Skip navigation
× You have 2 more free articles available this month. Subscribe today.

When Prison is Not Enough: The Rise (and Perhaps the Fall) of the Supermax Prison

by Keramet Reiter

In 1986, the Security Management Unit opened in Florence, Arizona. It was a new kind of prison designed for long-term, total isolation, for prisoners whom prison officials said simply could not get by in the general prison population: gang members, the extremely violent, some death row prisoners. The 424 modular, poured concrete cells of the SMU had no windows, only fluorescent lights that remained on 24 hours a day, every day. A cement ledge for a bed and a steel toilet-sink combination passed as furniture. Prisoners were only allowed out of their cells for, at most, an hour or two a day in an exercise yard hardly bigger than their 8x10-foot cells, with time for a shower, if they were lucky. There was no place for any congregate activity. Officers in central control booths looked out over multiple pods of cells at once, and could open the mechanized cell doors, one at a time, with the press of a button. The facilities essentially eliminated the need for human contact. The architects, who had worked with prison officials to design this feat of modern technology, touted their streamlined, efficient innovation as: “the kind of facility that could almost be patented.”1 And prison officials all over the country took notice. California became one of the first states to copy Arizona’s model. Only California prison officials built a bigger and “better” facility, with 1,056 beds designed for long-term and total isolation. The Pelican Bay State Prison Security Housing Unit (SHU) opened in 1989. Located in rural California, on the state’s northernmost border with Oregon, it would become the archetypal supermax.

Over the next quarter-of-a-century, supermaxes would captivate prison officials, horrify the public, terrorize prisoners, push the boundaries of constitutionally acceptable punishments and plague northern California courts – and eventually courts across the United States – with a seemingly endless stream of prisoner complaints. In my book, 23/7: Pelican Bay Prison and the Rise of Long-Term Solitary Confinement, I tell the history of the Pelican Bay SHU. Why and how was it conceived in the 1980s? How has it resisted judicial scrutiny? What happens to prisoners inside, and how have they organized to resist the harsh conditions? Do they survive? What about life after the SHU? I hope that, if you are interested in these questions, you will read the whole book. I have coordinated with the publisher to make an advance run of paperback copies available especially for prisoners. But here, I preview just a few of the explanations, arguments and stories I tell in the book.

The Origins of the Supermax

When Arizona and California built those first supermaxes in the 1980s, the United States, and especially Arizona and California, were in the middle of massive prison building projects. Arizona’s incarceration rate increased from 2,000 prisoners in the 1970s to 39,000 prisoners in the 2000s, and California’s skyrocketed from 20,000 prisoners in the 1970s to 160,000 prisoners in the 2000s. Across the United States, prison populations quintupled in these years.2 One way to understand supermaxes is in this context of mass incarceration: as states were building more prisons to house their burgeoning prison populations, they built more isolation units, too. As California’s Warden of New Prison Design and Activation explained to me, prison officials estimated that they needed about two percent of beds to be isolation beds, for the trouble making prisoners, so that general population prisoners could be left alone. So, California planners calculated that, as the state built dozens of new prisons, and added thousands of new beds, they needed at least one thousand-bed, dedicated isolation unit.

But supermaxes were more than a logical addition to an expanding prison system. They were newly harsh: an architectural innovation, designed to concentrate “the worst of the worst” in semi-permanent, if not permanent, solitary confinement. Moreover, rates of prisoner isolation in supermaxes quickly expanded from two percent of state systems to five percent, and, in some cases, up to ten percent. The Vera Institute of Justice noted that between 1995 and 2000, the rate of solitary confinement use across the United States increased far faster than the rate of incarceration: a 40 percent growth rate in solitary confinement populations versus a 28 percent growth rate in prison populations.3 Given their harshness and their widespread use, supermaxes have also been understood as integral to tough-on-crime political agendas. Over the past few decades, these agendas have included longer and harsher sentences – mandatory minimums, three-strikes-and-you’re-out-laws, juvenile life without parole terms – and policies that work to systematically exclude prisoners and former prisoners from social re-integration, like permanent sex offender registries, and prohibitions on access to welfare, public housing and federal education loans for some categories of former prisoners. But both the mass incarceration context and the tough-on-crime justifications fail to fully explain why and how supermaxes have developed and spread over the past few decades.

My research has revealed that the earliest supermaxes were not actually popular, public innovations, touted by legislators and approved by voters. Unlike California’s prison building bonds, passed by a majority of state voters and approved by legislators in the 1980s, or the state’s Three Strikes Law, also passed by a majority of state voters and approved by legislators in 1994, California’s Pelican Bay supermax was never put to the test of a popular vote, either in the state legislature or at the state’s polls. Instead, motivated by the need to build prisons fast, as prison populations expanded, California legislators delegated authority over prison design to prison administrators, like wardens and other managers. When the Pelican Bay SHU opened in 1989, legislators and public observers knew it was a high-security prison, but they knew little else about its newly harsh design. Even prison litigators and judges in California were surprised when they started to receive letters from Pelican Bay SHU prisoners describing the harsh conditions of confinement at the prison: lights that never turned off; a few hours a week in empty concrete exercise yards; no human contact, save the occasional beating. Thelton Henderson, a federal judge based in San Francisco, took notice in 1990, just a few months after the SHU opened its doors. Although Judge Henderson began investigating conditions at Pelican Bay in 1990, it would be years before any of the policies at the institution actually changed. In sum, supermaxes were an administrative innovation, designed and institutionalized by prison wardens and managers, rather than by tough-on-crime legislators.

Supermaxes were also an extension of policies targeting in-prison organizing and activism that long pre-dated mass incarceration and the tough-on-crime era. Specifically, prison officials built supermaxes to institutionalize a common prison operational practice that dated back to the 1970s: long-term lockdowns. Lockdowns started in prisons in California and New York following moments of violent unrest by prisoners organizing for more and better civil rights. In August of 1971, guards shot George Jackson on the prison yard of California’s San Quentin State Prison, alleging that he had a gun and was trying to escape. Jackson bled to death on the San Quentin prison yard that August day. At the time, he was widely known as a best-selling author (of Soledad Brother) and a vocal, if radical, advocate for more racially just and humane treatment of accused and convicted criminals. The other prisoners charged with plotting with Jackson to escape were locked down, in deteriorating cellblocks on San Quentin’s death row. Some of these prisoners remained in solitary confinement for decades, even though their roles in the alleged escape attempt were either never established or ambiguous. For instance, Hugo Pinell was locked-down at San Quentin until 1989, until he was moved into the Pelican Bay SHU when it opened. In total, Pinell spent more than 40 years in solitary confinement.

Two weeks after George Jackson’s death, prisoners at Attica took over the prison, protesting conditions of confinement there. Once the National Guard stormed the prison, many prisoners were beaten up and held in isolation, initiating another period of long-term lockdowns.4 This pattern of violent unrest was repeated across the United States. Infamously, in 1983, Tommy Silverstein murdered Officer Merle Clutts at the United States Penitentiary in Marion, Illinois. This initiated a long-term lockdown that was eventually institutionalized in the form of the federal supermax in Florence, Colorado – known simply as ADX – which opened in 1995. Tommy Silverstein remains in total solitary confinement at ADX to this day. The chronology of lockdowns and supermax building is important: this was not a federal innovation, but a state innovation, initiated in Arizona and California, and then copied by dozens of other states and the federal government. By the late 1990s, nearly every state had its own supermax.

The Legal Legitimizing of the Supermax

Readers of Prison Legal News are likely to be familiar with the fact that the birth of the supermax inspired plenty of legal challenges to both the harsh conditions of confinement in the facilities and to the often arbitrary procedures underlying prison officials’ decisions to place prisoners in isolation. The first major class-action case to assess the constitutionality of supermax confinement was litigated in California, led by prisoners in the Pelican Bay SHU and lawyers at the Prison Law Office (now famous for spearheading the Plata litigation about prison overcrowding in California). The case was Madrid v. Gomez. It began in 1990 when Judge Thelton Henderson, already renowned then for a wide range of civil rights affirming decisions, including a number of cases in which he upheld the rights of criminal defendants and prisoners, started receiving letters from prisoners housed at Pelican Bay complaining about the conditions there. Judge Henderson was so shocked by the number of letters, and the harsh conditions they described, that he “invited” the warden of Pelican Bay to a meeting in San Francisco to discuss what was going on up at Pelican Bay. Within a year, Judge Henderson had certified a class of all the prisoners at Pelican Bay, appointed counsel from the Prison Law Office (a nonprofit) and Wilson Sonsini Goodrich & Rosati (a local corporate law firm), and visited the prison himself. In 1993, he presided over a three-month trial interrogating conditions and uncovering abuse after abuse that had taken place at Pelican Bay, from prisoners being shot inside housing units, to months-long delays in medical care, to psychotic prisoners being housed outside naked in cages. Guards bathed one prisoner, who had a documented mental health problem and had smeared himself with his own feces, in scalding water, holding him down until his skin peeled off.

The gruesome stories of abuse were more like something out of a myth about a medieval torture chamber than something out of an actual Sixty Minutes episode (that aired on the eve of the Madrid trial) about a modern state-of-the-art prison. But these gruesome stories of abuse overshadowed the more fundamental challenge at the heart of the Madrid case – to the very practice of indefinite solitary confinement. In the end, Judge Henderson held that placing seriously mentally ill prisoners in the SHU at Pelican Bay violated Eighth Amendment prohibitions against cruel and unusual punishment, and he ordered many reforms to the way guards and staff treated prisoners, including better training for staff and more healthcare for prisoners. But he did not find that the conditions of isolation in the SHU, or the months and years prisoners were spending in those conditions, violated the U.S. Constitution.

After Judge Henderson’s decision in Madrid in 1995, many other prisoners across the United States challenged aspects of both the conditions in supermaxes and the procedures governing assignment to supermaxes. Most courts followed the precedent of Madrid and ordered the exclusion of mentally ill prisoners from long-term isolation. And in 2005, the U.S. Supreme Court – in the only Supreme Court case to directly consider the constitutionality of supermaxes – held that Ohio prisoners at least had a liberty interest in not being placed in the state’s supermax (Austin v. Wilkinson). If prisoners had a liberty interest in staying out of the supermax, then they were also constitutionally entitled to some minimal due process protections before being placed there, including notice about what justified the placement and some opportunity to respond to this notice. But no court has held that any length of time in solitary confinement, in a modern supermax, is per se unconstitutional.

In a sense, then, the Madrid decision also paved the way for other states to build supermaxes, and to leave prisoners languishing in these facilities for months, years and now, decades. As long as supermaxes provide the minimum necessities of survival – running water, a few hours a week outside with some access to natural light, a minimum daily dose of calories – and avoid physically harming prisoners, they are unlikely to be declared unconstitutional. As Prison Legal News reported presciently in describing the 1995 decision in Madrid, prison officials essentially won the right to maintain prisoners in long-term solitary confinement: “The court has given the green light for the proliferation of super-max control units, even those as harsh as Pelican Bay, and politicians and prisoncrats throughout the country are already moving forward with efforts to do just that.”5

Nonetheless, after Judge Henderson’s 1995 order in the Madrid case, lawyers from the Prison Law Office monitored conditions at Pelican Bay for the next 15 years, reporting regularly to Judge Henderson. And things improved. Mentally ill prisoners were diverted to isolation units that at least had windows. Prisoners at Pelican Bay received better healthcare. There were no more reports of shootings, nudity in outdoor cages or scalding baths.

But for the prisoners left in the Pelican Bay SHU – the ones with no pre-existing mental health problems, the ones whom prison officials had assigned to the SHU indefinitely, because they had been labeled as dangerous gang leaders – the years ticked on and on. In 2011, Judge Henderson found that Pelican Bay was finally in full compliance with his 1995 orders in Madrid, and he closed the case. At that moment, there were more than 500 prisoners in the Pelican Bay SHU who had been in total isolation there for more than ten years. And 78 prisoners had been in Pelican Bay more than 20 years, since the Madrid litigation had first begun.6 Many of those prisoners had started to feel as if they had nothing to lose. They would soon collaborate to lead the best-organized protest of prison conditions since the Attica uprising 40 years earlier. In part, few options were left to them.

Unsettling Solitary Confinement

Between the summer of 2011 and the summer of 2013, prisoners in isolation at Pelican Bay led three separate hunger strikes to protest the conditions of their confinement. The hunger strike leaders – all of whom were serving indefinite terms in isolation as “validated” gang members – collaborated across previously mortally divisive gang rivalries to coordinate the collective actions. In a system renowned for institutionalizing racial segregation, the prisoner hunger strikers set aside racial divisions, uniting behind the common goal of criticizing both the harsh conditions of their confinement and the administrative process by which prison officials (not judges) determined they were gang members. In 2011, three pieces of evidence, like a tattoo (Aztec tattoos might establish Mexican Mafia membership) or being in possession of revolutionary literature (the writings of George Jackson might establish Black Guerilla Family membership), could send a California prisoner to isolation forever, at least for the duration of his or her prison sentence. The third hunger strike, in August of 2013, involved 30,000 prisoners, some of whom refused food for 60 days.

Following the first hunger strike, the prisoners secured a team of civil rights counsel, including Legal Services for Prisoners with Children (based in San Francisco) and the Center for Constitutional Rights (based in New York City). In May of 2013, the legal team sought to certify the class of the 500 people who had been in isolation in the Pelican Bay SHU for 10 years or more. Throughout the hunger strikes and the early stages of the Ashker litigation, prison officials focused on maintaining their hard-won, total control over isolation in California. In 2012, prison officials initiated preemptive but superficial reforms, eliminating automatic assignment of gang members to indefinite isolation and instituting a new program to facilitate the transition of some gang-affiliated prisoners out of the SHU. Publicly, officials persistently reaffirmed the need for long-term isolation of gang members. They filed affidavits in the Ashker case characterizing the lead plaintiffs as vicious, manipulative and self-interested gang leaders. In a 2013 Los Angeles Times op-ed, Secretary of Corrections Jeffrey Beard described the Ashker plaintiffs as “convicted murderers who are putting lives at risk to advance their own agenda of violence.” The scale and duration of the nonviolent hunger strike, involving sustained collaboration of alleged rival gang leaders, provided a wordless but decisive rebuttal to Beard’s assertions.

After Judge Claudia Wilken certified the class of 500 prisoners who had spent more than 10 years in the SHU (in June of 2014), prison officials transferred 8 of the 10 named Ashker plaintiffs out of the Pelican Bay SHU. Undeterred, Judge Wilken issued an order in March of 2015 permitting these plaintiffs to continue to serve as class representatives. The prisoner plaintiffs then filed 10 damning expert reports assessing the destructive impact of long-term solitary confinement. The case was not going away. Talks of a settlement brewed.

On September 1, 2015, California prison officials agreed to a settlement in the case of Ashker v. Brown. Todd Ashker, together with the other prisoner plaintiffs housed in the SHU, alleged that the practice of assigning hundreds of prisoners to solitary confinement for indefinite terms dragging out 10 years and beyond violated the U.S. Constitution’s prohibition against cruel and unusual punishment, as well as basic due process rights. In the Ashker settlement, prison officials promised to drastically reduce the use of isolation in California, ending four years of intensive collective action and litigation about the practice.

Meanwhile, prison officials continued to move some prisoners out of isolation in the Pelican Bay SHU into general prison populations throughout California. Hugo Pinell was one of these prisoners. Nicknamed “Yogi Bear,” Pinell spent 45 years in total isolation deep within the California prison system. He was originally convicted of a rape, in 1965. But in prison, he was convicted of killing one correctional officer in 1970 and of assaulting two other correctional officers during George Jackson’s fatal escape attempt from San Quentin in 1971. (Of the “San Quentin Six” charged with conspiring to help Jackson escape, Pinell’s was the only conviction that was never overturned.) After Pinell was moved into the Pelican Bay SHU in 1989, he spent the next 25 years alone there, in an 8x10-foot windowless, poured concrete cell, under fluorescent lights that never completely turned off. Pinell’s last contact visit with a friend or relative was in December 1970.

On July 29, 2015, however, Pinell’s circumstances changed. Prison officials moved him into the general prison population at California State Prison, Sacramento. Pinell became a test case in the prison system’s nascent efforts to cut down on the use of indefinite solitary confinement. The test failed. On August 12, 2015, two prisoners stabbed Pinell to death. A riot followed; 11 more prisoners sustained non-fatal stab wounds.

Pinell’s death foreshadowed all the ways reform of solitary is and will be difficult. In general, prisoners who have spent decades in isolation are likely to suffer from some mental health problems – whether anxiety, hallucinations or depression – and to have difficulty adjusting to human contact, let alone life in a general population prison. Prisoners like Pinell, who spent 45 years in isolation, are especially likely to be well known to other prisoners who might have something to prove by planning a violent attack. Likewise, Pinell was well known to prison staff, who also might have had something to prove (about the necessity of keeping prisoners isolated) by permitting violence against him to take place.

Nonetheless, Pinell’s death hardly stalled the momentum of solitary confinement reform in California. Prison officials and the Ashker plaintiffs announced their settlement agreement just two weeks later. The 33-page agreement included five critical provisions, focused on the Security Housing Unit at Pelican Bay State Prison:

  1. Justifications for Confinement: Gang validation is no longer grounds for assignment to isolation in California; instead, prison officials may only assign prisoners who break specific in-prison rules to isolation.
  2. Durations of Confinement: Indeterminate isolation is now banned; five years is the new, hard limit on how long any prisoner can spend in isolation in the Pelican Bay Security Housing Unit.
  3. Conditions of Confinement: After five years, even dangerous or non-compliant prisoners must be provided with some opportunities to participate in education and rehabilitation programs, have regular contact with other prisoners, and have physical contact with approved outside visitors.
  4. Retroactive Application: California prison officials agreed to apply these new rules to all prisoners currently housed in the Pelican Bay SHU, moving all eligible prisoners into the general prison population within one year.
  5. Data Collection: Officials agreed to provide for 24 months thorough monthly data about the characteristics of the SHU population to plaintiffs’ lawyers, in order to document ongoing compliance with the agreement.

The settlement attracted national attention and is still being celebrated by prisoners, their families and legal advocates. Perhaps it will be a model for other states to reduce or eliminate prison conditions the United Nations has conclusively defined as torture. One settlement agreement, however, cannot sweep away decades of abusive prison policies. First, it is a settlement, not a legal opinion. At best, the settlement is a non-binding model of what other jurisdictions might attempt. Second, even though prison officials withdrew many of their claims about the dangerousness of SHU prisoners by agreeing to the provisions of the Ashker settlement, these beliefs have hardly been renounced. The genuine fear prison guards experience in coping with hunger strikes, managing mental illness or dealing with prisoners like Hugo Pinell must be acknowledged and addressed, so that they are motivated to strategize to support, rather than resist, reform. Third, the data collection and monitoring associated with the settlement is scheduled to conclude in two years – and may never be made public in the first place. The practice of solitary confinement has historically been defined by discretion and invisibility, and is therefore hard to investigate, control and reform. So the practice of solitary confinement could easily retreat back into the shadows in two years, absent longer-term requirements to institutionalize transparency. Hopefully, the Ashker settlement has actually unsettled solitary confinement practices in California. But that is just the first step on a long journey towards serious and sustainable reform.

In the end, the story of the supermax is the story of what happens when prison is not enough. Prison officials in the 1970s felt they needed more prison – longer, harsher conditions of confinement – in order to control activist prisoners. More broadly, voters and legislators supported more prisons in the form of longer sentences and more funding for building and running prisons. The story of the supermax forces us all to reconsider when and how prison reform might be possible. The abuses that have taken place deep inside the prisons within the prisons of supermaxes suggest that neither laws nor moral values have placed adequate limitations on punishment – in terms of its scale, its duration or its intensity.

Keramet Reiter is an assistant professor in the Department of Criminology, Law & Society and at the School of Law at the University of California, Irvine. Her latest book is 23/7: the Pelican Bay Supermax and the Rise of Long-Term Solitary Confinement, released by Yale University Press in October 2016. Portions of this article were previously published on the Social Justice Blog ( in September 2015.


1  Justice architect (Arizona). Telephone interview with the author, Feb. 10, 2011. Notes on file with the author.

2  Mona Lynch, “Punishment, Purpose and Place: A Case Study of Arizona’s Prison Siting Decisions,” Studies in Law, Politics & Society, Vol. 50 (2009): 109-37; Keramet Reiter, “Parole, Snitch, or Die: California’s Supermax Prisons and Prisoners, 1987–2007,” Punishment and Society, Vol. 14.5 (Dec. 2012): 530–63; Franklin Zimring, “The Scale of Imprisonment in the United States: Twentieth-Century Patterns and Twenty-First Century Prospects,” Journal of Criminal Law and Criminology, Vol. 100. 3 (2010): 1225–46.

3  John J. Gibbons & Nicholas J. Katzenbach, Confronting Confinement: A Report on the Commission on Safety and Abuse in America’s Prisons, New York, NY: Vera Institute of Justice, 2006: 52-3.

4  For a thorough review of the events at Attica in 1971 and the aftermath, see Heather Ann Thompson, Blood in the Water: The Attica Prison Uprising of 1971 and Its Legacy, New York, NY: Pantheon, 2016.

5  Prison Legal News, “Pelican Bay Ruling Issued.” Aug. 1995, 3.

6  Julie Small, “Under Scrutiny, Pelican Bay Prison Officials Say They Target Only Gang Leaders.” KPCC [Pasadena, California], Aug. 23, 2011,

As a digital subscriber to Prison Legal News, you can access full text and downloads for this and other premium content.

Subscribe today

Already a subscriber? Login