State v. Bowman

No. 4 January 30, 2025 213

IN THE SUPREME COURT OF THE

STATE OF OREGON

STATE OF OREGON,

Respondent on Review,

v.

RALIN KATHALINE BOWMAN,

Petitioner on Review.

(CC 19CR63060) (CA A175839) (SC S070412)

En Banc

On review from the Court of Appeals.*

Argued and submitted May 9, 2024.

Joshua B. Crowther, Senior Deputy Public Defender,

Oregon Public Defense Commission, Salem, argued the

cause and filed the reply brief for petitioner on review. John

P. Evans, Senior Deputy Public Defender, filed the brief on

the merits for petitioner on review. Also on the briefs was

Ernest G. Lannet, Chief Defender, Criminal Appellate

Section.

Adam W. Holobrook, Assistant Attorney General, Salem,

argued the cause and filed the brief for respondent on

review. Also on the brief were Ellen F. Rosenblum, Attorney

General, and Benjamin Gutman, Solicitor General.

FLYNN, C.J.

The decision of the Court of Appeals is reversed. The

judgment of the circuit court is reversed, and the case is

remanded to the circuit court for further proceedings.

Bushong, J., dissented and filed an opinion, in which Garrett,

J., joined.

______________

* Appeal from Lincoln County Circuit Court, Thomas O. Branford,

Judge. 326 Or App 565 (2023) (nonprecedential memorandum opinion).

214 State v. Bowman

Cite as 373 Or 213 (2025) 215

FLYNN, C.J.

The Oregon Evidence Code allows a witness qual-

ified as an expert by their knowledge, skill, experience,

training, or education to testify to that specialized knowl-

edge “in the form of an opinion or otherwise,” and it allows

the expert to base opinions on information made known to

the expert outside of the trial, if the information is “of a type

reasonably relied upon by experts in the particular field.”

OEC 702; OEC 703. As we will explain, those rules permit

an expert’s testimony to be based in part on hearsay, but

they do not permit an expert to simply repeat another’s out-

of-court statement as substantive evidence offered for its

truth. See OEC 801(3) (defining hearsay); OEC 802 (specify-

ing that hearsay is inadmissible absent a specific exception).

In this criminal case, we must decide whether the

trial court correctly overruled defendant’s “hearsay” objec-

tion when a law enforcement officer, who was qualified as an

expert in evaluating impaired driving, testified that babies

have limited peripheral vision and that a person impaired

by alcohol “could be said to be seeing just like a baby.” As we

will explain, we conclude that the challenged testimony was

hearsay because it was offered for the truth of the matter

asserted and merely repeated information that was beyond

the scope of the witness’s personal or specialized knowledge.

As we also explain, this is not a case in which we can affirm

the verdict despite the error, because we cannot say that

there was little likelihood that the erroneously admitted

opinion affected the jury’s verdict. Accordingly, we reverse.

I. BACKGROUND

A. Facts and Trial Court Proceedings

We summarize from the trial court record the facts

that are pertinent to the evidentiary issue presented here.

Oregon State Police Trooper Wertz stopped a car driven by

defendant and, based on defendant’s driving, her slurred

speech, and the odor of alcohol coming from the vehicle,

suspected that she was under the influence of intoxicants.

Defendant consented to Wertz administering a series of field

sobriety tests (FSTs), starting with a test commonly admin-

istered during a DUII investigation, called the horizontal

216 State v. Bowman

gaze nystagmus (HGN) test.1 Wertz then had defendant per-

form two other FSTs—the walk-and-turn and one-leg-stand

tests—before he repeated the HGN test. A breath test later

showed defendant’s blood alcohol content (BAC) to be 0.08

percent—above the legal limit for a driver.

The state charged defendant with Driving Under

the Influence of Intoxicants (DUII), ORS 813.010, Reckless

Driving, ORS 811.140, and Recklessly Endangering Another

Person, ORS 163.195. At defendant’s trial, the state pursued

the DUII charge on the alternative theories that defendant

had driven either with a blood alcohol level of at least 0.08

percent2 or while “under the influence of an intoxicant.” See

ORS 813.010(1) (describing alternative methods of commit-

ting the offense of DUII). Wertz testified on behalf of the

state as both a factual witness and as an expert witness.

In describing his qualifications as an expert at “detecting

impairment out on the road,” Wertz testified that he had

received general training at the police academy on admin-

istering FSTs and also had obtained certification as a drug

recognition evaluator (DRE) after attending an “intensive”

two-week training course that included education about how

FST results may reflect the body’s physiological response

to consuming alcohol or other intoxicants. He described

the HGN test as an eye examination where the person is

instructed to hold their head still and follow a stimulus

(here, a pen) from side to side with only their eyes. Wertz

explained that, when conducting that test, he looks for “nys-

tagmus,” which he described as a “rhythmic bouncing within

a person’s eyes.” Wertz further explained that, when alcohol

is present “at an impairing level,” the person will exhibit an

involuntary sideways rhythmic bouncing within their eyes.

When Wertz administered the HGN test to defen-

dant, he had instructed defendant to hold her “head perfectly

1

The HGN test is “designed to detect whether a person’s eyes demonstrate

nystagmus under certain conditions.” State v. O’Key, 321 Or 285, 294, 899 P2d 663

(1995). HGN is a physiological phenomenon involving an “involuntary rapid move-

ment of the eyeball.” Id. HGN occurs “when a person looks to the side at an object.”

Id. It arises from the eyes’ inability to “maintain visual fixation as they are turned

from side to side”—resulting in observable jerking or bouncing of the eyeball. Id.

2

A police forensic scientist testified that, by extrapolating based on the dis-

sipation rate of alcohol in the blood system, defendant’s BAC likely had been

between .08 and .11 while she had been driving.

Cite as 373 Or 213 (2025) 217

still,” focus on the tip of his pen, and follow the movement of

his pen with only her eyes. He testified that defendant had

been able to track the pen with her eyes without moving

her head, but that he had observed four of the six standard-

ized “clues” of impairment. Specifically, Wertz explained

that he had observed in both eyes a “lack of smooth pur-

suit” as defendant tracked the pen’s horizontal movement

(the first two clues). He described that lack of smooth pur-

suit as “like an old reel film” being played slowly, making

the vision appear “choppy.” Wertz also observed sustained

nystagmus in both eyes at the maximum point of deviation

as defendant’s eyes focused to the far left and right (the

second two “clues”). He concluded that, because defendant

had exhibited those “clues” of impairment on the HGN test,

there was “definitely evidence of impairment.”3 On redirect

examination, Wertz explained that the HGN test is useful

in detecting impairment because the perceptible “bouncing”

of the eyeball is an involuntary response that is not affected

by a person’s balance or fitness level. All of that testimony

was admitted without objection.

At the insistence of the prosecutor, however, Wertz

then provided additional testimony on redirect examination

that is at issue on appeal. The prosecutor first prompted

Wertz to confirm the statements of other officers that alco-

hol causes an adult’s eyes to “regress to like being a child, or

being a baby”:

“[PROSECUTOR:] Okay. And Trooper Wertz, I’ve

heard other officers describe it with a [HGN],4 that with

alcohol on board, your eyes regress to like being a child, or

being a baby.

“[WERTZ:] It’s something that Dr. * * * Citek teaches.

He’s an ophthalmologist that teaches—”

At that point, defense counsel objected based on a “lack of

foundation,” but the court overruled the objection. The court

reasoned:

3

Wertz testified to observing additional signs of impairment when he admin-

istered the walk-and-turn and one-leg-stand tests to defendant.

4

In the official transcript, the references to “HGN” are transcribed as “HgN.”

For precision and consistency, we have used brackets to change the abbreviation.

218 State v. Bowman

“We all learn from other people everything we know, but

this is not hearsay. He’s trying to explain the basis for the

opinion, so it’s admissible.”

When questioning resumed, Wertz explained that

Citek was a practicing ophthalmologist who taught at an

ophthalmological college as well as in the DRE school. Wertz

next stated, “[t]he way that [Citek] describes the effects of

alcohol on the eyes to students is that he talks about—,”

but defendant objected again, this time on the bases that

Wertz’s testimony was hearsay and that allowing the jury

to hear Citek’s opinion through Wertz would violate defen-

dant’s rights under the confrontation clauses of the state

and federal constitutions. The court again overruled the

objection.

After Wertz described additional information that

he had learned in his training about why the observation

of HGN indicates impairment, the prosecutor again sought

confirmation of the “baby-eyes” comparison, asking:

“And so I’ve heard it explained, or described as far as

with [HGN] present, it’s like a child unable to focus, or the

baby looking at—?”

Wertz answered:

“[I]n the development of a child, a child develops their eyes

in the first roughly nine months of child development, a

baby is not able to move their head independent of their

eyes. They have to look wherever they are—whatever they

want to see[,] they have to turn their entire head to do that.

At about nine months to a year is when a child develops

the ability to move their eyes independent of their head,

and then they also begin to develop a more robust periph-

eral vision until you finally develop the full use of your

eyes as you would as an adult with full peripheral about

120-degree field of vision, and the ability to move from a

distance object to a near object and back and forth.”

Wertz continued, comparing the vision of “a person with

enough alcohol on board” to the vision of a baby:

“Alcohol has the effect that it seems to cause a person to

regress backwards in that process, so * * * a person with

enough alcohol on board could be said to be seeing just

like a baby, or they are not able to even look at something

Cite as 373 Or 213 (2025) 219

independent of moving their head. Their eyes are moving

with their head now.”

At that point, the prosecutor commented: “Okay. That’s

what I wanted to get to.”

In closing argument, the prosecutor contended that

the jury could find defendant guilty of DUII either because

she had a BAC of .08 or more when she was driving or

because her physical or mental faculties had been adversely

affected by alcohol to a noticeable or perceptible degree. And

the same evidence that proved those charges, the prosecutor

argued, helped to prove the counts of reckless driving. The

prosecutor’s argument about impairment in part relied on

Wertz’s testimony comparing the vision of a person “with

enough alcohol on board” with the limited vision of a baby:

“So one of the ways of looking at that is the [HGN]. Probably

a lot of you until you came into * * * court here yesterday,

had never heard of the [HGN] test. And probably learned

something new in this trial as far as what [HGN] is, and

why it is important, that we have the full use of our physi-

cal and mental faculties when we’re operating a motor vehi-

cle. And we don’t want our eyes bouncing around. We don’t

want to be like Trooper Wertz talked about. We don’t want

to be regressing back to when we were a child and we have

to turn our head to focus on things. We want peripheral

vision.”

The jury found defendant guilty of DUII, reckless driving,

and reckless endangerment, and she appealed.

B. Appellate Proceedings

Defendant contended on appeal that the trial court

had erred in overruling her hearsay objection and allow-

ing Wertz to testify about what he had learned from the

ophthalmologist during his training. The Court of Appeals

disagreed, assessing that, “[a]bsent testimony from Wertz

that recounted a particular statement made by the ophthal-

mologist, there was no hearsay.” State v. Bowman, 326 Or

App 565, 570 (2023) (nonprecedential memorandum opin-

ion). The court reasoned that Wertz’s testimony “was not

hearsay” because he ultimately “discussed his training in

a way that did not refer to the ophthalmologist.” Id. at 570.

Even if Wertz’s testimony “could be construed as repeating

220 State v. Bowman

statements made by the ophthalmologist,” the court fur-

ther explained, the testimony was not hearsay because it

was offered only to “provide the foundation necessary to

explain” Wertz’s expert opinion, citing this court’s decision

in McCathern v. Toyota Motor Corp., 332 Or 59, 23 P3d 320

(2001). Id. Finally, the court concluded that, because the tes-

timony had not amounted to hearsay, “there was no violation

of defendant’s [constitutional] right to confront witnesses.”

Id. at 571.5

We allowed review and now reverse.

II. DISCUSSION

A. Standard of Review, Rules at Issue, and Parties’

Contentions

Whether the trial court correctly applied the rules of

evidence to decide that Wertz’s testimony was not hearsay is

a question that has “only one legally correct answer” and is,

thus, a question that we review for legal error. See Arrowood

Indemnity Co. v. Fasching, 369 Or 214, 250, 503 P3d 1233

(2022) (describing and applying that standard of review to

decision to admit hearsay evidence under the evidence code);

State v. Rogers, 330 Or 282, 315, 4 P3d 1261 (2000) (applying

that standard to review to question of whether an expert

is qualified to give testimony “relative to a particular topic”

(emphasis in original)).

As we will explain, the parties’ contentions hinge

on the interplay between four rules of evidence: Rule 602,

which sets out a general “personal knowledge” requirement

for witness testimony; Rules 702 and 703, which describe

special rules for a witness testifying as a qualified expert;

and Rule 802, which specifies a default exclusionary rule for

hearsay testimony.6 We describe those rules in more detail

before describing the parties’ arguments.

5

The Court of Appeals declined to reach, as unpreserved, defendant’s sep-

arate challenge that Wertz’s testimony about HGN was scientific evidence for

which the state had failed to lay an adequate foundation. On review in this court,

defendant does not challenge that aspect of the Court of Appeals’ decision. Nor

does she assert a violation of her constitutional right to confront witnesses as a

separate basis for reversal.

6

The rules are codified at ORS 40.315 (Rule 602); ORS 40.410 (Rule 702);

and ORS 40.415 (Rule 703); ORS 40.455 (Rule 802).

Cite as 373 Or 213 (2025) 221

Rule 602, as noted, sets out a “personal knowledge”

requirement, but with a noted exception:

“Subject to the provisions of [Rule 703], a witness may not

testify to a matter unless evidence is introduced sufficient

to support a finding that the witness has personal knowl-

edge of the matter.”

Rule 703 in turn, together with Rule 702, describe

special rules that govern the testimony of a qualified expert.

Rule 702, which governs the admissibility of expert testi-

mony, generally, provides:

“If scientific, technical or other specialized knowledge will

assist the trier of fact to understand the evidence or to

determine a fact in issue, a witness qualified as an expert

by knowledge, skill, experience, training or education may

testify thereto in the form of an opinion or otherwise.”

Rule 703, which pertains to the basis of expert testimony,

provides:

“The facts or data in the particular case upon which an

expert bases an opinion or inference may be those perceived

by or made known to the expert at or before the hearing. If

of a type reasonably relied upon by experts in the particu-

lar field in forming opinions or inferences upon the subject,

the facts or data need not be admissible in evidence.”

And both types of testimony—expert or otherwise—

are subject to the limitation of Rule 802, which generally

precludes a witness from offering “hearsay”—meaning “a

statement, other than one made by the declarant while tes-

tifying at the trial or hearing, offered in evidence to prove

the truth of the matter asserted.” OEC 801(3). That prohibi-

tion broadly extends to both oral and written assertions as

well as nonverbal conduct if intended as an assertion. OEC

801(1). Rule 802 provides that “[h]earsay is not admissible

except as provided in [Rules 801 to 806 of the evidence code]

or as otherwise provided by law.”

Defendant argues that Wertz’s testimony about

child development or eye anatomy was not admissible as

Wertz’s “personal knowledge” under Rule 602 and that there

is no basis in the record to conclude that Wertz’s expert qual-

ifications permitted him to testify to those matters under

222 State v. Bowman

Rule 702. She also disagrees with the conclusion of the trial

court and Court of Appeals that the testimony was admis-

sible under OEC 703 for the nonhearsay purpose of explain-

ing Wertz’s expert opinion. According to defendant, Wertz’s

testimony about information that he had learned from an

ophthalmologist about the visual acuity of babies had not

provided a basis for his expert opinion; rather, it had been

relevant only to prove the truth of the matter asserted. As a

result, defendant contends, that testimony was inadmissible

hearsay under OEC 802. Defendant further contends that

the error in allowing the jury to hear the challenged testi-

mony was not harmless because the opinion that a person

showing HGN has visual acuity that has regressed to the

functioning of a baby could persuade the jury to find impair-

ment in a way that Wertz’s own opinions did not.

The state counters that OEC 702 treats an expert’s

testimony about knowledge acquired outside of the court-

room as not hearsay. The state primarily urges this court

to view the challenged testimony as part of Wertz’s expert

knowledge and, thus, admissible for the truth of the matter

asserted. But the state also argues that OEC 703 permits

an expert to base opinions or inferences on information as to

which the expert may lack personal knowledge and permits

the expert to share that information for the “non-hearsay

purpose of explaining the basis of the expert’s opinion.” That

argument echoes the reasoning of the trial court that Wertz

was “trying to explain the basis for the opinion.” Finally, the

state contends that, in any event, any error in admitting the

challenged testimony was harmless and, therefore, provided

no basis for reversing the convictions.

B. Expert Testimony Under the Oregon Evidence Code

To address the parties’ contentions, we interpret

and explain how the various rules of evidence apply in this

context. Because the rules are adopted by the legislature,

we interpret their meaning through our traditional method

of statutory interpretation, by considering the relevant text,

context, and legislative history. Gollersrud v. LPMC, LLC,

371 Or 739, 745, 541 P3d 864 (2023) (citing State v. Gaines,

346 Or 160, 171-72, 206 P3d 1042 (2009)). The Oregon

Evidence Code was adopted in 1981, and the principal

Cite as 373 Or 213 (2025) 223

source to which this court has turned for legislative his-

tory is the 1981 Conference Committee Commentary. State

v. Serrano, 346 Or 311, 324, 210 P3d 892 (2009);7 Or Laws

1981, ch 892. We begin with an overview of how the Code

addresses expert testimony generally.

Under Rule 702, a witness who is “qualified as an

expert by knowledge, skill, experience, training or education”

may testify about subjects within their areas of expertise, “in

the form of an opinion or otherwise.” The party offering the

testimony of an expert bears the burden of establishing that

the expert possesses the requisite qualifications to provide

an opinion on a particular matter. Myers v. Cessna Aircraft,

275 Or 501, 520, 553 P2d 355 (1976). And an expert’s testi-

mony is admissible only if “scientific, technical or other spe-

cialized knowledge will assist the trier of fact to understand

the evidence or to determine a fact in issue.”8 OEC 702.

More pertinent to this case, however, proper appli-

cation of Rule 702 also requires an “assessment of the par-

ticular qualifications” of the testifying witness. Rogers, 330

Or at 316. As we have previously observed, “[n]o expert is

competent to express an opinion on every subject,” so “[t]

he object is to be sure that the question to the witness will

be answered by a person who is fitted to answer it.” Myers,

275 Or at 521 (internal quotation marks omitted), cited in

Rogers, 330 Or at 315.

Identifying the scope of an expert’s specialized

knowledge is critical, because, as described above, Rule

602 generally excludes testimony on any “matter” about

which the witness lacks “personal knowledge.” If, however,

a qualified expert is testifying to “specialized knowledge”

that is expressly admissible under Rule 702, then nothing

7

We explained in Serrano that, although the 1981 Conference Committee

Commentary is not an official part of the Oregon Evidence Code, “it nonetheless

provides highly useful background regarding each rule and guidance to courts

and attorneys in interpreting these rules.” 346 Or at 324 (internal quotation

marks omitted).

8

We have explained that the requirement in Rule 702 that expert testimony

must “assist the trier of fact” involves an inquiry into whether the testimony will be

helpful. See State v. Jesse, 360 Or 584, 593-94, 385 P3d 1063 (2016) (discussing what

is referred to as the “helpfulness inquiry” under Rule 702). To be helpful, the foun-

dation for an expert’s opinion must “intelligibly relate the testimony to the facts” of

the case. Id. at 594 (quoting State v. Brown, 297 Or 404, 409, 687 P2d 751 (1984)).

224 State v. Bowman

in Rule 602 makes that testimony inadmissible.9 And Rule

602 expressly is subject to the provisions of Rule 703. As

the Legislative Commentary explains, Rule 703 allows

an expert to “express opinions based on facts of which the

expert does not have personal knowledge,” and the express

carve-out in Rule 602 for Rule 703 is “designed to avoid any

question of conflict” between the two rules.10 Legislative

Commentary to OEC 602, reprinted in Laird C. Kirkpatrick,

Oregon Evidence § 602.02, 478-80 (7th ed 2020) (hereafter

Kirkpatrick on Oregon Evidence).

Identifying the scope of an expert’s specialized

knowledge also is critical because expert testimony remains

subject to the limitations that the Oregon Evidence Code

places on hearsay. The state contends, and we agree, that

“hearsay” does not include “scientific, technical or other spe-

cialized knowledge” that is admissible under OEC 702, even

if the expert acquired their knowledge through assimilating

information expressed by others outside of the courtroom.

Rule 702 expressly contemplates that an expert may acquire

specialized knowledge through “training or education,” both

of which involve absorbing information conveyed by others,

and may convey that knowledge either directly or in the

form of an opinion. See Legislative Commentary to OEC

702, reprinted in Kirkpatrick on Oregon Evidence § 702.02

at 634 (describing intent that Rule 702 allows both expert

testimony in the form of an opinion and expert testimony

in the form of a “dissertation or exposition” on “scientific

or other principles relevant to the case,” while “leaving the

9

The state understands the evidence code to treat an expert’s “scientific,

technical or other specialized knowledge” as part of the expert’s “personal knowl-

edge.” But, Rule 602 notwithstanding, it is unnecessary to endorse that expanded

concept of “personal knowledge” to conclude that the evidence code clearly per-

mits experts to testify to a matter within the scope of their “scientific, technical

or other specialized knowledge.”

10

Although Rule 703 does not specify what qualifies as “facts or data,” the

Legislative Commentary suggests that the legislature intended that phrase to

include opinions offered by a different expert if of the type reasonably relied

upon by experts in the field of the testifying expert when forming opinions. See

Legislative Commentary to OEC 703, reprinted in Laird C. Kirkpatrick, Oregon

Evidence § 703.02, 688 (7th ed 2020) (explaining that, under Rule 703, a doctor

testifying about a patient’s diagnosis may rely on “opinions from nurses, tech-

nicians and other doctors”); see also id. § 703.03[2] at 690 (asserting that the

Commentary “makes clear that an expert opinion may also be based in part upon

the opinions of other experts”).

Cite as 373 Or 213 (2025) 225

trier of fact to apply them to the facts”). And, as this court

has explained, “[i]t is clear that there can be no valid objec-

tion to the fact that a witness’s opinion rests upon hearsay in

the sense that the information [the expert] relies upon * * *

is derived in part from extrajudicial statements of others[,]”

because “[i]t is by assimilation of hearsay of this sort that

expert opinions are in fact, for the most part, made.” Jefferis

v. Marzano, 298 Or 782, 789-90, 696 P2d 1087 (1985).

But a witness’s qualification to testify as an expert

on a particular area of specialized knowledge does not make

the hearsay rules irrelevant. As described above, Rule 702

allows a witness to testify to “scientific, technical or other

specialized knowledge” only to the extent that the witness

is qualified as an expert with respect to that specialized

knowledge. Nothing in that rule permits the expert witness

to make assertions about matters beyond the scope of the

expert’s qualification merely because the witness heard the

information in an educational or training setting. And tes-

timony repeating such out-of-court assertions, if offered for

its truth, satisfies the definition of “hearsay.” See OEC 801.

In that sense, Rules 602 and 702 essentially func-

tion as a counterpoint to Rule 802: a witness expressing

their own personal or specialized knowledge often can do so

without repeating out-of-court statements for the truth of the

matter asserted and, thus, without directly implicating the

hearsay rules. But, if the witness’ testimony is merely a con-

duit for another person’s statement of personal or specialized

knowledge, then the testimony may be inadmissible hearsay

if offered for the truth of the matter asserted. For example,

a witness’ assertion that “the traffic light was green” can

be admissible as personal knowledge under Rule 602, if

the witness is describing what she observed with her own

senses. And because the testimony does not invoke an out-of-

court statement, it does not, on its face, trigger hearsay con-

cerns. But the same assertion—if offered for its truth—may

be inadmissible hearsay if the witness is merely repeating

something she learned from another observer at the scene.11

Similarly, an expert witness’s assertion that the traffic light

11

As the example illustrates, a question in aid of objection may be needed to

reveal whether testimony offered for its truth is a matter about which the wit-

ness possesses personal knowledge or if the assertion is potentially inadmissible

226 State v. Bowman

was programmed to turn green on a particular cycle can be

admissible as specialized knowledge under Rule 702, if the

witness is qualified as an expert on that subject. But the

same assertion—if offered for its truth—may be inadmissi-

ble hearsay if the witness is not qualified as an expert in the

functioning of traffic lights.

Turning to Rule 703, that rule expressly permits

experts to base their opinions on hearsay and other inad-

missible facts and data (if “of a type reasonably relied upon

by experts in the particular field”). For example, in the

scenario above, Rule 703 might permit an expert in acci-

dent reconstruction to rely on the assertion of a person at

the scene when forming an opinion about why the collision

occurred. But Rule 703 does not “render otherwise inadmis-

sible evidence admissible merely because it was the basis for

the expert’s opinion.” McCathern, 332 Or at 70. Thus, when

that “otherwise inadmissible evidence” consists of another

expert’s assertions that are being offered for the truth of

the matter, i.e., as hearsay, Rule 703 does not render that

evidence admissible “merely because it was the basis for

the expert’s opinion.” See Kirkpatrick on Oregon Evidence

§ 703.03[3] at 690-91 (emphasizing that “Rule 703 does not

allow one expert simply to be a ‘conduit’ for the opinions of

others”). As Professor Kirkpatrick explains:

“If the testifying expert is relying entirely on the conclu-

sions reached by other experts or technicians, such tes-

timony should not be allowed because it would simply be

parroting hearsay statements of third persons who are not

subject to cross-examination.”

Id.

Defendant points to this court’s decision in

McCathern as illustrating the line that the Kirkpatrick

quote is drawing under Rule 703, and we agree. The plaintiff

in McCathern had been injured when the Toyota 4Runner in

which she had been riding rolled over in an accident and

later brought a products liability action alleging that the

vehicle had been dangerously defective because of a design

that rendered it prone to rolling over. 332 Or at 62. Among

hearsay. Because the two concepts are intertwined, such testimony may draw

objections under both Rule 602 and the hearsay rules.

Cite as 373 Or 213 (2025) 227

other expert testimony to prove that allegation, the plain-

tiff presented testimony from a forensic engineer who testi-

fied that he had reviewed or investigated dozens of rollover

accidents involving the vehicle and concluded that approx-

imately 20 were “substantially similar” to the accident in

which the plaintiff was injured. Id. at 67. The trial court had

allowed the expert to briefly describe some of the data on

which he had based his opinion, over the defendant’s hear-

say objection. Id.

In affirming the trial court’s decision to allow the

expert a limited opportunity to explain the basis for his

opinion, we emphasized that the trial court had limited the

expert to describing the data “only to provide the foundation

necessary to explain [the expert’s] opinions, not for its truth.”

Id. at 70. Because the trial court had “admitted the evidence

solely for that purpose,” we concluded, “by definition, that

information was not ‘hearsay.’ ” Id. We agree with defendant

that implicit in the reasoning of McCathern is the princi-

ple that Rule 703 does not exempt expert witnesses from

the general prohibition against hearsay; an expert may not

merely parrot the statement of another for its truth.12 And

Rule 703 does not make hearsay admissible.

Thus, defendant’s objection that Wertz’s testimony

about the functioning of a baby’s eyes was inadmissible as

“hearsay,” and also was testimony for which the state had

laid no foundation, put at issue all of the evidence rules

described above. Had the challenged assertions been in the

form of a statement of Wertz’s personal knowledge admis-

sible under Rule 602, or in the form of a statement of his

specialized knowledge admissible under Rule 702, then the

testimony would not have been inadmissible hearsay even

though offered for the truth of the matter asserted. Or, had

Wertz offered the assertions about baby vision solely for the

12

Kirkpatrick cautions that, even when an expert is disclosing underlying

facts and data solely to explain a basis for the expert’s opinion, there is a danger

that “the jury will consider the statements for their truth rather than as a means

of evaluating the soundness of the expert’s opinion” and, thus, that any probative

value to such explanation should be weighed against the danger of “unfair prej-

udice, confusion of the issues, or misleading the jury under Rule 403,” and the

opponent should be entitled upon request to a limiting instruction. Kirkpatrick on

Oregon Evidence §703.03[4] at 691; see McCathern, 332 Or at 71-72 (emphasizing

that the record demonstrated that the trial court had complied with OEC 403).

228 State v. Bowman

purpose of explaining the basis for his opinion, and not for

the truth of the matter, then the testimony might well have

been admissible for that purpose; if so, it would not have

been hearsay subject to exclusion under Rule 802. We turn

to those questions in the context of this case.

C. Application of Evidentiary Rules to this Case

As summarized above, defendant challenges as

inadmissible part of Wertz’s testimony about the effect that

alcohol consumption can have on the appearance, movement,

and functioning of a person’s eyes. Although defendant does

not dispute that Wertz was qualified to offer opinion and

other testimony about his roadside evaluation of defendant’s

impairment, she contends that Wertz exceeded the bounds

of those qualifications, and offered hearsay, when the prose-

cutor prompted him to discuss the proposition that a person

“with enough alcohol” on board could be said to be seeing

“just like a baby.” The reasoning of the courts below, and the

state’s arguments on review, present a variety of possible

answers to defendant’s objection—that the challenged tes-

timony was admissible as expert testimony under Rule 702,

that the challenged testimony was admissible as an expla-

nation of Wertz’s expert opinions under Rule 703, or that

the challenged testimony was simply “not hearsay” because

Wertz had not “repeat[ed] any statement.” See Bowman, 326

Or App at 570. But on this record, none of those conclusions

is permissible.

Before explaining why we agree with defendant

that Wertz’s testimony about the functioning of a baby’s eyes

was not admissible under Rule 702, we briefly describe the

specialized knowledge regarding eye function that undis-

putedly fell within the scope of Wertz’s expert qualifications.

As described above, the state demonstrated that Wertz was

qualified as an expert at detecting impairment on the road,

including through the administration and evaluation of an

HGN test. We addressed the HGN test in detail in State v.

O’Key, in which we explained that HGN test evidence qual-

ifies as scientific evidence—because it “purports to draw

its convincing force from a principle of science, namely, the

asserted scientific proposition that there is a causal rela-

tionship between consumption of alcohol and the type of

Cite as 373 Or 213 (2025) 229

nystagmus measured by the HGN test.” 321 Or at 296. We

concluded that the “general proposition supporting HGN test

evidence—that alcohol consumption causes nystagmus—is

scientifically valid.” Id. at 319. And we further concluded

that law enforcement officers can be trained to observe hor-

izontal gaze nystagmus with sufficient precision to detect

alcohol impairment. Id. at 316-17. Ultimately, we held in

O’Key that a law enforcement officer’s testimony about a

driver’s performance on the HGN test is generally admissi-

ble in a DUII prosecution to “establish that a defendant was

under the influence of intoxicating liquor,” but not to “prove

that a defendant had a BAC of .08 percent or more.” Id. at

322-23. Additionally, the admissibility of HGN test evidence

is “subject to a foundational showing that the officer who

administered the test was properly qualified, that the test

was administered properly, and that the test results were

recorded accurately.” Id. at 289.

As should be apparent from the foregoing discussion,

much of Wertz’s testimony about the HGN test was admis-

sible under the rules of evidence and O’Key. For example, as

a trained law enforcement officer and certified DRE, Wertz

was qualified by his training and experience to testify that,

in his opinion, defendant had been impaired by alcohol. And

Wertz was permitted to explain the basis for that opinion

by describing what he had observed when he administered

the HGN test to defendant and how those observations were

consistent with what he had learned in training about the

relationship between the type of horizontal gaze nystagmus

that he had observed in defendant and alcohol consumption.

But Wertz’s testimony went further when the pros-

ecutor prompted him to talk about the vision of a baby and

to tell the jury that a person “with enough alcohol on board”

could be described as “seeing just like a baby.” The state

does not contend that Wertz was qualified as an expert on

the functioning of the eyes of a baby, and nothing in the

record would support such a contention. Thus, we disagree

with the state’s contention on review that Rule 702 made

Wertz’ testimony on that matter not hearsay.

Nor do we agree with the state’s contention that

the trial court correctly allowed the testimony under OEC

230 State v. Bowman

703 because Wertz was “trying to explain the basis” for his

opinion, such that his testimony was not hearsay. Wertz’s

permissible expert opinion was that defendant had been

impaired by her consumption of alcohol. And he permissibly

explained that he had based that opinion on his observa-

tion of defendant’s eyes during the HGN test and her per-

formance during the other two field sobriety tests. Wertz

also explained the foundational basis for his opinion that

the HGN observations demonstrated impairment was that

he had (1) been trained on the proper administration of the

HGN test; (2) learned in that training that alcohol con-

sumption causes the same type of nystagmus that he had

observed in defendant’s eyes when he had administered

the HGN test to her; and (3) administered the HGN test to

defendant properly and accurately recorded the results of

that test.

But Wertz at no point suggested that, in forming

his own opinion that defendant had been impaired by the

consumption of alcohol, he had relied on Citek’s proposi-

tion that babies must turn their heads to focus on objects in

their peripheral vision or that “a person with enough alco-

hol on board could be said to be seeing just like a baby, or

they are not able to even look at something independent of

moving their head.” If anything, those propositions are at

odds with the observations on which Wertz had based his

opinion, because he specifically testified that defendant did

follow the pen from side to side without turning her head

during the HGN test. And the state does not even attempt

to argue that information about the functioning of a baby’s

eyes qualifies as facts or data “reasonably relied upon by”

DRE experts in detecting impairment on the road, as Rule

703 requires.

Moreover, even assuming that such information

did qualify as a type that a DRE expert might reasonably

rely upon to form opinions about a driver’s impairment,

Wertz did not solely repeat what he had heard about the

functioning of a baby’s eyes for the purpose of explaining

his own impairment opinion. Rather, the prosecutor elicited

and relied upon that testimony for the truth of the matter.

The prosecutor repeatedly prompted Wertz to repeat what

Cite as 373 Or 213 (2025) 231

the prosecutor had “heard other officers describe” about the

eyes of a person with HGN present having regressed to “like

being a child, or being a baby,” finally announcing “[t]hat’s

what I wanted to get to” when Wertz eventually described

that concept. And later in closing argument, the prosecu-

tor pointed to Wertz’s testimony as if the proposition were

true, emphasizing: “We don’t want to be regressing back to

when we were a child and we have to turn our head to focus

on things. We want peripheral vision.” As explained above,

Rule 703 does not allow an expert to violate the prohibition

on hearsay set out in Rule 802. 373 Or at 227.

Finally, we disagree with the Court of Appeals that

the testimony was admissible simply because Wertz ulti-

mately offered the challenged testimony without explicitly

describing it as a statement (or statements) that he had

heard from Citek. Defendant contends that, even if Wertz

did not expressly describe the information in question as a

statement previously made by Citek, he effectively repeated

the substance of Citek’s teaching for its truth, in violation

of Rule 802. Defendant’s understanding of “hearsay” is cor-

rect. A “statement” for purposes of the rule refers to any

“oral or written assertion.” OEC 801(1). And the assertion

is “hearsay” if made by a declarant other than “while testi-

fying at the trial or hearing.” OEC 801(3). Significantly, the

definition of hearsay is not limited to assertions presented

with “air quote” gestures or introduced by the explanation “I

heard her say.” Although those classic cues may make it eas-

ier to identify possible hearsay, the definition is focused on

whether the assertion was made other than by the declarant

at trial (and for the truth of the matter asserted).

We have already explained that Wertz’s testimony

about the functioning of a baby’s eyes did not reflect his

own specialized knowledge. And it is clear from Wertz’s

exchange with the prosecutor that Wertz was repeating

assertions that Citek had made during trainings that Wertz

had attended. Wertz explained that “it’s something that

[Citek] teaches,” and he made the comparison immediately

before stating that that is “the way that [Citek] describes

the effects of alcohol on the eyes to students” in the DRE

class. The assertions did not become any less hearsay when

232 State v. Bowman

Wertz eventually repeated Citek’s assertions without explic-

itly describing them as what Citek “teaches” or “describes.”

Thus, the trial court erred when it overruled defendant’s

hearsay objection.

D. Harmless Error

Determining that the trial court erred is not the

end of the inquiry, however, because an error in admitting

evidence “does not require reversal if it is harmless—that

is, if it had little likelihood of affecting the verdict.” State v.

Henley, 363 Or 284, 307, 422 P3d 217 (2018). In general, “[a]

n evidentiary error is more likely to influence a verdict if

the error relates to a central factual issue in this case, and

is less likely to influence a verdict if it relates to a tangen-

tial issue.” State v. Bement, 363 Or 760, 779, 429 P3d 715

(2018) (internal citations omitted); see also State v. Willis,

348 Or 566, 572-73, 236 P3d 714 (2010) (emphasizing that

erroneously admitted evidence “went to “the heart of * * *

the case” (quoting State v. Davis, 336 Or 19, 34, 77 P3d 1111

(2003)). Other factors this court considers include whether

erroneously admitted (or erroneously excluded) evidence

“was ‘qualitatively different’ from the other evidence on that

point, rather than ‘merely cumulative.’ ” State v. Edmonds,

364 Or 410, 430, 435 P3d 752 (2019) (quoting Davis, 336 Or

at 34).13

With those general principles in mind, we turn to

the particular record in this case. Defendant contends that

the erroneously admitted testimony—Wertz’s comparison to

driving with vision that functioned like a baby’s—had more

than little likelihood of affecting the jury’s verdict on all

three counts, and we agree. The opinion that defendant’s con-

sumption of alcohol may have caused her peripheral vision

and focus to regress was central—not tangential—to the

state’s theory that defendant was guilty of DUII because she

had driven while under the influence of intoxicating liquor,

which in turn was central to the state’s theory of reckless

driving. The jury was instructed that “under the influence

13

The error in Edmonds consisted of erroneously admitting evidence, but

this court focused on the same considerations described in Davis, in which the

error consisted of erroneously excluding evidence. Edmonds, 364 Or at 430;

Davis, 336 Or at 34.

Cite as 373 Or 213 (2025) 233

of intoxicating liquor” meant that defendant’s “physical or

mental faculties were adversely affected by the use of intox-

icating liquor to a noticeable or perceptible degree.” And

Wertz’s testimony that the consumption of alcohol causes a

person’s vision to regress to limitations of a baby’s vision

went to that central issue.

The state does not contend otherwise, but it contends

that Wertz’s testimony was cumulative of, and not qualita-

tively different from, Wertz’s unchallenged testimony that

defendant had been impaired by the consumption of alcohol.

We disagree. Expert testimony—especially testimony that

is scientific in nature—must be closely scrutinized by trial

courts, because such testimony may have “an unusually

high degree of persuasive power” to jurors. O’Key, 321 Or at

291; see also Henley, 363 Or at 298 (noting that the concern

about scientific evidence is that a jury might “accord it signif-

icant persuasive value”). As we have explained, Wertz was

describing scientific principles when he testified to the effect

that alcohol consumption has on a person’s eyes. And when

the prosecutor prompted Wertz to also compare that effect

to the vision of a baby, Wertz described scientific concepts

that he attributed to an ophthalmologist and that went far

beyond the opinions that Wertz’s own expertise could justify.

To reiterate, after the trial court overruled multi-

ple objections, Wertz ultimately testified that babies lack

peripheral vision, because “they have to turn their entire

head” to see, and that alcohol “seems to cause a person to

regress backwards in that process, so * * * a person with

enough alcohol on board could be said to be seeing just like

a baby.” And he introduced those concepts in response to

the prosecutor’s question about descriptions of vision limita-

tions “with [HGN] present.” In other words, the testimony

allowed the jury to infer that defendant’s HGN meant that

she had “enough alcohol on board” to be “seeing just like a

baby” in terms of her peripheral vision.

That erroneously admitted hearsay, in addition to

having a potentially “high degree of persuasive power,” also

was evidence that the prosecutor highlighted to persuade

the jury in closing argument:

234 State v. Bowman

“We don’t want to be like Trooper Wertz talked about. We

don’t want to be regressing back to when we were a child

and we have to turn our head to focus on things. We want

peripheral vision. If somebody’s walking out in the middle

of the road we want to be able to see that. A child that’s

walking out * * * we want to be able to see that[.]”

We have emphasized in prior cases that “a party’s intro-

duction of facts not in evidence during closing arguments

can substantially affect the opposing party’s rights.” Cler

v. Providence Health System-Oregon, 349 Or 481, 491, 245

P3d 642 (2010). And the emphasis on facts that should not

have been in evidence presents similar concerns. The clos-

ing argument here not only emphasized scientific principles

that should not have been in evidence, it also encouraged

the jury to infer from those principles that defendant’s con-

sumption of alcohol, which was sufficient to cause observ-

able nystagmus, had caused her peripheral vision to regress

to that of a baby that must turn its head to focus. Given

the nature of the testimony and the prosecutor’s use of the

evidence in closing argument, the erroneously admitted evi-

dence was qualitatively different from the permissible evi-

dence of impairment and not merely cumulative.

The state also contends that the impairment evi-

dence was harmless because the state presented an inde-

pendent and alternative theory of impairment—that defen-

dant had driven with a blood alcohol level of at least .08.14

According to the state, the evidence to support that theo-

ry—a breath test performed 90 minutes later—was “signif-

icant and virtually unchallenged.” But the generic verdict

form does not reveal whether the jury was persuaded by the

state’s 0.08 theory of intoxication or by the general evidence

of impairment (tainted by the comparison to the vision of

14

At the time of defendant’s offense, ORS 813.010 defined DUII in part as

follows:

“(1) A person commits the offense of driving while under the influence of

intoxicants if the person drives a vehicle while the person:

“(a) Has 0.08 percent or more by weight of alcohol in the blood of the per-

son as shown by chemical analysis of the breath or blood of the person * * *;

[or]

“(b) Is under the influence of intoxicating liquor, cannabis, a controlled

substance or an inhalant[.]”

ORS 813.010 (2019).

Cite as 373 Or 213 (2025) 235

a baby). Nor does any other aspect of the record reveal the

jury’s basis for its decision.

Moreover, when making a determination of harm-

lessness, we do “not ask whether the evidence of guilt is sub-

stantial or compelling.” Henley, 363 Or at 307. Rather, we

consider the likelihood that the evidence affected the ver-

dict. Id. Here, contrary to the state’s suggestion, defendant

disputed that her BAC was at least 0.08 at the time when

she was driving. As we have explained, the results of chem-

ical analysis of a person’s breath through a later test “alone

never ‘shows’ the actual BAC of the driver at the time of

driving,” so expert extrapolation from those test results is

used “prove the amount of alcohol in the person’s blood at

the time of driving.” State v. Eumana-Moranchel, 352 Or 1,

9, 14, 277 P3d 549 (2012); see State v. Hedgpeth, 365 Or 724,

744, 452 P3d 948 (2019) (explaining that something more

than a “jury’s common knowledge of the generic (and incom-

plete) proposition that alcohol dissipates from the blood over

time is not sufficient” to permit a “nonspeculative connec-

tion” between a defendant’s BAC when he was driving and

test results taken two hours later showing a BAC of 0.09).

Defendant challenged the methodology by which the state’s

expert had extrapolated defendant’s BAC at the time she

was driving, and the expert acknowledged the possibility

that defendant’s BAC could have been lower at that time.

Given the record, we are unable to conclude that the erro-

neously admitted comparison to driving with the vision of a

baby had little likelihood of affecting the verdict.

The dissent proposes a different reason that the

inadmissible testimony was harmless. The dissent points to

the very reasons that we identified in explaining why the

trial court erred in concluding that the evidence “explain[ed]

the basis for” Wertz’s opinion, to argue that the evidence

was harmless—because it in no way explained Wertz’s opin-

ion that defendant had been impaired and, therefore, would

not have been “helpful” to the jury in assessing Wertz’s tes-

timony. But emphasizing the reasons that the court erred in

admitting the evidence does not explain why the error had

little likelihood of affecting the verdict. As discussed above,

the proposition that babies must turn their heads to focus

236 State v. Bowman

on objects in their peripheral vision does not explain Wertz’s

opinion, which was based on defendant’s performance

during the HGN test for which she did follow an object with-

out turning her head. But we are unable to accept the dis-

sent’s premise that jurors would have recognized the logical

flaw and simply disregarded the prosecutor’s invitation to

infer that HGN observations indicate that a person’s vision

has regressed to that of a baby. Indeed, the reason that we

impose on trial courts important gatekeeping requirements

regarding scientific evidence is precisely because lay jurors

are more likely to accept that evidence as persuasive with-

out scrutinizing it for logical flaws.

Nor are we persuaded by the dissent’s comparison

of this case to cases like State v. McAnulty, 356 Or 432,

338 P3d 653 (2014). In McAnulty, the defendant had made

admissions about the abuse of her daughter in a series

of interrogations, and this court held that the trial court

should have suppressed statements made during the earlier

interrogations, but not during the later interrogations. Id.

at 458-59. But, because the defendant ultimately entered an

unconditional plea of guilty, this court on review considered

the asserted error only as it related to the penalty phase.

Id. at 448. In that limited context, given both the defen-

dant’s plea—in which she had admitted to the “intentional

maiming and torturing” of her daughter—and the properly

admitted, “more substantial” later admissions of abuse, we

concluded that the jury “would have regarded the improp-

erly admitted evidence as duplicative or unhelpful” to the

question of punishment. Id. at 461.

Here, the prosecutor was determined to elicit from

Wertz the testimony that “a person with enough alcohol on

board could be said to be seeing just like a baby,” despite

repeated objections to alert the prosecutor that admitting

the testimony would create a potential error for appeal.

That determination, in combination with the prosecu-

tion’s explicit reliance on that hearsay assertion in closing

argument, persuades us that the prosecutor—unlike the

dissent—believed the evidence was important to proving

the case. See State v. Thompson, 370 Or 273, 301, 518 P3d

923 (2022) (in evaluating whether evidentiary error was

Cite as 373 Or 213 (2025) 237

harmless, looking to whether the parties’ arguments com-

municated that erroneously admitted statements “were

not considered significant by either party”). In light of the

unusually high persuasive power of scientific evidence, and

the fact that the inadmissible opinion in this case went to

a central theory and was qualitatively different from the

other evidence of impairment that the jury considered, we

are unable to conclude that the erroneously admitted evi-

dence, and the inferences that the jury could draw from it,

“were unlikely to influence the jury’s verdict.” See Bement,

363 Or at 781 (explaining that the erroneous exclusion of

emails was not harmless because “we cannot say that those

additional facts and inferences were unlikely to influence

the jury’s verdict”).

III. CONCLUSION

In sum, we conclude that the challenged testimony

was inadmissible hearsay, because it was offered for the

truth of the matter asserted and merely repeated informa-

tion that was beyond the scope of Wertz’s own personal or

specialized knowledge. We further conclude that the trial

court’s admission of that hearsay evidence did not qualify

as harmless error. Accordingly, we reverse both the decision

of the Court of Appeals and the trial court’s judgment of

conviction.

The decision of the Court of Appeals is reversed.

The judgment of the circuit court is reversed, and the case

is remanded to the circuit court for further proceedings.

BUSHONG, J., dissenting.

The majority opinion concludes that one aspect

of Officer Wertz’s testimony—that, according to an oph-

thalmologist who trained Wertz, alcohol consumption can

cause an adult’s eyes to function like those of a baby—was

inadmissible hearsay because it was offered and used for

the truth of that assertion. I agree. But I disagree with the

majority opinion’s conclusion that that evidentiary error

was not harmless. As I will explain, Wertz’s improper testi-

mony, when viewed in the context of this case, was unlikely

to have affected the jury’s verdict. In addition, the major-

ity opinion’s conclusion that the error was not harmless is

238 State v. Bowman

inconsistent with our prior cases that address when an evi-

dentiary error is harmless and when it is not. I begin with

Wertz’s testimony.

I. DISCUSSION

A. The Significance of Wertz’s Improper Testimony

I agree with the majority that the trial court erred

when it overruled defendant’s hearsay objection to Wertz’s

testimony that an ophthalmologist had taught him during

his training that the eyes of a person with “enough alcohol

on board” cannot track objects without moving their head,

comparable to how a baby’s eyes function during the early

stages of child development. But the nature of that error in

the context of this case reveals how unlikely it is that the

testimony could have affected the jury’s verdict, because it

was at most tangentially related to a relatively minor issue

in dispute.

Significantly, Wertz never testified that defendant

had been unable to track with her eyes without moving her

head. To the contrary, Wertz testified that he had adminis-

tered the horizontal gaze nystagmus (HGN) field sobriety

test to defendant twice, and both times, she had been able

to track with her eyes without moving her head. Thus, the

testimony comparing the eyes of a person with enough alco-

hol on board to those of a baby, as elicited by the prosecutor,

at most merely illustrated how alcohol can affect a person’s

vision in a hypothetical situation. Because that comparison

was not descriptive of what Wertz had observed in defen-

dant’s eyes when he administered the HGN test to her, it

was irrelevant and, as I will explain, it was also harmless.1

The prosecutor’s reference during closing argument

to that comparison—stating that “we don’t want [drivers]

regressing” to the point of having to turn their heads to see

something in their peripheral vision—had little likelihood

of affecting the jury’s verdict, because, again, the state had

presented no evidence that defendant had had to turn her

1

We have previously held that, to be relevant and helpful to the jury, expert

testimony must be “intelligibly related to a provable fact.” State v. Jesse, 360 Or

584, 596, 385 P3d 1063 (2016) (citing State v. Brown, 297 Or 404, 409, 687 P2d 751

(1984)).

Cite as 373 Or 213 (2025) 239

head to see something in her peripheral vision. The court’s

instruction that the arguments of counsel are not evidence

made it even less likely that the jury would have been led

astray by the prosecutor’s irrelevant reference in closing

argument. Moreover, the prosecutor’s reference to that testi-

mony during closing was so inconsequential that defendant’s

lawyer did not even mention it during his closing argument.

Wertz was testifying as a law enforcement offi-

cer—not as a medical professional—so the jury would not

have mistaken the ophthalmologist’s opinion for Wertz’s, nor

would it have given his testimony the “unusually high degree

of persuasive power” that it might have given to expert med-

ical testimony. State v. O’Key, 321 Or 285, 291, 899 P2d 663

(1995). The jury also would have understood that Wertz’s

reference to the ophthalmologist’s teaching that an adult

“with enough alcohol on board” cannot track with their eyes,

like a baby, did not apply in this case because the state had

presented no evidence that defendant had been unable to

track with her eyes.

The context in which that evidence was offered and

used at trial confirms, at least to me, that the erroneously

admitted evidence was, at most, tangentially related to a

minor issue in this case. The state’s main argument at trial

was that defendant was guilty of DUII because her blood

alcohol content (BAC) was .08 percent or more when she was

driving. The evidence supporting that argument included

Wertz’s testimony that the defendant’s BAC when she took

a breath test at the police station was .08 percent, and the

testimony of a police forensic scientist that, by extrapolating

backwards in time based on the dissipation rate of alcohol in

the blood system, defendant’s BAC likely had been between

.08 and .11 while she was driving. The closing arguments

focused on that evidence.

The evidence about how alcohol consumption can

affect a person’s eyes came up in the context of Wertz’s testi-

mony about how defendant had performed on the HGN test.

HGN evidence is generally admissible in a DUII prosecution

“to establish that a defendant was under the influence of

intoxicating liquor,” but not “to prove that a defendant had

a BAC of .08 percent or more.” O’Key, 321 Or at 323. Thus,

240 State v. Bowman

the HGN evidence—and the tangentially related testimony

about how alcohol consumption can cause a person’s eyes to

“regress” to being like those of a baby—could pertain only

to the state’s alternative theory that defendant was guilty

because she was under the influence of intoxicating liquor

when she was driving. And on that theory, there was ample

admissible evidence of defendant’s guilt, including testi-

mony about her poor driving, her slurred speech, the odor of

alcohol detected on her breath, and her poor performance on

three field sobriety tests.

Accordingly, I would conclude that the improp-

erly admitted evidence was not likely to have affected the

jury’s verdict, making the error harmless. That conclusion,

I believe, is consistent with our prior cases that analyzed

when evidentiary errors are harmless and when they are

not. I turn to those cases.

B. Harmless Evidentiary Errors

We concluded in three fairly recent cases that an

evidentiary error was harmless. Comparing the errors in

those cases with the error in this case helps illustrate why

this error was also harmless.

For example, we concluded in State v. Camarena,

344 Or 28, 176 P3d 380 (2008), that the trial court erred

in admitting into evidence certain out-of-court statements

made by a victim of domestic violence because the evidence

violated the defendant’s rights under the Sixth Amendment’s

Confrontation Clause, but we further concluded that the evi-

dentiary error was harmless. The defendant in that case had

been convicted of fourth-degree felony assault for punching

his live-in girlfriend in the face. At trial, in lieu of the vic-

tim’s live testimony, the court received as evidence a record-

ing of a conversation between the victim and a 9-1-1 operator.

On appeal, the defendant contended that the trial court had

erred because that evidence violated his rights under the

Confrontation Clause. Among other things, the victim had

told the 9-1-1 operator the defendant’s name and stated that

he was on probation for hitting her previously, but that she

did not want him to go to jail because he was her only source

of income. Id. at 31 (quoting the recorded 9-1-1 call). We

Cite as 373 Or 213 (2025) 241

concluded that those statements were testimonial because

they were “unnecessary to resolve an ongoing emergency”

and were “directed at establishing facts only relevant to a

subsequent criminal action.” Id. at 41. Thus, the trial court

erred in admitting those statements. However, we further

concluded that the error was harmless because “the 9-1-1

statements that were nontestimonial, and therefore admis-

sible, were of similar content and sufficient to establish the

elements of the charge against [the] defendant.” Id. at 42.

The defendant in State v. McAnulty, 356 Or 432, 338

P3d 653 (2014), pled guilty and was convicted of aggravated

murder after her 15-year-old daughter died from physical

abuse. A jury sentenced the defendant to death after hear-

ing evidence presented at the penalty-phase proceedings. We

determined that certain statements made by the defendant

during a police interrogation should have been suppressed

and were improperly received into evidence during those

proceedings. Id. at 457. The erroneously received evidence

included the defendant’s admissions that she had “ ‘spanked’ ”

her daughter multiple times with either a belt or stick, had

“controlled and limited” her daughter’s water supply, had

“cleaned up some evidence of abuse,” had “caused an injury

that had exposed * * * bone,” and had “attempted to treat [the]

injuries herself.” Id. at 461. In determining whether the erro-

neous admission of that evidence was harmless, we examined

“the effect of the improperly admitted evidence in light of the

admissions that were properly admitted and the [defendant’s]

guilty plea[.]” Id. We concluded that the error was harmless

because “the jury would have regarded the improperly admit-

ted evidence as duplicative or unhelpful.” Id.

In State v. Thompson, 370 Or 273, 518 P3d 923

(2022), the defendant was convicted of first-degree rob-

bery after he and an accomplice had used a knife to rob

the victim. The victim had resisted and ultimately shot the

defendant during the robbery. The police seized the defen-

dant’s cell phone while he was being treated for the gun-

shot wound and eventually obtained a warrant to search it.

The resulting search revealed records of calls and messages

related to the robbery and shooting. The police then used

that evidence in questioning the defendant. At trial, the

242 State v. Bowman

state offered statements made by the defendant during that

interrogation into evidence, in part to rebut the defendant’s

theory that his accomplice, not the defendant, had wielded

the knife. We concluded on appeal that “at least the three

interview statements that [the] defendant challenge[d]”

should have been suppressed. Id. at 292. And we noted that

the state had made a “brief reference” to those statements

during its closing argument. Id. at 301. But we nevertheless

concluded that the trial court’s error in failing to suppress

them was harmless. Id. at 302. In reaching that conclu-

sion, we rejected the defendant’s argument that the error

was not harmless because the jury could have been affected

by those statements when it found the defendant guilty of

the robbery. We explained that the inferences of guilt that

the defendant sought to draw from the improperly admitted

statements “are simply not there or are independently sup-

ported by unchallenged evidence[.]” Id.

Our harmless error analysis in Thompson was

based on “our review of the record and consideration of

[the] defendant’s statements” in the context of the case.

Id. Among other things, we noted that it would have been

“speculative to think that the jury would likely infer” that

one of the erroneously admitted statements—that the defen-

dant said he did not want the victim to “ ‘get in trouble’ ”

for shooting him—was effectively an admission that the vic-

tim “was acting to defend himself from [the] defendant,” not

from the accomplice, when he shot the defendant. Id. at 299-

300 (emphasis added). In addition, the erroneous admission

of the defendant’s statement that he “ ‘didn’t know’ ” if his

accomplice “ ‘had [the knife] on him or not’ ” was harmless,

because “the jury already knew” that the defendant’s theory

of the case was that the accomplice—not the defendant—had

wielded the knife. Id. at 298-300. We further noted, from

the evidence of all the statements that the defendant made

during the police interviews, that “[i]t would have been clear

to the jury” that the defendant’s statements “were often

evasive, vague, and ambiguous, but not otherwise incrim-

inating.” Id. at 301. And we stated that “the arguments

of the parties at trial would have confirmed for the jury”

that the erroneously admitted statements “were not consid-

ered significant by either party[.]” Id. For example, during

Cite as 373 Or 213 (2025) 243

closing arguments, instead of emphasizing the impermissi-

bly admitted statements, the parties “focused on whether the

victim’s extensive testimony about the robbery and shooting

and [the] defendant’s and [the accomplice’s] respective roles,

was credible.” Id.

The improperly admitted evidence in both Camarena

and McAnulty seems far more likely to have affected the

juries’ decisions in those cases than Wertz’s reference to

how a baby’s eyes function in this case. But in both of those

earlier cases, we concluded that the evidentiary error was

harmless when considered alongside the other evidence

establishing the defendant’s guilt. I do not understand how

we could conclude that the errors in those cases were harm-

less while the error in this case was not. Moreover, the fact

that the prosecutor referred to Wertz’s testimony in closing

argument is not enough, considering that we concluded in

Thompson that a brief reference to the improperly admit-

ted evidence during closing did not mean that the error was

harmful. In my view, the majority opinion’s conclusion that

the erroneously admitted testimony in this case could have

affected the jury’s verdict is even more speculative than the

inference at issue in Thompson.

In contrast with those cases, six other recent

cases concluded that evidentiary errors were not harmless.

Analyzing the evidentiary errors in those cases confirms

that the error in this case is different. I turn to those cases

next.

C. Harmful Evidentiary Errors

Our decision in State v. Davis, 336 Or 19, 77 P3d

1111 (2003), is the starting point for our modern harmless

error jurisprudence. There, we observed that our earlier

cases had not described the harmless error analysis “in a

consistent manner.” Id. at 28. Some of those earlier cases

had stated the test as consisting of a single inquiry, while

others had described a “bifurcated two-part formulation”

of the test. Id. at 31. We reiterated in Davis that “Oregon’s

constitutional test for affirmance despite error consists of a

single inquiry: Is there little likelihood that the particular

error affected the verdict?” Id. at 32.

244 State v. Bowman

Davis provides a good example of an evidentiary

error that is not harmless. The defendant in Davis was con-

victed of murdering his girlfriend. He contended on appeal

that the trial court erred in excluding the victim’s state-

ments showing that, “for years before her death, [the victim]

was obsessed with [the] defendant and threatened to kill

herself because of their relationship[.]” Id. at 26. The defen-

dant’s theory at trial was that the victim had committed

suicide. We concluded that the trial court erred in excluding

the statements and that the error was not harmless because

“the excluded evidence goes directly to the heart of [the]

defendant’s factual theory of the case.” Id. at 34.

The defendant in State v. Willis, 348 Or 566, 236

P3d 714 (2010), was charged with unlawful possession of

methamphetamine. We held that the trial court erred in

admitting a laboratory report without the testimony of the

criminalist who had prepared it. The report identified the

substance seized from the defendant as methamphetamine.

We concluded that the evidentiary error went to the heart of

the state’s case because the disputed evidence was the only

evidence the state had presented that the substance was, in

fact, methamphetamine. Id. at 573.

Another example of a harmful evidentiary error is

State v. Henley, 363 Or 284, 422 P3d 217 (2018). The defen-

dant in that case was convicted of first-degree sexual abuse

and attempted first-degree sodomy for sexually abusing his

11-year-old stepdaughter during a camping trip. At trial,

over the defendant’s objection, the trial court permitted

a forensic interviewer to testify that adults who sexually

abuse children often groom their victims before abusing

them and that the defendant’s earlier behavior towards the

victim was typical of this type of grooming. Id. at 290. We

concluded that the trial court had erred in admitting the

evidence because the disputed testimony was scientific evi-

dence that had been admitted without a proper foundation

establishing its validity. Id. at 304. We further concluded

that the error was not harmless “for at least three rea-

sons[.]” Id. at 308. First, the credibility of the defendant and

the victim “was a central factual issue, and the grooming

evidence bolstered [the victim’s] testimony.” Id. (internal

Cite as 373 Or 213 (2025) 245

quotation marks omitted). Second, the forensic interviewer’s

testimony was scientific evidence and, “as such, possessed

a high degree of persuasive power” that the prosecutor had

used during his closing argument. Id. Third, the jury “likely

viewed” the grooming testimony “as substantive evidence

that [the] defendant had sexually abused [the victim] in the

camper as charged.” Id. at 309. Without the “grooming” evi-

dence, we explained, “the prosecutor would have been left

with a much weaker argument to the jury[,]” and so we con-

cluded that the evidence was “likely influential in the jury’s

deliberations and verdict.” Id. at 309-10.

In State v. Bement, 363 Or 760, 429 P3d 715 (2018),

the defendant admitted that he had shot and killed the vic-

tim, but he contended that the shooting was in self-defense

when the victim had tried to rob him. The jury disagreed

and convicted the defendant of murder. On appeal, he con-

tended that the trial court had erred in excluding emails

showing that, at the time of the shooting, the victim had

been “in significant financial trouble and in desperate need

of money.” Id. at 762. The defendant contended that the

evidence of the victim’s financial troubles was relevant to

establish the victim’s motive for the robbery, thereby sup-

porting the defendant’s claim of self-defense. We concluded

that the trial court had erred in excluding that evidence

and that the error was not harmless because “the excluded

emails relate[d] to a central factual issue” in the case and

were not “merely cumulative” of other emails that had been

admitted into evidence. Id. at 780-81. To that latter point,

we explained that the excluded emails were not cumulative

because the admitted emails had provided “only a small pic-

ture of [the victim’s] financial distress” and had “fail[ed] to

convey the progression of [the victim’s] desperation.” Id. at

780.

The defendant in State v. Edmonds, 364 Or 410, 435

P3d 752 (2019), was charged with raping a child twenty years

before the trial. The victim, who was 24 years old at the time

of trial, testified that the defendant had raped her at a day

care facility when she was about five years old. The defen-

dant’s theory of the case was that the victim had a “false

memory” of the rape that had recently emerged. Id. at 430.

246 State v. Bowman

The state sought to support the victim’s testimony by intro-

ducing a transcript of a police interview that had occurred

in 2002 as part of an investigation of an unrelated sexual

offense. In that interview, the officer asked the victim if any-

one had previously touched her in a sexual way, and she

responded, “ ‘[w]hen I was at a babysitter’s but that was like

when I was really, really young.’ ” Id. at 413. We held that

the trial court had erred in admitting the transcript and

that the error was not harmless because the victim’s cred-

ibility was critical; the transcript rebutted the defendant’s

argument that the victim’s accusation had been based on a

false memory that had emerged only recently; and “[t]he fact

that the transcript predated the trial by 10 years cut at the

heart of that defense.” Id. at 430.

The defendant in State v. Skillicorn, 367 Or 464,

479 P3d 254 (2021), was convicted of reckless driving. We

held that the trial court had erred in receiving evidence of

the defendant’s bad driving on prior occasions because that

evidence amounted to character evidence that was inadmis-

sible under the particular evidentiary theory advocated by

the state. We further held that the erroneous admission of

that evidence was not harmless because it “related to the

central dispute in the case and, as used by the prosecutor,

carried significant risks that the evidence would prejudice

the jury against [the] defendant and be overvalued[.]” Id. at

495.

In summary, the evidentiary errors that we found

to be harmful in Davis and Edmonds went directly to the

heart of the defendant’s theory in those cases. The eviden-

tiary error in Willis involved the only evidence that the state

had presented relating to an element of the crime. The evi-

dentiary error in Henley involved the victim’s credibility,

and the error in Bement directly related to the defendant’s

claim of self-defense. Those were the key factual issues in

dispute in those cases. The “other acts” evidence erroneously

admitted in Skillicorn was highly prejudicial to the jury’s

perception of the defendant’s character.

In contrast, the evidence here was irrelevant and,

at most, only tangentially related to the state’s alterna-

tive theory that defendant had been driving under the

Cite as 373 Or 213 (2025) 247

influence of alcohol. That evidence was not highly prejudi-

cial to defendant, because it did not involve defendant at all.

Stated another way, we cannot say in this case, as we said

in Henley, that the prosecutor would have been “left with a

much weaker argument to the jury” without the testimony

explaining that the eyes of a person with “enough alcohol on

board” function like those of a baby.

II. CONCLUSION

In sum, the evidentiary error here was harmless,

because there was “little likelihood that the particular error

affected the verdict[.]” Davis, 336 Or at 32. That error is

more akin to evidentiary errors that we determined were

ultimately harmless in our prior cases, as opposed to those

errors determined to be harmful. The majority opinion is

wrong in concluding otherwise. Accordingly, I respectfully

dissent.

Garrett, J., joins in this dissenting opinion.