Skip to main content


Reset A A Font size: Print

Missouri Court of Appeals, Eastern District.

STATE OF MISSOURI, Plaintiff/Respondent, v. GEORGE A. ROGERS, Defendant/Appellant.

No. ED 104928

Decided: September 26, 2017


George Rogers appeals the trial court's judgment after a jury convicted him of statutory sodomy and child molestation. We reverse and remand for a new trial.


In January 2015, twelve year-old X.S. (Child) told her school guidance counselor (Counselor) that her grandfather (Rogers) had touched her inappropriately. Counselor contacted Children's Division, and Child was taken to the Child Advocacy Center for a forensic interview, conducted by Anthony Harper. Based on Child's revelations during the interview, the State charged Rogers with six counts of child sexual abuse. One was dismissed before trial, and Rogers was acquitted on three others. This appeal concerns the remaining two counts, alleging that Rogers undressed Child and touched her breasts and genitals during a family barbeque in 2013.

Before trial, pursuant to § 491.075, the State filed notice of its intention to offer the hearsay testimony of five witnesses to whom Child conveyed allegations of abuse: Counselor, Detective William Belcher, Child's mother, Child's friend, and Harper.

Counselor testified that Child, then in seventh grade, became upset upon entering Counselor's office and told her that Rogers had touched her inappropriately, over the clothes on an area covered by a swimsuit. Detective Belcher testified that he interviewed Child ten months later, when Child revealed new allegations that Rogers showed her pornography on his computer and molested her in a storage locker.1 Mother testified that she thought she overheard Child telling a cousin that she “touched” or “sucked” “pawpaw.” Friend testified that Child said that Rogers “used to touch her when she was younger.” Rogers objected to the witnesses' testimony on the basis that Child's statements to them lacked sufficient indicia of reliability to be admitted under the hearsay exception of § 491.075. The court overruled the objections and allowed the witnesses to testify at trial.

Harper described a progressive process of disclosure commonly observed in child victims of sexual abuse: denial, tentative disclosure, active disclosure, recantation, and reaffirmation. He explained that, during the tentative stage, children might minimize or distance themselves from the events in question. He also testified that Child provided sensory details and specific locations where events occurred. Rogers filed a motion in limine to exclude Harper's testimony about the stages of disclosure insofar as it invaded the province of the jury to determine when and whether Child was telling the truth or lying. The court denied the motion. Rogers renewed his objection at trial. The State argued that Harper possessed expertise about the process of disclosure, how children communicate differently than adults, and specifically how child sex abuse victims have certain characteristics that he could describe in his capacity as an expert. The State further assured the court that the testimony would be limited to generalized testimony. Rogers recorded a continuing objection, and proceedings continued. Harper described at length typical challenges in interviewing children, the stages of disclosure, the technique of source monitoring (distinguishing reality from fiction), the difference between scripted and episodic memory (recurring patterns versus discrete events), and the significance of sensory detail unrelated to the allegations. The jury then viewed a video of Harper's interview with Child.2 In that interview, as relevant to this appeal, Child told Harper that Rogers had touched her “chest” and “private area.” Harper asked Child to clarify those terms on an anatomical drawing, and Child circled the genital area and breasts. Child demonstrated the motion of Rogers's fingers on her upper thigh. And she said that Rogers “kept touching me everywhere” but she didn't remember anything more because she was “really tired” at the time.

After the video, Harper explained in itemized detail, excerpted infra, how Child's specific actions and statements in the video aligned with the aforementioned interview themes and were therefore reliable and consistent with sexual abuse. Rogers didn't object during questioning but reasserted his standing objection at the conclusion of testimony on the basis that Harper's opinion as to Child's reliability invaded the province of the jury. The State argued that Harper could opine whether Child's statements were consistent with sexual abuse as long as he didn't opine whether the child was actually abused. The trial court overruled the objection. In closing argument, the State reminded the jury that Harper was an expert on the process of disclosure and, in his expertise, Child showed indicators of reliability.

The jury found Rogers guilty of one count of statutory sodomy and one count of child molestation. The trial court sentenced him to consecutive prison terms of fifteen years and five years, respectively. Rogers appeals and asserts that the trial court erred by (1) allowing Harper to opine about Child's credibility, (2) allowing Harper to reconcile Child's statements with the stages of disclosure, (3) submitting the case to the jury without sufficient evidence, (4) allowing the other witnesses' hearsay testimony about Child's statements, and (5) denying a mistrial after the jury heard evidence of uncharged conduct.


Harper's Testimony (Points I and II)

For his first point, Rogers asserts that the trial court abused its discretion by allowing Harper to testify that Child's statements were reliable and consistent with sexual abuse. As a preliminary matter, the State asserts that the point wasn't properly preserved because Rogers's motion in limine, continuing objection, and post-trial motion challenged only Harper's general testimony about the stages of disclosure and not other communication characteristics he mentioned. Rogers's second point more directly targets Harper's testimony on the stages of disclosure. The State argues that this point is also unpreserved in that Rogers expanded the issue in his motion for a new trial.

“Allegations of error must be sufficiently definite to direct the trial court's attention to the particular acts or rulings asserted to be erroneous so that the trial court has an opportunity to correct them.” State v. Howery, 427 S.W.3d 236, 248 (Mo. App. E.D. 2014). We find the issues sufficiently preserved. Rogers referred to the “stages of disclosure” because the State presented Harper's 491 testimony in that framework. Though the State added other labels at trial, the record establishes that the disclosure stages and other characteristics are not separate inquiries; they are related. For example, Harper specifically confirmed that minimizing, distancing, and progressive revelations are characteristics of tentative disclosure. Regardless of nomenclature, Rogers's motions and objections consistently articulated a singular hazard: that Harper's testimony would invade the province of the jury as to Child's credibility. Rogers sufficiently directed the trial court's attention to this concern. Accordingly, we review points I and II together under the abuse of discretion standard. 3

Trial courts have broad discretion in determining the admissibility of evidence, and we review their rulings for an abuse of that discretion. An abuse of discretion occurs when a ruling is clearly against the logic of the circumstances and is so arbitrary and unreasonable as to shock one's sense of justice and indicate a lack of careful consideration. State v. Johnson, 207 S.W.3d 24, 40 (Mo. 2006). Where reasonable persons can differ about the propriety of the action taken by the trial court, no abuse of discretion will be found. Id. Further, we review for prejudice, not mere error, and will reverse only if the error was so prejudicial that it deprived the defendant of a fair trial. State v. Churchill, 98 S.W.3d 536, 539 (Mo. 2003).

The general purpose of expert testimony is to assist the jury in areas that are outside of everyday experience or lay experience. State v. Pickens, 332 S.W.3d 303, 321 (Mo. App. E.D. 2011). However, “the expert may not express an opinion as to the guilt or innocence of the defendant. To do so would usurp the decision-making function of the jury.” Id. at 322. Thus, when determining the admissibility of opinion testimony, expert witnesses should not be allowed to give their opinion as to the veracity of another witness's statement, because, in so doing, they invade the province of the jury. Id.

In cases involving the sexual abuse of a child, there are typically two types of expert testimony that give rise to a challenge: general and particularized. Churchill, 98 S.W.3d at 539. General testimony describes behaviors and characteristics commonly found in victims. Id. Particularized testimony concerns a specific victim's credibility as to the abuse. Id. The trial court has broad discretion in admitting general testimony, but particularized testimony must be rejected because it usurps the fact-finding role of the jury and thus is inadmissible. Id. “Expert testimony that comments directly on a particular witness's credibility” or “that expresses an opinion with respect to the credibility or truthfulness of witnesses of the same type under consideration invests ‘scientific cachet’ on the central issue of credibility and should not be admitted.” State v. Williams, 858 S.W.2d 796, 800 (Mo. App. E.D. 1993). Relevant case law illuminates the demarcation line for permissible inquiry.

On the safe side of the continuum, in State v. Baker, 422 S.W.3d 508 (Mo. App. E.D. 2014), a therapist named, without describing, the phases of disclosure and confirmed that they result in inconsistent statements, explaining, “you can see a child go through all of the phases․ based upon any varying consequence․” Then in closing argument, the prosecutor elaborated on each phase and used them to justify discrepancies in the victim's testimony. Baker argued that, by doing so, the State transformed the expert's general testimony into particularized testimony. This court rejected that theory, reasoning that the expert never expressed an opinion on the victim's credibility, and the State was entitled to argue reasonable inferences from the evidence. Id. at 514-515.

Similarly, in State v. Thomas, 290 S.W.3d 129 (Mo. App. S.D. 2009), a therapist described in more detail the process of disclosure, common problems or roadblocks, source monitoring, script memory, and tactile detail. Then in closing, like Baker, the prosecutor connected the victim's behavior and statements to the common characteristics described by the expert. Thomas argued that the expert's testimony was “tailored to the facts” of the case and thus particular. The court again rejected that theory, reasoning that the expert's testimony “did not speak to the Victims' credibility at all.” Id. at 135.

By contrast, in Williams (cited supra), a doctor testified that: children rarely lie about sexual abuse and “essentially don't lie” about who abused them; that if the child was not asked leading questions then her spontaneous response “declares who it was;” and that behavioral indicators can only support what the child says.” The court deemed this evidence inadmissible, reasoning, “Vouching too much for the victim's credibility, these statements supplied improper verisimilitude on the issue of whether the appellant was guilty.” Williams, 858 S.W.2d at 801. Even reviewing only for plain error, the court judged the testimony “demonstrably prejudicial” and “fundamental error,” because it “amounted to an impressively qualified stamp of truthfulness on the victim's story.” Id.

Likewise in State v. Foster, 244 S.W.3d 800 (Mo. App. S.D. 2008), a doctor testified that he “has a sense whether a child is telling me a lie or not” and 95% of his interviewees were abused though no physical evidence was present; he then specifically opined that the subject child had been abused. The court deemed this testimony inadmissible because it lent scientific cachet to the child's credibility. The court further noted that, while error may be harmless when the evidence of guilt is overwhelming, “this was a ‘he-said, she-said’ credibility case” with no other evidence. Id. at 803.

In Churchill (cited supra), a doctor testified that: the child exhibited several behavioral changes indicative of abuse, the changes meant that a significant event had occurred, and the child's change in demeanor when discussing the incident revealed that it “was real.” The State conceded that this testimony was improper, and the Supreme Court of Missouri deemed it prejudicial insofar as the only other evidence consisted of the child's behavior and conflicting statements, which the doctor bolstered in the State's favor. The Court qualified, however, that, although testimony such as this is always inappropriate, it may not necessitate a new trial in all situations.

Just such a situation was presented in State v. Collins, 163 S.W.3d 614 (Mo. App. S.D. 2005). In that case, the physical evidence of severe and prolonged abuse was undeniable; Collins signed a written confession, and other witnesses corroborated the victim's allegations. Though two experts' testimony invited numerous objections, none was asserted, so the appellate court examined only whether the trial court plainly erred in not excluding the testimony sua sponte. Specifically, one expert testified that the victim provided “credible idiosyncratic detail” about yellow rope and also exhibited a “hallmark of credibility” by correcting a misstatement by the expert. Despite these direct affirmations of credibility, the appellate court declined to fault the trial court for allowing this evidence, reasoning that Collins' failure to object was strategic, consistent with a defense theory implicating other assailants. The court further noted that any error was harmless in that the evidence of guilt was overwhelming. The experts also testified that the victim's behavior was “very common” and “extremely typical” among abuse victims. The court deemed this evidence “general profile” testimony in that it didn't “directly or implicitly” vouch for the victim's credibility, relying on State v. Silvey, upholding testimony that a child “displayed behavioral indicators consistent with child sexual abuse.” 894 S.W.2d 662, 671 (Mo. 1995) (abrogated on other grounds by State v. Porter, 439 S.W.3d 208 (Mo. 2014)). But while Silvey and its progeny permit an opinion that a child's behavior is “consistent with” sexual abuse, we remain wary that an expert's extensive personalization of general characteristics, factually dissected and aggregated on an otherwise equivocal record, inevitably bolsters a victim's credibility. That is precisely what occurred here.

In his generalized testimony, Harper explained typical challenges in interviewing children, the stages of disclosure, source monitoring, scripted versus episodic memory, and the presence of sensory detail. The State elicited the following general testimony to foreshadow and frame Child's interview.

Q: So you're trying to figure out if the kid really had this happen or if they were just told about it or heard about it, is that fair to say?

A: Yes.

Q: Are there different indicators of reliability to show that this really happened to a child?

A: Yes, there are.

Q: Tell me about some of those indicators of reliability.

A: If a child makes a spontaneous gesture during the interview ․ A child making a statement against their self-interest ․

Q: You also mentioned - we're talking about indicators of reliability – and then what about emotion?

A: Yeah, if a child displays emotion during the interview, that could also be an indicator.

Q: What if the child corrects you or asks for clarification?

A: Yes. ․ If the child corrects the interviewer, that shows that they're less likely to be suggestible or leading․

Q: What about details? Can details be an indicator of reliability?

A: Yes. ․

Q: Tell me more about that.

A: If a child is providing some type of a detail that's really not related to the – if they're talking about an abusive experience but they're also relating other things that were going on at the time, again those would be indicators that it's most likely something that they experienced.

Q: What about tactile details, things that they saw, felt, heard, that type of thing?

A: Yes, if a child is talking about how something on their body felt, those types of things would also be indicators.

Q: Indicators of?

A: Of reliability.

Q: Okay. Thank you.

Following this generalized testimony, the jury viewed the video of Harper's interview with Child. After the video, the State resumed Harper's examination, the prosecutor eliciting and often supplying specific examples of how Child's behavior fit the general description. Harper responded in detail, with the prosecutor furnishing additional insight. The following excerpts illustrate the reach of the State's inquiry.

Q: Did you see any examples of distancing or minimizing in this interview:

A: Yes. ․ She talked about being sleepy. ․ and I think she said she didn't know because her eyes were closed. ․

Q: So, when she talked about things like ․ I can't remember because I've let it go, put it in a vault in memory so she can never see it again or remember, those would be examples of that distancing or minimizing?

A: Yes.

Q: Was she also able to demonstrate how her grandfather rubbed her private area?

A: Yes, at one point when she was talking about how his fingers were, she had demonstrated on her leg, several motions on her leg.

Q: So you talked about some of the details that she remembers and some of those sensory details. Are those consistent with indicators of reliability?

A: Yes.

Q: One of the things that you said was an indicator of reliability is when the child will correct you or ask for clarification or ask questions. Did you see any of that in the interview with [Child]?

A: Yes, I did.

Q: Can you give me some examples?

A: There was one point where she had talked about ․ her mother's friend had moved to ․ Indiana ․ and she corrected me. There was another point where ․ she wasn't sure what my question was.

Q: Would another example be when she said that she was eleven when this barbeque thing happened, or ten, and then she corrected herself and said, no, I was eleven?

A: Yes. And there was another time․

Q: And those would all be indicators of reliability, correct?

A. Yes.

Q: What about idiosyncratic details ․ Did you hear any of those in this interview?

A: Yes.

Q: Can you tell me about those?

A: ․ she talked about not feeling well from eating cucumbers. ․

Q: So when she's talking about things like the cucumber salad ․ that's one of those indicators of reliability, correct?

A: Yes.

Q: Same thing when she's talking about her grandfather's jeans being big at the bottom. That would be another one of those kinds of details?

A: Yes.

Q: And when she describes the foam thing, the egg crate․ that would be another detail?

A: Yes.

Q: When she described how her grandfather took off her shirt and then her pants and then was touched her and rubbing her, that would be more of those details?

A: Yes.

Q: When she said that she was on her grandmother's side of the bed․ that's another detail?

A: Yes.

Q: When she talked about him sitting on the chair․ that would be another detail?

A: Yes.

Q: All of those things would be indicators of reliability, right?

Q: As you said before, demonstrated how her grandfather's legs were when he was sitting in the chair and told her to come touch his private, right?

A: Yes.

Q: And that kind of context, and detail, that's another indicator of reliability, isn't it?

A: Yes.

Q: You also talked about statements against the child's interests. Did you hear any of that in this interview?

A; Yes.

Q: Tell me about some of those.

A: She expressed concern about having to switch schools, not being around her friends any longer, not being able to participate in choir․

Q: So those would all again be consistent with indicators of reliability?

A: Yes.

Q: And then when she said things like she won't be with her friends ․ and she's worried about her birthday celebration․ That would be another one of those statements against her interests, right?

A: Yes.

Q: We also discussed emotion and how that can be an indicator of reliability.

A: Yes.

Q: Did you see any emotion during this video?

A: Yes. There was at least one part ․ where she began to tear up and I slid the Kleenex over to her.

Q: Based on your training and experience, if [Child] talked to other people in the future and told then more of what happened to her, would that be consistent with this process of disclosure?

A: Definitely.

Q: So if at later times [Child] revealed more incidents of abuse or more contacts that happened, would that be consistent with this process of disclosure?

A: Yes.

Q: Would that be consistent with what she told you about it, happening more than once?

A: Yes, it would.

Q: And would that be consistent with when she told you that, well, when she said that it had happened other times?

A: Yes.

Q: Based on your training and experience in the ChildFirst Protocol, were [Child's] disclosures to you consistent with those of a child who has been sexually abused?

A: Yes.

At this point, Rogers's counsel re-asserted his objection that Harper's testimony “invades the province of the jury. I mean, that's for the jury to decide whether it's consistent with sexual abuse and whether her statements are reliable and credible.” The trial court overruled the objection, and the examination continued.

Q: What kind of things are you paying attention to to decide if the child has actually experienced this event?

A: ․ Any details as far as sensory details; how things felt, things they may have heard or smelled, those types of things.

Q: So those things that we called indicators of reliability, right?

A: Yes.

Q: And you did that source monitoring with [Child], didn't you?

A: Yes.

Q: And you saw those indicators of reliability, didn't you?

A: Yes.

Q: When she was talking about [the barbeque], she was able to say that it was in the summer, right?

A: Yes.

Q: Then on this dining room incident ․ she talked about his private part, right?

A: Yes.

Q: And she could even describe how it felt. ․ Is that right?

A: Yes.

Q: And she did say that these things happened more than once, is that right?

A: Yes.

Q: When you did this source monitoring ․ and ․ asking her these questions and ․ listening to these details and ․ watching the gestures that she made, were her statements and her gestures and her emotions consistent with someone who has actually been molested as a child?

A: Yes.

In total, the State solicited at least twenty specific examples of how Child's statements or behavior evinced a common profile.4 This meticulous process of personalization cannot be reconciled with the concept of generalized testimony; it is pointedly particular. More troubling still, the phrase “indicator of reliability” appears sixteen times in the State's examination of Harper. While jurists in chambers may debate “indicia of reliability” as a term of art, jurors in a court room plainly understand reliability as a synonym for credibility.5 The State's excessive inquiry as to the reliability - i.e., credibility – of Child's statements was wholly improper. When the State “insists on walking the precipice of reversible error, it must be prepared to suffer the consequences of stepping over the edge - reversal and remand for a new trial.” State v. Moore, 352 S.W.3d 392, 404 (Mo. App. E.D. 2011). As in Moore, here, too, the prosecutor didn't just “drift a bit too close to the cliff's edge;” she “flung [herself] headfirst into the abyss.” Id.

Accordingly, the trial court abused its discretion by permitting particularized testimony to bolster Child's credibility. Moreover, prejudice resulted. “Error may be harmless if other evidence of guilt is overwhelming, or if the improper evidence was cumulative and not highlighted.” Foster, 244 S.W.3d at 803. Unlike Silvey where the evidence of guilt was overwhelming, this case more closely resembles the “he said, she said” facts of Foster and Williams in that the State relied heavily on testimonial evidence of Child's allegations to prove its case. The improper evidence was not cumulative, and it was certainly highlighted. The jury's verdict hinged on its impression of Child's credibility, and Harper's particularized testimony provided a “stamp of truthfulness” that invaded the province of the jury. Points I and II are granted and mandate reversal.

Sufficiency of the Evidence (Point III)

For his third point, Rogers contends that the trial court erred in overruling his motion for acquittal because the evidence was insufficient to comprise a submissible case. Specifically, Rogers argues that the record lacked evidence permitting a jury to find that he touched Child's genitals and breasts for sexual gratification.

On a claim of insufficient evidence, our review is limited to whether the State has introduced sufficient evidence for a reasonable juror to have been convinced of the defendant's guilt beyond a reasonable doubt. State v. Nash, 339 S.W.3d 500, 09 (Mo. 2011). This is not an assessment of whether this court believes that the evidence at trial established guilt beyond a reasonable doubt but instead whether, in light of the evidence most favorable to the State, any rational fact-finder could have found the essential elements of the crime beyond a reasonable doubt, accepting all evidence favorable to the State as true, including all favorable inferences drawn from the evidence. Id. This court does not act as a “super juror” with veto powers but gives great deference to the trier of fact, who may believe all, some, or none of the testimony. Id. Importantly here, appellate review of the sufficiency of the evidence is based on the entire record, including evidence adduced in error. State v. Wideman, 940 S.W.2d 18, 21 (Mo. App. W.D. 1997); State v. Wood, 596 S.W.2d 394 (Mo. banc 1980).

Mindful of these standards, we note the following evidence. During the CAC interview, Child told Harper that, when she was lying in her grandparents' bed resting, Rogers came to lie next to her, took off her shirt and touched her “chest,” and pulled down her pants and touched her “private area.” Clarifying her terminology on an anatomical drawing, Child circled the genital area and breasts. And as discussed above, Harper testified extensively about his interpretation of the interview, confirming that Child exhibited “indicators of reliability” and that her behavior and statements were consistent with sexual abuse. Although the record lacked specific testimony suggesting sexual gratification, given the circumstances of the encounter as described, viewing the entire record in the light most favorable to the State, and accepting all reasonable inferences, a juror believing all of the foregoing could conclude that Rogers touched Child's genitals and breasts for sexual gratification. Particularly given the import of Harper's testimony, we cannot say that the evidence on the whole record was insufficient to make a submissible case. Therefore, point III must be denied.

Remaining Points (IV & V)

For his fourth point, Rogers asserts that the trial court abused its discretion by allowing other witnesses – Counselor, Detective Belcher, Mother, and Friend – to testify about Child's statements to them under the hearsay exception of § 491.075. That statute allows such testimony when the trial court determines, after a closed hearing, that the time, content, and circumstances of the statement provide sufficient indicia of reliability. We review only for an abuse of discretion. State v. Ragland, 494 S.W.3d 613, 622 (Mo. App. E.D. 2014). “If reasonable persons can differ about the propriety of the action taken by the trial court, then it cannot be said that the trial court abused its discretion.” Id. Mindful of our standard of review, we cannot say that the trial court abused its discretion in allowing these witnesses to testify. Point IV is denied.

As a result of reversal on points I and II, we need not reach Rogers's point V regarding evidence of uncharged conduct inadvertently admitted through the CAC video. The trial court would have an opportunity to supervise the editing of the video in compliance with its evidentiary rulings on remand.


This court does not discount the gravity of harm suffered by child victims of sexual abuse. As a matter of law, however, the State may not prosecute such charges in a manner that usurps the role of the jury and thereby violates a defendant's right to a fair trial. The trial court's judgment is reversed and remanded for a new trial.6


1.   No pornography was found on Rogers's computer, and he was acquitted of the charge involving the storage locker.

2.   During the interview, Child also mentioned allegations involving Child's sister. Those portions of the video were ordered excluded from the record as evidence of uncharged conduct. However, at trial, the State failed to operate the audiovisual equipment in compliance with the court's order, causing the jury to hear that evidence, giving rise to Rogers's point V.

3.   Even were we to consider these points unpreserved, as discussed below, the trial court's permissiveness here would not withstand plain error review. See State v. Williams, 858 S.W.2d 796 (Mo. App. E.D. 2003).

4.   Defense counsel did not object on the basis of leading or testifying for the witness.

5.   “Reliable” is defined as “trustworthy, safe; honest, responsible; credible.” THE OXFORD AMERICAN DESK DICTIONARY AND THESAURUS 705 (2d ed. 2001). See also THESAURUS.COM, (last visited August 31, 2017).

6.   Where a conviction is reversed for trial error concerning the improper admission of evidence rather than for insufficiency of the evidence, the proper remedy is to reverse and remand for a new trial. State v. Taber, 73 S.W.3d 699, 707 (Mo. App. W.D. 2002), citing State v. Wood, 596 S.W.2d 394 (Mo. banc 1980). On remand, the State can either choose not to re-try the case or proceed to trial without the inadmissible evidence. Id.

Lisa Van Amburg, Judge

Colleen Dolan, P.J. Mary K. Hoff, J., concur.

Copied to clipboard