372 F.3d 346 | 6th Cir. | 2004
Lead Opinion
MOORE, J., delivered the opinion of the court, in which FORESTER, D.J., joined. ROGERS, J. (pp. 360-365), delivered a separate opinion concurring in part and dissenting in part.
OPINION
Plaintiffs-Appellees Charles E. Austin et al.,
On appeal, the ODRC Officials raise two claims of error: that the district court applied the wrong legal standards in finding a constitutional violation, and that even if a violation was correctly found, the remedial orders entered violate 18 U.S.C. § 3626, requiring particularized fact-finding by federal district courts interfering with state prison practices. Because we conclude that the district court did not err in determining that a liberty interest existed in the prison placement at issue and in modifying the procedures that govern that placement, we AFFIRM that portion of the district court’s judgments; because the district court erred, however, in modifying substantive Ohio prison regulations, we REVERSE AND REMAND that portion of the judgments.
I. BACKGROUND
In May 1998, the OSP, Ohio’s supermax prison facility, opened for business. Su-permax facilities, in operation in most of the states and in the federal prison system, represent an attempt to concentrate the “worst of the worst” in one facility, thereby making the rest of the general prison population more safe and easier to control. Ohio’s supermax, which is designated a high-maximum-security prison, was built in response to an April 1993 riot at the Southern Ohio Correctional Facility (“SOCF”). Prior to the construction of the 504-bed OSP, Ohio’s most secure prison was the SOCF, a maximum-security prison. Within the SOCF is an even more secure eellblock, the J-l cellblock, which houses twenty cells. Before the OSP opened, Ohio did not fill the J-l cells; instead, it did not have enough maximum-security cells to house inmates at that security designation. From these facts, the district court concluded that the surplus of high-maximum-security cells led to a “because we have built it, they will come” mentality, with the surplus of maximum-security inmates leading to placement of inmates at OSP who did not meet the high-maximum-security requirements, contrary to both corrections policy and constitutional norms. See Austin v. Wilkinson (Austin I), 189 F.Supp.2d 719, 724 (ND.Ohio 2002).
When the OSP first received inmates in May 1998, it did so in a concededly problematic and confused manner. Appellants’ Br. at 11. On August 31, 1998, the department “attempted to establish some predictability to placement at the OSP by
Under this policy, though, problems continued. Some of the more troubling instances of this haphazard system occurred when the Bureau would, without stating its reasons, overrule the recommendation of both the classification committee and the warden and either place or maintain the placement of an inmate at OSP; when inmates who would otherwise be recommended for parole were ineligible because of a suspect OSP placement; when multiple jumps in security levels happened as a result of a single incident; when decisions were made with little factual support; and when decisions were based solely on the use or smuggling in of small amounts of drugs. Id. at 734-36.
The goal of the OSP, to separate the most dangerous prisoners from the rest of the prison population, is achieved primarily through solitary confinement, of a type noticeably different than segregation at other Ohio prisons. Inmates at OSP spend twenty-three hours a day in their single cells, measuring approximately 89.7 square feet. Id. at 724. These cells are further isolated from the outside world by the installation of metal strips on the bottom and sides of the cell doors that prevent inmates from communicating with one another. Id. During the one hour per day that inmates can leave their cells, they have access to two indoor recreation rooms; most inmates have recreation alone, although a limited number may have recreation with one other prisoner. Id. Inmates having visitors are required to be strip-searched when they leave and reenter the cellblock, even though they are isolated from their visitors by solid windows. Id. at 725. Additional factual findings by the district court differentiate life at OSP from segregation conditions at other Ohio prisons, including extra limitations on personal property rights, access to telephones and counsel, outside recreation, and communication with other persons. Id.
On January 1, 2001, the Inmates filed their complaint, stating both procedural due process claims regarding placement at OSP, the claims at issue on appeal, and Eighth Amendment claims regarding inadequate medical and psychiatric care, inadequate outdoor recreation facilities, and harsh restraints used at OSP. The Eighth Amendment claims were settled below. See Austin v. Wilkinson, No. 4:01-CV-71 (N.D.Ohio Apr. 5, 2002) (order approving settlement agreement). A Fed.R.Civ.P. 23(b)(2) class was certified, and the case was scheduled for trial in January 2002. On the eve of trial, the ODRC released a new version of 111-07 (“new 111-07”), and
In its opinion, issued on February 25, 2002, the district court found the process specified in new 111-07 lacking. See Austin I, 189 F.Supp.2d at 750-52. The district court identified the following deficiencies: that inmates are not given notice of all the evidence that may be relied upon in their classification hearings; that inmates are not allowed to call witnesses; that the placement criteria give insufficient notice of the amount of drugs in possession that would trigger level five placement; that the placement criteria are unnecessarily vague with regard to the gang activity that would trigger a placement; that the final decisionmaker, the Bureau of Classification, was not required to describe the facts found and reasoning used in making its placement and reclassification decisions; that the inmate is not given adequate notice of the information to be considered at his reclassification hearing; and that adequate notice of the conduct necessary for the inmate to leave the OSP is not given. The district court then ordered the parties to file proposed injunctive orders; the ODRC Officials’ proposed injunctive order was one and a half pages in length. On March 26, 2002, the district court issued an injunction directing the ODRC Officials to correct each of the deficiencies it had found and issued an accompanying judgment terminating the action under Rule 58. See Austin v. Wilkinson (Austin II), 204 F.Supp.2d 1024 (N.D.Ohio 2002). The ODRC Officials filed a timely notice of appeal on April 17, which was docketed as Sixth Circuit Appeal Number 02-3429. On April 24, they filed with the district court a revised version of 111-07 (“revised 111-07”). On May 15, the district court ordered that policy adopted with slight changes. On June 6, 2002, the ODRC Officials filed a Rule 60(b) motion,
II. ANALYSIS
A. Jurisdiction
The underlying civil rights action was brought under 42 U.S.C. § 1983. The district court had original jurisdiction under 28 U.S.C. § 1331. This court has jurisdiction under 28 U.S.C. § 1291.
B. Standard of Review
The ODRC Officials appeal from the grant of a permanent injunction and the denial of a motion for relief from the judgment, both of which are reviewed for abuse of discretion. See Herman Miller, Inc. v. Palazzetti Imp. & Exp., Inc., 270 F.3d 298, 317 (6th Cir.2001); Mallory v. Eyrich, 922 F.2d 1273, 1279 (6th Cir.1991). In determining whether the district court abused its discretion, we give great deference to the district court, reviewing its legal determinations de novo, but only disturbing its factual findings if they are clearly erroneous. Elec. Workers Pension Trust Fund of Local Union # 58 v. Gary’s
C. Due Process Rights
1. Liberty Interest under Sandin v. Conner
Inmates challenge the procedures for classification at level five under the Due Process Clause of the Fourteenth Amendment, claiming that classification at that level and concomitant placement at OSP implicates a state-created liberty interest, and that the procedures in place before trial were inadequate to protect this interest. Therefore, our threshold inquiry is whether a state-created liberty interest exists with regard to placement in Ohio’s supermax prison. This inquiry is controlled by Sandin v. Conner, 515 U.S; 472, 115 S.Ct. 2293, 132 L.Ed.2d 418 (1995), which mandates that a state creates a liberty interest in avoiding certain prison conditions only where those conditions are an “atypical and significant hardship on the inmate in relation to the ordinary incidents of prison life.” Id. at 484, 115 S.Ct. 2293. Prior to Sandin, a state created a liberty interest through using “ ‘language of an unmistakably mandatory character’ such that the incursion on liberty would not occur ‘absent specified substantive predicates.’ ” Id. at 480, 115 S.Ct. 2293 (quoting Hewitt v. Helms, 459 U.S. 460, 471-72, 103 S.Ct. 864, 74 L.Ed.2d 675 (1983)). Sandin shifted the focus from parsing the language of state statutes and regulations to examining the severity of the conditions to which an inmate would be subject.
The district court thus properly made factual findings as to the conditions in OSP compared to the conditions in other Ohio prisons, specifically in the segregáted units of maximum-security prisons, the most severe non-OSP conditions in the Ohio system. The court found that the extreme isolation visited upon the inmates at OSP, the lack of any outdoor recreation, the limitations upon personal property rights and access to telephones and counsel, and, finally, the ineligibility of OSP inmates for parole, all combined to create a significant and atypical hardship. The ODRC Officials’ sole challenge on appeal to these careful findings is that the district court erred by comparing conditions at OSP to conditions at other Ohio prisons. They argue instead that the proper baseline in determining atypicality is the conditions at other supermax facilities around the country. Other circuits that have decided the question have split over whether the proper control group is the general prison population or inmates in typical segregation conditions. Compare Beverati v. Smith, 120 F.3d 500, 504 (4th Cir.1997) (finding “the conditions [at issue] were more burdensome than those imposed on the general prison population” although not sufficiently atypical), and Keenan v. Hall, 83 F.3d 1083, 1089 (9th Cir.1996) (“a major difference between the conditions for the .general prison population and the segregated population triggers a right to a hearing”), with Griffin v. Vaughn, 112 F.3d 703, 708 (3d Cir.1997) (administrative custody is not “extraordinary” and “stays of many months are not uncommon”), and Brooks v. DiFasi, 112 F.3d 46, 49 (2d Cir.1997) (explicit factual comparison between administrative segregation and disciplinary segregation is necessary). See also Hatch v. District of Columbia, 184 F.3d 846, 847 (D.C.Cir.1999) (appropriate comparison is “the most restrictive conditions ... routinely impose[d] on inmates serving similar sentences”); Wagner v. Hanks, 128 F.3d 1173, 1177 (7th Cir.1997) (appropriate comparison is to the conditions of nondisciplinary segregation in the
The ODRC Officials point only to the Supreme Court’s decision in Olim v. Wakinekona, 461 U.S. 238, 103 S.Ct. 1741, 75 L.Ed.2d 813 (1983), and dicta in Wagner, 128 F.3d at 1173, to support their argument. Olim involved a challenge to an interstate prison transfer; the Court held that no liberty interest was created by Hawaii state prison regulations, and that the Due Process Clause of its own force did not protect any liberty interest deprived by an interstate prison transfer. Olim, 461 U.S. at 245-49, 103 S.Ct. 1741. But Olim’s holding has limited applicability when dealing with a state-created liberty interest; that the Due Process Clause does not of its own force protect against interstate transfers has little to do with whether Ohio regulations create a liberty interest in remaining outside of a super-max prison. Sandin requires a situation-ally-based factual analysis; if it is typical that an Ohio prisoner experiences conditions similar to those of OSP, then state lines might truly be irrelevant. That, however, is not the case; of 44,000 prisoners in the Ohio system, only twenty to thirty have been transferred out of state — a number that itself might give rise to a typicality, if not hardship, if substantive state law limiting officials’ discretion in transfer existed — and not a single one has been shown to have been transferred to a supermax. See Hatch, 184 F.3d at 857 (“What matters, therefore, is not simply the possibility of transfer but also its likelihood”). Olim, to the contrary, relied heavily on the fact of interstate transfer as a common occurrence, suggesting that no reasonable expectation existed that any particular inmate would serve his sentence within his state of conviction. Olim, 461 U.S. at 245-47, 103 S.Ct. 1741. Even if the proper comparison in this case were nationwide rather than statewide, the appropriate question would be whether the OSP represented an atypical hardship as compared to, at most, the typical conditions in administrative or disciplinary segregation to which transferred Ohio prisoners are subject, not those in supermaxes.
Finally, Wagner is not to the contrary. First, as the Inmates point out, the language relied upon by the ODRC Officials in support of their claim is dicta. See Wagner, 128 F.3d at 1176. Second, and more important, the Wagner court itself in its analysis misses the important distinction between cases narrowly defining the contours of the protection of the Due Process Clause of its own force and those deciding when state laws create a liberty interest. The Supreme Court has repeatedly held that transfers in and of themselves do not implicate due process interests (although it has not to date dealt with transfer to a supermax prison), see Olim, 461 U.S. at 238, 103 S.Ct. 1741;
Ultimately, whether OSP is compared to the general prison population of Ohio, or instead to inmates in typical segregation conditions, which was the baseline used by the district court, OSP constitutes an atypical and significant hardship under Sandin, such that inmates enjoy a liberty interest in not being placed at OSP absent the state-mandated substantive predicates set out in new Policy 111-07. It is therefore unnecessary to determine which is the proper baseline for Sandin comparisons in order to decide this case, but we reject emphatically the ODRC Officials’ argument that the baseline should be out-of-state supermax prisons. Whatever the “ordinary incidents of prison life” may encompass, they must be decided with reference to the particular prison system at issue, and can only be truly “ordinary” when experienced by a significant proportion of the prison population.
2. The Process That Is Due
Once a liberty interest has been established under Sandin, we must turn to the question of what process is due to protect that interest. At issue on appeal is the propriety of a set of modifications made by the district court to the ODRC’s policy governing classification at level five. Most of these modifications concern the procedures used to classify inmates, but three modifications are substantive predicates to OSP placement and retention. We will analyze the procedural modifications in greater depth below under the familiar due process analysis of Mathews v. Eldridge, 424 U.S. 319, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976), but the substantive modifications can be dealt with more simply-
a. Modifications to Ohio’s Substantive Prison Regulations
Prior to the district court’s modifications, inmates could be placed at OSP for any contraband activity, no matter how minimal. As its first substantive modification, the district court directed that the policy be rewritten to specify a quantity of contraband activity, and for drug activity, the district court stipulated that the
The power of the federal courts to order modifications in state prison policies extends only as far as is necessary to protect federal rights. The Inmates do not argue and we do not decide whether placement at OSP implicates either the Eighth Amendment or the substantive portion of the Due Process Clause.
While the district court correctly identified adequate notice as a requirement of due process when making these changes, see Columbia Natural Res., Inc. v. Tatum, 58 F.3d 1101, 1104-05 (6th Cir. 1995), we conclude that each of the regulations provides sufficient notice and that the modifications made by the district court are in fact substantive modifications. For instance, new 111-07 states that any amount of drugs can trigger a reclassification hearing; altering the policy to require a specified amount does not improve upon that notice but instead limits the substantive discretion of the ODRC Officials. Similarly, the security group modification alters the substantive grounds for placement at OSP, rather than the process used in determining that placement. And the modification to the retention criteria also limits the substantive discretion of the Officials. While the due process requirement of notice applies in the prison context as well as outside of it, albeit slightly differently, see United States v. Chatman, 538 F.2d 567, 569 (4th Cir.1976), any deficiencies in the notice provided by these provisions are properly corrected through case by case “as applied” challenges to the regulations, rather than striking the regula
b. Procedural Modifications
That a liberty interest exists in avoiding classification at level five and concurrent placement in OSP is in many ways the easy half of the Sandin analysis. What is much less clear after Sandin is how to determine what process is due to protect that interest. Before Sandin, state-created liberty interests of prisoners were either protected by an adversary hearing on the record following Wolff v. McDonnell, 418 U.S. 539, 563-73, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974), or a more free-form hearing following Hewitt, 459 U.S. at 477, 103 S.Ct. 864. Wolff dealt with the process due in finding a disciplinary infraction punished by the rescission of good-time credits; Hewitt involved a challenged placement in administrative segregation, pending the outcome of an investigation into misconduct. Cases following the HeivittfWolff split have classified various factual situations depending upon the category into which the challenged process fell: disciplinary or administrative, historical or prospective, objective or subjective, Wolff or Hewitt. On appeal, the ODRC Officials assert that this mechanical dichotomy still has force after Sandin, that classification at level five and placement at OSP is a forward-looking, Hewitt-type procedure, and that our inquiry should end there, with a decision that only “Hewitt process” is due. We are convinced, however, that Sandin called into question not only the mechanistic way in which the circuit courts previously found liberty interests in prison regulations, but also the mechanistic fashion in which they applied the Hewitt/Wolff dichotomy.
requires consideration of three distinct factors: First, the private interest that*358 will be affected by the official action; second, the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards; and finally, the Government’s interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.
Id.
The district court made fifteen specific modifications to new 111-07, including the improper substantive modifications dealt with above. A first set deals with the classification hearing itself, and closely tracks Wolff. First, when classification proceedings are initiated, the notice already stipulated by new 111-07 shall include an exhaustive list of the reasons to be considered for placement and a summary of the evidence to be presented. Second, the inmate shall be allowed to present witnesses and documentary evidence at classification hearings, where “permitting him to do so will not be unduly hazardous or burdensome to institutional safety or correctional goals.” J.A. at 580 (citing Wolff, 418 U.S. at 566, 94 S.Ct. 2963). Finally, a record is to be made of the proceeding itself, and if the ODRC wishes to rely on confidential witnesses, it must indicate that reliance and disclose as much of the confidential testimony as possible.
A second group of modifications required by the district court centers on the administrative appellate procedure laid out in new 111-07. Because no comparable appellate procedure was at issue in Wolff, see 418 U.S. at 565, 94 S.Ct. 2963, these requirements do not track that case as closely. The district court found that the previous system of administrative review, in which each intervening appellate decisionmaker had plenary power to reverse the prior decisionmaker without any statement of reasons for the decision given to the inmate, had led to suspect inmate classifications. The specific changes ordered by the district court were: that the inmate is to receive the classification-committee’s recommendation and notice of his right to and method of appeal; that the warden is to engage in “independent review” of the committee’s recommendation, and if in doing so, relies upon a confidential witness statement not already made known to the prisoner, shall follow the procedure outlined above, including allowing the inmate to respond in writing; that the warden shall, if she approves the recommendation, send a copy of that recommendation to the inmate; and that the Bureau of Classification shall follow the same procedure in relying on new confidential witness statements, shall allow the inmate to submit documentary evidence, and will, if the inmate is recommended for level five placement, record a detailed and specific justification for the decision. Finally, the district court, in its May 15, 2002, order, required that none of the members of the original classification committee shall take part in the decision of the prisoner’s appeal to the warden or to the Bureau.
The first factor of the Mathews balancing test, the private interest at stake, is significant; placement at OSP is indefinite and reviewed only annually, unlike placement in disciplinary segregation in the Ohio prison system, which lasts only thirty days, or administrative segregation, which is reviewed every thirty days. Prisoners placed at OSP are deprived of all significant human contact and have other restrictions placed upon their movement and their personal privileges; they are also ineligible for parole during their stay at OSP. In this first factor, Sandin affects
Looking at each of the modifications ordered by the district court individually, we remain unconvinced that the district court abused its discretion in finding that each procedural modification it made was mandated by the weighty private interest at stake and the risk of error and was unmitigated by the governmental interests at stake. We examine first the requirement identified by the ODRC Officials both in their brief and at oral argument as the most burdensome: requiring officials to limit their placement decision to only those matters detailed in the notice given to the inmate. They argue that this requirement will constrict “substantive discretion” by disallowing reliance on “rumor, reputation, and even more imponderable factors.” Appellants’ Reply Br. at 8 (quoting Hewitt, 459 U.S. at 474, 103 S.Ct. 864). This argument is unavailing. Placement at OSP implicates a liberty interest because of the ODRC’s own regulations limiting the substantive discretion of prison officials; they can place inmates at OSP only in the presence of certain factual predicates, all of which are historical in nature. Having set out a detailed and restricted list of reasons why inmates can be put at OSP, the ODRC cannot turn around and argue that the district court’s order decreases their ability to rely on “rumor, reputation, and even more imponderable factors,” for those factors are illegitimate under their own placement scheme.
Having found that the additional procedural requirement identified by the ODRC Officials as the most burdensome passes muster under Mathews v. Eldridge, we conclude that those which pose a lesser burden on the ODRC are also appropriate. The ODRC Officials, both in their brief and at oral argument, did not in fact point to any other single procedural requirement as being particularly burdensome. We note, moreover, that many of the procedures ordered by the district court are an attempt to reconcile an elaborate administrative appeals scheme created by the ODRC Officials with the requirement that the inmate know the reason for any decision made about his fate; where a higher-up decisionmaker reverses the decision of the original factfinder, a brief description of the grounds for that reversal is constitutionally necessary.
D. Prison Litigation Reform Act
The ODRC Officials make a final argument that the district court failed to follow 18 U.S.C. § 3626, part of the Prison Litigation Reform Act (“PLRA”), governing prospective relief. Their complaints center around the substantive modifications made to new 111-07, which modifications are indeed invalid for the reasons discussed above. They also make a third, more general argument that the district court erred in failing to make findings that the remedial orders are necessary to correct “ ‘current and ongoing’ federal violations.” Appellants’ Br. at 60. But the “current and ongoing” language comes from § 3626(b)(3), governing the termination of relief, not from § 3626(a), governing requirements for initial relief. This argument therefore has no merit.
III. CONCLUSION
Because the Inmates have a liberty interest in avoiding placement at OSP, and because the procedural modifications ordered by the district court are necessary to protect that interest, we affirm those portions of the district court’s judgments that address procedural requirements and modifications. Because the district court was without power to reach the substantive prison regulations which were also modified, we reverse the district court’s judgments insofar as they rely on the following: Part II of the district court’s March 26, 2002 order, excepting the final paragraph;
. The additional named plaintiffs are Robert Baksi, Michael Benge, Alonzo L. Bonner, Au~ gust Cassano, David E. Clark, James DeJar-nette, Roy D. Donald, David Easley, Brian K. Eskridge, Keith Gardner, Roger Lee Hall, Frederick O. Harris, Jr., Daryl Heard, Edward O. Hodge, Orsino Iacovone, Kunta Ken
. Named additionally as defendants are Stephen J. Huffman, Bernard J. Ryznar, Todd E. Ishee, Bruce A. Martin, Deborah Nixon Hughes, Cheryl Jorgensen-Martinez, Manish B. Joshi, Patrick F. Biggs, Audrey Sandor Nietzel, and Matthew Meyer.
. The procedural information and quoted language are taken from new 111-07.
. When the district court made its findings and decision, only prisoners classified as level five were placed at OSP. As noted above, this is no longer the case, as the ODRC is now placing level four prisoners at OSP. The procedure for placement at level four is essentially the same as that set out in new 111-07 and found constitutionally inadequate by the district court. Therefore, inmates are being placed at OSP without the procedural protections ordered by the district court. Because the complained-of deprivation of liberty is not reclassification by itself, but placement at OSP which results from a level five classification, this seems a particularly disingenuous way to evade the district court's order in advance of appellate review.
. The ODRC Officials asked for relief under Rule 60(b)(6), (b)(1), and (b)(4).
. The ODRC Officials had moved on June 6, 2002, simultaneous with their Rule 60(b) motion, for additional time to file a notice of appeal from the district court's May 15 order. An extension was granted to July 15, 2002, pursuant to Federal Rule of Appellate Procedure 4(a)(5).
. The Fifth Circuit has also remanded at least one case to a district court with instructions to determine the proper baseline if Sandin is triggered. See Wilkerson v. Stalder, 329 F.3d 431, 436 (5th Cir.2003). But see Orellana v. Kyle, 65 F.3d 29, 31-32 (5th Cir.1995) (“it is difficult to see that any other deprivations in the prison context, short of those that clearly impinge on the duration of confinement, will henceforth qualify for constitutional liberty’ status”). Carson v. Johnson, 112 F.3d 818, 821 (5th Cir.1997), cited by the district court, does not in fact analyze state-created liberty interests under Sandin, but instead evaluates fundamental liberty rights under the three-strikes provision of the Prison Litigation Reform Act ("PLRA”).
. This distinction also partially explains why Moore v. Litscher, 52 Fed.Appx. 861 (7th Cir.2002), emá. Nash v. Litscher, 50 Fed.Appx. 317 (7th Cir.2002), cited by the concurrence for the proposition that avoidance of placement at a supermax is not in and of itself a liberty interest under Sandin, are inapposite. Moore cited to Olim in holding, "Prisoners do not have a federally protected liberty interest in being housed in a particular facility.” 52 Fed.Appx. at 861-62. With respect to a state liberty interest, the court held that “a failure to comply with state procedural rules does not violate the federal constitution.” Id, The Moore court failed to partake in any analysis under Sandin, and the case cannot stand for the proposition that placement at a supermax could not implicate a liberty interest created by the state. Nash noted simply that "Nash does not have a liberty interest in his prison placement,” and cited to an earlier case, Whitford v. Boglino, 63 F.3d 527, 532 (7th Cir.1995). 50 Fed.Appx. at 319-20. Whitford held that no federal interest existed in prison placement, and that Illinois state regulations, so far as the court "understood,” did not create such a liberty interest. Whitford also dealt with transfer to a maximum security prison, rather than a supermax. All three of these cases, Whitford, Moore, and Nash, were argued pro se in the Seventh Circuit. We do not think they can stand for the proposition that careful factual findings in the district court demonstrating severe hardship in prison conditions, atypical in a prison system, can never rise to the level of a protected liberty interest without a parole deprivation.
. While the Inmates challenged certain conditions at OSP under the Eighth Amendment, claims that were settled below, they do not argue that either the Eighth Amendment or the substantive protections of the Due Process Clause create liberty interests in freedom from transfer to OSP that require due process protection. We express no opinion as to the viability of such a claim.
. Like Judge Rogers, we are convinced that the ‘‘Hewitt/Wolff dichotomy” is not a viable form of analysis; we note later that the district court's procedural modifications track Wolff only because the ODRC officials had argued that the district court went beyond Wolff in ordering certain modifications, most of which were directed at the "appellate" process created by the ODRC.
. We do not think that we differ so much from Judge Rogers in describing this balance; we agree that the type of decision being made will affect the private interest, the government interest, and the value of certain procedural safeguards. We only emphasize that the type of decision is not, as the ODRC Officials would have us hold, the only factor necessary to determine what procedure is due. We believe, however, that in the face of the substantial factual findings of the district court as to haphazard ODRC placements, the procedural requirement of notice is particularly important, and give great weight to its value in increasing accuracy.
. We mean by this statement only to compare those liberty interests found to exist post-Sandin with those found to exist pre-Sandin. Because the Court has made clear that many of the liberty interests found by courts which would have in the past required certain pre-deprivation processes are no longer viable liberty interests, in looking over what courts have required of prison officials in the past, it is important to remember that many liberty interests which required less process in the past would require no process now. Any liberty interest which passes Sandin’s threshold comes with a higher presumption of process due than those which may have been found pre-Sandin.
. It is here we part ways with Judge Rogers — in determining what process is due, we believe reference to what the state substantively requires is the first step. In order to be placed at OSP, an inmate must fulfill one of those discrete, substantive historical predicates; the district court correctly required that ODRC Officials place an inmate on notice of what historical events will be used to demonstrate his fulfillment of one of those predicates. The state itself has limited its ability to place inmates at OSP; the combing through files predicted by Judge Rogers is unlikely in the face of the specific substantive predicates identified in the state scheme.
. That paragraph concerns the procedural modification of departmental notice to the inmate of the inmate's progress towards reclassification, and is a proper procedural modification.
. The ODRC Officials appeal from the district court’s July 12 denial of their Rule 60(b) motion; the only claims of error in that denial which they make on appeal concern the substantive modifications to 111-07 that we reverse in any case. There is therefore no need to evaluate the propriety of the district court's denial of the Rule 60(b) motion on its own merits.
Concurrence in Part
concurring and dissenting.
While I agree with much of the majority’s careful opinion, there are two areas where my analysis differs sufficiently to warrant a separate opinion, and one point upon which I respectfully dissent.
I agree that the Inmates have shown a protected liberty interest under Sandin v. Conner, 515 U.S. 472, 115 S.Ct. 2293, 132 L.Ed.2d 418 (1995), and that in applying Sandin we are not required to make an interstate, as opposed to intrastate, comparison. The record shows that inmates assigned to OSP not only are subjected to far more severe conditions of confinement, but they are also disqualified for parole while assigned to OSP. These two factors together permit the conclusion that a liberty interest is implicated under Sandin.
Because assignment to the OSP involves disqualification from parole, it is unnecessary for us to decide whether a prison classification that subjects an inmate to more restrictive conditions of confinement, without more, constitutes a deprivation of a liberty interest. Recent unpublished opinions of the Seventh Circuit hold that it does not, even where assignment to a su-permax prison was involved. See Moore v. Litscher, 52 Fed.Appx. 861 (7th Cir.2002) (prisoners do not have a federally protected liberty interest in being housed in a particular facility, and therefore, as a matter of federal constitutional law, prisoner was not entitled to any due process protection before he was moved to supermax facility); Nash v. Litscher, 50 Fed.Appx. 317 (7th Cir.2002) (same). If movement from one level of restrictive confinement to a significantly higher one triggers due process protections, then prison administration could be unduly burdened by the necessity of due process hearings. Courts would then have to struggle with just how much of a change in the severity of confinement triggers due process protection. Fortunately, we do not need to decide the issue. In this case, the decision to assign an inmate to OSP not only imposes extraordinarily strict conditions, but also suspends parole eligibility. While Ohio law does not create a liberty interest in parole, see Ohio Rev.Code § 2967.03; Wagner v. Gilligan, 609 F.2d 866, 867 (6th Cir.1979), a parole eligibility determination can indirectly affect the length of a prisoner’s incarceration, and is patently based on factors beyond the consideration of prisoner safety and prison management. See Ohio Admin. Code § 5120:1-1-07 (2004); Layne v. Ohio Adult Parole Auth., 97 Ohio St.3d 456, 780 N.E.2d 548, 555 (2002) (emphasizing that parole board may “consider any circumstances relating to the offense or offenses of conviction, including crimes that did not result in conviction, as well as any other factors [it] deems relevant”).
Even though assignment to a very restrictive prison might not by itself amount to a deprivation of a liberty interest, and even though a suspension of parole eligibility by itself may not amount to the deprivation of a property or liberty interest in Ohio, the combination of the two deprivations, in the context of the facts shown in the record of this case, amounts to a deprivation of a protected interest for procedural due process purposes. See Sandin, 515 U.S. at 487, 115 S.Ct. 2293 (finding no protected liberty interest in remaining free from disciplinary segregation, but noting that disciplinary record did not preclude parole); Neal v. Skimoda, 131 F.3d 818, 830 (9th Cir.1997) (finding that stigmatizing consequences of labeling inmate as “sex offender,” combined with parole ineligibility for non-completion of mandatory treatment program, triggered due process protections under Sandin). By way of analogy, the Supreme Court found a protected interest in Owen v. City of Independence, 445 U.S. 622, 633 n. 13, 100 S.Ct. 1398, 63 L.Ed.2d 673 (1980), from the combination of a loss of at-will employment (alone not a property interest under Board of Regents v. Roth, 408 U.S. 564, 92 S.Ct. 2701, 33 L.Ed.2d 548 (1972)) and defamatory statements (alone not a deprivation of a
B.
I also agree that the proper framework for evaluating whether the state procedures meet the requirements of procedural due process is the balancing test set forth in Mathews v. Eldridge, 424 U.S. 319, 334-45, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976). Such an analysis requires that each procedural protection sought for each category of administrative decisionmaking be evaluated independently under the Eldridge factors. We are not required to adopt for any prison-related decision the bundle of procedures required by one or another Supreme Court case dealing with different types of prison decisions. Thus discussion of the HewittfWoljf “dichotomy” is problematic. Many deprivations of liberty interests in prison, and certainly the ones in this case, are different in important respects both from the disciplinary rescission of good-time credits in Wolff v. McDonnell, 418 U.S. 539, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974), and from the administrative segregation in Hewitt v. Helms, 459 U.S. 460, 103 S.Ct. 864, 74 L.Ed.2d 675 (1983). The sought procedures may also vary. In short, the balance has to be context-specific. Thus whether or not a particular procedure imposed by the district court “tracks Wolff ” by itself does not tell us whether it is required by El-dridge. On the other hand, of course, the Supreme Court’s analysis with respect to what procedures are required to protect a particular liberty interest do provide guidance where the procedures or the interests are in relevant respects analogous. Thus to the extent, for instance, that Hewitt instructs that additional procedures with respect to “forward-looking” determinations are less likely to increase the accuracy of such decisions (the second Eldridge factor), 459 U.S. at 473-74, 103 S.Ct. 864, that guidance may appropriately be applied — not categorically but as part of the weighing — in other cases involving different procedures and different liberty interests.
I would also qualify the majority’s statement that “[i]t is not the nature of the decision which strikes the due process balance; it is the nature of the interests on both sides of that balance.” The Eldridge balance involves three factors, two of which can be characterized as “the nature of the interests on both sides.” The other, often dispositive factor, however, is the degree to which the desired procedures will increase the accuracy of agency deci-sionmaking. That determination often does depend on “the nature of the decision” in the sense that some types of decisions will be greatly benefited by certain procedures, while others may not.
C.
Applying the Eldridge analysis, I would uphold all of the procedural requirements imposed by the district court except the requirement that officials limit their placement decision to those matters detailed in the notice to the inmate.
At the outset, I note that our scope of review is de novo for legal issues such as whether procedural due process requires certain procedures, even though the issue is presented on appeal from the entry of an injunction. See Chao v. Hosp. Staffing Servs., Inc., 270 F.3d 374, 381 (6th Cir. 2001) (“A court abuses its discretion when
The comprehensive notice requirement imposed by the district court essentially provides inmates with notice of all of the evidence that may be relied on in determining his placement. Significantly, this requirement provides inmates appearing before classification committees with more notice than that received by criminal defendants at trial, where the liberty interests at stake are obviously more substantial. See Fed.R.Crim.P. 16; Weatherford v. Bursey, 429 U.S. 545, 559, 97 S.Ct. 837, 51 L.Ed.2d 30 (1977) (“There is no general constitutional right to discovery in a criminal case ... [and] the Due Process Clause has little to say regarding the amount of discovery which the parties must be afforded ... ”). As the Supreme Court has recognized, “[pjrison disciplinary proceedings are not part of a criminal prosecution, and the full panoply of rights due a defendant in such proceedings does not apply.” Wolff, 418 U.S. at 556, 94 S.Ct. 2963. In my view, the comprehensive notice requirement extends beyond what due process requires.
In upholding the district court’s requirement, the majority relies on Sandin to conclude that the liberty interest in this case is particularly weighty. If anything, however, the opposite inference is warranted. That is, under Sandin, a liberty interest arises from “atypical and significant hardships” not implicit in the original sentence. Typical or less significant hardships thus do not even rise to the level of a protected liberty interest. It is only the atypicality or the extraordinary significance of the hardship that is even enough to raise due process concerns. It follows that a hardship that is only marginally atypical and marginally significant should only be given marginal weight in an El-dridge analysis. It is illogical to say that any interest that meets the Sandin test must be of a weight requiring greater due process protection. On the contrary, a prisoner gains due process protection under Sandin only when the hardship exceeds typical hardship. Because in a sense it is the excess over typical hardship that warrants due process protection, it is logical that it be that excess that is weighed as the private interest in the El-dridge analysis. Since that excess may be very small, the fact that the interest was determined under Sandin may instead imply that the private interest be given a lesser weight than in the case of another type of protected property or liberty interest. In any event, the fact that the liberty interest is determined under a Sandin analysis cannot, without more, lead to the conclusion that the interests will be deemed to weigh particularly heavily.
Secondly, we must evaluate the increase in accuracy that will result from the procedural requirement that the decisionmakers limit their placement decision to the grounds and evidence detailed in the notice
Under the ORDC Officials’ new policy 111-07, inmates received written notice explaining the reasons they were referred for a classification hearing. See J.A. at 716, 731. The district court ordered that, not only must ORDC officials provide advance written notice of the reasons for the referral to a hearing, but that they must also provide “written notice of all the grounds believed to justify [placement] and a summary of the evidence that the [officials] will rely upon for the placement.” Austin v. Wilkinson, 204 F.Supp.2d 1024, 1026 (N.D.Ohio 2002). The district court added a footnote that appears to preclude consideration of evidence not described in the notice: “If [ORDC Officials] elect to use [a proposed form] to give an inmate notice, they must limit the grounds stated on the form and the evidence generally described on the form, to support placement at OSP.” Id. The majority fails to explain, however, how requiring ORDC officials to provide an inmate with such comprehensive notice increases the accuracy of the placement decision for a given inmate. The district court reasoned that
[requiring Department officials to give inmates specific notice of all of the grounds for placing and retaining them at OSP would cause minimal hardship. The officials would only need to expend the additional time to write out their reasons for making a specific classification decision. Furthermore, this minimal amount of additional time would increase the Department’s efficiency. Accurately summarizing all the grounds supporting an inmate’s placement at the OSP would later assist reviewing entities and avoid unnecessary prisoner assignments to the OSP.
Austin v. Wilkinson, 189 F.Supp.2d 719, 746 (N.D.Ohio 2002). The district court’s analysis misapprehends the burden that a comprehensive notice requirement imposes on the government. In making such decisions, ORDC Officials rely on a wide range of information. The hardship in expending additional time to write out reasons for making a specific classification decision may indeed be minimal. However, requiring that, prior to even conducting a hearing, ORDC Officials cull through often voluminous records and note every potentially relevant fact — on pain of barring them from considering any information, no matter how relevant, that was inadvertently omitted — is significantly more onerous. Such burdens have not been shown to be outweighed by the improvement in accuracy' — assuming there is any — asserted to arise from the comprehensive notice requirement.
Nor is the comprehensive notice requirement necessary to vindicate the interests cited by the district court. Under the procedural process required by the district court, the classification committee is responsible only for making the initial recommendation concerning whether an inmate should be assigned to OSP; both the warden (or the warden’s designee) and the Bureau of Classification must agree with the committee’s recommendation before an inmate can be placed at OSP. Austin v. Wilkinson, 204 F.Supp.2d 1024, 1026-28 (N.D.Ohio 2002). At each stage in the process, an inmate must be given a written statement explaining the justification for the placement recommendation and the evidence supporting it, as well as an opportunity to respond to the recommendation in writing. This process provides both a written record for review and ample op
As to the remaining procedures imposed by the district court, the Government has failed to articulate in any significant manner how they burden the government. On their face the additional procedural requirements appear to increase the accuracy and reliability of the decisionmaking process, and they were arrived at by careful consideration by the district court. The private interest is substantial. I therefore agree that the procedural requirements imposed by the district court, apart from the one discussed above, were properly imposed by the district court.
Conclusion
For the foregoing reasons, I concur in the judgment except to the extent that it upholds the requirement that ORDC Officials provide comprehensive notice to inmates appearing before classification committees.
. I agree that the substantive requirements must be reversed, for the reasons stated in Part 11(C)(2)(a) of the majority opinion.
