The state of the job: An embedded work role perspective on prison officer attitudes

Abstract

Although the United States has grown increasingly punitive in the last three decades, there is considerable variation across states. On a variety of indicators, California is much more punitive than Minnesota. Using data from two original, large-N surveys, we analyze whether these differences in the orientations of state correctional systems are reflected in the attitudes of workers who are tasked with the day-to-day oversight of state prisons. With respect to the purpose of imprisonment, we find that California prison officers are significantly more punitive than those in Minnesota. In contrast, officers in each state express similar levels of support for basic rehabilitation programs. Based on these findings, we propose an embedded work role perspective, which posits that officers across states reflect a shared position within the prison organization, but that the prisons in which they work are embedded in broader penal and political environments that predict differences in attitudes across state contexts. This conceptualization of prison officer orientations has implications for policymakers, prison administrators, and scholars concerned with the politics and practice of work and incarceration.

Keywords

imprisonment prison officer punitiveness rehabilitation work role

Introduction

Until the late 1970s, scholars did not systematically examine prison officers’ lives or worldviews.¹ In many respects, rank-and-file staff were the ‘invisible ghosts of penality’ (Liebling, 2000: 337), one-dimensional specters that loomed over the ‘prison community’ but did not warrant serious scholarly attention, much less ‘appreciative inquiry’ (Liebling et al., 1999, 2011: 6–13). About a quarter-century ago, Kauffman (1988: 3) rightfully criticized this scholarly elision: ‘Failure to understand officers – their characters and motivations, problems and perspectives – has inevitably undermined efforts to reform prisons and has contributed to the everyday misery of those who live and work behind the walls.’

Recent studies reinforce Kauffman’s claim about prison officers’ central role in shaping life on the inside (Crewe, 2011; Crewe et al., 2011; Liebling, 2000, 2011a; Liebling et al., 2011; Sparks et al., 1996). Vuolo and Kruttschnitt (2008: 309) argue that prison officers ‘are one of the primary actors in the penal system and the individuals who are directly responsible for implementing new penal policies’. Garland (1990: 210) similarly maintains that prison officers (along with other front-line penal personnel, such as probation and parole officers) are ‘the primary bearers of… penal culture, and the agents who do the most to transform cultural conceptions into penal actions’. In many respects, prison officers are ‘street-level bureaucrats’ (Lipsky, 1980) who are responsible for implementing policy on the ground. As scholars have emphasized recently, prison officers (like officers on the street) have extensive discretionary power (Crewe, 2011: 456; see also Liebling et al., 2011: ch. 6). How officers use this discretion greatly affects prisoner quality of life, program implementation, and the routine operations of the prison. Put simply, without buy-in from front-line prison staff, new programs, policies, and routines have little chance of success (Lin, 2000).

Yet, despite this increasing recognition of the centrality of officers to prison policy and practice, we still do not fully understand the factors that shape officers’ perspectives. In particular, we know very little about whether and how officers across different states vary in their conceptualizations of the purpose of imprisonment and in their support for the provision of rehabilitation programs. In this article, we analyze data from two original, large-scale surveys of prison officers in California and Minnesota, seeking to understand the influences that shape officers’ attitudes. In short, we ask: Do officers in more punitive states express more punitive attitudes toward incarceration?

In comparing data from these states, we test two competing predictions. Drawing on the theoretical notion of embeddedness (Bourdieu, 1980, 1997; Bourdieu and Wacquant, 1992; Krippner and Alvarez, 2007; Melossi, 2001; Polanyi, 2001 [1944]), we posit that prisons and prison officers are embedded in particular penal fields – relatively bounded social spaces with unique historical and cultural traditions. These fields, we predict, influence officers’ attitudes about the purpose of imprisonment. Because California’s penal field is significantly more punitive than Minnesota’s, we expect that officers in California will express different and substantially more punitive attitudes than officers in Minnesota. The competing prediction stems from the considerable research contending that occupational role shapes workers’ professional orientations (Crawley, 2004; Haney et al., 1973; Jacobs and Retsky, 1980; Liebling, 2008; Lin, 2000; Owen, 1988; Skolnick, 1966; Sykes, 2007 [1958]). If officers’ position in the prison (and the work they do), rather than the character of the broader penal field, is the primary determinant of prison officer attitudes, we should instead expect that prison officers in the two states will express similar attitudes about the purpose of imprisonment and related issues.

In our survey data, we find evidence to support both positions. Our data show that officers’ support for particular rehabilitation programs is clearly grounded in their professional role; when considering the specific policies and procedures that should govern their work place, officers are largely informed by their location within the organization. This differs little between states, but is significantly predicted by features of the particular facility in which officers work, such as the prevalence of violence and the perceived quality of management. Thus, in line with existing work on the importance of the occupational role, we suggest that prison officers are similarly situated across states, tasked primarily with maintaining order and security, and this shapes the programmatic dimension of their professional beliefs.

At the same time, prison officers labor in prisons that are embedded in social contexts, which affect their ideological attitudes concerning the purpose of imprisonment. Although prison officers have a unique job that greatly influences their attitudes about penal policies, they are also concerned with more abstract norms, ideas, and ideologies that are shaped by the broader settings in which they are situated. The particular policy environment in which they and their institutions are embedded also shapes their views, and this varies substantially across states.

In sum, our data show that prison officers hold both ideological and programmatic attitudes, which are distinct dimensions along which they form beliefs about rehabilitation. To understand prison officer orientations, it is therefore necessary to comprehend both the nature of prison work, and also the larger contexts in which prison systems exist. Toward this end, we propose an embedded work role perspective, which posits that both within state and between state contexts shape officers’ attitudes. By considering the bi-dimensionality of officer attitudes, and by differentiating the distinct sources of their ideological and programmatic orientations, our embedded work role perspective takes into account both similarities in work roles and variation in state-level environments.

This study represents a substantial step forward from existing analyses of prison officer attitudes. Although there is now a rather expansive (and growing) literature on prison officers, existing studies have been limited to focusing on one prison, facilities within individual prison systems, or single facilities in different prison systems (Cullen et al., 1989; Dial et al., 2010; Farkas, 1999; Hussemann and Page, 2011; Jurik, 1985; Lambert, 2010; Lambert and Paoline, 2008; Lerman, 2008; Paboojian and Teske, 1997). To the best of our knowledge, no studies compare prison officers’ attitudes across prison systems.

Our comparative data allow us to examine if state context affects how officers think about imprisonment. As such, we are able to empirically investigate theoretical claims about the embeddedness of punishment (Melossi, 2001; Melossi et al., 2011). Specifically, our data allow us to analyze whether embeddedness affects how prison staff perceive important penal issues. At the same time, we are able to examine if the prison officer role trumps social context, reducing or eliminating variation between officers in different state contexts. As leading scholarship demonstrates that ‘staff culture’ greatly affects staff–prisoner relations and other central features of the carceral experience (see, for example, Crewe et al., 2011; Liebling, 2008), our analysis has important implications for penological theory and practice.

Literature review

State-level punitiveness

Over the last three decades, the United States has become increasingly punitive. The most glaring indicator of the nation’s punitive turn is the imprisonment rate (number of persons incarcerated under state and federal jurisdiction per 100,000 population), which skyrocketed from 139 in 1980 to 506 in 2007 (Bureau of Justice Statistics, 2009). Other indicators of the USA’s penal transformation include the decline of the rehabilitative ideal and ascent of the ‘new penology’ (the value-neutral management of populations defined in terms of their dangerousness) and retribution as leading penal strategies (Feeley and Simon, 1992; Garland, 2001; Tonry, 2004). America's intensified punitiveness is clear in the country’s enormous parole and probation populations; austere penal facilities; vast resources spent on pre-trial detention, incarceration, and post-release supervision; harsh and expansive policies (e.g. Three Strikes and You’re Out, mandatory-minimum drug laws, and sex-offender restrictions); and pervasive, often-racialized images and rhetoric about types of offenders such as ‘sexual predators’ and ‘career criminals’.

While largely accurate, the typical portrait obscures extensive variation across states. Zimring and Hawkins (1991: 150) remarked about state variance in their classic book The Scale of Imprisonment, ‘There is in fact more diversity in rates of imprisonment among the cross section of American states than one finds when a comparison is drawn across the whole of Western Europe.’ A growing body of research supports Zimring and Hawkins’s contention, showing that states vary greatly with respect to imprisonment rates (Beckett and Western, 2001; Frost, 2008), imprisonment risk (Frost, 2008) and length of prison terms (Frost, 2008); parole revocation rates (Jacobson, 2005); spending on punishment (Jacobs and Helms, 1999); and punitive policies (Zimring et al., 2001).

Comparing California and Minnesota highlights this variation; the two states differ considerably in their relative punitiveness on a variety of measures. Studies that classify states according to imprisonment figures consistently rank California as more punitive than Minnesota. For example, Beckett and Western (2001) characterize California as ‘punitive’ and Minnesota as ‘non-punitive’ based on the states’ respective imprisonment rates, defined as the number of people in prison per 100,000 state residents (475 for California and 113 for Minnesota in 1997). Frost (2008) argues that imprisonment rates are not sufficient measures of punitiveness because they do not account for states’ propensity to send offenders to prison (as opposed to other sanctions) or the length of prison terms. Therefore, she ranks states according to ‘imprisonment propensity’ and ‘imprisonment intensity’. Imprisonment propensity is ‘the probability that an index or drug offense arrest will result in a new prison commitment’ (Frost, 2008: 283). Imprisonment intensity is the actual time served in prison – that is, length of sentence. Based on her expanded definition of punitiveness, Frost describes Minnesota as the only ‘underpunitive’ state in the nation while characterizing California as ‘average’ in terms of punitiveness.

In an earlier work, Hinds (2005: 48) similarly argued that imprisonment figures provide an incomplete picture of punitiveness. She maintains that parole policies are also important indicators of penal harshness. Even controlling for imprisonment rates, states that imprison parole violators – particularly ‘technical violators’, those who violate the terms of their parole but are not convicted of new crimes – at high rates are more punitive than those that imprison parole violators at lower rates. Jacobson (2005) ranks states according to the ‘percentage successful among state parole discharges’. By this measure, in 1999, California had the second lowest parole success rate in the nation (21 percent), while Minnesota had a higher than average parole success rate (approximately 55 percent, compared to the national average of about 45 percent).

Jacobs and Helms (1999) argue that expenditures of resources on corrections are another important indicator of punitiveness. Correctional spending is not necessarily a measure of punitiveness, as states may spend large portions of their budgets on improving prison conditions and providing rehabilitation programs to inmates. However, in practice, the vast majority of spending goes to operational costs, such as staffing. As a 2004 Bureau of Justice Statistics report noted, ‘Over three-fourths of the States spent 96% or more of prison funds on current operations such as salaries, wages, benefits, supplies, maintenance, and contractual services’ (Stephan, 2004: 4). In terms of spending on corrections, California is again very punitive relative to Minnesota. In 2008, California’s corrections budget was $9.66 billion, more than 9 percent of the state’s general fund (Pew Center on the States, 2009; Smart Politics, 2009). California spent approximately $263 per capita, which was the second highest amount in the country. In that same year, Minnesota’s corrections budget was $460 million, only 2.6 percent of its general fund. Minnesota spent $88 per capita, which was the second lowest amount in the nation (Pew Center on the States, 2009; Smart Politics, 2009).

Each of these measures indicates that California’s correctional system is significantly more punitive than Minnesota’s. Indeed, the Golden State has some of the nation’s harshest penal policies. Since the early 1990s, California voters have used the ballot initiative to approve penal measures that greatly increase sentences for an array of offense types, intensify and extend community supervision for sex offenders, make it easier to send juvenile offenders to adult prisons, and significantly increase the period between parole hearings for prisoners with indeterminate sentences (e.g. those serving x-years-to-life) (Page, 2011). California is also renowned for having the nation’s most expansive and punitive ‘Three Strikes and You’re Out’ law (Zimring et al., 2001). Vuolo and Kruttschnitt (2008: 330–331) conclude about contemporary California: ‘there is no doubt that the scope of its legislative reforms and the massive increases in its prison population over the last two decades of the twentieth century place it in a class by itself’.

Minnesota maintains far less draconian penal policies. For example, unlike California, the state does not have the death penalty or Three Strikes and You’re Out. The state likewise contrasts in sentencing. In 1978, Minnesota became the first jurisdiction in the USA to enact sentencing guidelines (the guidelines went into effect in 1980). The state's sentencing commission sought to establish guidelines that would reduce disparities in sentencing; make prison commitments proportionate to offenses; and avoid prison overcrowding by preserving prison beds for the most serious offenders. The commission committed to keeping the prison population from exceeding 95 percent capacity (Frase, 2005: 132, 135). Frase (2005: 212) concludes that, after 25 years, Minnesota’s guidelines have ‘not only survived, they have survived largely intact and continue to achieve their goals of fair and rational sentencing’. In short, Minnesota prides itself on its relatively non-punitive approach to criminal punishment, which its sentencing commission and guidelines exemplify.

Competing hypotheses

State context and the penal field

Theoretical concepts within and beyond the sociology of punishment lead to competing predictions about whether and how the state-level variation discussed in the previous section will be reflected in the attitudes of correctional staff. Our first prediction derives from scholarship on the embeddedness of social action. Scholars use the concept of ‘embeddedness’ to underscore the interrelationship between practice and social environment. Economic sociologists, for example, maintain that economic activity is embedded in relations between people and institutions. Against the atomistic idea that individuals are rational actors with set preferences who make decisions independently of others, they argue that social networks shape economic interests (or preferences) and behavior (Granovetter, 1985; Krippner and Alvarez, 2007: 223). Drawing on the work of Polanyi, (2001 [1944]), other scholars use the embeddedness concept to characterize the interrelationship between the market and other institutions, especially the state. Countering the argument that the market is an autonomous (‘free’) entity, scholars maintain that the economy and state constitute each other (Krippner and Alvarez, 2007: 228–229). Institutions and those who work inside of them are embedded in relationships that affect their orientation and operations. Thus, markets, schools, prisons, and other institutions (as well as the actors within them) must be understood within the particular context in which they are embedded.

Whereas the concept of embeddedness has a prominent place in economic sociology, it is a rather recent addition to the sociology of punishment (although, a foundational axiom of the sociology of punishment is that penality is deeply embedded in society, and penal practices and social institutions are mutually constitutive (Garland, 1990)). Melossi (2001) employs the concept of embeddedness to understand why the United States has such a higher imprisonment rate than Italy, even though the two countries have comparable crime rates. In defining embeddedness, Melossi (2001: 405) writes, ‘specific historical institutions… cannot be conceived separately from the historical evolution and development of the larger setting of social action within which they have emerged – a setting constituted also through given cultural traditions’. He argues that there is an ‘elective affinity’ between punishment (specifically imprisonment) trends and religious traditions in Italy and the US. Protestantism in America and Catholicism in Italy are ‘repertoires of motive’ that elites draw on (however consciously or unconsciously) to promote particular penal approaches (2001: 414–415). These religious repertoires resonate because they are deeply embedded in specific national contexts. The counterfactual claim, then, is the repertoires would not resonate in places with different historical and cultural traditions. To understand cross-national variation in imprisonment rates, we therefore must examine the ‘the larger setting of social action’ that produces it.

Because punishment is embedded, the same policies (or rather, policies in the same domain) often look much different across place and time. In other words, local context mediates the transfer of policies from one place to another, giving the imported policies a local flavor (Nelken, 2010, 2011). Likewise, penal concepts are embedded in specific social settings. On this issue, Melossi (2001: 405) writes, ‘the usage of identical words often obscures the degree to which they are embedded in the different history of different places, as well as being articulated through (partially) different discourses’. Therefore, notions such as ‘rehabilitation’ or ‘public safety’ do not have universal definitions – their meanings are developed within specific times and geographical regions.

Melossi’s conceptualization of penal embeddedness directs our attention to historical and cultural underpinnings of national trends in penality. However, studying national context is not sufficient for understanding punishment in the USA because of the country’s federalist political system (Miller, 2008). A growing body of scholarship demonstrates that state-level factors (e.g. political culture, legal institutions, history of race and ethnic relations, and interest group activity) affect the scope and nature of punishment (Barker, 2009; Campbell, 2009; Gilmore, 2007; Lynch, 2010; Page, 2011; Schoenfeld, 2010; Zimring et al., 2001). Together, these state-level factors translate national (and even international) developments (e.g. neo-liberal economics) into concrete penal outcomes. In other words, punishment is embedded in the social and political cultures of American states.

Drawing on the work of Pierre Bourdieu, we expect that prison officers’ perspectives are affected by their position in their respective penal field.² Although Bourdieu did not use the term embeddedness, he advanced an embedded conception of social action. He argued that practice is the product of both an actor’s habitus (an internal set of dispositions that shape perception, appreciation, and action) and position in social fields (Bourdieu, 1980: 52, 1997: 138). For Bourdieu, fields are semi-autonomous, relatively bounded spheres of action that have particular orientations, forms of capital, and hierarchies (Bourdieu and Wacquant, 1992: 16–17). Each field also has rules (or regularities) and taken-for-granted assumptions and values – a field-specific common sense that Bourdieu calls doxa (Bourdieu and Wacquant, 1992: 98). Experienced agents in a field intuitively grasp the mores, expectations, and acceptable actions of that field (Bourdieu, 1980: 67).

The position of individuals, groups, and organizations in social fields greatly influences actors’ perceptions and, ultimately, behavior. Actors in a field – particularly those who have similar positions in the field (as do prison officers) – tend to have similar viewpoints. Hence, scholars have described the pugilistic habitus particular to boxers (Wacquant, 2004), the military habitus common to US soldiers (Lande, 2007), and the wildland firefighting habitus of US Forest Service workers (Desmond, 2006). Because all practice is embedded within particular objective contexts, we cannot understand actors’ attitudes or actions solely by examining their biographies, demographic characteristics, or positions within organizations. Instead, we must understand the broader context and meaning in which they operate.

State-level penal fields have particular histories, hierarchies, and cultural traditions that affect contemporary practice and policy outcomes. Like neo-institutionalist scholars, we argue that habitus is the social-psychological link between individuals and social structure (Emirbayer and Johnson, 2008; Powell and Dimaggio, 1991; Vaughn, 2008). As Vaughn (2008: 73) explains, habitus is the ‘connective tissue between macro-, meso-, and micro-levels of analysis’. Filtered through the prison organization, field-specific penal orientations affect officers’ habitus; hence, we posit that there is a ‘prison officer habitus’, but one that varies to a degree across states because of the embedded nature of imprisonment. We expect that the historical struggles and cultural traditions of Minnesota and California’s respective penal fields affect officers’ taken-for-granted assumptions, feelings, and opinions about the purpose of imprisonment and related ideological issues.

Since approximately the early 1980s, ‘punitive segregation’ has been the prevailing orientation of California’s penal field. This orientation promotes long sentences in austere prisons and extensive and intensive post-release supervision as the proper responses to crime (Garland, 2000). As such, the Golden State’s sentencing laws, parole practices, and prison conditions grew increasingly punitive in the last three-plus decades (Barker, 2009; Gilmore, 2007; Page, 2011). In contrast, ‘managerialism’ is the prevailing orientation of Minnesota’s penal field. Above all else, managerialism promotes the efficient and effective containment and control of populations, defined in terms of dangerousness or risk (Cheliotis, 2006; Feeley and Simon, 1992). It is a ‘pragmatic, technologically-supported, and quantification-oriented’ approach to imprisonment (Cheliotis, 2006: 397). Because it is guided by a managerial orientation, Minnesota’s field is far less expressly punitive than California’s (Hesselton, 2007). Since the two states’ penal fields differ, we predict that officers in California will have different (and more punitive) attitudes than their Minnesotan counterparts regarding the purpose of imprisonment and related issues.³

Prison officers and the occupational role

Scholarship on prison officers’ ‘occupational role’ suggests that this initial prediction may be incorrect or incomplete. The central proposition of this stream of research is that officers ‘can be distinguished according to the kind of work they are assigned’ (Jacobs and Retsky, 1980: 55). That is to say, the work that prison officers do and the responsibilities they hold – rather than the policy environment, demographics, or other factors – largely shape their attitudes about imprisonment and other job-related issues. Jacobs and Retsky (1980: 56) maintain, ‘Prevention of escape and riot is the primary task around which the role of the guard is organized. Closely related is maintenance of a modicum of internal order and security.’ Writing nearly a quarter-century after Jacobs and Retsky, Liebling (2000: 338) similarly asserts that the principal job of a prison officer is to ‘maintain secure custody, in a context where people are held in confinement against their will’. Prison officers are charged with enforcing the deprivations (foremost the deprivation of liberty) that produce the ‘pains of imprisonment’ (Sykes, 2007 [1958]). In short, because officers view custody as their central responsibility, they typically see rehabilitation and care for inmates as the domain of treatment, counseling, and medical staff (Crawley, 2004; Crewe, 2011; Lin, 2000).

There is general agreement among scholars that prison officers develop a ‘working personality’ similar to that of street police officers, which is characterized by deep suspicion, hyper-vigilance, social isolation, cynicism, and an ‘us’ versus ‘them’ attitude (Skolnick, 1966). About officers’ working personality, Liebling (2008: 106) writes,

Prison officers share with the police a tendency to feel their work has a public mission (public safety), to express cynicism and pessimism, due to the hard-nosed nature of their work, to be suspicious, conservative, macho, internally cohesive, and pragmatic.

Complicating (and extending) earlier conceptions of staff culture (e.g. Crouch and Marquart, 1980; Haney et al., 1973; Jacobs and Retsky, 1980), which focused almost exclusively on negative elements, Liebling (2008: 108) argues that this culture has ‘both positive negative characteristics’. Positive aspects include ‘preparedness to work as a team, a problem-solving and decisive approach to their work, having “bottle” (confidence), being multiskilled and flexible, having a sense of humor, and being experts at talk and diplomacy.’ Liebling (2008: 110) extends earlier research by highlighting variation in prison staff culture due to facility-level factors such as ‘indirectly expressed organizational goals’ (e.g. ‘resocialization’ or ‘discipline’), levels of distress, and relations with management, as well as whether officers feel significantly valued, have appropriate amounts of authority and responsibility, and receive sufficient resources to do their jobs. Yet, although these factors affect the ‘precise shape [officer] culture takes (that is, the degree, intensity, and form of the negative aspects of staff culture)’, the working personality of officers varies in terms of degrees, rather than kind (Liebling, 2008: 108). After all, the fundamental work role of officers is largely consistent across prisons – officers must maintain order in ‘low-trust environments’ (Crewe, 2011: 459) where they keep people locked up against their will.

Classic studies of prison staff maintain that officers typically oppose rehabilitation. Jacobs and Retsky (1980: 71) argue, for example: ‘In general, prison guards are cynical about rehabilitation and the work of treatment agents within the prison.’ More recent research, however, indicates that officers’ views on rehabilitation are not altogether negative. In fact, officers tend to support (or at least not strongly oppose) the provision of basic programs to prisoners, because such programs help them achieve their fundamental objective (maintaining order) and lessen tension behind the walls (DiIulio, 1991; Lin, 2000; Logan, 1993). Whether or not rehabilitation programs can or should be expected to assist inmates in changing their lives, they are desirable to officers for other reasons, such as helping to secure order in the prison by giving inmates a productive way to fill free time. Moreover, providing access to programs does not make the prison or prison staff responsible for transforming prisoners. Instead, it shifts the responsibility to prisoners, a process that scholars call ‘responsibilization’ (Garland, 2001; Lynch, 2000). Crewe (2011: 464) argues that contemporary rehabilitative programming (particularly that which goes under the label ‘cognitive behavioral therapy’) supports officers’ traditional views of prisoners:

the cognitive behavioural assumptions that dominate offending behaviour programmes are consistent with widely held officer beliefs about the causes of offending. Crime is seen as a pure choice; prisoners are regarded as cognitively deficient and as the architects of their own predicaments.

To the extent that programs help officers do their job, do not alter the main purpose of imprisonment (custody), or seriously challenge officers’ views about their charges, we would expect staff to support them.

In contrast to the embeddedness hypothesis described above, then, the prison officer work role perspective leads us to predict that officers will favor punishment and public safety over rehabilitation as the dominant purposes of imprisonment. And since this orientation is based in officers’ work role, we expect no significant variation across California and Minnesota, even though the states’ respective prison systems are embedded in penal fields with different histories, policies, and cultural traditions. Likewise, we would expect that officers in both states will support the provision of basic rehabilitation programs to prisoners, because the provision of these programs helps officers do their job and does not challenge officers’ views of crime (as a rational choice) and imprisonment (as punishment of offenders and to provide for public safety). In brief, whereas the embeddedness hypothesis predicts significant variation between states, the work role perspective predicts similarity across contexts.

The embedded work role perspective

How might we adjudicate between these competing hypotheses? In this article, we propose a hybrid embedded work role perspective. Prison officers are embedded, first, in their workplace (the state prison system), and, second, in the penal field – and both positions significantly affect their views. Thus, we expect that officers will likely share attitudes that stem from their position within the prison, but differ on attitudes tied to larger debates and priorities within their respective penal field. Critical to this perspective is recognition that officers’ correctional orientations have both an ideological and programmatic dimension. The first dimension of officers’ support for rehabilitation is concerned with its utility as an ideological goal of the prison. The second is concerned with the utility of concrete programs within the prison environment. Our contention is that the former is primarily shaped by the officers’ position within the broader penal field, while the latter is influenced to a greater degree by the officers’ work role within the prison institution.

The first prediction of the embedded work role perspective, then, is that officers in both states will support basic rehabilitative programs that help them accomplish workplace duties and improve the prison environment. Because officers in California and Minnesota share the same work role, they will express similar views about rehabilitation programs. In addition, we expect officers’ attitudes toward rehabilitation programs to be predicted by features of their work role, such as the percentage of gang-affiliated inmates they work with and their perceptions of managerial competence.

The second prediction of the embedded work role perspective is that the ideological attitudes of prison officers will reflect the respective penal fields in which the officers work. Just as other indicators of punitiveness vary across states – indicators of the extent of imprisonment; rates of parole revocation rates; spending on punishment; and punitive policies – so, too, should we expect the ideological orientations of correctional officers to vary. Given the substantial differences across their penal fields that we have already described, we expect officers in Minnesota to be significantly less ideologically punitive than officers in California.

Data

In order to measure the professional attitudes of prison officers, we designed and implemented two large-scale surveys: the California Correctional Officer Survey (CCOS) and the Minnesota Correctional Officer Survey (MNCOS). The CCOS, conducted in 2006, measured the attitudes and experiences of 5775 prison officers working in 32 of California’s adult state prisons. The MNCOS was administered in 2007 and gathered data on 911 officers in Minnesota’s eight state prisons. The California survey achieved a response rate of 33 percent, while the Minnesota survey had a response rate of 51 percent. These response rates are in the normal range for mail-administered surveys (see, for example, Shih and Fan, 2008) and although response rates varied by institution, no prisons in either state had to be excluded from analysis due to a lack of respondents. Both surveys were administered via mail, and questionnaires were sent to officers’ home addresses in order to assure confidentiality. The CCOS sample was obtained through the membership database of the California Correctional Peace Officers’ Association, the union representing prison officers and counselors in the Golden State. The database contains contact information for all current employees in unionized positions in the California correctional system, as well as many retired employees. The CCOS sample is drawn from only those individuals currently employed, and excludes those who are not rank-and-file officers (i.e. parole agents, counselors and higher-ranking custody personnel.) The MNCOS sample was obtained through the membership database of the American Federation of State, County and Municipal Employees, the union that represents prison officers in Minnesota. The database contains contact information for current employees in unionized positions in Minnesota’s prison system. As with the California sample, the Minnesota sample includes only currently employed rank-and-file officers. As Table 1 shows, respondent demographics in both states were quite representative of their respective populations.

Table 1.

Demographics of the CCOS and MNCOS

	California officers		Minnesota officers
	Sample (%)	Population (%)	Sample (%)	Population (%)
Race
White	55	46	94	91
Black or African American	9	12	1	3
Hispanic	27	34	1	1
Asian	3	2	1	2
Other	4	5	1	1
Missing data	4	–	2	–
Gender
Male	84	82	70	77
Female	16	18	30	23
Party identification
Democrat	23		34
Independent	30		44
Republican	47		21
	N = 5775	N = 21,243	N = 911	N = 1918

Notes:

California population data are taken from the Department of Corrections and Rehabilitation Office of Personnel Services for 12 October 2006.

CCOS race sums to more than 100% due to some respondents identifying more than one racial category.

Each survey asked officers a variety of questions to gauge the range of ways that an officer might express his or her attitudes toward correctional goals and policy. The first set of variables tap officers’ ideological beliefs about corrections. A punitive ideology holds that the primary functions of imprisonment are to punish offenders – that is, to exact retribution for law breaking – and enhance public safety through incapacitation and deterrence. In contrast, a rehabilitation-oriented ideology contends that the purpose of incarceration is to provide offenders with skills or resources they lack, in order to increase the likelihood that they return to society as law-abiding citizens. The CCOS and MNCOS surveys included three indicators designed to engage ideological orientations toward prison practice: ‘Do you feel that the purpose of a prison is rehabilitation, punishment, or both?’; ‘The job of a prison is to keep the public safe, not to help inmates’; and ‘Rehabilitation should be a central goal of incarceration.’

In addition to a broad ideological perspective, orientations toward correctional work can be expressed through concrete policy preferences. Officers with more punitive attitudes support the principle of penal austerity and oppose the provision of rehabilitation programs to prisoners, except to the extent that such programs have managerial utility. Those with less punitive attitudes support prisoner access to treatment programs that facilitate the processes of recovery and reintegration. To gauge the relative punitiveness of officers in California and Minnesota on this programmatic dimension, the surveys inquired about officers’ support for the provision of three specific types of rehabilitation-oriented programs: vocational training; drug and alcohol treatment; and academic training up to and including GED. In addition to these measures, which comprise our dependent variables, we asked officers a set of questions concerning their demographics, work history, and professional experiences.

Results

We start by assessing whether questions about support for rehabilitation as an ideology and support for specific rehabilitation programs measure a single underlying dimension of punitiveness, or whether these two sets of questions tap distinct dimensions of officers’ correctional orientation. Table 2 presents the results of a principal component factor analysis that examines the underlying structure of our data. This analysis clearly indicates that the questions concerning correctional ideology measure a separate attitudinal dimension from questions about correctional programs. This bi-dimensionality holds in both states.

Table 2.

Two dimensions of rehabilitation

	Factor loading
	California		Minnesota
	Programs	Ideology	Programs	Ideology
Ideology
Do you feel that the purpose of a prison is rehabilitation, punishment, or both?	.018	.755	.120	.837
Rehabilitation should be a central goal of incarceration	.284	−.724	.333	−.668
The job of a prison is to keep the public safe, not to help inmates	−.171	.718	−.304	.568
Programs
Inmates who want it should have access to academic training at least up to and including GED preparation	.836	−.033	.751	−.045
Inmates who want it should have access to drug and alcohol treatment	.720	−.109	.620	−.256
Inmates who want it should be given access to vocational training	.813	−.172	.824	−.125

Note: Results from factor analysis with principal component extraction and oblimin rotation.

We then turn our attention to whether levels of support for each dimension of rehabilitation vary across states. Relative to prison officers in California, Minnesota officers are significantly more likely to voice support for rehabilitation as a goal that should guide penal practice. When asked whether they agreed that ‘Rehabilitation should be a central goal of incarceration’, 67 percent of Minnesota officers agreed, compared to about 47 percent of officers in California. Though the pattern was similar in regards to the statement ‘The job of a prison is to keep the public safe, not to help inmates’, differences were somewhat smaller. On this question, about a third of officers in each state (32 percent of California officers and 37 percent of Minnesota officers) disagreed.

There were likewise sizable differences between officers in the two states in response to the question, ‘Do you feel that the purpose of a prison is rehabilitation, punishment, or both?’ On this question, about twice as many Minnesota officers as California officers thought that the function of imprisonment should be rehabilitation or ‘more rehabilitation, but still punishment’ (15 percent in California and 26 percent in Minnesota). And while the largest proportion of officers in each state believes that prisons should be oriented more toward punishment or that punishment should dominate, this was more so the case in California than Minnesota. In California, 61 percent of officers took the position that prison should be ‘totally punishment’ or that the goals of incarceration should be ‘more punishment, but still rehabilitation’. In Minnesota, just 46 percent indicated the belief that punishment should be the dominant goal. Roughly equal proportions of officers in each state believe that some combination of more and less punitive goals should be pursued. When asked whether the ‘purpose of a prison is rehabilitation, punishment, or both’, 25 percent of officers in California and 28 percent of officers in Minnesota chose balance between the two objectives.

In contrast to the significant variation between officers in California and Minnesota on questions concerning the goals of corrections, there is much less divergence regarding specific correctional programs. A large majority of officers in each state supports the provision of academic training, vocational training, and drug and alcohol treatment to inmates. In California, 84 percent of respondents agree that inmates who want it should be given access to vocational training; 90 percent say that inmates should have access to drug and alcohol treatment; and 91 percent of respondents agree that inmates who want it should have access to academic training up to and including GED preparation. Levels of support in Minnesota are similar: 79 percent for vocational training (a difference between the two states of 5 percent), 96 percent for drug and alcohol treatment (a difference of 6 percent) and 94 percent for academic training up to and including GED (a difference of 3 percent).

Figure 1 clearly shows these patterns. Here, we combine the individual measures of each dimension (support for rehabilitation as philosophy, and support for rehabilitation programs) to create two indices. The Ideology Index includes the three questions gauging support for rehabilitation as a correctional philosophy: agreement that the purpose of a prison is rehabilitation; agreement that rehabilitation should be a central goal of corrections; and disagreement that the job of a prison is to keep the public safe, not to help inmates. The Programs Index includes measures of support for the three types of basic rehabilitation programs: academic training up to and including GED; vocational training; and drug and alcohol treatment. Both measures are additive linear indices scaled from 0 to 1.

Figure 1.

Ideology and program indices, by state and security level.

The data show significant state-level variation on the Ideology Index. On average, prison officers in Minnesota score significantly higher on this measure (F = 15, p < .001). Compared to a mean of .50 (with a standard deviation of .17) in California, which falls at the halfway point of the index, Minnesota officers hold average views about 8 percentage points higher on the scale, a mean of .58 (standard deviation .16). On the ideological dimension, measuring officers’ orientations toward the purpose of imprisonment, officers in California are significantly more punitive than officers in Minnesota. In contrast, differences across the two states are smaller on the index gauging support for rehabilitation programs. On the Programs Index, California officers have a mean score of roughly .76 (SD = .15) and officers in Minnesota have a mean score of about .78 (SD = .14), a difference of less than two percentage points (F = 11.29, p < .01).

This pattern differs only slightly across security levels (see Figure 1). Among officers working primarily with minimum and medium custody inmates, those in Minnesota are significantly more supportive of rehabilitation as an ideology than those in California (F = 8.02, p < .01 for Level I; F = 10.59, p = .001 for Level II; F = 55.349, p < .011 for Level III). In contrast, there is no significant difference across states on the Programs Index at any of these security levels. At the highest security level, however, there are differences across states on both indices. Minnesota officers working with maximum security inmates are more supportive of rehabilitation programs (F = 28.609, p < .001) and of rehabilitation as a correctional ideology than California officers (F = 96.764, p < .001). However, the size of the difference on the Ideology Index – a gap of about .09 – is almost twice as large as the between-state differences on the Programs Index.

In order to assess the independent effect of state context, we must verify that differences (and similarities) are not solely explained by the demographic characteristics of officers in the two states. For example, as shown in Table 1, officers in the California sample are considerably more racially heterogeneous than Minnesota officers. In order to account for these potential confounders, we estimate the effect of working in California relative to Minnesota in a set of Ordinary Least Squares regressions with fixed effects for institution.⁴ The first model regresses the Ideology Index on state of employment and controls for officer race, gender, age and education, and partisan identification. We also include measures that assess the officers’ length of experience within the state prison system. These indicators include an officer’s tenure at his current prison, and his tenure within his respective DOC. Finally, we include a variety of measures of the work context, including the security level to which an officer is primarily assigned, perceptions of the likelihood that officers will be assaulted by inmates, reported frequency of violent incidents, the percent of inmates an officer considers ‘very dangerous’, the percent of inmates perceived to be gang-affiliated at the institution, and two measures of perceived managerial competence: whether an officer feels that he or she has received sufficient job training, and whether he or she believes that there is someone who can help resolve work-related problems. The second model regresses the Programs Index on state of employment and the same set of controls (see Figure 2).⁵

Figure 2.

Predicting ideology and program support.

The results of these models, presented in Figure 2, make clear that the two dimensions of rehabilitation are predicted by a distinct set of factors. In terms of officer demographics, we find that male officers are less likely than female officers to express support for rehabilitation as a goal of corrections, and are less supportive of rehabilitation programs, all else equal. Both indices are likewise predicted by partisan identification; Republicans and Independents are less supportive than Democrats of both rehabilitation programs and ideologies. In addition, age appears to be positively associated with scores on the Ideology Index, but not the Programs Index, and education is not significantly predictive of either index. Being white relative to minority (including black, Latino and other) predicts slightly more support for rehabilitation programs, but less support for rehabilitation as an ideology that should guide correctional practice.

As predicted by the embedded work role perspective, significant differences emerge in the institutional and organizational variables that predict each dimension. Most centrally, we find that all else being equal, officers in California have a significantly more punitive correctional ideology than their Minnesota counterparts. Relative to working in Minnesota, and controlling for features of the institution, working in California predicts a decrease of nearly 20 percentage points on the Ideology Index. This effect is highly statistically significant (p < .001) and persists in various specifications of the model. In contrast, the models reveal that officers in California and Minnesota do not differ in their support for rehabilitation programs. Controlling for a variety of demographic and institutional factors, working in California relative to Minnesota predicts a decrease of less than one percentage point on the Programs Index, a difference that is not statistically significant at conventional levels.

There is not uniformity within each state, however. Instead, length of service and several facility-level factors are also predictive of officers’ ideological orientation toward corrections. These models indicate that officers’ relative tenure in their state Department of Corrections is predictive of their correctional ideology, such that those who have been working in the system longer are less supportive of rehabilitation as a philosophy of corrections.⁶ In contrast, tenure at a particular institution does not significantly associate with attitudes toward rehabilitation on either dimension. In addition, perceptions of managerial competence appear to predict greater support for both a rehabilitation ideology and rehabilitation programs. Officers who believe they have been well trained to carry out their job and who feel that they have help from superiors to resolve workplace issues score higher on each index. However, the magnitudes of these effects are small.

Finally, in line with previous research, perceptions of workplace violence are predictive of both indices. However, also in line with expectations, these effects are in different directions for each attitudinal dimension. Officers who express concerns about inmate–staff assault rates and report high incidence of violence are less supportive of rehabilitation as a correctional ideology. Conversely, officers who believe that a high proportion of inmates are very dangerous, who report higher frequency of violent incidents in their institution and who report more gang-affiliated inmates are actually more supportive of proffering rehabilitation programs. In these models, we do not find a significant effect for security level on either index.⁷

Discussion

Through our comparative analysis, we have investigated an important, hitherto unexamined question: Does state context matter in predicting prison officer orientations? We began by examining two sets of theories that attempt to explain and predict prison officers’ professional orientations. Scholarship on the embeddedness of social action suggests that officers’ positions in their respective penal fields significantly affect their attitudes. Therefore, officers in California should have substantially more punitive attitudes than officers in Minnesota. On the other hand, theories of the occupational role suggest that similarly situated workers hold comparable views about the nature of their work. Prison officers in different prison systems share the same basic job – to oversee the safety and security of the prison institution – and face similar challenges and constraints. Thus, both California and Minnesota officers should have similar professional orientations.

Our findings suggest that the guiding logic of the penal field, which prison officers are charged with implementing on the ground, affects officers’ attitudes regarding the ideology of corrections. Consistent with the orientation of their respective penal fields, California officers are more punitive than Minnesota officers concerning the purpose of incarceration. However, the orientation of the particular penal field is less important to officers’ attitudes about concrete programs for prisoners. The attitudes of officers concerning the provision of programs are consistent in both states.

In sum, our results provide support for both hypotheses. As regards the guiding philosophy of imprisonment, we contend that state context matters; on this dimension, we find support for the embeddedness hypothesis. In contrast, on questions concerning rehabilitation as a set of concrete programs that should or should not be made available to offenders, we do not find significant differences between officers in the two states. Therefore, with respect to attitudes toward concrete programs, we conclude that state context does not matter; on this dimension, we find support for the prison officer work role hypothesis.

Our data allow us to address the important question of state context and officer attitudes in ways that go beyond previous studies, which have focused on only one prison or one prison system. However, this study is just a first step toward understanding the complex set of forces that shape the prison officer experience and influence officers’ perceptions and practice. Additional comparative research is needed to further flesh out and test the validity of the embedded work role perspective. For example, additional studies are necessary to more clearly specify all the salient features of penal fields that differ across states. One way to do this might be to examine changes over time in officers’ attitudes as state policies, practices, and orientations change. It is also important to see whether the patterns we uncover hold in other states, particularly in states that fall along other points on the continuum of criminal punishment.

Future research also is needed to examine if and how prison officer unions mediate trends in the penal field and officer orientations. In states (and countries) with active officer unions, we would expect that these labor unions help shape officer attitudes about management, inmates, the purpose of imprisonment and related issues (Liebling et al., 2011; Page, 2011). Since the early 1980s, California’s officer union has been very engaged in struggles within the penal and political fields and has strongly reinforced particular views of crime and punishment among its membership. Until quite recently, these views were expressly punitive (Page, 2011). In contrast, Minnesota’s officer union largely refrains from participation in debates about sentencing and other hot-button penal matters, focusing primarily on ‘bread and butter’ issues like wages, job security, and benefits (interview with David Crist, deputy director of the Minnesota Department of Corrections by Joshua Page, 8 July 2010). Officers’ relationships to unions, along with their location in prisons and the larger penal field, might be an important determinant of officer attitudes. So, too, might the relative politicization of the crime issue across states matter for how officers’ attitudes develop, particularly in terms of the ideologies they espouse. A more intensive analysis of the politics of the penal field, and how it interacts with officers’ partisan identification, might help to further flesh out the mechanisms through which state context shapes the attitudes of line staff and others working in correctional departments.

This study also uncovers a bi-dimensionality in officer attitudes that demands further exploration. Officers in both states clearly distinguish between the ideological underpinnings of prison work and the practical policies implemented behind prison walls; while beliefs about these two dimensions are related, they are also distinct, and many officers support rehabilitation programs while expressing ambivalence or outright hostility toward rehabilitation as a prison ideology. One of our most interesting findings, which underscores this bi-dimensionality, is that perceptions of facility-level violence and dangerousness predict strong opposition to rehabilitation as an ideology, but support for the provision of rehabilitative programs. Based on the occupational role perspective, we would hypothesize that officers express favorable attitudes about prisoner access to programs because they believe that such programs can decrease violence and make their jobs less dangerous.

Researchers would do well to incorporate this bi-dimensionality into existing strains of research on officer attitudes. Which dimension is more important for the discretionary choices that officers make about how to do their job? Which governs relationships with inmates and thus has bearing on inmate outcomes? Which appears most correlated with officers’ levels of work-related stress and work–life conflict? And which is most likely to predict officer burnout and staff turnover?

Our findings and theoretical perspective also have implications for contemporary policymakers and prison administrators. As scholars in the UK have documented in a series of empirical analyses of both public and private prisons, officer attitudes greatly affect officer behavior. In turn, officer behavior is central to prison operations and climate (see, for example, Crewe, 2011; Crewe et al., 2011; Liebling, 2011a, 2011b). In this regard, Liebling (2008: 118) writes:

Staff attitudes, in other words, are linked to behavior, as studies of police culture have indicated, as well as to prisoner-related outcomes. Culture, then, is related to the likelihood of implementation of new policies, as well as to outcomes for prisoners … Staff attitudes translate into regime qualities that can make the difference between a survivable experience of imprisonment and an unbearable one.

Understanding the determinants of prison officer attitudes is therefore not simply an empirical or theoretical exercise. It is an essential first-step in developing strategies for shaping officer orientations in ways that promote positive changes behind the walls.

Likewise, the nature and effects of prison officer culture should be important to lawmakers and state officials throughout the United States, who are currently searching for ways to reduce prison populations and save resources. Politicians in some states believe that creating new and expanding old rehabilitation efforts can help achieve these aims.⁸ Indeed, a central motivation for the re-organization of California’s penal bureaucracy in 2005 was to put rehabilitation on equal, if not higher, footing as punishment.⁹ David Crist (interview by Joshua Page, 8 July 2010), deputy director of the Minnesota Department of Corrections, maintains that Minnesota is also re-prioritizing rehabilitation:

I think we’re seeing another paradigm shift, another shift of the pendulum. We’re seeing our department swing away from that just deserts notion that we’re not responsible for rehabilitating anybody to it is our responsibility to rehabilitate. And people expect that we’re going to put out a rehabilitated inmate, and not somebody that has just done his time.

In order to accomplish these goals, policymakers and prison administrators need buy-in from prison officers. Our data suggest that they will likely receive support from the majority of line staff for increasing basic rehabilitative programs and services. Although the prison officer role promotes skepticism about inmates and the rehabilitative enterprise, officers tend to support (or at least tolerate) basic programs that fit within their principle objectives – particularly maintaining order (Lin, 2000: 56). For example, in both California and Minnesota, offering programs to inmates who want them garners high levels of support among officers, and we expect they will continue to take this position so long as they believe it does not preclude or contradict prison officers’ primary role: to secure the safety and order of the institution. It will be more difficult to obtain backing from officers for fundamentally transforming the purpose of imprisonment from punishment to rehabilitation. After all, even in a state like Minnesota, which ranks consistently low on various indicators of punitiveness, a large proportion of officers believe that punishment should be the dominant goal of imprisonment. Thus the framing of new programs and programmatic shifts will be consequential. To succeed, new programs must be framed and structured in ways that are consistent with officers’ work roles and professional identity. Introducing these programs as part of general efforts to decrease in-prison misconduct and tension would likely be more fruitful than introducing them as changes in how officers do their jobs or changes in the goals of officers’ work.

Our findings align with other research that has demonstrated that facility-level factors affect officer culture and, therefore, the possibilities for program implementation (Liebling, 2008). Taken together, our data suggest that policymakers and administrators can shape officer views of rehabilitation by improving training, decreasing violence, and enhancing managerial competence and support for line staff. Based on our findings, positive change in these areas would indirectly increase officer support for helping inmates by making the prison officer work role less onerous and dangerous.

However, our embedded work role perspective suggests that improving facility-level conditions only on the margin is not sufficient for changing officers’ perspectives. As long as state-level penal fields focus much more on punishing and managing than on rehabilitating offenders, altering officers’ attitudes will be extremely difficult. After all, prison officers are embedded within particular objective contexts that influence how they view their function and that of the institutions in which they toil. Like Liebling (2008: 119), we believe that dominant views on the ‘moral status of prisoners’ (which are expressed and institutionalized through sentencing policies and imprisonment practices) reinforce traditional elements of the prison officer ‘working personality’ that promote opposition to programs and opportunities that significantly help prisoners. Without seriously altering prevailing sentiments about prisoners within and beyond the penal field, it will be extremely difficult to get buy-in from front-line officers for making contemporary prisons less punitive and austere and more ‘correctional’ and humane.

Footnotes

Acknowledgements

Thanks are due to the following organizations for their financial support: the Survey Research Center, Institute of Governmental Studies, IGERT Program, Goldman School of Public Policy, and Graduate Division at the University of California, Berkeley; the Center for Evidence-Based Corrections at the University of California, Irvine; the Fox Leadership Center at the University of Pennsylvania; the National Science Foundation; and the Life Course Center and College of Liberal Arts at the University of Minnesota. For their assistance in administering the CCOS, we are indebted to the California Correctional Peace Officers Association, particularly Mike Jimenez and Chuck Alexander; Jody Lewen of the Prison University Project; and Professor Henry Brady, Bob Lee, and the staff of the Survey Research Center at the University of California at Berkeley. For help administering the MNCOS, we are grateful to AFSCME Council 5. Thanks are also due to Jeannette Hussseman, Jennifer Onofrio, Lisa Steacy, and Natalie Torres for research assistance, and to Letta Page for editing. Finally, we are very much indebted to the prison officers of the CDCR and the Minnesota DOC for participating in this study.

Notes

Amy E Lerman is an assistant professor of politics and public affairs at Princeton University. Her research focuses on race and inequality; social identity; social policy; and political participation.

Joshua Page is an assistant professor of sociology at the University of Minnesota. He is the author of The Toughest Beat: Politics, Punishment, and the Prison Officers Union in California. His main research interests are crime and punishment, labor, organizations, and politics.

References

Austin

Clear

Duster

(2007) Unlocking America: Why and How to Reduce America’s Prison Population. Washington, DC: JFA Institute.

Barker

(2009) The Politics of Imprisonment: How the Democratic Process Shapes the Way America Punishes Offenders. New York: Oxford University Press.

Beckett

Western

(2001) Governing social marginality: Welfare, incarceration, and the transformation of state polic? In: Garland

(eds) Mass Incarceration: Social Causes and Consequences. London: SAGE.

Bourdieu

(1980) The Logic of Practice. Palo Alto, CA: Stanford University Press.

Bourdieu

(1997) Pascalian Meditations. Palo Alto, CA: Stanford University Press.

Bourdieu

Wacquant

(1992) An Invitation to Reflexive Sociology. Chicago, IL: University of Chicago Press.

Bureau of Justice Statistics (2009) Key facts at a glance: Incarceration rate, 1980–2008. Available at: http://bjs.ojp.usdoj.gov/content/glance/tables/incrttab.cfm (accessed 1 August 2010).

Campbell M (2009) Agents of Change: Law Enforcement, Prisons, and Politics in Texas and California. PhD Thesis, University of California, Irvine.

Cheliotis

(2006) Penal managerialism from within: Implications for theory and researc? International Journal of Law and Psychiatry 29: 397–404.

10.

Cloward

Cressey

Grosser

(1960) Theoretical Studies in Social Organization of the Prison. New York: Social Science Research Council.

11.

Crawley

(2004) Doing Prison Work: The Public and Private Lives of Prison Officers. Cullompton: Willan Publishing.

12.

Crewe

(2011) Soft power in prison: Implications for staff–prisoner relationships, liberty and legitimac? European Journal of Criminology 8(6): 455–468.

13.

Crewe

Liebling

Hulley

(2011) Staff culture, use of authority, and prisoner quality of life in public and private sector prison? Australian & New Zealand Journal of Criminology 44(1): 94–115.

14.

Crouch

Marquart

(1980) On becoming a prison guar? In: Crouch

(eds) The Keepers: Prison Guards and Contemporary Corrections. Springfield, IL: Charles C Thomas.

15.

Cullen

Lutze

Link

Wolfe

(1989) The correctional orientation of prison guards: Do officers support rehabilitation? Federal Probation 53: 33–42.

16.

Desmond

(2006) Becoming a firefighte? Ethnography 7(4): 387–421.

17.

Dial

Downy

Goodlin

(2010) The job in the joint: The impact of generation and gender on work stress in priso? Journal of Criminal Justice 38: 609–615.

18.

DiIulio

(1991) No Escape: The Future of American Corrections. New York: Basic Books.

19.

Emirbayer

Johnson

(2008) Bourdieu and organizational analysi? Theoretical Sociology 37: 1–44.

20.

Farkas

(1999) Correctional officer attitudes toward inmates and working with inmates in a ‘get tough’ er? Journal of Criminal Justice 27(6): 495–506.

21.

Feeley

Simon

(1992) The new penology: Notes on the emerging strategy of corrections and its implication? Criminology 30(4): 449–474.

22.

Frase

(2005) Sentencing guidelines in Minnesota, 1978–200? Crime and Justice 32: 131–219.

23.

Frost

(2008) The mismeasure of punishment: Alternative measures of punitiveness and their (substantial) consequence? Punishment & Society 10(3): 277–300.

24.

Garland

(1990) Punishment and Modern Society: A Study in Social Theory. Chicago, IL: University of Chicago Press.

25.

Garland

(2000) The culture of high crime societies: Some preconditions of recent ‘law and order’ policie? British Journal of Criminology 40: 347–375.

26.

Garland

(2001) The Culture of Control: Crime and Social Order in Contemporary Society. Chicago, IL: University of Chicago Press.

27.

Gilmore

(2007) Golden Gulag: Prisons, Surplus, Crisis, and Opposition in Globazing California. Berkeley, CA: University of California Press.

28.

Granovetter

(1985) Economic action and social structure: A theory of embeddednes? American Journal of Sociology 91: 481–510.

29.

Haney

Banks

Zimbardo

(1973) Interpersonal dynamics in a simulated priso? International Journal of Criminology & Penology 11: 69–97.

30.

Hesselton D (2007) Crisis and Reform at Minnesota’s Stillwater Prison, 1960–2000. PhD Thesis, University of Minnesota.

31.

Hinds

(2005) Crime control in western countries 1970 to 200? In: Pratt

Brown

Hallsworth

Morrison

(eds) The New Punitiveness: Trends, Theories, Perspectives. Cullompton: Willan Publishing.

32.

Hussemann

Page

(2011) Gender diversity and the prospects for progressive prison refor? Women and Criminal Justice 21(4): 267–289.

33.

Jacobs

Helms

(1999) Collective outbursts, politics, and punitive resources: Toward a political sociology of spending on social contro? Social Forces 77: 1497–1523.

34.

Jacobs

Retsky

(1980) Prison guar? In: Crouch

(eds) The Keepers: Prison Guards and Contemporary Corrections. Springfield, IL: Charles C Thomas.

35.

Jacobson

(2005) Downsizing Prisons: How to Reduce Crime and End Mass Incarceration. New York: New York University Press.

36.

Jurik

(1985) Individual and organizational determinants of correctional officer attitudes toward inmate? Criminology 2: 523–539.

37.

Kauffman

(1988) Prison Officers and Their World. Cambridge, MA: Harvard University Press.

38.

King

Honaker

Joseph

Scheve

(2001) Analyzing incomplete political science data: An alternative algorithm for multiple imputatio? American Political Science Review 95(1): 49–69.

39.

Krippner

Alvarez

(2007) Embeddedness and the intellectual projects of economic sociolog? Annual Review of Sociology 33: 219–240.

40.

Lambert

(2010) The relationship of organizational citizenship behavior with job satisfaction, turnover intent, life satisfaction, and burnout among correctional staf? Criminal Justice Studies 23(4): 361–380.

41.

Lambert

Paoline

(2008) The influence of individual, job, and organizational characteristics on correctional staff job stress, job satisfaction, and organizational commitmen? Criminal Justice Review 33(4): 541–564.

42.

Lande

(2007) Breathing like a soldier: Culture incarnat? The Sociological Review 55: 95–108.

43.

Lerman A (2008) The Citizens Prisons Produce: How Criminal Justice Policies Shape American Communities and Civic Life. PhD Thesis, University of California, Berkeley.

44.

Liebling

(2000) Prison officers, policing, and the use of discretio? Theoretical Criminology 4(3): 333–357.

45.

Liebling

(2008) Why prison staff culture matter? In: Byre

Hummer

Taxman

(eds) The Culture of Prison Violence. Boston, MA: Pearson.

46.

Liebling

(2011a) Distinctions and distinctiveness in the work of prison officers: Legitimacy and authority revisite? European Journal of Criminology 8(6): 484–499.

47.

Liebling

(2011b) Moral performance, inhuman and degrading treatmen? Punishment & Society 13(5): 530–550.

48.

Liebling

Prince

Elliott

(1999) Appreciative inquiry and relationships in priso? Punishment & Society 1(1): 71–98.

49.

Liebling

Price

Shefer

(2011) The Prison Officer. New York: Willan Publishing.

50.

Lin

(2000) Reform in the Making: The Implementation of Social Policy in Prison. Princeton, NJ: Princeton University Press.

51.

Lipsky

(1980) Street-Level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.

52.

Logan CH (1993) Criminal justice performance measures for prisons. In: DiIulio JJ (ed.) Study group on criminal justice performance measures. Bureau of Justice Statistics, Princeton Project.

53.

Lynch

(2000) Rehabilitation as rhetoric: The ideal of reformation in contemporary discourse and practic? Punishment & Society 2(1): 40–65.

54.

Lynch

(2010) Sunbelt Justice: Arizona and the Transformation of American Punishment. Palo Alto, CA: Stanford University Press.

55.

Melossi

(2001) The cultural embeddedness of social control: Reflections on the comparison of Italian and North-American cultures concerning punishmen? Theoretical Criminology 5(4): 403–424.

56.

Melossi

Sozzo

Sparks

(2011) Travels of the Criminal Question: Cultural Embeddedness and Diffusion. Oxford: Hart.

57.

Miller

(2008) The Perils of Federalism. New York: Oxford University Press.

58.

Neklen

(2010) Comparative Criminal Justice. New York: SAGE.

59.

Neklen

(2011) Theorizing the embeddedness of punishmen? In: Melossi

Sozzo

Sparks

(eds) Travels of the Criminal Question: Cultural Embeddedness and Diffusion. Oxford: Hart.

60.

Owen

(1988) The Reproduction of Social Control: A Study of Prison Workers at San Quentin. New York: Praeger.

61.

Paboojian

Teske

RHC

(1997) Pre-service correctional officers: What do they think about treatment? Journal of Criminal Justice 25: 425–433.

62.

Page

(2011) The Toughest Beat: Politics, Punishment, and the Prison Officers Union in California. New York: Oxford University Press.

63.

Page

(forthcoming) Punishment and the penal fiel? In: Simon

Sparks

(eds) The Handbook on Punishment and Society. New York: SAGE.

64.

Pew Center on the States (2009) One in 31: The Long Reach of American Corrections. Washington, DC: Pew Center on the States.

65.

Pew Center on the States (2010) Prison Count 2010: State Population Declines for the First Time in 38 Years. Washington, DC: Pew Center on the States.

66.

Polanyi

(2001 [1944]) The Great Transformation: The Political and Economic Origins of Our Time. Boston, MA: Beacon.

67.

Powell

Dimaggio

(1991) The New Institutionalism in Organizational Analysis. Chicago, IL: University of Chicago Press.

68.

Schoenfeld H (2010) The Politics of Prison Growth: From Chain Gangs to Work Release Centers and Supermax Prisons, Florida, 1955–2000. PhD Thesis, Northwestern University.

69.

Shih

Fan

(2008) Comparing response rates from web and mail surveys: A meta-analysi? Field Methods 20(3): 249–271.

70.

Skolnick

(1966) Justice without Trial: Law Enforcement in Democratic Society. New York: John Wiley & Sons.

71.

Smart Politics (2009) Should Minnesota‘s corrections budget be increased? Available at: http://blog.lib.umn.edu/cspg/smartpolitics/2009/05/should_minnesotas_corrections.php (accessed 1 August 2010).

72.

Sparks

Bottoms

Hay

(1996) Prisons and the Problem of Order. Oxford: Clarendon Press.

73.

Stephan

(2004) State Prison Expenditures, 2001. Washington, DC: Bureau of Justice Statistics, NCJ 202949.

74.

Sykes

(2007 [1958]) The Society of Captives: A Study of a Maximum Security Prison. Princeton, NJ: Princeton University Press.

75.

Tonry

(2004) Thinking about Crime: Sense and Sensibility in American Penal Culture. New York: Oxford University Press.

76.

Vaughn

(2008) Bourdieu and organizations: The empirical challeng? Theoretical Sociology 37: 65–81.

77.

Vuolo

Kruttschnitt

(2008) Prisoners’ adjustment, correctional officers, and context: The foreground and background of punishment in late modernit? Law & Society Review 42(2): 307–335.

78.

Wacquant

(2004) Body & Soul: Notebooks of an Apprentice Boxer. New York: Oxford University Press.

79.

Weisberg

Petersilia

(2010) The dangers of Pyrrhic victories against mass incarceratio? Daedelus 139(3): 124–136.

80.

Zimring

Hawkins

(1991) The Scale of Imprisonment. Chicago, IL: University of Chicago Press.

81.

Zimring

Hawkins

Kamin

(2001) Punishment and Democracy: Three Strikes and You’re Out in California. New York: Oxford University Press.