One obstacle to thinking clearly about violence is the way violence can crowd out or compete with other dimensions of our shared experience: the idea, for example, that rape is either sex or violence but not both, or the sense that serious violence is incompatible with genuine youth, and that therefore children or adolescents who carry out violent offenses should not be treated as juveniles. Prison violence presents the opposite challenge. Violence in carceral institutions can seem so natural that it can lose its ability to shock. It can seem normal. Prisons become arenas of tolerated violence.
There are other arenas of tolerated violence. Popular entertainment is one of those areas, although the violence in movies, television shows, and video games usually is fictional. As discussed in Chapter 1, many competitive sports also provide, quite literally, arenas of permissible violence, and the violence in a sport like boxing, football, or hockey is real, not imaginary. But the violence in sports is constrained by rules: even in “mixed martial arts,” knives and guns are off-limits. And sports violence is at least nominally consensual for the participants. The violence that is tolerated behind bars goes much further.
This chapter will explore the complicated roles that violence plays in the legal regulation of prisons and jails, as well as the more general idea of arenas of permissible violence. Formally, violence against persons in custody is categorically off-limits. As an official matter, corporal punishment doesn’t exist in the United States, leaving aside the gaping exception of the death penalty. Outside the context of capital punishment, the constitutional ban on “cruel and unusual punishments” is understood, almost universally, to disallow striking or otherwise physically assaulting someone as a criminal sanction. The same constitutional ban, imposed by the Eighth Amendment, can also be violated by failures to protect people locked up in jails and prisons from assaults by their fellow inmates, at least when those failures manifest “deliberate indifference.” We punish people for crimes by removing them from society and depriving them of liberty, not by causing them physical injury or corporal pain. That is the theory.
In practice, though, jails and prisons are notoriously violent. Moreover, the violence often seems tacitly accepted, if not actively encouraged. Vulnerability to assault has long seemed an almost ordinary component of incarceration—part of how a prison sentence works, and what it has come to mean. This is particularly true of sexual assault. The cultural understanding of prison rape has changed in recent years; it is not quite the object of mirth in popular culture that it used to be. Nonetheless, the social response to sexual assault in prison remains ambivalent. It is formally condemned but often, in practice, treated as being far less serious than sexual assault outside of carceral institutions. To some extent this is true more generally of violence behind bars. Officially, physical assault is as serious a crime in prison as it is in the outside world. Unofficially, though, violent crimes in prison are often treated as though they do not entirely count.
The ambiguous legal treatment of violence in jails and prisons is linked to a debate about the nature and origins of that violence: what it is that makes carceral institutions so dangerous. One well-established view is that prisons are brutal and sanguinary because of the people who are sentenced to serve time in them. In a frequently quoted opinion, Judge Frank Easterbrook of the United States Court of Appeals for the Seventh Circuit wrote that “prisons are dangerous places” because “they place violent people in close quarters.” He explained that “those who have difficulty conforming to society’s norms outside prison may find obedience no more attractive inside—and the threat of punishment is diminished for one already serving a long term.”1
The reality of prison violence, we will see, is more complicated. Many prisons are horrifically violent, but many others are not. Nonetheless, the idea of prisons as dangerous places, and of prisoners as inherently violent, plays a pervasive, outsized role in how prisons and prisoners are understood, both by the public and by courts and legislatures. If prisons are dangerous because the people sent to prison are dangerous, then to a great extent prison violence seems unavoidable. “Some level of brutality and sexual aggression,” Judge Easterbrook explained, “… is inevitable no matter what the guards do.”2 Moreover, if the violence in a prison is the fault of the people locked up there, their victimization can seem less serious: they are not “innocents.” And violence in prisons can validate the need not just for restrictive conditions of incarceration, but for punitive criminal justice policies more generally: aggressive policing, for example, and long mandatory sentences. If crimes are committed by people who are, by their nature, violent, the violence of the law seems more justified. Tit for tat.
There is a different way to understand carceral violence, though—another long-standing explanation for why jails and prisons are dangerous. Perhaps prisons make prisoners violent, as opposed to the other way around. This theory finds some support in how widely rates of violence vary from one prison or jail to another, and how they have fluctuated over time. And if this theory is accepted, it carries different implications for how the legal system should respond to violence behind bars.
The word “prison” usually refers to an institution for long-term incarceration; a “jail” typically is used to lock someone up for a shorter period—pending trial, or to serve a sentence of a year or less. Most prisons are run by states; most jails, by local governments. But the lines can be blurred. Since 2011, for example, jails in California have held many people serving long, multiyear sentences. In this chapter I will sometimes use the words “prison” and “imprisonment” to refer to incarceration more generally.
More than 100,000 people in the United States are locked up in facilities other than prisons or jails: immigration detention centers, juvenile halls and “camps,” and psychiatric hospitals. Immigration detention centers hold more than juvenile and psychiatric facilities combined. I will focus here on violence experienced by the more than two million people incarcerated in American prisons and jails, but much of what I will say also applies to other locked institutions.3
The Supreme Court has never actually held that striking or otherwise assaulting a prisoner violates the Eighth Amendment’s ban on “cruel and unusual punishments,” but for the past half century lawyers and judges, including the Justices of the US Supreme Court, have largely assumed this is the case. The leading authority for this proposition is a frequently cited Court of Appeals decision, authored in 1968 by then-Judge Harry Blackmun, which banned the use of the strap as a disciplinary tool in Arkansas prisons. The decision has come to stand for the broader rule that prisoners cannot be punished corporally, or at least for the “general abandonment” of that practice. The one, glaring exception is capital punishment, which remains constitutional. It is noteworthy, though, that the primary means of execution today is lethal injection, a method designed to downplay the violence of the death penalty. The use of lethal injection is one part of the dramatic, centuries-long shift away from punishments that make a spectacle of violence toward punishments that consist, at least in theory, of deprivations of liberty, not bodily assaults.4
The death penalty has long been deeply controversial. But no one defends corporal punishments for criminal offenses, or for violations of prison rules. As we saw in Chapter 5, corporal punishment of schoolchildren remains constitutional and is practiced in some parts of the United States. Outside the school context, however, the use of bodily violence for official disciplinary purposes is not simply assumed to be unconstitutional; it has become virtually unthinkable. Here is an exception that proves the rule: In 2009, Stephanos Bibas—then a law professor, now a federal appellate judge—wrote a paper with the title “Corporal Punishment, not Imprisonment.” He argued that transitory, “non-disfiguring” physical punishments would actually be more humane and less permanently damaging than our current practices of incarceration. Bibas circulated the paper to other scholars and presented it publicly. But he never published it, and he later repudiated the whole idea. During his judicial confirmation process, he told the Senate that corporal punishment of prisoners is never permissible. He renounced it even as a thought experiment, in a disavowal to end all disavowals: “It is wrong and deeply offensive.… [I]t is cruel, and it is un-American.… It is degrading, inhumane, and an affront to human dignity. I categorically, emphatically, and unequivocally reject it.”5
Violence against prisoners can violate the Constitution even when it is not officially sanctioned. So can other excessively cruel conditions of confinement, such as overcrowding or denial of adequate medical care. For the most part the Supreme Court has said that conditions of confinement violate the Eighth Amendment only if they cause “serious” harm and reflect “deliberate indifference” by prison officials. But when correctional officials use force against inmates, the standard is different: an “unnecessary and wanton infliction of pain” violates the Eighth Amendment even if it does not cause serious harm. Guards responding to a prison disturbance, for example, violate the Eighth Amendment if they use force “maliciously or sadistically,” as opposed to “a good-faith effort to maintain or restore discipline.” In 2015 the Supreme Court held that any “objectively unreasonable” use of force by jail guards against a pretrial detainee violates the due process clause of the Fourteenth Amendment.6
Failing to protect people behind bars from attacks by other inmates can also violate the Constitution. Endorsing the reasoning of lower courts, the Supreme Court has said that “prison officials have a duty … to protect prisoners from violence at the hands of other prisoners”—especially because prisoners have “demonstrated proclivities for antisocial … and often violent conduct,” and because the state deprives them “of virtually every means of self-protection and foreclose[s] their access to outside aid.” Consequently, failing to protect incarcerated people from physical attacks is one of the conditions of confinement that can constitute “cruel and unusual punishment” in contravention of the Eighth Amendment. Like other conditions of confinement, though, it is treated as a violation of the Constitution only if it satisfies two tests: it must be the result of the deliberate indifference of prison officials, and it must have caused serious harm.7
Regarding sexual assault in prison, the official legal response is more emphatic, albeit at the statutory rather than the constitutional level. The Prison Rape Elimination Act of 2003—often referred to by its acronym, “PREA”—was signed into law by President George W. Bush after receiving unanimous approval from both houses of Congress. PREA declares “a zero-tolerance standard for the incidence of rape in prisons in the United States” and expressly aims to “make the prevention of rape a top priority in each prison system.” Almost two decades after the statute’s passage, its practical effects remain uncertain, for reasons we will explore below. But PREA cannot be faulted for equivocating about the evil of sexual assault in prison. Prison rape, the statutory findings explain, is an “epidemic” and a “day-to-day horror.”8
Formally, then, the law treats violence in prison as being no less violent, and no less criminal, than violence in the outside world. At the level of practice, though, things are different. Prison violence is notoriously common, and it often seems to be officially tolerated, if not encouraged.
Reliable statistics are difficult to come by, in part because prison violence is drastically underreported. Correctional officials have strong incentives to underreport violence, and prisoners fear retaliation for reporting attacks—retaliation from prison officials for reporting violence by guards, and retaliation from other prisoners for reporting violence by prisoners. Even homicide rates in prison are subject to manipulation; there is reason to believe that a significant but unknown number of prison homicides are erroneously classified as suicides or accidents. For example, a 2019 report by the US Department of Justice on Alabama’s state prisons for men found “numerous instances” of deaths that were likely caused by prisoner-on-prisoner violence but were classified by state correctional officials as being due to “natural” causes.9
Still, homicide statistics are almost certainly more reliable than statistics on other violent crimes among prisoners, just as they are for the US population more generally. And statistics on prison homicides do not tell a simple story. As American prison populations began to skyrocket in the 1980s, the homicide rate for people behind bars plummeted, from close to 60 per 100,000 before 1980 to between 3 and 4 per 100,000 in the early 2000s. Then the rate began to rise again, reaching 7 per 100,000 in 2012, 2013, and 2014, the most recent years for which national statistics are available. (All of these numbers are for state prisons; federal prisons appear on the whole to have been comparably violent, but they hold vastly fewer people, and as a consequence the number of reported homicides each year is too low to allow calculating a meaningful rate.) The prison homicide rate thus appears to have roughly doubled in recent years, but it remains about 90 percent lower than it was a quarter century ago. Moreover, it is not dramatically different from the general homicide rate in the United States, which has hovered between 4.5 and 5.5 per 100,000 since the year 2000.10
These figures, though, obscure as much as they reveal. Although rates of nonfatal physical attacks in prison are much harder to determine than homicide rates, violence in prison appears to be much more commonplace than in society more generally. A landmark survey of more than 7,000 prisoners in 2005 concluded that more than 20 percent had been physically assaulted by other prisoners within the previous six months. This was true for female as well as male prisoners, although most of the attacks on male prisoners involved a weapon and most of the attacks on female prisoners did not. Nearly a quarter of male prisoners reported experiencing physical violence from correctional staff within the previous six months; for female prisoners the figure was slightly over 8 percent. Even adjusting for the socioeconomic characteristics of the victims, physical assaults are more than ten times more common in prison than in the outside world.11
Moreover, rates of prison violence—both fatal and nonfatal—vary widely, not just over time but between institutions. Alabama men’s prisons provide a chilling example: their homicide rate went from close to zero in 2007 to 35 per 100,000 inmates in 2018—roughly five times the national average in 2012, 2013, and 2014. In 2017 a reporter who toured the St. Clair Correctional Facility in Springville, Alabama, and interviewed current and former inmates, described the facility as “virtually ungoverned.” He wrote that “corrections officers disappeared from cellblocks for long periods. Those who were present were often disregarded. With officers absent or ignored, vulnerable inmates, including those who were wounded and bleeding, often pleaded in vain for help, several inmates said. Violence—robberies in dark tunnels, assaults in crowded dormitories, stabbings in cramped cells—was virtually unavoidable.” The federal Department of Justice concluded in 2019 that violence was pervasive in Alabama’s entire system of men’s prisons, comprising thirteen separate institutions, and that there was reason to believe that the state was violating the Eighth Amendment by its deliberate indifference to the problem. The New York Times summarized some of what the DOJ found: “One prisoner had been dead for so long that when he was discovered lying face down, his face was flattened. Another was tied up and tortured for two days while no one noticed. Bloody inmates screamed for help from cells whose doors did not lock.” During a single week in 2017, the federal investigators reported, Alabama men’s prisons were the site of four stabbings, one of which was fatal; five beatings, one with a sock full of metal locks; three sexual assaults; and an arson involving setting fire to a prisoner’s bed while he slept. A follow-up report by the Department of Justice in 2020 found that instances of excessive force by guards in Alabama prisons were so common that some officers appeared to view them as unremarkable. Supervisors and other guards would watch guards brutally attacking prisoners and not intervene. Often the attacks were carried out on prisoners who were handcuffed or compliant. The 2020 report linked the pattern of unconstitutional force in Alabama prisons to the same overcrowding and understaffing that contributed to high levels of violence between prisoners. Indeed the DOJ noted that violence between prisoners often led directly to uses of force by guards.12
Alabama’s prisons are an extreme case, but they are not entirely atypical. Five prisoners were killed in Mississippi in just the first week of 2020. And vulnerability to physical attack is a defining feature of daily life in many, if not most, American correctional institutions. Sexual assault plays a special role here, despite the fact that it constitutes, numerically, a small fraction of prison violence. For decades the danger of being raped by fellow inmates has seemed an implicit, if unofficial, part of what it means to be imprisoned in the United States, particularly for men. Politicians and entertainers joke about it. For especially reviled defendants, the public sometimes seems to rejoice in, or at least to take solace in, the danger of prison rape. And for many incarcerated men, the threat of rape feels ever present. Craig Haney, a psychologist who has spent decades studying incarceration and its effects, says that for many men behind bars the threat of rape operates as a “primordial fear” demanding “hyper-vigilance” and, often, preemptive aggression. Women behind bars—a growing percentage of the prison population—tend to experience a different source of sexual abuse: they are more likely to be victimized by guards than by fellow prisoners. For women as well as for men, though, sexual abuse often seems to be an implied, informal component of prison sentences. The formerly incarcerated writer Chandra Bozelko calls prison rape, of men and of women, “the cultural wallpaper of American correctional facilities.”13
Most of the sexual abuse of women in prison involves verbal harassment, groping, voyeurism, or “consensual” sex, but rape by force or the threat of force occurs as well. For male prisoners, too, sexual assault appears to be accomplished more often by threats and harassment than by the actual exercise of force, but the threats operate against the background of the “primordial fear” of physical sexual violence.14
The actual prevalence of rape and other forms of sexual abuse in prisons and jails is difficult to assess, because much of it goes unreported. This may be particularly true when the victims are men, given the stigma and sense of emasculation associated with being the victim of male-on-male rape. Estimates of the percentage of prisoners experiencing sexual threats or assaults range from a low of 1 percent to upward of 20 percent. When it enacted the Prison Rape Elimination Act of 2003, Congress concluded that at least 13 percent of prison and jail inmates had been sexually assaulted while incarcerated. Certain subpopulations are at elevated risk of sexual assault in prison; these include juveniles, sex offenders, and inmates who are physically slight, mentally ill, or gay or gender nonconforming.15
The chief significance of PREA may have been symbolic rather than practical. The statute provided no new rights to prisoners. Instead it directed the Department of Justice to collect statistics on prison rape; it created a National Prison Rape Elimination Commission and directed it to recommend standards for addressing the problem; and it threatened a 5 percent reduction in federal funding to state prison systems that failed to comply with the standards or work toward compliance. Congress further mandated that the standards promulgated by the commission could not impose “substantial additional costs” on prison systems that adopted them. The standards were promulgated in 2009; they address training and supervision of staff; reporting, investigation, and prosecution of abuse; and a wide range of other procedures. A decade later only two states—New Hampshire and New Jersey—had certified full compliance with the standards. Forty-six other states had committed to working toward compliance.16
Even where they have been adopted, the effectiveness of the PREA standards in reducing rates of sexual assault remains uncertain. At the symbolic level, the passage of PREA did signify a shift in public attitudes toward the offense. It is not as common as it once was to treat prison rape as risible, or to welcome it as a form of retribution. Still, old attitudes die hard. Jokes about prison rape are still common. So is the sense—drawing on the tit-for-tat logic often found in discussions about violence—that prison rape can function as a kind of rough justice. When Larry Nassar, the former team doctor for USA Gymnastics, was sentenced in 2017 to 175 years in prison for sexually assaulting many of the athletes under his care, the judge told him that she was signing his “death warrant,” and that if the Constitution allowed it, she would allow “other or many people to do to him what he did to others.” Her comments were criticized, but they were also applauded.17
Nassar was assaulted by other prisoners hours after being released into the general population of a high-security correctional facility.18 As a notorious sex offender, Nassar was at high risk for such an attack; it is unclear why he nevertheless was released into the general population or whether any special precautions were taken for his safety. Prison officials may well have been blameless with regard to Nassar. It is hard to be confident of that, however, precisely because correctional staff are so frequently accused of acquiescing in or even encouraging violence between inmates.
Prisoners and former prisoners routinely charge that correctional staff knowingly tolerate sexual assaults and other violence among inmates, partly as a strategy for pacifying prisoners, and partly out of a sense, shared by many members of the public, that prison violence is part of how we punish convicted criminals for their offenses. Scholars studying prisons, and judges presiding over lawsuits brought on behalf of prisoners, have often reached the same conclusion, and correctional staff themselves sometimes acknowledge having looked the other way. At times prison officials appear to do more than turn a blind eye to violence within their institutions; some actively promote it, stage it, or carry it out themselves.19
In Alabama and Mississippi, inmates and officers have blamed the startling high rate of prison violence on a correctional culture that tacitly accepts it.20 In Texas a federal judge found that prison officials “deliberately resist providing reasonable safety to inmates,” turning a blind eye to beatings and rapes.21 In a separate federal lawsuit, California prison officials were found to have staged “gladiator fights” between prisoners in what are officially known as “Special Housing Units” and unofficially called “supermaxes”—ultra-high-security facilities supposedly reserved for the “worst of the worst.” (We will return to these orchestrated fights below.) The supermax guards also beat prisoners, hog-tied them, and used excessive force in removing them from their cells; this happened so frequently that the trial court judge concluded that guards were using force, not just to maintain discipline, but “for the very purpose of inflicting punishment and pain.”22
The Supreme Court ruled in 1994 that failing to protect prisoners from violent attacks violates the constitutional ban on “cruel and unusual punishments”—but only if (a) prisoners face a “substantial risk of serious harm,” and (b) correctional officials are aware of that risk and consciously disregard it by failing to take reasonable steps to reduce it. In order to prevail in court, prisoners raising a constitutional challenge to unsafe prison conditions must therefore prove that officials have culpable mental states, not just that they have tolerated or even fostered high rates of violence within their institutions. Showing that prison officials had this kind of culpable mental state—that their hearts as well as their hands were unclean—is notoriously difficult. Most legal scholars who write about prisons view this requirement as the single most significant doctrinal obstacle faced by prisoners challenging unsafe conditions.23
Part of the reason it is so difficult for prisoners to prove that officials had a culpable mental state is that judges often assume that a good deal of prison violence is beyond the control of state officials—“inevitable no matter what the guards do,” in Judge Easterbrook’s frequently quoted words.24 Judges assume that prisons are dangerous because the people sent there are dangerous. This understanding of prison violence—that prison violence is largely characterological, not situational—is also part of the reason the Supreme Court required a showing of conscious disregard in the first place. The Justices reasoned in 1994 that the Eighth Amendment bars only cruel and unusual punishment, and punishment is partly a matter of mind-set. But lying behind that semantic argument was an understanding of what prison officials, as well as prisoners, were up against: the fundamental fact that prisons grouped together “persons who have a demonstrated proclivity for antisocial criminal, and often violent, conduct.” Concurring separately, Justice Thomas was even blunter. “Prisons are necessarily dangerous places,” he wrote, because “they house society’s most antisocial and violent people in close proximity with one another.”25
This idea—that prisons are inevitably violent because the people who are sent there are violent—might be thought to have gained explanatory power over the past several decades because the percentage of prisoners serving time for offenses classified as “violent” has grown. Beginning in the 1990s, the increasing focus of criminal prosecutions on violent offenders, and the proliferation of long mandatory minimum sentences for violent offenses—the trends discussed in Chapter 2 of this book—resulted in a dramatic increase in the proportion of prison inmates who had been convicted of violent offenses. Today most state prisoners fall into this category. (Federal prisoners are different: half are serving time for drug offenses, and less than 10 percent have been convicted of violent crimes. But state prisoners outnumber federal prisoners by more than five to one.) But the spread of mandatory minimum sentences has also meant that prisoners today are significantly older on average, and older prisoners are less violent. Moreover, prison homicide rates dropped during the 1990s and remain far lower today than they were in the 1980s, when a significantly smaller share of prisoners were serving time for violent offenses.26
The dramatic drop in prison homicide rates since the 1980s—a fall of roughly 90 percent—is one reason to view prison violence as situational rather than dispositional. Another reason is how widely rates of violence vary from prison to prison and from state to state, even leaving aside the nightmarish figures from Alabama. In the first decade and a half of the twenty-first century, the homicide rate per 100,000 in state prison systems ranged from zero in several states and between 2 and 3 in Illinois, Pennsylvania, New York, and Texas, to 9 in California, 12 in Maryland, and 13 in Oklahoma. Similar disparities can be observed in prison violence statistics from other countries. The wide variations in rates of prison homicides and prison assaults suggest that the level of violence behind bars is a function more of how prisons are operated than of who is incarcerated.27
A well-run institution can guard against violence even among inmates who come to prison with histories of violence or with mental illnesses that make them dangerous. A poorly run prison is likely not just to fail at that task but to make prisoners more violent and more dangerous than when they arrived. The psychologist Craig Haney points out that “homosexual rape in men’s prison is often committed by persons who have no experience with rape or with homosexuality. The environment of jail and prison renders them capable of both.” Poorly run prisons may also have a culture that tolerates violence, both by prisoners and by guards—the kind of culture that federal judges found in California and Texas prisons in the 1990s, and that inmates and prison officials have reported more recently in Alabama.28
The toleration of prison violence reflects ideas not just about the origins of that violence, that it is dispositional rather than situational, but also about what counts as violence in the first place. The legal scholar Bennett Capers argues, “We treat prisons as invisible zones, as lawless zones, as zones that need not concern us.”29 Partly because prisons are out of sight, and partly because the people sent there have been convicted of crimes, there can be a tendency to write off what happens inside prisons, not to treat it as a true violent offense—similar to how violence inside marriages used to be treated, and how violence toward children still is sometimes treated. In debates about penal policies, for example, it is common to hear suggestions that, however else prison fails, it at least is fully effective at preventing people who are incarcerated from committing further crimes while they serve their sentences—as though crimes committed against other prisoners do not really count.30
The legal test for when prison officials responsible for violence between inmates—the rule requiring prisoners to show that officials knew about a particular risk of violence and intentionally disregarded it—may itself reflect a willingness to tolerate prison violence, or at least a sense that it need not distress us greatly. Some Justices on the US Supreme Court have argued that officials should be responsible for violence (or, for that matter, other unacceptable prison conditions) that they should have known about and taken steps to prevent, even if simple carelessness or a callous lack of concern made them unaware of the danger. In the language of the law, prison officials could be made responsible for negligent failures to protect prisoners, not just reckless failures. Legal scholars who study prisons have made the same recommendation, as have some lower-court judges. A longtime federal magistrate judge suggested some years ago that if the courts actually were serious about reducing prison violence, they would require officials to have more than a “clean heart” and an “empty head.”31
The doctrinal basis for a more demanding standard would not be hard to find. The Supreme Court has said for decades that police officers violate the Constitution—specifically the Fourth Amendment—when they use force that is “objectively unreasonable,” even if the officers are well intentioned. In 2015 the Justices said the same is true for guards in pretrial detention facilities, although there the rule is based on the constitutional guarantee of due process of law, rather than on the Fourth Amendment. In prisons, though, the Supreme Court reasons that the use of force by officials violates the Constitution only if it is used “maliciously and sadistically for the very purpose of causing harm,” in which case it amounts to “cruel and unusual punishment.” The legal scholar Margo Schlanger has pointed out that there is no principled reason the due process standard—whether force is “objectively unreasonable”—could not also be applied to prison guards; nor would there be any obvious logical problem with applying the same standard to a failure to protect prisoners, whether before or after a criminal conviction, from attacks by other inmates. No lower court has pursued either of these possibilities, however.32
Over the past three decades guards in various correctional facilities, ranging from “supermax” prisons to local jails, have been accused of staging “gladiator style” fights between prisoners.33 The most notorious of these abuses occurred in Corcoran State Prison, in California’s Central Valley, in the 1990s, where, by one estimate, guards staged some 8,000 fights between inmates between 1988 and 1998. Surveillance cameras at Corcoran recorded guards releasing prisoners from rival gangs two at a time into small exercise yards, in the hopes that they would attack each other. Guards watched the fights from a distance, sometimes bringing along girlfriends or other prison employees as guests. The guards also wagered on the outcomes, and they used rifles to break up fights that spiraled out of control. At least five prisoners were shot to death in these circumstances. Corcoran guards were also alleged to have orchestrated the rape of prisoners for disciplinary purposes.34
In 2000, eight Corcoran guards were prosecuted in federal court on civil rights charges arising from the gladiator fights, but their trial ended in acquittals.35 Despite that verdict, even some prison reformers believed that Corcoran-style gladiator fights were fading into history. Recent years, though, have seen new complaints of guards staging fights between prisoners, at Corcoran and at other prisons and jails.36
Prisoners, prison reformers, and scholars who study prisons have all suggested that American carceral institutions unofficially tolerate violence because it serves a range of purposes. Prison violence functions as a tacit part of criminal penalties, and a disciplinary tool, a way of keeping order and retaliating against prisoners who are insufficiently compliant. Gladiator fights orchestrated by guards show that prison violence, like violence more generally, also serves as entertainment, as spectacle—not just a necessary evil, but something enjoyed by onlookers. The fights are a kind of morbid parody of combat sports like football or boxing, minus the rules and even a veneer of consent.
Punishment itself used to serve as spectacle, quite overtly. Open-air executions, pillories, public floggings: all of these drew crowds in search of diversion. Public punishment was supposed to edify spectators and deter them from wrongdoing, but there is no question that it also functioned as entertainment.37 Prisons moved punishment out of sight and, in theory, substituted deprivations of liberty for physical chastisement. Today the spectacle of violence is found chiefly in movies, television shows, and video games—including, of course, the hundreds of movies and television shows set in prisons.38 But the idea of places or times marked off as zones of violence is not limited to stories set behind bars.
The 2013 horror movie The Purge imagines that in the near future a totalitarian political party takes over the United States, and the government establishes an annual “holiday”—the Purge—during which all crimes including murder are legal for twelve hours. The official purpose of the holiday is to allow for a “country-wide catharsis … letting us release the aggression we all have inside of us.” The Purge is what sociologists call a “moral holiday”—a temporary suspension of normal constraints on behavior—but with the official sanction of the government.39 So far the movie has spun off three sequels plus a television show, with yet another theatrical version in the works.
The Purge and its offshoots are plainly dystopian; the audience is not supposed to think that the annual spree of violence is actually a good idea. The filmmakers behind the franchise say they wanted to make a “cautionary tale” about the dangers of gun culture and the need for gun control. The business about “social catharsis” is supposed to be understood as a cynical pretense put forward by an ultra-right-wing, homicidal government. But of course the movies themselves offer their own version of violent catharsis. As one critic put it, “the stated moral may be that violence is terrible,” but “the visceral message is that it’s a lot of fun.”40
Prisons themselves are such a frequent setting for on-screen storytelling that a small scholarly literature has emerged debating the reasons for it. The explanations likely are several. Part of it is that prisons are exotic, precisely because they are walled off from public view. Some of it is the stripped-down iconography of prisons, which lends itself to movies and television. Much of it, no doubt, has to do with the stark and elemental oppositions that prison stories can so easily draw upon: good and evil, captivity and freedom, power and abasement, sin and redemption. Some of it pertains to the frisson of physical danger.41
Beyond all of this, though, movies and television shows set in prisons may provide some of what public punishment used to provide: the spectacle of violence, especially when visited on those who can be thought to deserve it. (Movies and television shows about police—“cop shows”—often do the same. So do vigilante fantasies of the kind that proliferated on-screen in the 1970s and 1980s.) The stock prison movie involves a sympathetic hero, usually one whose incarceration is unjustified. Moreover, most prison movies cast prisons themselves in an unflattering light; the guards are often portrayed as incompetent, corrupt, or sadistic. On the whole, prison movies therefore might be said to have a reformist subtext. Nonetheless, the standard prison movie portrays most prisoners as brutal and dangerous, suggesting not only that prisons are necessary institutions but that the violence visited on them, like the violence visited on corrupt or sadistic guards, is justified. It is a version of the tit-for-tat logic explored in Chapter 1 and a pervasive trope in discussions of violence.42
But prison violence is not just a trope of video entertainment. In the real world, hundreds of thousands of people locked up in prisons and jails in the United States are subjected to daily risks of violence that would be deemed intolerable in the outside world, but that the correctional facilities often seem tacitly to tolerate or even to encourage. Prison violence—like violence by police officers, or like violence against children by parents and teachers—seems often to be viewed through a different lens than “ordinary” violence, the kind of violence that the criminal law, increasingly, takes as its principal concern.