Gender, Crime, and Desistance: Toward a Theory of Cognitive Transformation

TL;DR: In this paper, a symbolic interactionist perspective on desistance is developed as a counterpoint to Sampson and Laub's theory of informal social control, and life history narratives are used to illustrate the perspective.

Abstract: This article analyzes data derived from the first detailed long‐term follow‐up of a sample of serious adolescent female delinquents and similarly situated males. Neither marital attachment nor job stability, factors frequently associated with male desistance from crime, were strongly related to female or male desistance. A symbolic‐interactionist perspective on desistance is developed as a counterpoint to Sampson and Laub’s theory of informal social control, and life history narratives are used to illustrate the perspective. This cognitive theory is generally compatible with a control approach but (a) adds specificity regarding underlying change mechanisms, (b) explains some negative cases, and (c) fits well with life course challenges facing contemporary serious female (and more provisionally male) offenders.

Summary

BACKGROUND

• In a series of analyses that rely on data originally collected by Glueck and Glueck (1968) , Sampson and Laub (1993) documented that childhood predictors (e.g., early family experiences) failed to effectively distinguish male desisters from those who continued to offend in their adult years.
• They found that changes in levels of involvement are tied to variations in "local life circumstances," including living with a wife.

Gender Issues

• While the above studies differ in etiological emphasis, they coalesce around the idea that marriage matters, at least for male offenders (see especially Waite [1995] and Waite and Gallagher [2000] for more general treatments of this axiom).
• Contradictory themes and images coexist about the nature of young women's involvement in crimes and about whether theories designed to explain male delinquency are appropriate for theory-building in this area.
• Thus the authors might expect that (1) marital attachment may be even more critical as an influence on desistance for women than for men, (2) childbearing may represent a more life-changing transition for female than for male offenders, and (3) employment experiences will tend to be less important for women than for men.
• While peer involvement is an important element for both female and male delinquency, female adolescents are more likely to commit delinquent acts with a mixed-gender group, while males are typically accompanied by same-gender companions (Giordano and Cernkovich 1979) .
• Qualitative approaches are especially useful for developing new conceptual categories or lines of inquiry (question three) and can provide a window on mechanisms/processes (question four) that may be more difficult to elucidate using traditional quantitative procedures (Abbott 1992; Maines 1993; Morse 1994 ).

Toward a Theory of Cognitive Transformation

• The attempt to transport a theory typically used to explain juvenile behavior to the adult context works, but in their view it is not a perfect fit.
• Here the authors have the organism as acting and determining its environment.
• At a basic level, one must resonate with, move toward, or select the various catalysts for change.
• First, consistent with Mead's notion of opening the door to certain stimuli and closing it to others, the authors wish to emphasize the actor's own role in latching onto opportunities presented by the broader environment.

Types of Cognitive Transformations

• Conceptually, the authors distinguish four types of intimately related cognitive transformations.
• If, as Mead suggested, cognitions serve as an organizing process, then identity provides a higher level of organization and coherence to one's cognitions.
• The authors fundamental premise is that the various cognitive transformations not only relate to one another (an ideal typical sequence: an overall "readiness" influences receptivity to one or more hooks for change, hooks influence the shift in identity, and identity changes gradually decrease the desirability and salience of the deviant behavior), but they also inspire and direct behavior.
• Nevertheless, the traditional symbolic-interactionist focus on the actor's immediate social world has itself been justifiably criticized for unrealistically bracketing off the broader social forces that give shape and form to these interactions (Perinbanayagam 1985; Stryker 1980) .
• Both the ways in which respondents describe and actually accomplish (or fail to accomplish) life changes depends heavily on the particular repertoires (cognitive, linguistic, behavioral) to which they have access.

Sample

• In 1982, the authors conducted 127 interviews with the entire population of the only state-level institution for delinquent girls in Ohio; a comparable sample was drawn from the populations of three institutions for males ( ).
• An analysis of self-report data indicated that the female and male respondents in the institutional sample were significantly more delinquent not only in comparison to the average neighborhood respondent, but also when compared to the most delinquent youth in the neighborhood survey.
• In 1995, the authors attempted to locate and interview all of the respondents who had participated in the adolescent interview.
• This phase of the project involved extensive phone and street tracking of relatives, previous neighbors and friends, and searches of a variety of databases (e.g., Bureau of Motor Vehicles, the military).

Dependent Variables

• This scale indexes the respondent's report of level of involvement in property and violent crimes, as well as drug and alcohol use during the past year.
• -Another traditional index of desistance is arrest history.
• The authors conducted searches of all the jurisdictions in which respondents were known to have resided and also obtained data about incarceration histories from the Ohio State Department of Rehabilitation and Correction.
• It is quite possible that the absence of arrests represents for some merely a temporary lull, rather than a permanent shift to a more conforming status.
• In some cases, the authors may have had incomplete information regarding respondents' movements over the 13-year interval between data collection periods, and some individuals may have offended in areas other than where they resided.

Independent Variables

• Consistent with prior work (e.g., Sampson and Laub 1993) , the authors also examine the impact of time-1 predictors.
• The key adult social control variables include a measure of job stability, attachment to spouse, and attachment to child(ren).
• Respondents were categorized as having high job stability (and coded "2") if they were currently employed full-time and had a low likelihood of quitting/losing the job.

Eliciting and Analyzing the Qualitative Data

• The authors elicited open-ended life history narratives from 97 women and 83 men in the sample.
• Using the first two categories of respondent narratives, raters then attempted to extract from the life history account the respondent's view of the most important catalyst for changes they had made.
• Model 1 shows that gender is also a significant predictor-men self-report more criminal involvement at the time of the adult follow-up.
• While it appears that a large percentage of the Gluecks' respondents were both married and held a full-time job, the data shown in table 2 document the various combinations of marital and employment circumstances of the respondents in their sample.
• These percentage data, like the classification of various hooks for change, were thus derived from a somewhat subjective process and should be interpreted with caution.

FINDINGS: QUALITATIVE DATA

• The regression procedure, while dedicated to an exploration of variability, is rather limited in its ability to convey the range of adult life circumstances the authors actually encountered in the process of completing the followup.
• The narratives also reveal central tendencies that constitute an important background to their discussion.
• 22 The data described in table 2 are generally indicative of the respondents' lack of access to traditional markers of respectability; however, the following interviewer's description of one respondent's housing circumstances makes more concrete the kinds of contexts within which many of the respondents' change attempts have taken place: Quite a neighborhood.
• All the buildings were just destroyed-incredible squalor-and her house wasn't at all like that.
• Comparing her life to that of the friends she listed during the 1982 interview, Angie sees herself as doing much better "because I'm more settled down I always knew where I wanted to be and I'm basically there.".

Openness to Change

• All the women and men in their study experienced a highly problematic adolescence, and respondents' later lives are often characterized by an array of legal and other problems as they have matured into adulthood.
• And then the times I wasn't locked up, I was running with criminals.
• Other respondents adapt the basic outline of a change theme, but their stories lack depth and definition.
• This is of particular interest because as Linde (1993) notes, as stories unfold, both narrators and listeners share a desire for consistency and coherence.
• Respondent 6.-"I was a wild child" (31-year-old black female, never married, unemployed).

Hooks for Change

• Consistent with the quantitative findings and their discussion to this point, respondents in this sample, whether male or female, were very unlikely to build a story of change around the development of a rewarding career, and only a few focus heavily on stable employment.
• Two hooks that were more prominent link to experiences with formal organizational settings (prison or treatment and religion), and two relate to intimate networks (children and marital/romantic partner).
• Obviously, the authors included attention to the family in their quantitative analyses, where they determined that levels of attachment (to children and partner) were not strongly related to desistance.
• Thus the narratives are useful, not only because they reveal different hooks for change such as religion, but because they allow us to examine familiar variables like children and marriage using a different theoretical lens.
• This adds to their understanding of mechanisms of change, helps to explain some of control theory's negative cases (e.g., individuals with high attachment to a spouse who nevertheless persist in offending), and brings to light gender differences that were not apparent in their analysis of the quantitative data.

Prison/Treatment

• In the aggregate, prison and even most treatment strategies do not fare well as catalysts for lasting change.
• The respondent ties her change in direction to the prison experience, but she has focused heavily on her own shift in attitude, rather than actions of prison staff or a particular type of treatment program.
• I'm closer to my family and friends now than ever, and I do nothing spontaneous.
• Because I got off drugs and started meetin' people that didn't use drugs.
• It didn't matter what color you were, what was your career background, your home status didn't matter, didn't discriminate" (31-year-old black female, never married, full-time employment as a nursing assistant earning $20,800 a year).

Religion

• A large number of respondents within the sample make at least some reference to God, and women were somewhat more likely to consider religious experiences important catalysts for changes they have made (13% of the women as contrasted with 7% of the men).
• He takes me, that's where I'll go" (31-year-old white female, never married, currently unemployed).
• The following interviewer notes regarding respondent 19 (discussed below) illustrate the dominating quality of some of the lifestyles that become oriented around religious faith.
• She feels guilty if she's doing anything besides talking to Jesus and Mary.
• Next the authors apply this more "conditional-on-cognitive-transformations" perspective as they examine two hooks that figure even more prominently in the change stories, namely, children and the marital/intimate partner.

Children

• As noted in their earlier background discussion, Graham and Bowling (1996) found that for women in their British sample desistance often occurred abruptly and was tied directly to childbearing.
• Motherhood creates possibilities for a change in self-conception, but the internalization of this new status is far from automatic.
• One way respondents make this shift involves a reconfiguration of the meaning and impact of "shame.".
• I didn't want to do that to my kids" (29-year-old black male, married two times to the same person, currently divorced, employed full-time as computer operator earning $20,321 a year).
• Respondents who had not yet forged any meaningful connection between their behavior and their child's well-being contribute further variability regarding child effects.

The Marital Relationship

• Logically, marital partners could prove very powerful catalysts for changes in life direction.
• Thus, even while women describe how their marriages have been influential in the social bonding and investment sense, the narratives provide a window on the initial movement into this "conventionalizing" relationship form.
• I just figured that the authors could make the best of it" (30-year-old white male, cohabiting, employed full-time as masonry worker earning $27,100 a year).
• Respondent 34.-"Yeah, cause I mean, all the other guys I was ever with, was always drinking and drugging and drinking and drugs and that is all I knew.
• In other analyses (see, e.g., Giordano, Cernkovich, and Lowery 2001) , the authors point out a variety of ways in which the women offenders' adult lives differ significantly from those of their male counterparts.

Suggestions for Further Theory Building and Integration

• The authors have concentrated primarily on the women respondents in this analysis, but conjectured that "a theory of cognitive transformation" may also be a useful framework for understanding how it is that men manage to desist from criminal activity.
• In addition, male respondents, like their female counterparts, were frequently heavily involved in criminal and drug cultures that seemed to be more encapsulating and limiting of life chances-thus a high level of individual motivation or "up-front" commitment would seem to be required for successful and long-lasting change.
• Additional research could explore why some individuals who appear to have experienced significant cognitive shifts are nevertheless unable to move their behaviors into good alignment with them.
• These individually and socially structured differences in motivation and preference, the processes of interaction and communication that solidify them, and the gradual redefinitions that result are arguably as important as the "stake" itself.

Figures

TABLE 1 Adult Self-Report of Criminal Involvement Regressed on Adolescent and Adult Variables

Fig. 1.—Mechanisms associated with desistance

TABLE 2 Access to the Traditional Respectability Package (Marriage and Full-Time Employment) and %Desisting in Each Category

TABLE 3 Adult Self-Report of Criminal Involvement Regressed on Adolescent and Adult Variables

Full text

Bowling Green State University Bowling Green State University
ScholarWorks@BGSU ScholarWorks@BGSU
Sociology Faculty Publications Sociology
1-2002
Gender, Crime, and Desistance: Toward a Theory of Cognitive Gender, Crime, and Desistance: Toward a Theory of Cognitive
Transformation Transformation
Peggy C. Giordano
Bowling Green State University
, pgiorda@bgsu.edu
Stephen A. Cernkovich
Jennifer L. Rudolph
Follow this and additional works at: https://scholarworks.bgsu.edu/soc_pub
Part of the Sociology Commons
Repository Citation Repository Citation
Giordano, Peggy C.; Cernkovich, Stephen A.; and Rudolph, Jennifer L., "Gender, Crime, and Desistance:
Toward a Theory of Cognitive Transformation" (2002).
Sociology Faculty Publications
. 1.
https://scholarworks.bgsu.edu/soc_pub/1
This Article is brought to you for free and open access by the Sociology at ScholarWorks@BGSU. It has been
accepted for inclusion in Sociology Faculty Publications by an authorized administrator of ScholarWorks@BGSU.

990 AJS Volume 107 Number 4 (January 2002): 990–1064
䉷 2002 by The University of Chicago. All rights reserved.
0002-9602/2002/10704-0004$10.00
Gender, Crime, and Desistance: Toward a
Theory of Cognitive Transformation
1
Peggy C. Giordano, Stephen A. Cernkovich, and Jennifer L. Rudolph
Bowling Green State University
This article analyzes data derived from the first detailed long-term
follow-up of a sample of serious adolescent female delinquents and
similarly situated males. Neither marital attachment nor job sta-
bility, factors frequently associated with male desistance from crime,
were strongly related to female or male desistance. A symbolic-
interactionist perspective on desistance is developed as a counter-
point to Sampson and Laub’s theory of informal social control, and
life history narratives are used to illustrate the perspective. This
cognitive theory is generally compatible with a control approach
but (a) adds specificity regarding underlying change mechanisms,
(b) explains some negative cases, and (c) fits well with life course
challenges facing contemporary serious female (and more provi-
sionally male) offenders.
In a series of recent analyses, Robert Sampson and John Laub highlight
the importance of marital attachment and job stability as key factors
associated with desistance from crime (Laub, Nagin, and Sampson 1998;
Laub and Sampson 1993; Sampson and Laub 1993). While the delinquents
they studied were more likely than others to continue to offend as adults,
there was considerable variability in the success of their adult transitions
and in the timing of movement away from a criminal lifestyle. Sampson
and Laub develop a social control explanation that emphasizes the gradual
1
This research was supported by grants from the National Institute of Mental Health
MH29095 and MH46410. The authors wish to acknowledge H. Theodore Groat, a
coinvestigator on this project, who contributed greatly to the collection and analyses
of these data and the conceptual framework developed in this article. We have also
benefited from the comments and suggestions of Stephen Demuth, Michael Giordano,
Monica Longmore, Charles McCaghy, Wendy Manning, Robert J. Sampson, and the
AJS reviewers. We especially appreciate John Laub’s careful review and critique of
several drafts of the paper. Address all correspondence to Peggy C. Giordano, De-
partment of Sociology, Bowling Green State University, Bowling Green, Ohio 43403.
E-mail: pgiorda@bgnet.bgsu.edu
This content downloaded from 129.1.62.221 on Thu, 10 Oct 2013 19:22:42 PM
All use subject to JSTOR Terms and Conditions

Desistance
991
buildup of investments that tend to accrue in the presence of strong bonds
of attachment (“the good marriage effect”) and steady employment. This
focus on variability and on the impact of adult social bonds also adds to
the broader intellectual tradition that emphasizes the ways in which so-
cialization and development continue across the full range of the indi-
vidual life course (Claussen 1993; Elder 1998; Josselson 1996; Shanahan
2000).
A potential limitation of this important body of work is that the sample
on which the analyses were based was composed entirely of white male
offenders who matured into adulthood during the 1950s. Thus it is not
clear whether the findings described (or the theory that derives from them)
effectively capture the experiences of female or minority delinquents or,
more generally, offenders coming of age within the context of a more
contemporary social and economic landscape. We contribute to the lit-
erature on desistance processes by presenting results of the first detailed
long-term follow-up study of a cohort of serious adolescent female of-
fenders and a similarly situated male comparison group. We collected
both quantitative and qualitative data at the adult follow-up and have
found both “ways of knowing” (Polkinghorne 1988) useful in different
respects. In this article, we first examine the quantitative data to determine
whether factors such as marital attachment or job stability are associated
with female as well as male desistance from criminal activity. Because
our sample contains a significant percentage of minority respondents, and
follow-up data were collected in the mid-1990s, we can also consider how
race/ethnicity and (indirectly) historical changes further complicate the
picture.
We then turn to the relatively unstructured life history narratives we
elicited from respondents during the follow-up. Many of these narratives
exceed 100 pages in length. They are useful not only as an aid in inter-
preting the quantitative findings, but they also provide a close-in per-
spective on mechanisms through which actors indicate that changes in
life direction have been accomplished. It is primarily through our analyses
of these narratives that we developed a somewhat different perspective
on desistance. Our provisional theory centers on the cognitive shifts that
frequently occur as an integral part of the desistance process. For purposes
of exposition, we contrast this “theory of cognitive transformation” with
the social control framework Sampson and Laub and other scholars have
emphasized. While our ideas are not fundamentally incompatible with a
social control approach, we cover somewhat different conceptual terrain.
Social control theory emphasizes the ways in which a close marital
bond or stable job gradually exert a constraining influence on behavior
as—over a period of time—actors build up higher levels of commitment
(capital) via the traditional institutional frameworks of family and work.
This content downloaded from 129.1.62.221 on Thu, 10 Oct 2013 19:22:42 PM
All use subject to JSTOR Terms and Conditions

American Journal of Sociology
992
Social control is thus essentially a theory of constraint that is focused on
the long haul.
In our view, this provides an important but incomplete accounting of
change processes, because the perspective tends to bracket off the “up
front” work accomplished by actors themselves—as they make initial
moves toward, help to craft, and work to sustain a different way of life.
We wish to emphasize the actor’s own role in creatively and selectively
appropriating elements in the environment (we will refer to these elements
as “hooks for change”), including, but not limited to, such positive influ-
ences as a spouse. We argue that these elements will serve well as catalysts
for lasting change when they energize rather fundamental shifts in identity
and changes in the meaning and desirability of deviant/criminal behavior
itself. The latter notion contrasts with a basic assumption of control the-
ory—that an individual’s motivation or proclivity to deviate can be con-
sidered a constant, while it is the degree of external and internal control
that varies considerably (e.g., across individuals or across the period en-
compassed by an individual life course).
In emphasizing cognitive and identity transformations and the actor’s
own role in the transformation process, our perspective seems most com-
patible with the basic tenets of symbolic interaction. This more “agentic”
view of desistance balances some of the exteriority and constraint as-
sumptions implicit in a control approach. It is useful for (1) highlighting
the important period when actors make initial attempts to veer off a
deviant pathway (when, almost by definition, various forms of capital
have not had much chance to accumulate); (2) accommodating the ob-
servation that quite a few individuals exposed to prosocial experiences
like those associated with marriage or job opportunities fail to take ad-
vantage of them (they persist in offending anyway); and (3) focusing on
cognitive changes, rather than a small set of predictors. This provides a
measure of conceptual flexibility. That is, it takes into consideration in-
dividuals who manage to change their life direction, even in the absence
of traditional frameworks of support and resources like those provided
by a spouse or good job.
We conclude that this symbolic-interactionist perspective can in most
respects be integrated with social control notions. Such an integration
provides a more complete conceptual tool kit for understanding changes
in life direction than either perspective on its own. However, there are
also significant variations in the relative salience of these pro-
cesses—within samples, across different types of samples, or between dif-
ferent historical contexts. We developed our ideas about the importance
of cognitive processes and the role of “agentic moves” primarily through
analyses of one set of contemporary qualitative data. A preliminary dis-
cussion of these notions (Giordano, Cernkovich, and Rudolph 1997) was,
This content downloaded from 129.1.62.221 on Thu, 10 Oct 2013 19:22:42 PM
All use subject to JSTOR Terms and Conditions

Desistance
993
in the venerable tradition of symbolic interactionism, largely free of any
consideration of the broader structural underpinnings of the collected and
analyzed material. But while we continue to focus primarily on microlevel
processes, we have increasingly recognized that the form and content of
these narratives intimately connect to the social addresses of our respon-
dents. Thus, their discourse (and inferentially the character of their change
efforts) necessarily draws on themes that are within the reach of highly
marginal women and men attempting to navigate the specific conditions
and challenges of a late-20th-century environment.
For individuals, samples, or eras characterized by greater advantage,
perhaps the kinds of agentic moves we will emphasize may not have been
necessary (when things really do just tend to fall into place). In contrast,
our respondents’ frequent descriptions of efforts to, in effect, pull them-
selves up by their own cognitive “bootstraps” likely connect to the reality
that society has provided them with little in the way of raw materials
(i.e., structure). Feminist perspectives on agency and the broader literature
on structure-agency connections are useful correctives in this regard. We
believe our research contributes to these traditions as well; feminist the-
ories increasingly take into account the intersection or confluence of var-
ious kinds of disadvantage. The literature, however, contains a relatively
small number of longitudinal studies of women so positioned. Similarly,
the sociological literature contains numerous theoretical or formal dis-
cussions of the place of agency but fewer empirical investigations that
work directly with this elusive but important construct.
BACKGROUND
In a series of analyses that rely on data originally collected by Glueck
and Glueck (1968), Sampson and Laub (1993) documented that childhood
predictors (e.g., early family experiences) failed to effectively distinguish
male desisters from those who continued to offend in their adult years.
However, variables indexing the strength of adult social bonds (notably,
job stability and strong bonds of attachment to a partner) were found to
be important. In a recent dynamic analysis, Laub et al. (1998) demon-
strated that the good marriage effect tends to be gradual and cumulative
rather than abrupt, and they further articulated a control theory expla-
nation of these findings. Their analyses also lend support to the notion
that the differences in marital quality of desisters cannot be explained
entirely by initial differences between groups of men (i.e., by selection
effects), although this issue has been the source of controversy (Gottfredson
and Hirschi 1990). Farrington and West (1995) also concluded that while
the high-risk London boys they followed into adulthood were more likely
This content downloaded from 129.1.62.221 on Thu, 10 Oct 2013 19:22:42 PM
All use subject to JSTOR Terms and Conditions

Frequently asked questions

Q1. What contributions have the authors mentioned in the paper "Gender, crime, and desistance: toward a theory of cognitive transformation" ?

In this paper, a longitudinal study of young women offenders employed a similar inclusion criterion ; thus Laub and Sampson 's analyses of the Gluecks data and their theory of informal social control have provided a useful contrast with their findings and theoretical perspective. 

Q2. What have the authors stated for future works in "Gender, crime, and desistance: toward a theory of cognitive transformation" ?

Such young people are an important subgroup to study, however, because the individual and social costs of their actions are particularly high. Finally, the authors argued that this more self-conscious perspective is consistent with the greater freedom of movement and choice-making possibilities characterizing adulthood. Thus, future theory and research should add attention to emotions as they affect behavioral change directly or, indirectly, as they influence the nature and timing of cognitive shifts. The authors agree completely with the key premise that highly invested actors will develop a strong stake in conformity and will not wish to jeopardize what they have accumulated by reverting to criminal activity. 

Q3. What were the only buildings that seemed to have any hope for any residents in them?

The homeless shelter and the Catholic Mission on the end of her street, those were the only buildings that seemed to have any, even hope for any residents in them. 

Q4. What is the first example that supports the idea of a good marriage effect?

The first example that supports the idea of a good marriage effect reads like a Cinderella story, in that the male partner is seen as instrumental in directing the respondent away from a very negative environment. 

Q5. What are the three types of stories that are available within the genre of the personal narrative?

Gergen and Gergen (1986) note that there are only three types of stories available within the genre of the personal narrative, that is, those encompassing progressive, regressive, or stability themes. 

Q6. What is the role of hooks for change in the transformation process?

Hooks for change can provide an important opening in the direction of a new identity and concrete reinforcement during all phases of the transformation process. 

Q7. What is the role of the actor in selecting others who have the potential to be good influences?

Particularly as the authors focus on adult friendships and romantic liasons, the individual has an important role in selecting others who have the potential to be good influences, while “knifing off” undesirable companions (Emirbayer and Goodwin 1994). 

Q8. How did they study the patterning of offense involvement of recently convicted felons?

Osgood, and Marshall (1995) examined the patterning of offense involvement of recently convicted felons, using a shorter window of time, that is, month-to-month changes in the year following their release. 

Q9. What is the importance of elucidating the perspectives of these young women offenders?

Eliciting the perspectives of these young women offenders seems particularly important in light of the contradictory and incomplete images that can be derived from the existing literature. 

Q10. Why did the respondents not believe that a half-hearted approach to X or Y?

Due to their extremely marginal positions at the outset, such respondents may not believe (perhaps correctly) that a half-hearted approach to X or Y will be sufficient as a bridge to lasting change. 

Q11. What is the way to study hooks for change?

Additional research on specific hooks for change would ideally be carried out using a variety of samples, including data that contain a sufficient number of serious male and female offenders to allow for meaningful analysis.