman v. Schwarzenegger, 922 F. Supp. 2d 882, 1003–04 (E.D. Cal. 2009). These combined requisites are sufficiently rare that, for the most part, macro-level conditions remain undisturbed by the courts, which can create the misimpression that the baseline reality of life in prison must pose no constitutional problem. The marked disinclination of federal courts to find all but the most extreme conditions unconstitutional thus facilitates judicial findings for defendants in subsequent cases. It also helps to vindicate the seeming moral rightness of those holdings, since if prison conditions are known in the main to be constitutionally unproblematic, prisoners alleging unconstitutional treatment must only be trying to game the system to get more than they deserve. In this way, dispositional favoritism is self-reinforcing.

These dynamics, hidden in plain sight, had been present in the prison law doctrine for decades. Then came Covid-19. Suddenly, what may previously have seemed like the piecemeal shielding of prison officials from constitutional liability emerged as an undeniable uniform refusal on the part of the federal courts to seriously entertain any constitutional challenge to conditions plainly putting people in prison at outsized risk of illness and death.15×15. Brendan Saloner et al., Research Letter, COVID-19 Cases and Deaths in Federal and State Prisons, 324 J. Am. Med. Ass’n 602 (2020); Sharon Dolovich, Mass Incarceration, Meet COVID-19, U. Chi. L. Rev. Online (Nov. 16, 2020), https://ssrn.com/abstract=3766415 [https://perma.cc/Y7L6-NHZ6]. As Part III shows, the methods courts used to deny plaintiffs’ claims were not new: to find for the state, courts simply deployed the same mechanisms that had been used for years to deflect prisoners’ constitutional claims. At the same time, Covid exposed dynamics long present in prison cases, confirming that the federal judiciary, although providing constitutional relief in some marginal cases, in practice offers only the most minimal check,16×16. Dolovich, The Failed Regulation, supra note 7, at 164–68. constitutional or otherwise, on the abuse, neglect, and callous indifference that largely typify the administration of American prisons.17×17. Id. at 156–58, 160–64, 168–70. Covid, in short, definitively confirmed the terrible coherence of prison law.

That the Court has long been predisposed to favor some parties over others has been well documented, as has the general normative direction of the Court’s predilections in favor of the rich and powerful and against the poor and disenfranchised.18×18. See generally Erwin Chemerinsky, Presumed Guilty: How the Supreme Court Empowered the Police and Subverted Civil Rights (2021); Erwin Chemerinsky, Closing the Courthouse Door: How Your Constitutional Rights Became Unenforceable (2017); Erwin Chemerinsky, The Case Against the Supreme Court (2014); Adam Cohen, Supreme Inequality: The Supreme Court’s Fifty-Year Battle for a More Unjust America (2020). What I am mapping here is the way this troubling orientation manifests in prison law — and how the Court has managed to refashion its own evident sympathy for prison officials and hostility to the legal claims of incarcerated litigants into a governing ethos shaping judicial deliberations across the field.

I. Constructing the Doctrine: The Canons of Evasion

Prior to the 1960s, federal courts largely took a “hands-off” approach to prisoners’ constitutional claims.19×19. Margo Schlanger, The Constitutional Law of Incarceration, Reconfigured, 103 Cornell L. Rev. 357, 368 n.53 (2018) (tracing the lineage of the “‘hands-off’ phrase”). However draconian the challenged conditions, however extreme the abuse alleged, federal judges perceived themselves to lack authority over constit-utional claims arising from prison.20×20. See Malcolm M. Feeley & Edward L. Rubin, Judicial Policy Making and the Modern State: How the Courts Reformed America’s Prisons 30–34 (1998); Note, Beyond the Ken of the Courts: A Critique of Judicial Refusal to Review the Complaints of Convicts, 72 Yale L.J. 506, 506–07 (1963); see also United States ex rel. Atterbury v. Ragen, 237 F.2d 953, 954 (7th Cir. 1956) (finding that federal courts lacked jurisdiction to hear a case in which the plaintiff alleged having received “vicious beatings . . . at the hands of various defendants,” having been “placed in solitary confinement . . . for two months without clothes or blankets, . . . deprived of any food” for five days, and having been “denied mail including a copy of the constitution of Illinois sent to him at his request by the Secretary of State”). Then, in 1964, in Cooper v. Pate,21×21. 378 U.S. 546 (1964) (per curiam). the Supreme Court signaled a jurisdictional shift. Thomas X. Cooper, a Muslim, had “alleged that prison officials had blocked his access to religious services, ‘materials disseminated by the Black Muslim Movement,’ and the Koran.”22×22. Driver & Kaufman, supra note 1, at 529 (quoting Cooper v. Pate, 324 F.2d 165, 166 (7th Cir. 1963), rev’d, 378 U.S. 546 (1964) (per curiam)). The district court dismissed the case for failure to state a claim, but in a one-paragraph per curiam opinion, the Supreme Court reversed, finding that “the complaint stated a cause of action and [that] it was error to dismiss it.”23×23. Cooper, 378 U.S. at 546.

Cooper opened the courthouse door to incarcerated litigants. By the mid-1970s, the Court had decided a string of cases establishing federal court jurisdiction over a wide range of prisoners’ constitutional claims, including First Amendment speech24×24. See Procunier v. Martinez, 416 U.S. 396, 400 (1974), overruled in part by Thornburgh v. Abbott, 490 U.S. 401 (1989); Pell v. Procunier, 417 U.S. 817, 819 (1974). and association,25×25. See Jones v. N.C. Prisoners’ Lab. Union, Inc., 433 U.S. 119, 121 (1977). due process right of access to the courts,26×26. See Martinez, 416 U.S. at 419; Johnson v. Avery, 393 U.S. 483, 484 (1969). procedural due process,27×27. See Wolff v. McDonnell, 418 U.S. 539, 542–43 (1974). and Eighth Amendment medical neglect.28×28. See Estelle v. Gamble, 429 U.S. 97, 101 (1976). In these cases, plaintiffs did not always prevail. But the Court took for granted that the claims raised were among those the federal courts may properly consider and endorsed the courts’ role in enforcing the Constitution in prison.

Over the ensuing decades, the list of prison law claims the federal courts entertained as a matter of course expanded still further, coming to include asserted violations of the Fourth Amendment,29×29. See Hudson v. Palmer, 468 U.S. 517, 519 (1984). substantive due process,30×30. See McKune v. Lile, 536 U.S. 24, 37 (2002) (plurality opinion); Turner v. Safley, 482 U.S. 78, 83 (1987). and a range of Eighth Amendment claims, including excessive force,31×31. See Hudson v. McMillian, 503 U.S. 1, 4 (1992); Whitley v. Albers, 475 U.S. 312, 314 (1986). failure to protect,32×32. See Farmer v. Brennan, 511 U.S. 825, 829 (1994). and unconstitutional conditions.33×33. See Wilson v. Seiter, 501 U.S. 294, 296 (1991); Rhodes v. Chapman, 452 U.S. 337, 339 (1981). Yet even as the list of potential entitlements grew, the Court systematically increased the doctrinal burdens on incarcerated plaintiffs and repeatedly emphasized the deference courts owed prison officials, clearly signaling its unwillingness to provide more than minimal protection for prisoners.34×34. See Block v. Rutherford, 468 U.S. 576, 585 (1984); Bell v. Wolfish, 441 U.S. 520, 547 (1979); Jones v. N.C. Prisoners’ Lab. Union, Inc., 433 U.S. 119, 125 (1977); Pell v. Procunier, 417 U.S. 817, 827–28 (1974); Sharon Dolovich, Forms of Deference in Prison Law, 24 Fed. Sent’g Rep. 245, 246–49 (2012).

The Court’s retreat from the possibility of meaningful constitutional protections occurred over decades, accelerating most markedly during the Rehnquist Court. Over this period, in its prison law cases, the Court repeatedly established doctrinal standards with a striking effect: systematically allowing courts to maintain the appearance of meaningful judicial review — admitting evidence, hearing arguments, applying governing legal standards to the facts of the case, and so on — and yet readily finding for the defendants almost regardless of the facts before them.35×35. See Sharon Dolovich, Evading the Eighth Amendment: Prison Conditions and the Courts, in The Eighth Amendment and Its Future in a New Age of Punishment 133, 134 (Meghan J. Ryan & William W. Berry III eds., 2020) [hereinafter Dolovich, Evading the Eighth Amendment]; Dolovich, supra note 8, at 112–13. This effect is plainly evident in the key Eighth Amendment cases of Whitley v. Albers and Farmer v. Brennan,36×36. 511 U.S. 825 (1994). and also in Turner v. Safley, the case that governs the bulk of non–Eighth Amendment claims brought by prisoners. To be sure, some cases do not conform to this pattern. But these instances are rare and, as we will see, serve more to reinforce than to disprove the general point.

Take first the Eighth Amendment cases. There is no dispute that the Eighth Amendment — which prohibits the infliction of “cruel and unusual punishment”37×37. U.S. Const. amend. VIII. — obliges correctional officers (COs) to provide for prisoners’ basic needs. As the Court has repeatedly observed, under the Eighth Amendment, the state is duty-bound to provide people in custody with “the minimal civilized measure of life’s necessities,”38×38. Rhodes, 452 U.S. at 347. including physical safety.39×39. See Farmer, 511 U.S. at 833. Among the vital protections the Eighth Amendment thus accords the incarcerated is its proscription on excessive force.40×40. For an extended examination of the moral implications of COs’ use of force against prisoners, see Sharon Dolovich, Excessive Force in Prison (July 22, 2021) [hereinafter Dolovich, Excessive Force] (unpublished manuscript) (on file with the Harvard Law School Library). It is hard to overstate the urgency of this prohibition for those in custody. Without some meaningful external check, the hidden nature of prison life can invite COs to use violence against prisoners with impunity. And when COs exercise their power in this way, the pain, injury, and trauma inflicted on prisoners may be considerable.41×41. See, e.g., Whitley v. Albers, 475 U.S. 312 (1986) (COs shot plaintiff in the leg and left him bleeding for ten to fifteen minutes before attending to his wounds); Johnson v. McCowan, No. 20cv00582, 2021 WL 3027955, at *1 (W.D. Va. July 19, 2021) (COs unleashed a canine on plaintiff while he lay prone on the floor); Dawson v. Cook, 238 F. Supp. 3d 712, 714–15 (E.D. Pa. 2017) (defendant CO beat plaintiff on the head severely enough for plaintiff to lose vision in one eye); Banks v. Cnty. of Westchester, 168 F. Supp. 3d 682, 685 (S.D.N.Y. 2016) (defendant CO slammed plaintiff’s hand into a steel door, smashed his head on the floor, and kicked disinfectant into his eye).

Judicial review of Eighth Amendment excessive force claims is supposed to offer the requisite external check against such abuses. If it is to serve this purpose, a constitutional standard is required that would allow courts to fairly and independently assess whether force was warranted. In Whitley, however, the Court went in a different direction, holding that whether force exceeds Eighth Amendment limits “ultimately turns on ‘whether [it] was applied in a good faith effort to maintain or restore discipline or maliciously and sadistically for the very purpose of causing harm.’”42×42. Whitley, 475 U.S. at 320–21 (emphasis added) (quoting Johnson v. Glick, 481 F.2d 1028, 1033 (2d Cir. 1973)). In the universe of mens rea standards, it is hard to conceive of a higher one than this. Whitley’s “maliciously and sadistically” standard, moreover, is entirely subjective. After Whitley, the only question courts must answer when assessing the constitutionality of deliberate violence by COs against prisoners is whether the defendants themselves believed the force was warranted — a standard deemed met if the courts can identify any “plausible basis for the officials’ belief that [the] degree of force was necessary.”43×43. Id. at 323. If so, the force cannot have been used solely “for the . . . purpose of causing harm”44×44. Id. at 321 (quoting Johnson, 481 F.2d at 1033). and must therefore pass constitutional muster. In case this directive was not clear enough, Whitley was explicit that when the evidence suggests “a mere dispute over the reasonableness of a particular use of force or the existence of arguably superior alternatives. . . . [then] the case should not [even] go to the jury.”45×45. Id. at 322. In dissent, Justice Marshall took strong issue with the suggestion that no constitutional liability could lie even should defendants have had available to them “arguably superior alternatives.” Id. As Justice Marshall observed, “if prison officials were to drop a bomb on a cellblock in order to halt a fistfight between two inmates,” the Court would have no difficulty finding such action “sufficiently wanton to present a jury question.” Id. at 333 (Marshall, J., dissenting).

That this standard is intrinsically defendant friendly is undeniable. Perhaps most strikingly, it delegates to prison officials the power to set constitutional limits on the use of force in prison, although it is these very officers whose conduct the Eighth Amendment is supposed to constrain. To reach such a surprising and minimally protective outcome, the Court deployed three strategies of doctrinal construction I have elsewhere collectively termed the canons of evasion: the insistence on deference, a presumption of constitutionality, and the introduction of a substitute question.46×46. See Dolovich, supra note 8, at 112–13.

First, there is the inevitable call for deference. In almost all its prison law cases since 1974,47×47. The Court began hearing prison law cases in earnest in 1974. In that year alone, the Court decided Procunier v. Martinez, 416 U.S. 396 (1974), overruled in part by Thornburgh v. Abbott, 490 U.S. 401 (1989), Pell v. Procunier, 417 U.S. 817 (1974), Saxbe v. Washington Post Co., 417 U.S. 843 (1974), and Wolff v. McDonnell, 418 U.S. 539 (1974). the Court has emphasized the imperative of judicial deference to prison officials’ judgments, and Whitley is no exception. Writing for the Whitley majority, Justice O’Connor justified the “maliciously and sadistically”48×48. Whitley, 475 U.S. at 320 (quoting Johnson, 481 F.2d at 1033). standard by insisting that “[p]rison administrators . . . should be accorded wide-ranging deference in the adoption and execution of policies and practices that in their judgment are needed to preserve internal order and discipline and to maintain institutional security.”49×49. Id. at 321–22 (quoting Bell v. Wolfish, 441 U.S. 520, 547 (1979)). Justice O’Connor also emphasized that “when the ‘ever-present potential for violent confrontation and conflagration,’” id. at 321 — cue Driver and Kaufman’s “mythic prison,” Driver & Kaufman, supra note 1, at 512, which in Whitley becomes a tinderbox full of perpetually violent inhabitants unaccountably liable to explode at any moment — “ripens into actual unrest and conflict, the admonition that ‘a prison’s internal security is peculiarly a matter normally left to the discretion of prison administrators’ carries special weight,” Whitley, 475 U.S. at 321 (citations omitted) (quoting Jones v. N.C. Prisoners’ Lab. Union, Inc., 433 U.S. 119, 132 (1977); Rhodes v. Chapman, 452 U.S. 337, 349 n.14 (1981)). This degree of deference, Justice O’Connor insisted, “does not insulate from review actions taken in bad faith and for no legitimate purpose.”50×50. Whitley, 475 U.S. at 322. It does, however, “require[] that neither judge nor jury freely substitute their judgment for that of officials who have made a considered choice.”51×51. Id.

This highly deferential framing, in which the need for deference is explicitly invoked as a legal principle in its own right, reveals the second evasive maneuver employed by the Whitley Court: baking into the standard a presumption of constitutionality as to the defendants’ conduct. In foreclosing judicial scrutiny absent evidence that the defendant used force maliciously and sadistically, the Court effectively established an irrebuttable presumption that anything short of conduct evincing a “knowing willingness” to inflict unjustified harm is necessarily constitutional.52×52. See id. at 321 (directing courts to determine whether, on the evidence, “inferences may be drawn as to whether the use of force . . . evinced such wantonness with respect to the unjustified infliction of harm as is tantamount to a knowing willingness that it occur”). This move alone has the remarkable effect of shielding COs’ conduct from judicial scrutiny so long as they could have had some security-based justification for their conduct — even when that conduct involved the deliberate use of violence against prisoners and regardless of the degree of force used.

Whitley’s directive to courts to focus on the defendant’s subjective view of the matter produces the third evasive move employed in the case: the introduction of a substitute question that will allow courts to find for defendants without ever reaching what is arguably the real issue — whether COs’ actions were consistent with their constitutional obligations to those in their custody. Rather than considering when force in prison is constitutionally justified and crafting a standard that reflects constitutional limits, Whitley directs judges to assess whether any “plausible basis” existed for the defendants’ subjective view that “this degree of force was necessary.”53×53. Id. at 323. As might be expected, only in cases featuring the most extreme facts, on which no one could doubt the gratuitousness of the COs’ assault on the plaintiff,54×54. See, e.g., United States v. Walsh, 194 F.3d 37, 40 (2d Cir. 1999) (upholding defendant CO’s criminal conviction under 18 U.S.C. § 242 for multiple Eighth Amendment violations where the defendant, who weighed over 300 pounds, was found to have stepped on a prisoner’s penis, “‘mashing’ it as one might have extinguished a cigarette,” id. at 43). will courts generally decide this question in defendants’ favor.

One can see these same evasive canons at work in the 1994 case of Farmer v. Brennan. In Farmer, the Court established the operative standard for all Eighth Amendment prison conditions claims55×55. See Dolovich, Evading the Eighth Amendment, supra note 35, at 149–52; Sharon Dolovich, Cruelty, Prison Conditions, and the Eighth Amendment, 84 N.Y.U. L. Rev. 881, 943–48 (2009). For a detailed exegesis of the canons of evasion as they operate in Farmer v. Brennan, 511 U.S. 825 (1994), see Dolovich, supra note 8, at 130–35. except excessive force.56×56. Farmer’s deliberate indifference standard applies to claims of medical neglect, the deprivation of basic human needs, and the failure to protect people from physical or sexual assault by fellow prisoners. See Farmer, 511 U.S. at 825, 835, 837–38; see also id. at 837 (failure to protect); Wilson v. Seiter, 501 U.S. 294, 303 (1991) (deprivation of basic human needs); Estelle v. Gamble, 429 U.S. 97, 104 (1976) (medical neglect). In previous cases, the Court had held that plaintiffs bringing such claims must show that defendants were “deliberately indifferent” to the dangers prison conditions posed.57×57. See Wilson, 501 U.S. at 303; Estelle, 429 U.S. at 104. It had, however, left open the question of precisely what this showing requires. Farmer resolved the matter, defining Eighth Amendment deliberate indifference as the equivalent of criminal recklessness.58×58. Farmer, 511 U.S. at 837. As I have shown in detail elsewhere,59×59. See Dolovich, supra note 8, at 130–35. with this holding, the Court foreclosed constitutional liability for prison conditions absent evidence that defendants actually realized the risk of harm plaintiffs faced — even if the danger was substantial and glaringly obvious, even if a reasonable person in the defendant’s situation would have recognized the urgent need to act, and even if the defendant herself would have noticed the danger if only she had been paying proper attention. Thanks to Farmer’s highly deferential standard, the (substitute) question courts must ask when prisoners challenge their conditions of confinement is not whether COs failed to take adequate steps to protect plaintiffs from serious risks to their health and safety, but instead whether defendants personally realized the risk. Absent such a finding, even the most harmful prison conditions are presumed constitutional.

In sum, by deploying the canons of evasion in Whitley and Farmer, the Court established heavily pro-defendant standards for virtually all constitutional claims implicating prisoners’ health and safety. Courts apply these standards as if they provide meaningful protection against unconstitutional conditions. But in practice, barring extreme abuse or glaring neglect, defendants will generally prevail, although it seems plain that such a defendant-friendly regime will almost certainly lead to increased physical pain and suffering inflicted by the very officers duty-bound to keep prisoners safe. The Court’s failure even to consider this possibility well illustrates its overall blindness to the impact of its decisions on the human beings held in the facilities whose conditions the Court is so reluctant to scrutinize.60×60. See Dolovich, Evading the Eighth Amendment, supra note 35, at 137 (tracing the way the Court’s Eighth Amendment doctrine operates to shift judicial attention “away from the conditions themselves and toward what defendants did or did not know about the risk of harm to prisoners”).

What of prison law claims outside the Eighth Amendment? The incarcerated retain a wide range of fundamental constitutional liberties, violation of which entitles them to petition for judicial redress. In this broad sphere, as Driver and Kaufman explain, there is one case that with very few exceptions61×61. See infra note 88. has come to set the standard for all non–Eighth Amendment constitutional claims brought by incarcerated plaintiffs: Turner v. Safley.62×62. See Driver & Kaufman, supra note 1, at 523. It is hard to conceive of a more deferential standard than Turner, or one that creates a stronger presumption of constitutionality. Indeed, in crafting the Turner standard, the Court made such effective use of the evasive strategies of deference and presumption that the standard itself, although superficially recognizable as a species of rational basis review, effectively functions as its own substitute question. That is, simply by applying Turner, courts are able to find for defendants without needing to look too closely at either the facts of the case or the strength of the plaintiff’s arguments.

Turner dealt with two regulations in effect in a Missouri prison: a ban on correspondence between incarcerated people and a rule prohibiting prisoners from marrying without the warden’s permission.63×63. See id. at 535–36 (discussing the factual background of Turner v. Safley, 482 U.S. 78 (1987)). The question before the Court was the standard of review under which courts should decide constitutional challenges to regulations of this sort. For people outside of prison, both speech and marriage are considered fundamental constitutional rights. The unique and challenging nature of the prison environment may in some cases warrant greater scope for regulations burdening such rights than in society at large. Yet, given that in the prison context, these fundamental rights could in many instances be vindicated to a considerable degree without compromising significant state interests, one might readily imagine a standard of review that put some appreciable burden on prison officials to justify regulations hindering their exercise. But in Turner, the Court opted for a different approach, holding that prison regulations that undermine prisoners’ constitutional rights will nonetheless be upheld if they are “reasonably related to legitimate penological interests.”64×64. Turner, 482 U.S. at 89. This discussion of Turner is drawn in part from Dolovich, supra note 34, at 246.

The four factors the Turner Court specified for determining whether the state has met this burden betray a strikingly pro-state slant for a standard that, it bears repeating, is intended for use in determining whether state officials have violated fundamental rights. In such cases, per Turner, courts must ask (1) whether there is a “‘valid, rational connection’ between the prison regulation and the legitimate governmental interest put forward to justify it”;65×65. Turner, 482 U.S. at 89 (quoting Block v. Rutherford, 468 U.S. 576, 586 (1984)). (2) whether there are “alternative means of exercising the right that remain open to prison inmates”;66×66. Id. at 90. (3) what “impact accommodation of the asserted constitutional right will have on guards and other inmates, and on the allocation of prison resources generally”;67×67. Id. and (4) whether there are “ready alternatives”68×68. Id. by which prison officials can realize their interests while also affording prisoners the exercise of their rights.69×69. Id.; see id. at 90–91. Turner’s elaboration of each factor leaves no doubt that the test is intended to be extremely deferential to prison officials. The Court explained (1) that “a regulation cannot be sustained where the logical connection between the regulation and the asserted goal is so remote as to render the policy arbitrary or irrational”;70×70. Id. at 89–90. (2) that “[w]here ‘other avenues’ remain available for the exercise of the asserted right, courts should be particularly conscious of the ‘measure of judicial deference owed to corrections officials’”;71×71. Id. at 90 (citation omitted) (quoting Jones v. N.C. Prisoners’ Lab. Union, Inc., 433 U.S. 119, 131 (1977); Pell v. Procunier, 417 U.S. 817, 827 (1974)). (3) that “[i]n the necessarily closed environment of the correctional institution, few changes will have no ramifications on the liberty of others or on the use of the prison’s limited resources”;72×72. Id. and (4) that if a “claimant can point to an alternative that fully accommodates the prisoner’s rights at de minimis cost to valid penological interests, a court may consider that as evidence that the regulation does not satisfy the reasonable relationship standard.”73×73. Id. at 91 (emphasis added). In short, having explicitly directed lower courts to be deferential in assessing alternatives74×74. In Overton v. Bazzetta, 539 U.S. 126 (2003), the Court emphasized that, to satisfy Turner’s second factor, “[a]lternatives . . . need not be ideal, . . . they need only be available.” Id. at 135. (factor two) and having stipulated that any change to a prison regime will necessarily have ramifications for the institution (factor three), the Turner Court made clear that, unless the challenged policy is found to be an “arbitrary or irrational”75×75. Turner, 482 U.S. at 90. method for the state to achieve its stated goals (factor one) and claimants can identify an alternative means to fully accommodate their rights without any appreciable cost to the prison (factor four), the challenged regulation is to be presumed constitutional. Sure enough, it is a rare case decided under Turner in which the plaintiff ultimately prevails.

Turner, with its pro-state bent, has migrated well beyond cases involving First Amendment speech rights and the Fourteenth Amendment right to marry to become, as Driver and Kaufman put it, the “default standard for reviewing constitutional challenges to prison policy.”76×76. Driver & Kaufman, supra note 1, at 536. Since Turner was decided, the Court has applied this standard to cases involving First Amendment expression,77×77. See, e.g., Beard v. Banks, 548 U.S. 521, 528 (2006) (plurality opinion); Thornburgh v. Abbott, 490 U.S. 401, 414 (1989). association,78×78. See, e.g., Overton, 539 U.S. at 131–32. and free exercise,79×79. See, e.g., O’Lone v. Estate of Shabazz, 482 U.S. 342, 349 (1987), superseded by statute, Religious Freedom Restoration Act of 1993 (RFRA), Pub. L. No. 103-141, 107 Stat. 1488 (codified as amended at 42 U.S.C. §§ 2000b to 2000bb-4). the Fifth Amendment right against self-incrimination,80×80. See, e.g., McKune v. Lile, 536 U.S. 24, 37 (2002) (plurality opinion). the Fourteenth Amendment right against being involuntarily medicated,81×81. See, e.g., Washington v. Harper, 494 U.S. 210, 223 (1990). and even the due process right of access to the courts.82×82. See Lewis v. Casey, 518 U.S. 343, 361–63 (1996); see also id. at 367 (Thomas, J., concurring). The impact of Turner on the scope of prisoners’ constitutional claims cannot be overstated; according to David Shapiro, as of 2016 Turner had been cited in judicial decisions more than 8000 times.83×83. See David M. Shapiro, Lenient in Theory, Dumb in Fact: Prison, Speech, and Scrutiny, 84 Geo. Wash. L. Rev. 972, 975 (2016).

The application of Turner should not automatically foreclose plaintiffs’ success. To be sure, the plaintiffs’ burden is extremely high. Yet, as Shapiro has documented, there is seemingly no end to the prison policies burdening prisoners’ constitutional rights, and many such policies are hard to justify on any plausible account of the state’s legitimate interests.84×84. For a detailed set of examples, including my personal favorite — the removal of a map of the planets from the wall of a New York state prison library pursuant to a policy prohibiting all maps in case “they may prove useful to prisoners who manage to escape,” id. at 997 — see id. at 988–1005. When the state has no good reason for compromising such core constitutional rights as freedom of speech and expression, religious freedom, or due process, courts should as a matter of course strike down the offending policy or practice and restore the plaintiffs’ access to the enjoyment of these basic constitutional entitlements. In the prison context, however, honoring this imperative would require courts to recognize and affirm the value in enabling people in custody to exercise the civil liberties that the Bill of Rights is designed to protect — the right to read, debate, worship, interact with others, and even play games85×85. See Singer v. Raemisch, 593 F.3d 529 (7th Cir. 2010) (upholding Wisconsin prison’s ban on the playing of Dungeons and Dragons). For more on Singer v. Raemisch, 593 F.3d 529, see infra section II.C.2, p. 327. as one chooses without undue state interference. Instead, under Turner, courts hearing cases implicating these basic freedoms may — and often do — side with defendant prison officials on the flimsiest of grounds, however compelling the plaintiffs’ evidence and logically sound their arguments. In the vast majority of such cases, the personal toll this judicial posture inflicts on affected prisoners scarcely registers.

Turner’s deployment of the canons of evasion, and especially its extreme deference, facilitates such outcomes. Still, the fact that defendants so often win on Turner even when challenged regulations are seemingly impossible to defend on the merits suggests that defendants are benefiting from something more than favorable doctrinal standards.86×86. See infra section II.B, pp. 319–25. As Part II will shortly show, that something more is the morally dissonant orientation that the federal courts have come to adopt toward the parties in prison law cases.87×87. See infra Part II, pp. 316–32 (excavating the dispositional favoritism pervasive in prison law cases).

Not every legal claim raised by incarcerated plaintiffs is covered by Turner or the Eighth Amendment.88×88. See, e.g., Johnson v. California, 543 U.S. 499, 509 (2005) (holding that race-based equal protection claims brought from prison are subject not to Turner, but to strict scrutiny); Wilkinson v. Austin, 545 U.S. 209, 224 (2005) (applying procedural due process principles to transfers into solitary confinement and finding such principles afford some minimal procedural due process protections); Wolff v. McDonnell, 418 U.S. 539, 558 (1974) (same with respect to disciplinary hearings). Under the Religious Land Use and Institutionalized Persons Act of 2000 (RLUIPA), 42 U.S.C. §§ 2000cc to 2000cc-5, state prison policies burdening religious practice are subject to strict scrutiny; RFRA provides the same degree of statutory protection to the free exercise rights of federal prisoners. Procedural claims arising under the Prison Litigation Reform Act (PLRA), Pub. L. No. 104-132, 110 Stat. 1214 (1996) (codified as amended in scattered sections of the U.S. Code), have generated their own governing doctrine. See, e.g., Lomax v. Ortiz-Marquez, 140 S. Ct. 1721, 1723 (2020); Ross v. Blake, 578 U.S. 632, 635 (2016); Jones v. Bock, 549 U.S. 199, 202–03 (2007); Woodford v. Ngo, 548 U.S. 81, 93 (2006); Booth v. Churner, 532 U.S. 731, 734 (2001). And in some instances, the Court’s decisions run decidedly in plaintiffs’ favor. Here, two cases are especially notable. In Johnson v. California,89×89. 543 U.S. 499 (2005). the Court held that prison policies that facially discriminate based on race must satisfy, not Turner (for which defendants had advocated), but strict scrutiny.90×90. Id. at 513 (“Turner is too lenient a standard to ferret out invidious uses of race.”). And

in Brown v. Plata,91×91. 563 U.S. 493 (2011). a five–four opinion written by Justice Kennedy, the Court upheld a Ninth Circuit three-judge panel order directing the California Department of Corrections and Rehabilitation (CDCR) to reduce the population density of its facilities to 137.5% of rated capacity.92×92. Id. at 539. This decision, which turned on the three-judge panel’s application of a key provision of the Prison Litigation Reform Act93×93. Pub. L. No. 104-132, 110 Stat. 1214 (1996) (codified as amended in scattered sections of the U.S. Code). (PLRA),94×94. See Plata, 563 U.S. at 502. obliged the CDCR to release or find alternative housing for 46,000 people then in California state prison.95×95. See id. at 501; see also Margo Schlanger, Plata v. Brown and Realignment: Jails, Prisons, Courts, and Politics, 48 Harv. C.R.-C.L. L. Rev. 165, 183 (2013); Robert Weisberg & Joan Petersilia, The Dangers of Pyrrhic Victories Against Mass Incarceration, Daedalus, Summer 2010, at 124, 125.

But these cases were each exceptional in their own way. Johnson involved self-conscious race discrimination by government officials, thus giving those Justices concerned that the Court take a position opposed to state-sponsored racism a strong reason in that instance to side with the plaintiffs and against prison officials. Johnson therefore is best understood independently of the dynamics that more typically shape prison law cases. Plata, meanwhile, represents a different kind of exception: a highly consequential case, directly implicating the scope of state authority to run the prisons free of judicial interference, in which the Court, resisting resort to the canons of evasion, weighed the evidence impartially and put the state to its proof.96×96. See Dolovich, supra note 34, at 250–52. Plata offers a glimpse of what prison law could be were the Court not inclined to tilt the playing field so strongly in defendants’ favor — that is, a site of good faith legal analysis that takes seriously all the evidence and arguments in the record without automatically overcrediting defendants’ submissions or discounting those of plaintiffs.97×97. For another (rare) example of such evenhandedness on the part of the Supreme Court, see Cleavinger v. Saxner, 474 U.S. 193 (1985), in which the Court rebuffed efforts by defendant prison officials to assert absolute immunity for actions taken in their capacity as hearing officers in prison disciplinary matters, in part because, contrary to defendants’ representations, COs serving as hearing officers feel pressure to credit the testimony of fellow COs over that of prisoners, thus compromising the impartiality necessary for a fair process, id. at 204; see also Dolovich, supra note 34, at 252–53, 252 n.134.

Yet Plata is also exceptional in another sense: the extreme nature of the suffering and injury produced by carceral conditions so plainly unconstitutional that the violation — in this case, grossly inadequate medical and mental health care in a chronically overcrowded prison system — was in large part stipulated to by the defendants themselves.98×98. See Plata, 543 U.S. at 532. The case also dragged on for decades without implementation of an adequate remedy.99×99. In the majority opinion in Plata, Justice Kennedy rehearsed in detail some of the many disturbing facts of the case, including that, due to “a shortage of treatment beds, suicidal inmates may be held for prolonged periods in telephone-booth sized cages without toilets,” id. at 503; that one such individual, having “been held in such a cage for nearly 24 hours,” was found “standing in a pool of his own urine, unresponsive and nearly catatonic” because prison officials had “no place to put him,” id. at 504; and that, due to lack of space, “up to 50 sick inmates may be held together in a 12- by 20-foot cage for up to five hours awaiting treatment,” id. Justice Kennedy also found that, although the three-judge panel’s order to reduce the prison population was “of unprecedented sweep and extent,” the “medical and mental health care provided by California’s prisons” had “[f]or years . . . fallen short of minimum constitutional requirements” and that “[e]fforts to remedy the violation ha[d] been frustrated by severe overcrowding.” Id. at 501. Plata thus serves to mark the outer limits of prison law’s pro-state tilt: where the challenged treatment is glaringly indefensible, where dedicated and adept plaintiffs’ lawyers are willing to build the strongest possible case for their clients,100×100. In Plata — a consolidation of Plata and Coleman — class members were represented by Don Specter and his colleagues at the Prison Law Office (Plata), and Michael Bien, Ernest Galvan, Jane Kahn, and their colleagues at what was then Rosen Bien & Galvan LLP (Coleman), making the plaintiffs’ advocates among the most experienced and expert in the field. and where courts are open to taking plaintiffs’ claims seriously, prisoners may sometimes prevail.101×101. As to this last factor, Justice Kennedy’s late-career recognition of the failure of American carceral policy and the moral wrong it represents was plainly a key determinate of the way Plata, decided five–four, came out. See Anthony M. Kennedy, Assoc. Just., Sup. Ct. of the U.S., Speech at the American Bar Association Annual Meeting (Aug. 14, 2003), https://www.supremecourt.gov/publicinfo/speeches/sp_08-09-03.html [https://perma.cc/3DNZ-ARXC]. But Plata is an outlier. Ordinarily, it is the pro-state skew, evident in the Court’s opinions regardless of the merits of the cases, that is normative for this context.

II. Deciding the Cases: Dispositional Favoritism

A. Manifestations of Judicial Deference

As was noted above, starting in 1974, the Supreme Court began to expand the list of constitutional claims available to people in prison.102×102. See supra p. 306. Yet even while the Court was taking steps to open up the federal courts to the incarcerated, it insisted in almost every case on the need for judicial deference to prison officials.103×103. Dolovich, supra note 34, at 253.

As Part I has shown, the Court’s commitment to this deferential posture helped shape the substantive standards governing the vast majority of prisoners’ constitutional claims. There are, however, other ways the Court’s determination to defer manifests in the cases. Two additional deferential moves, once flagged, become easy to spot: the framing of relevant facts in ways favorable to the state, and the altering of procedural rules to the same end.104×104. Id. at 246–49. In this Part, I show the way these moves have played out in the Court’s prison law opinions. But these additional forms of deference must be understood as operating in concert with a further, more subtle manifestation of the Court’s normative commitments — one that, although less tangible and thus harder to pin down, is as consequential in shaping outcomes as the evasive moves described in Part I. I am speaking here of the general normative orientation with which, in its prison law cases, the Court approaches the parties’ submissions and even the parties themselves — an orientation that can best be described as a readiness to look upon prison officials and their evidence and arguments with favor and sympathy, while regarding incarcerated litigants and their evidence and arguments with skepticism and even hostility.

Reframing facts and altering procedural rules is only part of it. This orientation — call it dispositional favoritism — can also produce judicial reasoning that, among other things, automatically presumes good faith and expertise on the part of defendant prison officials, views prisoners in general with suspicion, and scarcely considers the real-life impact of case outcomes on the actual human beings who live behind bars. This normative posture generates a marked judicial elevation of defendants’ experience, perspectives, and interests, and a systematic devaluation of the plaintiffs’ experience, perspectives, and interests, not to mention their rights to basic liberties and a safe environment. The effect of such dispositional favoritism is a host of subtle findings and adjustments in whatever case is at hand, virtually all tilting in defendants’ favor. Unsurprisingly, when the Court adopts this posture, defendants generally win.

Space does not permit me to demonstrate the full reach of dispositional favoritism in prison law; the case law is too vast and varied to allow for a comprehensive analysis. My goals in this Part are more modest: to show how the Supreme Court has, through the force of its example, rendered dispositional favoritism broadly normative for prison law cases (section II.B), and to identify instances of dispositional favoritism in action in the lower federal courts (section II.C). In terms of federal court opinions, I am especially interested in those cases in which plaintiffs’ claims seem impossible to dispute and yet the court nonetheless sides with the state. These cases, with their puzzling logic and signs of flat-out judicial refusal to credit evidence that favors the plaintiffs, most effectively illustrate both the normativity of dispositional favoritism for prison law and the extent of the power it confers on prison officials in their interactions with prisoners. That such opinions can be issued at all — with (as it were) a judicial straight face — is proof enough that the usual judicial norms of impartiality and critical analysis do not govern here. Additional proof of the normativity of this inversion is found in the fact that even cases of this order are not generally greeted by incarcerated litigants and their advocates with shock or confusion. Instead, they tend merely to be met with resignation and a sense of frustrated disappointment that the court in question chose the path of least resistance rather than taking plaintiffs’ claims seriously and putting the state to its proof.

One hallmark of dispositional favoritism turns out to be the absence of the usual written indicators that a court has scrutinized and assessed the evidence presented. In the ordinary course, the unspoken expectation is that judges will approach each case with an open mind, prepared to hear and assess the arguments on the merits and deal with all litigants fairly and with respect. Such evenhandedness would not require uncritical acceptance of parties’ claims; on the contrary, judges are expected to critically examine all legal submissions and to assess the quality of the arguments. Signs of such reasoning are regularly sprinkled throughout judicial opinions, with courts observing that one party or other has “failed to show that” or “failed to meet its burden to” or “offered no sound reason why” or the like. Typically, phrases of this sort are so commonplace as to be largely invisible, indicating only that the court has fulfilled its baseline obligation of analyzing the evidence in light of extant legal standards. Yet in prison law cases, the absence of such indicia of judicial scrutiny — at least as directed toward the defendants’ evidence and arguments — is entirely normative, such that opinions finding for defendants with only cursory or conclusory analyses are unsurprising. Rather, it is the presence of such otherwise ordinary locutions that, when directed against defendant prison officials, seems to call out for explanation.105×105. One does occasionally see the Supreme Court deploying such indicia of judicial scrutiny in favor of incarcerated litigants. See, e.g., Holt v. Hobbs, 574 U.S. 352, 356, 365, 367 (2015) (“[T]he Department has failed to show,” id. at 365; “[t]he Department failed to establish,” id.; “[i]t has offered no sound reason why,” id. at 367; “the Department failed to show, in the face of petitioner’s evidence,” id. at 367); Turner v. Safley, 482 U.S. 78, 98 (1987) (“We are aware of no place in the record where prison officials testified that”; “petitioners have pointed to nothing in the record suggesting”; “[c]ommon sense likewise suggests”); Procunier v. Martinez, 416 U.S. 396, 415–16 (1974) (“Appellants have failed to show,” id. at 415; “[a]ppellants contend that . . . . [b]ut they do not suggest how . . . nor do they specify what,” id. at 416), overruled in part by Thornburgh v. Abbott, 490 U.S. 401 (1989). But each of these instances is unique in a way that helps explain the Court’s willingness to put the state to its proof. In Martinez, the first case in which the Supreme Court entertained a First Amendment freedom of expression claim brought by prisoners, the Court focused primarily on the First Amendment rights of non-incarcerated people wishing to correspond with people in custody. See Martinez, 416 U.S at 409 (“The wife of a prison inmate who is not permitted to read all that her husband wanted to say to her has suffered an abridgment of her interest in communicating with him as plain as that which results from censorship of her letter to him.”). In Turner, after establishing a strikingly defendant-friendly standard of review for prisoners’ constitutional claims, see supra pp. 311–14, the Court upheld without difficulty a regulation prohibiting people in Missouri prisons from writing to one another. But the case also considered a prohibition on prisoners’ marrying without the warden’s permission, and Justice O’Connor, who wrote for the majority, seemed to take particular umbrage at the gendered paternalism of a rule that had only ever been applied to women. Her analysis of that regulation under the new Turner standard reflected a consequent unwillingness simply to defer to prison officials’ characterization of the matter. As for Holt v. Hobbs, 574 U.S. 352 (2015), I discuss that case in some detail below. See infra section II.C.3, pp. 327–28. The matter to be resolved in such instances is why the Supreme Court (or an appellate or district court, as the case may be) here opted to fulfill the standard judicial obligation to critically evaluate the quality of defendants’ evidence.

B. Dispositional Favoritism in the Supreme Court

In this section, I map the way the Supreme Court models the dispositional favoritism with which federal courts approach prison law cases. The Court’s own disposition to favor prison officials will often go beyond the crafting of defendant-friendly doctrinal standards.106×106. See supra Part I, pp. 305–16. One manifestation of this inclination is emphasized by Driver and Kaufman: the “selective empiricism” through which the Court commits to factual claims about prisons and prison life that may directly conflict with claims made in other of its prison cases.107×107. See Driver & Kaufman, supra note 1, at 567. The impression created is that of a judicial body determined to subscribe to whatever factual account will best support defendants’ characterization of the issues.108×108. See id. at 568 (observing that, in the Court’s prison law opinions, “claims about penal institutions tend to shift in ways that benefit the government”). But inter-case contradiction is only one of several ways the Court’s approach to facts in its prison law cases — whether the facts on the record or those concerning the world in general — reflects its determination to favor the state. In Rhodes v. Chapman,109×109. 452 U.S. 337 (1981). for example,110×110. The examples in this paragraph are drawn from Dolovich, supra note 34, at 248–49. the majority opinion set a standard for determining when prison overcrowding constitutes an Eighth Amendment violation — and then held that the plaintiffs in that case did not meet this standard, while failing even to acknowledge the considerable evidence in the record suggesting otherwise.111×111. In Rhodes, the Court rejected an Eighth Amendment challenge to the practice of double-celling (that is, housing two men in sixty-three-square-foot cells originally intended for one person) on the ground that the double-celling did not “create other conditions intolerable for prison confinement.” Rhodes, 452 U.S. at 348. Yet the “conclusion of every expert who testified at trial” was that “a long-term inmate must have to himself” a minimum of fifty square feet of floor space “in order to avoid serious mental, emotional, and physical deterioration,” id. at 371 (Marshall, J., dissenting); see also id. at 371 n.4 (listing studies reaching the same conclusion). And after accounting for the bed alone, even “without making allowance for any other furniture in the room” (toilet, sink, locker, shelves, and so forth), the remaining square footage per person was approximately “20–24 square feet, an area about the size of a typical door,” id. at 371 n.3. In Lewis v. Casey,112×112. 518 U.S. 343 (1996). which dramatically narrowed the scope of prisoners’ due process right of access to the courts, the Court implicitly endorsed an account of what it takes to adequately litigate a legal claim that is wholly at odds with what any minimally competent lawyer (much less a Supreme Court Justice) knows to be true.113×113. In the controlling opinion in Casey, Justice Scalia maintained that prisoners’ right of access to the courts entails only the right to “bring to court a grievance that the inmate wished to present.” Id. at 354. Yet, as the Court had recognized in the prior case of Bounds v. Smith, 430 U.S. 817 (1977), “[i]t would verge on incompetence for a lawyer to file an initial pleading without researching such issues as jurisdiction, venue, standing, exhaustion of remedies, proper parties, [and so forth] . . . .” Id. at 825. Moreover, as Justice Scalia was surely well aware, the state’s response to any pro se pleading “will undoubtedly contain seemingly authoritative citations,” and “[w]ithout a library, [an incarcerated litigant] will be unable to rebut the State’s argument,” rendering the due process right of access to the courts largely empty. Id. at 826. Then there is Whitley, in which the procedural posture — an appeal from a directed verdict — required the Court to take the facts in the light most favorable to the plaintiff, Gerald Albers.114×114. See Whitley v. Albers, 475 U.S. 312, 322 (1986). Instead, in the majority opinion, Justice O’Connor adopted the state’s rendition of the facts — even though, as Justice Marshall pointed out in dissent, the record showed that Albers “bitterly disputed” the state’s characterization.115×115. Id. at 331 (Marshall, J., dissenting). Writing for the Whitley majority, Justice O’Connor emphasized that, at the time of the shooting, a riot was in progress, “a guard was still held hostage,” id. at 322 (majority opinion), and “[t]he situation remained dangerous and volatile,” id. at 323. Yet as Justice Marshall noted in dissent, Albers had presented “substantial testimony” at trial to show that by the time the officer shot him, “the disturbance had subsided.” Id. at 330 (Marshall, J., dissenting); see also id. at 331 (“Although the Court sees fit to emphasize repeatedly ‘the risks to the life of the hostage and the safety of inmates . . .,’ I can only point out that respondent bitterly disputed that any such risk to guards or inmates had persisted. The Court just does not believe his story.” (quoting id. at 323 (majority opinion))). It is no revelation that, in judicial hands, facts may be malleable. The point here is that whatever factual frame the Court adopts in its prison law cases always seems, as Driver and Kaufman put it, “to shift in ways that benefit the government.”116×116. Driver & Kaufman, supra note 1, at 567.

The Court’s disregard of Whitley’s procedural posture indicates yet another way the Court’s inclination to advantage prison officials manifests in this arena: through the recasting of procedural rules in ways that benefit defendants at plaintiffs’ expense. One sees this move in Beard v. Banks,117×117. 548 U.S. 521 (2006) (plurality opinion). in which the Court seemed to rewrite the rules of summary judgment to make it easier in prison cases for defendants to prevail.118×118. This discussion of Banks is drawn in part from Dolovich, supra note 34, at 247–48. For yet another example of the Court altering procedural rules in prison law cases to benefit defendants at plaintiffs’ expense, see id. at 246–47 (discussing Jones v. N.C. Prisoners’ Lab. Union, Inc., 433 U.S. 119 (1977)). Ordinarily, to succeed on summary judgment, defendants bear the burden of demonstrating that no genuine issue of material fact remains to be decided. If instead the plaintiffs’ evidence would allow a jury to decide in their favor, it would be inappropriate to deprive plaintiffs of the chance to make their case. In Banks, the facts strongly suggested the existence of open factual issues for trial. The case involved a ban on the possession of newspapers, magazines, and personal photo-graphs by people housed in level two of the Long Term Segregation Unit (LTSU-2), the most restrictive housing unit in Pennsylvania’s prisons.119×119. Banks, 548 U.S. at 524–25. The plaintiffs brought a First Amendment challenge, and the defendants sought summary judgment.120×120. Id. at 527. Under Turner, the first factor is paramount: unless plaintiffs can show that no valid rational connection existed between the challenged regulation and the state’s “asserted goal,” the court will generally find for defendants.121×121. See, e.g., Overton v. Bazzetta, 539 U.S. 126, 133–36 (2003) (focusing on the first Turner factor for each challenged regulation, with only glancing reference to the remaining three factors). Thus, in cases governed by Turner, the primary question the court must decide at summary judgment is whether a genuine issue of material fact remains as to the existence of such a connection.

In Banks, there seemed to be more than sufficient evidence to allow the plaintiffs to make this showing. The state had justified the challenged restrictions by the need to “incentiv[ize]” LTSU-2 residents “to move to”122×122. Banks, 548 U.S. at 531. LTSU-1 was the next-most restrictive unit, to which people in LTSU-2 stepped down en route to being released to the prison’s general population. See id. at 526. level one of the Long Term Segregation Unit (LTSU-1) and to minimize the amount of material in their cells, which might otherwise serve to hide contraband, be fashioned into weapons, catapult human waste, or be used “as tinder for cell fires.”123×123. Id. Yet as the record showed, LTSU-1 residents already had many other privileges LTSU-2 residents lacked, thus providing plenty of incentive for good behavior beyond the few additional items plaintiffs were requesting.124×124. See id. at 549 (Stevens, J., dissenting) (“Although conditions in LTSU-1 are also harsh” in several respects unrelated to the litigation, “they are far more appealing than the conditions in LTSU-2.”). Moreover, at least one of the items plaintiffs sought — photographs — was also prohibited in LTSU-1, so that a desire for photographs could not possibly incentivize LTSU-2 residents to try to get to LTSU-1.125×125. See id. (noting that although those “who ‘graduate’ out of the LTSU-1 and back into the general prison population do regain their right to possess personal photographs, . . . they also regain so many additional privileges . . . that it strains credulity to believe that the possibility of regaining the right to possess personal photographs if they eventually return to the general prison population would have any marginal effect on the actions of prisoners in LTSU-2”). And equally telling, LTSU-2 residents were already entitled to have many items in their cells that could in theory create the sorts of problems that had supposedly motivated the ban.126×126. As Justice Stevens further noted in dissent:

[E]ach [LTSU-2] inmate is given a jumpsuit, a blanket, two bedsheets, a pillow case, a roll of toilet paper, a copy of a prison handbook, ten sheets of writing paper, several envelopes, carbon paper, three pairs of socks, three undershorts and three undershirts, and may at any point also have religious newspapers, legal periodicals, a prison library book, Bibles, and a lunch tray with a plate and a cup. Many of these items are flammable, could be used [to start fires, catapult feces, or to create other dangers] as effectively as a newspaper, magazine or photograph, and have been so used by [LTSU-2] inmates.
Id. at 543–44 (alteration in original) (quoting Banks v. Beard, 399 F.3d 134, 143 (3d Cir. 2005)). This last fact in particular seemed to make refusal of plaintiffs’ request to have “one newspaper or magazine and some small number of photographs in their cells at one time” both unaccountable and arbitrary.127×127. Id. at 544 (quoting Banks, 399 F.3d. at 144).

On these grounds, the Third Circuit sided with the plaintiffs, but the Supreme Court reversed. Justice Breyer, writing for the plurality, acknowledged that on review of summary judgment, courts must ordinarily draw “all justifiable inferences” in favor of the nonmoving party (in this case, plaintiffs Banks et al.).128×128. Id. at 529 (plurality opinion) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986)). However, in cases brought by prisoners, a distinction must be drawn between “evidence of disputed facts and disputed matters of professional judgment.”129×129. Id. at 530. That is, when defendants’ proffered justifications represent their professional judgment, courts “must accord deference to the views of prison authorities.”130×130. Id. (citing Overton v. Bazzetta, 539 U.S. 126, 132 (2003)). And here, Justice Breyer found, the Third Circuit had “placed too high an evidentiary burden” on the defendant and accorded “too little deference to the judgment of prison officials” — for example, by “offer[ing] no apparent deference to the deputy prison superintendent’s professional judgment that the Policy deprived ‘particularly difficult’ [prisoners] of a last remaining privilege and that doing so created a significant behavioral incentive.”131×131. Id. at 535.

Assessing the record, the Third Circuit had found no evidence that the state’s “deprivation theory of behavior modification had any basis in real human psychology, or had proven effective with LTSU [residents].”132×132. Id. (quoting Banks, 399 F.3d at 142). Yet this lack of evidentiary support did not appear to trouble the Supreme Court, nor did the fact that the case plainly offered several other triable issues of material fact. Indeed, the Banks Court seemed simply to ignore any evidence tending to support Banks’s position and to elevate any assertions by the defendant, however unsupported by the facts, to the status of truth. As Justice Ginsburg succinctly put it in her dissent, the Court “effectively [told] prison officials they will succeed in cases of this order, and swiftly, while barely trying. It suffices for them to say, in our professional judgment the restriction is warranted.”133×133. Id. at 556 (Ginsburg, J., dissenting). After Banks, as Lisa Kerr aptly puts it, even on summary judgment, “[s]o long as the subject matter of a case concerns the judgment of prison administrators, then in almost no circumstance will the prisoner succeed.” Lisa Kerr, Contesting Expertise in Prison Law, 60 McGill L.J. 43, 71 (2014).

In Banks, the Court made use of both additional forms of deference identified here, recasting the facts and rewriting procedural rules in the defendant’s favor. But the deference displayed in Banks went beyond a pro-state move or two. It instead revealed what can only be described as an overall normative posture in favor of the state, one manifested in innumerable subtle ways. Consider that in Banks alone, the Court: ignored logical flaws in the state’s case; chastised the Third Circuit for evenhandedly weighing the evidence of the parties; rewrote the summary judgment rule to reduce the government’s burden in prison cases; and established what amounted to an irrebuttable presumption as to the validity of prison officials’ testimonial evidence. In addition, as with each of the Court’s prison law cases explored here, Banks gave no hint of ever having considered, much less factored into its analysis, either what day-to-day life was like for the plaintiffs or that its holding would only further immiserate people who already lived under conditions of extreme deprivation. Yet in virtually all its prison law opinions, Banks included,134×134. See, e.g., Banks, 548 U.S. at 525–26, 533 (plurality opinion). the Court has included language underscoring the challenges prison officials face on a daily basis and the need for courts to avoid decisions that might inadvertently make their jobs more difficult.

To treat these various features of Banks as separate and distinct would be to miss what becomes unmistakable when they are viewed in concert: the distinct normative cast to the Court’s divergent inclinations toward the parties in prison law cases, which has led the Court to routinely elevate the perspectives and experience of prison officials and to devalue the legal claims and lived experience of people in prison. This dispositional favoritism emerges clearly in the opinions themselves, with the Court approaching defendant prison officials and their evidence and arguments with sympathy, and incarcerated plaintiffs and their evidence and arguments with skepticism and even hostility. In case after case, the Court has found new ways to expand the power of prison officials over the incarcerated and to correspondingly shrink the scope of the constitutional protections prisoners can hope to receive.

It bears emphasizing just how directly this dynamic shapes the prison environment and, in particular, the degree of unchecked power that prison officials may exercise over prisoners.135×135. In this way, the Court’s dispositional favoritism manifests what Robert Cover famously described as the law’s violence. See Robert M. Cover, Essay, Violence and the Word, 95 Yale L.J. 1601, 1601 (1986) (“Legal interpretation takes place in a field of pain and death. . . . A judge articulates her understanding of a text, and . . . leave[s] behind victims whose lives have been torn apart by . . . organized, social practices of violence.” (footnote omitted)); Robert M. Cover, The Bonds of Constitutional Interpretation: Of the Word, the Deed, and the Role, 20 Ga. L. Rev. 815, 818 (1986) (“Even the violence of weak judges is utterly real . . . . Take a short trip to your local prison and see.”). In Overton v. Bazzetta136×136. 539 U.S. 126 (2003). and again in Banks, the Court gave its imprimatur to what Justice Stevens called “the deprivation theory of rehabilitation.”137×137. Banks, 548 U.S. at 546 (Stevens, J., dissenting); see also Overton, 539 U.S. at 137 (Stevens, J., concurring) (“Michigan, like many other States, uses withdrawal of visitation privileges for a limited period as a regular means of effecting prison discipline.”). On this approach — which, as Justice Stevens observed, “has no limiting principle”138×138. Banks, 548 U.S. at 546 (Stevens, J., dissenting). — even clear constitutional rights may be suspended if prison officials assert that doing so will motivate good behavior.139×139. See id. (observing that this justification “would provide a ‘rational basis’ for any regulation that deprives a prisoner of a constitutional right [if] there is at least a theoretical possibility that the prisoner can regain the right at some future time by modifying his behavior” (citing Kimberlin v. U.S. Dep’t of Just., 318 F.3d 228, 240 (D.C. Cir. 2003) (Tatel, J., concurring in part and dissenting in part))). In Casey, the Court dramatically narrowed prisoners’ right of access to the courts, effectively broadening the zone in which prison officials may act without fear of judicial censure.140×140. See Lewis v. Casey, 518 U.S. 343, 351–55 (1996). In Whitley, the Court established such a high standard for Eighth Amendment excessive force claims that COs may in all but the most extreme cases use violence with impunity.141×141. See Whitley v. Albers, 475 U.S. 312, 319 (1986). And in Woodford v. Ngo,142×142. 548 U.S. 81 (2006). the Court held that the PLRA143×143. See 42 U.S.C. § 1997e. required “proper exhaustion,” meaning that prisoners seeking to bring their constitutional claims in federal court must first comply with “all steps that the [prison] holds out, and do[] so properly.”144×144. Woodford, 548 U.S. at 90 (emphasis omitted) (quoting Pozo v. McCaughtry, 286 F.3d 1022, 1024 (7th Cir. 2002)). Among other things, Woodford’s effect has been to convert the short filing deadlines of prison grievance procedures — as short as thirteen days in many places145×145. See Giovanna Shay & Johanna Kalb, More Stories of Jurisdiction-Stripping and Executive Power: Interpreting the Prison Litigation Reform Act (PLRA), 29 Cardozo L. Rev. 291, 317 (2007). and in several states no more than three to five days146×146. See, e.g., Ky. Admin. Regs. 14.6 at 8 (2021), https://corrections.ky.gov/About/cpp/Documents/14/CPP%2014.6%20-%20Grievance%20-%20Effective%207-20-2021.pdf [https://perma.cc/MMZ2-9L4B] (five days); Mont. State Prison, Policies & Procedures: Policy No. MSP 3.3.3, Inmate Grievance System 3 (2005), https://cor.mt.gov/DataStatsContractsPoliciesProcedures/DataDocumentsandLinks/DOCPolicies/Chapter3/3.3.3-Offender-Grievance-Program.pdf [https://perma.cc/CL4W-2C6B] (three days); R.I. Dep’t of Corr., Policy No. 13.10-4: Inmate Grievance Procedure 7 (2019), https://doc.ri.gov/sites/g/files/xkgbur681/files/documents/policies/13.10-4-Final-Version.pdf [https://perma.cc/8Q8A-TRVY] (seven days); see also Derek Borchardt, Note, The Iron Curtain Redrawn Between Prisoners and the Constitution, 43 Colum. Hum. Rts. L. Rev. 469, 488, 494 & n.136 (2012) (discussing the “very strict time requirements,” id. at 494, of grievance systems). — into effective statutes of limitations on civil rights claims for incarcerated plaintiffs.147×147. For more detailed discussion of the impact of Woodford and its progeny on prisoners’ access to the federal courts, see Margo Schlanger & Giovanna Shay, Preserving the Rule of Law in America’s Jails and Prisons: The Case for Amending the Prison Litigation Reform Act, 11 U. Pa. J. Const. L. 139, 148 (2008); and Margo Schlanger, Inmate Litigation, 116 Harv. L. Rev. 1555, 1649–54 (2003). See generally Borchardt, supra note 146; Giovanna Shay, Exhausted, 24 Fed. Sent’g Rep. 287 (2012). In this way, Woodford ensures that many meritorious constitutional claims will fail long before they get to court, thereby affording legal impunity even to those COs who have plainly abused their authority.

It is worth pausing briefly over Woodford. As does the plurality opinion in Banks, Justice Alito’s majority opinion in Woodford demonstrates a commitment to deference that goes well beyond establishing a strikingly defendant-friendly standard for PLRA exhaustion. Among other things, the opinion: displays clear sympathy for prison officials (assumed to be interested only in providing “a meaningful opportunity for [the incarcerated] to raise meritorious grievances”);148×148. Woodford, 548 U.S. at 102. evinces hostility toward prisoners (framed as manipulative litigators who, given half a chance, would “deliberately and flagrantly [bypass] administrative review”);149×149. Id. at 97. writes as if prison grievance processes somehow embody the due process protections of more typical administrative procedures when they are in fact “more akin to lodging a complaint with the police” and almost entirely “lack the procedural protections usually associated with adversarial litigation”;150×150. Shay & Kalb, supra note 145, at 318. As Giovanna Shay and Johanna Kalb explain:

[Prison g]rievance policies never require a prisoner to spell out legal claims, and they often lack the procedural protections usually associated with adversarial litigation, such as formal discovery mechanisms and evidentiary hearings. . . . They are informal, non-adjudicative proceedings . . . . [In prison], submitting a grievance is more akin to lodging a complaint with the police than with filing a complaint in court; while a grievance initiates an investigative process, it is not intended to instigate adjudication of legal claims.

Id. and, as with all the cases canvassed here, never stops to consider the impact of its holding on those who are imprisoned, which in this case includes the real possibility that it might leave people in prison without recourse even for serious abuse or neglect at the hands of the very state officials charged to keep them safe.

That this opinion was written by Justice Alito, who now finds himself in the midst of a solid conservative majority on the new Roberts Court, provides some sense of the likely weight its orientation and tone will carry with the lower federal courts going forward.

C. Dispositional Favoritism in the Federal Courts

Given the marked tilt of substantive prison law doctrine in favor of the state, not to mention the many procedural obstacles incarcerated litigants must overcome to get a hearing on the merits,151×151. See Dolovich, supra note 8, at 114–16. it is only to be expected that defendant prison officials will easily prevail most of the time. This means that, in most prison cases, it is unnecessary for a court to engage in maneuvers at odds with basic norms of judicial reasoning. Where dispositional favoritism has its sharpest bite is in the adjudication of those claims in which, despite the defendant-friendly legal landscape, incarcerated plaintiffs still ought to succeed. It is here that, among other otherwise unaccountable moves, one is apt to see courts accepting legal or factual claims from defendant prison officials that are at best questionable and at worst “preposterous.”152×152. Remarkably, this characterization was used by a federal judge en route to finding for the state. See Transcript of Hearing on Motion for Temporary Injunction at 106, Holt v. Hobbs, No. 11-cv-00164 (E.D. Ark. Mar. 23, 2011) [hereinafter Transcript of Hearing], aff’d, 509 F. App’x 561 (8th Cir. 2013), rev’d and remanded, 574 U.S. 352 (2015). It is moments like this, which Driver and Kaufman dub “selective mythmaking,” Driver & Kaufman, supra note 1, at 572, that ground their claim that prison law is “incoheren[t],” id. at 571. And certainly, at least at this level — when courts endorse characterizations of prisoners and prison life that defy common sense — incoherence seems an apt description. Such cases must be understood not as bizarre aberrations but as the logical extreme of the normative posture the Court has been adopting — and implicitly and explicitly directing lower courts to adopt — for decades.

Consider the following:

1. Munson v. Gaetz. — In Munson v. Gaetz,153×153. 673 F.3d 630 (7th Cir. 2012). plaintiff James Munson experienced harmful side effects after being prescribed the wrong medication by prison medical staff.154×154. Id. at 631. Munson sought information on drug interactions and side effects from books in the prison library, but “long waiting lists and frequent prison lockdowns” impeded his access.155×155. Id. He therefore ordered some books on the topic “from a prison-approved bookstore,” including the Physicians’ Desk Reference (PDR) and the Complete Guide to Prescription and Nonprescription Drugs (Complete Guide).156×156. Id. at 631–32. As is clear from the table of contents and introductory material, the Complete Guide concerns only medications that are prescribed by physicians or are available over-the-counter. For each medication cataloged, the book also warns readers of problematic drug interactions with alcohol, marijuana, and cocaine. See generally H. Winter Griffith, Complete Guide to Prescription & Nonprescription Drugs (Stephen W. Moore ed., 2012). When the prison’s Publications Review Officer judged the books to be contraband for their drug-related content — aside from noting that the books were on the Disapproved Publications List, the officer wrote only one explanatory word, “DRUGS,” on the relevant form157×157. Munson, 673 F.3d at 632. — Munson brought a First Amendment challenge.158×158. Id. Although the drugs involved were plainly medical in nature, and although prisoners in the facility could access the same information contained in the two banned volumes in the prison’s own library,159×159. Id. the district court accepted defendants’ position that prison security required the prohibition. It therefore dismissed the case for failure to state a claim.160×160. Id. (quoting Munson v. Gaetz, No. 10-cv-00881, 2011 WL 692015, at *3 (S.D. Ill. Feb. 17, 2011). The Seventh Circuit upheld the dismissal.161×161. Id. at 638.

2. Singer v. Raemisch. — In Singer v. Raemisch,162×162. 593 F.3d 529 (7th Cir. 2010). prison officials banned the fantasy role-playing game (RPG) Dungeons and Dragons (D&D) on the grounds that the game’s “organized, hierarchical [status]”163×163. Id. at 535. promoted “gang . . . activity” and that it undermined players’ “rehabilitation and [the] effects of positive programming.”164×164. Id. (quoting Captain Bruce Muraski, Waupun’s Disruptive Group Coordinator and Security Supervisor). The only evidence offered by the state was an affidavit from Captain Bruce Muraski, the prison’s senior gang investigator and one of the co-defendants.165×165. Id. at 532. In response, plaintiff Kevin Singer submitted fifteen affidavits in addition to his own, including eleven from fellow incarcerated D&D players and three from free-world D&D and RPG experts.166×166. Id. at 533. This evidence directly challenged the state’s claims and described in detail the collaborative, prosocial, nonhierarchical, and even rehabilitative character of D&D gameplay.167×167. See id. Although the key issue in the case concerned the nature and effect of the game, about which all of Singer’s witnesses had direct personal knowledge and as to which Muraski had none, the Seventh Circuit dismissed much of Singer’s evidence as irrelevant because it originated with witnesses whose “experiential ‘expertise’ . . . is from the wrong side of the bars and fails to match Muraski’s perspective.”168×168. Id. at 536. Consequently, the panel found that no issue of material fact remained for trial and upheld the district court’s grant of summary judgment.169×169. See id. at 538.

3. Holt v. Hobbs. — In Holt v. Hobbs,170×170. 574 U.S. 352 (2015). the plaintiff, Gregory Holt,171×171. The plaintiff filed the lawsuit under the name Gregory Holt, but is also known as Abdul Maalik Muhammad. See id. This essay refers to him as Holt. sought an exception from the grooming standards of the Arkansas Department of Corrections to grow a half-inch beard for religious reasons.172×172. See id. at 355–56. Prison officials denied the request, and Holt brought suit under the Religious Land Use and Institutionalized Persons Act of 2000173×173. 42 U.S.C. §§ 2000cc to 2000cc-5. (RLUIPA), which established a strict scrutiny standard for any state prison regulation burdening religious practice.174×174. See Holt, 574 U.S. at 360–62. Defendant prison officials justified their refusal of Holt’s request on grounds of prison security, in particular that such beards could be used to hide or transport contraband.175×175. Id. at 359. The record of the hearing before the magistrate judge yielded several reasons for skepticism, including that people in Arkansas prisons are allowed to grow their head hair far longer than half an inch176×176. Arkansas Department of Corrections grooming standards allowed prisoners to grow their hair to a length just above their ears and just off the collar. Transcript of Hearing, supra note 152, at 58. and the obvious fact that they are permitted to wear clothes and shoes and thus have many more effective hiding places for contraband than a half-inch beard would afford.177×177. Holt, 574 U.S. at 359. Defendants also offered a second security-based justification — that prisoners might shave their beards to facilitate escape — which was equally unpersuasive, since as Holt observed, to address this problem the Arkansas Department of Corrections could simply take before-and-after photos as is the practice in New York state prison. Transcript of Hearing, supra note 152, at 23. Holt had been granted a preliminary injunction and was present at the merits hearing wearing a half-inch beard. During the hearing, the magistrate judge observed that “it’s almost preposterous to think that [Holt] could hide contraband in [his] beard.”178×178. Transcript of Hearing, supra note 152, at 106. Yet although the standard of review was strict scrutiny, the magistrate judge, the district court, and a unanimous panel of the Eighth Circuit all found for the state and against Holt.179×179. Holt, 574 U.S. at 360 (citing Holt v. Hobbs, 509 F. App’x 561 (8th Cir. 2013)). The Supreme Court took the case and reversed. For further discussion, see infra, pp. 329–30.

4. West v. Byers. — In West v. Byers,180×180. No. 13-cv-3088, 2015 WL 5603316 (D.S.C. Sept. 23, 2015). the plaintiff, Christopher West, brought an Eighth Amendment excessive-force claim arising from incidents that occurred while he was in administrative segregation.181×181. Id. at *2. In one incident, after West refused to remove his arm from the food slot during meal distribution, defendant CO Kevin Williams attempted to physically force West’s arm and shoulder back into the cell, “slammed” West’s arm several times with the metal door flap, and when West still did not remove his arm, pepper-sprayed him.182×182. Id. Subsequently, “an extraction team was assembled” by CO Williams, which forcibly removed West from his cell.183×183. Id. at *6. West was then placed in a restraint chair and not permitted a shower to remove the mace.184×184. Id. at *2. At the time of the incident, West’s only point of contact with others was through the food slot in an otherwise solid door, making it hard to see how his refusal to remove his arm from the slot created any danger to anyone. Yet the district court found that West’s conduct made force “necessary to restore order” and thus that CO Williams’s violent actions were not merely “for the very purpose of causing Plaintiff harm.”185×185. Id. at *6. On these and other grounds, the district court granted summary judgment to the defendants.186×186. Id. at *18.

5. Sixth Circuit Exhaustion Rules. — The PLRA instituted mandatory exhaustion of administrative remedies before people in custody may file a suit in federal court.187×187. 42 U.S.C. § 1997e(a) (“No action shall be brought with respect to prison conditions under section 1983 of this title, or any other Federal law, by a prisoner confined in any jail, prison, or other correctional facility until such administrative remedies as are available are exhausted.”). Following the PLRA’s passage, three district judges in the Sixth Circuit sua sponte created additional procedural burdens for prisoners, in each case heightening the pleading standards and making it easier for prison officials to defeat constitutional claims on exhaustion grounds.188×188. See Williams v. Overton, 136 F. App’x 859, 862 (6th Cir. 2005), rev’d and remanded sub nom. Jones v. Bock, 549 U.S. 199 (2007); Jones v. Bock, 135 F. App’x 837, 838 (6th Cir. 2005), rev’d and remanded, Jones, 549 U.S. 199; Walton v. Bouchard, 136 F. App’x 846, 848 (6th Cir. 2005), rev’d and remanded sub nom. Jones, 549 U.S. 199. In each case, the Sixth Circuit endorsed the rule change, effectively rewriting the circuit’s pleading rules exclusively for incarcerated plaintiffs189×189. Although exhaustion is typically treated as an affirmative defense, thanks to these new rules, prisoners in the Sixth Circuit would need to plead exhaustion in the complaint itself or risk dismiss-al. In addition, although it is normally the practice to dismiss only those claims that have not been exhausted, courts could dismiss every claim in a multiclaim complaint if the plaintiff had failed to exhaust even one of the listed claims, and courts in the circuit were empowered to dismiss prisoner claims if the complaint named defendants that had not been identified by name in the original internal prison grievance. Walton, 136 F. App’x at 848; Williams, 136 F. App’x at 862. and increasing the risk that prisoner suits otherwise entitled to be heard in federal court would be dismissed without a hearing.190×190. The Supreme Court struck down the three additional rules in Jones, 549 U.S. at 203.

In two of the situations just canvassed — Holt v. Hobbs and that involving the Sixth Circuit’s creative exhaustion rules — the Supreme Court intervened and reversed.191×191. See Holt v. Hobbs, 574 U.S. 352, 356 (2015); Jones, 549 U.S. at 199, 203, 224. In each instance, the ultimate outcome was plainly correct. The question is how the defendants prevailed in any of these cases in the first place. In Holt, a unanimous Court appropriately found the state’s arguments “hard to take seriously.”192×192. Holt, 574 U.S. at 363; see also id. at 363–64 (“An item of contraband would have to be very small indeed to be concealed by a ½-inch beard, and a prisoner seeking to hide an item in such a short beard would have to find a way to prevent the item from falling out.”). But for that case to have reached the Court, it first had to pass through three levels of federal court review. And at each stage, the court in question had to consider whether the state’s case satisfied not just rational basis review but RLUIPA’s mandated strict scrutiny. And in three separate proceedings, a federal court found that it did. The fact that it took the nine Justices of the United States Supreme Court to determine that, on those facts, the Arkansas Department of Corrections failed to meet its burden under strict scrutiny provides some sense of just how far the center of gravity in prison law cases tilts in favor of the state — and how prepared federal judges are to credit assertions by prison officials that would be roundly rejected if proffered by almost any other litigant.193×193. Prison officials are not the only state actors to enjoy the benefits of judicial deference. The Court has also, for example, historically shown deference to public school and military officials. See, e.g., Morse v. Frederick, 551 U.S. 393, 403 (2007) (holding that a school administrator’s decision to suspend a student for a banner displayed during an off-campus parade did not violate the First Amendment); Chappell v. Wallace, 462 U.S. 296, 300 (1983) (“Civilian courts must, at the very least, hesitate long before entertaining a suit which asks the court to tamper with the established relationship between enlisted military personnel and their superior officers . . . .”); see also Doe v. Pulaski Cnty. Special Sch. Dist., 306 F.3d 616, 619 (8th Cir. 2002) (holding that expelling an eighth-grade student for language in a letter found at his home did not violate the First Amendment); Ben-Shalom v. Marsh, 881 F.2d 454, 465 (7th Cir. 1989) (applying deferential rational basis review to an equal protection challenge brought by a gay U.S. Army Reserve sergeant alleging she had been discharged based on her sexuality). For an in-depth discussion of the Court’s demonstrated deference to public school officials, see generally Erwin Chemerinksy, The Hazelwooding of the First Amendment: The Deference to Authority, 11 First Amend. L. Rev. 291 (2013). And for one comprehensive take on the military deference doctrine, see generally John F. O’Connor, The Origins and Application of the Military Deference Doctrine, 35 Ga. L. Rev. 161 (2000). In these other contexts, however, this deference seems to come without the palpable skepticism and dismissiveness toward the plaintiffs’ arguments and concerns that are frequently in evidence when the plaintiffs are prisoners. Instead, one sees courts, even when deciding for the defendants, emphasizing the narrowness of the rulings, the minimally burdensome nature of the imposition, and the depth of the state interest in regulating the precise conduct at issue — all moves indicating a measure of respect for plaintiffs’ claims even when plaintiffs do not ultimately prevail. See, e.g., Parker v. Levy, 417 U.S. 733, 750–51 (1974) (upholding a court martial, although emphasizing that “enforcement of [the Uniform Code of Military Justice] in the area of minor offenses is often by sanctions which are more akin to administrative or civil sanctions than to civilian criminal ones”); Doe, 306 F.3d at 624 (finding no First Amendment violation against a student expelled based on content of a letter but going out of its way to emphasize the limits of the intrusion, noting among other things that the government “has no valid interest in the contents of a writing that a person, such as [the plaintiff], might prepare in the confines of his own bedroom”).

Likewise, in Jones v. Bock,194×194. 549 U.S. 199 (2007). the Court rejected each of the three heightened procedural burdens Sixth Circuit judges had been imposing in prison suits.195×195. See id. at 208–09, 211, 224. Yet the Court’s previously demonstrated readiness to ignore standard judicial procedure (as in Whitley) and to rewrite the rules of judicial review so as to benefit defendants (as in Banks) helps to explain how three district judges in the Sixth Circuit could have felt free to alter standard rules of pleading to the benefit of defendant prison officials and at the expense of incarcerated plaintiffs. It also helps to make sense of the Sixth Circuit’s otherwise unaccountable readiness to affirm in each case. From one angle, the Sixth Circuit ought not to have been surprised by the Court’s holding in Jones; it is, after all, the obligation of the federal courts to apply established pleading standards and not to revise them sua sponte to the detriment of disfavored parties. Yet in adopting these procedural innovations at prisoners’ expense, the Sixth Circuit might fairly be said to have only been following the Court’s lead.196×196. I read Jones as an announcement that the Court wished to reserve for itself the authority to change the rules governing prison cases. Others may read it differently. But whatever the explanation, it bears notice that it is the Court’s response to the Sixth Circuit’s creation of new rules governing PLRA exhaustion that seems to call out for explanation, and not the Sixth Circuit’s sua sponte establishment of additional procedural burdens on incarcerated litigants. That this is so only reinforces the main point: what is normative for this context is not fair treatment of all parties and evenhanded judicial scrutiny of their submissions but instead the readiness to shift the goalposts to benefit the state.

Taken together, the cases described above also serve to illustrate the way that in this legal arena, when it comes to the facts, the usual principles of judicial inquiry do not apply. Thus, one sees courts accepting factual claims presented by defendants that were either disputed by the record (following Rhodes) or contradicted common-sense understandings (as in Casey). In Munson, the Seventh Circuit endorsed with a straight face the prison’s labeling the PDR and the Complete Guide as contraband for their drug-related content, as if no difference existed between heart meds and heroin.197×197. See Munson v. Gaetz, 673 F.3d 630, 633–34 (7th Cir. 2012). And in Singer, the court put aside the considerable evidence in the record refuting defendants’ ill-informed claims regarding Dungeons and Dragons to reach the doubtful conclusion that no jury could possibly find for the plaintiffs at trial.198×198. See Singer v. Raemisch, 593 F.3d 529, 536, 540 (7th Cir. 2010). Similar dynamics are present in West,199×199. West v. Byers, No. 13-cv-3088, 2015 WL 5603316 (D.S.C. Sept. 23, 2015), aff’d sub nom. West v. Byars, 634 F. App’x 938 (4th Cir. 2016). in which on no plausible version of the facts could the force used have been thought warranted, and also in Holt, in which multiple federal courts ruled as if the prison’s asserted fear that Holt’s half-inch beard could be used to hide contraband was not patently absurd.200×200. Holt v. Hobbs, 574 U.S. 352, 360 (2015). These cases may seem — and are — extreme in terms of the courts’ discounting of the reality before them. But in their determination to approach the facts in ways that benefit defendants at the expense of incarcerated plaintiffs, the judges in these cases are arguably just following the Court in finding some way, however logically questionable, to frame the matter in ways sympathetic to the state. Certainly, there are plenty of prison law cases in which federal courts put the state to its proof and appropriately apply governing rules and standards.201×201. See cases cited supra note 14.. But this being the baseline judicial obligation, such cases should occasion no notice. The real puzzle is posed by cases of the sort just canvassed. Are they just statistical outliers? Merely evidence that federal courts may sometimes get it terribly wrong? In my view, to dismiss these opinions as marginal or of no moment would be to miss what they reveal about the field as a whole. These cases are decided by courts that have for decades looked to the Supreme Court’s prison law cases, not merely for the standards they establish or the holdings they reach, but for the tone they take and the signs they give as to how courts ought to regard the parties and their submissions. The Court, in other words, has led by example. That some federal judges, in their enthusiastic embrace of dispositional favoritism, may have somewhat slipped the traces is perhaps not surprising in a regime lacking clear boundaries or guidelines.202×202. See Dolovich, supra note 34, at 245 (“[T]aken as a body, the [prison law cases] reveal no principled basis for determining when deference is justified, what forms it may legitimately take, or the proper limits on its use. Instead, the mere mention of ‘deference’ has emerged as a catch-all justification for curtailing . . . the burden on prison officials to ensure constitutional prisons . . . .”). If, in these instances, courts have breached the limits of logic or common sense or exceeded their authority in imposing new burdens on incarcerated plaintiffs, it is the Court’s own handling of prison law claims that has invited them to do so.

III. The Covid Cases: Dispositional Favoritism in an Emergency

This was the judicial environment in March 2020 when Covid hit. From the first days of the pandemic, it was clear that the incarcerated would face an outsized risk of infection and death from the virus.203×203. See Dolovich, supra note 15; Saloner et al., supra note 15, at 602–03 (finding that, over the first four months of the pandemic, people held in state and federal prisons were experiencing Covid infection at five-and-a-half times the rate of the American population as a whole and that, adjusted for age, people in prison were dying of Covid at three times the national rate). The preventative measures that were emerging as key to self-protection — practicing social distancing, avoiding socializing indoors, maintaining good hand hygiene, making use of clean and effective personal protective equipment, and so on — were practically impossible for people in custody, who generally live in crowded, poorly ventilated facilities and lack any meaningful control over their environment.204×204. See Dolovich, supra note 15. To make matters worse, American penal institutions are full of people who, whether because of age, medical comorbidities, or both, are among those identified early in the pandemic by the Centers for Disease Control and Prevention (CDC) as disproportionately likely to develop severe complications from Covid.205×205. Id. Immediately recognizing the danger, advocates all over the country turned to the federal courts, arguing unconstitutional conditions and seeking relief.

These efforts produced a mountain of case law. The resulting judicial opinions addressed a wide range of thorny issues, perhaps chief among them the question of which procedural vehicle — habeas corpus or 42 U.S.C. § 1983 — ought to govern when the gravamen of the claim is unconstitutional conditions but the relief requested is release. At their core, however, these cases were asserting a constitutional claim: that subjecting people to living conditions well known to pose an inordinate risk of Covid infection and death violated the Eighth Amendment. It has long been established that prison conditions run afoul of this provision when prison officials are deliberately indifferent to prisoners’ basic human needs.206×206. See, e.g., Wilson v. Seiter, 501 U.S. 294 (1991); Rhodes v. Chapman, 452 U.S. 337 (1981); Estelle v. Gamble, 429 U.S. 97 (1976). And as we have seen, in Farmer, the Court held that the state of mind of Eighth Amendment deliberate indifference is equivalent to criminal recklessness, so that a prison “official [who] knows of and disregards an excessive risk to [the] health or safety” of people in custody satisfies this standard.207×207. See Farmer v. Brennan, 511 U.S. 825, 837 (1994); id. at 839–40; see also Helling v. McKinney, 509 U.S. 25, 35–36 (1993) (holding that plaintiffs satisfy the objective component of an Eighth Amendment claim if they can demonstrate exposure to “an unreasonable risk of serious damage to [their] future health,” id. at 35, and can show that the risk is one “society considers . . . so grave that it violates contemporary standards of decency to expose anyone unwillingly to such a risk,” id. at 36 — in other words, “that the risk of which he complains is not one that today’s society chooses to tolerate,” id.). Farmer’s holding is already extremely defendant friendly.208×208. For extended discussion on this point, see Dolovich, supra note 55, at 945–48 (arguing that Farmer’s actual knowledge standard enables both macro- and micro-level failures inflicting harm on prisoners to go unaddressed). Yet under the extraordinary circumstances of the pandemic, plaintiffs should have easily cleared the bar, given that, as prison officials were well aware, the basic living conditions in prison posed a substantial and disproportionate risk of serious harm to those inside. At least as to this aspect of the analysis, that is, plaintiffs in the Covid prison cases ought to have readily prevailed.

Yet this is not how it went — not by a long shot. In several Covid cases, plaintiffs had some initial success, as district courts, directly confronting the reality on the ground, granted preliminary injunctions or temporary restraining orders directing correctional officials to improve conditions inside and/or to identify facility residents at highest risk from Covid and prepare for their release.209×209. See, e.g., Wilson v. Williams, 455 F. Supp. 3d 467 (N.D. Ohio 2020), vacated and remanded, No. 20-cv-00794, 2020 WL 2542131 (N.D. Ohio May 19, 2020), vacated, 961 F.3d 829 (6th Cir. 2020); see also Wilson, No. 20-cv-00794, 2020 WL 2542131 (N.D. Ohio May 19, 2020) (Federal Correctional Institution Elkton); Valentine v. Collier, 455 F. Supp. 3d 308 (S.D. Tex.) (Texas Wallace Pack Unit Prison), rev’d, 993 F.3d 270 (5th Cir. 2021), motion to vacate denied, 140 S. Ct. 1598 (2020); Mays v. Dart, 453 F. Supp. 3d 1074 (N.D. Ill.) (Cook County Jail), rev’d in part, vacated in part, aff’d in part, 974 F.3d 810 (7th Cir. 2020); Swain v. Junior, No. 20-cv-21457, 2020 WL 1692668 (S.D. Fla. Apr. 7, 2020) (Miami-Dade Metro West Detention Facility), vacated and remanded, 961 F.3d 1276 (11th Cir. 2020). But some plaintiffs still lost in the district court.210×210. See, e.g., Money v. Pritzker, 453 F. Supp. 3d 1103 (N.D. Ill. 2020) (Illinois state prisons). And even when they won, in virtually every case framed as a constitutional class action,211×211. See Dolovich, supra note 15, at *24 n.46 (explaining that only two Covid-related cases seeking broad constitutional relief for people in prison actually yielded releases and that in both instances, the finding of deliberate indifference on the part of Federal Bureau of Prisons officials rested not on defendants’ general failure to take sufficient precautionary measures but on their failure to implement the clear directive, issued by then–Attorney General William Barr pursuant to the Coronavirus Aid, Relief, and Economic Security Act (CARES Act), 15 U.S.C. § 9001 (Supp. II 2020), to expand the use of home confinement as a response to the pandemic). decisions in plaintiffs’ favor were eventually overturned on appeal.212×212. See, e.g., Swain v. Junior, 958 F.3d 1081 (staying preliminary injunction); Valentine v. Collier, 956 F.3d 797 (5th Cir. 2020) (same). In case after case, appeals courts granted defendants stays of district court orders — and in the two instances where the circuit court declined to grant the stay, the Supreme Court stepped in and did it for them.213×213. Barnes v. Ahlman, 140 S. Ct. 2620 (2020); Williams v. Wilson, No. 19A1047, 2020 WL 2988458 (U.S. June 4, 2020) (mem.) (granting stay pending disposition of the Government’s appeal in the Sixth Circuit); cf. Valentine v. Collier, 140 S. Ct. 1598 (2020) (denying plaintiffs’ request to vacate stay granted by the Fifth Circuit).

How did they manage it? This is not the place for a full accounting of how, at this singular moment, when more than two million incarcerated people found themselves held captive under conditions facilitating the spread of a deadly pathogen, the federal courts ultimately rebuffed virtually all petitions for constitutional relief. For present purposes, the main point is that, in these cases, many appellate courts resorted to the familiar evasive moves of dispositional favoritism modeled so assiduously by the Court and internalized by the federal judiciary in the de-cades leading up to the pandemic. What we see in particular is frequent recourse to three main pro-state strategies: selective reading of the facts; a recasting of governing constitutional standards — here, Eighth Amendment deliberate indifference; and a studied disregard of prisoners’ lived experience during Covid and what the case at hand would mean in practical terms for incarcerated plaintiffs. These moves often appeared together, with courts ignoring plaintiffs’ evidence and accepting wholesale defendants’ framing of the facts in order to conclude, based on subtle (or not so subtle) revisions of Farmer’s deliberate indifference standard, that plaintiffs had failed to meet their burden.

I defer to another day a detailed catalog of the many ways appeals courts reframed Eighth Amendment deliberate indifference to increase the burden on incarcerated plaintiffs bringing Covid-related conditions challenges.214×214. For helpful discussion on this theme, see Brandon L. Garrett & Lee Kovarsky, Viral Injustice, 110 Calif. L. Rev. 117 (2022); and Michael L. Zuckerman, When the Conditions Are the Confinement: Eighth Amendment Habeas Claims During COVID-19, 90 U. Cin. L. Rev. 1 (2021). Here I will highlight just one such move, which emerged as an especially common means by which courts eased the state’s constitutional burden. This approach relied on Farmer’s observation that “prison officials who actually knew of a substantial risk to inmate health or safety may be found free from liability if they responded reasonably to the risk, even if the harm ultimately was not averted.”215×215. Farmer v. Brennan, 511 U.S. 825, 844 (1994). As articulated, this caveat is not inconsistent with Farmer’s mens rea holding — but here, the Farmer Court’s precise formulation is crucial. Under the logic of recklessness, it is only a reasonable response to the risk that should suffice as a defense. Otherwise, any act on the defendants’ part occasioned by the awareness of the risk would defeat a recklessness finding, even if that act were entirely ineffectual in averting the danger and the defendant knew it to be so. As Justice Sotomayor put it in her dissent from the Court’s refusal to vacate a stay in the Fifth Circuit case of Valentine v. Collier,216×216. 140 S. Ct. 57 (2020). the fact that “respondents took reasonable ‘affirmative steps’ to respond to the virus” cannot be sufficient to defeat a showing of deliberate indifference under Farmer “when officials know that those steps are sorely inadequate and leave inmates exposed to substantial risks.”217×217. See id. at 61 (Sotomayor, J., dissenting from the denial of application to vacate stay) (quoting Valentine v. Collier, 978 F.3d 154, 164 (5th Cir. 2020)). Yet, relying on the “reasonable response” language quoted above (and seemingly following the early lead of the Fifth Circuit, a first mover on this front), several courts proceeded as if evidence of any affirmative measures on the part of prison officials undertaken in response to Covid was sufficient to rebut deliberate indifference, regardless of whether the defendants knew full well that the danger persisted.218×218. Many courts hearing conditions challenges early in the pandemic conveniently elided this point. In the Middle District of Louisiana, defendants were found not constitutionally liable because they “clearly demonstrated that they ha[d] taken measures to implement precautions to protect inmates from the COVID-19 pandemic,” without the court making any finding as to whether those measures were sufficient to mitigate the threat or whether defendants subjectively believed they were. Belton v. Gautreaux, No. 20-00278, 2020 WL 3629583, at *5 (M.D. La. July 3, 2020). And for the Eleventh Circuit, it was enough that, “when faced with a perfect storm of a contagious virus and the space constraints in a correctional facility, the defendants here acted []reasonably by ‘doing their best’” — regardless of whether the actions the defendants had taken made any appreciable difference to the danger Covid posed to people in their custody. Swain v. Junior, 961 F.3d 1276, 1289 (11th Cir. 2020). It is true that the Covid threat required drastic remedies. To meaningfully reduce the risk to incarcerated people, state officials would have needed to release sufficient numbers of people to enable social distancing among those who remained. Yet, even in the post-PLRA world, this situation is hardly unprecedented. In Plata and Coleman, two class actions brought against the CDCR, federal courts found medical and mental health care delivery in CDCR prisons to be constitutionally deficient. Pursuant to those findings, there ensued a concerted effort by a range of state actors — corrections officials, legislators, public health experts, and even the governor — to institute court-ordered remedies. See Schlanger, supra note 95, at 184–91. Ultimately, the process required considerable population reduction in CDCR facilities, and for the same reason plaintiffs were giving in the Covid prison cases: without reductions in population density, CDCR officials would be unable to satisfy their constitutional obligation to protect the incarcerated against known risks of serious harm. Id. at 186. For CDCR officials to have defended against allegations of unconstitutional medical or mental health care on the grounds that they were “doing their best” would have been insufficient in Plata or Coleman to defeat a finding of deliberate indifference so long as the defendants realized that the danger posed to people in CDCR facilities persisted.

Still, in several cases, even this defendant-friendly gloss on Farmer’s deliberate indifference standard was insufficient to warrant a finding for the state. In some instances, the steps defendants attested to taking were on their face so glaringly inadequate that claims of a “reasonable response” rang especially hollow. In others, plaintiffs introduced evidence that, although defendants had announced policies to mitigate the risk of Covid spread, those policies were not actually being followed.219×219. See Dolovich, supra note 15, at *11–13. Prison officials cannot always be assumed to know everything going on in their facilities. But given the urgency of the need to limit Covid infection and what was well understood about Covid transmission from the first weeks of the pandemic, it can fairly be assumed that those running prisons in the spring of 2020 would have known whether, for example, those in their facilities were “socializing in dayrooms with no space to distance physically,” Barnes v. Ahlman, 140 S. Ct. 2620, 2621 (2020) (Sotomayor, J., dissenting from the grant of stay), or standing “shoulder-to-shoulder” while waiting to be seen in the medical clinic, or whether kitchen workers “only occasionally [wore] face masks . . . while serving food,” Marlowe v. LeBlanc, No. 18-63, 2020 WL 1955303, at *4 (M.D. La. Apr. 23, 2020), stay granted, 810 F. App’x 302 (5th Cir. 2020), motion to vacate denied, 140 S. Ct. 2823 (2020). This reality put heightened pressure on courts predisposed to side with the state and disinclined to look with favor on plaintiffs’ claims. But judges finding themselves in this situation were not without resources. It was in these moments of greatest tension between the weight of the evidence and the courts’ felt imperative to forebear from holding the state constitutionally liable that judicial recourse to the familiar moves of dispositional favoritism became most obvious. The resulting opinions often have the same bizarre quality as those canvassed above in section II.C. On the surface, they possess what might be thought of as the aesthetic of legal argument — they flag legal issues, cite precedent, name and apply standards, reach conclusions, and so forth. Yet in substance, in many instances, the analyses lack the signs of the open-minded and evenhanded weighing of the evidence and arguments that are supposed to define good faith judicial reasoning.

Below are just two examples of the way appeals courts faced with cases involving Covid in prison combined a recasting of the deliberate indifference standard with a defendant-friendly reading of the facts to justify finding for the state.

A. Wilson v. Williams

Wilson v. Williams220×220. 961 F.3d 829 (6th Cir. 2020). arose out of Federal Correctional Institution Elkton (FCI Elkton), where, by April 2020, six people had already died of Covid,221×221. See Dolovich, supra note 15, at *21. “more clung to life only with the aid of ventilators,”222×222. Wilson, 961 F.3d at 845 (Cole, C.J., concurring in part and dissenting in part). and at least thirty-five percent of the population was over the age of sixty-five or had “significant [preexisting] health conditions making them extremely vulnerable to COVID-19.”223×223. Wilson v. Williams, No. 20-cv-00794, 2020 WL 2542131, at *2 (N.D. Ohio May 19, 2020), stay granted, 207 L. Ed. 2d 168 (2020), vacated and remanded by No. 20-3547, 2020 WL 9813537 (6th Cir. Sept. 17, 2020). For further discussion of Wilson, see Dolovich, supra note 15, at *21–24. Reviewing the preliminary injunction on the merits, the Sixth Circuit conceded that the Federal Bureau of Prisons (BOP) “was aware of and understood the potential risk of serious harm” that Covid posed to Elkton residents.224×224. Wilson, 961 F.3d at 840. The issue, as the panel framed it, was whether “the BOP responded reasonably to the known, serious risks posed by COVID-19 to [the] petitioners.”225×225. Id. In finding that the BOP “‘responded reasonably to the risk’ and therefore ha[d] not been deliberately indifferent,” the panel emphasized the BOP’s implementation of a “six-phase action plan to reduce the risk of COVID-19 spread at Elkton.”226×226. Id. at 841 (quoting Farmer v. Brennan, 511 U.S. 825, 844 (1994)). Yet, taking a closer look at the details of this official response, it is hard to see how it could be thought a reasonable response, if “reasonable” means in any way adequate to mitigate the risk. As Chief Judge Cole explained in his Wilson dissent, the BOP’s plan, broken down, amounted to:

two different phases addressing the screening of inmates, an entire phase consisting of only taking inventory of the BOP’s cleaning supplies, a phase where the BOP confined inmates to their quarters where they cannot socially distance, and a final phase that just extended the previous one[, making it], for practical purposes, a four-phase plan where one phase is taking inventory of supplies and another involves the locking of inmates in 150-person clusters where they cannot access the principal method of COVID-19 prevention.227×227. Id. at 848 (Cole, C.J., concurring in part and dissenting in part).

The BOP also emphasized that it was “conduct[ing] testing in accordance with CDC guidance.”228×228. Id. at 841 (alteration in original). However, in May 2020, the district court had found that, up to that point, only 524 Covid tests had been administered in Elkton, a facility then housing 2357 people, and that of those tests performed, “approximately 24% . . . came back COVID-19 positive.”229×229. Wilson v. Williams, No. 20-cv-00794, 2020 WL 2542131, at *1 (N.D. Ohio May 19, 2020). This data suggests both a glaringly inadequate testing protocol and a worrying degree of viral spread in a facility housing a high proportion of Covid-vulnerable people.230×230. By this point in the pandemic, the Ohio Department of Corrections had already conducted mass testing at two facilities comparable in population size to FCI Elkton. See Dolovich, supra note 15, at *22. This data, moreover, was well known to BOP officials. It thus seemed plain that, regardless of any actions they had taken in mitigation to that point, officials at FCI Elkton would have “know[n] that those steps [we]re sorely inadequate and le[ft] inmates exposed to substantial risks.”231×231. Valentine v. Collier, 140 S. Ct. 57, 61 (2020) (Sotomayor, J., dissenting). Yet on the strength of the BOP’s representation of all it had done “to reduce the risk of COVID-19 spread at Elkton,” the Sixth Circuit found that defendants had “responded reasonably to the risk” and thus were not deliberately indifferent.232×232. Wilson, 961 F.3d at 841. As the Sixth Circuit observed, their “sister circuits have concluded that similar actions by prison officials demonstrate a reasonable response to the risk posed by COVID-19.” Id. (citing Swain v. Junior, 958 F.3d 1081 (11th Cir. 2020); Valentine v. Collier, 956 F.3d 797 (5th Cir.), motion to vacate denied, 140 S. Ct. 1598 (2020); Marlowe v. LeBlanc, 810 F. App’x 302 (5th Cir.), motion to vacate denied, 140 S. Ct. 2823 (2020)). Labelling this conclusion “dispositive,” the panel concluded that the plaintiffs had no likelihood of success on the merits of their Eighth Amendment claim and vacated the preliminary injunction.233×233. Wilson, 961 F.3d at 845.

B. Swain v. Junior

Swain v. Junior234×234. 961 F.3d 1276 (11th Cir. 2020). was brought by a class of detainees in the Miami-Dade Metro West Detention Center (Metro West).235×235. Id. at 1280. In its opinion, the Eleventh Circuit emphasized the testimony by defendant jail officials that they took “numerous measures . . . to mitigate the spread of the virus,” “including, among many other things,” “‘requiring staff and [detainees] to wear face masks at all times (other than when sleeping),’ . . . and ‘providing disinfecting and hygiene supplies to all [detainees].’”236×236. Id. at 1287–91. In addition, the panel cited “the court-commissioned expert report,” which found that jail officials had “put ‘tape on the floor to encourage social distancing in lines,’” “‘[staggered] bunks . . . with head to foot configuration’ . . . to maximize the distance between faces during sleep,” and ensured that “patients are staggered and appropriately distanced when going to medical.”237×237. Id. at 1287–88.

Yet, as Judge Martin noted in her dissent, the record contained numerous declarations from Metro West detainees directly contradicting these specific claims. Among other things, plaintiffs’ witnesses testified that “it is often difficult or impossible to clean shared surfaces, such as phones, because they are not provided with disinfectant or other cleaning supplies”;238×238. Id. at 1300 (Martin, J., dissenting). that although “detainees were each given a mask approximately once a week, . . . the masks are ‘soft,’ ‘rip a lot,’ . . . ‘get really dirty[,]’ [and sometimes] break after ‘two to three days’”;239×239. Id. that detainees were “lined up less than a foot apart for pat-down inspections”;240×240. Id. at 1301. and that, while waiting in medical, “detainees must wait ‘shoulder to shoulder,’ sometimes with people from other cells.”241×241. Id. As for the expert report commissioned by the court, plaintiffs’ witnesses attested that:

[I]mmediately before the inspection Metro West staff made numerous last minute changes, including moving people out of cells that were going to be inspected so they would be less crowded; restocking toilet paper and soap; painting bathroom walls to cover black mold; scrubbing down cells; and placing additional soap in the unit.242×242. Id. at 1301–02.

Despite evidence directly contradicting the defendants’ account and strongly suggesting manipulation of the fact-finding process, the panel found, on the strength of defendants’ representations, that jail officials had “acted reasonably” and thus that a finding of Eighth Amendment deliberate indifference was inappropriate.243×243. Id. at 1289 (majority opinion). The panel therefore vacated the preliminary injunction.244×244. Likewise, in Barnes v. Ahlman, the Supreme Court granted a stay to jail officials in Orange County despite evidence that “the Jail misrepresented under oath . . . the measures it was taking to combat the virus’ spread, and even though the Jail’s central rationale for a stay . . . ignore[d] the lower courts’ conclusion that the Jail’s measures fell ‘well short’ of . . . [CDC] Guidelines.” Barnes v. Ahlman, 140 S. Ct. 2620, 2621 (2020) (Sotomayor, J., dissenting from the grant of stay).

C. Dispositional Favoritism as a Normative Project

For those whose first encounter with prison law came through the Covid cases, and in particular the several appellate decisions that together foreclosed constitutional relief for incarcerated plaintiffs nationwide, the judicial reasoning they contain may have seemed puzzling. Yet as those familiar with the field well know, what one finds in these cases is only the logical extension of the normative posture courts have long adopted toward constitutional (or quasi-constitutional245×245. See, e.g., Holt v. Hobbs, 574 U.S. 352 (2015) (concerning the free exercise of religion under RLUIPA). ) claims brought by prisoners.

The judiciary is not the only public institution to regard the incarcerated with hostility. The legislative politics of the tough-on-crime era of the 1980s and 1990s were enabled by a sense — still persisting today — that people with criminal convictions, especially prisoners, are “a breed apart,”246×246. Kelsey Kauffman, Prison Officers and Their World 119 (1988). “a different species of threatening, violent individuals for whom we can have no sympathy and for whom there is no effective help.”247×247. David Garland, The Culture of Control: Crime and Social Order in Contemporary Society 136 (2001). And in the prisons themselves, innumerable dynamics reinforce the dehumanization and demonization of people in custody, which in turn shape the way prison officials treat the incarcerated.248×248. See Dolovich, Excessive Force, supra note 40. In sum, as I have argued at length elsewhere, there is in the United States no branch of democratic government in which people in custody are regarded or treated as human beings entitled to respect and protection from needless harm.249×249. See generally Dolovich, The Failed Regulation, supra note 7.

How to explain this troubling moral economy? People in prison are among the most socially marginalized and politically disenfranchised members of society, disproportionately likely to be undereducated, unskilled, and indigent and disproportionately likely to suffer from mental illness and substance use disorder.250×250. See Sharon Dolovich, Foreword: Incarceration American-Style, 3 Harv. L. & Pol’y Rev. 237, 244 (2009). Paradoxically, this extreme vulnerability may explain rather than confound the callous indifference toward the imprisoned that those in power too frequently display; as Nancy Isenberg documents, contempt and even disgust toward those most in need of compassion and help is a longstanding American cultural tradition.251×251. See generally Nancy Isenberg, White Trash: The 400-Year Untold History of Class in America (2016).

And then there is the matter of race. As is well documented, Black Americans are wildly overrepresented in the U.S. prison population,252×252. See, e.g., Ashley Nellis, The Sentencing Project, The Color of Justice: Racial and Ethnic Disparity in State Prisons 5 (2021), https://www.sentencingproject.org/publications/color-of-justice-racial-and-ethnic-disparity-in-state-prisons [https://perma.cc/SQ8W-4SR7] (reporting that Black Americans are incarcerated at nearly five times the rate of White Americans; that in seven states, the Black-White disparity is greater than nine to one; and that in twelve states, “more than half the prison population is Black”); Wendy Sawyer, Visualizing the Racial Disparities in Mass Incarceration, Prison Pol’y Initiative (July 27, 2020), https://www.prisonpolicy.org/blog/2020/07/27/disparities [https://perma.cc/JU8B-NHH5] (reporting that, although Black Americans comprise twelve percent of the U.S. population, they represent forty-eight percent of people “serving sentences of life, life without parole, or ‘virtual life’” and forty-two percent of people on death row). a disparity that only begins to hint at the deeply racialized character of the American carceral project.253×253. See Dolovich, The Failed Regulation, supra note 7, at 169–70 (tracing the role of incarceration as a system of racial control during slavery and continuing after Emancipation through Jim Crow and up to present day); see also Taja-Nia Y. Henderson, Property, Penality, and (Racial) Profiling, 12 Stan. J. C.R. & C.L. 177, 178 (2016) (describing the role local jails played in supporting enslavers and the institution of chattel slavery itself in the antebellum South); John Bardes, The Problem of Incarceration in the Age of Slavery 5, 43–47 (unpublished manuscript) (on file with the Harvard Law School Library) (describing a network of carceral institutions forming a “statewide penal system for enslaved convicts” in Louisiana, Mississippi, Tennessee, and elsewhere in the antebellum South, and describing the brutal methods of torture employed to humiliate and “discipline” the enslaved people held in those facilities). It is, moreover, no secret that race bias, whether implicit or explicit, predisposes individuals to respond to people of color, and Black Americans in particular, as somehow less than human and thus less morally worthy than similarly situated Whites.254×254. See generally L. Song Richardson & Phillip Atiba Goff, Interrogating Racial Violence, 12 Ohio St. J. Crim. L. 115 (2014); Kelly Welch, Black Criminal Stereotypes and Racial Profiling, 23 J. Contemp. Crim. Just. 276 (2007). While it is impossible to know how much of the Court’s disposition toward the incarcerated is shaped by these pernicious forces, it is equally hard to imagine that they do not play some appreciable role.255×255. I thank Rose Daeun Jung for pushing me to address the dynamics motivating the Court’s dispositional favoritism.

This brings us back to the Covid cases. Here, I intentionally leave to one side the question of why the federal courts may have been especially determined to deny constitutional relief to people who found themselves behind bars during the pandemic. For present purposes, the key point is that, even in this singular emergency, the orientation of the courts remained consistent with the normative tenor of prison law as a whole — and with the consequent dispositional favoritism that is its hallmark. The ultimate effect has been to leave the prison environment, even during the height of the Covid threat, almost wholly free from judicial regulation.


There is a striking coherence to the Supreme Court’s prison law doctrine: it is predictably pro-defendant, highly deferential to prison officials, and largely indifferent to the impact of judicial decisions on the lived experience of people in custody. This pronounced pro-state tilt, I have argued, is a function of the divergent normative inclinations evident in the Court’s treatment of the parties in prison law cases, which leads the Court to elevate the perspective, interests, and experience of prison officials and to devalue those of incarcerated litigants. The most obvious manifestation of this normative posture has been the crafting of doctrinal standards that systematically benefit defendant prison officials.256×256. See supra Part I, pp. 305–16. But as we have seen, there also exist several additional mechanisms by which prison law cases are made to shift in favor of the state. Some of these moves — the reframing of the facts, the remaking of procedural burdens — are relatively easy to spot.257×257. See supra section II.B, pp. 319–25. Others represent the more subtle effects of an overall moral orientation, here labeled dispositional favoritism, which disposes courts, following the Supreme Court’s lead, to take every opportunity to turn things in defendants’ direction at plaintiffs’ expense.258×258. See supra sections II.B–C, pp. 319–32. The effect has been a body of law that ensures only minimal constitutional protections for a class of legal subjects whose interactions with state actors take place behind high walls, away from public view, and in fraught and adversarial environments where, absent some meaningful external check, uniformed officers hold all the power.

It does not have to be this way. True, there is little likelihood of a shift any time soon in the overtly pro-state slant of the governing doctrinal standards; any such change must await a Supreme Court differently oriented than the new Roberts Court. But even as things stand, the judicial abandonment of dispositional favoritism would go far toward making the federal courts a site of fair adjudication of prisoners’ constitutional rights. And as this essay shows, courts already have the tools at their disposal to make this change. All it takes is a readiness on the part of judges to (1) acknowledge people in custody as full-fledged constitutional subjects entitled to the protections this status entails, (2) explicitly recognize prison officials as the state actors charged with fulfilling the state’s obligations to the incarcerated and thus whose official conduct demands careful scrutiny, and (3) approach each case with open-mindedness and evenhandedness in the treatment of litigants and the scrutiny of their evidence and arguments. These requisites are far from radical; they are indeed only the basic obligations that ought to guide such judicial deliberations. The current shape of the prison law doctrine, the product of decades of dispositional favoritism on the part of the courts, is thus at once a testament to the widespread judicial abdication of these basic obligations and a roadmap pointing the way out for any judge inclined to follow it.

* Professor of Law, UCLA School of Law. With thanks to Sasha Natapoff for extremely helpful comments and conversation, and Emma Maynard for excellent research and editorial assistance.