In the United States Court of Appeals For the Seventh Circuit ____________________ No. 23-1703 ABRE JACKSON, Plaintiff-Appellant, v. MARC T. ANASTASIO, et al., Defendants-Appellees. ____________________ Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 1:20-cv-06004 — Sara L. Ellis, Judge. ____________________ ARGUED APRIL 2, 2024 — DECIDED AUGUST 25, 2025 ____________________ Before ROVNER, HAMILTON, and SCUDDER, Circuit Judges. HAMILTON, Circuit Judge. Plaintiff Abre Jackson is a pris- oner in Illinois. After a physical altercation with several prison guards and a prison disciplinary hearing, he was placed in disciplinary segregation, more commonly known as solitary confinement, for three months. This appeal is the lat- est in a series that have required us to consider when prison- ers’ disciplinary placements in solitary confinement amount to deprivations of a liberty interest protected by procedural 2 No. 23-1703 due process, as well as the extent of process that may be due. The answer depends on the combination of the length of soli- tary confinement and the actual conditions of that confine- ment as compared to those the prisoners would otherwise ex- perience. The district court granted summary judgment for the de- fendant prison officials, finding that they did not deprive Jackson of a liberty interest by placing him in solitary confine- ment for three months despite Jackson’s evidence describing appalling conditions in the solitary cell. On that question, we respectfully disagree with the district court, but we affirm the judgment for defendants. Even if a trial were to establish that Jackson was deprived of a liberty interest, defendants would still be entitled to qualified immunity from any damages rem- edy because the applicable law was not clearly established at the time they ordered him into solitary confinement. I. Factual and Procedural History A. The Incident at Stateville & the Adjustment Committee Hearing Because we are reviewing a grant of summary judgment, we give Jackson as the non-moving party the benefit of con- flicts in the evidence and draw all favorable inferences from it. Hardaway v. Meyerhoff, 734 F.3d 740, 743 (7th Cir. 2013). On February 25, 2020, Jackson was incarcerated at Stateville Cor- rectional Center. A correctional officer approached Jackson’s cell to close its “chuckhole,” a small opening just large enough for Jackson to reach both of his arms through. As the officer reached for the chuckhole door, Jackson placed one arm through the opening. A second guard then approached and tried to help close the door. Jackson testified that one of the No. 23-1703 3 officers grabbed his arms and banged them against the cell door and that another guard forcefully bent his finger. Video footage depicts one officer grabbing Jackson’s arm, but a guard’s body obscures most of the rest of the interaction. A few seconds later, a third correctional officer, defendant Marc Anastacio, approached the cell, shaking a chemical agent in his right hand. Jackson asserts that Anastacio ground the spray canister into Jackson’s hand. About thirty seconds later, a fourth correctional officer, defendant Shadi Awad, approached the cell door and began speaking to the other officers. Jackson recalls that Awad alerted them to the presence of a camera and told them to stop. Moments later, Jackson broke free of the officers. Anastacio then discharged a chemical agent into Jackson’s cell. Jackson says that Anastacio continued spraying the chemical agent at Jackson even after he had retreated com- pletely into his cell. Defendants dispute this, and the video evidence does not conclusively resolve the dispute. The entire interaction lasted just over one minute, until a correctional of- ficer secured the chuckhole door. Within just hours of the incident, Jackson was transferred from Stateville to the Pontiac Correctional Center. Jackson tes- tified that he received medical treatment for his resulting in- juries at Stateville before transfer and at Pontiac afterwards. Based on the chuckhole incident at Stateville, Anastacio is- sued Jackson a disciplinary ticket for a major infraction. That ticket triggered an “adjustment hearing” for Jackson at Pon- tiac. The hearing was conducted on March 13, 2020 by a com- mittee of two, defendants Travis Bantista and Jesus Madrigal. At the hearing, Jackson was permitted to ask questions and to tell his side of the story, but he was not allowed to call 4 No. 23-1703 witnesses or to view the video of the incident. After the hear- ing, the committee issued its recommended disciplinary measures along with the reasons for its decision, which was based on statements by the Stateville guards. The recom- mended discipline included the loss of one month of good- time credit; three- or six-month losses of in-person visitation, commissary access, and other privileges; and most relevant to this appeal, three months of disciplinary segregation, which we also call solitary confinement. Pontiac’s warden, defend- ant Leonta Jackson, approved the hearing committee’s deter- mination. 1 B. Procedural History Jackson filed this suit under 42 U.S.C. § 1983, claiming in relevant part that Bantista, Madrigal, and Warden Jackson (collectively, “the Pontiac defendants”) violated his Four- teenth Amendment rights by sentencing him to disciplinary segregation without sufficient process. The defendants moved for summary judgment on several grounds, including failure to establish a protected liberty interest, failure to show 1 Jackson’s good-time credits were later restored, and for good reason. Revoking a prisoner’s good-time credits without permitting him to call witnesses violates the Supreme Court’s holding in Wolff v. McDonnell, at least when calling witnesses would not be “unduly hazardous to institu- tional safety or correctional goals.” 418 U.S. 539, 566–67 (1974); see also Donelson v. Pfister, 811 F.3d 911, 917 (7th Cir. 2016) (“Due process requires that prisoners in disciplinary proceedings, before being deprived of good time, be allowed to call witnesses and present other evidence.” (citing Wolff, 418 U.S. at 566)). Restoring the good-time credits, however, is a suf- ficient remedy for such a procedural error. See Adams v. Reagle, 91 F.4th 880, 896 (7th Cir. 2024) (majority opinion of St. Eve, J., on procedural due process issue). We need not discuss the good-time credits further. No. 23-1703 5 that the procedures afforded in his hearing were inadequate, and qualified immunity. 2 The district court granted defendants’ summary judgment motion as to Jackson’s procedural due process claim, finding that Jackson could not establish that defendants deprived him of a liberty interest subject to the due process requirements of the Fourteenth Amendment by forcing him into three months of segregation. Jackson v. Vasquez, No. 20 C 6004, 2023 WL 319530, at *7 (N.D. Ill. Jan. 18, 2023). The court first concluded that the length of Jackson’s segregation, three months, was in- sufficient to create a liberty interest standing alone. Id. at *6. But it also recognized that Jackson could still establish he was deprived of a liberty interest if the segregation caused him to suffer “unusually harsh conditions of confinement or addi- tional punishments….” Id., citing Kervin v. Barnes, 787 F.3d 833, 836–37 (7th Cir. 2015). Jackson provided evidence that the conditions of his seg- regation were indeed unusually harsh. In a declaration op- posing defendants’ motion for summary judgment, Jackson asserted that his disciplinary segregation cell, unlike Pontiac’s general population area, had feces and urine on the walls, constant noise with inmates banging on cell doors, water con- taminated with bacteria that causes Legionnaire’s disease, and roaches and mice. He also said that inmates in the disci- plinary segregation cells, unlike inmates in general 2 Jackson also alleged the Stateville guards used wanton and unnec- essary force in violation of the Eighth Amendment. A jury rendered a ver- dict against defendant Anastacio for $2,500 in compensatory damages and $5,000 in punitive damages. That claim is not at issue in this appeal. 6 No. 23-1703 population, throw feces and urine at other inmates when they are in the hallways. Jackson, 2023 WL 319530, at *6. Comparing the conditions of solitary confinement as testi- fied by Jackson to conditions considered in previous cases, the district court found that “when coupled with his relatively short segregation time, he has not sufficiently raised a genu- ine issue of material fact regarding whether he suffered” the sort of “atypical and significant hardship” required to estab- lish a liberty interest protected by procedural due process. Id. at *7; see generally Sandin v. Conner, 515 U.S. 472, 484 (1995) (adopting “atypical and significant hardship” standard for finding protected liberty interest). Because Jackson failed to raise a genuine dispute of material fact regarding the exist- ence of a liberty interest, the district court granted defendants’ motion for summary judgment on the merits of Jackson’s due process claim without reaching the defense of qualified im- munity. This appeal followed. 3 II. Analysis We review de novo a district court’s grant of summary judgment on the ground that the plaintiff was not deprived of a liberty interest. Thomas v. Ramos, 130 F.3d 754, 759 (7th Cir. 1997). We explain first why Jackson has raised a genuine issue of fact about whether he was deprived of a protected liberty interest. Our explanation of the applicable law will lay the foundation for our decision on qualified immunity because any transgression of Jackson’s due process rights did not vio- late clearly established law. 3 We thank Abigail R. Van Hook, Terrence P. Canada, and the law firm now known as Troutman Pepper Locke LLP for their capable service to plaintiff Jackson and to the district court and this court. No. 23-1703 7 A. The Two-Prong Qualified Immunity Framework Qualified immunity aims to strike a balance “between the interests in vindication of citizens’ constitutional rights and in public officials’ effective performance of their duties ….” An- derson v. Creighton, 483 U.S. 635, 639 (1987) (internal quotation marks omitted), quoting Davis v. Scherer, 468 U.S. 183, 195 (1984). To determine whether qualified immunity shields the actions of public officials, “the courts employ a two-prong test: (1) whether the facts, viewed in a light most favorable to the injured party, demonstrate that the conduct of the officers violated a constitutional right, and (2) whether that right was clearly established at the time the conduct occurred.” Harda- way, 734 F.3d at 743, citing Pearson v. Callahan, 555 U.S. 223, 232 (2009). We have discretion to address the two qualified immunity prongs in whichever order is appropriate. We may use that discretion to resolve step-one rights questions first, especially when avoiding them would “leave standards of of- ficial conduct permanently in limbo.” Smith v. Kind, 140 F.4th 359, 365 (7th Cir. 2025) (internal quotation marks omitted), quoting Camreta v. Greene, 563 U.S. 692, 706 (2011). In Camreta, the Supreme Court explained that such step- one merits rulings are not mere dicta or ‘statements in opinions.’ They are rulings that have a significant future effect on the conduct of public officials—both the prevailing parties and their co-workers— and the policies of the government units to which they belong. And more: they are rulings self-consciously designed to produce this effect, by establishing controlling law and preventing invocations of immunity in later cases. And still 8 No. 23-1703 more: they are rulings designed this way with this Court’s permission, to promote clarity— and observance—of constitutional rules. 563 U.S. at 704–05 (citations omitted), quoting California v. Rooney, 483 U.S. 307, 311 (1987). Whether and when solitary confinement deprives a pris- oner of a protected liberty interest has been and will continue to be a difficult problem for courts, so qualified immunity of- ten applies even where a prisoner’s rights have been violated. E.g., Hardaway, 734 F.3d at 744 (affirming grant of qualified immunity and agreeing with district court that liberty interest issue “is not at all clear except at the fringes”). We therefore address the first prong on the merits without skipping ahead to the “clearly established law” question. As instructed by Camreta, we have repeatedly thought hard about whether to decide the merits issue presented in this appeal, when solitary confinement affects a protected lib- erty interest. See 563 U.S. at 707. In Hardaway, more than a decade ago, we explained that while district courts “would benefit from a bright-line rule on the types of conditions and durations of segregation [that] give rise to a prisoner’s liberty interest, no such guidance has yet to be specifically addressed by this Court.” 734 F.3d at 745. We are still unable to set out a bright-line rule for solitary confinement cases, but this case is “one of those cases in which it is appropriate for the court to exercise its discretion to avoid avoidance.” Toevs v. Reid, 685 F.3d 903, 910 (10th Cir. 2012) (deciding first prong of qualified immunity in case concerning prisoner’s procedural due pro- cess right to periodic review of prolonged administrative seg- regation). No. 23-1703 9 B. Violation of a Constitutional Right We first ask whether the plaintiff has shown a violation of a constitutional right—or to be more precise, whether he has offered evidence that would allow a reasonable jury to find that he suffered a violation of a constitutional right. See Sauc- ier v. Katz, 533 U.S. 194, 201 (2001), overruled in part on other grounds by Pearson, 555 U.S. at 236. We note that this appeal does not present the closely related questions of whether and when conditions of solitary confinement amount to cruel or unusual punishment under the Eighth Amendment, regard- less of procedures used to impose those conditions. 1. Liberty Interests in Avoiding Disciplinary Segregation Prisoners may “claim the protections of the Due Process Clause. They may not be deprived of life, liberty or property without due process of law.” Wolff v. McDonnell, 418 U.S. 539, 556 (1974). To proceed on a due process claim, a prisoner must demonstrate: “(1) the deprivation of a liberty interest; and (2) the procedures he was afforded were constitutionally de- ficient.” Lisle v. Welborn, 933 F.3d 705, 720 (7th Cir. 2019). In Wolff, the Supreme Court held that a prisoner is entitled to limited procedural protections before he is deprived of his liberty interest in good-time credits as a disciplinary measure. 418 U.S. at 556–57. The Wolff Court also taught that those con- stitutional safeguards should apply when inmates face disci- plinary solitary confinement: The deprivation of good time and imposition of ‘solitary’ confinement are reserved for instances where serious misbehavior has occurred. This appears a realistic approach, for it would be diffi- cult for the purposes of procedural due process to 10 No. 23-1703 distinguish between the procedures that are required where good time is forfeited and those that must be extended when solitary confinement is at issue. The latter represents a major change in the condi- tions of confinement and is normally imposed only when it is claimed and proved that there has been a major act of misconduct. Here, as in the case of good time, there should be minimum pro- cedural safeguards as a hedge against arbitrary de- termination of the factual predicate for imposition of the sanction. Id. at 571 n.19 (emphases added). After Wolff, two Supreme Court decisions directly ad- dressed whether and when assignment to some form of seg- regation or solitary confinement could deprive a prisoner of a liberty interest that would trigger due process protections: Wilkinson v. Austin, 545 U.S. 209 (2005), and Sandin v. Conner, 515 U.S. 472 (1995). See Marion v. Columbia Correctional Institu- tion, 559 F.3d 693, 697 (7th Cir. 2009) (summarizing Wilkinson and Sandin). In Sandin, the Supreme Court held that thirty days of disciplinary segregation “did not present the type of atypical, significant deprivation in which a State might con- ceivably create a liberty interest.” 515 U.S. at 486. In Wilkinson, prisoners were assigned to a maximum-security prison, placed in segregated confinement for an indefinite duration, and “deprived of almost any environmental or sensory stim- uli and of almost all human contact.” 545 U.S. at 214. The Su- preme Court wrote that although “any of these conditions standing alone might not be sufficient to create a liberty inter- est, taken together they impose an atypical and significant hardship within the correctional context.” Id. at 224. No. 23-1703 11 As we have explained, “Sandin and Wilkinson establish that disciplinary segregation can trigger due process protec- tions depending on the duration and conditions of segrega- tion.” Marion, 559 F.3d at 697. In addition, “we have declined to read Wilkinson’s holding as being limited to its specific facts.” Id. We must therefore analyze “the combined import of the duration of the segregative confinement and the condi- tions endured by the prisoner during that period” to deter- mine whether solitary confinement deprives a prisoner of a liberty interest warranting due process protections. Id. A lib- erty interest may arise “if the length of segregated confine- ment is substantial and the record reveals that the conditions of confinement are unusually harsh.” See id. at 697–98. There is a lot of room between the thirty days of solitary confinement in Sandin and the indefinite, near-total sensory deprivation in Wilkinson. The Supreme Court has left it to lower courts to chart that territory. Our cases have empha- sized continued sensitivity to stints in solitary confinement measured in months. E.g., Ealy v. Watson, 109 F.4th 958, 964– 65 (7th Cir. 2024) (assuming liberty interest implicated by five months in disgusting solitary conditions); Marion, 559 F.3d at 698–99 & n.3 (roughly eight months of solitary confinement required remand for factual inquiry into liberty interest; col- lecting cases remanding where solitary confinement lasted as few as 75 days). But the combination of qualified immunity’s “clearly es- tablished” requirement layered on top of the multi-factor ap- proach from Sandin and Wilkinson means that for any term of solitary confinement lasting more than days but less than years, qualified immunity will often apply to claims for dam- ages. E.g., Hardaway, 734 F.3d at 745. For segregation terms 12 No. 23-1703 measured in months, even if an inmate’s segregation amounted to a deprivation of a liberty interest, it is unlikely that prison officials could be “held responsible for incorrectly guessing otherwise due to the ambiguity of the parameters of the law.” Id. 2. The Value of a Duration-Based Rule The Sandin, Wilkinson, and Marion line of cases requires us to consider both “the duration of the segregative confinement and the conditions endured.” Marion, 559 F.3d at 699. It is im- portant to emphasize that the significance of unusually appal- ling physical conditions diminishes when courts confront prolonged terms of segregation. When a “prisoner is sub- jected to a lengthy period of segregation, the duration of that confinement itself may be atypical and significant.” Marion, 559 F.3d at 699 n.4 (internal quotation marks omitted), quot- ing Trujillo v. Williams, 465 F.3d 1210, 1225 (10th Cir. 2006). Marion recognized that a long term of solitary confinement might suffice by itself to trigger a liberty interest. “Indeed, other courts of appeals have held that periods of confinement that approach or exceed one year may trigger a cognizable lib- erty interest without any reference to conditions.” Id. at 698– 99 & n.4, citing Iqbal v. Hasty, 490 F.3d 143, 161 (2d Cir. 2007) (segregation of 305 days or more necessarily implicates liberty interest), overruled on other grounds sub nom. Ashcroft v. Iq- bal, 556 U.S. 662 (2009); Trujillo, 465 F.3d at 1225 (reversing dismissal of claim involving 750 days’ solitary); Williams v. Fountain, 77 F.3d 372, 374 n.3 (11th Cir. 1996) (assuming that one year implicates liberty interest); see also Brown v. Oregon Dep’t of Corrections, 751 F.3d 983, 988 (9th Cir. 2014) (twenty- seven months of disciplinary segregation sufficed to implicate liberty interest); Perry v. Spencer, 94 F.4th 136, 159–60 (1st Cir. No. 23-1703 13 2024) (en banc) (same for fifteen months of administrative seg- regation); Colon v. Howard, 215 F.3d 227, 231 (2d Cir. 2000) (same for 305 days). 4 We have not yet affirmatively adopted any minimum du- ration of disciplinary segregation that automatically impli- cates a liberty interest, regardless of other conditions of con- finement. But some of our precedents have outlined durations that will not suffice by themselves—that is, without also showing that the conditions of the confinement were atypi- cally harsh. For instance, we held in Bryan v. Duckworth that a year in disciplinary segregation required remand to deter- mine whether “conditions in segregation were considerably harsher than those of the normal prison environment.” 88 F.3d 431, 433 (7th Cir. 1996), abrogated on other grounds as recognized by Diaz v. Duckworth, 143 F.3d 345, 346 (7th Cir. 1998). Similarly, in Wagner v. Hanks, we held that a year in sol- itary confinement could implicate a liberty interest, but we 4 In Colon, the Second Circuit held that more than 305 days in solitary confinement implicates a liberty interest and that durations between 101 and 305 days require the development of a detailed factual record. 215 F.3d at 231–32. In subsequent decisions, the Second Circuit has outlined a test akin to burden-shifting, where durations longer than approximately six months shift the burden of proving “typicality” back onto the officials seeking qualified immunity. See, e.g., J.S. v. T’Kach, 714 F.3d 99, 106 (2d Cir. 2013) (“confinement of 188 days is a significant enough hardship” to create a liberty interest “[i]n the absence of factual findings to the con- trary”). In Perry v. Spencer, 94 F.4th 136 (1st Cir. 2024) (en banc), the First Circuit adopted this burden-shifting approach in a case involving admin- istrative segregation. That court noted that “fifteen months of solitary con- finement” would suffice to establish a liberty interest “in the absence of the defendants making any contrary showing regarding the frequency of use of such prolonged solitary confinement for administrative reasons.” Id. at 159. 14 No. 23-1703 flagged issues with defining the relevant comparison group when assessing whether such conditions were “atypical.” 128 F.3d 1173, 1174, 1177 (7th Cir. 1997). The reasoning of those cases implies that a year in segregation does not automatically implicate a prisoner’s liberty interest. We have reached more explicit conclusions on shorter periods of segregation. See, e.g., Marion, 559 F.3d at 698 (“six months of segregation is ‘not such an extreme term’ and, standing alone, would not trigger due process rights”), quoting Whitford v. Boglino, 63 F.3d 527, 533 (7th Cir. 1995); Lisle, 933 F.3d at 721 (same for four months); Lekas v. Briley, 405 F.3d 602, 612 (7th Cir. 2005) (same for 90 days). In reasoning that solitary confinement of up to one year, standing alone, is not sufficient to implicate a liberty interest, Bryan and Wagner probably lie close to the outer limit under current law. A one-year threshold is consistent with our recognition in Marion that other circuits have held that “peri- ods of confinement that approach or exceed one year may trigger a cognizable liberty interest without any reference to conditions.” 559 F.3d at 699 (footnote omitted). It is also roughly consistent with the First Circuit’s fifteen-month threshold adopted in Perry, 94 F.4th at 159, and the Second Circuit’s 305-day boundary adopted in Colon, 215 F.3d at 231. 5 Long terms of solitary confinement are not necessarily un- constitutional, at least not yet. But they call for at least some due process protections that apply when prison discipline 5 For reasons addressed in the separate concurring opinion, we and other courts may find in the foreseeable future that we need to adopt a much shorter presumptive limit based on evolving research on the effects of prolonged solitary confinement, as well as international norms and do- mestic legal trends. No. 23-1703 15 deprives a prisoner of a liberty interest. See generally Wolff, 418 U.S. at 556–58 & 571 n.19. 3. Jackson’s Evidence of a Protected Liberty Interest Jackson’s assignment to solitary confinement was for three months, so whether he was deprived of a liberty interest de- pends, under Marion, on “the combined import of the dura- tion of the segregative confinement and the conditions en- dured by [him] during that period.” See 559 F.3d at 697. Jackson has raised a genuine issue of material fact as to whether his segregation at Pontiac was an “atypical or significant hardship in relation to the ordinary incidents of prison life” under Sandin. 515 U.S. at 484. Jackson swore in an affidavit that his disciplinary segregation cell, unlike Pontiac’s general population cells, had feces and urine on the walls, constant noise with inmates banging on cell doors, water contaminated with bacteria that causes Legionnaire’s disease, and roaches and mice. He also testified that inmates in the disciplinary segregation cells, unlike inmates in general population, throw feces and urine at other inmates. Jackson, 2023 WL 319530, at *6. On a motion for summary judgment, we find it reasonable to infer in Jackson’s favor that he personally experienced the deplorable conditions he described in his declaration opposing summary judgment. When considering a prisoner’s liberty interest in avoiding conditions of segregation, we have distinguished between al- legations of “disgusting conditions” versus conditions that “although restrictive, were not unsanitary or otherwise dis- gusting.” See Kervin, 787 F.3d at 836, citing Beverati v. Smith, 120 F.3d 500, 504 (4th Cir. 1997), and Hatch v. District of Colum- bia, 184 F.3d 846, 858 (D.C. Cir. 1999). Similarly, in Taylor v. 16 No. 23-1703 Riojas, which summarily reversed qualified immunity for prison officials, the Supreme Court held that “no reasonable correctional officer could have concluded that” it was consti- tutional under the Eighth Amendment to house a prisoner for six days in “deplorably unsanitary conditions.” 592 U.S. 7, 7– 9 (2020). While the Eighth Amendment question in Taylor is different from the Fourteenth Amendment question pre- sented in this case, the summary reversal in Taylor reinforced the principle that prison officials do not have unlimited dis- cretion to confine inmates in deplorable conditions. While the conditions described by Jackson are not quite as awful as those in Taylor, the Pontiac cell block described in Jackson’s affidavit was unequivocally “disgusting,” Kervin, 787 F.3d at 836, and his assignment to solitary confinement there lasted fifteen times longer than the plaintiff’s confine- ment in Taylor. The unsanitary conditions Jackson describes were far worse, and were described more specifically, than the “vague description of [the plaintiff’s] cell, including rust on the bars and ‘corroded feces’ in the toilet” that we found insufficient to implicate a liberty interest in Lisle, 933 F.3d at 721 (combined with four-month term in solitary). Jackson presents facts similar to those in Ealy v. Watson. The prisoner-plaintiff in Ealy was placed in segregation for five months and was confined in several cells, including one with bugs and spider webs, a lack of heat, and faulty plumb- ing “such that other inmates’ waste flowed up into [his] toilet, causing foul odors.” 109 F.4th at 963. We declined to decide expressly whether those conditions implicated Ealy’s liberty interests. We assumed that they did and instead resolved the case on the ground that “Ealy received the process he was due.” Id. at 966–67. Our assumption in Ealy provides support No. 23-1703 17 for the idea that segregation for a matter of months in repul- sive and unsanitary conditions can deprive a prisoner of pro- tected liberty. We intend to make that explicit today. A reasonable jury could conclude that the combined ef- fects of Jackson’s three-month assignment to disciplinary seg- regation and the conditions of his segregation imposed what Sandin called “an atypical and significant hardship.” 515 U.S. at 484. If so, then Jackson was deprived of a liberty interest entitling him to the “minimum procedures appropriate under the circumstances” to ensure the “protection of the individual against arbitrary action of government.” Wolff, 418 U.S. at 557, 558. He argues that he did not receive those procedures before this punishment was imposed. We do not decide that issue because, as we explain next, the defendants are entitled to qualified immunity on the liberty interest question. C. Whether Defendants Violated Clearly Established Law For the second prong of qualified immunity, we assume Jackson can prove he was deprived of a liberty interest. The question is whether the defendants violated law that was clearly established at the time they imposed this term of solitary confinement. The answer is no. Jackson fails to meet his burden on the second prong of qualified immunity, and defendants are entitled to summary judgment on his due process claim. “The affirmative defense of qualified immunity protects government officers from liability for actions taken in the course of their official duties if their conduct does not violate ‘clearly established statutory or constitutional rights of which a reasonable person would have known.’” Hardaway, 734 F.3d at 743, quoting Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982). 18 No. 23-1703 To defeat qualified immunity by showing that a constitutional right is “clearly established,” see Harlow, 457 U.S. at 818, the “contours of the right must be sufficiently clear that a reason- able official would understand that what he is doing violates that right.” Anderson v. Creighton, 483 U.S. 635, 640 (1987). Jackson “bears the burden of identifying ‘case law that has both articulated the right at issue and applied it to the factual circumstance similar to the one at hand.’” Hardaway, 734 F.3d at 744, quoting Boyd v. Owen, 481 F.3d 520, 526 (7th Cir. 2007). This “clearly established” requirement holds even where the constitutional merits depend on “fact-specific” questions. An- derson, 483 U.S. at 640–41. In sufficiently obvious cases of con- stitutional violations, however, qualified immunity will not be available even in the absence of a factually close precedent. Hope v. Pelzer, 536 U.S. 730 (2002). Under our case law at the time of Jackson’s hearing in 2020, for almost any term of disciplinary segregation meas- ured in months, consideration of the conditions of confine- ment was required to determine whether a liberty interest ex- isted. In Marion, we noted that six months of segregation, again standing alone, would not trigger due process rights. 559 F.3d at 698. And in Kervin, we cautioned that a liberty in- terest could be established for terms of segregation less than six months “depending on the conditions of confinement and on any additional punishments.” 787 F.3d at 836–37; see also Hardaway, 734 F.3d at 743 (“Although relatively short terms of segregation rarely give rise to a prisoner’s liberty interest, at least in the absence of exceptionally harsh conditions, such an interest may arise from a long term of confinement combined with atypical and significant hardships.”); Ealy, 109 F.4th at 964 (“five months in segregation, standing alone, is not enough to implicate a liberty interest”). No. 23-1703 19 Because the duration of Jackson’s disciplinary segrega- tion, three months, is not long enough to deprive him of a lib- erty interest on its own, defeating a qualified immunity de- fense required Jackson to show that violations have been es- tablished under “facts not distinguishable in a fair way from the facts presented in the case at hand.” Figgs v. Dawson, 829 F.3d 895, 905 (7th Cir. 2016), quoting Campbell v. Peters, 256 F.3d 695, 701 (7th Cir. 2001), abrogated on other grounds by Pearson, 555 U.S. 223. Put another way, the question is whether the defendant members of Jackson’s disciplinary committee should have known they needed to provide him due process protections. See Hardaway, 734 F.3d at 744. Jackson acknowledges this is not an easy task. He notes in his brief that “this Court has yet to establish clear standards for how harsh conditions in segregation need to be to satisfy the Sandin hardship analysis.” Appellant’s Br. at 9. Our comment over a decade ago in Hardaway remains true: the jurisprudence regarding what presents “an atypical and significant hardship” is “not at all clear except at the fringes.” 734 F.3d at 744 (internal quotation marks omitted), quoting Hardaway v. Meyerhoff, No. 10-CV-556-JPG, 2012 WL 2520953, at *3 (S.D. Ill. June 27, 2012). Jackson also acknowledges that he is not among the few plaintiffs whose particular combinations of duration and conditions of segregation have already come before this court. Appellant’s Br. at 10. Jackson’s acknowledgment that no case has previously considered factual circumstances closely comparable to his confirms defendants’ entitlement to qualified immunity. 20 No. 23-1703 Jackson’s liberty interest was not “clearly established” by the case law of this court or the Supreme Court in March 2020. 6 On the conditions factor, the conditions Jackson alleges here, though “more severe than those found in the general prison population,” are “hardly analogous to a confinement that deprives a prisoner of all human contact or sensory stim- uli,” Hardaway, 734 F.3d at 744, like the conditions the Su- preme Court found sufficient to create a liberty interest in Wil- kinson. And on the duration factor, Jackson has not “presented case law stating that a [three]-month period of confinement under conditions similar to [his] implicates a liberty interest.” See id. The closest cases are a handful of non-precedential orders from this court and federal district courts in Illinois that were identified in the district court’s order granting summary judg- ment. See Jackson, 2023 WL 319530, at *7, citing Obriecht v. Raemisch, 565 F. App’x 535, 540 (7th Cir. 2014) (no liberty in- terest where defendant “submitted a declaration recounting deplorable conditions (in particular having to sleep on a mat- tress placed directly on the wet floor)” but was “released from segregation after only 78 days”); Whitfield v. Atchinson, No. 13- cv-653-SMY-RJD, 2017 WL 3707180, at *5 (S.D. Ill. Aug. 28, 2017) (no liberty interest where segregation lasted three 6 See Reed v. Palmer, 906 F.3d 540, 547 (7th Cir. 2018) (“We look first to controlling Supreme Court precedent and our own circuit decisions on the issue. If no controlling precedent exists, we broaden our survey to include all relevant caselaw in order to determine whether there was such a clear trend in the caselaw that we can say with fair assurance that the recogni- tion of the right by a controlling precedent was merely a question of time.”) (internal quotations and alteration omitted), quoting Jacobs v. City of Chicago, 215 F.3d 758, 767 (7th Cir. 2000). No. 23-1703 21 months, with conditions including a steel door, cellmates with mental health issues, unpleasant odors, constant noise, and other inmates throwing feces at him); McKinley v. Atchison, No. 3:16-cv-661-NJR-MAB, 2019 WL 4744839, at *7 (S.D. Ill. Sept. 30, 2019) (liberty interest implicated where con- ditions including mice and cockroach infestation, cracked window, no heat, no hot water, and no cleaning supplies lasted more than three years). These cases do not defeat qualified immunity here. First, these cases are of limited value in demonstrating that Jack- son’s liberty interest was “clearly established.” See Anderson v. Romero, 72 F.3d 518, 525 (7th Cir. 1995) (district court and non-precedential circuit court decisions generally do not clearly establish law). More substantively, even among these non-precedential orders with comparable conditions, Jack- son’s three-month duration of confinement was closer to Obriecht and Whitfield in which the courts declined to find lib- erty interests, than to McKinley, where the court did find a lib- erty interest. Jackson has not identified any case predating his March 2020 adjustment hearing finding a liberty interest in circumstances resembling the three-month duration and de- plorable conditions of his segregation at Pontiac. Thus, he fails to overcome the defense of qualified immunity. * * * Jackson presented evidence showing that he was subjected to three months in appalling conditions of solitary confine- ment at Pontiac. Taking that evidence as true, Jackson had a liberty interest in avoiding the described conditions, so prison officials should have afforded him due process. Even so, “the Defendants should not be held responsible for incorrectly guessing otherwise due to the ambiguity of the parameters of 22 No. 23-1703 the law.” Hardaway, 734 F.3d at 745. The defendants are there- fore entitled to qualified immunity on Jackson’s due process claim. Moving forward, such short terms of solitary confine- ment combined with comparable disgusting conditions will suffice to show a loss of protected liberty requiring procedural protections, as will longer terms of solitary confinement alone. The judgment of the district court is AFFIRMED. No. 23-1703 23 HAMILTON, Circuit Judge, joined by ROVNER, Circuit Judge, concurring. In 2015, the United Nations General Assembly unanimously adopted the Standard Minimum Rules for the Treatment of Prisoners, also called the Nelson Mandela Rules. Those rules, which were sponsored by the United States, pro- hibit as a form of torture solitary confinement of more than fifteen consecutive days. 1 Yet federal courts in the United States routinely hold that terms of solitary confinement meas- ured in years do not violate the Eighth Amendment prohibi- tion on cruel and unusual punishment. We also find that in- mates rarely have a constitutionally protected liberty interest in avoiding such restrictive conditions. The lead opinion resolves this appeal correctly under cur- rent precedent and as the issues were framed for us. I write separately (a) to put this case in a broader context of constitu- tional law and prison practice, (b) to link procedural treat- ment of solitary confinement to substantive limits under the Eighth Amendment, and (c) to suggest how our case law un- der the Eighth and Fourteenth Amendments should develop in future cases. We know that prolonged periods of solitary confinement can have harmful, even devastating, effects on prisoners. Such prolonged punishments may be justified in extreme cases, but they should become much rarer than they are under current law and practice. We should begin bringing our Eighth Amendment law and actual prison practices into 1 See G.A. Res. 70/175, at 15–16 (https://perma.cc/8ZRB-6SA9) (Dec. 17, 2015); U.N. Comm’n on Crime Prevention and Criminal Justice, Mandela Rules, U.N. Doc. E/CN.15/2015/L.6/Rev.1 (https://perma.cc/TBA9-AW7E) (May 21, 2015) (U.N. Commission, including United States, recommend- ing adoption of Mandela Rules). 24 No. 23-1703 line with evolving standards of human decency and with the Nelson Mandela Rules. I. Ineffective Solitary Confinement Case Law As explained in the lead opinion, this appeal is framed to test only whether plaintiff Abre Jackson was entitled to pro- cedural protections before he was ordered to spend three months in solitary confinement in filthy, appalling conditions. E.g., Wilkinson v. Austin, 545 U.S. 209, 224 (2005) (finding lib- erty interest established and asking what process was due). 2 The Supreme Court has framed the due process problem for solitary confinement in two principal cases from a gener- ation ago, Sandin v. Conner, 515 U.S. 472 (1995), and Wilkinson v. Austin, 545 U.S. 209 (2005). Sandin held that a prisoner dis- ciplined with thirty days in solitary confinement was not en- titled to due process protections because prison officials did 2 Most cases cited in this opinion (like Jackson’s appeal) speak in terms of “segregation.” That is a euphemism for solitary confinement. In a typi- cal “segregation,” a prisoner is held in “a windowless cell no larger than a typical parking spot for 23 hours a day; and in the one hour when he leaves it, he likely is allowed little or no opportunity for conversation or interaction with anyone.” Davis v. Ayala, 576 U.S. 257, 287 (2015) (Ken- nedy, J., concurring); Wallace v. Baldwin, 895 F.3d 481, 482–83 (7th Cir. 2018) (noting euphemism and quoting plaintiff describing “segregation” as “akin to being sealed inside a coffin”). In 2015 and 2016, experts esti- mated that somewhere between 60,000 and 100,000 prisoners were held in solitary confinement. Judith Resnik et al., Punishment in Prison: Constitut- ing the “Normal” and the “Atypical” in Solitary and Other Forms of Confine- ment, 115 Northwestern Univ. L. Rev. 45, 77 (2020); see also Apodaca v. Raemisch, 586 U.S. 931, 935 (2018) (statement of Sotomayor, J., regarding denial of certiorari) (noting similar statistics). In 2017, almost 2,000 of those individuals had been in isolation for more than six years. Resnik, supra, 115 Northwestern Univ. L. Rev. at 78. No. 23-1703 25 not impose an “atypical and significant hardship” as com- pared to the “ordinary incidents of prison life.” 515 U.S. at 484. Wilkinson held that prisoners were entitled to modest pro- cedural protections when they faced transfers for indefinite durations to a “supermax” prison, which would subject them to “extreme isolation,” deprive them of “almost any environ- mental or sensory stimuli,” and render them ineligible for pa- role. 545 U.S. at 214–15. As the lead opinion explains, Sandin and Wilkinson leave much unclear. In due process cases like this one, the Supreme Court has left it to lower courts to figure out just when the duration and conditions of solitary confine- ment combine to impose a sufficiently “atypical and signifi- cant hardship” to require modest procedural protections. E.g., Marion v. Columbia Correctional Inst., 559 F.3d 693, 696 (7th Cir. 2009). The Sandin standard was adopted in response to problems with a different approach from Hewitt v. Helms, 459 U.S. 460 (1983), which focused on the phrasing of prison regulations and policies. The Sandin Court wrote that the Hewitt standard had encouraged States to give prison officials standardless discretion yet also invited undue federal court intrusion into day-to-day prison management. 515 U.S. at 481–83. Yet the Sandin standard has given rise to its own set of problems. For example, we have not been able to set effective parameters on when a period of solitary confinement is long enough to trigger due process protections regardless of other conditions. See ante at 12–14. Nor have we been able to pro- vide clear guidance on how bad conditions in solitary confine- ment must be to require procedural protections for shorter durations. See ante at 15–17. 26 No. 23-1703 A further debate addresses what procedures are due when a prisoner is punished with solitary confinement. In Wolff, the Supreme Court taught that its new standards for depriving a prisoner of good-time credit should also apply to solitary con- finement punishments. 418 U.S. at 571 n.19. In Sandin, the Court found that the punishment of thirty days in solitary confinement did not entitle the prisoner to “the procedural protections set forth in Wolff.” 515 U.S. at 487. In Wilkinson, however, the Court shifted focus from solitary confinement as an individual punishment to a State’s broader practices in de- ciding which prisoners to assign to its new “supermax” prison. After finding that the assignment would deprive pris- oners of a protected liberty interest, the Court held that pro- cedures less formal and adversarial than the Wolff procedures would suffice. 545 U.S. at 225–29. We followed that approach in Westefer v. Neal, 682 F.3d 679, 684–86 (7th Cir. 2012), allow- ing Illinois to use less formal and adversarial procedures in assigning prisoners to its own “supermax” prison, so long as they received some notice of the reasons for the assignment, an opportunity to respond to them, and a periodic review of their placements. Judge Scudder points to Adams v. Reagle, 91 F.4th 880, 895 (7th Cir. 2024) (majority opinion of St. Eve, J.), where the panel majority, without analysis of the point, extended the informal and non-adversarial Wilkinson and Westefer procedures for large-scale prison assignments to an individual punishment of many months of solitary confinement. In extending Wil- kinson and Westefer on this point, the Adams panel did not ad- dress the Supreme Court’s different teaching in Wolff and Sandin. Nor did it engage with our other precedents, includ- ing Kervin, Marion, and Lisle, indicating that the more formal No. 23-1703 27 Wolff procedures would apply to cases of individual punish- ment with prolonged solitary confinement. We have also struggled to explain what it means for a term of solitary confinement to impose an “atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life.” Sandin, 515 U.S. at 484. What is an ordinary inci- dent of prison life? Measured by which prisoner? In which prison? We do not know. See, e.g., Wilkinson, 545 U.S. at 223 (noting the “difficulty of locating the appropriate baseline”); Lekas v. Briley, 405 F.3d 602, 609–10 (7th Cir. 2005) (liberty in- terest implicated only if inmate’s confinement is worse than that authorized in “the most restrictive prison in the state pe- nal system”); Hardaway v. Meyerhoff, 734 F.3d 740, 745 (7th Cir. 2013) (quoting plaintiff’s brief and agreeing that “there is am- biguity among various Seventh Circuit cases regarding the proper baseline against which to measure conditions of disci- plinary confinement”) (internal quotation marks omitted). Judge Scudder’s opinion identifies another important problem in these due process claims—calling it a “legal para- dox”—albeit one that seems to be baked into the Supreme Court’s “atypical and significant hardship” standard from Sandin. The problem is that defendant prison officials who or- der a prisoner into solitary confinement may not be responsi- ble for, and may not even know, the actual conditions of soli- tary confinement they are imposing. Yet those actual condi- tions could be decisive under Sandin, Wilkinson, Marion, and similar cases looking at the combination of duration and con- ditions to determine whether the prison has been deprived of a protected liberty interest. See post at 57. And if a prisoner seeks damages under section 1983 against those officials, he will need to prove that they knew or perhaps should have 28 No. 23-1703 known of those conditions. Judge Scudder also calls this “backward” reasoning. Post at 52. There is some merit to that criticism, but again, it is inherent in the due process test of Sandin and Wilkinson, which teaches us to look at the combi- nation of duration and conditions of the solitary confinement to decide whether a prisoner has been deprived of a protected liberty interest. 3 3 Another complication in this field is that terms of solitary confine- ment can be divided into two broad categories: administrative and disci- plinary. Administrative segregation is not punitive but is imposed for prison management reasons, such as the safety of the prisoner. Jackson’s appeal challenges disciplinary segregation—solitary confinement im- posed as punishment for a prison infraction. Some courts have held that when state laws and prison guidelines authorize long terms of adminis- trative segregation, it should be seen as an ordinary and expected condi- tion of incarceration. As a result, administrative segregation raises fewer due process issues. See, e.g., Townsend v. Fuchs, 522 F.3d 765, 771 (7th Cir. 2008) (“Indeed, there is nothing ‘atypical’ about discretionary segregation; discretionary segregation is instead an ‘ordinary incident of prison life’ that inmates should expect to experience during their time in prison.”); but see Wilkinson, 545 U.S. at 224 (administrative assignment to supermax, with indefinite segregation, raised due process issue). Because adminis- trative segregation may be imposed at will, some courts have also held that terms of disciplinary segregation that do not exceed a possible term of administrative segregation do not raise due process issues. E.g., Sandin, 515 U.S. at 486. For clarity, the cases cited in the due process discussion of this opinion deal only with disciplinary segregation unless otherwise noted. But terms of both administrative and disciplinary segregation should pose Eighth Amendment issues, as discussed below. Some inmates have been sub- jected to unconscionable terms of solitary confinement for administrative reasons. See, e.g., Wilkerson v. Goodwin, 774 F.3d 845, 855 (5th Cir. 2014) (affirming denial of qualified immunity when plaintiff was in administra- tive segregation for 39 years); Estate of DiMarco v. Wyoming Dep’t of Correc- tions, 473 F.3d 1334, 1340–41 & n.5 (10th Cir. 2007) (collecting cases finding No. 23-1703 29 This court and other courts have struggled with these problems for decades. We have not yet been able to offer ei- ther clear guidance for prison officials or meaningful reme- dies for prisoners who are subjected to objectively inhumane conditions. Take the narrow question when a term of solitary confinement alone (with otherwise tolerable conditions) can entitle a prisoner to due process protections. We have ex- pressed approval of decisions from other circuits holding that “periods of confinement that approach or exceed one year may trigger a cognizable liberty interest without any refer- ence to conditions.” Marion, 559 F.3d at 699 (footnote omit- ted). In a more recent non-precedential order, we said that eight months was a “duration long enough to implicate a lib- erty interest….” Williams v. Brown, 849 F. App’x 154, 157 (7th Cir. 2021). But we have also held that inmates do not have a liberty interest in avoiding up to a year in solitary confine- ment, at least not without further inquiry into the conditions of confinement. E.g., Bryan v. Duckworth, 88 F.3d 431, 433–34 (7th Cir. 1996), abrogated on other grounds, Diaz v. Duck- worth, 143 F.3d 345, 346 (7th Cir. 1998). Decisions by our colleagues in other circuits are similarly unsettled. As the lead opinion explains, the First and Second Circuits have said that periods of disciplinary solitary con- finement exceeding roughly a year presumptively implicate an inmate’s due process liberty interest. Ante at 12–13, citing Perry v. Spencer, 94 F.4th 136, 159 (1st Cir. 2024) (en banc); Co- lon v. Howard, 215 F.3d 227, 231–32 (2d Cir. 2000). Other cir- cuits have assumed the same. E.g., Williams v. Fountain, 77 F.3d 372, 374 n.3 (11th Cir. 1996). The Third Circuit, on the no liberty interest in avoiding terms of administrative segregation lasting up to 1,825 days). 30 No. 23-1703 other hand, has said that disciplinary confinement for up to seven months “does not, on its own, violate a protected liberty interest….” Smith v. Mensinger, 293 F.3d 641, 654 (3d Cir. 2002). Yet courts still encounter periods of disciplinary soli- tary confinement longer than two years. See, e.g., Carrico v. Vanihel, No. 23-1381, 2023 WL 7015274, at *1 (7th Cir. 2023); Jackson v. Wexford Health Sources, No. 15-CV-920-NJR, 2015 WL 6663954, at *8 (S.D. Ill. Nov. 2, 2015) (plaintiff alleged more than five years of disciplinary segregation). Recall that these durations are not maximum limits on sol- itary confinement but only the times that trigger minimal pro- cedural protections. Under current law, prison officials may subject inmates to essentially indefinite solitary confinement so long as they have hearings with procedural safeguards as outlined in Wilkinson and Wolff v. McDonnell, 418 U.S. 539 (1974). Our precedents imply that prison officials may disci- pline inmates without any procedural protections by putting them in “humane” solitary confinement conditions for up to seven months, eight months or even a year, depending on the circuit. Case law from this circuit and other circuits has generally held that whether intermediate durations of solitary confine- ment—usually measured in months—implicate liberty inter- ests will depend on the facts of those conditions. See, e.g., Lisle v. Welborn, 933 F.3d 705, 721 (7th Cir. 2019) (analyzing condi- tions of confinement when plaintiff was placed in solitary for four months); Kervin v. Barnes, 787 F.3d 833, 836 (7th Cir. 2015) (period of solitary shorter than six months may require as- sessment of conditions, and collecting cases); Marion, 559 F.3d at 699 (eight-month term of disciplinary solitary required re- mand for assessment of conditions); Mitchell v. Horn, 318 F.3d No. 23-1703 31 523, 527, 531–33 (3d Cir. 2003) (remanding after prisoner was sentenced to 90 days in disciplinary confinement; cell had “human waste smeared on the walls” and was “infested with flies”); Gaines v. Stenseng, 292 F.3d 1222, 1226 (10th Cir. 2002) (remanding for consideration of conditions after 75 days in solitary); Colon, 215 F.3d at 232 (“in cases challenging SHU confinements of durations within the range bracketed by 101 days and 305 days, development of a detailed record will as- sist appellate review”) (footnote omitted). This makes sense in theory. The logic is that prisoners have a greater “liberty” interest in avoiding solitary confinement when they are held in appalling conditions, like those in this case, than when they are held in better conditions. But the upshot of these cases is that inmates have no liberty interest in avoiding solitary con- finement so long as the only thing that has changed is the fact that the prisoner is in solitary confinement—even if the pris- oner is segregated for months on end. More troubling, we have explained in non-precedential or- ders that “short” periods of solitary confinement—including “only 78 days”—do not support due process claims as a mat- ter of law even when the conditions are otherwise “deplora- ble.” See Obriecht v. Raemisch, 565 F. App’x 535, 540 (7th Cir. 2014) (no liberty interest where prisoner was placed in segre- gation for 78 days and made to sleep on mattress on wet floor); Younger v. Hulick, 482 F. App’x 157, 159 (7th Cir. 2012) (saying “relatively short” terms of segregation “do not re- quire inquiry into conditions” and collecting cases). These cases suggest that any term of solitary under three months will not implicate a liberty interest as a matter of law—that is, without even considering the condition of the cells (which, as this case makes clear, can be appalling). A review of the au- thorities cited for this premise in Obriecht and Younger makes 32 No. 23-1703 it seem like these cases reflect a “precedential drift,” where sound holdings morph slowly into the untenable and intoler- able. I hope our holding today will stop that drift, making clear that even less prolonged terms of disciplinary confinement warrant judicial attention and procedural protections. Again, under the Nelson Mandela Rules, anything more than fifteen days is prohibited as form of torture or cruel, inhuman, or de- grading treatment. It is time to modify and clarify our stand- ards in this area. II. Procedural Obstacles to Progress and the Need for Change Until now, change in the law of solitary confinement has been delayed by two barriers to meaningful judicial review: qualified immunity and the very limited availability of injunc- tive relief. As explained above, whether disciplinary solitary confinement implicates an inmate’s liberty interests is a fact- specific inquiry, meaning that qualified immunity will com- monly apply. In at least some of those cases, we will exercise our discretion under Pearson v. Callahan, 555 U.S. 223, 242 (2009), to bypass merits questions and address only whether the law was clearly established. E.g., Hardaway, 734 F.3d at 743–45; Carrico, 2023 WL 7015274, at *3. We also rarely have the opportunity to address claims for injunctive relief chal- lenging short-term prison disciplinary practices, which avoid qualified immunity defenses. By the time those cases make it to federal court, the claims for injunctive relief are moot. Cf. Black v. Brown, 513 F.2d 652, 655 n.8 (7th Cir. 1975) (“The claim for injunctive relief is, of course, mooted by plaintiff’s release from isolation and segregation.”); Hamner v. Burls, 937 F.3d 1171, 1175 (8th Cir. 2019) (“Hamner is no longer in No. 23-1703 33 administrative segregation, and he concedes that his claim for injunctive relief is moot.”). 4 Jackson’s case is one of the rare situations where a chal- lenge to a term of disciplinary solitary confinement is properly presented to this court. We correctly exercise our dis- cretion to address the merits issue and to “clarify [constitu- tional] law in our circuit and establish future constitutional boundaries…,” Smith v. Kind, 140 F.4th 359, 366 (7th Cir. 2025), but we are still obliged to affirm on the basis of qualified im- munity because the facts of Jackson’s case—although strik- ing—do not closely resemble facts we have confronted in any previous decision and a constitutional violation is not (yet) obvious on these facts. Our obligation to grant qualified immunity to the defend- ants here shows the benefits of a brighter-line rule. We should heed the Supreme Court’s warning that punting on a merits question “again, and again, and again” can “frustrate ‘the de- velopment of constitutional precedent’ and the promotion of law-abiding behavior.” Camreta v. Greene, 563 U.S. 692, 706 (2011), quoting Pearson, 555 U.S. at 237. The next time this court sees a due-process challenge to a punishment of a year or more in solitary confinement, we should join with the First and Second Circuits in presuming that solitary confinement of a year or more requires due-process protections regardless of the details of prison conditions. See Perry, 94 F.4th at 154 4 When punishment is imposed in a prison disciplinary proceeding, a civil challenge under § 1983 also must face the challenge posed by Heck v. Humphrey, 512 U.S. 477 (1994) (civil claim that implies invalidity of con- viction or sentence not actionable until conviction or sentence has been set aside), and Edwards v. Balisok, 520 U.S. 641 (1997) (applying Heck to prison discipline). 34 No. 23-1703 (“the showing can be made when the length of the confine- ment is as long as” 376 days of administrative segregation); Colon, 215 F.3d at 231 (same for 305 days). In fact, with an eye on the Nelson Mandela Rules and modern understanding about the effects of solitary confinement (addressed below), it would be better and more reasonable to move that boundary down from a year toward fifteen days. Drawing such a line in our case law would be helpful for two reasons. First, an outer boundary would limit prolonged periods of solitary confinement from falling into a “repetitive cycle of qualified immunity….” Camreta, 563 U.S. at 706 n.5. Forward-looking rules help safeguard constitutional rights in many areas of public law, and we should bring that approach to this doctrine as well. See, e.g., County of Riverside v. McLaughlin, 500 U.S. 44, 56 (1991) (adopting 48-hour standard for “prompt” probable cause determination by court after ar- rest); Miranda v. Arizona, 384 U.S. 436, 478–79 (1966) (requiring prophylactic measures to safeguard constitutional rights). Second, for practical reasons due process claims based on solitary confinement “are an especially appropriate context for a bright-line approach.” Colon, 215 F.3d at 233 (opinion of Newman, J.). Prison officials are entitled to the clearest guid- ance the courts can provide. Without a clear rule, prison offi- cials risk avoidable exposure to lawsuits that may or may not bypass qualified immunity depending on whether the plain- tiff has suffered conditions that have received prior appellate attention. As Judge Newman explained in Colon: This is a context in which a high degree of cer- tainty is extremely desirable. Prison officials conduct thousands of prisoner hearings each year, and they are entitled to know beforehand No. 23-1703 35 whether these hearings are subject to the proce- dural due process requirements of the Constitu- tion. It is better to alert these officials to the fed- erally enforceable requirements that must be observed than to leave the officials uncertain and merely afford damage remedies in those rare cases where courts determine after the fact both that the conditions were atypical and that the defendant officials were not entitled to qual- ified immunity. Id. (footnote omitted). In a footnote, the opinion continued: [T]the availability of a standard will alert the of- ficial to the need to provide federally required procedures in all hearings likely to result in con- finements of a duration equal to or exceeding the standard. In this respect, the situation is sim- ilar to criminal contempt hearings where the procedural protection of a jury must be pro- vided if the prosecutor intends to seek a penalty of more than six months. Id. at 233 n.6, citing Muniz v. Hoffman, 422 U.S. 454, 475–76 (1975). It is both appropriate and prudent for us to set forth pro- spective guidance on when solitary confinement requires pro- cedural protections, regardless of aggravating conditions of confinement. The facts of this case do not lie close enough to the outer boundary of our case law to justify drawing such a line today, but we should do so at the next opportunity. If it takes an en banc decision or a circulation under Circuit Rule 40(e), so be it. 36 No. 23-1703 III. Solitary Confinement and the Eighth Amendment This case did not present an Eighth Amendment challenge to Jackson’s solitary confinement. That is typical. Many (if not most) challenges to solitary confinement are brought under the due process clause of the Fourteenth Amendment because Wolff, Sandin, and Wilkinson make clear that such lawsuits can sometimes succeed—and because courts have declined to find Eighth Amendment violations resulting from prolonged periods of solitary confinement. E.g., Isby v. Brown, 856 F.3d 508, 521–24 (7th Cir. 2017) (no Eighth Amendment claim when prisoner was held in administrative segregation for more than ten years); see also Porter v. Pennsylvania Dep't of Corr., 974 F.3d 431, 457–58 & nn. 5 & 6 (3d Cir. 2020) (Porter, J., dissenting) (collecting cases: “the Supreme Court has not recognized psychological health, social interaction, or envi- ronmental stimulation as basic human needs in the Eighth Amendment context”); Grissom v. Roberts, 902 F.3d 1162, 1173–75 (10th Cir. 2018) (affirming summary judgment for de- fendants on Eighth Amendment claim; qualified immunity where prisoner was in solitary for twenty years). There have been some outlier cases, of course. See, e.g., Hutto v. Finney, 437 U.S. 678, 685–87 (1978) (approving district court’s conclusion that punitive isolation combined with par- ticularly heinous conditions violated Eighth Amendment and affirming remedial measures), abrogated on other grounds by Dep’t of Agriculture v. Kirtz, 601 U.S. 42, 56 (2024). And long- frozen Eighth Amendment doctrine on solitary confinement appears to be slowly thawing as some circuits have started to realize that prolonged solitary confinement is (or, at least, can be) “cruel and unusual.” E.g., Porter, 974 F.3d at 440–47, 450– 51 (finding issue of fact on Eighth Amendment rights No. 23-1703 37 violation based on 33 years in solitary but affirming on quali- fied immunity); Porter v. Clarke, 923 F.3d 348, 355–61 (4th Cir. 2019) (noting changing academic literature on the subject and affirming summary judgment for plaintiffs based on years in solitary confinement). But the weight of authority still holds that an inmate cannot show an Eighth Amendment violation with even a prolonged term of solitary confinement. We should reconsider whether and when solitary confine- ment violates the Eighth Amendment. Supreme Court Jus- tices have been calling for such a review for years. See, e.g., Apodaca v. Raemisch, 586 U.S. 931 (2018) (statement of So- tomayor, J., regarding denial of certiorari); Ruiz v. Texas, 580 U.S. 1191 (2017) (Breyer, J., dissenting from denial of stay of execution); Davis v. Ayala, 576 U.S. 257, 287–90 (2015) (Ken- nedy, J., concurring). We know that solitary confinement causes serious harm, regardless of whether it is imposed as punishment for the crime committed or as a discipline for a prison infraction. Recent case law, which has found protectable liberty inter- ests implicated by shorter and shorter periods of solitary con- finement, reflects a judicial recognition that prolonged puni- tive solitary confinement should be a thing of the past, or at least much rarer than it is now. But the harm caused by soli- tary confinement does not change based on the procedures used to impose it, making the due process clause an inade- quate (or at least imprecise) safeguard. Further doctrinal differences between Eighth Amendment law and Fourteenth Amendment law show that we should re- consider the question under the Eighth Amendment. Com- pare the two questions: 38 No. 23-1703 —Eighth Amendment: is this practice consistent with the evolving standards of decency that mark the progress of a maturing society? E.g., Walton v. Nehls, 135 F.4th 1070, 1072 (7th Cir. 2025) (explaining that conduct is objectively se- rious enough to constitute an Eighth Amend- ment violation if it does not meet this standard), quoting Hudson v. McMillian, 503 U.S. 1, 10 (1992)). —Fourteenth Amendment: does this punish- ment impose an atypical and significant hard- ship within the correctional context? E.g., Wil- kinson, 545 U.S. at 224. The Fourteenth Amendment procedural question is em- pirical. It asks us to determine whether conditions of confine- ment are irregular and extraordinarily harsh as compared to ordinary prison life. Few judges have spent a night in jail or prison. It’s easy for us to assume that life in any correctional context is hard, lonely, and isolating, and may commonly come with long periods of solitary confinement—making that punishment seem anything but “atypical.” See, e.g., Sandin, 515 U.S. at 485 (thirty days in solitary was “punitive” but did not implicate due process because it was not “a dramatic de- parture from the basic conditions of” the sentence); Judith Resnik et al., Punishment in Prison: Constituting the “Normal” and the “Atypical” in Solitary and Other Forms of Confinement, 115 Northwestern Univ. L. Rev. 45, 108 (2020) (“[M]any Four- teenth Amendment opinions take for granted long-term pro- found isolation as so normal an incident of prison life that no judicial oversight is needed.”). No. 23-1703 39 The Eighth Amendment question is more aspirational. It asks us to determine whether we have outgrown practices that we once tolerated. We should take up that question as applied to prolonged solitary confinement. The answer in most circumstances should be yes. To explain, I next outline how prison disciplinary tactics have evolved. I then summa- rize a small amount of the massive trove of evidence—from history, domestic and international legal sources, and scien- tific literature—showing that prolonged solitary confinement is inconsistent with evolving standards of decency. A. Winnowing Prison Disciplinary Tactics For decades, courts have held that some prison discipli- nary practices are off the table because they violate the Eighth Amendment. See Resnik, supra, 115 Northwestern Univ. L. Rev. at 59–67 (noting that in the 1960s, courts began rejecting the argument that “correctional need” justified prison disci- plinary tactics such as strip cells and deprivation of food and water). In Hutto v. Finney, the Supreme Court characterized as “cruel, unusual, and unpredictable” several prison discipli- nary tactics: (1) lashing prisoners with “a wooden-handled leather strap five feet long and four inches wide;” (2) using a hand-cranked device to “administer electrical shocks to vari- ous sensitive parts of an inmate’s body;” and (3) giving in- mates guns and tasking them with keeping order, resulting in beatings and shootings. 437 U.S. 678, 682 & n.4, n.5, and n.6 (footnotes omitted from first quotation). In Furman v. Georgia, Chief Justice Burger noted that courts found Eighth Amendment violations for “offensive punish- ments devised without specific authority by prison offi- cials …,” citing cases prohibiting whipping and placement of prisoners in freezing isolation cells. 408 U.S. 238, 384 (1972) 40 No. 23-1703 (Burger, C.J., dissenting), citing Jackson v. Bishop, 404 F.2d 571 (8th Cir. 1968) (whipping), and Wright v. McMann, 387 F.2d 519 (2d Cir. 1967) (filthy and freezing isolation). In Hope v. Pelzer, the Supreme Court famously held that prison officials acted with “obvious cruelty” when they hitched a prisoner to a post “for an extended period of time in a position that was painful, and under circumstances that were both degrading and dangerous.” 536 U.S. 730, 745 (2002). The Supreme Court found it particular notable that this “wanton treatment”— which was “antithetical to human dignity”—was “not done of necessity, but as punishment for prior conduct.” Id.; see also Smith v. Kind, 140 F.4th at 374 (Hamilton, J., dissenting) (“it is obviously unconstitutional to deliberately subject a na- ked prisoner to temperatures equivalent to and colder than a refrigerator”). Prolonged disciplinary solitary confinement should receive similar treatment to these practices that we have appropriately curtailed in the last century. The minimal procedures required in extreme cases by Wolff, Sandin, and Wilkinson should not be sufficient to justify prolonged solitary confinement. Historically, attempts to reg- ulate particularly harsh punishments with procedural protec- tions have shown that procedures are not sufficient, leading courts to declare later that the underlying practices are uncon- stitutional no matter how much process is afforded. Whip- ping is one example. For years, Arkansas law and prison guidelines had author- ized corporal punishment to “maintain prison discipline or to enforce respect for Penitentiary policies.” Talley v. Stephens, 247 F. Supp. 683, 687–88 (E.D. Ark. 1965). Courts were hesitant at first to declare the practice unconstitutional, explaining that “prison authorities must be given wide latitude and No. 23-1703 41 discretion in the management and operation of their institu- tions,” id. at 686, and that “corporal punishment ha[d] not been viewed historically as a constitutionally forbidden cruel and unusual punishment….” Id. at 689. A federal court tried to proceduralize whipping by regulating the maximum num- ber of lashes and the manner of administering lashes: each in- stance of whipping could not be “excessive” and had to be “inflicted as dispassionately as possible and by responsible people….” Id. at 689. These procedures look not so different from the ones we now confront in the solitary confinement context—the duration of the confinement cannot be too long, and the cell itself must be tolerable. Judicial experimentation with proceduralized whipping did not last long. The abhorrent nature of the punishment and its deleterious effects on prisoners soon proved too great to ignore. In Jackson v. Bishop, a case cited many times by the Su- preme Court, then-Circuit Judge Blackmun wrote an opinion holding that whippings were a cruel and unusual punish- ment: “the strap’s use, irrespective of any precautionary con- ditions which may be imposed, offends contemporary con- cepts of decency and human dignity and precepts of civiliza- tion which we profess to possess….” 404 F.2d 571, 579 (8th Cir. 1968). Noting that “[r]ules in this area seem often to go unobserved,” the Eighth Circuit reasoned that it was impos- sible to “ascertain the point which would distinguish the per- missible from that which is cruel and unusual[.]” Id. at 579– 80. Solitary confinement may prove to be a similar example. There is some value to improving procedural protections be- fore prisoners are sent to prolonged terms of solitary confine- ment. But we should not let procedural issues distract us 42 No. 23-1703 (courts, legislatures, and prison officials) from the larger sub- stantive issues, whether as matters of Eighth Amendment law or just sound public policy. Like early court decisions proce- duralizing whipping, we may look back on our procedural- ized solitary confinement cases as band-aids on a more fun- damental problem. B. Prolonged Solitary Confinement and Evolving Standards of Human Decency That brings me to my final point. There is a strong argu- ment that, in many circumstances, prolonged solitary confine- ment violates the Eighth Amendment, regardless of the pro- cedures used to impose it. 5 As noted above, the Supreme Court has explained that the Eighth Amendment prohibits punishments that are “incom- patible with the evolving standards of decency that mark the progress of a maturing society….” Walton, 135 F.4th at 1072 (internal quotation marks omitted), quoting Hudson, 503 U.S. at 10. The “clearest and most reliable objective evidence of contemporary values is the legislation enacted by the coun- try’s legislatures.” Atkins v. Virginia, 536 U.S. 304 (2002) (inter- nal quotation marks omitted), quoting Penry v. Lynaugh, 492 5 The Nelson Mandela Rules say that prisoners should “[i]n no cir- cumstances” be subjected to indefinite solitary confinement or prolonged solitary confinement as a disciplinary sanction. See G.A. Res. 70/175, at 15 (https://perma.cc/8ZRB-6SA9) (Dec. 17, 2015). The Rules define “prolonged solitary confinement” as a period longer than fifteen days. Id. at 16. This seems like a sensible approach to me. The sweeping “no circumstances” phrase does not, however, allow for the possible need to place a prisoner in solitary confinement for non-disciplinary reasons, such as if the pris- oner is a threat to other inmates or is at high risk of harm from other in- mates. No. 23-1703 43 U.S. 302, 331 (1989). When deciding Eighth Amendment is- sues, the Supreme Court has also considered views from the international community as evidence that “there is a consen- sus among those who have addressed the issue.” See Atkins, 536 U.S. at 316 n.21 (prohibiting death penalty for intellectu- ally disabled defendants; noting overwhelming worldwide disapproval of practice); see also, e.g., Graham v. Florida, 560 U.S. 48, 80 (2010) (prohibiting life-without-parole sentence for juveniles in non-homicide cases; international practices “not dispositive” but also “not irrelevant”) (internal quotation marks omitted), quoting Enmund v. Florida, 458 U.S. 782, 796 n.22 (1982). In the nearly thirty years since Sandin, sociological and medical research has laid bare the damaging mental and physical effects of solitary confinement. There is increasing scientific consensus and public awareness that the duration of isolation is itself a “condition” with a heavy bearing on how an inmate experiences confinement. That awareness is re- flected in case law. See Alexander N. Reinert, Solitary Troubles, 93 Notre Dame L. Rev. 927, 956 & n.163 (2018) (collecting cases and saying the “conceptual resistance” to considering human contact as a basic human need “is breaking down”); see also, e.g., Davis, 576 U.S. at 287 (Kennedy, J., concurring) (“The hu- man toll wrought by extended terms of isolation long has been understood ….”); Williams v. Pennsylvania Dep't of Cor- rections, 117 F.4th 503, 515–26 & n.98 (3d Cir. 2024) (violation of clearly established law to place mentally ill prisoner in long-term solitary confinement; collecting cases); Porter v. Clarke, 923 F.3d 348, 361 (4th Cir. 2019) (finding harm from prolonged solitary confinement obvious in light of “extensive scholarly literature describing and quantifying the adverse mental health effects of prolonged solitary confinement that 44 No. 23-1703 has emerged in recent years”); Hamner v. Burls, 937 F.3d 1171, 1181 (8th Cir. 2019) (Erickson, J., concurring) (recognizing “the developing science of mental health and what is now known [about] the profound detrimental and devastating im- pact solitary confinement has on an inmate’s psyche”); Pala- kovic v. Wetzel, 854 F.3d 209, 225–26 (3d Cir. 2017) (acknowl- edging “robust body of legal and scientific authority recog- nizing the devastating mental health consequences caused by long-term isolation in solitary confinement”); Kervin, 787 F.3d at 837 (“The serious psychological consequences of such quasi-solitary confinement have been documented.”), citing Elizabeth Bennion, Banning the Bing: Why Extreme Solitary Confinement is Cruel and Far Too Usual Punishment, 90 Ind. L. J. 741 (2015). The arc of American jurisprudence on solitary confine- ment shows that it has long been understood as a cruel prac- tice. Prolonged solitary confinement was uncommon in the early years of the United States. See David M. Shapiro, Solitary Confinement in the Young Republic, 133 Harv. L. Rev. 542, 581 (2019) (“[I]t does not appear that prisoners in the 1790s ever remained in solitary confinement for years on end.”). Begin- ning in the 1820s, a few states experimented with long-term solitary confinement to encourage individual repentance, but “after an initial phase of popularity, [it] came to be rejected by the 1860s because of its cruel effects.” John Stinneford, Exper- imental Punishments, 95 Notre Dame L. Rev. 39, 56, 61–64 (2019). The Supreme Court capped this rejection with In re Medley, 134 U.S. 160 (1890), finding that a state statute requir- ing a prisoner to remain in solitary confinement for two to four weeks prior to execution “added such significant suffer- ing to his sentence that it could not be imposed on a prisoner whose crime occurred before the statute took effect.” No. 23-1703 45 Stinneford, supra, 95 Notre Dame L. Rev. at 66, citing Medley, 134 U.S. at 168, 172 (noting that several decades ago, “the whole subject attracted the general public attention, and its main feature of solitary confinement was found to be too se- vere”). As one scholar noted: “The founding nation of the modern prison systems—the United States—was among the first to abandon large-scale solitary confinement.” Peter Scharff Smith, The Effects of Solitary Confinement on Prison Inmates: A Brief History and Review of the Literature, 34 Crime & Justice 441, 465 (2006). After Medley, solitary confinement survived only “at the very margins of penal practice during the twentieth century ….” Stinneford, supra, 95 Notre Dame L. Rev. at 56. That changed in the 1970s when prison overcrowding, a rise in prison riots, and the highly publicized October 1983 mur- ders of two prison guards resurrected solitary confinement “in the form of the ‘supermax’ movement in the 1980s and 1990s.” Id. at 56, 71–72; see also Scharff Smith, supra, 34 Crime & Justice at 442–43. Solitary confinement has remained a com- mon practice since then: according to estimates from 2021, at least 41,000 prisoners nationwide were isolated “in a cell for an average of twenty-two hours or more per day, for fifteen or more consecutive days.” Correctional Leaders Ass’n & Ar- thur Liman Ctr. for Pub. Interest at Yale L. School, Time-in-Cell 2021: A Snapshot of Restrictive Housing, vii, 60 (2022) (https://perma.cc/KSZ5-EY55); see also supra n.2 (providing even higher estimates from the mid-2010s). In the past decade, a great deal of sociological and scien- tific evidence of the harms of solitary confinement has 46 No. 23-1703 emerged. 6 As I said at the outset, in 2015, the United Nations General Assembly unanimously adopted the Standard Mini- mum Rules for the Treatment of Prisoners, also called the Nel- son Mandela Rules. Those rules prohibit disciplinary solitary confinement of more than fifteen consecutive days as a form of torture. G.A. Res. 70/175, at 15–16 (Dec. 17, 2015). 7 The Mandela Rules accompanied an international decline in the use of prolonged solitary confinement. Although each juris- diction’s laws are subject to certain nuances and exceptions, 6 The following is a small sample of recent work. Bruce Western et al., Solitary Confinement and Institutional Harm, 3 Incarceration 1, 20 (2022) (col- lecting empirical support for the premise that solitary confinement de- grades human dignity); Craig Haney, The Science of Solitary: Expanding the Harmfulness Narrative, 115 Northwestern Univ. L. Rev. 211 (2020) (collect- ing research); Justin D. Strong et al., The Body in Isolation: The Physical Health Impacts of Incarceration in Solitary Confinement, 15 PLoS One 10 (2020); Mimosa Luigi et al., Shedding Light on “the Hole”: A Systematic Re- view and Meta-Analysis on Adverse Psychological Effects and Mortality Follow- ing Solitary Confinement in Correctional Settings, 11 Frontiers in Psychiatry 840 (2020). Slightly older work includes Jeffrey L. Metzner & Jamie Fellner, Solitary Confinement and Mental Illness in U.S. Prisons: A Challenge for Medi- cal Ethics, 38 J. Am. Academy Psychiatry & L. 104, 104 (2010) (“(“[I]solation can be as clinically distressing as physical torture.”); Stuart Grassian, Psy- chiatric Effects of Solitary Confinement, 22 Wash. U. J. L. & Policy 325 (2006); and Craig Haney & Mona Lynch, Regulating Prisons of the Future: A Psy- chological Analysis of Supermax and Solitary Confinement, 23 N.Y.U. Rev. L. & Social Change 477 (1997). 7 On the State Department’s involvement, see Dan Sicorsky, The Nel- son Mandela Rules: Honoring a Prisoner Turned World Leader, U.S. Dep’t of State, July 18, 2017 (https://perma.cc/D4Y8-C3N3) (“A partnership be- tween the State Department and Colorado Department of Corrections has trained 1,050 corrections professionals from more than 20 countries in the operation of humane and transparent prisons.”). No. 23-1703 47 the following nations and provinces have broadly limited the disciplinary use of solitary confinement: - Brazil, Croatia, France, Japan, Peru, Switzer- land, and Western Australia (maximum of thirty days) - Albania, Belgium, Colombia, Croatia, the Netherlands, Sri Lanka, Venezuela, Italy, Yukon, Canada, Serbia, and Spain (fourteen or fifteen days) - Chile, Bolivia, and Romania (ten days) - Israel (seven days) - Ireland and Scotland (three days) Sharon Shalev, Mapping Solitary Confinement, 7–8 (2024) (outlining survey and noting that laws varied slightly depending on various factors including inmate’s age, severity of offense, and sentence length); see also id. at 17–21 (surveying international use of solitary as prison management and inmate-protection tool). That trend has also taken effect domestically. Between 2009 and 2022, 886 bills were introduced in 45 states to restrict or end solitary confinement. Forty states passed at least one of those bills. Unlock The Box, Banning Torture: Legislative Trends and Policy Solutions for Restricting and Ending Solitary Confinement throughout the United States 4 (2023) (https://perma.cc/8VRM-VLKM). New Jersey, for instance, has statutorily limited the use of solitary confinement to twenty consecutive days, subject to limited exceptions. See N.J. Rev. Stat. § 30:4-82.8(a)(9) (2025). Prison administrators have also made reforms in this direction. In 2017, the execu- tive director of Colorado’s Department of Corrections ended the state’s use of long-term solitary confinement beyond the 48 No. 23-1703 Nelson Mandela Rules’ limit of fifteen days, which he had helped draft. 8 A sitting president has called for the curtail- ment of solitary confinement, 9 and in testimony to a House Appropriations Subcommittee, so has a sitting Supreme Court justice. 10 8 See Rick Raemisch, Opinion, Why We Ended Long-Term Solitary Con- finement in Colorado, N.Y. Times (Oct. 12, 2017), https://www.ny- times.com/2017/10/12/opinion/solitary-confinement-colorado- prison.html (last accessed July 23, 2025); see also Amy Fettig, Why I Worked on the Mandela Rules, in Incarceration and the Law: Cases and Materials 225, 225–26 (Margo Schlanger et al. ed., 10th ed. 2020) (https://perma.cc/44W6-B7FB) (“At an early meeting, the U.S. government actually opposed any meaningful limits on solitary confinement,” but “this dynamic changed after U.S. advocates were able to bring to the ne- gotiating table two state corrections directors from Colorado and Wash- ington who pioneered reforms of solitary confinement in their own juris- dictions.”). As another example North Dakota derived inspiration from Norwe- gian practices and implemented dramatic reforms resulting in a 74% re- duction in the use of solitary confinement between 2016 and 2020. See Da- vid Cloud, et al., “We just needed to open the door”: a case study of the quest to end solitary confinement in North Dakota, 9 Health & Justice 28 (2021). 9 Barack Obama, Opinion, Why We Must Rethink Solitary Confinement, Wash. Post (Jan. 25, 2016), https://www.washingtonpost.com/opin- ions/barack-obama-why-we-must-rethink-solitary-confine- ment/2016/01/25/29a361f2-c384-11e5-8965-0607e0e265ce_story.html (last accessed July 23, 2025) (announcing executive directive “banning solitary confinement for juveniles” in federal prisons); The White House, Fact Sheet: Dep’t of Justice Review of Solitary Confinement (January 25, 2016) (https://perma.cc/3RDJ-K24A) (explaining presidential guidance). 10 Amy Fettig & Samuel Weiss, Supreme Court Justice Kennedy Invites Constitutional Challenge of Solitary Confinement, Solitary Watch (June 23, 2015) (https://perma.cc/SQ7J-NCMQ) (last accessed July 23, 2025) (Justice No. 23-1703 49 I do not mean to imply bad faith on the part of prison offi- cials, who have very difficult jobs. See, e.g., Wilkinson, 545 U.S. at 227 (acknowledging “the brutal reality of prison gangs”). My point is that our confusing case law on solitary confine- ment is not helpful to prison officials, not fair to prisoners, and not consistent with the Nelson Mandela Rules or “the evolv- ing standards of decency that mark the progress of a maturing society.” Hudson, 503 U.S. at 10. Today’s decision applies governing law correctly and, rec- ognizing the difficult situation that prison officials face, ap- propriately grants qualified immunity. In the longer run, however, even more specific guidance on procedures for dis- ciplinary solitary confinement will not adequately address the harms caused by prolonged solitary confinement, which do not depend on the procedures used to impose it. We should reconsider our application of the Eighth Amendment to pro- longed disciplinary solitary confinement, which should no longer be such a common feature of imprisonment. Kennedy telling Congress that “[s]olitary confinement literally drives men mad”). 50 No. 25-1703 SCUDDER, Circuit Judge, concurring in the judgment. To- day’s decision is a postcard example of a court getting ahead of itself. The majority opinion forgoes a narrow and straight- forward resolution of this appeal—on qualified immunity grounds—and reaches to resolve a much broader, much more complicated, and much more consequential issue of when a prisoner’s transfer to segregated confinement warrants con- stitutional protections. More concerning yet, the majority an- nounces a standard at odds with our case law and far afield from Supreme Court precedent. So I am left only to concur in today’s judgment. I Do not let the length and density of the majority or sepa- rate concurring opinion distract from what happened here and the ease of resolving the issue presented on qualified im- munity grounds. Following a hearing, Illinois prison officials ordered Abre Jackson to serve 90 days in segregated confinement as disci- plinary punishment for an incident at the Stateville Correc- tional Center. Everyone agrees that Jackson had a full oppor- tunity at the hearing to tell his side of the story and to ask questions of the prison officials. Jackson nonetheless invokes the Fourteenth Amendment’s Due Process Clause and claims he was owed more process—the right to review video evi- dence of the incident and to cross-examine correctional officer witnesses. Our recent decision in Adams v. Reagle defeats Jackson’s claim. “Our law is clear,” we explained, “that an inmate who is facing transfer to disciplinary segregation is entitled only to ‘informal, nonadversarial due process,’” meaning “notice of No. 23-1703 51 the reasons for the inmate’s placement and ‘an opportunity to present his views.’” Adams v. Reagle, 91 F.4th 880, 895 (7th Cir. 2024) (majority opinion of St. Eve, J.) (quoting Westefer v. Neal, 682 F.3d 679, 684–85 (7th Cir. 2012)). That informal process does not include the right to call witnesses or review video evidence, however. See id.; see also Westefer, 682 F.3d at 685 (explaining that “inmates do not have a constitutional right to call witnesses”). Adams is not an aberration. To the contrary, it aligns with direction from the Supreme Court. The Court rejected a nearly identical due process claim two decades ago when a group of inmates challenged the procedures that preceded their indef- inite transfer to Ohio’s Supermax facility. See Wilkinson v. Austin, 545 U.S. 209, 225–28 (2005). The Court found no pro- cedural due process violation where the inmates “receive[d] notice of the factual basis” for the transfer and “a fair oppor- tunity for rebuttal at a hearing,” although they could not call witnesses. Id. at 225–26, 228. That is exactly the process Jack- son received here: notice and an opportunity to be heard. Regardless, the qualified immunity inquiry is not close. At the time of Jackson’s disciplinary hearing, neither our court nor the Supreme Court had ever held that due process re- quires that an inmate be allowed to call witnesses or review documentary or video evidence prior to their placement in disciplinary segregated confinement. This is all we had to say to resolve Jackson’s appeal. II The majority charts a different course by seeing Jackson’s case as an opportunity to explore constitutional limits on the duration of segregated confinement. It is from that 52 No. 23-1703 perspective that the majority chooses to break new ground and announce a broad and unworkable rule of constitutional law—all purportedly in the name of procedural due process. A Jackson’s appeal should never have become a test case. He experienced a relatively short period of disciplinary segrega- tion. And his procedural due process claim makes very little sense: he points to the conditions of his segregated confine- ment and says those conditions proved so horrific that they should have informed the procedures afforded him in the first instance at the disciplinary hearing that led to the segregated confinement. If that reasoning seems backward, that is the correct reaction. Jackson’s position fares worse when measured against our precedent. Even the most cursory review of our cases shows that his challenge to his three-month term of disciplinary seg- regation cannot rise to the level of a protected liberty interest: • In Lekas v. Briley, the case perhaps most anal- ogous to this one, we declined to find a pro- tected liberty interest where an inmate re- ceived three months of segregated confine- ment which included a loss of work and pro- gramming privileges, no telephone usage or contact visits, and reduced access to per- sonal items and the commissary. 405 F.3d 602, 610 (7th Cir. 2005). We analogized the conditions in Lekas to Williams v. Ramos, 71 F.3d 1246 (7th Cir. 1995), where the prisoner experienced solitary confinement for 19 days, and Thomas v. Ramos, 130 F.3d 754 (7th No. 23-1703 53 Cir. 1997), where the prisoner’s 24-hour-a- day segregation lasted for approximately 70 days. In neither case did we identify a pro- tected liberty interest. Id. at 762; Williams, 71 F.3d at 1250. • In Ealy v. Watson, we considered a proce- dural due process claim where the plaintiff spent five months in segregation and experi- enced “cold temperatures, dirty cells, and faulty plumbing resulting in unsanitary con- ditions.” 109 F.4th 958, 961 (7th Cir. 2024). The majority quotes from Ealy at great length, but its reliance misses the mark. There we declined to find a protected liberty interest, instead taking the more obvious route and holding that there was no due pro- cess violation—leaving the liberty question unanswered. See id. at 965. But in past cases where we did reach the liberty question, we made clear the bar is high. We failed to iden- tify a liberty interest in Lisle v. Welborn, for example, where the prisoner challenged his four-month term of segregated confinement in a cell with “rust on the bars and ‘corroded feces’ in the toilet.” 933 F.3d 705, 721 (7th Cir. 2019) (explaining that a jury could not rea- sonably infer “that these conditions caused Lisle any significant hardship”). • As support for the very brief discussion of unsanitary conditions in Ealy and Lisle, we cited our earlier decision in Kervin v. Barnes, 54 No. 23-1703 787 F.3d 833 (7th Cir. 2019). See Ealy, 109 F.4th at 965; Lisle, 933 F.3d at 721. Yet we de- clined to find a protected liberty interest in Kervin, where the prisoner endured 30 days in segregated confinement and lost tele- phone and commissary privileges. See id. at 834, 837. In reaching that conclusion, how- ever, we described it as “sensibl[e]” that “the severity of treatment should be combined with its duration in assessing the gravity of the conditions complained of by the pris- oner.” Id. at 836 (citing out-of-circuit cases considering “unsanitary or otherwise dis- gusting” conditions in the context of a lib- erty interest analysis). • And in at least four cases where an inmate experienced a period of segregated confine- ment lasting anywhere between six months and one year, we remanded for further fac- tual development on whether the conditions of segregated confinement were sufficiently “atypical” to establish a liberty interest. See, e.g., Marion v. Columbia Corr. Inst., 559 F.3d 693, 698–99 (7th Cir. 2009); Wagner v. Hanks, 128 F.3d 1173, 1174, 1177 (7th Cir. 1997); Bryan v. Duckworth, 88 F.3d 431, 433 (7th Cir. 1996); Whitford v. Boglino, 63 F.3d 527, 533 (7th Cir. 1995). Right to it, we have never held that three months of segre- gated confinement constitutes a significant and atypical hard- ship. Nor have we ever found that unsanitary conditions No. 23-1703 55 within a term of segregated confinement have sufficed to es- tablish a protected liberty interest. Indeed, we have “de- scribed an inmate’s liberty interest in avoiding segregation as very limited or even nonexistent.” Marion, 559 F.3d at 697; see Wagner, 128 F.3d at 1175 (describing “the right to litigate dis- ciplinary confinements” as “vanishingly small”). Nothing improves for Jackson if the focus turns to the U.S. Reports. The Supreme Court has provided guidance in two cases occupying opposite ends of the spectrum. In Sandin v. Conner, the Court found no liberty interest where the inmate served 30 days in disciplinary segregation because the condi- tions in segregated confinement “mirrored those conditions imposed upon inmates in administrative segregation and protective custody.” 515 U.S. 472, 486 (1995). Ten years later in Wilkinson v. Austin, the Court encoun- tered facts on the opposite end of the spectrum. There a group of inmates challenged their transfers to Ohio’s Supermax fa- cility which had “more restrictive [conditions] than any other form of incarceration in Ohio,” depriving inmates “of almost any environmental or sensory stimuli and of almost all hu- man contact.” 545 U.S. at 214. Subjecting Ohio prisoners to these “harsh conditions” for years on end amounted to an “atypical and significant hardship,” resulting in a protected liberty interest for purposes of a procedural due process claim. Id. at 224. To my eye, Jackson’s term in segregated confinement falls much closer to the conditions described in Sandin than it does Wilkinson. Jackson’s commitment to disciplinary segregation was not unlimited in duration: he received three months as punishment which is similar to the one-month term of segre- gation in Sandin, and far shorter in duration than Wilkinson, 56 No. 23-1703 where an inmate’s placement in the Supermax was subject only to an annual review. See id. Even if I am mistaken, noth- ing in Sandin or Wilkinson comes anywhere close to suggest- ing a violation of Jackson’s procedural due process rights at his disciplinary hearing. The majority’s reliance on Taylor v. Riojas is most confus- ing. Riojas is an Eighth Amendment case—a conditions of con- finement case—where prison officials placed Texas inmate Trent Taylor in an extremely unsanitary cell covered “nearly floor to ceiling in ‘“massive amounts” of feces.’” 592 U.S. 7, 8 (2020). Taylor refused food and water for four days due to fear of contamination and was left to “sleep naked in sewage” on his cell floor. Id. The Court’s summary reversal of a grant of qualified im- munity in Taylor “reinforced the principle that prison officials do not have unlimited discretion to confine inmates in deplor- able conditions.” Op. at 16. But the conditions alleged by Jack- son fall short of those experienced by Taylor. Regardless, Ri- ojas is through and through an Eighth Amendment case, with the Justices saying not a word about procedural due process. But perhaps that is the aim of today’s majority opinion—to somehow force fit an Eighth Amendment claim into a due process mold. B Today’s decision is a legal paradox. Remember Abre Jack- son’s actual claim: he sued the prison officials who conducted his disciplinary hearing, contending they deprived him of a protected liberty interest without due process. His claim is all about the process he received at the disciplinary hearing—he believes he should have been able to call additional witnesses No. 23-1703 57 and review Stateville video evidence. But Jackson has never alleged that the officials who presided over his hearing were aware of the unsanitary conditions he came to experience in the segregation unit. If the hearing officials were not aware of the conditions, how could they have accounted for them in evaluating what process Jackson was due at the disciplinary hearing? By relying on Jackson’s allegations about the conditions of the segregation unit to find a protected liberty interest, the majority essentially conflates an Eighth Amendment condi- tions of confinement analysis with a Fourteenth Amendment procedural due process claim. It is far from clear to me, as a practical matter, what the difference would be after today’s decision. We will now be using the Due Process Clause to as- sess prison conditions unknown at the time prison officials decide that misconduct warrants a term of disciplinary segre- gated confinement. And how this will work in practice is an- yone’s guess. Do not misread what I am saying. To be sure, in Wilkinson the Supreme Court considered the “conditions of confine- ment” at the Supermax facility in assessing whether the in- mates had “a liberty interest in avoiding transfer” to those “more adverse conditions.” 545 U.S. at 221–22. But the Court in Wilkinson was not licensing courts to consider the full range of conditions that might support an Eighth Amendment claim when assessing whether segregated confinement gives rise to a protected liberty interest under the Fourteenth Amend- ment’s Due Process Clause. Instead, the Court focused solely on conditions that would accompany a prisoner’s transfer to a Supermax facility—lack of all human contact, undimmed lights, and limited exercise—all conditions which would have 58 No. 23-1703 been known to prison officials at the time of the transfer de- termination. See id. at 223–24; see also Lisle, 933 F.3d at 721 (“[I]f the disciplinary measures do not ‘substantially worsen the conditions of confinement’ of an inmate, then he has not been deprived of a protected liberty interest.” (emphasis added) (quoting Miller v. Dobier, 634 F.3d 412, 414–15 (7th Cir. 2011))). Go back to Jackson’s claim once again. Jackson com- plained about three aspects of his segregated confinement: loud and unruly inmates in his unit, unsanitary cell and shower conditions, and contaminated water. While we accept these allegations as true, Jackson’s observations about the conditions in segregation do not automatically lend support to his procedural due process claim. The question is whether the hearing officers knew of these conditions—the same way the conditions accompanying the Ohio Supermax program were known in Wilkinson. I am not doubting that there may be instances where an inmate adequately alleges that prison officials presiding over disciplinary hearing were on notice that segregated confine- ment included certain conditions, beyond the obvious loss of privileges it entails, that “impose[d] [an] atypical and signifi- cant hardship on the inmate in relation to the ordinary inci- dents of prison life.” Sandin, 515 U.S. at 484. That is not this case, though. After considering how the majority’s test would apply to the facts of Jackson’s claim, I am left to wonder whether the whole point of today’s decision is to make procedural due process coterminous with the protections afforded by the Eighth Amendment. Is the majority suggesting that deplora- ble prison conditions, known or unknown at the time of a dis- ciplinary hearing, mean that prisoners must receive mini No. 23-1703 59 trials prior to being housed in segregated confinement? If so, that conclusion will work a sea change in today’s law for prison officials conducting routine disciplinary hearings that end with terms of segregated confinement. And I see no way that Adams v. Reagle would remain good law in such a regime. The majority’s approach becomes all the more confounding when you compare it against well-established Eighth Amendment principles. In that context, a prisoner must demonstrate that the prison official “k[new] of and disregard[ed] an excessive risk to inmate health or safety.” Farmer v. Brennan, 511 U.S. 825, 837 (1994). But the majority enforces no similar knowledge requirement with respect to the prison officials who conducted Jackson’s disciplinary hearing. So the majority’s error today is at least two-fold: it applies an Eighth Amendment gloss to our analysis of procedural due process claims, while simultaneously lowering the standard that ordinarily governs claims challenging an inmate’s conditions of confinement. In its attempt to craft a bright-line rule, I worry that the majority’s decision will wreak havoc within routine disciplinary proceedings that happen all day, every day in federal and state prisons across the country. We could have easily avoided all of this by assuming Jack- son’s liberty interest and resolving this case in a few pages because, by any fair measure, Jackson received adequate pro- cess. And, at the very least, the defendants have qualified im- munity. III The majority’s desire to see more development in the law defining the limits of segregated confinement is a view I 60 No. 23-1703 share. And I also believe that in time the law can learn much from social science research informing the effects of pro- longed solitary confinement. But this case—brought only in the name of procedural due process and not under the Eighth Amendment—saves those difficult questions for another day. Instead of exercising cau- tion, today’s decision reaches to draw legal lines divorced from current law and with no prospect of sound practical im- plementation. In my respectful view, judicial restraint should have defined our path here.