State v. Shay

2026 UT App 83

THE UTAH COURT OF APPEALS

STATE OF UTAH,
Appellee,
v.
RAYMOND SHAY,
Appellant.

Opinion
No. 20240257-CA
Filed May 21, 2026

Third District Court, Salt Lake Department
The Honorable Elizabeth A. Hruby-Mills
No. 181904439

Emily Adams, Freyja Johnson, and
Rachel Phillips Ainscough, Attorneys for Appellant
Derek E. Brown and Daniel L. Day,
Attorneys for Appellee

JUDGE MICHELE M. CHRISTIANSEN FORSTER authored this Opinion,
in which JUDGE AMY J. OLIVER concurred. JUDGE RYAN M. HARRIS
concurred in part and in the result, with opinion.

CHRISTIANSEN FORSTER, Judge:

¶1 Raymond Shay appeals his convictions on seven counts of
aggravated sexual abuse of a child and one count of rape. He
argues that the trial court abused its discretion in allowing the
State to present testimony regarding sexual abuse allegations
asserted by a sibling of the victims in this case. We conclude,
however, that the trial court did not abuse its discretion in
determining both that this evidence had a proper, non-character
purpose and that the probative value of the evidence was not
substantially outweighed by a risk of unfair prejudice. We further
conclude that even were the evidence improperly admitted, the
State v. Shay

admission would have been only harmless error. We therefore
affirm Shay’s convictions.

BACKGROUND 1

¶2 In the summer of 2010, Shay became romantically involved
with a single mother (Mother) and moved in with her and her five
children in their house. Among Mother’s children were three
daughters: Ashley (age sixteen), Beth (age nine), and Clara (age
six or seven). 2

¶3 The following year, when Ashley was seventeen years old,
she alleged that Shay sexually abused her. According to Ashley,
the abuse occurred one afternoon when Shay called her into his
bedroom and began telling her he was proud of her for certain
recent accomplishments and behaviors. Ashley related that Shay
“pushed [her] down on the bed,” “reached under [her] shirt,” and
then “put his hands down [her] pants” and touched her vagina.

¶4 The next day, Ashley told Mother about the abuse.
Although Ashley initially thought Mother believed her
allegations, Mother’s subsequent actions and comments
suggested otherwise. Mother made Ashley move out of the house
that same day, and Mother never took Ashley to the police to
report the alleged touching. Mother not only “wouldn’t talk to”
Ashley, but Ashley was “no longer welcomed in the home,” and
from Ashley’s perspective, “[i]t was basically like [she] never
existed.” Ashley did, however, report the touching to the police

1. “We recite the facts in the light most favorable to the jury’s
verdict, and we present conflicting evidence as necessary to
understand issues raised on appeal.” State v. Black, 2015 UT App
30, ¶ 2, 344 P.3d 644.

2. We use pseudonyms to refer to the three children.

20240257-CA 2 2026 UT App 83
State v. Shay

herself, which eventually resulted in Shay pleading no contest to
one count of sexual battery.

¶5 In 2013, Mother and the remaining children moved to an
apartment, with Beth and Clara sharing a bedroom. Toward the
end of the year, Shay again moved in with the family, and Shay
and Mother married the following year. During the years at the
apartment, Shay would sometimes, while Mother was away at
work, watch scary movies with the children in the dark, and
when Beth (then twelve years old) would sit by Shay, he “would
lay his hand flat on the couch and then would grab [her]
butt[ocks].”

¶6 Also during this same period of time, Beth was diagnosed
with arthritis and could find relief from her joint pain only by
being massaged. Shay would often massage Beth’s legs and hips
when Mother was away at work, but he “would also go lower to
[her] hips and [her] thighs and get closer to [her] vagina.”
“Occasionally” during these massages, Shay touched Beth’s
vagina, sometimes over clothing and sometimes under it. Beth
“tried pushing his hand away” because she knew “it was wrong,”
but Shay “kept pushing his hand even closer.” On one occasion,
Mother observed Shay massaging Beth’s legs and noticed “he was
trying to get a little closer to [Beth’s] vaginal area,” resulting in
Beth looking “like she was uncomfortable.”

¶7 Shay additionally began asking Beth to “sit on his lap.”
And when she would do so, he would “move [Beth’s] hips around
over his penis” such that she could “feel his erection through his
pants.”

¶8 When Beth was thirteen years old and “had started to
develop” breasts, she and Shay “would have tickle fights”
together when Mother was not home. During these fights, Shay
would squeeze her breasts over her clothing. Shay would also
sometimes squeeze Beth’s buttocks and breasts when he went into
the girls’ bedroom either to say goodnight or to say goodbye in
the morning.

20240257-CA 3 2026 UT App 83
State v. Shay

¶9 Beth did not confront Shay about the inappropriate
touching or report it to Mother at this time because she “was
scared that [Mother] was not going to believe [her] like she did
with [Ashley].”

¶10 During the time that the family was living at the
apartment, Shay began to also inappropriately touch Clara, now
eleven years old. The abuse started one night when Shay came to
the girls’ room at bedtime and began rubbing Clara’s back but
then moved his hand down to rub her buttocks for “a couple
minutes.” A few nights later, this same activity happened a
second time, but this time Clara “tr[ied] to turn over so he would
stop” and Shay “started touching [her] vagina over [her] clothes.”
Then “maybe a week after that” second incident, the same pattern
recurred, with Shay rubbing Clara’s back and moving down to
her buttocks, but this time he “slipped his hand under [her]
clothes” and touched both her buttocks and vagina skin-to-skin.
Clara again “tr[ied] to turn over hoping he would stop,” and Shay
removed his hand. Clara remembers this kind of touching
happening “a couple times a week” while they lived in the
apartment.

¶11 For a few months when Clara was twelve years old, the
family moved to live with the girls’ grandfather. Shay’s
inappropriate touching of Clara at bedtime continued there, with
Shay sometimes putting “his hand . . . down [Clara’s] pants” and
“rubbing [her] vagina.” Clara did not tell anyone in her family
what was happening because she was “scared that no one would
believe [her],” similar to what had happened with Ashley.

¶12 Toward the end of 2015, the family then moved to another
house. Once they were “settled in” their new house, the pattern of
abuse with Clara continued, with Shay touching her under her
clothing when he came in at bedtime. But now, Shay additionally
“started shoving his fingers inside [Clara’s] vagina.” Clara
remembers this happening “[e]very night” that they lived in that
house with Shay.

20240257-CA 4 2026 UT App 83
State v. Shay

¶13 One night, when Clara was fourteen years old, the abuse
escalated. Clara was in bed trying to fall asleep when Shay entered
her bedroom, “took off the blanket,” “put [Clara’s] legs up on his
shoulders,” “slid [her] bottoms down,” and put “some of his
penis” into her vagina. Shay then heard something, causing him
to leave, and Mother “caught him coming out of [Clara’s] room.”
Mother asked Shay why he was coming out of the girls’ bedroom,
and Shay responded that he had been “just checking on” the girls.

¶14 On a different night when Clara was fourteen years old,
Mother, who was growing suspicious of Shay, decided to sleep on
the floor in the girls’ room, with the door locked. During the
middle of the night, Shay tried to enter the room and “was
banging on the door trying to get in.” Mother responded that she
was in the room, and Shay eventually left.

¶15 Mother was also uncomfortable when she saw Shay hug
Beth and Clara from behind, because she knew that when Shay
hugged her that way, he would rub his penis against her, and she
was hoping he was not doing the same to the girls. Mother
additionally noticed that during this time, Beth and Clara “would
try to avoid [Shay] as much as possible.”

¶16 Mother eventually kicked Shay out in 2017. A few months
later, Beth and Clara started talking with each other about the
abuse Shay had been inflicting on them. At first, they “were kind
of iffy” about whether they should tell Mother; they were worried
that she would not believe them because she had not believed
Ashley. But they thought that “if there [were] two of [them],”
things might be different, and they wanted to make sure that
Mother would not take Shay back again. Beth and Clara told
Mother about the abuse, and Mother promptly reported the abuse
to the police and took the girls in to be interviewed regarding the
abuse.

¶17 For his actions related to Beth and Clara, Shay was
ultimately charged with seven counts of aggravated sexual abuse
of a child and one count of rape. Before trial, the State provided

20240257-CA 5 2026 UT App 83
State v. Shay

notice that it intended to present evidence of “other bad acts”
committed by Shay, namely, the “prior sexual offense towards
[Ashley].” The State argued that evidence of this conduct was
being offered “not in an attempt to portray [Shay] as an individual
with a propensity to assault” but, rather, “to establish that [Shay]
had a continuing intent and a common scheme to sexually assault
these three siblings with whom he had a step-father relationship,”
that he “knew his behavior with [Beth and Clara] was illegal,” and
that he had an “intent to commit the crime with which he [was
then] charged.” Over Shay’s objection, the trial court granted the
State’s motion, concluding that the evidence was “relevant and
probative to show intent and a common plan or scheme to engage
in sexual contact with three minor siblings” to whom he had been
a parental figure. The court also determined that the probative
value of the evidence was “not outweighed by any risk of unfair
prejudice.”

¶18 The case proceeded to trial, with Shay choosing to
represent himself with the assistance of standby counsel. The
State presented its case largely through the testimony of Beth,
Clara, Mother, and Ashley. Shay presented no opening statement,
did not cross-examine Mother or the three girls, and did not call
any defense witnesses.

¶19 At several points throughout the proceedings, the jury was
reminded of the limited purposes for which it could use the
evidence regarding Ashley’s accusations of abuse. First, during
the State’s opening statement, the prosecutor previewed Ashley’s
expected testimony and said,

Now, you’re going to hear from [Ashley] not
because the conduct that [Shay] engaged in towards
her, the abuse he engaged in towards her isn’t
charged in this case, but because it shows that this
was not a mistake, that this was not an accident, that
this was the plan of [Shay], . . . to repeatedly sexually
abuse his stepdaughters.

20240257-CA 6 2026 UT App 83
State v. Shay

¶20 Second, the trial court provided a specific jury instruction
addressing the matter:

You have heard evidence that the defendant
committed an offense against [Ashley] before the
acts charged in this case. You may consider this
evidence, if at all, for the limited purposes of the
defendant’s intent, knowledge, and common
scheme. Keep in mind that the defendant is on trial
for the crimes charged in this case, and for those
crimes only. You may not convict the defendant
simply because you believe he may have committed
some other act at another time.

¶21 Third, during closing argument, the prosecutor again
acknowledged, “[T]he conduct that [Shay] did with [Ashley],
that’s not at issue here. Those are not charged counts. Those are
specifically to show that this wasn’t an accident and that this was
what [Shay’s] plan was. This is how he did it.”

¶22 Finally, Shay’s standby counsel, who gave the majority of
Shay’s closing argument, told the jury,

You also heard from [Ashley], but . . . none of the
charges have anything to do with her. And to be
clear about this, you heard from [Ashley] for very
specific reasons; to determine whether there was
knowledge, intent, or a common scheme in this case.

Our law is very clear that we don’t allow
evidence for what’s called propensity. We don’t
allow evidence of some other act to say, well,
because this person did that other act, they probably
did this similar act on this occasion. You cannot use
what [Ashley] said for that reason.

20240257-CA 7 2026 UT App 83
State v. Shay

What it came in for was to show that basically
because some of these things could have been sort
of incidental, touching, . . . tickle fights that resulted
in touching a buttocks or touching a breast or that
sort of thing, that this evidence of this other event
was intended to show that this wasn’t an accident,
if it happened. But you cannot say, well, because
something similar happened with [Ashley], that
makes . . . what [Beth and Clara] said more likely to
be true; that you can’t do.

¶23 The jury ultimately convicted Shay as charged. He
thereafter timely appealed.

ISSUE AND STANDARD OF REVIEW

¶24 Shay argues that the trial court abused its discretion in
allowing the State to present testimony of Ashley’s sexual abuse
allegations under rule 404(b) of the Utah Rules of Evidence. “We
review a [trial] court’s evidentiary rulings for abuse of discretion.
When a [trial] court applies the correct legal standard, its decision
to admit or exclude evidence is only an abuse of discretion if it is
beyond the limits of reasonability.” State v. Blackwing, 2025 UT 60,
¶ 16, 582 P.3d 829 (quotation simplified).

ANALYSIS

¶25 Rule 404(b) of the Utah Rules of Evidence provides that
“[e]vidence of a crime, wrong, or other act is not admissible to
prove a person’s character in order to show that on a particular
occasion the person acted in conformity with the character.” Utah
R. Evid. 404(b)(1). However, the rule also provides that this same
evidence “may be admissible for another purpose, such as
proving motive, opportunity, intent, preparation, plan,
knowledge, identity, absence of mistake, or lack of accident.” Id.
R. 404(b)(2). “And rule 404(b)’s use of ‘such as’ indicates that the

20240257-CA 8 2026 UT App 83
State v. Shay

list of non-character purposes is illustrative and not exclusive.”
State v. Green, 2023 UT 10, ¶ 70, 532 P.3d 930 (quotation
simplified). Thus, “under rule 404(b), the question is whether the
evidence has a plausible, avowed purpose beyond the propensity
purpose that the rule deems improper. If it does[,] then the
evidence is presumptively admissible (subject to rule 402 and 403
analysis).” Id. ¶ 63 (quotation simplified). See generally Utah R.
Evid. 402 (“Relevant evidence is admissible unless any of the
following provides otherwise: the United States Constitution; the
Utah Constitution; a statute; or rules applicable in courts of this
state.”); id. R. 403 (“The court may exclude relevant evidence if its
probative value is substantially outweighed by a danger of one or
more of the following: unfair prejudice, confusing the issues,
misleading the jury, undue delay, wasting time, or needlessly
presenting cumulative evidence.”). 3

¶26 “The difficulty in applying this simple rule, however,
springs from the fact that evidence of prior bad acts often will
yield dual inferences—and thus betray both a permissible
purpose and an improper one.” State v. Verde, 2012 UT 60, ¶ 16,
296 P.3d 673, abrogated on other grounds by State v. Thornton, 2017
UT 9, 391 P.3d 1016, and abrogated on other grounds by Green, 2023
UT 10. In circumstances “where the court concludes that the only
real effect of the evidence is to suggest likely action in conformity
with bad character,” the “avowed proper purpose may be rejected

3. Shay argued below that testimony regarding Ashley’s sexual
abuse allegations was not relevant under rule 402. But on appeal,
acknowledging that the rules of evidence “establish a very low
bar that deems even evidence with the slightest probative value
relevant and presumptively admissible,” State v. Richardson, 2013
UT 50, ¶ 24, 308 P.3d 526 (quotation simplified), he chose to not
develop a rule 402 argument; instead, he focused his appellate
argument on asserting that, even “assuming” relevance, this
evidence “was not used for a proper purpose” and “was unfairly
prejudicial.” We therefore do not separately address the trial
court’s determination as to relevance—a position that the separate
opinion also embraces, see infra ¶¶ 37–38.

20240257-CA 9 2026 UT App 83
State v. Shay

as a pretext or ruse.” Thornton, 2017 UT 9, ¶ 59 (quotation
simplified). “Short of that, however, the court’s job under rule
404(b) is not to balance or weigh competing (proper and
improper) inferences. Such weighing comes in under rule 403.” Id.
(footnote omitted). Thus, “even if the evidence may sustain both
proper and improper inferences under rule 404(b), the court
should balance the two against each other under rule 403,
excluding the bad acts evidence if its tendency to sustain a proper
inference is outweighed by its propensity for an improper
inference or for jury confusion about its real purpose.” Verde, 2012
UT 60, ¶ 18; see also Green, 2023 UT 10, ¶ 64 (“In sum, the rules of
evidence provide that evidence of prior crimes, uncharged
misconduct, or bad acts is admissible if it (1) is relevant to, (2) a
proper, non-character purpose, and (3) does not pose a danger for
unfair prejudice that substantially outweighs its probative value.”
(quotation simplified)).

¶27 Applying these rules to the instant case, we do not see that
the trial court abused its discretion in determining that the
testimony regarding Ashley’s allegations was relevant to the
proper, non-character purpose of showing Shay’s intent, that is,
that his touching of Beth and Clara was done with the specific
intent to “arouse or gratify [his] sexual desire.” Utah Code § 76-5-
404.1(2)(a)(ii)(B). This court has previously determined that
showing this particular intent is a valid, non-character purpose
under rule 404(b).

¶28 One example is State v. Bair, 2012 UT App 106, 275 P.3d
1050, wherein the defendant was charged with aggravated sexual
abuse of a child after abusing his daughter. See id. ¶¶ 2, 4. Prior to
the abuse, the defendant had disclosed that he was “addicted to
the touchy/feely-play around part of sex,” and the State sought
admission of that statement under rule 404(b), arguing that the
defendant’s “admitted addiction demonstrated his intent in
touching” his daughter. Id. ¶¶ 15, 19 (quotation simplified). The
trial court permitted the evidence to be admitted, and the
defendant thereafter appealed that decision, “arguing that the
[evidence] ‘amounted to inadmissible propensity evidence in

20240257-CA 10 2026 UT App 83
State v. Shay

disguise.’” Id. ¶ 15. This court disagreed, concluding that the
defendant’s prior disclosure did “provide evidence of [his]
specific intent to arouse or gratify his sexual desires, which is a
proper, noncharacter purpose under rule 404(b).” Id. ¶ 20. And we
explained that “even if admission of the [evidence] prompted the
jury to infer [the defendant] acted in conformity with his
addiction, rule 404(b) permits admission of prior acts evidence so
long as the sole purpose is not to prove that a defendant acted in
conformity with a character trait.” Id. ¶ 19.

¶29 Likewise, here the State’s use of evidence regarding
Ashley’s allegations of abuse was not an attempt to prove that
Shay acted in accordance with a general bad character but was,
instead, an effort to provide evidence of Shay’s specific intent to
seek sexual gratification through sexual contact with his minor
stepdaughters. The evidence shows that he targeted his minor
stepdaughters who lived in his household, exploited his parental
role, initiated abuse in private settings, and relied on secrecy and
authority—all as a means to gratify his sexual desires. This is a
proper, non-character purpose under rule 404(b), and the trial
court therefore did not abuse its discretion in so deciding. 4 See

4. The trial court did not view the State’s purported non-character
purpose of intent as a pretext or ruse. Both Shay and the separate
opinion argue that this was an abuse of the court’s discretion
because this case is, like State v. Verde, a case in which “intent is
uncontested and readily inferable from other evidence.” Verde,
2012 UT 60, ¶ 26, 296 P.3d 673, abrogated on other grounds by State
v. Thornton, 2017 UT 9, 391 P.3d 1016, and abrogated on other grounds
by State v. Green, 2023 UT 10, 532 P.3d 930. For two reasons, we
disagree.
First, it is not clear that intent was uncontested here. The
only record citation Shay provides for his assertion of uncontested
intent is to a statement made in closing wherein he explained that
he had been in prison for nine months during the time period
given for some of the charged offenses and then said, “For that, I
think my only defense would be, my argument would be it’s
(continued…)

20240257-CA 11 2026 UT App 83
State v. Shay

State v. Burke, 2011 UT App 168, ¶ 30, 256 P.3d 1102 (recognizing
that evidence of multiple acts “is offered for a proper,
noncharacter purpose if it is offered to prove intent,” and
concluding that multiple sexual offenses, “when considered
together with the circumstances that connect them, show[ed] that
[the defendant] possessed that specific sexual intent at the time
each offense was committed”). 5

fabricated.” (Emphasis added.) And Shay also made another
comment in closing that, while not very clear, could arguably be
interpreted as contesting intent: “My argument would be through
all from Count 1 to 8. Count 1 as it says, stated intentionally,
knowingly, recklessly, where No. 3, I touched my daughter. That
part I have not—I go against it too, Your Honor.” (Emphasis
added.)
Second, the situation in Verde was markedly different than
the situation here. There, the defendant not only “offered to
stipulate to his intent,” but his “intent [was] inferable from proof
that [he] groped [the victim’s] genitalia” because with such
groping, “it’s hard to imagine a jury that would conclude that [the
defendant] committed the actus reus but with an innocent intent.”
Id. ¶ 25. But here, Shay made no concession of intent, and at least
some of the alleged touching (i.e. those tied to the massaging and
tickle fights) commenced as welcome touching and then allegedly
included inappropriate touching—a situation in which there is a
possibility that a jury might conclude that the touching happened
without the requisite intent.

5. Although the trial court’s decision also found the evidence was
relevant to the additional non-character purpose of showing “a
common plan or scheme,” the State concedes on appeal that “the
approximate two-year gap between Shay’s sexual abuse of
[Ashley] and the beginning of the charged crimes likely fails the
temporal-proximity requirement necessary to establish a common
scheme or plan.” We therefore do not evaluate the applicability of
this alternate purpose further.
(continued…)

20240257-CA 12 2026 UT App 83
State v. Shay

¶30 As to the required rule 403 balancing, “trial courts have
wide discretion in determining relevance, probative value, and
prejudice. This means that we will not reverse the trial court’s
[rule] 403 ruling unless we find it was beyond the limits of
reasonableness.” State v. Beverly, 2018 UT 60, ¶ 56, 435 P.3d 160
(quotation simplified); see also De La Cruz v. Ekstrom, 2024 UT App
18, ¶ 11, 545 P.3d 285 (“Ultimately, as long as the court’s decision
was within the broad range of discretion entrusted to trial judges,
even if we think the trial judge made the wrong call we will affirm
because it was, in its essence, a discretionary call.” (quotation
simplified)); Burke, 2011 UT App 168, ¶ 42 (“We must recognize
that trial courts are allowed considerable freedom in applying
rule 403 to the facts of a case, freedom to make decisions which
appellate judges might not make themselves ab initio but will not
reverse.” (quotation simplified)).

¶31 And we do not see that the trial court abused its wide
discretion here. The evidence regarding Ashley’s allegations
carried substantial probative value by corroborating Beth’s and
Clara’s testimony that Shay intentionally targeted them for sexual
gratification. Although there were, admittedly, differences
between the events alleged by Ashley and the events alleged by

We do, however, note that, although not a purpose
highlighted by the trial court, the evidence of Ashley’s allegations
and the fallout therefrom was relevant to the additional non-
character purpose of providing context. The evidence regarding
Ashley’s allegations explained why Beth and Clara endured
abuse for years without telling anyone and explained their fear
that they would not be believed—they were afraid that Mother
would, like with Ashley, not believe them and that she might react
in a similarly unsupportive manner to how she reacted after
Ashley’s disclosure. Thus, this was yet another non-character
purpose that could have been pursued under rule 404(b). See
Thornton, 2017 UT 9, ¶ 57 (allowing other acts evidence that
“presented a narrative of relevance to the prosecution’s case,”
including by “suggesting why [the victim] may have waited to
come forward with accusations against [the defendant]”).

20240257-CA 13 2026 UT App 83
State v. Shay

Beth and Clara, the trial court’s assessment was that “[t]here
[were] sufficient similarities in conduct between the charged
offenses [and] the prior bad acts.” Cf. State v. Green, 2023 UT 10,
¶ 78, 532 P.3d 930 (“[W]e cannot say that the [trial] court abused
its broad discretion in concluding that the similarities among the
[victims’] accounts reduced the tendency for the jury to decide
upon an improper basis and that the danger of unfair prejudice
did not substantially outweigh the evidence’s probative value.”).
Further, the trial court provided the jury with a limiting
instruction, specifically listing the reasons for which the jury
could consider the evidence regarding Ashley’s allegations and
also clearly stating, “You may not convict the defendant simply
because you believe he may have committed some other act at
another time.” See id. ¶ 80 (recognizing that issuing a limiting
instruction could have “a material impact on the rule 403
balancing” (quotation simplified)).

¶32 Based on all this, the trial court did not exceed its
wide discretion in determining that the probative value of
the other acts evidence was not substantially outweighed by a
danger that the jury would base its decision on an
improper propensity inference. See Bair, 2012 UT App 106, ¶ 22
(“Even if some level of unfair prejudice exists, the mere fact
that evidence possesses a tendency to suggest a decision upon
an improper basis does not require exclusion; evidence may
be excluded only if the danger of unfair prejudice
substantially outweighs the probative value of the proffered
evidence.” (quotation simplified)). Thus, the trial court did
not abuse its discretion in allowing the State to present
testimony regarding Ashley’s prior allegations of abuse.

¶33 Finally, we also agree with the State’s assertion that even if
the trial court abused its discretion in admitting evidence under
rule 404(b), any such error was ultimately harmless. “An
erroneous decision to admit or exclude evidence cannot result in
reversible error unless the error is harmful.” Green, 2023 UT 10,
¶ 101 (quotation simplified). “For an error to require reversal, the

20240257-CA 14 2026 UT App 83
State v. Shay

likelihood of a different outcome must be sufficiently high to
undermine confidence in the verdict.” Id. (quotation simplified).

¶34 Beth and Clara each provided largely consistent, detailed
testimony of abuse spanning several years. Not only was their
testimony mutually corroborative, but it was additionally
corroborated by Mother’s testimony that Shay seemed to be
getting closer to the vaginal area when massaging Beth, that she
caught Shay trying to enter the girls’ room in the middle of the
night, that she was concerned about the manner in which Shay
hugged the girls, and that the girls increasingly tried to avoid
being around Shay. Further, this evidence went almost entirely
unchallenged; Shay chose to proceed pro se at trial and neither
offered an opening statement setting forth a defense nor engaged
in any cross-examination of Mother or the three girls. Under these
circumstances, we are not convinced that had the trial court
excluded all evidence regarding Ashley’s allegations, there is a
reasonable likelihood of a different outcome here. For this
additional reason, we affirm Shay’s convictions.

CONCLUSION

¶35 The trial court did not abuse its discretion in allowing
evidence of prior acts under rule 404(b) of the Utah Rules of
Evidence. And even were there any error related to this
admission, we would determine it to be only harmless error. We
therefore affirm.

HARRIS, Judge (concurring in part and concurring in the result):

¶36 I join Paragraphs 33 and 34 of the lead opinion, and I
concur in the result reached by the majority: that Shay’s
convictions should be affirmed. But I reach that conclusion
largely by way of very different reasoning than the lead opinion
does, and I write separately to explain that reasoning, an

20240257-CA 15 2026 UT App 83
State v. Shay

endeavor that requires me to set forth some concerns I have with
our supreme court’s decision in State v. Green, 2023 UT 10, 532
P.3d 930.

¶37 The issue in this case is whether the trial court erred by
overruling Shay’s objection to Ashley’s testimony that—when she
was seventeen, and several years before he sexually abused Beth
and Clara—Shay sexually abused her too. This testimony thus
constitutes “prior-bad-acts evidence,” the admissibility of which
is governed by rule 404(b) of the Utah Rules of Evidence. Under
Utah law, “such evidence is admissible if it (1) is relevant to, (2) a
proper, non-character purpose, and (3) does not pose a danger for
unfair prejudice that substantially outweighs its probative value.”
State v. Gallegos, 2020 UT App 162, ¶ 15, 479 P.3d 631 (quotation
simplified). The first step is not contested here: Shay at least tacitly
acknowledges that Ashley’s testimony is “relevant,” as that term
has been defined in our law. See id. ¶ 16 (noting that the relevance
inquiry “presents a low bar” (quotation simplified)); see also State
v. Murphy, 2019 UT App 64, ¶ 47, 441 P.3d 787 (Harris, J.,
concurring) (noting that prior-bad-acts evidence is presumptively
excluded by our rules “not because it has no appreciable
probative value, but because it has too much” (quotation
simplified)).

¶38 It is thus the final two steps that are contested here: that is,
Shay asserts that there is no proper non-character purpose for
admission of Ashley’s testimony and that (in any event) the
danger of unfair prejudice substantially outweighs any legitimate
probative value Ashley’s testimony might have. On these points,
I part ways with the lead opinion. In my view, no proper non-
character purpose was articulated to the trial court that would
justify admission of Ashley’s testimony in this case. And I also
believe, under the circumstances, that the bulk of Ashley’s
testimony had extremely low legitimate probative value and
that any such value was substantially outweighed by the risk of
unfair prejudice, namely, the risk that the jury would draw a
forbidden propensity inference. For these reasons, and as further
explained below, in my view the trial court exceeded its discretion

20240257-CA 16 2026 UT App 83
State v. Shay

by overruling Shay’s objection to the entirety of Ashley’s
testimony.

¶39 But I nevertheless agree that Shay’s convictions should be
affirmed because, on this record, the trial court’s error was
harmless. In a counterfactual trial in which Ashley did not testify,
Shay enjoys no reasonable likelihood of a different outcome. On
this basis, I join Paragraphs 33 and 34 of, and concur in the result
reached by, the lead opinion.

I. Non-Character Purpose

¶40 Under our evidentiary rules, “the admissibility of prior
misconduct evidence depends on its avowed purpose.” State v.
Verde, 2012 UT 60, ¶ 15, 296 P.3d 673, abrogated on other grounds by
State v. Thornton, 2017 UT 9, 391 P.3d 1016, and abrogated on other
grounds by State v. Green, 2023 UT 10, 532 P.3d 930. Our rules forbid
the introduction of prior-bad-acts evidence “to prove a person’s
character in order to show that on a particular occasion the person
acted in conformity with the character.” Utah R. Evid. 404(b)(1).
“This forbidden line of thinking is sometimes referred to as a
‘propensity inference’—that is, if jurors are told that a person has
acted in a certain way on previous occasions, they may conclude
that it is in that person’s character to act that way, and may
conclude that, due to this propensity, the person was much more
likely to have acted in conformity with that propensity on the
occasion in question.” Gallegos, 2020 UT App 162, ¶ 13; see also
Verde, 2012 UT 60, ¶ 15 (stating that, when prior-bad-acts evidence
“is offered to suggest action in conformity with a person’s alleged
bad character, it is inadmissible”).

¶41 But while prior-bad-acts evidence is not admissible under
rule 404(b) to prove propensity, “that rule allows admission of
evidence of such acts for other purposes.” Gallegos, 2020 UT App
162, ¶ 14. Under the rule, prior-bad-acts evidence “may be
admissible” to prove “motive, opportunity, intent, preparation,
plan, knowledge, identity, absence of mistake, or lack of
accident.” Utah R. Evid. 404(b)(2). Thus, “when past misconduct

20240257-CA 17 2026 UT App 83
State v. Shay

evidence is offered for any other purpose—other than to suggest
action in conformity with the bad character suggested by [a
defendant’s] prior bad acts—such evidence is admissible so long
as it satisfies rules 402 and 403.” Thornton, 2017 UT 9, ¶ 36
(quotation simplified); see also Verde, 2012 UT 60, ¶ 15 (“So long as
the evidence is not aimed at suggesting action in conformity with
bad character, it is admissible under rule 404(b).”).

¶42 But “evidence of prior bad acts often will yield dual
inferences—and thus betray both a permissible purpose and an
improper one.” Verde, 2012 UT 60, ¶ 16. And in such cases, it may
be difficult for “the court to differentiate the two inferences or to
limit the impact of the evidence to the purpose permitted under
the rule.” Id. Accordingly, our supreme court has instructed lower
courts to look closely at whether the evidence at issue “is genuinely
being offered for a proper, non-character purpose, or whether it
might actually be aimed at sustaining an improper inference of
action in conformity with a person’s bad character.” Id. ¶ 18
(emphasis added). “If such evidence is really aimed at
establishing a defendant’s propensity to commit crime, it should
be excluded despite a proffered (but unpersuasive) legitimate
purpose.” Id. ¶ 17 (quotation simplified). We must therefore
examine the purposes proffered by the State (and by the lead
opinion) for admission of Ashley’s testimony and consider
whether those purposes are genuine.

¶43 When this issue first arose at the trial court level, the State’s
initial position was that there were many proper non-character
purposes for Ashley’s testimony. Specifically, it asserted that
Ashley’s testimony was admissible to show “motive, intent,
knowledge, absence of mistake, lack of accident and common
scheme.” During briefing and argument in the trial court, the
number of potential non-character purposes was whittled down,
with the court eventually settling on two, ruling that Ashley’s
“testimony [was] relevant and probative to show intent and a
common plan or scheme to engage in sexual contact with three
minor siblings with [Shay as] the step-father or step-father
figure.” (Emphasis added.)

20240257-CA 18 2026 UT App 83
State v. Shay

¶44 In defending against Shay’s appellate arguments,
however, the State has further refined its position, abandoning
any argument that “common plan or scheme” is a proper non-
character purpose in this case. It does, however, defend the trial
court’s ruling that intent functions as such in this case. And it also
argues, as an alternative ground for affirmance, that rebuttal of
Shay’s fabrication defense—Shay argued at trial that Beth and
Clara were untruthful or mistaken in accusing him of abuse—is a
proper non-character purpose justifying admission of Ashley’s
testimony. Then, on its own, the lead opinion suggests a third
proposed non-character purpose, one not mentioned by the trial
court or argued as such by the State on appeal: that Ashley’s
testimony is “relevant to the additional non-character purpose of
providing context.” Supra note 5.

¶45 In my view, none of these arguments is persuasive enough,
either as a matter of logic or as a matter of interpretation of our
rules of evidence, to justify admission of the entirety of Ashley’s
testimony. I discuss each in turn.

A. Intent

¶46 “Intent” is indeed one of the purposes listed in rule 404(b)
as a potentially proper non-character purpose. But to gain
admission of prior-bad-acts evidence on the basis of intent, it is
not enough for the proponent of the evidence to merely incant the
word “intent.” Rather, that party must demonstrate that the
evidence is truly being admitted to prove intent and is not being
offered merely as cover for a back-door propensity inference. As
our supreme court put it, “if proof of intent is merely a ruse, and
the real effect of prior misconduct evidence is to suggest a
defendant’s action in conformity with alleged bad character, the
ruse is insufficient and the evidence should not be admitted.”
Verde, 2012 UT 60, ¶ 22.

¶47 In particular, the Verde court repudiated the “not-guilty
rule,” under which it was thought that, as long as a defendant
pled not guilty to the charged offense, the defendant thereby put

20240257-CA 19 2026 UT App 83
State v. Shay

his state of mind at issue. Id. ¶¶ 21–24. The court called the not-
guilty rule “an undisciplined substitute for careful analysis under
rule 404(b),” and it held that “the technical relevance of evidence
of a defendant’s intent is not enough to justify the admissibility of
evidence of prior bad acts purportedly aimed at establishing
intent.” Id. ¶¶ 22–23. Instead, the court instructed lower courts to
“evaluate the true purpose” of proffered prior-bad-acts evidence,
despite the existence of a not-guilty plea. Id. ¶ 24. And in engaging
in that same exercise in Verde, the court found it highly relevant
that the defendant “did not contest intent at trial.” Id. ¶ 25. Indeed,
the court stated that “[w]here intent is uncontested and readily
inferable from other evidence, [prior-bad-acts] evidence is largely
tangential and duplicative.” Id. ¶ 26. On the facts of Verde, where
the defendant was not contesting intent, the court held that “it
seems much more likely that [the prior-bad-acts evidence] was
aimed at sustaining an impermissible inference that [the
defendant] acted in conformity with the bad character.” Id. 6

¶48 Like the defendant in Verde, Shay did not contest the intent
element of the charged crimes in this case. In Verde, the defendant
put forth as his “primary defense” the argument that “he never
touched [the complaining witness’s] genitalia” at all. Id. ¶ 25.
Likewise here, Shay describes his defense as that “he did not touch
[Beth] and [Clara]” at all, “not that he touched them by mistake or
without the requisite intent.” The State agrees that this was Shay’s
main defense, acknowledging (as discussed more fully below)
that Shay raised a “fabrication” defense. Thus, just as in Verde, in
this case the intent element was “uncontested and readily
inferable from other evidence.” Id. ¶ 26. And that is a good clue
that the true motive for offering the evidence is not to establish

6. The main case upon which the lead opinion relies in this
context—State v. Bair, 2012 UT App 106, 275 P.3d 1050—came out
several months before Verde, and it must give way to Verde to the
extent the two opinions conflict. See Mathena v. Vanderhorst, 2020
UT App 104, ¶ 12 n.2, 469 P.3d 1144 (“We are bound by vertical
stare decisis to follow strictly the decisions rendered by the Utah
Supreme Court.” (quotation simplified)).

20240257-CA 20 2026 UT App 83
State v. Shay

intent but, instead, to invite the jury to draw a forbidden
propensity inference. See id.

¶49 My conclusion in this regard is further bolstered by the
manner in which the State, in its appellate brief, characterizes this
ostensible “intent” purpose. While the State uses slightly different
language at various points in its brief, its descriptions (with our
emphasis) are enough to give the game away:

• Ashley’s testimony “served the proper non-
character purpose of proving Shay’s specific
intent to seek sexual gratification through his
unusual predilections for his minor stepdaughters.”

• Ashley’s testimony “had the unique non-
character purpose of proving Shay’s specific
intent to sexually gratify himself through his
unusual motivations to force quasi-incestuous
relationships with his minor stepdaughters.”

• Ashley’s testimony was “admitted for the
proper, non-character purpose of establishing
Shay’s intent and motivations—his unusual
predilections toward seeking sexual gratification from
his future or current minor stepdaughters.”

• Shay had a “core pattern of exploiting his parental
role when Mother was not around to sexually
abuse three minor stepdaughters.”

¶50 Thus, taking the State’s description of its proffered purpose
at face value, it is apparent that the State wanted to admit Ashley’s
testimony—primarily if not exclusively—to invite the jury to
draw a forbidden propensity inference, namely, that Shay was the
type of person who had a propensity to sexually abuse young girls
in his care and that he acted in conformity with that character in
abusing Beth and Clara. On this record, I conclude that the State’s

20240257-CA 21 2026 UT App 83
State v. Shay

assertion that Ashley’s testimony operated to show Shay’s intent
was “merely a ruse” and that “the real effect” of the testimony
was “to suggest” a propensity inference. See id. ¶ 22. In my view,
the court erred, on this record, by concluding that “intent” was a
proper non-character purpose for admitting Ashley’s testimony.

B. Rebuttal of Fabrication

¶51 Perhaps recognizing the infirmities in its “intent”
arguments, the State shifts gears and invites us to affirm the trial
court’s ruling regarding non-character purpose on an alternative
ground. Specifically, the State asserts that because Shay raised a
fabrication defense—asserting that Beth and Clara were either
lying or mistaken—Ashley’s testimony was admissible, under
State v. Green, 2023 UT 10, 532 P.3d 930, to further the purpose of
rebutting that fabrication defense.

¶52 The State correctly observes that, in Green, our supreme
court did indeed hold that rebuttal of a fabrication defense
constituted a valid non-character purpose for the admission of
prior-bad-acts evidence. See id. ¶ 71. The State also correctly
observes that Shay was running a fabrication/mistake defense in
this case, in that he was asking the jury to disbelieve the testimony
from Beth and Clara. And the State correctly observes that, had
Green been decided prior to the trial court’s ruling in this case, the
State would almost certainly have argued—based on Green—that
Ashley’s testimony should be admitted to further the purpose of
rebutting that defense. And given Green’s analysis and holding,
the State would have had a very strong argument in that regard.

¶53 But for the reasons stated below, I do not think we can
affirm on this alternative basis, because—given the deferential
standard of review we apply in this context—the trial court
should have the first crack at deciding whether Ashley’s
testimony should be admitted pursuant to the rebuttal-of-
fabrication purpose. See State v. Dowhaniuk, 2025 UT App 100,
¶ 31, 574 P.3d 1000 (declining to affirm on an alternative ground
because that ground involved a determination that “should

20240257-CA 22 2026 UT App 83
State v. Shay

ordinarily be made in the first instance by the [trial] court”). At
best, the State’s argument about rebuttal of fabrication represents
something it could raise on remand, if the case were remanded.

¶54 However, even aside from the procedural problem of
affirming, right now, on the State’s proffered alternative ground,
I have reservations about the reasoning our supreme court
employed in Green in reaching its holding that rebuttal of
fabrication counts as a valid non-character purpose in this context.
I recognize, of course, that I am bound to follow our supreme
court’s holding in Green. But I nevertheless choose to take this
opportunity—as I have done on other occasions in this same
space, see State v. Lane, 2019 UT App 86, ¶¶ 36–50, 444 P.3d 553
(Harris, J., concurring); State v. Murphy, 2019 UT App 64, ¶¶ 45–
65, 441 P.3d 787 (Harris, J., concurring)—to express my view, for
whatever it might be worth, “that the governing law might
warrant re-examination in a future case.” See Murphy, 2019 UT
App 64, ¶ 45 (Harris, J., concurring).

* * * *

¶55 In Green, our supreme court finally did away with the
doctrine of chances, a legal doctrine against which I have, on a few
occasions, leveled some criticism. See Lane, 2019 UT App 86,
¶¶ 36–50 (Harris, J., concurring); Murphy, 2019 UT App 64, ¶¶ 45–
65 (Harris, J., concurring). I had been critical of the doctrine of
chances because, in my view, it was a metaphysical mess and
quite confusing for lower courts, not to mention lay juries, to
make any sense of. And I was critical of the doctrine not only
because it found no support in the text of the Utah Rules of
Evidence, but also because it served to materially undermine the
overarching reason for rule 404(b)’s existence: the ban on
propensity evidence. I noted that there may well be excellent
policy-based reasons to change our evidentiary rules to allow
admission of propensity evidence, at least in certain cases, and
that—for instance—the Federal Rules of Evidence had been so
amended to allow such evidence in all sexual assault and child
molestation cases. See Murphy, 2019 UT App 64, ¶¶ 49, 65 (Harris,

20240257-CA 23 2026 UT App 83
State v. Shay

J., concurring); see also Fed. R. Evid. 413(a) (“In a criminal case in
which a defendant is accused of a sexual assault, the court may
admit evidence that the defendant committed any other sexual
assault.”). But my overall point was that, if we as a society believe
that propensity evidence should be admitted either generally or
in certain types of cases, we should be straightforward about it
and amend our rules of evidence to textually allow for that
outcome, rather than continuing to profess fealty to rules that
include a propensity ban while at the same time inventing legal
doctrines that function as a contra-textual end run around that
same ban. See Murphy, 2019 UT App 64, ¶ 65.

¶56 In Green, our supreme court overruled its doctrine-of-
chances cases for precisely these reasons: the court stated that the
doctrine had “been difficult to apply in practice” and determined
that the doctrine’s “requirements deviate[d] from the plain text of
the rules of evidence.” 2023 UT 10, ¶ 49. I agree wholeheartedly
with this assessment, and I do not mourn the passing of Utah’s
doctrine-of-chances jurisprudence.

¶57 But the Green court put something else in its place that is,
in my view, equally erroneous and suffers from some of the
same flaws as the doctrine-of-chances jurisprudence did. Indeed,
the new regime continues to perpetuate the central problem that
I identified with the doctrine of chances: it mouths fealty to
the propensity ban, see id. ¶ 63 (noting “that the rule deems” a
“propensity purpose” “improper” (quotation simplified)),
while just as—perhaps even more—effectively removing all
practical limitations to admission of prior-bad-acts evidence.
Again, removing these limitations may be a perfectly acceptable
policy outcome; indeed, this outcome was achieved years
ago with regard to child molestation cases. See Utah R. Evid.
404(c). And this outcome was achieved just this year, by
legislative amendment, with regard to all other sexual assault
cases. See id. R. 404(d)(2) (“In a criminal case in which a defendant
is accused of sexual assault, the court may admit evidence that the
defendant committed any other acts of sexual assault. This
evidence may be considered on any matter to which the evidence

20240257-CA 24 2026 UT App 83
State v. Shay

is relevant, including to prove a propensity to commit the crime
charged.”). 7

¶58 But the way to properly enact such a policy is to change the
text of the rule—as occurred with rule 404(c) and, more recently,
with rule 404(d)—and not to continue to interpret the existing rule
(which still contains a propensity ban) in ways incompatible with
its text. And the fact that the court put this new regime into place
in an opinion professing to utilize a “plain text” formulation of
the rule makes the whole thing even stranger. In my view, the
court’s “plain text” analysis runs aground very quickly on the
second element of the three-part test for admissibility of prior-
bad-acts evidence—the one in which courts must analyze and
identify a purported proper non-character purpose for admission
of the prior-bad-acts evidence.

¶59 The purpose for which the State offered the prior-bad-acts
evidence in Green was for “rebutting” the defendant’s claims that
all six complaining witnesses were “fabricating” their allegations
of sexual assault. See 2023 UT 10, ¶¶ 66–67. The main problem
with the court’s “plain text” analysis is one it was candid enough
to acknowledge: this purpose—rebuttal of a defendant’s
fabrication defense—is not listed in the text of rule 404(b)(2). See
id. ¶ 70 (noting that “[t]he list of enumerated purposes . . . does

7. See also Joint Resolution Amending Rules of Evidence
Concerning Crimes or Other Acts, S.J.R. 001, 2026 Leg., Gen. Sess.
(Utah 2026). As noted again in note 11 below, this amendment—
whether one agrees with it or not—has the salutary benefit of
making our law in this area much more intellectually consistent.
And, at least as to sexual assault cases, this amendment moots
much of my criticism of Green as set forth herein. But the Green
regime is still in place as applied to this case—that is, no party
argues that the 2026 rule amendment applies to this case—and the
Green regime is still in place in cases that do not fall into either the
sexual-assault or child-molestation categories. See, e.g., State v.
Lane, 2019 UT App 86, 444 P.3d 553 (an example of a case in which
prior-bad-acts evidence was used in a non-sex-crime case).

20240257-CA 25 2026 UT App 83
State v. Shay

not include the rebuttal of fabrication”). To be sure, the rule does
contain a list of potential purposes for which prior-bad-acts
evidence might plausibly be offered, and this list is expressly non-
exhaustive. See Utah R. Evid. 404(b)(2) (stating that prior-bad-acts
evidence “may be admissible for” purposes other than proving
propensity, “such as proving motive, opportunity, intent,
preparation, plan, knowledge, identity, absence of mistake, or
lack of accident” (emphasis added)). While the rule’s language—
especially the words “such as”—provides textual support for the
general proposition that non-character purposes other than those
specifically listed in the rule might exist, it provides no textual
support for the specific proposition that “rebuttal of fabrication” is
properly considered one of them.

¶60 Against this backdrop, it is worthwhile to take a moment
to examine the Green court’s rationale for concluding that
“rebuttal of fabrication” constitutes a proper non-character
purpose. The court’s analysis on this point is, in the end, rather
spare and is contained entirely within one paragraph. The court
begins by stating that “the plain text of the rule, in conjunction
with the other rules of evidence, suggests that [prior-bad-acts]
evidence may be admissible to rebut a fabrication defense,” Green,
2023 UT 10, ¶ 70 (emphasis added); the court’s use of the verb
“suggests” is telling and itself suggests that the text of the rule
might not be quite as “plain” as the court describes. As
substantive support for its analysis, the court then proffers that
the rules of evidence are intended to support “an inclusionary
approach to admitting evidence under rule 404(b),” see id.—thus
quickly resolving, in a couple sentences and while citing only the
rules themselves, a complex historical debate about the reach of
rule 404(b), see Murphy, 2019 UT App 64, ¶ 48 n.16 (Harris, J.,
concurring) (describing the long debate over whether federal rule
404(b) is a rule of inclusion or a rule of exclusion) 8—and points

8. See also Dora W. Klein, “Rule of Inclusion” Confusion, 58 San
Diego L. Rev. 379, 415 (2021) (classifying as “substantively
misleading” the notion that “the whole of [r]ule 404(b) [is] ‘a rule
(continued…)

20240257-CA 26 2026 UT App 83
State v. Shay

out that “rule 404(b)’s use of ‘such as’ indicates that the list of non-
character purposes is illustrative and not exclusive,” see Green,
2023 UT 10, ¶ 70 (quotation simplified). But again, while these two
things do constitute support for the notion that purposes other
than the ones listed in rule 404(b)(2) might theoretically exist,
these things do not, in my view, constitute support—textual or
otherwise—for the specific proposition that “rebuttal of
fabrication” is one such purpose. To the contrary, they operate to
bring into sharp relief the absence of textual support for the
court’s specific conclusion.

¶61 To recap: the court’s “plain text” syllogism is as follows:

Major premise: Rule 404(b) is inclusionary.

Minor premise: Rule 404(b)(2)’s list of non-
character purposes is illustrative and not exclusive.

Ergo: “Rebuttal of fabrication” is a proper non-
character purpose.

Even as a non-plain-text argument, this syllogism is questionable.
It offers no limiting principles—textual or otherwise—on what
can be considered a proper non-character purpose under rule
404(b)(2). Indeed, this logic applies just as well to literally anything
(other than propensity—including the doctrine of chances) as it
does to “rebuttal of fabrication,” and it throws the door wide open
to consideration of any non-propensity label a creative lawyer can
think to put on prior-bad-acts evidence. Moreover, it does not for
a moment consider whether any new unenumerated non-
character purpose is likely to invite jurors to draw a propensity

of inclusion,’” and noting, for instance, the Ninth Circuit’s recent
attempt to more accurately characterize rule 404(b) by not
describing “the whole of [r]ule 404(b) as ‘a rule of inclusion’” but
by stating that the rule has an “exclusionary purpose” in
subsection (1) and an “inclusionary structure” in subsection (2)).

20240257-CA 27 2026 UT App 83
State v. Shay

inference, or the extent to which the new purpose might be at
odds with rule 404(b)(1)’s propensity ban.

¶62 But when the modifier “plain text” is placed on it, the
syllogism’s flaws become even more apparent. The Green court
identified no actual textual support—in any of the rules of
evidence, and certainly not in rule 404(b)—for the specific
proposition that “rebuttal of fabrication” is a proper non-
propensity purpose for admission of prior-bad-acts evidence.
And I submit that there is a very good reason that the text of the
rule contains no such support: because the drafters of the rule
likely understood that including “rebuttal of fabrication” in the
list of acceptable rule 404(b)(2) purposes would operate to
effectively eliminate the propensity ban that is rule 404(b)’s raison
d’être. After all, the plain text of the rule does quite plainly support
the concept of a ban on propensity evidence. See Utah R. Evid.
404(b)(1) (“Evidence of a crime, wrong, or other act is not
admissible to prove a person’s character in order to show that on
a particular occasion the person acted in conformity with the
character.”). And our supreme court has long observed that rule
404(b) contains a propensity ban. See, e.g., State v. Lucero, 2014 UT
15, ¶ 14, 328 P.3d 841 (cautioning that, under rule 404(b),
“evidence may be admitted despite its negative propensity
inference, but if such evidence is really aimed at establishing a
defendant’s propensity to commit crime, it should be excluded”
(quotation simplified)), abrogated on other grounds by State v.
Thornton, 2017 UT 9, 391 P.3d 1016; State v. Allen, 2005 UT 11, ¶ 17,
108 P.3d 730 (stating that evidence offered under rule 404(b) must
be “offered for a legitimate purpose other than to show the
defendant’s propensity to commit the crime charged”). Even the
court in Green professed continuing fealty to this principle. See
2023 UT 10, ¶ 63 (noting “the propensity purpose that the rule
deems improper” (quotation simplified)).

¶63 But here’s the rub: you can’t have anything approaching a
functional ban on propensity evidence if you let prior-bad-acts
evidence come in for the purpose of rebutting a fabrication
defense. In Green, the court noted that the prior-bad-acts evidence

20240257-CA 28 2026 UT App 83
State v. Shay

was “highly probative of the primary issue at trial—whether the
complaining witness is fabricating her allegation of sexual
assault.” Id. ¶ 67. But of course the prior-bad-acts evidence is
probative of this: if five other witnesses take the witness stand,
one after another, and swear, under oath, that the defendant
sexually assaulted them in similar ways, jurors are going to be
much less likely to believe that the complaining witness is lying
than they would have been if they had only heard the
complaining witness testify. We must, however, ask why this is so.

¶64 The reason jurors are less likely to believe a fabrication
defense after hearing six witnesses testify similarly is that, as a
result of hearing that testimony, they are more likely to believe that
the defendant is the type of person who would commit this sort
of crime. They disbelieve the fabrication defense precisely because
they come to espouse a propensity theory. See Murphy, 2019 UT
App 64, ¶¶ 57–59 (Harris, J., concurring) (quoting a law review
article (citation omitted) that observed that “each separate
accusation would have no bearing upon the accuracy of another
allegation but for the conclusion that the multiple accusations
demonstrate a cross-situational pattern of behavior, which is but
a variation on the taboo inference of a general propensity or
character trait”); see also 23 Stephen A. Hess, Colorado Practice
§ 413:3 (2025) (discussing a Colorado statute that allows prior-
bad-acts evidence to be admitted, in sex crimes cases, for the
purpose of rebutting a fabrication defense, and stating that this
asserted “‘non-propensity’ justification has a tendency to swallow
the [propensity ban] completely” because, after presentation of
prior-bad-acts evidence for this purpose, “one can hardly avoid
reasoning directly from the defendant’s propensity to commit
such crimes: the victim is telling the truth because the defendant
has done this sort of thing before”). Rebuttal of a fabrication
defense is thus nothing more—and, crucially, nothing less—than
the reverse side of the same logical coin as propensity evidence.
A litigant who is offering prior-bad-acts evidence for the
avowed purpose of rebutting a fabrication defense is, in most
cases, simply offering propensity evidence cloaked under a
different label.

20240257-CA 29 2026 UT App 83
State v. Shay

¶65 Our supreme court’s conclusion in Green—that rebuttal of
fabrication is a proper non-character purpose under rule
404(b)(2)—thus represents, for all practical purposes and in the
absence of a robust rule 403 inquiry, see infra Part II, the end of the
propensity ban in Utah. In the vast majority of criminal cases
(including this one), defendants defend cases, at least in part, by
asserting that the complaining witness should not be believed
because he or she is lying or is at least mistaken. Only the most
uncreative prosecutors will be unable to spin heretofore-
inadmissible prior-bad-acts evidence as information that will, at
least to some extent, rebut a claim that the complaining witness
should not be believed. And district courts, with the Green opinion
in hand, will be compelled to conclude that—despite the obvious
propensity inference jurors are likely to draw—the evidence has
a proper non-character reason for admission.

¶66 In short, the court’s conclusion in Green—that rebuttal of
fabrication is a proper non-character purpose under rule
404(b)(2)—is not specifically supported by anything in the text of
our rules of evidence, and in fact operates to eviscerate the main
purpose of rule 404(b), the ban on propensity evidence (which is
apparent in the rule’s text). As I see it, the court in Green, applying
an analysis consistent with the text of the rule, should have
concluded that rebuttal of fabrication is not a proper non-
propensity purpose under rule 404(b).

C. Context

¶67 The third potential purpose to be discussed here is
“context.” The State makes no argument that “context” is a proper
non-character purpose that would satisfy the second element of the
relevant test. See State v. Gallegos, 2020 UT App 162, ¶ 15, 479 P.3d
631 (noting that prior-bad-acts evidence “is admissible if it (1) is
relevant to, (2) a proper, non-character purpose, and (3) does not
pose a danger for unfair prejudice that substantially outweighs its
probative value” (quotation simplified)). To be sure, the State in
its brief does mention “context,” asserting that “Ashley’s
testimony was essential context for explaining [Beth’s] and

20240257-CA 30 2026 UT App 83
State v. Shay

[Clara’s] significant delay in reporting the abuse.” But the State
makes this argument as part of its rule 403 analysis—the third
element of the relevant test. I discuss this argument, in that
context, below. See infra ¶¶ 88–90.

¶68 But the lead opinion uproots the State’s rule 403 argument
about “context” and transplants it into entirely different soil,
suggesting that “providing context” can function as a proper non-
character purpose for admitting prior-bad-acts evidence. See supra
note 5. In so doing, the lead opinion acknowledges that
“providing context” was “not a purpose highlighted by the trial
court.” Supra note 5. And as noted, not even the State asks us to
take this step in this case.

¶69 There are several procedural problems with the lead
opinion’s suggestion. First and foremost, we should be extremely
cautious about deciding a case on the basis of an argument not
discussed by the trial court and not advanced or briefed by any
party. See State v. Johnson, 2017 UT 76, ¶ 40, 416 P.3d 443
(“Any time a judge raises an otherwise overlooked or unargued
issue, the judge arguably undertakes an advocacy role to some
extent, as it is the parties’ duties to raise and argue the issues.”).
There are exceptions to the general rule that, “if a party has
not raised an issue on appeal, an appellate court may not
consider the issue sua sponte.” Id. ¶ 48 (quotation simplified). But
the lead opinion does not explain whether, or how, any of those
exceptions are applicable here. See id. ¶¶ 49–52 (discussing
exceptions).

¶70 Second, the trial court did not consider the possibility that
“providing context” might be a potential non-character purpose
in this case, and therefore it did not include discussion of that
potential purpose in the instruction it gave to the jury about
Ashley’s testimony. See supra ¶ 20 (quoting the jury instruction).
Stated another way, the jury had no idea that Ashley’s testimony
could be, or was being, admitted for that purpose and, thus, was
unlikely to have so considered it. A jury instruction discussing
this purpose may have actually been helpful, because jurors are

20240257-CA 31 2026 UT App 83
State v. Shay

much more likely to be able to conceptually distinguish between
these two purposes (considering the evidence only for purposes
of explaining delayed disclosure, on the one hand, and
considering the evidence for propensity, on the other) than they
are to be able to distinguish between propensity and rebuttal-of-
fabrication. See infra ¶¶ 79, 81.

¶71 And third, because the trial court did not consider the
“context” argument, it had no opportunity to undertake the
analysis required by our supreme court in Verde, and assess
whether the evidence at issue “is genuinely being offered for a
proper, non-character purpose, or whether it might actually be
aimed at sustaining an improper inference of action in conformity
with a person’s bad character.” See 2012 UT 60, ¶ 18. And in this
same vein, as discussed below in connection with the rule 403
analysis, the trial court had no opportunity to consider whether
Ashley’s testimony—if truly admitted for “context” purposes and
not for propensity purposes—might warrant being limited (for
instance, by permitting Ashley to state simply that Mother did not
believe her when she accused Shay of improper behavior, but not
allowing testimony detailing the actual behavior).

¶72 I readily concede that, in some cases, prior-bad-acts
evidence may be admissible for some contextual purposes. See
State v. Thornton, 2017 UT 9, ¶ 57, 391 P.3d 1016. Inquiries of this
sort will inevitably be highly factual in nature and will depend on
the circumstances of each case. In this case, the trial court did not
identify “context” as a potential non-character purpose for
admission of Ashley’s testimony, and given the procedural
problems I identify, I do not think we can, on this record, affirm
the trial court’s analysis concerning the second step of the test—
articulating a proper non-character purpose—on the alternative
basis that the lead opinion sua sponte suggests.

* * * *

¶73 Accordingly, I do not think we can affirm the trial court’s
determination that there exists a proper non-character purpose for

20240257-CA 32 2026 UT App 83
State v. Shay

Ashley’s testimony in this case. The trial court erred by
concluding that this testimony could come in to prove Shay’s
intent. Rebuttal of fabrication can—under Green, even as much as
I don’t like it—constitute a proper non-character purpose, and so
can context, but the State did not make those arguments at the
trial court level and the trial court therefore did not have an
opportunity to exercise its discretion regarding those arguments
and, for that reason, affirmance (as opposed to remand) on those
alternative grounds is inappropriate.

II. Rule 403 Balancing

¶74 If there were a proper non-character purpose for admitting
Ashley’s testimony in this case, the next step would be to assess
whether the legitimate probative value of that testimony is
substantially outweighed by the risk of unfair prejudice. For the
reasons discussed, I think the risk of unfair prejudice posed by the
entirety of Ashley’s testimony—the risk that the jury would draw
the forbidden propensity inference—substantially outweighs the
legitimate probative value of that evidence.

¶75 My criticism of Green is not limited to its analysis of
whether rebuttal of fabrication constitutes a proper non-character
purpose for admission of prior-bad-acts evidence: I think Green
made errors in how it applied rule 403 as well.

¶76 A proper rule 403 analysis, in this context—where fear of a
forbidden propensity inference is present—consists of weighing
the prior-bad-acts evidence’s “valid non-character purpose on the
probative value side of the ledger,” and weighing “the evidence’s
value as propensity evidence on the prejudice side of the ledger.”
State v. Gallegos, 2020 UT App 162, ¶ 43, 479 P.3d 631 (quotation
simplified). Indeed, in its last major rule 404(b) case before Green,
our supreme court laid it out exactly that way:

The district court needed to identify the likely
inferences the jury would draw from the other-acts
evidence and then ask if the evidence’s probative

20240257-CA 33 2026 UT App 83
State v. Shay

value (the jury drawing a permissible inference) was
substantially outweighed by the danger of unfair
prejudice (the jury drawing an impermissible
inference). If the district court were to conclude that
the jury is substantially more likely to rely on an
impermissible inference, the evidence must be
excluded under rule 403.

State v. Richins, 2021 UT 50, ¶ 103, 496 P.3d 158.

¶77 Even assuming, for purposes of this part of the discussion,
that “intent” or “rebuttal of fabrication” or “context” could
somehow constitute a proper non-propensity purpose for
admission of the entirety of Ashley’s testimony in this case, a
proper rule 403 analysis involves comparing the value of that
evidence for the proper purpose—here, intent or rebuttal of
fabrication or context—with the danger that the jury would draw
an impermissible inference—here, propensity. See id. ¶ 104
(determining, in that case, that “[t]he risk of the jury making a
character-based inference substantially outweigh[ed] the
probative value of the other-acts evidence”).

¶78 In perhaps the most puzzling passage of its opinion in
Green, our supreme court refused to apply the rule 403 analysis
the way it had in Richins, stating that its “review in [Richins had
been] conducted under the doctrine-of-chances framework” and
was “therefore less relevant” because that doctrine had been
abolished (in Green itself). See Green, 2023 UT 10, ¶ 75. But the
court did not explain why its rule 403 analysis—presumably,
weighing the permissible inference against the impermissible
one—would be any different in a doctrine-of-chances case than in
a non-doctrine-of-chances case. After all, one of the commonly
claimed purposes for admitting evidence under the doctrine of
chances was for rebuttal of fabrication. See, e.g., Verde, 2012 UT 60,
¶¶ 44–62; Murphy, 2019 UT App 64, ¶¶ 56–59 (Harris, J.,
concurring). So the rule 403 balancing test—comparing the
probative value of the evidence submitted for “rebuttal of
fabrication” against the potential risk that the jury would draw an

20240257-CA 34 2026 UT App 83
State v. Shay

impermissible propensity inference—should be exactly the same
in the doctrine-of-chances context as it is outside that context. Yet
the Green court—on its own account—did not engage in any
comparison of the relative value and prejudice risk of the two
competing purposes; instead, it merely noted, almost in passing,
that the trial court had done such a comparison and had
determined that “the danger of unfair prejudice [was] slight.”
2023 UT 10, ¶ 76.

¶79 In my view, the court should have applied the same rule
403 analysis it applied in Richins, and had it done so, it likely
would have reached the same result it reached there: that the
evidence’s low legitimate probative value was substantially
outweighed by the risk that the jury would draw the forbidden
propensity inference. Even law-trained judges and attorneys, let
alone lay jurors, have the devil’s own time sussing out any
distinction between evidence coming in to rebut fabrication and
evidence coming in for a propensity purpose. I submit that this is
because the two purposes functionally and logically amount to
the same thing. Just as in Richins, in the Green case there was, in
my view, an extremely high risk that the jurors would draw an
improper propensity inference after hearing all six witnesses
testify. And for the reasons already discussed, the evidence had
minimal permissible probative value, because the bulk of its
persuasive value—which is admittedly a lot—comes because the
jury was likely to draw the impermissible propensity inference.
Again, jurors tend to disbelieve a defendant’s fabrication defense
precisely because, after hearing multiple witnesses tell a similar
story, they tend to draw a propensity inference. Just as in Richins,
“[t]he risk of the jury making a character-based inference” should
have been deemed to “substantially outweigh[] the [permissible]
probative value of the other-acts evidence.” See 2021 UT 50, ¶ 104.

¶80 Moreover, in support of its rule 403 analysis in Green, the
court noted that the stories told by the six witnesses were
strikingly similar, and it credited the trial court’s reasoning that,
“given the similarities among the women’s accounts, . . . it was
unlikely that a jury would find the evidence in one woman’s case

20240257-CA 35 2026 UT App 83
State v. Shay

to be lacking but find the evidence in another woman’s case
compelling enough to deliver a verdict on an improper basis.” See
2023 UT 10, ¶ 77. As a threshold matter, I must confess that I am
not sure I can follow that reasoning; at a minimum, I cannot
discern how that reasoning does not simply collapse into an
impermissible propensity inference. But more to the point here,
the trial court in Green—prior to the issuance of the supreme
court’s Green opinion, and at a time when the doctrine of chances
was robust—was considering the “similarity” of the six charged
crimes only because that was one of the four foundational
elements of the now-abolished doctrine-of-chances analysis; our
supreme court, in Green, did not explain why it rejected Richins’s
rule 403 framework because it was rooted in that now-abolished
doctrine, but relied on the trial court’s similarity analysis that was
rooted in that same doctrine. In reality, the similarity inquiry once
required by the court’s doctrine-of-chances jurisprudence is now
completely vestigial—not to mention that it is, and always was,
devoid of support in the text of rule 404(b)—whereas a rule 403
analysis is still required; the court should have therefore adopted
Richins’s rule 403 framework and discounted the trial court’s
similarity analysis, not done the opposite. And in addition,
viewing the similarity of the witnesses’ accounts as something that
weighs in favor of admissibility under a proper rule 403 analysis
is, in the propensity-ban context (although not in the rule 404(c)
and (d) context 9), precisely the wrong way to look at it. The more

9. The situation is, of course, different when the prior-bad-acts
evidence is properly admitted to prove propensity, as occurs in
child molestation and sexual assault cases. In those situations,
courts examine the similarities and differences between the prior-
bad-acts evidence and the charged crime, and they generally
conclude that when the similarities are strong and the differences
slight, the evidence has higher probative value and lower risk of
unfair prejudice. See, e.g., State v. Alvarado-Rodriguez, 2026 UT App
25, ¶¶ 21–22, 587 P.3d 47, petition for cert. filed, Apr. 17, 2026 (No.
20260455). But this is because, in rule 404(c) and (d) cases, the
prior-bad-acts evidence is properly admitted to prove propensity.
(continued…)

20240257-CA 36 2026 UT App 83
State v. Shay

similar the prior bad acts are to the charged crime, the more
likely—not less likely—jurors are to draw a forbidden propensity
inference and conclude that the defendant is just the sort of person
to commit precisely this sort of crime. See State v. Modes, 2020 UT
App 136, ¶ 18, 475 P.3d 153 (“[T]he more similar a previous act is
to the act the defendant is accused of committing, the more
commission of a previous act might suggest propensity.”).

¶81 Finally, our supreme court in Green took some comfort—
for purposes of its rule 403 analysis—in a “limiting instruction”
the trial court gave to the jury. See 2023 UT 10, ¶ 80. But that
instruction was, in my view, nowhere near as helpful as the court
seemed to assume. It told jurors that they could consider the prior-
bad-acts evidence, if at all, “for the limited purpose of
determining” whether the complaining witnesses “fabricated
[their] accusation[s],” but that they could not consider the
evidence “to show that [the defendant] has a general criminal
propensity, or to prove a character trait of the defendant, or to
show that he acted in a manner consistent with such a trait.” Id.
¶¶ 80, 91, 93–95. But as noted above, I am not at all convinced that

Stated another way, in rule 404(c) and (d) cases, propensity is a
proper purpose for admission of the evidence, and it therefore
weighs on the probative side of the rule 403 ledger. See State v.
Fredrick, 2019 UT App 152, ¶ 45, 450 P.3d 1154. But in rule 404(b)
cases, where propensity is not a proper purpose, “the evidence’s
value as propensity evidence should be weighed on the
‘prejudice’ side of the ledger.” Id. On at least one occasion, we
have noted the confusing role of “similarity” in a rule 403 analysis
in the rule 404(b) context. See State v. Simpson, 2025 UT App 32,
¶ 44, 566 P.3d 756 (“Utah’s courts have considered similarity in
relation to both the probative value prong of rule 403 and its
unfair prejudice prong.”), cert. denied, 578 P.3d 748 (Utah 2025).
But in my view, when conducting a rule 403 analysis in the rule
404(b) context, strong similarities between the prior-bad-acts
evidence and the charged crime—which lead inexorably toward
a propensity inference—should not be viewed as positively for
admission purposes as they were by the court in Green.

20240257-CA 37 2026 UT App 83
State v. Shay

even a law-trained human brain is capable of distinguishing
between these two purposes: they are essentially two sides of the
same coin. Our supreme court, in Richins, made this same point.
See 2021 UT 50, ¶ 53 (acknowledging the “substantial degree of
mental discipline” required of a jury when it is asked “to consider
a defendant’s past acts to assess whether his [or her] accuser is
making up the allegations, but to simultaneously not consider
whether the fact that the defendant has committed the prior acts
means he [or she] has a propensity to commit those crimes”). Yet
in Green, without really explaining why, the court changed its
tune and now apparently views this type of confusing instruction
as somehow comforting. I see no reason for this change of
direction; like the Richins court, I view such instructions, when
offered in the rebuttal-of-fabrication context, as extremely
unlikely to materially reduce the risk that a lay juror will draw an
impermissible propensity inference from prior-bad-acts evidence.

¶82 In short, the risk of unfair prejudice—from jurors drawing
a still-impermissible propensity inference—was extremely high in
Green, as it will be in any case in which prior-bad-acts evidence is
offered for the purpose of rebutting a fabrication defense. Jurors
were all but certain to draw a propensity inference after hearing
multiple witnesses give similar accounts on the witness stand. To
the extent that prior-bad-acts evidence helps rebut a defendant’s
fabrication defense, this likely occurs precisely because the jurors
come to espouse the forbidden propensity inference, and not for
any non-propensity reason. Because the risk of jurors drawing an
impermissible inference substantially outweighs any permissible
probative value that the prior-bad-acts evidence might have, a
rule 403 analysis, correctly undertaken, should have come out in
Green in favor of exclusion of the prior-bad-acts evidence.

* * * *

¶83 After Green, then, it is unclear to me what a rule 403
analysis in a rule 404(b) prior-bad-acts case is supposed to look
like. Is it supposed to look like what the court laid out in Richins,
where the evidence’s legitimate probative value is weighed against

20240257-CA 38 2026 UT App 83
State v. Shay

the risk of the jury drawing an illegitimate propensity inference?
Or is it supposed to look like what the court laid out in Green,
which is decidedly less robust? Indeed, under Green, proponents
of the admission of prior-bad-acts evidence can now apparently
meet their rule 403 burden by simply demonstrating that the prior
bad acts are similar to the charged crime and by persuading the
trial judge to give an incomprehensible limiting instruction.

¶84 As applied to this case, if Green sets the rule 403 standard,
then the trial court did not abuse its discretion in determining that
Ashley’s testimony passed rule 403 muster. The events Ashley
described were similar to the events Beth and Clara described,
and the court did instruct jurors that they could consider Ashley’s
testimony “for the limited purposes of [Shay’s] intent, knowledge,
and common scheme,” but that they could not vote to “convict
[Shay] simply because [they] believe[d] he may have committed
some other act at another time.” This was good enough under
Green, and I am bound to follow Green, so I suppose that must be
good enough here.

¶85 But it shouldn’t be. Under what I consider to be a proper
rule 403 analysis—the one our supreme court outlined in
Richins—Ashley’s testimony does not pass rule 403 muster,
regardless of whether the proper non-character purpose is
considered to be intent or rebuttal of fabrication.

¶86 Regarding intent, the evidence’s probative value is very
limited because intent was never contested at trial. Our supreme
court made this exact point in Verde, when it held as follows:

[E]ven if the past misconduct evidence in this case
could plausibly be deemed to have been aimed at a
legitimate purpose under rule 404(b), it would still
fail under the balancing framework required under
rule 403. Specifically, and for all the reasons detailed
above, we conclude that any legitimate tendency the
[rule] 404(b) evidence had to tell a narrative of [the
defendant’s] specific intent was minimal at best.

20240257-CA 39 2026 UT App 83
State v. Shay

And we likewise conclude that any such legitimate
purpose is far outweighed by the obvious,
illegitimate one of suggesting action in conformity
with bad character.

2012 UT 60, ¶ 31. And we made that same point in Gallegos. See
2020 UT App 162, ¶¶ 44–46.

¶87 Regarding rebuttal of fabrication, as I’ve already noted, the
risk of unfair prejudice is extremely high because all humans—
especially non-law-trained jurors—have a difficult time
ascertaining any difference between considering evidence for
rebuttal of fabrication and considering evidence for propensity.
Our supreme court understood this very well in Richins, when it
concluded that prior-bad-acts evidence proffered for rebuttal-of-
fabrication purposes did not pass muster under rule 403. In that
case, the court articulated its reasoning like this:

Under [a rebuttal of fabrication] articulation, there
is a little, but only a little, more room between the
permissible and impermissible inference. The non-
propensity-based inference is that because [the
defendant] had been accused of similar behavior on
four prior occasions, it is unlikely that [the
complaining witness] fabricated a story that closely
matched the other accusations. The impermissible
inference is the same as before—because [the
defendant] did this type of thing before, he did it
this time. But even in this posture, the risk of the jury
resorting to the impermissible inference
overwhelms the possibility that the jury will confine
itself to focusing on the probability of fabrication.
Stated differently, we have no confidence that,
having [been] told . . . that [the defendant] engaged
in this behavior previously, the jury would do
anything other than indulge the inference rule
404(b) exists to prevent.

20240257-CA 40 2026 UT App 83
State v. Shay

Richins, 2021 UT 50, ¶ 105. To be sure, the court’s rule 403 analysis
in Richins was undertaken in the doctrine-of-chances context, and
that doctrine has now been abolished. But the proffered non-
character purpose in Richins was rebuttal of fabrication, and (as
I’ve already noted) the rule 403 analysis should have been exactly
the same in Green as it was in Richins, and it should be the same
here. We are assessing the evidence’s probative value for
demonstrating a rebuttal of fabrication, and we are weighing that
probative value against the risk that the jury might draw the
forbidden propensity inference. When done correctly, in Richins,
the result was exclusion under rule 403. And that same result
should obtain here, as regards rebuttal of fabrication.

¶88 Finally, regarding the lead opinion’s suggestion that
“context” could have provided a proper non-character purpose,
the rule 403 analysis is somewhat more nuanced, and for that
reason should be conducted by the trial court in the first instance.
As the State suggests, there is appreciable probative value in
Ashley’s testimony as “context for explaining [Beth’s] and
[Clara’s] significant delay in reporting the abuse.” The State is
entitled to present evidence—which, at the State’s election, in
many cases includes a blind expert witness—to discuss reasons
for delayed disclosure, and the fact that Mother didn’t initially
believe Ashley’s allegations is a fact that does serve to explain the
girls’ delayed disclosure.

¶89 But a proper rule 403 balancing assessment should include
consideration of potential limitations on Ashley’s testimony,
designed to ensure that Ashley’s testimony comes in only for the
proper purpose and not for the forbidden propensity purpose. For
instance, if Ashley’s testimony were coming in only for this
limited “context” purpose, it would likely suffice for the jury to
learn that Ashley made unspecified allegations of improper
touching against Shay and that Mother didn’t initially believe
Ashley. Many of the additional details contained in Ashley’s
testimony—such as the number of and details of the events in
question, and whether Shay eventually pled guilty to any crime—

20240257-CA 41 2026 UT App 83
State v. Shay

are far less relevant to the purported proper purpose and are far
more likely to trigger a propensity inference among jurors.

¶90 In summary, I do not think the rule 403 balancing test—
properly conducted—comes out in the State’s favor regarding the
intent or rebuttal-of-fabrication purposes. And because “context”
was never suggested to the trial court as a potential proper
purpose, the trial court never got an opportunity to weigh the
probative value of that evidence against the potential risk of
unfair prejudice from an improper propensity inference.

III. Harmless Error

¶91 Despite my serious concerns with both Utah’s rule 404(b)
jurisprudence and the trial court’s analysis in this case, I
nevertheless believe that Shay’s convictions should be affirmed,
because even assuming that admission of Ashley’s testimony
was erroneous, any error was harmless on the specific facts of this
case.

¶92 “Prejudice analysis is counterfactual. To decide whether a
trial affected by error is reasonably likely to have turned out
differently we have to consider a hypothetical—an alternative
universe in which the trial went off without the error.” State v.
Ellis, 2018 UT 2, ¶ 42, 417 P.3d 86; see also State v. Cesspooch, 2024
UT App 15, ¶ 14, 544 P.3d 1046 (stating that, in this context, we
must “ask[] whether we remain confident that the verdict would
be the same had the improper information been excluded”
(quotation simplified)); State v. Soto, 2022 UT App 107, ¶ 25, 518
P.3d 157 (“Under a counterfactual analysis, we consider whether,
in the absence of the improperly admitted evidence, the likelihood
of a different outcome is sufficiently high to undermine our
confidence in the verdict.” (quotation simplified)).

¶93 The lead opinion ably explains why any error in admitting
Ashley’s testimony was harmless on the facts of this case. Beth
and Clara offered testimony that was detailed, consistent, and

20240257-CA 42 2026 UT App 83
State v. Shay

“mutually corroborative.” 10 Supra ¶ 34. And their accounts were
also at least partially corroborated by Mother, who witnessed
Shay massaging Beth in ways she felt might be inappropriate, and
who (on one occasion) caught Shay leaving Clara’s room during
the night and (on another occasion) trying to enter the girls’ room
during the night. These facts represent strong evidence in support
of conviction.

¶94 In addition, Shay didn’t offer much of a defense. He chose
to represent himself at trial (assisted by standby counsel), and he
presented no opening statement, called no witnesses, and opted
not to cross-examine Mother or either of the complaining
witnesses. Thus, the counterfactual trial that we must envision is
one in which Ashley does not testify but in which Shay again
represents himself and again chooses not to give an opening
statement, call witnesses, or cross-examine Mother or the two
complaining witnesses. There is simply no reasonable likelihood
that, in this counterfactual trial, Shay would enjoy a different
outcome than the one he received the first time.

CONCLUSION

¶95 In Green, our supreme court abolished the doctrine of
chances. And that’s a good thing, from my point of view. But in
many ways, Green leaves our law no better off, and no more
textually consistent, than it was before. We still have rules of
evidence—and case law—professing allegiance to the historical
ban on propensity evidence. But those rules mean even less than
they used to, because after Green it is arguably even easier for
litigants—often the prosecution in criminal cases—to obtain

10. Shay does not argue that there should have been separate trials
regarding Beth and Clara, and he makes no argument that the
same jury should have been prevented from hearing both girls’
accounts.

20240257-CA 43 2026 UT App 83
State v. Shay

admission of prior-bad-acts evidence despite the rules’
presumptive ban on propensity evidence.

¶96 Under our supreme court’s prior precedents, “[a] charge of
fabrication [was] insufficient by itself to open the door to evidence
of any and all prior bad acts.” See Verde, 2012 UT 60, ¶ 55. That
appears to no longer be the case, at least functionally. Now, a
prosecutor’s effort to rebut a defendant’s fabrication defense is
considered to be a proper non-propensity reason for admission of
prior-bad-acts evidence. And, judging by the analysis in Green,
prosecutors can now apparently meet their rule 403 burden by
simply demonstrating that the prior bad acts are similar to the
charged crime and by persuading the trial judge to give a
confusing limiting instruction.

¶97 It is therefore hard to read Green as anything other than the
functional elimination of the ban on propensity evidence in Utah,
in all factual contexts. Depending on one’s point of view, this may
or may not be a positive development. But in any event, we thus
find ourselves in more or less the same place we were in during
the doctrine-of-chances regime, when I lamented that “we
routinely allow character evidence to reach the jury while
maintaining the pious fiction that we follow the character
evidence rule.” See Murphy, 2019 UT App 64, ¶ 65 (Harris, J.,
concurring) (quotation simplified). In my view, there needs to be
either (a) an amendment to the rules that expands admissibility of
propensity evidence to all contexts (not just child molestation and
sexual assault cases) or (b) a reconsideration of Green (without a
resurrection of the doctrine of chances) to functionally restore the
propensity ban (at least in non-child-molestation and non-sexual-
assault contexts). 11 As I said in Murphy, the question of which one

11. As noted above in note 7, the 2026 amendment to rule 404—
regardless of what one might think of its merits—is a meaningful
step in the direction of intellectual consistency. But the problems
I identify here will still exist, going forward, in all cases except
child molestation cases and sexual assault cases. And that group
of cases is hardly a null set.

20240257-CA 44 2026 UT App 83
State v. Shay

of these “divergent pathways” to adopt is one that remains
“above my pay grade.” Id. It took us more than a decade to rid
ourselves of the doctrine of chances. I hope it doesn’t take us
another decade to fix, in one way or another, this latest misstep.

¶98 Despite my reservations about Green, I am bound to follow
it. In my view, it does not compel affirmance in this case, because
the trial court was not asked to weigh in on the rebuttal-of-
fabrication theory (or on the “context” theory) and was not given
an opportunity to exercise its discretion in deciding whether to
admit the contested evidence on those grounds. I am, however,
able to vote with my colleagues to affirm Shay’s convictions, but
I do so on harmless error grounds, because I believe that in a
hypothetical counterfactual trial that did not contain Ashley’s
testimony, there is no reasonable probability of a different
outcome for Shay.

20240257-CA 45 2026 UT App 83