CC BY-NC-ND
0JOURNAL OF CRIMINAL JUSTICE
ELSEVIER
Contents lists available at SciVerse ScienceDirect
Journal of Criminal Justice
The place of punishment: Variation in the provision of inmate services staff across the punitive turn
Michelle S. Phelps *
Department of Sociology, 107 Wallace Hall, Princeton University, Princeton, NJ 08544, United States
ARTICLE INFO ABSTRACT
Available online 18 July 2012 Despite the growing literature on the punitive turn, knowledge of how the experience of American imprison-
ment varied across time and place remains limited. This article begins to fill that gap, providing a nuanced portrayal of variation in the practices of rehabilitation.
Purpose: To examine how one aspect of the rehabilitative ideal in practice—the provision of staff dedicated to inmate services—varied across time and place over the past 30 years.
Methods: The article presents statistics on the inmate-to-staff ratios for inmate services staff (including teachers, counselors, doctors, etc.) between the years 1979 and 2005 for all 50 U.S. states. Results: The analyses reveal that while there was a substantial decline in the services staff ratio during the 1990s and 2000s, this shift across time paled in comparison to variation across place. Northeastern prison systems, for example, on average maintained higher inmate services staff ratios in 2005 than Southern states in any year. In addition, results suggest state variation is related to differences in prison crowding, inmates' racial composition, and political cultures.
Conclusions: The findings suggest the punitive turn was more variegated and partial than is often assumed and highlight the importance of exploring state variation in penal practices.
© 2012 Elsevier Ltd. All rights reserved.
Introduction
Scholars are by now familiar with the profound "punitive turn" in U.S. criminal justice policies, especially the massive expansion in the use of imprisonment: Between 1972 and 2010, the number of prisoners held in state facilities increased seven-fold, from 174,000 to 1.4 million (Pew Center on the States, 2010). As the prison system grew, other features of the carceral environment—including racial tensions, control technologies, and staff-inmate relationships—also shifted, altering the experience of imprisonment.
One of the most prominent changes in the nature of incarceration has been the decline of the rehabilitative ideal and the presumed devolution of prison rehabilitation programs. According to leading accounts, as prisons expanded, the mission of corrections departments transitioned from rehabilitation in "correctional" institutions to inca-pacitation in crowded, "warehouse" facilities (Garland, 2001; Irwin, 2005; Wacquant, 2001). Scholars argue that as rehabilitation declined as the dominant rhetoric of punishment, the practices of rehabilitation were dismantled. Thus, as the number of prisoners mounted to "mass" proportions, this account suggests that inmates became increasingly less able to access basic rehabilitative opportunities, such as educational programs, substance abuse treatment, and counseling services.
* Tel.: +1 609 258 4543; fax: +1 609 258 2180. E-mail address: phelps@princeton.edu.
0047-2352/$ - see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.jcrimjus.2012.06.012
Yet the leading theoretical accounts of this shift often exclude or gloss over empirical evidence on the presence of rehabilitative programming, analyzing changes in political narratives more often than changes in practices. When researchers have focused on changes in practices, the trends have often been quite different from a simple story of decline (Useem & Piehl, 2008; Phelps, 2011). Rather than discount these findings as relics of previous modes of punishment, an emerging literature argues that the complex relationship between the rhetoric and practices of rehabilitation is central to understanding the punitive turn both empirically and theoretically. These scholars criticize the conceptualization of distinct epochs in punishment and argue that penality has always been "variegated" or "volatile and contradictory," continually "braiding" together punitive and rehabilitative aims and practices (Goodman, in press; Hutchinson, 2006; O'Malley, 1999; O'Malley, 2000). In addition, this research documents how penal practices can survive seemingly contradictory changes in rhetoric—for example, by molding the framing of existing practices to fit new discourses (Cheliotis, 2006; Goodman, in press; Hannah-Moffat, 2005; Maurutto & Hannah-Moffat, 2006; Robinson, 2008). Thus, a shift from rehabilitative to punitive rhetoric does not necessarily produce a parallel move from rehabilitative to punitive practices.
In addition to showing continuity in practices across time, scholars have also described how the development of mass incarceration and the adoption of more punitive modes of punishment varied dramatically across state lines. Researchers have begun to explore how states differentially experienced the punitive trajectory, complicating the
standard narrative of the rise and fall of rehabilitation witnessed in progressive states by documenting the histories of states in the South and Southwest, where "tough" or "harsh" justice always dominated (Lynch, 2009; Perkinson, 2010). These works suggest that U.S. states (and regions) did not experience one collective punitive turn, but rather, had drastically different commitments to the rehabilitative ideal and experienced varied trajectories in the turn toward mass incarceration (Lynch, 2011).
This article merges these two areas of contestation—investigating change and continuity across time and variation across place—in order to understand how the practices of rehabilitation shifted across the country after the 1970s, as the national rhetoric around punishment turned increasingly punitive. The analyses focus on one measure of rehabilitation in practice—the number of inmate services professionals relative to the number of inmates—that provides an estimate of states' commitment to funding rehabilitation-related staff and ability to provide inmate services everyday. In line with the leading accounts of the punitive turn, the results suggest a real and important decline in the inmate services staff ratio during the 1990s and 2000s (although not during the 1980s), with most states losing a substantial number of inmate services professionals relative to the number of inmates. In contrast to the dominant accounts of a coherent national trend, though, the results also reveal profound statelevel variation, with differences across states overwhelming year-to-year changes in the average staff ratio. Many states provided more inmate services staff members well into the 2000s than other states did as far back as 1979, with a range of practices co-existing at every point in time. For example, at its lowest point in 2005, the median inmate services staff ratio for states in the Northeast was still higher than the median ratio for Southern states in any of the years between 1979 and 2005.
This importance of place, above and beyond changes across time, is consistent with theoretical accounts of punishment as "variegated" or "braided," with both rehabilitative and punitive elements flourishing during both the "rehabilitative" and "punitive" eras. Further, the results confirm how much there is to learn from examining state variation in the practices of punishment. At the end of the article, I sketch an example of this kind of research by evaluating ties between state characteristics and the inmate services staff ratio. The results suggest that state variation in rehabilitation practice is tied, in part, to factors such as overcrowding, prisoner racial composition, and political cultures, although no single factor satisfactorily explains state differentials in the trajectory of inmate services staff ratios.
Rehabilitation and the punitive turn
Punishment scholars have put forward a number of compelling explanations for the development of mass incarceration. Garland (2001) provides the richest account, describing sweeping changes in the U.S. and U.K. that have produced a new "culture of control," defined by "obsessive attempts to monitor risky individuals, to isolate dangerous populations, and to impose situational controls on otherwise open and fluid settings" (Garland, 2001, p. 194). Other scholars, like Beckett (1997), Tonry (1995), Simon (2007), and Weaver (2007), focus on the racial politics of punishment, arguing that mass incarceration emerged as a result of political maneuvering designed to capture white voters' support in the wake of the Civil Rights movement. Wacquant (2009) paints perhaps the bleakest picture, describing prisons a "peculiar institution" designed to permanently exclude poor African Americans from the polity.
In describing these changes in the scale of imprisonment, studies have also remarked on shifts in the experiences of punishment, cohering around the concept of the "decline of the rehabilitative ideal." This phrase, coined by Allen (1981), describes a shift in the logic of punishment away from a medical model of prisons as sites for inmate classification and treatment to a punitive model of imprisonment, in which
the most prisons can do is incapacitate offenders for increasingly long periods of time. This contemporary model is often referred to as the "warehouse prison," on account of its scale and the lack of rehabilitative opportunities (Irwin, 2005). Rather than transforming prisoners, this new model provides a semi-permanent site of exclusion for people deemed "social refuse" (Feeley & Simon, 1992; Wacquant, 2001).
Yet in crafting these sophisticated accounts of penal change, macro-level research often sacrifices details on the micro-level punishment practices that might paint a more complex and contradictory picture (Lynch, 1998; Lynch, 2011; Raynor & Robinson, 2009; Useem & Piehl, 2008). The strongest work often alludes to this omission; for example, Garland (2001) notes that "the distinctive technologies, powers and knowledges developed by the penal-welfare movement are still in daily use," and briefly cites statistics on in-prison treatment programs (170). However, the persistence of rehabilitative practices, despite the profound change in the scale and rhetoric of punishment, remains peripheral to the overall argument (Garland, 2004). Similarly, variation across place in the adoption and rejection of these practices often remains out of focus.
A new literature takes this disconnect between the narrative and practices of punishment as central to understanding the punitive turn and punishment more generally. This wave of scholars posits that there exists a "governmentality gap" between understandings of broad penal rationalities and specific practices and that a full account of the punitive turn must consider both (McNeill et al., 2009, p. 419). Rather than a "catastrophic" model of penal change which envisions distinct epochs of punishment, this literature conceives of punishment as inherently "volatile and contradictory," with penal interventions simultaneously entailing both punishment and reform (O'Malley, 1999, 2000). Hutchinson (2006), for example, argues that these two aspects of punishment are continually "braided" together, with change over time best described as a shift in the relative emphasis of one strand or the other, rather than a wholesale replacement.1
In addition to challenging the "catastrophic" narrative of the punitive turn and emphasizing continuity and contestation in penal practices, this group of scholars also argues that penal narratives and practices can come together (or disconnect) in surprising ways. In particular, work has shown that rehabilitative practices often survived shifts in penal logic by adopting new discourses. For example, researchers have documented how risk-need assessment tools combine both rehabilitative and risk management ideals, promoting services for inmates under the rhetoric of public safety (Hannah-Moffat, 2005; Maurutto & Hannah-Moffat, 2006). Other scholars have documented how prison administrators use the narrative of public safety to marshal support for prison programs (Schoenfeld, 2009) and how everyday discourses of rehabilitation have adapted to neo-liberal ideas around "responsibilization" (Goodman, in press; Lynch, 2000). Particularly given that correctional administrators themselves continued to support the goals of rehabilitation throughout the punitive turn (Cullen et al., 1993), there is little reason to assume that shifts in punishment discourses translated directly into changes in practices.
Regional and state variation
Scholars of punishment have long understood that there are important differences in practices across the U.S. This work focused in particular on disparities in the "scale of imprisonment" (Zimring & Hawkins, 1991). Using complex quantitative models, researchers have explored the state-level determinants of this variation, finding significant associations between imprisonment rates and state characteristics such as violent crime rates, racial composition, and state budget strength (Greenberg & West, 2001; Jacobs & Carmichael, 2001; Western, 2006; Spelman, 2009). More recently, scholars have begun to investigate how the history of the expansion of imprisonment varies across state, using case studies to illustrate the importance special interest groups, forms of political participation, and judicial interventions (Barker,
2009; Campbell, 2011; Lynch, 2009; Page, 2011; Schoenfeld, 2010). These factors help to explain why some states, such as Oklahoma, Mississippi, and Louisiana (clustered in the South), have the highest incarceration rates in the world, whereas states like Maine, Massachusetts, and Minnesota (located in the Midwest and Northeast) have rates on par with other Western countries.
A smaller literature focuses on variation in the experience of punishment. This research complicates the standard narrative of the rise and fall of rehabilitation witnessed in progressive states in the Northeast and elsewhere, including California, by documenting the penal histories of states in which "tough justice" remained pervasive. Lynch (2009) provides one of the best examples, documenting how penal discipline in Arizona constituted an especially harsh form of "frontier justice," largely untouched by progressive reforms. Similarly, studies of states in the South, including Texas (Perkinson, 2010) and Florida (Schoenfeld, 2009), highlight prison systems in which rehabilitation was never a strong goal in theory or practice. In Texas, for example, punishment in the "rehabilitative era" consisted mainly of back-breaking manual labor in the fields with "trusty" inmates serving as guards (Dilulio, 1987; Perkinson, 2010). Together, these works provide a sense of how states' penal trajectories have varied across the U.S. and over time, particularly how prison systems in the South and Southwest developed differently than states with deeper progressive roots (Lynch, 2011).
Defining rehabilitation
Definitions of rehabilitation have varied across time and place, encompassing nearly every aspect of punishment. Practices considered rehabilitative have ranged from contemporary markers like education and counseling programs, to religious practices (including quiet reflection and religious study) and even practices that many today consider punitive, such as corporeal punishment and forced labor (Raynor & Robinson, 2009). Therefore, when looking for changes in the practices of rehabilitation, scholars have examined many measures, including program participation rates (Lynch & Sabol, 2001; Phelps, 2011; Useem & Piehl, 2008), staff ratios (Phelps, 2011; Western, 2006), specialized facilities (Phelps, 2011), interactions between criminal justice employees and supervisees (Bosworth, 2007; Kruttschnitt & Gartner, 2005; Lynch, 2000), and prisoners' and staff members' own conceptualizations of rehabilitation (Goodman, in press). In some cases, these indicators measure fundamentally different dimensions of prisons' rehabilitation-orientation, such that even within a single facility trends can be discrepant—for instance, administrators' ideologies around rehabilitation may tell one story while the actual provision of services and programs tells another (Carroll, 1998; DiIulio, 1987; Jacobs, 1977; Kruttschnitt & Gartner, 2005). The one commonality from this research is that changes in rehabilitative practices have been much more complex than changes in the national rhetoric on punishment.
This article analyzes trends in staff ratios for inmate services professionals—a category of staff that includes teachers, counselors, doctors, and librarians—across time and place. Staffing ratios provide a proxy for states' commitment to rehabilitative practices by measuring the financial resources states are willing to spend on inmate services. While it may be easy for a state to pay "lip-service" to rehabilitation in mission statements or other materials, it takes considerably more political consensus to support staff devoted to inmate services. Further, this indicator offers the best available measure of the capacity of states to actually provide services to inmates. Without counselors, doctors, nurses, teachers, librarians, and other service professionals, prisons can provide few meaningful services.
Data and methods
This article relies primarily on data from the Bureau of Justice Statistics' Census of State and Federal Adult Correctional Facilities
(United States Department of Justice, 1997a, 1997b, 1998, 2001, 2004, 2010). Data were collected through questionnaires sent to administrators at all known correctional facilities in 1979,1984,1990, 1995, 2000, and 2005.2 The facilities include community corrections centers, youthful offender facilities, substance abuse treatment centers, and other specialized prisons, although the vast majority of prisoners are housed in general confinement (Phelps, 2011). Beginning in 1990, private prison facilities were also included in the census.3 The response rate for each year is 100 percent, except for the 2005 data, which is missing information on all Illinois facilities and staffing data for state-administered facilities in California. For these two states, data from 2000 was imputed to estimate 2005 values.
The questions on facility staffing levels distinguish between custodial (i.e., security), administrative and supportive (e.g., wardens, secretaries, culinary and maintenance workers), inmate services, and other staff. Inmate services staff include teachers, social workers and counselors, psychologists and psychiatrists, doctors and nurses, chaplains, librarians, and other similar professionals. Beginning in 1990, the survey queries about the number of academic and vocational education instructors, as distinct from the other inmate services staff. The survey instructs administrators to list the total number of staff employed by the facility, including part- and full-time staff, as well as staff members paid by other government agencies (e.g., teachers paid by education departments) and unpaid interns.4 Volunteers are excluded.
I examine variation across place and change over time in the staff to inmate ratio for all inmate services professionals and for the sub-category of teachers by state between 1979 and 2005. The staff ratios are calculated as the number of staff per 1,000 inmates in a given state-year. These ratios provide a proxy for states' commitment to rehabilitative practices by measuring the financial resources states are willing to spend on inmate services. In addition, this indicator provides the best available measure of the capacity of states to provide inmate services on a daily basis. Finally, the staff ratio data have the advantage of being reported consistently across time and place, allowing for reliable comparisons.
Still, there are two potential limitations of this measure. First, because inmate service professionals are hired to respond to inmates' needs, the indicator might reflect the average "neediness" of prisoners, rather than states' willingness to fund inmate services staff positions. However, research shows that prisoners in general tend to have a number of service needs, including education, medical care, counseling, and drug treatment, and prisons rarely provide enough services to meet them (Chamberlain, 2011). This suggests state variation in staffing levels is unlikely to simply reflect inmates' service needs. Second, rather than reflecting states' commitment to providing inmate services, this measure might reflect the ability of states to hire qualified staff. State administrators, however, can partially overcome such obstacles by choosing where to locate prisons and how to make such jobs competitive opportunities for professionals. In addition, as described below, the results support the face validity of this measure, documenting the highest inmate services staff ratios in the Northeast, where progressive reforms originated, and the lowest inmate services staff ratios in the South, the region with the weakest ties to rehabilitation.
In the final section, I examine four potential explanations for state variation in the inmate services staff ratio: prisoner population increases and overcrowding, inmate racial composition, court interventions, and political culture. Data for these analyses are drawn from the Census of State and Federal Correctional Facilities (various years), which provide information on facility crowding, inmate demographics, court orders, and "supermax" facilities. Additional data on states' racial composition come from the U.S. Census Bureau and information on the party affiliation of state governors is from the Partisan Balance of State Government data, available through State Politics and Policy Quarterly (Klarner, 2003).
Results and discussion
National trends
In this section, I examine state-level variation in the inmate services staff ratio (calculated as the number of inmate services staff members per 1,000 inmates) in states across the U.S. The results are presented in box plots, which visually convey a number of important features of the distribution of staff ratios in each year, including the median value (the line inside the boxes), the 25th and 75th percentiles (the lower and outer limit of the box), and the lower and upper adjacent values (the bottom and top of the "whiskers").5
If the dominant trend in prisons has been a switch from the rehabilitative ideal to a punitive warehouse model, we should expect to find the staff ratio for inmate services professionals in decline—i.e., fewer staff attending to a rapidly increasing number of inmates. Fig. 1 provides the first test of this hypothesis, displaying the distribution of states' inmate services staffing ratio between 1979 and 2005. Consistent with the leading accounts of the punitive turn, the results show a decline in the average inmate services staff ratio—the median inmate services staff ratio declined slightly between 1979 and 1984 (from 66 staff members per 1,000 inmates to 61), increased slightly between 1984 and 1990 (back up to 66), and declined between 1990 and 2005, reaching a median of 47 by 2005. Thus, there were on average nearly 30 percent fewer inmate services professionals per inmate in 2005 than in 1979, with most of this change occurring after 1990.
However, the second finding from Fig. 1 is that this decline across time is overwhelmed by the tremendous variation across place in each year. From the figure, it is clear that the distribution of states' inmate services ratio for each year overlap substantially. One way to quantify this overlap is to compare the decline in the median across years with a measure of the spread across states in each year, for example, the distance between the 25th and 75th percentiles. Using this measure, I find that even the smallest difference between the 25th and 75th percentile—22 staff members per 1,000 inmates in 2005— is greater than the decline in the median staff ratio of 19 staff members per 1,000 inmates between 1979 and 2005. Thus, in each year, even this conservative measure of state variation exceeded the total change between 1979 and 2005.
As described in the data section, the category of inmate services staff is quite diverse, including educational instructors, medical personnel, counselors, and other professional services staff. Beginning in 1990, the data provide counts for academic and vocational instructors separate from other inmate services staff, allowing for an analysis of this distinct sub-group. Among inmate services staff, teachers are the most strongly aligned with contemporary notions of rehabilitation, making trends in such staff ratios particularly important for study.
Fig. 2 displays box plots for the distributions of states' education staff ratios across years. The first result is the dramatic difference in scale compared to Fig. 1—whereas the upper adjacent staff ratios for inmate services staff reached values above 100 professionals per 1,000 inmates, the upper adjacent staff ratios for educational staff ratio barely exceed 30. This reflects the fact that educational staff comprise a minority of inmate services staff—across states, only a fifth of inmate services staff are teachers. This percentage declined over time, from a mean of 25 percent in 1990 to 19 percent by 2005. The decline in staff ratios for all inmate services staff, combined with the declining representation ofteachers, resulted in a particularly sharp decline in teaching staff. Between 1990 and 2005, the median staff to inmate ratio for teachers was cut nearly in half, from a median of 15 instructors per 1,000 inmates in 1990 to just 8 by 2005. Again, though, state-to-state variation was substantial, with a significant overlap in the distributions for each year. Using the 25th to 75th percentile spread as a measure of state variation, the decline across time in the educational staff ratio is smaller than variation within each year (with a decline of 7 teachers per 1,000 inmates between 1990 and 2005, compared to a gap between the 25th and 75th percentiles of 8 in each year).
It is also evident from Fig. 2 that in every year the lower adjacent reached the nadir of the graph—0 educational staff per 1,000 inmates. In some states, then, prison facilities did not report a single paid staff member (or unpaid intern)—hired by the Department of Corrections, another government agency, or a private contract—tasked with providing inmate education. It is difficult to know if these low figures represent an empirical reality or whether some facilities simply did not correctly report the total number of educational staff (perhaps because such staff are coordinated by outside agencies). While administrators are instructed to include all staff members in the survey, it is difficult to ascertain whether surveys were completed correctly since there is no source of comparable data available for all states.6 Despite these concerns, the coherence of the trends across place suggests a real decline in education staffing over this period. This is corroborated by a substantial decline in the percent of inmates participating in educational classes—between 1991 and 2004, the percent of inmates in national surveys reporting current or past participation in academic programs declined from 43 to 27 percent (Phelps, 2011).
These trends came in a period in which penal facilities increasingly privileged security staff over inmate service professionals (particularly educators). Between 1979 and 2005, the median staff ratio for correction officers remained nearly stable, declining by just 5 percent (from 225 to 214 officers per 1,000 inmates). Thus, the already large disparity between treatment and custody staff widened during this period across states, from a median ratio of 3.5 security officers for
Box-Plots of Variation Across States
Note: Box-plots display the median, 25th and 75th percentiles, and lower and upper adjacent values. Source: Census of State and Federal Correctional Facilities, various years.
Box-Plots of Variation Across States
m œ ra E
iE 20 -o o o
œ 10 -
Note: Box-plots display the median, 25th and 75th percentiles, and lower and upper adjacent values. Source: Census of State and Federal Correctional Facilities, various years.
Fig. 1. Inmate Services Staff Ratio, 1979-2005.
Fig. 2. Education Staff Ratio, 1990-2005.
every inmate services professional to a median ratio of 4.5. The results imply that as states struggled to rapidly build up staffing levels to keep pace with inmate numbers, prison guards were hired at a substantially faster rate than inmate services staff.
Together, the results presented in Figs. 1 and 2 suggest that, in terms of the provision of staff devoted to inmate services, prisons indeed became "harsher" places to do time after the 1970s. However, the results also support several qualifications. First, even in the more rehabilitation-oriented period of the late 1970s, the average inmate services staff ratios were quite low, with the median state still providing 3.5 times as many security officers as inmate services professionals. With this limited number of staff, the majority of inmates could not receive extensive rehabilitative programs or services (Phelps, 2011). Second, the decline began in the 1990s, rather than during the 1980s; changes in staffing levels lagged significantly behind shifts in the rhetoric of punishment. Third—and perhaps most importantly—in every year, a wide range of inmate services staff ratios were present across states, with variation across place overwhelming changes across time.
Regional trends
The national trends presented above suggest both a trend of decline across time and tremendous variation across states within year. But do staffing levels cluster by region? And are the same trends across time evident when states are disaggregated into regions? Fig. 3 begins to answer these questions, documenting state variation by region in the inmate services ratio between 1979 and 2005.
Fig. 3 documents that states in all four regions experienced declines in the inmate services staff ratio, although the starting and ending staffing levels varied significantly. Consistent with the literature, states in Northeast began and ended with the highest inmate services staffing levels, followed by states in the Midwest and West, and trailed by states in the South. Between 1979 and 2005, the median inmate services staff ratio declined by between 34 and 40 percent in all four regions—dropping from 99 to 63 inmate services professionals per 1,000 inmates in the Northeast, 82 to 54 in the Midwest, 73 to 47 in the West, and 50 to 30 in the South. During this period, the gap between the Northeast and South widened; by 2005, inmates in the median Northeastern state had access to more than twice the number of staff than their Southern counterparts.
Fig. 4 displays the same box-plots by region as Fig. 3, but is limited to academic and vocational instructors. We see that states in every region saw a decline in the staff ratio for teachers, with the largest declines outside the Northeast. Between 1990 and 2005, the educational staff ratio declined by just 17 percent in the Northeast (froma median of 18 to 15 teachers per 1,000 inmates), compared to declines of around 50 percent in the other three regions (from 18 to 8 in the
Box-Plots of Variation Across States Midwest Northeast South
Box-Plots of Variation Across States Northeast South
CD CD CD CD 0 O CD CD CD CD O O CD CD CD CD 0 O CD CD CD CD 0 O —•JQ^CDCDoO ^®CDCD0O ^®CDCD0O ^®CDCD0O CD O C71 O 5 CD O C71 O 5 CD O C71 O 5 CD O C71 O 5
Note: Box-plots display the median, 25th and 75th percentiles, and lower and upper adjacent values. Source: Census of State and Federal Correctional Facilities, various years.
Fig. 3. Inmate Services Staff Ratio by Region, 1979-2005.
Note: Box-plots display the median, 25th and 75th percentiles, and lower and upper adjacent values. Source: Census of State and Federal Correctional Facilities, various years.
Fig. 4. Education Staff Ratio by Region, 1990-2005.
West, 13 to 6 in the South, and 17 to 9 in the Midwest). To an even greater degree than in the broader inmate services staff ratio, the gap between teacher ratios in the Northeast and South expanded between 1990 and 2005; by 2005, inmates in the median Northeastern prison had access to 2.5 times the number of teachers than did inmates in the median Southern prison.
Together, Figs. 3 and 4 replicate the primary findings from Figs. 1 and 2, again showing staffing declines across time and tremendous variation across place. In addition, Figs. 3 and 4 document the regional clustering of inmate services staff ratios. The most striking result from these figures is the difference between Northeastern and Southern states. By 2005, there were fewer inmate services professionals everywhere, but the median state in the Northwest still provided more inmate services staff (and teachers) than the median state in the South in any year. Thus, in at least this respect, Northeastern prisons in the present decade are more rehabilitation-oriented than Southern states have been at any point in the past 30 years. This pattern is even more extreme than the stark differences in incarceration rates across regions—despite the generally lower incarceration rates in the Northeast and higher rates in the South, the median incarceration rate for Northeastern states in 2010 was still nearly 50 percent greater than the median for Southern states in 1980 (Pastore & Maguire, 2010).7
State trends
As displayed in Figs. 1 through 4, states varied widely in education and overall inmate services staff ratios in every year. In 1979, the inmate services staff ratio varied from a low of 15 inmate services professionals per 1,000 inmates in North Carolina and Texas to a high of over 145 in New Hampshire, Rhode Island, and Vermont. By 2005, the ratio ranged from a low of 7 services professionals per 1,000 inmates in Alabama and Arkansas to more than 70 in Minnesota, New Hampshire, and Vermont. Likewise, the staff ratio for academic and vocational instructors in 1990 ranged from 0 teachers to over 30 teachers per 1,000 inmates in Maine, Minnesota, New Hampshire, New York, and Vermont. By 2005, the highest staff ratios for teachers were above 20 in Maine, New Jersey, and Vermont.8
Despite this variation, almost all states (43) experienced a decline in the inmate services staff ratios between 1990 and 2005; 46 showed a decline in the education ratio. The median decline was 21 staff members per 1,000 inmates for all inmate services professionals and 7 educational staff members per 1,000 inmates. During the 1990s, these trends were driven by rapid increases in prisoners outpacing more tepid increases in inmate services staff. However, between 2000 and 2005, roughly half of states lost inmate services staff members while inmate totals continued to climb.
Despite the decline between 1990 and 2005, there remained substantial overlap across states, as displayed in the box-plot figures. One way to quantify this finding is to examine the overlap in the staff ratios in 1990 and 2005. If states' inmate services staff ratios from both 1990 and 2005 are ranked in order from smallest to largest, there are only 4 states in 2005 with inmate services staff ratios lower than the lowest value from 1990 and only 4 states in 1990 that had higher values than the highest value in 2005. In other words, these distributions overlap almost entirely—for nearly every state in 2005, there is another state that had a lower staff ratio in 1990, and vice versa. This suggests that throughout this time period—and in spite of the secular decline in the inmate services staff ratio—place remained the most important characteristic for understanding variation in staff ratios.
Understanding variation across states
This tremendous variation in staffing levels opens a new question— what explains states' diverse levels of providing rehabilitation-related staff? Here, I briefly consider the role of four factors—prisoner increases and overcrowding, inmate racial composition, court involvement, and political culture—by evaluating the correlations between these state characteristics and inmate services ratios. While the results cannot prove that these factors produced state variation, they provide an interesting descriptive pattern and suggestive evidence about the determinants ofstates' unique trajectories. Because the data represent a complete population enumeration (i.e., all states in a given year), formal statistical tests are unnecessary to evaluate the significance of differences. However, I analyze both the correlation coefficients and the p-values for each relationship as a method of determining which associations are largest and strongest.
Increases in prisoners
Rapid increases in prisoner populations may make it difficult for states to maintain inmate services staff ratios for two reasons. First, the rapid expansion of the inmate population requires states to dramatically expand staffing capacities, which can be logistically and politically difficult. Second, such rapid prisoner growth is often associated with overcrowding, as state administrators scramble to find enough beds. In prisons with inmates double and triple bunked, it may be especially hard for administrators to keep the increases in staff in step with the increases in prisoners, since there is little room for new programs or services (Carroll, 1998).
The results suggest a large and negative relationship between a state's inmate population and inmate services staff ratio in 2005, although the correlation does not reach statistical significance (r= -.17, p = NS). Similarly, the rate of crowding (measured as the ratio of prisoners to design capacity) in 2005 has a large and negative (but not-significant) relationship with the inmate service staff ratio (r=-.17, p = NS). There is also some evidence that states with greater increases in the crowding ratio between 1990 and 2005 experienced greater declines in the inmate services staff ratio over this period (r=-.15, p = NS). These relationships are not present for variation in teacher staff ratios, suggesting that variability in education staffing were not consistently related to population size or crowding rates.
Thus, rapid increases in prisoner populations and the resultant overcrowding may have been responsible for some of the decline in the inmate services staff ratio, with states more affected by rapid prisoner expansions and crowding less able to expand staffing capacities. However, this relationship was not strong enough to reach statistical significance, suggesting substantial variation across states, with some large, overcrowded states maintaining relatively high staff ratios while other states, with smaller prisoner increases and overcrowding problems, provided few inmate services staff.
Prisoner demographics
Much of the criminological literature around the punitive turn focuses on racial divides and the politics of race and punishment. Indeed, one of the robust predictors of higher incarceration rates across states is the percent of residents identified as black or African American (Beckett & Western, 2001; Greenberg & West, 2001; Jacobs & Carmichael, 2001). Scholars in social psychology have also demonstrated that individuals with negative perceptions of minority groups and anxiety about "racial threat" are more likely to support punitive criminal justice policies (Jacobs & Tope, 2007; King & Wheelock, 2006). This research suggests that some of the worst prison conditions are likely clustered in states in which the prison population is predominantly composed of black felons.
The correlation between the percent of inmates identified as black and the inmate services staff ratio in 2005 is very large, negative, and statistically significant (r = -.42, p<.01). On the high end, states like Louisiana, Mississippi, and South Carolina have some of the highest percentages of black prisoners and the lowest inmate staff ratios, while in states like New Hampshire and Vermont, where inmates are overwhelmingly identified as white, staffratios are on average much higher. There is a smaller, but still negative, relationship between percent black and the staff ratio for educational staff in 2005 (r = -.20, p = NS). These correlations are even stronger if percent black is measured as the percent of state residents identified as black, suggesting it is the racial composition of the state as a whole—rather than the demographics of inmates—that matters for this association.9 In addition, there is some evidence that states with larger increases in the percent of black prisoners between 1990 and 2005 experienced greater losses in the inmate services staff ratio (r = -.14, p = NS) and the teacher ratio (r = -.08, p = NS), although these relationships are less strong than the results for 2005 values (perhaps in part because most states saw little change in the racial composition of inmates across this short span).
Together, the inmate demographics results suggest a relationship between the racial composition of prisoners and the inmate services staff ratio. In large part, these racial differences are explained by—or contextualize—regional differences, with states in the South having, on average, the highest percent of black inmates and the lowest staff ratios. Within the South, the relationship between the inmate services and teacher staff ratios and percent of prisoners identified as black is even larger (r = -.64, p<.01 for both correlations). Outside the South, there is no relationship between the racial composition of prisoners and states' inmate services staff ratios and a positive relationship between percent black and the teacher staff ratio (r = .18, p = NS). This implies that there is little connection between inmates' racial composition and states' inmate services ratio outside ofthe South. Rather, higher percent ages of black inmates explains some of the variance between the South and other regions and differences within the South.
Court involvement
Examining the impact of federal court interventions in prison affairs, many legal scholars have concluded that prison litigation cases have been successful in improving the conditions of confinement (for a recent review, see Jacobs, 2003). One of the ways courts may have improved the conditions of confinement is in supporting rehabilitative programs, especially medical and psychiatric services, and, more generally, by limiting prisoner overcrowding (Carroll, 1998; DiIulio, 1990; Feeley & Rubin, 1998). If such advocacy had been successful, we would expect that states with court interventions would experience smaller losses in inmate services staff relative to other states with similar conditions.
In the facility census data, prison administrators report whether the facility is under court order (or consent decree) for crowding or any other conditions of confinement. I examine three different categories of court orders—any kind of court order, court orders for
staffing issues, and court orders for a cluster of issues potentially related to staff ratios (including overcrowding, medical care, education and training programs, and counseling). When 50 percent or more of a state's prisoner population resides in facilities under court order, I consider that state to have active court involvement.10 Consistent with the decline in inmate services staff ratios, court involvement in prison affairs declined in the 1990s and 2000s, in large part because the Prison Litigation Reform Act of 1996 limited inmates' ability to file claims (Schlanger, 2006). The number of states with any kind of court order that affected 50 percent or more of the inmate population increased from 11 in 1984 to 16 in 1990 and 1995, then declined to 9 by 2000 and reached a low of 5 states by 2005.
Given that court orders select for troubled prison systems, it would be inappropriate to compare staff ratios in states with and without court orders. Instead, I evaluate whether states with court orders in 1990 were more or less likely to experience relatively smaller or greater losses in inmate services staff between 1990 and 2005. I find that states with any kind of court order affecting the majority of the population in 1990 experienced greater losses in inmate services staff on average between 1990 and 2005 (r= -.23, p = NS), although this correlation is not statistically significant. While these results may simply reflect the kinds of prison systems that received court orders, they certainly do not suggest that the courts have played a protective role. This relationship was not replicated for the educational staff ratio. In addition, for both staff ratios, there was no relationship between having a staffing or staff-related court order in place in 1990 and the change in inmate services or educational staff ratios.
At best, differential court involvement was not responsible for shaping trajectories of staff loss across states. If anything, court intervention was perhaps a sign that the state system was on a downward trajectory (although this applied only to the overall inmate services staff ratio and broadest measure of court involvement).
Political culture
A growing literature holds that the punitive politics of mass incarceration flourished to varying degrees across states. If the inmate services staff ratio is a reflection of these broader punitive politics, there should be a correlation between these forces and states' staff ratios. Here, I evaluate two possible indicators of state political culture— the political party in control of the governorship and the use of "supermax" prisons.
Simon (2007) argues that the turn toward mass incarceration was propelled in part by the increasing power of executive authorities, especially governors, and the tendency to "govern through crime" (i.e., to prove political competency by enacting increasingly harsh criminal justice penalties). Republican leaders in particular adopted this "tough" model of justice, pioneering some of the most draconian sentencing and corrections policies (Beckett, 1997). Thus, if tougher political policies are driving the variable changes in the inmate services staff ratio, states led by Republicans should show greater declines.
The results suggest a large and negative (though not significant) correlation between having a Republican governor in 2005 and the inmate services staff ratio (r = -.31, p = NS) and teacher ratio (r = -.26, p = NS) in 2005. Although the results are not statistically significant, the differences across states are substantial—on average, states led by Republican governors provide approximately 20 percent fewer inmate services staff members (per 1,000 inmates) than states led by Democratic and Independent governors. In addition, states with Republican governors in 1990 were more likely to see relatively larger declines in the inmate services staff ratio (r=-.20, p = NS) and the educational ratio (r = -19, p = NS) between 1990 and 2005, although these correlations again do not reach statistical significance. The results suggest that while party politics may have been part of the
story behind declining inmate services staff ratios throughout this period, there is substantial state heterogeneity.
Another measure of states' political climate is the presence of "supermax" facilities, which hold inmates in near-permanent solitary confinement and represent the most extreme version of the "no-frills" prison movement (Lynch, 2011; Mears, 2008; Reiter, 2012). Although definitions of supermax facilities differ, making it difficult to identify which states operate such facilities (Naday et al., 2008), the facilities census data provide information on whether individual facilities within states are identified (by the administrator completing the survey) as "supermax." This indicator appears to provide a lower-bound on the number of states with supermax facilities (16 states in 2005), suggesting these are the states with the most unambiguous supermax prisons.
The data suggest a negative (but not significant) relationship between the presence of supermaxes in 2005 and the 2005 staff ratio— states with at least one institution identified as supermaximum security averaged 41 professional staff per 1,000 inmates, while states without any supermaxes had a mean ratio of 50 (r = -.22, p = NS). Similarly, states with supermaxes had an average education staff ratio of eight, compared to 10 for states without supermax facilities (r=-.13, p = NS). However, there were no significant relationships between operating a supermax facility by 2005 and changes in staff ratios between 1990 and 2005. Thus, states operating facilities identified as supermax prisons tended to have lower staff ratios, although there remained significant state heterogeneity in this relationship.
Conclusions
The findings suggest that the narrative of decline in inmate rehabilitation during the punitive turn is correct in at least one important respect—between 1979 and 2005, almost all states experienced a drop in the average staff to inmate ratio for inmate services professionals. In 1979, the median state provided roughly 70 services staff members per 1,000 inmates; by 2005, this had been reduced to 47 staff members per 1,000 inmates. Thus, as the number of individuals behind bars rapidly expanded, the average inmate's ability to access services, such as education, counseling, and medical care, declined. On this measure of rehabilitation in practice, there is little "gap" between the rhetoric and realities of punishment: both have become more punitive since the 1970s.
However, the results also suggest several crucial qualifications. First, even in the more rehabilitation-oriented period of the late 1970s, average staff ratios for inmate services were quite low; the median state in 1979 funded 3.5 times as many security officers as inmate services staff. With limited services staff, most inmates were not receiving extensive rehabilitative programs or services. In previous work, I estimate that this level of staffing translated into 25 percent of state inmates participating in academic programs, fewer than 15 percent participating in psychological counseling, and less than 15 percent receiving alcohol or drug treatment on any given day (Phelps, 2011). Clearly, while rehabilitation may have been the predominant narrative of punishment (in some places), its implementation was limited.
Second, the decline in rehabilitative staff took place during the 1990s and early 2000s, rather than in the immediate "post-rehabilitation" period (the 1980s). This suggests that there was a substantial lag between changes in the narrative of punishment and punishment in practice. Throughout the 1980s, prisons were rapidly expanding staffing levels and program capacity to meet the influx of prisoners (Phelps, 2011). Only in the 1990s, when the expansions continued, did inmate services staffbegin to fall behind. It seems penal practices were buffered from the changes in rhetoric for at least a decade, with the trend in rehabilitative staffing shifting only after a protracted struggle over penal values and in the face of a continuing flood of prisoners.
Third, and strikingly, while there was an important decline in the services staff ratio during the 1990s and 2000s, this shift across
time paled in comparison to the tremendous variation across place. Figs. 1 through 4 suggest the predominant trend in every year is enormous state variation, with distributions across years overlapping considerably. For example, only four states in 2005 had ratios lower than the lowest value from 1990, and only four states in 1990 had values higher than the highest value in 2005. Further, this variation was regionally clustered, with the median inmate services staff ratio for states in the Northeast higher in 2005 than the median ratio for Southern states in any of the survey years between 1979 and 2005. Thus, with respect to the provision of inmate services staff, prisons in the Northeast in 2005 remain more rehabilitation-oriented than Southern prisons were at any point in recent decades. For understanding variation in this indicator of penal practices, place seems to matter more than time.
Together, these results document that while prisons have indeed become "harsher" places to do time, with fewer inmate services professionals, rehabilitation in practice was not particularly pervasive in the 1970s, nor is it entirely gone in the 2000s. This supports the claim that conclusions about the punitive turn have often overstated the coherence of the decline of rehabilitation. The results also provide another dimension of empirical support to the theoretical understanding of punishment as "braided" or "volatile and contradictory." Alongside increases in many punitive practices, states have continued to provide inmate services professionals who diagnose, treat, counsel, and otherwise tend to inmates' wellbeing. In addition, throughout the punitive turn, inmates' experiences of punishment have been diverse, with some states providing a more progressive context in which inmates have greater access to medical care, counseling, and education programs, while others provide only the most minimal levels of custodial care (or less). Thus, the rise of punitive rhetoric did affect practices, but did not completely realign the "field" within prisons (Page, 2011).
In addition to supporting accounts of punishment as complex and contradictory, the results also document what scholars have to gain from studying state variation. While all states experienced the buildup of mass incarceration, the scale of those changes and the experiences of the imprisoned differ drastically. Researchers will profit from studying how the experiences of incarceration have varied across both place and time. Rather than experiencing a monolithic punitive turn, punishment began—and remained—a very different endeavor in states across the country.
At the end of this article, I provided a brief sketch of one such research agenda, exploring state-level correlates of variation in the decline in inmate services staff. The results suggest states with larger and more overcrowded prisons, a higher percentage of black prisoners, and more punitive political cultures provided fewer inmate services staff in 2005. There was no evidence that judicial intervention protected states from the decline in the inmate services staff ratio, with states with any kind of court order in 1990 more likely to experience large losses in the inmate services staff ratio. Thus, these results suggest that there are logistical, political, and sociological explanations behind the huge span in states' provisions of inmate services. However, with the exception of the relationship between the percent of inmates identified as black and the inmate services staff ratio (driven by states in the South), these relationships did not reach statistical significance—even within broad trends, unexplained state variation remains. Part of the story here likely involves factors that are difficult to quantify, perhaps most importantly, the role of individual correctional administrators and political leaders in maintaining support for rehabilitative programs and services. Research suggests that correctional administrators have continued to support and defend these services, often using narratives unhinged from rehabilitation, such as concern for inmate management, legal rights of prisoners, and public safety (Schoenfeld, 2009). Future research might explore these state-level administrative histories to gain a deeper understanding of the link between changes in the
rhetoric and practices of punishment and shifts in the implementation of rehabilitative practices inside prisons.
Looking to the future, where prison rehabilitation is headed remains an open question. To some extent, the new rhetoric emerging around punishment appears to be more promising, with state leaders increasingly promoting a "smart on crime" rather than a "tough on crime" agenda and scholars tentatively suggesting that rehabilitation might be "back on the table" (Simon, 2008; Steen & Bandy, 2007; Wool & Stemen, 2004). However, states continue to face extreme budget shortfalls and in tough political times, prison programs are often first on the chopping block. Already, states that had made gains in providing inmate services are scaling back as budget cuts impact criminal justice (Scott-Hayward, 2009). The only sure bet is that states will vary in their responses to both fiscal crises and broader discourses on punishment.
Acknowledgements
Special thanks to Devah Pager, Malcolm Feeley, Chris Uggen, Philip Goodman, Kim Lane Scheppele, Mona Lynch, attendees at the 2010 Law & Society Association Annual Meeting and 2011 American Society of Criminology Annual Meeting (where earlier drafts of this article were presented), and the journal editors for an array of useful comments and critiques. Thanks also to the Joint Degree Program in Social Policy, the Sociology Department, and the Office of Population Research at Princeton University for material and intellectual support during the completion of this work. This research was supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE-0646086 and by the Department of Education's Jacobs K. Javits Fellowship.
1. This braiding of punitive and rehabilitative strands is consistent with the way the public responds to opinion polls about punishment, with respondents reporting approval for both harsher sanctions and more rehabilitative elements (Cullen, Fisher, & Applegate, 2000). See also Lynch and Richards (2011) on the varied penal goals expressed in corrections departments' mission statements.
2. The Bureau of Justice Statistics also fielded a facility census in 1974, but the data do not identify states.
3. Privately-run state prisons before this point were very uncommon; by 1990, only 1 percent of state inmates were housed in private prisons.
4. In 2005, the survey instructs administrators to exclude staff paid through contractual agreements, whereas in earlier years, administrators were instructed to include these staff members in the total. However, in most cases, these individuals comprise a minority of the workforce and are unlikely to substantially affect trends. In 2000, contractual staff and unpaid interns together accounted for just 3 percent of the reported total nationally. Trends between 1995 and 2000 are broadly consistent with the trends between 2000 and 2005, suggesting this change did not create significant bias.
5. The lower and upper adjacent values are defined by Tukey (1977) and conceptually represent the lower and upper bounds of the majority of the distribution. For example, in the 2005 data, the lower adjacent value is roughly the 1st percentile and the upper adjacent value is roughly the 90th percentile. Mathematically, the lower adjacent value is defined as the smallest value greater or equal to the 25th percentile minus 1.5 times the difference between the 25th and 75th percentiles, while the upper adjacent value is the largest value equal to or less than the 75th percentile plus 1.5 times the difference between the 25th and 75th percentiles.
6. Examining individual cases suggests that, in some instances, these results are an error. For example, in 2005, Alabama's prisons reported a total of 0 educational staff. However, in Alabama's Corrections Department's 2005 annual report, administrators discuss a variety of both academic and vocational programs operating inside of prisons (State of Alabama, 2005).
7. The median incarceration rate in 1980 for Southern states was 172 prisoners per 100,000 residents, while the median incarceration rate for Northeastern states in 2005 was 247.
8. The highest outliers for the inmate services staff ratio and education staff ratio in 2005 are West Virginia, but these statistics appear to be erroneous. One facility in the state reported 117 teachers—an extreme outlier considering that the 99 th percentile value for the number of teachers per facility in 2005 is 40 teachers. In addition, this facility is neither the largest facility in the state nor the prison with the largest educational programs (State of West Virginia, 2006).
9. The correlation between percent black in the resident population and the staff ratio is -.51 (p<.01) for all inmate services staff and -.28 (p<.05) for educational staff.
10. Similar results are produced with different cut-points and a continuous measure of percent of inmates housed in facilities with court orders.
References
Allen, F. A. (1981). The Decline of the Rehabilitative Ideal: Penal Policy and Social Purpose. Storrs lectures on jurisprudence. New Haven: Yale University Press.
Barker, V. (2009). The Politics of Imprisonment. Oxford University Press.
Beckett, K. (1997). Making Crime Pay: Law and Order in Contemporary American Politics. Studies in crime and public policy. New York: Oxford University Press.
Beckett, K., & Western, B. (2001). Governing Social Marginality: Welfare, Incarceration, and the Transformation of State Policy. Punishment & Society, 3(1), 43-59.
Bosworth, M. (2007). Creating the Responsible Prisoner: Federal Admission and Orientation Packs. Punishment & Society, 9(1), 67-85.
Campbell, M. C. (2011). Politics, Prisons, and Law Enforcement: An Examination of the Emergence of "Law and Order" Politics in Texas. Law & Society Review, 45(3), 631-665.
Carroll, L. (1998). Lawful Order: A Case Study of Correctional Crisis and Reform. New York: Garland Pub.
Chamberlain, A. W. (2011). Offender Rehabilitation: Examining Changes in Inmate Treatment Characteristics, Program Participation, and Institutional Behavior. Justice Quarterly, 29(2), 183-228.
Cheliotis, L. K. (2006). How Iron is the Iron Cage of New Penology?: The Role of Human Agency in the Implementation of Criminal Justice Policy. Punishment & Society, 8(3), 313-340.
Cullen, F. T., Fisher, B. S., & Applegate, B. K. (2000). Public Opinion about Punishment and Corrections. Crime and Justice, 27, 1-79.
Cullen, F. T., Latessa, E. J., Burton, V. S., Jr., & Lombardo, L. X. (1993). The Correctional Orientation of Prison Wardens: Is the Rehabilitative Ideal Supported? Criminology, 31(1), 69-92.
DiIulio, J. J. (1987). Governing Prisons: A Comparative Study of Correctional Management. New York: Free Press.
DiIulio, J. J. (Ed.). (1990). Courts, Corrections, and the Constitution: The Impact of Judicial Intervention on Prisons and Jails. New York: Oxford University Press.
Feeley, M. M., & Simon, J. (1992). The New Penology: Notes on the Emerging Strategy of Corrections and its Implications. Criminology, 30(4), 449-474.
Feeley, M., & Rubin, E. L. (1998). Judicial Policy Making and the Modern State: How the Courts Reformed America's Prisons. Cambridge criminology series. Cambridge, U.K.: Cambridge University Press.
Garland, D. (2001). The Culture of Control: Crime and Social Order in Contemporary Society. Chicago: University of Chicago Press.
Garland, D. (2004). Beyond the Culture of Control. Critical Review oflnternational Social and Political Philosophy, 7(2), 160-189, http://dx.doi.org/10.1080/1369823042000266576
Goodman, P. (in press). 'Another Second Chance': Rethinking Rehabilitation Through the Lens of California's Prison Fire Camps. Social Problems.
Greenberg, D. F., & West, V. (2001). State Prison Populations and Their Growth, 1971-1991. Criminology, 39, 615.
Hannah-Moffat, K. (2005). Criminogenic needs and the transformative risk subject. Punishment & Society, 7(1), 29-51.
Hutchinson, S. (2006). Countering Catastrophic Criminology: Reform, Punishment and the Modern Liberal Compromise. Punishment & Society, 8(4), 443-467.
Irwin, J. (2005). The Warehouse Prison: Disposal of the New Dangerous Class. Los Angeles, California: Roxbury Publishing Company.
Jacobs, D., & Carmichael, J. T. (2001). The Politics of Punishment across Time and Space: A Pooled Time-Series Analysis of Imprisonment Rates. Social Forces, 80(1), 61-89.
Jacobs, D., & Tope, D. (2007). The Politics of Resentment in the Post-Civil Rights Era: Minority Threat, Homicide, and Ideological Voting in Congress. The American Journal of Sociology, 112(5), 1458-1494.
Jacobs, J. B. (1977). Stateville: The Penitentiary in Mass Society. Studies in crime and justice. Chicago: University of Chicago Press.
Jacobs, J. B. (2003). Judicial Impact on Prison Reform. In T. G. Blomberg & S. Cohen (Eds.), Punishment and Social Control (2nd ed.). New York: Aldine de Gruyter.
King, R D., & Wheelock, D. (2006). Group Threat and Social Control: Race, Perceptions of Minorities and the Desire to Punish. Social Forces, 85,1255.
Klarner, C. (2003). The Measurement of the Partisan Balance of State Government. State Politics & Policy Quarterly, 3(3), 309-319.
Kruttschnitt, C., & Gartner, R (2005). Marking Time in the Golden State: Women's Imprisonment in California. Cambridge, U.K.: Cambridge University Press.
Lynch, J. P., & Sabol, W. J. (2001). Prisoner Reentry in Perspective (Vol. 3). Crime Policy Report, Washington, D.C.: Urban Institute.
Lynch, M. (1998). Waste Managers? The New Penology, Crime Fighting, and Parole Agent Identity. Law & Society Review, 32, 839.
Lynch, M. (2000). Rehabilitation as Rhetoric: The Ideal of Reformation in Contemporary Parole Discourse and Practices. Punishment & Society, 2(1), 40-65.
Lynch, M. (2009). Sunbelt Justice: Arizona and the Transformation of American Punishment. Stanford, California: Stanford Law Books.
Lynch, M. (2011). Mass incarceration, legal change, and locale. Criminology & Public Policy, 10(3), 673-698.
Lynch, M. J., & Richards, T. N. (2011). State correctional policy survey: What state correctional departments say they do. Journal of Crime and Justice, 34(2), 139-158.
Maurutto, P., & Hannah-Moffat, K. (2006). Assembling Risk and the Restructuring of Penal Control. British Journal of Criminology, 46(3), 438-454.
McNeill, F., Burns, N., Halliday, S., Hutton, N., & Tata, C. (2009). Risk Responsibility and Reconfiguration: Penal Adaptation and Misadaptation. Punishment & Society,!! (4), 419-442.
Mears, D. P. (2008). An Assessment of Supermax Prisons Using an Evaluation Research Framework. The Prison Journal, 88(1), 43-68.
Naday, A., Freilich, J. D., & Mellow, J. (2008). The Elusive Data on Supermax Confinement. The Prison Journal, 88(1), 69-93.
O'Malley, P. (1999). Volatile and Contradictory Punishment. Theoretical Criminology, 3(2), 175-196.
O'Malley, P. (2000). Criminologies of Catastrophe? Understanding Criminal Justice on the Edge of the New Millennium. Australian and New Zealand Journal of Criminology, 33(2), 153-167.
Page, J. (2011). The Toughest Beat: Politics, Punishment, and the Prison Officers' Union in California. New York: Oxford University Press.
Pastore, A. L., & Maguire, K. (Eds.). (2010). Sourcebook of Criminal Justice Statistics, 6.29.2010 [Online]. Available: http://www.albany.edu/sourcebook/pdf/t6292010. pdf [April 26, 2012].
Perkinson, R (2010). Texas Tough: The Rise of America's Prison Empire (First ed.). New York, N.Y.: Metropolitan Books.
Pew Center on the States (2010). Prison Count 2010. Washington, D.C.: Pew Center.
Phelps, M. S. (2011). Rehabilitation in the Punitive Era: The Gap Between Rhetoric and Reality in U.S. Prison Programs. Law & Society Review, 45(1), 33-68.
Raynor, P., & Robinson, G. (2009). Rehabilitation, Crime and Justice. Palgrave Macmillan.
Reiter, K. A. (2012). The Most Restrictive Alternative: A Litigation History of Solitary Confinement in U.S. Prisons, 1960-2006. Studies in Law, Politics, and Society, 57, 71-124.
Robinson, G. (2008). Late-modern rehabilitation: The evolution of a penal strategy. Punishment & Society, 10(4), 429-445.
Schlanger, M. (2006). Civil Rights Injunctions Over Time: A Case Study of Jail and Prison Court Orders. New York University Law Review, 81(2), 550-630.
Schoenfeld, H. (2009). The Politics of Prison Growth: From Chain Gangs to Work Release Centers and Supermax Prisons Florida, 1955-2000. (Doctoral Dissertation). Retrieved from Proquest. Northwestern University.
Schoenfeld, H. (2010). Mass Incarceration and the Paradox of Prison Conditions Litigation. Law & Society Review, 44(3-4), 731-768.
Scott-Hayward, C. (2009). The Fiscal Crisis in Corrections: Rethinking Policies and Practices. New York: Vera Institute of Justice.
Simon, J. (2007). Governing Through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear. Studies in crime and public policy. Oxford, U.K.: Oxford University Press.
Simon, J. (2008). Introduction. In M. L. Frampton, I. Haney-Lopez, & Jonathan Simon (Eds.), After the War on Crime: Race, Democracy, and a New Reconstruction. New York: New York University Press.
Spelman, W. (2009). Crime, Cash, and Limited Options: Explaining the Prison Boom. Criminology & Public Policy, 8(1), 29-77.
State of Alabama (2005). Annual Report for Fiscal Year 2005: A Proud Past - A Bright Future. Montgomery, AL: Department of Corrections.
State of West Virginia (2006). Corrections in West Virginia: FY 2005-06 Annual Report. Charleston, WV: Department of Military Affairs & Public Safety, Division of Corrections.
Steen, S., & Bandy, R. (2007). When the Policy Becomes the Problem: Criminal Justice in the New Millennium. Punishment & Society, 9(1), 5-26.
Tonry, M. H. (1995). Malign Neglect: Race, Crime, and Punishment in America. New York: Oxford University Press.
Tukey, J. W. (1977). Exploratory Data Analysis. Reading, MA: Addison-Wesley.
United States Department of Justice. Bureau of Justice Statistics (1997a). Census of State and Federal Adult Correctional Facilities, 1979 [Computer file]. Conducted by U.S. Dept. of Commerce, Bureau of the Census. 2nd ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor].
United States Department of Justice. Bureau of Justice Statistics (1997b). Census of State and Federal Adult Correctional Facilities, 1984 [Computer file]. Conducted by U.S. Dept. of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor].
United States Department of Justice. Bureau of Justice Statistics (1998). Census of State and Federal Adult Correctional Facilities, 1995 [Computer file]. Conducted by U.S. Dept. of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor].
United States Department of Justice. Bureau ofJustice Statistics (2001). Census of State and Federal Adult Correctional Facilities, 1990 [Computer file]. ICPSR09908-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor].
United States Department ofJustice. Bureau ofJustice Statistics (2004). Census of State and Federal Adult Correctional Facilities, 2000 [Computer file]. Conducted by U.S. Dept. of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor].
United States Department ofJustice. Bureau ofJustice Statistics (2010). Census of State and Federal Adult Correctional Facilities, 2005 [Computer file]. ICPSR24642-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor].
Useem, B., & Piehl, A. M. (2008). Prison State: The Challenge of Mass Incarceration. Cambridge, U.K.: Cambridge University Press.
Wacquant, L. (2001). Deadly Symbiosis: When Ghetto and Prison Meet and Mesh. Punishment & Society, 3(1), 95-133.
Wacquant, L. (2009). Punishing the Poor: The Neoliberal Government of Social Insecurity. Durham, NC: Duke University Press. Weaver, V. M. (2007). Frontlash: Race and the Development of Punitive Crime Policy.
Studies in American Political Development, 21 (02), 230-265. Western, B. (2006). Punishment and Inequality in America. New York: Russell Sage.
Wool, J., & Stemen, D. (2004). Changing Fortunes or Changing Attitudes? Sentencing
and Corrections Reforms in 2003. New York, NY: Vera Institute of Justice. Zimring, F. E., & Hawkins, G. (1991). The Scale of Imprisonment. Studies in Crime and Justice. Chicago: University of Chicago Press.
