CLTL Home About CLTL Programs Key Issues Resources Outreach
Site Map
Site Index
Contact Us
Discussion Forum
Flash Version


"Combining Bibliotherapy and Positive Role Modeling as an Alternative to Incarceration"

Journal of Offender Rehabilitation, Vol. 28 (1/2), 1998, Pp. 127-139. Copyright 1998 by The Haworth Press, Inc., Binghamton, NY.
Indiana University, Indianapolis
University of Massachusetts, Dartmouth

ABSTRACT: This paper presents an evaluation of Changing Lives Through Literature, a program designed as an alternative to jail. The evaluation followed-up the first 32 men to complete the program as well as a matched comparison group of 40 probationers. Follow-up data indicate a reconviction rate of 18.75% in the study group, compared with 45% in the comparison group. In addition, information is presented from interviews with program participants. We consider the impact of the use of bibliotherapy, coupled with the provision of positive role models, on recidivism and the value of fitting such a program within a comprehensive system of punishments. [Article copies available for a fee from The Haworth Document Delivery Service: 1-800-HAWORTH. E-mail address:]

KEYWORDS: Bibliotherapy, alternative sanctions, recidivism, high-risk offenders, probation

Recidivists plague the criminal justice system, as indicated by the report that 61% of state prison inmates in 1991 previously had been incarcerated (Bureau of Justice Statistics, 1994a). In a follow up of 11,347 young adult parolees, 93% of those with six or more previous adult arrests had been rearrested within six years (Bureau of Justice Statistics, 1987). Imprisonment seems to have little effect upon recidivism while exacting a considerable cost to the public. Irwin and Austin (1994) note that currently we spend about $25 billion each year on corrections in this country.

Compounding this failure is the fact that nationally, jail and prison populations are at a record high and growing. At the end of 1993 there were 351 prison inmates per 100,000 residents in the United States, up from 139 per 100,000 in 1980 (Bureau of Justice Statistics, 1994b). The growth rate was similar for jails. In 1994, there were 188 jail inmates per 100,000 U.S. residents, up from 96 per 100,000 residents in 1983 (Bureau of Justice Statistics, 1995). One result of these increased imprisonment rates is that prisons and jails are operating at levels near or above capacity. At the end of 1993, 42 states and the federal prison system were operating prisons above capacity (Bureau of Justice Statistics, 1994b). On average, U.S. jails were filled to 97% of capacity (Bureau of Justice Statistics, 1995).

Overcrowding in prisons and jails has led to a markedly increased emphasis on alternative sentencing. Most discussions of alternative sentencing focus on punitive models-home detention, community service, shock incarceration, boot camps, and electronic monitoring. A program currently operating in southeastern Massachusetts is significantly different, however, in that its primary purpose is not punitive (Kane and Waxler, 1995), but focuses instead on individual growth. The executive summary of the Changing Lives Through Literature program proposal identified three factors which should, from the program founders' perspective, reduce recidivism: the literature content, growth of self-esteem, and the reintegration assistance.


The use of literature as a therapeutic tool has a long history, dating back to the Ancient Greeks' use of tragedy. More recently, such use of literature has been referred to as bibliotherapy, a term which entered the lexicon early in the twentieth-century (Rubin, 1978a). R. J. Rubin defines bibliotherapy as

"a program of activity based on the interactive processes of the use of print and nonprint material whether imaginative or informational, facilitated by a librarian or other professional, to achieve insight into normal development or to effect changes in emotionally disturbed behavior." (1978b, p. xi)

There are two ways in which literature may offer insight. The act of reading, itself, may provide a catharsis, or the lesson may arise during "guided dialogue about the material" (Hynes and Hynes-Berry, 1986, p. 11). The latter is referred to as interactive and bibliotherapy, as defined above, relies upon this interaction.

Initial use of bibliotherapy focused on institutionalized individuals: mental patients and prisoners. As bibliotherapy moved into clinical (noninstitutional) and community settings, it became useful to classify and define the variations. Rubin suggests three categories: institutional, clinical, and developmental. Developmental bibliotherapy is described as an interactive group of "normal" people who may be in crisis, led by a teacher who focuses discussion on "reactions and insights" (Rubin, 1978a, p. 7). The major populations served by developmental bibliotherapy include adolescents and children, senior citizens, handicapped persons, terminal patients, and public library patrons (Hynes and Hynes-Berry, 1986, pp. 15-17). Criminals have historically been served in correctional institutions. Bibliotherapy has been used, for example, to help adult prisoners adjust to being locked up (Miller, 1993).

Sheridan et al. (1984) used bibliotherapy within a school setting as a means of preventing behavior problems in youths from families experiencing major changes. The results of their project failed to show any behavioral effects of participation in the therapy, but the youths reported positive attitudes about the intervention. Miller (1993) used literature to improve the self-esteem of females identified as high risk for such problems as truancy, substance abuse, dropping out, and teen pregnancy. In this project, a group of teenage girls read and discussed novels which focused on women's issues. Miller found that the youths were able to relate to the characters in the books and, as a result of the discussions, evidenced more positive self-esteem. Miller (1994) also experienced success in using literature to enhance the self-esteem of youths with learning and behavior problems.

Self-esteem has been shown to be related to school performance and reading ability, and in turn, self-esteem has been shown to have an impact on the likelihood of delinquent involvement (Lawrence, 1985). Gold (1978) suggested that experiencing failure in school lowers self-esteem, which, in turn, increases the propensity for delinquency. According to Gold, "delinquent behavior.. is a defense against self-derogation" (1978, p. 293). Based on this argument, one might expect that enhancing self-esteem may reduce tendencies toward delinquent conduct.


The Changing Lives Through Literature program provides the court with an intermediate sanction for high-risk offenders. The standards for program admission are not specifically stated, but based upon general standards and the professional judgments of the judge and probation officer. To be eligible for the program, an offender must have a "fairly" long record (the average number of prior convictions for the first two groups of men participating in the program was 18.4 per person); pending cases, such that the conviction would carry a high likelihood of incarceration; the ability to read standard novels and short stories (approximately an 8th grade reading level); and stated willingness to participate. Participation in the program is in lieu of a jail sentence; electing not to participate, at any point in time, results in imposition of a sentence by the Court.

The program includes intensive probation, biweekly discussion groups focusing on contemporary literature, and pre-employment/job placement services to help participants plan and carry out successful reintegration into society. A novel or short story is assigned for each seminar meeting. Readings have included Banks's The Affliction, Boyle's short story "Greasy Lake," Dickey's Deliverance, Ellison's Invisible Man, Hemingway's Old Man and the Sea, London's Sea Wolf, Mailer's An American Dream and Morrison's The Bluest Eye. Two-hour meetings occur every other week. The participants are required to read the assignment and must engage in the group discussion. The discussion explores issues such as violence, masculinity, and individual identity as presented in the readings. The discussion topics mirror themes the participants may be dealing with in their own lives. Self-esteem, the program principles maintain, would be built by enhancing the participants' communication skills, sharpening their analytical skills, and providing them with a forum for discussing personal concerns without having to recount personal experiences.

The seminars are held in the Dean's Conference Room at the University of Massachusetts at Dartmouth. The location is important as it is removed from the environment of the criminal justice system. Use of the conference room, where participants sit around a table and engage each other, also may contribute to a sense of importance on the part of the offender. At the conclusion of the bibliotherapeutic portion of the program, area businessmen meet with the participants to share their own life stories. This activity is designed to illustrate real alternatives to the participants. Each businessman relates the ways in which he was ultimately successful.


In April 1993, we began an evaluation of the Changing Lives Through Literature program. At that time, the program had been operating for two years, had graduated five classes, and had 40 graduates. We elected to focus on the first four classes (N = 32), because there had been only enough time since course completion to determine recidivism with any accuracy for those participants. A sixth class was ongoing, enabling us to attend classes and observe the students' progress. Our evaluation was twofold. The first part was quantitative in nature, consisting of the collection and analysis of data on the program participants, such as age, prior record, and risk factor, and the comparison of those persons with a matched offender group. The second part of the evaluation was qualitative in nature, consisting of individual interviews with program participants. Our analysis, observations, conclusions, and recommendations follow.


We constructed a comparison group for the men in the first four groups through a three-step process. From the list of active probationers in the New Bedford District Court, we compiled a list of all of the potential matches for each subject in the study groups. For each man in the first four groups, we identified those male probationers who were classified at the same supervision level and placed on probation within six months of the start date of the bibliotherapy group in which the subject took part. We then obtained the completed Risk/Need Assessments for each of the potential matches.

The second step involved narrowing down the lists of potential matches by retaining only those probationers who were similar to the study group members in terms of individual risk category scores, total risk scores, and race. The third step involved obtaining the complete criminal history information on each of the potential matches and further narrowing down the selection based on age. We ended up with a comparison group of 40 men and a study group of 32 men (the entire first four groups).

The two groups are similar in terms of age and total risk. They also are similar in terms of most of the offender risk categories that make up the total risk score. As indicated in Table 1, there were three significant differences between these two groups. The members of the study group had a higher likelihood of being rated as more severe in terms of their prior record than did those in the comparison group. In addition, the study group members were significantly more likely to have been convicted of violent offenses than were members of the comparison group.

These differences are due to selection bias. Ideally, the appropriate comparison group for this evaluation would have been a group of matched offenders sentenced to the House of Correction, since the evaluated program is an alternative to incarceration, The offenders in the study group would have been committed to the House of Correction had they not been selected for the program. Accordingly, it is not surprising that in terms of prior record, particularly with regard to violent offending, the study group would be significantly different from a group of offenders sentenced to probation. Given this, if the study group commits fewer crimes after the program than the comparison group, the results will be even more impressive.

Table 1.

The other significant difference between the two groups has to do with the offender risk category pertaining to attitude. Offenders are scored as to whether or not they are "responsive/committed to suggested supervision plan." Differences between the two groups on this characteristic again point to selection bias. The members of the study group were selected because they were ready to make changes in their lives. They had the option of not participating in the program and being sentenced to the House of Correction. As such, we would expect that the members of the study group might be more likely to score higher on this dimension (scoring higher representing a more positive attitude) than members of the comparison group. This type of attitude will be discussed further below.

Members of the study group were followed from the beginning date for the group in which they participated. Members of the comparison group were followed from the beginning date of their probation. Criminal records were checked on each of these persons through the end of June 1993. In terms of convictions, six of the 32 men in the study group (18.8%) were convicted on new charges after completing the program. In the comparison group, 18 of the 40 men (45%) were convicted on new charges during the follow-up period. In terms of number of convictions per man, in the study group those men who were convicted on new charges averaged three convictions per person (a total of 18 convictions for the six men). In the comparison group, the average number of new convictions was 3.4 (62 new offenses among the 18 men).

We also looked at the subset of new offenses that can be classified as violent offenses. For the study group, only one man out of the 32 (3.1 %) was convicted of violent offenses during the follow-up period and he was convicted of two violent offenses. In the comparison group, five of the 40 amen (12.5%) were convicted on violent offenses in the same period. The average number of violent convictions per man in the comparison group was 1.8 (nine offenses among the five men). Thus, while the men in the comparison group were more likely to be convicted on new offenses than those in the study group, the number of convictions per man was not significantly different between the two groups.

During the evaluation, detailed information from the offender profiles of the men in the study group was compiled. This information was coded in such a way to allow for an examination and prediction of which men were most likely to fail (i.e., be convicted of a new offense) after participating in the program. In addition to the factors described above, we also considered race, the number of residence changes in the previous year, whether the person was employed, the nature of the man's family ties, the level of educational attainment, and the presence of alcohol and/or drug problems either past or present. Upon examining zero-order bivariate correlations between these factors and the number of convictions after the program (and the number of violent convictions after the program), we found three factors that seem to predict failure.

The most significant factor was the number of moves in the previous year. There was a significant, strong correlation between the number of moves in the previous twelve months (before the program) and the number of convictions after the program (r = .59). There also was a significant, strong correlation between number of moves and the number of violent convictions after the program (r = .65). Residential instability seems to be a factor that predicts those men who are likely to reoffend after participation in the program. Other significant correlations were found between the presence of a current alcohol problem and the number of violent convictions after the program (r = .47), as well as the presence of a current drug problem and the number of violent convictions after the program (r = .56). These results can help inform future decisions about whom to include in the program.

Participant Interviews

The participants who agreed to be interviewed as part of the evaluation spoke highly of the program. They all claimed that the program made what was likely to be a long-term impact on their lives. The participants were most impressed with the participation of the judge. His apparent respect for them in court as well as his participation as a member of the discussion group made them feel important - some of them for the first time. They also reported that the professor clearly cared about the men - this made a difference to them. The probation officer projected a level of commitment towards the men that was not always expected by the probationers. Participants also found the specific reading material interesting, finding they could relate to the issues raised in the books and in class. Unfortunately, the survey instrument suffered in terms of the information elicited about the impact of bibliotherapy on the lives of the men and as a result, we cannot point clearly to any specific effects of the bibliotherapy itself.

This program serves men who have long histories of criminal convictions. The fact that less than 20% of the participants return to crime after being so heavily involved in criminal activity is notable. A recidivism rate of less than 20% is quite impressive and certainly not a common finding in evaluations of alternative sanctions in adult corrections. Yet, the question of whether the success (or, perhaps the lack of failure) of these men is the result of the Changing Lives Through Literature program cannot be definitively answered in this evaluation. The majority of the participants in the study group seemed to be at a point in their lives when they were ready to make a change to a noncriminal lifestyle. Sampson and Laub (1993) provide evidence that adult offenders will move out of a criminal lifestyle if presented with the right motivations (e.g., employment, family ties). The men in the study group were clearly selected for these reasons. So one can argue that even without the bibliotherapeutic intervention, these men may have been successful at turning their lives around. Since the men were handpicked, it is impossible to separate out such confounding influences.

Results of our interviews suggest, however, that incarceration (the alternative to participation in this program) would have had deleterious effects on these men. One man in the study group commented that he never would have been successful had he been forced to give up the first real job he had ever obtained. He would have lost this job if he had been incarcerated. Upon release from the House of Correction, he would have been unemployed once again and may have turned back to using and dealing drugs. Instead, he participated in the literature program and was able to maintain his job and fulfill his responsibilities. Perhaps one of the most positive aspects of the program is that it facilitates responsible and noncriminal behavior for those offenders who are so inclined. This argument, of course, suggests that the program will not be effective with all offenders.


In addition to selection effects described above, there are other complicating factors which make it impossible for us to isolate the impact of the bibliotherapy on the subsequent offending behavior of the participants. This program, as designed and implemented, involved more than a simple bibliotherapeutic intervention. The positive role modeling provided by the presentations by businessmen may also influence the outcomes. We also cannot determine the extent to which variation in the outcomes can be attributed to the effect of the facilitators in this program, which will have implications for the success of replication efforts in other jurisdictions.

There are other issues to consider here. While it is true that for many adults the threat (or even the experience) of incarceration serves as a deterrent to future criminal behavior, for others it is not enough. As Meisenhelder (1977) suggests, equally important is the "offender's emerging desire to settle down" (p. 325). Once this desire exists, there are steps that need to occur to facilitate desistance from criminality. According to Meisenhelder, successful desistance requires the formation of social bonds (i.e., a good job, positive interpersonal relationships). This is in line with the theoretical position of Sampson and Laub (1993). In their theory on how adult social bonds lead to changes in criminal behavior, they stress:

"Adult social ties are important insofar as they create interdependent systems of obligation and restraint that impose significant costs for translating criminal propensities into action. It is unrealistic to expect that adults with a criminal background (or low self-control) can be wholly transformed by institutions (marriage or work), or that such institutions are even capable of imposing direct controls like surveillance. Nevertheless, we believe that adults, regardless of delinquent background, will he inhibited from committing crime to the extent that they have social capital invested in their work and family lives." (1993, p. 141)

In their reanalysis of the Glueck's Unraveling Juvenile Delinquency data, Sampson and Laub support the position that social capital is important. 1--or instance, in a sample of delinquents, job stability and attachment to a spouse had inhibiting effects on involvement in crime in the adult years. In addition, in the period from age 25 to age 32, occupational commitment, defined as "high aspirations and efforts to advance educationally and occupationally" (1993, p. 147), also contributed to desistance from offending.

The program may operate indirectly to affect desistance from offending. The majority of (lie participants in the study group seemed to be at a point in their lives when they were ready to make a change to a noncriminal lifestyle. They were clearly selected for this reason. Many of the men were enjoying good relationships with their spouses and other family members. Most of the men in the study group also were gainfully employed. Incarceration, which was the alternative to participation in this program, would have been a setback for these men. Participation in the Changing Lives Through Literature program, however, facilitated the realization of job stability. Also, any strains on family relationships that may have resulted from a period of incarceration were avoided. While the program cannot be credited with developing these social bonds, it was instrumental in avoiding any attenuation that may have resulted from incarceration.

At the same time, the program may have direct effects on the likelihood of future offending behavior. First, many of the men were motivated to seek further education upon completion of the literature classes. Also, each of the men was responsible for coming up with a life plan that set goals (e.g., educational as well as occupational) for the period immediately following completion of the program. This is an element of "occupational commitment" that has been shown by Sampson and Laub (1993) to contribute to desistence.

Second, to the extent that the, program served to enhance the participants' self-esteem, we might expect to find a corresponding reduction in the likelihood of criminality. Building self-esteem is a basic goal of bibliotherapeutic interventions (Miller, 1994). This evaluation does not provide any indicators of changes in self-esteem that may have occurred. Yet, it was clear from the interviews with the participants that many of the men felt empowered by the course. The participation of the judge led many of the men to feel more important. In addition, the end of the course was marked by a graduation ceremony in which each of the men was recognized as having been successful. The ceremony was designed to provide the participants with a sense of accomplishment and to leave them with a tangible symbol of their success. That the program may have achieved some of these goals can be found in comments by the judge and the professor:

"Many offenders also expressed an attitudinal change by returning to court family members or friends for the ceremony to reduce their probation period. Others exhibited their changed attitudes by atypically appearing in court with clothes reserved for special occasions. Still others expressed thanks for the court's faith through oral expression, written notes, or silent courtroom decorum. Many championed the program in the community, often enlisting other offenders' participation." (Kane and Waxler, 1995, p. 15)

Finally, providing these men with the opportunity to participate program is important in itself. Meisenhelder (1977) stressed that, in addition to forming social bonds, offenders seeking to reform require some form of "certification." Here, certification means,

"the social verification of the individual's "reform." Some recognized member(s) of the conventional community must publicly announce and certify that the offender has changed." (1977, p. 329)

Being part of a group which includes a college professor, a probation officer, and a judge seems to provide such "social verification." The group process involved making each member believe that he had just as much to offer as the next person. The tasks were such that each man was able to complete each reading assignment and be prepared to participate in discussions related to the readings. As such, the program is structured for building success experiences for each man. The recognition offered by the Changing Lives through Literature program may be what the offender needs to make a successful transition to a conventional, nonoffending lifestyle.


"[E]ffective sentencing will normally involve the curtailment of freedom either behind walls or in the community, large measures of coercion, and enforced diminutions of freedom; this is entirely properly regarded as punishment. The language of treatment, reform, and rehabilitation has been corrupted by unenforced and uncritically evaluated good intention" ( Morris and Tonry, 1989, p. 5)

The results of our evaluation suggest that this program can make an effective contribution as a component of a comprehensive system of sentencing alternatives. But before this program can be implemented on a wider scale, current political reality requires that we address the issue of degree of punishment. As Morris and Tonry (1989) comment, the perception of the general public is typically that incarceration (in jail or prison) is punishment and other options are treatment, usually viewed as letting offenders off easily.

In response, it is important to distinguish between bibliotherapy as a sanction, and its use as an adjunctive activity undertaken in the course of fulfilling a more traditional criminal sanction. Bibliotherapy, coupled with the provision of positive role models, could be a feature of a person's criminal sentence whether he or she were on unsupervised probation, electronic monitoring or some other form of intensive probation, or intermittent custody. To the extent that bibliotherapy and positive role modeling are meant to help better equip a person for independent living in the community, successful participation in such a program could relieve fears about the risks involved in keeping a person in a community rather than an institutional setting. In other words, the sanction may run the gamut of severity with this type of program serving as but one component.

Finally, it is important to make the conditions of participating in this program credible and enforceable. This program is offered as an alternative to jail. A suspended sentence could serve as a mechanism of enforcement should the offender fail to participate or successfully complete the program. Program administrators and participating court officials would then have the authority to impose more restrictive sanctions if necessary to protect the community.


Bureau of Justice Statistics. (1987). Recidivism of young parolees. Washington, DC: U.S. Department of Justice.

Bureau of Justice Statistics. (1994a). Comparing federal and state prison inmates, 1991. Washington, DC: U.S. Department of Justice.

Bureau of Justice Statistics. (1994b). Prisoners in 1993. Washington, DC: U.S. Department of Justice.

Bureau of Justice Statistics. (1995). Jails and jail inmates 1993-94. Washington, DC: U.S. Department of Justice.

Gold, M. (1978). Scholastic experiences, self-esteem, and delinquent behavior: A theory for alternative schools. Crime and Delinquency, 24, 290-308.

Hynes, A.M., and Hynes-Berry, M. (1986). Bibliotherapy, The Interactive Process. Boulder, CO: Westview Press.

Irwin, J., and Austin, J. (1994). It's about time: America's imprisonment binge. Belmont, CA: Wadsworth.

Kane R., and Waxler, R. (1995, January-February). Alternative sentencing: Changing Lives Through Literature Program. Community Corrections Report, 2, 1-3, 14-15.

Lawrence, R. (1985). School performance, containment theory, and delinquent behavior. Youth and Society, 17, 69-95.

Meisenhelder, T. (1977). An exploratory study of exiting from criminal careers. Criminology, 15, 319-334.

Miller, D. (1993). The Literature Project: Using literature to improve the self-concept of at-risk adolescent females. Journal of Reading, 36, 442-448.

Miller, D. (1994). Using literature to build self-esteem in adolescents with learning and behavior problems. Clearing House 67, 207-211.

Morris, N., and Tonry, M. (1990). Between prison and probation: Intermediate punishments in a rational sentencing system. New York: Oxford University Press.

Rubin, R.J. (1978a). Using Bibliotherapy. Phoenix, AZ: Oryx Press. Rubin, R.J. (Ed.). (1978b). Bibliotherapy Sourcebook. Phoenix, AZ: Oryx Press.

Sampson, R.J., and Laub, J.H. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press. Sheridan, J.T., Baker, S.B., and de Lissovoy, V. (1984). Structured group coining and explicit bibliotherapy as in-school strategies for preventing problems in youth of changing families. School Counselor, 32, 134-141.


G. Roger Jarjoura received his PhD in criminology in 1990 from the University of Maryland. He is presently an associate professor of public and environmental affairs at Indiana University, Indianapolis. His research interests currently focus on the cognitive thinking patterns of juvenile offenders and juvenile aftercare programming.

Susan T. Krumholz, previously an attorney in criminal defense, is presently a visiting lecturer in sociology at the University of Massachusetts at Dartmouth and a doctoral candidate at Northeastern University. Her research focuses on alternatives to incarceration and intimate violence.

The authors wish to thank Mark Mellberg, Jessica Miller, and Robert Allcock for their help in collecting and processing the data for this project.

Address correspondence to G. Roger Jarjoura, School of Public and Environmental Affairs, Indiana University, 801 W. Michigan Street, Room 4080, Indianapolis, IN 46202-5152.

An Official UMass Dartmouth Web Page/Publication © 2003 Board of Trustees of the University of Massachusetts. All Rights Reserved.
285 Old Westport Road | North Dartmouth, MA 02747-2300 | Phone: 508 999-8000 | Fax: 508 999-8901 | email comments to: