The People of the State of Colorado v. Nelson Melara
No. 23CA1125
Colorado Court of Appeals
May 15, 2025
2025COA48
Honorable Patrick Butler, Judge
Boulder County District Court No. 21CR21
Thе summaries of the Colorado Court of Appeals published opinions constitute no part of the opinion of the division but have been prepared by the division for the convenience of the reader. The summaries may not be cited or relied upon as they are not the official language of the division. Any discrepancy between the language in the summary and in the opinion should be resolved in favor of the language in the opinion.
SUMMARY
May 15, 2025
2025COA48
No. 23CA1125, People v. Melara — Criminal Law — Model Jury Instructions — Presumption of Innocence, Burden of Proof, and Reasonable Doubt — Lack of Evidence Presented
As a matter of first impression, a division of the court of appeals concludes that the trial court did not err by providing the jury with a definition of “beyond a reasonable doubt” that did not expressly inform the jury that it could base its verdict on the “lack of evidence” in the case. Although the trial court should have included the “lack of evidence” language, the omission did not lower the prosecution‘s burden of proof and therefore did not error given the facts of this case.
COLORADO COURT OF APPEALS
Court of Appeals No. 23CA1125
Boulder County District Court No. 21CR21
Honorable Patrick Butler, Judge
The People of the State of Colorado,
Plaintiff-Appellee
v.
Nelson Melara,
Defendant-Appellant.
Division VI
Opinion by JUDGE SCHUTZ
Kuhn, J., concurs
Welling, J., specially concurs
Announced May 15, 2025
2025COA48
Philip J. Weiser, Attorney General, Josiah Beamish, Assistant Attorney General, Denver, Colorado, for Plaintiff-Appellee
Polansky Law Firm, PLLC, Lisa A. Polansky, Boulder, Colorado, for Defendant-Appellant
¶ 1 Defendant, Nelson Melara, appeals the trial court‘s judgment of conviction entered on jury verdicts finding him guilty of two counts of sexual assault on a сhild (position of trust). We affirm the judgment.
¶ 2 Melara was convicted of sexually abusing his daughter and granddaughter. The jury heard the following evidence that supports the verdicts.
¶ 3 Melara began sexually abusing his daughter when she was about eleven years old and repeatedly assaulted her until she was approximately fourteen years old. When daughter was old enough, she moved out of the family home and limited contact with her parents.
¶ 4 In 2019, daughter and her two children, a son and a daughter (who is the victim we refer to as granddaughter), moved back in with her parents due to ongoing financial difficulties. Granddaughter was four years old when they moved in. Daughter routinely asked both children whether anyone had touched their private parts.
¶ 5 In September 2020, when daughter asked granddaughter whether anyone had touched her vagina, granddaughter responded, “[N]o, but grandpa did.” Daughter followed up with granddaughter later that evening and asked her the same question. Granddaughter replied in the same way, and when asked whether she was hurt, granddaughter stated that “it hurt when [Melara] pushed up.”
¶ 6 In October, daughter and her children moved out of her parents’ home. Shortly after, daughter reported the conversation with granddaughter to the police. After an investigation, Melara was charged with two counts of sexual assault on a child (position of trust) — one relating to daughter and one relating to granddaughter — along with an aggravated sexual offense sentencе enhancer.
¶ 7 The charges were joined for a single trial held in August 2022. Defense counsel moved to sever the separate charges arising out of the assaults against daughter and granddaughter. The trial court denied the motion and proceeded to trial on both counts. The trial court declared a mistrial after lengthy deliberations resulted in a hung jury.1
¶ 8 The trial court set the matter for retrial in March 2023. Between January and March, defense counsel and the prosecution discussed a possible community-based plea but were unable to reach an agreement. After a five-day trial, the jury convicted Melara as charged. The trial court sentenced him to a combined term of thirty years to life in the custody of the Department of Corrections.
¶ 9 On appeal, Melara asserts that the trial court erred by (1) failing to advise the jury that it could consider the lack of evidence in the case when assessing whether the prosecution proved the charges beyond a reasonable doubt; (2) denying his motion to enforce a plea offer; (3) improperly limiting his expert‘s testimony; (4) failing to sever the charges involving daughter and granddaughter; and (5) admitting child hearsay statements. We address these contentions in turn.
II. The Reasonable Doubt Instruction
¶ 10 Melara contends that the trial court reversibly erred in the second trial by giving the 2022 version of the Colorado Model Jury Instruction (COLJI) defining reasonable doubt, which did not expressly refer to the jury‘s ability to base its verdict on the lack of evidence in the case.
A. Additional Facts
¶ 11 In 2021, when Melara‘s first trial took place, the COLJI model instruction for reasonable doubt read as follows:
Reasonable doubt means a doubt based upon reason and common sense which arises from a fair and rational consideration of all of the evidence, or the lack of evidence, in the case. It is a doubt which is not a vague, speculative or imaginary doubt, but such a doubt as would cause reasonable people to hesitate to act in matters of importance to themselves.
COLJI-Crim. E:03 (2021) (emphasis added).
¶ 12 In 2022, after Melara‘s original trial, the Model Criminal Jury Instructions Committee (committee) revised the model instruction for reasonable doubt. In addition to modifying
¶ 13 The instruction that the trial court gave at the second trial tracked the new model instruction, reading as follows:
Every person charged with a crime is presumed innocent. This presumption of innocence remains with [Melara] throughout the trial and should be given effect by you unless, after considering all the evidence, you are convinced that [Melara] is guilty beyond a reasonable doubt.
The burden of proof in this case is upon the prosecution. The prosecution must prove to the satisfaction of the jury beyond a reasonable doubt the existence of eaсh and every element necessary to constitute the crime charged. This burden requires more than proof that something is highly probable, but it does not require proof with absolute certainty.
Proof beyond a reasonable doubt is proof that leaves you firmly convinced of [Melara‘s] guilt. If you are firmly convinced of [Melara‘s] guilt, then the prosecution has proven the crime charged beyond a reasonable doubt. But if you think there is a real possibility that [Melara] is not guilty, then the prosecution has failed to prove the crime charged beyond a reasonable doubt.
After considering all the evidence, if you decide the prosecution has proven each of the elements of a crime charged beyond a reasonable doubt, you should find [Melara] guilty of that crime.
After considering all the evidence, if you decide the prosecution has failed to prove any one or more of the elements of a crime charged beyond a reasonable doubt, you should find [Melara] not guilty of that crime.
Id.
¶ 14 In its comments to the 2022 amendment, the committee explained that it modified the longstanding reasonable doubt instruction for three reasons: (1) to eliminate negative phrasing; (2) to remove the “hesitate to act” language in the previous instruction because it was unhelpful; and (3) to provide additional context. Id. at cmt. 1. The committee did not provide any specific explanatiоn for its removal of the “lack of evidence” language. See id. at cmt. 3.
¶ 15 In 2023, the committee again revised E:03, this time by adding the following sentence at the end of the first paragraph: “A reasonable doubt can be based on the evidence presented or the lack of evidence presented.” COLJI-Crim. E:03 (2023). Its comments to that edition state simply that, “[i]n 2023, the [c]ommittee added the final sentence to the instruction‘s first paragraph regarding evidence or lack of evidence.” Id. at cmt. 8. The committee provided no rationale for the additional sentence. See id.
B. Standard of Review and Applicable Law
¶ 16 A trial court must accurately instruct the jury concerning the controlling law. Riley v. People, 266 P.3d 1089, 1092 (Colo. 2011). We review de novo a trial court‘s jury instructions, as a whole, to determine whether the court met this obligation. Id. If the trial court‘s instructions accurately describe the applicable law, we review the court‘s decision whether to give a particular jury instruction for an abuse of discretion. People v. Paglione, 2014 COA 54, ¶ 45. A trial court abuses its discretion if its decision is manifestly arbitrary, unreasonable, or unfair, or if it misconstrues or misapplies the law. People v. Vigil, 2024 COA 72, ¶ 19.
¶ 17 But these general principles are nuanced in the context of defining the prosecution‘s burden of proof on a criminal charge. A defendant is presumed innocent of any charges filed against him. “The presumption of innocence, although not articulated in the Constitution, is a basic component of a fair trial under our system of criminal justice.” Perez v. People, 2013 CO 22, ¶ 16 (quoting Estelle v. Williams, 425 U.S. 501, 503 (1976)). This bedrock principle of our jurisprudence is given life and vitality through the prosecution‘s obligation to “prove every factual element necessary to constitute the crime charged beyond a reasonable doubt.” Johnson v. People, 2019 CO 17, ¶ 10 (citation omitted). This requirement
¶ 18 The trial court is obligated to instruct the jury that the defendant‘s guilt must be proved beyond a reasonable doubt. Id. However, the “Constitution does not require that any particular form of words be used” in advising the jury of the government‘s burden of proof. Id. (quoting Victor v. Nebraska, 511 U.S. 1, 5 (1994)). These principles limit a trial court‘s discretion when crafting a burden of proof instruction.
¶ 19 A trial court‘s exercise of that discretion must be done with caution to ensure that the instruction does not inadvertently lower the prosecution‘s burden. See, e.g., Tibbels v. People, 2022 CO 1, ¶¶ 49-60 (reversing a defendant‘s conviction because the trial court impermissibly lowered the burden of proof by analogizing reasonable doubt to the decision whether to buy a home with a cracked foundation). As our appellate courts have repeatedly cautioned, “further attempts by courts or parties to define ‘reasonable doubt’ do not provide clarity.” Johnson, ¶ 13; see also Holland v. United States, 348 U.S. 121, 140 (1954) (“Attempts to explain the term ‘reasonable doubt’ do not usually result in making it any clearer to the minds of the jury.” (quoting Miles v. United States, 103 U.S. 304, 312 (1880))).
¶ 20 An instruction that lowers the prosecution‘s burden below reasonable doubt constitutes structural error and requires reversal. Tibbels, ¶ 60. To assess whether the trial court lowered the reasonable doubt burden, we evaluate the “likelihood the jury applied the instructions in an unconstitutional manner.” People v. Garcia, 2021 COA 80, ¶ 26, aff‘d, 2023 CO 30. We look at the instruction holistically and “view it in the context of the record as a whole.” Id.
¶ 21 The model instructions are intended to be “helpful templates when drafting instructions.” COLJI-Crim. ch. A, term definitions (2024). But they are not definitive statements of the law or binding on courts. Krueger v. Ary, 205 P.3d 1150, 1154 (Colo. 2009). Thus, using a model instruction does not provide trial courts with “a safe harbor that insulates instructional error from reversal.” Garcia v. People, 2019 CO 64, ¶ 22.
C. The Parties’ Contentions
¶ 22 Melara argues that using the 2022 COLJI reasonable doubt instruction violated his due process rights because the absence of language informing the jurors that they could consider a lack of evidence impermissibly lowered the prosecution‘s burden of proof. See Johnson, ¶ 9 (“We must determine whether the trial court‘s [extraneous] ‘hesitate to act’ instruction lowered the prosecution‘s burden of proof in violation of due process.“). He argues that failing to inform the jury that reasonable doubt may arise from the lack of evidence in a case undermines the presumption of innocence. As evidence of the material impact of the omission, Melara notes that the jury in the first trial — which received the reasonable doubt instruction informing it that it could consider the lack of evidence in the case — did not convict him. In contrast, when this language was not included in the instruction given to the jury in the second trial, it returned guilty verdicts on the same charges. This inconsistency, he argues, demonstrates that the omission of the “lack of evidence” language lowered the burden of proof and led to his conviction in violation of his due process rights.
¶ 23 The People reason that the burden of proof instruction given at his second trial accurately stated the law and did not lower the prosecution‘s burden. The People also note that several jurisdictions across the nation have long used a burden of proof instruction that does not expressly state that the jury may consider the absence of evidence in the case, and that courts in those jurisdictions have repeatedly upheld such instructions. See, e.g., United States v. Rogers, 91 F.3d 53, 56-57 (8th Cir. 1996) (“The additional language, in any event, says nothing that is not already obvious to people of common sense. That a lack of evidence may cause one to have a reasonable doubt is self-evident.“); Rault v. Louisiana, 772 F.2d 117, 136 (5th Cir. 1985) (“Nothing in the charge suggested that a reasonable doubt could not arise from the lack of evidence.“); State v. Wienke, 2022 MT 116, ¶ 29 (The trial court did not err by rejecting a reasonable doubt instruction tendered by the defense that included the “lack of evidence” language.). Consistent with these authorities, the People argue that it is not necessary to instruct jurors that they are allowed to consider the lack of evidence in a case because jurors will apply their common sense to consider a lack of evidenсe when determining whether the prosecution proved an offense beyond a reasonable doubt. Rogers, 91 F.3d at 56-57.
D. Application
¶ 24 Notwithstanding the noted authorities from other jurisdictions, we conclude that a trial court should inform the jury, as part of the reasonable doubt instruction, that it may consider the lack of evidence in the case. But we also conclude that the omission of the phrase did not impermissibly lower the prosecution‘s burden of proof in this case.
¶ 25 As a starting point, we are aware of no case from our appellate courts suggesting that the phrase “lack of evidence in the case” is confusing, inaccurate, or unhelpful to a jury. To the contrary, we have many appellate opinions supporting the conclusion that a jury may consider the lack of evidence in a case when assessing whether the prosecution has met its burden. See, e.g., People v. Vanrees, 125 P.3d 403, 410 (Colo. 2005) (approving instruction that allowed the jury to consider the evidence or lack of evidence presented in the case); People v. Rubio, 222 P.3d 355, 363 (Colo. App. 2009) (“The instruction that reasonable doubt can arise not just from the evidence but also from ‘lack of evidence’ strengthens rather than undercuts the presumption of innocence.“).
¶ 26 Thus, the committee‘s 2022 elimination of that phrase seems to have been a solution in search of a problem (or an oversight), and it runs contrary to the supreme court‘s repeated cautions against unnecessary modification of the reasonable doubt instruction. See Johnson, ¶ 13; Tibbels, ¶ 25.
¶ 27 Moreover, the committee did not provide any explanation or rationale for its removal of the “lack of evidence” phrase from the 2022 model instruction. Likewise, the committee did not provide any rationale for its decision to add that language back into the 2023 model instruction. In the absence of a contrary explanation, it seems the added sentence is at least an implicit acknowledgment that there was no need to remove the phrase from the 2022 model instruction or benefit associated with the removal.
¶ 28 In any event, the model instruction now includes both “the evidence presented” and “the lack of evidence” within the reasonable doubt instruction. COLJI-Crim. E:03 (2024). As previously noted, substantially similar language had been included in the pre-2022 definition of reasonable doubt for decades, without — to our knowledge — criticism from our appellate courts or legal scholars. Moreover, this language reflects an objective and balanced explanation that the jury may consider both the evidence presented and the lack of evidence when assessing whether a reasonable doubt exists. It also provides a defendant with an express legal foundation to encourage the jury to look at the absence of evidence in considering whether reasonable doubt exists. Thus, going forward (and absent contrary direсtion from the supreme court), we conclude that trial courts should include a statement within the reasonable doubt instruction that the jury may consider the lack of evidence in a case when determining whether the prosecution has met its burden. So while acknowledging that the trial court did not have the benefit of case law directly on point, we conclude that the court should have included the “lack of evidence” language in its reasonable doubt instruction.
¶ 29 But that does not answer the ultimate question before us: Was the trial court‘s omission of this language structural error? To answer this question, we must determine whether the omission lowered the prosecution‘s burden of proof when viewing the court‘s instructions as a whole along with the trial court record:
[I]n considering whether a court‘s statements to a jury regarding the meaning of
“reasonable doubt” (whether in formal instructions or not) unconstitutionally lowered the prosecution‘s burden of proof, an appellate court must ask whether there is a reasonable likelihood that the jury understood the court‘s statements, in the context of the instructions as a whole and the trial record, to allow a conviction based on a standard lower than beyond a reasonable doubt.
¶ 30 The 2022 model instruction did not expressly tell jurors how to evaluate the evidence or lack thereof. Rather, it required them to ensure that there is not “a real possibility that the defendant is nоt guilty” of “the crime beyond a reasonable doubt.” COLJI-Crim. E:03 (2022). That is an accurate statement of the law. Moreover, the instruction emphasized that, to find Melara guilty beyond a reasonable doubt, the jury must be “firmly convinced of [Melara‘s] guilt,” and that even proof that something is “highly probable” is not sufficient to satisfy the reasonable doubt burden. Id.
¶ 31 Furthermore, we agree with the People that the instruction did not forbid or even dissuade the jurors from applying their common sense when considering the evidence in this case, including issues or requirements that the evidence failed to address.
¶ 32 Because we have no basis to believe that the jury acted in an unconstitutional way, and because we conclude that the absence of an express instruction to consider the lack of evidence is not tantamount to a prohibition on doing so, we reject Melara‘s claim that the trial court denied his due process rights or otherwise reversibly erred by failing to include the “lack of evidence” language in its reasonable doubt instruction.
III. Withdrawal of Plea Offer
¶ 33 Melara next contends that the trial court erred by denying his motion to compel the prosecution to leave open for acceptance a plea offer predicated on a community-based sentence after he made a counteroffer. We disagree.
A. Additional Facts
¶ 34 In January 2023, three months before the second trial was set to begin, the prosecutor, a deputy district attorney, emailed Melara‘s counsel to see if Melara was open to accepting a guilty plea. In early February, defense counsel informed the deputy district attorney that she had spoken with Melara “and he does not want to plead. So we are proceeding to trial.” The deputy responded, “To be clear, he is not willing to take anything even if it involved a community-based sentence?” Melara‘s counsel replied, “I did not pose a probation only sentence to him as an option because you said you would not stip[ulate] to that. But if there‘s a probation stip[ulation] that would be a pretty different conversation.”
¶ 35 The same day, the deputy indicated that they were still having internal conversations but asked defense counsel to “let us know if he will consider a community-based sentence.” A week later, Melara‘s counsel responded,
Can you offer [class 5 felony] attempt[ed] [Sexual Assault of a Child] with a stip[ulation] to probation? As you recall the split was heavily in our favor last time. Mr. Melara is older and I think if prison is on the table there‘s no way he‘ll take a deal and would rather push this to trial. But if we can make an agreement to probation there‘s a chance at closing this out.
¶ 36 The prosecution responded with a formal offer on February 15:
We are willing to extend the following community-based offer:
1. Plea to [class 3 felony] Sexual Exploitation
2. Plea to [class 4 felony] 2nd Degree Assault (non-sex) naming both victims
3. Stipulation to 10-years of [Sex Offender Intensive Supervision Probation (SOISP)] for the [class 3 felony] Sexual Exploitation [with] the additional special probation conditions
4. Concurrent probation for the [class 4 felony] 2nd Degree Assault
5. [Sex Offender] Registration, MROG [Mandatory Restraining Order Granted], no contact with anyone under 18. . . .
Defense counsel responded, “Received, but I can all but guarantee he‘s not going to take this.” In response to the prosecutor‘s request
It is community based, but it‘s also a[] [class 3 felony] and [a class 4 felony]. I just think based on prior conversations he‘s unlikely to take it based on the level of offenses, plus 10 year SOISP. He is super reticent to accept any plea that is a sex offense plea, and I think all of that will play into his decision. I will convey.
¶ 37 In a series of subsequent emails, counsel engaged in the following exchange:
[Prosecutor]: To be clear, are you saying that he will accept 5-years of SOISP with pleas to the [class 3 felony] and [class 4 felony]?
[Defense Counsel]: Well, I guess I can‘t say for sure at this point. He maintains his innocence, so anything that can be done to sweeten the deal makes it more likely.
He does not like the sexual exploitation charge at all. I would counter with [a class 4 felony] contributing to the delinquency of a minor — as a sex offense and 2nd degree assault/ stip[ulation] 5 years SOISP.
[Prosecutor]: If that is a firm counte[r]-offer that he will accept, we will discuss it with the victims.
[Defense Counsel]: Ok, I can let you know Monday if it is something he will definitely take.
¶ 38 Four days later, Melara‘s counsel replied, “This is something [Melara] will take. So if you‘re able to offer contributing to the delinquency of a minor (sexual factual basis) [and] 2nd degree assault [and] stip[ulate to] 5 years SOISP, then he‘ll take it.” Two days later, the prosecutor responded, “We are not willing to accept this counteroffer. We plan to proceed to trial on March 13th.”
¶ 39 In subsequent emails to the deputy district attorney and her supervisors, Melara‘s counsel expressed frustration that the prosecution withdrew the February 15 offer. She argued that Melara was proceeding on the assumption that the negotiations were ongoing and that the February 15 offer had never been formally withdrawn and therefore the prosecution had an obligation to leave the offer open while the negotiations continued.
¶ 40 Despite these concerns, before the second trial began, Melara‘s counsel did not request entry of an order specifically enforcing the February 15 offer or otherwise addressing the plea negotiations. Only after the trial was completed, guilty verdicts were returned, and a sentencing hearing was set did Melara‘s counsel file a motion to specifically enforce the February 15 offer. The trial court summarily denied the motion.
B. Standard of Review and Applicable Law
¶ 41 We draw on contract principles to determine whеther a plea offer is enforceable through the remedy of specific performance. Keller v. People, 29 P.3d 290, 295 (Colo. 2000) (“[P]lea agreements are contractual in nature and should be interpreted in accordance with contract principles.“). Like a contract, the terms of a plea agreement are “interpreted in light of the reasonable expectations of the parties. In making such a determination, the court may consider the form and content of any written agreement containing the government‘s promise . . . .” People v. Mershon, 844 P.2d 1240, 1243 (Colo. App. 1992), aff‘d in part and rev‘d in part on other grounds, 874 P.2d 1025 (Colo. 1994), abrogated on other grounds by Melton v. People, 2019 CO 89.
¶ 42 “A defendant is entitled to specific performance of a plea agreement when ‘no other remedy is appropriate to effectuate the accused‘s legitimate expectation engendered by the governmental promise.‘” People v. Macrander, 756 P.2d 356, 361 (Colo. 1988) (quoting People v. Fisher, 657 P.2d 922, 931 (Colo. 1983)). In assessing the terms of a plea offer, our task is to construe the offer consistently with the parties’ intent and a defendant‘s right to be treated fairly by the government. Mershon, 844 P.2d at 1243 (“A promise by the government cannot be withdrawn if the defendant has reasonably and detrimentally relied thereon.“).
¶ 43 Like any contract, a plea offer must be accepted within a reasonable time, unless a specific time for acceptance is delineated in the offer. See Scoular Co. v. Denney, 151 P.3d 615, 618 (Colo. App. 2006). But if a defendant rejects an offer, he has no right to insist on its performance. Likewise, if a defendant makes a counteroffer, the prior offer is typically deemed rejected as a
C. Application
¶ 44 For several reasons, we reject Melara‘s contention that the trial court erred by denying his motion to specifically enforce the alleged contract.
¶ 45 First, Melara rejected the February 15 offer. The email communications between counsel indicate that Melara was unwilling to accept the People‘s offer to plead guilty to a class 3 felony and a class 4 felony in exchange for a community-based sentence. In immediate response to that offer, Melara‘s counsel indicated that “I can all but guarantee he‘s not going to take this.” She later followed up with the prosecutor, stating, “He does not like the sexual exploitation charge at all. I would counter with [class 4 felony] contributing to the delinquency of a minor — as a sex offense and 2nd degree assault/ stip[ulation to] 5 years SOISP.”
¶ 46 The first sentence of this email, coupled with the preceding communications from Melara‘s counsel, confirmed Melara‘s rejection of the prosecutor‘s offer. The second sentence drove the point home by offering to plead to one class 4 felony and a misdemeanor (rather than to separate class 3 and class 4 felonies) with a stipulated sentence of five years SOISP (as opposed to ten years SOISP). Both аs a matter of law and as an interpretation of the parties’ expressed intent, we conclude that the submission of this counteroffer — which decreased the high level of offense and materially varied the proposed sentence term — amounted to a rejection of the February 15 offer.
¶ 47 We also disagree with Melara‘s contention that it would be fundamentally unfair to treat the February 15 offer as rejected because it is allegedly common practice for defendants to attempt to negotiate a better deal without rejecting an existing offer. To begin, this contention is without record support, and beyond that, Melara points to no legal authority that reflects such a rule. Moreover, Melara‘s reasoning would be fundamentally unfair to the prosecution. In effect, the prosecution would be forced to negotiate against itself if a defendant were allowed to make counteroffers while still being able to fall back on accepting the previous unaccepted offer if the prosecution rejected the counteroffer.
¶ 48 The faults in Melara‘s argument are illustrated by the facts of this case. Not only did Melara fail to accept the February 15 offer, but he also failed to ask the court to specifically enforce it. He then proceeded to trial. Only after he was convicted did he request specific performance frоm the court. One can imagine that Melara would not have insisted on specific performance if he had prevailed at trial, or that he would have acquiesced in the enforcement of the February 15 offer had the jury returned not guilty verdicts. In effect, Melara demands a unilateral right to enforce the February 15 offer after the results of the trial became known. It strains credulity to suggest that the court should afford a defendant the right to submit a counteroffer rejecting the prosecution‘s offer, proceed to trial, and then enforce the prosecution‘s last offer after the jury returns guilty verdicts.
¶ 49 For these reasons, we conclude that the trial court did not err by denying Melara‘s motion for specific performance of the February 15 offer.
IV. Expert Testimony
¶ 50 Melara argues that the trial court erred by limiting the scope of his expert witness‘s testimony in a manner that excluded some opinions that the same expert was allowed to offer in connection with the first trial. The People counter that the trial court properly limited the expert‘s testimony to align with the scope of the endorsement at the second trial, which was more restricted than the endorsement made in advance of the first trial, and correctly excluded improper opinion testimony at the second trial.
A. Additional Facts
¶ 51 In both the first and second trials, the defense endorsed Dr. William O‘Donohue to testify. In advance of the first trial, a motions hearing wаs held, at which O‘Donohue was offered and qualified as an expert in “child abuse, child sexual assault, forensic interviewing, evaluation of forensic interviews,
¶ 52 At the second trial, the defense offered O‘Donohue as an expert in child sexual abuse generally, and more specifically, “how children remember abuse, common psychological reactions to abuse, behavior of child sexual abusers, and forensic interviewing with children who may have been sexually abused.” The prosecutor objected on the grounds that the proffer was beyond the endorsed areas of expertise in Melara‘s disclosure.
¶ 53 The court sustained the objection and accepted O‘Donohue as an expert only in the areas of child psychology and forensic interviewing. Subsequently, Melara‘s counsel asked O‘Donohue about pathways to false disclosure by children. The prosecutor objected and the court sustained the objection as beyond the scope of expertise for which O‘Donohue had been accepted as an expert, and on the grounds that the testimony was unduly prejudicial, and that such prejudice substantially outweighed the probаtive value of the testimony.
¶ 54 The court allowed O‘Donohue to testify concerning appropriate forensic interview practices and child sexual assault disclosures. He also testified specifically about the quality of the interviews investigators conducted with granddaughter, including how to judge the credibility of a child‘s disclosures. But the court did not permit O‘Donohue to testify concerning potential pathways to false disclosures by children, such as tensions in the family. The court also precluded O‘Donohue from testifying about the possibility of gradual disclosures by children who may be coached and typical child victim behavior after abuse.
¶ 55 On appeal, Melara argues that the trial court unduly focused on the specific content of the
¶ 56 The People respond by arguing that the trial court properly executed its role as a gatekeeper for the expert testimony by preventing O‘Donohue from improperly undermining granddaughter‘s testimony and addressing irrelevant matters.
B. Standard of Review and Applicable Law
¶ 57 We review a trial court‘s rulings on the admissibility of expert testimony for an abuse of discretion. People v. Collins, 2021 COA 18, ¶ 60.
¶ 58 The parties disagree on the appropriate standard of reversal. Melara argues that his presentation of O‘Donohue‘s testimony was so restricted that he was deprived of his constitutional right to defend against the charges. See, e.g., People v. Pronovost, 773 P.2d 555, 558 (Colo. 1989). Thus, he argues, we should apply a constitutional harmless error standard. See Golob v. People, 180 P.3d 1006, 1013 (Colo. 2008) (“Because a criminal defendant has the right to call witnesses in his defense, abridgment of that right is subject to a constitutional harmless error analysis.“). Under this standard, “we reverse if ‘there is a reasonable possibility that the [error] might have contributed to the conviction.‘” Hagos v. People, 2012 CO 63, ¶ 11 (quoting Chapman v. California, 386 U.S. 18, 24 (1967)).
¶ 59 The People counter that the court‘s rulings did not give rise to a constitutional violation, and therefore we should apply nonconstitutional harmless error. We reverse for nonconstitutional harmless error only “if the error ‘substantially influenced the verdict or affected the fairness of the trial proceedings.‘” Id. at ¶ 12 (quoting Tevlin v. People, 715 P.2d 338, 342 (Colo. 1986)).
C. Application
¶ 60 Initially, we reject Melara‘s argument that the trial court‘s rulings concerning the scope of O‘Donohue‘s permitted testimony at the first trial (or the preceding motions hearing) became the law of the case that was binding at the subsequent trial.
¶ 61 Second, we note that there are significant distinctions between the facts in the cases cited by Melara and the facts of this case. For example, Melara relies heavily on the supreme court‘s decision in Pronovost, 773 P.2d 555. But Pronovost is factually and legally distinguishable because the trial court there made a pretrial ruling completely prohibiting the defendant from calling an expert who was not endorsed until the Friday before a Monday trial. Id. at 556-57. Similarly, in People v. Melendez, 102 P.3d 315, 322 (Colo. 2004), the supreme court reversed a trial court order that precluded the defendant from calling a fact witness who violated a sequestration order.
¶ 62 Here, in contrast, the trial court did not wholly preclude O‘Donohue‘s testimony. Rather, the court limited or precluded certain lines of questioning. Even then, O‘Donohue was provided the opportunity to opine on the characteristics of an appropriate versus a coached or suggestive interview, which allowed the jurors the benefit of his expertise in evaluating the trustworthiness of granddaughter‘s disclosures to a forensic interviewer. We now turn to Melara‘s remaining arguments regarding O‘Donohue.
1. Scope of Disclosures
¶ 63 We agree with Melara that the trial court erred by resting its analysis on the incomplete nature of O‘Donohuе‘s endorsement.
Subject to constitutional limitations, and where the interests of justice would be served, the court may order the defense to disclose the underlying facts or data supporting the opinion in that particular case of an expert endorsed as a witness. If a report has not been prepared by that expert to aid in compliance with other discovery obligations of this rule, the court may order the party calling that expert to provide a written summary of the testimony describing the witness‘s opinions and the bases and reasons therefor . . . . The intent of this section is to allow the prosecution sufficient meaningful information to conduct effective cross-examination . . . .
¶ 64 During the first trial, the prosecutor heard many of the opinions that O‘Donohue was not allowed to provide at the second trial. Moreover, Melara provided the prosecution with a lengthy report that set forth at least some of O‘Donohue‘s anticipated opinions.
¶ 65 Under these circumstances, we conclude that the trial court erred by limiting O‘Donohue‘s testimony to “child psychology and forensic interviewing” without first conducting a рrejudice analysis. Cf. Pronovost, 773 P.2d at 558 (trial court erred by excluding, without first conducting a balancing test, testimony from defendant‘s expert based on the failure to timely endorse the expert).
¶ 66 But the perceived
2. Pathways to False Reports
¶ 67 Melara argues that “[t]he most important area about which O‘Donohue was prohibited from testifying was regarding pathways and factors that can lead to false reports.” But during his direct examination by Melara‘s counsel, O‘Donohue opined, “Most allegations of sexual abuse are true,
[Defense Counsel]: Dr. O‘Donohue, you had mentioned pathways to false disclosure. Can you talk to me a little bit about what that means?
[O‘Donohue]: Yes. It means what sort of variables can cause a child to make a false report.
[Defense Counsel]: And what types of variables do you see?
¶ 68 At this point the prosecutor objected and the court held a bench conferеnce. After hearing from both sides, the court ruled that the testimony was veering into specific circumstances that could be construed as commenting on granddaughter‘s credibility. See Venalonzo v. People, 2017 CO 9, ¶ 44 (“[The defendant] argues that the trial court erred in permitting the investigating police officer to testify that, in his experience as a school teacher, children only make up trivial stories, not serious accusations. Normally, this statement would constitute improper testimony that the children were telling the truth.“) (footnote omitted).
¶ 69 The trial court noted that it was proper for an expert to opine generally on the possibility that children sometimes fabricate reports of abuse. But the court also noted that both O‘Donohue and the People‘s expert had already opined that children sometimes falsely report abuse allegations. Getting into the specific circumstances that lead to such false reports, the court reasoned, was improper and prejudicial to the People. See
¶ 70 We acknowledge that the boundaries between general testimony about false reports and improper undermining by drawing direct lines between specific circumstances and false reports can sometimes be difficult. That boundary blurs when an expert purports to act as both a general expert addressing broad principles and also as a case-specific expert opining on the application of those general principles to the specific facts of a case. That is the dual role O‘Donohue played in this case, which created the concerns noted by the division in People v. Battigalli-Ansell, 2021 COA 52M.
The line between opinion testimony that improperly bolsters a witness‘s credibility and admissible testimony that may only collaterally enhance the witness‘s credibility is sometimes a difficult one to draw. See People v. Fortson, 2018 COA 46M, ¶¶ 105, 107, 114, 116, 421 P.3d 1236, 1250-51 (Berger, J., specially concurring). But when an “expert assumes the role of not only educating the jury on general [witness] characteristics but also opines that the particular [witness‘s] conduct is in conformity with those characteristics, the expert probably crosses the line.” Id. at ¶ 114, 421 P.3d at 1251.
¶ 71 We agree that reasonable judges can differ on whether expert testimony has crossed that line. But that is why we entrust these judgment calls to the trial court‘s sound discretion; that court is optimally positioned to make those decisions based on its personal observations of the evidence in the case. Given this deference, we cannot say the trial court erred by excluding this evidence under
3. Stages of Disclosure
¶ 72 Next, Melara objects to the trial court‘s exclusion of O‘Donohue‘s anticipated testimony about “stages of disclosure.” But the trial court did not prevent O‘Donohue from offering all opinions on this topic. In fact, on cross-examination the prosecutor examined O‘Donohue on stages of disclosure. During his redirect examination, Melara‘s counsel thoroughly examined O‘Donohue on his views of stages of disclosure, including the following testimony:
[O‘Donohue]: [A] forensic interview, all the protocols that I know, are designed to be a single interview, not multiple interviews, to
track some alleged process of disclosure — stage of disclosures. So all the protocols are are designed to be a single interview which captures the child‘s complete statement. And the reason why they design it to be one interview, not four interviews, is because experience in the field generally is they can get a full statement in the first interview. [Defense Counsel]: And so if there‘s new or additional disclosures later down the line, can that be in part because there wasn‘t a complete forensic interview?
[O‘Donohue]: Yes.
¶ 73 The prosecutor objected after Melara‘s counsel next asked, “Are those new or additional disclosures at greater risk for outside contamination?” The court sustained the objection “based on the facts presented in this case.” Melara‘s counsel did not make a further record or ask for an explanation of the court‘s ruling.
¶ 74 Melara‘s argumеnt is based only on the contention that his attorney was not allowed to question about “outside contamination.” But during his direct examination, O‘Donohue spoke extensively about outside contamination, including the following:
[O]utside contamination is the idea that prior to the forensic interview, the child has spoken to other adults about these alleged events. And it is possible that in those conversations, leading questions, suggestive questioning could have occurred that led to false memories.
¶ 75 Moreover, Melara does not further develop his claim about how the risk of contamination with additional disclosures related to the facts of this case, other than a conclusory suggestion that daughter coached or otherwise colored granddaughter‘s testimony.
¶ 76 Expert testimony was not necessary to enable the jury to consider the possibility that daughter — who also accused Melara of sexual assault — could have influenced granddaughter‘s testimony. See People v. Cooper, 2021 CO 69, ¶ 48 (The relevance of expert testimony depends on “a common sense inquiry: whether an untrained lay person would be qualified to determine a particular issue ‘intelligently and to the best possible degree without enlightenment from those having a specialized understanding of the subject involved in the dispute.‘” (quoting Lanari v. People, 827 P.2d 495, 502 (Colo. 1992))); accord
4. Victim‘s Behavior
¶ 77 Lastly, Melara challenges the trial court‘s exclusion of O‘Donohue‘s anticipated testimony about how children who are victims of sexual assault often behave. Defense counsel sought to elicit this testimony during the following exchange:
[Defense Counsel]: Psychologically, how would you expect the child to respond to the pain of sexual penetration or sexual intrusion?
[O‘Donohue]: They‘re very disturbed by it, that they don‘t like it, that they want it to stop, that they don‘t want it to occur again, that it‘s scary to them . . . . So it‘s quite disturbing to a child.
[Defense Counsel]: Is that something that even a child at around the age of four or five can describe?
[O‘Donohue]: In my experience, yes.
[Defense Counsel]: Do you see that pain affect[s] how they behave with other people as well?
[Prosecutor]: Objection. Relevance.
[The Court]: Sustained.
. . . .
[Defense Counsel]: In your experience, does the child typically fear the person who causes them pain?
[Prosecutor]: Objection. Relevance.
[The Court]: Overruled. You may answer that question.
[O‘Donohue]: Yes.
[Defense Counsel]: How does the fear of that person affect the child‘s behavior?
[Defense Counsel]: Will that also affect how the child might interact with other people?
[Prosecutor]: Objection. Relevance.
[The Court]: Sustained. Again, we‘re back to the forensic interviewing and/or child psychology only. And without a specific piece of information, it would be not relevant to the facts of this case.
¶ 78 Given the extensive testimony preceding this question, we do not perceive any abuse of discretion in the trial court‘s ruling to limit speculative testimony about how this experience may have impacted granddaughter‘s interactions with people other than Melara.
¶ 79 Therefore, we conclude that the trial court did not reversibly err by limiting O‘Donohue‘s testimony.
V. Trial Severance
¶ 80 As discussed above, at the first trial, Melara‘s counsel moved to sever the charges involving daughter from the charges involving granddaughter. The court denied that motion and the charges were tried together. Melara concedes that his attorney did not file a motion to sever after the first trial ended in a mistrial.
¶ 81 On appeal, Melara contends that at the second trial — which was presided over by a different judge than the first — the court erred by not severing the charges involving the different victims. The People argue that because the motion to sever was not renewed before the second trial, the issue was either waived or forfeited. Given the procedural context of this case, we conclude that, even if Melаra did not waive review of the severance issue at the second trial, the trial court did not err by failing to sua sponte sever the charges.
¶ 82 Waiver is “the intentional relinquishment of a known right or privilege.” People v. Rediger, 2018 CO 32, ¶ 39 (quoting Dep‘t of Health v. Donahue, 690 P.2d 243, 247 (Colo. 1984)). In contrast, forfeiture is the failure to timely assert a right. Forgette v. People, 2023 CO 4, ¶ 29. “The distinction between a waiver and a forfeiture is significant because ‘a waiver extinguishes error, and therefore appellate review, but a forfeiture does not.‘” Id. at ¶ 30 (quoting Rediger, ¶ 40).
¶ 83 Generally, objections made during a case that resulted in a mistrial do not preserve objections at a subsequent retrial. See, e.g., United States v. Palmer, 122 F.3d 215, 221 (5th Cir. 1997) (denying a motion to sever as unpreserved when the motion was made before the first trial, but not renewed prior to the second trial, because “objections made at [an] aborted trial have no bearing on the retrial, as the two are entirely separate affairs“). This general rule is particularly apt when, as here, the second trial is held before a different judge than the one who presided at the first trial. See United States v. Mann, 590 F.2d 361, 371 (1st Cir. 1978) (“[A] rule that one objection preserves an issue during any retrials would make the domain of a trial judge more of a mine field than ever.“).
¶ 84 Moreover, the absence of a renewed objection is particularly problematic in this case, given that obvious strategic factors could have impacted the decision whether to ask for a severance before the second trial. At the first trial, Melara avoided а conviction. Indeed, on one charge the jury was deadlocked ten to two in favor of acquittal but deadlocked eight to four in favor of conviction on the other charge. Viewing this outcome, Melara may well have made a strategic decision that he would benefit from trying the charges together so that the weakness of the evidence supporting one charge would adversely affect the jury‘s perception of the other.
¶ 85 On the other hand, the supreme court has made clear that a defendant who files a pretrial motion to sever charges need not renew that motion at trial to preserve the severance issue for appeal. See Bondsteel v. People, 2019 CO 26, ¶ 2. There is, however, a significant distinction between the procedural setting of Melara‘s case and the setting of the Bondsteel case. Bondsteel
¶ 86 Given the dynamics of the hung jury in the first case, there is certainly a reasonable inference that Melara made a strategic decision not to renew a motion to sever. Such strategic decisions may, in appropriate circumstances, give rise to a conclusion that the claimed error is waived. See Richardson v. People, 2020 CO 46, ¶ 26 n.2 (finding waiver based, in part, on fact that “counsel could have had sound strategic reasons” for decision not to exercise a peremptory challenge against a juror). But the supreme court has also counseled that we must indulge every reasonable presumption against waiver. Rediger, ¶ 39.
¶ 87 Balancing these considerations, we conclude that Melara forfeited, rather than waived, his argument that the trial court erred by not severing the charges involving the two victims at the second trial. See id. at ¶ 40 (forfeiture is the failure to timely assert a known right). But because severance was not mandated, and there were rational explanations why Melara may not have wanted the charges at the second trial severed, we perceive no error, much less plain error, associated with the trial court‘s decision not to sua sponte order severance of the charges at the second trial. See Hagos, ¶ 14 (an error is plain only if it is obvious and substantial); Vialpando, 954 P.2d at 624 (The trial court did not err by declining to order severance of charges in a second trial because “the original reason for severing the charges was no longer present, and defendant ha[d] not demonstrated prejudice as a result of the trial court‘s refusal to sever the escape charge at his second trial.“).
VI. Child Hearsay
¶ 88 Melara raises two claims of error related to the admission of multiple instances of child hearsay during the trial. First, he argues that the trial court reversibly erred by admitting granddaughter‘s initial disclosure and subsequent conversations with daughter in the fall of 2020, as well as the November 2020 video of granddaughter‘s forensic interview. Second, Melara contends that the court reversibly erred by departing from the COLJI child hearsay instruction. We address each contention below.
A. Admission of Granddaughter‘s Out-of-Court Statements
1. Additional Facts
¶ 89 Before the second trial, the prosecution filed a notice of its intent to introduce child hearsay statements. Shortly thereafter, Melara‘s attorney filed a motion challenging granddaughter‘s competency to testify under
¶ 90 The trial court set the prosecution‘s notice for a pretrial hearing to consider whether granddaughter would be permitted to testify in accordance with the requirements of
¶ 91 The court heard testimony at the hearing from daughter, granddaughter, the forensic interviewer, and O‘Donohue. The trial court found that granddaughter “described in age-appropriate terms what it means to tell the truth and lie and that she understood the consequences of lying.” The court also noted that she described Melara‘s house in a reasonable way for a seven-year-old and that despite being “a little bit squirmy,” she was reasonably focused.
- daughter testified credibly about granddaughter‘s September 2020 statements;
- even though granddaughter‘s statements about the abuse were not spontaneous and were prompted by daughter, the questions were not suggestive because daughter routinely asked the children whether they had been inappropriately touched;
- daughter was not trying to elicit a specific response from granddaughter when granddaughter made the statements;
- granddaughter did not make the statements when she was in pain, and she used language that would have been used by a child around her age;
- the allegation against Melara was not made in response to a leading question; and
- granddaughter had no bias against Melara or motive to lie.
¶ 93 With respect to granddaughter‘s November 2020 statements during the forensic interview, the trial court made similar findings about the content of granddaughter‘s statements and her demeanor during the interview.
2. Standard of Review and Applicable Law
¶ 94 A trial court‘s decision to admit a child‘s hearsay statements is reviewed for an abuse of discretion. People v. Rojas, 181 P.3d 1216, 1219 (Colo. App. 2008).
¶ 95 Hearsay statements are out-of-court statements, offered at trial, to prove the truth of the matter asserted. CRE 801(c); People v. Phillips, 2012 COA 176, ¶ 61. Such statements are inadmissible in a criminal prosecution unless an exception applies. CRE 802.
¶ 96 The admissibility of child hearsay depends on the assessment of the statement‘s reliability made after a pretrial hearing.
¶ 97 In District Court, 776 P.2d at 1089-90, the supreme court articulated eight factors that courts may consider in evaluating the reliability of a child‘s hearsay statements:
- whether the statement was made spontaneously;
- whether the statement was made while the child was still upset or in pain from the alleged abuse;
- whether the language of the statement was likely to have been used by a child the age of the declarant;
- whether the allegation was made in response to a leading question;
- whether either the child or the hearsay witness had any bias against the defendant or any motive for lying;
- whether any other event occurred between the time of the abuse and the time of the statement that could account for the contents of the statement;
- whether more than one person heard the statement; and
- the general character of the child.
3. Application
¶ 98 Melara argues that he was severely prejudiced by the admission of daughter‘s hearsay, particularly given the limitations the trial court placed on the сross-examination of O‘Donohue about pathways and factors that can lead to false reporting.
¶ 99 The People respond that the court made proper findings regarding granddaughter‘s competence and the reliability of the statements, and therefore the court properly exercised its discretion by admitting the statements. We agree with the People.
¶ 100 The record supports the trial court‘s finding that the hearsay statements were sufficiently reliable to be admitted.
¶ 101 The court also appropriately considered whether granddaughter understood the difference between the truth and a lie. When asked, granddaughter was forthcoming when she did not understand a question. She was candid about not remembering specific details about Melara‘s house or the 2020 interview. Granddaughter also admitted to lying in the past, and that she understood the importance of telling the truth in court.
¶ 102 Furthermore, because granddaughter and daughter were available to testify and were cross-examined, Melara‘s counsel could — and did — raise at trial these concerns about the credibility of the hearsay statements.
¶ 103 Despite the trial court‘s detailed findings, Melara argues that the court did not adequately defer to O‘Donohue‘s expert opinions concerning the alleged defects in the methodology used by the interviewer. Specifically, O‘Donohue testified that the interviewer engaged in repetitive and leading questions and did not eliminate the possibility that threats or bribes may have colored granddaughter‘s testimony. It is true that the interviewer asked targeted and focused questions, but the questions were not leading because they did not suggest a particular answer. And the interviewer allowed granddaughter to fully answer, asked her clarifying questions, and did not coerce her into providing a particular response.
¶ 104 The trial court was free to give O‘Donohue‘s opinions the weight it deemed appropriate in light of all the evidence. See People v. Rivera, 56 P.3d 1155, 1164 (Colo. App. 2002) (The fact finder “retains its authority to determine the facts from the evidence and accept or reject” expert opinion testimony.). After considering O‘Donohue‘s testimony, the trial court found that the content and circumstances of granddaughter‘s statements to the interviewer provided “sufficient safeguards of reliability” and that the interviewer did not engage in “leading, suggestive or otherwise inappropriate questioning.”
¶ 105 These findings are supported by the record. We therefore reject Melara‘s argument that the trial court abused its discretion by admitting granddaughter‘s out-of-court statements.
B. The Jury Instruction
¶ 106
during the proceeding the jury heard evidence repeating a child‘s out-of-court statement and that it is for the jury to determine the weight and credit to be given the statement and that, in making the determination, the jury shall consider the age and maturity of the child, the nature of the statement, the circumstances under which the statement was made, and any other relevant factor.
(Emphasis added.)
¶ 107 The court instructed the jury as follows:
In this case you heard evidence of out of court statements made by granddaughter. . . . [I]t is for you to determine the weight and credit to be given any such statements. In making this determination you should consider the age and maturity of the child, the nature of the statements, the circumstances under which the statements were made, and any other evidence that has been admitted that you choose to consider for this purpose.
¶ 108 Melara contends that the instruction contained three errors because it (1) omitted the word “repeating“; (2) omitted the word “allegedly“; and (3) allowed the jury to consider “any other evidence that has been admitted” rather than “any other relevant factor.”
1. Standard of Review
¶ 109 As discussed, supra Part II.B, we review de novo a trial court‘s jury instructions, as a whole, to determine whether they properly instructed the jury on the controlling law. Riley, 266 P.3d at 1092. If the trial court‘s instructions accurately described the applicable law, we generally review its decision to give a particular instruction for an abuse of discretion. Paglione, ¶ 45.
¶ 110 Melara‘s first contends that the instruction did not follow
¶ 111 As best we understand Melara‘s second contention, he argues that the word “allegedly” should have been included in the instruction, as suggested by the COLJI instruction when the hearsay statement has not been otherwise admitted via video recording. COLJI-Crim. D:12 cmt. 4 (2022). But the jury viewed the video recording of granddaughter‘s interview. We acknowledge that the jury also received evidence of granddaughter‘s parallel hearsay statements to daughter, but Melara does not explain, and we do not perceive, how he was prejudiced by the omission of the word “allegedly” with respect to the similar statements made to daughter.
¶ 112 Finally, Melara argues that the court erred by instructing the jury that it may consider “any other evidence that has been admitted” rather than “any other relevant factor.” He argues this allowed the jury to consider irrelevant factors about granddaughter‘s credibility, such as Melara‘s alleged conduct toward daughter. But, as Melara recognizes, the jury was instructed to consider the charges against each victim separately, uninfluenced by its verdict on other charges. We presume the jury followed that instruction. People v. Snelling, 2022 COA 116M, ¶ 42 (“We presume that juries follow a court‘s instructions absent evidence to the contrary.“). Moreover, presumably the evidence admitted by the trial court was relevant to the issues presented in this case. Therefore, by limiting the jury‘s consideration to other admissible evidence, the court limited the jury‘s consideration to relevant evidence.
¶ 113 The trial court did not abuse its discretion by giving the child hearsay instruction.
VII. Cumulative Error
¶ 114 Finally, Melara contends that the cumulative effect of the asserted errors warrants reversal. “For reversal to occur based on cumulative error, a reviewing court must identify multiple errors that collectively prejudice the substantial rights of the defendant, even if any single error does not.” Howard-Walker v. People, 2019 CO 69, ¶ 25. Even though we have concluded that the trial court should have included the “lack of evidence” language in the reasonable doubt instruction and erred by relying on the incomplete nature of O‘Donohue‘s endorsement to limit the scope of his testimony, the omission and error did not have a synergistic or compounding effect and did not substantially prejudice Melara. See People v. Serna-Lopez, 2023 COA 21, ¶ 47 (“The doctrine of cumulative error is based on the notion that multiple errors, in isolation, may be viewed as harmless, but the synergistic effect of the multiple errors may be so prejudicial that they deprive a defendant of a fair trial.“). Thus, we reject his cumulative error claim.
VIII. Disposition
¶ 115 The judgment of conviction is affirmed.
JUDGE KUHN concurs.
JUDGE WELLING specially concurs.
JUDGE WELLING, specially concurring.
¶ 116 I am fully on board with the disposition and rationale of the majority. I write separately because I would go one small step further with respect to the reasonable doubt instruction: I would conclude that the denial of Melara‘s request to include the “or lack of evidence” language in the reasonable doubt instruction was error, albeit harmless error.
¶ 117 I agree with the principles guiding our review as set forth by the majority. That is, “[w]e review jury instructions de novo to determine whether they accurately inform the jury of the governing law,” Hoggard v. People, 2020 CO 54, ¶ 12, and “[a]s long as the instruction properly informs the jury of the law, a trial court has broad
¶ 118 To be sure, abuse of discretion is a highly deferential standard. Id. (“A trial court‘s ruling on jury instructions is an abuse of discretion only when the ruling is manifestly arbitrary, unreasonable, or unfair.” (quoting Day, 255 P.3d at 1067)). The touchstone of the abuse of discretion standard is “whether the trial court‘s decision fell within a range of reasonable options.” People v. Salazar, 2012 CO 20, ¶ 32 (Bender, C.J., dissenting) (quoting E-470 Pub. Highway Auth. v. Revenig, 140 P.3d 227, 230-31 (Colo. App. 2006)). Accordingly, our inquiry isn‘t whether we agree with the trial court‘s decision, id. (citing Streu v. City of Colorado Springs, 239 P.3d 1264, 1268 (Colo. 2010)); instead, “our role is simply to review the trial court‘s decision to ensure that it did not ‘exceed[] the bounds of the rationally available choices,‘” id. (quoting Streu, 239 P.3d at 1268).
¶ 119 Here, I would conclude that the trial court‘s denial of Melara‘s request to include the “or lack of evidence” language, though understandable based on the then-recent change to the model instruction, nevertheless lacked a rational basis. I reach this conclusion because I can‘t find a rationale for refusing defense counsel‘s request that survives scrutiny. As the majority points out, there are a host of solid reasons for including the requested language: it‘s an accurate statement of the law, supra ¶ 25; it clarifies a principle that, at times, is difficult for jurors to grasp, supra ¶ 28; People v. Young, 16 P.3d 821, 825 & n.3 (Colo. 2001) (“It is not unusual for a juror to be confused or uncertain about the presumption of innocence because it is a difficult legal concept.“); it has been included in the model reasonable doubt instruction in Colorado for decades, supra ¶ 28; and there is no indication that the requested language has been confusing, misleading, or otherwise unhelpful to jurors, supra ¶¶ 25, 28.
¶ 120 On the other side of the balance, as best I can divine, there are two potential rationales for declining to include the requested language: (1) it had been omitted from the then-recently modified model instruction, and (2) the principle was already adequately covered by other instructions. I am not persuaded that either rationale provides a valid reason for denying the request.
¶ 121 With respect to the first reason, it‘s certainly understandable why the court was inclined to faithfully cleave to the language in the model instruction. See Galvan v. People, 2020 CO 82, ¶ 38 (“The model instructions . . . have been approved in principle by [the supreme] court and serve as beacon lights to guide trial courts.“). But following the model instructions doesn‘t insulate a trial court‘s decision from appellate scrutiny. See, e.g., Garcia v. People, 2019 CO 64, ¶ 22 (following the modеl instruction doesn‘t provide trial courts with “a safe harbor that insulates instructional error from reversal“). This, in my view, is one of those exceptionally rare circumstances where tracking the model instruction alone isn‘t enough to save the court‘s decision even under an abuse of discretion standard. Cf. id. at ¶ 23 (holding that tracking model instruction isn‘t always enough to avoid even plain error); Auman v. People, 109 P.3d 647, 660-61 (Colo. 2005) (concluding that the felony-murder instruction that the court gave was erroneous notwithstanding the fact that it tracked the model instruction in effect at the time). This is particularly so where the change had yet to be tested on appellate review and the omission of the requested language lacked any clear rationale. Indeed, as observed by the majority, although other changes in the reasonable doubt instruction were explained in the comment, there is no explanation for the omission of the requested language. Supra ¶¶ 25-27. Thus, as a standalone rationale, I would conclude that relying on the requested language‘s omission from the model instruction doesn‘t survive scrutiny.
¶ 122 An alternative rationale for declining to include the requested language is that it‘s duplicative of other concepts already adequately covered in the instructions that the court gave. This is a closer call. After all, a trial court “need not give a supplemental
¶ 123 I know this conclusion is in tension with the rationale undergirding the conclusion that the omission of the language doesn‘t constitute structural error. To put a sharper point on it, one might say the requested language is either adequately covered by another instruction so it‘s not an abuse of discretion to omit it or it‘s not adequately covered so its omission is structural error. I‘m not persuaded that this is a binary choice. I think the concept is sufficiently covered by the instructions as a whole that the omission of the language doesn‘t lower the burden of proof, but I still conclude that there is no rational basis to declining a request to include it. I‘d walk that tightrope.
¶ 124 For all of these reasons, I would conclude that the trial court‘s refusal to include “or lack of evidence” language upon request was an abuse of discretion and, therefore, erroneous. This conclusion means that I would also have to consider whether this error was harmless. See McDonald, ¶ 55 (“Where an error exists and ‘a defendant . . . object[s] to an instruction, a harmless error standard applies.‘” (quoting People v. Garcia, 28 P.3d 340, 344 (Colo. 2001))). That is where I turn next.
¶ 125 “Under a harmless error standard, reversal is required unless the error does not affect substantial rights of the defendant.” Id. (quoting Garcia, 28 P.3d at 344). The substantial right implicated by the refusal to include this language upon request is that its omission would lower the prosecution‘s burden of proof. But as noted at the outset of this special concurrence, I agree with the entirety of the majority‘s rationale, including that the omission of the requested language doesn‘t constitute structural error. See supra ¶¶ 24-32. For the same reason the majority here and the division in People v. Schlehuber, 2025 COA 50, ¶¶ 19, 21-25, conclude that the omission of the language didn‘t lower the prosecution‘s burden of proof, I would conclude that the erroneous omission of the requested language was harmless.
¶ 126 Finally, I would note that, as a practical matter, I don‘t think my preferred resolution of the question before us encourages trial courts to do anything different than what the majority and the division in Schlehuber suggest. As I read the opinions, all five of the judges who have signed on to those opinions are sending the same, strong message: It‘s better practice for trial courts to include the “or lack of evidence” language in the reasonable doubt instruction. See supra ¶¶ 24, 28; Schlehuber, ¶¶ 20, 25. And just a few months after the trial in this case, the Model Criminal Jury Instructions Committee sent the same message when it further amended the reasonable doubt instruction to again include the previously omitted language. I would send the same message, just on the added basis that failure to include this language upon a party‘s request is error.