concurring specially.
Although I concur fully with Division 2, I write separately to specially concur with Division 1.
*510In Tyler v. State,14 we recognized that evidence that a defendant charged with child molestation was himself molested as a child is both irrelevant and prejudicial.15 It is prejudicial because it permits the jury to infer that “any molestation of [the defendant] as a child would have increased his propensity to become a child molester.”16
In this case, the trial court allowed the prosecutor, over objection, to question Osmer about the fact that he had been molested as a child by his stepfather. Notwithstanding our holding in Tyler, the majority concludes that the trial court did not err because (1) evidence of Osmer’s history of molestation was relevant; and (2) the prosecutor did not elicit hearsay testimony about matters not in evidence, as occurred in Tyler. The majority also concludes that any error in allowing the prosecutor to question Osmer about his history of molestation was harmless because such evidence was cumulative. I disagree with the majority’s conclusion that no error occurred. Nevertheless, under the facts of this case, I am compelled to conclude that the error was harmless.
The trial court should not have allowed the questioning at issue. Evidence of Osmer’s history of molestation was not relevant. The state argues that such evidence showed that Osmer knew that he could have stopped the abuse of his female cousin (the victim’s younger sister) by reporting it to the proper authorities, because Osmer’s own abuse ended after he reported it to a school counselor years before. Thus, the state contends, such evidence was relevant to undermine Osmer’s credibility by refuting his theory that his victim made up the molestation allegations to retaliate against him for speaking out about the victim’s molestation of the victim’s younger sister. But the salient point is that Osmer knew how to report allegations of child molestation, not that he had been molested as a child. The prosecutor could have elicited testimony that Osmer had previously made a report of child molestation without delving into his personal history of molestation.
And even though this case did not involve the prosecutor’s intentional eliciting of hearsay testimony from a state’s witness, as did Tyler, the fact remains that evidence of a defendant’s history of child molestation is irrelevant and prejudicial, no matter how the prosecutor introduces that history into evidence. The majority opinion obscures that fact.
The trial court’s error in allowing the evidence, however, is not reversible. As the majority points out, a defense witness already had *511referred, without objection, to Osmer’s history of molestation. During cross-examination of Osmer’s friend Loden, the prosecutor asked, “In the course of you disclosing that you had been molested, did [Osmer] disclose that he had been molested?” Loden responded, “Yes.” Osmer did not object to this testimony. Later, on direct examination, Osmer testified he and Loden “talked about some things that me and her had in common.” Because this statement immediately followed Osmer’s observation that Loden had been molested as a child, it seemed to allude to Osmer’s own history of molestation.
Decided September 15, 2005. Berry & Reynolds, D. Victor Reynolds, for appellant. James R. Osborne, District Attorney, Fred A. Lane, Jr., William M. Clark, Assistant District Attorneys, for appellee.In addition, the evidence of Osmer’s guilt, although not overwhelming, appears to have been stronger than the evidence against the defendant in Tyler. Although the case ultimately boiled down to a credibility contest between Osmer and the victim, multiple state witnesses corroborated various aspects of the victim’s testimony, and there were inconsistencies in Osmer’s testimony. Finally, the prosecutor did not argue to the jury that Osmer’s history of molestation increased the likelihood that he would become a child molester. Instead, the prosecutor contended that Osmer was not a credible witness and that his version of the events was less believable than the victim’s. Under these circumstances — where the prejudicial evidence was cumulative, the victim’s testimony was corroborated, and the prosecutor did not draw improper inferences from the evidence — it is “highly probable that the error did not contribute to the judgment.”17
266 Ga. App. 221 (596 SE2d 651) (2004).
Id. at 223 (2).
Id.
Johnson v. State, 238 Ga. 59, 61 (230 SE2d 869) (1976) (citation and punctuation omitted) (setting forth “highly probable test” as standard for determining whether nonconstitutional error was harmless).