The New York Times reports today on research that demonstrates a very high correlation between use of child pornography and the actual molesting of children. The Times did a good job of reporting why it is so important to be cautious about interpreting a study like this one. And it also does a good job of reporting on the need for continued research on child molestation.
Because of the tremendous moral panic risks that are attached to publishing anything about htis kind of research I am going to focus entirely on the cautions. There will be lots of voices out there focusing on the tentative conclusions of the study itself, so here lets just focus on the limitations:
1. Remember when thinking about these results that they were produced using only already-incarcerated men convicted of child pornography charges. These men may well not be representative of all people who have ever downloaded or viewed child pornography.
2. The men who were studied were not only incarcerated, they were voluntary participants in treatment programs for sex offenders. It is quite reasonable to ask whether men who volunteer for sex offender treatment are like other users of child pornography. There are several ways in which they could be different. They could be more likely to be men who had in fact molested children and thus believed they could benefit from treatment, for example.
3. The Times reported that the study found that 85% of the child-porn convicts in their sample also admitted to “acts of sexual abuse with minors,from inappropriate touching to rape”. But we can’t tell what to make of this statistic. We don’t know whether the use of child pornography came after the acts of sexual contact with kids or before it. (The study has been at least temporarily blocked from publication by the Federal Bureau of Prisons whose psychologists conducted the research, so we can’t yet evaluate it in its entirely.) Given the lack of complete information, it would be dangerous to interpret the statistic reported in the Times. Correlations are notoriously misinterpretable. For example, what if there is a correlation between use of child pornography and likelihood of molesting children. Does it matter whether it is the inclination toward molesting children that causes the use of child pornography, or whether it is the use of the child pornography that causes the molestation? Of course it does. It also matters whether there is some external variable that causes a person to be inclined toward both of those other activities.
The limitations of the study that the Times reported today should not be cause for putting down the research itself. Rather, they should be used as a guide for interpreting the findings and for highlighting where more study needs to be done.
The Times has, in the past, discussed the difficulties with studying sex offenders. While some of the challenges are methodological, and some are ethical, in an article published in March, a professor from a law school in Minnesota pointed out that some are cultural:
Professor Janus said he hoped for “an explosion of knowledge” about how to prevent sexual violence before it happened, which he said would prevent far more sex crimes than civilly committing offenders.
That sort of research is unlikely to happen in the United States, Dr. Berlin and other experts said, because so many Americans believe that the only investment in sex offenders should be punitive.
Research on sex offenders, on their treatment, and on preventing sex crimes is all very important and needs to be encouraged. It’s difficult to encourage research in an environment like ours, where findings — whatever they indicate — are so potentially explosive because of the moral panic that characterizes our approach to policy around kids and sex and crime. If as Dr. Berlin suggests many of us believe that punishment is the only thing to consider when we address sex offenders, we will never get any clear understanding of how to prevent those crimes in the first place. Such an attitude essentially guarantees that more kids will be harmed and more adults will become criminals.
In encouraging more research on sex offenders and sex crimes, we need to keep the following goals in mind:
1. To develop prevention strategies that work so that harm is avoided in the first place.
2. To develop treatment strategies that work to reduce rates of reoffending.
3. To better understand adult sexuality, childhood and teen sexuality, and to better understand consent so that we can distinguish between criminal acts with real victims, and loving, affectionate or simply playful acts that harm no one.
This last is a controversial goal to be sure. When Bruce Rind and colleagues published an article in Psychological Bulletin (a peer-reviewed and widely respected academic journal) in which they found that not every instance of sex between a child and an adult caused harm to the child, they were the subject of a firestorm that even led to their being “unanimously condemned by Congress.” And when Judith Levine published Harmful to Minors, perhaps the clearest discussion of kids, sex, and policy out there, she writes that “overnight I became the author of ‘the pedophilia book,’ even though the book only touches on pedophilia in a few of its 300+ pages. University of Minnesota Press, which published the first edition, was overwhelmed with calls “demanding that the press’s management resign and Harmful to Minors — and maybe its author — be burned.” (p. 229, Afterword, Harmful to Minors, 2002 edition.) The book went on to win the 2002 Los Angeles Times Book Prize and its 2002 edition, published by Thunder’s Mouth Press, has a foreword by Dr. Joycelyn Elders.
Yet this last goal is ultimately important if we are to avoid the kinds of harm we cause to children, teens, and adults when we make policy based on fear rather than on evidence. Prevention, treatment, and a clearer understanding of the sexuality of kids, teens, and adults are all essential if we’re going to get a handle on sex crimes.