People v. Peterson

450 Mich. 349 (1995)

PEOPLE v. PETERSON PEOPLE v. SMITH

Docket Nos. 98941, 99981, (Calendar Nos. 12-13).

Supreme Court of Michigan.

Argued April 5, 1995.

Decided August 22, 1995.

Frank J. Kelley, Attorney General, Thomas L. Casey, Solicitor General, Joseph P. Kwiatkowski, Prosecuting Attorney, and J. Ronald Kaplansky, Assistant Attorney General, for the people in Peterson.

*352 Frank J. Kelley, Attorney General, Thomas L. Casey, Solicitor General, Charles H. Koop, Prosecuting Attorney, and William E. Molner, Assistant Attorney General, for the people in Smith.

Jeanice Dagher-Margosian for Peterson.

Earl R. Spuhler for Smith.

Amicus Curiae:

Frank J. Kelley, Attorney General, Thomas L. Casey, Solicitor General, Donald Martin, President, John D. O'Hair, Prosecuting Attorney, and Timothy A. Baughman, Chief, Research, Training and Appeals, for the Prosecuting Attorneys Association of Michigan.

MALLETT, J.

In these consolidated cases, we are asked to revisit our decision in People v Beckley, 434 Mich 691; 465 NW2d 391 (1990), and determine the proper scope of expert testimony in childhood sexual abuse cases. The question that arises in such cases is how a trial court must limit the testimony of experts while crafting a fair and equitable solution to the credibility contests that inevitably arise. As a threshold matter, we reaffirm our holding in Beckley that (1) an expert may not testify that the sexual abuse occurred, (2) an expert may not vouch for the veracity of a victim, and (3) an expert may not testify whether the defendant is guilty. However, we clarify our decision in Beckley and now hold that (1) an expert may testify in the prosecution's case in chief regarding typical and relevant symptoms of child sexual abuse for the sole purpose of explaining a victim's specific behavior that might be incorrectly construed by the jury as inconsistent with that of an actual abuse victim, and (2) an *353 expert may testify with regard to the consistencies between the behavior of the particular victim and other victims of child sexual abuse to rebut an attack on the victim's credibility.

Therefore, for the reasons set forth below, in Peterson, we find that the trial court erred in allowing certain aspects of the expert testimony to be presented to the jury. However, in light of the overwhelming evidence against Peterson, we find that any error was harmless. In Smith, we find no error and therefore affirm the decision of the Court of Appeals in both cases.

I

PEOPLE v PETERSON

The victim in this case was the defendant's daughter. She lived with the defendant and her mother and siblings in rural Cheboygan County. The victim testified about incidents of anal intercourse, cunnilingus, and fellatio perpetrated on her by the defendant in 1986 and 1987. She said the incidents took place "about five or six times." The jury found the defendant guilty of first-degree criminal sexual conduct.[1]

The victim was the first prosecution witness. She was eleven years old at the time. She drew on a chalkboard and described various incidents of how the defendant performed sexual acts on her and directed her to perform fellatio on him.

The victim also testified that at some point, she reported these incidents to her mother, who reportedly did not believe her. She testified that she eventually told her foster parent, Thomas O'Melia, about a month after she began living with his family.

*354 The defendant testified on his own behalf and denied ever sexually abusing the victim. The picture he painted of his family life was a very unstable one. The defendant acknowledged having various fights with his wife that resulted in him leaving the area for weeks or months at a time. The defendant was arrested after someone made an anonymous call to the police indicating that there was a Cheboygan County warrant for the defendant's arrest for CSC I.

During the trial, five prosecution expert witnesses testified. For present purposes, the testimony of the last two witnesses is less important than that of the first three. The latter two were Drs. David Hickok and Anselma Ramilo. Dr. Ramilo physically examined the complainant about a year and a half after the alleged criminal sexual conduct. He found that she had a large vaginal opening for a child of her age and that, although there were no scars or gaping of her anal canal, there was some irregularity of the fold immediately surrounding her anal opening. Dr. Hickok opined that a tongue, finger, or penis could have caused findings consistent with the dimension of the child's genitals. He noted the obliteration of the hymen.

Dr. Ramilo testified that she examined the victim on the basis of a Department of Social Services referral that was made pursuant to allegations of sexual abuse. She did not find any physical evidence of sexual abuse, but said she strongly believed the victim had been sexually abused on the basis of the contents of the medical history in which she described in detail the anal intercourse and that the person "put his penis into her mouth and white junk came out."

The prosecution's other three experts were Kathy Jo Gillan, a social worker, Margaret *355 Green, a social worker, and Thomas O'Melia, a clinical psychologist who was also the victim's foster parent. The experts' testimony was presented during the prosecution's case in chief.

Gillan testified regarding the profile of a sexual abuse victim. She also described the existence of a prototype for sexual assault perpetrators, and said that such individuals often display a symptomatology of denial. She was permitted to testify about several studies that indicate that children lie about sexual abuse at a rate of about two percent. Gillan described how she worked directly with the Peterson family to assess the extent of their dysfunction and to educate them about child abuse. Those contacts occurred between June and December 1987. Over defense objection, Gillan was permitted to testify that the victim showed behavior manifestations that were symptomatic of sexual abuse.

Green testified that she had provided psychotherapy for the victim. She said she had treated about a hundred children who were sexually abused, and that one of the ways that she verified their stories was to look for reenactments in play therapy, such as drawings, and that the victim in this case engaged in this activity. She testified that she came into contact with the victim after the victim was identified as a sexual abuse victim. The trial judge overruled the defendant's objection to Green's testimony regarding whether the victim's behavior was consistent with children who have been sexually abused.

O'Melia was called by the prosecution as an expert witness on child sexual abuse. He stated that the victim lived with him in foster care, along with her younger brother, for about seven months ending in July, 1988. He testified that she made disclosures to him about sexual abuse on four *356 occasions, the first in May, 1988. O'Melia stated that there were prototypes for both sexual abuse victims and perpetrators. He stated that the symptomatology includes most of the same factors mentioned by the two previous experts. Over defense objection, O'Melia was permitted to testify that the victim's symptoms were consistent with those of a sexual abuse victim.

Again, over defense objection, O'Melia was also permitted to testify that, of the cases and studies he was familiar with, there is about an eighty-five percent rate of veracity among child abuse victims. He testified that of the last twenty cases he had worked on, "the sexual abuse has also been confirmed by the perpetrator" in about eighty percent of the cases, but he concluded his testimony by stating that perpetrators generally do not admit sexually abusive conduct, primarily because of fear of prosecution.

After the testimony of O'Melia, defense counsel requested that he be allowed to bring in evidence of sexual touchings of the victim. The defense's argument was that, in light of all the evidence of sexual abuse behaviors and symptomatology put forth by the prosecution, defendant was entitled to show that there may be other reasons for her to experience that behaviors similar to other abuse victims. Although the trial judge initially granted that motion, when defendant later attempted to call the victim to examine her concerning those incidents, he denied the request.

In her closing argument, the assistant prosecutor referred to the case as a "classic one on one credibility test." She emphasized that the victim exhibited the symptoms delineated by all expert witnesses as being indicative of sexual molestation. She argued that a child of this age could not have made up a story in such graphic detail. On rebuttal *357 closing, the assistant prosecutor emphasized that there is a low rate of lying among children regarding allegations of sexual abuse. Defense objection was overruled. The defendant was subsequently convicted and sentenced to a twenty- to thirty-year term.

Defendant appealed to the Court of Appeals. In a two to one unpublished decision, the panel majority remanded the case for a determination whether the expert testimony "was admitted consistent with the requirements of Beckley, supra, and if not, whether any error was harmless." Dissenting Judge NEFF thought a reversal and new trial were in order because the prosecution shifted the burden of proof to the defendant in dwelling impermissibly on the assertion that child sexual abusers deny their guilt. Issued February 11, 1993 (Docket No. 123450), slip op at 2.

On remand, the trial judge concluded that Beckley was not violated.

The case was remanded to the Court of Appeals, and, in its second unpublished opinion, the panel affirmed in a two to one decision. The majority found no error requiring reversal under Beckley on these facts. Dissenting Judge NEFF incorporated by reference her earlier reasons for concluding that a reversal and new trial were warranted. Issued February 3, 1994 (Docket No. 123450). Defendant filed the present timely application for leave to appeal. 447 Mich 1041 (1994).

PEOPLE v SMITH

In 1986, the seven-year-old victim lived with her mother and her stepfather, the defendant, in an apartment in Mancelona, Michigan. The period of marital cohabitation was short lived. Several years later, the victim became the focus of protective *358 services inquiries when it was suspected that her natural father was abusing her sexually. Her natural father had shown her pornographic videos. During the course of inquiries from protective services about those activities, the victim related several incidents in which the defendant forced her to engage in fellatio. Additionally, she related that the defendant made her pose naked while she was holding her genitals. She reported that pictures were taken with a Polaroid camera.

The defendant was charged with four counts of CSC I,[2] four counts of child sexually abusive commercial activity,[3] and with being a fourth-felony offender.[4] A two-day jury trial took place in August, 1991. The defendant was found guilty on all counts. He then pleaded guilty of being a fourth-felony offender.

The trial judge imposed a life term on the defendant, noting that he had been previously convicted of CSC involving another stepdaughter, as well as a manslaughter conviction involving the death of the defendant's six-month-old son. There was also a previous conviction for malicious destruction of property over $100. The Court of Appeals unanimously affirmed.

The victim's testimony was direct and offered explicit descriptions of the defendant's activities forcing her to perform fellatio and to pose for photographs. No questions were asked of her regarding the delay in reporting the activity.

Defense counsel's ostensible strategy was not to ask the victim any questions that would have brought into issue her behavior that the jury would arguably consider to be inconsistent with the behavior of an abused person. The prosecuting *359 attorney took the position that the questions of defense counsel during voir dire (generally raising the issue of credibility of children) and the nature of such allegations in general were sufficient to make expert testimony about the victim's behavior more probative than prejudicial. In resolving the issue, the judge cited Beckley and stated:

Here it is the Court's view that while the issue of delay has not been argued by the Defendant, there is a common misperception that delay is inconsistent with the commission delay of reporting would be inconsistent with the commission of the offense and, therefore, inconsistent with the behavior of one who was actually abused. The Court does not see and the People are apparently no longer interested in offering any testimony regarding a behavioral relationship between observing films and reporting promptly or not, and that request has been withdrawn and the Court did not find that to be of merit, in any event. The Court will allow limited testimony solely dealing with the issue of delayed reporting by one child abuse expert in this case. The Court has clearly been cautioned about the use of such testimony in the majority opinion offered by Justice BRICKLEY, and while the Court recognizes there are various concurring opinions in this case as well as a dissent, there certainly is a recognition by the entire court of the caution which must be used in introducing such testimony.

The judge agreed with defense counsel that there would be no testimony from the expert about the credibility of the complainant.

The expert witness, Barbara Cross, was called to testify. After being qualified as an expert, she was asked whether the several-year delay between the occurrence of the alleged CSC and the victim's telling anybody about it had any significance. The expert testified that there are a variety of explanations *360 regarding why a particular victim might delay reporting, which include the victim's relationship with the perpetrator, the living arrangements, the age of the victim, coercive threats, and childhood fears.[5]

*361 The defendant testified on his own behalf. He denied having had the opportunity to sexually abuse the victim and explained that he and his wife worked the same shifts at a nearby resort. He denied taking any pictures of the victim. Cross-examination brought only admissions that she had come into the bathroom while the defendant was showering. He denied ever having taken the victim for a ride and telling her that if she ever mentioned the sexual activity, he would drop her off and not pick her up. However, that testimony was impeached with testimony from the victim's brother who recalled such an incident.

II

Pursuant to MRE 702, expert testimony will be allowed as follows:

*362 If the court determines that recognized scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.[[6]]

In child sexual abuse cases, we have previously determined that expert testimony concerning syndrome evidence is sometimes necessary to explain behavioral signs that may confuse a jury so that it believes that the victim's behavior is inconsistent with that of an ordinary victim of child sexual abuse.[7] However, we note from the outset that we *363 do not endorse or adopt the use of the term "syndrome." As stated in Beckley, "the admissibility of syndrome evidence is limited to a description of the uniqueness of a specific behavior brought out at trial." Id. at 725 (opinion of BRICKLEY, J.). See also People v Christel, 449 Mich 578, 591; 537 NW2d 194 (1995).

It is undisputed that the expert witness' testimony in each of these cases "assist[ed] the trier of fact to understand the evidence or to determine a fact in issue" in accordance with MRE 702. However, to be admissible under MRE 702, the testimony must first be determined to be legally relevant. MRE 402.[8]

People v Beckley and People v Badour were consolidated to consider whether evidence of specific behaviors common in abuse victims is admissible in child sexual abuse cases in order to rebut the inference that the victim's behavior was inconsistent with that of an actual sexual abuse victim. In Beckley, the fifteen-year-old victim did not reveal that her father engaged in sexual intercourse with her until a year later. At that time, she wrote about the incident in a high school journal. She was cross-examined about her delay in disclosing the assault, her disclosure to a third party outside the family, her continual contact with her father, and her initial reporting only of "passes." An expert witness subsequently testified that these were typical behavioral characteristics of a victim of sexual abuse. On cross-examination, defense counsel asked the expert about the complainant's failure to remember certain conversations. The *364 expert said that these inconsistencies were not necessarily indicative of lying, but were an attempt by the victim to minimize the event. On redirect examination, the witness elaborated on the indicators of sexual abuse. The testimony was not limited to the specific behavioral characteristics that had been attacked by the defendant on initial cross-examination of the complainant. The Court of Appeals and this Court affirmed the conviction.

In Badour, the defendant was accused of forcing her six-year-old daughter to perform sexual acts with the defendant's boyfriend. The defendant's involvement did not come to light until six months after the child revealed the boyfriend's assault. The revelation came during a group counseling session. A psychiatrist testified for the prosecution regarding sexually abused children generally, and specifically with regard to observable behavior of the complainant. On cross-examination, defense counsel tried to get the witness to agree that the symptoms the child was experiencing were the result of being placed in foster care, and that the child was lying in an attempt to get back at her mother for breaking up with the boyfriend. The witness agreed that she could not rule out the possibility of a lie. On redirect examination, however, the witness testified that children have no basic knowledge of sexual acts and could not invent such an act without experiencing it. A second expert witness, who testified over defense objections, corroborated the first expert's testimony regarding fabrication. The Court of Appeals affirmed the conviction; however, this Court reversed.

The lead opinion in Beckley-Badour was authored by Justice BRICKLEY, and was signed by Justices LEVIN and GRIFFIN. Justice BOYLE wrote a *365 separate concurrence, which was signed by Chief Justice RILEY. Justice ARCHER, along with Justice CAVANAGH, concurred in the result in Badour (a new trial), but dissented in Beckley (in which a majority affirmed).

The lead opinion stated that

(1) a properly qualified expert witness may testify regarding characteristics of sexually abused children, "`for the narrow purpose of rebutting an inference that a complainant's postincident behavior was inconsistent with that of an actual victim of sexual abuse, incest or rape.'" Id. at 710.

(2) expert testimony is admissible only to the extent that it is directed toward providing an explanation of specific behavior. The purpose of the testimony is to provide the jury with background information that it could not otherwise bring to its evaluation of the child's credibility. Id. at 728.

(3) as "long as the purpose of the evidence is merely to offer an explanation for certain behavior, the Davis/Frye[[9]] test is inapplicable." Syndrome evidence is not a technique or principle that can predict abuse, but merely an expert's opinion that explains and describes probable responses to a traumatic event. Id. at 721.

(4) an expert witness may not render a legal conclusion regarding whether sexual abuse actually occurred. Id. at 725.

In writing to affirm in Beckley, Justice BRICKLEY noted that the trial court permitted expert testimony because "it ... appear[ed] that defendant would raise these issues by attack on the credibility of the complainant...." Id. at 700. The trial court limited the testimony to whether any of the *366 observed and relevant behavior was consistent with the profile of an incest victim. Although the jury later was exposed to a wide range of general "syndrome" characteristics, it was defense counsel who opened the door by inquiring whether there was "anything that is in fact inconsistent with sexual abuse" and whether the witness had seen "any of those things" in the case. Id. at 731. Thus, had the prosecution introduced the evidence regarding any consistencies with sexual abuse victims in its case in chief, such testimony would have been deemed improper.

In writing to reverse in Badour, Justice BRICKLEY noted that the trial court permitted the expert witness to "make observations and give ... their conclusions about those observations, even to the degree that, in their opinion, these children, or this child, showed indicia of being sexually abused." Id. at 732. In other words, the trial court allowed expert testimony without consideration whether the behavior of the victim was at issue, and made no determination regarding whether the evidence would be helpful to the jury. Because the testimony was not limited to background information, the expert's role became that of advocate rather than advisor.

Justice BOYLE agreed with Justice BRICKLEY

to the extent that [the opinion] holds 1) that the Davis/Frye test is inapplicable to the expert testimony in question, 2) that syndrome evidence is not admissible to prove that sexual abuse occurred, 3) that an expert may not testify that a child is telling the truth, 4) that an expert may testify that the behavior of the complainant in the particular case is consistent with that of children who report sexual abuse. [Beckley at 734.][[10]]

*367 Justice BOYLE wrote separately because of her concern "that the rationale employed in the lead opinion may create restrictions on the use of expert testimony in child sexual abuse cases that unnecessarily limit an expert's ability to assist the factfinder." Id. at 734. Justice BOYLE agreed that expert testimony should be admitted only where it is relevant and would help the jury evaluate the child's credibility. However, she endorsed a broader view of how such evidence becomes relevant in the first place:

Among the many ways in which relevance may be made clear are motions in limine, voir dire, opening statement, the child's direct examination, the testimony of other prosecution witnesses, cross-examination of the child, or by the defendant's proofs. [Id. at 734.]

Justice BOYLE disagreed with the lead opinion that the expert testimony in child sexual abuse cases properly is characterized as "rebuttal" or "rehabilitative." She argued that when credibility is at issue, expert testimony is admissible as substantive evidence if it will assist the jury and would not relegate the expert to an "advisory" role:

The limitation suggested by the lead opinion's reasoning is inconsistent with its ... finding that expert testimony in child abuse cases is admissible to help the factfinder, since the jury will presumably be most helped by an expert who is allowed to apply expert knowledge to the case at hand. [Id. at 739-740.]

*368 Justice BOYLE agreed with the affirmance in Beckley, saying that the case "demonstrates a near-perfect model for proper procedure in the use of such evidence." Id. at 742. She also agreed with the reversal in Badour. However, her agreement in that case was based on the fact that the defendant objected to the testimony on the ground that it was being offered to show that the child had been abused, and that the prosecutor's purpose in using the testimony was to show the defendant's participation. Although defense counsel later raised a question that might have put the complainant's credibility at issue, and declined a limiting instruction, reversal still is required because of the child's halting and incomplete testimony and the fact that the expert testimony was directed at the fact of abuse and the defendant's involvement. Beckley at 741.

Justice ARCHER agreed with the reversal in Badour, but dissented in Beckley "[b]ecause the lead opinion would permit an expert to refer to the complainant in this particular case and does not limit the expert's testimony to a discussion of the specific postincident behavior traits at issue without reference to the complainant or the specific facts of this case...." Id. at 744-745. Justice ARCHER said that "the danger is too great that the trier of fact will improperly infer that an expert testifying on the basis of syndrome evidence is, in effect, concluding that the particular complainant before the court has been abused." Id. at 747.

In Justice ARCHER'S view, the question in Beckley that required reversal came during the prosecutor's direct questioning of the expert witness:

Aside from the obvious prejudice deriving from the prosecution's repeated references to the complainant as a "victim," a jury could reasonably *369 infer that Ms. Smietanka, who testified that she had personally examined the complainant for purposes of diagnosis and treatment, was, at the very least, predicting that the complainant in fact had been sexually abused. The proper purpose of Ms. Smietanka's testimony was merely to dispel the misconceptions that the specific postincident behavior traits at issue, such as delayed disclosure and the complainant's initial tendency to deny sexual intercourse, were inconsistent with those postincident behavior traits exhibited by children who in fact had been sexually abused. The prosecution could have dispelled these misconceptions just as effectively without the repeated references to the child or the facts of this case. [Id. at 750.]

Thus, the following may be discerned from the opinions in Beckley: Seven justices agreed that syndrome evidence is not admissible to demonstrate that abuse occurred and that an expert may not give an opinion whether the complainant is being truthful or whether the defendant is guilty. At least five justices agreed that where syndrome evidence is merely offered to explain certain behavior, the Davis/Frye test for recognizing admissible science is inapplicable. We continue to adhere to these holdings and reaffirm their application to child sexual abuse cases.

However, this Court was unable to agree on the foundations for and parameters of expert testimony. Justice BRICKLEY would limit its admission "`for the narrow purpose of rebutting an inference that a complainant's postincident behavior was inconsistent with that of an actual victim of sexual abuse, incest or rape.'" Id. at 710. Justice ARCHER concurred in part, but would hold that an expert may only testify in generalities and cannot discuss whether the victim's behavior is consistent with that of other abuse victims. Id. at 744. On the other hand, Justice BOYLE would allow an expert *370 to testify about these similarities and would allow expert testimony as long as it would assist the jury in deciding a fact at issue. Id. at 736.

III

Since our decision in Beckley, the majority of jurisdictions considering the proper use of expert syndrome evidence testimony in child sexual abuse cases have limited its admissibility. See, generally, State v Gokey, 154 Vt 129; 574 A2d 766 (1990), State v J Q, 130 NJ 554; 617 A2d 1196 (1993), State v Chamberlain, 137 NH 414; 628 A2d 704 (1993), State v Jones, 71 Wash App 798; 863 P2d 85 (1993), Frenzel v State, 849 P2d 741 (Wy, 1993), and People v Patino, 26 Cal App 4th 1737; 32 Cal Rptr 2d 345 (1994). See also Myers, Expert testimony in child sexual abuse litigation, 68 Neb L R 1, 68 (1989). Most courts will not allow expert testimony in this area if offered to prove that the abuse occurred. See State v Gokey, State v J Q, and Frenzel v State, supra. Several jurisdictions, consistent with our holding in Beckley, limit the admissibility of expert testimony solely to rehabilitate the victim's credibility. People v Nelson, 203 Ill App 3d 1038; 561 NE2d 439 (1990), Frenzel v State, and State v Jones, supra. A few jurisdictions never admit expert testimony concerning syndrome evidence because it has not attained scientific acceptance. See State v Foret, 628 So 2d 1116 (La, 1993).

Because we adopt the position that the admission of expert testimony regarding evidence of behaviors common in other abuse victims should be limited in these cases, we begin our discussion of our sister jurisdictions with a caveat from the United States Supreme Court. In Maryland v Craig, 497 US 836; 110 S Ct 3157; 111 L Ed 2d 666 *371 (1990), four members of the United States Supreme Court cautioned that courts should be particularly insistent in protecting innocent defendants in child sexual abuse cases because of the reliability problems created by children's suggestibility in child-sexual abuse prosecutions. Id. at 868 (Scalia, Brennan, Marshall, and Stevens, JJ., dissenting). Thus, these cases have special considerations that do not exist in other syndrome-type cases that include both the concerns of suggestibility and the prejudicial effect an expert's testimony may have on a jury. See Beckley, supra at 721-722.

In State v Chamberlain, supra, the New Hampshire Supreme Court considered whether the presentation of limited expert testimony in the prosecution's case in chief was error. While finding that the trial court had erred in admitting testimony that was offered to prove that the abuse occurred, it also noted that it is not error to allow expert testimony that is "offered to preempt or rebut an inference, based on the child victim's actions and behaviors following the alleged abuse, that the child has lied about the abuse." Id. at 417. Therefore, the pertinent inquiry is not the timing of the admission, but rather, the reason for the use of the evidence.

In Frenzel v State, supra at 749, the Wyoming Supreme Court adopted this Court's reasoning in Beckley, stating:

CSAAS[[11]] does not assist the trier of fact on the issue of whether abuse actually occurred. Additionally, *372 we believe that admission of CSAAS evidence, without limitation, would run too high a risk of misleading the jury and therefore be more prejudicial than probative under [Wyoming Rule of Evidence] 403. However, we also find that CSAAS evidence is relevant and admissible to dispel myths the public might hold concerning a child sexual abuse victim's post-abuse behavior if that behavior is an issue in the case. Qualified experts on child sexual abuse may, therefore, use evidence of CSAAS characteristics of sexually abused children for the sole purpose of explaining a victim's specific behavior which might be incorrectly construed as inconsistent with an abuse victim or to rebut an attack on the victim's credibility. For example, if the facts of a particular case show that the victim delayed reporting the abuse, recanted the allegations, kept the abuse secretive, or was accommodating to the abuse, then testimony about that particular characteristic of CSAAS would be admissible to dispel any myths the jury may hold concerning that behavior. [Emphasis added.]

In People v Patino, supra, the California Court of Appeals applied the holdings of several California cases using the following language that is pertinent to the issues presented here. The court stated:

Identifying a "myth" or "misconception" has not been interpreted as requiring the prosecution to expressly state on the record the evidence which is inconsistent with the finding of molestation. It is sufficient if the victim's credibility is placed in issue due to the paradoxical behavior, including a delay in reporting a molestation. Admission of evidence such as CSAAS is not error merely because it was introduced as part of the *373 prosecution's case-in-chief rather than in rebuttal. The testimony is pertinent and admissible if an issue has been raised as to the victim's credibility. [Id. at 1744-1745 (citations omitted, emphasis added).]

We adopt the position of these courts and therefore clarify our holding in Beckley regarding the admission of expert testimony in child sexual abuse cases. An expert may testify regarding typical symptoms of child sexual abuse for the sole purpose of explaining a victim's specific behavior that might be incorrectly construed by the jury as inconsistent with that of an abuse victim or to rebut an attack on the victim's credibility.

However, at issue in these consolidated cases and, apparently a question left unanswered by Beckley, is whether the prosecution may present an expert witness in its case in chief to describe certain behavioral characteristics recognizable in victims of child sexual abuse. We hold that the prosecution may present evidence, if relevant and helpful, to generally explain the common postincident behavior of children who are victims of sexual abuse.[12] The prosecution may, in commenting on the evidence adduced at trial, argue the reasonable inferences drawn from the expert's testimony and compare the expert testimony to the facts of the case. Unless a defendant raises the issue of the particular child victim's postincident *374 behavior or attacks the child's credibility,[13] an expert may not testify that the particular child victim's behavior is consistent with that of a sexually abused child. Such testimony would be improper because it comes too close to testifying that the particular child is a victim of sexual abuse. Again, we reiterate the concerns about such testimony given in the lead opinion in Beckley:

The use of expert testimony in the prosecution of criminal sexual conduct cases is not an ordinary situation. Given the nature of the offense and the terrible consequences of a miscalculation the consequences when an individual, on many occasions a family member, is falsely accused of one of society's most heinous offenses, or, conversely, when one who commits such a crime would go unpunished and a possible reoccurrence of the act would go unprevented appropriate safeguards are necessary. To a jury recognizing the awesome dilemma of whom to believe, an expert will often represent the only seemingly objective source, offering it a much sought-after hook on which to hang its hat. [Id. at 721-722 (emphasis added).]

Moreover, we do not think that just because evidence is helpful and relevant, it is necessarily admissible. Often, relevant evidence may be far *375 too prejudicial to present to a jury in its crudest form. In such cases, we apply the exclusionary rule found in MRE 403.

We believe that in the present case we strike the appropriate balance by allowing an expert to testify about behavioral traits that may, by their very nature, create confusion in the minds of the jury. Because the pertinent inquiry is not the timing of the admission, but rather the reason for the use of the evidence, the admission of expert testimony is not confined to the rebuttal stage of proofs and thus may be introduced, as limited by this opinion, in the prosecution's case in chief. When the credibility of the particular victim is attacked by a defendant, we think it is proper to allow an explanation by a qualified expert regarding the consistencies between the behavior of that victim and other victims of child sexual abuse.[14]

IV

A

Applying this to the cases at bar, we first hold that in Peterson, the trial judge erred in the following areas. First, the experts in that case improperly vouched for the veracity of the child *376 victim. For example, Gillan was allowed to testify that children lie about sexual abuse at a rate of about two percent. O'Melia was allowed to testify, over defense objection, that of the cases and studies he was familiar with, there is about an eighty-five percent rate of veracity among child abuse victims. Although we have no basis to dispute these numbers, their inherent inconsistency shows the difficulties that arise when attempting to vouch for the credibility of a witness. Certainly neither witness stated that the child victim was telling the truth. However, the risk here goes beyond such a direct reference. Indeed, as we have cautioned before, the jury in these credibility contests is looking "to hang its hat" on the testimony of witnesses it views as impartial. Such references to truthfulness as go beyond that which is allowed under MRE 702.

Second, the experts in Peterson were allowed to make numerous references to the consistencies between the victim's behavior and the behavior of typical victims of child sexual abuse. The experts, in various ways, testified regarding the dysfunctional family that each had observed and about some of the bizarre behavior exhibited by the victim. However, the defendant never challenged the prosecution's case regarding whether this behavior was inconsistent with that of an actual victim of child abuse. Nevertheless, over defense objection, Gillan was allowed to testify that the victim exhibited behavior manifestations that were symptomatic of sexual abuse. Green was permitted to testify that the victim had posttraumatic stress syndrome. O'Melia was allowed to testify that the victim's symptoms were consistent with those of a sexual abuse victim. Because the defendant never argued that the victim's behavior was inconsistent *377 with that of a typical victim of child sexual abuse, such testimony is error.

Because we recognize the errors below, we must next consider whether those errors were harmless. The testimony by the experts concerning behavior traits exhibited by the victim that were consistent behavior traits of other victims of child sexual abuse, although error, did not result in a miscarriage of justice. MCL 769.26; MSA 28.1096.

Parts of the testimony of the prosecution's expert witnesses actually served to question the victim's credibility.[15] For example, there was conflicting expert testimony concerning the physical evidence of sexual abuse. Dr. Hickok testified that after the alleged abuse, the victim had a large vaginal opening for a girl her age. He also remarked on the obliteration of the victim's hymen. However, Dr. Ramilo testified that she did not find any physical evidence of abuse. As to the victim's credibility, each expert testified about the victim's anti-social behavior. They described her behavior as erratic and aggressive. Green testified that the victim "has a great deal of difficulty with trust and with control." O'Melia testified that her behavior was "bizarre." None of these descriptions in any way bolsters the victim's credibility.

As noted above, Gillan testified that children lie *378 about sexual abuse at a rate of two percent. O'Melia testified that about eighty percent of child sexual abuse allegations are true. This conflicting testimony concerning the veracity rates among children of sexual assault, presented by two prosecution experts, actually served to question each other's accuracy.

Kathy Jo Gillan testified about her observations of what the prosecutor characterized as "incestral [sic] families." However, at the time she used this term, it was in reference to background information regarding the indicators of an abusive home and not a direct reference to defendant's home. She also stated that all these observations were made "without the defendant being there." She testified about the general conditions of the home and, without objection, was allowed to testify about her personal knowledge of the chaotic manifestations within such homes. Her lengthy testimony concerning her direct observations of the Peterson home were properly admitted.

The prosecution stated during closing arguments that this case was a "`classic one on one credibility test." Contrary to the contention of the dissent, this statement did not result in focusing the jury's attention on the expert testimony that was admitted erroneously. Instead, exactly the opposite is true. The statement highlights only the testimony of the victim and the defendant and asserts that the resolution of the case centers on which of the two they should believe. Lastly, the trial judge properly instructed the jury throughout the trial with respect to questionable testimony and properly charged the jury concerning its use of the expert testimony during deliberations.

In light of the foregoing, we do not think that leaving the jury verdict undisturbed results in a miscarriage of justice. The errors in the Peterson *379 record did not particularly favor either side, and, hence, did not affect the jury's decision to convict.[16] There is no purpose in sending this case back through the system for the victim and family to once again relive these tortuous events where the errors are harmless and justice is served.

B

In People v Smith, we hold that the trial court exercised the proper discretion in allowing a single expert to clarify why child victims of sexual assault frequently delay reporting the incident. Although the expert testimony was admitted into evidence during the prosecution's case in chief, the trial judge properly limited the testimony to an explanation of the commonality seen in victims of child sexual abuse.

Because Beckley has been interpreted to allow expert testimony only to rebut an inference created by the defendant, Smith used the strategy not to directly attack the credibility of a particular victim. Therefore, jurors are often left with misgivings regarding particular behavior of the victim, as in this case regarding the delayed reporting. We therefore find that where there are common misperceptions regarding the behavior of the victim on which a jury may draw an incorrect inference, such as the delayed reporting, the prosecutor may present limited expert testimony dealing solely with the misperception. However, as we noted in Beckley, the evidence has very limited use and should be admitted only to state that certain behavioral characteristics are common among victims of child sexual abuse.

It is a logical argument that the jurors would *380 have made an improper inference upon learning that the victim failed to report the sexual assault for five or six years. As it was stated by Justice BRICKLEY in Beckley, "[t]he findings of professional research suggest that there are many seemingly inconsistent responses to the trauma of the incident which require some form of explanation.... [One] possible misconception concerning the child victim is the belief that when a child suffers an injury it will be reported immediately." Id. at 716-717. Thus, according to this statement, when the victim on direct examination stated that she did not report the incident until five or six years after the fact, the jurors may have improperly inferred that the witness was incredible. Therefore, the expert testimony was proper because it rebutted an inference that the child's postincident behavior was inconsistent with that of an actual victim of sexual abuse.

Furthermore, the testimony presented did not unfairly prejudice the defendant and did not improperly bolster the credibility of the victim. Showing that the victim's delay in reporting was consistent with other sexual assault victims was mere background information that allowed the jury to better understand the victim's testimony. Lastly, this evidence did not improperly bolster the child witness' testimony. Because the child's testimony could easily be misconstrued, Beckley at 716-717, the expert testimony was proper to rebut possible inferences by the jurors. "This uniqueness put[] the evidence beyond the jury's ability to properly evaluate the facts in issue absent expert testimony." Id. at 715. See MRE 702.

V

Although we find error in the record in Peterson, *381 we find this error harmless in light of the overwhelming evidence against the defendant. People v Bahoda, 448 Mich 261; 531 NW2d 659 (1995).

We affirm the conviction in Smith, and note that the conduct of the trial presents an almost perfect model for the limitations that must be set in allowing expert testimony into evidence in child sexual abuse cases.

BOYLE, RILEY, and WEAVER, JJ., concurred with MALLETT, J.

BRICKLEY, C.J. (concurring in part and dissenting in part).

I concur in the rationale of the majority opinion, and in the holding in People v Smith. I dissent from the holding in People v Peterson, because I do not agree that the error was harmless.

CAVANAGH, J. (dissenting).

In these cases, we revisit the issue of the use of expert testimony with respect to behavioral evidence[1] regarding child sexual abuse victims. See People v Beckley, 434 Mich 691; 465 NW2d 391 (1990). I dissent because the majority will allow an expert to testify that the particular child complainant's behavior was "consistent with" that of other child sexual abuse victims. This diagnostic testimony fails to satisfy the scientific or medical foundation requirements under MRE 702 because the behavioral "symptoms" of child sexual abuse are too varied and too unreliable to be used as accurate detectors of sexual abuse. Additionally, I would continue to limit the use of behavioral reaction testimony to *382 rebuttal purposes. I would also preclude an expert who is testifying about the reactions exhibited by child sexual abuse victims in general, based either on personal observations of a general class or on consensus findings from the psychological or medical professions, from making any reference to the particular complainant or defendant.

I dissent in Peterson, because the experts were impermissibly allowed to testify that the complainant had been sexually abused. This preserved constitutional error requires reversal because it invaded the province of the jury as ultimate factfinder. I dissent in Smith, because the complainant's credibility had not been attacked before the "rehabilitative" expert testimony was introduced. Consequently, I would reverse and remand for a new trial in both cases.

I. "CONSISTENT WITH" TESTIMONY

I first want to distinguish those behaviors that are exclusively associated with sexual abuse from those behaviors that are also associated with other types of traumatic events.[2] There may come a day when a particular behavioral reaction that is exclusively associated with sexual abuse becomes sufficiently accurate, reliable, and standardized as *383 a detector of sexual abuse so that it obtains general scientific acceptance by the medical and psychological professions as a diagnostic tool under the Davis/Frye test.[3]If that day arrives, then diagnostic testimony may be permissible, provided that all other evidentiary requirements have been satisfied. My research has revealed that we have not yet reached that day. Therefore, I will confine my discussion to testimony concerning behavioral reactions that are not exclusive to sexual abuse.

I will also concede that courts that have addressed this issue have taken divergent approaches to the thresholds of admissibility, the permissible purposes of the testimony, and the scope of such testimony. See Steward v State, 652 NE2d 490, 495-498 (Ind, 1995) (grouping cases by approach taken).

In the instant cases, the expert witnesses described many behavioral reactions that may be caused by other nonsexual events.[4] When the behavioral reactions that are exhibited by the complainant may be caused by traumatic events unrelated to sexual abuse, they are not sufficiently reliable as diagnostic tools.[5] Consequently, any testimony that a child who exhibits x, y, and z *384 reactions is a sexually abused child fails the Davis/Frye test's scientific foundational requirements under MRE 702. In other words, a child exhibiting depression, withdrawal, and night-mares, cannot be diagnosed as a sexual abuse victim because such behaviors just as likely can be present with any number of stressful events.

Accordingly, child sexual abuse accommodation syndrome (CSAAS), which was created by Dr. Roland Summit to identify and group commonly observed behaviors in child sexual abuse victims, cannot be used as a diagnostic device. Myers, Expert testimony in child sexual abuse litigation, 68 Neb L R 1, 66-67 (1989).[6] Dr. Summit's intent was to provide a method of putting professionals on the same page. Id. at 67. The syndrome's sole function was to start with a known child victim of sexual abuse, and then to explain the child's behavioral reactions to the abuse. The syndrome cannot be used in reverse, which would be to start with the behavioral reactions identified by the syndrome, check off the ones that the child exhibits, and conclude that, on the basis of the checklist, the child was sexually abused.[7] Similarly, behavioral reactions, *385 which may be produced by nonsexual abuse causes, cannot be legally probative of sexual abuse because they are also probative of other causes. The court in State v J Q, 130 NJ 554, 573; 617 A2d 1196 (1993), acknowledged:

The most pointed criticism of the theory is that the same traits may equally appear as the result of other disorders.... Even extreme poverty or psychological abuse can produce the sense of entrapment or accommodation. In other words, the existence of the symptoms does not invariably prove abuse. [Citations omitted, emphasis added.][[8]]

This very point prompted one court to reject profile evidence altogether, reasoning that because there was no consensus in the medical field with respect to any particular psychological profile of a sexually abused child, such evidence is too unreliable to pass muster under the evidentiary requirements of scientific evidence. State v Ballard, 855 SW2d 557, 562 (Tenn, 1993).[9]

Here, the majority admits that when an expert testifies "that the particular child victim's behavior is consistent with that of a sexually abused child[, s]uch testimony ... comes too close to testifying that the particular child is a victim of *386 sexual abuse." MALLETT, J., ante at 374. Yet, it finds such testimony acceptable if the defendant has attacked the credibility of the complainant. Id. at 375. However, as State v Cressey, 137 NH 402, 407; 628 A2d 696 (1993), recognized, there is no material distinction between express testimony that the child has been sexually abused, and implicit testimony that outlines the unreliable behavioral reactions found with sexually abused victims, followed by a list of the complainant's own behavioral reactions, that points out that the two are consistent, and then invites the jury to add up the points to conclude that the child has been sexually abused. The majority is asking this type of evidence to perform a task that it cannot do.[10] More importantly, the foundation of such testimony does not become more reliable simply because it is offered after an attack on credibility by the defendant and not before.

The lack of reliability has caused some courts to limit this type of testimony to rebuttal and to "not inconsistent" testimony. State v Jones, 71 Wash App 798, 819; 863 P2d 85 (1993), reasoned:

Because the use of testimony on general behavioral characteristics of sexually abused children is still the subject of contention and dispute among experts in the field, we find that its use as a general profile to be used to prove the existence of abuse is inappropriate. However, we agree with the current trend of authority that such testimony may be used to rebut allegations by the defendant that the victim's behavior is inconsistent with abuse. [Citations omitted.]

Many courts that have accepted expert testimony for limited purposes have expressly rejected *387 this sort of "consistent with" sexual abuse testimony because its foundational basis is too unreliable to satisfy the expert testimony scientific basis requirements as diagnostic testimony.[11] In State v Chamberlain, 137 NH 414, 418; 628 A2d 704 (1993), the expert testified that the complainant's behaviors "were consistent with" other child sexual abuse victims. The court continued its earlier holding that such testimony was prejudicial error. It reasoned that syndrome evidence cannot be used as a diagnostic device. Accordingly, the only admissible purpose that the expert testimony could be offered for was "to preempt or rebut an inference from some specific behavior, action, or inconsistency displayed by the child victim that otherwise may be misinterpreted by a jury." Id. (citation omitted). Another court, which has traditionally allowed "consistent with" testimony, has recently questioned continued approval of such evidence, because such evidence "can have an impact on the minds of the jury far disproportionate to its foundation in science." Toro v State, 642 So 2d 78, 83 (Fla App, 1994) (citation omitted).

Indeed, Beckley held that behavioral reactions were too unreliable for an expert to testify that the complainant was in fact sexually abused. Id. at 733-734 (BRICKLEY, J.); id. at 740 (BOYLE, J.); id. at 744, n 2 (ARCHER, J.). Justice BRICKLEY, in justifying *388 the admissibility of some expert testimony, stated: "so long as the purpose of the evidence is merely to offer an explanation for certain behavior, the Davis/Frye test is inapplicable." Id. at 721.

I would argue that "consistent with" testimony is no longer merely offering an explanation. It is based on the expert's opinion, as an expert, that this child's behavior matches the behavior exhibited by other child sexual abuse victims. This is no different from a physician testifying that this child's broken leg, as revealed by the x-rays, is "consistent with," i.e., "was caused by," falling from a second-story roof. We would not allow a physician to make such testimony in the absence of sufficiently reliable medical evidence, first, that x-rays are accurate pictures of broken bones, second, that a fall from a second-story roof will cause fractures of this type, and, finally, that this type of fracture is unique to multistory falls.

Further, "consistent with" testimony is fundamentally different from testimony that a given set of behavioral reactions are not inconsistent with child sexual abuse. "Not inconsistent" is rebuttal-type testimony; it merely states that it could be true. "Consistent with" is positive testimony that it is true. "Not inconsistent" testimony can convey to the jury all the information that it needs to assess the issues in the case.

To illustrate, suppose the facts of the case indicate that the child has withdrawn, is exhibiting severe depression, initially reported the abuse to a school counselor, but has now recanted her story in fear that her stepfather-abuser will be taken away from the family, and her mother will be very sad. This information will be placed before the jury by the child's own testimony or through witnesses with personal knowledge of the child's behavior. If the defense then focuses on the recantation *389 and argues that her initial allegations were made in retaliation or rebellion to house rules, an expert could testify that child sexual abuse victims have been known to withdraw, become depressed, and recant their allegations when they realize that the family may be torn apart. If appropriate, the prosecutor could pose a hypothetical question to the expert paralleling the facts of the case, and ask whether such testimony is "inconsistent with" sexual abuse behavioral reactions. The expert could reply, "No, it is not inconsistent." At that point, the jury has all of the information that it needs to evaluate the child's credibility. The expert has offered no diagnostic evidence, and therefore, Davis/Frye remains inapplicable. The prejudicial effect has been minimized because the expert will be viewed as a disinterested person. A limiting instruction would further minimize prejudice.[12]

The majority has offered no reason why this approach is insufficient. Therefore, I would hold that "consistent with" diagnostic testimony should not be allowed, regardless of when it is presented. Such testimony violates the Davis/Frye test because its foundation is too unreliable as a detector of sexual abuse.

*390 II. REBUTTAL LIMITATION

I agree with the majority that expert testimony may be presented in the prosecution's case in chief. MALLETT, J., ante at 375. However, I would maintain the Beckley course that this testimony should be used solely for rehabilitative purposes. Consequently, I would hold that it must follow an attack on the complainant's credibility, either by the prosecution impeaching the complainant's testimony on the stand, or by the defense challenging the complainant's testimony or postincident behavior.[13] The alternate course that the majority navigates here, which is that the postincident behavior, in and of itself, may justify expert testimony, will create more questions than it answers. The majority has offered no direction to trial courts regarding how to decide which behaviors qualify.

Basically, the majority's wake will make expert testimony a matter of course in child sexual abuse cases and, if the defendant presents any kind of defense, then the expert can offer diagnosis testimony that is inherently unreliable. The majority is ignoring the original premises that justified the use of this type of testimony, and is opening a door here that should not be opened.

III. TESTIMONY WITH RESPECT TO THE PARTICULAR COMPLAINANT

I continue to believe that Justice ARCHER'S approach in Beckley was appropriate. Id. at 748. I would not allow an expert to testify about the *391 particular complainant, even if the testimony is only with respect to whether the child's behavioral reactions are not inconsistent with sexual abuse. Once the expert has given testimony disabusing the seemingly inconsistent behavioral reaction, the jury has all the information that it needs to assess the complainant's credibility. The marginal probative value of allowing the expert to further testify with respect to the particular complainant is substantially outweighed by the danger of unfair prejudice that the jury will misuse the testimony. It invades the province of the jury to assess credibility.[14] It invites the jury to give undue weight to testimony that is foundationally and fundamentally unreliable merely because it is cloaked with the expertise of an expert. It also invites the jury to believe that the expert knows more than he is telling, by letting the jurors infer that the expert, who works with sexually abused children every day, must believe this child's story or else the expert would not be testifying.[15]

IV. SUMMARY

In conclusion, I believe that the Indiana Supreme Court has recently offered a comprehensive and persuasive approach to behavioral reaction testimony involving child sexual abuse. Steward, supra. The court summarized:

It is possible that foundational support may be discovered in the future which will expand the *392 purposes for which such expert testimony may be deemed reliable. We echo the Mississippi Supreme Court in noting that "what should be emphasized about this view is that it renders dynamic the realm of expert opinion testimony. Should subsequent empirical research and scientific investigation yield a [diagnostic] child sexual abuse profile or syndrome." Goodson [v State], 566 So 2d [1142] 1146, n 6 [Miss, 1990], our trial courts may consider admitting that evidence for any use to which it may reliably be put. Furthermore, we decline to distinguish between expert testimony which offers an unreserved conclusion that the child in question has been abused and that which merely uses syndrome evidence to imply the occurrence of abuse. Where a jury is confronted with evidence of an alleged child victim's behaviors, paired with expert testimony concerning similar syndrome behaviors, the invited inference that the child was sexually abused because he or she fits the syndrome profile will be as potentially misleading and equally as unreliable as expert testimony applying the syndrome to the facts of the case and stating outright the conclusion that a given child was abused. The danger of the jury misapplying syndrome evidence thus remains the same whether an expert expresses an explicit opinion that abuse has occurred or merely allows the jury to draw the final conclusion of abuse. Exclusion of such evidence is authorized by Indiana Rule of Evidence 403. However, we recognize that, once a child's credibility is called into question, proper expert testimony may be appropriate. Daubert notes the importance of "a valid scientific connection to the pertinent inquiry as a precondition to admissibility." Daubert [v Merrell Dow Pharmaceuticals, Inc,] [509] US 579; 113 S Ct [2786] 2796; 125 L Ed 2d [469] 482 [1993] (emphasis added). Because research generally accepted as scientifically reliable recognizes that child victims of sexual abuse may exhibit unexpected behavior patterns seemingly inconsistent with the claim of abuse, such evidence *393 may be permissible under Indiana Evidence Rule 702(a)'s authorization of "specialized knowledge [which] will assist the trier of fact to understand the evidence." Therefore, if the defense discusses or presents evidence of such unexpected behavior by the child, or if during trial testimony the child recants a prior allegation of abuse, a trial court may consider permitting expert testimony, if based upon reliable scientific principles, regarding the prevalence of the specific unexpected behavior within the general class of reported child abuse victims. To be admissible, such scientific evidence must assist the finder of fact in understanding a child's responses to abuse and satisfy the requirements of both Rule 702(b) and the Rule 403 balancing test. We agree with the Arizona Supreme Court's assessment that when the relevant inquiry is the syndrome's reliability and probative value for rehabilitative and related purposes, "[s]uch evidence may harm defendant's interests, but we cannot say it is unfairly prejudicial; it merely informs jurors that commonly held assumptions are not necessarily accurate and allows them to fairly judge credibility." [State v] Moran, [151 Ariz 378, 382] 728 P2d [248 (1986)]. [Steward, supra at 499.]

I would adopt this approach.

V. APPLICATION

In Peterson, the majority properly finds that there were numerous errors even under the rule that it announces today. MALLETT, J., ante at 377. Nevertheless, it concludes that the errors were harmless because there was overwhelming evidence against the defendant. The majority's attempt to identify this supposedly overwhelming nontainted evidence is insufficient. There was certainly no convincing physical evidence. As the majority notes, Dr. David Hickok testified about *394 physical alterations of the complainant's vagina, hymen, and anal opening. Id. at 354. In contrast, Dr. Anselma Ramilo "did not find any physical evidence of sexual abuse...." Id. Indeed, the majority notes that the prosecutor characterized this case as a "`classic one on one credibility test.'" Id. at 356.

If that is the case, then one cannot reasonably conclude that the testimony of three experts, all testifying that the complainant was in fact sexually abused, did not prejudicially taint the jury's verdict beyond a reasonable doubt. The defendant did not deny that sexual abuse may have occurred. He argues here that the trial court denied him access to a meaningful expert to refute the prosecution's experts, and that the trial court also denied his request to present witnesses with respect to other potential perpetrators of sexual abuse to the complainant,[16] and that this denied him his constitutional right to a fair trial.

When the defense theory was to admit that there may have been sexual abuse, but to deny that he was the perpetrator of it, then the credibility of the complainant became the only issue in the case. As such, improper bolstering of her testimony cannot be considered harmless beyond a reasonable doubt.

In Beckley, Justice BOYLE agreed with reversal in Badour, where she stated that on the basis of the unconvincing testimony of the child complainant, "and the fact that the expert testimony was directed not only to the question whether abuse had occurred but to the defendant's participation in it, I cannot conclude that the trial did not constitute a miscarriage of justice. MCL 769.26; MSA 28.1096." Id. at 742 (emphasis added).

*395 Similarly, in this case, the expert testimony was directed to both the existence of sexual abuse and to the defendant's participation in the sexual abuse. Kathy Jo Gillan testified that she referred the complainant's case to the Department of Social Services by a "thirty-two hundred, which is a suspected sexual or physical or neglect [abuse form]." She testified at length about incestuous families and that the Peterson home symptomatology indicated incestuous relationships and that she saw very little improvement in those incestuous relationships over the course of her involvement. She also testified that most perpetrators stay in a state of denial, while most children rarely lie about sexual abuse. She was, in effect, making a direct comment on the complainant's credibility. Moreover, as argued by the Court of Appeals dissent in this case, the expert shifted the burden of proof to the defendant to prove that he was not denying the abuse. Unpublished opinion per curiam of the Court of Appeals, issued February 11, 1993 (Docket No. 123450).

Gillan also testified:

Q. As a result of your work was there any other symptomatology that you noted that I interrupted you on? A. The extreme ag[g]ressiveness at home, her teaching younger children how to be a[g]gressive in the home. When I first became involved with the family she had bruises on her buttocks where her father had hit her with a stick and I also observed some more bruising on her very deep bruising on her forearms. [Emphasis added.]

This testimony was of course irrelevant to the allegations of sexual abuse, moreover, it was direct testimony without foundation that the defendant had hit her. The direct inference being: because *396 the defendant hit the complainant, he also sexually abused her.

Margaret Green testified that the complainant had been identified as a sexual abuse victim. MALLETT, J., ante at 355. She almost immediately then testified:

[The complainant] came into treatment with a very overt set of symptoms and really what I would call in trouble emotionally. She was doing a lot of acting out behavior in school. She was doing a lot of re-enactment of the abuse. She was feeling very unsafe. At the time I got her, her father's whereabouts were not known. She was very fearful of him seeking retribution. [Emphasis added.]

Even though the defendant's objection to this testimony was sustained, the jury still heard the expert directly testify that the defendant committed the sexual abuse.

This testimony was immediately followed by testimony of the complainant's reenactment of the sexual abuse:

She would take Play Dough and make a buttocks and a rectum and a penis and re-enact anal intercourse. She would form a penis and mutilate it, you know, cut it up and throw it around and this type of thing. She would often draw pictures on the I have a white board where you can erase. She would draw pictures of herself in a bed with her father over her and write, "My father touching me."

Again, even though the defendant's objection was sustained and a limiting instruction was given to the jury, the jury still heard the expert testify that this defendant committed the sexual abuse.

Green also referred to the defendant when she testified:

*397 The first concern would be maintaining her emotional stability in terms of her thought processes. That would be one. The second would be assisting [the complainant] with resolution of abandonment issues with the parental termination of rights, assisting with adoption placement.

Once again, even though the defendant's objection was sustained and a limiting instruction was given, the court could not erase from the jury's mind that the expert testified that this defendant had had his parental rights terminated.

The third expert, Thomas O'Melia testified that the complainant would score very high on a checklist of symptoms of sexual abuse. He too testified that children do not lie and that most perpetrators deny the abuse.

Even under Justice BOYLE'S standard in Beckley, the improper expert testimony cannot be harmless. The untainted evidence was not overwhelming. Unlike People v Christel, 449 Mich 578; 537 NW2d 194 (1995), in which there was convincing physical evidence, here, there was simply the child complainant's own testimony. Her credibility therefore was critical. When the jury was allowed to hear, even if some objections were sustained, that the defendant had hit the complainant, that his parental rights had been terminated, that most perpetrators deny sexual abuse and that this defendant had denied abuse, and testimony from three experts that the complainant had in fact been abused, one cannot reasonably come to the conclusion that the error was harmless. The majority's conclusion is simply incredulous.

In Smith, I would reverse and remand for a new trial because the defendant never attacked the complainant's credibility by challenging her delay in reporting. Because her credibility did not yet *398 need rehabilitating, expert testimony was inappropriate.

VI. CONCLUSION

For the reasons stated above, I dissent in both cases and would reverse and remand for new trials.

LEVIN, J., concurred with CAVANAGH, J.

NOTES

[1] MCL 750.520b; MSA 28.788(2).

[2] MCL 750.520b(1)(a); MSA 28.788(2)(1)(a).

[3] MCL 750.145c(2); MSA 28.342a(2).

[4] MCL 769.12; MSA 28.1084.

[5] The relevant testimony of the expert follows:

Q. And why would a child delay reporting of sexual abuse?

A. Well, there's a multitude of reasons and it depends on several variables, depending on the relationship between the child and the perpetrator, the relationship between them, the living arrangements between them. There's a lot of contingencies.

* * *

Q. I believe the question was what type of relationships would or would not cause delayed reporting.

A. It is from my own experience, as well as the literature, that clearly indicates the closer the relationship between the child and the perpetrator, the less likely that abuse is going to be disclosed at the time of offense. If an offender is living in the same dwelling where the victim resides, the chance of that sexual abuse being disclosed at the time that it is occurring are unlikely.

Q. Now, what, if you have any experience or any opinion, as to what effect the removal of the person causing the abuse or doing the abuse from the home would trigger or cause a person to report, a child to report?

A. It increases the likelihood of that child revealing the sexual abuse once the offender has left the home.

Q. Now, does the age of a child victim have anything to do with the timing of reporting?

A. Yes; it does.

Q. And what effects do age have?

A. In my experience, the younger the victim, the less likely they are going to tell at the time the abuse is occurring. The older a victim becomes, the more empowered and liberated from home that victim becomes, the more likely they are to tell.

Q. What effect, if any, do threats made to a child affect their ability to report or not report?

A. Threats are very effective in keeping a child quiet.

* * *

Q. Other than threats and the offender being in the home, are there any other reasons which a child would delay in reporting?

A. Many.

Q. What type of reasons?

A. Well, aside from fear of the person abusing them, very often children feel responsible for holding the family together, and most children, however small they are, are smart enough to know what's going on, if it is disclosed could very well split the family apart, and so keeping the family together very often is one reason why they do not disclose. In addition to feelings that they are responsible for the other family members' emotional welfare, very often children feel that is their responsibility. Very often they feel they won't be believed. Very often they are frightened that if the abuse has occurred more than once, that someone is going to blame them for letting it happen more than once. They feel guilty. Somehow they are made to feel responsible, quite often, for that abuse occurring, that there's something they did to cause it.

Q. How do children exhibit I'll withdraw that question. In your experience, is there any magical number on how long a child would delay in reporting?

A. Not necessarily. One particularly common thread that I see is very often well, two things come to mind, one, that once a child feels empowered or safe enough to tell, they may, and the second thing is very often I find children tell when they are provoked, when either something new is introduced to them sexually or they have been provoked by something that they will tell. There's a third thing that I would like to mention, too.

Q. Sure.

A. Very often children will disclose if they are directly asked.

[6] the situation is a proper one for the use of expert testimony is to be determined on the basis of assisting the trier. "There is no more certain test for determining when experts may be used than the common sense inquiry whether the untrained layman would be qualified to determine intelligently and to the best possible degree the particular issue without enlightenment from those having a specialized understanding of the subject involved in the dispute." [People v Smith, 425 Mich 98, 106; 387 NW2d 814 (1986) (citations omitted). See also Beckley, supra at 715.]

[7] See Beckley, supra at 715 (opinion of BRICKLEY, J.), at 735 (opinion of BOYLE, J.), and at 744-745 (opinion of ARCHER, J.). Some of the behavior identified as signs of child sexual abuse:

overt or subtle and indirect disclosures to a relative, friend, or teacher,

highly sexualized play,

withdrawal and excessive daydreaming,

low self esteem, feelings of shame or guilt,

falling grades,

pseudomature personality development,

sexual promiscuity,

poor peer relationships,

suicide attempt,

positive relations exhibited toward the offender, and

frightened or phobic reactions, especially toward adults.

See State v J Q, 130 NJ 554, 564-565; 617 A2d 1196 (1993), citing AMA Diagnostic and Treatment Guidelines Concerning Child Abuse and Neglect, 254 JAMA 796, 798 (1985).

[8] Some jurisdictions have approached the problems presented in these cases as involving inquiries into whether the otherwise relevant evidence should be excluded because of the danger of unfair prejudice. See MRE 403. Such inquiries, however, should be left to the sole discretion of the trial judge, and we do not believe that it is appropriate to set mandates regarding what is or is not unfairly prejudicial.

[9] People v Davis, 343 Mich 348; 72 NW2d 269 (1955), and Frye v United States, 54 App DC 46; 293 F 1013 (1923).

[10] The choice of the language "to the extent" may have added to some of the difficulties experienced by the lower courts in interpreting the decision of this Court because it is somewhat unclear to what extent the lead opinion would hold that an expert may testify that the behavior of the complainant in a particular case is consistent with that of children who report sexual abuse.

[11] Child sexual abuse accommodation syndrome (CSAAS) is a term that has become common place in other jurisdictions. It is used to describe the different reactions observed in child victims of sexual abuse. The term originated in an article published by Dr. Roland Summit entitled The child sexual abuse accommodation syndrome, 7 Child Abuse & Neglect 177 (1983). The article described five common characteristics observed in child victims of sexual abuse: (1) secrecy, (2) helplessness, (3) entrapment and accommodation, (4) delayed, conflicted, and unconvincing disclosure, and (5) retraction. See Myers, Expert testimony, supra at 66.

[12] This expert testimony, however, may be introduced only if the facts as they develop would raise a question in the minds of the jury regarding the specific behavior. The behavior must be of such a nature that it may potentially be perceived as that which would be inconsistent with a victim of child sexual abuse, i.e., delay in reporting, recantation, accommodating the abuser or secrecy. The court must determine whether the particular characteristic is one that in fact calls for an expert explanation. MRE 702. The expert is then only allowed to testify regarding the behavior at issue and may not testify regarding CSAAS characteristics that are not at issue.

[13] The credibility of the victim is attacked when the defendant highlights behaviors exhibited by the victim that are also behaviors within CSAAS and alludes that the victim is incredible because of these behaviors. The scope of the testimony, however, is again limited to the behavior at issue, and the expert may not testify about behavior manifestations that are not at issue. In other words, it does not matter how the behavior trait came into evidence. The expert may not proffer testimony of other behaviors unless the facts as they develop make the specific behavior relevant or if the defendant attacks the victim's credibility based on the behavior.

While the specific behaviors of delay in reporting, recantation, accommodation, and secrecy are widely recognized as those that may create doubt regarding the credibility of the victim, we believe that it is the province of the experts in this field to delineate what behaviors are considered part of CSAAS, and we leave the door open to further development in this area.

[14] See also Commonwealth v Hudson, 417 Mass 536, 543; 631 NE2d 50 (1994), in which the Massachusetts Supreme Court held that where expert testimony is used solely for background explanation of the victim's testimony, without an opinion about truthfulness of testimony, it does not improperly bolster the victim's credibility. See also Frenzel v State, supra at 746. In People v Bowker, 203 Cal App 3d 385; 249 Cal Rptr 886 (1988), the court interpreted the "rebuttal limitation" and determined that expert testimony explaining the victim's postabuse behavior is admissible in the prosecution's case in chief regardless of how or by whom the behavior in question was introduced to the jury. In addition, in State v Cardany, 35 Conn App 728, 731-732; 646 A2d 291 (1994), the court held that testimony that explains to the jury why a minor victim of sexual abuse might delay in reporting the incidents of abuse should be allowed as part of the state's case in chief.

[15] The victim's testimony was strong and consistent throughout the trial. As noted by the dissent, Justice BOYLE agreed with the reversal in Badour because the expert referred to the defendant's participation in the abuse and because of the unconvincing testimony of the child complainant. At the time of trial, the victim was eleven years old. She was subject to a lengthy direct examination and was required to repeat some of the most disturbing aspects of her story on cross-examination. During cross-examination, the defendant's attorney attempted to elicit responses that may have been viewed as inconsistent with some of her earlier statements. She provided explanations for each instance. She testified in graphic detail without any inherent inconsistencies on direct or cross-examination, and the record is devoid of any indication that the victim was prone in any way to suggestibility.

[16] We note that this trial took place before the release of this Court's decision in Beckley, and therefore the trial judge was without its guidance.

[1] The evidentiary requirements of behavioral reactions should not be confused with the evidentiary requirements of other types of evidence, such as battered child syndrome, which is based on physical evidence, or hearsay statements by the child complainant under MRE 803A. See State v J Q, 130 NJ 554, 581-582; 617 A2d 1196 (1993).

[2] See Myers, Expert testimony in child sexual abuse litigation, 68 Neb L R 1, 75 (1989). The authors found that there exists a general consensus that certain factors can indicate sexual abuse. Such factors include

age-inappropriate sexual knowledge; sexualized play; precocious behavior; excessive masturbation; preoccupation with genitals; indications of pressure or coercion exerted on the child; the child's story remains consistent over time; the child's report indicates an escalating progression of sexual abuse over time; the child describes idiosyncratic details of the abuse; and physical evidence of abuse.

See also State v Jones, 71 Wash App 798, 819; 863 P2d 85 (1993) (distinguishing sexual acting-out behavior).

[3] See People v Davis, 343 Mich 348; 72 NW2d 269 (1955), and Frye v United States, 54 App DC 46; 293 F 1013 (1923). I do not predict what effect Daubert v Merrell Dow Pharmaceuticals, Inc, 509 US 579; 113 S Ct 2786; 125 L Ed 2d 469 (1993), may have on the Davis/Frye test.

[4] In Peterson, Kathy Jo Gillan testified that indicators of sexual abuse include depression, withdrawal, anger, aggression, nightmares, age-inappropriate behavior, delayed disclosure, running away, hitting, kicking, spitting, and dysfunctional family role reversals and chaos. She also testified about age-inappropriate sexual verbiage. Margaret Green testified about similar reactions, including guilt, shame, distorted self-esteem, and gender identity changes. She too testified about age-inappropriate sexual behavior. In Smith, Barbara Cross testified regarding delayed reporting.

[5] Compare Jones, n 2 supra at 819, which distinguished the types of behavioral reactions:

We further note that sexual acting out behavior has been viewed as more logically and clinically indicative of sexual abuse than other generalized reactions to emotional traumas such as nightmares and phobic behaviors. [Citing Myers, n 2 supra at 59-60.]

[6] He identified five behavioral reactions commonly observed: "(1) secrecy, (2) helplessness, (3) entrapment and accommodation, (4) delayed, conflicted, and unconvincing disclosure, and (5) retraction." Id.

[7] J Q, n 1 supra at 579 (citing Myers, supra at 66-67) (syndrome evidence may explain abuse, not detect it). In State v Foret, 628 So 2d 1116, 1127 (La, 1993), the court noted:

Criticisms include the varying reactions children have to abuse and the fact that behavior often attributed to abuse is sometimes the result of other emotional problems that do not stem from abuse.

[8] State v Sims, 158 Vt 173, 179; 608 A2d 1149 (1991), noted:

The fact that a complainant displays symptoms similar to those displayed by children who have been sexually abused is not usually sufficient to allow an expert to testify that that particular complainant has been sexually abused. [Citations omitted.]

[9] Accord Jones, n 2 supra at 818:

The prevalent objection to such testimony is that a general profile of typical behaviors is not specific to sexual abuse, but may be produced by other traumatic events in the life of the child. [Citations omitted.]

[10] See J Q, supra at 582 (testimony that victims' symptoms "were consistent with sexual abuse" was prejudicial error).

[11] See Foret, n 7 supra at 1123, in which the court adopted the less stringent Daubert scientific evidence standard, and concluded that syndrome behaviors still fail the test as diagnostic devices, id. at 1127. However, with respect to rebuttal purposes, the court stated:

The proper presentation of this sort of expert testimony must focus on explaining to a jury why "superficially bizarre" reactions such as delayed reporting, etc. take place in such cases. [Id. at 1130 (citation omitted).]

The court limited the testimony to a discussion of the general characteristics of the behavioral reaction at issue, and precluded testimony concerning the particular complainant.

[12] People v Patino, 26 Cal App 4th 1737, 1745; 32 Cal Rptr 2d 345 (1994), noted:

The trial court in the instant action handled the matter carefully and correctly. The jury was immediately admonished after [the expert's] testimony it was to consider CSAAS testimony only for the limited purpose of showing, if it did, that the alleged victim's reactions as demonstrated by the evidence were not inconsistent with her having been molested.

The instructions also required the jury to give the defendant the presumption of innocence, and further admonished that the prosecution still had the burden of proof of guilt, that the syndrome evidence is based on the assumption that abuse has occurred, and that the testimony could not be used as proof that the complainant's allegation of sexual abuse was true. Id. at 1745-1746.

[13] See Foret, n 7 supra at 1129, which adopted the rule of admitting testimony for the very limited purpose "of rebutting attacks on the victim's credibility based on inconsistent statements, limited disclosures, or recantations of the testimony." (Citations omitted.) See also People v Nelson, 203 Ill App 3d 1038; 561 NE2d 439 (1990) (limiting admissibility to rebuttal after the complainant's credibility has been attacked).

[14] testimony concerning the particular complainant must be approached with caution, as it too often slips into impermissible comment on the complainant's credibility. [Sims, n 8 supra at 181.]

[15] See id. at 179 (improper questions required answers that implicitly conveyed the expert's belief that the complainant was truthful).

[16] The defendant suggests that a member of the foster family may have been one perpetrator of the abuse.