![]()
Prisons are institutions where individuals convicted of crimes are forcibly confined and subjected to the supervision and authority of prison staff. The number of persons imprisoned in the United States increased considerably between 1970 and 2000, and prisons operating in the United States now confine over 1.5 million persons (Carson, 2018). The number of prison officers and available bed capacity did not increase at the same rate as the inmate population (compare Stephan, 1997 to Stephan, 2008), and so most prisons in the United States are understaffed and operate over their intended capacities (Stephan, 2008). Placing so many offenders in close proximity to one another under the supervision of limited prison staff maximizes the convergence in time and space of persons more vulnerable to violent victimization and those more likely to commit violence given the opportunity (Edgar, OāDonnell, & Martin, 2003; Wooldredge, 1998; Wooldredge & Steiner, 2014).
The study described herein is about individual-on-individual violence in prison. Prison violence more broadly includes intrapersonal (e.g., self-harm), interpersonal (e.g., offending, victimization), and collective violence (e.g., riots). Our focus for this study is interpersonal violence perpetrated by inmates and experienced by inmates and/or staffāthe violent events that take place within the everyday framework of a prisonās social organization (Bottoms, 1999).1
Prisons can be violent places. Data collected from the Census of Adult State and Federal Correctional Facilities in the United States indicates that the average within-prison rate of assaults on inmates housed in U.S. prisons was 16.25 assaults per 1,000 inmates in 2005 (Bureau of Justice Statistics, 2010). In contrast, the arrest rate for assaults among adults in the U.S. general population was 6.54 in 2005 (Federal Bureau of Investigation, 2006). Based on self-report data from the 2004 Survey of Inmates in State and Federal Correctional Facilities, 12.5 percent of all surveyed prisoners physically assaulted another inmate during their incarceration, 2.8 percent physically assaulted a staff member, and 15.0 percent were intentionally injured by another person during their incarceration (Bureau of Justice Statistics, 2007). Data from the 2008 National Former Prisoner Survey revealed that approximately 7.5 percent of former state inmates were sexually victimized during their most recent imprisonment (Beck & Johnson, 2012), while data from the 2011ā2012 National Inmate Survey (NIS) showed that approximately 4 percent of inmates were sexually victimized during a 12-month period (Beck et al., 2013). In comparison, findings from the corresponding National Crime Victimization Surveys showed that less than 2 percent of the U.S. population age 12 and over experienced a physical assault during the previous year (Catalano, 2005), and approximately 1 percent of the general population were sexually assaulted during this same period (Rand, 2009; Truman et al., 2013).
Despite the tenuous nature of some of the comparisons discussed above (i.e., some estimates reflect the occurrence of events over different periods), the rates of violent offending and violent victimization among inmates appear to be higher in U.S. prisons than in the general population. This situation is not unique to inmates. Prison officers experience higher rates of workplace violence relative to those working in most other occupations (Harrell, 2011), and prison officers rank among the highest occupational groups for nonfatal injury rates and days absent from work due to those injuries (Konda et al., 2012). However, researchers have found that the likelihood of perpetrating violence and/or experiencing violent victimization is not consistently high among all inmates or prison staff. Studies of prison inmates and staff perceptions also consistently reveal that inmates generally feel safe most of the time and most officers feel safe while at work (e.g., Bottoms, 1999; Edgar et al. 2003; Steiner & Wooldredge, 2017). Rates of violence also differ across prisons (e.g., Beck et al., 2013; Camp et al., 2003; Steiner, 2009; Steiner & Wooldredge, 2008, 2017; Wolff et al., 2007; Wooldredge, 1998; Wooldredge & Steiner, 2014). Considered together, these observations suggest that prison violence is not a random process. Some inmates perpetrate violence frequently or experience violent victimization repeatedly, while others are never involved in a violent event (as either an offender or a victim). Prison staff exhibit similar variation in the frequency with which they are involved in violent events. Violence is widespread in some prisons but rarely occurs in others. Differences between individuals and the facilities in which they live or work contribute to variation in the odds of perpetrating violence and experiencing violent victimization in prison, and an understanding of the sources of this variation is important for several reasons.
First, institutional safety is a high priority for prison administrators, and yet prison violence threatens the safety and order of a prison (DiIulio, 1987; Gendreau et al., 1997; Steiner & Wooldredge, 2009b; Toch et al., 1989). Violent incidents not only cause harm to the inmates and/or staff involved, they can also disrupt the daily routines in prison and further limit the services and amenities that officials can provide to other inmates (DiIulio, 1987). Second, prison officials are legally responsible for protecting the confined (Park, 2000), and officials are evaluated in part by the level of violence and other disturbances in their institution (Camp et al., 2003; DiIulio, 1987; Lombardo, 1989). Further underscoring this point is the volume of civil liability for inmate-on-inmate assaults, not just in the United States but also in other industrialized countries. The environments of prisons necessarily create some very unique opportunities for violence, such as officers provoking fights between inmates in adjacent cells (e.g., Pinkston v. Madry, 2006), placing violent habitual offenders in open dormitories (e.g., Pierson v. Hartley, 2004), failing to protect inmates when they were witnesses in other cases involving offenders housed in the same facility (e.g., Rangolan v. County of Nassau, 2004), permitting transsexual inmates to interact with sexual predators (Greene v. Bowles, 2004), and placing inmates in close proximity to others who consistently threaten them (Borello v. Allison, 2006). From a practical standpoint, research on prison violence is essential for uncovering the most effective means of violence prevention in prison in order to reduce harm to both inmates and staff, and to reduce incumbent costs to the state.
Finally, prison officials have a moral responsibility to protect both the confined and the public (i.e., it is the right thing to do). Scholars have documented the direct and collateral consequences of incarceration (e.g., Clear, 2007; Travis & Western, 2014), but exposure to prison violence has additional consequences for those who live and work in prison, as well as for the public. For example, inmates who perpetrate violence are more likely to suffer victimization in prison (Boxer et al., 2009; Wooldredge & Steiner, 2014), and those who are victimized are more likely to experience psychological problems such as PTSD, depression, anxiety, hostility, and aggression both during their incarceration and after release (Boxer et al., 2009; Cohen & Wills, 1985; Hochstetler et al., 2004; Listwan et al., 2010; Struckman-Johnson & Struckman-Johnson, 2006; Wolff & Shi, 2011; Wooldredge, 1999; Zweig et al., 2015). Violent victimization has also been associated with greater stress and burnout among prison staff (e.g., Schaufeli & Peters, 2000; Spinaris et al., 2012; Steiner & Wooldredge, 2015b). Further, inmates who perpetrate violence or suffer victimization in prison are more likely to offend after they are released (e.g., Cochran, 2013; Cochran et al., 2014; Listwan et al., 2010, 2013). Prison officials have an ethical responsibility to ensure that their fellow staff, those in their care, and the public are not harmed, and uncovering the sources of prison violence would aid in the development of strategies designed to reduce harm stemming from these events. The reduction of prison violence could also go a long way toward increasing the legitimacy of prisons in the eyes of inmates, staff, and society (Sparks et al., 1996).
The study described in this book is about understanding the sources of prison violence and why violence is more or less common among individuals confined in the same prison as well as why levels of violence differ across prisons. As noted, we focus specifically on interpersonal violenceāviolence perpetrated by inmates and experienced by inmates and staff. The study was carried out in 46 prisons across two of the United States (Ohio and Kentucky), and data were collected from inmates, officers, and prison administrators.
Explaining prison violence
Researchers have long been interested in the study of prisons, inmates, and prison staff (e.g., Clemmer, 1940; Reimer, 1937; Sykes, 1958), which, when coupled with the priority that prison administrators place on promoting order and safety in their institutions, have generated numerous studies of the causes and correlates of prison violence. Scholars have typically relied on three perspectives and the interactions between them when framing potential predictors of violence. Deprivation theory suggests that inmate behaviors are manifestations of how inmates adapt and cope with the āpainsā inflicted by the prison environment, whether through participation in a social system that helps to reduce these deprivations (Clemmer, 1940; Sykes, 1958), or through individual-level choices that help to facilitate need satisfaction (Goodstein et al., 1984; Goodstein & Wright, 1989). Drawing from this perspective, scholars have emphasized the relevance of environmental features of prisons (e.g., population size, security levels) for understanding inmate violence (e.g., Cao et al., 1997; Lahm, 2008; Steiner & Wooldredge, 2008; Thomas, 1977).
Aside from deprivation theory, importation theory holds that prisons are not completely closed systems and that inmate behaviors are shaped primarily by individualsā pre-institution characteristics, beliefs, attitudes, and experiences (Irwin & Cressey, 1962; Irwin, 1980). Scholars have often framed the link between pre-incarceration individual-level characteristics of inmates (e.g., age, race) and prison violence within importation theory (e.g., Cao et al., 1997; Goetting & Howsen, 1986; Mears et al., 2013; Harer & Steffensmeier, 1996). Others have also considered the interactions between the deprivation and importation perspectives, recognizing that the impact of environmental factors may be conditioned by the personal attributes of prisoners (e.g., Mears et al., 2013; Wooldredge et al., 2001).
A third perspective is actually a set of perspectives focusing on prison management, such as administrative control or inmate balance theory. These explanations de-emphasize the impact of prison environments and inmates, suggesting that differences in levels of violence across prisons are primarily the result of differences in how prisons are managed (Camp et al., 2003; Colvin, 1992; DiIulio, 1987; Useem & Kimball, 1989; Useem & Reisig, 1999). Researchers who have assessed the empirical validity of management perspectives have found that indicators of managerial practices (e.g., use of disciplinary housing or facility programming) are related to levels of violence across prisons (e.g., Camp et al., 2003; Huebner, 2003; Steiner, 2009; Useem & Reisig, 1999). More recent discussions of legitimate authority and the exercise of power by prison officers also fall within this perspective, even if not formally stated as such (e.g., Crewe & Laws, 2018; Steiner & Wooldredge, 2018).
Empirical studies have provided support for each of these ātraditionalā explanations of prison violence; however, researchers who have examined the relevance of individual pre-incarceration characteristics, environmental characteristics of prisons, and management practices in the same empirical model have found that variables reflecting each of these three domains are relevant to an understanding of prison violence, not to mention interactions between at least two of the three domains (e.g., Camp et al., 2003; Cao et al., 1997; Huebner, 2003; Jiang & Winfree, 2006; Lahm, 2008; Steiner & Wooldredge, 2008, 2009a, 2009b, 2015; Steiner, 2009; Wooldredge et al., 2001; Wooldredge & Steiner, 2014). Comprehensive reviews of the evidence pertaining to prison violence have also found that individual characteristics, prison characteristics, and management practices are all linked to inmate violence (Crewe & Laws, 2018; Steiner et al., 2014, 2017), suggesting that each of the three theories of prison violence are inadequate as standalone explanations of the phenomenon. As we discuss in Chapter 2, this is because none of the existing theories of inmate behavior can account for variation in institutional violence at both the inmate and facility levels of analyses, while also incorporating all of the relevant sources of violence based on related empirical studies (Crewe & Laws, 2018; Steiner et al., 2014). A conceptual framework inclusive of concepts depicting individual characteristics, environmental characteristics of prisons, and management practices has yet to emerge (Byrne et al., 2008).
A multilevel social controlāopportunity perspective
To the end of providing a more comprehensive framework for unde...
