Introduction
Human beings need sleep. We all know this. When we do not get enough sleep or when the sleep we get is repeatedly interrupted, our ability to function—to focus, to treat others appropriately, to respond effectively to life’s daily challenges—will be severely compromised. And these are just the effects we notice. Over the past decades, advances in sleep science have made clear just how necessary adequate sleep is to every dimension of human health. A chronic insufficiency of sleep increases the risk of, among other medical conditions, “obesity, diabetes, cardiovascular disease, and infections.” As noted sleep scientist Matthew Walker puts it, “Every major system, tissue, and organ of your body suffers when sleep becomes short. . . . [T]he shorter your sleep, the shorter your life.” And persistent inadequate sleep takes an equally great toll on mental health, exacerbating “all major psychiatric conditions, including depression, anxiety, and suicidality.”
For adults, the consensus recommendation is a minimum of seven hours of uninterrupted sleep per night; otherwise, the processes operating at each stage of the sleep cycle will be unable to perform their restorative work. It is no secret that many Americans from all walks of life routinely fail to meet this target. Judging from the size of the sleep aid market—an estimated $65 billion is spent annually in the United States alone—access to financial resources is no guarantee of adequate sleep. Yet, as Keisha Ray observes, the more limited one’s access to the social determinants of health—including, among other things, “food security, stable housing,” and “the freedom to choose where and how we sleep”—the less well-slept a person is likely to be. Access to adequate sleep, in other words, exists along a continuum, on which those lacking socioeconomic power and control over their physical environment are likely to experience a greater measure of sleep deprivation than society’s more privileged members.
This Article focuses on a group at the extreme end of this spectrum of disempowerment: those incarcerated in American prisons. Drawing on original data collected through in-depth, semi-structured interviews with people who served time in prisons all over the country, this Article maps the myriad overlapping conditions that systematically prevent people in prison from getting adequate sleep. Whether through luck, privilege, or their own affirmative choices, some people sometimes manage to get reasonably adequate sleep inside. But as my interviews show, most people in prison routinely sleep far less per night than the minimum seven hours that public health experts recommend. And for just about everyone living in prison, the quality of the sleep they get is consistently poor. In this Article, I offer a detailed account of why this is so.
To my knowledge, this study represents the first systematic effort to understand a destructive and dehumanizing experience that, although an inherent feature of prison life, has gone almost entirely unnoticed even by those academics, advocates, and policymakers invested in ensuring humane carceral conditions. Its conclusions are unambiguous: for a host of reasons explored in these pages, the incarcerated are chronically, perpetually sleep deprived—a condition they endure every single day, for years and sometimes even for decades.
The consequences of this situation are dire. Sleep deprivation compromises prisoners’ physical and mental health, making them age faster and die younger. It leaves people brittle, prone to frustration and anger, and without the cognitive resources to make good decisions, thus promoting conflict and violence in already volatile environments. And it strips them of the ability to think deeply and to exercise reasoned deliberation, capacities that are necessary for personal growth and healthy interpersonal bonds. Sleep is a basic human need, as vital to human survival and proper functioning as food, water, shelter, and personal safety. Yet persistent sleep deprivation—the product of what might be termed systematic sleep interference by the actors and institutions charged with administering carceral penalties—is a constitutive feature of life in custody. It heaps hardship, injury, and all manner of suffering on top of the formal punishment. And it represents a key, underappreciated reason why American carceral facilities are so noxious and inhumane.
Recognizing chronic sleep deprivation as a standard feature of prison life sheds new light on aspects of the carceral experience that have long seemed fixed and inescapable, including the high level of interpersonal tension, the pervasiveness of mental illness, and even the seemingly illogical willingness of people inside to put themselves into debt—a dangerous posture to adopt in prison—for the sake of a few candy bars or a bag of chips. In prison, moreover, residents are not the only ones unable to get enough sleep. Thanks to shiftwork and overtime (whether mandatory or voluntary), correctional officers (“COs”) too are chronically tired. In short, in the often tense and high-octane environment of the prison, many if not most people in a facility at a given time, staff included, are wrestling with the debilitating effects of persistent sleep deprivation. This shared state of impairment cannot help but impact the functioning of the institution and poison interpersonal interactions between and among residents and staff.
In future work, I will explore at greater length the implications of pervasive chronic tiredness on the character and operation of the modern American prison. I will also map the occupational health and safety effects on COs of a work schedule that often precludes anything close to sufficient sleep. Here, I focus on the basic phenomenon of sleep deprivation as a constitutive feature of the prison experience for those we incarcerate—a feature that, I argue, is as central to the punitive character of a prison sentence as is grossly inadequate medical and mental health care, the use of solitary confinement, the risk of physical and sexual assault, and the threat of excessive force by COs.
Two additional introductory points: First, this project focuses primarily on prisons, where time served is measured in years and decades. Yet sleep deprivation is also a problem in jail. If anything, it is worse. All my interview subjects did jail time before winding up in prison. And they universally reported that, however hard it was to sleep in prison, it was exponentially harder to sleep in jail. In jail, people may be detoxing. They are on average younger, more impulsive, and generally lack familiarity with the code of respect that can prevail in prison among people with long sentences. The constant churn means that people are always surrounded by strangers, a situation certain to exacerbate fear. The majority of those in jail at any given time are pretrial and understandably anxious about their cases. There is no programming to speak of, and mental illnesses will mostly go untreated. The combined result of these dynamics is a cacophony of noise and disruption in the housing units that, I am told, never stops. Among the troubling implications of this situation is that people detained pretrial routinely go to court exhausted, even to the point of falling asleep at counsel table during their trials. For most people, a stint in jail is relatively short. But even short periods of insufficient sleep can take a toll on physical and mental health. And for many people, a jail stay can last months or even years. For these reasons, much of what I say in this Article about sleep in prison should be taken to apply equally, if not more, to the context of jail.
Second, that COs are also chronically sleep deprived equally demands our attention. As a basic ethical matter, we ought to care about the experiences of COs, fellow human beings tasked with doing an impossible and often traumatizing job under difficult and at times dangerous circumstances. The least we can do is ensure that they have the opportunity to get adequate sleep, and thus avoid the harms that may come to them and their families and communities if they do not. That staff routinely get insufficient sleep is also extremely salient for those living behind bars. When staff are sleep deprived, prisoners suffer. As noted, in future work, I will investigate sleep deprivation among staff both as a matter of occupational health and safety and for its impact on the prison environment. I flag this issue now so that, by focusing here on the experience of sleep deprivation among the incarcerated, I do not create the impression that I imagine the problem to be exclusive to prisoners.
The argument of this Article proceeds as follows. Part I offers a short overview of the sleep science, focusing on the physiological and psychological harms and cognitive impairments traceable to insufficient sleep. Part II briefly describes the study parameters and methodology. Parts III and IV represent the sociological heart of the Article, offering a thick description of the experience of trying to sleep while incarcerated. Part III describes the impact of a set of concrete conditions that systematically prevent people in prison from getting adequate sleep: uncomfortable beds, hunger, extremes of temperature, noise, and excessive light. Part IV identifies a set of what I call “meta-conditions”—specifically, fear of violence, trauma, poverty, overly-intrusive rules enforcement, and daily humiliations. These conditions, which likewise routinely compromise sleep, are also products of institutional choices, but they have been so naturalized to the life of the prison as to leave us blind both to their destructive effects and to their ultimately contingent character. Part V begins to explore the normative implications of the phenomenon identified here, including what it means for our understanding of prisons and of carceral punishment, the prospects for Eighth Amendment conditions claims grounded in sleep deprivation (and, very briefly, for claims of torture under international human rights law), and the policy challenges likely to confront efforts to address the problem. On each of these topics, there is much more that could be said. The goal here is to start the conversation.
I. A Brief Overview of the Sleep Science
Over the past twenty-five years, the field of sleep science has exploded. Although the discipline is still developing, more than two decades of research strongly attests to the profound negative health effects—physical, psychological, and cognitive—of less-than-optimal sleep. Total sleep deprivation—enforced wakefulness over days—is “understood to be particularly detrimental.” But interference with sleep need not be total to compromise physical and psychological health. Even people who regularly sleep five to six hours per night—considered “short sleepers” in the literature—exhibit notable adverse health effects, in some cases after only a few nights of restricted sleep.
Perhaps the most jarring finding to date has been confirmed in three separate meta-analyses. Encompassing studies comprising 1.32 million, 3.58 million, and 5.17 million subjects, respectively, all three investigations found short sleep duration to be associated with an increased mortality risk of 12–13%. No single driver of this mortality effect has been identified. Instead, the data shows significant associations between short sleep and a host of serious medical conditions, including heart attack, hypertension, cardiovascular disease, coronary heart disease, and stroke. Short sleep and sleep disturbance have been associated with an increased risk of developing Alzheimer’s disease, as well as a host of metabolic disorders, including a heightened risk of insulin resistance, obesity, metabolic syndrome, and Type 2 diabetes. Insufficient sleep has also been found to cause imbalances in the autonomic nervous system, the indicia of which include increased heart rate, elevated cortisol levels, decreased heart rate variability, and increased levels of norepinephrine. Among other adverse physiological effects, this collection of symptoms increases the risk of cardiovascular disease and represents a predictive marker for “sudden cardiac death.”
Sleep deprivation has also been found to generate adverse psychological effects. For example, sympathetic bias in the operation of the autonomic nervous system has been linked with conditions indicating “emotional dysregulation, such as anxiety, depression, and rigid attentional processing of threat.” Persistent sleep disturbance is consistently associated with a heightened risk of developing major psychiatric disorders, including depression, anxiety, and suicidality. Concerning depression, one major meta-analysis, incorporating research from twenty-one qualifying studies, found that “nondepressed subjects” who experienced “difficulties in initiating/maintaining sleep or non-restorative sleep” have “a twofold risk to develop depression, compared to people with no sleep difficulties.” Research also indicates a greater likelihood of mental health issues among those who experience sleep disturbance over time. One early study found that respondents with persistent sleep disturbance “had significantly higher rates of new cases of both major depression and anxiety disorders compared with those whose insomnia resolved.” Other studies have begun to map the link between sleep disturbance and the symptoms of bipolar disorder. There is even some evidence suggesting that sleep disturbance increases proneness to substance use disorder.
Inadequate sleep also compromises cognitive functioning—and the more sleep deprived a person is, the greater the cognitive effect. Multiple well-controlled studies have demonstrated the impact of sleep deprivation on the capacity for vigilance, working memory, and sustained attention, with one experimental study finding that “chronic restriction of sleep to [six] h[ours] or less per night produced cognitive performance deficits equivalent to up to two nights of total sleep deprivation.” Sleep deprivation also slows reaction time and compromises performance in the completion of cognitive tasks. When sleep restriction is chronic, these cognitive defaults are cumulative, and when people routinely sleep less than seven hours per night, the resulting cognitive impairments are as marked as in those experiencing “severe acute total sleep deprivation.”
Finally, sleep deprivation impairs one’s ability to successfully navigate stressful or frustrating situations. These effects are evident after just one night of inadequate sleep. As Andrea Goldstein and Matthew Walker put it in their survey of the field, even “one night of experimentally controlled sleep loss increases subjective reports of stress, anxiety and anger in response to low-stress situations and increases impulsivity towards negative stimuli.” Insufficient sleep is also “associated with a lack of flexibility and capacity to respond to emotional challenges,” and thus with a decrease in emotional resilience. Together, these emotional regulatory failures undermine the capacities necessary if one is to get along with others. In one study, researchers showed that sleep deprivation was “associated with increased outward expression of aggressive responses and a greater tendency to assign blame to others when confronted with frustrating situations.” Subjects tended towards “antagonistic, self-serving, and uncooperative behavioral responses, such as increased blame and hostility, reduced willingness to make concessions and restitutions, and a general increase in atypical or impulsive responses, all of which are antithetical to harmonious interpersonal relations.”
The aim here is not to provide a comprehensive account of all relevant findings related to the negative health effects of sleep deprivation. The field is too vast to make such a goal workable. The point is rather to make clear that interference with sleep is not simply a matter of losing access to the “creature comforts” widely considered inconsistent with carceral penalties. Chronic sleep deprivation carries an increased risk of physiological harm, early mortality, and a range of psychopathological effects. It takes a toll on cognitive functioning and on a person’s capacity for prosocial interpersonal interaction. To the extent that prison systematically impedes a person’s ability to sleep over months, years, and even decades, this feature of a prison sentence can inflict substantial long-term damage. This brings us to the central question of this Article: How much and how well do people sleep in prison?
II. The Study
In prison, it is virtually impossible for people to get the recommended minimum seven hours of uninterrupted sleep. On most nights, most people inside will get only four to six hours of light sleep punctuated by frequent interruptions—a pattern that persists night after night, for years.
This is the unmistakable conclusion that emerged from over seventy‑four hours of interviews I conducted in 2023 with thirty-nine people formerly incarcerated in American prisons nationwide. Of these thirty-nine interviews, thirty-five were formal, semi-structured interviews based around a questionnaire with roughly 200 questions, with the average interview lasting just under two hours. Subjects were asked about their experiences of sleeping/trying to sleep in prison, the factors that may have prevented them from getting adequate sleep inside, and the ways their own persistent sleep deprivation—or that of staff or other residents—may have affected their daily lives while incarcerated. In addition, I conducted four somewhat more freeform on-the-record conversations (labeled “ad hoc” or “AH” interviews to distinguish them from the more formal interviews) with subjects who were outside the study parameters for the formal interviews, but whose insights and experiences promised to provide valuable perspective and context. During these conversations, I asked many of the same questions I asked during the formal interviews. In the end, much of what I heard in all my interviews, ad hoc interviews included, proved remarkably uniform. As a result, I wound up drawing on all thirty-nine interviews when analyzing the data.
With one exception, each interview subject had spent at least four years in prison. The average length of incarceration among my interviewees was thirteen and a half years, with fourteen individuals in the sample having served more than twenty years and eight serving more than twenty-five years. In terms of jurisdictional diversity, eighteen interviewees served time in four states: California (five), New Jersey (five), New York (four), and Louisiana (four). Between them, the remaining twenty-one participants had experience of prisons in fourteen other geographically diverse states plus the federal Bureau of Prisons.
Everyone I interviewed was incarcerated in multiple prisons over the course of their confinement. This broad experience proved valuable in two ways. First, it allowed me to learn about conditions in many more facilities than I had interview subjects. Collectively, my interview subjects did time in at least 185 separate institutions, including 126 state prisons, 18 federal prisons, 33 jails, and 8 youth detention centers. By speaking to thirty-nine people, I was thus able to hear firsthand about conditions in 8.5% of all prisons in the United States, including 8% of state prisons and 18% of federal prisons. Second, it meant that virtually everyone was able to compare conditions among institutions. Although the interviews revealed some notable regional variation, perhaps most striking was just how nationally uniform were experiences bearing on sleep—with any jurisdictional differences for the most part proving more a matter of degree than kind.
In terms of race, my interview sample self-identified as follows: 15 Black (38%), 13 White (33%), 7 Hispanic (18%), 1 Asian (3%), 1 Native American (3%), and 2 mixed race (6%). The goal was to replicate as nearly as possible the racial distribution of the American prison population, and in the end, I came close. As for gender, four of my thirty-nine interview subjects were housed in women’s prisons, comprising 10% of the total, with the remainder doing their time in men’s prisons. These numbers also came close to the gender breakdown of the American prison population overall.
It would be impossible to reproduce in full the richness of the narratives the interviews yielded. Instead, in what follows, I distill the key points that emerged around the experience of sleeping in prison and the conditions that impede sleep. In part, the power of the methodology employed—the long-form, semi-structured interview—is the degree of confidence it can afford in the accuracy of what one hears when the same thing is reported by multiple subjects in a diverse sample. On this measure, confidence is warranted as to many of the topics covered in my interviews, which yielded a high degree of consistency despite the diversity of the sample across, among other factors, race, gender, jurisdiction, housing configuration, and security level. Where my interviews revealed variation as to a particular issue or experience, I indicate as much in the text.
To my knowledge, this is the first study to attempt to capture what it is like to try to sleep in prison in the United States. The goal here is to begin to excavate a phenomenon that has to this point eluded scrutiny, despite the centrality of sleep—and, it appears, sleep deprivation—to the American carceral experience. Inevitably, some relevant aspects will remain unaddressed. One central contribution of this work is to provide an initial account that may be built out through future research.
III. Findings I: Carceral Conditions Impeding Sleep in Prison
In this Part, I describe specific, discrete conditions persistently impeding the ability of people to sleep in prison. Representing features of the physical plant (e.g., lighting, HVAC, beds, and bedding) and institutional operations (e.g., housing configurations, methods of conducting count and security checks, mealtimes), they are of the sort typically considered appropriate targets for more conventional policy reform. Of these sleep-compromising conditions, two categories—beds and bedding, and food and hunger—appear from my interviews to be experienced across the board by pretty much everyone, regardless of security level, jurisdiction, or any other factors. A further set of conditions—those salient during periods of extreme heat or cold—impact people differently depending on region and time of year. Finally, two other issues—noise and excessive light—are experienced differently depending on a range of factors, including housing configuration, security level, staff behavior, and the happenstance of the unit in which one is housed and where one’s bunk is located.
In what follows, I explore each factor in turn, describing how each impacts the ability of people to sleep in prison. But this approach carries a risk, as it makes it seem as if the culprit is exclusively poor system design. To some extent this is so, and for this reason, implementing changes on the ground is one obvious policy response to the findings reported here. At the same time, this focus on concrete institutional arrangements risks obscuring a deeper, more pernicious truth, one that I address in more detail in Part IV. There, I explore a second set of carceral conditions—termed “meta-conditions”—which more directly reflect the prison’s toxic moral foundations and COs’ consequent blindness to the humanity of those in their custody. If prison officials cannot see the people in the beds at night as human beings who need sleep to survive and properly function, they will not act, or run the prisons, in ways that protect and promote adequate sleep. And sure enough, in American prisons, they do not.
A. Uncomfortable Beds/Bedding/Mattresses
In prison, the construction and nature of the beds and bedding alone would be enough to keep most people from getting decent sleep. The beds are rock hard—typically just a metal or concrete slab—and so narrow and short that people who are taller or heavier than average often will not fit. “Mattresses” are nothing more than thin pads covered with a rubbery plastic that is sticky in the heat and cracks in the cold. But equally if not more disruptive of sleep is the construction of the bunkbeds, the main sleeping configuration for people in prison. Bunkmates feel and are disturbed by every slight movement, which, depending on how people get along, can make sleeping a stressful experience for those afraid to spark conflict simply by turning over. For this reason, or simply out of respect for a bunkmate’s sleep, those on top bunks may hesitate to leave their beds at night even to urinate. Not only do those who try risk waking frustrated bunkmates who are themselves desperate for sleep, but the high likelihood of disrupting their neighbor’s personal property as they try to get in and out of their bunks in the dark while half asleep generates further stress and potential conflict. None of this is conducive to enabling sleep.
In prison, the beds are “rock hard.” For the most part, the design is the same: metal frames with a solid concrete or metal pallet—a “large . . . piece of sheet metal with four legs bolted to it.” They are also narrow, so “[y]ou[’ve] got to train yourself to sleep in this small space and not roll over . . . onto the floor.” People who are bigger or taller than average have an especially hard time. For those who are overweight, it can be “hard to turn over . . . because the beds are small.” One person, who at one point weighed 330 pounds, described how he needed to “lift up and flop” to turn over “[b]ecause you can’t roll. If I roll, I’m going [to wind up on] the floor.” Those who are tall have a different problem. One especially tall person “had to sleep on the bottom bunk with his feet on the toilet.” Another “big guy,” the cellmate of an interviewee, “had to sleep [in] reverse [of] the normal way so his feet could hang off the edge of the bed.” To do this, “he had to tuck his head underneath the shelf” which would ordinarily be at the foot of the bed.
Were the mattresses sufficiently thick and springy, the concrete or metal pallets might not cause problems. This, however, is not the case. Each time I asked about mattresses, I was treated to the same visual representation, with people using thumb and index finger to indicate roughly three to four inches, or five inches at most. And this was when mattresses were new. As one person explained, “at one point they were about four or five inches thick, but then . . . as time[] went on, the middle part bec[ame] smaller . . . [and] at some points, it can become nothing more than two pieces of fabric.” The thinness of the mattresses was an issue for everyone: “you’re sleeping pretty much on metal,” and “you’re gonna feel that iron underneath you every frickin night.” Many people reported experiencing perpetual discomfort, with each night an ordeal of tossing and turning. The hardness of the mattresses was “difficult on your back, so you wind up with back problems” or “sciatic nerve damage.” In some cases, people reported such problems persisting after release.
The mattresses’ material also causes problems. The covers are a plastic, rubbery substance to which bodies stick, especially on hot nights. One person described preferring to “sleep on the middle piece of steel,” because on the mattress he “would feel the heat and the sweat.” Even on temperate nights, the rubbery material was unpleasant to the touch. One could cover the mattress with a sheet, but the “vinylized rubber” material “will [still] make you very hot and sweaty.” And as the mattresses age, “the plastic [gets] all cracked and ripped,” and “the person who slept in it before you [was] sweating [so] it gets moldy—the foam holds the mold, and you can smell it.” Replacing bedding, including mattresses, is at the COs’ discretion. As one person put it, “[y]ou could request a new mattress, and depending upon the officer . . . on duty at the time, he’ll put your request through, or he’ll just toss it into the garbage can or run it through the shredder.”
Blankets are generally thin: “[S]ummer blanket[s]” are “the lowest quality of . . . cotton”—“more like fish nets, . . . maybe crocheted or something but . . . they weren’t solid, so air can go through it.” Whatever the ambient temperature, people tend to be allocated one blanket only. As for pillows, most prisons appear not to supply them. In many places, the best one might get is a pillowcase, which allows for handmade pillows stuffed with “hoodies and sweatpants.” This approach does not afford a comfortable sleep. If pillows are supplied, they “could be hard as a rock or . . . flat as a pancake.” As one person with a gift for metaphor put it: “[T]he pillows you’re sleeping on . . . might as well stack a couple sheets of paper together and put your head on that.” In some prisons, the mattresses came with “a little bump at one end that was supposed to be a pillow. . . . [A]n inch high or something.” When I asked if that did the trick, people just scoffed.
Then there is the fact that people largely sleep on double bunks that share a metal frame. It is hard to overstate how much this arrangement compromises sleep. When the beds are connected this way, you feel every movement the other person makes. Being “crummy old metal bunks,” they “move and squeak a lot,” which means “you’re going to wake up when your bunkmate gets up or even just moves around.” When people put their weight on the bed, it is not unusual for the flat metal of the pallet to dip slightly and then make a loud pop when the weight is lifted. People described strategizing ways to shift position in bed that would not make noise, and even staying put rather than, say, getting up to urinate, out of a desire not to disturb their cellmate. Even the ordinary movement of a restless sleep could prevent a bunkmate from sleeping. One person described an instance when he was in terrible pain from a toothache, “I would wake up in the middle of [the] night, I’m like [makes a face like in pain, moving head around], but I would make sure I wouldn’t toss and turn too much [out of] respect for my cell[mate] . . . so he gets his sleep.” Disturbing a cellmate or neighbors in the dorms at night can be a serious source of friction “[s]o it was very important to be quiet.” The burden on sleep here is twofold: people hesitate to make adjustments that would help them fall (back) asleep, and they experience stress at the thought of disturbing a frustrated, sleep-deprived cellmate. The physiological effects of this stress alone can cut against achieving the relaxed state that sleep requires.
In some prisons, bunks are not attached to one another but only to the wall. Although these bunks “weren’t as sensitive, they definitely still connected through the ladder and everything. So . . . one hundred percent you could feel it” when the other person moved in bed. And with beds connected to the wall, the bed of “the person in the cell next to you” may be connected through the same wall, “so when that person moved, your bed moves.” If that person happened to be “a very deep and robust snorer . . . you could actually feel the wall and your bed vibrate when they snore[d].”
Whatever the structure of the bunks, the person on the top bunk is typically forced to find some creative way to get in and out of bed. There are small ladders attached to the beds—really just “a couple of rungs attached to the bunk”—but if you use them, “your bottom bunkmate is going to feel the bed move.” So people descending from top bunks may instead step on the desk, commode, or other available surface, which might lead them to jostle or even displace the property of their cellmate. The risk this operation entails is particularly elevated in the middle of the night. For this reason too, those on the top bunks might forbear from getting out of bed at night to relieve themselves, even if the need were pressing, simply to avoid provoking conflict. Bottom bunks are widely preferred, although people on bottom bunks also need to take care that in moving, they don’t disturb their bunkmates. One person reported that “in the middle of the night, when [he would] go to turn over, if [he] was on the bottom bunk, [he would] make sure to keep [his] hands down low, so [he] won’t accidentally hit [the top bunk].” And bottom bunks present their own issues. More than one person described the jarring impression of waking from sleep to find a dark shadow looming over them. By the time they realized it was just their cellmate using the commode, they were wide awake, the effect of an adrenaline rush sparked by fear.
Just about everyone tries in some way to increase the comfort level of their sleeping arrangements. The most common approach is to angle for a second mattress, blanket, or pillow, usually by being the first to snag one left behind by someone being transferred or released. But in every facility, having anything beyond what was issued by the prison—generally one mattress, one blanket, two sheets, and possibly a pillow or pillowcase—is a clear rules violation and could result in a write-up. In many cases, COs won’t bother writing the ticket, but they will likely confiscate the extra items—“if you get caught with doubles, they’ll take it.” Those who manage to procure an extra mattress might pursue a more effective (if more audacious) approach: stuffing one mattress inside the other. As one person explained, “[T]here’s a seam on the mattress, so I unstitched it and slid the other one in it and then sewed it back up.” These measures could make a difference on the margins: after a while, one’s “mattress would be like pancakes. . . . [T]here’s a difference between two mats and one mat for sleeping.” Still, the difference is only a matter of degree, and even doubled up, “it’s still bad.” As one mattress-doubler put it, “I was limiting the amount of damage that the regular mattress was doing to me by a small percentage. But when you’re gasping for air, you’ll take any air you can get.”
B. Hunger
Three basic aspects of food in prison contribute to the hunger that, according to my interviews, universally impedes sleep: quality, quantity, and timing. The quality of the food is so poor that people often have to force it down and will often opt not even to try. The quantity is frequently insufficient to induce satiety. And most significantly, the timing of the meals is such that there is typically a gap of at least five or six hours without food between “dinner” and lights out. Every single person I spoke with affirmed that if you have no money on your books for food from commissary, you will go to bed hungry, a condition that makes it difficult if not impossible to sleep. Those with family support will usually have food to eat before they go to sleep. The rest scramble to make money with side hustles (e.g., cleaning cells, making and selling food, doing legal work), or resort to less salutary strategies that could put them at risk (e.g., stealing, gambling, selling sex). The desperate desire not to go to bed hungry appears to be a driver of some of the most dangerous and pathological behaviors adopted by people in prison.
The other obstacle to sleep universally attested to in my interviews relates to food. Together, three food-related issues—quality, quantity, and timing—ensure that those able to eat only what the prison serves will routinely go to bed hungry, a condition that can make it hard to fall or stay asleep. First, the quality of the food served in prison is notoriously poor and often inedible. One person I spoke to, capturing the general sentiment, described prison food as “garbage, . . . like what they feed a pig. If you put [slop] in a bucket and throw it in a pig’s face . . . [t]hat’s what it looks like.” So unappetizing is the food served that those with other options will choose to forgo the prison fare altogether. The quantity is also frequently insufficient, and, as a result, even people who eat everything served to them still may not get enough to feel satiated. One person reported losing “thirty or forty pounds just on the food. . . . It was just very, very, very small portions.”
But in terms of impact on sleep, the most consequential issue is the timing of meals. At some point, I began asking when meals were served. As to breakfast, answers varied from as early as 3:00 a.m. to as late as 8:00 a.m., with the majority reporting breakfast times between 4:00 a.m. and 7:30 a.m. Of still greater significance in terms of sleep is the timing of dinner. Here, answers ranged from 2:00 p.m. to 6:30 p.m., with the overwhelming majority reporting a dinner time between 4:00 p.m. and 5:00 p.m., or earlier. In most prisons, people appear to be locked in for the night around 9:00 p.m. or 9:30 p.m., with the lights dimmed for sleep around 10:00 p.m. Even those who go immediately to sleep at 10:00 p.m. will have gone at least five hours without eating, and for those who stay up later, the delay without food will be correspondingly longer—what one person called “intermittent fasting before it became fashionable.”
Those with the good fortune to work in the kitchen have “access to extra food left over.” Otherwise, thanks to the timing of dinner, “[a] whole lot of people went to sleep hungry.” To avoid this situation, people try to have food on hand to eat before going to sleep. In facilities with a central dining area, although “you’re not supposed to,” people would “bring the food out [of the chow hall] . . . . Might be rice and beans. I’ll bag that up and bring that back.” In facilities where food is served in cells or day rooms, people might keep back food from their trays. But in many prisons, being caught with prison-issued food in one’s cell is a rules violation and “sometimes, you’re going to get caught.” When that happens, even if you are not written up, the food will be confiscated, leaving the problem of nighttime hunger unaddressed.
There is, however, one thing that allows people to stave off hunger: access to commissary (also known as “canteen” or “store”). In every prison, people can purchase a range of items from what is in effect a virtual prison shop, which typically sells a mix of food, personal hygiene products, and other miscellaneous items. Each facility has an approved vendor, and people fill out forms indicating what they want to buy, with the funds coming out of the purchaser’s prison account. As in society in general, there are all kinds of reasons to want money in prison. But judging from my conversations—which again, involved people from all regions of the country, diverse as to race, gender, jurisdiction, housing configuration, and security level—the single biggest reason is that having money or its equivalent allows you to get the food you need so that you do not go to bed hungry. As one person recalled, “When I first [got to prison], I didn’t have as much money coming to me. So there were nights where I would go to bed starving. And you just couldn’t sleep because of how hungry you were, or [you would] wake up early waiting for breakfast . . . just because you’re so hungry.” People described “nights that you would be tossing and turning because your stomach was hurting” from hunger; “nights of not having food,” when “you [were] so hungry, you[’d] take your hands and push in on your stomach to make [it] tighter, and try[] to fall asleep like that”; and nights of “being so hungry to the point where you can’t even sleep.”
In prison, commissaries typically price goods close to or even higher than market rates, yet people who work in prison will earn well below $1 an hour. A key determinant of a person’s ability to buy items from the store is thus whether someone on the outside is putting money on their books or whether they have some other way to make money while inside. I spoke to one person who had a relatively well-paying job with a private company run with prison labor. But in most cases, people who received no funds from family found some side hustle in the prison. People would “make and sell wine,” “homemade . . . cards,” or “prison burritos” crafted from ingredients bought from commissary or kitchen workers and sold for a markup. Or they might “wash other people’s clothes” or “bowls [as a] kind of dishwasher,” “sell their phone time,” “do legal work for people,” or “type papers” for those in school. One person I spoke to “fixed all electronics like headphones, fan, Walkman, TV, radio, anything. . . . Even the officers would bring [him] broken fans and watches and stuff.” Like others with side hustles, he took his payment “[i]n commissary, in food.”
Those without family support or some effective way to earn money found other ways to get food or the money to buy it. One person I spoke to, who “didn’t have [any money] for a long time, . . . started gambling and stealing . . . to provide for [her]self.” People “sold drugs to make ends meet” or exchanged “sex for food” or “g[o]t jobs in the kitchen so they could steal.” Others would “panhandle, ask your neighbors, your buddies,” or they might resort to a more general appeal:
[T]here were people who had no money and they would stand out by where we get our trays. And they’re asking people, you know, “Oh, do you want the rest of that?” and they’re collecting it, and putting it into a bowl to try to have what they need for that night.
However one does it, the main goal is to have provisions available so that, when 10:00 p.m. rolls around, a person has something to eat before they try to sleep.
Even people with the resources for commissary may sometimes go to bed hungry. People may run out of food before the next canteen day. Or they might have had extra expenses that ate into their stash for the month. COs might have tossed their cell, and in the process rendered whatever food they had inedible—tearing open packages, contaminating their provisions, and so on. A rules violation sometimes carries the penalty of loss of canteen privileges. People are often transferred between facilities, and it can take time for the finances to catch up, which can mean missing several cycles of commissary while you wait. One person reported being transferred to “a very restrictive facility,” where “you come in [and] you can’t get commissary for the first thirty days.” And perhaps most notably, access to property is contingent on remaining in general population. When people are sent to solitary (a.k.a. administrative segregation or “ad seg”), whether for discipline, protective custody, or suicide watch, they are not permitted to bring any property with them and are highly restricted in what they can get while they are there. Generally, in ad seg, people “don’t have any access to commissary,” so people are completely dependent on whatever food is served. As one person explained, people in ad seg will “get fed [dinner] around 3:30 or 4:00 [p.m.], and they don’t get anything [else] [un]til about 6:30 [a.m.].” To make matters worse, in ad seg, you can’t even “save food from your evening meal to try to hold it [t]o have something to eat in the morning or in the middle of the night” because doing so is a rules violation—“if they find you with that, then they add extra time to the time that you stay in seg.” Whatever the obstacles to accessing commissary, when people are unable to buy food to supplement prison-issue meals, they go to bed hungry, which makes it hard to sleep—even assuming a sleeping environment that is temperate, quiet, and dark, which, as will be seen, prison environments frequently are not.
C. Extremes of Heat and Cold
To get adequate sleep, the body needs to be at ease. Yet for months every year, people in prisons nationwide are forced to try to sleep in sweltering heat or freezing cold. Predictably, this leads to conflict. Men in dorms fight over where to direct the few available fans—battles generally won by those whose threats of violence are most credible. People struggling to get comfortable in their bunks toss and turn, drawing the ire of bunkmates whose own efforts at sleep are thereby disturbed. And cellmates with different sleeping preferences clash over whether to close windows or block vents. It would be too simple to put this down to changing climate. The ambient temperature in carceral facilities is an operational choice. Although average annual temperatures continue to climb, most prisons have no air conditioning in the housing units (even while providing air conditioning in the spaces where COs congregate, including staff lounges and officers’ booths). And in the coldest months, poorly maintained physical plants—broken windows, ineffective or broken heating systems, and so on—leave people too cold to sleep.
In prison, the ability to sleep is greatly impacted by the ambient temperature. In society in general, people who have control over their sleeping environments will adjust their thermostats to a comfortable setting. When two or more people share living space, there may be competing preferences as to the best temperature for sleeping. But it is rare that, given the choice, people would opt to sleep in the freezing cold or in sweltering heat. The reason is obvious: to get adequate sleep, the body needs to be at ease. In conditions of extreme heat or cold, the physical relaxation necessary to fall and stay asleep becomes unattainable.
Apart from those few fortunates who did their time in facilities with air conditioning, functional heating systems, and well-maintained physical plants, my interviewees consistently reported long stretches each year when extremes of heat or cold greatly impeded sleep. Consider first excessive heat. Multiple interviewees from across the country described enduring many months every year when it was too hot to sleep, when it got “so hot and humid that the walls are sweating [and] [t]he floors constantly have that dewiness on them,” when you “sweat so much, [i]t wears you out,” “your shirts would be dripping wet. You actually could wring your shirts out.” As one person succinctly put it: “In the summer, it’s so hot, it’s like you’re in a blowtorch.”
Because heat rises, those on the top tiers suffer most. One person described “working in the infirmary,” and “hav[ing] to respond to guys that were living up there. And as you walked up five flights of stairs, it was like . . . [you were] in a sauna all of a sudden, . . . like [a] forty to fifty degree temperature change sometimes. It was ridiculously hot.” In the summer months, the dorms can also get “exceedingly hot”: “If it’s 100 degrees outside, it’s gonna be 110 degrees in the dormitories.” When it was this hot, no one was sleeping. “[All that] body heat. . . . Everybody’s tossing and turning.” One person I spoke to described being “up all times a night.” He would “get up soaking wet. Go to the bathroom, get a cold towel, wipe my skin off. Go lay back down. Wake up an hour and a half later [and] can’t sleep.” Even outside the South, this situation was hardly rare; one person in a federal facility in the Northeast described being “super-duper hot and miserable” during the height of summer. In the South, heat this intense could last six months or more each year.
Predictably, this degree of discomfort sparks frustration and even threats of violence. In the dorms, prison officials sometimes set up industrial-strength fans, but their placement and direction can provoke anger and even lead to fights as people desperate for sleep struggle to get or retain access to the relief the fans provide. As one person described it, “[e]verybody’s fighting for the fan. The fan has been cocked this way, cocked that way. You got guys waking up in the middle of the night, ‘Man, don’t turn that fan no more,’ which caus[es] a fight [and] now everybody waking up. ‘Man, listen, leave the fucking fan alone.’ ” If tensions did not run as high in the cells, periods of extreme heat still sparked issues. In the heat, people who are unable to sleep find themselves shifting in their bunks trying to get comfortable. The mattresses, as we have seen, are typically wrapped in a kind of rubber to which bodies would stick in the heat. And as we have also seen, when people are double-bunked, every move one person makes can be heard and felt by the other, especially when the bunks share a single frame. Out of respect, people try to minimize movement as much as possible. But the cost of this courtesy is being forced to remain immobile even amidst the great discomfort excessive heat creates, making sleep even more elusive.
It is not that prison officials are unaware of the heat. More than one person I spoke to described housing units so hot that staff avoided coming inside. One interviewee noted that COs would “sit outside in an air conditioned booth all night, and they would come inside of the dormitory, make their rounds, . . . and then go back into an air conditioned booth while we stayed in the dormitory.” The lack of air conditioning in Southern prisons in particular has already become the subject of constitutional litigation, with Eighth Amendment challenges to excessive heat grounded in the substantial risk of serious harm posed to people with certain medical conditions, the symptoms of which are exacerbated by heat exposure. In Ball v. LeBlanc, a case concerning excessive heat on death row at the Louisiana State Penitentiary (a.k.a. “Angola”), the district court appointed outside experts to monitor the temperature over three weeks in July and August. During that time, the heat index in the facility rose as high as 107.78 degrees Fahrenheit and remained over 100 degrees for a full week. The district court ordered the Louisiana Department of Corrections (“DOC”) to install air conditioning, but the Fifth Circuit vacated that part of the order. In 2018, the case settled, with the DOC committing to providing a minimum of fifteen minutes of shower time each day, ice machines and ice containers, fans, water, icy breeze units, and diversion of cool air from the guard’s pod to the area of the plaintiffs’ cells on any day when the heat index in the plaintiffs’ cells exceeds 88 degrees. Judging from what I heard, whatever else these half-measures might accomplish, they are unlikely to do much to improve the sleep of people in Louisiana prisons during the hot summer months.
Being too cold can also obstruct sleep. Virtually all facilities have heating systems for cold winter months, yet judging from my interviews, excessive cold remains a pervasive issue. In many cases, the problem stems from a poorly maintained physical plant. One person incarcerated in the Northeast told me that, in his prison, the windows had no glass, so “[w]hen it snowed, it snowed [directly into the] room.” It was like “sleep[ing] on a bench in Central Park in the wintertime.” Likewise, in one Midwestern prison, thanks to broken windows in the cells, people would wake up “with snow on [their] blankets.” In a prison in New England, “a lot of windows were broken out” and people resorted to using newspapers to cover windows and block the draft coming in under the doors. In one Southern prison, “the place was so old and rundown” that the “window closing mechanism . . . was just broken.” This meant that “in January and February, the windows were literally stuck open and it was against the rules to stuff blankets or towels in the windows . . . so it was freezing in there all the time.” In a facility on the West Coast, “you could almost see your breath in the air,” and in some cells, “you could even see ice forming on the wall.” One person in the Northeast reported having actual “icicles . . . forming inside the room.”
Plainly, under such extremity, it is very difficult to sleep. I asked one person how he slept in the cold he was describing. His answer: “You didn’t. . . . [W]e were bundled up—we put our sweatsuits on. You put [on] your full tans [i.e, the prison-issued uniform] and you’d have your jacket on with your hat and your gloves, and you’d sit there and try to get as warm as you could.” Others painted a similar picture: “[B]asically you’re laying there with this wind that’s blowing constantly through into your cube, and you’re just trying to stay warm. . . . Not very good [quality of sleep]. Because you’d wake up in the morning . . . frozen.”
As with extreme heat, conditions of extreme cold can spark tensions among people frustrated by their inability to sleep. One person who lived in a dorm in a Southern prison described
power struggles with the older generation [who] wanted the heater on [while] the younger generation didn’t. Sometimes . . . the older generation would win [and] the heater would stay on, . . . which would make it warm in there. So some guys would open their window, but the gust from the cold is coming through the window and [now] you’re cold. . . . [and you’re] trying to convince him, “Hey, man, I’m cold, could you close the window?”, which could lead to a fistfight.
Excessive cold was frequently a problem for people in ad seg. One person explained that, in ad seg, “you don’t get [any] blankets,” even though “it was freezing”—so cold, in fact, that he “had to cut [his] mat open and . . . go to sleep inside [the] mat.” This move did not keep him warm enough to sleep, but “it kept [him] a lot warmer than it would have been otherwise.” For those on suicide watch, the discomfort is intensified by the fact that they are typically forced to wear nothing but tear-resistant gowns known as “suicide smocks.” While these garments are made of durable fabric, they are entirely lacking in warmth. People on suicide watch are also denied standard blankets, sheets, and mattresses (although they may receive so-called suicide blankets). The justification for these deprivations is the need to prevent people contemplating suicide from accessing anything they might use to hang themselves. But the combined impact of these conditions leaves people too cold to sleep—a counterproductive effect, given that sleep deprivation appears to exacerbate suicidality.
D. Noise
Prisons are extremely noisy places. Although the nature and extent of the noise varies by housing configuration, having one’s sleep disrupted by noise is a standard part of the carceral experience. Sources vary widely. The regular nighttime soundtrack may include loud voices, music, televisions, toilets flushing, the alarm clocks of night workers, or people “screaming and hollering.” Dorms, housing as many as eighty or a hundred people, are especially loud and chaotic. Some housing units have an ethos of nighttime quiet, maintained by mutual respect (as in honor dorms) or threats of violent reprisal (as in some maximum-security cellblocks). But some pathological sources of noise are not so easily quelled. People with mental illness can be loud at all hours. Some screams may come from victims of violence. And by far the most persistent and resented source of nighttime noise comes from staff who move through housing units at regular intervals doing count or security checks. Some COs try to do their checks without waking people up. But too often, COs on the night shift conduct themselves as if completely oblivious to the fact that they are surrounded by fellow humans desperately trying to sleep.
Prisons are extremely noisy places. In the daytime, the cacophony can be deafening. Although the decibel level drops considerably once people are locked in for the night, the nighttime soundtrack is still far from peaceful. This is especially so in dorms, where as many as eighty or a hundred people may live together in one cavernous room. One person compared it to “being outside at a parade” or “on [a] construction site,” with “a lot of guys having a nightmare, screaming and hollering.” In a dorm “people are moving around, people are going to work even in the middle of the night,” and “almost never would there be a night with nobody shouting or nobody making noise.” There were times when “people would stay up all night, and they’[d] leave their radio [or TV] on kind of loud, . . . or somebody [would be] partying, drinking, and making a lot of noise, up all night with the music playing.” Although the rule—and the strong norm—is that headsets must be worn by anyone watching TV or listening to music, in practice, this rule is often ignored. If only one or two people in a dorm flout the norm, everyone’s sleep is disrupted. And even when people wear headsets, those close by may hear enough of the sound—“like a fly buzzing in your ear all night”—to make it hard for them to fall or stay asleep. One person mentioned sometimes using earplugs to try to cut the noise. But as he explained, “[E]arplugs are tricky in prison. It’s a little easier if you’re in a cell [and] especially fine if you’re in your own cell. But in a dorm, it’s not that comfortable to not know what’s going on around you.”
In some dorms, the bathroom is connected to the living space, so people whose beds are nearby may be disturbed all night by others going back and forth, or by flushing toilets and running water. As one person put it, “imagine eighty-seven people going to the restroom at all different times of the night. And sometimes it’[s] just one person going to the restroom starts a chain reaction, because they hear the water flow and everything. And so if you’re by the restroom, it’s not fun.”
In more orderly dorm communities, there may be less of a problem with other people making noise at night. In one dorm setting I heard about, “there was this common respect that everyone [had] to go to work call in the morning and if you make[] noise in [the] dormitory, somebody’s going to get up and say something and tell you something.” This ethos is especially evident in the honor dorms, where people do their best to avoid disturbing others, creating an environment that can be more “peaceful” and “laid back” with “less drama.” Yet even still, people invariably move around at night, especially older residents, who tend to be overrepresented in honor dorms and who may need to visit the restroom multiple times a night, creating frequent disruption.
A similar ethos of keeping noise creation to a minimum also appears to prevail in at least some high-security cellblocks housing people serving long sentences. In these units, there is a particular premium placed on displays of mutual respect, and making noise while others are sleeping is considered a strong show of disrespect. In such units, people may be quieter at night “because they don’t want to have to deal with the [violent] consequences.” Certainly, even in high-security facilities, extant norms will vary. And outside maximum security, no such collective enforcement is likely, especially when people may have “a release date . . . within five years” and are “more likely to be on [their] [best] behavior because [they] can see the door.” If others are disruptive at night, “somebody might say something to them [and] at worst it would be a fight. . . . But more likely, it would[] be nothing. . . . People don’t want to risk it by stabbing some guy for being noisy.”
Viewed normatively, threats of violence against noisemakers are hardly desirable. But where no anti-disruption norm operates, a cellblock at night can be extremely loud. One person described trying to sleep in a housing unit with five galleries—a hundred men to a tier—where “individuals [would be] yelling from one gallery to [another] all night.” Another spoke of “people . . . banging on the door, yelling through the doors, talking to each other.” Or “[t]here are guys who will have radio battles all night long.” And in “open tiers, . . . you take all that [exposure to people’s sound] and you multiply it, so there’s probably a small window of time at night where there is no noise.” Other sources of noise may add to the din. One person described “big industrial fans” in his housing unit that “would just blow and make so much noise.” Another spoke of air blowing through the vents so loudly that it sounded like “a jet engine going off.” Some nighttime noise in prison mirrors the ordinary soundtrack of life in any setting. You might hear “different things fall[ing] in the middle of the night” (which “is gonna ring”), “other people’s alarm clocks going off,” “their TVs when they’[re] up at night,” or—in facilities with “stronger flushes”—people flushing their toilets. This is only to be expected in places where people live. But multiplied by hundreds of people, the effect is constant, pervasive interference with sleep.
Other sources of nighttime noise reflect the deepest pathologies of American prisons. For example, even where some norm of respectful nighttime silence obtains, people with untreated mental illness may be unable to comply with or even process expectations around noise. As a result, those trying to sleep may find themselves disturbed by screams or other vocalizations coming from those unable to remain quiet. One person described being housed in her first months in prison with someone who was “schizophrenic and [who] was up all night, having . . . arguments with herself . . . like really heated arguments.” As a result, she “could not sleep.” And people with mental health issues need not scream or yell to disturb the sleep of those around them. Another woman spoke of how, in her dorm of 150 people, “there were the people who would get up super early and make noise . . . cook in the microwave, or get their breakfast ready, or be the first ones to shower.” These, as she explained it, were “people with mental illness” who “didn’t want to be bothered at 6:00 a.m. [when the lights came on]” and were rigid about “doing everything before [others woke up].” Although this was “an issue of inconvenience” to others, people “just learn[ed] to tolerate that stuff,” because “that’s how they are.” This realization may have alleviated resentment, but it did not help reduce the disturbance at a time when everyone else was trying to sleep.
Mental illness among residents is also one reason why some solitary confinement units can be so unremittingly loud. Judging from my interviews, ad seg can vary from extremely noisy to extremely quiet. Those who experienced the former reported that people would be “screaming [and] . . . kicking the doors,” or “yelling from door to door, talking to the homeboys down the hall, [and if] somebody thinks they can sing—they wanna sing.” “Everybody’s making noise, can’t nobody sleep.” If it was not the other residents, it was the doors: “[T]he [Secure Housing Unit] was always loud because the doors . . . are metal [and] they slam. Anytime anybody comes through—nurse, counselors, whatever—it’s always a bunch of noise.” As one person vividly described it, “There were these automated doors that would make this loud ‘ahhhh’ slam.” Depending on the location of your cell or who was working, “all night long, you[’re hearing] ahhhh clang, ahhhh clang, ahhhh clang.”
Then there are the sounds—the yells, the cries, the screams—produced by victims of violence. Based on my narrow sample, it is impossible to know how frequently this occurs. I did, however, hear it enough to credit that it happens. When asked about nighttime noise, people mentioned “cellies fighting each other,” or “[s]omebody might be attacking somebody [or] somebody might be screaming and hollering in the shower.” One person described being awakened one night by the harrowing sounds of a man who, it transpired, was being killed in a nearby cell.
But by far the most persistent, intrusive, and resented source of nighttime noise is that made by staff. In every prison, whatever the housing configuration, COs are expected to come through each unit multiple times a night—as frequently as every fifteen or thirty minutes—to scan each cell and bed and see that all is well. The intended purpose of these rounds (known as “security checks”) is to prevent assaults and suicides, to break up fights, to have staff available for anyone who may need medical attention, and so on. In addition, at certain set intervals, COs fan out through the facility to conduct “count,” an event occurring several times over the course of the day—and night. At these designated times, COs count every person in the facility to make sure the number of people inside corresponds to the official tally.
COs’ performance of these rituals means constantly interrupted sleep. Virtually everyone I spoke with recalled being woken up repeatedly at regular intervals by those staff members doing rounds who chose to conduct this task by banging on the doors or bars of their cells, kicking their bunks, or shining flashlights in their eyes, seeking (and sometimes loudly demanding) some indication from the sleeping person that they are still alive. People told of COs who would “deliberately pound on the door or flash the light in our face until we move,” or who might “take their stick, or . . . the chirp thing for their rounds [i.e., the wand some COs must use to electronically record the completion of mandated security checks], and bang it on the bar or bang it on the door until you move.” As one person described it,
[E]very two hours or three hours . . . the officers do security check[s]. . . . They can just kick your door, boom, boom, boom, say, “Move your legs! Move your legs! Let me see you alive!” . . . So any given day, the officers . . . will wake you up . . . . [And] if you sleep with a cover over your head or something, the police will hit your window, boom, boom, boom, “let me see something.”
Often, people are woken up long before the COs reach their cells—whether because the officers “just slam the doors” when they walk through, “walk around with their walkie-talkies or radios turned all the way up,” or “have a bunch of different big, humongous keys on their belt, on a keychain that jangle[s] all the time when they come by.” In prisons not generally designed for effective sound absorption, when COs “knock on someone else’s door . . . you can hear the echoes from the whole hallway. . . . So you have to cover your ears because there’s always [constant] noise.”
In the dorms, in addition to flashlights in eyes, the kicking of bunks, and the shouts of “let me see you move,” several people described CO behavior reflecting seeming obliviousness to the sleeping people around them. They might “stand by your bunk and just start talking in the radio . . . extra loud,” or “bang[] [keys] on that corner of the bed [to wake someone up], . . . [which] pretty much wakes the whole dorm up.” Or a pair of COs “may just come to the dormitory talking to one another. One would be on one side of the dormitory and the other one would be on the other side. And they’re screaming across people while they’re making rounds.” One person reported that, rather than counting quietly, some staff would yell out the numbers: “[yells] 12345678!” As a result, “you could never get a deep sleep there . . . because there was always noise.” And it is not only during rounds that staff on the night shift conduct themselves in ways that interfere with sleep. People also described COs in the officers’ booth laughing and joking without making any effort to keep the noise down. They would be in “their bubble, and they’d have a [fellow officer] come over, and then they [would] talk . . . so loud that you can hear them inside the dorms, you know, and they’re telling the funniest stories [and] [a]ll you hear is them laughing.”
In fairness, when officers conducting security checks insist on evidence that a person is present and alive, they are only following policy. As one CO I spoke to explained,
For me to be able to do my job confidently, I have to see flesh or body movement. So . . . that requires me . . . to knock on the door and wake them up. “Hey, are you alright?” “Yeah, man, can’t you see me?” “No, your bed sheets are up. I have to be able to see you, man. It’s nothing personal. I’m just trying to do my job.”
And notably, COs’ intrusive behavior on the night shift may also arise in part from their own struggles with chronic exhaustion arising from sleep deprivation. As my research has shown, COs too are often sleep deprived, and this is especially true of those on the graveyard shift (typically 10:00 p.m. to 6:00 a.m.). The resulting fatigue appears to drive at least some of the behaviors that disturb prisoners’ sleep. For example, “most [COs] are constantly talking to [co-workers] to get them through the night.”
Even still, the degree of disruption repeatedly described by my interviewees appears excessive, even accounting for institutional requirements and COs’ own exhaustion. Some considerate COs prove this point nightly by making affirmative efforts not to wake people up while doing rounds (with the proportions varying widely depending on the facility), thus showing it is possible to do so. These courteous COs do not yell or bang the bars or kick the bunks or shine their flashlights directly into people’s eyes. They instead “hold their keys as they’re doing count and . . . put the light up toward the ceiling”—using the ambient light to establish that people are present and breathing—“[s]o it’s not really affecting our sleep. And they kind of walk on by, . . . come by real quick, [look in], and take off” or they might “just tap real gently enough to make you move subconsciously.” Such efforts were not always successful, but they were always appreciated. At least some people reported that, when COs approached count in this way, they were able to sleep right through it. And respectful behavior by COs conducting count had a second sleep-promoting effect: when a CO makes an obvious effort not to disturb, those who are nonetheless woken up are at least spared the feelings of resentment, frustration, and anger so often provoked when COs seemed to be gratuitously inconsiderate—feelings that would in turn impede efforts to fall back asleep.
E. Light
In some units in some prisons, the lights stay on 24/7. In such environments, deep, restorative sleep can be impossible. In most cases, the lights are dimmed at night, but they never fully go off, leaving it “dark, but . . . not . . . sleep dark.” Those who suffer most are people on top bunks, especially those stuck directly below the lights. But the perpetual glow makes it hard for everyone to sleep, leading many to attempt creative strategies for blocking the light. Unfortunately, most such strategies are only marginally successful and quickly run up against several pathological features of the prison environment. In this way, exploring the phenomenon of excessive nighttime light effectively illustrates the way that, in prison, the causes of sleep deprivation are never only about a single isolated factor.
Excessive light also impedes sleep in prison, although, as with noise, the effects vary depending on housing configuration. In the worst cases, people are housed in units where the lights are on full blast 24/7. This is so in some ad seg units, where “very bright lights are on twenty-four hours a day.” In general population, in some instances, prison administrators may decide that certain units are too dark at night. One person told me that, in one facility he was in, “they used to cut the lights out” at night, but then there was “a fight and [the COs] really couldn’t pinpoint who did what, . . . so the sergeant actually made us keep the lights on . . . in the dorm. . . . Full strength. All night.” Whatever the reason, when the lights stay bright all night, sleep eludes pretty much everyone.
In most housing units in most prisons, the standard practice is to dim the lights, usually an hour or so after everyone is locked in for the night. If this situation is far preferred to round-the-clock brightness, for many people it still remains too light to sleep. This is especially an issue in dorms, in which several lights stay on all night and “it [is] never anything other than like twilight in that big space”—“it’s dark, but it’s not . . . sleep dark.” It is worse for people with bunks close to lighted areas of the dorm, typically those near “the guard booth and bathrooms.” Cells generally have light switches that turn off the cell lights completely. But the hallway lights, while dimmed, remain on. In at least some places, rather than all the lights being dimmed to the same degree, the lights are calibrated, with some portion remaining at full strength. It might be “half a tube that would still be on, and that could be directly in front of your cell”; understandably, those cells “are the least coveted.” “[E]ven if they turned off the big light, the small light was still there. And so it was never ever dark at all,” still “[t]oo bright to sleep.” Multiple people reported cellblock night lighting “bright enough to read by.” Hallway night lighting was especially a problem for those with cell bars rather than a solid door, because with this setup “[there’s] nothing filtering the light.” And even those with solid doors could find their sleep disturbed, because the “lights that stay[] on at night . . . still have a tendency of coming in through the window and shining something in your room.”
In both dorms and cells, those on bottom bunks have an advantage when it comes to light. Depending on the angle of the bed, “if you’re on the bottom bunk, the top bunk would shed some of the light off,” while “if you’re on the top bunk, you get all the light.” People on bottom bunks have an additional advantage: the configuration allows them to “put a sheet under the [top bunk] mattress” to make a tent. And assuming the COs on duty are not sticklers, this move may afford “some type of shade, to shade your face from the actual light.” But putting up such tents is against the rules and if the COs are unwilling to look the other way, the tents must come down, letting in light even for those on the bottom bunks.
Those on top bunks suffer most from excessive nighttime light. This is especially so in the dorms, in which some have the bad luck to be in beds directly under lights that remain on all night—so-called “bad bunk[s].” People likened the experience of a bad bunk to feeling “like you’re [a] cockroach [under] the light . . . and [you] can’t get away,” or like a “rotisserie [chicken]” sleeping under “rotisserie lights.” In terms of lighting, the worst moment—mentioned by several people I spoke to—is when all the lights come on first thing in the morning, often as early as 5:00
a.m. or 6:00 a.m. Especially for those on top bunks, once that happens, further sleep becomes impossible—“like trying to sleep with the sun on your forehead.”
Although the intensity of the intrusion varies, pretty much everyone must find a way to block the light in order to sleep. In the cells, people have somewhat more control over their environment and may employ hacks that involve tweaking the physical plant. When the lights themselves are easily reachable, the most common move is to make paper light covers and place them directly over the lights to cut the brightness. Sometimes these efforts could be very involved. One interviewee explained that he would use
a cardboard box—like a Ritz cracker box—you open it and flatten it out. [Y]ou would get three or four of those [boxes . . . and use] tape. Or . . . if you don’t have tape, you could get toothpaste [as an adhesive] and get paper, and that will dim [the light] out a lot. . . . I used to order art supplies . . . like construction paper [to] make a light block at night.
Another person described an even more audacious approach: “In max facilities, if the light is directly in front of the cell, people have gone so far as to actually paint that side panel [of the light itself] to block out as much [light] as they can.”
These strategies are not open to people in dorms, where the lights will generally be too high to cover or paint. Even if someone succeeded in getting at the lights to cover them, the COs would spot it immediately. As one person put it, his frustration apparent, “[T]here were nights [in the dorm] whe[n] I literally had to put my coat over my face, just to get some darkness.” Sleeping with something covering the face appears to be the primary way people try to block the light. Various strategies are employed to this end, but none is without downsides. If you use a T-shirt or a towel, it is likely to fall off in the night. When this happens, the sudden light exposure will wake you up. As one person described it, using a towel “would block the light, but when you’re sleeping [and] moving around, the towel comes off, and you are dreaming about light and then you open your eyes to see just why [and find] it’s not a dream.” And if one’s T-shirt or towel falls from a top bunk, there is no way to retrieve it without disturbing the occupant of the bottom bunk, who will feel the bed shifting as you get down and get back up. As we have seen, such maneuvers can create conflict that is best avoided. So, unless you have something else to hand, you are out of luck. If instead you try to cover your head with a blanket, you are likely to be woken up the next time a CO does rounds and insists on seeing your face: “If you slept with a sheet or a blanket [motions with hands over his face/head] . . . to cover your face, [makes knocking motion and imitates CO saying,] ‘I can’t see you. I can’t see you.’ ” Some people try to forestall this disruption by sleeping with one bare foot outside the blanket, but during the winter this approach may leave one too cold to sleep. And even should a person find an effective way to cover their face, it won’t necessarily do the trick, because you “could still feel the light.”
In some facilities, enterprising craftspeople make eye masks—a homemade version of the type people get when flying first class. They may even sell eye masks to others. Using an eye mask appears to be the best way for people on top bunks to reduce the intrusive effects of the lights when they are trying to sleep. But this strategy too has its challenges. One problem is that, in many cases, the masks are made from materials taken from items issued by the prison: fabric from T-shirts, cotton from pillows, elastic from boxer shorts. In every facility, destruction of state property is a rules violation, which means that the mask you may rely on if you are to have any hope of sleeping may also be the basis for a write-up. Then there is the fact that, in prison, fully covering one’s eyes (or blocking one’s ears) while asleep can impede a person’s ability to quickly recognize danger. As one person put it,
Sleep is great, but [in prison,] what you don’t want to be is unaware. So . . . wearing earplugs or putting an eye mask on—and actually I did use an eye mask . . . made out of socks—[is not a good idea]. [P]rison’s a place where you want to be aware of your surroundings, and especially if you’re in a dorm environment.
All this goes to show how, as prisons currently operate, something as simple as trying to block the light to enable sleep quickly collides with many pathological aspects of prison life, including the delicate dance of conflict avoidance, overly intrusive nighttime CO rounds, the perpetual fear of violence, and the enforcement of what prisoner-turned-criminologist John Irwin labeled “chickenshit” rules. As with all the causes of sleep deprivation explored in this Part, the sleep-compromising effects of excessive light in prison go well beyond those experienced by nonincarcerated people who may face this issue in their own lives.
IV. Findings II: Meta-Conditions Impeding Sleep in Prison
The conditions discussed in Part III are specific and concrete, of the sort typically considered appropriate targets for more conventional policy reform. But it is also possible to identify a number of what might be called “meta-conditions” that also tend to impede sleep in prison. By “meta-conditions,” I mean aspects of the carceral experience that are institutionally pervasive and highly constitutive of life inside, yet so deeply embedded in the life of the prison—so wholly naturalized—that it can be hard to recognize either their destructive impact or the institutional role in their production. In this section, I highlight the sleep-compromising effects of five such conditions: fear of violence, trauma, poverty, overly intrusive rules enforcement, and daily humiliation. As will be seen, there is some thematic overlap here with the discussion in Part III, a function of how, in prison, all aspects of the experience are fundamentally interconnected.
A. Fear of Violence
In prison, the fear of violence is endemic. It compromises sleep in many ways, most profoundly when people are forced into close quarters with others not of their choosing. Whether in a cell or a dorm, it takes time to feel comfortable enough around strangers to fall asleep. People in dorms, who live in company with scores of others, are especially at risk and thus especially apprehensive when arriving in a new environment or when new people enter their unit. But rarely is anyone—even a long-time cellmate with whom one has a good relationship—ever fully trusted, and people pretty much always sleep “with one eye open.”
It might be imagined that, in prison, a perpetual fear of assault would keep people awake 24/7. This is not the case. At the same time, in all but the most functional lowest-security facilities, the fear of violence is real, and undermines sleep in numerous ways. In my interviews, I asked: Did you ever have difficulty sleeping because you were in close proximity to someone you didn’t know or trust? In response, everyone said pretty much the same thing. In double cells, people sleep little when first encountering a new cellmate; it takes time to be comfortable enough to be able to sleep in close quarters with someone who is basically a stranger. “If you’re just entering a cell with somebody new that you don’t know, if this person moved, or you heard the mattress twitch or anything—any slight movement at all—your eyes are wide open.” But eventually, “you kind of get comfortable with the person you’re with.” As one person described it, when assigned to a cell with “a roommate that [he] did not know,” at first all he could think was, “I do not want this guy to go crazy and try to kill me in my sleep [or] to take my stuff in my sleep.” It took “a few days to a week to really feel like okay [signals with hand lowering]; the tension is okay and I can sleep.” It is not unusual for people to spend years in the same cell with the same bunkmate. In these instances, this initial interference with sleep is brief enough not to be especially significant. But it is also not uncommon for people to be frequently transferred between cells, units or institutions, or to have a rotating series of cellmates. For those in this situation, the disruption to sleep from being in close quarters with others neither known nor trusted can be considerable.
In dorm settings, any influx of new people generates an even greater sense of insecurity. Because of the open setup, “you had to watch your back” because “most people know that when a person is asleep, that’s his most vulnerable time because he can’t see the attack coming. And there have been a lot of guys who have been attacked in their sleep.” New arrivals are thus always viewed with suspicion, and people will be hesitant to sleep deeply until they can be sure of their own safety:
You’re sleeping with one eye open and one eye closed because you don’t know who you’re in a dormitory with . . . . [I]f a new guy is coming to [the] dorm, everybody’s kind of leery of who this cat is. Everybody’s watching him [to] see how he’s moving, what he’s about. And then it doesn’t happen until you begin to talk to him. You realize, okay, he’s cool. I can go to sleep.
Depending on the character of the dorm, people may never feel secure enough to give in to deep sleep. Or they may band together with trusted associates and sleep in shifts, taking turns to watch over each other to ensure that all is well. I heard of this strategy from two people housed in dorms in two very different jurisdictions, one in the Northeast and one in the South. As one person who did his time in the South described it:
[I]n a dormitory setting, so you have a lot of chaos going on. So you might have four guys [who are] really close. And they’ll be like, “Man, look, we’re gonna go take a nap [so] keep an eye out on things,” [which] means, watch over us. So they do for a certain amount of time. . . [a]nd then . . . the guys [who were watching] are going to go to sleep, and [the] other guys [will] stay up.
The fear of violence also compromises sleep in other ways. In prison, particularly in high-security facilities, people who are perceived as weak are especially vulnerable to being victimized. In dorm settings, where people must sleep out in the open, the stress of feeling oneself to be at risk from innumerable potential assailants can make it very hard to sleep. I asked people this question: Did you ever have trouble sleeping because you felt physically vulnerable or were afraid for your safety? In response, several told the same story: they had gone into prison young or physically diminutive and were only able to gain a sense of security by showing they could fight. As one person described their experience, “I went into prison at 18 years old, 185 pounds, and a very effeminate male. So . . . I was beaten up pretty badly a couple of times [but then] I learned how to very quickly overcome that by becoming the person that nobody wanted to mess with, whether I could back it up or not.” For those who did not manage to build a reputation of this sort, the ongoing sense of vulnerability would make it very hard to sleep.
Then there are the sleep-disrupting effects of knowing that others around you may be targeted for violence. One man I interviewed lived for twelve years in a dorm setting in a Midwestern prison. Although he did not feel himself to be at risk, he described being assigned a bunk adjacent to that of a “young kid” who was repeatedly sexually assaulted after lights out. Even if one were not inclined to come to the defense of someone being attacked (and in prison, there are many good reasons not to intervene, including the desire to avoid becoming a target oneself), people need to be able to achieve some measure of physical relaxation in order to sleep. In environments where such violations are possible, the fear of victimization will be a perennially disruptive force, whether the potential victim is you or someone in your vicinity.
For those in two-person cells, even when you know someone well, trust in a cellmate can only ever be provisional. “In a double cell, even if you become friends with a person, there is still that unsurety of safety. So even the heaviest sleeper [is] sleeping with one eye open.” This is because “no matter how high of a level of trust you may have for somebody inside prison, you don’t ever [fully] trust them. So if you hear somebody moving in the middle of the night, you’re paying attention, whether it’s consciously or subconsciously.”
Although most of the time, most people in prison will not be assaulted while they sleep, the fear of violence, including sexual violence, remains pervasive and never wholly disappears. Living with such insecurity, in an environment where violence is an ever-present possibility, means that when people sleep, they sleep “light”: “In prison, there is no way in hell you should ever sleep [so] hard that a man can come in the cell and leave and you not know.”
This is especially the case for those who are gang-involved or who have affiliations with groups that may require them to engage in collective violence at a moment’s notice. When people in prison talk about “sleeping with their shoes on,” this is what they mean. For example, in the federal system, your state of origin may dictate who you “run with” and thus on whose behalf you may be expected to fight. As one person who spent years in a federal penitentiary explained, for those in this situation, there is no sleeping until the cell doors lock for the night: “[Y]our homeboys, everybody’s on duty. You gotta keep your boots on and just sit there until the CO says it’s time to go lock the doors.” And in the morning, the moment “the doors pop, you have to be awake because somebody might run in the cell [and attack you] because something happened in another unit with somebody from the same state you’re [from].” Those committed to this arrangement learn not to let themselves sleep deeply so they can always be ready to act quickly if need be.
Those who lack such commitment but who are nonetheless expected to participate in collective violence can lose sleep for different reasons: the stress of being unsure whether to respond when the moment comes, and the fear of retaliatory violence if they opt not to join in. One of my interviewees described exactly this conundrum:
I had some friends in gangs. . . . They would get into situations where it’s like, “Yo, we’re about to move on this dude. . . . [Y]ou a Blood just like we Blood[s] [so] you gotta move with us.” And I see my friends sleep uncomfortable. Because it’s like, man, I’m going to school. I got something good going on. I’m trying to leave that [and] change my life. But if I don’t move with them, . . . I become one of the enemies. There [are] so many different factors . . . to this life in [prison] that sleep is not something that comes by easy.
As this person explained it, given all the stresses of life in prison—including the pressure from the gangs and the dangers people may face if they try to remain unaffiliated—“If you’re able to sleep in [prison], you’re almost looked at as [an] extra-terrestrial. You must be [an] alien. No seriously, something is wrong with you if you’re able to sleep [inside].”
B. Trauma
People in prison are exposed to countless traumatic events, whether experiencing them directly or as witnesses to them. In my interviews, people described seeing people stabbed or beaten to death, flayed open, burned alive, or committing violent acts of self-harm. Experiences like these can leave people terrified for their own safety and may also interfere with sleep as people struggle to process psychologically and emotionally what they have witnessed. Not everyone has this reaction to witnessing senseless violence; many reported becoming inured to the brutality around them. This desensitization may serve a protective function, allowing people to survive impossible circumstances—and perhaps to sleep a little more easily. At the same time, it reveals a core mechanism by which the carceral experience alienates people from their own humanity.
Judging from my interviews, the experience of trauma further compromises the quality and quantity of sleep that people get inside. This dynamic, moreover, is self-reinforcing, as those who sleep poorly after traumatic experiences appear to be more likely to develop post-traumatic stress disorder (“PTSD”) and other mental health conditions—conditions that, among other unwelcome effects, may in turn impede sleep. True, poor sleep is a diagnostic component of many mental health disorders, sometimes raising the question of which came first. But in the case of PTSD, the evidence suggests that sleep disruption itself—in particular, disrupted REM sleep—may provoke the development of PTSD following trauma exposure. This possibility is especially concerning given that both persistent trauma exposure and systematically disrupted sleep are hallmark features of the prison experience.
In my interviews, I heard countless stories from people who witnessed deeply disturbing events while incarcerated. These included seeing “somebody get beat to the point where they’re laying in a puddle of blood and they have to be carried away because they can’t walk”; seeing “a kid get slammed on his neck and, then you see him getting a white sheet over his body and he’s getting carted off the unit. Or somebody leaking, somebody’s white T-shirt is now red”; seeing “one man that was literally cut open from the bottom of his chin to the top of his hips, cut and flayed wide open because they were looking for something they assume that he had swallowed”; and seeing “another guy running down the tier with his eyeball bouncing off of his cheek.” At some point, trying to process the weight of all these accounts, I began asking some version of this question: Did you ever find it hard to sleep because you were traumatized by things that you had seen?
For those who responded in the affirmative, the answers suggested two ways that witnessing traumatic events can interfere with sleep. First, there was the sleep-disrupting effect of the fear engendered by exposure to terrible violence. Seeing the awful things that happened to others made it hard to sleep, “knowing that any given time somebody can choose me as their victim.” This effect is of a piece with the fear of violence previously explored. But there was also a second theme that emerged in the answers: the sleep-disruptive effects of the psychological or emotional processing triggered by exposure to trauma, as when “you just keep seeing that same image over and over and over.” One person, recounting a particularly vivid and distressing event, effectively captured how the brain’s need to process can make it hard to sleep:
[S]ee[ing] guys have . . . padlocks put on their doors, flammables thrown in on [them], and one guy had been burnt alive. I witnessed this man, that’s gone from someone 6’1”, 6’2” . . . [to] not even eighteen inches long, you know, charcoal, just burnt alive. . . . [W]hen you’re seeing all of these things . . . [s]leep is hard. . . . [
Y]
ou’re trying to sleep, but you have all these things going on in your head, and any little sound that you hear, you[
‘re]
coming out of this sleep.
Not everyone connected the trauma they experienced to difficulty sleeping. But it is hard to imagine that the quality of one’s sleep would remain unaffected in an environment where such exposure is common.
At the same time, many people reported that the distressing things they saw while they were in prison did not affect their ability to sleep. Over time, they instead became inured to the violence and abuse to which they were exposed:
[While in prison] I saw a lot of violence. But it became, when you see it, you just see and you don’t see, you just sort of get out of the way . . . [and] you keep it moving. Prison teaches you to literally stay in your lane. . . . If it doesn’t affect you, oh well.
Whatever the example may be—seeing somebody cut, seeing somebody stabbed, seeing an officer beat on somebody, break their arm, or whatever it may be, you get so desensitized that this is natural.
Eventually you get numb to all the stuff that happens.
It may be that these individuals remained genuinely unaffected emotionally by what they saw and experienced. But the terms they used to describe their state of mind, which included becoming “numb” and “desensitized,” suggested their recognition of the defense mechanisms they were relying on to get through. To experience a state of equanimity in such a traumatizing environment, the level of denial would have to be considerable. If one’s sleep is thereby protected, it may also indicate an alienation from one’s own moral core, a feature of prison’s dehumanizing effect that carries its own psychic and emotional costs.
C. Poverty
Having money in prison does not guarantee a good night’s sleep. But it helps. There are innumerable items that those with money can procure, many directly from the commissary, that may enhance sleep inside: fans to drown out noise, extra clothes for cold nights, high-quality eye masks, softer sheets, and warmer blankets, not to mention sleep-inducing drugs available on the prison’s black market. And those with money can stock up on commissary items, ensuring they will not go to bed hungry. We have already seen that people who lack family support or are otherwise unable to earn what they need may opt to engage in survival strategies that put them in danger, including stealing, gambling, or selling sex or drugs. The official practice of paying incarcerated workers nominal wages—in those jurisdictions that pay them at all—can leave the least well-off people in prison forced to choose between forgoing items that might improve sleep or adopting strategies that could put them at risk of serious physical or sexual harm.
Most people enter prison without financial means. Although most institutions require prisoners to work, the hourly wage—when paid at all—is a pittance: on average, between $0.14 and $0.63 an hour. Depending on the prison, some people may secure positions with the private industries that operate behind the walls. But even in such cases, although the hourly wage may be considerably higher than the standard prison wage, it will remain far lower than the federal minimum wage—typically somewhere between $0.33 and $1.41 an hour. Some fortunate few can make even more: one person I spoke to trained call center agents for a private company. He reported making $285 per month (roughly $1.80 an hour), in contrast to the kitchen workers in his facility, who made around $30 per month (roughly $0.19 an hour). But these relatively well-paid opportunities are rare, and as we have seen, if people are to have access to money while in prison, they must either get it from loved ones outside or find off-the-books ways to earn it.
Although everyone in prison, regardless of financial status, faces serious obstacles to getting adequate sleep, access to money can help improve one’s odds. Many people mentioned buying fans from the commissary, “the whole point [of which is] to drown out the noise from the tier.” As one person explained, echoing an experience shared by many, “I could not sleep without my fan. I pretty much set it up right by my head . . . to help drown out the other ambient noise . . . outside the cell.” But fans that do the trick do not come cheap in prison—the most durable of them can cost up to $40. Given the extremely low wages paid to incarcerated workers, a fan could cost almost a month’s earnings. And if a person still chooses to buy one—which they well might, if it means the difference between sleeping and not sleeping—they would have little left over for other necessities like food or basic hygiene products.
Money, or the lack thereof, impacts sleep in other ways. Those with resources to spare can buy extra clothes—sweatshirts, warm socks, etc.—to help keep themselves warm on cold nights. They can buy softer sheets and warmer blankets. They can buy marijuana, heroin, Seroquel, or other sleep-inducing substances on the prison’s black market, as several of my interviewees reported doing. If they are not themselves skilled with a needle, they can pay someone to make them a top-shelf eye mask. They can buy a second fan to cool their bodies on hot nights. And if they use a CPAP machine, assuming power strips are allowed, they can buy one to guarantee themselves access to a plug without risking getting into a fight over control of the wall sockets with a cellmate or others in the dorm. These advantages may seem minor. But for people in prison, they may mean the difference between eking out a few solid hours of sleep or being consigned to sleepwalking through another day.
Still, when it comes to being able to sleep, the primary reason to want money or its equivalent is to ensure you have ready access to food. As we have seen, “if you don’t have the money, you can’t eat, and . . . you can’t
sleep if you’re hungry.” As one person explained, if he didn’t have much money on his books,
I might not have been able to buy as much canteen [as I needed], so I would go through it. . . . When you’re hungry it’s hard to budget it and make it through. Like one night . . . I’m trying to go to sleep and I’d be hungry so that I would go in and eat some [food items that had been] budgeted for later on [in] the week. But I just I had to eat it then. . . . Eventually, though, when I started getting more money, [hunger interfering with sleep] wasn’t a factor.
The lengths to which people will go to get money for food are a testament to the urgency of the need. One person I spoke to, who had no family support, “started gambling and stealing while [she] was in there to provide for [herself],” because “[o]nce you[’re] full, you can sleep better.” It is hard to overstate the desperation that would drive a person in prison to engage in either of these strategies. Someone caught stealing may earn a violent reprisal from the victim of the theft. And gambling is equally risky. It is a sure-fire way to get into debt, and in prison, the penalty for nonpayment of debts is often physical violence or forced sexual servitude.
Another risky, last-resort strategy is “two-for-one,” the prison loan-sharking practice whereby someone with a load of commissary will provide you with something you want—say, a candy bar, a bag of chips, or a soup. The catch is that, when the next day for canteen rolls around, you have to pay back double what you borrowed—two candy bars, two bags of chips, or two soups. If you can’t, the quantity owed doubles again. This is another way people in prison can get into debt, which, again, can result in violence or pressure for sex. “[Y]ou have dudes who are wealthier than others who max out every canteen . . . , and they sell stuff to guys. ‘Well I’ll give you one for two back.’ ” Or “[t]hey won’t tell you it’s two-for-one. [T]hey’ll say, ‘Here’s a box of [cookies].’ Then . . . [the] next week you owe double, . . . they’d say, ‘[W]ell, we could exchange it for sexual favors.’ That’s how they were trapping these young kids coming in.” According to my interviews, the desire to avoid going to sleep hungry is a major reason why people risk putting themselves in this position. Note that at least some of this felt pressure to access commissary, which can drive people to act in ways that could put them in serious danger, would be alleviated if prisons simply made decent food available to everyone in the evenings before lights-out.
These dynamics mean that those without money—the prison’s poorest, who lack outside support or the ability to accrue funds while inside—experience appreciably worse conditions than those with means. To some, this situation may seem unproblematic, just how the world is. But as the example of sleep makes clear, what is at stake is not simply differential access to luxury items. Sleep is a basic human need, and insufficient sleep causes considerable harm to physiological, psychological, emotional, and cognitive health. To the extent that poverty leaves some people less able to mitigate the sleep-disrupting aspects of the prison environment, poor people are experiencing a harsher, more destructive punishment than that experienced by those with access to funds.
D. Overly Intrusive Rules Enforcement
In prison, virtually every aspect of a person’s life is governed by rules enforced by COs. Many of the strategies people adopt to try to improve the quality of their sleep run afoul of these rules. This situation leaves people forced to choose between breaking the rules—and possibly getting written up, but perhaps eking out slightly improved sleep—or following the rules and forgoing small comforts (an extra mattress, a light cover, some food held back from a tray to eat before lights out) that might help them sleep. The stress involved in making this choice itself compromises sleep, especially for those who opt to roll the dice and break the rules. Compounding the problem is the fact that penalties for minor rule violations of the sort at issue here often include loss of commissary or yard privileges, thus depriving people of access to food and exercise—two pathways for improved sleep available to those in prison.
Pretty much every aspect of a prisoner’s life is governed by rules. Enforcement authority belongs to line officers, who, when they witness a rules violation, may issue a ticket (also known as a “shot”). The most extreme violations, which generally involve physical violence, may earn the perpetrator delayed release and even additional charges. In cases involving less serious offenses, penalties are less severe but still deeply undesirable. They may include time in ad seg, lockdown in one’s own cell, or temporary loss of yard time, commissary, phone privileges, or even shower access. As with law enforcement on the outside, COs have considerable discretion when enforcing prison regulations. As one person put it, discussing light covers, “[M]aybe when you go to sleep at night, the CO doesn’t care [so] they let you do it. But then the next shift comes on, and then they’ll bang on the door, [and] you’ll take that down.” If the violation involves possession of contraband—which generally means any item not directly issued by the prison or purchased directly from the commissary—a CO may choose only to confiscate the offending item, or they may confiscate it and issue a write-up, potentially exposing the offender to some form of penalty.
Many rules enforced through this system serve valid institutional purposes; few people would quarrel with prison prohibitions on physical violence or the distribution of narcotics. But innumerable other “chickenshit rules” penalize conduct that is less obviously necessary to maintaining institutional order. As Irwin has observed, “[M]any of [the facility’s] rules intrude into prisoners’ ordinary practices and significantly interfere with their attempts to carry on their already excessively reduced life routines.” In other words, people inside may find themselves at risk of punishment simply for trying to make intolerable conditions slightly more bearable.
This effect is certainly evident in the context of sleep. One person I spoke to clearly captured this dilemma. As he explained, “[I]f the light was right in front of my window, I would cover it up. And then that’s usually when the CO would bang on the window, . . . but then I would just take it down and put it right back up. . . . I was basically being forced to break the rules to be able to sleep.” The use of light covers is only one example of a
standard strategy for improving sleep that violates prison rules and can expose people to sanctions should they get caught. Among other things, it is generally against the rules to:
- Drape a sheet from the top bunk to curtain off one’s bottom bunk (a.k.a. make a “tent”);
- Cover the vents or windows;
- Shower outside designated hours (especially on cellblocks);
- Possess an extra mattress;
- Possess an extra blanket or sheet;
- Destroy state property (including the shorts, T-shirts, or stuffing from mattresses or pillows used for making eye masks);
- Remove food from the dining hall;
- Store food from the dining hall in one’s cell;
- Hold back food from trays.
Rules thus prohibit precisely the strategies that many people use to try to improve the quality and quantity of their sleep in the face of excessive light, excessive heat or cold, uncomfortable beds or inadequate bedding, and insufficient food. To compound the problem, the penalties for such violations may include loss of commissary privileges and loss of yard time. Yet, as we have seen, without access to food, people will often be forced to go to bed hungry, a state that considerably impedes sleep. And for many people, vigorous exercise is a way to “exhaust themselves, [so it’s] that much easier to go to sleep at night.” Without access to the yard, this strategy is more difficult to operationalize, and sleep becomes that much harder.
The perverse structure of this arrangement tells us everything we need to know about the moral orientation of the carceral system toward those inside. First, the prison operates in innumerable ways to undermine prisoners’ ability to sleep. Then, COs police the housing units to ferret out any items beyond each person’s allotment—items they may have expended considerable effort to assemble to try to improve their sleeping environment and perhaps increase their chances of getting a little more sleep. These items are then confiscated, a move that strips the possessor of the benefits, however slight, the seized items may have offered. COs are then empowered to impose penalties likely only to compromise still further a person’s ability to sleep. It is tempting to call this system Kafkaesque, except that, for Kafka, the bureaucracies that outrage and dehumanize are so infuriating in part because their procedures have no moral valence and are simply manifesting their own internal imperatives. In the prison, the cruelty of the process just sketched is of a piece with the callous indifference with which the institution and its COs seem to regard the daily hardships faced by those in their custody.
For those on the receiving end, COs’ interference with their efforts to improve the sleep they get may seem spiteful and even sadistic. And in some instances, this impression may be apt. It does, however, bear noting that COs’ decisions as to how to exercise their considerable discretion vis-à-vis rule violations may also be shaped to some degree by the institutions’ own pathological impact on the COs themselves—including, ironically, the chronic sleep deprivation that is a standard comorbidity experienced by those in the role. When COs are sleep deprived—a condition almost certainly exacerbated by the host of other comorbidities to which COs are disproportionately prone, including depression, alcohol overuse, PTSD, and suicidality—they become less able to treat the incarcerated with consideration and sympathy. To the extent that prisoners’ sleep is compromised by exposure to rules unevenly and unpredictably enforced by people who at times seem incapable of basic human sympathy and understanding, the resulting frustration may be less a product of individual spite than the broader operational logic of a system that is blind to the humanity of all parties, incarcerated and COs alike.
Yet however incapacitated COs may be by the hours they must keep and the conditions in which they work, it is the prisoners who are forced to daily navigate a challenging context in which any efforts they make to alleviate conditions of extreme discomfort are liable at any moment to be nixed by those COs who for whatever reason are not inclined to look the other way. As a consequence, not only do people in prison expend considerable energy strategizing how they might improve their sleeping environment, but a good part of this effort turns out to involve a constant process of weighing the benefit to be gained from a rules violation against the odds of getting caught and incurring a penalty. As I wrote in my field notes after one interview, “even if you weren’t going to get written up, even if the staff weren’t going to give you a ticket for an extra blanket or an extra mattress or whatever, there’s always the stress of knowing they could do it.” In this situation, “you were always trading the ability to be slightly more relaxed and not able to sleep”—because, in a bid to reduce stress, you opted not to avail yourself of the strategies that might help make sleep possible—against “put[ting] yourself in a position where you were stressed about whether you were going get into trouble just to improve [your] sleep a little bit.”
The stress this situation occasions can be both considerable and relentless. For example, as one interviewee reported, there were days when, as frequently happens, the people on his tier were not able to shower:
And you don’t want to go to sleep sweaty and dirty [because] that really impedes the ability to sleep. [So] you would have to find a way to sneak out of your cell to get a shower just so you would be able to sleep. So how would you do that? Well, if there was a time when tier 1 was supposed to be on the floor and the mezzanine, [and] tier 2 is supposed to be in their cells, you might sneak out of your cell even when you were supposed to be in it. Or you would try to get them to open your cell door for some reason and then grab a shower.
In other words, “it was hard to sleep when you weren’t clean and [there was] also stress involved in trying to get a shower when there is no time available for you to do it.”
Note that the goals driving this stressful negotiation of prison rules—sleeping, eating, keeping warm, showering—are at once basic requisites of human functioning and integral to maintaining one’s own humanity and sense of self. And, of course, the frustration occasioned by the enforcement of what are widely experienced as petty rules is also likely to interfere with the ability to get restful sleep.
E. Daily Humiliations
Every day, people in prison receive multiple reminders that the system regards them as undeserving of consideration and respect. At night, as we have seen, staff members conduct count and security rounds often without seeming to care whether they wake up the people sleeping around them. And similar demonstrations of callous disregard persist all day long: the incarcerated find their health problems minimized, their reasonable requests peremptorily denied, and their valid grievances ignored by prison officials at all levels. These routine humiliations understandably leave people frustrated, humiliated, resentful, and outraged. These feelings do not simply dissipate once an interaction is over. They persist and fester, making it even harder for people already coping with countless sleep-impeding obstacles to fall or stay asleep at night.
On a daily basis, people in prison are treated in myriad ways as if they are morally worthless. One manifestation of this attitude recurs nightly, when many COs appear to make no effort to do their rounds quietly. Instead, many behave as if the beds around them are empty, as if they are not surrounded by human beings who are trying to sleep—as if, as one person put it, “[T]hey don’t see you [or] your head right by the pillow that [they]’re standing next to.”
Nighttime noise by staff doing rounds wakes people up, thereby directly interfering with prisoners’ sleep. But there is a second, more subtle sleep-disrupting effect of this intrusive conduct: it forces the incarcerated to perpetually confront the disdain in which they are held, an experience that can spark feelings of helplessness, frustration, and rage that themselves impede sleep. Imagine how hard it would be to fall asleep if, every night, thoughts like these were swirling around in your head:
So on top of already fighting to get some type of sleep, those little couple of hours that you may get, then it’s interrupted by [the COs] . . . . [A] lot of them . . . have no type of empathy at all. . . . [Y]ou’re asleep [and] they don’t care. You’re an animal. This is the[ir] house [so] you got to do what they say, when they say, how they say.
[F]rom the officer’s perspective, they were . . . told like, “[H]ey, look, your job, and the way this job is supposed to be done, is more important than any of these inmates’ sleep.” W]e were . . . pieces of trash. We were manipulators, and we weren’t to be trusted so who cared what we thought and . . . that’s what [we] get for being in prison
It’s a power issue. You ask them to keep it down or something like that, they’re gonna look at you like, . . . ‘[W]ho the hell are you? Yeah. Right.’ And they’ll spite you [by being] even louder. . . .
As one person I spoke to observed, “[S]leep is more than just laying down and resting your body. [To] sleep, your mind needs rest as well—being able to actually go to sleep and not have to worry about anything, [to] just be comfortable in that moment and let your body and . . . mind rest.” Far from promoting a restful mind, daily humiliation seems far more likely to
achieve the opposite. Compounding this effect is the way those in this position must swallow these feelings—however justified—because to express them would be to court retaliation that among other hardships would only guarantee still further sleep disruption.
The ethos of contempt and callous disregard reflected in the behavior of many COs on the night shift also manifests in multiple ways during the day. Staff ignore the needs of those in their custody, refuse their reasonable requests, and target them for abusive treatment, whether out of malevolence, caprice, or poor judgment sparked by their own intense sleep deprivation. Medical personnel dismiss prisoners’ symptoms, deny requests for x-rays or specialist access, and pass them off with ibuprofen and pamphlets. Access to loved ones, both by phone and in person, is highly restricted, often for no reason other than official caprice. Life is constantly governed by “chickenshit rules,” which can seem designed more to demean than to address genuine institutional needs. Grievance procedures are generally byzantine, and complaints, however meritorious, are almost always denied. When disciplinary infractions require a hearing, these hearings often feel like shams, with those officials serving as adjudicators virtually always siding with the CO who wrote the ticket.
The relentlessness of these humiliations can take a deep emotional toll, leaving people feeling frustrated, resentful, disrespected, and outraged. Such feelings are not conducive to sleep, to say the least. In my interviews, I asked people what they thought would have to change to enable people in prison to get adequate sleep. Multiple people spoke about the dehumanizing treatment that they experienced daily, the constant reminders of their own powerlessness, and the pleasure staff seemed to take in ignoring their needs. One interviewee described what this treatment looks like in real time:
[For example,] I gotta go to school. Now I gotta ring the doorbell on the cell one thousand times for this individual to let me out. When you coming back from school [you might] want to use the restroom, but the officer got me out here waiting forty-five minutes. When we’re going to yard, we[’re] delay[ed] forty minutes and now when we get to the yard, instead of getting an hour and a half, we get thirty minutes in the yard. . . . It’s just little components of them treating us like [we]’re not human beings.
Or consider this account of a cell search:
When [COs] come in to search your cell, they don’t just nicely pick things up and put them back down. It looks like a hurricane ran through your room. They literally flip your bed over, your pillow is on the ground, your sheets are on the ground, your stuff gets mixed in with your bunkie’s stuff. Your clothes are all over the floor, your food’s mixed everywhere. Some [COs] will take the jelly and just squeeze it on your bed for no reason. . . . [T]hey come in and they destroy. That’s what they do.
When this is your daily experience, when nighttime rolls around, the state of relaxation requisite for sleep remains elusive. One woman I spoke to was especially eloquent on this point. What she said is worth quoting at length:
[I]f we’re constantly having these bad interactions, these bad experiences, and the culture isn’t conducive, or it always . . . undermine[s] you and . . . makes you less than and makes you feel worthless and all these other things—you’re not going to sleep like oh, I conquered the world today. No, you’re going to sleep feeling like shit, feeling even worse than you’ve felt. . . . I go to the freaking clinic to get medication for . . . a toothache and I get treated like a fucking drug addict. And [I] don’t even get any Motrin, so I’m not going to sleep feeling too good. . . . And then when you’re constantly looking at me and talking down to me as if I’m less than you, [as] if I’m not a person, . . . all those things don’t make for a good night’s sleep. You don’t feel good at the end of the night. . . . [A]nd not only that, but the setting in general . . . doesn’t make it conducive for family reunification. They make it even harder for you to have relationships [and] more so [for] women. [T]he women’s prison, geographically where it’s placed, is in the middle of nowhere. So all the people that often are coming into the prison settings live so freaking far away. . . . [T]hese inner cities are so far away from this rural fucking place in [location of prison] in the middle of nowhere, that it makes it hard for you to have visits, to have a constant contact with your family or the . . . how much is it? Fifty cents a minute a phone call? When we’re making one dollar a day on our freakin’ wages? Like, seriously? So all these things just work to produce even more stressors and, as we all know, when I’m super stressed throughout the day, all that’s happening at the end of the night when it comes time for me to sleep is these things [are] replaying in my head . . . and making me feel even more incompetent and worthless.
In short, sleep will not come easily to people who are constantly humiliated.
Like those conditions canvassed in Part III, the conditions discussed in this Part contribute significantly to the inability of people in prison to get adequate sleep. If there is a difference to draw, it lies in the pervasiveness of these “meta-conditions,” the extent to which each is woven into the carceral project. While the more concrete and specific conditions described in Part III appear tied to policy decisions concerning prison administration and operations, the conditions described in this Part seem to arise directly from the culture and ethos of the American prison. Yet, as far as moral valence goes, the distinction I have drawn between the more concrete sleep-impeding conditions and what I have labelled “meta-conditions” may not much signify. Whatever the moral orientation of individual COs, all the conditions that systematically compromise the potential for sleep in prison reflect a system shaped at its core by a callous indifference to the well-being of the incarcerated, a blindness to their humanity, and a refusal to accord those in custody the basic consideration and respect owed to any fellow human. This moral injury is part of the daily experience of imprisonment. And as the foregoing has shown, among the innumerable practical manifestations of this systematic institutional disdain is the persistent interference with the fundamental human need for sleep.
V. Implications: Prisons, Punishment, Law, Policy
This Part begins to explore the normative implications of chronic sleep deprivation in prison. Section A considers, as a descriptive matter, what this phenomenon adds to our understanding of the prison experience. Section B flags several ways the inevitability of chronic sleep deprivation deepens the punitive character of the carceral penalty. Section C offers a first cut at the core constitutional question—whether the conditions mapped in this Article violate the Eighth Amendment’s prohibition on cruel and unusual punishment. It also touches on a question that frequently arises when I discuss this work: whether the conditions chronicled here constitute torture. Finally, Section D offers some thoughts on the prospects for policy change and the challenges that await efforts to tangibly improve sleep in prison. For reasons of space, my contributions on these topics will necessarily be brief. The aim here is twofold: to begin mapping the normative implications of the problem, and to offer preliminary thoughts on the four dimensions identified here in the hope of sparking a broader conversation.
A. Prisons: Sleep Deprivation and Carceral Life
Prior to the interviews, many of my subjects had given little thought to their experiences of sleep inside, or to how those experiences shaped their time in prison. Yet what I learned during those conversations brought into focus certain basic realities of prison life that have, until now, been largely unrecognized, not only among many of my interviewees, but also among those who study incarceration. Below, in no particular order, I identify four such implications.
First, chronic sleep deprivation is a constitutive feature of life in prison. Regularly getting insufficient sleep—night after night, for months, years, and even decades—is as central to the daily experience of incarceration as lousy food, crushing boredom, grossly inadequate medical and mental health care, solitary confinement, and the perpetual fear of physical or sexual violence from staff or fellow prisoners. Being sleep deprived is an intrinsic part of what it means to be inside.
True, people in prison sometimes manage to get decent sleep. In some cases, housing configuration makes the difference. Virtually everyone I spoke to agreed that people in single cells sleep much better than anyone else. As one person memorably put it, “[T]he best cellie is no cellie.” In other instances, people can benefit from their own wise choices. One person I spoke to spent thirty-seven years in prison. He reported that, especially in later years, he was able to get sufficient sleep and wake feeling rested. He credited several aspects of the life he built in prison for this atypical experience, including staying out of prison politics (“not trying to be king of the jail”), making peace with not being able to control the behavior of loved ones on the outside (not “trying to live one foot in prison and one foot in the world”), and building a supportive religious community of fellow Orthodox Muslims who bunked together and provided one another with “comfort and security.” He was also an accomplished artist, which compounded his sense of purpose and seemed to enhance the personal equanimity that helped him sleep. Yet he also described multi-year stretches over his long incarceration during which he slept far less. And even in single cells—the equivalent of temporarily winning the prison sleep lottery—people struggle with noise, light, extremes of heat and cold, and many other obstacles to sleep chronicled here. In other words, in prison, even the outliers only sleep so well.
In any case, the fact that some people may sometimes beat the odds scarcely goes to controvert the core finding that sleep deprivation is endemic in prison. To take a parallel example, even though some people may sometimes receive reasonably acceptable and even effective medical or mental health treatment while incarcerated, there is no question that a systematic failure to provide adequate health care constitutes a definitive feature of the prison experience. A central claim of this Article is that chronic sleep deprivation, produced by persistent, systematic interference with sleep, is likewise intrinsic to the carceral experience—even if some people in prison occasionally manage to sleep reasonably well.
Second, at any given time, most people living in prison will be operating from a sleep deficit, possibly a considerable one, and this situation will have profound effects on what happens inside. In future work, I will take up the question of how sleep deprivation in prison impacts interpersonal dynamics and institutional functioning. For now, it is enough to say that, judging from my interviews (and as we would expect), the fact that virtually everyone in prison is chronically tired greatly heightens institutional tension and volatility, along with the likelihood of interpersonal conflict. And because COs too are frequently sleep deprived, the heightened potential for conflict and all that goes with it also implicates COs—who, it bears noting, are legally authorized to use force. At this point, it is hard to know just how much the instability and threat of violence that shapes life in the contemporary American prison is attributable to the poor judgment and short tempers emblematic of fatigue. Yet I am certain that the impact is far greater than heretofore recognized.
Third, the overwhelming weight of the sleep science, combined with the striking unidirectionality of the findings, strongly suggests that, in the aggregate, people in prison will suffer considerable physiological and psychological harm traceable to the long-term systematic interference with their ability to get adequate sleep. Again, at this stage, it is impossible to say with any precision what shape this harm takes. However, in light of what is known about the health-compromising effects of even short periods of sleep deprivation, it is hard to imagine that no such effects occur. In addition, the health-compromising effects of sleep deprivation suggest several other troubling possibilities, each of which merits serious consideration:
- The inability to sleep properly in prison suggests a vicious circle as to the health of the incarcerated: thanks to chronic sleep deprivation, people inside are vulnerable to illness and disease that well-rested people might avoid, and when they get sick, the constant sleep deprivation and disruption that define the carceral experience are likely to undermine the body’s natural healing processes.
- Sleep deprivation may help to explain why people age faster in prison. Research suggests that the biological age of incarcerated individuals is as much as ten to fifteen years greater than their chronological age. This effect is traceable in part to “the high prevalence of risk factors for poor health” among the incarcerated, including “a history of substance abuse, head trauma, poor health care, and low educational attainment and socioeconomic status.” Yet given the breadth of the harms generated by insufficient sleep, it seems hard to imagine that chronic sleep deprivation does not also play a central role in the accelerated aging process prisoners experience.
- If, as some studies suggest, sleep deprivation constitutes a risk factor for—or exacerbates the symptoms of—mental illness, the systematic interference with sleep that people routinely experience inside could help to explain the strikingly high incidence of mental illness among those in custody. These psychopathological effects mean that those concerned with the epidemic of mental illness behind bars should pay particular attention to the quality and quantity of the sleep people get inside.
Fourth and finally, the causes of sleep deprivation in prison are inextricably bound up with the normative design and operation of the modern American prison. Every person alive, whether inside prison or out, knows what it is like to try to sleep in the face of obstacles like noise, excessive light, extremes of temperature, and so on. For this reason, it may at first seem that people struggling to sleep in prison are on a continuum with others across society who fight for sleep in unconducive situations. But if sleep deprivation is not unique to prisoners, the impediments to sleep behind bars have a particular normative cast, reflecting the distinctive animus towards the incarcerated that shapes virtually all aspects of prison life.
For example: Prison beds are not fiercely uncomfortable by happenstance. People are forced to sleep on metal slabs with thin, cracked, moldy, sticky mattress pads because those who decide what form the beds will take view the resulting discomfort as appropriate for people serving time as criminal punishment and would begrudge as undeserved the cost of anything more substantial. People sleep on flimsy, creaky bunkbeds for the same reason, and fear moving in their sleep in case they provoke frustrated, angry bunkmates who, like themselves, are forced to live in moral ecosystems where they are routinely humiliated and dehumanized and who may lash out at small indignities simply to feel some small measure of control. Every night, people are woken up by COs who, “hav[ing] zero respect for prisoners,” make no effort to do their rounds quietly. During the height of summer and the depths of winter, people labor to sleep in intense heat or extreme cold, a hardship that could be largely addressed were state corrections agencies to invest in air conditioning and routine winter maintenance in the housing units where prisoners sleep, as they already do as a matter of course in the parts of the prison frequented by staff. And so on. Obstacles to sleep that may at first seem of a piece with what people routinely experience outside prison turn out on further examination to constitute specific, sleep-compromising manifestations of the callous indifference and even hostility towards the incarcerated that in countless ways shape the American carceral system.
B. Punishment: Sleep Deprivation and the Carceral Penalty
Recognizing the fact of sleep deprivation in prison and the shape it takes also deepens our understanding of the nature of the punishment inflicted on people sentenced to prison time. In this section, I identify four related dimensions of one key insight: the way the inability to get adequate sleep heightens the punitive character of the carceral penalty.
First, it is already well understood that in practice, a prison sentence carries with it a raft of noxious conditions beyond simply the deprivation of liberty. This Article strongly suggests that, among these conditions, we must include the experience of being chronically sleep deprived for the duration of the stipulated term. Not merely a side effect of being in prison, chronic sleep deprivation is a constitutive feature of the punishment itself, a part of the penalty imposed when the judge pronounces sentence.
Second, the punishment incarceration represents also includes being subjected to the harms that arise from chronic sleep deprivation. To put the point more concretely, when we sentence people to prison time, we are sentencing them to insufficient sleep, which is very likely to (1) trigger a degradation of their body’s proper functioning and defenses against illness and disease, (2) expose them to an increased risk of early mortality, and (3) ignite psychological harms, activating or exacerbating the symptoms of a wide range of psychopathologies. We are also sentencing them to perpetually live with the cognitive deficits that daily plague people who are poorly slept—the irritability, ready frustration, fuzzy-headedness, impaired judgment, proneness to conflict, and general incapacity that collectively make it difficult to get through the day, much less make good decisions, achieve one’s goals, or build healthy relationships. To the extent that sleep deprivation is an in-built feature of the prison experience, all these harmful effects become part of the punishment prison time represents.
Third, the punitive character of the carceral penalty is intensified by the effects of pervasive sleep deprivation on the prison’s social ecosystem. Although sleep-impeding conditions are experienced individually, they are simultaneously endured by everyone. In other words, people in prison are routinely forced into close quarters with scores and perhaps hundreds of others who are themselves struggling with the psychological challenges and cognitive deficits produced by chronically inadequate sleep. People in prison are already likely to be bored, frustrated, resentful, and sometimes distraught, a situation that helps to explain the constant undercurrent of tension, conflict, and volatility that defines the carceral environment. These dynamics will only be exacerbated when everyone is denied access to restorative sleep. Being forced to live in this environment, with all its stress, instability, and incipient danger, is yet another essential component of the carceral penalty traceable to collective sleep deprivation—part of what the punishment of prison time entails.
Fourth, being subjected to chronic sleep deprivation is dehumanizing, which adds a further dimension to the punitive character of prison. Dehumanization is the process of “stripping people of human qualities” so that “they are no longer viewed as persons with feelings, hopes[,] and concerns but as sub-human objects.” The mechanisms of dehumanization experienced by people in prison do not only lead others to deny their moral worth. They also turn the process inward, undermining a person’s most elemental capacities, including the internal resources necessary for self-reflection, reasoned judgment, moral fortitude, and personal growth.
Sleep deprivation is not generally recognized as among the core dehumanizing aspects of the carceral experience. But it should be. As we all know from personal experience, when we do not get enough sleep, our most fundamental faculties are impaired. It can be hard to think, to reason, and to read situations and other people. Even things that might otherwise feel urgent or meaningful can seem flat or insufficiently important to motivate action. These effects are regularly experienced by people in prison, where the stakes may be especially high. Those who are sleep deprived may react with irritation or hostility even to those they love the most—including the friends and family on the outside who are their lifeline. Like people everywhere, when they are tired, people in prison make poor choices. They opt not to participate in activities they would otherwise value—even those that might improve their chances of successful reentry. Thanks to diminished self-restraint, they sabotage their prospects.
In short—again, like people everywhere—prisoners who are chronically tired may find themselves sleepwalking through life. Obviously, by virtue of being incarcerated, the human potential and personal agency of people in prison are necessarily drastically curtailed. Yet it is also true that people make meaning in custody. Within the constraints of their incarceration, to a greater or lesser degree, people can be agents of their own situation. But the less sleep they get—and the more they are thereby deprived of the restorative effects of adequate sleep—the less equipped they will be to engage in meaningful, productive interactions, and the more readily they will be dragged into the dark, negative, hostile space that so frequently traps those lacking the resources to escape it. In all these ways, being sleep-deprived severs people from central aspects of their own moral characters and those features that define their humanity. This severing too is a part of the punishment traceable to—and inflicted by—systematic sleep deprivation in prison.
C. Law: Is Systematic Interference with Sleep Unconstitutional? Is it Torture?
The conditions described in Parts III and IV raise two obvious legal questions: (1) Do these conditions violate prisoners’ constitutional rights? and (2) Do they amount to torture under international human rights law? This Section considers these questions in turn. It focuses primarily on the constitutional question and the likely prospects for successful constitutional claims in the courts, before briefly touching on the matter of torture, which frequently arises in conversations about this research.
The primary provision for assessing the constitutionality of prison conditions is the Eighth Amendment prohibition on “cruel and unusual punishments.” At its most basic, the question is whether plaintiffs challenging the conditions catalogued here could satisfy current Eighth Amendment standards—and the answer, I argue, is yes. To see why requires a short overview of the governing law.
Under existing doctrine, to prevail on a prison conditions challenge, plaintiffs must satisfy two components, styled by the Supreme Court as “objective” and “subjective.” The objective component considers the challenged treatment, asking whether “the deprivation [was] sufficiently serious.” The subjective component focuses on the responsible officials, asking whether they acted “with a sufficiently culpable state of mind.” Of the two, the objective component may require somewhat more doctrinal explication to make the case. However, as currently defined, neither component should pose insurmountable obstacles for plaintiffs challenging the conditions that impede prisoners’ sleep.
As the Court made clear in Wilson v. Seiter, the objective component requires a showing that the challenged conditions “have a mutually enforcing effect that produces the deprivation of a single, identifiable human need.” Writing for the Wilson majority, Justice Scalia identified “food, warmth, or exercise” as examples of “human need[s]” that might ground such claims. A complaint, he observed, might allege “a low cell temperature at night combined with a failure to issue blankets,” thus implicating multiple “mutually enforcing” conditions depriving the plaintiffs of the “single, identifiable human need” for warmth.
Sleep deprivation seems a textbook case of the sort of objective component Wilson explicitly approved. It is beyond peradventure that sleep is a basic human need, as intrinsic to human survival as “food, warmth [and] exercise.” Even should plaintiffs choose to focus their complaint only on the more concrete factors explored in Part III—uncomfortable beds, hunger, extreme heat and cold, noise, and excessive light—these conditions together make getting adequate sleep close to impossible and would thus seem to readily satisfy the Wilson standard.
Yet incarcerated plaintiffs will not typically claim the complete denial of sleep but rather the inability, over months and years, to get even close to the recommended seven hours per night. And as Justice Powell observed in Rhodes v. Chapman, “the Constitution does not mandate comfortable prisons.” This means that, when incarcerated plaintiffs bring constitutional claims grounded in sleep deprivation, there is a need to determine the point at which insufficient access to adequate sleep ceases being simply the denial of those “creature comforts” of the sort incarceration necessarily entails, and instead qualifies as the “deprivation of a single, identifiable human need” that satisfies the objective component of Eighth Amendment conditions challenges.
In Garrett v. Lumpkin, an ongoing case out of Texas, the district court offered one answer. It held that “because Garrett failed to show that his sleep deprivation—which was undisputed—actually caused his health issues, he had not satisfied the objective element of cruel and unusual punishment.” Seeing why this answer fails as a matter of basic Eighth Amendment doctrine helps make clear why sleep deprivation need not be total to satisfy this hurdle. When Michael Garrett filed his claim, he had been in prison for more than thirty years. At the time he drafted his complaint, he was living in the McConnell Unit and was later transferred to the Estelle Unit. In his complaint, Garrett challenged a set of conditions that, he alleged, interfered with both the quantity and quality of his sleep and left him chronically sleep deprived. As to quantity, the schedule operating in the Estelle Unit afforded residents only three and a half hours of sleep a night, with bedtime set for 10:30 p.m. and breakfast starting “around 2:00 a.m.” (The hours in the McConnell Unit were only slightly better, with bedtime at 10:30 p.m. and breakfast commencing around 2:30 a.m.) And “even during this three and a half hour window, sleep is not continuous,” because residents must “be awake for a 1:00 a.m. bed-book count,” which meant that “the most continuous sleep Garrett can theoretically receive is two and a half hours” (assuming he falls asleep instantly at 10:30 and sleeps undisturbed until the bed-book count). As to sleep quality, the “nighttime prison conditions—namely, the hallway lighting, heavy doors slamming, and prisoners yelling—further imperil” the sleep residents are able to get during the brief period the prison’s compressed schedule allows.
In his complaint, Garrett challenged this raft of conditions on Eighth Amendment grounds and sought an injunction “that would mandate a prison schedule with six hours per night designated for sleep.” Yet the district court found that Garrett had failed to make out “the objective element of the Eighth Amendment inquiry.” The reason? Having failed to present “ ‘any expert testimony establishing that a lack of sufficient uninterrupted sleep has, within reasonable medical probability,’ caused his health conditions,” Garrett could not establish a cause-and-effect relationship between his sleep schedule and any medical complaint.”
But this notion—that a plaintiff in Garrett’s position can satisfy the objective component of an Eighth Amendment conditions claim only by showing actual harm—is directly at odds with governing doctrine, specifically the Supreme Court’s clear holding in Helling v. McKinney. Helling was decided just two years after Wilson. While in prison in Nevada, William McKinney was “assigned to a . . . [cellmate] who smoked five packs of cigarettes a day.” McKinney filed suit, arguing that exposure to this degree of cigarette smoke would unconstitutionally “jeopardiz[e] his health.” At trial, the magistrate judge granted the state’s motion for a directed verdict in part on the ground that McKinney “had failed to present evidence showing . . . medical problems that were traceable to [exposure to] cigarette smoke.”
Helling wound up in the Supreme Court. The state, following the magistrate, argued that “unless McKinney can prove that he is currently suffering serious medical problems caused by exposure to [environmental tobacco smoke (“ETS”)], there can be no violation of the Eighth Amendment.” The Court, however, rejected this suggestion, with its implication that “prison authorities . . . may ignore a condition of confinement that is sure or very likely to cause serious illness and needless suffering the next week or month or year.” As Justice White put it, writing for the majority, “[A] remedy for unsafe conditions need not await a tragic event.” Instead, the Court held, it is enough to show that McKinney was exposed “to levels of ETS that pose an unreasonable risk of serious damage to his future health.” The following year, the Court decided Farmer v. Brennan, which phrased the general inquiry slightly differently: whether the plaintiffs faced a “substantial risk of serious harm.” In the three decades since Farmer was decided, it is this latter formulation that has governed.
When Garrett came up on appeal, the Fifth Circuit made short work of the district court’s reasoning. In her opinion on behalf of a unanimous panel, Judge Clement drew directly on Helling, which declared it “ ‘cruel and unusual punishment to hold convicted criminals in unsafe conditions,’ regardless of whether those conditions actually cause injury.” Because “[i]t would be odd to deny an injunction to inmates who plainly proved an unsafe, life-threatening condition in their prison on the ground that nothing yet had happened to them,” the Fifth Circuit had previously held that, to satisfy the objective component of an Eighth Amendment conditions claim, the plaintiffs “need not show that death or serious injury has already occurred.” Instead—in keeping with Farmer—they “need only show that there is a substantial risk of serious harm.” In short, when sleep in prison is sufficiently restricted as to expose people to a substantial risk of serious harm, plaintiffs will have satisfied the objective component, whether or not the danger has yet manifested and even if they still manage to get some sleep each night.
What of the subjective component? Here too, were a court to faithfully follow governing precedent, there is a clear pathway for plaintiffs alleging unconstitutional sleep deprivation to meet their burden. The requisite state of mind for Eighth Amendment conditions claims is “deliberate indifference.” In Farmer v. Brennan, the Court held that to satisfy this standard—and thus the subjective component of an Eighth Amendment claim—plaintiffs must show that defendants “kn[ew] of and disregard[ed] an excessive risk to inmate health or safety.” As the Farmer court explained, Eighth Amendment deliberate indifference is equivalent to criminal recklessness as defined by the Model Penal Code: “[T]o act recklessly . . . a person must ‘consciously disregar[d]’ a substantial risk of serious harm.”
To satisfy this standard, plaintiffs bringing an Eighth Amendment sleep deprivation claim must show that prison officials subjectively realized the substantial possibility that, because of the challenged conditions, those incarcerated in their facility were unable to get adequate sleep. As always, plaintiffs must take care to name defendants who will have some knowledge of the challenged conditions. But given the pervasiveness of the conditions described here and the many operational decisions that daily compromise the quality of prisoners’ sleep—the nightly counts, the double-bunking, the crowded dorms, the early dinner hour, the lack of air conditioning, and so on—it would not be hard to identify prison officials with full knowledge of the conditions that deprive the incarcerated of the basic human need for sleep.
Defendants in such cases might try to argue that, although they knew of the challenged conditions, they were not aware that these conditions posed “a substantial risk of serious harm” to prisoners. Yet as the Farmer Court made clear, “[w]hether a prison official had the requisite knowledge of a substantial risk is a question of fact subject to demonstration in the usual ways, including . . . the very fact that the risk was obvious.” At a minimum, prison officials who work the night shift are well acquainted with the conditions under which those living in their facilities are forced to try to sleep. And because they are human themselves, and thus equally in need of the full complement of sleep necessary for human functioning (not to mention equally likely to have wrestled with environmental conditions impeding sleep), it would be reasonable to infer that they realized that people subjected to this set of conditions would be unable to get adequate sleep. Given the way the Court defined the Eighth Amendment deliberate indifference standard in Farmer, the plaintiffs should be readily able to identify defendants as to whom they can satisfy the subjective component of an Eighth Amendment sleep deprivation claim.
I am aware that, in this discussion, I have glossed over innumerable issues that plaintiffs may run up against in litigating these two aspects of an Eighth Amendment conditions challenge. For example, defendants might argue that plaintiffs failed to exhaust administrative remedies. They might invoke qualified immunity or “Turner deference,” or deny that plaintiffs suffered actual injury. They might claim relief is foreclosed absent a showing of physical injury. Or they might claim—as prison officials repeatedly did during COVID—that, despite taking no meaningful steps to reduce the risk of harm, they reasonably responded to the risk prisoners faced from lack of sleep and could thus not be found deliberately indifferent. At least as to some challenged conditions—hunger in particular comes to mind—they might insist that they lacked knowledge of the condition itself and thus of the risk of harm it might have posed. Litigators will need to address these issues as they arise. To some extent, their force will depend on the particulars of individual cases. But my point here is a more general one: assuming a well-litigated case, so long as courts apply the governing Eighth Amendment doctrine in an evenhanded and straightforward way, plaintiffs alleging unconstitutional sleep deprivation should have a clear doctrinal pathway to prevailing on the merits.
All this means that in theory—assuming the ability to overcome the many procedural hurdles to getting into court and getting a hearing on the merits—an affected class that meets its burden under governing Eighth Amendment standards should be entitled to injunctive relief designed to remedy the many sleep-impeding conditions producing harm or risk of harm. However, in practice, those seeking relief in the courts from the conditions undermining sleep are still likely to face a steep uphill climb.
One notable challenge would arise from the nature of the relevant evidence. On the one hand, the considerable body of sleep science demonstrating unequivocally the myriad harms—physiological, psychological, and cognitive—arising from sleep deprivation will help incarcerated plaintiffs make their case. On the other hand, making such a showing demands expertise and resources that pro se plaintiffs are likely to lack, which means that, to succeed, plaintiffs will require outside counsel with deep pockets and a commitment to building the strongest possible case. And even still, such efforts may run up against what Margo Schlanger has characterized as “a general hardening of attitudes about causation” on the federal bench, producing demands by courts for “more rigorous proof on harm and causation.”
Then there is the fact that today’s federal courts are extremely reluctant to order meaningful changes to how prisons are run. Except for one brief period in the 1970s and 1980s, federal courts have been notoriously unwilling to issue broad injunctive orders in prison cases. And as we have seen, the concern with sleep deprivation implicates wide swaths of prison operations—the design of housing units, maintenance of the physical plant, a prison’s ability to accommodate large populations in relatively small spaces, the way staff shifts are organized, how COs do their jobs, and even the fundamental matter of how staff perceive and treat the incarcerated. In today’s judicial climate, the breadth of any sufficient remedy is likely to incline courts to try to find some way to dispose of the case long before the remedial stage—however strongly supported the constitutional claim.
For these and other reasons, litigation will not be a cure-all. Yet even granting the impediments, these cases are still very much worth bringing. For one thing, with the right case, dedicated and adept advocates with sufficient resources to mount the litigation may well find a way to win. And partial successes are still worth pursuing. Even if a court were unwilling to order sweeping relief encompassing all the issues explored here, it would still make for better sleep inside if they were at least to require, say, adequate air conditioning during hot summer months or that prisoners be provided real beds with higher quality mattresses. To this, some may object that, given the cruelty of forcing people to try to sleep in the dehumanizing and inhumane conditions mapped here, we ought to accept nothing less than complete institutional overhaul. But when conditions are this dire, advocates do prisoners no favors by letting the best be the enemy of the good.
Moreover, we should not discount the possible virtuous circle that might arise were courts to order—and prison officials to implement—even partial relief. Of the obstacles to sleep catalogued here, those I have labeled meta-conditions are perhaps least likely to form the basis for injunctive relief. Yet if a carceral facility took steps to improve the more concrete conditions impeding sleep, some of the meta-conditions may at least partly take care of themselves. If a prison provided decent beds, served a meal before bed, reduced the frequency of nighttime counts, or ensured an ambient temperature conducive to sleep, these efforts might mitigate the daily humiliations that can themselves compromise sleep. Depending on the nature of the policy changes, people in such facilities might also feel less compelled to break the rules to try to improve the quality of their sleep, or to engage in risky behavior for the sake of having something to eat before bed. This possibility of felicitous knock-on effects is another reason Eighth Amendment claims grounded in sleep deprivation are worth bringing, even if the results may fall short of total institutional transformation.
Sleep deprivation may also ground claims beyond class actions seeking broad structural change. As with other pathological features of the carceral experience—medical neglect, inadequate mental health care, the risk of physical or sexual assault, excessive force, and so on—sleep deprivation is both a shared experience produced by macro-level institutional design and a personal experience the precise shape of which will vary according to the individual and their context. Depending on the circumstances, people in custody may face particular obstacles to adequate sleep for which individual prison officials may be liable. For example, prison officials may persistently refuse to provide testing and treatment for people exhibiting obvious signs of sleep apnea, or to authorize a cell move for someone whose cellmate is a kitchen worker with a 2:00 a.m. wake-up time, or to provide a longer or wider bunk for someone of greater than average height or weight. Such situations too deprive people of the basic human need for sleep and thus raise potential constitutional claims, along with claims under other legal frameworks—perhaps most notably the Rehabilitation Act and the Americans with Disabilities Act, the “two principal federal disability anti-discrimination statutes.”
Again, I do not wish to overstate the likelihood of success. Given the dispositional favoritism federal courts routinely exhibit toward defendant prison officials in prison law cases, incarcerated plaintiffs will always face long odds. Yet advocates should still bring these claims—and not only because they may sometimes win. They should also do so because, just by filing cases, they can put the issue of perpetual sleep deprivation in prison on the public radar. If change is to be possible, the public first needs to be aware of what goes on behind the walls. And litigation, with its built-in drama and familiar script, offers a way to expose the realities of prison life to public view. Legal cases can become the focus of media attention; as reporters cover the various stages of the case, they also educate the public about the underlying issues. Judicial proceedings offer inflection points for organizing, around which advocates can build political campaigns. And lawsuits themselves can open pathways for sympathetic engagement by legislators, who may in turn hold hearings and push for policy change. Eighth Amendment litigation alone cannot achieve these aims. It can, however, play a crucial supporting role in a multi-pronged, multi-stakeholder strategy.
One other legal issue, implicating not the United States Constitution but international human rights law, bears addressing here. That is, some readers may be inclined to regard the systematic interference with sleep in American prisons as tantamount to torture. Whether such a claim could succeed under international law is an open question. There is a high bar to a finding of torture under the United Nations Convention Against Torture (“CAT”), which explicitly excludes from the definition “pain or suffering arising only from, inherent in or incidental to lawful sanctions”—a provision that would greatly complicate efforts to apply CAT to sleep deprivation in American prisons (if not for cases focusing on pretrial or administrative detention). For this and other reasons, advocates wanting to argue that the conditions mapped here violate international law bans on torture may also face a steep uphill climb.
Those pursuing this legal pathway might find it more promising to try framing sleep deprivation in prison, not as torture, but as “cruel, inhuman or degrading treatment” (“CIDT”)—a category of treatment also prohibited under international human rights law. CIDT falls somewhat below torture on the schedule of human rights abuses, and includes “forms of punishments . . . that cause ‘serious mental and physical suffering,’ ” which persistent interference with sleep of the sort described here arguably does.
I leave it to experts in international human rights law to determine how best to categorize the conditions described here within CAT and other international human rights frameworks. Here, I would simply note that, at this stage, how this doctrinal debate (torture or only CIDT?) would come out under international law may matter less than the fact that, for many, this is the question that most immediately comes to mind when confronting the fact of systematic sleep deprivation in American prisons. That we find ourselves in a moral universe where these are the stakes provides the strongest possible evidence that the conditions identified here are among the features of American prison life most worthy of condemnation and challenge.
D. Policy: Challenges and Complications
The conditions chronicled in this Article are the product of multiple failures: of institutional design, of moral obligation, of constitutional protection and enforcement. In these ways, the phenomenon of sleep deprivation is no different from many other toxicities constitutive of the American carceral experience. All the unconscionable conditions that shape prison life, sleep deprivation included, have the same moral driver: an inability (or unwillingness) to recognize the humanity of the incarcerated, and a consequent callous indifference to their health and well-being. They also have considerable mutual interaction effects, each reinforcing and amplifying the harms prisoners endure.
This interconnection means that none of prison’s noxious aspects can be fully resolved while the others persist. Yet, however much we might wish it, we will not witness the wholesale transformation of American carceral practice any time soon. Of course, we ought still to push for sweeping change. But this hard truth means that, for now, the best we may realistically be able to achieve for those currently living behind bars is to ameliorate their suffering as much as possible.
When it comes to improving the conditions in which prisoners sleep, there are some obvious policy fixes. To name just a few: people should have real beds, decent mattresses and pillows, and sufficient blankets to keep them warm at night. Meals of reasonable quality and quantity should be served at appropriate times, and food should be available in the evening for people who are still hungry. Facilities should be air-conditioned in summer, well-heated in winter, and generally well maintained to ensure a temperate environment. Nighttime counts should be reduced and staff trained to respect prisoners’ sleep. Approaches to lighting should be found that do not interfere with sleep and leave some people feeling like “rotisserie chicken[s].” And ideally, people would have their own rooms, so that their sleep will not be disrupted by the movements and noises of a cellmate—or ninety-nine other dormmates. Even if policymakers balk at implementing the policy reforms just itemized, these are changes advocates should push for on every available front.
No doubt, these proposals and others in a similar vein would be condemned by abolitionists (and other advocates of sweeping change) as woefully inadequate. But given the actual distribution of power in this space, what is of more practical moment is that such proposals are also likely to be fiercely resisted by prison officials and policymakers. Such changes would not come cheap, and prisoners, largely viewed as undeserving across society at large, are not a constituency in which legislators are generally eager to invest. Although the desire of politicians to seem “tough on crime” has waned somewhat since its peak in the 1990s, there is still a palpable reluctance among policymakers to address even the most pressing needs of the incarcerated. This is not the place to fully address these troubling dynamics. Here, I would simply note that the state’s decision to incarcerate carries with it the duty to provide for the basic needs of people in custody. This is the state’s carceral burden. If efforts at economy produce conditions that deprive people of what Justice Powell once labeled “the minimal civilized measure of life’s necessities,” the state is obliged to reorient its priorities. If it finds the costs too great to bear, the appropriate response—both morally and constitutionally—is not to subject the incarcerated to dehumanizing and affirmatively harmful conditions, but instead to reduce the carceral footprint.
It is well beyond the scope of this Article to fully explore the policy challenges facing those seeking to ameliorate the sleep-disrupting conditions prisoners face. I will, however, flag one additional complication that advocates will invariably need to confront: efforts to improve the conditions enabling sleep in prison will at times run up against policy initiatives intended to promote unrelated dimensions of prisoners’ interests and needs. For example, as we have seen, one central source of nighttime disruption comes from COs conducting count and performing regular security checks. But in the view of many prisoners’ advocates, these rounds are vital to ensuring the health and well-being of those in custody, by preventing or disrupting suicide attempts and intra-prisoner violence and allowing timely intervention in the event of medical crises. In my view, this justification is questionable. Yet even assuming its validity, it is also the case that these nighttime checks exact a considerable cost to the health and well-being of the very people they are supposed to help—a cost measured in the disrupted sleep, every single night, of hundreds of thousands of people incarcerated in locked facilities around the country. It may be that nighttime rounds should
be maintained, but at the very least, their considerable downside effects on sleep should be taken seriously and addressed.
A similar tension will also confront efforts to improve sleep in prison by increasing access to single cells. Judging from my interviews, it is impossible to overstate the sleep-enhancing effects of having one’s own room. Given the choice, virtually everyone I spoke with would opt for a single. Yet since the 1980s, when the American prison population exploded, double-celling—that is, housing two people in cells designed for one—has been the norm. This practice allowed states to cram many more people into each facility, and solo housing became relatively rare. But although the size of the incarcerated population remains high, it peaked nationwide in the early 2000s and continues to exhibit appreciable declines. Among other advantages, this drop has somewhat eased pressures on housing and, in some jurisdictions at least, could allow for a greater number of people to be housed alone. However, for some advocates, the preferred response when jurisdictions reduce their numbers is not to reduce population density but instead to close facilities, thus shrinking the carceral capacity of the system. If overall decarceration were the only goal, seeking prison closures would make sense. Yet viewed through the lens of sleep, shuttering prisons carries a significant downside for those who remain behind: the missed opportunity to expand the number of people able to live—and sleep—alone. This too is a serious tension advocates must navigate when considering how best to fight for change in this arena.
This brief policy discussion has focused primarily on challenges likely to attend efforts to make concrete changes to the conditions affecting prisoners’ sleep. But as I have repeatedly emphasized, the phenomenon of sleep deprivation in prison cannot be fully understood independently of the moral character of the carceral enterprise, specifically, the normative hostility and callous indifference that drive institutional choices and shape official behavior. If those with the power to shape the carceral environment are blind to or unmoved by the humanity of people in custody, they will easily dismiss demands for the concrete changes that would help those inside get adequate sleep. Any strategy for achieving practical change on the ground must therefore include efforts to generate a broad cultural recognition of the shared humanity of the people society chooses to incarcerate.
Conclusion
It may be tempting for some to dismiss concerns with the quality or quantity of the sleep people get in custody as simply part of the loss of those so-called creature comforts that necessarily comes with a carceral penalty. But this notion misapprehends the stakes. Getting adequate sleep is not a matter of personal indulgence or a luxury only to be enjoyed by those at liberty to chart their own course. It is a basic human need, as fundamental to human survival and adequate human functioning as is access to food, water, and shelter from the elements.
For those who live in prisons, being denied the ability to sleep night after night takes a substantial toll. Most immediately, the experience is frustrating, infuriating, and even humiliating. Chronically insufficient sleep also inflicts substantial physiological, psychological, and cognitive harm. Yet when aspects of the American prison are singled out for condemnation, sleep deprivation is virtually never mentioned. In this Article, I have argued that chronic sleep deprivation should be counted as among the fundamental harms of incarceration, explicitly acknowledged as a core part of what makes prison so unlivable. And the claim might be put still more forcefully: the persistent inability of people in custody to get adequate sleep is not only a product of some of the most damaging and degrading features of the prison experience, but also in turn helps to produce them. Not to reckon with this endemic aspect of prison life is to miss a key driver of the toxicity of the prison environment.
Appendix A
Research Methodology
The findings presented in this Article are based on thirty-nine interviews conducted primarily over four months in 2023 with people formerly incarcerated in prisons across the United States. These interviews were of two types. Thirty-five were semi-structured interviews conducted on Zoom using a 200-item questionnaire developed for the purpose. These interviews were video-recorded after obtaining consent. Each research subject had spent at least four years in prison, and all but three had been released within five years of the interview. I established these parameters to ensure clearer memories and a depth of experience with the issues.
The remaining four interviews—which I came to label “ad hoc interviews”—were more free-form. Participants in these interviews were people I connected with who seemed to have insight and experience to contribute, but who did not meet the study parameters at the time or who otherwise seemed unsuited for formal enrollment. These interviews were audio-recorded after obtaining consent. To get context and perspective, I asked these individuals many of the same questions as appeared in the questionnaire, and their answers were entirely consistent with what I was hearing in the formal interviews. As a result, I wound up drawing on all thirty-nine interviews when analyzing the data.
The primary focus of the research, as well as the main source of the data, was the thirty-five formal interviews conducted using the 200-item questionnaire. What follows is a description of the protocol I followed in undertaking this core aspect of the research.
I began by compiling a draft questionnaire, initially shaped by many conversations with people with firsthand carceral experience as well as by my own knowledge of prison conditions accrued over years of study. I then enlisted three volunteers, each of whom had spent time behind bars (two in California and one in Louisiana), to do individual mock interviews based on my draft questionnaire. After each mock interview, the volunteer shared feedback on which aspects warranted revision. I also incorporated feedback from Joe Doherty, former Director of the Empirical Research Group at the UCLA School of Law, who advised me throughout this project on matters of research methodology.
After receiving IRB approval, I began recruiting participants. I used two main channels. First, I put out a call for participants through the networks I have developed as part of my long-standing research into prisons and prison conditions. Targeted recipients included the Prisoners’ Rights listserv, a national list populated primarily by advocates for the incarcerated. This call briefly described the study and the format of the interviews and asked recipients to share the information with anyone they knew who might fit the parameters. I also posted a notice on Twitter, advertising the study and inviting participants. In each notice, I included a link to a Microsoft form asking for basic demographic data and contact information. I also indicated that study participants would receive a $30 Amazon gift card. The idea here was to provide some compensation for the time people spent doing the interview, while keeping the amount low enough that it would be unlikely to draw people who would not otherwise be inclined to share personal details or revisit prior experiences likely to have been traumatic. Unfortunately, I did not anticipate that the up-front promise of a gift card might incline some people without prior carceral experience to pretend to have been incarcerated in an effort to receive this benefit. The first few iterations of the sign-up form proved to be a mix of genuinely qualifying individuals and fraudsters, and it took some time to figure out how to sort them out. Eventually, I began scheduling “pre-calls” with those who filled out the form. During these brief calls, I described the study and then explained that I had been hearing from people who were only pretending to be formerly incarcerated and that, before we could schedule an interview, I needed to hear enough to confirm that the person was in fact formerly incarcerated. In each instance, it took only a few sentences of the person recounting their carceral history for me to know they were in earnest.
Once I got through this rough patch, it did not take long to receive expressions of interest from more potential participants than I had time to interview. My aim was to build a research sample diverse as to race and roughly proportionate as to gender. The reason for seeking a diverse sample was to ensure that my findings would not reflect only the experiences of one group of people. I also sought a sample that had broad jurisdictional/geographic representation. I did so because, as is common knowledge among people who study prisons or who work in the field, carceral systems vary considerably across the country on a variety of metrics, including facility size, general conditions of confinement, and—most importantly—moral orientation toward the health, safety, and well-being of those in custody.
I started by interviewing people in the order in which they reached out, but as my sample took shape, I began to pick and choose among those who had filled out the initial form, seeking people who would broaden the sample on the metrics I sought. Some states were better represented than others in my participant pool, and at some point, I realized that a certain degree of jurisdictional depth would enhance the data. I therefore set out to interview four or five people from four states well-represented in my pool: California, Louisiana, New Jersey, and New York. This approach helped strengthen confidence in my findings, as there proved to be considerable overlap in what I was hearing regarding the conditions impacting sleep from these four very different prison systems. I then did my best to speak to people from as many other state systems as I could to build out the picture more broadly.
Each interview proceeded as follows:
First, prior to beginning to record, I briefly described the study and what types of questions the interview would cover. I explained that the person was free to decline to answer any question they would prefer to skip over. I also explained that if, at the end of the interview, they decided they did not want their data to be used in the study after all, they would be given an opportunity to retroactively withdraw their consent. Finally, I explained that the content of our exchange would be, not only confidential, but also fully anonymous. To that end, with the permission of each participant, I changed the name in their Zoom box to the random code I had previously selected for their interview, each taking the form FI.# (with FI standing for “Formerly Incarcerated”). Doing so ensured that even the person who transcribed the interview would not know the subject’s name. And once the interviews were transcribed, the (anonymized) transcript would be the only record of the interview. I then solicited consent to participate in the study and for the session to be recorded. Only after consent was granted on both fronts did we start the interview.
The interviews were semi-structured. On this approach, the questionnaire becomes the vehicle for introducing issues, any one of which may invite deeper inquiry depending on how the conversation goes. Each interview began with the interviewee providing their carceral history, including which facilities they were in, how long they spent in each, and the housing configuration of each. After asking them to estimate how much time they slept in each housing context, I asked one series of questions about issues that may have impaired sleep, and another about how if at all being chronically tired (or being surrounded by other chronically tired people) may have impacted their carceral experience or the functioning of the prison. A final set of questions addressed demographics and related issues.
Over the course of the research, I began to get a feel for which questions were most fruitful, and which yielded little of interest. Consequently, I stopped asking some questions altogether. I also started asking additional questions as salient issues emerged. This meant that, as to some issues—those that only emerged over time—I received input from fewer people than on those issues I had been asking about from the beginning. Among the questions that emerged in this way were:
- Did you ever have trouble sleeping due to the presence of insects, rodents, or other vermin in your cell or dorm? If so, tell me about it.
- Did you ever find it hard to sleep because of untreated physical pain? If so, tell me about it.
- Did you have any difficulty sleeping in the bunks because of above-average weight or height, or did you ever see others in this situation? If so, how did you/they handle it?
- Do you think staff were also tired/sleep deprived? If so, what makes you say so? If so, how did staff being tired affect your daily life or that of others in the prison?
During the initial process of drafting the questionnaire, an issue came up that I thought might be fruitful to pursue more systematically: the experience of people diagnosed with sleep apnea who used/tried to use CPAP machines in custody. I therefore designed a second, shorter questionnaire intended for use with this population. It contained targeted questions about sleep apnea and CPAP use, and also asked what I thought were likely to be the core questions for all interview subjects: those concerning how long and how well people slept, the specific obstacles to sleep, and how being sleep deprived or being surrounded by people who were sleep deprived affected their carceral experience. After conducting two such interviews, I realized that I would not get sufficiently consistent data on the sleep apnea/CPAP piece to be able to say anything reliable on these issues. I therefore folded these two interviews into my larger data set and ceased soliciting participation from those in this category.
A question arose early on about whether doing the interviews over Zoom might disadvantage some participants. But it quickly became clear that, in a post-COVID world, virtually everyone was comfortable with the Zoom platform. Occasionally the start of an interview would be delayed by technical difficulties, but these were readily overcome.
At the close of each interview, after I ceased recording, I gave each person the opportunity to offer any thoughts they might have on the experience we had just shared. Once or twice, the subject began talking about an entirely new issue, one we had not discussed during the interview. Each time this happened, I hesitated to risk interrupting the flow by asking to restart the recording. I therefore took as detailed notes as I could and immediately after ending the call recorded my recollection of the exchange as part of the field notes for that interview.
At the close of each interview, I also gave each person the opportunity to withdraw their initial consent to participate and not to have their data included in the study. No one took me up on this offer. After each interview, I followed up by email to share my contact information and the contact information for the UCLA Office of the Human Research Protection Program, in case they had any questions or concerns about the research that they preferred not to go through me. I also shared the unique code for the $30 Amazon gift card that had been promised to each participant as a token of thanks for their participation.
Once the interviews were completed, the recordings were shared with a team of research assistants (“RAs”) for transcription. There were two steps to the transcription process. First, each interview was handed off to an RA who ran it through Otter.ai, which produced an initial transcript that caught roughly ninety to ninety-five percent of the content. That same RA then went simultaneously through the recording and the initial transcript and filled in any missing pieces. Second, the interview was handed off to another RA on the team, who went through the recording and transcript to catch any remaining errors.
The coding process proceeded as follows: I created an initial, lengthy list of coding categories, each taking the form #issue (e.g., #noiseinadseg, #extremeheat, #mattresses). My RAs then went through each transcript, identifying any passages related to each category. Whenever they identified a relevant passage, they dropped the code in the interview text (producing a set of coded interviews searchable by issue) and also added the text to a master coding document that grouped interview excerpts according to categories. To ensure quality control, I reviewed the first set of coded transcripts and an early version of the master coding document. As they went, my RAs sometimes added to the initial list of coding categories. We also created tables, populated with any data amenable to being preserved in this way (e.g., race, gender, state of origin, length of incarceration, age at initial incarceration, age on release, weight on entry, weight on release, and so on). I then went through the master coding document to identify relevant issues and representative passages. I also frequently went back to the coded transcripts to read for context and confirm impressions. In this Article, I only quoted passages that were representative of what I heard more generally. Where the experiences varied, for example by security level or housing configuration, I indicated as much in the text.
Appendix B: Descriptive Statistics of the Sample
In this Appendix, I provide tabular representation of several aspects of the research sample amenable to quantification. Unless otherwise indicated, each table includes pertinent data for the thirty-five subjects formally enrolled in the study and the four individuals with whom I conducted ad hoc interviews.
Table 1. Racial Distribution (n=39) |
Identified as | # in sample | % of sample | Overall % in U.S. Prison and Jails |
Black | 15 | 38% | 42% |
White | 13 | 33% | 36% |
Hispanic | 7 | 18% | 20% |
Asian | 1 | 3% | N/A |
Native American | 1 | 3% | 3% |
Mixed Race | 2 | 5% | N/A |
Table 2. Gender Distribution (n=39) |
Housed in facilities for | # in sample | % of sample | Overall % in U.S. Prisons |
Men | 35 | 90% | 93% |
Women | 4 | 10% | 7% |
Table 3. Jurisdictional Breakdown (n=39) |
As this Table shows, the data on which this Article is based derived from interviews with people with firsthand experience of incarceration in seventeen state systems, as well as the D.C. Department of Corrections and the Federal Bureau of Prisons. In addition, six members of my sample served time in private prisons located in five different states. |
Alabama | 2 |
California | 5 |
Colorado | 2 |
Illinois | 2 |
Louisiana | 4 |
Michigan | 1 |
Missouri | 1 |
New Jersey | 5 |
New York | 4 |
Ohio | 1 |
Oregon | 2 |
Pennsylvania | 2 |
Rhode Island | 1 |
Tennessee | 1 |
Virginia | 3 |
Washington, D.C. | 2 |
Wisconsin | 1 |
Wyoming | 1 |
Federal Bureau of Prisons | 4 |
Private prisons | 6 (located in Arizona, California, Colorado, Ohio, Oklahoma, and Virginia) |
Table 4. Housing Configuration (n=39) |
The figures in this table represent the total number of study subjects who experienced each type of housing configuration. If, as was common, a subject experienced multiple housing configurations over their incarceration, then all these configurations were counted in this table. It was not always possible from the interview transcripts to determine with precision the full range of a person’s experience, so these numbers are most likely undercounts. |
Housing Type | Total # Interview Subjects Reporting Experience |
Solitary confinement | 29 |
Single cell | 29 |
Double cell | 30 |
3+ cell (range: 3–10 occupants) | 10 |
Cubicle in dorm | 12 |
Open dorm | 25 |
Table 6. Misc. Data Points |
| Minimum Value | Q1 | Q2 | Q3 | Maximum Value |
Hours of sleep per night (n=39) | 3 (n=3) | 4.5 | 5.1 | 6 | 11.5 |
Total years incarcerated (n=39) | 3 | 9.5 | 13.5 | 23.5 | 45 |
Interview length (n=35) (hours: mins) | 1:03 | 1:44 | 1:58 | 2:14 | 2:48 |
Years between release and interview | 0 | 1 | 2 | 4 | 12 (n=2) |
Table 6 features several data points as to which the answers reflected a range. To capture the breadth of responses, I used the standard “five number summary,” indicating as to each category the minimum value, Q1 (25th percentile/lower quartile), Q2 (50th percentile/median value), Q3 (75th percentile/upper quartile), and the maximum value. As to each data point, some further explication is in order, as follows:
Hours of sleep per night: The effort to calculate the hours of sleep per night my subjects received was complicated by many factors, including the long duration of the sentences served by many of my subjects and the wide range of experiences each had across housing configurations, not to mention inevitable variability night to night. I asked people to give me their best estimate of how many hours per night they slept on average in each housing configuration they experienced. When respondents reported variation across their incarceration, I took an average of the reported hours slept. If someone gave a range (e.g., four to six hours per night), but characterized the sleep quality as very poor or described themselves as waking frequently during the night, I took the low end of the stated range as more accurately reflecting the actual amount of sleep they likely received. These limitations mean that the numbers reported should be considered rough estimates only, although the considerable sleep deficit they indicate seems entirely consistent with the narrative accounts of the extremely limited and poor-quality sleep reported in the interviews.
The maximum value for hours of sleep per night is 11.5. The person who reported this outlier value served time in two prisons and reported sleeping eight hours a night in the first facility and fourteen to sixteen hours per night in the second. To calculate his hours slept for purposes of this chart, we took the average of eight hours (from prison #1) and the middle of the range he named for prison #2.
Total years incarcerated: In building my sample, I sought to enroll only people who had spent at least six years in prison. I wound up enrolling two individuals who had done four years each, because each offered diversity as to other metrics. The minimum value noted in the table (three years in custody) reflects the experience of one of the individuals with whom, in part for this reason, I opted to conduct an ad hoc interview rather than formally enrolling them in the study.
Interview length: For this data point, I included only the thirty-five interviews conducted with those individuals formally enrolled as subjects in the study. I did so because the four ad hoc interviews took a different form and did not follow the path shaped by the questionnaire. For reference, the four ad hoc interviews ranged in duration from 1:05 to 2:09.
Years between release and interview: The interviews were conducted in 2023. Two members of my sample were released that same year and were coded as having had zero years between release and interview. Two members of my sample were released in 2011, twelve years prior to the interview. This gap was considerably greater than my intended parameter of five years. Each of these individuals was interviewed very early in the process, when I was still unsure whether I would find enough participants to assemble an adequate sample. Once things progressed and I had more people expressing interest in participating than I had time to interview, I reverted to requiring a release date within five years of the interview. One other enrolled participant was outside this stipulated limit. He had entered the wrong release date when he filled out the initial form expressing interest in participating, substituting 2018 for 2015. When I enrolled him in the study, I believed he was five years
out, and was surprised when, at the end of the interview, he named 2015 as his year of release. Having already conducted the interview, I opted to retain his data as part of the full data set.
