Managing Prisons Through Extended Solitary Confinement: A Necessary Approach or a Signal of Prison System Failure?

Abstract

In contemporary American corrections, extended solitary confinement (ESM) as a management tool has emerged as a strategy for avowedly controlling the most violent individuals and, in so doing, creating a safer prison system. We theorize that the emergence of this unique form of housing may also be viewed as a signal of prison system failure. To advance this argument, we identify how different theoretical perspectives can be used to anticipate the effects of ESM on prison system violence and order and then investigate the plausibility of this account by grounding it in analysis of qualitative data from a study of one state’s prison system. The analysis suggests theoretical and empirical warrant for both views of ESM—as an effective tool and as a symptom of system failure. Implications of the study research and policy are discussed.

The rise of, and greater reliance, on solitary confinement over the past three decades both in the United States and in other countries has led to concerns about its potentially harmful effects on incarcerated persons, such as adverse effects on mental health, rule compliance, and recidivism, and whether such incarceration is humane (Garcia, 2016; King, 2005; Kurki & Morris, 2001; Labrecque & Smith, 2019; Shalev, 2009). Some studies raise questions as well about systemwide increases in violence and disorder that may arise from reliance on this housing (Briggs et al., 2003; Mears & Watson, 2006). Juxtaposed against such concerns is a belief among corrections officials that extended solitary confinement for managerial purposes, what we refer to as “ESM” for short, may create greater systemwide order through incapacitation and deterrent effects (Frost & Monteiro, 2016; Mears & Reisig, 2006; Riveland, 1999; Ross, 2013).

This argument has assumed even more salience in contemporary debates about extended solitary confinement because of accounts that suggest that some prison systems may increasingly face challenges in maintaining order. The U.S. Department of Justice, for example, recently investigated the Alabama Department of Corrections and found that “illegal drugs and weapons were rampant, cellblocks were overcrowded and dilapidated and the few poorly trained officers on duty appeared powerless to establish any semblance of control” (Johnson & Phillips, 2019). Published accounts identify similar conditions in California, Colorado, Delaware, Kansas, New Mexico, Texas, and other states (Fifield, 2016; Galvin, 2017; Irby, 2019; McCullough, 2019).

In such a context, ESM can be argued to be especially important and, indeed, necessary for system safety and order. Solitary confinement may sometimes be used for punishment or protective purposes, or to undertake administrative reviews to determine what actions to take with certain incarcerated individuals (Labrecque, 2016; Shalev, 2009). Each of these types of housing may or may not entail solitary confinement. As used here, ESM refers specifically to the placement of individuals in solitary confinement, typically for 22 hours or more per day for 6 months or longer, for the purposes of managing the incarcerated individuals and creating greater systemwide order and safety. ¹ Different terms, such as supermax incarceration, are used by states to designate such housing (Garcia, 2016; Mears et al., 2019; Riveland, 1999).

The argument for ESM is rooted in a broader punitive turn in crime and correctional policy (King, 2018; Pizarro et al., 2006; Pratt, 2018; Sundt, 2016). This shift guided early efforts to use solitary confinement as a managerial tool and remains a central guiding justification for continued and expanded reliance on it (Cochran et al., 2018; King, 1999; Mears et al., 2019; Kurki & Morris, 2001; Stickrath & Bucholtz, 2003; Ward & Werlich, 2003). A different view, however, stems from critiques of solitary confinement (King, 1999, 2018; Venters, 2019) and from accounts that emphasize the salience of such dimensions as governance, deprivations, and perceptions among incarcerated persons about the legitimacy of prison authority (Bottoms, 1999; Butler & Maruna, 2016; Crewe & Laws, 2018; Sparks et al., 1996; Steiner & Wooldredge, 2018; Sykes, 1958/2007). From these perspectives, reliance on solitary confinement can be viewed as a potential symptom of system failure and as a management tool that in the long term cannot effectively contribute to order or safety and may worsen it. Despite frequent debates about these competing views, there exist few empirical accounts that examine them from the perspective of those who work in prison systems. Such individuals may have greater insight into the realities and challenges of prison operations and of how to understand the role of ESM in contributing, or not, to order and safety.

The goal of this paper is to contribute to scholarship on prisons, solitary confinement, and the diverse implications of the punitive turn in corrections. We seek to illuminate in particular the contested nature of solitary confinement that is used for managerial or punishment purposes and to identify how it can be viewed both as a tool that effectively improves safety and order in prisons and as a symptom of prison system failure. To this end, we situate the study in the context of the broader movement toward reliance on punitive responses to crime and the manifestation of this trend in mass incarceration and greater reliance on punitive regimes, including the use of solitary confinement, within prison systems. We then discuss theories of behavior of incarcerated persons and prison order, and argue that they have direct implications for understanding the potential impacts of solitary confinement on order and safety in prison systems. Finally, we describe the qualitative study on which the paper draws to illuminate empirically how reliance on solitary confinement can be viewed as a necessary tool for maintaining order and safety in prisons but also as a symptom, or signal, of systems failure.

Background

The Punitive Era, Prisons, and Solitary Confinement

Approaches to crime and justice have evolved over the past two centuries in America (Friedman, 1993). However, beginning in the 1970s, policies became increasingly punitive at state and national levels. This shift can be seen in the more than quadrupling of the rate of incarceration over the last four decades (Travis et al., 2014, p. 1). It can be seen, more generally, in the enactment of get-tough sentencing policies, investment in punishment rather than treatment, and the creation of a spectrum of new punishments, including restrictions on the ability of felons to vote, secure employment in certain occupations, or reside in public housing. There was more generally a turn toward more retributive and less rehabilitative sanctioning and intervention (Beckett et al., 2018; Garland, 2013; Gottschalk, 2006; Simon, 2007).

The punitive era permeated all aspects of the criminal justice and correctional systems. In corrections, prison populations did not just increase, they increasingly served more time and did so under more austere conditions. Although the number of individuals incarcerated escalated during the get-tough era, investment in rehabilitative programming did not (Lynch & Sabol, 2001; Phelps, 2011). As Travis et al. (2014, p. 163) have observed, “Eventually, advocates of these more punitive policies began to focus explicitly on daily life inside the nation’s prisons, urging the implementation of a ‘no frills’ approach to everyday correctional policies and practices.” This approach included restrictions on educational, recreational, and rehabilitative programming, personal possessions, and other amenities, and a concomitant increased emphasis on harsher, more punishment-oriented treatment of incarcerated persons. “This ethos of vindictiveness and retribution,” as Johnson et al. (1997, p. 25) have emphasized, ran “counter to that of previous decades, which emphasized humane treatment of prisoners and the rehabilitative ideal.”

One manifestation of this change within prison systems can be seen in the rise of extended solitary confinement for management purposes, or what we refer to as “ESM.” Prison systems increasingly turned to ESM to manage individuals deemed to be difficult to control any other way and, through incapacitation and general deterrence, to increase prison system safety and order (King, 1999, 2018; Kupers, 2017; Kurki & Morris, 2001; Reiter, 2016). In the 1980s, states increasingly turned to lockdowns of prison units, which requires that incarcerated individuals remain in their cells until administrators determine that safety can be restored (Frost & Monteiro, 2016; Ward & Werlich, 2003).

In the ensuing decades, they built wings and stand-alone units dedicated to super-maximum, or “supermax,” confinement (Shalev, 2009). Not all states used this terminology, but almost all of them substantially increased their investment in housing that entailed the same types of conditions—including isolating one individual per cell for 22-or-more hours per day with limited visitation or other types of privileges and minimal access to programming—and use of it to control those deemed to be unmanageable.

What differed primarily from the past was the reliance on this type of incarceration as “normal routine” (Riveland, 1999, p. 5). A shift in terminology in the last decade has on occasion led to description of solitary confinement as “restrictive housing” (Beck, 2015; Garcia, 2016). However, the latter housing allows for the possibility of two individuals to share a cell and thus in these instances does not include isolation, and it includes housing that is used for any of a variety of purposes, including management, punishment, protection, and administrative goals (Mears et al., 2019). Regardless of the terminology, incarcerated persons in contemporary corrections, especially since the 1980s and the growth of supermax prisons, appear to be subjected to more control- and punishment-oriented approaches. Short-term solitary confinement—or restrictive housing that does not entail isolation but does entail many restrictions—for punishment has become a central part of American corrections (Garcia, 2016). And reliance on long-term solitary confinement for management purposes, or ESM, has become institutionalized into state and Federal prison system operations (Frost & Monteiro, 2016; King, 1999, 2018; Reiter, 2016; Ross, 2013; Shalev, 2009).

As with the shift toward punitive criminal justice policies, the greater reliance on long-term solitary confinement was justified in large part on pragmatic grounds—it was needed to control the most violent and difficult individuals and to ensure that the prison system operated in an orderly and safe manner (Riveland, 1999; Stickrath & Bucholtz, 2003). Yet, reliance on such confinement clearly reflected a turn toward the “ethos of vindictiveness and retribution” that scholars observed as prison systems evolved throughout the get-tough era (Johnson et al., 1997, p. 25; see, generally, Kupers, 2017; Rhodes, 2004; Shalev, 2009; Venters, 2019).

In fact, the justification was broader than one of managing violent individuals. Prison system officials thought that use of solitary could improve prison order and safety through a variety of mechanisms, such as general and specific deterrence, incapacitation, and normalization of prison system routines (Mears & Reisig, 2006). These justifications have continued, but the empirical basis in support of them remains limited and contested. For example, few studies have empirically examined systemwide impacts of the housing. Some, such as Crouch and Marquart’s (1989) study of Texas’ supermax housing, suggest that it may reduce violence. Others, such as Briggs et al.’s (2003) four-state study, point to mixed results, with some evidence for potential increases in violence, no effect on violence, and decreases in it (see, generally, Sundt, 2016). Still other work identifies different competing dynamics, some helpful and some harmful, that would produce such results (e.g., Mears & Watson, 2006).

Although studies point to potential harmful effects on individuals, such as increasing misconduct or recidivism or worsening mental health symptoms, there exists debate about what research to date shows. Some reviews suggest that solitary confinement contributes to misconduct and recidivism as well as to mental illness (see, generally, Garcia, 2016; Haney, 2018; Kupers, 2017), yet different conclusions have been reached in other reviews and studies. For example, Mears and Bales (2009) found that supermax incarceration was associated with violent recidivism, while Lovell et al. (2007) did not. Similarly, while Kapoor and Trestman’s (2016) review found that individuals in the housing are more likely to exhibit maladaptive behaviors and characteristics, other reviews suggest that the evidence for such adverse effects is weak (Gendreau & Labrecque, 2018; Labrecque & Smith, 2019). Morgan et al.’s (2016, p. 17) study, which synthesized results from two meta-analyses, concluded that harms to inmate mental health were “in the small to moderate range.” Labrecque and Smith (2019, p. 300) undertook a systematic review and concluded that various types of restrictive housing have a “null to weak effect on behavioral outcomes,” but emphasized the limited number of studies that have examined them. A central concern in many studies is that relatively few rely on strong research designs that incorporate comparison groups and controls. This concern is especially relevant given that Morgan et al.’s (2016) review highlighted that there were smaller effects among studies that relied on more rigorous research designs.

The mixed results reflect in part the paucity of studies that employ rigorous research designs that would allow for adjudicating not only the range but also magnitude of impacts of solitary confinement (Frost & Monteiro, 2016; King, 2005; Pizarro & Narag, 2008; Shalev, 2009; Steiner & Cain, 2016). Further complicating matters is that studies frequently conflate what amounts to different types of housing, such as incarceration that permits two individuals to a cell versus only one and incarceration for short durations, such as for punishment, versus incarceration for longer durations, such as to manage individuals deemed to be unmanageable any other way (Labrecque et al., 2021; Mears et al., 2019).

Theories of Prison Order and Safety: Implications for Assessing Solitary Confinement

Although the bulk of studies to date have focused on incarcerated individuals placed in ESM, a systems focus remains of particular relevance given that this practice has been justified in part on the basis that it allows for the orderly and safe operations of facilities throughout the prison system, and, in particular, greater protection to officers and incarcerated persons (Garcia, 2016; Mears & Reisig, 2006; Riveland, 1999; Stickrath & Bucholtz, 2003). Proponents claim that it normalizes the prison environment and provides a deterrent effect (see, e.g., Hershberger, 1998). Few accounts, however, have examined the theoretical logic for how reliance on ESM might increase or decrease systemwide order and safety.

One exception is a conceptual analysis by Mears and Reisig (2006). They identified arguments for how normalization and deterrence might run in both directions and for why the anticipated theoretical pathways were questionable. They noted, for example, that specific deterrent effects might arise, but could be offset to the extent that incarcerated persons seek to be placed in solitary confinement to accrue a reputation of toughness (see, e.g., Laws, 2021). Morris (2016) has highlighted that direct or indirect exposure to solitary confinement may have deterrent effects at the individual level, while Pizarro and Stenius (2004, p. 258) have highlighted that although deterrent effects may occur, they may be unlikely because “the perceived certainty of placement in supermax facilities is likely to be low.” A similar argument can be made for incapacitation—it may reduce violence and disorder if it prevents individuals from aggression or inciting misconduct, but such a benefit may be offset by replacement effects, as when gang leaders placed in solitary confinement and other gang members take on their leadership roles. Finally, Morris (2016), citing Sherman’s (1993) defiance theory, and others (e.g., Ward & Werlich, 2003) have argued that if individuals placed in solitary perceive the placement as unfair or the treatment in it as inappropriate, they may act out either while in solitary confinement or after release. Conversely, a similar theoretical line of reasoning suggests that should they perceive the placement as appropriate, those incarcerated in the housing may be more likely to comply with prison rules (see, generally, Bottoms, 1999).

A different approach from focusing on individual-level theoretical arguments is to apply established theories of prison order to extended solitary confinement to understand its potential impacts on prison systems. Several of the most prominent theories include Clemmer’s (1940) theory of inmate prisonization, Sykes’ (1958/2007) account of deprivation and its effects, importation theory (Irwin & Cressey, 1962), work on the salience of governance (DiIulio, 1987), and accounts that emphasize the importance of incarcerated persons’ perceptions of prison authority legitimacy (Bottoms, 1999). Prisonization is “the taking on, in greater or lesser degree, the folkways, mores, customs, and general culture of the penitentiary” (Clemmer, 1940, p. 270). That includes learning to accept an inferior status and how to dress, eat, and act. From this perspective, the nature of prisons can lead to submission, but it can as well lead incarcerated persons to become accustomed to violence, defiance, manipulation, and “acceptance of the dogmas and mores of the primary group and the general penal population” (Clemmer, 1950, p. 318). Sykes’ (1958/2007) model extended Clemmer’s (1940) work and points to the ways in which incarcerated persons react to the pains of imprisonment. Viewed in this light, prisonization reflects an adaptation to the deprivations that inhere to prison life, which in turn encompasses adoption of an inmate code that facilitates survival and opposition to prison authority. Importation theory highlights that behavior among incarcerated persons stems from pre-existing characteristics, including attitudes and beliefs, that individuals bring with them into incarcerative settings. DiIulio (1987), as well as other scholars (e.g., Reisig, 1998), has drawn attention to the salience of management. He identified three models: control (the operation of prisons in a paramilitary-like manner, including strict adherence to enforcing rules), which he argued was most effective, responsibility (reliance on informality between staff and inmates, officer discretion, and minimal restrictions on inmates), and consensual (a mix of the control and responsibility models). Finally, a large body of work underscores the importance of legitimacy for prison order. From this perspective, when incarcerated persons view authority as wielded in a fair and consistent manner, they are more likely to comply with rules (Bottoms, 1999; Crewe & Laws, 2018; Sparks et al., 1996; Steiner & Wooldredge, 2018).

Collectively, these theoretical perspectives suggest that systemwide order and safety result from a constellation of factors. Viewed in that light, for ESM to be effective, it would need to foster a prison experience that promotes compliance with rules and to provide deprivations so severe that they create a deterrent effect. It also would need to be used only with the most violent or disruptive incarcerated persons. It would need to free prison system personnel to enforce rules consistently, per DiIulio’s (1987) administrative control model. Not least, it would need to occur in a manner that led incarcerated persons throughout the prison system to hold a more favorable view about the legitimacy of the wardens, staff, and officers. Under the opposite conditions, greater disorder and violence would be expected. For example, if the ESM experience socializes individuals into, or entails deprivations that contribute to, defiance and hostility toward prison authority, then they likely will act out both during and after release (Morris, 2016; Pizarro & Stenius, 2004). They then “import” their attitudes, beliefs, and behaviors into general population facilities. If “nuisance” individuals, or incarcerated persons who do not fit the targeted “worst of the worst” profile, are housed in ESM, this problem would be amplified. In addition, if individuals in ESM and in general population facilities view ESM as used in an inconsistent manner, they may act out more, in what amounts to an anti-general deterrent effect. And, finally, if those in ESM and throughout the prison system view officials and officers as acting in an illegitimate manner—in general or in their use of ESM—then they can be anticipated to become more defiant and violent rather than less so.

Juxtaposed against these theoretical stances for understanding the impacts of long-term solitary confinement on system order and safety is the fact that little empirical work exists that sheds light on whether and how ESM may constitute a necessary feature of a well-managed and safe prison system or may contribute to the very problems that it seeks to address. Although many studies have relied on administrative records data to examine the use and impact of ESM, such studies have not examined system-level impacts. The few administrative records studies to do so (e.g., Briggs et al., 2003) have advanced the literature but have not documented the different ways that beneficial or harmful effects of ESM may arise. Finally, while some accounts have identified potential impacts of ESM on system operations, these have been theoretical (e.g., Mears & Reisig, 2006) or have not systematically and empirically examined the potentially competing effects (e.g., Mears & Watson, 2006; Rhodes, 2004). There remains a need, therefore, to empirically investigate the different possible ways in which ESM may improve or worsen prison system order and safety and to empirically identify their plausibility.

The present study addresses this research gap by investigating how ESM may be a necessary approach to effective prison system management and, alternatively, how it may signal prison system failure. To this end, we rely on a qualitative study that draws on the views of those who oversee and work in prisons and thus have a unique vantage point from which to gain insight into how such confinement may signal an effective or ineffective prison system.

Data and Methods

Our study draws on qualitative data collected from focus groups, interviews, and observations for a federally funded project examining the broader effects of extended solitary confinement for management purposes in the Florida Department of Corrections (FDC). The data collection occurred from 2017 through 2020. One hundred forty-four personnel across ten prisons, and including two individuals with statewide administrative positions, participated in the study. All of them worked in ESM units or managed individuals who had spent time in ESM. In Florida, ESM consists of “close management 1” housing, which at the time of the study existed in five of the state’s correctional institutions. This housing typically entails a minimum 6-month stay, serving almost the entirety of each day alone in a cell, with few privileges. The housing serves to control or manage individuals deemed not to be manageable any other way and who constitute a threat to system order and safety. It is similar to accounts of ESM in other states, though the specific terminology to describe the housing may vary (Butler et al., 2013; King, 1999, 2018; Kurki & Morris, 2001; Labrecque et al., 2021; Naday et al., 2008; Riveland, 1999).

The sample included personnel with varying roles within the prison system and whose experiences positioned them to provide insight into the impacts of solitary confinement on the prison system. They were selected to obtain diverse perspectives about the uses and impacts of this specialized housing. For example, correctional officers work 12 to 16 hours a day, with much of their time spent directly managing incarcerated individuals. Medical and mental health personnel attend to the physical and mental health needs of incarcerated persons. Classification personnel make decisions about prison job and housing assignments; they therefore are familiar with each individual’s case. Wardens and assistant wardens oversee the daily operations of the prison and interact with incarcerated persons on a nearly daily basis. They also play an intricate role in housing placement decisions. In addition, program staff and system administrators occupy positions that give them different vantage points from which to understand the role, operations, and impacts of ESM.

Personnel were provided with an informed consent document approved by the Institutional Review Board at the researchers’ university, which explained the purpose of the study, and were asked if they would like to participate in the study. Data collection occurred in two steps. First, upon arrival to each of the prisons, we toured and observed the solitary units as well as general housing units and, for some visits, the recreational and visiting areas. We took notes of our observations during and after each tour. Second, we conducted semi-structured focus groups and interviews with the prison personnel. Twenty focus groups occurred across the ten prisons, with two occurring at each prison. One focus group typically included junior ranking correctional officers while the other typically included senior correctional officers, classification personnel, and medical and mental health personnel. Focus groups lasted 1 to 2 hours and included five to ten respondents. Interviews with wardens, and sometimes their deputies, typically lasted 45 minutes.

Following recommended guidelines for qualitative research (Krueger & Casey, 2014; Morgan, 1996), we used a semi-structured approach to ask questions about respondent’s experiences managing incarcerated persons either currently in or who had been in solitary confinement. We also asked questions about their experiences working within the prison system in general. Due to the level of prison security, recording devices were not allowed. Therefore, hand-written notes were taken. Throughout the focus groups and interviews, at minimum, two researchers took notes, while one researcher served as moderator.

After each visit or interview, the members of the research team compared notes and discussed major themes that surfaced. Within 24-hours of each site visit, each notetaker transcribed their own handwritten notes and then the two notetakers compared the transcribed notes for accuracy. Next, one researcher synthesized both sets of notes into one document. All three researchers then reviewed the document for accuracy and completeness.

For this study, we used thematic analysis to identify themes and insights about the contribution of ESM to prison order and safety (Bryman & Burgess, 2002; Charmaz, 2006; Krueger & Casey, 2014; Lofland et al., 2005). We do not and cannot demonstrate causal effects. Rather, we seek to show, and what the qualitative nature of the study affords an opportunity to illuminate, is how ESM can be viewed from different perspectives and how its effects, as theory and research on prisons suggest, may be variable in predictable ways.

In what follows, we examine first the ways in which respondents identified reliance on ESM as a necessary means by which to maintain order and safety in the prison system. We then turn to ways in which they identified that such reliance reflected a failure of the prison system to operate effectively and how it potentially contributes to less rather than more order and safety. In each instance, we proceed by identifying ways in which site visits, focus groups, and interviews for the study lent support to as well as countered claims about the effectiveness or ineffectiveness of ESM in contributing to order and safety. The findings are organized around different theoretical perspectives and then around a final perspective, “systems problems,” because it emerged as a prominent theme during the course of the study. Accordingly, the analyses proceeded in a “top-down” and “bottom-up” approach. We anticipated that these different theories may help to illuminate ways in which extended solitary confinement as a management tool may be effective or ineffective. We then revisited the qualitative data to determine whether insights from the participants lent support to one or another theoretical perspective about the effects of ESM. In addition, we responded to insights from the study participants that led to a focus on “systems problems.”

Results

Extended Solitary Confinement as Necessary for Maintaining Order and Safety

Many of the study participants reported that ESM (close management 1 housing in Florida) contributes to system order and safety through incapacitation and deterrence. Focusing on incapacitation, one administrator emphasized, “Some inmates are just that vicious,” and so require extended confinement in solitary housing to protect staff and other inmates. “[ESM] is,” an officer stated, “one of the safest units to work in.” Another reason is that they viewed the housing as effective in providing a specific and general deterrent effect. Inmates, in their view, underestimate the restrictions in ESM and come to view compliance as the best option for reducing the deprivations they face. It was thought that ESM was most effective for inmates who are the most violent and have difficulty interacting with others; they were viewed as best able to function and follow rules in a highly controlled setting. Participants also said that inmates in general population facilities feared being placed with the “animals” in ESM and that “most inmates just want to do their time without any issues and go home.” What is unique in the present study is the pervasiveness of these views from respondents across the different prisons.

During the focus groups and interviews, many if not most participants viewed the new generation of inmates as “jits”—in contrast to “convicts” who want to do their time and get out—that is, as younger, more impulsive and violent, and less deterred by traditional punishments, such as disciplinary confinement or suspension of privileges. They viewed ESM thus as a critical approach for managing the “jits.” They also viewed it as a critical tool for controlling increasingly problematic gang activity and related violence and contraband. Solitary was said to interrupt gang influences. As one staff member stated, “We can’t completely stop the issues, but we can control them with [ESM].” This, too, was a pervasive theme across the different focus groups and interviews. Respondents described a systems context that in their view fundamentally differed from what existed in previous decades. And they highlighted that one of the most critical aspects of that context was the seeming incorrigible or unchangeable nature of a new generation of younger inmates.

Participants also highlighted that the availability of ESM increased their authority with inmates. Without it, inmates would cycle in and out of short-term stays in disciplinary segregation. With it, inmates know that officers are not issuing empty consequences—threats and violence can result in long-term stays in solitary confinement. In addition, officers gain greater credibility with general population inmates by protecting them from the most violent or, as one officer emphasized, “deviant,” prisoners. An administrator underscored that “getting rid of [ESM] would be like getting rid of the county jails”—that is, it serves a critical role in normalizing the day-to-day routines of the prison system and enabling it to operate effectively. ESM also sends a message to officers and staff that officials prioritize their safety.

Part of that normalization entailed a greater ability to interact with general population inmates. Across the different focus groups and interviews, respondents stated that dangerous, violent, disruptive inmates consumed a great deal of their time. As a result, personnel reportedly had little time for any meaningful interactions with other inmates. ESM was said to provide a means by which to at least somewhat mitigate this problem.

Throughout the focus groups and interviews, another theme consistently emerged—participants felt that ESM was essential in a context of declining staffing levels and programming. Administrators and front-line officers alike pointed to staff vacancies of 10% to 15% or more as highly problematic. They reported that inadequate staffing includes too few, and inexperienced, personnel. Less experienced staff have little control over the inmates. At one facility, for example, two officers had worked in corrections for only 18 months but were viewed as “senior” staff. The inmates push limits because they know what they can get away with while avoiding consequences. One officer stated, “Inmates do not mess with experienced officers who go by the book. Inmates know which officers they can push and which officers they can’t.”

Limited staffing also limits the ability to conduct random contraband checks, which respondents said allows more contraband into the prison system. This problem was especially concerning to respondents. They said that dramatic increases in contraband led to substantially greater inmate access to cell phones, illegal drugs, and weapons, which reportedly have led to increases in violence, drug-related injuries, and assaults on staff. Limited staffing also reportedly precludes having time to develop relationships with and an understanding of the idiosyncrasies of individual inmates. The end result is poor staff morale. One respondent noted about staff retention, “The only thing holding this [system] together is the benefits.” Recourse to ESM in such a context was almost universally viewed therefore as an essential tool for incapacitating the most violent inmates and sending a deterrent message to others when approaches that rely on adequate staffing and programming are largely precluded.

Extended Solitary Confinement as a Symptom of Systems Failure

Participants indicated that ESM served a critical role in maintaining an orderly and safe prison system. However, many of them also consistently pointed to a seeming contradiction—namely, that reliance on it could and did contribute to problems and that increased reliance on it reflected a failure of the prison system to operate effectively.

One prominent example was that stays in ESM were viewed as sometimes contributing to a culture of defiance and manipulation. Participants noted that inmates in ESM have nothing else to do but to identify officers’ weaknesses as a source of “mental stimulation.” Staff reported that when an inmate in ESM becomes upset with staff, he or she will say, “If I don’t get what I want, I’m going to make you work today” by claiming to have psychiatric emergencies, filing complaints against staff, and pitting officers against one another. Gangs reportedly commit violence to be sent to ESM so that, in turn, they are released to designated transitional facilities where other gang members reside. Officers reported that other inmates do similar “check-in moves,” committing an infraction that they know will land them in ESM so that they can avoid association, or sharing a cell, with other inmates. Here, then, ESM serves as a tool for and a means of learning how to manipulate prison rules.

Several participants emphasized that sending disruptive inmates to ESM does not create a general deterrent and that, due to substitution—such as when a gang member steps up to replace a gang leader—there is little in the way of an incapacitation effect. In making these points, some participants highlighted that larger forces, such as inadequate staffing and limited programming, serve as the primary factors that contribute to systemwide order and safety. Such observations were linked to a dynamic view of the role of ESM. For example, although many participants said the new generations of inmates were more impulsive and unmanageable, they saw ESM as sometimes worsening that problem. One respondent stated, “[ESM inmates] are not more manipulative than general population inmates, they are just better manipulators.” Those in ESM reportedly learn much more quickly different techniques for manipulating staff and the prison system. A stay in ESM also is sometimes reportedly viewed as a symbol of toughness and status. To illustrate, a general population facility officer described an inmate who said, “I just got off [ESM]. You can’t break me.” Another stated that some individuals sent to ESM come to accept the idea that they are “bad” and then, when returning to general population units, “wear [the stay in ESM] like a badge of honor.” Ironically, then, housing intended to send a deterrent message could serve to engender the very behavior it sought to reduce, and it did so in a manner that could increase not only misconduct among those sent to ESM but also to others who might come to see ESM in a similar light. Put differently, it could have an anti-specific deterrent and anti-general deterrent effect.

Study participants consistently pointed to inadequate staffing and insufficient inmate programming as a root cause of disorder and violence. Without a correction to this problem, ESM was viewed as a necessary tool for operating the prison system. What is perhaps most striking about this finding was the fact that although most respondents expressed the view that ESM inmates differed markedly in being more violent and impulsive than other inmates, many also said that there are ESM inmates who can be rehabilitated. They noted, though, that without sufficient programming—a commonly identified problem in ESM—rehabilitation could not occur.

Not surprisingly, given the importance that they placed on staffing and programming, study participants emphasized that ESM did not substitute for staffing or programming shortfalls. They also highlighted that these shortfalls contributed to the need for ESM. For example, at one facility, almost two-thirds of staff reportedly had worked in corrections for less than 2 years. In prior decades, promotions could not happen until an employee had 5 years of on-the-job experience. Promotions were said to now occur much sooner. Reliance on inexperienced “senior” and front-line officers reportedly creates opportunities for inmates to violate prison rules. These officers also were seen as more likely to escalate incidents in ways that ultimately would lead to a referral to ESM.

Compounding this problem was a high rate of staff turnover, which reportedly has led to inconsistent rule enforcement, limited officer rapport, and a reduced ability among officers and staff to defuse potential conflict among inmates. In such a context, increased reliance on ESM as a panacea was viewed as problematic because it diverted attention from addressing staffing problems, limited programming opportunities for inmates, and the ability to rely on privileges to motivate compliance with prison rules. According to one participant, “If we had more staff, we would not need [ESM] as much because we could manage inmates better and stop issues from escalating.” Another emphasized that incentives within solitary are essential. Without them, inmate behavior was held to worsen; with them, inmates’ behavior was held to improve. As one administrator stated, “Positive reinforcement is key to control.” In short, with “[general population] inmates being warehoused,” as a participant put it, and with ESM units providing minimal programming, the end result was viewed by some study participants to be a worsening of, rather than an improvement in, system order and safety.

Viewed in this light, ESM not only was seen as contributing to problems with order and safety, but also as reflecting a systemic failure to operate the prison system in a way that would more effectively promote these outcomes. Many of the study participants emphasized that with fewer “carrots” (e.g., privileges) and intermediate “sticks” (e.g., random cell checks, reductions in various privileges), the prison system must rely more on ESM as a tool of punishment and management. Across several of the focus groups and interviews, participants pointed to an irony. Greater reliance on ESM creates waiting periods for entry. Its deterrent value dissipates. There is then nothing left to motivate or deter inmates. In this situation, the prison system has run out of programming, services, treatment, adequate staffing, and punishment possibilities, and so must rely even more on ESM, which in turn contributes to a vicious cycle that reinforces this very situation. To our knowledge, such observations have not been identified in prior work and suggest that personnel, from front-line officers to senior administrators, see ESM as a problematic feature when they exist without adequate staffing to prevent their need in the first place.

Conclusion

Despite the widespread use of extended solitary confinement to manage incarcerated individuals deemed to be unmanageable any other way and, in turn, to improve systemwide order and safety, there exist few accounts that examine (a) whether ESM in fact improves or worsens these systems outcomes or (b) the empirical plausibility of how these impacts occur. To address this research gap, we drew on the insights of those who work in corrections to illuminate its potential impacts on system order and safety and how these impacts arise. Responding to calls to understand better the theory, practice, and effects of solitary confinement (e.g., Frost & Monteiro, 2016; Kupers, 2017; Mears et al., 2019; Ross, 2013; Shalev, 2009), this study identified theoretical grounds for anticipating that reliance on ESM may be a necessary strategy for maintaining safety and order in prison systems and, conversely, for viewing it as a signal of system failure and as a practice that itself may contribute to the very problem that it is designed to address. We then drew on a qualitative study that sought insights from those who work in ESM (close management 1) and general population facilities in Florida, and who could offer views from the “top” (e.g., wardens, deputy wardens, colonels) and from the front lines (e.g., corrections officers, sergeants, mental health staff).

We identified support for the view of ESM as necessary to prison order and for the view of such confinement as an indicator of prison system failure. Indeed, across all the site visits and almost all interviews, respondents provided a nuanced account of ESM that reflected support for both views. On the one hand, ESM was characterized as essential. Respondents expressed in no uncertain terms that the prison system would fall into chaos without the ability to use such housing to manage inmates. This view aligns with Labrecque and Mears (2019) study, which found that objective risk factors, such as misconduct, contribute to the use of solitary confinement. There is, though, an important caveat. In the present study, almost all respondents, from administrators to front-line staff, viewed ESM as necessary given the conditions and staffing levels in the prison system. They emphasized that with proper staffing levels—including more, and more experienced and better trained, officers and staff—there would be far less need to rely on it. However, that observation was followed by an equally emphatic assessment that understaffing would not change and so the use of ESM would remain essential to maintaining order and safety through incapacitation of what they viewed as the most violent and disruptive inmates and through assumed general deterrent effects on the general inmate population. Although some prior work suggests that personnel view ESM as essential (e.g., Mears & Castro, 2006), it has not illuminated the conditional support that may exist for this view. That is, personnel might be less inclined to see ESM as a critical part of corrections if they believed that alternatives, such as improved staffing and programming, could reasonably be expected.

A second major finding is the critical role of staffing and its salience for understanding the use and effects of ESM. Respondents echoed the insights of theoretical accounts of prison order that emphasize the salience of staffing to operating a prison system. For example, all three governance models—control, responsibility, and consensual—identified by DiIulio (1987) place a premium on sufficient staffing. Without it, misconduct and disorder occur. Similarly, sufficient staffing must exist to enforce rules fairly. Without it, staff cannot do so, and incarcerated persons come to view prison authority as illegitimate, resulting in less compliance (Lombardo, 1989; Sparks et al., 1996). To illustrate, respondents highlighted that with more experienced staff and training, there is a greater ability to enforce rules consistently, an approach that can foster greater deterrence but also contribute to a greater likelihood that inmates will view prison authority as legitimate and in turn comply with rules. They highlighted as well that more programs and services could be provided, thereby reducing some of the deprivations of life in prison. They noted that with more and well-trained and experienced staff, there is a greater ability to use balanced approaches to intervene with inmates. Rather than rely on “sticks,” they can rely more on “sticks” and “carrots,” or, in behavioralist terms, a combination of negative and positive stimuli. In short, this study of a contemporary correctional system reinforces theoretical arguments that place staffing at center stage (see, e.g., Bottoms, 1999; DiIulio, 1987). It also highlights that at least some who work in prison systems want to approach order and safety maintenance in a way that does not rely solely on punitive measures, but that they see little alternative without sufficient staffing. Research to date has not systematically sought insights about this issue from personnel, and the present study is, to our knowledge, the first to document the ways in which personnel view staffing as so centrally implicated in the need for, use of, and (in)effectiveness of ESM.

Third, the study points to the potential harm that can come from long-term reliance on ESM as a substitute for adequate staffing and programming. Respondents did not view solitary as an effective substitute; rather, it simply was a logical last resort when no other levers exist to motivate compliance with rules. At the same time, reliance on it in lieu of adequate staffing and programming contributes to a need to rely increasingly on control-oriented strategies for managing inmates. That leads to greater inmate deprivations and to “prisonization,” which can contribute to more misconduct. Inmates who leave solitary confinement may, for example, create problems for the facilities to which they return. That may result from any of a variety of mechanisms, but not least would be defiance bourne of concern about unfair placements. Consider the Labrecque and Mears (2019) study, which found that ascriptive characteristics (e.g., gender and age) contribute to placements in solitary confinement. Should individuals feel that they are singled out for ESM, or that they receive differential treatment while in it, they may be more likely to act out both during and after release from ESM. It is possible, too, that some individuals may actively seek ESM placement (Laws, 2021). With such possibilities, excessive reliance on ESM in lieu of adequate staffing and programming, as well as a more balanced use of incentives and punishment, can delegitimize prison authority, leading to reduced cooperation and compliance among incarcerated persons. Not least, it may undermine deterrent effects, especially when these individuals know that stays in ESM are unlikely (Mears & Reisig, 2006).

These findings accord with several theoretical and empirical accounts of prison systems and of solitary confinement. For example, writing about Rikers Island, home to one of the largest jail systems in the world, Venters (2019, p. 157) has described the ways in which systemic problems led to a “deeply ingrained ethos that relies on violence and intimidation to conduct the business of the jails.” This culture, Venters (2019) argued, contributed to organizational pathologies, including a situation in which the health service was turned “into a tool of the security service” (p. 74). His account documents structural factors, like staffing and organizational culture, that contribute to the behavior and health risks of incarcerated persons.

Our study echoes this assessment and that of a broader literature on prisons. When systemic failures exist, the harms—such as increased misconduct and violence or injuries to staff or incarcerated persons—may result more from the structure and operations of prison systems than from the individuals who are controlled by or work in them (Crewe & Laws, 2018; Kurki & Morris, 2001; Mears, 2017; Sparks et al., 1996). That does not mean that governance alone matters (DiIulio, 1987). It means that how prisons are operated—including their ability to operate in ways that incarcerated persons view as legitimate—the resources that they have available, the training and experience of staff, and the availability of programming may play a role and likely need to adapt to changing prison populations (Labrecque & Mears, 2019; Sundt, 2016). A unique contribution of the present study is to identify ways in which reliance on ESM may signal not only a different, get-tough orientation toward corrections, as many research accounts suggest (e.g., King, 2018; Rhodes, 2004; Shalev, 2009), but also may reflect a necessity that stems from over-extended, understaffed prison systems.

The findings accord as well with the assessment that while sometimes ESM may provide a short-term benefit (Briggs et al., 2003; Mears & Castro, 2006), greater reliance on it may worsen rather than improve order and safety in the long term. As Bottoms (1999, p. 270) has said, “To impose additional physical restrictions, especially of a severe character, will almost certainly lead to a legitimacy deficit; and that deficit may well in the end play itself out in enhanced violence.” This insight is important, as it does not imply that extended solitary confinement is intrinsically a practice that contributes to disorder and violence. Rather, it implies that when used appropriately and in limited instances, it may promote order and safety. And when used inappropriately and on a large scale—and in place of investment in improving the structure and operations of the prison system—it instead may lead to more violence and rule violations among incarcerated persons.

Structural problems, such as understaffing, may contribute to persistent problems that become increasingly difficult to address. In our study, most of the respondents dismissed out of hand the idea that staffing levels and quality would appreciably improve. Their lived experience said as much. This situation has the potential to activate several different theoretical pathways to greater disorder and safety. For example, it forces personnel into more of an “us” (e.g., officers) versus “them” (“inmates”) approach, which can delegitimize them in the eyes of incarcerated persons. At the same time, this approach can contribute to a more punitive, control-oriented strategy for working with or punishing incarcerated persons. In the case of medical and health care, that can mean, as Venters (2019) has argued, that security not only takes precedence over physical or mental health outcomes, but also that the near-exclusive focus on it subverts efforts to address these outcomes.

Extended solitary confinement for management purposes is striking in this respect. It is a tool that serves the interests of security. Used on a delimited basis, it may be effective, though critics might still point to it as inherently unethical to rely on incarceration that relies on isolation (see, generally, Haney, 2018; Kupers, 2017). In addition, without programming, those who leave the housing appear unlikely to fare well upon release to general population facilities or society (cf. Ward & Werlich, 2003). Support for this view, and for the notion that programming within restrictive housing may be important for improving outcomes, can be seen in several studies. For example, Batastini et al. (2021) undertook a pilot study of the Stepping Up, Stepping Out program and found that participants had less emotional distress, but found no consistent evidence of reduced criminal risk. Other studies suggest that cognitive-behavioral or therapeutic programming in restrictive housing may improve outcomes, though evidence suggests that implementation challenges may preclude achieving them (Butler et al., 2018; Meyers et al., 2018).

On a larger scale, though, ESM appears unlikely to achieve the goal of systemwide order and safety. It does not, for example, address organizational factors known to contribute to behavior among incarcerated persons, such as staffing levels and quality as well as the culture among officers and incarcerated persons (Bottoms, 1999). And it does not entail behavioral programming but rather interferes with it—individuals in the housing cannot easily receive behavioral program interventions and greater investment in the housing can, alongside of chronic understaffing, contribute to a reduced ability to invest in such interventions on a large-scale, systemwide basis. This potential is notable given French and Gendreau’s (2006, p. 208), meta-analysis, which found that “77% of the time behavioral programs produced a better result than nonbehavioral and educational/vocational programs.” As those authors emphasized, “even a small reduction (e.g., 10%) in misconducts for a group of disruptive inmates might mean the difference between having a chaotic prison environment and having one that is coping adequately with usual pressures” (pp. 208–209).

From this perspective, reliance on ESM represents a paradox. In a context of understaffing and resource shortages, it may be essential for day-to-day prison operations. An individual who violently attacks others requires intervention. Yet, this reliance both appears to reflect a systems failure, including insufficient staffing or programming, and to contribute to it. That contribution may come in the form of increased reliance on control-oriented strategies and less on systems changes and effective programming as well as through adverse effects on incarcerated persons or delegitimization of prison authority in the eyes of this population.

This idea has broader relevance. The punitive turn in criminal justice that began in the late 1970s and accelerated in subsequent decades, for example, arguably involves a similar tension. Reliance on more and tougher punishment may have been necessary. However, it simultaneously contributed to more individuals under some form of correctional supervision and a reduced capacity to implement crime prevention programs, supervise correctional populations closely, or offer rehabilitative programming (Clear & Frost, 2014; Garland, 2013; Gottschalk, 2006; Huebner & Frost, 2019; Mears, 2017; Stuntz, 2011; Travis et al., 2014).

It has broader relevance within prison systems as well. For example, this study focused on extended solitary confinement used to create greater safety and order. Studies are needed as well that examine whether the findings here extend to arguments that might be made about reliance on short-term isolation or various forms of restrictive housing for punishment or protective purposes (Frost & Monteiro, 2016; Shalev, 2009). The goals that these types of housing serve differ, but the general reliance on a control-oriented and more punitive approach is similar. Early studies of supermax prisons tended, as the present study did, to examine long-term solitary confinement used to manage prison systems. However, the emerging body of work on restrictive housing suggests warrant for understanding how a broader spectrum of housing, and how it is used, may both reflect and contribute to the problems that prison systems face.

It warrants emphasis that this study makes no claim to demonstrate that reliance on extended solitary confinement for management purposes improves or worsens prison order and safety. Indeed, the findings should be viewed with caution until more studies, including those relying on quantitative methods, are undertaken. On theoretical grounds, and as this study suggests, ESM could have a beneficial or an iatrogenic effect (Mears & Reisig, 2006). But that does not mean that it has either effect. It might produce both beneficial and harmful effects. Regardless, what this study suggests is that ESM cannot, as Sparks et al. (1996, p. 313) have argued, “magically . . . unlock the problem of order for a prison system as a whole.” Beyond a focus on particular individuals who may present management challenges, prison system order and safety ultimately may depend on effective governance strategies (DiIulio, 1987; Reisig, 1998), attention to adequate staffing and programming, and administrative and management approaches that incarcerated persons view as legitimate (Butler & Maruna, 2016; Irwin, 2005).

This study’s findings should be interpreted in light of its limitations. The individuals who participated in the focus groups and interviews may not have represented the views of all other personnel. It is possible, too, that, although they reported many problems with ESM, they may have felt uncomfortable identifying other problems with it. Not least, Florida’s prison system may differ qualitatively from other prison systems. Like many states, it relies on ESM, but the state differs in having one of the largest prison systems in the country. Its size, as well as the composition of the prison population and of personnel, and the larger political context within which the system operates may have influenced the results.

Going forward, research on ESM still is needed on many fronts. In addition to the calls that other researchers have made for understanding who is placed in the housing and what its psychological and behavioral impacts are for different social and demographic groups (Garcia, 2016; Labrecque et al., 2021; Mears et al., 2019), this study highlights the need for studies that examine system effects of ESM. Few studies to date have done so (e.g., Briggs et al., 2003) and are hampered by the limitations that attend to administrative records data. At the same time, gaining access to prison systems to obtain qualitative data on prison system operations can be challenging (Ward & Werlich, 2003). Such data, however, may be especially important for capturing system-level dynamics and changes. For example, studies should investigate how the culture among personnel at different prisons may influence reliance on ESM or on how they assist individuals in transitioning from ESM to general population units. They should study as well how personnel may vary in their views of and support for ESM. Forty years ago, relatively few personnel would have viewed ESM as conventional practice. In contemporary prison systems, however, that view likely differs because several generations of personnel now have known nothing but a system context in which ESM exists. How do generational differences among personnel at different prisons influence views about the housing? What factors shape their outlook when they become administrators? What alters these views? Alongside of qualitative studies that answer such questions, there remains a need to undertake rigorous empirical studies that can determine if, and under what conditions, reliance on ESM improves or worsens systemwide order and safety.