State v. Bowman

373 Or. 213

Opinion

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* Appeal from Lincoln County Circuit Court, Thomas O. Branford, Judge. 326 Or App 565 (2023) (nonprecedential memorandum opinion).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 qualified as an expert by their knowledge, skill, experience, training, or education to testify to that specialized knowledge “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 (specifying that hearsay is inadmissible absent a specific exception).

In this criminal case, we must decide whether the trial court correctly overruled defendant’s “hearsay” objection 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 administered during a DUII investigation, called the horizontal 216 State v. Bowman

gaze nystagmus (HGN) test.1 Wertz then had defendant perform 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 committing 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 administering 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 “nystagmus ,” 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 defendant , 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 movement 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 dissipation 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 standardized “clues” of impairment. Specifically, Wertz explained that he had observed in both eyes a “lack of smooth pursuit ” as defendant tracked the pen’s horizontal movement (the first two clues). He described that lack of smooth pursuit 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 alcohol 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 administered 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 defendant ’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 peripheral 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 physical and mental faculties when we’re operating a motor vehicle . 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 allowing 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 ophthalmologist , there was no hearsay.” State v. Bowman, 326 Or App 565, 570 (2023) (nonprecedential memorandum opinion ). 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 further 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 testimony 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 separate 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 knowledge 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 testimony , 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 particular 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 testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.” OEC 801(3). That prohibition 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 qualifications 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 admissible under OEC 703 for the nonhearsay purpose of explaining 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 testimony 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 impairment 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 courtroom 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 history 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 testimony is admissible only if “scientific, technical or other specialized 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 application of Rule 702 also requires an “assessment of the particular 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 foundation 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 specialized 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 knowledge .” But, Rule 602 notwithstanding, it is unnecessary to endorse that expanded concept of “personal knowledge” to conclude that the evidence code clearly permits 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, technicians 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 objection 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 testimony 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 function 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 conduit 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 concerns . 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 witness 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 inadmissible 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 inadmissible 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 accident 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 inadmissible 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 conclusions reached by other experts or technicians, such testimony 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 plaintiff presented testimony from a forensic engineer who testified that he had reviewed or investigated dozens of rollover accidents involving the vehicle and concluded that approximately 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 hearsay 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 principle 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 admissible 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 prejudice , 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 prosecutor 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 testimony was admissible as expert testimony under Rule 702, that the challenged testimony was admissible as an explanation 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 undisputedly 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 qualifies 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 relationship 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 horizontal 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 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.” 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 admissible 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 prosecutor 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 observation of defendant’s eyes during the HGN test and her performance 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 consumption 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 proposition that babies must turn their heads to focus on objects in their peripheral vision or that “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 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 prosecutor 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 ultimately 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 correct . 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 testifying 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 easier 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 explicitly 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 tangential 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 consumption 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 intoxicating 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 qualitatively 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 significant 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 multiple 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 limitations “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 introduction 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 closing 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 consumption of alcohol, which was sufficient to cause observable 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 evidence was qualitatively different from the permissible evidence of impairment and not merely cumulative.

The state also contends that the impairment evidence was harmless because the state presented an independent and alternative theory of impairment—that defendant had driven with a blood alcohol level of at least .08.14 According to the state, the evidence to support that theory —a breath test performed 90 minutes later—was “significant 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 person 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 harmlessness , we do “not ask whether the evidence of guilt is substantial or compelling.” Henley, 363 Or at 307. Rather, we consider the likelihood that the evidence affected the verdict . 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 chemical 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 incomplete ) proposition that alcohol dissipates from the blood over time is not sufficient” to permit a “nonspeculative connection ” 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 erroneously 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 opinion that defendant had been impaired and, therefore, would not have been “helpful” to the jury in assessing Wertz’s testimony . 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 without turning her head. But we are unable to accept the dissent ’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 without 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 defendant ’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 improperly 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 prosecution ’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 communicated 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 evidence , 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 information 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 ophthalmologist 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 testimony , when viewed in the context of this case, was unlikely to have affected the jury’s verdict. In addition, the majority opinion’s conclusion that the error was not harmless is 238 State v. Bowman

inconsistent with our prior cases that address when an evidentiary 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 administered 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 alcohol 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 defendant ’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 testimony 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 officer —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 medical 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 testimony 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 testimony 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 improperly 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 evidentiary 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 victim ’s live testimony, the court received as evidence a recording 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 admissible , 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 hearing 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 erroneous 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 admitted evidence as duplicative or unhelpful.” Id.

In State v. Thompson, 370 Or 273, 518 P3d 923 (2022), the defendant was convicted of first-degree robbery 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 defendant ’s cell phone while he was being treated for the gunshot 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 conclusion , 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 supported 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 defendant said he did not want the victim to “ ‘get in trouble’ ” for shooting him—was effectively an admission that the victim “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 incriminating .” 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 considered significant by either party[.]” Id. For example, during

Cite as 373 Or 213 (2025) 243

closing arguments, instead of emphasizing the impermissibly 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 harmless 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 admitted 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 convicted of murdering his girlfriend. He contended on appeal that the trial court erred in excluding the victim’s statements 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 defendant ’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 defendant 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 evidence 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 reasons [.]” 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” evidence , we explained, “the prosecutor would have been left with a much weaker argument to the jury[,]” and so we concluded 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 victim , 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 contended 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 supporting 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 picture 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 defendant ’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 introducing 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 anyone 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 credibility 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 inadmissible 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 evidentiary error in Willis involved the only evidence that the state had presented relating to an element of the crime. The evidentiary 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 alternative theory that defendant had been driving under the

Cite as 373 Or 213 (2025) 247

influence of alcohol. That evidence was not highly prejudicial 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.