REHABILITATING CRIMINAL SELVES: Gendered Strategies in Community Corrections

Abstract

As the community corrections system has moved away from a focus on rehabilitation, it has been suggested that criminal offenders are no longer understood psychologically, but rather as rational actors for whom criminality is a choice. Rehabilitative efforts thus aim to guide these choices. Utilizing mixed methodology that draws on observational, interview, and case note data collected within the probation/parole system of a western U.S. state, I suggest that both officers' conceptualizations of the criminal self and the rehabilitative strategies they use are gendered. I find that officers view the male criminal self as flawed or underdeveloped and the female as permeable and amorphous, that is, lacking firm boundaries. In response to these constructions, officers aim to rehabilitate men largely by encouraging economic roles and responsibilities, while for women, rehabilitation aims to solidify boundaries: discouraging relationship formation and containing emotions. The differences identified point to ways in which gendered concepts of the criminal self contribute to gender disparities in contemporary supervision.

The increasingly punitive nature of the American penal system and its rapid expansion have been well documented (Chesney-Lind and Pasko 2004; Travis 2005; Western 2006). Though a less frequent topic of research, these changes have also affected community supervision, where one of every 48 American adults is currently on probation or parole (Glaze and Bonczar 2011).¹ As resources dwindle and legislators strive to appear “tough on crime,” community supervision has moved away from rehabilitation as traditionally conceived and towards a strategy of containment, risk management, and efforts to enlist offenders in their own regulation (Feeley and Simon 1992; Garland 2001). One important strand of this strategy aims to encourage offenders to accept personal responsibility for their criminality and processes of reform (Lynch 2000; Rose 2000). Underlying this approach is a conception of the criminal offender as a self-interested, rational actor for whom criminality is a choice (Garland 1997; Lynch 2008; O'Malley 1996).

While this line of work pays little attention to the gendered cast of these conceptions and the processes they shape, a growing literature has documented the ways in which, for women, both in the prison (Hannah-Moffat 2001; McCorkel 2003, 2004) and community (Haney 2010; McKim 2008), gender shapes the characterization of a criminal self and the nature of punishment. This work rests on an implicit comparison with the male criminal subject although, particularly within the context of community supervision, little empirical research has addressed men as gendered criminal subjects.

In this article, I explore how contemporary supervision is structured in response to offenders' gender. I uncover the gendered beliefs and strategies officers rely on as they aim to rehabilitate criminal men and women, and in so doing shape their (gendered) subjectivity, and extend the research on offender management, suggesting ways in which distinct concepts of the criminal self contribute to gender disparities in treatment.

This study relies on and extends the literature classified by the Foucauldian concept of governmentality. Governmentality literature aims to deconstruct and make sense of contemporary practices, such as those exhibited by agencies of social control, revealing how the exercise of power depends on “specific ways of thinking (rationalities) and specific ways of acting (technologies) as well as on specific ways of `subjectifying' individuals” (Garland 1997, 174). Subjectivity can be understood to reflect the ways in which offenders are made sense of, acted on, and taught to act on themselves by agents of social control (Garland 1997; Rose 2000). Casting an individual as a particular criminal subject shapes the crime control strategies relied on and, more broadly, the nature of punishment. The governmentality approach thus provides a useful means of theorizing the linkages between everyday practice and systems of governance (Weir, O'Malley, and Shearling 1997).

Community corrections serves as a particularly compelling site from which to study these processes for several reasons. In contrast to the prison, women represent a substantial share of the supervised, comprising 24 percent of probationers and 12 percent of parolees (Glaze and Bonczar 2011). Unlike the reduction in discretion marking sentencing and release decisions, officers maintain substantial discretion to tailor their response to clients, and these responses can be both deeply personal and highly significant, ranging from rights limitations to possible reincarceration (National Research Council of the National Academies 2007). Finally, unique among correctional institutions, community corrections officers supervise men and women within the same organizational context, allowing for direct comparison of the treatment male and female offenders receive.

Using mixed methodology that draws on observational, interview, and case note data collected within the probation/parole systems of a western U.S. state, I suggest that both officers' conceptualizations of the criminal self and the rehabilitative strategies they employ are gendered. I find that officers view the male self as flawed or underdeveloped and the female as permeable and amorphous, that is, lacking firm boundaries by which to contain emotions and function independently of others. In response to these constructions, officers' rehabilitative efforts with men emphasize economic roles and responsibilities, while for women, officers aim to solidify boundaries: encouraging emotional containment and discouraging relationship formation. Qualitative data illustrate officers' gendered beliefs and supervision strategies, while quantitative data confirm that these accounts are representative.

I first discuss historical shifts in strategies of crime control, which is followed by a review of the literature on women and corrections, pointing to ways in which criminal subjectivity is gendered. Next, I identify differences in officers' beliefs about the basis of criminality for men and women, and then link these beliefs to the supervision strategies they employ. Quantitative analyses of officers' case notes are woven throughout. I conclude by summarizing the primary findings and suggesting possible implications.

GENDER AND THE RESPONSIBLE SELF

Social understandings about crime, and the policies and practices used in attempts to control crime and reform offenders, are not static but transform over time. Underlying these transformations are changing beliefs about the causes of crime and “solutions” to criminality, beliefs that reflect particular understandings of the nature of the self of those deemed criminal (Garland 1997). Response to crime must then be understood as situated within a particular social context. Foucault (1977) detailed how, across the eighteenth century, punishment was transformed from a public spectacle, designed to emphasize the power of the sovereign, to a private, ideologically oriented endeavor whose primary aim was the alteration of individual subjectivity. Techniques of knowledge and discipline were used in efforts to produce submissive, useful subjects. The late twentieth century saw another change in correctional philosophy, characterized as a decline of the rehabilitative ideal (Allen 1981) and moving toward a more punitive correctional philosophy (Garland 2001). This shift was partially premised on the notion that individually focused rehabilitative strategies simply did not work (Martinson 1974). The dramatic expansion of the criminally supervised population can be directly tied to policy changes that were responsive to this reorientation (Blumstein and Beck 1999; Mauer 1999).

In line with these broader shifts were alterations that occurred within the community correctional system. Feeley and Simon (1992) and Simon (1993) suggest that individually tailored rehabilitation has largely been supplanted by strategies of incapacitation that aim to manage the risks posed by the most dangerous offenders. Under this model, the individuality of the criminal subject, his motivation and potential for reform, are no longer explicit concerns. In contrast, Lynch (2000) maintains that while rehabilitation remains a prominent rhetorical aim of parole, the institutional and financial investment needed for its realization are lacking. Rather, officers emphasize offenders' accountability for criminal actions and reform, a process of “responsibilization” in which experts' interventions entice the subjects of governance into self-regulation, rather than through the use of force (Garland 1997; Lynch 2000; Rose 2000). The criminal self at whom such interventions are aimed is understood to be an autonomous, responsible choice-maker (Fox 1999; Garland 2001; O'Malley 1996).

While this body of work takes little notice of if (or how) gender shapes techniques designed to “responsibilize” the rational subject, a related literature, centered around the disposition of criminal cases, suggests that it may. This work argues that court officials employ mental images of offender or criminal “types” that both guide their perceptions of the individuals they encounter and shape their treatment of them (Emerson 1969; Sudnow 1965). Bridges and Steen (1998), for instance, reveal differences in probation officers' attributions about the causes of crime between white and minority youth, and link these differential attributions to disparities in the assessments of reoffense risk and sentence recommendations. Gender, as a primary system of difference, is likely to play a similar role in shaping perceptions and treatment.

Indeed, recent scholarship has begun to uncover the ways in which responsibilizing strategies and beliefs about criminal subjectivity are gendered. In an ethnographic study of a mandated, community-based drug treatment program for female offenders, McKim (2008) finds that staff conceive of criminal women as lacking an adequate self, in response, they emphasize therapeutic forms of governance such as monitoring of self and others, therapeutic and emotional disclosure, and working through disordered emotions. McCorkel (2003), in her study of a drug treatment program, instituted in a women's prison, identifies a therapeutic discourse locating the source of women's criminality in deviant, non-“habilitated” selves, which she links with supervision techniques of embodied surveillance. In the Canadian context, scholars find that the female criminal subject is understood psychologically rather than structurally, as “fractured” (Turnbull and Hannah-Moffat 2009) and “disorderly and disordered” (Pollack 2005). Female parolees are then governed through conditions that encourage self-regulation and normativity (Turnbull and Hannah-Moffat 2009), while female prisoners are governed through psychiatric labels and treatments (Pollack 2005). In one Canadian “woman-centered” prison, Hannah-Moffat (2001) finds that women's self-control and self-esteem are the primary targets of reform. Finally, Haney (2010) describes how correctional workers in one alternative-to-incarceration facility for women rely on a therapeutic discourse focused on righting women's cognitive distortions and regulating appropriate desire.

While this literature has revealed a great deal about the project of correcting women, gaps nonetheless remain. First, although not comparative in design, a comparison with the male criminal subject is often assumed. Absent research detailing the gendered construction and treatment of the male subject, it may seem that criminal men are conceptualized rather simply as that which women are not: psychologically “whole,” autonomous individuals, lacking cognitive or emotional deficits, and rich in self-understanding and esteem. Second, while there is general consensus in the characterization of female offenders as psychologically and emotionally flawed, the nature of this flaw and its connection to “responsibilizing” practice remains elusive: What does it mean to say that a self is fragmented or disordered, and how is supervision responsive to such beliefs? Direct comparison of officers' beliefs about and treatment of male and female offenders provides analytical leverage for understanding the gendered nature of contemporary criminal justice practice, and the ways in which responsibilization takes on gendered meaning.

DATA AND METHODS

The analyses that follow draw on interviews with community correctional officers, observations of meetings between officers and their clients, and quantitative analyses of a sample of officers' case notes, all collected within a western U.S. state. In “Western State,” while community corrections is operated semiautonomously at the county level, offenders' administrative records are maintained by the state's Department of Corrections (WDOC). I thus pursued access to the quantitative data and field sites independently. After reaching out to WDOC's research division, I was granted access to administrative records containing basic demographic, criminal, and supervision history information for all offenders on probation and parole supervision within the state. From these data, I drew a sample of offenders whose case notes I would review.² The interview and observational data were drawn from two county community corrections systems within Western State. I initially contacted several counties across the state by letter, explaining the goals of my research and inquiring as to their interest in participating. Within days I received responses from the counties I refer to as Greendale and Riverside.

While Greendale and Riverside are similar in some respects—they border one another and are the most populous counties in the state—because each county system is administered separately, they also differ. While I was in the field, both county systems were facing fiscal pressure resulting from the recession, but Riverdale was hit harder. As a result, officers were preparing for changes, including supervising a greater proportion of high-risk offenders, and the transfer of low-seniority officers to different offices. These changes had stirred up animosity between management and staff. The other important difference was Greendale's reputation for lenience, an approach characterized as “hug a thug” by one Riverdale officer.³ While this seemed to have been true historically (and continued to be reflected in the fact that officers in Riverdale could choose to be armed while those in Greendale could not), interviews with officers in both counties suggested that the differences were not nearly as stark as presumed. Rather, it seemed that Riverdale, in line with recent trends in correctional practice, was following Greendale's lead as it moved away from a “law and order” orientation toward evidence-based practices, including motivational interviewing and graduated sanctions.

I interviewed 26 officers and staff in the sole county office in Greendale and 24 officers and staff across the five offices in Riverside. Interviews averaged between 45 minutes and one hour and addressed a variety of topics, from beliefs about the causes of crime, to the personal mission of the job, to what gender-specific treatment meant in practice. Questions were designed to gain an understanding of how officers conceptualized their work, their role, and their relationship with offenders.

In Greendale, I also observed more than 50 meetings between community corrections officers and their clients, after obtaining consent from the officers. These meetings averaged between 15 and 45 minutes. In each case, I approached offenders in the lobby as they waited for their appointment, and sought their consent to observe their meeting; no offender declined. As a whole, offenders seemed relatively indifferent to my presence, although some offenders inquired about the project or my field of study. I attribute offenders' apparent lack of concern with my presence in part to the fact that I was a student, relatively young, and a woman, which may have made me appear less threatening. However, it is also important to acknowledge that offenders simply have little expectation of privacy in this setting. Officers discuss the details of offenders' crimes and lives with doors wide open and stand in the bathroom while offenders give urine samples. In this context, desire for privacy is often interpreted as something to hide. To the extent that I benefited from this dynamic I became complicit in it, a tension that must be balanced against the importance of revealing the inner workings of this system.

In both counties, how offenders are managed is important to mention. First, officers' caseloads are composed of both parolees and probationers, with relatively few distinctions made between the two populations. Second, officers' caseloads are composed of offenders sharing a particular identity (women, Spanish speaking) or category of crime (generic/drug offender, domestic violence, mentally ill).⁴ This allows officers to become familiar with the resources and regulations pertinent to their caseload. Because nearly all officers cycle through diverse caseloads during their tenure, they are able to speak to differences across populations. Officers also often “matched” the identity of their caseload in at least one respect; for instance, women were more likely to supervise women. Such matching reflected the fact that offenders were assumed to be more comfortable with those who shared their identity grouping.⁵ I use these data to illuminate officers' gendered beliefs about criminal offenders and offenders' needs, and the gendered supervision strategies they employ in response.

Turning to the quantitative data, from the administrative records I had received, I drew a random sample, stratified by gender, of offenders with low-risk scores, no criminal history, no recorded mental health problems, and a (low-level) sentence of either Drug I or Theft I.⁶ Sampling only offenders in these categories controls for gender differences along these domains. Because of the sensitive nature of these data, I was allowed access to them for just one week within a WDOC office and was restricted to recording handwritten notes. I was thus able to access the case notes of only 100 offenders. For each offender, I began the case note review at the beginning of their supervision period and reviewed notes across their first full year on supervision. I tallied the number of notes recorded on key topics over the course of this year as well as the total number of distinct text entries. These data allow me to examine how frequently officers mention particular categories of comments in men's versus women's case notes, one measure of the relative importance they place on these domains.⁷

I examine these gender differences using regression models in which the independent variables are male officer and male offender. As dependent variables, I use the number of times the topic of interest occurs in the offender's case notes. Because each dependent variable is a count of the number of times a specific topic is mentioned, I use negative binomial models. Control variables include offenders' age, race, and urban county. Recall that I have implicitly controlled for differences in criminal history, crime, risk score, and mental health by means of my sampling technique. I use clustered standard errors to correct for the fact that multiple offenders have the same officer. Because the total number of entries varies among offenders, dependent on the number of interactions (meetings or phone calls), I adjust for this variability by specifying offenders' differential “exposure” to text entries in the model. For ease of interpretation, I have reported exponentiated results.

In the analyses that follow I ask whether, net of officer gender, there is a statistically significant difference in the number of comments recorded in the case notes of men and women regarding employment, crime/criminality, fee payment, sanctioning, timeliness to meetings, and romantic relationships. Revealing a gender difference in this formal record establishes that officers maintain distinct foci in their work with men and women and hints at the ways in which rehabilitation is differentially understood. I embed these quantitative findings within the qualitative analysis of interviews and observations in order to reveal that the gendered processes I discuss are typical representations of offender management.

RESULTS

Becoming Criminal

Officers' work with offenders is structured in response to their beliefs about the origin of criminal behaviors and reasons for their continuity, discourses pointing to distinct ways in which the criminal self is conceived for men and women. Asked to explain why offenders initially became criminally engaged, officers' responses, though diverse, clustered around a few key ideas. Officers commonly expressed the belief that offenders had been exposed to criminality and substance abuse through their family of origin. Linked with this explanation was the recognition that offenders' social positioning contributed to their criminality. Officers noted that poverty, oppression, homelessness, and feelings of hopelessness, for instance, contributed to offenders' criminal “choices.”

Yet when asked to identify pathways by which women specifically became criminally engaged, or when officers who worked primarily with female offenders were queried, the answers differed. In these cases, the most frequently cited explanation was that of a relationship to a drug abusing and/or criminally engaged man. The next most cited explanation was social context factors, such as few job skills, poverty, and lack of child care, which left women with few resources or options. Finally, female offenders' histories of abuse and victimization were also frequently mentioned, whether in terms of explaining their drug abuse (a form of “self-medication”), low self-esteem, or relationships with abusive partners. Thus, while men's early life context was more frequently relied on to explain criminality, women's current relationships and context were the focus.

This distinction also underlies officers' conceptions of why men and women remain criminally engaged. Having learned criminal behaviors in their youth, men's thought processes and self-concepts remain immature, narcissistic, and antisocial in adulthood. Ryan, a white officer supervising offenders with mental health problems, explains that men make criminal choices rather than take on adult responsibility:

Some of it is immaturity … people get stuck emotionally where they're at when they're 14, 15…. It's easier to be a failure for a lot of these guys too, to just … couch surf and not have responsibility and not have people relying on you because when they start feeling that sense of responsibility, that's uncomfortable for them.

The criminal self at the heart of such a description is painted as under-developed or flawed.

In contrast, a woman's criminality results from an inability to contain her emotions and a tendency to be drawn into relationships with others who wield undue influence over her decisions and choices. Carol, a white officer supervising domestic violence offenders, described this as “getting involved with men and selling out on their core self to please a man, do what the criminal behavior he's involved in.” Underlying these descriptions is a concept of an insufficiently bounded, permeable self. Next, I discuss how these concepts shape officers' reformative efforts.

Supervision Meetings: A Site for Gendering the Criminal Subject

In supervision meetings with both men and women, officers discussed alcohol and drug treatment, housing, and access to transportation, and encouraged conventional behaviors. Despite these similarities, gender differences were substantial, reflecting differences in officers' primary goals for men and women. With men, officers focused on the formal rules of supervision, including abstaining from crime, obtaining employment, and paying fees, suggesting that officers view a rehabilitated man as having exchanged criminality for a role in the marketplace. With women, officers focused on social networks (particularly romantic relationships) and emotions, suggesting that officers view a rehabilitated woman as socially independent and emotionally contained.

Constructing Men

Within the context of the supervisory meeting, officers worked with offenders to identify conventional long-term goals and encouraged men to the take initial steps along a noncriminal path. Officers used a technique known as motivational interviewing that is designed to lower client resistance and strengthen commitment to personal change (Miller and Rollnick 2002). Bill, a white officer supervising a generic caseload, explains how he uses the technique:

I try and find out … what they want out of life and from that ask them how they expect to get it … whether it's having a good job or having a car or having a house or having a family…. I work with them to figure out how they can attain those things in a legal way so they don't have to worry about losing it…. If it's one thing at a time, whether it's the shirt on their back … and work up from there.

For Bill, work with men entails reframing both the goals the offender is striving toward and the strategies used to meet these goals. The officer encourages men to conceive of an alternative reality, in which criminal choices are passed over in favor of conventional adult responsibilities.

For officers, conventional responsibility seemed largely to entail men's assumption of an economic role. Indeed, discussion of employment (job search, work hours, wages) was often the central focus of meetings with men, regardless of whether the offender voiced other significant needs or faced substantial barriers to obtaining employment. This was exemplified in one meeting I observed between one young offender and his white supervising officer, Debra.

The offender had recently been released from prison and was homeless. As the meeting began, he informed the officer that without family or friends to offer him a place to stay, he had no place to go. Rather than responding to this question, the officer asked the offender what he had done to seek out a job and whether he had a resume, and informed him of a help wanted sign she had seen posted nearby. The client and officer then discussed how he should best present the crime for which he was incarcerated, a serious beating of another teenager, when future employers inquired about it. Only then did she inform him of a counselor he could speak with to get help finding both employment and housing. At the end of the meeting, the officer encouraged the offender, noting that he was smart and presented well (not like her other clients), and that she was sure he would be able to get a job. This client's homelessness and adjustment from prison to the community receded into the background in comparison to the primary focus: employment.

Another example of this approach can be seen in an excerpt from my field notes, recorded following the observation of a meeting between Jerry, a black officer supervising a generic caseload, and his client:

Young Mexican American man. Looks down, says little. Reports that he recently got a job in a local factory…. Correctional Officer (CO) is very pleased, tells him so repeatedly. How did he get that job? Are they still hiring? He just kept calling. He works five days a week but will soon be working seven days a week at minimum wage. Has he finished up his community service hours? He … [is] not quite done. CO states that he must get those hours done before [he] starts working seven days a week. Agrees to do so. CO asks how things are going at home? Much better now that he has a job. [CO informs me after the meeting that his crime was meth related.]

The offender's employment status dominates this conversation, almost as if employment in a conventional job has rectified his criminal status. Of secondary importance, the officer inquires about his community service, in recognition that full-time employment may clash with other supervision conditions. Finally, the offender's family relationships are inquired after (in a way that implies that something had been problematic in the past), but are clearly not of central importance to the conversation.

While both male and female officers emphasized employment when talking with men, this was particularly true of male officers. Indeed, quantitative analysis of the case notes reveals that male officers are significantly more likely to mention employment and timeliness than are female officers. Results are displayed in Table 1, models 2 and 4. As shown in model 2, male officers' case notes have 1.43 times as many comments referencing employment and 1.46 times as many comments referencing timeliness as do the case notes recorded by female officers. The focus on employment may reflect the central role that work tends to play in men's self-concepts more generally (Lamont 2000; Thompson and Walker 1989). The focus on timeliness may reflect a similar belief about the importance of employment, as showing up on time is a prerequisite for maintaining a job.⁸

TABLE 1.

Negative Binomial Regressions of Counts of Case Notes: Crime, Employment, Fee Payment, Timeliness, Sanction, and Romantic Relationships

	Dependent Variables
Independent Variables	Crime	Employment	Fee Payment	Timeliness	Sanction	Romantic Relationships
	(1)	(2)	(3)	(4)	(5)	(6)
Male offender	2.02***	1.28*	1.44**	1.24	0.90	0.66+
Male officer	0.7	1.43**	1.4	1.46†	0.92	1.57*
Age	1.00	1.00	1.02**	1.04***	1.01	1.01
Race	1.05	0.75*	0.96	1.03	1.10	0.67
Urban county	2.62***	0.87	0.68	2.07***	1.61*	1.18
Number of offenders	100	100	100	100	100	100

NOTE: Dependent variables indicate the number of times a topic was mentioned in offenders' case notes. Results are reported in an exponentiated format. Exposure is total number of text entries recorded during the year. Control variables include age, a dummy variable for race where white is the reference category and a dummy variable for urban/rural where urban is the reference category. Sampling method implicitly controls for risk score, crime, criminal history, and mental health.

^†p < .10;

^*p < .05;

^**p < .01;

^***p < .001.

Such an interpretation was suggested by my interview with Jerry, when he explained that as an African American man who experienced racism both on and off the job, he identified with the frustrations and hopelessness his clients often expressed. Nevertheless, he stuck with his job, deriving a sense of purpose and masculine identity from it: “If I didn't feel like I made a difference with my clients I wouldn't be here. There's been days when I … wanted to bag it, but I'm a fighter.” He worked to instill a similar positive identification in his clients as a substitute for the hopelessness that he believed fueled criminality:

Teaching guys how to be responsible fathers, making better choices, understanding that, okay, we live in a society that is not really set up for us to win, but how do you survive, how do you make the best of it? You do that by getting an education … getting into a job at the entry level but working your way up through the system. Being a responsible person, being a law-abiding person, drugs aren't going to get you there.

He encouraged offenders to persevere, as he had.

Closely linked with the focus on employment was officers' attention to fee payment. In meetings, officers often began the conversation by asking whether men were up-to-date on their payment and stressed the importance of paying fees. Barriers to payment were frequently discounted. As Carol noted, “They can find money for what's important to them … maybe 75 percent of our cases, they're smokers, if we made it to their homes they'd have big-screen TVs and they all have cell phones.” Some officers stressed the importance of paying fees whether the client was employed or not. I observed one officer counseling an undocumented worker who was having trouble finding a job to pick up cans by the side of the road in order to pay his fees. I observed another officer mention that, given the offender's recent payment, he would rescind the warrant he had issued. Although the offender had not obtained a job, the officer did not inquire as to where he had obtained this money. It was the fulfillment of economic obligations that mattered to many officers, and not the social relationships or circumstances that made this possible.

This focus on economic responsibilities was also evident in my examination of the case notes, where I also found greater attention to men's crimes. Table 1 shows the main results. Displayed in models 1 through 3, I find that men's case notes showed more frequent discussions of crime, employment, and fee payment than did the case notes of female offenders. Specifically, being male increases the number of comments about crime by a factor of 2, comments about employment increase by 1.28 times, and comments referencing fee payment increase by 1.44 times. Thus, officers' work with men seems more focused on criminal behaviors and economic roles and responsibilities than is their work with women. However, gender differences are not evident in two related areas: discussion of timeliness of arrival to supervision appointments and comments about sanctions imposed (models 4 and 5). Regarding timeliness, this lack of difference may reflect officers' views of the supervision meeting as a crucial aspect of their work with both male and female offenders. The equal number of comments about sanctions is more understandable when informed by my observations, which suggest that differences in sanctioning between men and women are not those of frequency, but of type of sanction imposed.

The emphasis on normative behaviors and responsibilities was also reflected in officers' style of interaction with men. Officers aimed to forge a respectful, officious rapport, as Scott, a white supervisor of a generic caseload, explained: “The rapport is based on just a basic level of trust and respect, that I'm gonna treat you as a human being, and I expect to be treated the same. And as long as we don't violate that, then we have somewhere to work from.” If the offender trusted the officer, the officer could then “start to hold up the mirror a little bit and maybe create a little cognitive dissonance,” opening up the possibility for cognitive shifts away from criminal thought patterns and identities, and toward conventional goals. This approach also tells the offender that the officer considers him a mature adult, albeit one whose behavior has gone astray.

In contrast to work with women, rapport was not built on engagement with men's emotions or traumas. Rather, officers described their quite limited attempts to work through traumatic histories and engage in therapeutic discourse. Men would generally be allowed to unload feelings of anger and frustration for a short time, but were quickly encouraged to move on to other issues. Erin, a white supervisor of a domestic violence caseload, explains:

I let them … talk about how frustrated they are or why they shouldn't be here: “I've seen worse things, I mean, there's other guys that do worse than I do.” … Let them get that all out … but if it continues to be really unuseful … I can say, “Okay, you're here now, let's focus on what you need to do here. Can't change that.”

In other words, listening was intended as a starting point to forge a positive working relationship with men, but generally not as a mini-therapy session. Deborah, a white officer supervising a generic caseload, noted that she was careful not to delve too deeply into clients' personal lives:

I want to make sure that we address the problems that are important to them, but … whatever can of worms you open in a session you have to be able to close, you can't let them walk out this door raw…. I try to keep it a little more superficial because … our job is not as counselors.

I was able to observe several meetings between this officer and her male clients and confirmed that although some men seemed to be seeking a connection with her, she largely stuck to the script, focusing on the formal conditions of supervision. A field note excerpt reveals her approach:

He comes in looking for love. He's exhausted, working two jobs, a swing shift and regular day shift. Short blond hair, sort of a frat-boy look … slumped in his chair. Towards the end of the meeting … he gives her a speech about how different things are this time, he didn't really see the light before, he was still blaming other people, but now he realizes it is all him…. She doesn't really respond…. She is quite reserved, telling him he has to go to aftercare, and giving him the list with the options, etc.

In spite of his attempts to elicit sympathy from her and forge a connection, she maintained a business-like style that left little space for talk of emotions or relationships.

As men's criminality resulted not from emotional or relational problems, but rather developmental deficits, officers' rehabilitative efforts encouraged men to assume economic roles and responsibilities, markers of a conventional, masculine self.

Constructing Women

Within supervisory meetings, officers aimed to harness women's emotional disorder, build self-esteem, and monitor their relationships. Officers addressed these goals by probing, often deeply, into women's emotional lives and attachments. While officers also inquired after women's progress in meeting the formal goals of supervision, it was these socioemotional aspects of women's lives that seemed to be at the heart of the supervisory relationship. This focus on emotions was revealed in the following meeting I observed between Leslie, a white officer who supervised primarily women, and a young woman on the cusp of completing supervision:

Blue-tipped nails, tapping on the table. CO asks her if she's excited, how she's feeling. She says she's really excited, but really nervous too. What are you nervous about? Everything. What are you going to do about those feelings? Who are you going to talk to about things? She says she has [a friend], going to group, there are girls in the house, but [her eyes fill with tears] what makes her upset is that people she thought were her friends turned out not to be, she didn't know that could happen even in recovery. CO asks if she has other friends and she says that she does. Asks what is her plan? How much longer is she in school?

As in many meetings with women, discussion of women's housing, employment, and long-term plan is only part of the conversation, while the emotions that are understood to drive women's criminality are central.⁹

Female offenders' perceived self-esteem deficits were also a key area of concern for officers. Officers working with women saw the meeting as an opportunity to address and begin to work through these emotional needs. Rita, an African American officer supervising a caseload of women, explains:

And I have to tell my clients every day, you have to work on your self-esteem. Every day you have to work on your self-worth, and when you start feeling that low self-esteem and low self-worth, go in the backyard, dig it outta the dumpster, put it back on, and keep pushing.

One way, Rita explained, that she helped women build self-esteem was by discouraging her clients from using language like “babymama” to describe themselves; another was to encourage her clients to wear less revealing clothing. In contrast, officers working with men worried that men's self-esteem was already too high, that their criminality resulted from a narcissistic personality. Because narcissism is one feature of “criminal thinking,” these officers believed that building men's self-esteem could then prove counterproductive.

For some officers, women's emotional lives were seen as so central that the meeting took on the character of a therapy session. As Gloria, a Latina officer supervising a generic caseload, explains:

Finally, I built a relationship with her enough that it's, like, “Come on … I know your pattern. You're dating these men because of your father.” Her father abandoned her when she was a young child, and she's always had this wanting to find a man.

Discussion of women's problematic personal patterns becomes an integral part of court-ordered supervision.

Officers' intense focus on women's emotional lives was also evidenced in the tone of the supervisory meeting. While officers working with men aimed for a respectful but emotionally contained relationship, officers working with women believed that women required empathetic, emotionally responsive treatment in order to engage in processes of personal change. Melinda, a white officer supervising a caseload of women, explains:

Women are much more into sharing, and they wanna know they can trust you…. They love to talk, so it's a lot of listening and relationship building. They wanna know that you know their kids' names; they need that connection. Men, not so much.

Without such a reception, these psychologically troubled women would continue to “self-medicate” with drugs or unhealthy, “addictive” relationships.

Officers forged this emotional connection by interacting informally with offenders, engaging in a chatty style that minimized social distance, and remarking on successes at work or school and changes to appearance, like a new haircut or color. One particularly engaged female officer on a women's team spent time on the weekends and evenings with her clients, texting, taking them out to dinner, and even overseeing a family visit at the zoo. Another officer on a women's team led a knitting class for her clients once a week. Officers felt that this was an important aspect of working with women, in part, because their clients had so few positive relationships in their lives. The criminal activity that led to supervision thus availed women of one person (their corrections officer) whom they could rely on. Lisa, a white officer supervising a women's team, explains:

I've had people use me as a reference on jobs, as the emergency contact number on applications…. Sometimes there's no one else left, so I'm like this authoritative, responsible person they actually have a relationship with that they can refer other people to.

Jenny, a white officer supervising a women's team, concurs: “I don't mind them needing a hug at the end. I don't mind being the one person they call if I'm the one person they have right now.” Of course, like male officers' focus on employment, female officers' emotional engagement with women may not be just a reaction to offenders' gender but also a way for officers to do gender themselves, aligning supervisory style with normative conceptions of femininity.¹⁰ Building a relationship with women was also seen as important because of the widespread assumption that the other relationships in women's lives, particularly those with men, were a fundamental cause of women's criminality.

Indeed, only one domain was noted more frequently in female offenders' case notes: romantic relationships (model 6), which show a marginally significant difference. Specifically, being male decreases the number of mentions of romantic relationships by a factor of 0.66. This suggests that officers' work with women focuses less on women's own behaviors, and more with the potentially negative influence of those with whom a woman becomes involved.

In the relationship model, officer gender also has a significant effect, with male officers more likely to mention relationships. Though counterintuitive, this may be explained by differences in the gender pairing of officer and offender. Although small sample size prohibits testing for the significance of cross-gender pairings, the raw numbers suggest that, while both male and female officers frequently mentioned female offenders' relationships, female officers paired with men rarely mentioned relationships and male officers paired with men fell somewhere in the middle. Thus, as a whole, male officers discussed relationships with greater frequency. Because male offenders occasionally propositioned female officers, it could be that female officers avoid discussion of relationships in order to maintain a professional distance.

In contrast, romantic relationships were a big part of the discussion with female offenders, and inquiring about boyfriends and partners was routine. Officers tried to discover whether women were romantically involved with male friends or associates they mentioned in passing, and commonly asked whether men they were in relationships with were also on supervision or had a substance abuse problem. Lisa explained that officers generally assumed that the men criminal women were involved with were problematic: “I mean, what we call it here is that their picker is broken … that part of the brain that picks the man you're attracted to, that picker is broken and you only pick bad men, either abusive or addicted or all of the above.” Officers focused on the men in women's lives because of the widely held belief that men had led women into criminality initially, and were likely to do so again. Thus, involvement in relationships posed a threat to women's rehabilitation.

In one meeting, Lisa inquired initially about the offender's housing and job search. Yet, both of these questions led to further inquiries about the nature of relationships. On hearing that the offender planned to reside with her father, the officer asked about the history of the relationship. She was interested not just that the offender had secured housing, but in the emotional implications this residence might have. Similarly, when the offender noted that she had obtained a job at a pizza parlor through a friend, the following conversation ensued:

CO [Correctional Officer] asks, who is this friend? An old friend [offender replies]. She worked there [at the parlor] in the past. He saw her go through the whole bad cycle of doing well and then falling apart…. CO asks if he is interested in a relationship with her—“He was!”—but the client explains that she straightened him out, she doesn't want a relationship with him ever, and not with anybody right now.

Later, the officer brings up relationships again:

CO asks if she is looking for a boyfriend right now. “No way. I'm not looking for that at all, not for at least a year.” CO asks why. “I just have really bad taste in boyfriends…. ” CO highlights that not getting involved in relationships is a good choice right now.

Lisa makes it clear to the offender that staying out of relationships is important to her success on supervision. Notably, this offender had already secured a job and housing and was attending Narcotics Anonymous meetings twice weekly, achievements that, for men, often meant there was little else to discuss.

In the case notes not just the frequency, but how relationships were discussed differed as well. For male offenders, relationships seemed to signal progress towards conventional goals: “O [offender] moved in with his GF, working at McDonalds, wants to go to college! Overall I am very impressed w/O's progress, by all appearances doing very well.” Officers generally assumed that partnership for men was positive. In contrast, notes about women's relationships were largely negative. While the resources boyfriends offered were mentioned, boyfriends' drug use, supervision histories, and roles in encouraging the female offender's own criminality were also frequently noted:

Told O that she could not live with a male who is also on supervision due to her relapses in the past. I reviewed O's past relapses with her, how she gets lost in relationships w/males who have criminal history and then she relapses.

PO told O that any contact with husband [who continues to sell marijuana] would result in loss of custody of child.

Admits she is still with Jake, states he drinks but no law enforcement involvement.

In these quotes, it is not imminent danger posed by the male partner that the officer seems to be concerned with; rather, involvement in a relationship itself seems to run counter to officers' conceptions of proper female rehabilitation.

Because women's criminality is believed to result from insufficient boundaries, officers' rehabilitative efforts center on policing these boundaries, encouraging emotional independence, and discouraging romantic relationships.

CONCLUSION

As the carceral net has expanded, drawing an astonishing number of men and women under criminal justice supervision, often for years at a time, it has become increasingly important to understand how such governance is both shaped by and constitutive of particular notions of men and women as criminal subjects. Building on past literature, which finds that criminal women are viewed as emotionally and psychologically damaged, I suggest that the criminal woman's self is understood to lack firm boundaries, what I characterize as a permeable or amorphous self. Women's crime is understood to occur in reaction to her current social context: when she is overcome by emotions and influenced by relationships with criminal others. While the particularly gendered nature of men's subjectivity has received less attention, I diverge from that which suggests that men's criminality is a rational choice made by a whole self. Rather, I find that officers view the criminal male self as flawed and underdeveloped and understand this self to be emergent from patterns learned in childhood. As an adult, these criminal thought processes and self-concepts lead men to criminal choices. In addition to describing the root causes of criminality, these gendered typologies contribute to distinct goals in work with men and women.

While it has been established that contemporary supervision aims primarily to govern offenders by guiding their choices (choices for which they are seen as ultimately responsible), this article suggests ways in which such responsibilizing practices are gendered. I find that while officers' efforts to normalize men aim to encourage the assumption of economic roles and responsibilities, viewed as the counterpoint to criminal engagement, efforts to normalize women aim to solidify boundaries: discouraging romantic engagements and containing women's emotions. Thus, beliefs about the context in which criminal choices are made, and the rehabilitative response needed to encourage conventional choices, are distinctly gendered.

While contributing to an understanding of how offenders' gender structures supervision goals and practices, the data presented here also raise additional questions, highlighting the need for future research. First, hinted at throughout this article and suggested by related literature (Watkins-Hayes 2009), officers draw on their own identities and experiences in their work with clients. Future work, ideally with a larger sample of cross-gender supervision pairs, should elaborate the role of officer gender in structuring supervision. Second, future work should link the disparities I have identified to gender differences in outcomes, such as recidivism and relapse. And third, future work should attempt to replicate these findings in offices managed by alternative discursive and policy regimes.

I suggest that future work continue to pursue the topic of gendered strategies within community supervision because of the significant implications such strategies likely have for those supervised. While the collateral consequences of imprisonment have been well documented, less is known about the consequences of community supervision. This study suggests that, for women, whose needs are defined largely in relational and emotional terms, community supervision extends far beyond the presumed objectives of preventing crime and protecting public safety to bring officers' powers of social control to bear on deeply personal aspects of women's lives. While there may be truth to officers' beliefs about women, for instance in the role relationships play in women's criminality (Daly 1994; Richie 1996), it remains important to consider whether such regulation should be a central goal of criminal justice supervision. Though it may be that women are reliant on problematic romantic partners and lacking in self-esteem, if women have no source of income and no place to stay (other than with this partner), self-esteem building alone will not translate into real independence for women. In contrast, for men, whose needs are defined largely in economic terms, officers often fail to consider the barriers men may face in fulfilling supervision conditions, as well as how men's social and emotional lives influence their success or failure. While men's lives face less scrutiny, ignoring the very real obstacles men encounter in attempts to obtain employment, secure housing, and pay fees may then prove discouraging to male offenders, and even hold the potential to spark a reinitiation of substance abuse and/or criminal offending. Although the gendered attributions I have identified are not unique to this setting, their significance is: while officers view themselves more as social workers than correctional officers, their words are backed by the threat of force. These gendered strategies may then have significant implications, as offenders, responding to the power officers hold over their lives, shape their conduct in response to officers' expectations.

NOTES

1. The dramatic increase, and women's rising share of those supervised, is responsive largely to changes in sentencing guidelines and penalties such as those characterized by the War on Drugs, rather than to an increase in criminal offending (Britton 2011; Chesney-Lind and Pasko 2004).

2. A complete description of the sampling procedure and details of the analytic strategy are included in the methodological appendix, which is available as an online supplement at http://gas.sagepub.com/supplemental.

3. Greendale gained this reputation in the early 1990s when a new director was hired who, rejecting the “nail 'em and jail 'em” approach, instituted reforms that aimed more for behavioral change than punishment. While innovative at the time, these practices had increasingly been adopted by county systems across the state.

4. Women are represented across these categories.

5. While I report officer race and gender in the quotes, I do not discuss race differences in the nature of gendered treatment because the observational sample was simply too small to draw conclusions about race–gender interactions. This remains an important topic for future research.

6. The majority of the offenders I selected for the case note analysis were convicted of Drug I.

7. One limitation of these data is that they cannot reveal the source of these differences. However, as most of these topics reflect important aspects of supervision, we would expect officers to bring them up and comment on them regardless of whether offenders did so. Officers are also unlikely to take note of topics that they do not feel are importantly linked with offenders' progress on supervision. Thus, case notes do not represent simply a record of what is discussed in meetings, but officers' distillation of important themes.

8. Although I could not test for the significance of cross-gender pairing because of small sample size, my data did allow for comparison of male officers paired with male offenders to female officers paired with female offenders. These models revealed significant differences across gender pairs only for employment. Male officers working with male offenders made more frequent mention of employment than did female officers working with female offenders.

9. This extract also suggests an alternative interpretation, that officers' focus on emotionality is largely a response to women's greater emotional expression. My observations suggest that women do express emotions more freely in meetings than male offenders, but that this could be in response to officers' inquires. Further, men's visible signs of distress and hints to officers that all was not well were, more frequently than not, passed over in favor of discussion of topics viewed as more central to supervision.

10. Overall, I did not find that officer gender was strongly linked with supervisory style; for instance, female officers working with men were as officious as male officers. Rather, style seemed responsive to the gender of the offender they were working with. That being said, officers did seem to identify more with offenders of their own gender, and female officers could be somewhat less empathetic towards male offenders.

Biography

Jessica Wyse is an NIA Postdoctoral Fellow at the Institute for Social Research at the University of Michigan. Her research centers on the role of gender and family relationships within the context of criminal justice institutions and among the criminally involved.