Male Homosexual Attraction to Minors Information Center
Home Using this site Print documents Full Bibliography Mission Contact us

Article details

Title: The abused/abuser hypothesis of child sexual abuse: A critical review of theory and research
Author(s): Randall J. Garland and Michael J. Dougher
Affiliation: Department of Psychology, University of New Mexico
Citation: Garland, R.J. & Dougher, M.J., “The abused/abuser hypothesis of child sexual abuse: A critical review of theory and research,” in Feierman, J. (ed.), Pedophilia: Biosocial Dimensions, New York: Springer-Verlag, 1990, pp. 488-509.


This article assesses the widespread belief among the general public and professionals that children and adolescents who have sex with an adult or significantly older adolescent are at risk in later years of acting sexually with children. This belief is referred to as the “abused/abuser hypothesis.”

The authors write that given its popularity, it is surprising that there is so little evidence to support it. Only a few studies have actually investigated it, and they suffer from significant flaws. The uncritical acceptance of the abused/abuser hypothesis may lead to premature and incorrect conclusions about the causes of childhood sexual abuse as well as to ineffective policies.

Speculation has been the basis for several theories put forth to support the abused/abuser hypothesis. In 1965, McGuire and colleagues proposed a classical conditioning theory. They believed that early childhood sexual experiences with peers or adults might supply the material for fantasies which were conditioned by pleasant orgasms during masturbation. These fantasies then may have become more deviant over time due to memory distortion and feelings of inadequacy.

Since then, Alter-Reid, Finkelhor, and Yates have also promoted this theory. Some indirect support for it has been lent by studies finding that sexual response can be conditioned, and that some offenders are aroused by descriptions of their own childhood sexual experiences with adults.

In the 1980s Freeman-Longo and Howells proposed that sexual behavior was learned through a modeling process. According to their theory, a child or adolescent who had sex with an adult saw that adults did this, enjoyed it, and were not punished. As a result, he may have grown up to believe such behavior was acceptable.

Both of these behavioral theories are limited. There is no systematic evidence that conditioning or modeling processes operate in the development of sexual behavior with children. In addition, it is obvious that some other variable is necessary to explain why a minor who is sexually involved with an adult would remain sexually interested in minors when he becomes an adult. Many children experience arousal and orgasm with other children but do not have sex with children when they reach adulthood.

Since the 1960s, several researchers have proposed psychodynamic theories to support the abused/abuser hypothesis. One theory speculated that an emotionally gratifying sexual experience with an adult in childhood (especially when the child was neglected by parents) could cause him or her to identify with the adult and therefore repeat the behavior in adulthood. As an adult, he or she would identify with the child who was the object of his or her affection.

This theory has some indirect support from studies which have found that (1) emotional deprivation from the father correlates with sexual interaction of boys with men, (2) not all men report childhood sexual contact with men negatively, (3) some men report that they experienced the disruption in their boyhood sexual relationship with a man as a loss, and (4) pedophiles attracted to boys often retrospectively report poor relationships with their fathers. However, there is no direct empirical support for this psychodynamic theory.

In the 1970s, an alternative psychodynamic theory was proposed which claimed that any kind of childhood sexual trauma caused some kind of adult “perversion.” This theory held that the traumatized child, through “identification with the aggressor,” converted his or her passive experience into an activity done to others in order to gain revenge and mastery over the trauma.

Studies of boys involved in sexual activity with adults and clinical observations of sex offenders against children provide some indirect support for this theory. Aggressive (sometimes sexual) behavior in boys is commonly correlated with disclosure of sexual involvement with adults. Several studies of adult offenders have found that they often replicate the sexual experience they had as boys.

While all of the above theories have some indirect empirical support, none have been shown to offer superior explanatory and predictive power. The area needs systematic research.

However, it can be concluded that by itself, childhood or adolescent contact with adults is inadequate to explain subsequent adult behavior with minors. Numerous studies suggest that adult-minor sex has no inevitable consequences. It is necessary to study other variables to determine under what conditions the abused/abuser hypothesis has merit.

Many researchers have studied the prevalence of childhood sexual involvement with adults among sex offenders. All studies with adolescent offenders lack comparison groups so it is impossible to determine if the prevalence of childhood sex with adults is higher among them than among non-offenders with similar demographic characteristics. Also, these studies mix all types of offenders together—exhibitionists, voyeurs, rapists, and offenders against minors. Among these groups, prevalence estimates have ranged from 18% to 47%.

Not all studies of adult offenders suffer from the same flaws, but those that do find prevalence rates of 20% to 40%. Those studies that have used comparison groups have used various kinds, including other kinds of offenders as well as non-offenders. They have found widely varying prevalence rates for both offenders against minors and the comparison groups: 0% to 57% among offenders against minors, 8% to 57% among other offenders, and 3% to 16% among non-offenders. Differences in sampling methods and definitions of abuse account for these varying findings.

Despite these differences, self-reported child and adolescent sexual behavior with adults is more prevalent among sex offenders against minors than among non-offenders. However, it appears to have the same prevalence among all types of offenders (including those convicted of non-sexual crimes).

Even this conclusion is tentative because of the methodological problems of the studies. They used adjudicated and incarcerated offenders—probably not representative of all men who have sex with minors. Men who are most “pathological” or who come from disadvantaged backgrounds are the most likely to be incarcerated. In addition, no method was used to control for differences among compared groups on ethnicity, psychopathology, or other factors that may have contributed to differences in childhood experiences with adults.

A reasonable estimate of adjudicated offenders against minors who experienced childhood sexual contact with an adult is 30%, so that such contact still characterizes only a minority of them.

Furthermore, even though population samples have typically found that 10% of minors have had sex with an adult, it is unlikely that 10% of all adults have sex with minors. Thus, many minors who experience sex with adults do not grow up to have sex with minors. The apparent conclusion is that “childhood and adolescent sexual contact with adults is neither a necessary nor a sufficient cause for becoming an adjudicated sex offender of children or adolescents.”

The abused/abuser hypothesis is therefore inadequate, incorrect, simplistic, and misleading. Additional factors need to be considered. Studies suggest several that may be important.

First are characteristics of the child or adolescent. These may include age (although data conflict on which ages are more effected), pre-existing emotional disturbances (which can exacerbate trauma or alternatively cause affectionate sexual interaction to be experienced exceptionally positively), interpersonal skills, stability of gender identity, knowledge and attitudes about sex, parental mental health, parental attitudes about sex, and family dysfunction.

Characteristics of the sexual interaction may also be important. These may include the quality of the relationship between the minor and the adult, the minor’s perception of control over interaction, the amount of force involved, the gender of the adult (which determines whether the interaction is heterosexual or homosexual), the duration and frequency of contact, the kind of sexual contact involved (findings disagree on the effect of this factor), the amount of age difference, the social visibility of the relationship, and the circumstances under which the contact or relationship ended (which may cause a feeling of loss).

Finally, effects of the interaction may be important. These may include the minor’s behavior after the contact or relationship ends (such as more sexual behavior), the consequences for the adult (whether or not he is punished), and the reactions of others to disclosure of the contact (denying, minimizing, blaming the child, or overreacting which may heighten trauma).

Future studies should be designed to take into account all these factors in an attempt to determine under which cases the abused/abuser hypothesis is valid.

Top of page
Home Using this site Print documents Full Bibliography Mission Contact us