Changing Perceptions of the Etiology of Crime: The Relationship Between Abuse and Female Criminality

Susan Marcus-Mendoza
Elizabeth Sargent
Yu, Chong Ho

Abstract

This study examines the relationship between female criminality and the experience of abuse. A survey was administered to 557 female inmates at two correctional centers in Oklahoma. Eighty percent of the subjects reported being abused, and 72.3 percent of the subjects surveyed experienced at least two types of abuse. Abused inmates were more likely to report substance abuse problems, emotional problems, interpersonal problems, and have a pessimistic attitude towards life after prison. Our findings suggest the need for more research about the connection between women's criminality and abuse, and the development of programs for incarcerated women who have been abused.

Introduction

Most of the research available on the criminality of women suggests that there is a relationship between crime and environmental factors such as economic opportunity or attitudes towards women. However, many aspects of women's experiences have yet to be examined, especially in relation to the rapidly increasing number of incarcerated women. This study examines the relation between the criminality of women and the experience of emotional, physical, and sexual abuse.

Historical View of Women in Prison and Prison Reform in the United States

An examination of the limited literature on the history of the incarceration of women reveals a complex set of social, political, and personal problems experienced by women in this country over the last two centuries. By the early nineteenth century, the reasons for incarcerating women and their plight once in prison were of concern to prison reformers. According to Freedman (1981) in Their Sisters' Keepers, three conditions began to emerge in the 1820s which gave rise to the prison reform movement for women. First, most northern states adopted the prison as a primary means of punishing and reducing criminal activity. Second, a small but significant number of women became inmates of these prisons, especially after 1940. Finally, middle-class American women, motivated both by benevolence and their growing consciousness as a sex, became active in reform movements that brought them into contact with their imprisoned sisters. Two social forces--the increasing rate of incarcerated women and a growing social consciousness among middle class women--merged to create a prison reform movement which is still evolving today.

Freedman (1981) suggests that the growing rate of women in prison between 1815 and 1860 can be linked to social change, especially urbanization, and new agents of social control such as urban police and moral reformers. Under these influences, "not serious crimes against persons or property, but unlawful personal behavior--drunkenness, idle and disorderly conduct, and vagrancy--brought the majority of criminals of both sexes into the courts and prisons (Freedman, 1981, p. 14)." However, Freedman states that the moral codes for women were stricter, and therefore, women were more likely to be convicted of such crimes. In addition, she points out that fewer job opportunities and lower wages for women resulted in economic marginalization and increased the need for women to resort to crimes such as prostitution, especially during wars, when men were not able to support their families. Prostitution was often the most readily available way for women to support themselves and their children. Once convicted or even suspected of a crime, a woman became even further marginalized. The penalty for the nineteenth century female criminal was the label "fallen woman," and both men and women shunned anyone suspected of being a "fallen woman." Because of this stigma, the female prisoner was largely neglected and often subjected to "overcrowding, harsh treatment, and sexual abuse . . . " (Freedman, 1981, p. 15).

This attitude towards women can be traced to our European ancestors. According to Feinman (1980), in classical Greece and Rome and medieval Europe, the primary function of a woman was to provide heirs for her husband to continue his name and property line. Therefore, adulterous women could be executed because by being unfaithful, they threatened the legitimacy of the heirs. In seventeenth century England, unmarried mothers were imprisoned because their children were dependent upon the parish. And, by the late 1600s, homeless women and mothers of several illegitimate children were being sent to the American colonies.

In the late nineteenth century, Lombroso (1900) developed a theory of criminology based on Social Darwinism. Lombroso theorized that women, nonwhites, and the poorer classes were less evolved than white, upper-class men, and therefore, were more likely to commit crimes. He further states that for a women to commit crime and stray from the "normal" path of "maternity, piety, and weakness, . . . her wickedness must have been enormous . . . (Lombroso & Ferrero, 1900, p. 150)." This theory helped perpetuate the "fallen woman" notion.

To effectively help women prisoners, women reformers had to free themselves of the long-held societal biases against "fallen women." They had to step over the "sexual purity" line and identify both themselves and imprisoned women as being part of the same class:

. . . a sexual class, an identity that contradicted and potentially weakened the purity boundary. Eventually some women would find the concept of a common womanhood stronger than the boundary of moral purity. A few would cross the line and cautiously enter the "gloomy abode" of women prisoners. (Freedman, 1981, p. 21).

These early reformers focused on conditions women inmates were subjected to and they were largely responsible for the establishment of separate prisons for women.

During the Progressive Era, at the beginning of the twentieth century, women reformers turned their attention to the causes of female criminality. They rejected Social Darwinism and began to develop a sociological theory of female criminality that assailed the concept of a physiological criminal type, examined the relationship of mental ability and crime, and eventually "argued for an economic interpretation of women's crime." (Freedman, 1981, p. 111). In fact, Freedman states that many of the first women social scientists actually conducted research in women's prisons, and further, a new theory of criminality gradually emerged which provided a shift from the "fallen women" notion.

The new sociological theory identified environmental sources of crime, including poverty, lack of education, and low-paying jobs. Consequently, it became apparent that prisons could not resolve the social problems associated with women's criminality. Reformers took several approaches toward solving the social problems that resulted in the incarceration of women and toward helping women once incarcerated.

On the one hand, Progressive reformers who favored the extra-institutional, preventive services over incarceration concentrated on changing criminal justice practices before the stage of imprisonment. On the other hand, . . . [other reformers] tried to improve the women's prisons through better classification and education, and diversified training (Freedman, 1981, p. 126). Thus, these growing environmental reform theories gave rise to preventive services, mostly aimed at keeping economically marginalized women from using criminal activities such as prostitution to solve economic problems (Freedman, 1981).

Current Theories of the Etiology of Crime by Women

Social Darwinist theories of female criminality still exist (Wilson & Herrnstein, 1985; Wilson, 1975), and contemporary society still espouses these ideas. Judge Lois Forer, a trial judge for 16 years and a noted author, stated that over the past two decades:

In both the United States and the United Kingdom the emphasis was on the offender as a wrongdoer or sinner who must be punished . . . Criminals soon came to be perceived as an evil class who must be punished for their sins, not rehabilitated. The fault lay not in society but in the felon and in his or her family. Genetic defects, a thinly disguised racist theory, again became a popular explanation for the causes of crime. (Forer, 1994, pp. 48-49).

Other theorists have continued to explore the environmental explanations of women's criminality as proposed by the Progressive reformers, and economics is still at the center of much of that theory. As the female rate of incarceration increases, researchers continue to develop theories of female criminality. According to Simon and Landis (1991), there are four major theories of why women commit crimes. The first, based on the work of Adler, is the masculinity thesis. Specifically, this thesis states that as women are liberated and assume more traditional male social roles, they tend to assert themselves in typically male ways--that is, they become aggressive, pushy, and hardheaded . . . Good girls are still those who maintain their allegiance to traditional social roles, while bad girls are those who act like men (Simon & Landis, 1991, p. 2). Increased criminality is equated with being "masculinized." The roots of this thesis can be seen in the historical perspective offered by Freedman--women are being incarcerated because they do not adhere to the strict moral codes or social roles assigned to them.

The second theory, the opportunity thesis, posits that as women attain social positions similar to men, their pattern of criminality will also become similar. It is women's objective locations in the social structure and particularly within the occupational sphere, as well as in the private, family sphere, that influence the nature of their criminality. Thus, this thesis argues that as the employment patterns of men and women become similar, so too will their patterns of employment related crimes (Simon & Landis, 1991, p. 3). Opponents of this thesis assert that women do not have the same employment opportunities as men, and further, that women are socialized differently than men and so would not necessarily commit the same crimes. Messerschmidt (1986) states that a review of crime statistics actually reveals that the rise in crime is predominately among younger offenders who are not part of the white-collar work force, and that the crimes being committed are nonoccupational, thereby discrediting the opportunity theory. He explains that crimes such as shoplifting and fraud (check and welfare) account for most of the rise in crimes committed by women.

The third view, the marginalization thesis, offers the opposite assumption of the opportunity thesis--women commit crimes due to the lack of opportunities to make money. The proponents of the marginalization thesis contend that:

1. Greater participation in the labor force does not necessarily mean either more equality between the sexes or an improved economic situation for women.
2. The bulk of female offenders, if employed at all, are concentrated in a pink collar ghetto, and their positions are characterized by poor pay and unrewarding, insecure work.
3. Female crime, the bulk of which is petty property crime, constitutes a rational response to poverty and economic insecurity (Simon & Landis, 1991, p. 9).

Although this theory would imply that as women's economic opportunities rise, their property crime rate will fall, Simon and Landis point out that the reverse has occurred--women are attaining more white-collar positions, and are committing more property crimes. However, according to Messerschmidt, these two phenomena are not necessarily connected.

Finally, the chivalry thesis suggests that in response to the women's movement, the criminal justice system has lessened their leniency (chivalry) towards women who commit crimes, "creating the 'if it's equality they want, it's equality they'll get' mentality (Simon & Landis, 1991, p. 12)." However, Simon and Landis point out there is little evidence of chivalry in the court system, and that any favors granted are probably granted to white, upper-class women. Since the typical female offender is not a white, upper-class woman, chivalry probably does not have any great impact on the crime rate among women.

Messerschmidt posits that the criminal justice system is not only lessening their leniency, but also is punishing women for the women's movement. "Increasingly, those females who do not act in a 'feminine' way--that is, those whose behavior indicates an erosion of traditional female gender roles--are viewed as stereotypically nontraditional and therefore deserving of punishment (Messerschmidt, 1986, p. 80)." In addition, Messerschmidt suggests that the criminal justice system publicizes and exaggerates the crimes of women relative to men, which creates the impression of a larger increase in the female crime rate than actually exists.

Crime and Abuse

There are still unexamined environmental factors which may be contributing to the increase in crimes committed by women. Abuse is one such factor. Herman (1992) explains that the adult abused as a child will have difficulty negotiating the demands of adult life . . . the personality formed in an environment of coercive control is not well adapted to adult life. The survivor is left with fundamental problems in basic trust, autonomy, and initiative. She approaches the tasks of early adulthood--establishing independence and intimacy--burdened by the major impairments in self-care, in cognition and memory, in identity, and in the capacity to form stable relationships. She is still a prisoner of her childhood (p. 110). Given the limited capacity of the survivor of abuse to cope with the demands of adult life, it seems plausible that women who were abused as children would be unable to negotiate the environmental stressors that have been previously associated with crime. This is evident in Pollack-Byrne's (1990) description of women inmates, which is similar to Herman's portrayal of survivors of abuse. Pollack-Byrne notes that the typical female inmate abuses herself with drugs and alcohol, is perpetually involved in abusive relationships, and is facing financial hardship.

In her review of the research on women inmates, Pollack-Byrne (1990) finds a connection between abuse and crime. She found a range of 35 percent to 63 percent for sexual abuse, and 35 percent to 53 percent for physical abuse. Our study examines whether abuse might be a correlate to crime in Oklahoma, which has the second highest overall incarceration rate in the country, and the highest incarceration rate for women (Bureau of Justice Statistics, 1993).

Method

Subjects and Procedures

Surveys were administered to a total of 557 female inmates at two correctional centers in Oklahoma as part of a larger project (The Project for Recidivism Research and Female Inmate Training; PRRFIT). The surveys were administered under the supervision of the research team to large groups of inmates. Assistance was available for inmates who had questions or difficulty reading and writing.

Survey

The survey was developed by a research team-- researchers with a variety of interests--to collect preliminary data on a number of issues. The team's survey contained 142 questions in a variety of formats, including Likert scale, multiple choice, and open-ended questions. The survey included questions about demographics, criminal history, history of abuse, alcohol and drug abuse, medical and psychological problems and treatment, and self-concept. To obtain a history of abuse, the inmates were asked, using six separate questions, whether they had been physically, sexually, or emotionally abused either before or after the age of 18. Inmates were also given space to elaborate on their responses.

Results

Categorization of Abuse

To examine the characteristics of abused inmates, and compare them to those who did not report abuse, the data were divided into groups on the basis of their answers to the questions asking if they had ever experienced physical, emotional, or sexual abuse before or after the age of 18: 1) those inmates who responded "never" to all six questions on abuse, and 2) those who reported one or more types of abuse.

Categorizing the respondents into "never" abused and "ever" abused provided a logical way to characterize the differences between the inmates reporting a history of abuse and those never abused for several reasons. First, some inmates did not distinguish between the types of the abuse experienced on the open-ended questions about abuse. For example, some inmates reported instances of physical abuse and when asked how they had been physically abused, they responded "sexually." Second, the fact that the majority of the inmates who were abused reported experiencing two or more types of abuse suggests that types of abuse often co-exist and that it would be extremely difficult to distinguish between the effects of the different types of abuse. Third, distinguishing between inmates by age at the time of abuse (before or after 18) seemed unwarranted since many inmates were abused both before and after the age of 18. Finally, since many of the women reported having their first child by the age of 18, it seemed inaccurate to have an arbitrary divider between childhood and adulthood, Therefore, the only meaningful distinction seemed to be between women who had never been abused and those who had experienced one or more types of abuse.

Given the small number of inmates in the "never" category, we combined the data from the two correctional centers to conduct the analyses. Overall, 110 (20 percent) reported never experiencing any type of abuse, and 439 inmates (80 percent) reported being abused. Forty-three inmates (7.7 percent) reported experiencing one type of abuse, 95 (17.3 percent) reported two types of abuse, 84 (15.3 percent) reported three types of abuse, 76 (13.8 percent) reported four types of abuse, 87 (15.8 percent) reported five types of abuse, and 54 (9.8 percent) reported six types of abuse.

Demographics

We obtained descriptive statistics on the demographics for each group and looked for significant differences between the two groups using t-tests and Chi-squares. Table 1 summarizes the demographics of the two groups.

In examining the demographic variables, we found several significant differences. When we collapsed Question 7 to combine those who have either a high school diploma or a GED versus those who had neither, we found a significantly higher percentage (64.10 percent) of those with a diploma or GED reporting abuse than those not reporting abuse (52.38 percent; Chi-square=4.909, p=0.027, phi=0.096). We also found a significant difference in the racial composition of the two groups (Chi-square=14.061, p=0.001, phi=0.170). The group reporting abuse was mainly Caucasian (47.7 percent Caucasians, 10.07 percent Native Americans and 37.0 percent African Americans) while the group not reporting abuse was mainly African American (31 percent Caucasians, 7 percent Native Americans, and 57 percent African Americans). Overall, a higher percentage of the Caucasian inmates (86.34 percent) and Native American inmates (86.27 percent) reported being abused than did African American inmates (72.73 percent). Since Caucasians, Native Americans, and African Americans account for 95.3 percent of the inmates, they were the only groups with enough representation to be included in the analyses.

Table 1: Demographics of Inmates Reporting Abuse and Inmates Not Reporting Abuse

Abused inmates received significantly fewer visits from their children than inmates not reporting abuse (Chi-square=4.537, p=0.033, phi=-0.119). We compared those who received visits from their children on a regular basis (weekly, monthly, or every few months) to those who reported never receiving visits from their children and found 75.7 percent of those who had been abused received regular visits from their children as compared to 85.38 percent of those who did not report being abused.

Legal History

The only significant difference found between the legal histories of the two groups was on the number of convictions. We found that a higher percentage of the abused inmates (54.92) had been convicted of two or more crimes than those not reporting abuse (40.91; chi-square=5.717, p=0.017, phi=-0.106).

Substance Abuse, Emotional and Physical Problems

Next, we examined the histories of abuse, drug and alcohol problems, and mental and physical health problems. Table 2 summarizes the descriptive statistics.

Several significant differences were found when comparing the histories of the two groups in relation to drugs, alcohol, emotional and physical problems, and counseling. More inmates who had been abused reported drinking before incarceration (Chi-square=4.475, p=0.034, phi=-0.093), using street drugs before incarceration (Chi-square=6.407, p=0.011, phi=-0.112), thinking they had a problem with drugs (Chi-square=19.160, p=0.000, phi=0.193), and thinking they had a problem with alcohol (Chi-square=15.423, p=0.000, phi=0.174).

In addition, more inmates who had been abused reported receiving treatment or counseling for an alcohol problem before incarceration (Chi-square=6.753, p=0.009, phi=-0.114), and while incarcerated (Chi-square=10.248, p=0.001, phi=-0.142), and treatment or counseling for an emotional or mental problem before incarceration (Chi-square=12.926, p=0.000, phi=-0.158), and while incarcerated (Chi-square=7.910, p=0.005, phi=-0.126). Similarly, more inmates who reported abuse had been prescribed medication for an emotional problem before incarceration (Chi-square=22.538, p=0.000, phi=0.210), and while incarcerated (Chi-square=4.958, p=0.026, phi=0.099). Abused inmates also reported more health problems both before incarceration (Chi-square=8.173, p=0.004, phi=-0.128), and while incarcerated (Chi-square=16.124, p=0.000, phi=-0.181).

Table 2: History of Abuse, Drugs, Alcohol, and Mental and Physical Health Problems of Inmates Reporting Abuse and Inmates Not Reporting Abuse by Percentage

 Open-Ended Questions

We examined the response rates of the two groups of inmates and found that in all cases, significantly higher percentages of inmates who reported having been abused answered the questions. Significantly different response rates were found between the groups on the questions about their current conviction (95.82 percent versus 83.33 percent; Chi-quare=21.616, p=0.000, phi=0.200); what contributed to their on-going trouble with the law (51.74 percent versus 29.63; Chi-square=16.937, p=0.000, phi=0.177); how they felt about returning to important relationships (92.11 percent versus 68.52 percent; Chi-square=43.603, p=0.000, phi=0.284); what would help them stay away from unhealthy relationships (85.85 percent versus 65.74 percent; Chi-square=23.468, p=0.000, phi=0.209); and what is waiting for them when they get out of prison (94.9 percent versus 80.56 percent; Chi-square=24.190, p=0.000, phi=0.212).

We further examined Question 78, "What were you convicted of for your current sentence," by coding the responses according to the crime committed and category of crime. Some inmates listed more than one crime. The frequencies and percentages for each type and category of crime is reported for both the groups of inmates in Table 3. T-tests were conducted on categories A (murder and murder related crimes), D (drug related crimes), and E (money related crimes). We found a higher number of inmates who had not been abused committed drug related crimes (T=2.2291, p=0.0261).

Table 3: Categories of Crimes Reported

 Question 82 asked "What do you feel contributed to your ongoing trouble with the law?" To examine the differences on this question between the inmates who reported abuse and those who did not, three coders placed responses in 16 coding categories. The responses were coded separately by two of the coders, and then the three coders met to decide the final code for each response. As with Question 78, some inmates gave more than one response. The frequencies and percentages for each group are presented in Table 4. No further analyses were possible on this question due to the small number of responses (31) from the inmates who did not report abuse.

We also obtained frequencies of the number of negative responses or responses that reflected mixed feelings on Question 138, "What is waiting for you when you get out of prison?" Thirty of the abused inmates who answered the question gave negative responses such as "nothing," "death," or "uncertainty," and 31 gave mixed responses such as "a darling son, a loving mother-in-law, and a sorry husband." These 61 responses represent 14.91 percent of the responses given by the abused inmates. Only two negative responses and one mixed response were given by the non abused inmates, representing 3.45 percent of the responses.

Discussion

Our study may seem limited by the small number of women who did not report abuse; that in itself is our most salient finding. The majority of the women inmates surveyed in our study reported having experienced two or more types of abuse. Often the abuse was of an extremely violent nature such as, "My father beat me with anything he could get his hands on such as bats, guns, etc.," and "Threatened he would kill me and my folks, screaming in my face, held a gun to my head while having sex."

The abused inmates were more likely than inmates not reporting abuse to report histories of substance abuse, emotional problems, problems with family and friends (including fewer visits from children), and a pessimistic outlook about what their life will be like when they leave prison. This pattern is similar to the profile of the adult survivor of abuse described by Herman (1992) and to the typical female inmate described by Pollack-Byrne (1990).

Table 4: Reasons for Continuing Problems with the Law

 In addition, the abused inmates were more likely than inmates who had not been abused to have been convicted of at least two crimes, and were more likely to cite the factors listed above as causing their ongoing problems with the law. The major reasons cited by the abused inmates for ongoing problems with the law were drug-related activities (35.37 percent), money (12.92 percent), emotional problems (11.56 percent), peers (11.22 percent), and family problems other than abuse (5.44 percent), and abuse by spouse or family (4.76 percent). The top reasons cited by inmates who had not been abused were drug-related activities (35.48 percent), money (22.58 percent), self-specific trait (12.9 percent), emotional problems (9.68 percent), and problems with police or legal system (6.45 percent). Although both groups reported the top two reasons for ongoing problems as being drug-related activities and money, emotional and interpersonal difficulties were more problematic for survivors of abuse. Caution must be used in interpreting these results, since only 51.74 percent of abused inmates and 29.63 percent (31 inmates) of the inmates not reporting abuse answered the question "What contributed to ongoing problems with the law?"

Many inmates answered "drugs" to the question discussed above without any further explanation. Both groups of inmates had high rates of conviction for drug-related offenses (abused: 35.43 percent and never abused: 46.49 percent), although the rate of conviction for inmates not reporting abuse was significantly higher. In addition, 42.8 percent of the abused inmates and 19.6 percent (a significant difference) of the inmates not reporting abuse reported having a drug problem. Therefore, whether they were referring to substance abuse or legal problems related to drugs is unclear and needs to be researched further.

The fact that a large number women who had been abused cited abuse as contributing to their legal difficulties suggests that abuse may, indeed, be an environmental cause of female criminality. Research is needed to further clarify the role that abuse plays in female criminality. The emphasis on money problems, specified by inmates with and without a history of abuse as a reason for their ongoing legal problems, suggests that economic marginalization may also be a factor for this group of women and should also be studied further.

To be effective, inmate services during incarceration must address the needs of women who have been abused. Abused women experience a profound lack of control during their abuse, have difficulty with authority figures, engage in unhealthy coping strategies (Herman, 1992), all of which may be exacerbated by incarceration. Therefore, abused women are likely to be retraumatized by their incarceration, making them unable to benefit from the usual rehabilitative services unless the issues surrounding trauma and abuse are addressed. Thus, survivors should be offered therapy to help them better adapt to the prison environment. Inmates with a history of abuse should also receive therapy that addresses how the abuse contributed to becoming incarcerated. Because of the numerous difficulties faced by abuse survivors, programming should focus on alternative ways of coping. While some of these services may already be available, the findings of our research indicate they need to be expanded to accommodate the large number of inmates who report a background of abuse. Finally, prison staff should be educated about the potential difficulties experienced by survivors, so that the interactions between staff and inmates do not exacerbate abuse-related difficulties.

The abuse of women is a serious societal issue; the fact that many incarcerated women share a history of abuse indicates that abuse and its resulting problems must be addressed by the criminal justice system. Judge Forer described the vicious cycle in which many abused women who commit crimes now become caught:

Dorita's crime was shoplifting. She had been convicted three times and, . . . she was sentenced under a habitual offender statute mandating a prison term. She was poor, black, unemployed, a single mother on welfare. She fit the profile of the female prisoner. On release she will still be poor, unemployable, and on welfare. Her daughter, who is a ward of the state, will be shuttled from one foster home to another. When Dorita is released, her daughter will not know her. Dorita will be alone and embittered. (Forer, 1994, p. 80).

The number of the embittered, unemployable women released from prison will probably continue to increase unless proper interventions are employed during incarceration.

Biography

Susan Marcus-Mendoza, Ph.D., is assistant professor of human relations and women's studies at the University of Oklahoma, Norman.

Elizabeth Sargent, M.A., is a doctoral student in English at the University of Oklahoma, Norman.

Yu Chong Ho, M.H.R., is a doctoral student in educational psychology at Arizona State University, Tempe.

Bibliography

Feinman, C. (1980). Women in the criminal justice system. New York: Praeger.

Forer, L. (1994). A rage to punish: The unintended consequences of mandatory sentencing. New York: W. W. Norton.

Freedman, E. B. (1981). Their sisters' keepers. Ann Arbor: The University of Michigan Press.

Gillard, D. K. (1993). Prisoners in 1992. (Report No. NCJ-141874) Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Herman, J. L. (1992). Trauma and recovery. New York: BasicBooks.

Lombroso, C., & Ferrero, W. (1900). The female offender. New York: Appleton.

Messerschmidt, J. W. (1986). Capitalism, patriarchy, and crime: Towards a socialist feminist criminology. Totowa, NJ: Rowman & Littlefield.

Pollock-Byrne, J. M. (1990). Women, prison, and crime. Pacific Grove, CA: Brooks/Cole.

Simon, R. J., & Landis, J. (1991). The crimes women commit, the punishments they receive. Lexington, MA: Lexington Books.

Wilson, J. Q. (1975). Thinking about crime. New York: Random House.

Wilson, J. Q., & Herrnstein, R. (1985). Crime and human nature. New York: Simon & Schuster.