State v. Adam P.

351 Conn. 213

• Court: Supreme Court of Connecticut
• Decided: February 11, 2025
• Docket: SC20849
• Status: Published

Opinion

February 11, 2025 CONNECTICUT LAW JOURNAL Page 3

351 Conn. 213 FEBRUARY, 2025 213 State v. Adam P.

STATE OF CONNECTICUT v. ADAM P.* (SC 20849) McDonald, D’Auria, Mullins, Ecker and Dannehy, Js.**

Syllabus

Convicted of multiple counts of sexual assault in the first degree and risk of injury to a child in connection with the sexual abuse of the minor victims, D and T, the defendant appealed to this court. He claimed, inter alia, that the trial court had violated his due process right to a fair trial by instructing the jury, in accordance with this court’s directive in State v. Daniel W. E. (322 Conn. 593), that it was not to consider the victims’ approximately nine year delay in officially reporting the abuse at issue in assessing their credibility. Held:

This court overruled that portion of Daniel W. E. that modified the constancy of accusation doctrine, as set forth in State v. Troupe (237 Conn. 284), and returned to the standard previously articulated in Troupe, which provides that a person to whom a sexual assault victim has reported the assault may testify only with respect to the fact and timing of the victim’s complaint, that constancy evidence is admissible only to corroborate the victim’s testi- mony and not for substantive purposes, and that a defendant is entitled to an instruction that a victim’s delay in reporting is a factor that the jury can consider in evaluating the victim’s credibility.

This court concluded that Troupe, along with other avenues available to negate juror biases, such as expert testimony and the application of the rules of evidence, sufficiently balanced the defendant’s interest in being free from the undue prejudice that may result from the presentation of multiple constancy witnesses with the state’s interest in overcoming poten- tial jury bias against sexual assault victims who delay in reporting.

The defendant was not entitled to a new trial because any error with respect to the trial court’s instructing the jury in accordance with Daniel W. E. was harmless, as any such error was not constitutional in nature and it was not reasonably probable that the instruction misled the jury in arriving at its verdict.

* In accordance with our policy of protecting the privacy interests of the victims of sexual abuse and the crime of risk of injury to a child, we decline to use the defendant’s full name or to identify the victims or others through whom the victims’ identities may be ascertained. See General Statutes § 54-86e. ** The listing of justices reflects their seniority status on this court as of the date of oral argument. Page 4 CONNECTICUT LAW JOURNAL February 11, 2025

214 FEBRUARY, 2025 351 Conn. 213 State v. Adam P. The jury charge, as a whole, directed the jury to consider the victims’ general credibility on the basis of factors other than their delay in reporting and emphasized that it was the state’s burden to prove the requisite elements of the charged offenses, it was not reasonably probable that the jury believed that the challenged instruction precluded it from considering the victims’ credibility more generally, neither the jury’s verdict nor the primary theory of defense turned on the victims’ delayed reporting, and the state’s case was strong.

The trial court did not abuse its discretion in permitting D to testify that the defendant told her that he had played the same sex ‘‘games’’ with his daughter that he played with D and T, as D’s testimony was probative of the defendant’s attempts to groom or pressure her to engage in sexual intercourse with him, and the trial court took measures to mitigate any prejudicial effect of D’s testimony. (One justice concurring in part and dissenting in part) Argued September 26, 2024—officially released February 11, 2025

Procedural History

Substitute information charging the defendant with seven counts of the crime of sexual assault in the first degree and eight counts of the crime of risk of injury to a child, brought to the Superior Court in the judicial district of Fairfield and tried to the jury before Dayton, J.; verdict and judgment of guilty of five counts of sexual assault in the first degree and eight counts of risk of injury to a child, from which the defendant appealed to this court. Affirmed. Lisa J. Steele, assigned counsel, for the appellant (defendant). Timothy J. Sugrue, assistant state’s attorney, with whom, on the brief, were Joseph T. Corradino, state’s attorney, and Edward L. Miller, senior assistant state’s attorney, for the appellee (state). Opinion

D’AURIA, J. The defendant, Adam P., appeals directly to this court from his conviction of five counts of sexual assault in the first degree in violation of General Stat- utes § 53a-70 (a) (2), four counts of risk of injury to a February 11, 2025 CONNECTICUT LAW JOURNAL Page 5

351 Conn. 213 FEBRUARY, 2025 215 State v. Adam P.

child in violation of General Statutes § 53-21 (a) (1), and four counts of risk of injury to a child in violation of General Statutes § 53-21 (a) (2). He first claims that the trial court violated his due process rights when, based on our decision in State v. Daniel W. E., 322 Conn. 593, 629, 142 A.3d 265 (2016), it instructed the jury not to consider the delay of the twin minor victims, D and T, in reporting the defendant’s sexual abuse when assessing their credibility because that credibility was central to the outcome of the case, therefore misleading the jury. He then claims that the trial court abused its discretion by permitting D to testify that the defendant told her that he previously had played the same sexual ‘‘games’’ with his daughter, A, that he was playing at the time with the victims. We conclude that we must overrule Daniel W. E., to the extent that it modified our constancy of accusation doctrine set forth in State v. Troupe, 237 Conn. 284, 304–305, 677 A.2d 917 (1996), so that jurors understand more precisely the parameters for when and how they may consider a victim’s delayed reporting when assessing the victim’s credibility. Further, we hold that the alleged instructional error was nonconstitutional in nature and that, based on the charges against the defendant and the record in its entirety, it was not reasonably probable that the trial court’s Daniel W. E. instruction misled the jury in arriving at its verdict. Finally, we reject the defendant’s second claim and conclude that the trial court did not abuse its discretion by permitting D to testify that the defendant had told her that he played the same sexual ‘‘games’’ with A that he had with the victims. We therefore affirm the trial court’s judgment. We summarize the facts that the jury could have reasonably found, and the relevant procedural history, as follows. From 2005 to 2013, the defendant was in a romantic relationship with the victims’ mother, Q. Dur- ing most of that time, the defendant lived with Q, D Page 6 CONNECTICUT LAW JOURNAL February 11, 2025

216 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

and T, first in Bridgeport, and then in Stratford. He did not live with them during summers, when the victims stayed with extended family in South Carolina, or for ten months in 2010, when Q was evicted from her Strat- ford apartment. When the defendant lived with the victims, Q was rarely home because she often worked twelve hour shifts, seven days a week, as a nursing assistant. While Q was at work, the defendant, who was unemployed, looked after the victims. There were other adults in the home intermittently during this period, with one adult living with Q and the victims for three years. The defendant sexually abused the victims, beginning when they were five years old. At trial, the victims described graphic incidents in which the defendant engaged in cunnilingus, fellatio, vaginal-penile inter- course, and other sexual contact with the victims, indi- vidually or simultaneously, as well as showed them pornography. The defendant described the sexual abuse as ‘‘a game’’ to the victims. The defendant continued to sexually abuse the vic- tims until he moved to North Carolina after he and Q had ended their relationship. The victims were eleven years old when the abuse ended. They never told anyone about it while it was happening. After the breakup, Q moved to a different location in North Carolina than the defendant and sent the victims to live with extended family in South Carolina. Approximately two years later, the victims joined Q in North Carolina. Q and the defen- dant kept in touch during this period, and, in 2017, Q asked the victims, then almost fourteen years old, how they would feel about the defendant’s coming to live with them again. It was then that the victims disclosed to Q that the defendant had sexually abused them. Q did not report the victims’ disclosure to the police but cut off contact with the defendant. February 11, 2025 CONNECTICUT LAW JOURNAL Page 7

351 Conn. 213 FEBRUARY, 2025 217 State v. Adam P.

The victims’ behavior changed in the months follow- ing their disclosure to Q. D struggled at school, engaged in self-harm, ran away from home, and researched sex trafficking. T became withdrawn. D was hospitalized for a mental health crisis, at which point she told a hospital worker that the defendant had sexually abused her. The hospital worker reported this disclosure to the North Carolina authorities, who investigated the claims and, because the incidents had taken place in Connecti- cut, shared the relevant forensic interviews and medical evaluations with Connecticut law enforcement. The defendant was then arrested in North Carolina and brought back to Connecticut to stand trial. The defendant was charged with seven counts of sexual assault in the first degree and eight counts of risk of injury to a child. More specifically, the defendant was charged with two counts of sexual assault in the first degree based on an incident involving sexual inter- course with the victims in the apartment where they lived in Bridgeport. The additional five counts of sexual assault in the first degree involved incidents in the apart- ment where they lived in Stratford. Four of the eight risk of injury to a child charges were based on the events in Bridgeport; the other four were based on the Stratford incidents. The defendant’s defense at trial was that he had never abused the victims and that they fabricated their accusa- tions to maintain Q’s attention once it became apparent to the victims that the defendant might return to live with them. He advanced this defense on cross-examina- tion by demonstrating his lack of opportunity to assault the victims, the victims’ motive to fabricate to gain Q’s attention, and the lack of physical evidence suggesting sexual abuse. Following the close of evidence and the parties’ closing arguments, and consistent with our decision in State v. Daniel W. E., supra, 322 Conn. 629, the trial court’s charge to the jurors included an instruction that D and T’s eight year delay in reporting Page 8 CONNECTICUT LAW JOURNAL February 11, 2025

218 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

the sexual abuse ‘‘should not be considered by you in evaluating their credibility.’’ The defendant previously had objected to this instruction outside the jury’s pres- ence, claiming that it unfairly undermined his primary defense at trial that the victims had fabricated their accusations. The jury found the defendant guilty on five counts of sexual assault in the first degree and eight counts of risk of injury to a child. It found the defendant not guilty of two counts of sexual assault in the first degree, both involving incidents of sexual intercourse with T alone. The trial court rendered judgment of conviction based on the jury’s verdict and sentenced the defendant to a total effective term of forty-eight years of incarcera- tion, twenty-five years of which was a mandatory mini- mum, followed by thirty-five years of special parole and lifetime sex offender registration. This appeal followed. I The defendant first claims that the trial court violated his due process rights when, consistent with Daniel W. E., it instructed the jury not to consider the victims’ delay in reporting the abuse when evaluating their credi- bility. The contested instruction arose on the third day of trial, when the court, while discussing its draft final jury instructions with the parties outside the jury’s pres- ence, noted that it intended to deliver a delayed reporting instruction based on Connecticut Criminal Jury Instructions 7.2-1.1 Defense counsel first objected 1 Instruction 7.2-1 of the Connecticut model criminal jury instructions provides in relevant part: ‘‘If no constancy of accusation witness testified, but there was a delay in officially reporting the offense,’’ the trial court should instruct the jury that ‘‘[t]here was evidence in this case that the complainant delayed in making an official report of the alleged sexual assault. There are many reasons why sexual assault victims may delay in officially reporting the offense, and to the extent the complainant delayed in reporting the alleged offense here, the delay should not be considered by you in evaluating (his or her) credibility.’’ Connecticut Criminal Jury Instructions 7.2-1, available at https://jud.ct.gov/JI/Criminal/Criminal.pdf (last visited February 4, 2025). February 11, 2025 CONNECTICUT LAW JOURNAL Page 9

351 Conn. 213 FEBRUARY, 2025 219 State v. Adam P.

orally, arguing that the instruction would invade the jury’s domain and bolster the testimony of the state’s expert witness on the behavior of child victims of sexual assault, Danielle Williams, who testified later that day that sexual assault victims often delay reporting abuse. The court overruled the objection, stating that delay ‘‘is not one of the things that should be considered in evaluating credibility’’ and that, with respect to the impact of the instruction on the expert testimony, ‘‘[the jury is] told that . . . [expert] testimony is not binding [on] them and they can disregard it completely.’’ Defense counsel then submitted a written objection to the instruction, which the court overruled the follow- ing morning, again relying on Daniel W. E. After the court ruled that the instruction was appropriate, the jury heard counsels’ closing arguments and the court’s final instructions, including the instruction now at issue, which read in full: ‘‘There was evidence in this case that [D and T] delayed in making an official report of the alleged sexual assaults. There are many reasons why sexual assault victims may delay in officially reporting the offense, and, to the extent [D and T] delayed in report- ing the alleged offense here, the delay should not be considered by you in evaluating their credibility.’’ The defendant asks that we overrule the portion of our decision in Daniel W. E. that directed trial courts to give this instruction because requiring that jurors disregard any delay in reporting abuse when assessing the victims’ credibility constitutes an invasion of the jury’s fact-finding role. He contends further that the instruction improperly shifted the state’s burden of proof to him by signaling to the jury that the victims’ delay in reporting was a symptom of the defendant’s sexual abuse, which, in turn, undermined his fabrication defense. He claims that the instruction necessarily harmed him because the victims’ credibility was dispositive of the case’s outcome, and the jury might have found him Page 10 CONNECTICUT LAW JOURNAL February 11, 2025

220 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

not guilty on all counts if it could have considered the delayed reporting when assessing his fabrication defense. In response, the state argues that the trial court did not provide the instruction in error because, read in the context of the entire jury charge, the instruction precluded the jury from considering only the fact of the victims’ delay in reporting the sexual abuse, but not the reasons for that delay, when assessing their credibility. The state also asserts that, even if the instruc- tion was erroneous, it was harmless because the jurors understood, based on the evidence and arguments pre- sented at trial, that they could consider the victims’ reasons for delay when assessing their credibility and that the instruction did not inescapably signal that D and T were, in fact, victims because they delayed reporting. Regardless of whether the instruction was deficient in the present case, however, the state joins the defendant in requesting that we reconsider our holding in Daniel W. E., specifically asking that we ‘‘overrule Daniel W. E. in its entirety, scrap the apparently unique procedure adopted therein, and return to the constancy doctrine as it existed under Troupe.’’ We first address and reconsider the jury instruction in Daniel W. E. that we directed trial courts to provide when facing circumstances such as those the court confronted in the present case, and we further overrule the changes to the constancy of accusation doctrine contained in that opinion. Next, we consider the impact the instruction had on the defendant in the present case and hold that any impact was harmless. A As noted, the instruction at issue derives from our holding in Daniel W. E., which built on our constancy of accusation doctrine. Any discussion of that doctrine requires some context. February 11, 2025 CONNECTICUT LAW JOURNAL Page 11

351 Conn. 213 FEBRUARY, 2025 221 State v. Adam P.

Beginning in the early nineteenth century, Connecti- cut common law allowed the state in a sexual assault case to offer constancy evidence, which generally included proof that the victim had disclosed the sexual assault to third parties prior to making a formal com- plaint to authorities. See State v. Troupe, supra, 237 Conn. 294–99 (reciting history of constancy of accusa- tion doctrine). The law permitted the court to admit this evidence to negate the inference a fact finder might draw, namely, that a complainant was fabricating her accusation if she failed to officially report the alleged sexual abuse promptly after the incident. Id., 296. In Troupe, we rejected the defendant’s request that we abandon our constancy of accusation doctrine alto- gether and held that the biases supporting that doctrine still existed in our society. Id., 303. But we narrowed the admissibility of constancy evidence to better serve ‘‘the interests of the defendant, the state and the victim,’’ who, in that case, was thirty years old at the time of the crime. Id., 287, 303. Specifically, we held that a constancy witness ‘‘may testify only with respect to the fact and timing of the victim’s complaint; any testimony by the witness regarding the details surrounding the assault must be strictly limited to those necessary to associate the victim’s complaint with the pending charge, including, for example, the time and place of the attack or the identity of the alleged perpetrator. . . . Thus, such evidence is admissible only to corroborate the victim’s testimony and not for substantive purposes.’’ (Footnote omitted.) Id., 303; see also Conn. Code Evid. § 6-11 (c) (codifying Troupe rule). To further accommo- date the interests of the defendant, we directed that ‘‘the defendant is entitled to an instruction that any delay by the victim in reporting the incident is a matter for the jury to consider in evaluating the weight of the victim’s testimony.’’ State v. Troupe, supra, 237 Conn. 305. Page 12 CONNECTICUT LAW JOURNAL February 11, 2025

222 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

In Daniel W. E., two decades after Troupe, we again were asked to either modify or abandon the constancy of accusation doctrine, particularly in child sexual abuse cases. The defendant in Daniel W. E. first asserted that the trial court’s jury instruction on the correct use of constancy evidence misled the jury because it failed to adequately distinguish constancy evidence from sub- stantive proof. State v. Daniel W. E., supra, 322 Conn. 608. He next claimed that, based on increased public understanding of child sexual abuse along with other evidentiary rules permitting the introduction of proba- tive evidence, he was unfairly prejudiced by the hypo- thetically unlimited number of constancy witnesses that the state could call under Troupe, even though only two constancy witnesses testified at his trial. Id., 617–18. We concluded that the instruction concerning the use of constancy evidence did not mislead the jury and that implicit juror biases against victims, including children, who delay in officially reporting sexual abuse remained, and, therefore, we declined to abolish the constancy of accusation doctrine entirely. Id., 616, 618. We neverthe- less modified the doctrine to alleviate ‘‘the potential prejudice to defendants caused by the testimony of multiple constancy witnesses.’’ Id., 618. To achieve ‘‘a proper balance between the interests of the state and the defendant’’; id., 630; and after considering ‘‘the myr- iad ways in which other courts have attempted to bal- ance these competing interests,’’ we directed that ‘‘the victim in a sexual assault case should continue to be allowed to testify on direct examination regarding the facts of the sexual assault and the identity of the person or persons to whom the incident was reported. . . . Thereafter, if defense counsel challenges the victim’s credibility by inquiring, for example, on cross-examina- tion as to any out-of-court complaints or delayed reporting, the state will be permitted to call constancy of accusation witnesses subject to the limitations estab- February 11, 2025 CONNECTICUT LAW JOURNAL Page 13

351 Conn. 213 FEBRUARY, 2025 223 State v. Adam P.

lished in Troupe, as modified in this opinion. If defense counsel does not challenge the victim’s credibility in any fashion on these points, the trial court shall not permit the state to introduce constancy testimony but, rather, shall instruct the jury that there are many rea- sons why sexual assault victims may delay in officially reporting the offense, and, to the extent the victim delayed in reporting the offense, the delay should not be considered by the jury in evaluating the victim’s credibility.’’ (Citation omitted; emphasis added; inter- nal quotation marks omitted.) Id., 629. The parties accu- rately observe that this procedure, now at issue in this appeal, is unique to this state. We agree with both parties that we should reconsider the balance we struck in State v. Daniel W. E., supra, 322 Conn. 617–30, in our attempt to mitigate the potential prejudice to the defendant of multiple constancy of accusation witnesses. We recognize that doing so neces- sarily implicates stare decisis, which ‘‘counsels that a court should not overrule its earlier decisions unless the most cogent reasons and inescapable logic require it.’’ (Internal quotation marks omitted.) State v. Sala- mon, 287 Conn. 509, 519, 949 A.2d 1092 (2008). Appro- priate reasons and logic include ‘‘[w]hen a previous decision clearly creates injustice . . . .’’ (Internal quo- tation marks omitted.) Id., 520. Then, ‘‘[t]he court must weigh [the] benefits of [stare decisis] against its burdens in deciding whether to overturn a precedent it thinks is unjust’’; (internal quotation marks omitted) id.; under- standing that stare decisis is ‘‘not a mechanical formula of adherence to the latest decision, however recent and questionable, when such adherence involves collision with a prior doctrine more embracing in its scope, intrin- sically sounder, and verified by experience.’’ (Internal quotation marks omitted.) Boys Markets, Inc. v. Retail Clerks Union, Local 770, 398 U.S. 235, 241, 90 S. Ct. 1583, 26 L. Ed. 2d 199 (1970); see also State v. Moulton, Page 14 CONNECTICUT LAW JOURNAL February 11, 2025

224 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

310 Conn. 337, 362 n.23, 78 A.3d 55 (2013). We ‘‘should be especially wary of overturning a decision that involves the construction of a statute’’; (internal quotation marks omitted) State v. Bischoff, 337 Conn. 739, 762, 258 A.3d 14 (2021); or claims ‘‘expressly . . . raised and rejected by this court’’; State v. Ray, 290 Conn. 602, 614, 966 A.2d 148 (2009); however, our decision in Daniel W. E., involved an issue of judicial policy, specifically, how to instruct the jury on a sexual assault victim’s delayed reporting, and when constancy of accusation evidence may be introduced at trial. Although we do not suggest that we should ever treat the stare decisis doctrine cavalierly, we are less reluctant under these circum- stances to correct what we conclude to be an error of judicial policy that might mislead the jury at trial. See Payne v. Tennessee, 501 U.S. 808, 828, 111 S. Ct. 2597, 115 L. Ed. 2d 720 (1991) (‘‘[c]onsiderations in favor of stare decisis are at their acme in cases involving prop- erty and contract rights . . . [and] the opposite is true in cases . . . involving procedural and evidentiary rules’’ (citations omitted)). Although we reject the defendant’s argument that the instruction we directed trial courts to deliver in Daniel W. E. violates due process, as we will discuss in greater detail in part I B of this opinion, we conclude that our modification of the constancy of accusation doctrine in Daniel W. E., although well-intentioned, unduly con- fuses what a jury may and may not consider when assessing a victim’s credibility in a sexual assault case, and conflicts with aspects of Troupe that we left untouched. In light of these conclusions, we have deter- mined that we will not allow principles of stare decisis to prevent our reconsideration and reformulation of an area of the law so important to the fairness of a sexual assault trial. The instruction at issue refers to evidence ‘‘that [D and T] delayed in making an official report’’ of the February 11, 2025 CONNECTICUT LAW JOURNAL Page 15

351 Conn. 213 FEBRUARY, 2025 225 State v. Adam P.

assaults and explains in the same sentence that there are ‘‘many reasons why sexual assault victims may delay,’’ all the while admonishing that any delay ‘‘should not be considered by [the jury] in evaluating their credi- bility.’’ Considering the instruction closely, as the defen- dant requests, we understand how the sentence may create confusion. The state argues that, because the court’s instructions as a whole emphasized the jury’s role in assessing general credibility and the state’s bur- den of proof, the jury understood that it could consider the victims’ reasons for delay, but not the fact of the delay itself, when assessing their credibility. Although our cases consistently direct that we must read a trial court’s charge as a whole, and not myopically, we do not believe that the jury would understand that the reasons why a victim might delay reporting sexual assault are necessarily separate from the delay itself based on the instructions provided, given that we have trouble discerning or articulating this distinction con- vincingly ourselves. Rather than only enjoining trial courts from providing this portion of the instruction, which we had directed courts to provide in Daniel W. E., we choose to overrule the entirety of our modification of the constancy doc- trine in that case because we are convinced that incom- patibilities remain between Daniel W. E. and Troupe. For example, Daniel W. E. instructs that, if the state is permitted to call constancy of accusation witnesses, the testimony of those witnesses should be governed by the limitations established in Troupe. See State v. Daniel W. E., supra, 322 Conn. 629. But Troupe provides that the defendant is entitled to an instruction that ‘‘any delay by the victim in reporting the incident is a matter for the jury to consider’’ when evaluating witness credi- bility. State v. Troupe, supra, 237 Conn. 305. Reading Troupe and Daniel W. E. together thus means that, if a defendant challenges the victim’s credibility based on Page 16 CONNECTICUT LAW JOURNAL February 11, 2025

226 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

a delay in reporting and the state then calls constancy witnesses, the jury must be instructed to consider any delay; but, if a defendant does not challenge the victim’s credibility and the state therefore does not call con- stancy witnesses, the jury will be instructed to disregard any delay. We cannot reconcile this disparate outcome considering the otherwise parallel nature of these cases. See E. Prescott, Tait’s Handbook of Connecticut Evi- dence (6th Ed. 2019) §§ 6.33.4 and 6.33.5, pp. 420–21. Moreover, if we were to abandon only the Daniel W. E. instruction but leave in place the rest of our holding, all that would remain of Daniel W. E. is the requirement that a defendant must attack a victim’s credibility for the state to introduce constancy testimony. This requirement does little in practice, as a defendant will almost invari- ably attack a victim’s credibility in sexual assault cases, whether through cross-examination, as part of the defen- dant’s own presentation of evidence or in argument.

We therefore return our constancy of accusation doc- trine to its earlier status under Troupe. We believe that Troupe, along with the other avenues available to negate juror biases, such as expert testimony, sufficiently bal- ance the defendant’s interest in not being unreasonably burdened by constancy testimony with the state’s inter- est in overcoming potential jury bias against victims of sexual abuse who delay in reporting. As illustrated by the testimony of Williams, we specifically allow expert testimony about the general behavioral characteristics of sexual abuse victims, including delayed reporting. This testimony, if credited by the jury, combats any skepti- cism the jury might harbor toward sexual assault vic- tims who delay in reporting, without requiring the jury to take delay entirely off the table when assessing the victims’ credibility or risking that the jury will so con- strue the court’s instructions. See State v. Favoccia, 306 Conn. 770, 780, 51 A.3d 1002 (2012). February 11, 2025 CONNECTICUT LAW JOURNAL Page 17

351 Conn. 213 FEBRUARY, 2025 227 State v. Adam P.

It is also important to point out that Troupe already directs that trial courts substantively provide the reformed jury instruction the defendant in the present case requests, namely, one that instructs that delay is a factor jurors may consider when evaluating witness credibility. See State v. Troupe, supra, 237 Conn. 305 (‘‘the defendant is entitled to an instruction that any delay by the victim in reporting the incident is a matter for the jury to consider in evaluating the weight of the victim’s testimony’’). Thus, the defendant’s argument against a return to Troupe and, in turn, for maintaining part of the constancy of accusation doctrine we articu- lated in Daniel W. E., relies on the possibility of undue prejudice to the defendant from the state’s presentation of multiple constancy witnesses. But, in Troupe, we expressly addressed this concern, concluding that a trial court’s responsibility to carefully consider and limit the detail that constancy witnesses may provide suf- ficed to offset the prejudice a defendant might suffer from multiple constancy witnesses. See id., 302–303. We further observe that ordinary rules of evidence already safeguard defendants against this prejudice. Section 4-3 of the Connecticut Code of Evidence pro- vides, for example, that relevant evidence may be excluded if it wastes time or needlessly presents cumu- lative evidence. Thus, if constancy testimony is unnec- essarily duplicative, trial courts already have the tools to vigilantly protect the defendant from undue preju- dice. We must, however, distinguish between what the defendant in Daniel W. E. described as the state’s ‘‘ ‘pil- ing on’ ’’ of multiple constancy witnesses; State v. Dan- iel W. E., supra, 322 Conn. 618; and the state’s presentation of a strong case involving several con- stancy witnesses who testify in a limited manner, con- sistent with Troupe. The former requires that trial courts not admit needlessly cumulative2 evidence, 2 We highlight that our Code of Evidence specifies that evidence may not be needlessly cumulative. See Conn. Code Evid. § 4-3. Two constancy Page 18 CONNECTICUT LAW JOURNAL February 11, 2025

228 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

whereas the latter strongly negates any inference that a sexual assault victim’s delayed reporting is evidence of fabrication. Troupe also already contains other limitations meant to protect defendants from undue prejudice. For exam- ple, under Troupe, constancy witnesses ‘‘may testify only with respect to the fact and timing of the victim’s complaint; any testimony by the witness regarding the details surrounding the assault must be strictly limited to [the details] necessary to associate the victim’s com- plaint with the pending change . . . .’’ State v. Troupe, supra, 237 Conn. 304. Additionally, trial courts must witnesses who testify about a victim disclosing an assault to them on sepa- rate occasions, as in Daniel W. E., does not necessarily result in needlessly cumulative testimony. See State v. Parris, 219 Conn. 283, 293–94, 592 A.2d 943 (1991) (‘‘[E]ach item of the state’s constancy evidence . . . pertained to a different statement that the victim had made to a different person at a different point in time. . . . [T]herefore, the evidence covered new matter by demonstrating . . . that the victim previously had reported the incident . . . in a constant and consistent fashion.’’); see also State v. Kelly, 256 Conn. 23, 38–40, 770 A.2d 908 (2001) (overlapping constancy testimony still was not necessarily unfairly duplicative following Troupe). Rather, such testimony serves the purpose Troupe contemplated: it undercuts any notion that a victim must be lying if she delayed making an official report. See State v. Kelly, supra, 38; State v. Parris, supra, 293–94. This is not to say that constancy testimony can never be needlessly duplicative. For example, if a victim told a group of individuals about a sexual assault at a sleepover party, it might be needlessly cumulative for each individual to testify because each report would only reiterate the same disclosure, given in the same set of circumstances. However, if, at the same sleepover party, a victim told multiple individuals about different sexual assaults, or told one individual about a sexual assault and then told another person about the same sexual assault at a later date, that testimony would not be needlessly cumulative because each report would present ‘‘new matter . . . .’’ (Internal quotation marks omitted.) State v. Parris, supra, 293. But see R. Block, Note, ‘‘The New Face of Connecticut’s Constancy of Accusation Doctrine: State v. Troupe,’’ 29 Conn. L. Rev. 1713, 1738–39 (1997) (‘‘[T]he court [in Troupe] admitted that cumulative testimony could be prejudicial [and addressed this concern by holding that] detail testimony would no longer be allowed . . . [but] the change does not correspond to the concern. . . . There will still be the enhanced risk that the jury may be unduly swayed by the repeated iteration of the constancy of accusation testimony.’’ (Footnotes omitted; internal quotation marks omitted.)). February 11, 2025 CONNECTICUT LAW JOURNAL Page 19

351 Conn. 213 FEBRUARY, 2025 229 State v. Adam P.

provide a limiting instruction, stating that, because it is hearsay and the rules of evidence therefore consider it unreliable, any constancy testimony admitted pursu- ant to Troupe should not be considered substantively— that is, jurors are not permitted to use, and should not use, the testimony to help determine whether the accusation is true. Id., 305. By returning to Troupe, we affirm that trial courts, subject to appellate review, are in the best position to properly ‘‘balance the probative value of the evidence against any prejudice to the defen- dant,’’ pursuant to the standards articulated in Troupe and the ordinary rules of evidence. Id. Accordingly, we overrule the modification of the con- stancy of accusation doctrine established in Daniel W. E. and return to the standards we previously articulated in Troupe. Going forward, we encourage the Judicial Branch’s Criminal Jury Instruction Committee to mod- ify the constancy of accusation instruction to more closely follow the language used by the New Jersey courts in their 2013 revision of the charge. See New Jersey Model Criminal Jury Charges, Fresh Complaint: Silence or Fail- ure to Explain (revised April 15, 2013), available at https:// www.njcourts.gov/sites/default/files/charges/non2c029.pdf (last visited February 4, 2025);3 see also State v. Daniel 3 The April 15, 2013 revision of the New Jersey model charge provides: ‘‘The law recognizes that stereotypes about sexual assault complainants may lead some of you to question [complaining witness’s] credibility based solely on the fact that [he or she] did not complain about the alleged abuse sooner. You may or may not conclude that [complaining witness’s] testimony is untruthful based only on [his or her] silence/delayed disclosure. You may consider the silence/delayed disclosure along with all of the other evidence including [complaining witness’s] explanation for his/her silence/delayed disclosure when you decide how much weight to afford to [complaining witness’s] testimony.’’ (Footnote omitted.) New Jersey Model Criminal Charges, supra, Fresh Complaint: Silence or Failure to Explain. We note that the New Jersey model instruction includes a footnote that states that this instruction should not be used when there is testimony about ‘‘child sexual abuse accommodation syndrome,’’ akin to expert testimony about behaviors common to child sexual abuse victims, because New Jersey provides a separate, special instruction for such testimony. See id., n.1. Because our model jury instructions do not presently contain a special Page 20 CONNECTICUT LAW JOURNAL February 11, 2025

230 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

W. E., supra, 322 Conn. 615–16 and n.11 (rejecting defen- dant’s claim that Troupe instruction on limited purpose of constancy evidence misled jury and encouraging trial courts, going forward, to provide jury instructions in line with more detailed New Jersey model instruction, as revised to February 5, 2007, to inform jurors about limited purpose of constancy evidence). B Having concluded that we should abandon the approach to constancy evidence that we adopted under Daniel W. E., we must next address whether the instruction we directed trial courts to provide in that case was harmful, consequently entitling the defendant to a new trial. We exercise plenary review over a challenge to jury instructions because it presents a question of law. See, e.g., State v. Campbell, 328 Conn. 444, 529, 180 A.3d 882 (2018). To determine whether the defendant is entitled to a new trial because of an erroneous jury instruction, ‘‘we review the entire charge to determine if, taken as a whole, the charge adequately guided the jury to a cor- rect verdict. . . . In appeals not involving a constitu- tional question [we] must determine whether it is reasonably probable that the jury [was] misled . . . . [I]n appeals involving a constitutional question, [how- ever, the standard is] whether it is reasonably possible that the jury [was] misled.’’ (Citation omitted; emphasis in original; internal quotation marks omitted.) State v. Ortiz, 343 Conn. 566, 594, 275 A.3d 578 (2022); see State v. Tatum, 219 Conn. 721, 734, 595 A.2d 322 (1991) (‘‘ ‘[t]he ultimate test of a court’s instructions is whether, taken as a whole, they fairly and adequately present instruction for that type of expert witness, but only a standard instruction for expert testimony generally, we encourage trial courts to rely on the body of the New Jersey model charge without paying heed to the exception in the footnote. February 11, 2025 CONNECTICUT LAW JOURNAL Page 21

351 Conn. 213 FEBRUARY, 2025 231 State v. Adam P.

the case to a jury in such a way that injustice is not done to either party under the established rules of law’ ’’). If the instruction involves a constitutional violation, the state bears the burden of demonstrating harmlessness, whereas, if the instruction does not, the defendant bears the burden of demonstrating harm. See State v. John- son, 316 Conn. 45, 58, 111 A.3d 436 (2015) (‘‘ ‘[i]f an improper jury instruction is of constitutional magni- tude, the burden is on the state to prove harmlessness beyond a reasonable doubt’ ’’). Instructional errors vio- late due process—and, therefore, are of constitutional dimension—when they confuse the burden of proof, the presumption of innocence, or one of the essential elements of the crime. See State v. LaBrec, 270 Conn. 548, 557, 854 A.2d 1 (2004) (‘‘[w]e previously have con- sidered an instructional impropriety to be of constitu- tional dimension only when it has gone to the elements of the charged offense, the burden of proof or the pre- sumption of innocence, concepts that undeniably are fundamental to the notion of a fair and impartial jury trial’’); see also State v. Dash, 242 Conn. 143, 152, 698 A.2d 297 (1997); State v. Tillman, 220 Conn. 487, 503, 600 A.2d 738 (1991), cert. denied, 505 U.S. 1207, 112 S. Ct. 3000, 120 L. Ed. 2d 876 (1992). The defendant argues that the instructional error in the present case is of constitutional dimension, requir- ing the state to show that the error was harmless beyond a reasonable doubt. We are not persuaded. We have repeatedly held that, when a claim ‘‘does not involve the violation of a constitutional right, the burden rests upon [the defendant] to demonstrate the harmfulness of the court’s error.’’ State v. Cooper, 182 Conn. 207, 212, 438 A.2d 418 (1980); see also State v. Patterson, 276 Conn. 452, 471–72, 886 A.2d 777 (2005). More partic- ularly, we have expressed that general credibility instructions regarding constancy of accusation testi- mony are nonconstitutional in nature because they do Page 22 CONNECTICUT LAW JOURNAL February 11, 2025

232 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

not, on their own, confuse the elements of a crime, shift the state’s burden of proof to the defendant, or undermine the defendant’s presumption of innocence. See, e.g., State v. Jones, 337 Conn. 486, 509, 254 A.3d 239 (2020); State v. Daniel W. E., supra, 322 Conn. 610; see also State v. Roberto Q., 170 Conn. App. 733, 743, 155 A.3d 756 (citing Appellate Court cases), cert. denied, 325 Conn. 910, 158 A.3d 320 (2017). Credibility is often particularly important in sexual assault cases, and victims’ delayed reporting may be relevant to their credibility. In fact, what to make of delay is often a point of dispute between the parties in sexual abuse cases, with one side positing that delay is a symptom of a victim’s trauma, and the other claiming that it is consistent with a motivation to fabricate accusa- tions. Even so, neither delay nor credibility is an ele- ment of the charged offenses; nor do they shift the state’s burden of proof to the defendant. Cf. State v. Devalda, 306 Conn. 494, 507–508, 50 A.3d 882 (2012) (instructional error was of constitutional dimension when jury was told it could find restraint element, which is essential to kidnapping, proven if it found that victim had ‘‘ ‘acquiesced,’ ’’ but court failed to define ‘‘restraint by acquiescence’’ as limited to when victim is younger than sixteen or incompetent). And, although we have held that instructions mandating that jurors draw partic- ular inferences might violate due process, we have done so when those instructions ‘‘[relieve] the state of the burden of proving an essential element of the offense . . . .’’ (Citation omitted; emphasis added.) State v. Wil- liams, 199 Conn. 30, 36, 505 A.2d 699 (1986); see also State v. Bunkley, 202 Conn. 629, 657–58, 522 A.2d 795 (1987) (instruction on mandatory legal effect of police officers’ conduct as to defendant’s liability did not vio- late due process because ‘‘the challenged statement was not given as part of the court’s instruction on an element of the crimes charged’’). Because the instructional error February 11, 2025 CONNECTICUT LAW JOURNAL Page 23

351 Conn. 213 FEBRUARY, 2025 233 State v. Adam P.

we have found in the present case neither shifted the burden of proof to the defendant, implicated the defen- dant’s presumption of innocence, nor dealt with an ele- ment of the charged crimes, we reject the defendant’s argument that the error is of constitutional dimension. We must now consider whether the defendant has demonstrated that it is not just reasonably possible, but reasonably probable—meaning more probable than not—that the instructional error misled the jury in arriv- ing at its verdict, therefore requiring reversal. See State v. Chapman, 229 Conn. 529, 544, 643 A.2d 1213 (1994) (‘‘[u]nder the circumstances . . . we cannot conclude that it is more probable than not that the court’s instruc- tional impropriety affected the result’’). This determina- tion is necessarily fact intensive, requiring that we look to the instructions provided, the parties’ theories of the case, and the larger record, all in their entireties, to ensure that ‘‘injustice is not done to either party under the established rules of law.’’ (Internal quotation marks omitted.) State v. Gomes, 337 Conn. 826, 849, 256 A.3d 131 (2021). ‘‘We evaluate the harmfulness of a noncon- stitutional error on the record as a whole. . . . An improper instruction is not automatically harmful merely because it adversely affects a defense.’’ (Cita- tions omitted.) State v. Ross, 230 Conn. 183, 215, 646 A.2d 1318 (1994), cert. denied, 513 U.S. 1165, 115 S. Ct. 1133, 130 L. Ed. 2d 1095 (1995). The defendant’s claim of harm requires that we conclude that the jury’s verdict turned on its misunderstanding, based on the Daniel W. E. instruction, that it could not consider the victims’ delayed reporting of the sexual abuse when assessing their credibility. See State v. Patterson, supra, 276 Conn. 473 (‘‘in the absence of [an instruction on the jailhouse informant’s credibility] the jury reasonably could not have found the defendant guilty . . . [b]ecause [the informant’s] testimony was so critical to the state’s case, and because the other evidence on which the state Page 24 CONNECTICUT LAW JOURNAL February 11, 2025

234 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

relied was so weak, we cannot say that the trial court’s failure to charge the jury specially regarding [the infor- mant’s] credibility was harmless’’ (citation omitted)). Considering the record as a whole, including the jury instructions, the theory of defense, and the strength of the evidence presented, we conclude that it is not reasonably probable that the erroneous instruction mis- led the jury because it had sufficient tools with which to judge the victims’ credibility, and neither the verdict nor the defendant’s theory of the case turned on delay. Beginning with the jury instructions, notwithstanding our rejection of the Daniel W. E. instruction that the victims’ delay ‘‘should not be considered . . . in evalu- ating their credibility,’’ we cannot conclude that it is reasonably probable that the jury believed that it was precluded, based on this portion of the charge, from considering the victims’ credibility more generally, and therefore, we are not convinced based on the instruc- tions alone that the jury was misled in arriving at its verdict. The instructions as a whole directed the jury to consider the victims’ credibility based on factors other than delay and emphasized that the state’s burden of proof does not shift to the defendant. There is no reason to believe that the jury, based purely on the Daniel W. E. instruction, would disregard the credibility factors the court provided as examples in its final charge or its instruction on the state’s burden of proof. See, e.g., State v. Beavers, 290 Conn. 386, 408, 963 A.2d 956 (2009) (in absence of contrary evidence, reviewing court presumes jurors followed trial court’s jury instructions). Specifically, the trial court, immediately before pro- viding the instruction at issue, stated that, to obtain a guilty verdict, the state must prove the requisite ele- ments beyond a reasonable doubt for each charge, and that that burden cannot shift to the defendant during trial. It then instructed that the jury was the sole judge February 11, 2025 CONNECTICUT LAW JOURNAL Page 25

351 Conn. 213 FEBRUARY, 2025 235 State v. Adam P.

of credibility and that it alone should determine ‘‘which testimony to believe, and which testimony not to believe.’’ It proceeded to describe the factors a jury might look to when considering a witness’ credibility, including whether ‘‘the witness [was] able to see, hear, or know the things about which the witness was testifying, how well was the witness able to recall and testify to these things . . . the witness’ manner while testifying, [whether] the witness [had] an interest in the outcome of the case, or any bias or prejudice concerning any party or any matter involved in the case, how reasonable was the witness’ testimony when viewed in light of all the evi- dence in the case, and [whether] the witness’ testimony [was] contradicted by what that witness [had] said or done at another time, or by the testimony of other witnesses, or by other evidence.’’ After the trial court provided the instruction that we have today disavowed,4 it instructed the jury on how to assess the credibility of expert witnesses, noting that, ‘‘[i]t is for you to consider the testimony with the other circumstances in the case and, using your best judg- ment, to determine whether you will give it any weight, and, if so, what weight you will give it.’’ Finally, the trial court again discussed with the jurors the elements of each charged offense, reminding them that, to find the defendant guilty of any charge, they must find that the state has proven every element of the charge beyond a reasonable doubt. 4 We emphasize that the trial court in the present case laudably undertook its best efforts to faithfully implement the direction that this court provided in Daniel W. E. and that the model criminal jury instruction for constancy of accusation testimony sought to implement. See Connecticut Criminal Jury Instructions 7.2-1, supra. For this reason, at least, we cannot conclude that the trial court committed plain error, as the defendant asks us to conclude in the alternative. See, e.g., State v. Turner, 334 Conn. 660, 684, 224 A.3d 129 (2020) (‘‘ ‘[i]t is axiomatic that the trial court’s proper application of the law existing at the time of trial cannot constitute reversible error under the plain error doctrine’ ’’). Page 26 CONNECTICUT LAW JOURNAL February 11, 2025

236 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

Our conclusion that it is not reasonably probable that the jury was misled by the erroneous instruction is supported further by the fact that the defendant’s princi- pal defense—that no sexual abuse had occurred and that the victims had fabricated such accusations—cen- tered on his insufficient opportunity to sexually abuse the victims and the absence of physical evidence, cate- gories wholly separate from any delayed reporting. The defendant’s reliance on delay in support of his fabrica- tion theory was minimal, and the jury was able to appreciate his defense even if it arguably believed it was constrained to disregard delay in assessing the victims’ credibility. See State v. Quintana, 209 Conn. 34, 47, 547 A.2d 534 (1988) (erroneous jury instruction was harmless when jury’s verdict hinged on factual determi- nations that were ‘‘ ‘not classically dependent’ ’’ on sub- tleties of legal issue); see also State v. Helmedach, 306 Conn. 61, 80, 48 A.3d 664 (2012) (instructional error was harmless when instruction could not have changed jury’s verdict). Specifically, while cross-examining the various wit- nesses, defense counsel repeatedly stressed the lack of physical evidence, highlighting, for example, that T claimed that the defendant had bitten her ear during a sexual abuse incident but did not leave a mark and that, although Q regularly took her and D to a doctor as children, T’s doctor had never asked about sexual or physical abuse. Counsel also questioned the validity of the police investigation, as neither Connecticut nor North Carolina detectives had made any effort to obtain the victims’ medical records from the time during which the alleged abuse occurred to corroborate their accu- sations. Defense counsel, again on cross-examination, also highlighted the defendant’s lack of opportunity to assault the victims because other adults lived in the home dur- ing the time the abuse occurred, with at least one adult February 11, 2025 CONNECTICUT LAW JOURNAL Page 27

351 Conn. 213 FEBRUARY, 2025 237 State v. Adam P.

living with them for three years. Defense counsel also pointed to the lengthy periods of time during which the victims and the defendant did not live in the same household, with D stating that they spent summers in South Carolina with extended family, away from both Q and the defendant. This focus on the other adults present in the home during the time the abuse occurred, and the distance between the victims and the defendant, necessarily encouraged the jury to question whether he would have had the opportunity to abuse the victims at all. Defense counsel’s closing argument further illus- trates that the primary theory of defense did not center on delay but, rather, on the defendant’s lack of opportu- nity to assault the victims and the lack of physical evidence. It also demonstrates that the jury was well equipped to assess the victims’ credibility using other factors, despite the erroneous instruction. Defense coun- sel suggested that it would not be possible for the vic- tims to be assaulted in such a violent manner without the other adults living in the home knowing about it, considering the small size of the apartments in which the abuse occurred. She laid out the many inconsisten- cies between the victims’ testimony that did not involve delay in reporting and explicitly asked the jury to con- sider those inconsistencies when assessing their credi- bility. For example, she noted that the victims’ testimony did not match up as to where the first incident had occurred, whether the defendant had threatened them, and whether the victims had talked to each other about the abuse. Defense counsel further highlighted the com- plete lack of physical evidence and the results of the victims’ medical examinations, emphasizing that there were no findings consistent with sexual abuse and going so far as to posit, ‘‘[w]e’re talking about . . . very rough sex on small children. . . . [The defendant is] not a small man. . . . Does this make sense?’’ Page 28 CONNECTICUT LAW JOURNAL February 11, 2025

238 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

Thus, notwithstanding the defendant’s efforts on appeal to portray the delay as the centerpiece of his defense, our comprehensive review of the testimony elic- ited and arguments made at trial reveals that the victims’ delay in reporting the defendant’s abuse played a rela- tively unimportant role in the defendant’s theory of the case. Even if we assume that the jury did not credit either of the victims’ testimony, which offered some alternative reasons for their disclosure that could be considered separately from delay,5 or the testimony of the state’s expert, who explained why many victims do not more promptly disclose the abuse they have endured at a very young age, the victims’ delay in reporting the defendant’s abuse was arguably relevant to their motive to fabricate their accusations. Yet, evidence of delay was only one piece of the defendant’s fabrication the- ory. On cross-examination, and again during closing, defense counsel questioned why the victims would not disclose the abuse earlier, once they were not living with the defendant during the summers and it was there- fore physically safe to do so. In closing argument, coun- sel suggested to the jury that the victims’ accusations were merely a recent bid for attention from Q. In other words, the defendant argued that the victims’ motiva- tion to fabricate was based on their relationship with Q, a credibility factor that the jury might consider as separate from delay. The record supports this distinction. On cross-exami- nation, defense counsel elicited that Q was absent for much of the victims’ childhoods because of her work schedule, that D believed that Q did not care that the defendant was sexually abusive, that nothing happened as a result of D’s disclosure until she was hospitalized, and that Q did not immediately come to see D in the hospital, despite having been notified. Defense counsel highlighted this testimony during closing argument, ask- 5 One such reason was the victims’ fear of pregnancy. February 11, 2025 CONNECTICUT LAW JOURNAL Page 29

351 Conn. 213 FEBRUARY, 2025 239 State v. Adam P.

ing the jury to consider that the victims made up their accusations because they did not want to give up their long sought-after attention from Q and blamed Q for their deteriorated emotional state leading up to and after their disclosure based on her absence from their lives. Defense counsel also stressed the damage that constantly moving between homes, starting at a new school, and not having a stable mother figure would have on the victims, and how that damage might mani- fest itself in the victims’ fabricating accusations to get attention from Q. Thus, the jurors were able to assess the defendant’s fabrication theory, even without consid- ering delay, by relying on the victims’ relationship with Q to inform their motivation to fabricate, as well as on general credibility factors provided by the defendant and the court. The only explicit mention during closing arguments of delayed disclosure as related to credibility was brief. Defense counsel recounted Williams’ expert testimony that victims might delay disclosure because of their love for the abuser and that they do not want to hurt the abuser, or because the victim lives with the abuser and fears retaliation. She contended that the victims’ proffered reasons for their delayed disclosure were inconsistent with those Williams described because the victims felt no loyalty toward the defendant and were not in close proximity to him for extended periods of time. Therefore, as we previously discussed, even if defense counsel did not highlight delay during closing argument because of the Daniel W. E. instruction, the jury was able to assess the victims’ credibility by consid- ering other factors, such as their consistency while testi- fying, that both the trial court and defense counsel had emphasized. Lastly, we consider the strength of the evidence pre- sented at trial. The defendant contends that the state’s case was weak and, therefore, that the instruction we Page 30 CONNECTICUT LAW JOURNAL February 11, 2025

240 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

now reject was more likely to be harmful because ‘‘[t]here was no forensic evidence, no physical evidence, and no witnesses other than the [victims].’’ Upon our review of the record, however, we conclude to the con- trary. The strength of the state’s case buoys our determi- nation that it is not reasonably probable that the erroneous instruction misled the jury. See State v. Pat- terson, supra, 276 Conn. 472 (instructional error was exacerbated by, and therefore harmful because of, lack of corroborative evidence); State v. Ruth, 181 Conn. 187, 197, 435 A.2d 3 (1980) (harmlessness of instruc- tional error was supported principally by ‘‘the over- whelming nature of the evidence’’). The state’s case, even without physical or forensic evidence, was strong. See State v. Felix R., 319 Conn. 1, 18–19, 124 A.3d 871 (2015) (‘‘a [sexual abuse] case is not automatically weak just because . . . [of] the lack of conclusive physical evidence corroborating sex- ual assault, especially given the corroborating evidence introduced at trial’’). The victims testified extensively, and largely consistently about the defendant’s actions, at least as to the type of idiosyncratic details that Wil- liams noted were indicative of children truthfully recall- ing sexual abuse.6 The state also presented multiple 6 The victims testified that the first incident of sexual abuse began with the defendant’s asking if they would like to play a game that T described as ‘‘licking cups.’’ D then testified that the defendant orally, vaginally, and anally assaulted them in the dining room, and T testified that the defendant orally and vaginally assaulted them after forcing them to watch a porno- graphic film. D testified to another incident when the defendant forced her to watch a pornographic film while he vaginally assaulted T and, following the film, orally and vaginally assaulted D. T also testified that the defendant vaginally assaulted her twice without D present. The victims further testified about an incident that started with the defen- dant’s asking if they wanted to play the game and ended with Q’s returning from work. D said that she saw the defendant take T into Q’s bedroom and, worried for her sister, went inside to ask what they were doing. After D entered the room, the defendant forced her to digitally penetrate him while he orally assaulted T. D then described the defendant’s ordering her to leave the room because she had digitally penetrated him ‘‘so badly,’’ at which point she sat outside Q’s bedroom door. T’s account of this incident was February 11, 2025 CONNECTICUT LAW JOURNAL Page 31

351 Conn. 213 FEBRUARY, 2025 241 State v. Adam P.

witnesses who bolstered the victims’ credibility, includ- ing three detectives and Williams. The detectives cor- roborated the victims’ timeline of events, noted that, in their experience, the lack of physical evidence was common in cases involving late disclosures by children of sexual assault, and detailed how they investigated and assessed the reliability of the victims’ allegations. With respect to reliability, Detective Thomas Russell Varnadoe, the lead investigator into the victims’ claims in North Carolina, testified that the victims underwent separate forensic and physical examinations conducted by a multidisciplinary team of prosecutors, police offi- cers, social workers, and others who have worked on sexual assault investigations. Varnadoe stated that he observed the interviews of the victims, transcripts of which were admitted into evidence at the time of trial, and that the interviews informed his investigation. Fol- lowing those examinations, law enforcement was able to establish a reliable timeline of when and where the incidents occurred, in part by corroborating certain details with Q, such as the color or makeup of the house in Stratford, and by confirming where the victims lived and for what period of time through the use of a police database that provided information that was largely consistent with the information provided by Q and the victims. Williams also bolstered the victims’ credibility by testifying at length about the general behavior of child sexual abuse victims, including changes in behav- ior and delayed reporting. She explained the common largely consistent with D’s account. T testified that the defendant put on a pornographic film and proceeded to orally assault D while T watched and, in short order, attempted to force T to engage in oral sex with D. T further testified that the defendant, after D said she did not like it and did not want to continue, told D to leave the room and stand as a lookout for when Q returned from work. Once D left the room, T testified, the defendant vaginally and orally assaulted her. Although the victims testified only as to the details of the previously mentioned incidents, D further stated that, at least at the Stratford apartment, the abuse happened ‘‘every day . . . .’’ Page 32 CONNECTICUT LAW JOURNAL February 11, 2025

242 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

triggers for reporting abuse and factors that might make a disclosure more or less credible, including a recounting of ‘‘important idiosyncratic details, contextual details, peripheral details . . . [that] wouldn’t be there if some- thing didn’t happen to them because they’re not typi- cally details [someone would have] if . . . it didn’t happen . . . .’’ Additionally, although there are limits to what infer- ences we may properly draw from a jury’s verdict, the jury’s split verdict in the present case increases our confidence that the erroneous instruction was harmless because what the jury split its verdict on indicates that it assessed the victims’ credibility. See, e.g., State v. Angel T., 292 Conn. 262, 294, 973 A.2d 1207 (2009) (‘‘the split verdict suggests that the jury had doubts concern- ing [the complainant’s] credibility as a general matter, as it failed to credit [some of] her testimony’’). Specifi- cally, the jury found the defendant not guilty of two counts of sexual assault in the first degree as to inci- dents involving only T. The state alleged that both inci- dents took place in Stratford. The jury found the defendant guilty of one count of sexual assault in the first degree in connection with an incident in which D was the sole complainant that the state also alleged took place in Stratford. The jury came to different con- clusions about the veracity of the victims’ claims, despite their other similarities: both sexual assault inci- dents allegedly took place in Stratford and involved one victim without the other present. Therefore, we are comfortable inferring, based on the present record, that the jury must have made a credibility determination based on something besides delay to find the defendant not guilty of those two counts involving only T. Given our careful review of the trial record, including the victims’ extensive testimony and corroborating tes- timony by other witnesses, we are confident that the jury had sufficient grounding on which to make a credi- February 11, 2025 CONNECTICUT LAW JOURNAL Page 33

351 Conn. 213 FEBRUARY, 2025 243 State v. Adam P.

bility determination and that the jury’s verdict turned on that general credibility determination rather than specifically on the victims’ delayed reporting. Because the trial court’s jury instructions were adequate as a whole, and the strength of the state’s case corroborated the victims’ credibility, we conclude that it is not reason- ably probable that the instructional error we identify today misled the jury in arriving at its verdict. II The defendant also claims that the trial court abused its discretion by permitting D to testify that he had told her that he previously played the same sexual games with his daughter, A, that he was ‘‘playing’’ with the victims. He contends that the prejudicial effect of D’s testimony, admitted as a statement of a party opponent to show that the defendant was grooming D for sexual abuse, outweighed its probative value because it sug- gested to the jury that the defendant also had sexually abused A.7 We disagree. Prior to trial, the state provided written notice that it intended to offer evidence that the defendant had told D that the ‘‘game’’ he played with her and T was the same ‘‘game’’ he played with A, and that D was the only one who did not like the game. The defendant responded that the evidence was not probative of a charged offense and was unduly prejudicial because ‘‘the jury’s knowledge of these unsupported allegations 7 The defendant alternatively contends that D’s testimony ‘‘would have been inadmissible as uncharged [sexual] misconduct.’’ The defendant, how- ever, concedes that the trial court admitted D’s testimony as a statement by a party opponent, not as uncharged sexual misconduct, and that the trial court instructed the jury not to consider D’s testimony as uncharged misconduct evidence. The trial court therefore never exercised its discretion as to whether to admit or exclude D’s testimony as evidence of uncharged sexual misconduct, and we will not review a hypothetical ruling of the court. See, e.g., State v. Juan J., 344 Conn. 1, 13–14, 276 A.3d 935 (2022) (reviewing court cannot determine whether trial court abused its discretion in admitting evidence for different purpose). Page 34 CONNECTICUT LAW JOURNAL February 11, 2025

244 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

could clearly tip the balance in favor of the state and unfairly alter the outcome of the trial.’’ After hearing argument, the court orally ruled that the evidence was not overly prejudicial because D’s testimony was not ‘‘blatantly sexual in nature’’ com- pared to the allegations in the case that ‘‘the defendant was having . . . oral, anal, and vaginal sex with two little girls . . . .’’ The court indicated that it would per- mit the state to introduce the evidence as a statement by a party opponent for the limited purpose of showing a ‘‘form of bizarre peer pressure’’ or ‘‘grooming activity’’ but that, to minimize any prejudice to the defendant, it would instruct the jury that it should not use the evidence for propensity purposes or as evidence of uncharged sexual misconduct and would restrict any attempt by the state to argue that the defendant in fact sexually abused A. At trial, D testified on direct examination that the defendant had described the sexual abuse as a ‘‘game’’ to the victims. D testified that the defendant had asked her why she did not ‘‘like the game’’ and that, when she answered, ‘‘it feels uncomfortable,’’ he said that ‘‘it’s supposed to be.’’ Consistent with the court’s ruling, D further testified that the defendant had ‘‘told [her] that he did it before with another girl named A. And he was telling me that A had also enjoyed the game, that [D] was the only one that did not enjoy the game. So, he tried to make it seem—our conversation was trying to persuade me to enjoy the game more. So, I remember just seeming shocked that there was another girl out there that also was going through what me and T were going through. So, that was basically the concept of the conversation.’’ D also testified that she knew that A was the defendant’s daughter and that they had met before. The next morning, the court instructed the jury that D’s testimony regarding the ‘‘game’’ was not offered to February 11, 2025 CONNECTICUT LAW JOURNAL Page 35

351 Conn. 213 FEBRUARY, 2025 245 State v. Adam P.

establish that the defendant had engaged in any type of uncharged misconduct but, rather, for the limited purpose of describing the nature of the discussions or communications between the defendant and D. Later that day, A testified, stating that the defendant never acted inappropriately toward her. In its final charge, the court again directed the jury to use D’s testimony for the limited purpose for which it was offered. ‘‘It is well established that [s]tatements made out of court by a [party opponent] are universally deemed admissible when offered against him . . . so long as they are relevant and material to issues in the case.’’ (Internal quotation marks omitted.) State v. Tomlinson, 340 Conn. 533, 576, 264 A.3d 950 (2021). ‘‘Relevant evi- dence may be excluded if its probative value is out- weighed by the danger of unfair prejudice or surprise, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time or need- less presentation of cumulative evidence.’’ Conn. Code Evid. § 4-3. ‘‘Evidence is probative if it has any tendency to make the existence of any fact that is material to the determi- nation of the proceeding more probable or less probable than it would be without the evidence.’’ (Internal quota- tion marks omitted.) Doe v. West Hartford, 328 Conn. 172, 195–96, 177 A.3d 1128 (2018); see also Conn. Code Evid. § 4-1. ‘‘To be unfairly prejudicial, evidence must be likely to cause a disproportionate emotional response in the jury, thereby threatening to overwhelm its neutrality and rationality to the detriment of the opposing party. . . . All evidence adverse to an opposing party is inher- ently prejudicial because it is damaging to that party’s case. . . . For exclusion, however, the prejudice must be unfair in the sense that it unduly arouse[s] the [jurors’] emotions of prejudice, hostility or sympathy . . . .’’ (Citation omitted; internal quotation marks omitted.) State v. James K., 347 Conn. 648, 672–73, 299 Page 36 CONNECTICUT LAW JOURNAL February 11, 2025

246 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

A.3d 243 (2023). We review the trial court’s balancing of the potential prejudicial effect against its probative value for an abuse of discretion. See State v. Patterson, 344 Conn. 281, 298, 278 A.3d 1044 (2022). We conclude that the court did not abuse its discre- tion by permitting D to testify that the defendant had told her that he played the same sexual games with A. The testimony was probative of the defendant’s attempts to groom or pressure D to engage in sexual intercourse with him. The testimony was especially probative when paired with Williams’ expert testimony that child sex abusers often use grooming as a mechanism to coerce children to engage sexually with them and that referenc- ing the sexual abuse of another child in the same manner could be a form of grooming. The trial court mitigated any prejudicial effect of D’s testimony by admitting the testimony only for the limited purpose of establishing grooming and by repeatedly instructing the jury to use it only for that purpose. In the absence of evidence to the contrary, we presume that the jury properly fol- lowed the court’s instructions and used D’s testimony for that limited purpose. See, e.g., State v. Beavers, supra, 290 Conn. 408. Both of these mitigating measures mili- tate against a conclusion that the trial court abused its discretion. See, e.g., State v. Patterson, supra, 344 Conn. 301 (limiting instructions minimize prejudicial effect evidence might have); State v. Bermudez, 195 Conn. App. 780, 793, 228 A.3d 96 (2020) (limiting purpose of admitted evidence quells potential for unfair prejudice), aff’d, 341 Conn. 233, 267 A.3d 44 (2021). In light of these mitigating measures, we cannot say that the trial court unreasonably balanced the probative value of D’s testimony with any remaining unfair preju- dice. We agree with the trial court that D’s testimony did not likely cause a disproportionate emotional response in the jurors or unduly arouse their emotions because the substance of D’s challenged testimony was not bla- February 11, 2025 CONNECTICUT LAW JOURNAL Page 37

351 Conn. 213 FEBRUARY, 2025 247 State v. Adam P.

tantly sexual in nature compared to the bulk of the evidence in the case bearing on the state’s allegations of the defendant’s sexual contact with the victims. D did not explain the graphic details of the defendant’s physical contact with A, only that he had told her that A enjoyed playing the same game. See, e.g., State v. Jacobson, 283 Conn. 618, 639–40, 930 A.2d 628 (2007) (testimony that defendant had slept in same bed with another child to establish grooming behavior was not inflammatory or otherwise unfairly prejudicial because jury already had heard graphic testimony of abuse of sexual victims). Consequently, indulging every reason- able presumption in favor of the trial court’s ruling, we conclude that it was not a manifest abuse of discretion for the court to permit D to testify that the defendant had told her that he played the same sexual games with his daughter. The judgment is affirmed. In this opinion McDONALD, MULLINS and DAN- NEHY, Js., concurred.

ECKER, J., concurring in part and dissenting in part. I agree with the majority that State v. Daniel W. E., 322 Conn. 593, 142 A.3d 265 (2016), must be overruled because the jury instruction that we adopted and approved in that case incorrectly informs the jury that ‘‘there are many reasons why sexual assault victims may delay in officially reporting the offense, and, to the extent the victim delayed in reporting the offense, the delay should not be considered by the jury in evaluating the victim’s credibility.’’1 Id., 629; see part I A of the 1 Consistent with our holding in Daniel W. E., this instruction subsequently was incorporated into our model criminal jury instructions. See Connecticut Criminal Jury Instructions 7.2-1, available at https://jud.ct.gov/JI/Criminal/ Criminal.pdf (last visited February 4, 2025) (‘‘If no constancy of accusation witness testified, but there was a delay in officially reporting the offense,’’ then the following instruction should be given: ‘‘There was evidence in this case that the complainant delayed in making an official report of the alleged sexual assault. There are many reasons why sexual assault victims may Page 38 CONNECTICUT LAW JOURNAL February 11, 2025

248 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

majority opinion. This instruction was misleading because it prohibited the jury from considering a complainant’s delay in reporting sexual abuse allegations to assess the complainant’s credibility, while simultaneously instructing the jury that delayed reporting is consistent with credible allegations of sexual abuse in that ‘‘there are many reasons why sexual assault victims may delay in officially reporting the offense . . . .’’ (Emphasis added.) State v. Daniel W. E., supra, 629. The majority rightly rejects the state’s argument that the jury instruc- tions as a whole cured the defect by permitting the jury to consider the reasons for the delay, if not the fact of the delay itself, cogently reasoning that ‘‘the jury would [not] understand that the reasons why a victim might delay reporting sexual assault are necessarily separate from the delay itself based on the instructions provided, given that we have trouble discerning or articulating this distinction convincingly ourselves.’’ Part I A of the majority opinion. Unlike the majority, however, I cannot conclude that the erroneous jury instruction was harmless. See part I B of the majority opinion. The credibility of the com- plainants, T and D, was the central issue in the case, and the jury’s assessment of the truthfulness of their testimony was critical to the theory of the defendant, Adam P., that they had fabricated the sexual abuse allegations. There was no corroborative evidence of the sexual abuse; the jury’s verdict hinged entirely on the credibility of the complainants’ testimony. By forbid- ding the jury from drawing a negative inference about the complainants’ credibility from their nine year delay in reporting the abuse, the tainted instruction necessar- ily deprived the jury of a legitimate basis on which to form a reasonable doubt about the veracity of the delay in officially reporting the offense, and to the extent the complainant delayed in reporting the alleged offense here, the delay should not be consid- ered by you in evaluating (his or her) credibility.’’). February 11, 2025 CONNECTICUT LAW JOURNAL Page 39

351 Conn. 213 FEBRUARY, 2025 249 State v. Adam P.

complainants’ allegations. The erroneous instruction deprived the jury of information relevant and material to its resolution of the critical factual issue in the case— the credibility of the complainants’ sexual abuse allega- tions—and, on this record, I have no trouble concluding that it likely impacted the jury’s verdict. I would reverse the judgment of conviction and remand for a new trial before a properly instructed jury.2 The test for assessing harm depends on whether the instructional error was of constitutional magnitude. If the instructional error was ‘‘of a constitutional magni- tude, the state bears the burden of establishing that there is no reasonable possibility that the error affected the verdict. If, on the other hand, the error does not rise to the level of a constitutional violation, then a new trial is required only if the accused can demonstrate that the error probably affected the verdict.’’ State v. Breton, 235 Conn. 206, 243, 663 A.2d 1026 (1995). Under either standard, we have observed that child sexual abuse cases that boil down to a ‘‘credibility contest characterized by equivocal evidence . . . is a category of cases . . . far more prone to harmful error.’’ (Inter- nal quotation marks omitted.) State v. Favoccia, 306 Conn. 770, 816–17, 51 A.3d 1002 (2012). The majority concludes that the instructional error in this case was not of constitutional magnitude because it did not ‘‘confuse the elements of [the] crime, shift the state’s burden of proof to the defendant, or under- mine the defendant’s presumption of innocence.’’ Part I B of the majority opinion. But these are not the only circumstances in which we have recognized that an erroneous jury instruction implicates a defendant’s con- 2 To the extent that the issue is likely to arise on remand, I agree with the majority that the trial court did not abuse its discretion in admitting D’s testimony regarding the defendant’s out-of-court statement that he pre- viously had played sexual games with his daughter. See part II of the major- ity opinion. Page 40 CONNECTICUT LAW JOURNAL February 11, 2025

250 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

stitutional rights. In particular, we have held that ‘‘[a]n improper instruction on a defense, like an improper instruction on an element of an offense, is of constitu- tional dimension.’’ (Internal quotation marks omitted.) State v. Prioleau, 235 Conn. 274, 284, 664 A.2d 743 (1995); see also State v. Gomes, 337 Conn. 826, 845- 56, 256 A.3d 131 (2021) (erroneous jury instruction on defense of inadequate police investigation was of con- stitutional magnitude); State v. Heinemann, 282 Conn. 281, 298, 920 A.2d 278 (2007) (fundamental constitu- tional right to present defense of duress ‘‘includes proper jury instructions’’ (internal quotation marks omitted)). The erroneous instruction at issue in this case struck at the very heart of the defendant’s defense, which was that the complainants’ allegations of sexual abuse were not credible. The discussion of the New Jersey Supreme Court’s analysis of a substantially similar jury instruction is instructive. In State v. P.H., 178 N.J. 378, 840 A.2d 808 (2004), the jury was instructed in relevant part that ‘‘a child may not complain or tell anyone of sexual abuse for a myriad of reasons including fear, ignorance or confusion. You therefore may not consider the child’s failure to complain as evidence weighing against the credibility of the child because silence is one of the many ways a child may respond to sexual abuse if it has occurred. . . .’’ (Emphasis in original; internal quotation marks omitted.) Id., 387. The court held that this instruction deprived the defendant in that case, P.H., of his constitutional right to present a complete defense because it ‘‘effectively barred the jury from considering [the complainant’s] delayed disclosure when assessing her credibility.’’ Id., 399. Because the issue of the complainant’s credibility was central and material both to P.H.’s defense and the outcome of the case, the court held that the erroneous jury instruction deprived P.H. of his constitutional right to a fair trial. Id., 400. February 11, 2025 CONNECTICUT LAW JOURNAL Page 41

351 Conn. 213 FEBRUARY, 2025 251 State v. Adam P.

To arrive at its conclusion to the contrary, the major- ity cites to State v. Jones, 337 Conn. 486, 509, 254 A.3d 239 (2020), State v. Daniel W. E., supra, 322 Conn. 610, and State v. Roberto Q., 170 Conn. App. 733, 743, 155 A.3d 756, cert. denied, 325 Conn. 910, 158 A.3d 320 (2017), to support the proposition that ‘‘we have expressed that general credibility instructions regarding con- stancy of accusation testimony are nonconstitutional in nature because they do not, on their own, confuse the elements of a crime, shift the state’s burden of proof to the defendant, or undermine the defendant’s presumption of innocence.’’ Part I B of the majority opinion. I believe that these cases are distinguishable, however, because none of them involved jury instruc- tions that expressly and affirmatively prohibited the jury from considering evidence relevant to the wit- nesses’ credibility. In Jones, we held that it was errone- ous to reject a request for a jury instruction on the special considerations pertinent to assessing the credi- bility of a jailhouse informant, but the omission of this instruction did not bar the jury from taking those con- siderations into account when evaluating the infor- mant’s credibility. See State v. Jones, supra, 491–92 and n.3, 494–95. Likewise, in Daniel W. E. and Roberto Q., the jury was instructed on the constancy of accusation doctrine, pursuant to which a complainant who is impeached on the ground of a delayed report of sexual assault may have his or her credibility rehabilitated through the introduction of evidence indicating that a timely informal allegation of sexual assault had been made. See State v. Daniel. W. E., supra, 605–607; State v. Roberto Q., supra, 740–41. The juries in those cases, however, were not instructed that the delay in reporting the sexual assault could not be considered by them at all in assessing the credibility of the complainants’ tes- timony. Ultimately, I need not decide whether the erroneous jury instruction in this case was of constitutional dimen- Page 42 CONNECTICUT LAW JOURNAL February 11, 2025

252 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

sion because, even under the nonconstitutional stan- dard, it is clear to me that the error likely affected the jury’s verdict. The credibility of the complainants, who were children when the sexual abuse occurred, was the central and outcome determinative issue in this case. The complainants were nineteen years old at the time of trial, and they testified in graphic detail about sexual abuse that took place many years earlier, when they were between the ages of five and eleven. The complainants’ testimony was not corroborated by any physical, forensic, or contemporaneous evidence; nor was the jury presented with any constancy of accusa- tion testimony or evidence of any uncharged sexual mis- conduct attributed to the defendant. The only evidence of the crimes committed by the defendant came from the complainants themselves.3 A child sexual abuse case that lacks physical evidence is not automatically weak, but neither is it automatically strong merely because the complainants testified that the abuse happened, especially if the prosecution rested exclusively on the uncorroborated testimony of com- plainants who were young children when the sexual abuse occurred years earlier. On the basis of my careful review of the evidentiary record in this case, I cannot agree with the majority that ‘‘[t]he state’s case, even 3 Aside from the complainants and the state’s expert witness, Danielle Williams, only four other witnesses testified for the state: (1) Heineman Gonzalez, a former detective with the Stratford Police Department who did not participate in the forensic interviews of the complainants in North Carolina but applied for a warrant for the defendant’s arrest in Connecticut; (2) Jonathan Policano, a detective with the Stratford Police Department who transported the defendant from North Carolina to Connecticut; (3) Thomas Russell Varnadoe, a detective with the Raleigh Police Department in North Carolina who investigated the complainants’ allegations of sexual assault, ascertained where the alleged sexual assaults had occurred, and notified the pertinent Connecticut authorities; and (4) the complainants’ mother, Q, who confirmed the locations where the complainants lived during the relevant time periods and testified that the complainants had disclosed the sexual abuse to her around the time of their fourteenth birthdays, approx- imately five months before an official complaint was reported. February 11, 2025 CONNECTICUT LAW JOURNAL Page 43

351 Conn. 213 FEBRUARY, 2025 253 State v. Adam P.

without physical or forensic evidence, was strong.’’ Part I B of the majority opinion. I recognize that an absence of physical evidence is not uncommon in child sexual abuse cases, but the problem in this case is not just an absence of physical or forensic evidence, but an absence of any corroborative evidence to support the credibility of the complainants’ testimony. Compare State v. Felix R., 319 Conn. 1, 19, 124 A.3d 871 (2015) (‘‘the state’s case was not weak due’’ in part to corroborative evi- dence, such as ‘‘the defendant’s purchase of a pregnancy test and morning after pills’’ for complainant), with State v. Freddy T., 200 Conn. App. 577, 599, 241 A.3d 173 (2020) (state’s case was weakened by lack of cor- roborative evidence). The state’s witnesses corrobo- rated the locations where the complainants lived when the alleged sexual assaults occurred, but those loca- tions were not in dispute at trial. See footnote 3 of this opinion. The disputed factual issue was not where the alleged sexual assaults had occurred but whether they had occurred at all. The majority contends that the complainants them- selves provide corroboration, because there are two of them rather than just one. See part I B of the majority opinion. This logic might have some force except for the fact that, contrary to the assertion of the majority, the complainants’ respective versions of events cannot be characterized as ‘‘largely’’ consistent. Id. To the con- trary, my review of the record persuades me that the complainants’ testimony was largely inconsistent as to key factual issues, such as the location in the home where the abuse occurred, the sex acts that were per- formed, and the participants involved.4 These inconsis- 4 For example, with respect to the first instance of sexual abuse that occurred in Bridgeport when the complainants were five years old, T testified that the sexual assaults occurred in her mother’s bedroom, where the defen- dant took T and D to watch a pornographic movie. According to T, the defendant put his penis in the complainants’ mouths and then told D ‘‘to get up and put her clothes on, and walk out.’’ After D left the room, the defendant proceeded to rape T vaginally. D’s description of the sexual Page 44 CONNECTICUT LAW JOURNAL February 11, 2025

254 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

tencies undermine the claim of corroboration and simultaneously strengthen the defendant’s argument that the complainants’ testimony was not credible.5 In addition to the complainants’ testimony, the jury heard expert testimony from Danielle Williams, a foren- sic interviewer with extensive experience working with child victims of sexual abuse. Williams testified about child sexual abuse accommodation syndrome, informing the jury that delayed reporting is not uncommon. In fact, according to Williams, ‘‘60 to 80 percent of the time, we’re going to expect a delay of a child coming forward.’’ Children delay reporting sexual abuse for a variety of reasons—because they love or fear their abuser, the abuser is the primary breadwinner in the family, or they do not understand that they are being abused. Williams acknowledged, however, that there are times when a child discloses sexual abuse immedi- ately. Williams also explained that she had not met the complainants in this case or watched their forensic interviews. T and D did not disclose the sexual abuse until around the time of their fourteenth birthday, approximately nine assaults differed. According to D, the first instance of sexual abuse did not take place in her mother’s bedroom but on a mattress in the dining room. D did not mention the viewing of a pornographic movie and testified that, in addition to performing oral sex on and vaginally raping T, the defendant performed oral sex on D, penetrated D’s vagina digitally, and raped D anally. Both versions describe horrific criminal conduct, of course, but that is not the point. The point is that the testimony of T and D contains substantial differences that cannot be characterized as corroborating in nature. There are additional illustrations of nontrivial inconsistencies between the com- plainants’ descriptions of their abuse, but I have attempted to minimize the need for graphic detail. 5 I wish to emphasize that my observations in this concurring and dis- senting opinion regarding the nature and quality of the evidence should not be understood to reflect my own personal views about the credibility of the complainants. It is not my role in this case to make credibility determina- tions; that role is reserved exclusively for the jury. See, e.g., State v. Favoccia, supra, 306 Conn. 786 (‘‘[t]he determination of the credibility of a witness is solely the function of the jury’’ (internal quotation marks omitted)). February 11, 2025 CONNECTICUT LAW JOURNAL Page 45

351 Conn. 213 FEBRUARY, 2025 255 State v. Adam P.

years after the abuse began and three years after it had stopped. The disclosure occurred a couple days after the complainants’ mother, Q, informed them that she had rekindled her romantic relationship with the defen- dant and was considering asking him to come live with them. T explained that she did not tell anyone about the sexual abuse earlier because the defendant had told her not to tell anyone about it, she did not want to upset Q when she was happy in her relationship with the defendant, and she feared backlash from the defen- dant. D testified that she did not tell anyone about the sexual abuse earlier because the defendant ‘‘was always there’’ and had told her that, if she disclosed the abuse, she would ‘‘go to jail, and bad things happen to people in jail.’’ During closing argument, counsel for both parties relied on the complainants’ delayed disclosure of the sexual abuse to support their respective theories of the case. The prosecutor posited that the complainants’ delayed disclosure was consistent with their testimony regarding the sexual abuse, citing the expert testimony of Williams. The prosecutor argued: ‘‘And what percent- age of child sexual abuse cases did . . . Williams say involved late disclosure? I believe she said 60 to 80 percent. Now, you may have been asking yourself, if this had been going on as long as it did, why didn’t the girls tell someone sooner? And to that, I would tell you to go back to the testimony of our expert. According to . . . Williams, there could be a litany of reasons why a child might not disclose to others right away. She mentioned secrecy. The girls testified that the defendant told them to keep it a secret, otherwise they would get in trouble, jail[ed] even, and that, if they kept it a secret, they would be rewarded. Also, the relationship between [them] and the defendant—up until that point, the defendant had been the one watch- ing the girls while their mother was working. There Page 46 CONNECTICUT LAW JOURNAL February 11, 2025

256 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

could have also been some confusion. According to the testimony of the girls, the defendant asked them if they wanted to participate in a game. According to . . . Wil- liams, at their age, the concept of sexual intercourse is limited, and I would argue it’s nonexistent. And, with- out a point of reference, they would have no reason to believe that what they were engaging in was not play, especially since it was with someone [who] was sup- posed to be looking after them, someone they had no reason not to trust up until that point.’’ Defense counsel urged the jury to find that the com- plainants’ delay in reporting was not due to sexual abuse accommodation syndrome but, instead, resulted from their recent fabrication of the sexual assault allegations to gain their mother’s undivided attention. Defense counsel argued: ‘‘And I want to pause and consider the testimony of the state’s expert . . . Williams. She talked a lot about the delay of disclosure. She talked about how it’s very common. She gave several reasons for it. She suggested that . . . the child knows and loves [the] abuser and does not want to hurt them. She also indicated that it could be because the child lived with the abuser and is in fear of retaliation. . . . [C]lose proximity can be a reason that the child is fearful. How- ever, when we consider this timeline, T and D were away from [the defendant] for long periods of time, and they weren’t in close proximity. They were far away. They were states apart. And we also heard from them that both girls felt no loyalty to him. They indicated they did not love him. They did not consider him a father figure. They just wanted their mother, and Q said what they would say about him is, ‘why don’t you have more time for us? Why is he always here? We want to be with you.’ D also indicated that they spent most summers in South Carolina. So that’s plenty of time away. There was the ten month period while they were in South Carolina, and then the two year period [when] February 11, 2025 CONNECTICUT LAW JOURNAL Page 47

351 Conn. 213 FEBRUARY, 2025 257 State v. Adam P.

they did not see Q at all. On top of that, Q testified that there were multiple breakups [during which the defendant] had moved out for periods of time. So, after this two year period of being moved back and forth between houses, Q . . . swoops in and brings the girls back to North Carolina. They’re finally back with their mother. They finally have her attention.’’ Defense coun- sel argued that Q ‘‘sacrificed everything in her life for this relationship [with the defendant], including her daughters,’’ and that the jury should consider ‘‘[w]hat . . . the girls possibly [could] say to their mother to keep [the defendant’s return] from happening.’’ During rebuttal argument, the prosecutor urged the jury to consider the complainants’ explanation as to why they did not disclose the abuse earlier, after they no longer lived with the defendant: ‘‘Now, I know there was discussion at some point [that] they were far apart from the defendant, and I don’t know if the point was being made, you know, why didn’t they disclose and say we’re no longer in proximity to him, but do you remember T’s testimony? She said that, despite the abuse . . . [the defendant] was now out of their life. [The defendant] was out of their life. They had no fear of abuse, and T specifically mentioned that she was concerned [about] what would happen to her mother if she told her. Her mental health. She also said that, while this was going on, she had concern[s] about telling her mom because she had never seen her mom that happy with someone. She was in love with him. I don’t think there’s any denying that.’’ In its charge to the jury, the trial court instructed the jury on the general credibility considerations applicable to all witnesses. Then it gave the erroneous instruction specifically relating to the issue of delayed reporting in this case: ‘‘There was evidence in this case that the complainants delayed in making an official report of the alleged sexual assaults. There are many reasons Page 48 CONNECTICUT LAW JOURNAL February 11, 2025

258 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

why sexual assault victims may delay in officially reporting the offense, and, to the extent the complain- ants delayed in reporting the alleged offense here, the delay should not be considered by you in evaluating their credibility.’’ (Emphasis added.) Immediately thereafter, the trial court instructed the jury on the credibility considerations applicable to the testimony of Williams, the state’s expert witness on sexual abuse accommodation syndrome. During its deliberations, the jury submitted multiple notes asking to rehear the testimony of T and D describ- ing the alleged sexual assaults. The jury’s careful con- sideration of the complainants’ testimony indicates that this evidence, and specifically the credibility of this evidence, was of vital importance to the jury’s delibera- tions. See, e.g., State v. Devalda, 306 Conn. 494, 510, 50 A.3d 882 (2012) (‘‘[w]e have recognized that a request by a jury may be a significant indicator of [its] concern about evidence and issues important to [its] resolution of the case’’ (internal quotation marks omitted)); State v. Miguel C., 305 Conn. 562, 577, 46 A.3d 126 (2012) (‘‘the jury’s note to the court during deliberations provides insight into the facts that the jury considered when it was reaching its verdict’’); State v. Moody, 214 Conn. 616, 629, 573 A.2d 716 (1990) (jury’s note requesting to rehear evidence was ‘‘a clear indication that [the jury] deemed the evidence . . . as being important’’). The importance of the credibility of the complainants’ testi- mony also is reflected by the jury’s split verdict finding the defendant not guilty on two counts of sexual assault in the first degree involving T alone but guilty of the remainder of the charged crimes. ‘‘[T]he split verdict suggests that the jury had doubts concerning the [com- plainant’s] credibility as a general matter, as it failed to credit [some of] her testimony . . . .’’ State v. Angel T., 292 Conn. 262, 294, 973 A.2d 1207 (2009). Given the importance of the complainants’ credibility to the jury’s February 11, 2025 CONNECTICUT LAW JOURNAL Page 49

351 Conn. 213 FEBRUARY, 2025 259 State v. Adam P.

verdict, it is likely that the improper instruction on credibility ‘‘might have tipped the balance.’’ State v. Favoccia, supra, 306 Conn. 813.6 In light of the evidence before the jury, the central role the complainants’ credibility played in the outcome of the case, and the parties’ focus on the reasons for the complainants’ delay in reporting the sexual abuse either to strengthen or undermine the complainants’ credibility, I cannot conclude that the erroneous jury instruction was harmless. The reason the jury instruc- tion is erroneous, in short, is precisely the same reason that it was harmful on the present factual record. As explained by the majority, ‘‘the jury would [not] under- stand that the reasons why a victim might delay reporting sexual assault are necessarily separate from the delay itself based on the instructions provided . . . .’’ Part I A of the majority opinion. Stated another way, the complainants’ delay in reporting the sexual abuse and their reasons for the delay in reporting the sexual abuse were so inextricably intertwined, or at least so conceptually indistinct, that, by instructing the jury to disregard one (the delay), it is likely that the jury was misled to believe that it must also disregard the other (the reasons for the delay). The assistant state’s attorney candidly acknowledged during oral argument before this court that the reasons for the 6 I agree with the majority that we should exercise caution when drawing inferences regarding what a split verdict tells us about a jury’s thought process. See part I B of the majority opinion. But I disagree that the split verdict in the present case plausibly could increase our confidence that the erroneous jury instruction was harmless. The majority reasons that the split verdict demonstrates that the jury did, in fact, make credibility assessments notwithstanding the erroneous instruction. My point, however, is not that the erroneous instruction deprived the jury of any ability to make credibility assessments, or that the jury did not, in fact, make credibility assessments in reaching its verdict. Rather, the problem is that the erroneous instruction deprived the jury of the ability to include, as part of its credibility assess- ment, one potentially important consideration, namely, the complainants’ delay in reporting the assaults. Page 50 CONNECTICUT LAW JOURNAL February 11, 2025

260 FEBRUARY, 2025 351 Conn. 213 State v. Adam P.

complainants’ delay were ‘‘the centerpiece of the evi- dence . . . [and] both parties’ arguments’’ at trial, and that it would be ‘‘hard’’ to say that a misleading jury instruction regarding those reasons ‘‘didn’t make any difference’’ to the jury’s verdict on this record. I take this to be a significant concession with respect to the harmless error analysis. The instruction was harmful because it likely misled the jury regarding its consideration of the reasons for the complainants’ delayed reporting, and it did so in a manner that favored the state’s theory of the case. Revisiting the precise language of the challenged instruc- tion, I observe that it informed the jury that ‘‘[t]here are many reasons why sexual assault victims may delay in officially reporting the offense, and, to the extent the complainants delayed in reporting the alleged offense here, the delay should not be considered by you in evaluating their credibility.’’ This instruction, which explicitly refers to ‘‘sexual assault victims,’’ informs the jury that the reason it is prohibited from considering the complainants’ delay in reporting the sexual abuse is because the delay may be indicative of credible allega- tions of sexual assault pursuant to the testimony of the state’s expert witness. In this way, the erroneous instruction impermissibly bolstered the credibility of the state’s witnesses in two ways: by suggesting, first, that the complainants’ delayed reporting was consistent with that of ‘‘sexual assault victims,’’ and, second, that Williams’ expert testimony as to the ‘‘many reasons why sexual assault victims may delay in officially reporting the offense’’ must be believed. The error was particu- larly harmful in this case because the identity of the perpetrator was not at issue and the sole question for the jury was not who had committed the sexual assaults, but whether the sexual assaults had occurred at all (i.e. whether they had been fabricated). The faulty instruc- tion permitted the state to use the complainants’ delayed February 11, 2025 CONNECTICUT LAW JOURNAL Page 51

351 Conn. 213 FEBRUARY, 2025 261 State v. Adam P.

reporting as a sword to prove that the sexual assaults had occurred but did not permit the defendant to use it as a shield to defend against the crimes charged. This court has observed that, when a child sexual abuse case presents the jury with a ‘‘credibility contest characterized by equivocal evidence . . . [it] is far more prone to harmful error.’’ (Internal quotation marks omitted.) State v. Favoccia, supra, 306 Conn. 816–17. Additionally, we have cautioned that ‘‘[i]t cannot be harmless error to remove from the fact finder the very tools by which to make a credibility determination . . . . [When] credibility is an issue and, thus, the jury’s assessment of who is telling the truth is critical, an error affecting the jury’s ability to assess a [witness’] credibility is not harmless error.’’ (Citations omitted; internal quotation marks omitted.) State v. Fernando V., 331 Conn. 201, 223–24, 202 A.3d 350 (2019); see also State v. Ritrovato, 280 Conn. 36, 57–58, 905 A.2d 1079 (2006) (improper exclusion of complainant’s prior sex- ual conduct to impeach her testimony that she was virgin at time of sexual assault was not harmless error because ‘‘there was no independent physical evidence of the assault and no other witnesses to corroborate [the complainant’s] testimony, [and] her credibility was crucial to successful prosecution of the case’’); State v. Iban C., 275 Conn. 624, 641–42, 881 A.2d 1005 (2005) (improper admission of expert testimony that five year old complainant had been diagnosed with child sexual abuse was not harmless error because complainant’s ‘‘credibility was central to the state’s case,’’ and there was lack of corroborative evidence). This case hinged entirely on the credibility of the complainants’ allega- tions of sexual assault, and the erroneous jury instruc- tion deprived the jury of the tools that it needed to evaluate the truthfulness of their testimony. On this factual record, the erroneous jury instruction cannot be deemed harmless, and, therefore, I dissent from part I B of the majority opinion.