The alleged association between homosexuality and pedophilia is a contentious issue that is addressed on this page.

Basic concepts

Age and sex preferences - In psychology, the terms homosexual, heterosexual, and bisexual refer to people with a preference for sexual partners of the same sex, a preference for sexual partners of the opposite sex, and a preference for sexual partners of both sexes, respectively; these terms do not provide any information as to the preferred age of the sexual partners. Some terms that provide information as to the preferred age of the sexual partners include pedophilia (sexual preference for pre-pubescents), ephebophilia (sexual preference for pubescents), hebephilia (sexual preference for adolescents), teleiophilia (sexual preference for adults), teleiohebephilia (sexual preference for adults and adolescents), and so on.
Pedophile vs. child molester - A pedophile is an individual with intense, recurrent sexual attraction toward pre-pubescent children that often borders on obsession-compulsion and is sometimes obligatory.1 Not all child molesters are clinically diagnosed as pedophiles. Hence, child molestation is neither necessary nor sufficient for a pedophilia diagnosis.
Pederast - A man sexually interested in teenage or younger boys. Pedophiles are pederasts but pederasts are not necessarily pedophiles.


Keeping the aforementioned definitions in mind, plenty of evidence indicates that homosexuals and bisexuals are notably relatively overrepresented among pedophiles and hebephiles. On the other hand, the majority of homosexuals and bisexuals are teleiophiles and are unlikely to molest children because they have no sexual interest in children. Alternatively, if one ignores the above terminology and focuses on the phenomena of sexual interest in the same sex, sexual interest in the opposite sex, sexual interest in children, and sexual interest in adults, then a considerable amount of evidence indicates that sexual interest in the same sex is notably relatively elevated among individuals with any level of sexual interest in children compared to individuals with sexual interest in adults but not children, although the majority of adults with any level of sexual interest in the same sex are sexually interested in adults but not children.

Evidence from history

Historically, male homosexuality has typically been described in the context of the military and the priesthood, both usually involving pederasty. Many academics or textbooks on sexuality use examples of tolerance of homosexual behaviors from history to make the point that homosexual relationships between adults should be considered normal and accepted even though their citations involve tolerance of pederasty, not sexual relations between adult men.2 Even in cultures where it was acceptable for cross-dressing adult men to engage in receptive anal intercourse with other men -- e.g., the berdache among the American Indians and the xaniths among the Oman Arabs -- such individuals typically began homosexual relations with older boys and men before or around puberty.3, 4

The question is what happened to these pederasts? What made them largely disappear, i.e., end up being replaced by homosexual teleiophiles? The social constructionists have a convenient answer. They typically portray the history of male homosexual behavior in terms of a clear line of development from ancient pederastic relationships through early/modern patron/protégé relationships to modern egalitarian (teleiophilic) relationships. This issue has been addressed well by Norton:5

It is not a coincidence that this portrayal resembles the Marxist dialectic leading from feudalism through capitalism to a classless society. Foucault's epistemological “breaks” require a radical difference between certain classes of people, and also that these different classes not coexist. The social constructionist notion of “radical breaks” comes from “Marx's theory of the end of the ‘immanent critique,’ i.e. the revolutionary transformation of society, which required a radical break with history itself.” Such “radical breaks” have not occurred with respect to homosexuality.
Pederasty has far from disappeared in modern times. A pederastic magazine, Der Eigene, was published in Germany during 1898-1930. Pederastic themes dominate late-Victorian and Edwardian homosexual literature.6 Young rough trade was a common feature of homosexuality in England up to the 1930s. In France, pédérasts were “the archetypal homosexual roles” through the 1950s and the pédérast is still an important “queer cultural paradigm.” Significant age differences between male partners continue to be common in southern Europe today. Early gay liberation collections of poetry such as Winston Leyland’s Angels of the Lyre or Ian Young’s The Male Muse or Paul Mariah’s Manroot journal are filled with pederastic verse. Slim adolescent males dominated homosexual photographic magazines in London through most of the 1970s; “chunky rough types” were not common until the 1980s, and “older men” became common objects of desire only in the late 1980s, only as long as they wore leather.

One could add that during the early years of The Advocate, a major mainstream American publication for homosexuals, its unofficial mascot was a sexually provocatively posed male figure that had a roughly 10-year-old boy’s face and a muscular adult body.

The following point by Thorstad is also pertinent:7

A point I’ve been trying to make is that if we reject the boy-lovers in our midst today we’d better stop waving the banner of the Ancient Greeks, of Michelangelo, Leonardo da Vinci, Oscar Wilde, Walt Whitman, Horatio Alger, and Shakespeare. We’d better stop claiming them as part of our heritage unless we are broadening our concepts of what it means to be gay today.

Compared to the past, pederasts haven't largely disappeared in contemporary Western societies; they are just mostly closeted. Among the pedophiles and hebephiles that are not closeted, we observe them forming a group such as the North American Man-Boy Love Association (NAMBLA), but not a North American Man-Girl Love Association. Homosexual pedophiles, in general, are much more organized than heterosexual pedophiles with respect to forming groups advocating the lowering of the age of consent.

GLBT groups sometimes condemn pederasty, but the following incident explains the context:8

Prior to the 1960s, Dutch law (Section 248bis) considered homosexuality and pedophilia as equivalent. When Dutch homosexual activists did not stand a chance toward integrating homosexuality into mainstream society, the Netherlands Society of Homophiles COC did not display any special attitude toward pedophilia. However, with decreasing hostility toward homosexuals, the chances of repealing Section 248bis increased. Around the same time (1960s), the COC considered any association with pedophilia too great a liability, and displayed marked animosity toward pedophilia. To help integration, the COC changed its name to The Netherlands Society for Gay Integration COC in 1971, thereby de-emphasizing a separate identity. However, de-emphasizing a separate identity on the outside does not translate to a sense of similarity to heterosexuals on the inside. Some Dutch homosexuals reacted by forming separate groups emphasizing identity and pride, but the integrationists dominated. By the 1980s, the Dutch were ready to pass laws with the intention of reducing discrimination against homosexuals and providing other benefits to them. De-emphasizing a separate identity at this point would not help because the laws would appear to confer special rights to a group that was not very different from the great majority of people (heterosexuals); hence, homosexuals reemphasized a separate identity. Pedophilia became a gay-rights issue. In the mid 1980s, the Dutch Minister of Justice proposed lowering the age of consent to 12 years. The Public was outraged; homosexuals had overestimated the tolerance of the public toward homosexuality, and the issue was rapidly buried. Later on, it became legal for adults to have sex with children in the age range 12-16 years, provided that the children involved or their parents did not complain.

Evidence from individuals arrested for child molestation

Excluding individuals that molest both boys and girls, the proportion of incarcerated men that molest boys but not girls is about 33%.9, 10, 11, 12 Therefore, as a first approximation, homosexuals appear considerably relatively overrepresented among child molesters.

To compare the proportion of pedophiles among homosexual and heterosexual men, one needs to adjust the above figure for 1) the proportion of child molesters that are pedophiles, 2) the average number of children molested by homosexual and heterosexual pedophiles, and 3) how boys compare to girls with respect to reporting molestation.

  1. Pedophiles are expected to molest more children, on average, than non-pedophile child molesters. Among incarcerated child molesters, more homosexuals than heterosexuals have committed multiple offenses against children.13 This suggests that a greater proportion of homosexual child molesters are pedophiles compared to heterosexual child molesters.14 For instance, in a sample of 100 expert-appraised child molesters, half the homosexual/bisexual child molesters were pedophiles, whereas only a quarter of the heterosexual child molesters were pedophiles.15 Consistent with such results are phallometric data (penile arousal following erotic stimuli), which Freund and Watson used to calculate that among pedophiles, 41-43% are homosexual, assuming that homosexual and heterosexual pedophiles molest an equal number of children, on average, and that boys and girls are equally likely to report molestation.16
  2. Homosexual pedophiles are about twice as likely to be repeat child molesters as heterosexual pedophiles (DSM-IV-TR). Abel et al. reported that among non-incarcerated pedophiles, the average heterosexual pedophile had molested 20 girls, whereas homosexual pedophiles had molested an average of 150 boys each.17 Freund and Watson used this figure to calculate that among pedophiles, 10-11% are homosexual, assuming that boys and girls are equally likely to report molestation. However, in a sample of 91 non-incestuous child molesters, 34 homosexual child molesters (all but 3 molested boys exclusively) had molested an average of 3.3 boys each, and 57 heterosexual child molesters (all but 4 molested girls exclusively) had molested an average of 4.7 girls each.18 Hence, Abel et al.’s figures, given that the high status of the researchers may have been responsible for the referral of the worst pedophiles to them, likely does not represent the typical difference in the number of children molested by homosexual and heterosexual pedophiles. Therefore, the proportion of pedophiles that are homosexual, assuming that both boys and girls are equally likely to report molestation, is certainly higher than 10-11%.
  3. Boys are less inclined than girls to talk to their parents about their sexual adventures with grown-ups, and are more likely to sexually cooperate with male adults.19 In a probability sample of 585 men and women sexually abused in childhood, the men had experienced more severe sexual abuse, i.e., repeated assaults over a prolonged period involving force and actual or threatened violence, but were less likely to report it than the abused women.20 There is plenty of additional evidence that boys are far less likely to report molestation than girls.21, 22 This fact significantly increases the proportion of pedophiles that are homosexual, although Freund and Watson failed to adjust for this factor.

It is a rare female child molester that is diagnosed with pedophilia. Chow and Choi described a female pedophile that had molested two 4-year-old girls; this woman was a 23-year-old single woman with two sons who described herself as bisexual and more interested in women than men and also more interested in young girls than adult women, mostly preferring girls in the age range 3-4 years.23 Cooper et al. described a 20-year-old obese female pedophile that had been charged with molesting two girls, a 4-year-old and a 5-year old; this woman admitted to having molested several young boys and girls, and having participated in incestuous sex with some adult male relatives; she was also diagnosed with sexual sadism and zoophilia.24 Two cases don't help one compare the rates of pedophilia between heterosexual and homosexual/bisexual women, but the following data are consistent with an association between nonheterosexuality in women and increased likelihood of sexual interest in sexually immature individuals:

In an assessment of all registered adult female sex offenders in Texas (n = 471), the offenders fell into 6 clusters: 146 “heterosexual nurturers” with male victims averaging age 12 years; 114 “noncriminal homosexual offenders,” 96% of whose victims were females averaging age 13 years; 112 “female sexual predators,” 40% of whose victims were females averaging age 11 years; 50 “young adult child exploiters” with male or female victims averaging age 7 years; 22 “homosexual criminals,” 73% of whose victims were females averaging age 11 years; and 17 “aggressive homosexual offenders,” 88% of whose victims were females averaging age 31 years.25

Let us also address adolescent sex offenders. In a phallometric examination of 40 adolescent male sex offenders, ages 14-17 years, who had child victims, adolescents with only female victims did not differ from the comparison group of young adults in the age range 18-21 years, but adolescents with male victims displayed significantly greater sexual arousal to children than the young adults.26 Homosexual interests are considerably elevated among male adolescent sex offenders as evident in three different samples of male adolescent sex offenders in the age range 12-17 years:27

  • Sample 1 (n=45; 38%, male victims only; 24%, female victims only; 38%, victims of both sexes).
  • Sample 2 (n=158; 23%, male victims only; 52%, female victims only; 25%, victims of both sexes).
  • Sample 3 (n=69; 15%, male victims only; 63%, female victims only; 22%, victims of both sexes).

One can conclude that there is an association between same-sex sexual interest and sexual interest in children.

Evidence from general population samples

Age preference studies

Freund et al.,28 Bailey et al.,29 and Jankowiak et al.30 have studied partner age preferences among homosexual men and others. Freund et al. selected homosexuals that were attracted to adult but not teenage men, and we cannot draw meaningful conclusions from their study. Bailey et al. did not present visual stimuli to their subjects, but merely asked homosexuals about their age preferences. Jankowiak et al. presented participants with controlled visual stimuli, but restricted their homosexual participants to “middle-aged professionals” and the stimuli presented to “University to middle-aged men.” Both Bailey et al. and Jankowiak et al. reported that male homosexual responses showed an increased variability more characteristic of heterosexual women than heterosexual men. A possible reason for the increased variability of male homosexual responses compared to that of heterosexual men could be that some homosexuals preferred young men while others preferred older men. Neither Bailey et al. nor Jankowiak et al. examined this possibility. To overcome such shortcomings, Silverthorne and Quinsey examined the age preferences of 192, equal groups of homosexual and heterosexual men and women presented with pictures of human faces arranged into five categories within the age range 18-60 years.31

The results are shown in Fig. 1. Note the response of homosexual men compared to heterosexual men. Silverthorne and Quinsey wrote, “The current study was intended to expand Jankowiak and co-workers’ study by recruiting a wider age range of homosexual male participants and by categorically arranging the stimuli into specific age groups between 15 and 50 years of age.” They also criticized Freund et al. for selecting male homosexuals that were attracted to adult but not teenage men; yet reported results for stimuli in the age range 18-60 years. One wonders why? Given the trend in Fig. 1, the male homosexuals would likely have reported strong attraction for men in the age range 15-18 years, too.

Homosexual men are more strongly attracted to men in their late teens than heterosexual men are attracted to women in their late teens.
Fig. 1: The sexual appeal of faces (ages 18-60 years) of the desired sex to homosexual and heterosexual men. The mean ages (in years) of the participants were: homosexual men, 33.24 (SD (standard deviation) = 9.46, range = 18-50); heterosexual men, 33.14 (SD = 9.91, range = 18-52). The intensity of attraction is on a scale from 1-5, with higher values implying stronger attraction. Homosexual and heterosexual women did not differ with respect to partner age preference.

Nordic studies

In an investigation of 10,828 adolescents (94.3% response rate), ages 14-17 years, and representative of the private and public school systems in Oslo, Norway, 2.1% of the boys and 0.6% of the girls reported selling sex for money, mostly before age 16.32 These figures are consistent with a previous Norwegian survey and unlikely to be the result of overreporting or underreporting. The mean age for first episode of trading sex for money was 12.6 years in boys and 14.1 years in girls.33 The great majority of the clients were obviously men, and the figures suggest that nonheterosexual men are relatively overrepresented among hebephiles and pedophiles to an extent greater than what one would infer from prison statistics. A similar find has been reported in Swedish studies.34

Consider data from a representative sample of Swedish adolescents in school below and a literature review of child molestation prevalence in the U.S.A. (Table 1). As in the Swedish study, estimates of child molestation should be based on anonymously quizzing individuals rather than police reports given that boys are far less likely than girls to report molestation. In late adolescence or adulthood, a reasonable estimate of the proportion of individuals who engage in homosexual behaviors is 5%; i.e., the ratio of the prevalence of individuals engaging in exclusively heterosexual behaviors to those engaging in homosexual behaviors is at least 19 in the general population, which easily exceeds a similar ratio among child molesters.

Table 1: Prevalence of molestation in boys (B) and girls (G) in Sweden and the U.S.A.


Molestation Type



G/B ratio

Het/Hom (M)

Het/Hom (F)








Except non-contact






Oral/anal/vaginal *













Except non-contact






Estimate **






* Note that the molestation prevalence discrepancy between boys and girls appears to be reduced for the more serious forms of molestation. In the Swedish sample, sexual abuse in the form of anal sex was experienced by 3/25 (12%) of the molested boys but none among 126 molested girls; i.e., more boys had experienced the worst form of molestation than girls. In a review of physical trauma among molested children seen by physicians, 3.7% of the girls but 54.5% of the boys had sustained anorectal injury.35 ** Estimated prevalence of any form of molestation, adjusted for response bias. Non-contact forms of molestation include exhibitionistic acts. The Het/Hom ratio for male molesters (M) and female molesters (F) is the ratio of molestations of a heterosexual nature to that of a homosexual nature on the part of the molesters; it is assumed that homosexual and heterosexual molesters molest an equal number of children. In the Swedish sample, half the molested boys were molested by women. The Het/Hom ratio for molested boys in the American sample assumes that all girls have been molested by males and half the boys have been molested by females, i.e., the Het/Hom ratio is likely to be an overestimate rather than an underestimate. The Swedish study assessed a representative sample of 17-year-old Swedish adolescents (814 males, 1129 females).36 The American study found a higher prevalence of molestation in studies with smaller response rates, and the figures reported have been adjusted for response bias.37 However, given that the best response rates averaged 69%, the estimate (marked with two asterisks) is likely more reliable. The Swedish study had a 92.2% response rate among the school students.

Homosexual/homophile sources

There are studies from sources that are not anti-gay that document high rates of sexual experience with children among male homosexual adults:

  • Bell and Weinberg reported that in the 1960s, 25% of white homosexual men and 14% of African-American homosexual men had had as sexual partners boys who were 16 years or younger when they themselves were 21 years or older.38
  • In The Gay Report, Jay and Young reported that 23% of male homosexuals and 6% of female homosexuals admitted to sexual interaction with youth less than 16 years of age.39
  • In the SIGMA study of British male homosexuals (late 1980s to early 1990s), 3% of the participants “had their first sexual experiences” with a man before the age of 10 years, about 10% by the age of 10 years, about 25% by the age of 12 years, and approximately 50% by the age of 14 years.40

A large college student sample

Lee Ellis and colleagues have anonymously surveyed thousands of college students in the U.S. and Canada, and have published many analyses on the dataset (e.g.41). The following finds are from this extensive dataset; data courtesy of Jill Hellberg.

The students were not recruited with the understanding that they will be quizzed about sexual/pedophilic interests. They were asked about sexual interest in adults, children, the same-sex, the opposite-sex, and self-identified sexual orientation.

Analyses indicated an ordered ranked stratification of self-identified sexual orientation groups, based on both proportion with and intensity of pedophilic interest; ranked from highest to lowest and figures in parentheses indicating proportion with sexual interest in children: self-identified homosexual men (1 in 4); self-identified homosexual women (1 in 10) and self-identified heterosexual men (1 in 10), both similar to each other; and self-identified heterosexual women (1 in 50). These figures should not be extrapolated to the general population because the sample consisted of college students, not the general population, and unusual sexual interests have been shown to be higher among college students than in the general population. Nevertheless, the rank order undoubtedly applies to the general population, as suggested by other evidence cited on this page.

The authors then used cluster analysis to assess how the variables examined clustered, if at all, without pre-specifying the number of clusters they wanted the computer to group the data into, and the computer came up with two clusters among both men and women, best described as heterosexual vs. nonheterosexual, and it was found that the variables assessing pedophilic interest were largely confined to the nonheterosexual cluster among both men and women; see the two figures and table below.

Clustering of same-sex attraction and sexual interest in children in males.
Fig. 2: Dendrogram using complete linkage for men. Note that sexual interest in children clusters with same-sex attraction/fantasy.
Clustering of same-sex attraction and sexual interest in children in females.
Fig. 3: Dendrogram using complete linkage for women. Note that sexual interest in children clusters with same-sex attraction/fantasy.
Table 2. Legend for Figures 2 and 3
PEDO_BOYSexual Attraction to Boys less than 10 years of age
PEDO_GRLSexual Attraction to Girls less than 10 years of age
PEDO_S_FFantasized about sex with a child of the same sex
PEDO_O_FFantasized about sex with a child of the opposite sex
PEDO_ASexual attraction to young children
SAMSEX_ASexually attracted to members of the same sex
SAMSEX_FSame-sex sexual fantasy
OPPSEX_ASexually attracted to members of the opposite sex
OPPSEX_FOpposite-sex sexual fantasy

Special populations: the clergy and fashion designers

Homosexual/bisexual individuals are vastly overrepresented among child molesting clergy and top fashion designers.

The Clergy

The majority of cleric-sex offenders are homosexual pedophiles, ranging from as low as 70% to as high as 90% of all cleric-sex offenders.42, 43, 44, 45, 46

The 2004 John Jay College of Criminal Justice report on child molesting Catholic clergy revealed that 81% of the recorded molestation allegations during 1950-2002 had been of a homosexual nature; about 4% of the priests who had served during this time had been accused of child molestation.

Gay fashion designers

The fashion business is dominated by homosexual men.47 One only has to look at enough female high-fashion models(face, physique) and it would be clear that a lot of these homosexuals find adolescent boys appealing.

Similarities between male homosexual teleiophiles and male homosexual pedophiles

Clearly, there can be no disputing that there is an association between same-sex attraction and sexual interest in children or that there is a huge discrepancy between the proportion of pedophiles that are homosexual and the proportion of men that are homosexual. However, some will be quick to point out that the path to homosexual teleiophilia (sexual preference for mature adults) differs from that to homosexual pedophilia. In support of this possibility, one notes that a number of homosexual teleiophiles exhibit childhood sex-atypical (feminine) behavior, but this is not true of homosexual pedophiles.48 In addition, only homosexual teleiophiles manifest feminine identification.49 However, sex-atypical childhood behavior is far from a universal correlate of homosexuality. Male homosexuals display considerable heterogeneity with respect to feminine identification and the extent of sex-atypical childhood behaviors.50 Another difference reported between homosexual pedophiles and homosexual teleiophiles was that unlike homosexual teleiophiles, homosexual pedophiles did not appear to experience earlier puberty compared to heterosexual men,51 but there is no clear evidence of earlier puberty among male homosexuals in general.52

On the other hand, both male homosexual teleiophiles53, 54 and male homosexual/bisexual pedophiles/hebephiles55, 56 tend to be later born, especially with respect to sons compared to their heterosexual counterparts. This suggests some level of developmental pathway similarity between these two groups, a suggestion that is bolstered by the following finds:

  • Several studies have shown that mentally retarded individuals are more likely to be sex offenders than people with normal IQ.57 Sex offending on the part of mentally retarded individuals is a manifestation of deviant sexual interests rather than impulsive behavior.58 Mentally retarded male sexual offenders are more likely to offend against males than men with normal IQ.59, 60 Additionally, pedophiles tend to have a lower IQ than normal individuals.61 Furthermore, high fraternal birth order in men is associated with an elevated likelihood of mental retardation.62, 63, 64
  • It appears that there exists an association between zoophilia and mental retardation, 65 which is of some interest given the relative overrepresentation of homosexuals among zoophiles.
  • Phallometric assessment has revealed that deviant sexual interests such as sexual arousal to violence among mixed groups of male sex offenders (homosexual, heterosexual, and bisexual; offenders against adults and children) is positively related to elevated fraternal birth order.66, 67 Furthermore, in a comparison of 69 incarcerated men who had raped women and 244 men taken from the general community, which included both men who had committed sexual aggression against women and men who had not done so, fraternal birth order significantly predicted rapist vs. non-rapist group membership.68

The age preference of male homosexual pedophiles tends to cluster in the 10-12 range (DSM-IV-TR). Going by the appearance of boys in this age range and adult men, it is difficult to believe that nothing substantial pertaining to erotic target orientation is shared between male homosexual pedophiles and male homosexual teleiophiles. Those who disagree need to provide an age preference that demarcates a developmental pathway to teleiophilia from the developmental pathway to other age group preferences, and, say, if this age cutoff is 15, then offer some hypothesis regarding what developmental processes make the erotic orientation of a man with a preference for 14-year-old boys strikingly different from that of a man with a preference for 15- to 16-year-old boys.

Taxonomy considerations

Consider Table 3 below, which describes 129 convicted male child molesters in a treatment program.69

Table 3: Characteristics of 129 convicted male child molesters.
Type N P H B M > 1
Daughter molesting fathers 38   1    
Non-incestuous child molesters Prefer boys 34 11.8% 2 3 32.4%
Prefer girls 57 14% 3 4 31.5%
N = total number, P = multiple paraphilias , H = number having sex with adult men, B = number molesting both boys and girls, M > 1 = masturbation more than once a day.

Table 3 mentions a man who was arrested for molesting his daughter and also found to have had sex with men in restrooms. What is the sexual orientation of this man? To answer this question, start with the clustering of multiple unusual sexual behaviors within this group and note in particular that at least 39 (30%) indulged in recent homosexual behavior, which considerably exceeds the incidence of recent homosexual behavior in the general population.

The tendency for unusual sexual interests to cluster has been well-documented, and anyone who goes through the sexuality section here will note that this is especially true of homosexuals and bisexuals. Taxometric analyses of sexual orientation have shown that there is a taxonic demarcation between nonheterosexuals and heterosexuals, where the nonheterosexual group comprises of homosexuals, bisexuals and self-identified heterosexuals with a homosexual component to their behaviors/attractions at some point of their lives.70 Cluster analysis (Figures 2 and 3) shows that sexual interest in children is strongly associated with the nonheterosexual taxon. And, as previously mentioned, male homosexual pedophiles and male bisexual pedophiles appear to cluster into the same taxon by virtue of the fraternal birth order evidence. Therefore, one thing is clear, namely that either processes leading to same-sex attraction increase the likelihood of sexual interest in children or vice versa.

The daughter-molesting father found to be having sex with men in restrooms is clearly a nonheterosexual individual.

Groth and Birnbaum (1978) & Jenny et al. (1994)

Groth and Birnbaum (1978)71 & Jenny et al. (1994)72 are frequently cited to disprove an association between homosexuality and pedophilia/child molestation. These studies need to be addressed at some length.

Groth and Birnbaum (1978)

Groth and Birnbaum reported on a sample of 175 male child molesters selected at random (Table 4).

Table 4: Characteristics of 175 randomly selected male child molesters.   Pedophiles Non-pedophiles
Sex F M B F M B
Number having sex with children   28 35 20 65 15 12
Number also having sex with adults F * 62 5 3
M * 0 0 0
B * 3 10 9
Notes: * data not provided, F = female only, M = male only, B = both female and male.

They did not find a single man among the non-pedophile child molesters that manifested a preference for adult male but not adult female sexual partners, and claimed that “The possibility emerges that homosexuality and homosexual pedophilia may be mutually exclusive and that the adult heterosexual male constitutes a greater risk to the underage child than does the adult homosexual male.”

This conclusion does not follow from their data:

  • They did not provide data on any sexual relations that the pedophiles may have had with adults, saying that it is not applicable. Some of these pedophiles likely had had sexual relations with adults, including homosexual pedophiles that had sexual contact with adult men but not adult women.
  • The fraternal birth order evidence argues against homosexual teleiophilia and homosexual pedophilia being mutually exclusive.
  • The point about the appearance of 10- to 12-year-old boys also holds, namely that this is the age group that homosexual pedophiles largely prefer, and it is difficult to believe that most of them don't share something major pertaining to sexual orientation with homosexual teleiophiles.
  • It is clear that Groth and Birnbaum are using a strict definition of male homosexuality, i.e., no contact with women. What proportion of men satisfy this definition? Whereas up to 2% of men may self-identify as homosexual, only a small minority of them are lifetime-exclusive homosexual. For instance, in a random 1990 British sample of 8,778 men and 10,758 women, ages 16-59 years,73 whereas 5.73% of men reported any level of attraction to men, only 0.52% reported exclusive attraction to men; whereas 5.42% of men reported any level of sexual experience with men, only 0.46% reported exclusive sexual experience with men; and the correlation between reported same-sex attraction and reported homosexual behavior was 0.605 for men. In other words, there are few exclusively homosexual men and even fewer lifetime-exclusive homosexual men. A very generous estimate of the proportion of men who are lifetime-exclusive homosexual is 0.3%. If these men are overrepresented by a factor of 3 among child molesting men, then they would constitute 0.9% of child molesting men, i.e., in the sample of 92 non-pedophile child molesters reported in Table 4, one would easily not obtain a single such child molester. In other words, Groth and Birnbaum had a very inadequate sample size to address whether homosexual pedophilic and homosexual teleiophilic but not heterosexual interests can occur in the same individual.

Jenny et al. (1994)

Jenny et al. examined all sexually abused children over a period of one year at a children’s hospital. There were 352 children; 276 girls and 76 boys. 30 children had been molested by other children less than age 13, 44 children had been molested by teenagers; Jenny et al. did not provide data about the sex of these molesters. The offender could not be identified in 9 cases. Among the remaining 269 children, there were 219 girls and 50 boys. 206 girls had been molested by a man only, 8 girls had been molested by a woman only, and 5 girls had been molested by both a man and a woman; 169/219 girls, i.e., 77% of the girls had been molested by a man who was or had been in a sexual relationship with a female relative of the child. Among the 50 boys, 42 had been molested by a man only, 3 had been molested by a woman only, and 5 had been molested by both a man and a woman; 37/50 boys (74%) of the boys had been molested by a man who was or had been in a sexual relationship with a female relative of the child.

Jenny et al. obtained information about the molesters from parents, foster parents, child care workers, or sometimes the children themselves, but not by interviewing the molesters. They determined that one of the 13 women that had molested at least one girl each was a lesbian, i.e., 7.7% of the female molesters who had molested girls were lesbian. If we add the remaining 8 women who had molested boys, none of whom were identified as lesbian, 4.8% of the entire sample of female child molesters consisted of lesbians, which notably exceeds the proportion of lesbians in the general population. However, note that the proportion of women that had engaged in homosexual child molestation in the entire female child molester sample is 62%, which is considerably in excess of the proportion of women in the general population that have recently engaged in homosexual behavior. Of the 47 boys that had been molested by a man, one was identified as gay; however, note that among all the men in the entire male child molester sample, 18.2% of the molestations had been of a homosexual nature, which considerably exceeds the prevalence of recent male homosexual behavior in the general male population. In the entire sample of the child molesters, 237 (88%) were found to engage in heterosexual behavior, and in 32 cases, homosexual, bisexual, or heterosexual orientation could not be inferred. Jenny et al. concluded that children are not at a greater risk for molestation by “identifiable homosexuals.”

The shortcomings of the Jenny et al. study:

  • What about non-“identifiable homosexuals”? Are all homosexuals openly living as homosexuals? Homosexuals claim playwright Oscar Wilde as one of their own even though he was a married man with two children; Oscar Wilde had sex with several adolescent males.
  • Whereas many of the molesters that had molested a child of the same sex were involved in a heterosexual relationship with the adult parent/caretaker of the child, it is a reasonable expectation that men interested in underage boys will attempt to befriend single mothers with a son that they take fancy to. These men are clearly some type of bisexuals and may lean more toward homosexuality than heterosexuality, but this will not be clear unless they are properly examined, which Jenny et al. never did.
  • Jenny et al. wrote, “A review of the literature reveals few references to persons living a homosexual lifestyle and committing child sexual abuse,” and they supported this statement by citing a book by Finkelhor74 and a case-study by Halpern,75 which also reviewed some of the literature on this issue. The choice of citing the case-study by Halpern was remarkable given that Halpern described the case of a 16-year-old boy who had been repeatedly raped by his adoptive father since the boy was adopted at age 12 years; his adoptive father had been and was a single man who had a male partner, i.e., was a homosexual.
  • In a sample of 611 consecutively assessed sex offenders at a treatment center, there were 229 men that had molested children less than age 14 years. Erickson et al.76 reported that 70% of these men had molested girls, 26% had molested boys, and 4% had molested both boys and girls. Once again, note the elevated prevalence of recent homosexual behavior in this sample. When asked to describe their sexual orientation, 86% of the men that had molested boys described themselves as homosexual or bisexual, which clearly reveals the major shortcoming of the likes of Jenny et al., who didn't bother to assess the self-described sexual orientation or sexual desires of child molesters. Erickson et al. also noted that whereas 41.2% of the men that had molested boys also fellated their victim, only 19.2% of the men that had molested girls had performed cunnilingus (p < 0.001). Presumably, most men that had molested boys in Erickson et al.’s sample also had had sexual relations with adult women; yet even if one buys into homosexual apologetics that some heterosexual men may anally rape boys because prepubertal boys and girls are not as physically dimorphic as adult men and women and their anorectal region is similar, does one expect heterosexual men to want to fellate boys? In this regard, Spencer and Dunklee77 described a sample of 140 molested boys seen between 1979 and 1984 at a children’s hospital in San Diego. All perpetrators except two were male. 85% of the molesters were relatives or acquaintances, and only 4% of the molesters were known homosexuals. However, 46% of the perpetrators had fellated the boys (53% of the boys had been anally raped, an additional 9% were victims of attempted anal rape, 5% had foreign bodies inserted in their rectum, and 9% had feces smeared on them).

The definition of a homosexual

Some semantic issues need to be discussed. If a homosexual is defined as someone with an exclusive sexual preference for same-sex adults, then by definition, a homosexual could not be a pedophile. Alternatively, someone who insists that there is an association between homosexuality and pedophilia could be accused of arguing for an association between homosexual teleiophilia and homosexual pedophilia. Again, if homosexual teleiophilia is defined as an exclusive sexual preference for same-sex adults (as opposed to a predominant preference for adults), then by definition it could not be associated with pedophilia or child molestation.

The important issues are the phenomena that are being observed, not how one labels them. Do homosexual teleopihilic interests and homosexual pedophilic interests co-occur, and if so, do they co-occur more frequently than the co-occurrence of heterosexual teleopihilic interests and heterosexual pedophilic interests? Teleiophilic and pedophilic interests obviously co-occur in some individuals, and this co-occurrence is much more strongly characteristic of nonheterosexual individuals (Figures 2 and 3). Lifetime-exclusive homosexuals (no attraction or sexual contact with women ever) with both teleiophilic and pedophilic interests are absolutely rare because lifetime-exclusive homosexuals are rare to start with, but this does not mean that these individuals are not overrepresented -- relative to the proportion of lifetime-exclusive homosexuals in the general population -- among individuals with both teleiophilic and pedophilic interests compared to their lifetime-exclusive heterosexual counterparts.

The best hope of those arguing against an association between homosexuality and pedophilia is to either resort to semantic sophistry as in the definitions issue or steer the debate toward lifetime-exclusive homosexuals since few studies/datasets specifically address them, but one need only point out that the homosexual community itself includes as part of self a lot more people than lifetime-exclusive homosexuals, just as many academics do.

Comment: theoretical implications

The evidence for a tendency for same-sex attraction and sexual interest in children to cluster is very strong. This is the phenomenon that is of interest rather than whether there is an association between homosexuality and pedophilia since one could define a homosexual in a manner that makes it impossible for a homosexual to be a pedophile or child molester.

What accounts for the association between same-sex attraction and sexual interest in children? It is beyond the scope of this article to answer this question, but it is the same factor responsible for the clustering of atypical/paraphilic sexual interests, i.e., prenatal developmental disturbances leading to increased odds of multiple anomalous outcomes. Both sexual interest in children and same-sex attraction result from prenatal developmental disturbances.


Note for Gregory Herek: Paul Cameron is not cited.

  1. ^ DSM-IV-TRTM. (American Psychiatric Association, 2000) # 302.302.
  2. ^ Rind B. Biased use of cross-cultural and historical perspectives on male homosexuality in human sexuality textbooks. J Sex Res 1998;35(4):397-407.
  3. ^ Williams WL. The spirit and the flesh: sexual diversity in American Indian culture. Boston: Beacon Press, 1992.
  4. ^ Wikan U. Behind the veil in Arabia: women in Oman. Baltimore: Johns Hopkins University Press, 1982.
  5. ^ Norton R. The myth of the modern homosexual: queer history and the search for cultural unity: Continuum Books, 1997.
  6. ^ Reade B. Sexual heretics: male homosexuality in English literature from 1850 to 1900. New York: Coward-McCann, 1970.
  7. ^ Thorstad D. Man/Boy love and the American gay movement. J Homosex 1990;20(1-2):251-74.
  8. ^ Schuijer J. Tolerance at arm's length: the Dutch experience. J Homosex 1990;20(1-2):199-229.
  9. ^ Mohr JW, Turner RE, Jerry MB. Pedophilia and exhibitionism.: University of Toronto Press, 1964.
  10. ^ Gebhard P, Gagnon J, Pomeroy W, et al. Sex offenders: an analysis of types. New York: Harper & Row, 1965.
  11. ^ Freund K, Heasman G, Racansky IG, et al. Pedophilia and heterosexuality vs. homosexuality. J Sex Marital Ther 1984;10:193-200.
  12. ^ Freund K, Watson RJ. The Proportions of heterosexual and homosexual pedophiles among sex offenders against children: An exploratory study. J Sex Marital Ther 1992;18(1):34-43.
  13. ^ Freund K, Watson R, Rienzo D. A comparison of sex offenders against female and male minors. J Sex Marital Ther 1987;13(4):260-4.
  14. ^ Freund K, Watson RJ. Assessment of the sensitivity and specificity of a phallometric test: an update of phallometric diagnosis of pedophilia. Psychol Assess 1991;3:254-60.
  15. ^ Beier KM. Differential typology and prognosis for dissexual behavior: A follow-up study of previously expert-appraised child molesters. Int J Legal Med 1998;111(3):133-41.
  16. ^ Freund K, Watson RJ. The Proportions of heterosexual and homosexual pedophiles among sex offenders against children: An exploratory study. J Sex Marital Ther 1992;18(1):34-43.
  17. ^ Abel G, Becker J, Mittleman M, et al. Self-reported sex crimes of nonincarcerated paraphiliacs. J Interpersonal Violence 1987;2:3-25.
  18. ^ Marshall WL, Barbaree HE, Eccles A. Early onset and deviant sexuality in child molesters. J Interpersonal Violence 1991;6(3):323-36.
  19. ^ Brongersma E. Boy-lovers and their influence on boys: distorted research and anecdotal observations. J Homosex 1990;20(1-2):145-73.
  20. ^ Gordon M. Males and females as victims of childhood sexual abuse: an examination of gender effect. J Family Violence 1990;5:321-33.
  21. ^ Holmes WC, Slap GB. Sexual abuse of boys: definition, prevalence, correlates, sequelae, and management. Jama 1998;280(21):1855-62.
  22. ^ Holmes GR, Offen L, Waller G. See no evil, hear no evil, speak no evil: why do relatively few male victims of childhood sexual abuse receive help for abuse-related issues in adulthood? Clin Psychol Rev 1997;17(1):69-88.
  23. ^ Chow EW, Choy AL. Clinical characteristics and treatment response to SSRI in a female pedophile. Arch Sex Behav 2002;31(2):211-5.
  24. ^ Cooper AJ, Swaminath S, Baxter D, et al. A female sex offender with multiple paraphilias: a psychologic, physiologic (laboratory sexual arousal) and endocrine case study. Can J Psychiatry 1990;35(4):334-7.
  25. ^ Vandiver DM, Kercher G. Offender and victim characteristics of registered female sexual offenders in Texas: a proposed typology of female sexual offenders. Sex Abuse 2004;16(2):121-37.
  26. ^ Seto MC, Lalumiere ML, Blanchard R. The discriminative validity of a phallometric test for pedophilic interests among adolescent sex offenders against children. Psychol Assess 2000;12(3):319-27.
  27. ^ Seto MC, Murphy WD, Page J, et al. Detecting anomalous sexual interests in juvenile sex offenders. Ann N Y Acad Sci 2003;989:118-30; discussion 144-53.
  28. ^ Freund K, Langevin R, Cibiri S, et al. Heterosexual aversion in homosexual males. Br J Psychiatry 1973;122(567):163-9.
  29. ^ Bailey JM, Gaulin S, Agyei Y, et al. Effects of gender and sexual orientation on evolutionary relevant aspects of human mating psychology. J Person Soc Psychol 1994;66:1081-93.
  30. ^ Jankowiak W, Hill E, Donovan J. The effect of sex and sexual orientation on attractiveness judgments: an evolutionary interpretation. Ethol Sociobiol 1992;13:73-85.
  31. ^ Silverthorne ZA, Quinsey VL. Sexual Partner Age Preferences of Homosexual and Heterosexual Men and Women. Arch Sex Behav 2000;29(1):67-76.
  32. ^ Pedersen W, Hegna K. Children and adolescents who sell sex: a community study. Soc Sci Med 2003;56(1):135-47.
  33. ^ Pedersen W, Hegna K. [Children and adolescents selling sex]. Tidsskr Nor Laegeforen 2000;120(2):215-20.
  34. ^ Svedin CG, Priebe G. Selling sex in a population-based study of high school seniors in sweden: demographic and psychosocial correlates. Arch Sex Behav 2007;36(1):21-32.
  35. ^ Black C, Pokorny W, McGill C, et al. Ano-rectal trauma in children. J Pediatr Surg 1982;17:501-4.
  36. ^ Edgardh K, Ormstad K. Prevalence and characteristics of sexual abuse in a national sample of Swedish seventeen-year-old boys and girls. Acta Paediatr 2000;89(3):310-9.
  37. ^ Gorey KM, Leslie DR. The prevalence of child sexual abuse: integrative review adjustment for potential response and measurement biases. Child Abuse Negl 1997;21(4):391-8.
  38. ^ Bell A, Weinberg M. Homosexualities: a study of diversities among men and women. New York: Simon and Schuster, 1978.
  39. ^ Jay K, Young A. The gay report. New York: Summit Books, 1979.
  40. ^ Weatherburn P, Hunt AJ, Hickson FCI, et al. The sexual lifestyles of gay and bisexual men in England and Wales. London: Project SIGMA, 1992.
  41. ^ Ellis L, Robb B, Burke D. Sexual orientation in United States and Canadian college students. Arch Sex Behav 2005;34(5):569-81.
  42. ^ Haywood TW, Kravitz HM, Grossman LS, et al. Psychological aspects of sexual functioning among cleric and noncleric alleged sex offenders. Child Abuse Negl 1996;20(6):527-36.
  43. ^ Haywood TW, Kravitz HM, Wasyliw OE, et al. Cycle of abuse and psychopathology in cleric and noncleric molesters of children and adolescents. Child Abuse Negl 1996;20(12):1233-43.
  44. ^ Plante TG, Manuel G, Bryant C. Personality and cognitive functioning among hospitalized sexual offending Roman Catholic priests. Pastoral Psychology 1996;45:129-39.
  45. ^ Loftus JA, Camargo RJ. Treating the clergy. Ann Sex Res 1993;6:287-304.
  46. ^ Langevin R, Curnoe S, Bain J. A study of clerics who commit sexual offenses: are they different from other sex offenders? Child Abuse Negl 2000;24(4):535-45.
  47. ^ Goss F. Stars of style (Influential gay individuals in fashion). The Advocate. Vol 735 (June 10), 1997:27, 28, 37, 39.
  48. ^ Freund K, Blanchard R. Is the distant relationship of fathers and homosexual sons related to the sons' erotic preference for male partners, or to the sons' atypical gender identity, or to both? J Homosex 1983;9(1):7-25.
  49. ^ Freund K, Blanchard R. Feminine gender identity and physical aggressiveness in heterosexual and homosexual pedophiles. J Sex Marital Ther 1987;13(1):25-34.
  50. ^ Phillips G, Over R. Adult sexual orientation in relation to memories of childhood gender conforming and gender nonconforming behaviors. Arch Sex Behav 1992;21(6):543-58.
  51. ^ Blanchard R, Dickey R. Pubertal age in homosexual and heterosexual sexual offenders against children, pubescents, and adults. Sex Abuse 1998;10:273-82.
  52. ^ Savin-Williams RC, Ream GL. Pubertal onset and sexual orientation in an adolescent national probability sample. Arch Sex Behav 2006;35(3):279-86.
  53. ^ Blanchard R. Quantitative and theoretical analyses of the relation between older brothers and homosexuality in men. J Theor Biol 2004;230(2):173-87.
  54. ^ Blanchard R, Lippa RA. Birth order, sibling sex ratio, handedness, and sexual orientation of male and female participants in a BBC internet research project. Arch Sex Behav 2007;36(2):163-76.
  55. ^ Blanchard R, Barbaree HE, Bogaert AF, et al. Fraternal birth order and sexual orientation in pedophiles. Arch Sex Behav 2000;29(5):463-78.
  56. ^ Bogaert AF, Bezeau S, Kuban M, et al. Pedophilia, sexual orientation, and birth order. J Abnorm Psychol 1997;106(2):331-5.
  57. ^ Lindsay WR. Research and literature on sex offenders with intellectual and developmental disabilities. J Intellect Disabil Res 2002;46(suppl I):74-85.
  58. ^ Parry CJ, Lindsay WR. Impulsiveness as a factor in sexual offending by people with mild intellectual disability. J Intellect Disabil Res 2003;47(Pt 6):483-7.
  59. ^ Brown H, Stein J. Sexual abuse perpetrated by men with intellectual disabilities: a comparative study. J Intellect Disabil Res 1997;41 ( Pt 3):215-24.
  60. ^ Blanchard R, Watson MS, Choy A, et al. Pedophiles: mental retardation, maternal age, and sexual orientation. Arch Sex Behav 1999;28(2):111-27.
  61. ^ Cantor JM, Blanchard R, Christensen BK, et al. Intelligence, memory, and handedness in pedophilia. Neuropsychology 2004;18(1):3-14.
  62. ^ Ackerman PT, Goolsby CM, Paal NP. A test of the immunoreactive theory of selective male affliction. J Pediatr Psychol 1988;13(1):49-53.
  63. ^ Lord C. Birth order effects on nonverbal IQ in families with multiple incidence of autism or pervasive developmental disorder. J Autism Dev Disord 1992;22(4):663-6.
  64. ^ Flannery KA, Liederman J. A test of the immunoreactive theory of the origin of neurodevelopmental disorders: Is there an antecedent brother effect? Dev Neuropsychol 1994;10:481-92.
  65. ^ Duffield G, Hassiotis A, Vizard E. Zoophilia in young sexual abusers. J Forensic Psychiatry 1998;9:294–304.
  66. ^ Côté K, Earls CM, Lalumière ML. Birth order, birth interval, and deviant sexual preferences among sex offenders. Sex Abuse 2002;14(1):67-81.
  67. ^ Lalumière ML, Harris GT, Quinsey VL, et al. Sexual deviance and number of older brothers among sexual offenders. Sex Abuse 1998;10(1):5-15.
  68. ^ Lalumiere ML, Quinsey V, Harris G, et al. Birth order and rape. Paper presented at the Annual Forensic Day of the University of Toronto. Penetanguishene, ON, Canada, 2000.
  69. ^ Marshall WL, Barbaree HE, Eccles A. Early onset and deviant sexuality in child molesters. J Interpersonal Violence 1991;6(3):323-36.
  70. ^ Gangestad SW, Bailey JM, Martin NG. Taxometric analyses of sexual orientation and gender identity. J Pers Soc Psychol 2000;78(6):1109-21.
  71. ^ Groth AN, Birnbaum HJ. Adult sexual orientation and attraction to underage persons. Arch Sex Behav 1978;7(3):175-81.
  72. ^ Jenny C, Roesler TA, Poyer KL. Are children at risk for sexual abuse by homosexuals? Pediatrics 1994;94(1):41-4.
  73. ^ Johnson AM, Wadsworth J, Wellings K, Field J. Sexual Attitudes and Lifestyles. (Blackwell Scientific Publications, Oxford, 1995).
  74. ^ Finkelhor D. Child sex abuse: new theory and research. New York: Free Press, 1984;47.
  75. ^ Halpern J. Family therapy in father-son incest: a case study. Social Casework 1987;68:88-93.
  76. ^ Erickson WD, Walbek NH, Seely RK. Behavior patterns of child molesters. Arch Sex Behav 1988;17(1):77-86.
  77. ^ Spencer MJ, Dunklee P. Sexual abuse of boys. Pediatrics 1986;78(1):133-8.