Date: Sep 3, 2026

Harmless-Error Review for Limited 404(b) and CAC-Interview Evidence, Unaffected by Jury Impasse — State v. Giroux (N.H. 2026)

1. Introduction

In State of New Hampshire v. Max Giroux (No. 2024-0505), the New Hampshire Supreme Court affirmed three convictions for aggravated felonious sexual assault (AFSA) following a jury trial. The defendant challenged two evidentiary rulings: (1) the admission of testimony suggesting a prior incident (about a year before the charged assaults) where the defendant was observed “laying down” with the child in bed, which he argued was improper “other acts” evidence under N.H. R. Ev. 404(b); and (2) the admission of testimony that the complainant participated in a Child Advocacy Center (CAC) forensic interview and testimony describing the general CAC interview process, which he argued was irrelevant and unfairly bolstered the complainant’s credibility even though the jury heard no interview content.

The Court resolved the appeal by order under Sup. Ct. R. 20(3) and did not decide whether the trial court erred in admitting either category of evidence. Instead, it held that any assumed error was harmless beyond a reasonable doubt given the totality of the trial record.

2. Summary of the Opinion

The Court assumed (without deciding) that the trial court may have erred by admitting: (a) the “in bed with A.A.” testimony offered as other-acts evidence; and (b) evidence that A.A. participated in a CAC forensic interview and general testimony about how CAC interviews are conducted. Applying the State’s burden to prove harmlessness beyond a reasonable doubt, the Court concluded the verdicts were unaffected because:

The State’s case included overwhelming direct evidence—detailed testimony by A.A. describing each charged act of digital penetration.
Additional corroboration supported the State, including testimony reflecting consciousness of guilt (“maybe I did do something wrong,” and “lay low … until this all blew over”).
A.A.’s emotional disclosures to her best friend and later to her mother provided further context and support.
The challenged evidence comprised only a small portion of the State’s case; the State did not reference the “in bed” testimony in closing, and CAC references were limited and were subject to a curative instruction after an improper argument.
The defendant’s reliance on the jury’s reported impasse was speculative; the Court noted it was unaware of any precedent treating an impasse as a harmless-error factor, and held that even if considered, it did not outweigh the other circumstances.

The convictions were therefore affirmed, and any unbriefed issues were deemed waived.

3. Analysis

3.1 Precedents Cited

The Court’s analysis is primarily a synthesis of its modern harmless-error framework, anchored in State v. Boudreau, and illustrated through comparisons to multiple recent evidentiary-harmlessness decisions.

State v. Boudreau, 176 N.H. 1 (2023)

Boudreau supplies the controlling harmless-error rule: the State must prove beyond a reasonable doubt that the alleged evidentiary error did not affect the verdict. The Court also relies on Boudreau for the “totality of the circumstances” approach and for the non-exhaustive list of factors used to assess harmlessness (strength of the case, cumulative/inconsequential nature, frequency, corroboration, nature of the defense, circumstances of introduction, curative steps, inflammatory nature, and overwhelming evidence).

In Giroux, the Court applies these factors to characterize the disputed evidence as limited in scope, not inflammatory, and overshadowed by direct, detailed testimony and additional corroboration.

State v. Rouleau, 176 N.H. 400 (2024), 2024 N.H. 2

Rouleau is used to emphasize that, in harmless-error analysis, a case grounded in direct victim testimony (rather than circumstantial proof) can support a finding that the impact of an evidentiary mistake is minimal when compared to the victim’s account. The Court also adopts Rouleau’s reasoning that harmlessness is more likely when disputed evidence is a small portion of testimony and is not emphasized in closing argument.

State v. Lemieux, 136 N.H. 329 (1992)

Lemieux supports the idea that even credibility-bolstering evidence may be harmless when the victim provides a detailed descriptive account of the assaults. In Giroux, the Court analogizes A.A.’s sensory and contextual details (physical sensations, defendant’s conduct and statements, sequence of events) to the kind of compelling direct testimony that can render a challenged evidentiary item “inconsequential” by comparison.

State v. Thibedau, 142 N.H. 325 (1997)

Thibedau is cited for two points: (1) the significance of a victim’s detailed account in evaluating harmlessness, and (2) the role of defendant statements evidencing consciousness of guilt. The Court treats the defendant’s reaction to learning of the accusation (“maybe I did do something wrong” and advice to “lay low”) as meaningful corroboration that further diminishes any marginal effect of the challenged evidence.

State v. Sleeper, 177 N.H. 665 (2025), 2025 N.H. 52

Sleeper is used to frame the “inflammatory nature” factor. The Court reasons by comparison: if highly graphic or emotionally charged evidence (e.g., admissions about a killing) can dwarf disputed evidence, then here the challenged items—brief other-acts context and the fact of a CAC interview—were not inflammatory relative to A.A.’s own detailed testimony of sexual penetration and the surrounding circumstances.

State v. Zuzelo, 176 N.H. 499 (2024), 2024 N.H. 14

Zuzelo supplies a contrast case: harmlessness is less likely when disputed evidence is repeated through multiple witnesses and repeatedly referenced in closing to support credibility and guilt. Giroux distinguishes itself: one witness provided the other-acts testimony, it was not used in closing, and CAC references were few and partly neutralized by a curative instruction.

State v. Wells, 166 N.H. 73 (2014)

Wells is invoked for the presumption that juries follow curative instructions. This mattered because the prosecutor argued that A.A. was “consistent with what she told the CAC,” prompting an objection and an instruction that the jury had “heard no facts in evidence” regarding CAC statements and must disregard the remark. The Court treated that instruction as effective in preventing the jury from using nonexistent CAC “content” as proof.

State v. Anctil, 134 N.H. 623 (1991)

Anctil informs the Court’s handling of delayed disclosure and initial denial. The defense asked the jury to credit A.A.’s denial at age seven and discredit her later disclosure as a teenager. Relying on Anctil, the Court notes that evidence explaining delay (fear of consequences, concern for a sibling, disclosure to a trusted confidant) can blunt the force of delay-based attacks—and that the “most damning” evidence typically remains the victim’s descriptive account of the abuse, not the reporting timeline.

State v. Blackmer, 149 N.H. 47 (2003)

Blackmer is cited for appellate practice: issues raised in the notice of appeal but not briefed are waived. The Court used it to dispose of any unbriefed claims.

3.2 Legal Reasoning

The Court’s reasoning proceeds in three deliberate steps.

Assume error, then test prejudice rigorously. The Court bypassed the contested admissibility questions under Rule 404(b) and relevance/unfair prejudice principles as to CAC participation, and moved directly to whether any error affected the verdicts.
Measure the challenged evidence against the trial’s evidentiary “center of gravity.” The Court identified the core proof as A.A.’s direct, detailed testimony establishing each element of AFSA—three separate acts of sexual penetration (digital intrusion) when she was under 13. It then emphasized corroborative proof: (a) the defendant’s reaction suggesting consciousness of guilt, and (b) emotional, separate disclosures to a best friend and later to her mother.
Apply “inconsequential and limited use” indicators. Several Boudreau-type factors pointed toward harmlessness: the disputed evidence was limited in volume, not emphasized in closing (or was corrected by a curative instruction), and not inflammatory relative to the charged conduct described by the complainant. The Court treated the defense’s “jury impasse” theory as speculative and, critically, declined to let it re-weight the record where the remaining factors strongly favored harmlessness.

3.3 Impact

Although the Court issued an order rather than a full opinion, Giroux meaningfully reinforces several practical lessons for New Hampshire evidentiary appeals:

Harmless error can carry the case even when evidentiary questions are close. Litigants should expect the Court to affirm where the State can show the verdict was independently supported by direct and corroborated evidence.
“CAC interview occurred” evidence is scrutinized for bolstering risk, but prejudice turns on use. The Court did not endorse admitting CAC-participation evidence; it held only that, on this record—no content admitted, minimal references, and a curative instruction—any assumed error did not affect the verdict.
Closing-argument emphasis is pivotal. Consistent with Rouleau and contrasted with Zuzelo, the Court treated limited or absent closing-argument reliance as a strong indicator of harmlessness.
Jury impasse is not (yet) a standard harmless-error factor in New Hampshire. The Court stated it was unaware of prior cases using a reported impasse as part of harmless-error analysis and rejected speculation about why the impasse occurred. Future appellants may cite Giroux against attempts to infer prejudice merely from deliberation dynamics absent a clearer doctrinal hook.

4. Complex Concepts Simplified

AFSA (Aggravated Felonious Sexual Assault): A serious sexual assault offense. Here, the State had to prove the defendant purposely committed “sexual penetration” (defined to include even slight intrusion of any part of the body) against a child under 13, on three separate occasions.
N.H. R. Ev. 404(b) (“Other Acts” Evidence): A rule that generally forbids using prior acts to show a person’s propensity (“he did it before, so he did it again”), while allowing other-acts evidence for limited purposes (e.g., motive, intent, identity) if relevant and not unfairly prejudicial. In Giroux, the Court assumed (without deciding) that the “in bed” testimony might have been improper other-acts evidence.
CAC Forensic Interview: A structured interview, typically conducted by trained professionals, used in child abuse investigations to reduce repeated questioning and to preserve a reliable account. Even mentioning that such an interview occurred can raise a “bolstering” concern if jurors infer professionals believed the complainant. Here, the jury heard that an interview happened and how CAC interviews work generally, but heard no statements A.A. made in the interview.
Harmless Error Beyond a Reasonable Doubt: Even if the trial court made a mistake, the conviction stands if the State proves beyond a reasonable doubt that the mistake did not affect the verdict. The Court looks at the whole trial, not the disputed evidence in isolation.
Consciousness of Guilt: Conduct or statements suggesting the defendant believed he was culpable (e.g., evasive behavior or statements implying wrongdoing), which can corroborate the State’s case.
Curative Instruction: A direction from the judge to the jury to disregard certain statements or to limit how evidence may be used. Appellate courts usually presume juries follow such instructions.
Waiver on Appeal: If a party raises an issue but fails to brief it adequately, the appellate court will treat it as abandoned.

5. Conclusion

State v. Giroux reinforces New Hampshire’s modern harmless-error doctrine: when the record contains overwhelming direct proof (a detailed complainant account) plus corroboration (consciousness-of-guilt statements and emotionally supported disclosures), limited and minimally emphasized evidentiary items—such as brief other-acts context and the mere fact of a CAC interview—will not warrant reversal even if their admission is assumed to be erroneous.

The order also signals caution against reading too much into deliberation dynamics: a reported jury impasse, without more, is not a recognized basis to infer prejudice in harmless-error review, and speculation about the reason for an impasse will not outweigh stronger record-based indicators that the verdict was independently supported.