State of Iowa v. Alexander Shantee Thomas Ross

IN THE SUPREME COURT OF IOWA

No. 20–0914

Submitted December 15, 2022—Filed February 24, 2023

STATE OF IOWA,

Appellee,

vs.

ALEXANDER SHANTEE THOMAS ROSS,

Appellant.

On review from the Iowa Court of Appeals.

Appeal from the Iowa District Court for Madison County, Martha L. Mertz,

Judge.

The State seeks further review of the court of appeals decision reversing

the defendant’s convictions for sexual abuse in the second degree based on

erroneous noncorroboration jury instructions. DECISION OF COURT OF

APPEALS AFFIRMED; DISTRICT COURT JUDGMENT VACATED AND CASE

REMANDED.

Oxley, J., delivered the opinion of the court, in which McDonald,

McDermott, and May, JJ., joined. Mansfield, J., filed a dissenting opinion, in

which Christensen, C.J., and Waterman, J., joined.

John C. Heinicke of Kragnes & Associates, P.C., Des Moines, for appellant.

Brenna Bird, Attorney General, and Thomas J. Ogden, Assistant Attorney

General, for appellee.
 2

OXLEY, Justice.

In State v. Kraai, 969 N.W.2d 487, 490–92 (Iowa 2022), and State v.

Mathis, 971 N.W.2d 514, 516, 520 (Iowa 2022), we held that it is error in sexual

abuse cases for a trial court to instruct jurors that “[t]here is no requirement

that the [alleged victim’s testimony] be corroborated.” Although a correct

statement of law, such an instruction, without more, unduly emphasizes the

alleged victim’s testimony. Kraai, 969 N.W.2d at 491–95. The district court in

this case used a variation of the noncorroboration instruction that had been

percolating through our court of appeals on a parallel course to the Kraai

instruction:

You should evaluate the testimony of [the alleged victim] the
same way you evaluate the testimony of any other witness. The
law does not require that the testimony of [the alleged victims] be
corroborated in order to prove that [they were] sexually abused.
You may find the Defendant guilty of Sexual Abuse if [the alleged
victim’s] testimony convinces you of guilt beyond a reasonable
doubt.

We must now decide whether the additions to the instruction cure the errors we

found in the instruction at issue in Kraai and Mathis. On our careful review of

the instructions provided in this case, we conclude they do not.

I. Factual and Procedural History.

Alexander Ross and Tykeshia McCuen began dating in 2011. Eventually,

the two moved in together in Patterson, Iowa. Tykeshia and Ross have two

children together. Tykeshia also has two daughters from a previous

relationship—L.C. and K.C.—who were babies when Ross moved in with
 3

Tykeshia. From the time Ross and Tykeshia began dating until Tykeshia left in

April 2019, Ross acted as the girls’ stepfather.

The couple had a tumultuous and physically abusive relationship. On

April 11, they got into a heated argument, which left Tykeshia afraid for her

safety. So, in the middle of the night, she left Ross and took the children to

Westminster, Colorado to stay with her mother, Jackie. According to Tykeshia,

it was at this point that L.C. and K.C. first accused Ross of having sexually

abused them. Jackie and Tykeshia reported the allegations to Deputy Don

Kinney of the Madison County, Iowa Sheriff’s Office, who began an investigation.

Ross was charged with two counts of second-degree sexual abuse. See

Iowa Code § 709.3(1)(b) (2019). At trial in March 2020, the State’s primary

evidence against Ross was the testimony of L.C. and K.C.; no physical evidence

was presented to corroborate the charges. Each testified that Ross put his private

part in her private parts (both front and back) on multiple occasions when

Tykeshia was working night shifts as a nurse. Each testified she told her mother,

who did nothing. Each also testified to screaming, but Ross put his hand over

her mouth. In addition, L.C., who was in fourth grade, testified that Ross said

inappropriate things and described “white stuff” that came out of Ross’s penis

as “gooey” and “disgusting.” K.C., who was in second grade and is diabetic,

testified that Ross gave her suckers afterward. Neither girl testified about Ross’s

abuse of the other. The defense’s only witnesses were: Tykeshia, who testified

that the girls never made her aware of any sexual abuse prior to the April 2019
 4

trip to Colorado and that she had never herself noticed signs of such abuse, and

a friend of Ross’s, who testified Ross was a good stepparent to L.C. and K.C.

At the close of trial and over Ross’s objection,1 the court gave jurors

Instructions 16 and 17, which were identical other than identifying each girl

separately:

You should evaluate the testimony of [L.C. or K.C.] the same
way you evaluate the testimony of any other witness. The law does
not require that the testimony of [L.C. or K.C.] be corroborated in
order to prove that she was sexually abused. You may find the
Defendant guilty of Sexual Abuse if [L.C.’s or K.C.’s] testimony
convinces you of guilt beyond a reasonable doubt.

The jury deliberated for a little over ninety minutes and returned a guilty verdict

on both counts.

Ross was sentenced to two indeterminate, consecutive twenty-five-year

sentences (with a combined mandatory minimum sentence of thirty-five years),

along with a lifetime special sentence under Iowa Code § 903B.1. Ross appealed,

alleging three points of error: (1) insufficient evidence supported the verdict;

(2) the district court considered an improper sentencing factor; and (3) the

noncorroboration instructions were improper.

We transferred the case to the court of appeals. The court rejected Ross’s

sufficiency challenge but agreed, over a dissent, that the noncorroboration

instructions were improper based on our recent decisions in Kraai and Mathis.

In her dissent, Judge Greer noted that the noncorroboration instructions used

1Ross’s counsel initially asked the court to remove the third sentence of the proposed
instructions as “lead[ing] the jury to put more emphasis on one witness’s testimony than it does
on the other witness’s testimony.” He then objected to giving the instructions at all but asked
that the last sentence be removed if they were included.
 5

here were virtually identical to the noncorroboration instruction approved by an

earlier court of appeals panel in State v. Altmayer, No. 18–0314,

2019 WL 476488 (Iowa Ct. App. Feb. 6, 2019). In Kraai, we cited Altmayer in a

manner that left Judge Greer uncertain whether the noncorroboration

instruction at issue in Altmayer would stand up to post-Kraai scrutiny.2

We granted the State’s application for further review to decide that issue.

We elect to let the court of appeals’ opinion stand on the sufficiency of the

evidence challenge. See Farnsworth v. State, 982 N.W.2d 128, 135 (Iowa 2022)

(“When we grant further review, we may exercise our discretion to let the court

of appeals decision stand as the final decision on particular issues.” (quoting

State v. Fogg, 936 N.W.2d 664, 667 n.1 (Iowa 2019))). Because we affirm the

court of appeals’ remand for a new trial, we also decline to consider Ross’s

sentencing challenge.

II. Noncorroboration Jury Instructions.

We review challenges to jury instructions for correction of errors at law.

State v. Rohm, 609 N.W.2d 504, 509 (Iowa 2000) (en banc). Jury instructions are

considered “as a whole to determine their accuracy” and are “judged in context[,]

2We referenced Altmayer as follows:
An instruction that stated no witness’s testimony needs to be corroborated
(with some exceptions not applicable here) would correctly state the law and
help dispel any misconceptions regarding uncorroborated witness testimony.
Cf. State v. Altmayer, No. 18–0314, 2019 WL 476488, at *5 (Iowa Ct. App.
Feb. 6, 2019) (approving noncorroboration instruction that provided the jury
“should evaluate the testimony of [the alleged victim] the same way [it]
evaluate[d] the testimony of any other witness”).

Kraai, 969 N.W.2d at 495.
 6

. . . not in isolation,” meaning improper instructions “can be cured ‘if the other

instructions properly advise the jury as to the legal principles involved.’ ” Kraai,

969 N.W.2d at 490 (first quoting State v. Donahue, 957 N.W.2d 1, 10 (Iowa

2021); then quoting State v. Liggins, 557 N.W.2d 263, 267 (Iowa 1996); and then

quoting Thavenet v. Davis, 589 N.W.2d 233, 237 (Iowa 1999) (en banc)). Where

an instruction is incorrect as a matter of law and not cured by other instructions,

we presume prejudice “unless the record affirmatively establishes there was

no[ne].” Donahue, 957 N.W.2d at 6 (quoting State v. Hanes, 790 N.W.2d 545, 551

(Iowa 2010)). In the absence of such record evidence, reversal is required. Id.

A. Propriety of the Noncorroboration Instructions. We repeat the

challenged noncorroboration instructions as given by the district court:

You should evaluate the testimony of [L.C. or K.C.] the same
way you evaluate the testimony of any other witness. The law does
not require that the testimony of [L.C. or K.C.] be corroborated in
order to prove that she was sexually abused. You may find the
Defendant guilty of Sexual Abuse if [L.C.’s or K.C.’s] testimony
convinces you of guilt beyond a reasonable doubt.

The court of appeals has previously approved of this same instruction. See

Altmayer, 2019 WL 476488, at *5. The court of appeals had also approved of a

similar but truncated noncorroboration instruction that included only the

second sentence, State v. Barnhardt, No. 17-0496, 2018 WL 2230938, at *4 (Iowa

Ct. App. May 16, 2018) (approving noncorroboration instruction providing that

“[t]he law does not require that the testimony of the alleged victim be

corroborated” (alteration in original)), but we abrogated that decision in Kraai,

see 969 N.W.2d at 496 (“[W]e hold the district court erred in instructing the jury
 7

that ‘[t]here is no requirement that the testimony of a complainant of sexual

offenses be corroborated.’ ” (second alteration in original)).

Our “long-standing precedents” provide “that instructions regarding

witness credibility should apply equally to witnesses for the state and the defense

and that instructions should not draw attention to specific evidence.” Kraai,

969 N.W.2d at 495. In Kraai and Mathis, we held that the truncated

noncorroboration instruction failed on both points—it drew undue attention to

the victims’ testimony and (or, indeed, because) it omitted discussion of the

symmetrical principle that no witness’s testimony needs corroboration to be

believed.3 Id. at 492–93; see also Mathis, 971 N.W.2d at 519–20. Thus, there

were two problems with the Kraai instruction: (1) it emphasized the victim’s

testimony, which could lead jurors to conclude that the victim’s testimony

deserved special consideration; and (2) it provided an incomplete discussion of

the noncorroboration principle, which could confuse jurors and mislead them

into believing that although victims’ testimony does not require corroboration,

other witnesses’ testimony does. The noncorroboration principle applies equally

to all witnesses, and any instruction on the issue should include this symmetry

to avoid misleading the jury into believing it applies only to certain witnesses.

See Kraai, 969 N.W.2d at 493 (“The omission of a symmetrical noncorroboration

3There are some exceptions—not applicable here—where witness testimony requires
corroboration; for example, under the Iowa Rules of Criminal Procedure, the testimony of an
accomplice to a crime requires corroboration “tend[ing] to connect the defendant with the
commission of the offense” in order to convict the defendant. Iowa R. Crim. P. 2.21(3). This case
does not involve accomplices, so an instruction that “no witness’s testimony needs corroboration”
would have been an accurate statement of the law.
 8

instruction may have permitted the jury to infer that Kraai’s testimony required

corroboration to be believed.”); see also State v. Dever, 508 P.3d 158, 170 (Utah

Ct. App. 2022) (“[B]y mentioning only the ‘testimony of a witness to a crime,’ the

jurors could have believed that the testimony of other witnesses, particularly

Dever, did require corroborating evidence to be believed.”).

As compared to the Kraai instruction, the first sentence in this instruction

helps dispel the first improper assumption: that alleged victims’ testimony

requires special consideration. Cf. Tyler J. Buller, Fighting Rape Culture with

Noncorroboration Instructions, 53 Tulsa L. Rev. 1, 24 (2017) (“[N]oncorroboration

instructions can be prefaced with language that tells the jury to evaluate the

credibility of victims the same way they evaluate other evidence. Including that

verbiage takes the wind out of the sails for this concern [of singling out victims’

testimony].”). The first sentence addresses the way the victims’ testimony is

evaluated and naturally takes the jurors back to the general instructions that

they should “[g]ive all the evidence the weight and value [they] think it is entitled

to receive” and “may believe all, part, or none of any witness’s testimony.” Thus,

the first sentence reinforced to the jurors the need to consider the victims’

testimony in the same way they considered other witnesses’ testimony.

But whether a witness’s testimony needs corroboration to support a

conviction is a different issue from how the testimony is evaluated against other

evidence in the case. And importantly, no other instruction here addressed

corroboration of any other witness’s testimony, leaving in place the asymmetry

problem we identified in Kraai. On this point, the second sentence of these
 9

instructions is materially indistinguishable from the Kraai instruction in that it

fails to explain the rest of the noncorroboration principle: that no witness’s

testimony requires corroboration to be believed. Therefore, as in Kraai and

Mathis, there is a risk jurors could have inferred that other witnesses’ testimony

does require corroboration.

The State argues that the first sentence cures any deficiencies in the

second sentence that are inconsistent with Kraai and Mathis. Of course, jury

instructions need not be perfect. See Moser v. Stallings, 387 N.W.2d 599, 605

(Iowa 1986) (“Although the instructions are not perfect, they did not deprive the

plaintiff of a fair trial.”). Ultimately, we are looking for whether the “instructions

are misleading and confusing” when read as a whole. Rivera v. Woodward

Res. Ctr., 865 N.W.2d 887, 902 (Iowa 2015). “[A]n instruction is misleading or

confusing if it is ‘very possible’ the jury could reasonably have interpreted the

instruction incorrectly.” Id. (quoting McElroy v. State, 637 N.W.2d 488, 500 (Iowa

2001)). “On the other hand, if a review of the instructions ‘leads to the inevitable

conclusion that the jury could not have misapprehended the issue,’ then the

challenge is without merit.” Id. (quoting Moser, 387 N.W.2d at 605).

We are not convinced that the first sentence’s direction to evaluate the

victims’ testimony the same as other witnesses cures the express reference to

corroboration of a victim’s testimony in the second sentence. As written, the jury

could read the second sentence to be an example of the principle contained in

the first sentence, or it could be read as an exception to that principle. They

could be read together as:
 10

You should evaluate the testimony of [L.C. or K.C.] the same
way you evaluate the testimony of any other witness. [In other
words,] [t]he law does not require that the testimony of [L.C. or
K.C.] be corroborated . . . .

Or they could be read together as:

You should evaluate the testimony of [L.C. or K.C.] the same
way you evaluate the testimony of any other witness. [However,
(or But,)] [t]he law does not require that the testimony of [L.C. or
K.C.] be corroborated . . . .

These examples reflect that the first sentence does not provide the clarification

the State suggests. Because we do not expect jurors to parse jury instructions

the same way we parse statutes, we cannot be certain how they attempted to

navigate the ambiguity. Either interpretation was plausible, but one was legally

erroneous.

In any event, we must also consider the third sentence of Instructions 16

and 17, which was the primary concern Ross raised at trial. That sentence says:

“You may find the Defendant guilty of Sexual Abuse if [L.C.’s or K.C.’s] testimony

convinces you of guilt beyond a reasonable doubt.” As Ross argued in his original

objection to the district court, the third sentence highlights the girls’ testimony

over other evidence in the record. The overarching premise of our concern in

Kraai was that “instructions that set apart, highlight, or accentuate the

testimony of a particular witness or a particular piece of evidence are improper.”

969 N.W.2d at 492. The third sentence of Instructions 16 and 17 runs afoul of

that concern. See, e.g., Ludy v. State, 784 N.E.2d 459, 460–62 (Ind. 2003)

(holding that jury instruction “unfairly focuse[d] the jury’s attention on and

highlight[ed] a single witness’s testimony” where it instructed the jury: “A
 11

conviction may be based solely on the uncorroborated testimony of the alleged

victim if such testimony establishes each element of any crime charged beyond

a reasonable doubt.”); Veteto v. State, 8 S.W.3d 805, 816 (Tex. Ct. App. 2000)

(holding that jury instruction singled out the alleged victim’s testimony and

improperly commented on the weight of the evidence where it informed the jury:

“The law provides the testimony of the victim alone, if believed by you beyond a

reasonable doubt, need not be supported by other evidence before a finding of

guilt can be returned. That is to say, the testimony of [A.L.], standing alone, if

believed by you beyond a reasonable doubt, is sufficient proof to support a

finding of guilt.” (alteration in original)), abrogated on other grounds by State v.

Crook, 248 S.W.3d 172, 176–77 (Tex. Crim. App. 2008).

Although the truncated noncorroboration instruction in Kraai did not

include this reasonable doubt sentence, we did consider whether the general

reasonable doubt instruction cured the taint from the asymmetrical

noncorroboration instruction. See 969 N.W.2d at 496. Not only did we conclude

it did not, we reasoned that, “[i]f anything,” it accentuated the error, explaining:

“While the jury was correctly instructed that the State bore the burden of proving

Kraai guilty beyond a reasonable doubt, the jury was determining whether the

State satisfied its burden in light of the noncorroboration instruction that

uniquely accentuated N.F.’s testimony over all others.” Id. Here, the third

sentence of the instructions even more directly accentuated the alleged victims’

testimony: “You may find the Defendant guilty of Sexual Abuse if [L.C.’s or K.C.’s]

testimony convinces you of guilt beyond a reasonable doubt.” (Emphasis added.)
 12

Review of the other, general jury instructions does not alleviate the

concern that Instructions 16 and 17 improperly focused the jury on the girls’

testimony. Those instructions were the same here as they were in Kraai and

Mathis, where we concluded they may have actually amplified, rather than cured,

the error in the noncorroboration instruction. Kraai, 969 N.W.2d at 496 (“If

anything, the instructional error necessarily was channeled into the general

instructions. For example, instruction 10 told the jury they could believe ‘all,

part or none of any witness’s testimony,’ but, in determining which witnesses to

believe, the jury evaluated [the victim’s] testimony in light of the

noncorroboration instruction that uniquely accentuated her testimony over all

others.”).

Likewise, the instructions here may have compounded any uncertainty

and pushed jurors to make the improper inference. For example, Instruction 13,

dealing with expert testimony, contained a clause similar to the first sentence in

the noncorroboration instructions: “Consider expert testimony just like any

other testimony.” That instruction did not, however, follow up with a sentence

explaining that expert witness testimony does not need to be corroborated to be

believed. When comparing Instruction 13 to the noncorroboration instructions,

a juror could reasonably conclude that since the instructions started from an

equal premise (i.e., treat all witness testimony the same) but Instructions 16 and

17 contained an additional point of law (i.e., the law does not require

corroboration for victims’ testimony specifically), the additional point about

noncorroboration was exclusive to L.C.’s and K.C.’s testimony. See State v.
 13

Davis, 951 N.W.2d 8, 19 n.1 (Iowa 2020) (“When the marshaling instructions for

the other nine offenses cross-referenced the insanity defense but not the

instruction for first-degree murder, the jurors would reasonably conclude that

omission was intentional and the defense was unavailable.”). We cannot say the

jurors would necessarily read the first two sentences of Instructions 16 and 17

to apply the corroboration principle to all witnesses. And more critically, read

together with the third sentence, those instructions single out the girls’

testimony, thereby improperly highlighting some evidence (their testimony) over

all other evidence. We conclude that Instructions 16 and 17 were misleading or

confusing to the jury. See Rivera, 865 N.W.2d at 902 (instructions are misleading

or confusing if it is “ ‘very possible’ the jury could reasonably have interpreted

the instruction incorrectly” (quoting McElroy, 637 N.W.2d at 500)).

As in Kraai, we acknowledge the State’s interest in sex-abuse cases in

dispelling the misconception that alleged victims’ testimony requires

corroboration to support a conviction.4 969 N.W.2d at 495. But “those interests

can be advanced by a nonparticularized instruction applicable to all witness

testimony.” Id. (emphases added); see also Mathis, 971 N.W.2d at 520 (“In the

absence of any other instruction regarding the uncorroborated testimony of other

4We note, however, that the State’s need to do so here is somewhat questionable. On
appeal, the State gave examples of three jurors’ responses to voir dire questions about the need
for victim-witness corroboration to support its position that the noncorroboration instruction
was necessary here. None of those three jurors were selected for the final jury panel. And the
State candidly admits that “other panelists said they understood that they could rely on
testimony as proof that abuse occurred.” Presumably, then, the voir dire process worked as
intended to remove jurors who harbored such attitudes and were unwilling to set them aside.
But for the jurors that remained, the State’s repeated emphasis through its voir dire questioning
on whether corroboration is required may have only worked to sow more confusion than it
settled.
 14

witnesses, the instruction unduly and improperly emphasized the complainant

witnesses’ testimony.” (emphasis added)). The instructions given here missed

that mark; they were still particularized to the victims’ testimony, and there was

still no other instruction telling jurors the noncorroboration principle applies to

all witnesses. We meant what we said in Kraai: “[a]n instruction that state[s] no

witness’s testimony needs to be corroborated” would adequately balance the

competing interests. Kraai, 969 N.W.2d at 495.5 Without endorsing it as an

exclusive method of doing so, we suggest courts simply use that language when

the State requests a noncorroboration instruction: “No witness’s testimony needs

to be corroborated to be believed.” Such an instruction “correctly state[s] the law

and help[s] dispel any misconceptions regarding uncorroborated witness

testimony” without unduly emphasizing any witness’s testimony over another’s.

Id.

B. Prejudice. Having determined the district court erred by giving an

improper noncorroboration instruction that inappropriately highlighted the

5On this point, we recognize Kraai’s “cf.” citation to Altmayer caused some confusion as
to whether we intended to approve of the entire instruction used there. As the majority opinion
in the court of appeals here correctly concluded, we did not intend to tacitly approve of the entire
instructions used in Altmayer through a “cf.” citation. That citation is best understood as
building on Kraai’s earlier discussion of nonparticularized noncorroboration instructions
generally, where we cited State v. Ludwig, 305 N.W.2d 511 (Iowa 1981), as an example of a case
where a court used an instruction that “by its terms [applied] to all witnesses who testif[ied] at
trial” and therefore “did not carry with it the danger of distinguishing [the] defendant’s testimony
from that of other witnesses.” Kraai, 969 N.W.2d at 493 (alterations in original) (quoting Ludwig,
305 N.W.2d at 512).
In contrast, the instructions used here and in Altmayer contain a sentence that, by its
terms, applies to all witnesses—“evaluate the testimony of [an alleged victim] the same way [you]
evaluate the testimony of any other witness,” Altmayer, 2019 WL 476488, at *5—followed by a
sentence that, by its terms, applies only to the victims’ testimony. Approving the use of the first
sentence generally does not mean that the second sentence—explicitly disapproved of in Kraai
and Mathis—becomes proper by association.
 15

alleged victims’ testimony, we must address whether that error requires a new

trial. Kraai, 969 N.W.2d at 496–97. We presume prejudice “unless the record

affirmatively establishes there was no[ne].” Donahue, 957 N.W.2d at 6 (quoting

Hanes, 790 N.W.2d at 551). “When [an instructional] error is not of constitutional

magnitude, the test of prejudice is whether it sufficiently appears that the rights

of the complaining party have been injuriously affected or that the party has

suffered a miscarriage of justice.” State v. Plain, 898 N.W.2d 801, 817 (Iowa

2017) (quoting State v. Marin, 788 N.W.2d 833, 836 (Iowa 2010), overruled on

other grounds by Alcala v. Marriott Int’l, Inc., 880 N.W.2d 699 (Iowa 2016)). Our

prejudice inquiry “looks . . . to the basis on which ‘the jury actually rested its

verdict.’ ” State v. Kennedy, 846 N.W.2d 517, 527 (Iowa 2014) (omission in

original) (quoting Sullivan v. Louisiana, 508 U.S. 275, 279 (1993)). “The inquiry

‘is not whether, in a trial that occurred without the error, a guilty verdict would

surely have been rendered, but whether the guilty verdict actually rendered in

this trial was surely unattributable to the error.’ ” State v. Shorter, 945 N.W.2d

1, 9 (Iowa 2020) (quoting Kennedy, 846 N.W.2d at 527). The presumption of

prejudice may also be “overcome when the jury received ‘strong evidence’ of a

defendant’s guilt.” Kraai, 969 N.W.2d at 497 (quoting Plain, 898 N.W.2d at 817).

In Kraai, this court found “strong evidence of Kraai’s guilt” overcame any

prejudice that the erroneous instruction may have caused. Id. at 497–99. The

victim’s testimony in that case was corroborated by several items of physical

evidence, making “the challenged instruction, in essence, . . . a moot point.” Id.

at 499 (quoting Garza v. State, 231 P.3d 884, 891 (Wyo. 2010)).
 16

In Mathis, on the other hand,

neither [victim’s] testimony [wa]s corroborated by other evidence.
Both children testified to separate instances of sex abuse, but
neither witnessed nor testified about the sex abuse allegedly
committed against the other child. Both children eventually told
their mother about the sexual abuse, but there [wa]s no evidence
their mother independently corroborated the children’s accounts
of abuse. And there [wa]s no physical evidence of any sort
corroborating the children’s testimony.

The noncorroboration instruction in th[at] case thus did not
pertain to a moot point; instead, the noncorroboration instruction
in th[at] case was a focal point.

971 N.W.2d at 521.

The facts of this case put it closer to Mathis than Kraai, meaning the

presumption of prejudice was not overcome. All of the State’s evidence of guilt

derived—either directly or indirectly—from L.C.’s and K.C.’s allegations. No

physical evidence was introduced; in fact, the State’s only exhibits at trial were

a photo of Ross (used for identification purposes) and the resumes of its two

expert witnesses. The State’s experts testified only as to issues regarding

sex-abuse victims generally; neither was even familiar with L.C. or K.C. prior to

trial. Other than discussing his interview with Ross, in which Ross generally

denied the charges, Deputy Kinney’s testimony only relayed what he learned

about the allegations through conversations with Tykeshia, Jackie, and

investigators in Colorado, all of whom acquired their knowledge of the allegations

from L.C. and K.C. Although by all accounts Jackie was supportive of the girls’

claims, she did not testify at trial. Finally, Tykeshia testified, in Ross’s defense,

that she had never noticed signs of abuse and that the girls never reported any

abuse to her until their April 2019 trip to Colorado. Notably, the prosecutor
 17

highlighted the noncorroboration instruction during closing argument, telling

the jury:

Now, Instruction No. 16. Can everybody read that? Everybody see
that? “You should evaluate the testimony of [L.C.] the same way
you evaluate the testimony of any other witness.” The law does
not require that the testimony of [L.C.] be corroborated in order
to prove that she was sexually abused. It doesn’t have to be DNA
evidence. There doesn’t have to be. “You may find the defendant
guilty of sexual abuse if [L.C.’s] testimony -- if what she told you
convinces you of guilt beyond a reasonable doubt.”

Instruction No. 17 is the exact same instruction except for
[K.C.]. If you believe [K.C.’s] testimony, if it convinces you beyond
a reasonable doubt, then you must find Alexander Ross guilty.

The asymmetrical noncorroboration instructions could have led jurors to believe

Tykeshia’s testimony needed to be corroborated while also improperly

highlighting the importance of L.C.’s and K.C.’s testimony over Tykeshia’s.

Accordingly, although “there is sufficient evidence to sustain [Ross’s]

convictions” over his sufficiency-of-the-evidence claim, here, where the

noncorroboration instruction improperly tainted the jury’s consideration of

L.C.’s and K.C.’s testimony, and where that consideration was the “focal point”

of the State’s case, we cannot conclude that the record “affirmatively establish[es]

the absence of prejudice.” Mathis, 971 N.W.2d at 521.

III. Conclusion.

The State presented sufficient evidence to sustain Ross’s convictions, but

the district court erred by giving the jury asymmetrical, particularized

noncorroboration instructions focusing on the victims’ testimony. Ross’s

convictions are hereby vacated, and the case is remanded to the district court

for further proceedings.
 18

DECISION OF COURT OF APPEALS AFFIRMED; DISTRICT COURT

JUDGMENT VACATED AND CASE REMANDED.

McDonald, McDermott, and May, JJ., join this opinion. Mansfield, J., files

a dissenting opinion, in which Christensen, C.J., and Waterman, J., join.
 19

#20–0914, State v. Ross

MANSFIELD, Justice (dissenting).

I respectfully dissent and would affirm the sexual abuse convictions of the

defendant Alexander Ross. I question whether there was any error in the

instructions; if there was, it was inconsequential and nonprejudicial. There is no

reason to subject the child victims in this case to a second trial.

In State v. Kraai, 969 N.W.2d 487, 490, 492–94 (Iowa 2022), we

disapproved of a different jury instruction that said, simply, “There is no

requirement that the testimony of a complainant of sexual offenses be

corroborated.” We criticized that instruction because it “focused the jury on the

testimony of a single witness, [i.e., the complaining witness],” and it

“particularized [the complaining witness’s] testimony as not requiring

corroboration in the absence of a universal instruction regarding the

noncorroboration of all other witness testimony.” Id. at 493. In other words, the

single-sentence instruction was “asymmetrical and particularized.” Id.

We could have followed a different path in Kraai. We noted in Kraai that

courts are divided and several jurisdictions have approved of this single-sentence

instruction. Id. at 494–95. We also observed that the instruction “was a correct

statement of the law.” Id. at 491–92. Nonetheless, we held that the giving of this

instruction was error, although in the end we did not disturb the jury verdict

because we concluded that the error was nonprejudicial and harmless. Id. at

496, 499.
 20

This case involves a far less objectionable instruction than the single-

sentence instruction at issue in Kraai. And it’s another case in which the

evidence of the defendant’s guilt was strong. See id. at 497. Accordingly, I would

find that any minimal error here—if there was error—was nonprejudicial and

harmless.

Instructions must be read as a whole. Id. at 490. In the present case,

Instruction 16 said,

You should evaluate the testimony of [L.C.] the same way you
evaluate the testimony of any other witness. The law does not
require that the testimony of [L.C.] be corroborated in order to prove
that she was sexually abused. You may find the Defendant guilty of
Sexual Abuse if [L.C.]’s testimony convinces you of guilt beyond a
reasonable doubt.

Instruction 17 contained the same directive for K.C.

This form of instruction did not originate in this particular trial. It was

proposed in a thoughtful 2017 law review article by a prosecutor who now serves

on our court of appeals. See Tyler J. Buller, Fighting Rape Culture with

Noncorroboration Instructions, 53 Tulsa L. Rev. 1, 27–29 (2017) [hereinafter

Buller].

By way of further context, the preamble to all of the instructions stated

that the State has “the burden of proving beyond a reasonable doubt each and

every allegation of the Trial Information.” Instruction 2 added, “If the State does

not prove the Defendant guilty beyond a reasonable doubt, your verdict must be

not guilty.” Instruction 6 said, “In determining the facts, you may have to decide

what testimony you believe. You may believe all, part, or none of any witness’s

testimony.” Instruction 7 told the jury: “You must consider all of the instructions
 21

together. No one instruction includes all of the applicable law.” And Instruction

9 reiterated, “The burden is on the State to prove Alexander Shantee-Thomas

Ross guilty beyond a reasonable doubt.”

All things considered, I seriously doubt that Instructions 16 and 17 had

any risk of misleading the jury. Unlike the instruction challenged in Kraai, which

consisted of only a single sentence highlighting the testimony of the complaining

witness as not needing corroboration, the instruction here had that sentence

sandwiched between a statement that “[y]ou should evaluate the testimony of

[the complaining witness] the same way you evaluate the testimony of any other

witness” and a statement that “[y]ou may find the Defendant guilty of Sexual

Abuse if [the complaining witness’s] testimony convinces you of guilt beyond a

reasonable doubt.” In order to go astray, the jury would have had to disobey both

of these additional sentences.

The majority says that the jury could have understood the second sentence

as contradicting the first. Respectfully, that makes no sense. “We presume jurors

follow instructions.” State v. Fontenot, 958 N.W.2d 549, 562 (Iowa 2021). Also,

we ourselves follow a rule that we try to avoid construing statutes, contracts, or

whatever so they conflict with each other; shouldn’t we credit a jury of twelve as

having the same discernment? Not only was the first sentence of Instructions 16

and 17 clear and unambiguous, but the third sentence also steered the jury back

to the critical point that the complaining witness’s testimony—uncorroborated

or not—needed to convince the jury beyond a reasonable doubt in order to justify

a guilty verdict. See Buller, 53 Tulsa L. Rev. at 27–28 (explaining how the first
 22

and third sentences of this form of instruction counter any criticism that could

be leveled at the second sentence).

Notably, Kraai cited an unpublished court of appeals opinion that had

approved of the same three-sentence instruction:

An instruction that stated no witness’s testimony needs to be
corroborated (with some exceptions not applicable here) would
correctly state the law and help dispel any misconceptions regarding
uncorroborated witness testimony. Cf. State v. Altmayer, No.
18-0314, 2019 WL 476488, at *5 (Iowa Ct. App. Feb. 6, 2019)
(approving noncorroboration instruction that provided the jury
“should evaluate the testimony of [the alleged victim] the same way
[it] evaluate[d] the testimony of any other witness”).

969 N.W.2d at 495 (alterations in original). According to The Bluebook, the tea

leaf “Cf.” means, “The authority is different from the main proposition but

sufficiently analogous to lend support.” The Bluebook: A Uniform System of

Citation B. 1.2(a), at 5 (Columbia L. Rev. Ass’n et al. eds., 21st ed. 2020). Thus,

from our signal, one could conclude that we were blessing the same three-

sentence instruction given in Altmayer; we certainly didn’t suggest it amounted

to reversible error.

Moreover, there is a key point disregarded by the majority: Ross didn’t take

the stand and deny the charges. So this wasn’t a situation, as in Kraai, where

“[t]he omission of a symmetrical noncorroboration instruction may have

permitted the jury to infer that [the defendant]’s testimony required

corroboration to be believed.” 969 N.W.2d at 493. This wasn’t a “she said, he

said” case; it was a “they said” case, and the question was whether “they” should

be believed beyond a reasonable doubt.
 23

And the State’s case was strong. To recap, at the time of trial, L.C. was ten

years old; K.C. was nine. They described fairly recent events. To my reading, their

testimony comes across as quite compelling. But don’t take my word for it. Ask

the trial judge. As she put it, “[T]hese two little girls . . . were about as credible

as any child witness I’ve ever heard.”

Here is an example from K.C.’s testimony:

Q. . . . Where did he touch you?

A. In my lower area.

Q. Okay. Do you know what the doctors and teachers call that
lower area, what the word is for it?

A. Yes.

Q. What is it? What’s that word?

Are you embarrassed to say it?

A. Yes.

Q. Why don’t you just tell us anyway.

If I had you stand up, could you point to it?

A. No.

Q. Why not?

A. Because it makes me feel uncomfortable.

Q. Okay. Why does it make you feel uncomfortable?

A. Because -- I don’t know.

Q. Well, what do you call that part?

A. The “V” word.

Q. And what is the “V” word? Can you spell it? Would it be
easier to spell it?
 24

A. I don’t know how to spell it.

The defense took depositions of both girls before trial, yet we did not hear

anything at trial about those depositions. What does that tell you? Presumably,

the girls’ deposition testimony was consistent with their trial testimony.

The defense called two witnesses. I do not believe they helped the defense.

One was a longtime friend of Ross who testified basically that he observed Ross

to be a good father. The other was the girls’ mother, Tykeshia. She admitted she

had been working nights as a nurse during the alleged abuse. Therefore, there

would have been plenty of opportunity for Ross to molest the girls at night, as

they testified. Tykeshia also admitted that Ross couldn’t hold down a job, that

he basically hung around the house, and that he had been physically abusive to

her. She denied that either girl had mentioned the sexual abuse to her (contrary

to the girls’ testimony). Yet she had a motive to deny that the girls had told her

about the sexual abuse. Admitting to awareness of sexual abuse would reflect

poorly on her since she had not done anything about it.

And unlike K.C. and L.C., Tykeshia was impeached. She admitted that she

had not been honest in her testimony in the child-in-need-of-assistance

proceeding several months before the criminal trial, where she testified that Ross

had called her “maybe once or twice.” In fact, there were over one hundred

complete phone conversations, and she was still telling Ross she loved him.

As I’ve noted, Ross exercised his constitutional right not to testify. But it

is also significant that he went AWOL after the trial started.
 25

Trial was supposed to begin on the morning of Wednesday, March 11,

2020. Ross was not present in the courtroom. His attorney reported that Ross

claimed his wife and daughter had been in a car accident and he had been

unable to get a ride.6 The defense asked for a one-day continuance. The

continuance was not opposed, and the trial court granted it. Ross appeared in

court the next day, March 12, which was the first day of trial. Jury selection took

place, opening statements were given, and one witness testified.

But on the second day of trial, March 13, the defendant again was absent.

Defense counsel reported that the defendant claimed to be ill. The district court

excused the jury but ordered the defendant to report on Monday, March 16, with

a note from a doctor explaining his absence. The guardian ad litem objected to

the delays, noting the stress on the girls, and it was put on the record that the

girls would be testifying on Monday.

On Monday, March 16, the defendant was again a no-show. Evidence was

presented that Ross had removed his GPS monitor. The court issued a warrant

for Ross’s arrest and ordered that the trial proceed in his absence. The girls and

other witnesses testified. Then, on Tuesday, March 17, Ross again failed to

appear for the final day of trial, which included more testimony, closing

arguments, jury deliberations, and the jury’s verdict.

At the prosecution’s request, the district court instructed the jury as

follows regarding Ross’s absence:

6The prosecutor later made a record that “it was told to [him] that it was a very minor

accident.”
 26

The Defendant has a Constitutional Right to be present at his
trial. The defendant also can choose to waive this right and not
attend all or part of the trial proceedings. In this case, the
defendant’s partial attendance at the trial should not be considered
in the jury’s deliberations. You are to draw no inference of guilt from
Defendant’s partial attendance. The burden of proof remains upon
the State to prove the guilt of the defendant.

In my view, this instruction (Instruction 14) was a gift to Ross. Under the

circumstances here, I think a jury would have been entitled to infer that Ross

didn’t want to face his accusers. See People v. Tafoya, 833 P.2d 841, 843 (Colo.

App. 1992) (upholding jury instruction telling jurors they could consider

defendant’s flight from trial as an indication of guilt); Sorrell v. State, 554 A.2d

352, 354–55 (Md. 1989) (finding court properly instructed jury that it may infer

consciousness of guilt from the defendant’s voluntary absence). As Texas’s

highest criminal court recently put it, “A defendant could be present and still not

testify, or he could testify and be absent part of the time, so a comment on a

defendant’s absence from the proceedings is not itself a direct comment on the

failure to testify.” Sandoval v. State, No. AP–77,081, 2022 WL 17484313, at *39

(Tex. Crim. App. Dec. 7, 2022).

I do not fault the prosecutor’s decision to ask for Instruction 14;

presumably, he wanted to avoid creating an issue for appeal. But I don’t believe

the defendant was entitled to this instruction. In fact, Instruction 14 is the only

instruction in the case that actually misstated the law because the defendant

doesn’t have a “choice” to waive his presence at trial. So to sum up, the court’s

jury instructions—considered in their entirety—gave Ross more than he was

entitled to.
 27

One final point. This case is not like State v. Mathis, 971 N.W.2d 514, 516,

519–21 (Iowa 2022), where we reversed for a new trial based on the giving of the

same single-sentence instruction that we condemned in Kraai. The present case,

of course, doesn’t involve the problematic Kraai instruction. In addition, we have

different facts. In Mathis, the defendant took the stand and denied the abuse, so

the case came down to a swearing contest between the alleged victims and the

alleged perpetrator. Id. at 517–18. In Mathis, it was prejudicial to emphasize that

the word of the alleged victims didn’t need to be corroborated—without also

pointing out that testimony of all witnesses should be treated the same and the

testimony of the alleged victims had to be believed beyond a reasonable doubt to

support a guilty verdict. Id. at 520. The Mathis circumstances don’t exist here.

For all the reasons stated, I would affirm Ross’s convictions and sentence.7

Christensen, C.J., and Waterman, J., join this dissent.

7Ross’s other arguments for reversal—(1) that the evidence was insufficient and (2) that
the district court considered an improper factor during sentencing—are insubstantial, and I
would reject them as well.