Full opinion text
OPINION AND ORDER GWIN, District Judge. In this case, a class of current and former prisoners at the Ohio State Penitentiary (“OSP”) says the defendants, all employees of the Ohio Department of Rehabilitation and Correction (the “Department”), violated their constitutional rights by denying them due process in their placement and retention at the OSP. The plaintiffs say conditions at the OSP give rise to a liberty interest because they impose an atypical and significant hardship on the prisoners in relation to the ordinary incidents of prison life. See Sandin v. Conner, 515 U.S. 472, 484, 115 S.Ct. 2293, 132 L.Ed.2d 418 (1995). Despite the existence of this liberty interest, the plaintiff prisoners claim the procedures used by the defendants in transferring them to the OSP and retaining them at the institution deny them due process. Responding, the defendants deny that conditions at the OSP are atypical or impose a significant hardship. In major part, the defendants argue that the Court should compare the plaintiffs’ conditions to those of other inmates at the OSP or to inmates in similar facilities in other states when determining whether the conditions are atypical. The defendants alternatively argue that, if a liberty interest is found, they afford sufficient process to the inmates. From January 2, 2002 through January 10, 2002, the Court conducted a bench trial on this matter at which it heard from twenty witnesses and accepted over one thousand pages of exhibits. After considering all of the evidence, and as hereinafter described, the Court finds that the nature and duration of restrictions at the OSP are conditions not expected by those serving similar incarcerations. The Court makes this determination despite finding that the current operation of the OSP, under the progressive stewardship of Warden Todd Ishee, has greatly improved inmates’ treatment. Instead, the Court finds that inmates at the OSP face an atypical and significant hardship even under Warden Ishee’s sensible leadership. In laying out its decision, the Court first describes the conditions at the OSP. Next, the Court discusses its holding that confinement at the OSP is an atypical and significant hardship in relation to the ordinary incidents of prison life. After discussing the nature of the confinement at OSP, the Court considers the process afforded to inmates in challenging their initial placement and subsequent retention at the OSP. Finally, the Court discusses the appropriate remedy for the constitutional violation that it finds. I. Factual Background and Discussion In this case, the plaintiffs represent a class of current and former inmates at the OSP. They sue certain state officials for violation of their constitutional rights under 42 U.S.C. § 1983. The plaintiffs sue the named defendants in their official capacities only for purposes of injunctive relief. Originally, the plaintiffs’ suit alleged that the defendants operation of the OSP was a violation of a number of the inmates’ Eighth and Fourteenth Amendment rights. The parties have settled most of the claims. The only claim left at trial was the plaintiffs’ claim that the defendants violated their right to due process in the selection and retention of inmates for the OSP. A. Purpose of the Ohio State Penitentiary The OSP is a high maximum security facility, also known as a “supermax” facility, located in Hubbard, Ohio, near the city of Youngstown. Constructed in reaction to the April 1993 riot at the Southern Ohio Correctional Facility at Lucasville, the OSP supermax prison first received inmates in May 1998. The OSP was designed to house 504 male inmates in single-inmate cells. The OSP was designed as “a more secure facility, to handle prisoners who were hellbent on disrupting the orderly operation of our correctional institutions.” (Wilkinson Dep. at 8). Ohio intended the OSP be “a location in the state that we can separate the most predatory and dangerous prisoners from the rest of the Department of Rehabilitation and Correction’s general population.” (Wilkinson Dep. at 24-25). The OSP carries out this goal primarily through solitary confinement — extended periods of incarceration in which the inmate is kept alone in his cell and has minimal contact with the outside world. In other prisons, this type of confinement is commonly referred to as “segregation.” The stark conditions and psychological consequences of solitary confinement at the OSP are noticeably different than at other Ohio prisons. Before describing the conditions at the OSP, the Court notes that it was presented with evidence at trial suggesting that Ohio does not need a high maximum security prison or does not need one with the capacity of the OSP. Peter Davis, a member of the Ohio Parole Board and former executive director of the Correctional Institution Inspection Committee of the Ohio General Assembly, testified about the Department of Rehabilitation and Correction’s use of the J-l eellblock at the Southern Ohio Correctional Facility. The Southern Ohio Correctional Facility is Ohio’s only maximum security prison, the security level immediately below the OSP’s high maximum security level. The J-l area is a self-contained eellblock of twenty cells that has tighter access requirements and allows less movement of inmates than a typical maximum security cell at the Southern Ohio Correctional Facility. The J-l eellblock is the most restrictive cell-block within the Southern Ohio Correctional Facility. Before the OSP opened, the J-l eellblock was the most restrictive and isolated eellblock in the Ohio prison system. (Davis Test, at 117-19). Suggestive that Ohio never needed the 504-inmate capacity of the OSP, before the OSP was built, Ohio did not fill the J-l cells at the Southern Ohio Correctional Facility. (Davis Test, at 119). Instead, Ohio faced a different problem. It did not have a sufficient number of maximum security cells, the level below the OSP’s high maximum security cells. The deficit of maximum security cells and the surplus of high maximum cells causes an imbalance in assigning inmates to appropriate confinement. In December 1998, a Department of Rehabilitation and Correction quality review team made up entirely of correctional officials reviewed the operations at the OSP. The quality review report supports the plaintiffs’ claim that no clear standard describes which inmates would be placed at the OSP: When asked about the inmate population for which OSP is intended most all respondents cite “the worst of the worst.” This concept has proven difficult to op-erationalize, particularly when we go beyond the 200 or so inmates who are clearly OSP material. Identifying those inmates who represent the “lighter” end of high maximum has become clouded by the overlap and similarity in characteristics among high close, maximum, maximum A.C. and high maximum inmates. (Pis.’ Ex. 10 at 2). The Department of Rehabilitation and Correction assigns inmates to the OSP from a conflicted position. The opening of the OSP has created too much capacity for the highest level of security. At the same time, Ohio lacks sufficient capacity at maximum security, the level of confinement below the OSP’s high maximum security level. After the huge investment in the OSP, Ohio risks having a “because we have built it, they will come” mind set. As a result, the defendants consider inmates for placement at the OSP who do not need its level of restrictions. B. Conditions at the Ohio State Penitentiary Conditions at the OSP are significantly more restrictive than at other correctional facilities of the Department of Rehabilitation and Correction. First, inmates at the OSP are kept in their single cells for twenty-three hours a day. An inmate’s cell measures approximately 89.7 square feet and is sealed with a solid metal door. The cell door has a small, thick glass window. The door also has a “food slot” or “cuff port” that an officer may unlock to insert food or mail, and that is also used to put handcuffs on the prisoner before the door is opened. Inmates eat all meals alone in their cells. The Department has made the cells more isolated by installing metal strips to the bottom and sides of the cells doors at the end of 2000. Before installation of the metal strips, the doors had half-inch gaps along the sides and two- to three-inch gaps along the bottom. The Department installed the strips ostensibly to stop the throwing of urine or feces, although the defendants did not give specific testimony of the frequency of such incidents. The Department does not use these metal strips at any other institution. The OSP cells have a narrow outside window that cannot be opened. These small windows do not comply with the square footage standard established by the American Correctional Association. An inmate has no control over the heating and cooling of his cell or the amount and temperature of air blowing through the cell. The cells are sparsely furnished, containing only a sink, a toilet, a small desk, and an immovable stool. Inmates sleep on a narrow concrete slab with a thin mattress, a pillow, a pillow case, a blanket, and a set of sheets. A light remains on at all times, although the light can be dimmed. The Department strictly limits the personal property an inmate may have at the OSP. An inmate at the OSP may only leave his cell for one hour a day. During that hour, inmates have access to a recreation area with two rooms. In one of these rooms, the inmates can use minimal exercise equipment. In the other room, a slot with a grate that is approximately six inches wide by four feet high allows outside air to circulate through the room. The room gives the sensory impression of a closed room with a small screen opening to the outdoors. Both recreation rooms are cells within the building. Until recently, recreation for all inmates at the OSP was solitary. Now, however, a limited number of inmates at the OSP may have recreation with one other prisoner. The ceiling height of the recreation rooms do not meet the American Correctional Association’s established standard. After auditing the conditions at the OSP, the American Correctional Association also found that the OSP did not comply with its standards for outdoor recreation. The OSP is significantly more restrictive than other Ohio correctional facilities, including those facilities housing inmates under administrative control. The other Ohio Department of Rehabilitation and Correction facilities do not lack outdoor recreation. In contrast to inmates at the OSP, inmates in administrative control on death row at the Mansfield Correctional Institution have outdoor recreation, more access to personal property, more access to telephone usage, and more access to counsel. (Jason Robb Test, at 153-54). Administrative control prisoners at the Lebanon Correctional Institution, Orient Correctional Institution, and Lorain Correctional Institution live in open-faced cells. The Lebanon facility allows administrative control inmates access to basketball courts and work-out areas. At the Southern Ohio Correctional Facility, administrative control inmates live in cells with bars, not solid doors, and have outside recreation available every day. Similarly, the Trumbull Correctional Facility has outdoor recreation. Most important to the quality of conditions, Ohio’s other facilities all afford inmates a much greater ability to communicate with other persons. The conditions at the OSP do not allow any amelioration of the prolonged isolation designed into the OSP’s structure. The Department of Rehabilitation and Correction also restricts inmates at the OSP more intensely during the brief periods they are outside their cells. For example, until very recently, the defendants required inmates receiving visitors to be shackled into an uncomfortable “black box” that contorted their wrists. In addition, inmates having visitors are still required to be stripped searched when they leave their cellblock and when they reenter the cellblock. The OSP subjects inmates to these searches even though the visits are made through solid windows that do not allow any possible contact with the visitor. The Department also limits contact between fellow OSP inmates. As described above, inmates always eat alone. They usually exercise alone, although now some inmates may occasionally exercise with one other inmate. Inmates are not permitted to share books, magazines, or other personal property. With the exception of some recently introduced group counseling sessions, inmates have almost no verbal communication with any other individual. The defendants’ expert, James Austin, described this: Q: When you were there, did you observe inmates communicating with one another? A: No. I observed there was some indications where they have two inmates who are recreating together, but I didn’t — I didn’t notice much communication because the inmates were almost always being brought out in isolation with no — I didn’t see any contact with other inmates except for those situations in the housing units where they could recreate together. (Austin Test, at 1020). The OSP’s limitation upon communication contrasts with conditions at other correctional facilities, including administrative control units at Ohio’s other prisons. For example, inmate James DeJarnette testified that inmates in administrative control at the Orient Correctional Institution can easily communicate through their bar doors. Such communication is next to impossible in the OSP with metal strips around the edges of the solid metal cell doors. Once at the OSP, an inmate faces different levels of restriction based on the classification level to which he is assigned. There are three classification levels, and the levels are assigned to discipline and to reward. After orientation, the OSP classifies most inmates at Level II. After approximately one or two years without problems, inmates can progress to Level III with attendant additional privileges. Similarly, the OSP assigns inmates who violate prison rules to Level I. Level I is the most restricted confinement possible and may last for an indefinite period of time. At Level I, prisoners receive no visitors except for their attorneys and receive no phone calls apart from calls with their attorneys or for family emergencies. At the OSP, inmates are normally allowed to make one ten-minute phone call a week. Under the Department’s new policy, effective March 1, 2002, inmates will only be allowed two ten-minute calls a month. Under the same policy change, the Department will reduce recreation outside the cell from seven one-hour periods per week to five one-hour periods a week. At the OSP, inmates do not participate in any prison-based work. One inmate per cellblock is responsible for keeping the area tidy. Except for the rare visitor and one hour a day of recreation, inmates remain in their cells. The OSP has no educational programs beyond the GED level. Instructional programs come to inmates through closed-circuit television and self-study workbooks. Most mental health programs available to the inmates also come through the small black and white television sets inmates are allowed in their rooms. In conclusion, inmates at the OSP live under significantly different conditions than prisoners at Ohio’s other correctional facilities, including those prisoners with maximum security classifications and those prisoners with maximum security classifications who are currently in administrative control. The Court now turns to a discussion of the defendants’ current procedures for selecting which inmates are placed at the OSP. C. Selection Procedures and Consequences of Placement at the Ohio State Penitentiary The plaintiffs challenge the procedures the Department of Rehabilitation and Correction uses to select inmates for placement and retention at the OSP. Regarding placement at the OSP, the Department first transferred inmates to the OSP in early May 1998. At the time of this initial transfer, the Department had no policy in effect identifying which inmates could suitably be placed at the OSP. Without any transfer policy in place, the Department simply relied upon wardens to choose inmates for transfer. Without any set criteria, similarly situated inmates were often treated differently. Peter Davis, an Ohio Parole Board member and former director of the Ohio General Assembly’s prison oversight committee, testified: The only thing that’s clear is, as I’ve said here, for every inmate that was cited to me and the reasons why that person was sent there, for this particular act, this assaultive behavior, if you will, we knew for a fact of plenty other inmates that were at other institutions, even close security institutions, that were not transferred there. My frustration was trying to understand how the criteria was being applied, whether or not it was being applied consistently at all institutions on referral, whether or not it was being applied consistently at central office level in those determinations. Just simply was no way to understand how one assaultive inmate could get high max placement and other assaultive inmates could not. (Davis Test, at 126). On August 31, 1998, the defendants attempted to establish some predictability to placement at the OSP by issuing Department of Rehabilitation and Correction Policy 111-07 (“Policy 111-07”). The defendants issued this policy after a large number of inmates had already been transferred to the OSP. The policy described behavioral criteria that referral committees at Ohio’s prisons should consider before recommending an inmate to the OSP. The policy was based on a April 15, 1998, memorandum by Regional Directors Norm Hills and Eric Dahlberg. The current version of Policy 111-07 used to transfer inmates to the OSP became effective on January 28, 1999. The Department’s review of an inmate’s placement at the OSP differs from the review inmates receive when they are placed into administrative control at other Ohio prisons. At other prisons, the administrative control committee can release an administrative control prisoner without a classification review. In contrast, release from the OSP requires an inmate to be reclassified from high maximum security to maximum security. An inmate at the OSP is considered for reclassification only once a year. This reclassification process includes several levels of review. Initially, an OSP inmate has a hearing before a three person reclassification committee. The reclassification committee is composed of an OSP deputy warden, a designee of the Department’s North Regional Director, and a Department mental health professional. Under the Department’s reclassification process, the OSP reclassification committee makes an initial recommendation of whether an inmate should stay at the OSP or be reclassified and transferred to a maximum security prison. After the committee makes its recommendation, the OSP warden reviews the decision and reaches his own conclusion. He makes this review without notice to the inmate of any additional factors not considered by the reclassification committee. After review and decision by the OSP warden, the chief of the Bureau of Classification reviews both the committee’s recommendation and warden’s decision. Finally, after the chief of the Bureau of Classification has approved or disapproved of the committee’s recommendation, the Department’s North Regional Director makes the final decision. After reviewing the evidence presented at trial on the issue, the Court questions how much consideration is given to the review of each inmate’s reclassification recommendation. Within the Bureau of Classification, Chief Bernard Ryznar reviews all reclassifications involving the OSP inmates. In addition to this task, he supervises a staff of nine and reviews each administrative control placement in Ohio’s 44,000 inmate population. (Ryznar Test, at 638). Although his co-workers assist him, Chief Ryznar is also responsible for authorizing each of the thousands of transfers within the Department of Rehabilitation and Correction each year. (Ryznar Test, at 618, 638). He personally reviews each maximum security placement. He also reviews other placement issues that are brought to him, and he supervises the placement recommendations of his staff. Finally, he alone makes at least 400 high maximum security classification decisions each year. (Ryznar Test, at 638-39). North Regional Director Stephen J. Huffman has even more responsibilities than Chief Ryznar that distract him from time needed to review high maximum security classifications. The Department’s Regional Directors essentially manage all aspects of the correctional facilities within their geographic area. Classification decisions are a very small part of their responsibilities in running a huge organization. Placement and retention at the OSP has immense consequences for some inmates. As described above, the OSP conditions are significantly more restrictive than the conditions at other prisons. In addition, initial placement at the OSP denies some inmates the chance to be considered for parole because of a Department policy. The Department has a policy, approved by Director Wilkinson, that prevents inmates in maximum security facilities from being paroled. Any inmate with a maximum security classification or any other classification title utilized to denote the most serious security risk inmates, at the time of release eligibility, shall not be granted release ... Release at any such projected release date shall be conditioned upon the inmate receiving a security classification less than maximum, or any other security classification title utilized to denote the most serious security risk inmates. (Pis.’ Ex. 3 at 8). Therefore, no inmate in the OSP’s high maximum security classification can be paroled. 1. James DeJarnette Inmate James DeJarnette’s case exemplifies the effect placement at the OSP has on parole eligibility. Convicted of armed robbery with a firearm specification, De-Jarnette began serving an indeterminate sentence of three to fifteen years on April 22, 1993. While housed at the Orient Correctional Institution on a medium security classification, DeJarnette assaulted a correctional officer while intoxicated. The Orient Correctional Institution discipline committee unanimously agreed DeJarnette should be punished by being placed in administrative control at his current prison. The discipline committee also unanimously agreed against increasing his classification level and transferring him to the OSP. The Orient Correctional Institution’s warden agreed with the discipline committee. Despite these recommendations and the dismissal of the criminal indictment against DeJarnette arising from the assault, the chief of the Bureau of Classification increased DeJarnette’s security classification three levels to high maximum and transferred him to the OSP in October 1998. DeJarnette was sent to the OSP without notice or explanation of why the Orient Correctional Institution’s discipline committee and warden’s recommendations were ignored. While at the OSP, DeJarnette participated in programming and mainly complied with all prison rules. The only exceptions were attempting to share a newspaper with another inmate and being found with an altered radio because a screw had come out of the radio cover. After fourteen months at the OSP, DeJar-nette received a classification review. In December 1999, the reclassification committee, made up of three Department employees pursuant to Policy 111-07, examined DeJarnette’s security classification. The committee recommended that he remain at the OSP. Warden Ishee and the chief of the Bureau of Classification concurred with the recommendation. As the result of DeJarnette’s three-level increase to high maximum security, he was automatically denied parole when he appeared before the Ohio Parole Board in August 2000. Parole board guidelines recommend that a first-time offender like DeJarnette receive parole after serving forty-eight to sixty months. Because of DeJarnette’s high maximum security classification, he could not be paroled even though he had served over ninety months. The parole board’s records explain its thinking: “Inmate serving 1st adult commitment. He is currently high max security with next screening^] next security screening 12/2000. He has served above the recommended range. However, his security status prevents a release recommendation. Time assessed to get inmate to earliest appointment for reduction to close [security] ... Requiring 19 additional months to serve until next [hearing] eligibility.” (Pis.’ Ex. DeJarnette-9 at 2) After receiving this indication from the parole board that he would qualify for release if he could move to an appropriate security level, DeJarnette again appeared before the OSP reclassification committee in November 2000. This time the reclassification committee recommended that De-Jarnette’s security classification be reduced and he be removed from the OSP, saying DeJarnette had made a “good adjustment” and was “not a behavioral problem.” (Pis.’ Ex. DeJarnette-11). Despite this recommendation for a reduction in security level classification, the OSP warden, Chief of the Bureau of Classification, and the North Regional Director all decided against reducing DeJarnette’s security classification. Importantly, the North Regional Director never heard from DeJar-nette before deciding to keep him at the OSP, never fully explained his reasons behind the decision, and never told DeJar-nette what issues prevented a reduction in his security level. DeJarnette’s case is especially troubling because increases or decreases in an inmate’s security classification level usually occur one level at a time. Jumping multiple security levels is the exception and not the rule. (Ryznar Test, at 628). After the Department reviewed the OSP’s operation in December 1998, the review team reported: Some Wardens believe that an inmate must progress through close security and maximum security before being considered an appropriate high maximum placement. Although this requirement is not found in either the original Memorandum regarding high maximum (Appendix C) or the policy on maximum security (111-07, Appendix D), there appears to be a strong perception among institutional personnel that this procedure must be followed. (Pis.’ Ex. 10 at 4-5). 2. Daryl Heard Inmate Daryl Heard’s experiences also highlight the consequences of placement at the OSP. His case further demonstrates the defendants’ willingness to disregard a reclassification committee’s recommendation concerning an inmate’s security level. After conviction in 1982 for aggravated robbery, aggravated burglary, and kidnap-ing, Heard worked his way down to a minimum security classification. In March 2000, the Orient Correctional Institution Rules Infraction Board brought rule violation charges against Heard for involvement with a scheme to bring marijuana into the prison. He was convicted of those charges and disciplined by being placed into local control for fifteen days. Additionally, his security classification was increased from minimal to medium. In June 2000, his case was submitted for possible high maximum placement. Because of his attempt to smuggle in marijuana seven months earlier, the chief of the Bureau of Classification ultimately recommended his security level be increased three levels to high maximum. In October 2000, Heard was transferred to the OSP. When deciding to increase Heard’s security classification four levels, the Department noted that Heard had not been involved in violence in the last sixty months. He served over three years at the minimum security classification without any incident prior to the marijuana involvement. On December 18, 2001, Heard had a parole hearing. At the time of the hearing, Heard had served 235 months of incarceration. The parole board guidelines suggested Heard should be paroled after serving between 156 and 192 months. (Heard Test, at 289). Heard was denied parole because .of his high maximum security classification. Recognizing the impact of the Department’s rule against paroling inmates classified as maximum or high maximum security, the parole board asked Heard to contact it immediately if he obtained a close security classification so another hearing could be scheduled. (Heard Test, at 293). 3. Keith Gardner Inmate Keith Gardner’s case gives a further example of the effect placement at the OSP has on parole eligibility. Now forty-four years old, Gardiner has been in prison since age nineteen for a murder conviction. Having served twenty-five years, he has served more time than the parole board guidelines suggest. The Department transferred Gardner to the OSP in November 1998 after an incident in which another prisoner was stabbed. The state prosecuted Gardner for the stabbing. At trial, Gardner argued self-defense, an affirmative defense to which Gardner had the burden of proving by a preponderance of the evidence. See Ohio Rev.Code § 2901.05(C); see also State v. Barnes, 94 Ohio St.3d 21, 24, 759 N.E.2d 1240, 1244 (2002). The jury acquitted Gardner, finding that he acted in self-defense. Despite the acquittal, the Department increased Gardner’s security classification and transferred him to the OSP in April 1999. While at the OSP, Gardner had no rule violations and he participated in numerous programs. In both 2000 and 2001, the reclassification committee reviewed his security classification and recommended that the Department reduce Gardner’s security classification. In 2000, Mansfield Correctional Institute Warden Baker, serving as a member of the reclassification committee, told Gardner he “shouldn’t even be here.” (Gardner Test, at 387). In addition, written on Gardner’s 2000 form filled out by the reclassification committee is the comment: “Placement to OSP questionable. No recent disciplinary action after A/C release 1/98.” (Pis.’ Ex. Gardner-3). On both occasions, the committee’s recommendation was rejected, and Gardner was kept at a high maximum security classification. In rejecting the committee’s recommendation to decrease Gardner’s security classification, the Department principally based its decision upon the stabbing incident of which he was acquitted years earlier. As with Heard and DeJarnette, the Department’s decision to retain Gardner at the OSP rendered him ineligible for parole. Gardner has served more than twenty-five years of his original sentence of life with a possibility of parole after fifteen years. However, the Department’s rule forbidding parole release from high maximum security denied the parole board an opportunity to exercise its discretion. D. Review of Inmates Once at the Ohio State Penitentiary As demonstrated by these accounts, the Department’s procedures for reviewing an OSP inmate’s classification do not provide the prisoner a hearing or even access to the individual deciding the inmate’s security classification. The Department’s procedures for initially moving someone to the OSP also suffer from the same lack of notice and opportunity for hearing. For instance, the Department sent more than one hundred inmates to the OSP before adopting its Policy 111-07 concerning such transfers. While lacking a formal policy, the Department transferred inmates under the guidance of the April 15, 1998, memorandum from Regional Directors Hills and Dahlberg. The memorandum informs wardens that inmates will be assigned to high maximum security when: The inmate is or is about to be classified as maximum security; The inmate has demonstrated behavior which meets high maximum security criteria; and/or The inmate presents the highest level of threat to the security and order of the department, in the professional judgment of the classifying official. (Joint Ex. 19). The memorandum’s lack of specificity and guidance made it almost worthless in describing which inmates should be assigned to the OSP. In addition, the vast majority of inmates initially transferred to the OSP in 1998 neither received advance notice that they were being considered for transfer nor a hearing to contest their transfer. Only a few inmates received posttransfer hearings. In support of their argument that some transfers and reclassification decisions are so irrational as to violate due process, the plaintiffs presented strong evidence showing that many of the Department’s decisions to transfer and retain inmates at the OSP were made with little support. Perhaps there are no transfers to the OSP with less support than those of Kevin Roe and Lahray Thompson. 1. Kevin Roe On January 26, 1989, the Cuyahoga County Court of Common Pleas sentenced Roe on a rape conviction. During his twelve years in prison, Roe had only minor rule violations, principally involving the use of marijuana. He was not charged with any rule violations in the two years before his transfer to the OSP. In January 1999, the Southern Ohio Correctional Facility dealt with increased tensions caused by a series of assaults among inmates identified as gang members of either the Aryan Brotherhood or the Crips. No evidence suggested Roe was directly involved in any of the incidents. During this same time, Roe received his annual classification review. After a hearing on January 20, 1999, the reclassification committee recommended raising his security classification from maximum to high maximum even though the board’s behavior worksheet indicated he should be considered for a security level decrease. The committee’s justification for its recommendation was that Roe was a longtime member of a gang and had participated in a racial disturbance over five years ago. (Pis.’ Ex. Roe-2 at 2). Roe was not given notice that these issues would be used to increase his security classification. Furthermore, the committee does not explain why this evidence, without any proof of Roe’s involvement in incidents at the prison, was sufficient to increase Roe’s classification in January 1999, but that same evidence did not warrant an increase in his security classification in May 1998, when the OSP first began accepting prisoners. Four days after his classification review, another inmate assaulted Roe by hitting him over the head with a spatula while Roe was standing in a food line. The control unit hearing board presiding over the disciplinary charge brought against the other inmate acknowledged that Roe did not fight back. (Pis.’ Ex. Roe-1). The Department never charged Roe with a rule violation for having gotten hit in the head from behind. Furthermore, the Department’s Security Threat Group Coordinator Matthew Meyer testified: Q: Did he [Roe] get in any fights before that? Before he got hit in the head, was he involved with anything? A: Not that we are aware of, sir. (Meyer Test, at 1106). Once Roe left the Southern Ohio Correctional Facility’s medical area he was transferred to the OSP. He did not receive notice of his impending transfer or an opportunity to challenge his upgraded security status. In an attempt to quiet sentiments at the Southern Ohio Correctional Facility, along with Roe, the warden sent approximately twenty other inmates to the OSP. Many of these inmates were transferred even though they had no current misconduct and the Department never made out or proved a rule violation associated with gang membership or other security threat group. At a closed hearing, Security Threat Group Coordinator Meyer testified regarding the evidence the Department relied on in its decision to assign Roe to high maximum security. Coordinator Meyer described violence and gang member gatherings in 1999 that did not involve Roe. Also, Coordinator Meyer never talked with Roe, and the evidence did not show that Roe had any connection with the incidents surrounding the 1999 gang disputes. Even considering the evidence generously, at best it shows that Roe may have had some past connection to the Aryan Brotherhood. However, the Department’s own rules require evidence of a leadership position in a security threat group before sending an inmate to the OSP. The defendants offered no credible evidence that Roe held a leadership position. Nonetheless, Roe’s high maximum security classification has continued for more than two years after the tension at the Southern Ohio Correction Facility has passed. Roe received a reclassification committee hearing in April 2000. Under the Department’s formula, he scored a negative three on his supervision review form, the best score possible. The reclassification committee recommended his release from high maximum security, commenting: [Roe w]as sent to OSP from general population 2 months after incidents. Was not involved in incident in library [word or words redacted] only involvement. He was hit with a spatula in chow hall. It appears he was assigned to OSP based on [word redacted] history in 97. Only 4 tickets during incarceration. No assaults or violence. (Pis.’ Ex. Roe-6). The warden approved the committee’s recommendation but Chief Ryznar denied it, stating that the “[inmate’s security threat group] activity directly led to assaults, fights and disturbance at [Southern Ohio Correctional Facility].” (Pis.’ Ex. 6). In February 2001, Roe received another reclassification committee hearing. Once again, Roe scored a negative three on his supervision review form. Once again, the reclassification committee recommended reducing his security classification. The warden agreed. However, once again, Chief Ryznar disagreed with the recommendation to reduce Roe’s security classification, stating “[r]ecommend continue high max. Involved in conduct that resulted in disturbance at SOCF.” (Pis.’ Ex. Roe-10). The Court is perplexed by Chief Ryz-nar’s decision keep Roe at the OSP. The Court is even more troubled by Roe’s lack of notice and opportunity to contest the reasons for which he was transferred. Although nothing presented in the trial’s closed session appeared confidential, the Court does not more fully describe the testimony out of an abundance of caution. It is sufficient to say that the Court has reviewed all of the Department’s records on Roe and finds nothing to support Chief Ryznar’s assertion he was involved in the 1999 incidents at the Southern Ohio Correctional Facility. The evidence of Roe posing an ongoing threat is exceedingly weak. Roe’s behavior and threat do not even meet the standards of the defendants’ own classification expert, James Austin. At trial, Austin describes these standards with regard to gang affiliation: Q: Could you give us examples of the type of thing [conduct that warrants long-term placement at OSP] you are thinking of? A: Inciting a riot, killing an officer, killing an inmate; you know, a history, a history of assaults on staff or inmates even though death may not occur, but every time this inmate comes out; a member of a Security Threat Group who is clearly the organizer of this group, and whenever this person is released a lot of bad things start happening wherever he or she may be. Q: Not someone who is just identified as a leader, but there’s got to be clearly some concrete evidence that this person in a particular prison leads to riots or leads to something bad happening? A: Yeah, that type of person. (Austin Test, at 1045). Nonetheless, Chief Ryznar decided to keep Roe at Ohio’s most secure, and most expensive, classification without much evidence and without giving Roe a hearing or informing him of the evidence leading to his retention at the OSP. 2. Lahray Thompson The defendants offered even less evidence justifying the placement and retention of inmate Lahray Thompson at the OSP. In January 1999, Thompson was present at a fight at the Southern Ohio Correction Facility involving members of antagonistic gangs. The defendants never brought rule violations against Thompson to the Rule Infraction Board for a hearing. The Department’s confidential security threat group file on Thompson makes no mention of what role, if any, Thompson played in the fight. Nonetheless, Thompson was transferred to the OSP. In transferring Thompson to OSP, the Department never clearly defined Thompson’s role in the incident but determined that he was affiliated with a gang. (Pis.’ Ex. Thompson-2 at 3). Thompson was not given notice that he might be classified to high maximum or an opportunity to defend against the gang affiliation charge before his transfer. Justifying the process used to send Thompson to the OSP, the defendants first say that Thompson indicated affiliation with the Crips while in his early teens living in California. Second, the defendants presented evidence Thompson once had a tattoo often associated with the Crips. Third, the defendants offered evidence that Thompson once wrote a letter using the letter “b” in a fashion sometimes used by Crips members to disrespect rival gangs. Finally, the defendants produced a summary report that said Thompson was present at the time of the January 1999 fíght. The report did not further describe his role, if any, in the incident. The defendants have not shown the Court that Thompson was involved in gang related conduct, save the uncertain evidence that he may have been involved in the January 1999 fight. Even if Thompson had participated in the January 1999 fight, under the Department’s own guidelines, his placement to the OSP appears unjustified. The edition of Policy 111-07 relevant to Thompson’s transfer says that an inmate can be recommended for placement at the OSP if his behavior shows he “functions as a leader or enforcer of a security threat group.” (Pis.’ Ex. 1). Chase Riveland, one of the defendants’ experts, supports this standard, giving his opinion that mere gang membership does not justify placement in an overly restrictive prison like the OSP. (Riveland Test, at 924-26). Despite the minimal evidence used to support Thompson’s placement to the OSP, the defendants have continued to keep him at a high maximum security classification. Thompson has not had a rule infraction citation while at the OSP. In both 2000 and 2001, the OSP’s reclassification committee recommended reducing Thompson’s security classification. However, on both occasions, he was ultimately denied a classification reduction. The treatment of Thompson and Roe reflects a troublesome trend where the defendants deny reclassification based upon gang activity without giving the inmates notice and an opportunity to respond. Equally troublesome, reclassification is denied based on exceedingly weak evidence and alleged activity years in the past. In Thompson’s case, he has been held in near solitary confinement for more than three years based on nothing more than the way he writes the letter “b,” the fact that in 1999 he was with a group involved in a dispute, and his association with Crip members more than fifteen years ago while growing up in southern California. Warden Ishee better described what the rule should be: Q: [These particular words] require the Security Threat Group coordinator to determine that a particular prisoner is an active member of a Security Threat Group or a Security Threat Group behavior. And just as a matter of clarification, do you understand the word “active” to mean current member of a Security Threat Group or currently engaging in Security Threat Group behavior? A: I think currently within a reasonable time frame. You know, if we, if we use the word “Current” exactly, that would have had to have been an action that just occurred or occurred within the last few days. Q: Well, what would seem to you a reasonable time frame? A: Within the-the period of review that applies to this instrument. Q: That is, within the past year? A: Generally, yes. (Ishee Test, at 586-87). Challenging the process afforded to inmates transferred to the OSP, the plaintiffs say the absence of process results in the OSP using a large amount of its expensive and restrictive capacity to house inmates involved with using or bringing drugs into a correctional facility. Currently, the OSP houses more than fifty inmates whose only rule violations are their involvement with drugs while in an Ohio prison. In an unusually large percentage of cases, Chief Ryznar overrules the OSP reclassification committee’s recommendations. Over a one-year period, the reclassification committee held 369 hearings on inmates at the OSP. Of those 369 hearings, the reclassification committee recommended reducing 157 inmates’ (43%) security classifications. (Defs.’ Ex. 0 at 9). But ultimately, the Regional Director only recommended seventy-one inmates (19%) for a security reduction. (Defs.’ Ex. 0 at 9). In contrast, Riveland, the defendants’ expert and former head of the State of Washington’s prison system, testified that he did not often overturn a committee recommendation concerning a high maximum security classification. In defending his reviews, North Regional Director Huffman simply says that he is more conservative and might have security threat group information not available to the reclassification committee. (Huffman Test, at 1141). The Court was not overly persuaded by this testimony as the security threat group information involved in the cases of Roe and Thompson hardly shows an ongoing threat even if one were to stretch and find a gang association in the first place. While the Court agrees that reviewing individuals should be conservative, the- issue here is not about releasing these inmates from prison. The inmates would simply be classified as maximum security, a level at which they remain subject to severe restrictions. Against this factual backdrop, the Court now turns to a discussion of the law controlling its determination of the plaintiffs’ procedural due process rights. In this discussion, the Court considers new policies that the defendants plan for the future. As discussed below, the Department’s new rules improve upon its current procedures but still do not provide the plaintiffs with adequate due process. II. Discussion of Procedural Due Process Criminal incarceration does not end all constitutional protections. See Wolff v. McDonnell, 418 U.S. 539, 555-56, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974); see also Estelle v. Gamble, 429 U.S. 97, 103-04, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976) (right to be free from cruel and unusual punishment); Cruz v. Beto, 405 U.S. 319, 321-23, 92 S.Ct. 1079, 31 L.Ed.2d 263 (1972) (right to freedom of religion); Johnson v. Avery, 393 U.S. 483, 485, 89 S.Ct. 747, 21 L.Ed.2d 718 (1969) (right of access to the courts). Although prisoners do not lose all constitutional rights, they are subject to additional restrictions upon those rights. See Turner v. Safley, 482 U.S. 78, 95-97, 107 S.Ct. 2254, 96 L.Ed.2d 64 (1987) (holding that the fundamental right to marriage is subject to limitations as a result of incarceration); Pell v. Procunier, 417 U.S. 817, 827-28, 94 S.Ct. 2800, 41 L.Ed.2d 495 (1974) (restricting inmates’ first amendment right to access to news media). Generally, prisoners retain those rights compatible with the objectives of incarceration. See Hudson v. Palmer, 468 U.S. 517, 523, 104 S.Ct. 3194, 82 L.Ed.2d 393 (1984); Vitek v. Jones, 445 U.S. 480, 493, 100 S.Ct. 1254, 63 L.Ed.2d 552 (1980) (noting that incarceration deprives inmate of right to freedom from confinement). Despite prisoners’ retention of certain constitutional rights, federal courts have been hesitant to interfere with the administration of prisons. See Sandin v. Conner, 515 U.S. 472, 482, 115 S.Ct. 2293, 132 L.Ed.2d 418 (1995) (noting that “federal courts ought to afford appropriate deference and flexibility to state officials trying to manage a volatile environment”); Turner, 482 U.S. at 85, 107 S.Ct. 2254 (stating that the separation of powers doctrine should caution judicial intervention in prison administration because it is an area traditionally governed by the state legislative and executive branches); Rhodes v. Chapman, 452 U.S. 337, 352, 101 S.Ct. 2392, 69 L.Ed.2d 59 (1981) (stating that courts should exercise judicial discretion rather than assume prison administrators are insensitive to prisoners’ constitutional rights). In this case, the plaintiffs say that the defendants violated their right to procedural due process. The Fourteenth Amendment to the U.S. Constitution provides that no state shall “deprive any person of life, liberty or property, without due process of law.” U.S. Const, amend. XIV, § 1. As the plaintiffs were not deprived of life or property, they are only entitled to due process if they were deprived of “liberty” within the meaning of the Fourteenth Amendment. The consideration of a due process claim goes through two steps. First, the Court asks whether a liberty or property interest exists with which the state has interfered. See Ky. Dep’t of Corr. v. Thompson, 490 U.S. 454, 460, 109 S.Ct. 1904, 104 L.Ed.2d 506 (1989); Pusey v. City of Youngstown, 11 F.3d 652, 656 (6th Cir.1993). Second, the Court determines whether the procedures attendant upon that deprivation were constitutionally sufficient. See Thompson, 490 U.S. at 460, 109 S.Ct. 1904; Pusey, 11 F.3d at 656. More specifically, the plaintiffs must plead and prove that state remedies for redressing the alleged violation are inadequate. See Hahn v. Star Bank, 190 F.3d 708, 716 (6th Cir.1999) (citing Vicory v. Walton, 721 F.2d 1062, 1066 (6th Cir.1983)). The Sixth Circuit has elaborated on this requirement: In other words, a party may maintain a procedural due process § 1983 case in federal court if he alleges and proves that there was a constitutional violation under color of law and: (1) The state did not have a remedy; or (2) the state had a remedy but it was deemed inadequate; or (3) the state had an adequate remedy in form, both procedurally and in damages, but the state did not apply it or misapplied its remedy. Id. With these elements in mind, the Court now analyzes the plaintiffs’ due process claim. A. The Plaintiffs’ Protected Liberty Interest To succeed on their procedural due process claim under 42 U.S.C. § 1983, the plaintiffs must first demonstrate they possessed a protected liberty interest and were deprived of that interest without due process. See Thompson, 490 U.S. at 460, 109 S.Ct. 1904; Pusey, 11 F.3d at 656. In Sandin, the Court revisited its earlier decision in Hewitt v. Helms, 459 U.S. 460, 103 S.Ct. 864, 74 L.Ed.2d 675 (1983). In Hewitt, the Court found that an inmate confined to administrative segregation did not have a liberty interest “independently protected by the Due Process Clause” because “the transfer of an inmate to less amenable and more restrictive quarters for nonpunitive reasons is well within the terms of confinement ordinarily contemplated by a prison sentence.” Hewitt, 459 U.S. at 468, 103 S.Ct. 864. Nevertheless, the Court went on to hold that because Pennsylvania state law had established “specific substantive predicates” that must exist before restrictive confinement could be imposed, the inmate had a protected liberty interest in avoiding such confinement. Id. at 471-72, 103 S.Ct. 864. Sandin rejected much of Hewitt’s reasoning. Specifically, the Court rejected Hewitt’s conclusion that a court had to ask whether “the State had gone beyond issuing mere- procedural guidelines and had used ‘language of an unmistakably mandatory character’ such that the incursion on liberty would not occur ‘absent specified substantive predicates.’ ” Sandin, 515 U.S. at 480, 115 S.Ct. 2293 (quoting Hewitt, 459 U.S. at 471-72, 103 S.Ct. 864). The Court found Hewitt had created disincentives for states to codify their prison management procedures and led to the involvement of federal courts in the day-today management of prisons. See id. at 482, 103 S.Ct. 864. In order to, correct these unwanted side effects, the Court held that liberty interests in the prisoner context “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, 103 S.Ct. 864. Although Sandin rejected Hewitt’s, methodology, the Court continued to “recognize that States may under certain circumstances create liberty interests which are protected by the Due Process Clause.” Id. at 483-84, 103 S.Ct. 864. In deciding whether the state had created a liberty interest, the Sandin Court emphasized that “the real concerns undergirding the liberty protected by the Due Process Clause,” id. at 483, 115 S.Ct. 2293, were whether the state deprived the prisoner of “an interest of ‘real substance,’ ” id. at 480, 115 S.Ct. 2293 (quoting Wolff, 418 U.S. at 557, 94 S.Ct. 2963). Sandin “shifted] the focus of the liberty-interest inquiry” away from “the language of a particular regulation” and back to “the nature of the deprivation.” Id. at 481, 115 S.Ct. 2293. Therefore, the Court must determine whether the nature of confinement at the OSP is an “atypical and significant hardship” giving rise to a protected liberty interest. This inquiry raises “many complex and fact-specific issues.” Brown v. Plant, 131 F.3d 163, 170 (D.C.Cir.1997). In deciding whether placement at the OSP is atypical, the Court must first choose an appropriate comparison. The plaintiffs argue the conditions at the OSP are atypical when compared to prisoners in administrative control. On the other hand, the defendants argue that the appropriate comparison is with the conditions at the OSP itself: “Ohio often transfers ‘high security risk’ inmates to OSP, the conditions of which are more restrictive than the sending prisons. Subjecting inmates to those conditions, which are no different from those ordinarily experienced by other similarly situated inmates in OSP is not ‘atypical.’ ” (Defs.’ Final Argument Br. at 12). The Court rejects the defendants’ suggested comparison. In Sandin, the Court said no claim could arise when the conditions of confinement were within the range that could reasonably be expected as a normal incident of incarceration. See Sandin, 515 U.S. at 484, 115 S.Ct. 2293. However, the Court did not suggest that the existence of a condition somewhere within a state’s prison system automatically made such a condition normal. The courts of appeals have used varied comparisons. The Fourth and Ninth Circuit use the general prison population as the comparative baseline. See Beverati v. Smith, 120 F.3d 500, 504 (4th Cir.1997); Keenan v. Hall, 83 F.3d 1083, 1089 (9th Cir.1996). In contrast, the Second and Third Circuits use the typical conditions of administrative segregation when deciding whether conditions are atypical. See Griffin v. Vaughn, 112 F.3d 703, 708 (3d Cir.1997); Brooks v. DiFasi, 112 F.3d 46, 49 (2d Cir.1997). Taking a different approach, the Seventh Circuit has defined the baseline as the conditions of nondisciplinary segregation in the state’s most restrictive prison. See Wagner v. Hanks, 128 F.3d 1173, 1175 (7th Cir.1997). According to the Fifth Circuit, segregation never implicates a liberty interest unless it lengthens a prisoner’s sentence. See Carson v. Johnson, 112 F.3d 818, 821 (5th Cir.1997). The remaining circuits have applied Sandin’s “atypical and significant hardship” test without characterizing the comparative baseline. See Bass v. Perrin, 170 F.3d 1312, 1318 (11th Cir.1999); Perkins v. Kan. Dep’t of Corr., 165 F.3d 803, 809 (10th Cir.1999); Kennedy v. Blankenship, 100 F.3d 640, 642 (8th Cir.1996); Dominique v. Weld, 73 F.3d 1156, 1160 (1st Cir.1996). The Sixth Circuit has not clearly described the appropriate comparison that the Court should use. See Mackey v. Dyke, 111 F.3d 460, 463 (6th Cir.1997). The Court finds that the better approach is to compare the range of prison conditions experienced by the plaintiffs against the complete range of conditions experienced by a broad range of similarly situated inmates. Placement in administrative control for limited periods of time is not an atypical change that imposes a significant hardship. But to determine whether the plaintiffs’ placement at the OSP is atypical in relation to ordinary prison life, the Court must consider the duration and extent of the deprivation because “especially harsh conditions endured for a brief interval and somewhat harsh conditions endured for a prolonged interval might both be atypical.” Sealey v. Giltner, 197 F.3d 578, 586 (2d Cir.1999). When viewed against this standard, the Court finds that the conditions at OSP are atypical and impose a significant hardship. In reviewing this issue, the Court begins with the time an inmate is held at the OSP. The vast majority of inmates placed at the OSP will remain for a minimum of two years, with only an annual review of their status. To date, two hundred OSP prisoners have been there for more than three years. This lengthy stay is a function of the current OSP procedures. Upon arrival at the OSP, inmates complete a month long orientation program and are then assigned to Level II. At the inmate’s reclassification hearing, conducted approximately a year after arrival, the inmate may progress from Level II to Level III. Only once an inmate is at Level III may he be reclassified to the less severe maximum security status. Because reclassification hearings are only held annually, even inmates with exemplary behavior rarely progress through OSP in less than two years. In contrast, inmates in administrative control or disciplinary control at other Ohio institutions, including the maximum security Southern Ohio Correctional Facility, are reviewed every thirty days. Terry Collins, the Department’s deputy director of institutions, testified that: The most you can do in a disciplinary control cell at one time is 30 days. Typically the arrangements there are anywhere from one to 15 days. You could be placed there — if there’s a spree of offenses that you go before the Rules Infraction Board on, you could be placed there for 30 days. Typically that stay is one to 15 days. (Collins Test, at 1176-77). He also testified that high maximum security incarceration at the OSP is qualitatively different than any other placement in the Department’s prisons: Q: And those categories of incarceration, Mr. Collins, those will all be philosophically institutionally distinct from high max status, won’t they? A: There — yes, most definitely different than high max facility. Q: High max will be intended for a different sort of inmate? A: Yes. Q: Will have a different sort of restrictive environment? A: Yes. (Collins Test. 1175-76). Even with good behavior, inmates at the OSP serve indefinite terms at the institution. With the exception of the end of their sentences, there is no other indication of how long an individual inmate will be kept at the OSP. In addition to inmates’ indefinite retention at the OSP, the Department treats inmates at the OSP qualitatively differently than it treats inmates at other Ohio prisons. With regard to isolation, death row inmates have access to true outdoor recreation and have direct access to attorneys. Most significantly, death row prisoners can interact with other inmates during recreation or by conversations in their cells. In contrast, inmates at the OSP have extremely limited contact with other individuals. The inmates remain alone in their solid-door cells for twenty-three hours a day. Metal strips along the cell doors do not allow conversation with adjacent inmates. Only a small, inoperable window allows the inmate to view the outside his cell. A correctional officer only opens the small “cuff-port” in the cell door to affix handcuffs or to drop off food. An OSP prisoner only goes outside the building in the rare case of a necessary medical procedure or a court appearance. Without contradiction, inmate Jason Robb testified that he had not been outside the walls of the OSP for four ye