RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 25a0153p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

┐ CARVIN L. THOMAS and TERRELL LAWRENCE, on │ behalf of themselves and others similarly situated, │ Plaintiffs-Appellants, │ > No. 24-5637 │ v. │ │ RICHARD MONTGOMERY, as Chairman of the │ Tennessee Board of Parole; ZANE DUNCAN; GARY M. │ FAULCON; TIM GOBBLE; MAE BEAVERS; ROBERTA │ NEVIL KUSTOFF; BARRETT RICH, as Members of the │ Tennessee Board of Parole, │ Defendants-Appellees. │ ┘

Appeal from the United States District Court for the Middle District of Tennessee at Nashville. No. 3:23-cv-01204—William Lynn Campbell Jr., District Judge.

Argued: March 18, 2025

Decided and Filed: June 9, 2025

Before: COLE, STRANCH, and READLER, Circuit Judges. _________________

COUNSEL

ARGUED: Drew Justice, JUSTICE LAW OFFICE, Murfreesboro, Tennessee, for Appellants. Joshua Daniel Minchin, OFFICE OF THE TENNESSEE ATTORNEY GENERAL, Nashville, Tennessee, for Appellees. ON BRIEF: Drew Justice, JUSTICE LAW OFFICE, Murfreesboro, Tennessee, for Appellants. Joshua Daniel Minchin, OFFICE OF THE TENNESSEE ATTORNEY GENERAL, Nashville, Tennessee, for Appellees. No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 2

_________________

OPINION _________________

COLE, Circuit Judge. In their class-action complaint against the members of Tennessee’s Board of Parole, plaintiffs allege that Tennessee’s use of a computer test to determine parole eligibility violated their constitutional right to due process. The district court found that plaintiffs failed to state a plausible claim for relief because Tennessee’s parole statutes do not confer a protected liberty interest in parole. We affirm.

I.

Because this appeal arises from an order granting a Rule 12(b)(6) motion to dismiss, “we recite the facts as they are alleged in the complaint.” Savel v. MetroHealth Sys., 96 F.4th 932, 937 (6th Cir. 2024).

A.

This case concerns how Tennessee grants parole to eligible inmates. Tennessee’s parole scheme is governed by a seven-member Board of Parole. Tenn. Code Ann. § 40-28-103(a). The Board is responsible for determining inmates’ fitness for parole. Id. § 40-28-118(c). Inmates become parole-eligible when they have served either half of their prison sentence (for determinate sentences) or their minimum sentence (for indeterminate sentences involving a range). Id. § 40-28-115(a), (b)(1).

When an inmate becomes parole-eligible, the Board begins to review the inmate’s fitness for parole. The Tennessee Department of Correction notifies the Board of eligible inmates, and the Board compiles and distributes a list of inmates who shall have a hearing. Tenn. Comp. R. & Regs. 1100-01-01-.08(1). The Board appoints hearing officers who conduct hearings, take testimony, and propose findings and recommendations to the Board regarding whether an inmate should be paroled. Tenn. Code Ann. § 40-28-105(d)(2). Thereafter, the Board votes to adopt, modify, or reject the hearing officer’s recommendations. Id.; Tenn. Comp. R. & Regs. 1100-01- 01-.08(2). No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 3

At issue here is how the Board determines whether an eligible inmate receives parole. The Board uses a computer test, “STRONG-R,” to assess an inmate’s fitness for parole. STRONG-R reviews an inmate’s information and assigns a score of “Low,” “Moderate,” or “High” risk. Tennessee law requires that each inmate annually receives a STRONG-R test. The test “decides whether the inmate’s housing situation, family life, or mental health is problematic” and recommends prison programs to assist with any issues it identifies. (Am. Compl., R. 12, PageID 85.)

Along with prison programming, the Board uses STRONG-R as a test for determining whether an inmate receives parole. The Board interprets an assessment of “High” or “Moderate” risk as grounds for denying parole. And when the Board denies parole and advises the inmate how to improve his chances for future parole, the Board’s denial notifications state only, “Complete Programming As Recommended by [STRONG-R] Assessment.” (Id. at PageID 86– 87.)

The Board’s reliance on STRONG-R, however, has produced inaccurate results. Correctional employees are generally “not adequately trained to administer the [STRONG-R] test properly.” (Id. at PageID 88.) For example, the test asks about the inmate’s life and history, and correctional employees should relay those questions to the inmate, then enter the responses into the test on the inmate’s behalf. But an inadequately trained employee may fail to gather or input the necessary information, yielding a STRONG-R score inconsistent with the inmate’s true record and personal history.

The Board keeps the STRONG-R results “secret,” sharing them with the inmate at only their parole hearing. The inmate cannot subpoena the detailed results from their own tests, nor can they subpoena the prison officer who inputted their information into STRONG-R. Plaintiffs allege that this policy “works to ensure that no inmate can meaningfully challenge or address the [STRONG-R] results.” (Id. at PageID 96.) The Tennessee Department of Correction has a similar policy of refusing to release STRONG-R results, but some inmates have nonetheless successfully managed to acquire some of their STRONG-R information from prison officials. No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 4

B.

Plaintiffs Carvin Thomas and Terrell Lawrence are two parole-eligible inmates who were denied parole due to their STRONG-R scores. Their experiences with the test demonstrate STRONG-R’s shortcomings. For several years, Thomas received a “Low” or “Moderate” risk score from STRONG-R assessments. When he was transferred to a different prison in 2022, however, his assessment scores began to change despite no new negative behavior or activity. His new STRONG-R results listed him as “High” risk, with a likelihood of becoming violent. Thomas alleges that his change in risk score was due to counselors “who had inferior training about how to administer the assessment.” (Id. at PageID 89.)

Thomas’s specific results revealed several errors. They falsely reported that Thomas had been confined to a mental asylum, showed signs of mental illness, and had committed crimes in the past because of non-compliance with mental health medications. But Thomas was never confined to a mental asylum, diagnosed with a mental illness, or prescribed psychological medication. The results also stated that Thomas committed his decades-old crime “for thrill or pleasure,” but Thomas states he committed those crimes to obtain money. (Id. at PageID 90.) And the results stated that Thomas had become addicted to drugs in the preceding six months, despite Thomas’s long-term sobriety and his incarceration during that time.

Thomas attended a parole hearing in January 2023. He requested a continuance and a re- test of the STRONG-R assessment. The Board postponed his hearing, and prison officials audited Thomas’s STRONG-R test result. The Board did not, however, order any retesting. During this time, Thomas alleges that the prison counselor who performed his assessment told him to cease the request for an audit or retesting, “or that things might get ‘worse.’” (Id. at PageID 91.)

Thomas never received the results of the audit, but the hearing officer announced at his parole hearing in May 2023 that prison officials had concluded that his “High” STRONG-R risk result was accurate. Thomas challenged the result at the hearing. One of his supporters “also sent a letter to the Board[] listing about a dozen concrete ways that Thomas’s STRONG-R result was factually wrong.” (Id. at PageID 92.) But, as is common practice, no witness testified to No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 5

verify the accuracy of the STRONG-R results. The hearing officer announced a denial of parole based on the STRONG-R assessment, and the Board agreed. The Board advised Thomas that, for a higher likelihood of future parole, he should follow the STRONG-R recommendations.

Similarly, Lawrence had historically scored “Low” on STRONG-R, but despite no new issues or adverse developments, his score in the spring of 2022 rose to “High.” Lawrence had completed many rehabilitative programs, including the Cognitive Behavioral Intervention Program (CBIP). The Board denied Lawrence parole, citing his STRONG-R results and the seriousness of his offense. His denial notice recommended that Lawrence could improve his parole chances by completing CBIP, which he had already completed.

While appealing his parole decision, Lawrence was retested under STRONG-R and received a “Low” score, despite no new information. Lawrence notified the Board of his new score, and then a prison counselor inputted new data into Lawrence’s test and produced a third STRONG-R score, showing Lawrence’s risk as “Moderate.”

C.

Plaintiffs filed a class action amended complaint on behalf of inmates who sought parole in Tennessee. They sued the seven members of the Board, all in their official capacity. Plaintiffs brought one claim for relief under 42 U.S.C. § 1983, alleging denial of procedural due process in violation of the Fourteenth Amendment. Defendants moved to dismiss plaintiffs’ amended complaint, arguing that plaintiffs failed to state a claim under Federal Rule of Civil Procedure 12(b)(6).

The district court concluded that plaintiffs failed to state a claim for relief. First, the district court concluded that inmates in Tennessee “do not have a constitutionally protected liberty interest in parole.” (Mem. Opinion, R. 26, PageID 235.) And it concluded that any liberty interest in parole would not save plaintiffs’ claims, as they received the required due process in their parole proceedings because the Board gave plaintiffs a chance to be heard and informed them why it denied parole.

Plaintiffs timely appealed. No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 6

II.

We review de novo a district court’s dismissal of complaints under Federal Rule of Civil Procedure 12(b)(6). Mattera v. Baffert, 100 F.4th 734, 739 (6th Cir. 2024). We “construe the complaint in the light most favorable to the plaintiff, accept its allegations as true, and draw all reasonable inferences in favor of the plaintiff.” Watson Carpet & Floor Covering, Inc. v. Mohawk Indus., Inc., 648 F.3d 452, 456 (6th Cir. 2011) (internal citation and quotation marks omitted). To survive a Rule 12(b)(6) motion to dismiss, a plaintiff must allege sufficient facts that, taken as true, state a plausible claim for relief. Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007); Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). A complaint states a plausible claim for relief where its alleged facts “allow[] the court to draw the reasonable inference that the defendant is liable for the misconduct alleged.” Iqbal, 556 U.S. at 678.

III.

The Fourteenth Amendment prohibits deprivation of “life, liberty, or property, without due process of law[.]” U.S. Const. amend. XIV, § 1. A plausible Fourteenth Amendment procedural due process claim requires plaintiffs to allege “(1) that they have a life, liberty, or property interest protected by the Due Process Clause of the Fourteenth Amendment [], (2) that they were deprived of this protected interest within the meaning of the Due Process Clause, and (3) that the state did not afford them adequate procedural rights prior to depriving them of their protected interest.” Cooperrider v. Woods, 127 F.4th 1019, 1042 (6th Cir. 2025) (quoting Hahn v. Star Bank, 190 F.3d 708, 716 (6th Cir. 1999) (internal quotation marks omitted)).

“A liberty interest may arise from the Constitution itself, by reason of guarantees implicit in the word ‘liberty,’ . . . or it may arise from an expectation or interest created by state laws or policies[.]” Wilkinson v. Austin, 545 U.S. 209, 221 (2005) (cleaned up). Plaintiffs asserting a liberty interest must demonstrate “more than a unilateral expectation of it.” Greenholtz v. Inmates of Neb. Penal & Corr. Complex, 442 U.S. 1, 7 (1979) (quoting Bd. of Regents of State Colls. v. Roth, 408 U.S. 564, 577 (1972)). Rather, “[they] must, instead, have a legitimate claim of entitlement to it.” Id. And “[t]o determine whether due process requirements apply in the No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 7

first place, we must look not to the ‘weight’ but to the nature of the interest at stake.” Id. (quoting Roth, 408 U.S. at 570–71).

We first outline the Supreme Court’s precedent on liberty interests in the context of inmates seeking parole. Then, we consider how Tennessee’s parole scheme fits within that precedent. In doing so, we conclude that Tennessee’s scheme does not create a constitutionally recognized entitlement to parole.

A.

The Supreme Court first addressed the liberty interests of inmates seeking parole in Greenholtz. There, a group of inmates sued members of the Nebraska Board of Parole, alleging that the Board denied them procedural due process in the inmates’ pursuit of parole. Id. at 3–4. The Court granted review to determine whether the Fourteenth Amendment’s Due Process Clause applied to discretionary parole-release determinations in Nebraska. Id. at 3. The plaintiffs advanced two primary arguments in support of a liberty interest in parole. Id. at 8. First, they argued that the possibility of parole creates a liberty interest. Id. at 9. Second, they argued that the text of Nebraska’s parole statute created a legitimate expectation of parole. Id. at 11.

The Court rejected the plaintiffs’ first argument but accepted its second. Id. at 11–12. It concluded that “[t]here is no constitutional or inherent right of a convicted person to be conditionally released before the expiration of a valid sentence.” Id. at 7. But, while it held that the mere possibility of parole could not create a liberty interest, it also concluded that a state can create a liberty interest if it conveys to an inmate a legitimate expectation of parole. Id. at 7, 12. So while the mere existence of a parole system itself does not establish a liberty interest in parole, a state statute can give rise to a liberty interest where it creates a legitimate expectation of parole. Id. The key issue is whether the state statute contains mandatory language transforming a possibility into an expectation. See id.

A decade later, the Court applied Greenholtz to Montana’s parole statutes. Bd. of Pardons v. Allen, 482 U.S. 369, 377 (1987). In Allen, the Court found that Montana’s statute, like the Nebraska statute in Greenholtz, contained mandatory language that curtailed the state No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 8

board of parole’s discretion. Id. at 377–78. The statute thus gave inmates a legitimate expectation that they would receive parole. Id. at 378 n.10. In so holding, the Court reiterated that “[w]hen statutes or regulatory provisions are phrased in mandatory terms or explicitly create a presumption of release, courts find a liberty interest.” Id. (collecting cases).

Greenholtz and Allen require us to examine Tennessee’s parole system and determine whether Tennessee sufficiently constrains the Board’s discretion such that inmates have a legitimate expectation that they will be paroled. See Crump v. Lafler, 657 F.3d 393, 399 (6th Cir. 2011) (“Allen and Greenholtz teach that the salient factor is whether the statute contains mandatory language that creates a presumption of release when the designated findings are made.”).

B.

This is not the first time our circuit has evaluated whether Tennessee confers a liberty interest in parole. As Tennessee’s parole statutes—and accompanying Board rules—have changed over time, so too have our conclusions.

In Mayes v. Trammell, 751 F.2d 175 (6th Cir. 1984), this court concluded that Tennessee conferred a liberty interest in parole. We determined that “[t]he presence of the decisive language in the Board’s rules” established a liberty interest. Id. at 179. At the time, the Rules of Tennessee Board of Parole stated, “The Board operates under the presumption that each resident who is eligible for parole is a worthy candidate and thus the Board presumes that he will be released on parole when he is first eligible.” Id. at 178 (quoting Tenn. Bd. of Parole Rule 1100– 1–1–.06(1)). This language “appear[ed] to create the kind of presumption that the Supreme Court found significant in Greenholtz.” Id. Thus, Tennessee inmates had a legitimate expectation of parole. Id. at 179.

We concluded that the statutory text separate from the Board of Parole Rules, however, would not create a liberty interest. Id. at 177. Unlike the Nebraska statute in Greenholtz, the Tennessee statutes did not use “shall” or “unless,” and the statutes provided for parole as a privilege, rather than a right. Id. at 177–78. Thus, without the Board of Parole Rules imposing a No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 9

presumption of release, this court would have concluded that the inmates had no liberty interest in parole. See id.

A few years later, statutory amendments to Tennessee’s parole system dictated a different outcome. See Wright v. Trammell, 810 F.2d 589, 590 (6th Cir. 1987) (per curiam). The state amended the Rules of the Tennessee Board of Parole, removing Rule 1100-1-1-.06’s reference to a presumption of parole. Id. at 590–91. This court determined that, by removing those words, Tennessee no longer gave inmates an expectation of parole. Id. And without an expectation of parole, Tennessee conferred no liberty interest. Id.

Tennessee most recently amended its parole statutes when it passed the Reentry Success Act of 2021. Tennessee’s parole system is governed by a series of statutory provisions. The primary parole statute states:

Parole being a privilege and not a right, no prisoner shall be released on parole merely as a reward for good conduct or efficient performance of duties assigned in prison, but only if the board is of the opinion that there is reasonable probability that the prisoner, if released, will live and remain at liberty without violating the law, and that the prisoner’s release is not incompatible with the welfare of society. If the board so determines, the prisoner may be paroled[.]

Tenn. Code Ann. § 40-28-117(a)(1). Additionally, “no inmate convicted shall be granted parole if the board finds that”:

(1) There is a substantial risk that the incarcerated individual will not conform to the conditions of the release program; (2)(A) The release from custody at the time would depreciate the seriousness of the crime of which the incarcerated individual stands convicted or promote disrespect for the law[]; (3) The release from custody at the time would have a substantially adverse effect on institutional discipline; or (4) The incarcerated individual’s continued correctional treatment, medical care or vocational or other training in the institution will substantially enhance the incarcerated individual’s capacity to lead a law-abiding life when given release status at a later time. No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 10

Id. § 40-35-503(b). This system shall be “liberally construed” to treat inmates in accordance with their “individual characteristics, circumstances, needs and potentialities[.]” Id. § 40-28- 101(a).

Additionally, the Board cannot deny parole to certain nonviolent inmates on the sole basis that they committed a serious crime. Id. § 40-35-503(b)(2)(A). And inmates with lower- class nonviolent felonies receive an explicit presumption of parole. Id. § 40-35-503(h) (“[T]here is a presumption that an inmate convicted of a Class E or Class D nonviolent felony offense is to be released on parole upon the inmate reaching the inmate’s release eligibility date unless good cause is shown as to why the inmate should not be released.”). When the Board denies parole, it must state the reason for doing so in writing and advise the inmate how they can improve their chances for parole. Id. § 40-35-503(j).

Today’s Tennessee parole scheme more closely mirrors Wright than Mayes. After the “privilege” language, the statutes state that if the Board determines the inmate “will live and remain at liberty without violating the law, and that the prisoner’s release is not incompatible with the welfare of society[,]” he “may be paroled[.]” Id. § 40-28-117(a)(1) (emphasis added). And today’s Board of Parole Rules lack the critical presumption language in Mayes. See Tenn. Comp. R. & Regs. 1100-01-01-.07(4); Mayes, 751 F.2d at 177–78. Thus, the statutory scheme permits parole but does not mandate it.1

Considering the statutes governing parole, we conclude that inmates in Tennessee do not have a constitutionally recognized expectation of receiving parole because the statutory scheme does not sufficiently constrain the Board’s discretion to deny parole. Consequently, Tennessee does not confer a protected liberty interest on inmates.

1Contrary to defendants’ argument, Tennessee’s “privilege and not a right” language alone would not preclude us from finding a liberty interest. See Mayes, 751 F.2d at 179. Rather, we must consider the statutory scheme as a whole to determine whether the Board’s discretion is sufficiently constrained. See Crump, 657 F.3d at 399 (“Allen and Greenholtz teach that the salient factor is whether the statute contains mandatory language that creates a presumption of release when the designated findings are made. . . . The mandatory language may be found in a statute, a regulation, or even ‘policy statements . . . or other official promulgations’ by parole or prison officials.” (citations omitted)). That the court in Mayes found a liberty interest despite the “privilege and not a right” language shows that such language is not dispositive. No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 11

Plaintiffs offer two arguments to the contrary. First, they argue that the parole statutes use the mandatory word “shall” and use the four key considerations in Greenholtz. See Tenn. Code Ann. § 40-35-503(b); Greenholtz, 442 U.S. at 11 (citing Neb. Rev. Stat. § 83-1,114(1) (1976)). Although the statutes use words like “shall,” the use of “shall” serves the opposite effect as in Greenholtz. See 442 U.S. at 11 (“[T]he Board of Parole . . . shall order [an inmate’s] release unless it is of the opinion that his release should be deferred because [conditions are met.]”). Here, the four Greenholtz considerations are grounds for mandatory denial of parole, not the mandatory grant of parole. See Tenn. Code Ann. § 40-35-503(b) (“[N]o inmate . . . shall be granted parole if [one of the conditions is present.]”); § 40-28-117(a)(1) (“[N]o prisoner shall be released . . . but only if [conditions are met]. If the board so determines, the prisoner may be paroled[.]” (emphasis added)); Tenn. Comp. R. & Regs. 1100-01-01-.07(4) (“[T]he Board shall deny the inmate’s release on parole if it determines that [one of the conditions is present.]”). Dictating when the Board shall deny parole is different from dictating when it shall grant it.

Second, plaintiffs argue that the Board’s customs in granting and denying parole creates a liberty interest. Plaintiffs cite no caselaw where a parole board’s customs—rather than policies and state statutes—created an expectation of parole sufficient to establish a liberty interest.

Though we have typically conducted a statutory analysis, see Crump, 657 F.3d at 399, the Supreme Court has suggested that a parole board’s practices could confer a liberty interest. See Greenholtz, 442 U.S. at 4 (“The procedures used by the Board to determine whether to grant or deny discretionary parole arise partly from statutory provisions and partly from the Board’s practices.”); see also Allen, 482 U.S. at 373 n.3 (“There is far more to liberty than interests conferred by language in state statutes.”). A future case may present allegations that demonstrate practices conferring a liberty interest in parole, even where the state’s statute does not contain the presumption language in Greenholtz or Allen. But plaintiffs do not develop this argument or include plausible allegations of customs that create a legitimate expectation of parole. Accordingly, plaintiffs’ customs-based arguments do not demonstrate a protected liberty interest.

Although plaintiffs do not demonstrate a protected liberty interest, they identify serious issues with STRONG-R. Recognizing that STRONG-R is a new technology that continues to be refined, we nonetheless remind the Board, in its oversight role regarding the state parole system, No. 24-5637 Thomas, et al. v. Montgomery, et al. Page 12

of the Tennessee legislature’s instruction that the Board consider an inmate’s “individual characteristics, circumstances, needs and potentialities” in making parole decisions. See Tenn. Code Ann. § 40-28-101(a) (2025); Zachary Hamilton, et al., Never Going to Let You Down: Preventing Predictive Shrinkage via the STRONG-R Assessment Method, Just. Q. 3 (Aug. 7, 2024). Relatedly, the lack of opportunity to address the inaccuracies with STRONG-R could demonstrate a lack of due process. Supreme Court precedent, however, precludes us from reaching those due process concerns. Tennessee’s statutes do not confer a liberty interest, regardless of plaintiffs’ legitimate belief that had their STRONG-R results been accurate, they would have received parole. See Greenholtz, 442 U.S. at 30 (Marshall, J., dissenting) (“[I]n light of the role that parole has assumed in the sentencing process, I believe the Court misapplies its own test . . . by refusing to acknowledge that inmates have a legitimate expectation of release whenever the government establishes a parole system.”).

Finally, defendants raise a litany of issues on appeal that they did not raise below, likely forfeiting those issues for this appeal of the district court’s Rule 12(b)(6) order. See Bannister v. Knox Cnty. Bd. of Educ., 49 F.4th 1000, 1011 (6th Cir. 2022) (“A forfeiture occurs when a party fails to timely assert a claim, even if the party does so unintentionally[.]”). In any case, because we conclude that the district court correctly dismissed plaintiffs’ claim for failure to state a claim, we need not address defendants’ remaining arguments.

IV.

For these reasons, we affirm the district court’s determination that plaintiffs fail to plead a plausible liberty interest in parole.