RICHARD JOHNSON, Plаintiff-Appellant, v. CHARLES L. RYAN, named as Charles Ryan, Director of Arizona Department of Corrections; STACEY CRABTREE, Administrator of Offender Services Bureau; P. DAYS, Deputy Warden of Browning Unit; MONTONO, AKA Unknown Montano, First name unknown; Deputy Warden of Central Unit on or about 4-12-18 till 4-23-18, alias added pursuant to Doc #53; BELT, First name unknown; SSU Sergeant B, Special Security Unit Sergeant at Central Unit, Defendants-Appellees.
No. 20-15293
UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
December 15, 2022
D.C. No. 2:18-cv-03055-MTL-ESW
FOR PUBLICATION
OPINION
Appeal from the United States District Court for the District of Arizona
Michael T. Liburdi, District Judge, Presiding
Argued and Submitted April 13, 2022
San Francisco, California
Filed December 15, 2022
Before: Jay S. Bybee and Ryan D. Nelson, Circuit Judges, and Jed S. Rakoff,* District Judge.
Opinion by Judge Bybee;
Partial Concurrence and Partial Dissent by Judge
SUMMARY**
Prisoner Civil Rights
In an action brought pursuant to
Johnson‘s complaint alleged, in part, that after he was validated as a member of a Security Threat Group (“STG“), he was moved to maximum security confinement where he is confined to his cell for twenty-four hours per day, strip searched every time he leaves his cell, takes meals in his cell, and has limited access to rehabilitation programs. These conditions are substantially more restrictive than the general population from which he was moved. Johnson also alleges that he was denied the opportunity for restoration of lost earned release credits.
Addressing Johnson‘s claim that the ADC‘s annual review process of his confinement status violates due process, the panel held that Johnson has a liberty interest in avoiding assignment to maximum custody as a consequence of his STG validation. Nevertheless, Johnson failed to state a claim for a due process violation under the three-prong framework set forth in Mathews v. Eldridge, 424 U.S. 319, 335 (1976). Johnson did not challenge the procedures by which he was initially validated as an STG member. Instead, he argued that ADC‘s annual reviews of his confinement status do not afford him adequate process because they are based solely on his alleged gang affiliation, without regard to his criminal history, propensity of violence, or disciplinary record within his past reclass review period or his disciplinary record overall. The panel held that ADC is entitled to substantial deference in its determination that an inmate‘s STG membership and failure to renounce and debrief poses a continuing security threat. Although Johnson disagreed with ADC‘s judgment, he failed to plausibly allege how that judgment created a risk that he will be erroneously classified as a security threat.
The panel next Johnson‘s claim that his removal from thе Step-Down Program (“SDP“) in 2018 violated his due process rights under the
Addressing Johnson‘s
Concurring in part and dissenting in part, Judge Rakoff stated that while Arizona provides the mirage that a once validated member of an STG can later escape solitary confinement, the reality is that he will be kept there for the entire duration of his sentence. Believing that this is unconstitutional, as well as contrary to past holdings of this Court, Judge Rakoff dissented from the majority‘s analysis in Part III.A of its opinion, pertaining to Johnson‘s claim that Arizona‘s refusal to consider factors other than his initial STG validation and his subsequent failure to debrief denies him due process. And while Judge Rakoff concurred in Part III.B of the majority‘s opinion, which reverses and remands the district court‘s entry of summary judgment against Johnson on his claim that his removal from the SDP program violated due process, he wrote separately to emphasize that his claim is validly broader than the majority contends. Finally, Judge Rakoff concurred fully in Part III.C of the majority opinion with respect to Johnson‘s retaliation claim.
COUNSEL
Brian Wolfman (argued) and Daniel Wassim (argued), Hannah M. Mullen, Madeline H. Meth, Samuel Myers, Sanders Keyes Gilmer, and Rebecca Van Voorhees, Georgetown University Law Center Aрpellate Courts Immersion Clinic, Washington, D.C., for Plaintiff-Appellant
Patrick J. Boyle (argued), Assistant Attorney General; Mark Brnovich, Attorney General of Arizona; Office of the Arizona Attorney General, Phoenix, Arizona; for Defendants-Appellees.
OPINION
BYBEE, Circuit Judge:
Arizona Department of Corrections (ADC) inmate Richard Johnson is a validated member of a Security Threat Group (STG) and, pursuant to ADC‘s policy, has been assigned to maximum custody confinement. Johnson challenges two aspects of ADC‘s STG determination. First, he argues that ADC‘s annual reviews of his maximum security confinement are insufficient to satisfy the Due Process Clause of the
I. BACKGROUND AND PROCEEDINGS
A. Regulatory Background
1. ADC‘s security threat group policy
In 2005, the Supreme Court observed that “[t]he use of Supermax prisons has increased over the last 20 years, in part as a response to the rise in prison gangs and prison violence.” Wilkinson v. Austin, 545 U.S. 209, 213 (2005) (citing Chase Riveland, U.S. Dep‘t of Justice, Nat‘l Inst. of Corr., Supermax Prisons: Overview and General Considerations 1 (1999)). These facilities, sometimes referred to as “jails within prisons,” Riveland, supra, at 1, are “more restrictive than any other form of incarceration,” Wilkinson, 545 U.S. at 214.
In 1991, ADC addressed its own prison gang challenges by adopting a Security Threat Group policy, referred to as Department Order (DO) 806. The policy‘s purpose is to “minimize the threat that inmate gang or gang like activity poses to the safe, secure and efficient operation of institutions.” The policy defines an STG as “[a]ny organization, club, association or group of individuals, either formal or informal (including traditional prison gangs), . . . whose members engage in activities that include . . . committing or attempting to commit unlawful acts or acts that violate the Department‘s written instructions, which detract from the safe and orderly operation of prisons.” Ariz. Dep‘t of Corr., Dep‘t Order Manual, Dep‘t Order 806, at 25 (2017) [hereinafter DO 806]. The policy contemplates “[m]inimizing gang or gang like activity” through two programs: (1) “the debriefing and segregation of inmates who disavow gang membership,” and (2) “a step-down process for gang members who participate in programming, reject gang activity and affiliation, and remain disciplinary free.” DO 806 at 1.
ADC‘s process for confirming an inmate‘s membership in an STG is called “validation,” and entails the following process. When an inmate is first suspected of being an STG member, the inmate is monitored for any STG-related activity. ADC collects any documentation and physical evidеnce in support of the inmate‘s STG membership in an STG Suspect File held by the Special Security Unit (SSU). DO 806.03. Once an inmate accrues at least ten points in two or more of ADC‘s validation criteria, a “validation packet” is prepared and sent to the STG Validation Hearing Committee. In the hearing, SSU presents the validation packet, the inmate presents his defense, and the STG Validation Hearing
Validated STG members who refuse to renounce are assigned to maximum custody. DO 806.07. Male validated STG members, like Johnson, are housed in the ASPC-Eyeman Browning Unit (Browning Unit) for the duration of their incarceration. Validated STG inmates are ineligible for “restoration of forfeited time credits [and] . . . rescission of Parole Class III time.” Id. 806.07.1.1. Furthermore, Johnson alleges that ADC conducts an annual review of maximum security inmates’ status, but for STG inmates, these annual reviews consider only whether they are validated STG members and whether they have renounced and debriefed.
Conditions in maximum custody facilities such as the Browning Unit are very restrictive. Johnson alleges that he and other STG inmates in maximum custody are confined to their cells twenty-four hours per day, strip searched, and handcuffed behind their backs every time they leave their cells. In the Browning Unit, STG inmates are entitled to between $60 to $160 in weekly allowance at the store, one to three fifteen-minute phone calls per week, one to three two-hour non-contact visit blocks per week, and three three-hour recreation blocks per week in a chute enclosure with the possibility that one block may be in a ten-by-ten foot enclosure. They also have access to the library, television, and mandatory rehabilitation programming, with the possibility of further access to hobby supplies, the job of pod porter, and eligibility for unrestrained escorts out of their cells. ADC has promulgated an Inmate Maximum Custody Management and Incentive System. Ariz. Dep‘t of Corr., Dep‘t Order Manual, Dep‘t Order 812 (2019) [hereinafter DO 812]. The program uses a “step incentive system” that provides inmates “the opportunity to participate in jobs, programs, and other out of cell activities” by which “inmates may progress from controlled based housing to open privilege based housing.” DO 812.1.
2. Pathways out of maximum security
STG-validated inmates have two distinct paths to become eligible for custody reductions and housing status change. They must either renounce their STG membership and debrief or they must сomplete the SDP. DO 806.07.1.2.
a. Renouncing and debriefing
Renunciation is “[t]he process, in which a validated STG member agrees to renounce STG affiliation, successfully completes a debriefing, and is considered a former member.” DO 806 at 25. Debriefing is a process by which the inmate provides ADC with information to convince ADC that the inmate has withdrawn from the STG. Debriefing permits ADC to get additional information about the STG so that ADC may manage the threat and to determine whether the inmate will need to be protected from other STG members and suspects. DO 806.06.1.1. A validated STG member is “[p]ermitted to renounce and debrief at any time.” DO 806.07.1.1.9. If the inmate has debriefed to the satisfaction of the STG Validation Hearing Committee, the inmate is placed in protective custody pending review of eligibility for lower custody housing or a double cell environment. DO 806.07.1.5.1. Inmates who successfully debrief may also request out-of-state placement. DO 806.07.1.5.2.
The SDP “permits active inmates who have been validated as STG members, to remove themselves from STG activity and demonstrate to Department staff that they are no longer involved with STG activity.” ADC considers it to be an “alternative, indirect way of demonstrating a disassociation with gang activity [that] does not require renunciation and debriefing.” In order to participate in the SDP, a validated STG member must notify ADC in writing of his desire to participate in the program and must do so after a twenty-four-month period in which the inmate did not participate in STG activity or receive documented violations. DO 806.1.2.
The SDP is split into five “Phases.” Phases I through III take place in the Browning Unit and each lasts 180 days. DO 806.08. At Phase I, inmates complete “general evidence based programs” such as diversity training, high school equivalence preparation, cognitive thinking, and other rehabilitative programming. At Phase II, inmates are allowed outside of their cells individually and unrestrained in order to complete job assignments, and may walk to and from the shower and recreation unrestrained. DO 806.08.1.5.2. They may also participate in town hall meetings in non-contact cells, restorative justice programming, and other rehabilitative programming. DO 806.08.1.5.2.1. At Phase III, inmates are allowed two-person recreation periods, job assignments, and one meal each day with other SDP inmates in which they are unrestrained. DO 806.08.1.5.3. They also continue programming, such as substance abuse activities, conflict resolution, domestic violence, and other treatment programs. DO 806.08.1.5.3. Phase IV involves transfer from the Browning Unit to a close custody unit and a four-week transition period in the close custody unit. DO 806.10.1.2. In the first week, SDP inmates may only eat and have recreation periods with other SDP inmates. DO 806.10.1.3. In the second, they may eat with other close custody inmates. DO 806.10.1.4.2. In the third, they may also have recreation periods with close custody inmates. DO 806.10.1.5.3. In the fourth week, inmates begin employment. DO 806.10.6.4. Upon completion of Phase IV, inmates begin Phase V, which is an indefinite period of monitoring. DO 806.10.1.7.
Inmates may be removed from the SDP for participating in STG activity or having a documented incident involving violence, a threat, a weapon, an improper use of a cell phone, or drug usage. DO 806.08.1.2. They may also be removed for participating in activities that adversely affect the safety of staff and the public, failing to complete all programming, and for disciplinary behavior that changes the inmate‘s classification or housing assignment. DO 806.11.1.4. At Phases I through IV, removals are documented, and the deputy warden for the inmate‘s assigned housing unit makes the final decision of whether to reinstate the inmate or terminate them from the SDP. DO 806.11.1. In these phases, inmates are not entitled to a revocation hearing or an appeal before the STG Appeals Committee, but the regular inmate grievance procedure remains available. Inmates removed at Phase V are entitled to a hearing before the STG Validation Hearing Committee with the right to an appeal to the STG Appeals Committee. DO 806.11.5.
An inmate terminated from the SDP must remain in maximum security for two years before he is eligible to participate again. DO 806.11.12. If an inmate is removed twice, he is permanently ineligible to participate in the SDP, although he may still renounce and debrief at any time. DO 806.11.13.
In October 2014, the STG Validation Hearing Committee classified Johnson as a validated member of the Warrior Society STG. He appealed, and the STG Appeals Committee affirmed his validation. He was transferred to the ASPC-Eyman Browning Unit in November 2014.
Johnson enrolled in the SDP in November 2016. He was terminated from the program in December 2017, due to an inconclusive polygraph. After Johnson submitted an inmate grievance and received a hearing, ADC determined that he successfully passed a polygraph and was eligible for Phase IV. In April 2018, Johnson began Phase IV and was transferred to the ASPC-Florence Central Unit, a close custody unit.
Pursuant to ADC policy, SSU officers searched Johnson and his belongings when he arrived at the close custody unit. SSU Sergeant Belt prepared a memorandum that described three STG-specific documents found in Johnson‘s belongings: (1) a calendar code specific to the Warrior Society with the name and ADC number of another Warrior Society validated inmate written in code; (2) a roster of Warrior Society members housed in Wing IV at the Browning Unit; and (3) a micro-note from one Dine Pride STG member to another that included a roster of Dine Pride members in the Browning Unit and described other Dine Pride activities.1 Then-Deputy Warden Ruben Montano stated that Belt‘s memorandum and evidence was sufficient to remove Johnson from the SDP becаuse it demonstrated that he violated the program‘s prohibition on STG activity. Johnson was terminated from the SDP on April 23, 2018. When Johnson asked Belt why he was terminated from the SDP, Belt told him that “higher-ups” wanted Johnson off the yard and “jailhouse lawyers” were not welcome in Belt‘s unit. After Johnson asked Belt why he was labeled a “jailhouse lawyer,” Belt said, “you know why.”
On April 23, 2018, Johnson received a notice of hearing regarding his proposed placement in maximum custody. He also submitted a written statement in connection with the hearing, but the copies provided by both parties are illegible. Johnson attended the hearing on April 25, 2018. ADC staff determined that he would be returned to the Browning Unit because of his role as an active, validated STG member. Johnson was transferred back to the Browning Unit on April 30, 2018.
Johnson contested his transfer back to the Browning Unit in two ways: through the grievance process and through an appeal of his maximum custody placement. On May 2, 2018, he filed an informal grievance, and on May 15, 2018, he filed a formal grievance. Both grievances argued that he did not receive adequate process for his removal from the SDP and raised the possibility of retaliation. They further alleged that Johnson was not told the reason for his reassignment to maximum custody
In the meantime, on June 17, 2018, Johnson filed an appeal of his maximum custody placement. His appeal form repeated that he was not told the reason for his reassignment other than his status as an active STG member. The administrator of the Offender Services Bureau, Stacey Crabtree, denied the appeal, finding that Johnson was correctly classified to maximum custody based on the discovery of STG-specific documents in his belongings. After Johnson returned to the Browning Unit, SSU officers refused to tell him why he was terminated from the SDP, with one officer telling him that he “would need a ‘court order’ to know why [he] was terminated.” Johnson also sent a letter to ADC Director Charles Ryan. Ryan never responded, but Johnson received a response from a division director stating that the current SDP policy did not require a hearing for inmates removed from Phases I-IV.
C. Proceedings Below
Johnson filed suit under
On cross-motions for summary judgment, the district court granted summary judgment to Defendants on Johnson‘s remaining two claims. With respect to Johnson‘s due process claim against Ryan, the district court held that Ryan was not involved in the decision to remove Johnson from the SDP and Johnson failed to allege that his removal was attributable to an unconstitutional policy, practice, or custom that could be attributed to Ryan. With respect to Crabtree and Days, the district court found that neither defendant had the authority to order additional procedural safeguards for Johnson‘s termination. Finally, with respect to Montano, the district court held that Johnson‘s liberty interest in avoiding the conditions of solitary confinement
Johnson timely appealed.
II. STANDARD AND SCOPE OF REVIEW
We review de novo a district court‘s dismissal under
To survive § 1915A review, a complaint must contain sufficient factual matter, accepted as true, to state a claim to relief that is plausible on its face. Moreover, we have an obligation where the petitioner is pro se, particularly in civil rights cases, to construe the pleadings liberally and to afford the petitioner the benefit of any doubt.
Id. at 642 (citations and quotations omitted).
We review de novo a district court‘s grant of summary judgment. Transgender L. Ctr. v. Immigr. & Customs Enf‘t, 46 F.4th 771, 778 (9th Cir. 2022). “We therefore employ the same standard used by the district court and must ‘view the evidence in the light most favorable to the nonmoving party, determine whether there are any genuine issues of material fact, and decide whether the district court correctly applied the relevant substantive law.‘” Id. (quoting Animal Legal Def. Fund v. FDA, 836 F.3d 987, 989 (9th Cir. 2016) (en banc) (per curiam)).
Johnson raises three issues on appeal. First, he raises a challenge under the Due Process Clause to ADC‘s policy of conducting annual reviews of the validation of STG members. Second, he claims the he had a liberty interest in the Step-Down Program and that his removal without a hearing violated his due process rights under the
III. WHETHER ADC‘s STG REVIEW VIOLATES DUE PROCESS
We first consider whether the district court erred when it screened Johnson‘s claim that ADC‘s annual reviews of his confinement status violate his due process rights under the
A. Johnson‘s Liberty Interest in Avoiding Maximum Security
A liberty interest “may arise from the Constitution itself . . . or it may arise from an expectation or interest created by state laws or policies.” Wilkinson, 545 U.S. at 221. The Constitution does not give rise to a liberty interest in avoiding transfer to more adverse conditions of confinement, but such an interest may “arise from state policies or regulations.” Id. at 221-22; Meachum v. Fano, 427 U.S. 215, 225 (1976) (“[T]he Due Process Clause [does not] . . . protect a duly convicted prisoner against transfer from one institution to another within the state prison system. Confinement in any of the State‘s institutions is within the normal limits or range of custody.“). However, an interest in avoiding certain conditions of confinement “will be generally limited to freedom from restraint which, while not exceeding the sentence in such an unexpected manner as to give rise to protection by the Due Process Clause of its own force, nonetheless imposes atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life.” Sandin v. Conner, 515 U.S. 472, 484 (1995) (citations omitted). “After Sandin, it is clear that the touchstone of the inquiry into the existence of a protected, state-created liberty interest in avoiding restrictive conditions of confinement is not the language of regulations regarding those conditions but the nature of those conditions. . . .” Wilkinson, 545 U.S. at 223.
Johnson has alleged sufficient facts to establish a liberty interest in avoiding maximum security confinement in the Browning Unit as a result of his STG validation. Johnson was not initially placed in the Browning Unit as a consequence of the crimes for which he was convicted and sentenced. See Wilkinson, 545 U.S. at 216 (noting that initial assignment to Ohio‘s supermax might follow “if the inmate was convicted of certain offenses“); Myron v. Terhune, 476 F.3d 716, 718 (9th Cir. 2007) (holding that an inmate has no liberty interest in his initial classification); see also Prieto v. Clarke, 780 F.3d 245, 254 (4th Cir. 2015) (rejecting a due process challenge to harsh conditions on death row). Johnson‘s complaint states that after he was validated as an STG member in 2014, he was moved to the Browning Unit, where he is confined to his cell for twenty-four hours per day, strip searched every time he leaves his cell, takes meals in his cell, and has limited access to rehabilitation programs. These conditions are substantially more restrictive than the general population from which he was moved. Johnson also alleges that he was denied the opportunity for restoration of lost earned release credits. The Supreme Court has held that similar conditions in an Ohio Supermax facility created a liberty interest because they “impose[d] an atypical and significant hardship under any plausible baseline.” Wilkinson, 545 U.S. at 223–24; see also Brown v. Or. Dep‘t of Corr., 751 F.3d 983, 988 (9th Cir. 2014) (holding that solitary confinement for a fixed and irreducible twenty-seven month period created a liberty interest). We hold that Johnson has a liberty interest in avoiding assignment to maximum custody as a consequence of his STG validation.
B. Sufficiency of Process
Although we find that Johnson has a liberty interest, he fails to state a claim for a due process violation under the Mathews three-prong framework. In this
1. Balancing inmate and prison interests
In the context of prison gangs, the balance of private and public interests—that is, the first and third prongs of the Mathews analysis—weigh heavily in favor of ADC. On the inmate side, Johnson must accept that almost any placement in the Arizona prison system is a severely restricted environment, and that “the procedural protections to which [inmates] are entitled are more limited than in cases where the right at stake is the right to be free from confinement at all.” Wilkinson, 545 U.S. at 225. On the state‘s side, “[p]rison security, imperiled by the brutal reality of prison gangs, provides the backdrop of the State‘s interest” and is a
“dominant consideration.” Id. at 227. Indeed, we have described prison gangs as a “chronic problem” in which “gangs engage in extortion, drug trafficking, assault, and murder within the prisons.” Griffin v. Gomez, 741 F.3d 10, 12 (9th Cir. 2014); see United States v. Silverstein, 732 F.2d 1338, 1341 (7th Cir. 1994) (describing the court‘s “horrifying glimpse of the sordid and lethal world of modern prison gangs“). We have observed that the assignment of gang affiliates to administrative segregation “is not а disciplinary measure, but an administrative strategy designed to preserve order in the prison and protect the safety of all inmates” and that “the assignment of inmates within the [State‘s] prisons is essentially a matter of administrative discretion.” Munoz v. Rowland, 104 F.3d 1096, 1098 (9th Cir. 1997); see also Bruce v. Ylst, 351 F.3d 1283, 1289 (9th Cir. 2003) (acknowledging that “prisons have a legitimate penological interest in stopping prison gang activity“); Mark S. Fleisher & Scott H. Decker, An Overview of the Challenge of Prison Gangs, Corr. Mgmt. Q., vol. 5, issue 1, at 2–5 (2001) (recounting the history, organization, and impact of prison gangs). And in this context, “courts must give substantial deference to prison management decisions before mandating additional expenditures for elaborate procedural safeguards when correctional officials conclude that a prisoner has engaged in disruptive behavior.” Wilkinson, 545 U.S. at 228.
In this case, ADC has determined that an inmate‘s STG status—the fact of membership in a gang, irrespective of his prison disciplinary record—is a sufficient indication of that inmate‘s security risk to justify continuing solitary confinement. Arizona is not alone in this assessment. As a report from the National Institute of Justice at the U.S. Department of Justice found: “Numerous responses to combat gangs have been implemented throughout U.S. prison systems, but only one has been described as a ‘silver bullet:’ removing gang affiliates from the general population and placing them in restrictive housing.” David C. Pyrooz, U.S. Dep‘t Justice, Nat‘l Inst. of Just., Using Restrictive Housing to Manage Gangs in U.S. Prisons (June 30, 2018), https://nij.ojp.gov/topics/articles/using-restrictive-housing-manage-gangs-us-prisons (footnote omitted). As a consequence, “Corrections officials have, overwhelmingly, endorsed the use of restrictive housing for gang affiliates.” Id.; see, e.g., Madrid v. Gomez, 889 F. Supp. 1146, 1240–44 (N.D. Cal. 1995) (describing California‘s program for segregating prison gang affiliates). The proposition is not without controversy, even among corrections
2. Risk of erroneous deprivation
With these considerations in mind, we turn to the second prong of the Mathews analysis. Given the procedures that ADC has in place, what is the risk that ADC officials will erroneously determine that Johnson remains a security risk to the prison? Johnson, with vigorous support from our dissenting colleague, argues two points. First, citing our decision in Toussaint v. McCarthy, 801 F.2d 1080, 1102 (9th Cir. 1986) (Toussaint III), overruled in part on other grounds, Sandin v. Connor, 515 U.S. 472 (1995), he argues that reviews of his confinement status cannot be “meaningless gestures” and must be held more frequently than once a year. See Dissenting Op. at 4 (Toussaint III “requires review of [segregated] placement more than once per year“). Second, he argues that irrespective of the frequency of such retention reviews, the ADC‘s review process is inadequate because periodic reviews have never considered whether he remains a threat to the institution, but “consider[] only an initial STG validation and the subsequent failure to debrief.” See id. at 9. We will address both points.
The frequency of review of Johnson‘s gang status. In Hewitt v. Helms, the Supreme Court set forth the general standard:
[A]dministrative segregation may not be used as a pretext for indefinite confinement of an inmate. Prison officials must engage in some sort of periodic review of the confinement of such inmates. This review will not necessarily require that prison officials permit the submission of any additional evidence or statements. The decision whether a prisoner remains a security risk will be based on facts relating to a particular prisoner—which will have been ascertained when determining to confine the inmate to administrative segregation—and on the officials’ general knowledge of prison conditions and tensions, which are singularly unsuited for “proof” in any highly structured manner.
459 U.S. 460, 477 n.9 (1983), abrogated in part on other grounds, Sandin, 515 U.S. 472.
We addressed the question of retention reviews in Toussaint III, 801 F.2d at 1101. To understand our statements, some context is required. Toussaint was a class action spanning nearly two decades and broadly addressing the conditions of confinement at four California prisons—San Quentin, Folsom, Deuel Vocation Institute at Tracy, and Correctional Training Facility at Soledad. The litigation had an extensive history in the district court and in our court.3 Relevant to this case, the district court addressed the segregated housing practices at San Quentin and Folsom and
information’ from a source of unproven reliability identifying the inmate as a gang member” and that inmates had been held continuously in lockup for more than one year, and “[some] had been so held for over two years.” Id. at 1407 (footnote omitted). The district court issued a permanent injunction “to correct [the prisons‘] arbitrary practices in imposing segregation for ‘administrative’ reasons, and in retaining prisoners in segregation” and required such inmates to “be released from segregation upon his Minimum Eligible Release Date (MERD), if any, or at the expiration of twelve (12) months” unless the inmate is afforded a hearing and it is determined that he fits one of the three justifications described above. Id. at 1424.
Both parties appealed the district court‘s injunction. We held that “the state‘s interest in maintaining security in San Quentin and Folsom is at least as great, if not greater, than the state‘s interest shown in Hewitt,” as California‘s facilities were “composed of the most violent and anti-social offenders in the California prison system.” Toussaint III, 801 F.2d at 1100. With respect to the decision to make the initial placement, we held that “due process only requires . . an informal nonadversary hearing within a reasonable time after the prisoner is segregated.” Id. (footnote omitted); see id. at 1100 n.20 (stating that “the district court‘s requirement that a hearing be held within 72 hours of segregation constitutes a ‘reasonable time‘” and “intimat[ing] no view as to whether due process would tolerate a more lengthy delay.“). With respect to retention reviews, we noted recent decisions of the Third and Eighth Circuits approving4 monthly reviews
On remand, the district court addressed objections from the parties in response to a report from the prison monitor appointed to supervise the injunction. Toussaint IV, 711 F. Supp. at 537. The monitor recommended that inmates in “indeterminate segregation” be reviewed every 90 days. The state objected and contended that review every 120 days was sufficient. Although the district court had previously approved up to a one-year retention review policy, the district court ordered the state to review inmates’ segregated status every 90 days. Id. at 539–40. We reversed, with little analysis and without citation to any decision of any court, including our prior decision in Toussaint III:
The Constitution does not support a nice distinction between 90 days and 120 days. The question is one of discretion. Is it to be the discretion of the prison administrators or the discretion of a district court? Nothing in the Constitution invests the district court with discretion to override the discretion of the prison officials on this administrative point. Here administrative discretion must prevail; 120 days satisfies due process.
The Toussaint litigation is of limited usefulness in this case. Between Toussaint III and Toussaint V, we are left with bare ipse dixit that 120-day review of segregated status is constitutionally acceptable, a one-year period is too long, but anything in between those poles is to be left to the discretion of prison officials. Nevertheless, our declaration in Toussaint III that an annual review period was too long would bind us—if annual review were the exclusive form of relief for Johnson. It is not. Johnson may renounce his gang status and debrief at any time. At a minimum, his willingness to renounce and debrief would earn Johnson a hearing to determine whether he had satisfied the criteria for renounce-and-debrief relief.
Because Johnson has not sought to renounce and debrief, we are hard-pressed to understand why annual rеview of STG status is not sufficient to satisfy the Due Process Clause. We need to be very precise here so that we cannot be misunderstood. We do not have inmates before us who are subject to maximum custody in Arizona for violations of disciplinary rules or have been segregated for their own protection, to use the language of the Toussaint litigation. See Toussaint II, 597 F. Supp. at 1393–94. Instead, Johnson is subject to segregation because he has been classified as a threat to the safety of the prison staff, other inmates, or the general security of institution. See id. That classification may encompass a variety of concerns. See Clark, 776 F.2d at 228 (inmate was subject to long-term segregation after numerous disciplinary reports and after he killed a guard and an inmate); Mims, 744 F.2d at 948 (inmate was subject to long-term segregation after killing a deputy warden). In Johnson‘s case, Arizona has articulated a particular concern. It has not alleged that Johnson has violated disciplinary rules or is an immediate threat to the staff or other inmates. Rather, Arizona is concerned that Johnson‘s gang membership presents an ongoing threat to the
As we discuss in greater detail in the next section, Johnson‘s argument really amounts to a disagreement about what criteria the prison should consider in determining whether he remains a security risk and only secondarily challenges how often ADC reviews his STG status. But as the Supreme Court noted in Hewitt, periodic reviews do not necessarily require additional evidence and may rely on facts that were ascertained when the initial decision to confine the inmate to administrative segregation was made. 439 U.S. at 477 n.9. Unless Johnson can show that his initial validation as an STG was in error—a claim Johnson does not make—then his recourse for the time being is to renounce his membership, thereby altering his status as a Warrior Society member.5 Neither Johnson nor the state has anything to gain by conducting monthly, 90-day, or 120-day reviews of Johnson‘s status as a gang member because nothing about his STG status has changed. Such periodic reviews would be useful to review conduct-based threats to prison seсurity, but that is not the basis for Johnson‘s segregation. See Madrid, 889 F. Supp. at 1278 (noting that the “lack of continuing evidence of gang membership or activity is
simply considered irrelevant since the justification for administrative segregation is the fact of gang membership itself, not any particular behavior or activity” and “the premise for finding that the inmate is a security risk—gang membership or association—is not affected by the lack of subsequent gang activity“). Because Johnson can initiate review of his segregated status at any time by indicating that he is prepared to renounce his membership, Johnson has some control over the review process. We will not engage in the empty formalism of requiring ADC to conduct a review at some point less than annually—every 364 days, for example—just to satisfy our broad statement in Toussaint III.6 We hold that annual review of Johnson‘s gang status plus the possibility of the opportunity to renounce and debrief is sufficient to satisfy the demands of the Due Process Clause. See Wilkinson, 545 U.S. at 217 (noting that once an inmate has been placed in the Ohio supermax facility, “his placement is reviewed on at least an annual basis“).
Johnson has framed his argument as a Mathews v. Eldridge challenge to Arizona‘s procedures. Properly considered, his argument does not sound in procedural due process. Rather, it is a fundamental disagreement with ADC‘s judgment that gang status is a sufficient grounds for placing Johnson, or any other gang member, in segregated housing. Johnson and the dissent challenge ADC‘s judgment about what criteria the prison should consider in determining whether Johnson remains a security risk. Johnson‘s argument sounds in substantive due process, not procedural due process.7 See United States v. Salerno, 481 U.S. 739, 746 (1987) (describing the difference between the substantive and procedural components of the Due Process Clause); Mills v. Rogers, 457 U.S. 291, 299 (1982) (similar);
United States v. Quintero, 995 F.3d 1044, 1051–52 (9th Cir. 2021) (similar). No additional process would satisfy the dissent‘s claim that status alone is not sufficient cause for maximum custody—hence the dissent‘s candid acknowledgment that it wouldn‘t matter “if Arizona reviewed Johnson‘s placement monthly or even daily” Dissenting Op. at 69—because it would not change ADC‘s judgment that STG status is a critical, determinative fact. See Reno v. Flores, 507 U.S. 292, 301–02 (1993) (stating that due process “includ[es] a substantive component, which forbids the government to infringe certain ‘fundamental’ liberty interests at all, no matter what process is provided, unless the infringement is narrowly tailored to serve a compelling state interest“); Daniels v. Williams, 474 U.S. 327, 331 (1986) (characterizing certain substantive due process rights as “bar[ring] certain government actions regardless of the fairness of the procedures used to implement them“).
The proposition that gang membership alone threatens prison security is both outside of our expertise and too well established for us to consider sua sponte a substantive due process challenge to ADC‘s judgment. Hewitt established that
Johnson‘s confinement in maximum security is based on the well-documented, near existential threat that gangs pose to prison order. See Wilkinson, 545 U.S. at 227 (“Clandestine, organized, fueled by race-based hostility, and committed to fear and violence as a means of disciplining their own members and their rivals, gangs seek nothing less than to control prison life and to extend their power outside prison walls.“); Fleisher & Decker, supra, at 2–5. In a lengthy, thorough opinion in Madrid v. Gomez, the district court for the Northern District of California reviewed California‘s approach to prison gangs and considered the conditions at Pelican Bay Prison, a modern facility designed to hold “the worst of the worst.” 889 F. Supp. at 1155. The court found that “[a]lthough both prison gangs and
disruptive groups [such as motorcycle and street gangs] pose threats to prison security, prison gangs are considered the greater threat.” Id. at 1240. Under California‘s policies, gang members or associates could be placed in segregated housing for an indefinite term, unless an inmate debriefs. Id. at 1241. The district court found that although segregated housing “contains an element of punishment and creates a deterrent effect,” restrictive “conditions in the SHU serve to undermine [gang] networks and opportunities by separating gang members from one another.” Id. at 1275. The court also found that California “considered [an inmate] to be a security threat so long as the inmate is validated as a gang affiliate and has not yet debriefed . . . even if the inmate has, for some period of time ‘remained clean.‘” Id. at 1278. It further explained that “[t]he lack of continuing evidence of gang membership or activity is simply considered irrelevant since the justification for administration segregation is the fact of gang membership itself, not any particular behavior or activity. . . . Therefore, the premise for finding that the inmate is a security risk—gang membership or association—is not affected by the lack of subsequent
As in Madrid, it is Johnson‘s continuing status as a gang member that is critical to the state‘s interest in maintaining him in maximum security. It is appropriate for ADC to rely on Johnson‘s STG validation status as justification for its conclusion that he remains a security threat, and ADC is not required to consider additional evidence such as his criminal history, propensity for violence, or his immediate past disciplinary history. If the only evidence in Johnson‘s file was “for specific, serious misbehavior,” then “more formal, adversary-type procedures might be useful” to avoid the possibility of administrative error. Wilkinson, 545 U.S. at 228.
Johnson has not alleged any facts that would demonstrate that ADC‘s determination that he is a member of the Warrior Society is erroneous or that his STG status is being used pretextually. For instance, Johnson does not allege that the ADC determination in his case is based on stale information or is so outdated as to be irrelevant to a current risk analysis. To the contrary, ADC‘s evaluation focuses on information that remains unchanged: Johnson was properly validated as a member of the Warrior Society, he has not renounced his membership and debriefed, and the Society still operates as a prison gang and has active members in Arizona prisons. See Ariz. Dep‘t of Corr., Certified and Monitored Security Threat Groups, https://corrections.az.gov/warrior-society-0 (last visited Oct. 24, 2022); but see Dissenting Op. at 74 (“Arizona would still need to show that a years or decades-old STG validation, coupled with a prisoner‘s subsequent failure to debrief, actually establishes current gang status.“). Thus, Johnson does not adequately allege that there exists a risk that ADC has erroneously declared him to be a security threat—Johnson (like the dissent) simply disagrees with ADC‘s judgment about what criteria are relevant. But Johnson has not brought a substantive due process claim, and the Suprеme Court has said that periodic reviews may be based on facts ascertained when initially assigning the inmate to administrative segregation. See Hewitt, 459 U.S. at 477 n.9.
Johnson also suggests that ADC‘s reliance on his STG status and failure to debrief does not meet the “some evidence” standard because these periodic reviews consider only whether the prisoner “was previously validated” and “has debriefed.” The “some evidence” standard requires courts to determine “whether there is any evidence in the record that could support the conclusion reached by the disciplinary board.” Superintendent, Mass. Corr. Inst., Walpole v. Hill, 472 U.S. 445, 455–56 (1985). This evidence must have “sufficient indicia of reliability.” Bruce, 351 F.3d at 1288. Although the age of evidence could affect its weight, Castro v. Terhune, 712 F.3d 1304, 1315 (9th Cir. 2013), we do not reweigh evidence when determining whether there is “some evidence” for due process purposes, Bruce, 351 F.3d at 1287. Johnson does not challenge his initial validation or otherwise allege that this evidence is unreliable.
* * *
Our review of the three Mathews prongs shows that ADC‘s annual reviews of Johnson‘s STG status are not constitutionally deficient. ADC is entitled to substantial deference in its determination that an inmate‘s STG membership and failure to
C. Response to the Dissent
Our dissenting colleague comes to the conclusion that ADC‘s description of its system for dealing with gang membership is unconstitutional, see Dissenting Op. at 68 (“I do not believe that the possibility of debriefing suffices to render Arizona‘s otherwise unconstitutional practice constitutional“), and that additional factfinding is required to bring Arizona‘s program in line with the Constitution, see Id. at 73–79. In addition to accepting Johnson‘s arguments, the dissent makes two additional points that merit response.
1. Renouncing and debriefing as process
The dissent argues that renouncing and debriefing is not an effective way out of segregated housing. Id. at 66, 73–75. The dissent calls debriefing “euphemistic,” “practically impossible,” “a pseudo remedy,” and a “mirage.” Id. at 66. Where the ADC regulations specify that a validated STG member is “[p]ermitted to renounce and debrief at any time,” DO 806.07.1.1.9, and will then be considered to be a “former [gang] member,” DO 806 at 25, the dissent dismisses renouncing and debriefing as “theoretical eligibility” that may not offer inmates “a plausible path” out of segregated housing. Dissenting Op. at 76. The dissent‘s principal explanation is a hypothetical that Johnson “may not be able to successfully debrief even if [he] wished.” Id. at 77. The dissent explains that under ADC‘s regulation, an inmate who is renouncing and debriefing must “provide additional information regarding the STG‘s structure, activity and membership that would adversely impact the STG and assist in management of the STG population.” DO 806.061.1.2. Even though Johnson has never attempted to renounce and debrief, the dissent hypothesizes that Johnson will not be able to satisfy these criteria because “it is not at all clear how a prisoner who was validated eight years ago and has been held in solitary confinement ever since could possibly be in a position to provide any information that would ‘adversely impact the STG’ or ‘assist’ the prison in ‘management of the STG population.‘” Dissenting Op. at 77.
These claims are not established anywhere in this record or the record of any other case, but are entirely of the dissent‘s own imagination. Debriefing has been widely used in prisons in this circuit. See, e.g., Hinojosa v. Davey, 803 F.3d 412, 416–17 (9th Cir. 2015) (describing California‘s debriefing program), rev‘d on other grounds, 578 U.S. 412 (2016); Griffin, 741 F.3d at 12 (same), Madrid, 889 F. Supp. at 1240–44 (same); Nev. Dep‘t of Corr., AR 446.03(3) (providing for a debriefing process); Wash. Dep‘t of Correc., DOC 470.500(IV) (providing for a debriefing process); see also Pyrooz, supra, (“Debriefing . . . remains an established route out of restrictive housing . . . .“).9 And the record in one of our cases disclosed that debriefing has been used
and were successfully debriefed by the STG Unit.” Security Threat Group (STG) Program Evaluation, supra, at iii.
So long as Johnson refuses to renounce and debrief, the dissent‘s objection is not ripe. We will not make up objections to the debriefing program in the name of due process.
2. Protective custody and segregated housing
The dissent claims that renouncing and debriefing is ineffective as a way out of segregated housing because once an inmate renounces his gang membership he may be targeted by the gang and will end up in segregated housing as a protective measure. Dissenting Op. at 65–66, 75–76. The dissent objects that even if Johnson renounces and debriefs, he would be given “the opportunity to trade one form of solitary confinement for another.” Id. at 76.
The problem the dissent identifies is real and is sometimes referred to as the “snitching” dilemma. See Hinojosa, 803 F.3d at 416 & n.3; Griffin, 741 F.3d at 13; see also Gonzales v. Calif. Dep‘t of Corr., 739 F.3d 1226, 1234–35 (9th Cir. 2014). In the 2001 Arizona study of STGs, the authors found that “[t]he rate of renouncement . . . [was] low in part due to the threat of retaliation from members of the gang, and in part to the lack of a strong incentive to renounce, i.e., most renounced members remain in a supermax security unit.” Security Threat Group (STG) Program Evaluation, supra, at iii. ADC is not blind to the problem. Its regulations contemplate that an inmate who has renounced and debriefed will be eligible for lower custody housing or transfer out of state, but ADC has acknowledged the reality that some inmates who renounce their gang membership may have to be placed in protective custody, voluntarily or involuntarily. DO 806.07.1.5.1, 806.07.1.5.2; Report and Recommendations Concerning the Use of Restrictive Housing, supra, at 23 (observing that many inmаtes in protective custody have requested their removal from the general population; others are involuntarily removed).
The dilemma identified by the dissent cannot be avoided. But protective custody in segregated housing is only necessary as long as ADC or the inmate believes it is necessary to guarantee the safety of the inmate. It is not intended to be punishment for violation of prison rules or to protect others from the segregated inmate. It is for the protection of the former gang member and is the direct result of the inmate‘s own unfortunate past associational choices. The Supreme Court has reminded us that “[t]he safety of the institution‘s guards and inmates is perhaps the most fundamental responsibility of the prison administration.” Hewitt, 459 U.S. at 473. Accordingly, “prison officials have a duty . . . to protect prisoners from violence at the hands of other prisoners.” Farmer v. Brennan, 511 U.S. 825, 833 (1994) (quotation marks and citation omitted). Prison officials must strike a careful balance to determine who must be protected
Moreover, the dissent‘s ultimate conclusion cannot be correct. Having decided that renouncing and debriefing as a way out of segregated housing is illusory and that an inmate who debriefs will end up in segregated housing anyway as a protective measure, the dissent concludes that
“Arizona cannot satisfy due process.” Dissenting Op. at 76. And if the Due Process Clause cannot be satisfied, then the inmate must be released from maximum custody. Put another way, in the dissent’s view, if Johnson does not have a way out of segregated housing, he cannot be placed in segregated housing at all. The conclusion is contrary to the judgment of federal and state correctional officials and inconsistent with the studies showing that segregated housing is an effective means for controlling the threat of prison gangs to the safety of correctional officers and inmates. We cannot find any warrant in the Due Process Clause for this line of reasoning.
IV. REMOVAL FROM SDP AND DUE PROCESS
We now consider whether the district court erred in granting summary judgment to Defendants on Johnson’s claim that his removal from the SDP violated his due process rights under the
A. Participating in the SDP as a Liberty Interest
As we discussed in the prior section, a liberty interest “may arise from the Constitution itself . . . or it may arise from an expectation or interest created by state laws or policies.” Wilkinson, 545 U.S. at 221. Not every program or policy implemented by a state, however, creates a life, liberty, or property interest protected by the Due Process Clause itself. See, e.g., Wolff v. McDonnell, 418 U.S. 539, 557 (1974). We are not aware of any principle of constitutional law that would require Arizona to create a program such as SDP to permit a prisoner to exit solitary confinement. Thus, we turn to the doctrine of state-created liberty interests.
1. The law of state-created liberty interests
The doctrine of state-created liberty interests, which developed primarily in the prisoners’ rights context, has evolved parallel to the Supreme Court’s doctrine of state-created property interests that began in the 1970s. See id. at 557–58. The “new property” revolution began with Goldberg v. Kelly, in which the Supreme Court acknowledged that welfare
In Wolff v. McDonnell, the Court addressed for the first time the extent to which incarcerated inmates have state-created liberty interests. A Nebraska statute awarded inmates “good-time credits,” which allowed inmates credit toward early release for good behavior. 418 U.S. at 546 n.6. Nebraska’s statutory disciplinary scheme allowed for forfeiture or withholding of such credit only for serious misconduct. Id. at 546–47. The Court held that nothing in the
The Court further developed the doctrine in Meachum v. Fano, which addressed a claim by Massachusetts inmates that they were entitled to due process protection before being transferred from the general prison population to a maximum security institution for administrative reasons. 427 U.S. at 216–22. Looking to thе nature of the interest at stake, the Court found that “[c]onfinement in any of the State’s institutions is within the normal limits or range of custody. . . . That life in one prison is much more disagreeable than in another does not in itself signify that a Fourteenth Amendment liberty interest is implicated when a prisoner is transferred.” Id. at 225. The Court acknowledged that Wolff’s approach to state-created liberty interests was consistent with Roth, Sindermann, Kelly, and other state-created property interest cases. Id. at 226.
Unlike in Wolff, however, in Meachum, Massachusetts had not created a right for prisoners to remain in a particular prison or security level—Massachusetts did not condition transfer between prisons on the occurrence of specific events, and instead vested transfer decisions to the discretion of prison officials. Id. at 226–28. Thus, the Court held that the inmates did not have a liberty interest in avoiding transfer to less favorable conditions. Id. at 228–29. The Court reasoned that recognizing a liberty interest in any change in prison conditions “would subject to judicial review a wide spectrum of discretionary actions that traditionally have been the business of prison administrators rather than of the federal courts.” Id. at 225.
After Meachum, the Court began to employ a methodology that required parsing
prison context. 515 U.S. at 483 (“[W]e believe that the search for a negative implication from mandatory language in prisoner regulations has strayed from the real concerns undergirding the liberty protected by the Due Process Clause.”). Instead, the Court “return[ed] to the due process principles . . . applied in Wolff and Meachum” that turned on the “nature of the deprivation.” Id. at 481, 483. After Sandin, states may create liberty interests, but “these interests will be generally limited to freedom from restraint which . . . imposes atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life.” Id. at 484 (citations omitted). The Court ultimately held that disciplinary segregation was not a dramatic departure from the ordinary incidents of prison life and therefore did not implicate a constitutional liberty interest. Id. at 486. After Sandin, we no longer parse state statutes and regulations for “mandatory language,” but we look to the nature of the deprivation to determine if the state has created some “underlying substantive interest” that rises to the level of a legitimate claim of entitlement. See Town of Castle Rock, 545 U.S. at 757.
The Supreme Court has also made clear that, for purposes of due process analysis, substantive rights—life, liberty, and property—are distinct from the procedures that are designed to protect them. See id. at 758–68; id. at 772 (Souter, J. concurring) (“[T]he property interest recognized in our cases has always existed apart from state procedural protection.”). In Olim v. Wakinekona, the Court held that an inmate did not have an independent liberty interest in processes that might protect him from transfer from a prison in Hawаii to a prison in California. 461 U.S. at 250. The Court noted that “[p]rocess is not an end in itself. Its constitutional purpose is to protect a substantive interest to which the individual has a legitimate claim of entitlement.” Id. The mere fact that the State had provided procedures “d[id] not create an independent substantive right” in those procedures. Id. at 250–51. Similarly, in Cleveland Board of Education v. Loudermill, the Court found that a state civil service statute created a property right, the scope of which was not defined by the procedures described in the statute. 470 U.S. 532, 538–41 (1985). The Court distinguished between the substantive rights of life, liberty, and property guaranteed by the Due Process Clause and constitutionally adequate procedures. It noted, “The categories of substance and procedure are distinct. Were the rule otherwise, the [Due Process] Clause would be
In sum, although “[a] state-created right can, in some circumstances, beget yet other rights to procedures essential to the realization of the parent right,” Dumschat, 452 U.S. at 463, a plaintiff does not have an independent right to those procedures. See Dist. Att’y’s Off. for Third Jud. Dist. v. Osborne, 557 U.S. 52, 67–68 (2009) (holding that because a plaintiff did not have a liberty interest in state executive clemency, he therefore did not have an interest in “any procedures available to vindicate an interest in state clemency”). Only if a party has shown a liberty interest does the Due Process Clause require procedural protections before the state may deprive the party of his state-conferred interest. See Dumschat, 452 U.S. at 463 (“[T]he underlying right must have come into existence before it can trigger due process protection.”). In this way, the Court’s consistent adherence to the distinction between a substantive interest and procedure is consistent with the two-step analysis for the Due Process Clause. At the first step, we must be careful not to confuse procedure with the underlying substantive interest that gives rise to a legitimate claim of entitlement, because we will consider the adequacy of those procedures in protecting that entitlement at the second step.
Johnson argues that he has an independent liberty interest in remaining in the SDP under twо theories: completion of the SDP is one way for him to secure good-time credits and parole eligibility, and completion of the SDP allows him to return to close custody and avoid the harsh conditions of maximum custody. We will address each theory separately.
2. Liberty interests and continued participation in the SDP
We disagree that ADC has created a liberty interest in Johnson’s participation in the SDP. Completion of the SDP may be a means for acquiring eligibility for good-time credits, parole, and avoiding maximum security, but that does not establish an independent liberty interest in mere participation in the SDP. The SDP is one of several programs that ADC has provided Johnson to permit him to change his confinement status, including renouncing his STG membership and debriefing, which he may do at any time. But the mere fact that ADC has provided Johnson with these programs does not create a liberty interest in them—the SDP “is not an end in itself.” Wakinekona, 461 U.S. at 250. The SDP is no different than the procedures protecting an inmate from transfer in Wakinekona, the post-discharge review in Loudermill, or the arrest warrant in Town of Castle Rock. That is, the SDP is a process by which Johnson can leave maximum custody and regain eligibility for good-time credit and parole. It is not itself a liberty interest, but only one means by which Johnson can prove that he is prepared to return to the general prison population.
Removal from the SDP does not itself constitute an “atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life,” for two reasons. See Sandin, 515 U.S. at 484. First, removal from the SDP during Phases I–III does not result in any significant change in an inmate’s conditions of confinement. Phases I–III all take place in the Browning Unit, where STG validated inmates are housed. So underlying conditions of confinement are the same throughout these phases. Second, inmates in the general population and in other forms of administrative segregation do not have access to the SDP. The inmate removed from SDP has only lost аccess to one of several procedures by which he might change his conditions of confinement, and that alone is insufficient to create a liberty interest independent of any underlying change to Johnson’s conditions. Our understanding is consistent with Sandin’s instruction that state-created liberty interests in the prison context are “generally limited to freedom from restraint.” Id.
Our dissenting colleague disagrees. The dissent catalogues changes in an inmate’s living conditions as he moves from Phase I to Phase III and concludes that “participation in any stage of the SDP . . . entails significantly more freedom from restraint and social exposure than ordinary placement in maximum custody.” Dissenting Op. at 83. In addition, the Dissent argues that Johnson’s underlying housing assignment is merely “collateral” to the SDP. Id. at 86.
We are not persuaded. As we noted supra, following Sandin and Wilkinson, only a change in placement that works an “atypical and significant hardship” creates a liberty interest protected by the Due Process Clause. Wilkinson, 545 U.S. at 224; Sandin, 515 U.S. at 484. In Wilkinson, the Supreme Court observed that “[i]n Sandin’s wake the Courts of Appeals have not reached consistent conclusions for identifying the baseline from which to measure what is atypical and significant in any particular prison system.” 545 U.S. at 223; see Aref v. Lynch, 833 F.3d 242, 253–56 (D.C. Cir. 2016) (surveying the cases). In Ramirez v. Galaza, however, we addressed what kinds of circumstances count as “atypical and significant.” 334 F.3d 850 (9th Cir. 2003). Acknowledging that “[t]here is no single standard,” we said that the inquiry should be guided by three considerations:
1) whether the challenged condition “mirrored those conditions imposed
upon inmates in administrative segregation and protective custody,” and thus comported with the prison’s discretionary authority; 2) the duration of the condition, and the degree of restraint imposed; and 3) whether the state’s action will invariably affect the duration of the prisoner’s sentence.
Id. at 861 (quoting Sandin, 515 U.S. at 486–87); see also Aref, 833 F.3d at 255 (adopting similar criteria). Not every transfer accompanied by marginally harsher conditions creates a liberty interest. As the Court said in Hewitt, “transfer of an inmate to less amenable and more restrictive quarters for non-punitive reasons is well within the terms of confinement ordinarily contemplated by a prison sentence.” 459 U.S. at 468. And in Meachum, the Court held that no liberty interest was implicated by a transfer that “place[d] the prisoner in substantially more burdensome conditions” because such transfers may be made on the basis of “informed predictions as to what would best serve institutional security or the safety and welfare of the inmate.” 427 U.S. at 215, 225.
We have had few occasions to apply the guideposts we set forth in Ramirez, and the cases we have decided are not particularly helpful here.11 Nothing, however, in our cases would suggest that denying an inmate a two-person recreation period, favorable job assignments, unrestrained meals, unrestrained walks and access to the showers and recreation areas, or access to a GED program rises to the level of an “atypical or signifiсant hardship.” See Dissenting Op. at 83–84. It is true that these changes to Johnson’s circumstances in Phases II and III of the SDP are perquisites of the program, put in place by Arizona to encourage inmates to continue participating in the SDP. These benefits may not feel trivial to an inmate who has been isolated and experienced only limited social contact. But they do not represent a beyond-standard deviation from the ordinary circumstances of prison life. See Sandin, 515 U.S. at 484.
In these initial phases, Johnson remains in maximum custody. Thus, there has been no material change in the underlying condition of his confinement. Depriving an inmate like Johnson of these incidental, fleeting benefits does not introduce an “atypical and significant hardship” that would trigger a liberty interest. Id. If, as the dissent would hold, any incidental deprivation counts under Sandin, then we would have to “subject to judicial review a wide spectrum of discretionary actions that traditionally have been the business of prison administrators rather than of the federal courts.” Meachum, 427 U.S. at 225. We decline to
follow the dissent down that road. We thus disagree with the dissent that removing an inmate from “any prior phase” implicates a liberty interest under the Due Process Clause. Dissenting Op. at 84.12
Although we conclude that Johnson has no liberty interest created by the SDP, we think that Johnson has adequately stated a liberty interest in avoiding a return to maximum custody from close custody. Once Johnson attained Phase IV, he was moved from maximum custody in the Browning Unit to a close custody facility—and that
move constituted a material change in his living conditions. His expulsion from Phase IV of the SDP meant that he was removed from close custody and returned to maximum custody. Johnson’s liberty interest in avoiding maximum custody is clearly established: “After Sandin, it is clear that the touchstone of the inquiry into the existence of a protected, state-created liberty interest in avoiding restrictive conditions of confinement is . . . the nature of those conditions themselves ‘in relation to the ordinary incidents of prison life.’” Wilkinson, 545 U.S. at 223 (quoting Sandin, 515 U.S. at 484).13
In considering whether conditions of confinement impose an “atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life,” Sandin, 515 U.S. at 484, we may consider “whether the challenged condition mirrored those conditions imposed on inmates in administrative segregation and protective custody”; “the duration of the condition, and degree of restraint imposed”; and “whether the state’s action will invariably affect the duration of the prisoner’s sentence.” Brown, 751 F.3d at 987 (quoting Ramirez v. Galaza, 334 F.3d 850, 861 (9th Cir. 2003)). In Brown, we held that an inmate had a liberty interest in avoiding confinement in an “Intensive Management Unit” because inmates in the unit were held in
solitary confinement for over twenty-three hours a day, with limited exceptions for recreation and non-contact visits, and inmates were confined for minimum periods of twenty-seven months without meaningful annual review. Id. at 985, 988. In Wilkinson, the Supreme Court held that similar conditions, coupled with a loss of parole consideration, constituted an atypical and significant hardship that gave prisoners a liberty interest in avoiding assignment to Ohio’s Supermax facility. 545 U.S. at 223–24.
The facts of this case parallel Brown and Wilkinson. Johnson’s declaration states that, in maximum custody, he is confined to his cell twenty-four hours per day and is strip searched and handcuffed
By contrast, inmates in close custody have significantly greater freedom than those in maximum custody. Close-custody inmates live at different facilities from maximum-security inmates and enjoy reduced security protocols. See generally Ariz. Dep’t of Corr., Dep’t Order Manual, Dep’t Order 801 [hereinafter
Thus, it is not Johnson’s removal from the SDP per se that creates an atypical and significant hardship, but the change in Johnson’s underlying conditions of confinement when he was moved from close custody and returned to the Browning Unit. See Wolff, 418 U.S. at 571 n.19 (noting that the “imposition of ‘solitary’ confinement” requires procedural protection because it “represents a major change in the conditions of confinement”). We conclude that Johnson’s return to maximum security from close custody implicates a liberty interest under the Due Process Clause. We so conclude even though the SDP only contemplates a four-week duration for Phase IV.
B. Removing Johnson from SDP and Sufficiency of Process
Because we find that Johnson has alleged a liberty interest in avoiding the conditions of maximum custody, we must determine whether the procedures he was provided when he was moved from the Florence Central Unit to the Browning Unit in April 2018 were constitutionally adequate. We proceed with the Mathews analysis.
As we noted above, in the prison context, the first and third Mathews factors—Johnson’s interests and ADC’s interests—weigh heavily in favor of ADC. Because this case involves the assignment of an inmate to maximum custody based on his membership in a prison gang, Johnson’s private interest is limited, and ADC has a strong interest in mitigating the threat of STGs to prison security. Compare Wilkinson, 545 U.S. at 225 (“The [inmate’s] private interest at stake here, while more than minimal, must be evaluated, nonetheless, within the context of the prison system and its attendant curtailment of liberties.”), with id. at 227 (“In the context of prison management, . . . [the state’s] interest is a
Under the second Mathews factor, we consider the risk that under its procedures, ADC will erroneously reassign Johnson to maximum security. We may also consider “the probable value, if any, of additional or alternative procedural safeguards.” Id. at 225 (quoting Mathews, 424 U.S. at 335). The district court held that Johnson’s liberty interest was adequately protected by ADC’s annual reviews of his confinement in addition to his ability to renounce and debrief at any time. Although annual reviews and the opportunity to renounce and debrief might be adequate to protect an inmate’s liberty interest in avoiding retention in solitary confinement, we conclude that they are insufficient for protecting a liberty interest in avoiding reassignment to such conditions. See Toussaint III, 801 F.2d at 1098–1101 (analyzing separately the adequacy of procedures for placement and retention in solitary confinement).
The Supreme Court’s “procedural due process cases have consistently observed that [notice of the factual basis for a decision and a fair opportunity for rebuttal] are among the most important procedural mechanisms for purposes of avoiding erroneous deprivations.” Wilkinson, 545 U.S. at 226. Under the SDP policy, an inmate removed from Phase IV—which will result in the inmate’s transfer from close custody to maximum custody—is entitled to a hearing including ten days’ notice “to enable the inmate time to prepare a defense,” written notice of the decision, and a right of appeal.
Before Johnson was transferred back to the Browning Unit, he was given notice of and attended a hearing on his maximum custody placement.15 However, the notice only states that Johnson was being returned “as an active validated security threat group inmate.” Nothing in that notice would have apprised Johnson of the reason for his
reassignment. Johnson had been a validated STG member since 2014; thus, the notice did not proрose a change in his status, only a continuation of his status.
Johnson filed grievances indicating that he was never told why he was being moved back to the Browning Unit other than that his SDP status was revoked and he had been determined to be an “active validated [STG] inmate.” Deputy Warden Days denied Johnson’s formal grievance on June 1, 2018. Days acknowledged that Johnson had asked for a revocation hearing. For the first time, Johnson learned that he had been removed from close custody because he had violated “one or several of the criteria” in
I was . . . led to believe that I would need a court order to see what is being used against me to have me transferred back to maximum security. (I was perfectly clear in my I/M grievance that I haven’t been told why I was rolled up from central unit). . . . I don’t have an administrative appeal process available to me to dispute, rebut, and/or appeal this arbitrary decision that (puts me) a level away from being kept (in maximum security indefinitely for a non-disciplinary reason.)
(spelling and punctuation in original). His grievance appeal was denied, and he was told only that “[y]our removal from the STG Step Down Program was done in accordance with Department Policy.”
Johnson’s notice of appeal from the hearing decision, filed on June 17, 2018, tells a similar story. He complained that he “was given no information other than box 5 being checked on [the notice of hearing] and a statement saying I am being returned to Browning as ‘an active validated [STG] inmate.’” He further stated that he learned after the hearing that he was accused of violating several criteria in
I have learned that Central Unit’s deputy warden removed me arbitrarily from the STG Step-Down Program (SDP) due to violating 1 or several criteria’s outlined in Dept. Order 806. . . . Prison officials are refusing to tell me how many incident(s) I have allegedly violated and what are the substance of those alleged incident[s]. . . . I am not being given a revocation hearing to view, to dispute, to rebut, or to appeal these alleged incident(s) of violations.
His appeal was denied on July 24, 2018. In the denial, Crabtree advised him in the most general terms that the factual basis for his return to maximum custody was “your recent [STG] documented activity found in your belongings during a search by SSU officers on April 13, 2018.” This bare-bones explanation came three months after Johnson’s hearing.
In light of this record, the procedure that it appears Johnson was given was not adequate to satisfy the Due Process Clause. Johnson was not given a meaningful opportunity to learn of the factual basis for his transfer from close custody to maximum custody or to prepare a defense to the accusations. Johnson may or may not have violated the criteria for remaining in close custody, but the prison officials making that decision should make informed decisions—and the record available to us does not specify whether Johnson was made aware of the allegations against him. Without notice of the evidence against him, Johnson could not meaningfully respond and his hearing could not constitute an informed one. See Hewitt, 459 U.S. at 476 (holding that an inmate being considered for transfer to administrative segregation is entitled to “some notice of the charges against him and an opportunity to present his views to the prison official charged with deciding whether to transfer him to administrative segregation”); cf. Melnik v. Dzurenda, 14 F.4th 981, 986 (9th Cir. 2021) (holding, in the context of a prison disciplinary hearing, that a prisoner had a due process right to access evidence to be used against him). We have some frustration with the quality of the copies in the excerpts of record that deal with Johnson’s complaints concerning his hearing. From the evidence before us, Johnson was not given an adequate hearing before he was reassigned to maximum custody. At the very least, the district court should not
The district court addressed the individual liability of Ryan, Crabtree, Days, and Montano based on their authority for implementing the SDP, rather than their individual responsibility for denying Johnson notice of the factual basis for his change in conditions and an opportunity to present a rebuttal. On remand, the district court should decide in the first instance whether each defendant is individually liable for the constitutional deprivation as described in this opinion.
In sum, the district court was correct in finding that Johnson does not have an independent liberty interest in participation in the SDP process, but it erred in concluding that Johnson‘s liberty interest in avoiding reassignment to the restrictive conditions of the Browning Unit was adequately protected by the procedures he was provided when he was moved from close custody to maximum custody in April 2018. We also conclude that Johnson was likely denied due process in the procedures that resulted in his return to maximum custody. We reverse the district court‘s grant of summary judgment for Defendants on Johnson‘s Fourteenth Amendment due process claim and remand for further proceedings.
V. JOHNSON‘S FIRST AMENDMENT RETALIATION CLAIMS
Johnson alleges that he was removed from the SDP and returned to maximum custody because of his lawsuits against various ADC defendants. The district court granted summary judgment to Defendants, finding that the record did not support Johnson‘s assertions that his lawsuits were a substantial or motivating factor behind his removal.
“The most fundamental of the constitutional protections that prisoners retain are the First Amendmеnt rights to file prison grievances and to pursue civil rights litigation in the courts.” Entler v. Gregoire, 872 F.3d 1031, 1039 (9th Cir. 2017) (footnotes omitted). “[B]ecause purely retaliatory actions taken against a prisoner for having exercised those rights necessarily undermine those protections, such actions violate the Constitution quite apart from any underlying misconduct they are designed to shield.” Rhodes v. Robinson, 408 F.3d 559, 567 (9th Cir. 2005). We have said that a First Amendment claim in this context has five elements: (1) adverse action by a state actor against the inmate (2) because of (3) that prisoner‘s protected conduct, and the action (4) chilled the inmate‘s exercise of his First Amendment rights and (5) did not reasonably advance a legitimate correctional goal. Chavez v. Robinson, 12 F.4th 978, 1001 (9th Cir. 2021).
The parties agree that Johnson‘s removal from the SDP and transfer to maximum custody was an adverse action, that his lawsuits were protected First Amendment conduct, and that the adverse action chilled his exercise of such conduct. For purposes of summary judgment, a factual issue is genuine “if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Ochoa v. City of Mesa, 26 F.4th 1050, 1055 (9th Cir. 2022) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986)). The parties disagree on the existence of a dispute of material fact regarding whether Johnson‘s removal was in retaliation for his First Amendment activity and whether removal advanced a legitimate penological goal.
A. Retaliatory Motive
To establish a retaliatory motive, an inmate “must show that his protected conduct was the substantial or motivating
In this case, Johnson had a pending appeal before us when he was removed from the SDP and transferred back to the Browning Unit. See Johnson v. Bendel, 745 F. App‘x 750, 751 (9th Cir. 2018). His declaration also described an encounter with Belt—the author of the memorandum that provided the basis for Johnson‘s removal from the SDP—in which Belt told Johnson that “higher-ups” wanted Johnson off the yard and that “jailhouse lawyers” were not welcome in Belt‘s unit. Defendants characterize these facts as speculative, but we must view this evidence in the light most favorable to Johnson. See Transgender L. Ctr., 33 F.4th at 1193. These facts, if true, would allow a reasonable jury to return a verdict in Johnson‘s favor, so Johnson has raised a genuine dispute of material fact. See Bruce, 351 F.3d at 1289.
B. Reasonable Advancement of a Legitimate Correctional Goal
Johnson bears the burden of proving the absence of a legitimate correctional goal for the adverse action. See Pratt v. Rowland, 65 F.3d 802, 806 (9th Cir. 1995). Defendants argue that Johnson‘s removal advanced the legitimate correctional goal of curtailing prison gang activity and that Johnson‘s removal was supported by the evidence described in the Belt memorandum. In Bruce, we acknowledged that prisons have a legitimate interest in stopping prison gang activity, but held that this general justification was insufficient to show reasonable advancement of a legitimate correctional goal on summary judgment. 351 F.3d at 1289. We noted that, in light of the genuine dispute of material fact as to whether the action was taken with retaliatory motive, the defendants could not “assert that Bruce‘s validation served a valid penological purpose, even though he may have arguably ended up where he belonged.” Id. (emphasis in original).
As in Bruce, even if Johnson “arguably ended up where he belonged,” the presence of a genuine dispute of material fact with respect to a retaliatory motive means that Defendants’ general justification for the action is not sufficient to defeat summary judgment. Id. (emphasis omitted). Put differently, if Belt and Montano used procedures outlined in ADC‘s policies to remove Johnson from the SDP and transfer him back to the Browning Unit in order to punish Johnson for his lawsuits, their use of these procedures was pretextual and not a reasonable advancement of the legitimate penological goal of stopping prison gang activity. Johnson contests whether the documents described in the Belt memorandum are actually STG-specific. He provided alternative characterizations of the documents in question and evidence of SSU officers seizing materials that they had mistaken for STG material.
Viewing this evidence in the light most favorable to Johnson, we find that there is a genuine dispute of material fact with respect to whether Johnson‘s removal from the SDP and return to the Browning Unit reasonably advanced a legitimate penological purpose.
*
*
We reverse the district court‘s grant of summary judgment to Defendants on Johnson‘s First Amendment retaliation claim and remand for further proceedings.
VI. CONCLUSION
We affirm the district court‘s screening and dismissal of Johnson‘s Fourteenth Amendment due process claim regarding the adequacy of ADC‘s annual review process. We reverse the district court‘s grant of summary judgment for Defendants on Johnson‘s Fourteenth Amendment due process claim, holding that Johnson had a liberty interest in avoiding reassignment to maximum security and that there is a genuine dispute of material fact with respect to whether he was afforded constitutionally adequate process. Finally, we reverse the district court‘s grant of summary judgment on Johnson‘s First Amendment retaliation claim. We remand this case to the district court for further proceedings consistent with this opinion.
AFFIRMED in part, REVERSED and REMANDED in part.
RAKOFF, District Judge, concurring in part and dissenting in part:
Since 2014, Richard Johnson has been held in highly restrictive conditions approximating solitary confinement because Arizona at that time “validated” him as a member of a prison gang or Security Threat Group (“STG“). Although there is no basis to believe that Johnson has been involved in any STG activity since then -- indeed, the restrictive nature of his confinement virtually precludes such involvement -- Arizona, Johnson alleges, provides no reasonable way he can demonstrate he is not a threat to prison security and exit solitary confinement. It thus violates his constitutional liberty rights. Specifically, Johnson alleges that Arizona offers validated STG members no viable way out of these extremely restrictive conditions short of what is euphemistically termed “debriefing.” But “debriefing” consists not only of renouncing gang membership, but also informing on other gang members, which is practically impossible for someone who has had no access to other gang members for eight years. In short, it is a pseudo remedy. Moreover, Johnson alleges, since meaningful “debriefing” is premised on being an informant, the fact that Johnson would then be regarded as a “snitch” (even if he weren‘t) would mean that he would face potentially deadly threats to his safety that could only be protected against by putting him back in solitary confinement or its equivalent. In short, while Arizona provides the mirage that a once validated member of an STG can later escape solitary confinement, the reality is that he will be kept there for the entire duration of his sentence. Believing that this is unconstitutional, as well as contrary to past holdings of this Court, I dissent from the majority‘s analysis in Part III.A of its opinion and would instead reverse the district court‘s dismissal on the pleadings of Count III of Johnson‘s complaint. And while I concur in Part III.B of the majority‘s opinion, which reverses and remands the district court‘s entry of summary judgment against Johnson on his claim that his removal from the Step Down Program (“SDP“) violated due process, I write separately to emphasize that his claim is validly broader than the majority contends. Finally, however, I do concur fully in Part III.C of the majority opinion with respect to Johnson‘s retaliation claim.
I. Whether Arizona may confine prisoners in maximum custody based solely on prior STG status and the failure to debrief.
A. The Alleged Deficiencies in Arizona‘s Process
In Count III of his complaint, Johnson alleges that Arizona has denied him due
I agree with the majority that Johnson has adequately alleged a liberty interest in avoiding his extremely restrictive conditions of confinement. Opinion at 18-20. I likewise agree with the majority that Arizona, in order to deprive Johnson of this liberty interest, must make some judgment that Johnson remains a threat to prison safety. Opinion at 32. But I disagree that Johnson “has failed to plausibly allege how that judgment creates a risk that he will be erroneously classified as a security threat.” Opinion at 34-35. This is so for two reasons.
First, and very simply, we have previously held that the risk that a prisoner will be wrongly confined in solitary confinement or similarly restrictive conditions requires review of that placement more than once per year. Toussaint v. McCarthy (Toussaint III), 801 F.2d 1080, 1101 (9th Cir. 1986) (“We do not believe that annual review sufficiently protects plaintiffs’ liberty interest.“). See also Toussaint v. McCarthy, 926 F.2d 800, 803 (9th Cir. 1990) (”Toussaint V“) (holding that review every 120 days satisfied due process). Here, Johnson alleges that Arizona reviews his placement in what he calls solitary confinement only once per year. FAC ¶ 37. The majority describes our conclusion in the Toussaint litigation -- that prisons must review prisoners’ placement in extremely restrictive custodial conditions such as Johnson‘s more than once a year, but that review every 120 days is sufficient -- as “bare ipse dixit.” Opinion at 27. But this sensible balancing of the benefits and costs of additional process is precisely the balancing required under the Supreme Court‘s familiar framework for evaluating procedural due process claims. See Mathews v. Eldridge, 424 U.S. 319, 335 (1976).
The majority nonetheless acknowledges that Toussaint III is good law and that, under it, Arizona‘s annual review process would deny Johnson due process except (the majority contends) for the fact that Johnson could supposedly exit maximum security by “renounc[ing] his gang status and debrief[ing] at any time.” Opinion at 27. For the reasons described below, I do not believe that the possibility of debriefing suffices to render Arizona‘s otherwise unconstitutional practice constitutional, and especially not at the pleading stage, where Johnson‘s plausible allegations regarding the limitations of the debriefing process must be taken as true.
Second, and relatedly, it would not in any case matter if Arizona reviewed Johnson‘s placement monthly or even daily, because Johnson plausibly alleges that Arizona‘s
In this regard, the majority argues, bewilderingly, that “[t]his is not a Mathews v. Eldridge challenge to Arizona‘s procedures” and instead “sounds in substantive due process.” Opinion at 31. But Johnson‘s claim is that Arizona‘s categorical refusal to consider any evidence other than two facts -- a near-decade old STG validation, and Johnson‘s subsequent failure to debrief -- results in an unreasonably high probability that Johnson will be wrongfully deprived of his acknowledged liberty interest in avoiding the conditions of maximum custody. He contends that other procedures -- namely, procedures allowing consideration of other factors bearing on Johnson‘s dangerousness -- might lead to a different outcome. It is hard to imagine a more typical procedural due process argument. See, e.g., Mathews, 424 U.S. at 333-34 (discussing, as part of the Court‘s evaluation of the probable value of additional procedure, the types of evidence that would be relevant to a given type of determination, and what procedures would be necessary to enable consideration of the relevant evidence); Larita-Martinez v. INS, 220 F.3d 1092, 1095-96 (9th Cir. 2000) (discussing, as an element of procedural due process, the requirement that a decisionmaker consider relevant evidence).
Indeed, it is the majority that abandons the familiar Mathews balancing framework in favor of categorical presumptions favoring one party. It insists that Arizona‘s “subjective evaluation . . . that an inmate‘s STG membership and failure to debrief represents a continuing and significant risk to prison safety such that it justifies the inmate‘s confinement to maximum custody” must be accorded “significant deference.” Opinion at 32-33. But the majority in fact affords that “subjective” determination dispositive weight, allowing Arizona‘s mere assertion of it -- at the pleading stage, no less -- to trump any and all other considerations, including the likelihood that considering additional evidence might result in a different judgment as to whether Johnson poses a threat, and the likely cost of considering such additional evidence. Indeed, under the majority‘s telling, even if Johnson could show that some other factor (perhaps a prisoner‘s criminal or disciplinary history, or the subjective opinion of prison staff) was both easily ascertainable and had a 100% track record in predicting whether or not that prisoner‘s release from maximum custody would pose any danger to other prisoners
To support this counter-intuitive result, the majority cites language from the Supreme Court‘s decision in Hewitt v. Helms, 459 U.S. 460 (1983), arguing that ”Hewitt established that prison officials’ judgment that an inmate represents a threat to the safety of the prison may ‘turn[] largely on purely subjective evaluations and on predictions of future behavior’ and may be appropriate ‘even if [the inmate] himself has committed no misconduct.‘” Opinion at 32 (quoting Hewitt, 459 U.S. at 474, abrogated in part on other grounds, Sandin v. Conner, 515 U.S. 472 (1995)). But while Hewitt certainly afforded significant weight to the government‘s interest in prison security -- and real deference to prison officials’ judgment about how best to achieve that interest -- it also carefully applied the Mathews framework, independently weighing the government‘s security interests alongside the prisoner‘s interest in avoiding solitary confinement and the probable value of additional procedural safeguards. Hewitt, 459 U.S. at 474.
In the context presented in that case -- the decision whether to continue to segregate a particular prisoner mere weeks after a prison riot, when the situation in the prison continued to be volatile -- the Court found the private interest at stake (continuing to be held in segregation for a few more weeks “pending completion of an investigation into misconduct charges against” that prisoner, id. at 463-64, 476) not particularly weighty, and the proposed requirement of a live adversary hearing unnecessary. Id. at 474-75. It also acknowledged that predictions about when a particular prisoner‘s presence in the general population might cause violence are “subjective” and may turn in part on facts about general prison conditions not attributable to a particular prisoner‘s conduct. Id. at 474. But the Court in Hewitt made absolutely clear that “administrative segregation may not be used as a pretext for indefinite confinement of an inmate,” that “[p]rison officials must engage in some sort of periodic review of the confinement of such inmates,” and that such a review should take into account both “facts relating to a particular prisoner,” “the officials’ general knowledge of conditions and tensions,” “the progress of [any] investigation,” and “a wide range of administrative considerations.” Id. at 477 n.9.2 In other
words, Hewitt established that even for relatively brief periods of administrаtive segregation, due process requires consideration of a wide variety of factors. It provides no support for Arizona‘s
For these reasons, it seems clear to me that Johnson should at least be entitled to proceed past the pleadings on his claim that Arizona‘s refusal to consider factors other than his initial STG validation and his subsequent failure to debrief denies him due process. He has alleged that the consideration of additional factors -- specifically, his “criminal history, propensity for violence, or disciplinary record” -- might lead Arizona to determine, with greater accuracy, whether he poses a threat to prison security. FAC ¶ 37. Perhaps he is wrong that consideration of these factors would lead to a more accurate determination, or perhaps Arizona could show that it is overly burdensome to consider these factors, even if they are predictive of violence. But to settle these questions against Johnson at the pleading stage usurps the factfinder‘s role and eliminates from the Mathews analysis any consideration of the probable value of additional procedures, instead collapsing it into a one-pronged inquiry as to whether the government has asserted a legitimate security interest. That inquiry stacks the deck against the person asserting a liberty interest and in favor of the government.
B. The Significance of Debriefing
The majority‘s response to all of this is to argue that Johnson‘s “recourse for the time being is to renounce his membership, thereby altering his status as a Warrior Society member.” Opinion at 29; id. at 34 (“[I]t is appropriate for ADC to rely on Johnson‘s STG validation status as justification for its conclusion that he remains a security threat. . . .“); id. at 37-39 (discussing the debriefing process). But here, even granting the majority‘s premise that consideration of current gang affiliation standing alone and to the exclusion of all other factors might justify prolonged solitary confinement, Arizona would still need to show that a years or decades-old STG validation, coupled with a prisoner‘s subsequent failure to debrief, actually establishes current gang status. That premise is far from clear, and cannot be ascertained at the pleading stage. See Boquist v. Courtney, 32 F.4th 764, 773-74 (9th Cir. 2022) (“[D]ismissal is proper under Rule 12(b)(6) if it appears beyond doubt that the non-movant can prove no set of facts to support its claims.“). Indeed, “where, as here, a plaintiff proceeds pro se,” the district court was obliged to “construe the pleadings liberally” and “afford [Johnson] the benefit of any doubt.” Id. at 774; Erickson v. Pardus, 551 U.S. 89, 94 (2007) (“A document filed pro se is ‘to be liberally construed,’ and ‘a prо se complaint, however inartfully pleaded, must be held to less stringent standards than formal pleadings drafted by lawyers.‘“).
Rather than draw plausible inferences in Johnson‘s favor and construe his pro se complaint liberally, the district court -- and, now, the majority -- rush to resolve the disputed factual question of whether a years-old STG validation suffices to establish current gang status or dangerousness. There is of course no factual record as yet as to Johnson‘s claim that by conditioning his release from maximum custody on debriefing, Arizona effectively denied him any “meaningful opportunity” to demonstrate he was no longer a threat. FAC ¶ 39. And it was not Johnson‘s obligation at this preliminary stage of the litigation to allege anything more than that by conditioning his placement on debriefing, Arizona had effectively denied him the opportunity to leave maximum custody. But even looking ahead to future stages of litigation, it seems likely -- and, at the very least, plausible -- that Johnson could demonstrate
First, as the Supreme Court has emphasized, “[t]estifying against, or otherwise informing on, gang activities can invite one‘s own death sentence.” Wilkinson v. Austin, 545 U.S. 209, 227 (2005). Indeed, the concern that prisons cannot easily investigate gang violence and prevent retaliation after the fact is one of the reasons the Supreme Court has instructed courts to be reasonably deferential to prison procedures designed to prevent gang violence in the first place. Id. At the pleading stage, it certainly seems plausible that Johnson could demonstrate that he could not debrief without facing deadly danger. See Madrid v. Gomez, 889 F. Supp. 1146, 1241 (N.D. Cal. 1995) (“[A] number of prison staff agree that inmates who debrief and gain release from the SHU are considered ‘snitches,’ and thus face serious risks of being attacked or even killed by other inmates.“).
Second, largely because of the acknowledged danger that prisoners who have debriefed face from other prisoners, Arizona has pointed to prison regulations requiring that any prisoner who debriefs be placed in a form of “protective custody,” the purpose of which is to continue to separate the debriefed prisoner from other prisoners.
The majority acknowledges that this problem -- that a prisoner may not be able to debrief without either risking death or else simply trading one form of restrictive custody for another -- “is real . . . [but] cannot be avoided.” Opinion at 40. But if it really is true that debriefing requires braving violent retaliation or indefinite solitary confinement, it seems reasonable to ask whether consideration of any other factors beyond a prisoner‘s failure to debrief might satisfy a prison‘s legitimate security needs. Whatever the answer to that question, I feel confident it cannot be resolved at the pleading stage.
Third, Arizona‘s regulations (again, assuming they are even properly before us at this stage) on their face raise the plausible inference that many prisoners (likely including Johnson) may not be able to successfully debrief even if they wished. Under those regulations, no prisoner can debrief without first “provid[ing] additional information regarding the STG‘s structure, activity and membership that would adversely impact the STG and assist in management of the STG population.”
Johnson was validated in 2014 and Arizona has since twice allowed him to begin the SDP program, which he would not have been allowed to do on either occasion under Arizona‘s regulations if he had had any STG-related activity within the past two years. FAC ¶¶ 4, 9;
possibilities is clear from the face of Johnson‘s complaint, and none justifies dismissing it.
The majority dismisses this concern about the potential unavailability of debriefing to a prisoner who has been confined for years on-end and who therefore lacks current information about the STG into which he was validated, arguing that “[t]hese claims are not established anywhere in this record” and belong instead to the “dissent‘s own imagination.” Opinion at 38. This, once again, ignores the fact that this case is still at the pleading stage, not to mention that it involves a pro se pleading. Johnson‘s claim that Arizona must consider more than his failure to debrief in order to hold him in maximum custody for years on end has never proceeded to discovery, so there is, of course, no factual record affirming or disputing this point. But the possibility that debriefing might not be a viable option for a prisoner who has been held in maximum custody for many years more than plausibly follows from Arizona‘s regulations requiring satisfactory debriefing to establish information about the STG.
Instead, it is the majority that takes it upon itself to imagine facts beyond the pleadings (and, for that matter, an entire record as to this claim), where no such facts have yet been established by either party.4 Johnson has alleged that
by “conditioning
Indeed, in dismissing Johnson‘s claim at the pleading stage, the district court cited cases sustaining Arizona‘s system of indefinite maximum security placement because prisoners could theoretically leave maximum security by debriefing at “any time.” But every case cited by the district court came at the summary judgment stage, after the plaintiffs (usually proceeding pro se and in prison) had failed
to meet their burden to affirmatively adduce evidence demonstrating that debriefing was not a viable option. See Hernandez v. Schriro, No. 05-cv-2853, 2011 WL 2910710, at *8-9 (D. Ariz. July 20, 2011) (rejecting plaintiff‘s “unsupported assertion that risks to debriefed inmates are ‘common knowledge’ . . . [because] [t]o defeat summary judgment, Plaintiff must present evidence” (emphasis added)); Mendez v. Ryan, No. 10-cv-1867, 2013 WL 6408389, at *8-*11 (D. Ariz. Aug. 13, 2013) (similar); Standley v. Ryan, No. 10-cv-1867, 2012 WL 3288728, at *9-10 (Aug. 13, 2012 D. Ariz. 2012) (similar); Faulkner v. Ryan, No. 10-cv-2441, 2012 WL 407452, at *9-10 (D. Ariz. Feb. 9, 2012) (similar). District courts that have considered similar allegations at the pleading stage have routinely found them sufficiently plausible to proceed to discovery. See Askher v. Brown, No. 09-cv-5796, 2013 WL 1435148, at *7 (N.D. Cal. April 9, 2013) (finding allegations that the possibility of debriefing did not offer a meaningful path out of solitary confinement аt California prisons plausible at the pleading stage); Fanaro v. Cty. of Contra Costa, No. 09-cv-03247, 2019 WL 5191018, at *5 (N.D. Cal. Oct. 15, 2019) (similar); Gonzales v. Guirbino, No. 14-cv-00173, 2016 WL 1599449 at *2-5 (E.D. Cal. April 21, 2016) (similar).
Similarly, the majority relies extensively on a “lengthy, thorough” 1995 district court decision in Madrid v. Gomez, 889 F. Supp. 1146 (N.D. Cal. 1995) that held that California prisons could rely on the fact of a relatively old STG validation -- even in the “absence of gang-related activity or association over some period of time” -- to continue to hold a prisoner in solitary confinement. Opinion at 34 (quoting Madrid, 889 F. Supp. at 1278). But the court in Madrid reached that determination only after a trial in which it “heard testimony from 57 lay witnesses, including class members, defendants, and correctional employees at all levels,” and “received into evidence over 6,000 exhibits, including documents, tape recordings, and photographs, as well as thousands of pages of deposition excerpts,” and even “spent two days touring [the prison], accompanied by counsel for both parties and prison officials.” Madrid, 889 F. Supp. at 1156. The court‘s determination that a prisoner‘s status as a gang member might continue to demonstrate that prisoner‘s dangerousness and might be determined by an old STG validation, notwithstanding the absence of recent gang-related activity, was made only “in light of [the court‘s] factual findings,” and its resulting determination that “the
Madrid, in other words, demonstrates precisely why Johnson‘s claim cannot be dismissed on the pleadings, even assuming, as the majority does, that current gang status alone demonstrates dangerousness. Whether a years-old STG validation, coupled with a subsequent failure to debrief -- but also an apparent absence of any gang-related activity -- actually demonstrates present gang affiliation and dangerousness is a factual question, subject to factual dispute. The fact that a district court resolved this factual question one particular way with respect to a specific California prison‘s policies almost 30 years ago following a trial and the consideration of extensive evidence cannot justify resolving this question the same way against Johnson at the pleading stage.
For the reasons described above, I would hold that Arizona must review Johnson‘s placement in maximum custody more than once a year, and I would also allow Johnson to proceed to discovery as to his claim that Arizona violates due process by failing to consider any factors beyond a many-years-ago STG validation and a prisoner‘s subsequent failure to debrief before confining him indefinitely in maximum custody. In my view, if Johnson can demonstrate that consideration of other facts beyond a prisoner‘s current gang affiliation would both lead to more accurate determinations about whether that prisoner poses a threat and are not overly burdensome to consider, then Arizona must then consider those other factors. But even if that were not the case, and the majority were correct that a prisoner‘s current gang status, standing alone, justifies indefinite solitary confinement, I do not view the bare fact of a years-old STG validation and the mere possibility of debriefing as sufficient to establish current gang status at the pleading stage.
II. Whether Johnson‘s removal from the SDP violates due process.
I concur in the majority‘s judgment reversing the district court‘s grant of summary judgment as to Count I of Johnson‘s complaint (which alleged that he was terminated from the SDP program and reassigned to extremely restrictive conditions of maximum custody in the Browning unit without adequate due process). I also agree with the majority that “the procedure that it appears Johnson was given was not adequate to satisfy the Due Process Clause.” Opinion at 59. However, I do not agree with the majority‘s conclusion that only Johnson‘s reassignment of housing units -- rather than his underlying termination from the SDP program, which is what led to his reassignment -- gave rise to a constitutionally protected liberty interest. This is so for two reasons.
First, the majority is wrong that “removal from the SDP during Phases I-III does not result in any significant change in an inmate‘s conditions of confinement.” Opinion at 49. Participation in any phase of the SDP has significant implications for a prisoner‘s current living conditions, rendering those conditions materially freer than those of most prisoners in maximum custody. See
As Arizona‘s regulations make clear, participation in any stage of the SDP -- and, at the very least, participation in Phases II and III, as well as Phases IV and V -- entails significantly more freedom from restraint and social exposure than ordinary placement in maximum custody. So, to the extent that Johnson‘s removal from Phase IV led to “a material change in [Johnson‘s] living conditions” implicating a liberty interest, Opinion at 53, so too would a prisoner‘s removal from any prior phase -- and, at the very least, from Phases II and III. The majority argues that eliminating the freedoms enjoyed by prisoners in Phases I, II, and III of the SDP would not “rise[] to the level of an ‘atypical or significant hardship.‘” Opinion at 51 (quoting Sandin, 515 U.S. at 484). But, as described, above, many of the freedoms allowed in at least Phases II and III quite literally involve giving prisoners limited “freedom from restraint,” which is the quintessential sort of liberty interest as to which due process rights attach. Sandin, 515 U.S. at 484.
Second, the majority is wrong to conclude that removal from the SDP simply results in merely the loss of “one means by which Johnson can prove that he is prepared to return to the general prison population.” Opinion at 48. As described above, except for the single alternative of debriefing (the material shortcomings of which have already been outlined), removal from the SDP effectively guarantees that a prisoner will be confined in maximum custody for at least 24 more months or, in the case of a second removal from SDP, indefinitely. Whether or not the participating prisoner‘s transfer from maximum to close custody at the beginning of Phase IV has already occurred simply sidesteps the central issue, which is that at any phase of SDP, an SDP participant remains eligible to ultimately “step down” from maximum custody and ultimately transition to the general population, whereas once that participant is removed from SDP, he will face at least two more years of segregation in the case of a first removal and indefinite segregation in the case of a second removal unless he debriefs.5
from SDP without providing due process.
For the foregoing reasons, I respectfully dissent from the majority‘s decision affirming the district court‘s dismissal of Count III of Johnson‘s complaint. Put simply, the very real problem of maintaining prison security in the face of gang activity is not a blank check allowing prison officials to overrule the Constitution and keep a prisoner once identified as a gang member in solitary confinement forever without any genuine possibility of release. And while I concur in the majority‘s judgment reversing the district court‘s grant of summary judgment against Johnson as to Count I, in my view Johnson‘s removal from the SDP itself, and not just his collateral housing reassignment, implicated a liberty interest. Finally, I concur fully in the majority‘s holding that the district court erred in entering summary judgment against Johnson as to his First Amendment retaliation claim.