State v. Jose T. (Concurrence & Dissent)

************************************************ The "officially released" date that appears near the beginning of an opinion is the date the opinion will be published in the Connecticut Law Journal or the date it is released as a slip opinion. The operative date for the beginning of all time periods for the filing of postopinion motions and petitions for certification is the "officially released" date appearing in the opinion.

All opinions are subject to modification and technical correction prior to official publication in the Connecti- cut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports. In the event of discrepancies between the advance release version of an opinion and the version appearing in the Connecticut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports, the latest version is to be considered authoritative.

The syllabus and procedural history accompanying an opinion that appear in the Connecticut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports are copyrighted by the Secretary of the State, State of Connecticut, and may not be reproduced or distributed without the express written permission of the Commission on Official Legal Publications, Judicial Branch, State of Connecticut.

************************************************

1

McDONALD, J., with whom D'AURIA and ECKER, Js., join, concurring in part and dissenting in part. If the defense and the state agreed on one thing at trial, it was that, in the words of the prosecutor, "[t]his has always been a case about delayed disclosure." In such a case, where the jury must assess the relative credibility of the complainant and the defendant, there are only two possibilities. One possibility is that the complainant is telling the truth, that the defendant sexually assaulted her long ago, and that the complainant held off report- ing the crimes for one or more of the many reasons why, as we have come to understand, children often do delay reporting instances of sexual abuse. The other possibility is that, in this particular case, the defendant is telling the truth and the complainant falsely accused him of decade-old crimes, long past the point when he could convincingly defend himself. The more plausible the one theory, the less so the other. It is often difficult to assess these matters in isolation.

Here, however, it is virtually impossible. This case is unique, insofar as the defense argued that the prosecu- tor had offered at least two different sets of explana- tions as to why the complainant had delayed so long in reporting the alleged crimes. On one theory, the pros- ecutor suggested that the complainant had buried or repressed any recollection of the sexual assaults, until a school assignment triggered a series of flashbacks and the recovery of her memories. On another theory, or set of theories, the prosecutor argued that she had never forgotten the assaults. Rather, she stayed silent because of threats, fears, bribery, or other incentives. Unlike in other cases we have considered, then, a core component of the defense's theory of the case was that the com- plainant's account not only of the alleged crimes but of the delay itself was inconsistent, laden with lapses of memory, and rendered more problematic by what the defense argued was an inadequate police investigation. That is to say, the issue of delayed disclosure was more than simply one line of attack, interwoven with these other strands of the defense through the common thread

2

of credibility. It was a microcosm of the theories of the case of both the defendant, Jose T., and the state. Under these circumstances, although I agree with the major- ity that the trial court improperly instructed the jury regarding delayed disclosure; see part II A of the majority opinion; I cannot agree with the majority's conclusion that the instructional error was harmless. See part II C of the majority opinion. Therefore, I concur in part and dissent in part.

I

I begin with some brief background. In our prior cases, we explained how our modern assumptions regarding, and legal rules governing, the delayed disclosure¹of sexual assaults represent a departure from the common- law approach. Specifically, the modern approach repu- diates the common-law "fresh complaint" and "hue and cry" rules, which assumed (and, effectively, required) that crime victims would immediately cry out, alert their neighbors, and report any genuine crime. (Internal quotation marks omitted.) State v. Troupe, 237 Conn. 284, 294-306, 677 A.2d 917 (1996); see also, e.g., State

v. Daniel W. E., 322 Conn. 593, 618-28, 142 A.3d 265 (2016), overruled by State v. Adam P., 351 Conn. 213,

330 A.3d 73 (2025); State v. Marshall, 246 Conn. 799, 807-10, 717 A.2d 1224 (1998).

Although those traditional assumptions may still hold true for many crimes,²sex crimes, especially against

3

children, are different. Over time, we have come to under- stand that there are reasons why children, in particular, may not immediately reveal that they have been sexually abused. This was the principal reason why, in 2019, the legislature eliminated the statute of limitations for any crime involving the "sexual abuse, sexual exploitation or sexual assault" of a child. Public Acts 2019, No. 19-16, § 17 (P.A. 19-16); see also, e.g., State v. Sanderson, 239 Conn. App. 379, 395, 359 A.3d 847 (2026) (discussing P.A. 19-16, § 17).

But that choice was not without its trade-offs. Stat- utes of limitations serve to "prevent the unexpected enforcement of stale and fraudulent claims" and to "aid in the search for truth that may be impaired by the loss of evidence, whether by death or disappearance of wit- nesses, fading memories, disappearance of documents or otherwise." (Internal quotation marks omitted.) Neuhaus

v. DeCholnoky, 280 Conn. 190, 206-207, 905 A.2d 1135 (2006). When the state prosecutes a defendant for crimes that involve allegations of sexual abuse that occurred a long time ago, the defendant may find that the lengthy delay has made it effectively impossible to establish an alibi or to muster the physical evidence to defend against spurious charges.³

For these reasons, as this court has grappled over the years with the fairest and most appropriate judicial rules to apply to sexual assault prosecutions, we consistently have sought to strike a balance, "accommodati[ng] . . . the interests of the defendant, the state and the victim . . . ."

State v. Troupe, supra, 237 Conn. 303; accord State v. Adam P., supra, 351 Conn. 226; State v. Daniel W. E., supra, 322 Conn. 630. This need for balance is at the core of the issue of delayed disclosure.

4

II

Although I largely agree with the majority's statement of the harmless error standard that governs nonconsti- tutional errors; see part II C of the majority opinion; a brief discussion of the standard, as applied to cases of alleged child sexual abuse, will be helpful. These are tough cases. There typically is no physical or forensic evidence, no eyewitness or adverse codefendant, no video recording or cell phone data, and no confession. In short, these cases usually have none of the things that courts sometimes count among the ingredients that make up a

"strong" case. State v. Grenier, 257 Conn. 797, 807-808,

778 A.2d 159 (2001); see also, e.g., People v. Henderson,

9 Cal. 5th 1013, 1030, 470 P.3d 71, 266 Cal. Rptr. 3d 365 (2020). Rather, these cases often come down to a credibility contest between the complainant and the defendant. See, e.g., State v. Grenier, supra, 807-808. We know from the conviction whom the jury believed, but we can only speculate as to whether jurors might have formed some reasonable doubt if key evidence had been admitted or excluded, or misleading jury instruc- tions were not given.

Given the complexities, it is perhaps not surprising that we have struggled to reach consensus when assessing whether errors likely tipped the outcome in such cases. Compare, e.g., State v. Fernando V., 331 Conn. 201, 202 A.3d 350 (2019) (three to two decision, reversible error), and State v. Favoccia, 306 Conn. 770, 51 A.3d 1002 (2012) (five to two decision, reversible error), with State

v. Adam P., supra, 351 Conn. 213 (four to one decision, harmless error). In large part, these divergent outcomes reflect the highly "fact intensive" nature of the inquiry. State v. Adam P., supra, 233. But this court also has been less than consistent in how it has articulated the showing that a defendant must make to establish that a nonconstitutional error is so harmful as to warrant reversal.⁴

5

In any event, I agree with the majority that, in assess- ing the harmfulness of an incorrect jury instruction, we must consider, among other things, whether the defense's and/or the state's theory of the case hinged on the erroneous instruction, the overall strength of the state's case against the defendant, and whether the erroneous instruction was misleading in the context of the instructions as a whole. See part II C of the majority opinion; see also, e.g., State v. Adam P., supra, 351 Conn. 233-34; State v. Grenier, supra, 257 Conn. 807-12. To my mind, regardless of the precise standard applied, this is a relatively straightforward case, insofar as each of these factors clearly points in the direction of reversible error.

A

I first consider the parties' theories of the case and the extent to which those theories turned on the roughly ten year delay between the first charged sexual assault, it is more probable than not that the erroneous action of the [trial] court affected the result. . . . A second line of cases indicates that the defen- dant must show that the prejudice resulting from the impropriety was so substantial as to undermine confidence in the fairness of the verdict."

(Citations omitted; internal quotation marks omitted.) State v. Grenier, supra, 257 Conn. 807; see also, e.g., State v. Sawyer, 279 Conn. 331, 355, 904 A.2d 101 (2006) ("[f]ew areas of doctrinal development have been marked by greater twisting and turning than the development of standards for applying the harmless error rule" (emphasis in original; internal quotation marks omitted)), overruled on other grounds by State v. DeJesus, 288 Conn. 418, 953 A.2d 45 (2008). In 2006, this court in Sawyer purported to resolve these inconsistencies, settling on one single formulation of the standard: "[A] nonconstitutional error is harmless when an appellate court has a fair assurance that the error did not substantially affect the verdict." (Internal quotation marks omit- ted.) State v. Sawyer, supra, 357. Notwithstanding Sawyer, however, we have continued to use different language to articulate the governing standard. See, e.g., State v. Adam P., supra, 351 Conn. 233 (whether

"it is not just reasonably possible, but reasonably probable—meaning more probable than not—that the instructional error misled the jury"); State v. Fisher, 342 Conn. 239, 259, 269 A.3d 104 (2022) (whether

"the impropriety was so prejudicial as to undermine confidence in the fairness of the verdict" (internal quotation marks omitted)); State v. Fernando V., supra, 331 Conn. 223 (whether impropriety "could have made a substantial impact on the jury").

6

which allegedly occurred in 2008 or 2009 in the basement of the complainant's childhood home in Stamford, and the complainant's 2018 disclosure. From start to finish, the issue of delayed disclosure was central to how both parties litigated the case.

Long before trial began, the prosecutor gave notice that she intended to call a delayed disclosure expert. On the first day of trial, the prosecutor closed out her direct examination by asking the complainant, "[w]hy didn't you tell anyone sooner?" Defense counsel picked up right where the prosecutor left off, trying to pin down the complainant's answer to one question: "Why did you not tell it to anybody else, ever?"

The complainant gave what the jury could have under- stood to be differing answers to this key question. In her initial disclosures, the complainant explained that she did not report the sexual abuse until a school reading assignment triggered flashbacks, resurfacing long buried memories. She repeated this explanation at trial, testify- ing that the assignment had "triggered . . . memories that [she] kept buried inside [her]" and that, prior to February 2018, she had not thought about what happened to her. The complainant also testified that recollections of other incidents of abuse had come back to her in the years since her initial disclosure, although she had not mentioned them to anyone prior to trial. Her exact words were, "all the memories keep coming back, slowly."

In response to further questions regarding the delayed disclosure issue, however, the complainant gave what the jury could have understood to be a different set of explanations for not disclosing the alleged abuse sooner. She said that she "never forgot" about the sexual abuse but that she "was constantly told not to" disclose, and that she was afraid to do so, in light of the negative reac- tions two former schoolmates had when she told them about the abuse. The complainant testified that she "was in constant fear" and that she had been told by unnamed parties that, if she disclosed the abuse, she "would have to talk about it" in court. The complainant then offered

7

a third set of explanations, stating that the defendant had "kept bribing [her] by buying [her] stuff . . . to keep [her] quiet." When defense counsel asked her about these apparent contradictions, the complainant did not attempt to reconcile her various explanations. She simply agreed with counsel's statement that "having flashbacks" is

"different from being reminded to tell your story . . . ."

Following this line of questioning, defense counsel began to cross-examine the complainant about other alleged changes and inconsistencies in her story.⁵One aspect of the complainant's testimony highlighted by the defendant is the pronounced discrepancy between the complainant's answers to the prosecutor's questions and her answers to defense counsel's questions, especially with respect to matters of memory. On direct examina- tion, the complainant was able to recall details of events that had transpired approximately fifteen years prior, when she was about five or six years old. For example, when describing the 2008/2009 basement incident, the complainant recalled giving the defendant a hug, where he was in the laundry process, the specific small talk they engaged in, and the details of her mother's work schedule. By contrast, her response to dozens of defense counsel's questions was some version of "I don't remem- ber." Of particular note, when questioned about impor- tant ways in which her trial testimony diverged from the statements she had made during her initial forensic interview,⁶the complainant was unable to recall what she had said in that interview, even though she had recently

8

reviewed it in preparation for trial. During both cross- examination and recross-examination, defense counsel expressly related these potential credibility issues with the complainant's testimony to her claim that she had recovered buried memories of sexual abuse. This case is like other delayed disclosure cases, in that the jury had to resolve whether the lengthy delay in reporting reflected a typical child's response to sexual trauma or a recent fabrication. But it is also unique, in that the reasons for, and the nature of, the delay itself became a credibility issue. The defense's primary chal- lenges to the complainant's allegations—her inconsis- tencies, her memory lapses, her alleged animus toward the defendant—all played out in tandem with respect to her differing accounts as to whether she had delayed disclosing because the memories of sexual abuse were repressed or because of various carrots and sticks. The delayed disclosure issue remained front and center when the prosecutor questioned the other state wit- nesses. The complainant's oldest sister, N, testified on direct examination that she had asked the complainant several times why she "hadn't said anything before, and [the complainant] responded that she did not want to leave her sisters . . . ." When defense counsel raised a hearsay objection, which the trial court overruled, the prosecutor countered that the testimony was relevant to the complainant's state of mind because "it's come out several times about the delayed disclosure and the fact that she did not really disclose until 2018." Defense counsel likewise pursued the delayed disclosure issue early in his cross-examination of N.

Defense counsel's cross-examination of the state's next witness, Leslie Lugo, a police sergeant, focused almost entirely on Lugo's failure to pursue lines of inquiry that might have verified or discredited the complain- ant's claim that a schoolbook had triggered the recovery of her memories of sexual abuse. Once again, defense counsel expressly linked this line of attack—potential

9

shortcomings in the state's investigation—to the com- plainant's delayed disclosure of the alleged sexual abuse.⁷ The prosecutor also called the teacher to whom the complainant initially disclosed, Kaylin English, as a witness to establish, among other things, that the com- plainant was fearful of what would happen to her family when she revealed the sexual abuse, thus reinforcing an explanation for the delayed disclosure. Defense counsel objected unsuccessfully to that testimony, arguing that

"it would be highly prejudicial for the jury . . . to hear that and think, 'oh, that must be why [the complainant] didn't tell it before' . . . ."

The state's last witness was Monica Madigan, an expert on the delayed disclosure of child sexual abuse. Madigan testified at length about typical child sexual abuse dis- closure dynamics, connecting common patterns with the details of the complainant's story and the stated ratio- nales for her delay in reporting the alleged sexual abuse. Needless to say, Madigan's testimony centered around delayed disclosure and allowed the state to fully present its perspective on that issue to the jury. Defense counsel cross-examined Madigan about the scientific literature on delayed disclosure and the recovery of repressed memo- ries. Madigan acknowledged on cross-examination that inconsistencies in successive delayed disclosures, such as those that characterized the complainant's various accusations, "could be" "an indication of a false allega- tion . . . ."

10

After the state rested, defense counsel moved for a judgment of acquittal on the sexual assault and risk of injury to a child counts. See footnote 1 of the major- ity opinion and accompanying text. The prosecutor responded that both English's and Madigan's testimony

"goes toward the credibility of the [complainant] . . . [and] puts the [complainant's] testimony into context."

In short, both the prosecutor and defense counsel high- lighted the issue of the complainant's delayed disclosure of her allegations with respect to all but one of the state's witnesses. In addition, the defense expressly tied its other lines of attack—the complainant's inconsistent testimony, her flawed memory, the inadequate police investigation—to the delayed disclosure and the com- plainant's inconsistent and shifting rationales therefor. Indeed, the prosecutor repeatedly acknowledged that the complainant's delayed disclosure was central to the case. In seeking the trial court's leave to ask Madigan whether a school assignment relating to sexual assault could lead to delayed disclosure, the prosecutor noted that defense counsel "ha[d] made quite a [big] deal about the word 'trigger' and repressed memories and things like that." The prosecutor argued that "the reasons why a child may not disclose earlier . . . would certainly be rel- evant to the jury's determination and assessment of the credibility of the [complainant]." In response to defense counsel's objection to English's testimony, the prosecu- tor elaborated: "[T]he issue of why [the complainant] did not disclose earlier and her current state of mind at that time . . . is relevant . . . to the jury's determination of the facts in this case, the credibility of the witnesses, and the circumstances surrounding disclosure. . . . [T]he jury has to assess the [complainant's] credibility; that is essentially what this entire case is about. And . . . the fact that she maintains that fear even after she disclosed

. . . goes directly to her credibility in this . . . case, par- ticularly when that gets placed in the context of what

. . . Madigan is going to educate the jury about . . . ."

11

Ultimately, the prosecutor argued, "it can be probative of whatever the jury wants it to be probative of."

After the prosecutor repeatedly sought to introduce evidence suggesting that the complainant's testimony was consistent with a delayed disclosure of a bona fide childhood sexual assault, she requested a jury instruction that the jury could not consider the delay in evaluating the complainant's credibility. Defense counsel "strenu- ously" objected, arguing that the injustice of making the defendant defend against decade-old allegations would be compounded if the jury were allowed to consider only the state's perspective on "how the delay could bear on the veracity of the allegation[s]." Counsel summarized the defense's objection to the erroneous instruction in six words: "The delay is the whole defense."⁸(Emphasis added.) The prosecutor's response: "This has always been a case about delayed disclosure." (Emphasis added.) Although defense counsel believed that the trial court's instruction effectively precluded him from arguing the delayed disclosure issue to the jury, it was an important focus of closing arguments. The prosecutor, having just persuaded the trial court to issue the erroneous instruc- tion, repeatedly argued that the complainant's credibility was bolstered by the parallels between her account of the alleged sexual abuse and typical delayed disclosure pat- terns in bona fide child sexual assaults. And, although defense counsel never used the phrase "delayed disclo- sure," he argued at some length that the complainant's conflicting explanations for the delay exemplified her lack of credibility, her failure to testify truthfully, and the inadequacy of the state's investigation. For these reasons, I cannot agree with the majority's conclusion

12

that the issue of delayed disclosure was not central to this case. See part II C of the majority opinion.

B

The next consideration in the harmlessness analysis is the strength of the state's case against the defendant, which, the majority acknowledges, was "not particularly strong." Id. I agree with that assessment. As I will dis- cuss, the evidence here, although sufficient to persuade the jury of the defendant's guilt, was on par with that in other cases where we have found the state's case to be too weak to survive harmless error analysis. See part II D of this opinion.

As I discussed, the state's case against the defendant was founded almost entirely on the complainant's tes- timony. This court has characterized such cases as espe- cially prone to harmful error, given that any additional reason to question the complainant's veracity potentially could have allowed in some reasonable doubt. See, e.g., State v. Marcello E., 351 Conn. 345, 367, 330 A.3d 577 (2025) ("[c]redibility contest[s] characterized by equivo- cal evidence . . . [are] a category of cases that [are] far more prone to harmful error" (internal quotation marks omitted)); accord State v. Favoccia, supra, 306 Conn. 816-17; see also, e.g., State v. Sawyer, 279 Conn. 331, 359, 904 A.2d 101 (2006) ("a sexual abuse case lacking physical evidence is not particularly strong, especially when the victim is a child" (internal quotation marks omitted)), overruled on other grounds by State v. DeJe- sus, 288 Conn. 418, 953 A.2d 45 (2008). There were aspects of that testimony that a jury rea- sonably could have viewed with skepticism,⁹and, as the

13

defendant catalogs in his briefs to this court, the defense was able to identify and highlight various inconsisten- cies in the complainant's accounts, as well as significant memory lapses. Those inconsistencies and lapses extended to key issues in the case, ranging from the reasons for, and the nature of, the delayed disclosure to the types of sexual contact alleged.

When the only direct evidence of guilt is a complain- ant's long delayed accusation, constancy of accusation witnesses can help bolster the credibility of the accusa- tion. See, e.g., State v. Troupe, supra, 237 Conn. 296-98, 302-304. In this case, however, the prosecutor's attempt to elicit constancy testimony arguably backfired. The complainant could not recall the last names of the two former best friends in whom she allegedly had confided, and the jury was left to wonder why the state was unable to locate either classmate from school records in order to secure their corroborative testimony.

The same can be said of uncharged misconduct. A com- plainant's credibility can be substantially bolstered by independent testimony that the defendant abused other individuals in a manner similar to that charged. See, e.g., State v. DeJesus, supra, 288 Conn. 463, 468-70. Here, however, none of the defendant's other daughters or stepdaughters testified that he had sexually abused them, and there was no independent evidence of uncharged misconduct. The only such allegations were from the complainant's own account of a third incident that alleg- edly took place in Bridgeport in 2008 or 2009. But that account also evolved over time, and important, verifiable details of the complainant's story were directly rebutted by several defense witnesses.

In fact, the only corroborating evidence that the pros- ecutor offered¹⁰was testimony from N that, on the day assaulted the complainant in the front seat of a parked pickup truck while her youngest sister, M, then approximately nine or ten years old, sat in the backseat, with a clear view of the front.

14

of the basement incident, the complainant had stated that she and the defendant had been playing a game in which she was "his girlfriend." This testimony certainly was consistent with the complainant's story and could have bolstered her credibility.¹¹Its value was limited, however, in that any support for the complainant's alle- gations of sexual abuse was indirect, at best. See, e.g., State v. Fernando V., supra, 331 Conn. 215-16 and n.6 (state's case against defendant was not strong, despite state's presentation of comparable, corroborating tes- timony from complainant's brother); see also, e.g., id., 234-36 and n.7 (Robinson, C. J., dissenting) (discuss- ing testimony); cf. State v. William G., 232 Conn. App. 317, 334, 335 A.3d 511 (2025) (testimony of propensity witnesses did not specifically corroborate complainant's allegations).

A final consideration is the fact that the defendant voluntarily took the stand, testifying in his own defense and unequivocally denying the complainant's allegations of sexual abuse. Although the defendant also was far from an ideal witness, he was able to present the jury with a reasonable explanation of why the complainant might have fabricated the allegations in the wake of their heated dispute over money that he took from her to reimburse himself for some of the thousands of dol- lars that he spent on her quinceañera,¹²after which she became very angry and never spoke to him again.

C

A third relevant consideration is the charge itself:

whether, "taken as a whole, the charge adequately guided the jury to a correct verdict." (Internal quotation marks omitted.) State v. Adam P., supra, 351 Conn. 230. In this case, the delayed disclosure instruction was an incorrect corroborates her account of the underlying sexual assault allegations. See part II C of the majority opinion.

15

statement of the law, and this court must presume that the jury followed the trial court's erroneous instruction. See, e.g., State v. Carlos G., 354 Conn. 21, 28, 349 A.3d 1 (2026). The state offers two theories as to why, despite this presumption, the jury was not misled in the present case, neither of which I find persuasive. First, the state argues that the jury was properly instructed as to other legal matters: the burden of proof, the presumption of innocence, evaluation of expert tes- timony, and the jury's role as the ultimate arbiter of credibility. That is all true but largely beside the point. The defendant does not dispute that the trial court's instructions accurately set forth much of the relevant law, including the law addressing the jury's role in assess- ing witness credibility. Rather, the defendant contends that the erroneous delayed disclosure instruction made it impossible for the jurors to properly and thoroughly carry out that role. They were precluded from considering one of the primary factors that bore on the credibility assessment, one that, if properly considered alongside the others, likely could have given rise to reasonable doubt.¹³ Second, the state argues that the delayed disclosure instruction was not harmful because, rather than com- pletely foreclosing jurors from considering the complain- ant's delay in disclosing the alleged sexual abuse, the trial court instructed that they could not consider the

"mere fact" of delay. The gravamen of this argument is not entirely clear, at least to me. To the extent that the argument is simply that the delayed disclosure instruc- tion did not otherwise preclude the jury from considering

16

the complainant's credibility, separate and apart from delay related issues, that is true enough but, again, fails to address the defendant's claim.

To the extent that the state's claim is that the "mere fact" language barred jurors from considering the fact of the delay per se but permitted them to consider the reasons for the delay, this court recently rejected that very argument in Adam P. (Emphasis added.) See State

v. Adam P., supra, 351 Conn. 225 ("we do not believe that the jury would understand that the reasons why a victim might delay reporting sexual assault are neces- sarily separate from the delay itself based on the instruc- tions provided, given that we have trouble discerning or articulating this distinction convincingly ourselves"). Although the language of the erroneous instruction in Adam P. was different from the one in the present case; see id., 219; the same reasoning applies. The majority construes the state's second argument more ambitiously to mean that, under the challenged instruction, jurors could not consider delay in a vacuum, but they could consider it in the context of the broader case. See part II C of the majority opinion. That argument should be rejected for the same reasons that we rejected the legalistic distinction that the state proposed in Adam P. I do not know what it means for a jury, deliberating over all of the evidence in a criminal case, to consider delay in context but not in isolation, and, if I cannot make sense of it, presumably the jury could not either. If we take the instructional language seriously, then the mere fact that the complainant reported the alleged abuse after some delay, standing alone, becomes poten- tially relevant only in conjunction with other consid- erations, such as the length of the delay, the possible reasons for the delay, the background assumption that bona fide victims of sexual assault will not delay overlong in reporting such crimes, and valid doubts as to the per- sistence and reliability of memory over time. Are those things that the jury was permitted to consider under the trial court's delayed disclosure instruction? If yes,

17

then it was a superfluous instruction, and it is difficult to understand why the parties debated it at such length and with such urgency.¹⁴If no, if all of the key reasons why jurors may be dubious of lengthy reporting delays were off-limits, then the inclusion of the word "mere"

gives me far less solace than it does the majority. More- over, it is unclear to me how we are to ascertain which of the many delay-adjacent factors the jury would have understood to still be fair game under the trial court's modified instruction.

D

Tying all of this together, I believe that this case is much closer to child sexual assault cases such as Fernando

V. and William G., in which this court and the Appellate Court, respectively, found reversible error, than cases such as Adam P., in which we concluded that the error was harmless. In State v. Fernando V., supra, 331 Conn. 201, the state's delayed disclosure expert had testified as to the typical characteristics exhibited by victims of child sexual abuse following disclosure; see id., 205-206; and the issue was whether the trial court's improper exclusion of a defense witness—who would have testified that the complainant did not exhibit those characteristics—was reversible error. See id., 203, 206-209. In that case, as here, the only evidence of guilt was the complainant's testimony. See id., 215-16. That testimony suffered from various inconsistencies and contradictions; see id., 231 and n.5 (Robinson, C. J., dissenting); but found circumstantial corroboration in the testimony of a sib- ling, who twice saw the defendant and the complainant

18

acting in ways that suggested that they had been sexually engaged. See id., 215-16 and n.6; see also id., 234-36 and n.7 (Robinson, C. J., dissenting). In other ways, the state's case in Fernando V. was much stronger. There was strong constancy of accusation evidence, insofar as the complainant had disclosed the sexual abuse to her mother on several occasions before she finally was believed; see id., 204-205; and the defendant did not take the stand to present an alternative narrative. Nonetheless, we reversed the conviction, holding that "[i]t cannot be harmless error to remove from the fact finder the very tools by which to make a credibility determination" and that, 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'] cred- ibility is not harmless error." (Internal quotation marks omitted.) Id., 223-24.

In State v. William G., supra, 232 Conn. App. 317, the Appellate Court likewise concluded that the trial court's erroneous delayed disclosure instruction constituted reversible error. See id., 318, 325, 341. The complaining witness in that case had delayed disclosing the sexual abuse for only one month or so; see id., 319-21; and the state's case was stronger than in this case, insofar as two other girls testified as propensity witnesses that the defendant had engaged in similar, inappropriate behav- ior and touching. See id., 323, 334. Nevertheless, the error was held to be harmful because the complainant's credibility was the primary issue in the case and the ques- tion of delay was closely intertwined with the defense's focus on the complainant's inconsistent testimony and its contention that she had fabricated her allegations of sexual abuse. See id., 331-32, 334-35.

By contrast, in State v. Adam P., supra, 351 Conn. 213, the relevant factors pointed much more clearly in the direction of harmless error. We characterized the state's case there as "strong," as there were two complaining witnesses who "testified extensively, and largely consistently about the defendant's actions . . . ."

19

Id., 240. The complainants' independent forensic inter- views in Adam P. were specific and consistent, and other witnesses corroborated important details. See id., 240-

41. The defendant did not testify, and the defense's theory of the case rested heavily on his insufficient opportunity to sexually abuse the complainants and the lack of physical evidence to support the accusations. See id., 236. We characterized the defense's reliance on the complainants' delayed reporting as "minimal"; id.; and the parties made only one brief reference to delayed disclosure during closing arguments. Id., 239. Consider- ing the entire record, then, the rationales that animated our decision in Adam P. are largely absent here, and I see no meaningful basis for distinguishing the present case from Fernando V. and William G.

Accordingly, I concur in part and dissent in part.

20