Child molestation has been much in the news in the past few months, and as always in such debates, the issue of homosexuality is never far from the surface.  For decades, conservative activists have argued that homosexual behavior is closely related to molestation and pedophilia, so that tolerating homosexuals ultimately endangers children.  According to the slogan, “Homosexuals aren’t born, they recruit.”  Homosexual activists angrily deny these charges.

Recently, though, the substance of the debate has changed significantly, as the homosexual rebuttal has become much more aggressive.  As expressed in op-ed pieces and talk shows across the country, a new liberal orthodoxy not only denies the gay/pedophile link but explicitly charges that pedophiles are virtually always heterosexual, even when they victimize boys.  When pressed on this point, activists respond that this is something that “studies have shown.”  In fact, studies have shown no such thing, and only one significant study has even claimed to prove such a point.  What we have here is a classic example of a popular myth drawn from the misinterpretation of social science.

The legend grew out of an article in Pediatrics magazine in 1994 by Carole Jenny, Thomas Roesler, and Kimberly J.  Poyer, who wrote a study entitled “Are children at risk for sexual abuse by homosexuals?”  Based on an examination of 352 young children who had been diagnosed as sexually abused, the authors answered that question with a resounding “No.”  In only two of these cases was the perpetrator clearly a homosexual—a homosexual man and a lesbian, respectively.  Overall, the authors found that the children studied “were unlikely to have been molested by identifiably gay or lesbian people.”  As the study entered popular discourse, however, the claims based on it expanded somewhat, to boast that pedophiles were virtually never homosexuals.  In this case, it was necessary to determine why heterosexuals were so likely to abuse: Just what kind of pathology followed from that deeply unnatural state?  I have not yet seen the further claim that “heterosexuals recruit”—but it should not be too far down the road.

Unfortunately, the famous “Jenny Study” is deeply flawed, at least in what it can possibly tell us about any linkage between homosexuality and molestation.  The problems of the study should be immediately apparent from the sample used.  Reviewing their cases, the authors found 269 children who had clearly been abused by adults, including 219 girls and 50 boys.  Right there, we know there is something wrong, since no study of molestation victims has ever suggested such a large disparity between the sexes.  Probably as many boys as girls are victims of abuse, and the hospitals and clinics used by Jenny and the others must have been disproportionately treating female victims, while missing boys.  Many reasons can be suggested for this fact—perhaps the doctors referring children were better able to recognize abuse in girls, perhaps parents were more sensitive to the dangers faced by their daughters—but whatever the reasons, the boys are lacking.  Since men are far more likely to abuse than women, this logically means that the sample is also ignoring the men who abused these boys.  If men abuse little girls, we call them “heterosexual pedophiles”; if men abuse little boys, we call them “homosexual pedophiles.”  By definition, the study is simply failing to detect a large number of homosexual pedophiles, a type of deviant that undoubtedly does exist.

Though this logic seems unassailable to me, it would not be accepted by Jenny et al., whose article is frankly presented as a polemical contribution to political debates over homosexual rights.  Even in the cases in which men clearly abused boys, the authors still deny that this act involved any element of homosexuality.  In their view, the term “homosexual” (or “lesbian”) can only be applied to individuals who are clearly and identifiably “out”: It reflects social status rather than behavior.  With such an incredibly narrow definition, and such a flawed sample to begin with, it is remarkable that Jenny and Co. were prepared to admit the possibility of any homosexual misconduct with minors occurring at all.

Based on the Jenny Study, we can say absolutely nothing positive or negative about any possible association between homosexuality and child molestation.  My impression of the evidence accumulated over the years is that there is no terribly direct linkage either way, that homosexuals are neither more nor less likely than heterosexuals to molest.  A desire to avoid false charges against homosexuals, however, should not lead us to the opposite extreme, to favor an exoneration that is ultimately derived from deeply flawed social science.  If you ever hear the “Jenny Legend” in the media, recognize what it is, and be prepared to confront it.