The Role of Transformation Narratives in Desistance Among Released Lifers

Abstract

Research on desistance emphasizes the importance of the transformation narrative, in which the individual has replaced his old, criminal self with a new, law-abiding self. Key elements of the transformation narrative are generative motivations, the core self, and a sense of agency. Thus far, it is not known what role these elements play in desistance among released lifers. To fill this caveat, we conducted in-depth life interviews with 67 individuals who had served a life sentence. Almost all interviewees presented a transformation narrative that included a good core self and generative motivations, including those who persisted in criminal behavior. We found that individual agency was a key factor distinguishing the paroled lifers from the re-incarcerated lifers. Findings suggest that rather than learning to present a transformation narrative focused on reflecting a good core self and generative motivations, (post-)prison programs should focus on restoring agency to ensure successful re-entry.

Keywords

lifers life imprisonment desistance transformation narrative redemption script agency

Introduction

Empirical qualitative research on desistance suggests that people who abandon criminal activity make identifiable changes to their personal identity and self-narrative, and produce a new, “improved” self that no longer cognitively or emotionally coheres with offending (Giordano, Cernkovic, & Rudolph, 2002; Laub & Sampson, 2003; Maruna, 2001; Sampson & Laub, 1993; Stevens, 2012). In his research on 65 convicted offenders, Maruna (2001) found that people who were desisting from crime had established a coherent and forgiving narrative, a “redemption script,”¹ which made sense of their offending past. In the “prototypical reform story” (Stevens, 2012), those who were desisting from crime portrayed their past offending as something their “old self” had done, someone who was never the “real me.” In Maruna’s (2001) research, those who were still engaging in crime, however, did not present such a redemption script, but rather a condemnation script: These individuals felt themselves to be perpetually defeated by their personal failing and the structural barriers to rehabilitation. They engaged in “negative phrasing” by viewing themselves as passive objects with very little control over their lives. Maruna (2001) emphasized three elements that are central to the redemption script: the idea of a core self, generative motivations, and a sense of agency. First, a key element of the redemption script is the idea of a normal core self. Despite what others think of the ex-offender, there “lurks within him a core of being that is normal” (Maruna, 2001, p. 140). This idea is parallel to Braithwaite’s (1989) notion of reintegrative shaming. According to Braithwaite, shaming can be communicated as either reintegrative or stigmatizing. At its essence, this theory predicts that stigmatizing shame increases crime whereas reintegrative shaming reduces crime. The idea is that through repeated confessions offenders learn to take responsibility for their actions while at the same time are given a “way out.” Offenders should understand that all people have the potential to lapse into deviant or criminal behavior from time to time and that these behaviors occur in a social context not entirely under their control, but that those behaviors do not define the individual as a whole. Thus, if the shaming is reintegrative and offenders are given a way back into the moral mainstream, they should have an idea of a normal, non-criminal, core self, which facilitates reintegration (Maruna, 2001). To maintain the idea of a normal core self, the ex-offender refers to his criminal past in terms of “failure events” (Maruna, 2001, p. 133). In doing so, the ex-offender uses neutralization techniques, most commonly denial of responsibility (Sykes & Matza, 1957) or diffusion of responsibility (Bandura, 1977), such as justifications, excuses, and other explanations for past shortcomings—the most common excuse involving blaming past behavior on the effects of drugs or alcohol (Maruna, 2001; Maruna & Mann, 2006).

Second, Maruna (2001) argued, in the redemption script, generative motivations include a sense of fulfillment—perceiving one’s existence as meaningful. Some of the interviewees in his study engaged in criminal behavior as a way to fill a perceived void or emptiness in their lives (Maruna, 2001). Incidentally, the stigmatizing (non-reintegrative) shaming associated with criminal behavior can cause feelings of despair and a belief that there is “no way out” of deviance. Generative activities thus serve a dual purpose: They can provide offenders a sense of fulfillment in life, and can help clear one’s conscience associated with previous criminal acts. As Marsh (2011) showed in his recent work, much like sober individuals in Alcoholics Anonymous, generative activities can also be therapeutic for offenders, which helps them maintain a path of desistance from crime. Maruna’s (2001) desisters described the fulfillment they derived from being able to contribute to a social establishment they had long seen as their adversary, for example, by volunteering, public speaking, or becoming a counselor or mentor. These generative activities help the individual to deal with the shame and guilt of their past behavior, and serve to re-frame their identity—shedding their old past and make way for a new, re-invented self. The desisters perceived their new identity as incompatible with ongoing criminality. This finding has been supported by others studying narratives of desistance (Appleton, 2010; Giordano et al., 2002; Laub & Sampson, 2003; Marsh, 2011; Paternoster & Bushway, 2009; Sampson & Laub, 1993; Vaughan, 2007).

A third part of the redemption script consists of a sense of agency, or the capability of individuals to act independently and to make their own choices within the social structure. While attracting increasing interest within desistance research, the concepts “agency” and “choice” are used differently by different authors (Bottoms, Shapland, Costello, Holmes, & Muir, 2004). Whereas Laub and Sampson (2003), for example, referred to men with agency as “active participants in the decision to give up crime” (p. 146), rational choice theorists hold that criminal behavior arises out of an opportunity-focused choice. Following Bottoms et al.’s (2004) sage advice to use the concept “agency” with greater precision, we refer to “agentic” action as voluntary action that can be modified by the actor on request. While the desisters in Maruna’s (2001) study emphasized their successes as a result of their own free will, and were hence referred to as “super-agents,” the non-desisting interviewees lacked this sense of control, or a “language of agency.” They perceived themselves as subjected to outside social forces, rather than as actors in control over their lives. The desisters, however, had a sense of control over their destiny that made them more likely to succeed post-release than their non-agentic counterparts. Compared with Maruna (2001), who downplays the role that structure plays in individual agency in favor of “super-agents,” we recognize that structure can indeed influence or constrain agentic action.

The shortcomings of prior research on the role of the transformation narrative, however, are at least twofold. First, with the exception of Appleton’s (2010) doctoral study, these studies are largely based on general delinquents who have served relatively short prison sentences or samples that contain only a small number of individuals serving long prison sentences (e.g., Bottoms et al., 2004; Cid & Marti, 2012; Giordano et al., 2002; Laub & Sampson, 2003; Maruna, 2001; Stevens, 2012). We do not know to what extent these findings are applicable to those who have served much longer prison sentences, such as individuals who have been sentenced to life in prison or “lifers.”

Second, earlier studies either rely on incarcerated offenders who have not had the ability to “prove” their desistance (e.g., Cid & Marti, 2012; Stevens, 2012) or, for studies relying on released offenders, take for granted what respondents say about desistance (e.g., Appleton, 2010). In other words, we have no way of knowing to what extent those with a transformation narrative are, in fact, truly transformed.

The aim of this study is to assess the role of transformation narratives in the desistance process of lifers who were released. In doing so, we take the perspective of the individual, while taking into account the social structure within which his or her actions unfold.

Method

Following Cid and Marti (2012), we used the narrative interview to identify how interviewees constructed their narratives and evaluated their lives before, during, and after incarceration. Second, by making use of a life-history calendar (Laub & Sampson, 2003), we used the interview to obtain information on the interviewees’ life events, as well as determining the context in which these events took place. This dual approach allowed us to analyze both objective and subjective factors related to the formation of the narratives.

Design

This study is part of a larger research project on the influence of imprisonment and life events on desistance among lifers (references omitted to ensure blind review) in two American cities: Boston, Massachusetts, and Philadelphia, Pennsylvania. Individuals were eligible for inclusion if (a) they committed a homicide in the Boston or Philadelphia metropolitan area; (b) they had been paroled or released from a life sentence for this offense over the past 15 years; and (c) they were either not currently incarcerated or were currently re-incarcerated.

City Demographics

The cities chosen for this study are Boston, Massachusetts, and Philadelphia, Pennsylvania. In 2010, Boston had a total population of about 636,000 while Philadelphia’s population was about 1,526,006 (Census Bureau, 2010). Whites and African Americans comprised about 54% and 25%, respectively, of the total population in Boston. Philadelphia has a much higher (about 43%) African American population than Boston and fewer Whites (about 41%). In Boston, about 21% of individuals were living below the poverty line compared with 24% in Philadelphia.

Operationalizing Desistance

In operationalizing desistance, we did not take being out in the free world as a direct measure of desistance. Following Maruna (2001), we recognize that there is no neat dichotomy of desisters and non-desisters. Rather, we consider going “straight enough”—that is, staying crime-free for several years after release—as the primary indicator of desisting. We classified those who were re-incarcerated as non-desisters. As “recidivism” is a term typically used in a criminal justice setting, we use “desistance” and “non-desistance” to emphasize the criminological and process-oriented nature of not engaging in criminal behavior. The terms “desisters” and “non-incarcerated lifers,” and their counterparts “non-desisters” and “re-incarcerated lifers” will thus be used interchangeably to describe the interviewees’ status.

Re-Incarcerated Lifers

The Pennsylvania Department of Corrections and the Massachusetts Department of Corrections selected individuals who met the above-described inclusion criteria and were currently re-incarcerated for a criminal or technical offense. In Pennsylvania, after identifying the individuals that qualified, superintendent’s assistants at each facility presented a letter to the individuals on our behalf, inviting them to participate in the interview. In Massachusetts, corrections program officers (CPOs) at each facility presented a letter to the individuals, on our behalf, inviting them to participate in the interview.

With the exception of two, all individuals who were contacted consented to participate. One did not participate out of a lack of interest, and another individual was only able to conduct the interview in Spanish. In total, we interviewed 33 re-incarcerated individuals: 8 in Pennsylvania and 25 in Massachusetts.

Non-Incarcerated Lifers

In Pennsylvania, we selected currently non-incarcerated, ex-lifer offenders in two ways. First, the Pennsylvania Department of Corrections identified individuals who met the above-described inclusion criteria and were currently not incarcerated. Subsequently, following Institutional Review Board regulations, we asked the Research Division of the Pennsylvania Department of Parole and Probation to contact those individuals who were currently under parole supervision. They presented a letter to the qualifying individuals on our behalf, inviting them to contact the principal investigator if they were willing to participate in the interview. In two cases, these individuals made contact, after which we scheduled a life-history interview by phone.

Second, the Philadelphia Prison Society helped us reach out to individuals who met the inclusion criteria. We supplied the Prison Society with a recruitment letter they could present to the individuals concerned. Two individuals were identified. Upon their consent, these interviews took place in private at the Philadelphia Prison Society.

In Massachusetts, we selected currently non-incarcerated offenders by contacting local organizations that provide services for ex-offenders. We gave these organizations a letter to present to the individuals who met the inclusion criteria, and were currently not incarcerated. Upon the individual’s consent, the principal investigator (initials omitted to ensure blind review) informed the participants about the study and they were given an opportunity to opt in. These interviews typically took place at their attorney’s office or at one of these local organizations. We interviewed a total of 34 non-incarcerated individuals: 4 in Pennsylvania and 30 in Massachusetts.

Interview Procedure

All individuals were given an opportunity to ask questions about the study and their participation. All gave informed consent to participate prior to the interview, on the basis that their disclosed material would be made anonymous and unidentifiable. After the interview, the participants were de-briefed. Findings were shared where this was requested.

In-person interviews with re-incarcerated individuals took place over the course of 10 days (October 2012). The interviews were approximately 2 hr long, depending on the participant’s responses.

In-person interviews with non-incarcerated individuals took place over the course of 3 months (October-December 2012). Two interviews took place at the Philadelphia Prison Society and two interviews were conducted by phone.

We conducted in-depth, one-on-one, semi-structured life history interviews. The questions were developed to obtain a thorough description of factors that characterized the individual’s life before, during, and after incarceration. The majority of the questions were open-ended (e.g., “Could you describe some of the challenges you initially faced after being released?”). In a study on desistance among serious adolescent delinquents, Giordano and colleagues (2002) found that the role of agency emerged almost entirely from the authors’ exposure to qualitative data. As opposed to Giordano et al.’s study, we included explicit questions related to agency. In doing so, we did not only ask about their initial moves toward desistance but also about its subsequent maintenance. In addition, we included specific questions on the other two key elements of the redemption script: the core self (e.g., “Could you describe yourself?”) and generative motivations (e.g., “Do you use your story when you talk to other people?”).

Both in-person and telephone interviews with non-incarcerated individuals were audio-recorded and transcribed ad verbatim. As audio recording was prohibited inside the correctional facilities, interviews with incarcerated individuals were conducted by the principal investigator, while a graduate student took detailed notes that were immediately transcribed after the interview.

Supporting Documentation

As outlined above, studies that emphasize transformation narratives rely upon qualitative data and take for granted what respondents say about their desistance as being an accurate representation of what has happened, or as something that they hope will happen (e.g., Cid & Marti, 2012; Stevens, 2012). This is not to say, as Farrall and Bowling (1999) pointed out, that respondents are actively lying, but rather that they may attempt to make sense of their lives with rationality and intent during the interview. Desistance can only truly be determined retrospectively, after significant crime-free periods (Bottoms et al., 2004). Therefore, to create categorical distinctions between the desisters and non-desisters, we utilized a variety of supporting documentation. We were able to search public court dockets for each individual in our sample to determine whether they had been adjudicated after their release from prison. Furthermore, we cross-referenced the interviewees’ statements with background information consisting of newspaper accounts of the homicide, Pennsylvania Police Department records on the homicide, Parole Board documentation as well as—with the interviewees’ permission—disciplinary infractions that occurred while in prison.

Participants

Participants’ ages ranged from 37 to 75 (M = 53.80; SD = 8.394). Thirty-one interviewees were African American, 28 were White, 7 were Hispanic, and 1 was Asian. Sixty-two of the 67 interviewees were male. Interviewees in both groups committed the homicide² in their teens (n = 28), or in their 20s (n = 33), between the 1970s and 1990s. The time spent in prison for the homicide was on average 19 years (SD = 8.45), ranging from 1 to 40 years.

About half of the sample (n = 34) had an extensive criminal history before the homicide, being involved in activities such as property crimes, selling drugs, and (armed) robberies. Only seven individuals (10%) had no history of delinquent behavior prior to the homicide. Both during the time of the homicide as well as after release, most participants in this group resided in (poor) urban and suburban locations. There was little divergence among the group in socio-economic terms—Most were unemployed at the time of the homicide or were working in manual jobs. At the time of the homicide, the majority lived alone or with non-family members.

At the time of the interviews, the non-incarcerated interviewees were on lifetime parole. The 33 re-incarcerated interviewees were detained in maximum-security or medium-security facilities. Of the re-incarcerated individuals, approximately half were re-incarcerated because of a criminal violation, and the other half were re-incarcerated for a technical violation. The re-incarcerated groups did not differ in terms of demographics or criminal record.

Data Analyses

Following content analyses used in previous work (Appleton, 2010; Giordano, Longmore, Schroeder, & Seffrin, 2008), initial data analyses consisted of reading the interview transcripts several times, and then noting connections, associations, and preliminary interpretations. Analytic conclusions were formulated by coding and then categorizing similar statements of experiences from data, replicating qualitative methodologies described in other studies on paroled offenders (Appleton, 2010; Liem & Kunst, 2013). With the aid of qualitative software (NVivo Version 9; QSR International, 2010), these statements were grouped into categories and were then compared across all transcripts to identify connections, patterns, or contradictions.

Results

To determine the role of the transformation narrative in the lives of both desisting and non-desisting released lifers, we specifically examined the role of the core self, generative motivations, and a sense of agency. In Table 1, the columns reflect the incarceration status at the time of the interview—dichotomized in “desisting lifers” (those who were not re-incarcerated) and “non-desisting lifers” (those who were re-incarcerated). For the desisting lifers, the average time staying crime-free since their most recent release was 5 years, therefore we are able to say with confidence that these lifers were indeed desisting. The rows in the table reflect elements of the redemption script. Our findings show a lack of a one-on-one relation between the presence/absence of elements of the redemption script and desistance. The large majority of re-incarcerated interviewees exhibited a sense of a core self and a sense of generative motivations. What was absent among most re-incarcerated lifers, however, was a sense of agency. The large majority of desisting lifers, however, showed a sense of a core self, agency, and generative motivations.

Table 1:

Elements of the Redemption Script in Desisting and Non-Desisting Lifers.

	Lifer Status
Elements of the Redemption Script	Desisting Lifers	Non-Desisting Lifers
Core self	Good core self: “You know I’m a good person. I’m a good human being, I’m a good friend, I’m a good son. I’d describe that I’m not defined by the labels that might be put on me because of my past.” (Male, age 41)	Good core self: “I am selfless. I know I could give all I have to someone to better themselves.” (Male, age 42) “I’m honest, open-minded, determined, grateful, but sometimes also disappointed with myself for having put myself and my wife through this.” (Male, age 62)
Generative motivations	Making good: “Using my story, my experience, and just like building the bond, I can see a lot of the clients [at the re-entry center], they can really identify with me, even though I don’t have the big degrees, and all that, but because I have the experiences, they can create a bond. They know what I am coming from and I understand a lot better where they are coming from . . . So it’s been really good. Fulfilling, you know, so fulfilling, because I know that I am helping someone.” (Male, age 46)	Making good: “When I was out, I volunteered in church and at the food bank . . . And when I was in minimum [security prison], I did public speaking at schools. If only I could touch one kid, prevent one kid from going through what I have been through . . .” (Male, age 52) “I was doing good. I was helping old people, and homeless people. I’d pass out clothing and food and stuff. And I shoveled snow for old people.” (Male, age 52)
Sense of agency	Individual agency: “I have control over my life within my ability . . . There are many solutions to solve a problem. The key mistake that people make is that there’s only one way to solve a problem. (Male, age 50)	External locus of control: “I know God’s running my life . . . I don’t go to church, but . . . I have God in me.” (Male, age 52) “I came back to prison because of the economy.” (Male, age 52)

Core Self

The large majority of interviewees (both non-incarcerated and re-incarcerated) presented the so-called “failure events,” in which they attributed past criminal behavior to emotional and physical neglect, resulting in drug and alcohol use, which in turn resulted in juvenile delinquent behavior. Many attributed criminal behavior in adolescence to having weak bonds with parents.

I was raised by my grandma and my mom. I never knew my father. I didn’t get any excitement from my brothers so I got my role models from the street. That’s when I started learning violence. (Re-incarcerated male, age 56)

I was beginning [hanging in the streets] you know at thirteen, fourteen, fifteen years old I became a little rebellious because I wanted to be with my friends and they [my parents] wanted me to go to church so it was a clash between me and my father, my parents. And that just pushed me further into the streets . . . So you move from that and you get into alcohol, drugs and then eventually you get into crime. (Non-incarcerated male, age 61)

Most of the interviewees reported having been involved in the “street life” involving the use and sale of drugs and alcohol. However, it was clear from the interviews that the “street life” did not represent their “core self.” It was only their behavior, but did not reflect who they “really were.” They stressed that they became involved in these activities because of environmental factors such as pressure from a group or growing up in an impoverished area where crime and drug use are rampant.

I smoked weed when I got into my teenage years because it was easy to get into those types of things . . . that played a major part of me not doing certain things and not going to school. So I always was a follower and that’s how I’ve always looked at myself. I wouldn’t always be the one who wanted to get into mischief but I did what I needed to do for my substance abuse; . . . I did it because I wanted to do it and it made me feel good even though I thought it would help with the problems that were going on and what I dealt with, with my stepfather or school. (Non-incarcerated male, age 57)

Almost all interviewees, however, emphasized that this was who they had been—now, they were different: They had shed their old, delinquent selves and were now the person they were supposed to be all along, or as one interviewee expressed, “Now, I’m more stable, more situated, now I know to be a law-abiding citizen” (Re-incarcerated male, age 59). Another re-incarcerated interviewee told us, “I hope the parole board is convinced, because I’m a good person and I can be a good contributing member of society” (Re-incarcerated male, age 40). They portrayed their former offending self as a false identity, denying that it was their “real self.” This false identity runs parallel to Paternoster and Bushway’s (2009) concept of the “feared self”—An image of what the person does not want to become, or turn back to. Another interviewee states,

My life had been out of balance, you know, because I focused heavily on the negative activity. And that who I was, that’s what people saw me as, as a criminal, as a street thug. And I was trying to redeem myself by showing a side of myself and to demonstrate that I had sociable redeemable qualities and the pains that I did wasn’t who I was, it was just things that I did. (Non-incarcerated male, age 61)

Another interviewee phrased the contrast between her old self and new self by stating that 10 years ago, “I would say: ‘Fuck everything and everybody.’ . . . Nobody gives a fuck about me, so why the fuck should I care about anybody else?” Today she describes herself as, “I am a better person. I’m a stronger person, nothing can break me down” (Non-incarcerated female, age 49).

Generative Motivations

Both desisting and non-desisting interviewees expressed how they were involved, or wanted to become involved, in generative activities. These activities were mostly centered on their life stories, and how they could help others with the life lessons they learned. For those who are re-incarcerated, generative motivations were mostly aimed at younger inmates: “I talk to people about the consequences [of joining a gang] . . . There is no work for these young people, and they want easy money, so they go on and sell drugs” (Re-incarcerated male, age 59). Another interviewee sought out programs not only for himself but also for his future children: “I want to be able to instill in them pro-social behaviors” (Re-incarcerated male, age 50). Others expressed the desire to “give back” professionally once they were released, for example, as a paralegal: “I want to get a paralegal license ’cause I want to try to make some type of change, some type of impact” (Re-incarcerated male, 50). The same is mentioned among those who were released:

After having spent so much time in prison, one of my passions is to become a prisoner’s advocate, so I can help teenagers and adults not make the same decisions and choices that I made. Just to be a role model, I remember one of my personal models reinforced the person I am today. (Non-incarcerated male, age 53)

Being involved in generative activities strengthens the new, conformist, and pro-social identity some have assumed. By being a counselor, acting as an expert and “sharing their story,” they use their past to reinforce their newly invented self:

. . . since I came out I might have spoken at 35 different places . . . I ran four different groups from all of my certifications from inside prison doing life skills I learned out here . . . I was content and I was happy. (Non-incarcerated male, age 57)

“Giving back” seemed to have contributed to an identity as a “good person” both among non-incarcerated as well as re-incarcerated interviewees, such as the following interviewee who volunteered at a father’s group in his community when he was out on parole:

Volunteering is something I wanted to do, to give back to society because of what I did. I told them [these men] my story, and how I dealt with it. It made me feel good, to help someone else and preventing them from doing this. (Re-incarcerated male, age 44)

Sense of Agency

The above-mentioned results illustrate that those who re-offended did not differ in terms of their idea of their “core self,” nor in terms of generative motivations, from those who did not re-offend. Rather, the two groups differed in terms of their sense of agency—the capability of individuals to act independently and to make their own choices within the social structure. The sense of agency, or a lack thereof, was expressed in the individuals’ narratives of their lives before, during, and after imprisonment. The majority of interviewees who were re-incarcerated reflected a low degree of agency, exemplified by minimizing their involvement in the events that brought them to prison or by attributing the causes to external factors beyond their control. They used neutralization techniques when referring to the homicide they were first incarcerated for, as well as later recidivist offenses. One interviewee, for example, describes how the victim of the homicide “. . . was coming at me with a knife. We fought over the knife and the knife pierced his heart through his hand” (Re-incarcerated male, age 52). Police records of this case, however, indicate,

Witnesses told police that the decedent was being chased by three . . . males. [The suspect] gave a statement . . . admitting stabbing the decedent when they got into an altercation over the decedent bumping into him in front of the bar and refused to apologize.

The interviewee explains other future crimes as cases “I picked up.” Another describes, “The case was over a younger girl. I got her pregnant and her ex-boyfriend had something for her. He tried to stab me and the gun accidentally went off” (Re-incarcerated male, age 59). This provides a different narrative than that reflected in the police report, stating, “[The interviewee] gave a signed statement . . . admitting to shooting [the victim] after [the victim] took off his jacket and produced a knife.” Another refers to the homicide as “an accident with a rifle.” He stated that the rifle was broken and when he tried to fix it, it went off and killed the victim: “Anyway, she passed away” (Re-incarcerated male, age 59). Another interviewee stated, “One of my associates got into a fight with another individual and when I went to help him, things got out of hand” (Re-incarcerated male, age 50). Other interviewees claimed they were not guilty of the homicide they were convicted for, which is not to say that they denied being involved in other criminal activities surrounding the homicide. Their denial of the initial offense is still compatible with their present idea of “having changed”—even though they denied being involved in the homicide, they acknowledged living a life characterized by crime and associated behaviors at the time of the crime: “It’s probably a good thing I came to prison, because if I didn’t, I’d probably be dead” (Re-incarcerated male, age 41).

We do not intend to pathologize these individuals for minimizing and rationalizing their involvement in a lethal act of violence (see also Maruna & Mann, 2006), but rather wish to emphasize that the re-incarcerated participants expressed a lack of choice in their involvement in the homicidal event. Differences between individual agency in the desisting group and the re-incarcerated group were also encountered when asked how they spent their time in prison. Whereas the desisting group emphasized that they became involved in prison programs that increased their sense of agency, re-incarcerated interviewees either completed prison programs to enhance their chances of parole or did not become involved in prison programs at all. For example, asked if he was involved in any prison programs during incarceration, one interviewee responds, “They did not tell me I had to do any programs” (Re-incarcerated male, age 52).

The differences between the two groups became most clearly visible, however, in their sense of agency after release, and could be centered around four themes that were most prevalent among the re-incarcerated group: the idea of chance, the notion of self as passive object, “negative phrasing,” and finally, the inability to reflect.

First, the notion of chance was voiced by several re-incarcerated individuals, who considered the outcome of their lives as a matter of chance, rather than individual choice. When asked how someone can be successful in staying out on parole, for example, one interviewee answered that you should “keep hoping . . . you have to be patient and hopefully, good things will happen” (Re-incarcerated male, age 52). This interviewee further expresses the role of chance by referring to a lottery system to obtain employment. At the time of the interview, his brother had put his name in a lottery for a manufacturing job: “I didn’t get in last time, but this time I might be able to get a job. If you get lucky, you may get a job that can give you $12-$15 an hour” (Re-incarcerated male, age 52, italics added).

Laub and Sampson (2003) suggested that agency is reactive, rather than purposive, in that desistance most frequently occurs by default, as a result of social ties with family and employment, with little reflection and without meaningful intention. Similarly, it may be argued that agency flourishes in a context of some social factors present in the case of desisters and absent in the case of non-desisters.³ Both groups of lifers, however, seemed to be “off-time” in terms of traditional turning points in the life course; they were either single, or involved in problematic, rather than supportive, intimate partner relationships. With the exception of one, none of the interviewees was able to maintain a relationship with their intimate partner throughout their confinement. Over time, visits diminished and in both groups of interviewees, were typically limited to visits from parents. They had lost contact with their children during their decades-long incarceration, or never had children. Interviewees in both groups were struggling to secure and maintain employment. The groups did not differ in their access to social support networks, such as religious institutions or substance-related programs. In short, neither desisters nor non-desisters had access to potentially life-changing social bonds. Desisting lifers, accordingly, went through an agentic, internally motivated, rather than externally motivated, desistance process.

Second, the absence of agency was closely tied to the idea that others exerted control over their lives. In line with Maruna’s (2001) concept of the “condemnation script,” these individuals engaged in “negative phrasing” in that they described themselves as subjected to social forces. One interviewee, for example, said that he hoped to get in contact with an organization post-release: “I don’t know which organization, there are so many, you know what I mean. They can help me . . . they can steer me in the right directions” (Re-incarcerated male, age 59). Another inmate expressed the lack of control he felt by stating, “Parole is a system designed for failure. There’s no way to change since the odds are stacked against you.” This perception of a lack of control over one’s life is also expressed in how he felt about younger inmates’ ability to change. When asked if he had advice for younger inmates just entering the system he stated, “I don’t really have much advice cause I probably won’t be able to reach them anyways” (Re-incarcerated male, age 50).

Another element that emphasizes the lack of control over their own lives includes the strong belief in higher powers controlling their lives. Several re-incarcerated interviewees believe that if they pray for change, change will occur: “It keeps me stable, to believe in a higher power . . . I need this [religion] in my life, to get out and to stay out . . . If there’s a God, I’ll get paroled again” (Re-incarcerated male, age 59).

It is worth noting that religion does not necessarily mean that someone gives up individual agency. The difference between the two groups lies in the relation with religion: Whereas individuals in the re-incarcerated group voiced the idea that others (including God) were in control of their lives, those in the desisting group experienced religion in an active, rather than a passive way:

It [Islam] became my way of life . . . it always taught you to understand more about yourself, your family, and those around you because for you to get into it you have to really practice it within yourself . . . God won’t help people unless they change themselves and that’s why I learned to change myself. (Non-incarcerated male, age 57)

Third, the absence of agency among re-incarcerated lifers became apparent in their “negative phrasing”: In describing their lives and the circumstances that resulted in their incarceration as well as their ideas for the future, many of the re-incarcerated interviewees used a passive, rather than active voice. They further emphasized that being successful in staying out depended on not to act, rather than to act. This was exemplified by an interviewee, who said that while out on parole he was “trying to do the right thing. You know, working and staying out of trouble” (Re-incarcerated male, age 52). For another re-incarcerated interviewee, the keys to success were to “stay busy. I try not to get into anything” (Re-incarcerated male, age 59).

For the desisters in our sample, maintaining desistance after release seemed important to them. Unlike the re-incarcerated individuals in the preceding paragraph, the desisters used an active rather than passive voice when discussing desistance maintenance. In other words, desistance was seen as contingent upon acting rather than not acting. Much of this maintenance consisted of their continued participation in programs both before and after incarceration. As one of the desisters expressed,

I was doing work for [a re-entry program] for a number of years [while incarcerated] . . . When I got out I wanted to take advantage of the services and resources that they provided because I know they provided them. (Non-incarcerated male, age 61)

Their initial involvement in programming while incarcerated was the catalyst of change: “I really submerged myself into prison programming, re-entry programming, trying to understand why I after having served so much time the first time thought about committing a crime. That’s when I started to realize my criminal insanity” (Non-incarcerated male, age 53).

Fourth, the absence of agency among re-incarcerated lifers appeared to be strongly related to the inability to reflect. Agency, as Deitz and Burns (1992) pointed out, is associated with intentional and reflective actions—the latter referring to the notion that “the actor has enough awareness of the effects of actions to monitor those effects and use information about the perceived effects to modify their rule system” (p. 192). Several of the re-incarcerated interviewees, however, simply did not seem to be able to reflect on available options to be successful on parole. The lack of introspection in one’s behavior and alternatives to criminal behavior was largely attributable to a lack of self-awareness, and strongly related to low intelligence. One interviewee whose IQ seemed to be well below average, for example, was therefore convinced that he would be successful on the outside, because “in this prison, there is no smoking, so all the bad is out of me” (Re-incarcerated male, age 59). His narrative, however, lacked a realistic sense of agency, as much of what he said seemed scripted: “Nobody influences my behavior but myself,” and “On the inside, I used programs to better myself,” and “I try to read self-help books in prison.” Whereas he successfully refashioned a narrative in which he emphasized a new “replacement self,” he was not able to tie in a sense of agency in his prototypical redemption narrative. Even though these interviewees may have agentic intentions, a lack of self-awareness acts as a constraint in successfully expressing agency.

Negative Cases

Out of the total 33 re-incarcerated interviewees, three individuals presented all elements of the redemption script, but were nevertheless considered as “non-desisters.” Consider, for example, the following interviewee, who was re-incarcerated for a technical violation: When his Parole Officer showed up at his house, he was at his brother’s house without notification. He describes that, in his 17-year long sentence, “I decided to change, and it wasn’t handed to me on a silver platter. I had to do the work.” Throughout the interview, he emphasizes, “There is always a choice.” Even in prison, he says,

I can choose not to eat. I can choose not to wear this monkey suit [pointing to his jumpsuit]. They can say what they want, but I have a choice. I don’t have to listen to him [pointing to the Correctional Officer in the adjacent office]. You will be sent to the hole [solitary confinement] for that, but it’s a choice. I won’t let them take away my freedom, my choice, my individuality, even if I’m wearing this suit like a monkey, I’m still who I am. (Re-incarcerated male, age 41)

What characterizes these “negative cases” is that these individuals presented all elements of the redemption script, but were still re-incarcerated, typically involving a triviality. To give a few examples, one parolee was re-incarcerated after testing positive for a small amount of alcohol, another was violated after a non-physical, verbal altercation between him and his girlfriend and still another for a curfew violation.

Conversely, there were a total of four negative cases involving non-incarcerated lifers, who presented few elements of the redemption script, and a low degree of agency. The following interviewee, for example, who had been out of prison for 10 years openly questioned whether he even felt “transformed,” and pointed out that the reasons he was staying out of prison was because of being on parole:

I just think I take my life more serious than I did before going away. Did I transform? . . . I take my life more seriously. I’m more mature, you know what I’m saying, I, I want certain things in life, so, I want to live comfortably. I was thinking, . . . if I wasn’t on parole . . . I would smoke a joint, you know? But, I know I don’t have that, that freedom to do that. You know what I’m saying? So, I mean in that respect I can appreciate that I am on parole, so someone can kick me in the straight line. You know what I’m sayin’, I am not, willing to get into anything and everything. (Non-incarcerated male, age 52)

For these men, in spite of a lack of agency, or without the idea that they had gone through a personal transformation, the reasons for staying out of trouble were related to Parole having a tight grip on them, or simply because they “aged out”: “I don’t get involved in things because of my age. I am seventy-five [years old], I don’t drive anymore” (Non-incarcerated male, age 75). His movements and behavior were largely limited by his age, making desistance neither a conscious choice nor a result of individual agency, but rather a side effect of aging out.

In summation, almost all interviewees, including both desisters and non-desisters, did not appear to differ significantly in their idea of a normal core self nor in their generative motivations. What was strikingly different between the desisting and non-desisting groups was their sense of agency. Almost all interviewees from the non-desisting group expressed a low degree of agency, characterized by minimizing their involvement in the circumstances that brought them to prison. This group also displayed a rather passive attitude about their outlook on life by attributing their potential for success or failure to external forces such as God or their parole officer. Those from the desisting group, however, expressed a strong sense of agency and control over their own lives. While recognizing that a variety of social and environmental factors (e.g., substance abuse, lack of financial resources, and low educational achievement) influenced their behavior and prospects upon release, the desisters still displayed high levels of agency as evidenced by their belief that they are able to act independently and make their own choices.

Discussion

We set out to assess the role of transformation narratives in desistance among those who have served a life sentence. Previous research on desistance suggests that those who stay away from crime have made identifiable changes to their personal identity and self-narrative, and produce a new, “improved” self which no longer cognitively or emotionally coheres with offending (Laub & Sampson, 2003; Maruna, 2001; Paternoster & Bushway, 2009; Sampson & Laub, 1993; Stevens, 2012). The above-mentioned accounts from both “desisters” and “non-desisters,” however, suggest otherwise; individuals from both groups present a prototypical reform story—almost all of them had consciously fashioned a “better” version of themselves. The large majority expressed that they were now different from the person they were at the time of the homicide, all subscribed to conventional moral values, and almost all emphasized that they had redeemed themselves and were now “the person they were always meant to be.” Our findings thus speak against Maruna’s (2001, 2004) observations that desisting former offenders were more likely than active offenders to employ external attributions to explaining past criminal behavior. Arguably, our finding that participants from both groups used neutralization techniques to minimize their responsibility in past criminal behavior enabled them to reconstruct their biography and modify their past in a manner that is conducive to their present self-concept (Maruna & Mann, 2006). Both groups, in other words, presented some of the key elements of the redemption narrative, including the notion of a good core self and generative motivations. To maintain a self-concept that was essentially good, many used techniques of neutralization (Sykes & Matza, 1957) or, what Scott and Lyman (1968) termed “sad tales,” to highlight an extremely dismal past that serves to “explain” their past actions and present state. What differed between the two groups was the individual’s level of agency or self-efficacy. Agency, we argue, was not a necessary requirement for individuals to express such “transformational power,” and create a narrative of change, as suggested elsewhere, but it is an important element of the redemption narrative in desisting from crime (King, 2012). The large majority of interviewees in this study discarded their past identity in favor of an alternative, pro-social one, and used generative motivations, ranging from speaking in public, running in-prison programs, and expressing a desire to “give back.” The finding that both desisters and non-desisters present such elements of the transformation narratives may be traced back to the role of (post-)prison programming that encourage therapeutic discoveries and emotional disentanglement from ex-offenders “old” selves (Stevens, 2012). Another explanation for the omnipresence of these narratives can be found in lifer groups in prison, who help each other prepare for parole board hearings. These programs and interactions thus aid in the creation of new identities, produced through the autobiographical work these lifers engaged in for so many years. We do not argue that these stories are fictions, in the sense of being “made up,” but rather, as Haraway (1997, cf. Lawler, 2008) suggested, devices to produce certain kinds of meaning. This is not to say that these interviewees are purposely lying—rather, we suggest, they have been schooled in presenting a narrative of redemption, change, and “giving back” (see, for example, California Department of Corrections, 2010; Kentucky Department of Corrections, 2013)—and for such a narrative to arise, our results suggest, agency is not a crucial factor.

This blurs the line between the concepts “primary” and “secondary” desistance. While primary desistance refers to “any lull or crime-free gap . . . in the course of a criminal career” secondary desistance involves “the movement from the behavior of non-offending to the assumption of a role or identity of a non-offender or ‘changed person’” (Bottoms et al., 2004, p. 371). All of our participants described themselves as “changed” into “better persons.” In spite of their identity change, most were not crime-free after being paroled from their life sentence and were, as a result, re-incarcerated. We thus argue that primary and secondary desistance are inadequate terms to describe the conformity versus non-conformity among this group of lifers. The conundrum can be solved by emphasizing the role of agency, rather than the mere presence of a transformation narrative: Those who are successful in staying out expressed a higher degree of agency, or self-efficacy, compared with those returning to prison. The process of desistance is thus not solely the result from a change in male coming-of-age societal forces (e.g., parenthood, marriage, employment, military service; Carlsson, 2013), as emphasized by life-course theorists (Bersani, Laub, & Nieuwbeerta, 2009; Cid & Marti, 2012; Ganem & Agnew, 2007; Laub & Sampson, 2003; Lopoo & Western, 2005; Sampson & Laub, 1993, 2003a, 2003b; Sampson, Laub, & Wimer, 2006; Uggen, 2000), or a resolution of an individual to change. Even through it is increasingly recognized that both these factors are necessarily implicated in change (King, 2012; Vaughan, 2007), previous research has still downplayed the role of individual agency, while overestimating the power of social environment to instill law-abiding behavior. This study emphasized the role of agency in desistance. We stress that those who re-offended did so out of a lack of agency, giving way to impulsivity and a lack of knowledge of future consequences rather than—as Sampson and Laub (2004) observed—a choice for a criminal lifestyle because of the rewards of crime itself (Katz, 1988).

No analysis of agency, however, is complete without considering the role of social structure, such as social class, religion, gender, ethnicity, and so on, that determine or limit an individual and his or her decisions. Empirical research on desistance from offending has treated individuals as either “super-agents” or as “super-dupes” (Farrall & Bowling, 1999). In this view, “super-agents” are free to act as they choose and can directly influence the outcome of their lives through their decision making, while the lives of “super-dupes” are an outcome of their role as a passive play ball, subjected to societal forces (see, for example, Appleton, 2010; Maruna, 2001). This divide between agent’s own actions on one hand and structural constraints on the other, Farrall and Bowling (1999) argued, has tempted to explain experiences of crime on one side of this false dichotomy. In interpreting these results, we should not neglect the societal circumstances that affect the lives of the interviewees. As Carlsson (2013) suggested, even though adult men are expected to have “aged out” of crime and live conventional lives, some offenders fail to desist—offending could be seen as a means of achieving a masculine status at times when economic independence were problematic. Among the men in our sample, the reasons for being re-arrested, re-convicted, and subsequently re-incarcerated, for example, were often of trivial nature, a product of harsh parole guidelines and in some cases, the result of misguided justice—an observation also encountered elsewhere (Hanrahan, Gibbs, & Zimmerman, 2005). We do not intend to minimize the structural inequalities ex-offenders face and the possible implications for a lack of (masculine) status that is compensated by criminal behavior (Carlsson, 2013). Rather, we argue that those individuals who are able to stay out successfully have acquired the sense that they have at least some choice and some amount of power that enables them to make a difference in their lives. Both groups of lifers faced a number of various constraints within their immediate social contexts, and it is how the individual responds to these that is central to the argument presented here. We thus adapt Mouzelis’ (2008) view that social actors make choices based on the structures that surround them (see also Farrall, Sharpe, Hunter, & Calverley, 2011) and agentic individuals, in turn, have the ability to shape their lives within these structures.

Limitations

While this study has attempted to provide unique insights into the role of transformation narratives on desistance, it is not without its limitations. The first constitutes the demographics of our sample, a large proportion of the interviewees were African American. When interpreting these results, one should keep in mind the relative overrepresentation of this group. Nationwide statistics reveal a similar pattern showing the disproportionate representation of African Americans as offenders and victims in lethal outcomes (Farrington, 2012) as well as in recidivism after a life sentence (reference omitted to ensure blind review). This may be explained by a variety of factors that disadvantage African Americans, ranging from sentencing practices (Alexander, 2012) to structural disadvantages, including disadvantages associated with the neighborhoods they return to (Laub & Sampson, 2001; Sampson & Morenoff, 2004; Western & Pettit, 2010). In addition, our sample was skewed in terms of gender. Women are generally underrepresented as offenders in violent crime, including homicide (Cooper & Smith, 2011). Desistance research, however, suggests that women experience different pathways into and out of crime compared with men and are subject to a different set of structural and psychological factors in the desistance process (Farrall et al., 2011; Giordano et al., 2002). To shed further light on these differences, future studies should attempt to oversample underrepresented racial and gender groups.

A second limitation concerns some of the individuals in the “desisting” category. A handful of individuals in this category have only been out of prison for just several years. Thus, their desisting may be an artifact of time rather than a “true” change in behavior. At the same time, this seems unlikely given what we know from previous research, showing that those who recidivate mostly do so within 3 years of release (see, for example, Baaij, Liem, & Nieuwbeerta, 2012; Neuilly, Zgoba, Tita, & Lee, 2011; Roberts, Zgoba, & Shahidullah, 2007; Vries & Liem, 2011; and for an overview, see Liem, 2013). Nevertheless, future studies should attempt to address these limitations by increasing the sample size to include a representative number of individuals from different ethnic and racial groups and allowing more time to lapse after release before an observation is made.

A third limitation concerns a possible selection effect when it comes to who is likely to have been paroled, and hence, included in our study and who is not. First, our study only includes those who have been convicted of second-degree murder (Massachusetts) and third-degree murder (Massachusetts and Pennsylvania), and those committing first-degree murder are not eligible for parole. More importantly, however, a selection process may be at work as a result of the expression of agency during incarceration and during the Parole Board hearings. This poses a challenging paradox. Incarceration transforms the potency of an individual’s agency (Farrall & Calverley, 2005), making him an obedient prisoner (Sykes, 1958). Those who show too many signs of independent decision making to advance their situation may be seen as anti-authoritarian, while those who show too little may be accused of not taking initiative to better themselves. In other words, the idea that others, rather than the self, are in control over their lives is reinforced in the prison system. The paradox lies in the observation that after having been institutionalized for decades, and being encouraged to shed all sense of agency, the inmate is expected to display a strong sense of agency when facing the Parole Board. This brings about the question who, in the end, gets paroled: It is a fine line to walk between being a model prisoner who has shed his individual willpower over years of institutionalization and the prisoner who has to prove he possesses a high degree of agency. He needs to prove that he has the capability to make plans for employment and housing, the ability to build a life post-release, to have insight into the cause of his prior criminal behavior, and the strength to resist the temptations to return to the life he left behind (Burns, Kinkade, Leone, & Phillips, 1999). Those who have not convinced the Parole Board that they can successfully walk this line, have never been paroled and hence, not included in this study. This may inhibit the generalizations of our findings to larger groups of released prisoners, including those long-term incarcerated offenders who “wrapped up” their sentence and never had to face the Parole Board.

A fourth limitation is that we did not factor in self-reported crime for making categorical distinctions between the desisting group and the non-desisting group, which has the potential to complicate the distinction between the two groups. It could be argued that interviewees in both groups committed crimes, and that those in the desisting group reported a high level of agency because of their success in not getting caught, rather than their avoidance of illegal behavior.⁴ Even though we acknowledge that people on parole may commit crimes for various reasons, we hold that self-reported crime is highly unlikely to go unnoticed among the desisting group given their tight parole restrictions and constant monitoring and supervision.

Furthermore, it should be noted that individual narratives constitute our unit of analysis, and hence should be interpreted as narratives. While constructing this narrative, non-incarcerated individuals may over-emphasize the role of agency and downplay the influence of other factors, while re-incarcerated individuals would do the opposite. Future research should attempt to assess potential changes in narrative through longitudinal design—examining individual narratives at initial release from prison (in which all participants arguably have similar prospects of re-entry), as well as after a follow-up of several years.

Finally, not all of the interviews were conducted under ideal circumstances. Two interviews were done by phone and several interviews were carried out by two interviewers. Phone interviews were conducted to reduce travel-related costs for both the interviewers and the interviewees. Also, given that audio recording devices were not allowed inside the correctional facilities, a note-taker assisted in the interview process. Thus, these less-than-ideal circumstances could potentially jeopardize an atmosphere of confidence we hoped to achieve that would guide this kind of interview.

One of the key conclusions drawn from this article is that non-incarcerated lifers exert agency, or the power to exert control over the direction of their lives, while most re-incarcerated lifers lack such power. It is not unreasonable to imagine that, while constructing a narrative, non-incarcerated individuals may over-emphasize the role of agency and downplay the influence of other factors. Conversely, re-incarcerated individuals may over-emphasize the role of external factors to explain their lack of success in re-entry. A lack of future-oriented agency may then be interpreted as an expression of a current state of helplessness in being re-incarcerated. Future, longitudinal research designs should seek to overcome this limitation by measuring agency at different points in the lifers’ criminal history—before initial parole, during parole, and upon re-incarceration—to accomplish a full overview of the development of such agentic versus non-agentic narratives.

Implications

Treatment programs for ex-offenders are typically deficit-based and aim to eradicate or reduce the various problems associated with criminal behavior (Ward & Brown, 2004); here, the story is that ex-offenders are people with multiple deficits, ranging from drug and alcohol use, poor educational history, to psychological problems and anger issues (Maruna, Immarigeon, & LeBel, 2004). To reduce crime, conventional programs focus on teaching prisoners to eliminate these deficits by education, addiction counseling, anger management, and the like. Once inmates are released, there are a variety of re-entry programs designed to help them ease their transition back into the community including finding housing, gaining employment, and maintaining sobriety (Petersilia, 2009; Travis, 2005). Maruna (2001), instead, has argued for an institutionalization of the redemption ritual as a necessary part of successful re-entry; by being rewarded for participation in programs and by re-entry courts recognizing redemption (Maruna & LeBel, 2002), offenders would feel recognized in their efforts to change. Our findings, however, suggest that acquiring a transformation narrative alone may not touch on the quintessential part of desistance post-release and may do little in the way of changing actual behavior. Our findings suggest that the distinguishing factor between desisters and non-desisters is agency, or a lack thereof, rather than other parts of the transformation narrative, such as a good core self or generative motivations (Maruna, 2001). Rather than learning to present a transformation narrative focused on presenting a good core self and generative motivations, (post-)prison programs should focus on restoring a sense of agency to released inmates to ensure successful re-entry. More important than holding someone responsible for something they have done in the past—what Maruna and Mann (2006, p. 167) termed “passive responsibility,” is whether individuals feel a sense of control over their lives and take “active responsibility” for putting things right for the future. Control over the future, then, might be a valuable therapeutic tool (Ward & Brown, 2004). Years of incarceration—most often decades for convicted murderers—result in stripping away any sense of individual agency and replacing it with institutional-level control (Sykes, 1958). Once these lifers are paroled, they are expected to, automatically, get back into the driver’s seat. This seems to be an unreasonable expectation if not for the fact that these individuals have spent all of their time being told when to exercise, work, see visitors, eat or sleep. This disempowerment creates a mentality that is not well suited to overcoming the structural impediments that many face upon release from prison (Farrall & Calverley, 2005). Thus, to undo years spent stripping away an inmate’s sense of agency, there ought to be intensive, post-release programs that help inmates regain control over their lives and actions. As a result, they may be more likely to make a successful re-entry into the society they left many years ago.

Footnotes

Authors’ Note:

This research was supported by a Marie Curie Outgoing Fellowship for Career Development in the project 299875.

Notes

Marieke Liem is a Marie Curie fellow with the Program in Criminal Justice Policy and Management (PCJ) at Harvard Kennedy School (HKS). Her research interests involve homicide and international comparative research in lethal violence. She currently works on a book on the effects of long-term imprisonment.

Nicholas J. Richardson is a doctoral candidate in the Department of Sociology and Anthropology at North Carolina State University. His research interests include mass incarceration, neighborhoods and crime, and neighborhood context and recidivism.

References

Alexander

(2012). The new Jim Crow. New York, NY: New Press.

Appleton

C. A.

(2010). Life after life imprisonment. Oxford, UK: Oxford University Press.

Baaij

Liem

Nieuwbeerta

(2012). Ex-imprisoned homicide offenders: Once bitten, twice shy? The effect of the length of imprisonment on recidivism for homicide offenders. Homicide Studies, 16, 219-237.

Bandura

(1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.

Bersani

B. E.

Laub

J. H.

Nieuwbeerta

(2009). Marriage and desistance from crime in the Netherlands: Do gender and socio-historical context matter. Journal of Quantitative Criminology, 25, 3-24.

Bottoms

Shapland

Costello

Holmes

Muir

(2004). Towards desistance: Theoretical underpinnings for an empirical study. The Howard Journal of Criminal Justice, 43, 368-389.

Braithwaite

(1989). Crime, shame and reintegration. Cambridge, UK: Cambridge University Press.

Burns

Kinkade

Leone

M. C.

Phillips

(1999). Perspectives on parole: The board members’ viewpoint. Federal Probation, 63, 16-22.

California Department of Corrections. (2010). Inmate programs. Retrieved from http://www.cdcr.ca.gov/Facilities_Locator/CIM-Inmate_Programs.html

10.

Carlsson

(2013). Masculinities, persistence, and desistance. Criminology, 51, 661-693.

11.

Census Bureau. (2010). American Fact Finder. Washington, DC: U.S. Department of Commerce.

12.

Cid

Marti

(2012). Turning points and returning points: Understanding the role of family ties in the process of desistance. European Journal of Criminology, 6, 603-620.

13.

Cooper

Smith

E. L.

(2011). Homicide trends in the United States, 1980–2008. Washington, DC: U.S. Department of Justice.

14.

Deitz

Burns

T. R.

(1992). Human agency and the evolutionary dynamics of culture. Acta Sociologica, 35, 187-200.

15.

Farrall

Bowling

(1999). Structuration, human development, and desistance from crime. The British Journal of Criminology, 39, 253-268.

16.

Farrall

Calverley

(2005). Understanding desistance from crime. New York, NY: Open University Press.

17.

Farrall

Sharpe

Hunter

Calverley

(2011). Theorizing structural and individual-level processes in desistance and persistence: Outlining an integrated perspective. Australian & New Zealand Journal of Criminology, 44, 218-234.

18.

Farrington

(2012). Longitudinal and experimental research in criminology. Paper presented at the American Society of Criminology, Chicago, IL.

19.

Ganem

N. M.

Agnew

(2007). Parenthood and adult criminal offending: The importance of relationship quality. Journal of Criminal Justice, 35, 630-643.

20.

Giordano

Cernkovic

S. A.

Rudolph

J. L.

(2002). Gender, crime, and desistance: Toward a theory of cognitive transformation. American Journal of Sociology, 107, 990-1064.

21.

Giordano

Longmore

M. A.

Schroeder

R. D.

Seffrin

P. M.

(2008). A life-course perspective on spirituality and desistance from crime. Criminology, 46, 99-132.

22.

Hanrahan

Gibbs

Zimmerman

S. E.

(2005). Parole and revocation: Perspectives of young adult offenders. The Prison Journal, 85, 251-269.

23.

Haraway

(1997). Modest_Witness@Second_Millennium.FemaleMan©Meets_OncoMouse™: Feminism and technoscience. New York, NY: Routledge.

24.

Katz

(1988). Seductions of crime. New York, NY: Basic Books.

25.

Kentucky Department of Corrections. (2013). Inmate programs, exiting class. Retrieved from http://corrections.ky.gov/depts/AI/KCIW/Pages/InmatePrograms.aspx

26.

King

(2012). Transformative agency and desistance from crime. Criminology & Criminal Justice, 13, 317-335.

27.

Laub

J. H.

Sampson

R. J.

(2001). Understanding desistance from crime. In Crime and Justice: An Annual Review of Research, 18, 1-69.

28.

Laub

J. H.

Sampson

R. J.

(2003). Shared beginnings, divergent lives: Delinquent boys to age 70. Cambridge, MA: Harvard University Press.

29.

Lawler

(2008). Identity: Sociological perspectives. Cambridge, UK: Polity.

30.

Liem

(2013). Homicide offender recidivism: A review of the literature. Aggression and Violent Behavior, 18, 19-25.

31.

Liem

Kunst

M. J. J.

(2013). Is there a recognizable post-incarceration syndrome among released “lifers”? International Journal of Law and Psychiatry, 33, 333-337.

32.

Lopoo

L. M.

Western

(2005). Incarceration and the formation and stability of marital unions. Journal of Marriage and Family, 67, 721-734.

33.

Marsh

(2011). Narrating desistance: Identity change and the 12 step script. Irish Probation Journal, 8, 49-68.

34.

Maruna

(2001). Making good: How ex-convicts reform and rebuild their lives. Washington, DC: American Psychological Association.

35.

Maruna

(2004). Desistance from crime and explanatory style: A new direction in the psychology of reform. Journal of Contemporary Criminal Justice, 20, 184-200.

36.

Maruna

Immarigeon

LeBel

T. P.

(2004). Ex-offender reintegration: Theory and practice. In Maruna

Immarigeon

(Eds.), After crime and punishment: Pathways to offender reintegration (pp. 3-26). Portland, OR: Willan Publishing.

37.

Maruna

LeBel

T. P.

(2002). Welcome home: Examining the reentry court concept from a strengths-based perspective. Western Criminology Review, 4, 91-107.

38.

Maruna

Mann

R. E.

(2006). A fundamental attribution error? Rethinking cognitive distortions. Legal and Criminological Psychology, 11, 155-177.

39.

Mouzelis

N. P.

(2008). Modern and postmodern social theorizing: Bridging the divide. Cambridge, UK: Cambridge University Press.

40.

Neuilly

M. A.

Zgoba

K. M.

Tita

G. E.

Lee

S. S.

(2011). Predicting recidivism in homicide offenders using classification tree analysis. Homicide Studies, 15, 154-176.

41.

Paternoster

Bushway

(2009). Desistance and the “feared self”: Toward an identity theory of criminal desistance. The Journal of Criminal Law and Criminology, 99, 1103-1156.

42.

Petersilia

(2009). When prisoners come home: Parole and prisoner reentry. New York, NY: Oxford University Press.

43.

QSR International. (2010). NVivo Qualitative Data Analysis Software (Version 9). Burlington, MA: Author.

44.

Roberts

A. R.

Zgoba

K. M.

Shahidullah

S. M.

(2007). Recidivism among four types of homicide offenders: An exploratory analysis of 336 homicide offenders in New Jersey. Aggression and Violent Behavior, 12, 493-507.

45.

Sampson

R. J.

Laub

J. H.

(1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press.

46.

Sampson

R. J.

Laub

J. H.

(2003a). Desistance from crime over the life course. In Mortimer

J. T.

Shanahan

(Eds.), Handbook of the life course (pp. 295-310). New York, NY: Kluwer Academic.

47.

Sampson

R. J.

Laub

J. H.

(2003b). Life-course desisters? Trajectories of crime among delinquent boys followed to age 70. Criminology, 41, 301-340.

48.

Sampson

R. J.

Laub

J. H.

(2004). A general age-graded theory of crime: Lessons learned and the future of life-course criminology. In Farrington

D. P.

(Ed.), Advances in theoretical criminology: Testing integrated developmental/life course theories of offending (pp. 165-181). New Brunswick, NJ: Transaction Publishers.

49.

Sampson

R. J.

Laub

J. H.

Wimer

(2006). Does marriage reduce crime? A counterfactual approach to within-individual causal effects. Criminology, 44, 465-508.

50.

Sampson

R. J.

Morenoff

J. D.

(2004). Spatial (dis)advantage and homicide in Chicago neighborhoods. In Goodchild

Janelle

(Eds.), Specially integrated social science (pp. 145-170). New York, NY: Oxford University Press.

51.

Scott

Lyman

(1968). Accounts. American Sociological Review, 33, 46-62.

52.

Stevens

(2012). I am the person now I was always meant to be: Identity reconstruction and narrative reframing in therapeutic community prisons. Criminology & Criminal Justice, 12, 527-547.

53.

Sykes

C. M.

(1958). The society of captives: A study of a maximum security prison. Princeton, NJ: Princeton University Press.

54.

Sykes

C. M.

Matza

(1957). Techniques of neutralization: A theory of delinquency. American Sociological Review, 22, 664-670.

55.

Travis

(2005). But they all come back: Facing the challenges of prisoner reentry. Washington, DC: Urban Institute Press.

56.

Uggen

(2000). Work as a turning point in the life course of criminals: A duration model of age, employment, and recidivism. American Sociological Review, 67, 529-546.

57.

Vaughan

(2007). The internal narrative of desistance. The British Journal of Criminology, 47, 390-404.

58.

Vries

Liem

(2011). Recidivism of juvenile homicide offenders. Behavioral Sciences and the Law, 29, 483-498.

59.

Ward

Brown

(2004). The good lives model and conceptual issues in offender rehabilitation. Psychology, Crime & Law, 103, 243-257.

60.

Western

Pettit

(2010). Incarceration and social inequality. Daedalus, 139, 8-19.