These findings appear in “Disclosure of Child Sexual Abuse: What Does the Research Tell Us About the Ways That Children Tell?”, by Kamala London, Maggie Bruck, Stephen J. Ceci, and Daniel W. Shuman. Their article appears in the March 2005 issue of Psychology, Public Policy and the Law, which may be found here in Adobe PDF format.

Rather than conducting fresh experiments, the authors analyzed 17 studies bearing relevant data.

The authors trace the popularity of the notion that molested children commonly deny or recant to the 1983 publication of the Child Sexual Abuse Accommodation Syndrome, by psychiatrist Roland Summit. Dr. Summit did not intend that CSAAS be a template for diagnosing molestation, but, as the authors note, many child professionals have used it as one. The authors also note that many courts have admitted such diagnoses in evidence, even though they have no sound scientific basis.

The authors acknowledge the impossibility of knowing whether a particular child has been sexually abused. (They suggest that some child abuse researchers have concluded that abused children are prone to deny their abuse or to recant because they have categorized as victims children who, in fact, were never molested.) They also do not dispute that many children who are molested never say so because they never undergo an investigative interview.