State v. Skillicorn

479 P.3d 254, 367 Or. 464

• Court: Oregon Supreme Court
• Decided: January 14, 2021
• Docket: S066822
• Status: Published

Opinion

Argued and submitted January 15, 2020; decision of Court of Appeals reversed, judgment of circuit court reversed, and case remanded to circuit court for further proceedings January 14, 2021

STATE OF OREGON, Respondent on Review, v. DAVID JOHN SKILLICORN III, Petitioner on Review. (CC C152791CR); (CA A162831); (SC S066822) 479 P3d 254

To prove that defendant intentionally damaged another’s property with a vehicle, the state introduced evidence that defendant had driven recklessly on a prior occasion. The Court of Appeals affirmed the admission of the evidence, accepting the state’s argument that the evidence was not barred by OEC 404(3), which prohibits propensity evidence, because, under State v. Johns, 301 Or 535, 725 P3d 312 (1986), the evidence was admissible on a theory of relevance, the “doctrine of chances,” which the parties regard as a nonpropensity theory of rel- evance. Held: (1) OEC 404(3) prohibits admission of uncharged misconduct evi- dence to argue that a person has a propensity to commit certain acts, and there- fore, it is more likely that the person committed a similar act; (2) the doctrine of chances does not create an exception to that prohibition; (3) to the extent the doctrine of chances provides a basis for the admission of uncharged misconduct evidence, it only supports arguments based on the objective improbability of the recurrence of unusual events, like accidents; (4) Johns held that uncharged mis- conduct evidence could be admitted under the doctrine of chances to support a propensity argument, a conclusion that was erroneous; (5) the trial court in this case erred in admitting the prior driving evidence because the state proffered it to make a propensity argument, and (6) the error was not harmless. The decision of the Court of Appeals is reversed. The judgment of the cir- cuit court is reversed, and the case is remanded to the circuit court for further proceedings.

On review from the Court of Appeals.* Emily P. Seltzer, Office of Public Defense Services, Salem, argued the cause and filed the briefs for petitioner on review. Also on the brief was Ernest G. Lannet, Chief Defender. Lauren P. Robertson, Assistant Attorney General, Salem, argued the cause and filed the brief for respondent on ______________ * Appeal from Washington County Circuit Court, Rick Knapp, Judge. 297 Or App 663, 443 P3d 683 (2019). Cite as 367 Or 464 (2021) 465

review. Also on the brief were Ellen F. Rosenblum, Attorney General, and Benjamin Gutman, Solicitor General. Before Walters, Chief Justice, and Nakamoto, Flynn, Duncan, Nelson, and Garrett, Justices.** DUNCAN, 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.

______________ ** Balmer, J., did not participate in the decision of this case. 466 State v. Skillicorn

DUNCAN, J. In this criminal case, defendant was charged with first-degree criminal mischief, among other crimes. The state’s theory regarding the charge was that, after a dis- agreement with his girlfriend, defendant intentionally drove a truck into her car. Defendant admitted that he had hit the car but claimed that he had done so accidentally. Specifically, he claimed that the truck had malfunctioned and that he had lost control of it. To rebut that claim, the state sought to introduce evidence that, after a prior disagreement with his girlfriend, defendant had driven recklessly. Over defen- dant’s objection, the trial court admitted the evidence. The state used the evidence to argue that, when defendant “gets angry, he acts out,” and that, therefore, the jury should find that, on the night of the charged crimes, defendant had acted out by intentionally damaging his girlfriend’s car. The jury convicted defendant of first-degree criminal mischief and other crimes. Defendant appealed, asserting that the trial court’s admission of the evidence of his prior driving violated OEC 404(3), which provides, in part, “Evidence of other crimes, wrongs or acts is not admissible to prove the character of a person in order to show that the person acted in conformity therewith.” The state argued that the evidence was admis- sible under the “doctrine of chances,” as applied in State v. Johns, 301 Or 535, 725 P2d 312 (1986). The Court of Appeals observed that the evidence appeared to be propensity evi- dence, which is prohibited by OEC 404(3), but concluded that it was admissible under Johns. State v. Skillicorn, 297 Or App 663, 681, 443 P3d 683 (2019). We allowed review to consider whether evidence of uncharged misconduct can be admitted under the doctrine of chances to support an argu- ment like the one the state made in this case. For the reasons explained below, we conclude (1) OEC 404(3) prohibits the admission of uncharged misconduct evi- dence for the purpose of arguing that a person has a pro- pensity to commit certain acts, and therefore, it is more likely that the person committed such an act during the incident at issue; (2) the doctrine of chances does not cre- ate an exception to that prohibition; (3) to the extent that Cite as 367 Or 464 (2021) 467

the doctrine of chances provides a basis for the admission of uncharged misconduct evidence, it does so only to support arguments based on the objective improbability of the recur- rence of unusual events, like accidents; (4) Johns held that uncharged misconduct evidence could be admitted under the doctrine of chances to support a propensity argument, a conclusion that was erroneous; (5) the trial court in this case erred in admitting the evidence of defendant’s prior driving because the state proffered the evidence to make a propensity argument, and (6) the trial court’s error was not harmless. Therefore, we reverse and remand.

I. HISTORICAL AND PROCEDURAL FACTS

We begin with a brief description of the undis- puted historical facts. On the night of the charged crimes, defendant was using his employer’s truck. After complet- ing a work project, defendant drove the truck to visit his girlfriend, Walker, who was staying at her mother’s house. Defendant wanted Walker to leave with him, but Walker declined to do so because she and her mother, Peterson, had a work obligation the next morning. Defendant left the house and got in the truck. Moments later, defendant hit the back of Walker’s car, which was parked in Peterson’s driveway. Defendant got out of the truck, returned to the house, and apologized to Walker and Peterson, who told him to leave. Defendant got back in the truck and drove away. As he did, he hit a car parked on the street. The car belonged to one of Peterson’s neighbor’s, Howard. One of the truck’s wheels lodged in Howard’s car, and the truck crashed into some nearby trees. Defendant got out of the truck, collapsed, and was taken to the hospital. After being released from the hospital, defendant was arrested. He told the arresting offi- cer that the truck had malfunctioned. Specifically, he told the officer that, “[w]hen he put the truck in drive and tried to leave, the truck just took off on him and it jumped for- ward,” causing him to hit Walker’s car. Defendant also told the officer that, on the street, the truck “pulled to the right,” causing him to hit Howard’s car. In addition, defendant told the officer that the truck belonged to his employer and that it “had been loaned to him so he could fix issues that were already going on with [it].” 468 State v. Skillicorn

The state charged defendant with unauthorized use of a vehicle, first-degree criminal mischief, second-degree criminal mischief, and failure to perform the duties of a driver. The first-degree criminal mischief charge alleged that defendant had intentionally damaged Walker’s car, and the second-degree criminal mischief charge alleged that defendant had recklessly damaged Howard’s car. After jury selection, but before the presentation of evidence, the prosecutor made a motion for a ruling on the admissibility of evidence of defendant’s prior driving. The prosecutor told the trial court that the state’s theory regard- ing the first-degree criminal mischief charge was that defen- dant “got angry and rammed [Walker’s] car * * * on purpose.” She also told the court that, because she “need[ed] to prove that the defendant intentionally damaged * * * Walker’s car,” she wanted to introduce evidence that defendant had “driven in the same [or] similar manner in the same neigh- borhood before, after leaving [Walker’s] residence.” She stated that, when Howard and another neighbor, Hout, were interviewed by the police on the night of the charged crimes, they reported that defendant had “blazed through the neighborhood before.” She also stated that Howard and Hout would testify that defendant had previously “crashed somewhere on the street.” Defense counsel objected to the admission of the evidence of defendant’s “prior driving in the neighborhood.” He also told the trial court that the state had not provided any information about a prior crash. The court asked the prosecutor whether the discovery that it had provided to defendant contained any information about a prior crash, and the prosecutor said that it had not, but that she had spoken to Hout and he had told her more about defendant’s prior driving in the neighborhood. The prosecutor also told the court that Hout was in the courthouse and that defense counsel could speak with him. At that point, the trial court ruled that evidence of defendant’s prior driving in the neighborhood was relevant, stating: “With regard to the incident that occurred prior to this at * * * Walker’s residence, I do find that that is relevant Cite as 367 Or 464 (2021) 469

because we’re back at—we’re at her—this incident on November 7th is at—at her house, it involves a car that he—a Toyota Tundra which is an—the unauthorized use and he’s there at the house on November 7th and then that’s when the criminal mischief in the first degree occurs and the state has to prove the intent. And his prior conduct with regard to the issues with regard to Ms. Walker and what he’s done before then is relevant.” (Emphasis added.) The court then balanced the probative value of the evidence against the risk of unfair prejudice, pursuant to OEC 403, which provides that courts may exclude relevant evidence if, among other things, “its proba- tive value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury[.]” The court then ruled that the evidence was admissible. After opening statements, defense counsel asked the trial court to revisit its ruling. Defense counsel told the court that he had spoken with Hout and learned that, although the police had been called in response to a prior incident during which a car had gone off the street and onto a grassy area in the neighborhood, they had not identified defendant as the driver of the car in that incident. The pros- ecutor did not dispute that, but she said that Hout had told her that he knew defendant was the driver because he saw defendant later and confronted him about the incident. The court adhered to its ruling. At trial, the state presented evidence about the events on the night of the charged crimes. It also presented evidence about defendant’s prior driving in the neighbor- hood. That evidence included testimony about defendant’s driving in the neighborhood in general, as well as testimony about the incident during which defendant drove onto the grassy area. On direct examination, the prosecutor asked Peterson to “describe [defendant’s] driving in the neighborhood.” Peterson responded: “Well, on one occasion, [defendant and Walker] had a fight and [defendant] left, got into his truck and just screamed, I mean, just—it was so loud and it was so fast it scared me and so I kept [Walker] in the house. And my 470 State v. Skillicorn

neighbors after that event, I think he went up a grassy knoll area, but came over and said, ‘We have children and we have pets and we don’t want him in the neighborhood anymore. We’ve heard his truck. We know the way he drives and, you know, we’re—we’re going to bring your name up to the Board.’ And so I thought I might have to move for a while.” When questioning Hout, the prosecutor elicited additional details about defendant’s prior driving: “Q: Okay. Now, you said that you know the defendant. And have you witnessed him driving dangerously in the neighborhood before? “A: As a matter of fact, I have. “Q: Okay. Describe that. “A: After doing a little mental searching around September 14th, it’s a nice summer day. I’ve got a deck out front and I’m sitting on my deck. I hear burning rubber. You know, a car tearing loose. Straight across from me is a green space. You’ve got a sidewalk, there’s a green berm and on the other side of that berm is a little water drainage. I hear—I see the car again coming north down 178th * * *. It burns out, hits the curb, goes sideways up into the green space and then kind of launches back down into the street, gets squirrely, almost hits Mr. Howard’s car that was inev- itably hit in the later incident and I—I run down to the street just in time to make—to make out the vehicle make and model as it careens across Walker Road. No stop, prob- ably doing 35 to 40 miles an hour, full accelerator. Never let off the accelerator.” Hout testified that Walker came down the sidewalk and indicated that the driver of the car was her boyfriend. Hout kept an eye out for defendant and, when he saw defendant again, he walked up to him and said, “ ‘Hey, man, are you the guy that came playing Dukes of Hazard through my neighborhood a couple of weeks ago?’ ” and defendant “said something to the effect of, ‘What if I am?’ ” Hout then threat- ened defendant, and defendant threatened him back. The prosecutor also asked Howard, who owned the second car that defendant hit, about defendant’s prior driving: Cite as 367 Or 464 (2021) 471

“Q. Okay. And you spoke to the police about [defen- dant] previously driving recklessly in the neighborhood? “* * * * * “A. * * * I didn’t know who [Walker] was, but I knew that the house on the corner where [Peterson] lived at, that whoever was visiting that residence had a bad habit of rac- ing in and racing out of my neighborhood. And my house is right on the corner, so my house is the first house that a car comes to as it pulls into my neighborhood. “And I’ve got, you know, four sons at the house that play either—not in the street, but, you know, when you’re play- ing in the street or you’re running across the street, there’s a green space directly across the street from my house and all the neighborhood kids gather there and play, so it— it’s—it’s very upsetting to me when anybody drives in or out of my neighborhood extremely fast. “Q. Okay. And was that what you were talking about when you told—I think it was Officer Mansfield that you spoke to. Does that sound right? “A. Right. “Q. That he’d blazed out of the neighborhood before. You used the word ‘blazed.’ Does that sound about right? “A. Correct.” In her closing argument, the prosecutor told the jury that, in order to find defendant guilty of the first-degree criminal mischief charge, they had to find that “defendant intentionally damaged or destroyed the property of another,” specifically, Walker’s car. She also told the jury that “intent” seemed “to be what the defense is contesting the most” and that they could rely on defendant’s prior driving to find that defendant had acted intentionally. She noted that Peterson had testified that when defendant “gets angry, he acts out.” As an example of that behavior, the prosecutor referred to the prior incident when defendant drove onto the grassy area, which Peterson had mentioned in her testimony. The prosecutor urged the jury to rely on that incident to infer that defendant had acted intentionally when he hit Walker’s car, arguing, “So it makes a little more sense now why [Peterson] would have said on the stand that when he gets angry, he acts out. So just like he did prior, proving the intent 472 State v. Skillicorn

after he got into an argument. He took off because he was angry.” (Emphasis added.) The prosecutor also referred to Hout’s testimony about the prior incident. She noted that Hout was concerned about defendant’s driving and that, when Hout confronted defendant about the prior incident, defendant did not take responsibility for it: “[Hout] was mad, you know. You could tell on the stand that he was mad, but, you know what, you can understand why. I mean, after [the prior incident] happened, you know, they’re concerned about the neighborhood. It’s a quiet neighborhood with children running around. Mr. Howard has six children. Right? I think four are still home and able to play. So, after that happened, he apparently asked around to find out who it was. He found out that it was Ms. Walker’s boyfriend who liked to show up. “So the next time he showed up, they had this confronta- tion and * * * Hout said, ‘You’re not—don’t drive like that in my neighborhood anymore. You know, that’s it.’ What does defendant say? Something like, ‘What are you going to do about it?’ or ‘What if it was me?’ You know, just doesn’t take any responsibility or apologize at all. So, you can kind of (inaudible) * * * Hout was upset on the—stand. But, again, evidence of his intent in this case that he (inaudible) down there when he was angry.” (Emphases added.) Finally, the prosecutor turned to Howard’s testi- mony that defendant had “ ‘blazed’ out of the neighborhood before.” She told the jury that they could “consider that evi- dence for the specific purpose of proving the defendant’s mental states.” The jury found defendant guilty of unlawful use of a vehicle, first-degree criminal mischief for intentionally damaging Walker’s car, and second-degree criminal mis- chief for recklessly damaging Howard’s car. It found him not guilty of failing to perform the duties of a driver. Defendant appealed, challenging, among other things, the trial court’s admission of evidence of the prior incident when he had driven onto the grassy area. Defendant argued that the evidence was propensity evidence, which is barred Cite as 367 Or 464 (2021) 473

by OEC 404(3). In response, the state argued that the evi- dence was admissible under the doctrine of chances, which, it asserted, is a nonpropensity theory of relevance that, as applied in Johns, allows for the admission of uncharged mis- conduct evidence to support an argument that, because a person engaged in similar behavior on other occasions it is more likely that the person engaged in the behavior at issue. The Court of Appeals observed that “[i]t is an understatement to say that the line between propensity and nonpropensity inferences is difficult to discern under Oregon law.” Skillicorn, 297 Or App at 678. It noted that, when uncharged misconduct evidence is admitted—as it was in this case and Johns—to prove that a person acted with a particular intent on a prior occasion and, therefore, likely acted with the same intent on the charged occasion, the relevance of the evidence appears to rely “on a classic propensity theory.” Id. But, because it was bound by Johns, the court held that the evidence of defendant’s prior driving was admissible. Id. at 681. II. PARTIES’ ARGUMENTS On review, defendant renews his argument that the trial court erred in admitting the evidence of the prior inci- dent in which he drove onto the grassy area. He contends that, contrary to the Court of Appeals decision based on Johns, the evidence was not admissible under the doctrine of chances. Defendant asserts that the doctrine is based on “the improbability of recurring inadvertent events: that it is objectively improbable that the same accident will befall the same person again and again.” Therefore, he reasons that “[a] deliberate prior act is not admissible under the doc- trine of chances because its relevance does not depend on that probabilistic inference. Instead, it depends on a propensity inference. The prior act is relevant because if a person acted deliberately—with bad intent or guilty knowledge—on a prior occasion, it is likely that he acted with the same bad intent on a later, similar occasion.” Defendant urges that, because the doctrine of chances is premised on the proposition that “multiple similar accidents are highly improbable,” this court should overrule Johns to the extent that it allows for the admission of evidence of 474 State v. Skillicorn

deliberate acts. According to defendant, “[a]pplying the doc- trine of chances solely to * * * acts that are accidents, acts that are claimed to be accidents, or acts for which no cause is known limits the doctrine to its core logical inference and prevents the admission of propensity evidence under the guise of the doctrine of chances.” Regarding the specific evidence in this case, defen- dant contends that the evidence of the prior driving incident was not admissible under the doctrine of chances because, as proffered by the state, its relevance was “depend[ent] solely on a propensity inference—that if the defendant drove dangerously once, he had a propensity to drive dangerously again—rather than on the proposition that it is unlikely that multiple, unusual similar accidents befell defendant multiple times.”1 In response, the state renews its argument that the evidence was not barred by OEC 404(3) because it was admissible under the doctrine of chances. In the state’s view, the doctrine can be used to support the admission of evi- dence of deliberate uncharged misconduct to argue that, because defendant acted deliberately before, it is more likely that he acted deliberately again. Thus, as framed by the parties, the issue in this case is whether the evidence of the prior incident in which defendant drove onto the grassy area was admissible under the doctrine of chances. Both parties regard the doctrine as a nonpropensity theory of relevance. There is a debate among commentators regarding whether the doctrine actu- ally is a nonpropensity theory of relevance. See Skillicorn, 297 Or App at 680-81 (observing that “over the years, legal

1 Defendant makes two alternative arguments. First, he argues that, even if deliberate acts can be admissible under the doctrine of chances, the evidence of the prior driving incident in this case was not admissible under the doctrine because it was insufficient to establish an extraordinary coincidence; according to defendant, in order to establish such a coincidence, uncharged misconduct and charged misconduct must have been “unusually frequent, highly similar, and involve the same mental state,” and here they were not. Second, he argues that, even if the evidence met the requirements for admission under the doctrine of chances, the trial court had to exclude it under OEC 403. Because we conclude that the evidence of the prior incident was not admissible under the doctrine of chances, given the state’s theory of relevance, we need not, and do not, address those arguments. Cite as 367 Or 464 (2021) 475

commentators have criticized the ‘doctrine of chances’ under multiple theories” and summarizing critiques). But we do not understand defendant to categorically challenge the doctrine of chances as a basis for the admission of uncharged mis- conduct evidence; instead, we understand him to argue only that the doctrine is more limited than Johns and its progeny hold and, as properly understood, could not be used in this case to admit the evidence of the prior driving incident. III. ANALYSIS Because defendant argues that the evidence of the prior driving incident was barred by OEC 404(3), we begin our analysis with a discussion of that rule. We then examine the doctrine of chances. A. The Requirement of a Nonpropensity Theory of Relevance In order to be admissible, evidence of uncharged misconduct must be relevant under OEC 401, which provides, “ ‘Relevant evidence’ means evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less prob- able than it would be without the evidence.” Therefore, the proponent of uncharged misconduct must articulate a the- ory of relevance. To do so, the proponent must identify the inferences that it wants the factfinder to draw based on the evidence and explain how those inferences make the exis- tence of a fact of consequence more or less probable than it would be without the evidence. The proponent’s theory of relevance is critical. That is because, even if evidence is relevant under OEC 401, it may be barred by another rule, including OEC 404(3), which provides: “Evidence of other crimes, wrongs or acts is not admissi- ble to prove the character of a person in order to show that the person acted in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.” “ ‘Character’ for purposes of evidence law means a per- son’s disposition or propensity to engage or not to engage 476 State v. Skillicorn

in certain types of behavior.” Laird C. Kirkpatrick, Oregon Evidence § 404.03, 213 (7th ed 2020). Thus, OEC 404(3) pro- hibits the use of uncharged misconduct evidence to prove that a person has a propensity to engage in certain types of behavior and that the person acted in conformance with that propensity on a particular occasion. In short, it prohib- its “propensity evidence.” If the proponent’s theory of relevance requires the factfinder to employ propensity reasoning, then the trial court cannot admit the evidence based on that theory under OEC 404(3). State v. Johnson, 340 Or 319, 338, 131 P3d 173 (2006). Evidence is barred by OEC 404(3) if “the chain of logical relevance” connecting the evidence to the fact it is proffered to prove relies on “an inference relating to [a per- son’s] character or propensities.” Id.2 Consequently, in criminal cases, OEC 404(3) pro- hibits the prosecution from using uncharged misconduct evi- dence to argue that the defendant has either a general pro- pensity to engage in misconduct or a specific propensity to engage in misconduct like the charged crime and, therefore, it is more likely that the defendant committed the charged crime. The prosecution may not use uncharged misconduct evidence to prove “that the defendant is either generally a criminal or more particularly a rapist or burglar.” Edward J. Imwinkelried, 1 Uncharged Misconduct Evidence § 2:19, 2-139 (2013). As this court has explained, “If the only theme is ‘once a burglar, always a burglar,’ the evidence cannot be used as a ticket for admission. That concept does not qualify prior crime evidence for admission.” Johns, 301 Or at 549.

2 Although OEC 404(3) bars propensity evidence, this court has held that OEC 404(3) has been superseded by OEC 404(4) in criminal cases, except as otherwise provided by the state or federal constitutions. State v. Williams, 357 Or 1, 15, 346 P3d 455 (2015). Williams involved charges of child sexual abuse, and this court held that, in such cases, “ ‘other acts’ evidence to prove character and propensity” may be admissible, depending on “whether the risk of unfair prejudice outweighs the probative value of the evidence under OEC 403.” Id. at 20. At the same time, this court commented that, in cases involving crimes other than child sexual abuse, the Due Process Clause might prohibit “the admission of ‘other acts’ evidence to prove propensity.” Id. at 17. In this case, the state has expressly disclaimed any reliance on OEC 404(4), and therefore we do not address it. Like the parties, we focus on whether the evi- dence of defendant’s prior driving was barred by OEC 404(3). Cite as 367 Or 464 (2021) 477

The prohibition against propensity evidence can be traced back “more than three centuries.” Imwinkelried, 1 Uncharged Misconduct Evidence § 2:32 at 2-215. References to the exclusion of propensity evidence can be found in cases decided in the seventeenth century, including Harrison’s Trial, 12 How St Tr 833 (Old Bailey 1692) and Hampden’s Trial, 9 How St Tr 1053 (KB 1684), which evidence scholars frequently cite when describing the origins of the prohibi- tion. State v. Williams, 357 Or 1, 8, 346 P3d 455 (2015) (so stating). That prohibition is a fundamental aspect of our legal system. As this court observed in State v. Baker, 23 Or 441, 442-43, 32 P 161 (1893), “[t]he general rule is unques- tioned that evidence of a distinct crime unconnected with that laid in the indictment, cannot be given in evidence against the prisoner,” and “under no enlightened system of jurisprudence can a person be convicted of one crime on proof that he has committed another.” Similarly, in People v. Molineux, 168 NY 264, 291, 61 NE 286 (1901), the Court of Appeals of New York explained: “The general rule of evidence applicable to criminal trials is that the state cannot prove against a defendant any crime not alleged in the indictment, either as a foun- dation for a separate punishment, or as aiding the proofs that he is guilty of the crime charged. * * * This rule, so universally recognized and so firmly established in all English-speaking lands, is rooted in that jealous regard for the liberty of the individual which has distinguished our jurisprudence from all others, at least from the birth of Magna Charta. It is the product of that same humane and enlightened public spirit which, speaking through our com- mon law, has decreed that every person charged with the commission of a crime shall be protected by the presump- tion of innocence until he has been proven guilty beyond a reasonable doubt.” (Internal citations omitted.) See also State v. Houghton, 43 Or 125, 130, 71 P 982 (1903) (describing the prohibition against propensity evidence as “a universal rule of law”). The purpose of the prohibition is to protect the fair- ness of trials and the accuracy of verdicts. As the Supreme Court summarized in Michelson v. United States, 335 US 478 State v. Skillicorn

469, 475-77, 69 S Ct 213, 93 L Ed 168 (1948), the prohibition is based on the view that, even though propensity evidence may be relevant, it is not admissible, because it is unfairly prejudicial and likely to be overvalued: “Courts that follow the common-law tradition almost unan- imously have come to disallow resort by the prosecution to any kind of evidence of a defendant’s evil character to establish a probability of his guilt. * * * The State may not show defendant’s prior trouble with the law, specific crimi- nal acts, or ill name among his neighbors, even though such facts might logically be persuasive that he is by propensity a probable perpetrator of the crime. The inquiry is not rejected because character is irrelevant; on the contrary, it is said to weigh too much with the jury and to so overpersuade them as to prejudge one with a bad general record and deny him a fair opportunity to defend against a particular charge.” (Emphasis added; footnotes omitted.) Similarly, in State v. Pinnell, 311 Or 98, 105-06, 806 P2d 110 (1991), this court explained that “[b]ad character evidence (such as other crimes by the accused) is excluded under the propensity rule, not because it is irrelevant, but because of the risk of unfair prejudice to the accused.” Among other things, propensity evidence can cause factfinders to “convict for crimes other than those charged” or “give more weight to the evidence than it deserves in assessing the guilt of crime charged.” Id. at 106. Propensity evidence can have numerous harmful effects, including those detailed below. It can (1) impair the opposing party’s ability to present its case; (2) distract and confuse the factfinder; (3) prejudice the factfinder against a person; and (4) result in verdicts based on erroneous assumptions. First, propensity evidence can impair the oppos- ing party’s ability to present its case because it forces the opposing party to defend itself against allegations beyond those in the pleadings. The evidence may take the opposing party by surprise, a concern reflected in Hampden’s Trial, in which the court commented that, in a forgery case, it would not allow the prosecution to present “evidence of any other forgeries, but that for which [the defendant] was indicted, because we would not suffer any raking into men’s course Cite as 367 Or 464 (2021) 479

of life, to pick up evidence that they cannot be prepared to answer to.” 9 How St Tr at 1103. Similarly, in Baker, this court commented that, in a criminal case, “[i]t is of utmost importance to a defendant that the facts given in evidence by the prosecution shall consist exclusively of the transac- tion which forms the subject of the indictment, and which he has come prepared to answer.” 23 Or at 443.3 Second, propensity evidence can distract factfind- ers. It can result in “confusion of issues and undue consump- tion of time through what may be, in effect, a trial within a trial to ascertain the relationship between the purported other crime and the defendant.” Pinnell, 311 Or at 106. Third, and perhaps most importantly, propensity evidence can give rise to prejudice, which can detract from the factfinder’s ability to neutrally and thoroughly assess the evidence in the case. It creates a risk that verdicts will be affected by bias at a conscious or subconscious level. For example, a juror could decide a case against a party because, based on evidence of the party’s uncharged mis- conduct, the juror believes that the party is a bad actor. Or the juror could believe that the party should be punished for the uncharged misconduct, especially if the party has not been held accountable for that misconduct. Or the juror could believe that, given both the uncharged and charged misconduct, there is a risk that the party will engage in misconduct in the future and a verdict against the party will reduce or prevent that possibility. In a criminal case, prejudice arising from the admission of evidence of uncharged misconduct can under- mine fundamental constitutional protections, including the presumption of innocence and the requirement that the 3 The concern about surprise may be less now than when the decisions in Hampden’s Trial and Baker were issued, given modern requirements for pretrial discovery, but it remains a concern, as this case illustrates. In order to prop- erly respond to uncharged misconduct evidence, a party needs the opportunity to investigate the purported misconduct and to prepare to litigate the admissibility of the evidence under OEC 404(3) and OEC 403. But here, the prosecutor raised the issue of the admissibility of the evidence of defendant’s prior driving on the morning of trial, and the evidence included information that had not been pro- vided to defendant in discovery. As a result, the trial court made its initial ruling on the admissibility of the evidence before defense counsel could investigate the state’s new information. 480 State v. Skillicorn

prosecution prove the elements of each charged crime beyond a reasonable doubt. Uncharged misconduct evidence may cause a factfinder to shift the burden of proof to the defendant. A juror may not believe that a defendant who has engaged in other misconduct should be presumed innocent of the charged misconduct. In addition, uncharged miscon- duct evidence may have the effect of lowering the standard of proof. A juror may not afford a defendant who has engaged in other misconduct the benefit of a reasonable doubt. “Whereas [jurors] might have agonized over the possibility of convict- ing an innocent person, hearing that the accused committed similar bad acts might make jurors less cautious. At the very least, jurors’ consciences are eased because they know that the defendant is not a blameless character.” Tamara Rice Lave & Aviva Orenstein, Empirical Fallacies of Evidence Law: A Critical Look at the Admission of Prior Sex Crimes, 81 U Cin L Rev 795, 799 (2013) (footnotes omitted). Fourth, uncharged misconduct evidence can result in verdicts based on erroneous assumptions because fact- finders may give “more weight to the evidence than it deserves[.]” Pinnell, 311 Or at 106. They may “misuse the evi- dence by overvaluing its persuasiveness.” Lave & Orenstein, 81 U Cin L Rev at 798. “Objections arise from psychological questions surround- ing the reliability of character evidence, particularly the way such evidence is generated in a courtroom. People are not predictable characters and so psychologists question whether we can reliably determine how someone behaved on one particular occasion by reviewing the person’s past deeds.” Id. (footnotes omitted). Thus, there is a concern that fact- finders “will take what is essentially a weak circumstantial argument—‘he did it once, he probably did it again,’ or ‘he’s the type of person who would do such a thing’—and prove too much with it.” Id. Professor Imwinkelried provides a figure to illus- trate the dangers posed by propensity reasoning. The figure breaks down the reasoning into two steps, and Imwinkelried explains that each of those steps involves a “probative dan- ger” that “creates the risk of a verdict on an improper basis.” Cite as 367 Or 464 (2021) 481

Imwinkelried, 1 Uncharged Misconduct Evidence § 1:3 at 1-17. The figure, which is based on the use of uncharged misconduct by the prosecution against a criminal defendant, identifies the “intermediate” and “ultimate” inferences involved:

Item of Evidence Intermediate Inference Ultimate Inference Uncharged act The accused’s per- The accused’s con- by the accused sonal subjective bad duct on the charged character occasion consis- tent with the bad character

Edward J. Imwinkelried, Criminal Minds: The Need to Refine the Application of the Doctrine of Objective Chances as a Justification for Introducing Uncharged Misconduct Evidence to Prove Intent, 45 Hofstra L Rev 851, 859 (2017); Imwinkelried, 1 Uncharged Misconduct Evidence § 2:19 at 2-141. The first step in propensity reasoning is inferring the defendant’s subjective character, disposition, or pro- pensities from the uncharged misconduct. Imwinkelried, 1 Uncharged Misconduct Evidence § 2:19 at 2-141. In the words of OEC 404(3), it is using the uncharged misconduct evidence “to prove the character of a person.” That first step gives rise to the risk that the factfinder’s verdict will be affected by prejudice, as discussed above. For example, Imwinkelried explains that, “[I]t tempts the jury to decide the case on an improper basis. The jury may try the defen- dant for being a criminal rather than for the specific crime he or she is charged with.” Imwinkelried, 1 Uncharged Misconduct Evidence § 2:19 at 2-142. The second step in propensity reasoning is “infer- ring the defendant’s conduct on a particular occasion from his or her subjective character.” Id. § 2:19 at 2-143. That is, in the words of OEC 404(3), using the evidence of the defendant’s character “to show that the person acted in con- formity therewith.” This step gives rise to the risk that the factfinder will “overestimate the probative value of character evidence.” Imwinkelried, 1 Uncharged Misconduct Evidence § 1:3 at 1-29. As discussed above, the factfinder may believe 482 State v. Skillicorn

that the character is a better predictor of behavior than it is and, as a result, may give the character evidence “more weight than it deserves.” Id. § 2:19 at 2-145.4 Although OEC 404(3) prohibits the admission of uncharged misconduct evidence for propensity purposes, it does not prohibit the admission of such evidence for other purposes. Again, it provides: “Evidence of other crimes, wrongs or acts is not admissi- ble to prove the character of a person in order to show that the person acted in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.” (Emphasis added.) By its terms, the list of permissible purposes is not exclusive. Thus, OEC 404(3) allows for the admission of uncharged misconduct evidence to prove any relevant fact other than that a person has a propensity to commit certain acts and acted in accordance with that pro- pensity on a particular occasion. Id.; Johns, 301 Or at 549. But, as mentioned, OEC 404(3) prohibits the admis- sion of evidence to prove a fact if proof of the fact relies on “an inference relating to the [person’s] character or pro- pensities.” Johnson, 340 Or at 338. Consequently, when ruling on the admissibility of evidence under OEC 404(3), a trial court must determine the facts—intermediate and ultimate—that the proponent wants the factfinder to infer from the evidence. If the inferences involve whether the defendant has a propensity to commit certain acts and whether the defendant acted in accordance with that pro- pensity, the evidence is inadmissible. Such evidence is

4 The dangers posed by propensity evidence are so significant that admission of such evidence in criminal cases may violate the Due Process Clause of the Fourteenth Amendment. In Williams, this court noted that “ ‘historical practice’ is the primary guide for determining whether an evidentiary rule is so funda- mental as to be embodied in the federal constitution,” and that, in United States v. LeMay, 260 F3d 1018 (9th Cir 2001), “the Ninth Circuit considered the ‘histori- cal practice’ prohibiting the use of ‘other acts’ to prove the charged crime and con- cluded that ‘the general ban on propensity evidence has the requisite historical pedigree to qualify for constitutional status.’ ” 357 Or at 17 (quoting LeMay, 260 F3d at 1025). Accordingly, in Williams, this court indicated that, in most crimi- nal cases, the Due Process Clause might “preclude[ ] the admission of ‘other acts’ evidence to prove propensity.” Id. Cite as 367 Or 464 (2021) 483

propensity evidence, and thus barred by OEC 404(3), even if the proponent asserts that it is being offered to prove, for example, “intent” or “absence of mistake or accident.” Trial courts “must not jump immediately” to the purposes listed in OEC 404(3). Johns, 301 Or at 549. Instead, they must first determine how the proponent intends to use the evidence and whether that use involves proving a propensity to commit certain acts. Id.; see also State v. Hampton, 317 Or 251, 257 n 12, 855 P2d 621 (1993) (“[C]ourts must be on guard to prevent the motive label from being used to smuggle forbidden evidence of propensity to the jury.” (Internal quotation marks omitted.)). A proponent of uncharged misconduct evidence might make a general assertion that the evidence is admis- sible for one or more of the purposes listed in OEC 404(3). But such an assertion, by itself, is insufficient to allow a trial court to determine whether the evidence is supported by a nonpropensity theory of relevance. A proponent should identify the logical path that it will be asking the factfinder to follow. As Imwinkelried’s figure, set out above, illus- trates, the path may have multiple steps. If the proponent does not identify each step, the proponent (and the trial court) may fail to realize that the relevance of the evidence depends upon propensity reasoning. “Unfortunately,” as Imwinkelried has observed regarding the use of uncharged misconduct evidence to prove a criminal defendant’s intent, “in many cases—especially cases in which the prosecution’s only tenable theory is the doctrine of objective chances— the courts often do not demand that the prosecution explain how the uncharged act is relevant to intent without posit- ing a forbidden, intermediate inference of bad character.” Imwinkelried, 1 Uncharged Misconduct Evidence § 5:2 at 5-9. Instead, courts simply conclude that “evidence is admis- sible to prove intent—a generalization that is false when the prosecution must rely on an intermediate bad character inference to connect the uncharged act to the ultimate con- clusion of intent.” Id. In this case, the state’s theory of relevance was that the evidence of defendant’s driving was admissible for the related purposes of proving his intent and disproving his 484 State v. Skillicorn

claim of accident. Specifically, the state’s theory was that the evidence was relevant to prove that defendant intention- ally drove the truck into Walker’s car and that it did not malfunction as he claimed. Thus, the state proffered the evi- dence for two of the purposes listed in OEC 404(3). But, as just discussed, even if evidence is offered for one of the listed purposes, it is barred by OEC 404(3) if the proof depends upon propensity reasoning. Thus, the question becomes whether the state’s the- ory of relevance involved propensity reasoning. The state asserts that it did not, because the evidence was relevant under the doctrine of chances, which, the state asserts, is a nonpropensity theory of relevance. In response, defendant argues that the doctrine does not allow the admission of evi- dence to support the type of argument that the state made in this case. To resolve that dispute, we must examine the reasoning that underlies the doctrine. B. The Doctrine of Chances The doctrine of chances is a theory of relevance based on the objective improbability of the recurrence of uncommon events. The idea underlying the doctrine is that, at some point, it becomes unlikely that each event in a series of similar events can have the same uncommon cause; there- fore, if the number of events in a series claimed to have the same uncommon cause exceeds the number that can reason- ably be expected to have that cause, a factfinder can infer that not all of the events actually have that cause. So, for example, if a party asserts that all the events in a series of similar events were accidents, an opponent might rely on the doctrine of chances to argue that the number of events exceeds the number of accidents that the party was likely to suffer, and the factfinder should therefore infer that not all the events were accidents.5

5 As the Court of Appeals noted in this case, the doctrine of chances is the subject of debate. Skillicorn, 297 Or App at 680. Some critics argue that it is not actually a nonpropensity theory. But the parties in this case assume that the doc- trine can be used as a nonpropensity theory of relevance to justify the admission of uncharged misconduct evidence. And they both rely on Imwinkelried’s descrip- tion of the logic underlying the doctrine. They disagree about whether that logic justifies the admission of the evidence of defendant’s prior driving in this case. Consequently, we—like the parties—rely on Imwinkelried’s description of the Cite as 367 Or 464 (2021) 485

Imwinkelried traces the use of the doctrine of chances to prove “intent” to Dean Wigmore, who described the use as follows: “ ‘§ 302. Theory of evidencing Intent. To prove Intent, as a generic notion of criminal volition or willfulness, includ- ing the various non-innocent mental states accompanying different criminal acts, there is employed an entirely dif- ferent process of thought. The argument here is purely from the point of view of the doctrine of chances—the instinctive recognition of that logical process which eliminates the ele- ment of innocent intent by multiplying instances of the same result until it is perceived that this element cannot explain them all. Without formulating any accurate test, and with- out attempting by numerous instances to secure absolute certainty of inference, the mind applies this rough and instinctive process of reasoning, namely, that an unusual and abnormal element might perhaps be present in one instance, but the oftener similar instances occur with similar results, the less likely is the abnormal element likely to be the true explanation of them.’ ” Imwinkelried, 1 Uncharged Misconduct Evidence § 5:6 at 5-30-31 (quoting 2 John H. Wigmore, Evidence in Trials at Common Law § 302, 196 (3d ed 1940) (emphases added)). To illustrate his theory, Wigmore provided an example involv- ing two hunters: “ ‘[I]f A while hunting with B hears the bullet from B’s gun whistling past his head, he is willing to accept B’s bad aim or B’s accidental tripping as a conceivable expla- nation; but if shortly afterwards the same thing happens again, and if on the third occasion A receives B’s bullet in his body, the immediate inference (i.e. as a probability, perhaps not a certainty) is that B shot at A deliberately; because the chances of an inadvertent shooting on three successive similar occasions are extremely small; or (to put it in another way) because inadvertence or accident is only an abnormal or occasional explanation for the discharge of a gun at a given object, and therefore the recurrence of a similar result (i.e. discharge towards the same object, A) excludes the fair possibility of such an abnormal cause and points out the cause as probably a more natural and usual one, i.e. a deliberate discharge at A. In short, similar

doctrine, and we focus on whether the doctrine, as he describes it, justifies admis- sion of the evidence of defendant’s prior driving in this case. 486 State v. Skillicorn

results do not usually occur through abnormal causes; and the recurrence of a similar result (here in the shape of an unlawful act) tends (increasingly with each instance) to negative accident or inadvertence or self-defense or good faith or other innocent mental state, and tends to establish (provisionally, at least, though not certainly) the presence of the normal, i.e. criminal, intent accompanying such an act; and the force of each additional instance will vary in each kind of offence according to the probability that the act could be repeated, within a limited time and under given circumstances, with an innocent intent.’ ” Id. § 5:6 at 5-31 (quoting 2 Wigmore, Evidence § 302 at 196-97). According to Imwinkelried, under Wigmore’s the- ory, “the inference of mens rea arises from the implausi- bility of the defendant’s claim of ‘successive similar’ inno- cent acts.” Id. § 5:8 at 5-36. “The defendant claims that he accidentally discharged the rifle in the victim’s direction on each occasion; but as the number of ‘accidental’ discharges increases, the claim of accident becomes less believable.” Id. § 5:11 at 5-64. As Imwinkelried explains, “In isolation, it might be plausible that the defendant acted accidentally or innocently; a single act could easily be explained on that basis. However, in the context of other mis- deeds, the defendant’s act takes on an entirely different light. The fortuitous coincidence becomes too abnormal, bizarre, implausible, unusual, or objectively improbable to be believed. The coincidence becomes telling evidence of mens rea.” Id. § 5:6 at 5-29-30 (emphasis added). Thus, according to Imwinkelried, the doctrine of chances can be used to justify the admission of uncharged misconduct evidence to disprove a defendant’s claim of successive similar accidents or other innocent acts when the evidence is sufficient to support an inference that the defendant’s claim involves a “fortu- itous coincidence” that is too “objectively improbable to be believed.” Id. As Imwinkelried further explains, “The doctrine of chances theory is an example of reasoning by process of elimination. The proponent uses the theory to eliminate random chance as an explanation for the set of outcomes.” Id. § 4:1 at 4-31. But, “properly construed, the doctrine of Cite as 367 Or 464 (2021) 487

chances recognizes the limited probative value of the dis- proof of the random chance hypothesis.” Id. § 4:1 at 4-34. “The only direct inference from the doctrine of chances is that one or some of the incidents were not accidents.” Id. It does not prove that any particular incident was intentional, much less that they all were. Id. Given the idea that underlies the doctrine of chances—viz., the objective improbability of the recurrence of similar, uncommon events—there are two foundational requirements for its use: similarity and unusual frequency. First, evidence of uncharged misconduct must be similar to the charged misconduct. That is because “[a] dissimilar uncharged incident has at most a negligible effect on the probability of an accidental occurrence of the social harm.” Edward J. Imwinkelried, The Use of Evidence of an Accused’s Uncharged Misconduct to Prove Mens Rea: The Doctrines Which Threaten to Engulf the Character Evidence Prohibition, 51 Ohio St LJ 575, 589-90 (1990); State v. Leistiko, 352 Or 172, 186, 282 P3d 857, adh’d to as modified on recons, 352 Or 622, 292 P3d 522 (2012) (observing that the number of other events required will depend on the circumstances, and that, in Johns, this court stated that “[a] simple, unremarkable single instance of prior conduct probably will not qualify, but a complex act requiring several steps, particularly pre- meditated, may well qualify”). Second, the number of events in the series must exceed the number of events that could reasonably be expected to share the uncommon cause. Thus, in a case where a person claims that events were caused by accident, the number of events in the series would have to exceed the number of accidents that could reasonably be expected to befall the person. In other words, the number of similar, uncommon events has to be sufficient to establish a “for- tuitous coincidence” that is too “abnormal” or “bizarre” to be believed. The number of events required will depend on how rare they are. Wigmore suggested that a single other event may not be sufficient. Imwinkelried, 1 Uncharged Misconduct Evidence § 5:7 5-33-34. But Imwinkelried posits that, if the event is very rare—a “once in a lifetime event”—a single other event may be sufficient. Id. at 5-35. “The key is the relative frequency rather than brute number of 488 State v. Skillicorn

incidents.” Imwinkelried, The Use of Evidence of an Accused’s Uncharged Misconduct to Prove Mens Rea, 51 Ohio St LJ at 591. When determining how many other events are neces- sary to support an inference of an improbable coincidence, it may be necessary to consider the circumstances of the per- son accused of the misconduct. For example, some events that might be rare for most people might be more common for people who do certain types of work. Id. The proponent of the evidence bears the burden of proving that the foundational requirements are satisfied. But if either of the two foundational requirements is unmet, the proponent “has not triggered the doctrine of chances[.]” Id. at 592. Unfortunately, according to Imwinkelried, when courts are asked to rule on the admissibility of uncharged misconduct evidence, they may fail to determine whether the proponent has satisfied the requirements: “[I]n a large number of cases in which the courts admit uncharged misconduct to establish intent and the prosecu- tion’s only conceivable non-character theory is the doctrine of chances, the court’s analysis is conclusory in the extreme. Rather than invoking the doctrine and inquiring whether the prosecution has satisfied the doctrine’s requirements, the courts advance the broad generalization that similar misdeeds are admissible to prove intent.” Imwinkelried, Criminal Minds, 45 Hofstra L Rev at 857 (footnotes omitted). When applying the doctrine of chances, a factfinder is asked to consider the likelihood of the recurrence of the uncommon events. Therefore, Imwinkelried asserts, the admission of evidence of uncharged misconduct under the doctrine of chances does not violate the prohibition against propensity reasoning. To illustrate the difference between propensity reasoning and doctrine-of-chances reasoning, Imwinkelried sets out another figure: Item of Evidence Intermediate Inference Ultimate Inference An uncharged Considered together The probability of the event involving with the charged event, accused’s criminal the accused an objectively improba- state of mind at the ble coincidence time of one or some of the events Cite as 367 Or 464 (2021) 489

Id. at 865. According to Imwinkelried, doctrine-of-chances reasoning differs from propensity reasoning “with respect to both of the probative dangers inspiring the character evidence prohibition.” Id. First, doctrine-of-chances rea- soning “does not require the jurors to consciously advert to the question of the accused’s personal, subjective character. Rather, they are asked to assess the objective improbabil- ity of so many accidents or inadvertent acts.” Id. Second, it “does not require jurors to use character as a predictor of conduct.” Id. Rather, “the second step necessitates that the jurors do what the judge will tell them to do in another part of the jury charge, namely, draw on their common sense and knowledge to assess the relative plausibility of the parties’ competing versions of the events.” Id. The fact that the doctrine of chances is based on the objective improbability of the recurrence of similar, uncom- mon events leads us to two conclusions relevant to this case. First, in cases like this, where the doctrine is used to prove “lack of accident,” the application of the doctrine requires an assessment of the odds that all of the events in a series were accidental; therefore, it does not make sense to include events in the series that are known not to have that cause or explanation. Doing so creates a false set. For example, if the defendant in a criminal case was charged with theft and claimed that he took the property at issue by accident, evidence that he had committed theft on two prior occasions would not be relevant on a doctrine-of- chances theory. Because the prior thefts were intentional, there would be no reason to ask, “What are the odds that all three thefts were accidental?” The answer to that question would be “zero,” because the first two are known to have been intentional. But the answer would not enable the fact- finder to make any determination regarding whether the defendant has suffered more accidents than could reason- ably be expected—which is what the factfinder must deter- mine under a correct application of the doctrine of chances. To be sure, the prior thefts might be relevant on a propen- sity theory, but such a theory is prohibited by OEC 404(3). They might also be relevant on a nonpropensity theory, such as to prove the existence of a plan, if, for example, the items that the defendant stole during the prior thefts were used 490 State v. Skillicorn

to commit the charged theft. But they are not relevant on a doctrine-of-chances theory. Second, the doctrine of chances can be used only to support a particular type of argument: an argument about the objective improbability of certain events. According to Imwinkelried, that focus is what distinguishes doctrine- of-chances reasoning from propensity reasoning. Thus, the doctrine of chances cannot be used as a basis for arguing that, because a person acted in a certain way before, the factfinder should find that the person has a propensity to act in that way and, therefore, it is more likely that the person did so on a particular occasion. It is often difficult to apply the doctrine of chances correctly. Just because a factfinder is using a series of events to assess the odds that the charged event was an accident does not mean that the factfinder is employing doctrine-of- chances reasoning. The factfinder could be employing pro- pensity reasoning. As we will explain, that is what occurred in Johns. C. The Application of the Doctrine of Chances in Johns In Johns, the defendant was charged with murder for shooting his second wife. The defendant’s defense was that the shooting was accidental. Johns, 301 Or at 537-38. Immediately after the shooting, the defendant called the police and reported that his wife had fired a gun at him and that, when he tried to take the gun from her, the gun went off and she was hit. Id. To rebut that defense, the state sought to introduce evidence that, six years earlier, the defendant had intentionally assaulted his first wife. Id. at 539-40. The state argued that the two incidents were similar because, around the time of each incident, the defendant was hav- ing problems in his career and marriage, was financially dependent on the victim, and had threatened the victim. Id. at 556. The trial court admitted the evidence over the defendant’s objection. Id. at 557. The defendant was con- victed, and he appealed. On review, this court affirmed the admission of the evidence of defendant’s assault on his prior wife. In doing so, this court discussed the doctrine of chances, quoting both Cite as 367 Or 464 (2021) 491

Wigmore and Imwinkelried, as we have above. Johns, 301 Or at 552-55. This court also observed that, according to Imwinkelried, “Wigmore’s theory of logical relevance does not depend on a character inference because the proponent is not asking the trier of fact to infer the defendant’s conduct (entertain- ing a particular mens rea) from the defendant’s subjective character. The intermediate inference is an objective likeli- hood under the doctrine of chances rather than a subjective probability based on the defendant’s character.” Id. at 554. And this court set out Imwinkelried’s figure describing the inferences involved in doctrine-of-chances reasoning. Thus, in Johns, this court observed that the doc- trine of chances is based on the objective improbability of the recurrence of uncommon events, like accidents. But this court did not employ doctrine-of-chances reasoning. That is, it did not focus on whether the defen- dant had been involved in more purportedly accidental shootings or assaults than could reasonably be expected. That, of course, was because there was no evidence that the defendant had been involved in any purportedly accidental shootings or assaults other than the charged one. There was no claim that the defendant’s prior assault was accidental. The state’s theory (and all the evidence) was that the prior assault had been intentional. Thus, the state had used a prior intentional act to argue that it was more likely the defendant had acted intentionally when he committed the charged act. The state had used propensity reasoning, and this court followed suit. It affirmed the trial court’s admis- sion of the defendant’s assault of his first wife, apparently on the theory that the assault was relevant “to show that when similarly agitated in a domestic setting [the] defendant will act violently and intentionally.” Id. at 551. In short, this court in Johns described the doctrine of chances but did not properly apply it. The opinion is inter- nally inconsistent and has created confusion. Although it acknowledged that propensity evidence is prohibited, it affirmed the admission of such evidence. And it has led to overly broad statements regarding the admissibility of uncharged misconduct evidence, which essentially state 492 State v. Skillicorn

that, if a defendant has intentionally committed other acts similar to the charged act, evidence of those other acts can be admitted to prove that the defendant intentionally com- mitted the charged act. See, e.g., State v. Tena, 362 Or 514, 524, 412 P3d 175 (2018) (quoted by the state for the propo- sition that the simple recurrence of an act “ ‘increases the likelihood of a mens rea or mind at fault’ ”); Leistiko, 352 Or at 182 (describing the doctrine of chances as allowing for the admission of uncharged misconduct evidence on the theory that “the more often [a] defendant performs the actus reus, the smaller is the likelihood that the defendant acted with an innocent state of mind”). Those statements are false, to the extent that they state that uncharged misconduct evi- dence can be admissible to show that a defendant has a pro- pensity to act in a certain way, and, therefore, it is more likely that the defendant did so when he committed the charged act. The state urges us to adhere to Johns. It invokes the principle of stare decisis, which “promotes our legal system’s compelling interests in ‘stability,’ ‘predictability,’ and ‘con- sistency in the law.’ ” (Quoting Farmers Ins. Co. v. Mowry, 350 Or 686, 698, 261 P3d 1 (2011).) “[S]tare decisis is a prudential doctrine that is defined by the competing needs for stability and flexibility in Oregon law.” Farmers Ins. Co., 350 Or at 698. When pre- sented with a challenge to precedent, this court “begin[s] with the assumption that issues considered in our prior cases are correctly decided.” Id. “At the same time,” however, “this court’s obligation * * * is to reach what we determine to be the correct result in each case.” Id. Consequently, if a party can demonstrate that we erred in deciding a case, “because we were not presented with an important argu- ment or failed to * * * adequately analyze the controlling issue, we are willing to reconsider the case.” Id. Here, several considerations weigh against adhering to Johns. First, Johns has not led to clarity in the law. It is internally inconsistent; it describes the doctrine of chances, but then reaches a result that is at odds with the logic that underlies the doctrine. Courts struggle to apply, in a princi- pled manner, decisions that are internally inconsistent. As Cite as 367 Or 464 (2021) 493

a result, affirming or extending internally inconsistent deci- sions often undermines the stability, predictability, and con- sistency that stare decisis intends to promote. Second, Johns has created confusion, which this court has already had to correct. See, e.g., State v. Turnidge (S059155), 359 Or 364, 434-37, 374 P3d 853 (2016) (noting misapplications of Johns); Tena, 362 Or at 522 n 1 (observing that, although this court had previously applied Johns to cases in which uncharged misconduct evidence was admitted to prove a defendant’s hostile motive, it applies only when such evidence is used to prove a defendant’s intent on a doctrine-of-chances theory). Third, and most importantly, Johns has opened the door to propensity evidence, which is inconsistent with the long- standing and fundamental prohibition against such evidence. Therefore, we overrule Johns to the extent that it holds that evidence of uncharged misconduct can be admitted under the doctrine of chances for the purpose of arguing that, because the defendant engaged in deliberate conduct before, it is likely that he engaged in it again during the charged incident.

D. The Admissibility of Evidence of Defendant’s Prior Driving

With that understanding of the doctrine of chances, we return to the particular issue in this case. As mentioned, defendant argues that the evidence of the prior driving was not admissible under the doctrine. When reviewing a trial court’s ruling admitting uncharged misconduct evidence, this court focuses on the theory of relevance relied on by the trial court. State v. Garrett, 350 Or 1, 6, 248 P3d 965 (2011). Here, the state’s theory, which the trial court relied on, was not based on the objective improbability of repeated accidents. The state did not argue, for example, that the defendant had experienced an extraordinary number of vehicle malfunctions. Instead, as recounted above, the state sought to introduce the evidence to support an argument that, because defendant had acted a certain way during a prior incident, the jury should infer that he acted in a sim- ilar way during the charged incident. In other words, the state sought to introduce the evidence to prove defendant’s character—that when he “gets angry, he acts out”—in order 494 State v. Skillicorn

to prove that he acted in conformance therewith. The state’s theory was not a doctrine-of-chances theory. It was a pro- pensity theory. Consequently, the admission of evidence on that theory violated OEC 404(3). Because we conclude that the trial court erred in admitting the evidence of defendant’s prior driving, we turn to the question of whether the error was “harmless.” Evidentiary error is harmless only when “there is little like- lihood that the particular error affected the verdict[.]” State v. Davis, 336 Or 19, 32, 77 P3d 111 (2003). The state argues that admission of the evidence was harmless because there was other evidence of defendant’s intent, including the locations of the cars when they were hit and the amount of damage caused. That is certainly true, but when determining whether the erroneous admission of evidence was harmless, the question is not whether the state presented sufficient other evidence to support a conviction. Instead, it is whether there is little likelihood that the error affected the verdict. Id. (when determining whether eviden- tiary error was harmless, this court focuses “on the possible influence of the error on the verdict rendered, not whether this court, sitting as a factfinder, would regard the evidence as substantial or compelling”). In this case, we cannot conclude that the admission of the evidence of defendant’s prior driving was harmless. The evidence related to the central dispute in the case. See State v. Marrington, 335 Or 555, 566, 73 P3d 911 (2003) (holding error related to a “central factual issue” was not harmless). And the state used it as propensity evidence, encouraging the jury to decide the case based on impermis- sible character-based reasoning. Such reasoning carries a risk of causing the verdict to be based on unfair prejudice, and here, the prosecutor highlighted defendant’s past driv- ing, noting that he had not taken responsibility for it and that he posed a danger to children in the neighborhood. Propensity evidence also carries a risk of causing the ver- dict to be based on an overestimation of the probative value of the evidence. That is particularly true in cases where the uncharged misconduct involves a different mental state than the charged misconduct, like this one, where the prosecutor Cite as 367 Or 464 (2021) 495

argued that the jury should rely on defendant’s prior reck- less driving to infer that he intentionally damaged property. See Imwinkelried, 1 Uncharged Misconduct Evidence § 5:9 at 5-51 (“perhaps the most important factor” as to whether one act is sufficiently similar enough to be probative of intent is whether the acts involved the same state of mind). Here, because the evidence 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 defen- dant and be overvalued, we conclude that the admission of the evidence was harmful as to all of the counts on which defendant was convicted and, therefore, it is necessary to reverse those convictions. 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.