NELSON v. CAIN (2026)

Petitioner, who was convicted of various forcible sex offenses following a jury trial, challenges the denial of a post-conviction claim in which he asserted that his defense attorney had provided constitutionally inadequate and ineffective assistance of counsel. Petitioner contends that his attorney failed to exercise reasonable professional skill and judgment by failing to move for a mistrial or otherwise make a record after three jurors had seen him, during the course of trial, in the courthouse hallway in jail clothes, a belly chain, and shackles. The post-conviction court denied relief, concluding that petitioner had not established that counsel's performance fell below constitutional standards or that any such deficiency caused him prejudice, and the Court of Appeals affirmed. We allowed review to consider whether the jury's out-of-court observation of petitioner in shackles, like unwarranted in-court shackling, obligated defense counsel to take further action on behalf of petitioner than counsel did here. As we will explain, we conclude that the out-of-court observation that occurred during the course of trial in petitioner's case required counsel to seek a mistrial.

Specifically, we conclude that, given the then-existing case law regarding in-court shackling and the principles underlying that case law, an attorney exercising reasonable professional skill and judgment would have concluded that, as with most in-court shackling cases, allowing jurors to see a criminal defendant in shackles outside the courtroom is grounds for a mistrial. Defense counsel was therefore obligated to choose a course of action with that understanding in mind. And under the circumstances of this case, as reflected in the post-conviction court's factual findings and the underlying record, the only reasonable course of action was for counsel to seek a mistrial on the record. By failing to do so, counsel prejudiced petitioner by denying him an opportunity to persuade the trial court that a mistrial was necessary to ensure his right to a trial by an impartial jury as guaranteed by Article 1, section 11, of the Oregon Constitution. Accordingly, the post-conviction court erred in denying petitioner's claim for relief on this ground. We therefore reverse the judgment of the post-conviction court and the decision of the Court of Appeals with respect to this aspect of its decision.¹

I. BACKGROUND

We review the denial of post-conviction relief for legal error. Green v. Franke, 357 Or 301, 312, 350 P3d 188 (2015). A post-conviction court's findings of historical fact are binding on this court if there is evidence in the record to support them. Lichau v. Baldwin, 333 Or 350, 359, 39 P3d 851 (2002). If the post-conviction court has not expressly made findings as to all such facts and there is evidence from which those facts could be decided in more than one way, we presume that the court decided the facts consistently with its conclusions of law. Id. (citing Ball v. Gladden, 250 Or 485, 487, 443 P2d 621 (1968) (so stating)). As background for our analysis, we begin by describing the general legal standards that govern post-conviction claims like this one, then provide an overview of the relevant procedural and factual history from petitioner's underlying criminal trial. Finally, we discuss the proceedings in the post-conviction court and the Court of Appeals.

A. Legal Standards

A post-conviction court must grant relief if a petitioner establishes a substantial violation of their state or federal constitutional rights in the proceeding that led to their conviction. ORS 138.530(1)(a). As relevant here, Article I, section 11, of the Oregon Constitution guarantees criminal defendants adequate assistance of counsel. Krummacher v. Gierloff, 290 Or 867, 871, 627 P2d 458 (1981).² To be entitled to post-conviction relief on the basis of inadequate assistance of counsel, a petitioner must prove by a preponderance of the evidence that counsel failed to exercise reasonable professional skill and judgment. Lichau, 333 Or at 359. If a petitioner meets that burden, the court must then determine whether counsel's failure caused the petitioner prejudice. Id. This court has often emphasized that, in determining whether counsel's performance was constitutionally adequate, we “must make every effort to evaluate a lawyer's conduct from the lawyer's perspective at the time, without the distorting effects of hindsight.” Id. at 360. That, the court has explained, is because “[t]he constitution gives no defendant the right to a perfect defense—seldom does a lawyer walk away from a trial without thinking of something that might have been done differently or that he would have preferred to have avoided.” Krummacher, 290 Or at 875.

To exercise reasonable professional skill and judgment on behalf of a criminal defendant, counsel must investigate the facts and familiarize themselves with the applicable law to the extent necessary to enable them to “advise * * * and represent the defendant in an informed manner.” Id. Counsel's recommendations and decisions must be based on “an evaluation of the nature and complexity of the case, the likely costs and potential benefits of the contemplated action, and other factors.” Stevens v. State, 322 Or 101, 109, 902 P2d 1137 (1995). Thus, to determine whether counsel's investigation and preparation were adequate, a court “must understand the nature and complexity of the issues presented,” including “the stakes for petitioner” and what “defense counsel knew.” Richardson v. Belleque, 362 Or 236, 257, 406 P3d 1074 (2017).

Further, merely investigating the relevant facts and law does not guarantee that counsel's tactical decisions will reflect reasonable professional skill and judgment. As this court has previously held, even if each of counsel's choices are made with the appropriate facts and law in mind, counsel's assistance is considered constitutionally inadequate if their understanding of the facts and law was not reasonably accurate. See Farmer v. Premo, 363 Or 679, 700, 427 P3d 170 (2018) (where counsel was aware of the relevant facts but did not fully grasp their significance for the defendant's case, counsel did not provide adequate assistance). As we stated in Farmer, “it would make little sense to demand that attorneys gather sufficient information to make a decision, but not require them to reasonably understand and assess it.” Id. at 697. Thus, we explained, “[a]n appropriate consideration of the risks and benefits of a strategic decision requires a reasonably accurate and adequate consideration of the facts produced by an investigation.” Id. at 699.

Where the premise of a post-conviction claim is that counsel failed to even consider an available option, our assessment “will turn on, among other things, whether the strategy that defense counsel did employ was reasonable, the relationship between the evidence or theory that defense counsel failed to consider and the strategy that counsel did pursue, and the extent to which counsel should have been aware of the strategy that petitioner now identifies.” Pereida-Alba v. Coursey, 356 Or 654, 674, 342 P3d 70 (2015). Where a claim is that “counsel should have been aware of” a particular legal strategy, id., we have explained that the failure to identify and argue an unsettled or ambiguous question of law on behalf of a defendant could constitute inadequate assistance of counsel, at least where the forgone argument, if successful, would “so obviously offer possible benefits to a defendant that any lawyer exercising reasonable professional skill and judgment would raise it,” Burdge v. Palmateer, 338 Or 490, 500, 112 P3d 320 (2005). And in assessing such a claim arising from an overlooked statutory argument, we have elaborated that “the post-conviction court may grant a petitioner relief * * * if it can say affirmatively that a lawyer exercising reasonable professional skill and judgment would have recognized the statutory ambiguity, would have seen an interpretation that could benefit the defendant, and would have concluded under the circumstances that the potential benefits of advancing that interpretation exceeded any risks.” Id. at 497.

With the legal standards for defense counsel's performance thus framed, we turn to the specific factual and procedural history of petitioner's criminal case.

B. Events and Proceedings in Criminal Trial

We take the relevant factual and procedural history from the post-conviction record, as reflected in the post-conviction court's judgment and summarized in the Court of Appeals’ opinion, Nelson v. Cain, 334 Or App 326, 556 P3d 1 (2024). In 2007, petitioner was tried for charges arising out of a domestic violence incident “in which he punched his domestic partner in the face, had sex with her against her will, and took her vehicle.” Id. at 327. A jury found petitioner guilty of various charges, including assault, sexual abuse, rape, and unauthorized use of a motor vehicle, and petitioner appealed. Id. In 2012, the Court of Appeals reversed and remanded petitioner's convictions for first-degree rape and three counts of first-degree sexual abuse because of an erroneous jury instruction, but it otherwise affirmed. State v. Nelson, 241 Or App 681, 251 P3d 240 (2011), rev. dismissed, 354 Or 62 (2012). In 2013, the state retried petitioner on the charges of first-degree rape and sexual abuse, ultimately giving rise to the post-conviction claim now on review.

In that second trial, the jury began deliberations on a Friday, but they did not reach a verdict by the end of the day, so the trial court dismissed them for the weekend. As a group of four or five of petitioner's jurors were walking down the stairs towards the courthouse exit, two sheriff's deputies happened to be leading petitioner out of a courthouse holding cell into the hallway to be transported back to the jail. While petitioner was being escorted into the hallway, at least three of the jurors viewed petitioner as they walked down the stairs. Upon realizing that some jurors had seen petitioner, the deputies quickly pulled petitioner back into the holding cell, where they held him for another ten minutes before resuming his transport.

Before the jurors saw petitioner in the hallway, he had changed out of his courtroom attire and into his jail clothing, which he described as a blue jumpsuit with a pink t-shirt and pink socks. Petitioner also had been placed in shackles, including handcuffs and ankle restraints that were attached to a belly chain wrapped around his waist, all of which were visible outside of his clothing.³ There is no dispute that petitioner's jail clothes and shackles were plainly visible to the jurors.

When the jury returned to resume deliberations the following Monday, petitioner told defense counsel what had happened. To the best of counsel's recollection, he then asked the sheriff's deputies about the incident and brought it to the “court's attention,” but no discussion of the matter took place or was memorialized on the record. Further, counsel neither raised any objections on the record as to what had occurred nor requested that the trial court declare a mistrial, inquire of the jury, or take any other action.

Later that day, the jury returned their verdict. The jury found petitioner guilty of two counts of first-degree sexual abuse, one count of third-degree sexual abuse as a lesser-included offense, and one count of first-degree rape. Petitioner appealed. The Court of Appeals reversed the sexual abuse convictions on merger grounds, holding that the trial court should have entered a single conviction for first-degree sexual abuse, and remanded for sentencing; it otherwise affirmed. State v. Nelson, 282 Or App 427, 386 P3d 73 (2016). Petitioner was resentenced, resulting in an amended judgment.

C. Post-Conviction Relief Proceedings

Petitioner sought post-conviction relief alleging various instances of inadequate assistance of counsel, including a claim that counsel during his second trial provided inadequate assistance of counsel by failing to move for a mistrial after the jurors viewed him shackled in the hallway.⁴ Petitioner testified in support of that claim at his post-conviction trial:

“[Petitioner's counsel]: On the last day of your second trial in 2013, after trial was finished, but before deliberations, did something happen to make you think that jurors had seen you in the hall?

“[Petitioner]: Well, yes. When * * * they were escorting me out of the holding cell, the two guards that were escorting me they—they seen the—the jurors first and was, like, oh, * * * the jurors see him and then they just opened the door and snatched me back into the holding cell for about ten minutes. And then that's when I * * * realized what had happened.

“[Petitioner's counsel]: And what had happened? What was the concern?

“[Petitioner]: Well, * * * the concern was that they seen me in—in cuffs. It was probably about four or five of the jurors walking down the steps at the time. * * * But that was it.

“[Petitioner's counsel]: And what * * * cuffs would they have seen? Where were they on your body?

“[Petitioner]: Well, I was shackled with belly chains that wrap around your belly and cuffs your—around your wrists and I had shackles on my feet.

“[Petitioner's counsel]: And were those outside of your clothing or were they inside your clothing?

“[Petitioner]: They were outside and I was in the jail jumpsuit.

“[Petitioner's counsel]: Okay. And what color was the jumpsuit.

“[Petitioner]: I believe it was blue with the orange t-shirt and orange—orange—pink t-shirt and pink socks.

“* * * * *

“[Petitioner's counsel]: Okay. And how many jurors do you think saw you in that moment—or those moments?

“[Petitioner]: It was about four or five and I * * * remember seeing probably three of them that actually looked at me. The other two were just, like, walking. They were still walking down the steps. But I know three, for sure, seen me.

“* * * * *

“[Petitioner's counsel]: Okay. And did you tell [trial counsel] what had happened when you saw him?

“ [Petitioner]: Absolutely.

“[Petitioner's counsel]: What do you recall you said?

“[Petitioner]: Well, * * * I explained everything to him how I just explained it to the court that jurors—a few of the jurors had seen me coming out in cuffs being escorted by the officers and how they rushed me back and kind of made a big commotion. And I told him—he made it—he let the court[ ] know. He let the—the judge know and the DA know and that's about—that's all that happened.”

In response, the state submitted a declaration from petitioner's trial counsel, which stated only the following with regard to that encounter:

“As to my conduct alleged by the petitioner, I have searched my memory concerning the petitioner's report to me that he had been seen by members of his jury while being escorted to/from court by sheriff's deputies and that at that time he was bound by handcuffs and leg irons. I have no doubt that the petitioner reported that event to me but I roundly disagree with his assertion that I ‘seemed unconcerned.’ I am certain that I made inquiries of the deputy sheriffs involved and brought the matter to the court's attention. While that may not have been on the record it surely occurred. If the court declined to take action the court must have been satisfied that there was no violation of [petitioner's] rights.”

Although the same attorney testified at the post-conviction trial regarding petitioner's other inadequate-assistance claims, neither party elicited further testimony from counsel about the shackling issue.

In its judgment, the post-conviction court found that petitioner's testimony had been credible “except to the extent that he testified that his trial attorney did nothing after he repo[r]ted the possible contact with jurors while restrained in the courthouse hallway.”⁵ The post-conviction court also made the following findings of fact:

“The second trial went to the jury on Friday April 12, 2013. The jury did not reach verdicts on April 12 and returned to continue deliberations on Monday April 15. As Petitioner was leaving the courthouse on Friday, Petitioner was being escorted out of the holding cell into the courthouse hallway. The guards saw jurors in the hallway and pulled him back into the holding cell. There were 4-5 jurors going down the steps. Petitioner testified that three of the jurors looked at him and saw him. Petitioner was escorted by guards [wearing] his jail clothing and was in handcuffs and shackles. Petitioner was kept in the holding cell for about 10 minutes after the episode and then taken to the jail. Petitioner reported the incident to [his] attorney on Monday morning. [Counsel] remembers Petitioner telling him about the incident and believes that he discussed it with the guards and then the judge and the district attorney in chambers. No discussion regarding the incident appears on the record and counsel made no motions on the record. It does not appear that trial counsel or the court inquired of jurors to determine whether they had observed the Petitioner. Petitioner makes no[ ] claim that he was ever in shackles or cuffs in the courtroom.”

Ultimately, the court denied relief on the basis that petitioner had failed to establish either that defense counsel's performance had been deficient or that any such deficiency had resulted in prejudice to petitioner.

As to counsel's purportedly deficient performance, the post-conviction court observed that petitioner had not cited any Oregon appellate decision that stood for “the proposition that there is basis for a mistrial when a defendant in cuffs and shackles may have been inadvertently and briefly seen by jurors in a courthouse hallway.” The court therefore concluded that petitioner had not established that the law on that issue was so clear that defense counsel's decision not to seek a mistrial had constituted inadequate assistance. The court also reasoned that counsel's actions might have been strategic; that is, counsel might not have wanted a mistrial or might have anticipated that “a [motion for] mistrial would not [be] successful.” Thus, the court concluded, petitioner had not established that “all reasonable trial attorneys in that situation would have filed a motion for a mistrial.”

With regard to whether petitioner had established that counsel's performance had caused him prejudice, the post-conviction court explained that, for similar reasons, petitioner had not. The court reasoned that petitioner had not shown that there was a tendency that the trial court would have granted a motion for mistrial, nor that the jury's sighting of petitioner would have tended to affect the outcome of deliberations.⁶ The court cited federal and state case law from outside Oregon holding that brief and inadvertent juror sightings of a defendant in restraints outside the courtroom only entitled a defendant to a mistrial upon a showing of actual prejudice, which petitioner had not made here. That is, petitioner had not offered “evidence from the viewpoint of the jurors” and, therefore, had not made the requisite showing that the exposure had actually biased the jury against him.

D. Appeal

Petitioner appealed the post-conviction court's judgment, renewing his arguments that counsel in both trials had provided him with inadequate assistance of counsel. As relevant on review, petitioner renewed his assertion that defense counsel had provided inadequate assistance in not taking appropriate steps to protect petitioner's right to a fair trial by moving for a mistrial or making a record of the jurors’ observation of him in shackles.

Petitioner argued that the post-conviction court's assessment of counsel's performance was erroneous because, in reasoning that counsel might have made a strategic decision not to move for a mistrial, the court had necessarily relied on post hoc rationalization that the record did not support. That, petitioner argued, was impermissible under the Court of Appeals’ decision in Delgado-Juarez v. Cain, in which the court explained that a post-conviction court may not engage in post hoc rationalization in a manner that does not reflect “counsel's actual strategic reasoning.” 307 Or App 83, 96-97, 475 P3d 883 (2020) (internal quotation marks and citation omitted). Instead, petitioner argued, the post-conviction court should have recognized that trial counsel had performed inadequately, because the constitutional and common law rights of a criminal defendant to be free from shackles during trial was well founded at the time of petitioner's trial. Thus, petitioner argued, all reasonable counsel would, at a minimum, have sought a mistrial so as to preserve the issue for direct appellate review.

Petitioner further contended that he had established prejudice, because, under another decision of the Court of Appeals, prejudice is presumed when a defendant's restraints are visible to the jury. Sproule v. Coursey, 276 Or App 417, 424, 367 P3d 946, rev. den., 359 Or 777 (2016) (“[W]here the record shows that a criminal defendant was restrained in a manner that could not be effectively shielded from the jury's view, there is a presumption that the shackles are seen by the jury and prejudice results.” (Internal quotation marks omitted.)). In petitioner's view, there was a “reasonable probability” that, had trial counsel moved for a mistrial, either the trial court would have granted the motion or the Court of Appeals would have reversed on direct appeal under Sproule.⁷ According to petitioner, the post-conviction court denied relief based on a distinction not rooted in Sproule—or in any Oregon decision—that viewing a defendant in shackles in the courtroom is somehow distinguishable from a similar viewing in a courthouse hallway, such that the principles warranting a presumption of prejudice are limited to in-court shackling.

For its part, the state acknowledged there to be “ample available case law” establishing that criminal defendants have the right to be free from shackling during trial and that visible shackling in front of the jury may impair the defendant's constitutional rights. Yet, in the state's view, the governing case law at the time of petitioner's trial would not have led all reasonable attorneys to conclude that a mistrial was warranted under the circumstances of petitioner's case. Further, like the post-conviction court, the state cited authority from other jurisdictions holding that a juror's brief glimpse of a defendant in shackles outside the courtroom is not presumptively prejudicial. And without the benefit of a presumption of prejudice, the state argued, petitioner had not established actual jury bias. It followed, the state reasoned, that he also had not satisfied the prejudice prong of his post-conviction claim.

The Court of Appeals affirmed the post-conviction court's decision to deny relief on all of petitioner's claims. Concerning counsel's performance with regard to the shackling issue, the Court of Appeals reasoned that, because petitioner had cited no Oregon case law addressing “the issue of jurors briefly observing a criminal defendant in restraints outside of the courtroom and whether that constitutes grounds for a mistrial[,]” it was “not clear that every trial counsel exercising reasonable skill and judgment would have necessarily sought a mistrial under the circumstances * * *.” 334 Or App at 329. And as to prejudice, the court rejected petitioner's contention that, under Sproule, the facts of petitioner's case gave rise to a presumption of jury prejudice. Id. at 330. And, because petitioner had presented no direct evidence demonstrating that the jurors who saw him in shackles had recognized him and were prejudiced as a result, he had not established that defense counsel's acts or omissions at trial had tended to affect the outcome of his case. Id.

Petitioner sought review in this court, which we allowed. Before this court, he renews only his argument that counsel was inadequate in failing to move for a mistrial or make a record of the jurors’ observations or resulting perceptions in his second trial, and he contends that counsel's failure caused him prejudice. As to the latter contention, petitioner asks this court to recognize a presumption of prejudice that arises whenever jurors in a criminal case see the defendant in shackles during the course of trial, regardless of whether that sighting occurs inside or outside of the courtroom. In petitioner's view, such a presumption is necessary to give effect to the right to an “impartial jury” under Article I, section 11. As we will explain, we do not consider it necessary to recognize a presumption of prejudice to conclude that petitioner was prejudiced in this case, but we otherwise agree with petitioner.

II. ANALYSIS

A. Trial counsel failed to exercise reasonable professional skill and judgment.

We begin with the performance aspect of petitioner's post-conviction claim. That requires us to determine whether petitioner has established by a preponderance of the evidence that, by not moving for a mistrial, trial counsel failed to exercise reasonable professional skill and judgment. Lichau, 333 Or at 359. Like the post-conviction court and the Court of Appeals, we recognize that, at the time of petitioner's trial, no Oregon appellate decision had held that jurors viewing a defendant wearing shackles outside the courtroom impairs the defendant's right to an impartial jury or otherwise implicates the same concerns that in-court observations raise. Nonetheless, in light of the shackling case law available to counsel at the time of petitioner's trial, as well as the principles underlying that case law, we conclude that competent counsel would have both viewed the circumstances in this case as providing colorable grounds for a mistrial motion and acted to protect petitioner's right to be tried before an impartial jury. We further conclude that, in failing to move for a mistrial in this case, counsel failed to provide petitioner with the adequate assistance that Article I, section 11, guarantees.

1. The state of the law at the time of petitioner's trial

As explained above, when evaluating the adequacy of trial counsel's performance, our focus is on the circumstances as counsel reasonably understood them to be, and not how, in retrospect, we conclude counsel might have perceived them. See Lichau, 333 Or at 360 (in determining whether counsel's performance was constitutionally adequate, we “must make every effort to evaluate a lawyer's conduct from the lawyer's perspective at the time, without the distorting effects of hindsight”). With respect to counsel's understanding of the law that applied under the circumstances, our approach is similar. That is, when assessing whether trial counsel should have made a particular legal argument, we look at the law “as a lawyer would have seen it at the time.” Burdge, 338 Or at 498 (so stating with regard to whether counsel should have recognized statutory ambiguity later identified by the courts). Thus, in this case, we begin our assessment of counsel's performance with a review of the state of the law regarding shackling as it existed at the time of petitioner's trial.

By that time, Oregon case law had firmly established that, at least when a criminal defendant is brought in shackles before the jury within the courtroom, there are grounds for objection or even a mistrial. That case law dates back well over a century, to State v. Smith, in which this court reversed the conviction of a defendant who had been sentenced to death for murder, because he had been “kept with irons on his feet during trial * * *.” 11 Or 205, 207, 8 P 343 (1883). During that defendant's trial, defense counsel had sought to have the irons removed, but the trial court had denied that request because the irons had been “put on the defendant at the penitentiary and could not be removed without much delay, and it would require the work of a blacksmith to move them * * *.” Id. (internal quotation marks omitted). Relying on even older case law from California, this court held that “it was error to keep the prisoner in fetters during the trial” in violation of the common law rule “first * * * considered and enforced” in People v. Harrington, 42 Cal 165, 10 Am Rep. 296 (1871). Id.

Then, in State v. Long, we explained that a criminal defendant has the right to be tried free from shackles “ ‘unless [the defendant is] shown to be a desperate character whose restraint is necessary to the safety and quiet of the trial.’ ” 195 Or 81, 91, 244 P2d 1033 (1952) (quoting Eaddy v. People, 115 Colo 488, 491, 174 P2d 717, 718 (1946)). We explained that one rationale for the rule was that “[a] prejudice might be created in the minds of the jury against a prisoner who should be brought before them handcuffed and shackled, which might interfere with a fair and just decision of the question of the guilt or innocence of such prisoner.” Id.⁸ Long further entrenched the rule that there are grounds for a mistrial when a defendant is shackled in the presence of the jury during trial. In adhering to that rule, this court appears to have recognized that a defendant's appearance before the jury in shackles can instill an inference of guilt in the minds of jurors or, put another way, that appearing in shackles in the presence of the jury impinges upon the presumption of innocence afforded all criminal defendants.

Smith, Long, and the common law cases they relied upon have been cited by Oregon appellate cases to support the principle that, absent a record showing a substantial justification for shackling a defendant, it is reversible error to keep the defendant in shackles during trial. See, e.g., State v. Wall, 252 Or App 435, 437-39, 287 P3d 1250 (2012), rev. den., 353 Or 280 (2013) (identifying that case law and related principles). Indeed, before petitioner's trial in this case, the Court of Appeals had also developed its own body of case law regarding shackling built, in part, upon this court's decisions. For example, in State v. Kessler, the Court of Appeals held that the trial court had erred when it simply accepted the prosecutor's assertion that the defendant presented a security risk warranting shackling during trial rather than requiring the prosecutor to put on “evidence of an immediate and serious risk of dangerous and disruptive behavior.” 57 Or App 469, 473, 645 P2d 1070 (1982) (citing State v. Moore, 45 Or App 837, 839-40, 609 P2d 866 (1980)). Without the requisite showing of substantial necessity, requiring the defendant to wear shackles in the presence of the jury had deprived him of a fair trial. Id. at 475. In reaching that conclusion, the Court of Appeals identified four types of “manifest prejudice” that result when a person is tried in shackles: first, shackles “impinge on the presumption of innocence”; second, they detract from “the dignity of the judicial proceedings”; third, shackles “may inhibit consultation” with defense counsel; and fourth, they may inhibit the defendant's “decision whether to take the stand as a witness.” Id. at 474.⁹

In concluding that shackling leads to those harms, Kessler relied on the following reasoning set forth in People v. Duran, 16 Cal 3d 282, 290, 545 P2d 1322, 1327 (1976), in which the California Supreme Court adhered to the principles first articulated by that court in Harrington:

“We believe that it is manifest that the shackling of a criminal defendant will prejudice him in the minds of the jurors. When a defendant is charged with any crime, and particularly if he is accused of a violent crime, his appearance before the jury in shackles is likely to lead the jurors to infer that he is a violent person disposed to commit crimes of the type alleged. * * * The removal of physical restraints is also desirable to assure that ‘every defendant is [* * *] brought before the court with the appearance, dignity, and self[-]respect of a free and innocent man.’ Eaddy, 115 Colo at 492 * * *. We believe that possible prejudice in the minds of the jurors, the affront to human dignity, the disrespect for the entire judicial system which is incident to unjustifiable use of physical restraints, as well as the effect such restraints have upon a defendant's decision to take the stand, all support our continued adherence to the Harrington rule.”

Duran, 16 Cal at 290, 545 P2d at 1327; Kessler, 57 Or App at 472 (quoting that passage from Duran).

The Kessler court made it clear that shackling invites the jury to consider improper influences—such as inferring that a defendant has violent tendencies—unrelated to the question of guilt or innocence or the evidence presented at trial. The court further suggested that the unwarranted shackling of a defendant in front of a jury can be deemed such an indignity as to detract from jurors’ respect for the judicial system, respect that is essential to a trial court's ability to ensure a fair and functioning trial. And it is the need to avoid such consequences that appears to animate the prohibition against unnecessarily allowing jurors to view defendants in shackles, a prohibition of sufficient magnitude that disregarding it constitutes reversible error on direct appeal.

Moreover, by holding that the prejudice resulting from such a violation is “manifest,” the Kessler court rendered it unnecessary for a defendant who has been shackled during trial without a showing of substantial necessity to prove individualized prejudice—the mere fact that they were shackled without any apparent justification was sufficient to establish a “violation of [their] due process right to a fair trial.” Id. at 475.¹⁰

At the time of petitioner's trial, this court had not expressly considered Kessler’s presumption of prejudice nor adopted its reasoning as to why shackling a defendant in the presence of the jury without a showing of substantial necessity deprives the defendant of a fair trial. It had, however, relied on Kessler as imposing an obligation on appellate counsel to raise the issue when the record shows that a defendant was brought before the jury in shackles without a finding of substantial necessity having been made. Guinn v. Cupp, 304 Or 488, 498, 747 P2d 984 (1987) (stating, in post-conviction case alleging inadequate assistance of appellate counsel for failing to raise shackling issue, that a petitioner can establish a prima facie post-conviction claim simply by showing that they were “shackled during the trial and that no record was made in the trial court to support the decision to shackle” the defendant). In reaching that conclusion, we reiterated that, “[a]t common law, a criminal defendant had the right to appear at trial free of restraint, absent a substantial showing of necessity for confinement[,]” id. at 493 (citing Smith, 11 Or at 205), and we noted that the federal Due Process Clause likewise protects that right, id. (citing Estelle v. Williams, 425 US 501, 505, 96 S Ct 1691, 48 L Ed. 2d 126 (1976) (trying defendant in prison clothing impinges upon presumption of innocence, thereby denying a fair trial); and Illinois v. Allen, 397 US 337, 344, 90 S Ct 1057, 25 L Ed. 2d 353 (1970)). By relying on Kessler as having established the elements of a prima facie post-conviction claim in the shackling context, see id. at 498, our opinion in Guinn effectively endorsed the principles articulated in Kessler—including its articulation of how shackling can impair a defendant's right to a fair trial—and made it clear that failing to take action to protect that right could constitute inadequate assistance of counsel.¹¹

2. Competent counsel would have recognized the serious constitutional concerns that arose at petitioner's trial and moved for a mistrial.

Given the state of Oregon's shackling case law at the time of petitioner's trial, we conclude that, in opting not to seek a mistrial—and seeming instead to rely on the trial court to act sua sponte if it believed that petitioner's rights had been violated—counsel did not exercise reasonable professional skill and judgment. As we will explain, under that case law, there was, at a minimum, a reasonable argument to be made that allowing jurors to view a defendant in shackles outside the courtroom violates the defendant's right to trial before an impartial jury under Article I, section 11, thereby warranting a mistrial, and all competent counsel would have seen that argument and recognized its obvious beneficial value. By not making that argument under the circumstances of this case and instead simply deferring to the trial court to act if it were so inclined, counsel failed to provide constitutionally adequate assistance. Petitioner has therefore satisfied the performance prong of his inadequate assistance of counsel claim.

To exercise reasonable professional skill and judgment, defense counsel must investigate the facts and familiarize themselves with the applicable law sufficiently to permit them to make informed decisions on behalf of the defendant. Krummacher, 290 Or at 875. Here, after petitioner told counsel that three jurors had seen him outside the courtroom in shackles, counsel was obligated to investigate those circumstances and familiarize himself—if he were not already familiar—with Oregon shackling law as discussed above. See id. (facts and law with which counsel must familiarize themselves depend upon “the nature and complexity of the case”). The post-conviction court found that counsel “believe[d] that he discussed it with the guards[,]” and petitioner neither disputes that counsel spoke with the sheriff's deputies nor suggests that counsel should have undertaken some further factual inquiry. Thus, to the extent that counsel provided petitioner with inadequate assistance, petitioner does not appear to contend that counsel did so by failing to sufficiently investigate the facts.¹²

The extent to which counsel prepared himself on the applicable law is less clear. Neither counsel's declaration nor his oral testimony addressed whether he had ever conducted any legal research regarding shackling, and the post-conviction court made no findings as to whether counsel had previously familiarized himself with Oregon shackling law or did so before electing not to move for a mistrial. The circumstances would have alerted competent counsel that a potentially significant issue had arisen, and a baseline understanding of Oregon shackling law would have been essential for counsel to make an informed choice on petitioner's behalf. See Richardson, 362 Or at 257 (counsel's failure to investigate was unreasonable given the factual information that defense counsel knew and the high stakes for the petitioner).

Implicit in the post-conviction court's opinion is an assumption that counsel would or should have been aware of Kessler and the general rule that a defendant has a constitutional right to appear in front of the jury free of physical restraints. Although nothing in the record directly supports a finding that counsel was, in fact, generally aware of Oregon shackling law, petitioner does not contend that counsel was inadequate for not having familiarized himself with the applicable law to at least that extent. Rather, petitioner contends that, based upon that general shackling law, constitutionally adequate counsel would have moved for a mistrial under the circumstances presented here. The post-conviction court concluded otherwise. As explained above, that court reasoned that, although cases like Kessler and Wall left “no doubt” regarding petitioner's right to be tried free of physical restraints absent substantial justification, no Oregon case law would have alerted counsel that he should move for a mistrial here. Further, because petitioner had “cite[d] no cases for the proposition that there is basis for a mistrial when a defendant in cuffs and shackles may have been inadvertently and briefly seen by jurors in a courthouse hallway,” petitioner had not established “that the law on the issue was so clear” that counsel had provided inadequate assistance in electing not to move for a mistrial.¹³

The post-conviction court erred in limiting counsel's obligation to advancing only “clear” statements of existing law on behalf of petitioner. As this court has previously explained, competent representation is not limited to being familiar with and applying “clear” statements of law; counsel may also be inadequate for not making an argument based on an unsettled or ambiguous question of law where the argument, if successful, would “so obviously offer possible benefits to a defendant that any lawyer exercising reasonable professional skill and judgment would raise it.” Burdge, 338 Or at 500. That is, if (1) a legal issue arises in the course of representing a defendant; (2) the legal issue has not yet been decided under controlling case law; (3) the issue is susceptible of an interpretation that so obviously would offer potential benefits to the defendant that all reasonable counsel would identify it; and (4) all such counsel would conclude that, under the circumstances, the potential benefits of advancing that interpretation exceeded any risks, then the failure to make that argument constitutes inadequate assistance of counsel. Id. at 497 (so stating with regard to whether counsel was inadequate for failing to identify a statutory ambiguity and argue for resolving ambiguity in favor of the defendant).

We have not limited the principle articulated in Burdge to unsettled issues of statutory law. See, e.g., Antoine v. Taylor, 368 Or 760, 775-79, 499 P3d 48 (2021) (discussing Burdge in connection with potential ambiguity in case law). Further, we conclude that an attorney exercising reasonable professional skill and judgment would have argued that, for much the same reason that having a defendant shackled in the presence of the jury in the courtroom is grounds for a mistrial, a mistrial is warranted when jurors observe a defendant in shackles in the courthouse hallway. Although this court has never expressly held that the right to be tried free of physical restraints is applicable outside the courtroom, it also has never limited that right to in-court observations.¹⁴ And at least some of the concerns that Kessler and Guinn recognized as resulting from in-court shackling could be reasonably understood to arise in the out-of-court context as well, and nothing in the reasoning of our shackling cases compels the conclusion that the right to be free of shackles is limited to the in-court setting. Furthermore, although the post-conviction court described the jurors’ observation of petitioner in the hallway as brief and inadvertent, neither that court nor the state in defending its ruling have identified anything in the record to support an inference that the jurors would have perceived things differently based upon petitioner's location at the time. See Kessler, 57 Or App at 474 (noting that prejudicial inferences that a defendant is a dangerous person logically follow when the defendant is “see[n] in shackles at all[,]” not merely when jurors are “constantly exposed” to that sight).

Given the circumstances of petitioner's trial, a lawyer exercising reasonable professional skill and judgment would have recognized an opportunity for “advancing [an] interpretation” of the law that benefitted petitioner. Burdge, 338 Or at 497. Counsel in petitioner's case does not appear to have recognized that opportunity. Instead, rather than moving for a mistrial on that basis, counsel at most had a conversation with the judge in chambers, seemingly deferring to whether the trial court thought petitioner's rights had been compromised. And because nothing in the record suggests that counsel was concerned that advancing that argument would carry any risks for petitioner, much less any risk that outweighed the benefits of obtaining a mistrial, the failure to do so here constituted inadequate assistance of counsel. See id.¹⁵

Resisting the conclusion that counsel provided inadequate assistance, the state posits that counsel might have had strategic reasons for not moving for a mistrial or taking any other action on the record. Echoing the post-conviction court, the state reasons that counsel may have believed that a motion for mistrial would not succeed. That argument falls short for two reasons. First, nothing in the record suggests that counsel considered moving for a mistrial, weighed the risks and benefits of doing so, and then made a strategic decision not to seek a mistrial based upon the perceived likelihood that the motion would simply fail. The record does not permit us to conclude that counsel's performance reflected a reasonable tactical decision when it does not suggest strategic reflection of any kind. Cf. Montez v. Czerniak, 355 Or 1, 27-28, 322 P3d 487, adh'd to as modified on recons., 355 Or 598, 330 P3d 595 (2014) (where there was evidence of counsel's actual reasoning, declining to “indulge in posthoc rationalizations of trial counsel's decisions”). Second, standing alone, the belief that a motion would not succeed in the trial court does not relieve counsel of their duty to advance an obviously beneficial argument based on a reasonable interpretation of the law. Even if the motion might have failed, the state does not suggest that an unsuccessful motion could in any way have rendered petitioner worse off at trial. Moreover, as we discuss in greater detail in addressing whether counsel's performance resulted in prejudice to petitioner, even an unsuccessful mistrial motion could have led to the trial court taking other protective measures.¹⁶ Thus, even if we could impute a tactical rationale to counsel's decision not to seek a mistrial despite there being no suggestion of any rationale in the record, we could not conclude that counsel's decision reflected a reasonable exercise of professional skill and judgment here.

The dissent likewise argues against the conclusion that petitioner received inadequate assistance of counsel and contends, among other things, that we have not properly applied the standard of review. Specifically, the dissent suggests that we have not properly accounted for the possibility that, after speaking with the transport deputies, the judge, and the district attorney, “trial counsel * * * concluded from his investigation that jurors had not recognized the person being transported” as petitioner and, consequently, opted not to seek a mistrial. ––– Or at –––– (Bushong, J., dissenting) (slip op at 7:9-10). The dissent suggests that the post-conviction court implicitly made that finding, because it would be consistent with that court's ultimate legal conclusion that counsel provided adequate assistance to petitioner. Id. at –––– (slip op at 6:8 - 7:11).

We disagree. It is true that, had counsel's limited investigation led him to conclude that the jurors had either not seen or not recognized petitioner, then his apparent decision not to seek a mistrial might well have been reasonable. But the standard that the dissent refers to would require us to presume that the post-conviction court made that finding only if it were a disputed factual matter and the post-conviction record could support it. Ball, 250 Or at 487 (where findings are not made on all relevant historical facts, “and there is evidence from which such facts could be decided more than one way,” the court presumes that the factfinder decided the facts consistently with its ultimate legal conclusion). In the dissent's view, there necessarily was a dispute about what counsel had “learned and concluded as a result of his investigation, because that undoubtedly affected his determination that there was ‘no reason’ to request a mistrial.” ––– Or at –––– (Bushong, J., dissenting) (slip op at 7:17 - 8:1). Respectfully, that view is wrong. It is axiomatic that, for a fact to be in dispute, there must be evidence of that fact in the record. See id. at –––– (slip op at 9:5 - 6) (applying principle when “there is evidence from which such facts could be decided more than one way”). And as the dissent acknowledges, there at least “is no direct evidence of what counsel had learned and concluded from his investigation.” ––– Or at –––– (slip op at 8:1 - 2). Because the record is silent as to what counsel concluded after speaking with petitioner and the others, we cannot manufacture a factual dispute simply because its resolution in the state's favor would support the post-conviction court's ultimate ruling.

At best, had the post-conviction court thought at all about counsel's conclusion in that regard, it could perhaps have speculated that, because counsel might have learned that the transport deputies thought that the jurors might not have recognized petitioner, counsel might have concluded that the incident had not prejudiced petitioner. We do not understand the applicable standard of review to permit us to endorse that level of speculation, much less require us to assume that the post-conviction court implicitly made that finding and to account for it here. That is particularly so given that, in this case, the post-conviction court's ultimate legal conclusion does not appear to have relied on what counsel did or did not conclude about the jurors’ actual perceptions; instead, that court emphasized the state of the law at the time and potential strategic justifications for counsel's inaction.¹⁷

Thus, we respectfully decline the dissent's suggestion that we account for the possibility that counsel concluded that the jurors had not recognized petitioner before opting not to seek a mistrial. Rather, we conclude that counsel's obligation to act arose when petitioner informed him that the jurors had seen him in shackles ¹⁸ and that, given the record created in the post-conviction court, there is no nonspeculative basis on which to conclude that counsel's failure to seek a mistrial constituted an exercise of reasonable professional skill and judgment.

The dissent separately appears to argue that not all reasonable counsel would have recognized an opportunity under the existing case law to argue that the circumstances warranted a mistrial, but again we disagree. The dissent acknowledges that, at the time of petitioner's trial, it was well established that a mistrial is warranted when a criminal defendant is brought before the jury in the courtroom in shackles. ––– Or at –––– (Bushong, J., dissenting) (slip op at 24:13 - 14). The dissent emphasizes, however, another undisputed point that, at that time, it was not so well established what happens if a juror happens to see a criminal defendant in shackles outside the courtroom, while being transported to jail at the end of the day. ––– Or at –––– (slip op at 24:14 - 17). From that undisputed premise, the dissent turns to this court's opinion in State v. Walton, 311 Or 223, 809 P2d 81 (1991), and case law from other jurisdictions, contending that those cases provide relevant guidance as to whether petitioner met his burden of proving that counsel failed to exercise reasonable professional skill and judgment in his case. Id. at –––– (Bushong, J., dissenting) (slip op at 20:8 - 23:21).

To the extent that the dissent reasons that the absence of Oregon case law holding that a mistrial was warranted under petitioner's circumstances—or the existence of case law from other jurisdictions distinguishing between in-court and out-of-court encounters—means that defense counsel provided adequate assistance in this case, we respectfully disagree. First, as we have explained, defense counsel's professional obligations are not limited to asserting a defendant's “firmly established” rights; rather, where there is a legitimate question whether a right exists under the law, counsel may be obligated under certain circumstances to advance an argument for that right, and we have concluded that those circumstances were present here. Second, nothing in Walton or any other decision that the dissent relies on refutes that principle or precludes its application here.

Starting with Walton, counsel in that case, unlike defense counsel here, had moved for a mistrial after a sitting juror and an alternate juror saw the defendant, along with other prisoners, in handcuffs.¹⁹ 311 Or at 250. Although we held that the trial court had not erred in denying the defendant's motion, Walton is still wholly consistent with our conclusion in this case. First, although that case, like petitioner's, appears to have involved a relatively brief, out-of-courtroom encounter, this court evaluated the trial court's decision whether to grant a mistrial for an abuse of discretion. See id. (citing State v. Montez, 309 Or 564, 789 P2d 1352 (1990)); see also Montez, 309 Or at 595 (“A motion for mistrial is addressed to the sound discretion of the trial judge[.]”).²⁰ Second, as relevant to both the performance and prejudice prongs of petitioner's inadequate-assistance claim, the trial court in Walton responded to defense counsel's mistrial motion with action: addressing the potentially tainted jurors, confirming that they had not been influenced by the incident and had not discussed it with other jurors, instructing “both jurors that they were to disregard their observations, that they were not to discuss their observations with anyone, and that [the] defendant's custody status had no bearing on his guilt or innocence[,]” and, notwithstanding all those protective measures, further offering to replace the seated juror with an alternate juror who had not experienced the concerning exposure. Walton, 311 Or at 250. Third, because Walton merely held that a mistrial was not required when those measures had been taken in response to defense counsel's motion—including the trial court's offer to ensure a jury consisting wholly of jurors who had not seen the defendant in handcuffs—Walton in no way foreclosed counsel from arguing on petitioner's behalf that the circumstances in this case warranted a mistrial, especially in the absence of any protective measures by the trial court.

Similarly, we are not persuaded by the dissent's reliance on case law from other jurisdictions holding that, absent a showing of actual prejudice, jurors’ out-of-court observations of defendants in shackles do not provide grounds for reversal.²¹ Cases that apply different prejudice standards to in-court and out-of-court observations have no bearing on the adequacy of counsel's performance, a distinction that the dissent itself seems to recognize. See ––– Or at –––– (Bushong, J., dissenting) (citing State v. Hardy, 230 Ariz 281, 292, 283 P3d 12 (2012) (“A juror's inadvertent exposure to the defendant in handcuffs outside the courtroom * * * is not inherently prejudicial, and a defendant is not entitled to a new trial absent a showing of actual prejudice.”)) (slip op at 22:22 - 23:8). And here, the performance-related question is not whether the trial court would have been required to grant a mistrial had counsel requested one; the question is whether counsel was obligated to move for a mistrial under the circumstances, a question that we have answered affirmatively. Moreover, as we explain more fully below in addressing prejudice, in many of the out-of-jurisdiction cases, the trial courts, like the trial court in Walton, had afforded the defendants at least some form of relief following an out-of-court shackling incident where defense counsel had sought relief. Thus, that nonbinding case law discussing the prejudice standard for out-of-courtroom encounters fails to provide even persuasive authority that counsel's decision not to seek relief at all constituted adequate performance in this case.

B. Trial counsel's performance prejudiced petitioner.

Having concluded that petitioner's trial counsel failed to exercise reasonable professional skill and judgment, we turn to whether that failure resulted in prejudice to petitioner, entitling him to post-conviction relief. To establish prejudice in an inadequate-assistance-of-counsel claim under Article I, section 11, a petitioner must demonstrate that counsel's deficient performance “had a tendency to affect the result of his trial.” Green, 357 Or at 312 (internal quotation marks omitted). And because many things can affect the outcome of a jury trial, a petitioner is not required to establish as a matter of probability that counsel's inadequate performance affected that outcome. Id. at 322 (discussing that principle in reference to the court's decisions in Lichau and Stevens). Rather, “the tendency to affect the outcome standard demands more than mere possibility, but less than probability.” Id. Thus, to satisfy the prejudice requirement here, petitioner must show that trial counsel's failure to move for a mistrial “could have tended to affect” the outcome of his case. Id. at 323 (internal quotation marks and emphasis omitted). We conclude that petitioner has met that standard. That is, given the state of the law at the time of petitioner's trial, there is more than a mere possibility that the trial court, in its exercise of discretion, would have granted a mistrial had counsel requested one.

As an initial matter, the state and the dissent each argue, for slightly different reasons, that we should be circumspect in applying Green’s “tendency to affect the outcome” standard. Both contend that petitioner cannot establish prejudice without first showing that the circumstances would have entitled him to a mistrial had counsel properly moved for one, but they arrive at that conclusion differently. The dissent agrees generally that Green’s “tendency to affect the outcome” standard applies to the trial court's hypothetical decision whether to grant a mistrial, had one been requested.²² Where the dissent parts ways with us, however, is in our understanding that, in this case, a “tendency to affect the outcome” means that there was “more than a mere possibility” that the trial court would have granted a mistrial motion. In the dissent's view, Green limited the application of the “more than a mere possibility” articulation to cases in which “the effect of inadequate assistance of counsel on the outcome of a jury trial is at issue,” Green, 357 Or at 322 (emphasis added); see id., 357 Or at 323 n 13 (contrasting cases in which the prejudice inquiry relates to other outcomes, such the entry of a guilty plea, where the court has applied a probability standard). ––– Or at –––– (Bushong, J., dissenting) (slip op at 27:15 - 28:9). Relying on that distinction, the dissent argues, in relevant part, that “to establish prejudice from counsel's failure to move for a mistrial, a petitioner must show that the trial court would have granted the motion * * *.” ––– Or at –––– (slip op at 31:7 - 8). At least for purposes of the present dispute, we disagree with the dissent's understanding of the “tendency to affect the outcome” standard.

Whatever merit the dissent's distinction may have in other contexts, we disagree that Green supports applying a probability standard in this case. In Green, the petitioner had gone to trial on allegations that he had engaged in nonconsensual sexual behavior with nine different victims, some of whom were incapable of consent due to their age, and some of whom alleged that the petitioner had used forcible compulsion against them. Green, 357 Or at 303-05. At issue on post-conviction was whether defense counsel had performed deficiently by failing to request a limiting instruction that would require the jury to consider only the evidence related to each individual victim in determining whether the state had proved its case as to that victim. Id. at 312-13. After concluding that neither the post-conviction court nor the Court of Appeals had had the benefit of this court's then recent decision in Pereida-Alba and that it should reverse and remand the case for further consideration, id. at 320, the court turned to the issue that the dissent focuses on here—the appropriate standard for evaluating prejudice, id at 321.

In its assessment of prejudice, the Court of Appeals had required the petitioner “to show that his trial counsel's deficient exercise of skill and judgment had a ‘tendency to affect the result of the prosecution,’ ” relying on a phrase that this court had first used more than 30 years earlier in Krummacher. Id. The state had argued in favor of the probability-based standard that the post-conviction court had applied, which would have required the petitioner to show what would have happened had counsel performed adequately, rather than merely what could have happened. Id. After a brief survey of its own case law, this court rejected that argument and held that, “because many different factors can affect the outcome of a jury trial, in that setting, the tendency to affect the outcome standard demands more than mere possibility, but less than probability.” Id. at 322. The dissent reads into that passage the court's intent to apply that standard only to jury outcomes, but the dissent reads Green too narrowly.

To be sure, this court in Green observed in a footnote that, in another context, the court had “described the tendency to affect the outcome test in terms of actual probability.” Id. at 323 n 13 (citing Moen v. Peterson, 312 Or 503, 513, 824 P2d 404 (1991) (discussing standard in context of petitioner's decision to enter a guilty plea)). But despite that observation, the court did not purport to create a bright-line rule, with the “tendency to affect the outcome” standard applicable only to jury verdicts. And, to date, no appellate decision of which we are aware has considered whether such a limitation should exist, much less held that Green had established one.²³

Here, although the state argues for a nuanced application of Green, neither it nor any other party or court besides the author of the dissent has suggested that the “more than a mere possibility” articulation of the “tendency to affect the outcome” standard does not apply to petitioner's claim; under those circumstances, we are hesitant to begin that inquiry at this stage. Moreover, to the extent that this court might someday recognize the limitation that the dissent posits, we would question whether this case is of the sort in which to make such a distinction. As noted above, the trial court's decision whether to grant a mistrial is a matter of discretion. That is because the trial court is in the best position to assess prejudice and to determine what, if any, corrective measures are appropriate. In this case, no one has suggested that, had the trial court declared a mistrial, whether at the request of counsel or on its own motion, it would have abused its discretion. Nor is this court likely to have second guessed the trial court's reasoning if it had denied such a request. Thus, like the outcome of jury deliberations, the outcome of a trial court's discretionary choice regarding a mistrial motion can be affected by many different factors. See id. at 322 (reasoning that use of a probability standard to assess prejudice is inappropriate in the context of a jury trial because, in that setting, many different—and presumably unknown—factors can affect the jury's decision). In that regard, petitioner's claim is similar to a claim related to a jury verdict—where the reasons for a decision are largely unknowable—and dissimilar to, for example, a decision whether to accept a plea offer, a decision that a post-conviction petitioner is in a unique position to explain. See id. at 323 n 13 (distinguishing cases involving decisions to enter guilty pleas). We therefore apply the same standard here.

Applying Green’s “more than a mere possibility” standard to counsel's failure to move for a mistrial, we conclude that petitioner has established that counsel's performance resulted in prejudice. Had counsel requested a mistrial, there is more than a mere possibility that the trial court would have declared a mistrial, thereby affecting the outcome of his case. As we will explain, that conclusion follows primarily from the nature of mistrial motions—particularly from the fact that such motions trigger the discretion of a trial court—and the case law that would have informed the trial court's exercise of that discretion.

Here, if counsel had sought a mistrial, it would have been within the discretion of the trial court to grant that request. This court has long held that “[w]hether to grant a mistrial is a determination committed to the sound discretion of the trial court.” State v. Rogers, 313 Or 356, 381, 836 P2d 1308 (1992), cert. den., 507 US 974 (1993); id. (citing State v. Smith, 310 Or 1, 24, 791 P2d 836 (1990) (“The question thus is not whether this court would have granted a new trial to [the] defendant, but whether the trial court abused its discretion in refusing to do so.”)). We afford courts that discretion because trial judges are in the best position to “assess and to rectify the potential prejudice to the defendant.” State v. Farrar, 309 Or 132, 164, 786 P2d 161, cert. den., 498 US 879 (1990). The question here, of course, is not whether the trial court erred in denying a motion for mistrial, because no motion was made. Instead, the question for purposes of post-conviction relief is whether, if defense counsel had moved for a mistrial, there is more than a mere possibility that the trial court would have granted that request or, if it concluded that counsel's concerns could be addressed through corrective action short of a mistrial, taken such action. Given the specific circumstances of petitioner's trial and the Oregon case law that then existed, we conclude that the answer is yes.

As with our assessment of defense counsel's performance, we find it instructive to consider the state of the law at the time of petitioner's trial in determining whether petitioner has established that counsel's performance was prejudicial. The post-conviction court and the Court of Appeals similarly considered the then-existing case law in connection with both aspects of petitioner's post-conviction claim. But unlike those courts, we do not consider that case law to determine whether the trial court would clearly have been obligated to declare a mistrial if one had been requested; rather, we consider the then-existing case law for its relevance in assessing whether the trial court's potential understanding of—and response to—the existing case law means that counsel's failure to seek a mistrial had a tendency to affect the outcome of petitioner's case.

As we have explained, at the time of petitioner's trial, the Court of Appeals had well established case law providing that, if jurors see a defendant shackled in the courtroom during trial without a finding of substantial necessity on the record, the defendant is entitled to a new trial upon request. Kessler, 57 Or App at 475. Further, in Guinn, this court effectively endorsed Kessler and its underlying rationale in concluding that a post-conviction petitioner makes a prima facie case of ineffective assistance of counsel by showing that they were “shackled during the trial and that no record was made in the trial court to support th[at] decision[.]” 304 Or at 498.

A key concern in Kessler was the potential effect of such a sighting on the presumption of innocence. And, as we have noted in this case, the effect that shackling a criminal defendant has upon the presumption of innocence is not inherently limited to in-court shackling. Kessler itself did not suggest such a distinction. See Kessler, 57 Or App at 474 (“It is not a question whether the jury will be constantly exposed to the vision of a shackled defendant, but whether they see the shackles at all. If any of the jurors could see the leg irons, the prejudicial inferences logically follow.”).²⁴

Before this case, the only circumstance in which Oregon courts have limited the application of Kessler’s principles is when there was a question whether the defendant's restraints had been seen at all, and not when they were seen in one location as opposed to another. See, e.g., State v. Bowen, 340 Or 487, 496, 135 P3d 272 (2006), cert. den., 549 US 1214 (2007) (on a plain error challenge to the use of a stun belt during a murder trial, declining to find that defendant had been prejudiced where there was “no evidence in the record that the stun belt * * * was visible to the jury, and, therefore, [the] defendant [could not] claim that the jury was biased by its presence”); State v. Bates, 203 Or App 245, 251, 125 P3d 42 (2005), rev. den., 340 Or 483 (2006) (on direct appeal, any error in compelling the defendant to wear a stun belt was harmless because the jury was not aware that the defendant was wearing the belt; therefore, use of belt was not likely to cause prejudice by “impinging on the presumption of innocence and the dignity of the judicial proceeding”).

In light of that case law, the trial court in petitioner's case may not have made the distinction that the state urges here—that the jury incident could be disregarded because of where petitioner was seen in shackles. Instead, the trial court may have understood the relevant question to be only whether jurors had seen the shackles at all. Although the court would not have had the benefit of the Court of Appeals’ later decision in Sproule, that case applies both Bowen and Bates and illustrates why the trial court might have had that understanding. In Sproule, the court held that the petitioner's trial counsel had not been inadequate in failing to object to the use of a leg brace as a restraint when the jurors had not been aware that the petitioner was restrained. 276 Or App at 426 (“[G]iven that the record supports the post-conviction relief court's implicit finding that the jury was not aware that petitioner was restrained, we will not presume that the restraint improperly impinged on the presumption of innocence or the dignity of the proceedings, and thereby prejudiced petitioner.”). But had the trial court in Sproule not taken pains to prevent jurors from inadvertently observing the restraints, the Court of Appeals would have presumed both that the jury had seen the restraints and that the petitioner had been prejudiced as a result:

“Thus, where the record shows that a criminal defendant was restrained in a manner that could not be effectively shielded from the jury's view, there is a presumption that the shackles are seen by the jury and prejudice results. In contrast, if a defendant is restrained in a manner that is not visible to the jury, prejudice will not be presumed.”

Id. at 425-26 (internal quotation marks omitted).

That reasoning reflects the Court of Appeals’ understanding of the case law that existed at the time of petitioner's trial, and it suggests that prejudice to a defendant occurs when jurors become aware that the defendant is wearing restraints, which could occur inside the courtroom, in the courthouse hallway, or in some other place. And even though the court in Sproule was not addressing an out-of-court encounter, it is not easy to square its focus on the jury's awareness that a defendant is shackled with a limitation on where the jury reaches that awareness. More significantly for present purposes, we cannot conclude that a trial court, confronted with the same body of case law that the Court of Appeals relied on in Sproule, would necessarily choose to focus on something different than the Court of Appeals focused on there. Once again, as stated in Kessler, “[i]t is not a question whether the jury will be constantly exposed to the vision of a shackled defendant, but whether they see the shackles at all.” 57 Or at 474.

It follows that, if counsel had moved for a mistrial, the trial court may well have understood that Oregon law entitled petitioner to a new trial because the jury encounter had “prejudice[d] him in the minds of the jurors” and “impinge[d] on the presumption of innocence.” Id. at 472-74 (discussing Duran); see also Long, 195 Or at 91 (explaining that one rationale for the common law rule prohibiting the use of shackles at trial was that “[a] prejudice might be created in the minds of the jury against a prisoner who should be brought before them handcuffed and shackled, which might interfere with a fair and just decision of the question of the guilt or innocence of such prisoner”). And that conclusion would be especially understandable in petitioner's case, where the risk arising from seeing petitioner in shackles—namely, that jurors might perceive him to be a dangerous threat—was closely related to a disputed issue at trial—whether petitioner had forcibly subjected the victim to sexual contact.²⁵ See Lichau, 333 Or at 365 (finding prejudice when omitted evidence could have tended to affect the jury's consideration of key credibility issues); cf. Rhoden v. Rowland, 172 F3d 633, 637 (9th Cir 1999) (where at least some jurors saw the petitioner in shackles during trial for violent crimes and disputed issue at trial was whether petitioner had used force or fear to overcome the victim's will, the shackles adversely affected the jury's verdict by “essentially brand[ing]” the petitioner “as having a violent nature in a case where his propensity for violence was the crucial issue”; further, jury's lengthy deliberations “suggest[ed] that they did not find the case to be clear-cut”).

Arguing against the conclusion that petitioner has established prejudice here, the state stands by its argument that, unlike in-court shackling, out-of-court shackling does not impinge upon the presumption of innocence or warrant a mistrial. Like the post-conviction court, the state points to case law from outside jurisdictions to argue that jurors would know that “any shackling during transport is merely a precaution that the sheriff ha[s] taken” and “impl[ies] only that the defendant is being held in custody during trial” and, thus, petitioner would not have been entitled to a mistrial. For two reasons, we are not persuaded.

The first difficulty we have with the state's argument is that, like part of its argument related to counsel's performance, it is somewhat misguided. As noted above, the issue in this case is not whether the trial court would have been obligated under the law to grant a mistrial if defense counsel had requested one; the issue is whether there is more than a mere possibility that the trial court would have done so under the existing circumstances. For the reasons just stated, we conclude that there is such a possibility and that petitioner has therefore established that counsel's omission had a tendency to affect the outcome.

Our second and related difficulty is that we cannot conclude, based on the out-of-state case law that the state and post-conviction court have pointed to, that the trial court would necessarily have made the distinction that those cases call for. Most of the cases that the post-conviction court relied on assert, with limited discussion, that witnessing a defendant in shackles outside the courtroom is somehow different and unremarkable. See, e.g., State v. Cassel, 48 Wis 2d 619, 625, 180 NW 2d 607, 611 (1970) (stating that “the situation is psychologically different and less likely to create prejudice in the minds of jurors”; further, “[p]eople normally expect to see a prisoner under some restraints in situations where he is not able to escape if not in restraints”); State v. Elliott, 137 NC App 282, 285, 528 SE 2d 32, 35, rev'd on other grounds, 352 NC 663, 535 SE2d 32 (2000) (“The handcuffing of defendants, as they are transferred between the courtroom and the jail, is a common practice well known by the general public.”). Because the reasoning of those cases is somewhat cursory and not particularly persuasive, the trial court may not have afforded them the weight that the post-conviction court appears to have given them.

The cases that the state relies on suffer similar deficiencies. Some of them summarily characterize the out-of-courtroom encounters as “brief and accidental,” e.g., Wharton v. Chappell, 765 F3d 953, 964 (9th Cir 2014), while others, like those cited by the post-conviction court, reason that jurors are aware that some defendants are unable to afford bail, e.g., United States v. Halliburton, 870 F2d 557, 561 (9th Cir 1989), or that “[i]t is a normal and regular as well as a highly desirable and necessary practice to handcuff prisoners while they are being taken from one place to another, and the jury is aware of this[,]” United States v. Leach, 429 F2d 956, 962 (8th Cir 1970). Although some of those decisions acknowledge that out-of-court encounters may affect jurors’ perceptions of the presumption of innocence, they summarily dismiss that concern as being adequately mitigated by the assumed fact jurors know that some defendants are in custody.²⁶

This court has never discussed the jury's likely understanding of pretrial detention practices, such as shackling detainees during transport, nor has it suggested that jurors would be indifferent to witnessing it. Cf. State v. Cavan, 337 Or 433, 447, 98 P3d 381 (2004) (“Our 200-year American jury trial tradition informs us that” jurors are not “indifferent to their surroundings.”). Even if we were to assume that jurors are aware that some defendants are detained pretrial simply because they cannot post bail, we are reluctant to further assume that their awareness of that fact can eliminate the negative consequences of seeing a defendant in shackles, including the creation of unconscious bias. For present purposes, however, we need not—and do not—conclusively decide whether there is a material difference between in-court and out-of-court viewings for purposes of petitioner's constitutional and common law rights. It suffices, instead, to conclude that, although it is possible that the trial court would have reached the same conclusion as those out-of-jurisdiction decisions and therefore denied any request for a mistrial, their persuasive value is not so great that we can say that there is only the mere possibility that the trial court would have granted a mistrial if counsel had requested one. Accordingly, petitioner was prejudiced by counsel's failure to do so.

Finally, although we decide petitioner's case on the basis of counsel's failure to move for a mistrial and the possibility that the trial court would have declared a mistrial if one had been requested, we briefly address the dissent's observation that the trial court might instead have taken corrective action short of declaring a mistrial. ––– Or at –––– (Bushong, J., dissenting) (slip op at 33:4 - 34:10). It is true that, given this court's opinion in Walton, the other case law from Oregon and other jurisdictions, and the discretion afforded trial courts with regard to mistrial motions, the trial court in this case might have chosen to do something other than declare a mistrial to address what had occurred. See, e.g., State v. Evans, 211 Or App 162, 166, 154 P3d 166 (2007), aff'd, 344 Or 358, 182 P3d 175 (2008) (stating that, “[i] n ruling on a motion for a mistrial, a trial court must decide whether to grant the motion, to cure the effect of inappropriate conduct or testimony by giving a proper instruction instead, or to do nothing at all”); see also Walton, 311 Or at 250 (listing corrective measures taken by trial court to address jurors’ observation of the defendant in shackles); United States v. Moreno, 933 F2d 362, 368 (6th Cir 1991) (observing that courts have expressed a preference for remedial action after an accidental observation of a defendant in custody); id. (approving instruction to jury “that the custodial status of the defendants was not indicative of guilt or innocence and, therefore, should be disregarded”); United States v. Pina, 844 F2d 1, 8 (1st Cir 1988) (approving actions taken by the trial court to ensure jury impartiality, including individual questioning of potentially tainted jurors, obtaining confirmation from jurors that they remained unbiased, and instructing jurors not to discuss the matter with anyone). But the possibility that the trial court would take different action to address its concerns that the jury had been tainted does not transform the possibility of a mistrial into a “mere” possibility. And because any such action would reflect the trial court's view that measures were necessary to ensure an impartial jury, the absence of such measures, like the failure to seek a mistrial, would, in our view, have a “tendency to affect the outcome” of the jury trial. Green, 357 Or at 322.

III. CONCLUSION

We conclude that, although no Oregon appellate court had previously held that a mistrial is warranted when jurors briefly see a defendant in shackles outside the courtroom, the state of the law at the time of petitioner's trial was such that an attorney exercising reasonable professional skill and judgment would have seen an opportunity to argue for a mistrial under the circumstances of this case. Moreover, even if counsel might have had reasonable tactical or other reasons not to seek a mistrial, the post-conviction court may not rely on any such reasons when the record cannot support the finding that counsel in fact considered them.

We further conclude that counsel's performance was prejudicial. Had counsel moved for a mistrial, there is more than a mere possibility that the trial court would have granted that request or taken other protective measures to address petitioner's harms. Thus, petitioner has shown that counsel's inadequate performance had a tendency to affect the outcome and that petitioner therefore suffered prejudice. Accordingly, the post-conviction court erred in denying petitioner's claim of inadequate assistance of counsel, and we reverse the post-conviction court's judgment on that basis.

The decision of the Court of Appeals is affirmed in part and reversed in part. The judgment of the circuit court is reversed, and the case is remanded to the circuit court for further proceedings.

Today, this court grants petitioner post-conviction relief, concluding that his criminal defense lawyer rendered constitutionally inadequate assistance during his criminal trial. It is well established that, in evaluating whether counsel's assistance was inadequate in violation of Article I, section 11, of the Oregon Constitution, we first must determine “ ‘whether petitioner demonstrated by a preponderance of the evidence that his lawyer failed to exercise reasonable professional skill and judgment.’ ” Montez v. Czerniak, 355 Or 1, 7, 322 P3d 487, adh'd to as modified on recons., 355 Or 598, 330 P3d 595 (2014) (quoting Lichau v. Baldwin, 333 Or 350, 359, 39 P3d 851 (2002) (brackets omitted)). If we conclude that petitioner met that burden, “we further must determine whether he proved that counsel's failure had a tendency to affect the result of his trial,” id.,—or in other words, whether counsel's failure “prejudiced” petitioner, Lichau, 333 Or at 363.¹

The majority opinion holds that petitioner met his burden in both respects. First, the majority opinion concludes that petitioner's lawyer failed to exercise reasonable skill and judgment when he did not move for a mistrial after petitioner told him that some jurors had seen him in the hallway shackled and dressed in jail clothes. Second, the majority opinion concludes that counsel's failure was prejudicial, because, if counsel had requested a mistrial, “there is more than a mere possibility that the trial court would have declared a mistrial, thereby affecting the outcome of his case.” ––– Or at –––– (slip op at 41:8-10). I disagree with both conclusions and the majority opinion's analysis. Accordingly, I respectfully dissent.

I. FAILURE TO EXERCISE REASONABLE SKILL AND JUDGMENT

As noted, it was petitioner's burden to establish, by a preponderance of the evidence, that his trial counsel failed to exercise reasonable skill and judgment when he did not move for a mistrial. Montez, 355 Or at 7. The post-conviction court concluded that petitioner failed to meet his burden. In determining whether the post-conviction court erred, this court must “ ‘make every effort to evaluate a lawyer's conduct from the lawyer's perspective at the time, without the distorting effects of hindsight.’ ” Id. (quoting Lichau, 333 Or at 360). This court must not “ ‘second-guess a lawyer's tactical decisions in the name of the constitution unless those decisions reflect an absence or suspension of professional skill and judgment.’ ” Id. (quoting Gorham v. Thompson, 332 Or 560, 567, 34 P3d 161 (2001)). That is because the constitution “ ‘gives no defendant the right to a perfect defense * * *.’ ” Id. (quoting Krummacher v. Gierloff, 290 Or 867, 875, 627 P2d 458 (1981)).

As I will explain, because petitioner has the burden of proof, this court on review must analyze whether petitioner established facts showing that moving for a mistrial was so obviously beneficial to him that it was the only acceptable course of action; whether petitioner cited any case decided before this incident that had clearly held that jurors seeing a criminal defendant in shackles outside the courtroom required a mistrial; and, if not, whether petitioner established that the law so obviously foreshadowed such a ruling that all defense attorneys exercising reasonable professional skill and judgment would have recognized at that time that the circumstances required a request for a mistrial. I agree with the post-conviction court and the Court of Appeals that petitioner did not meet his burden. In concluding otherwise, the majority opinion does not, in my view, fully engage with the required analysis of the facts and law consistent with petitioner's burden.

I begin with the majority opinion's treatment of the facts. Petitioner's criminal trial occurred in 2013; his post-conviction trial was in 2020. Given that seven-year gap, the factual record related to jurors’ potential exposure to petitioner in shackles and jail clothes at petitioner's criminal trial is understandably sparse.

In his testimony at the post-conviction trial, petitioner described the incident, recalling that it happened on a Friday, after the jury had finished its first day of deliberations without reaching a verdict. Petitioner testified that, when the deputies began to escort him out of a holding cell into a courthouse hallway to take him to jail, the deputies saw the jurors as they were going down the stairs to leave the courthouse. They then “opened the door and snatched [petitioner] back into the holding cell for about ten minutes.” Petitioner described where he was when this happened. He stated that there were “some real big steps that lead up to each floor of the courtrooms” and that the holding cells were “all the way at the end of the hall.” He remembered seeing “probably three” jurors who “actually looked at” him, and that “the other two” jurors that he had seen were “still walking down the steps.” Petitioner testified that he told his defense counsel that jurors had seen him in shackles and jail clothes, that counsel then informed the court and the prosecutor, but “that's all that happened.” He testified, that, at the time, he “didn't think the jurors seeing [him] in handcuffs was all that serious” and that he was more focused on the possibility of a “hung jury.”² Petitioner also testified that, when the jury came back into the courtroom to return a verdict, he asked his lawyer about a mistrial, and counsel responded that “there's no reason” to request one.

Petitioner's trial counsel—an experienced criminal defense lawyer ³ —submitted a declaration stating that he had “searched [his] memory” and had “no doubt” that petitioner had told him that jurors had seen him in shackles and jail clothes as he was being transported to or from court. Counsel stated that he was “certain” that he made inquiries of the sheriff's deputies and then brought the matter to the court's attention, but he did not offer any details of what was discussed or why he did not move for a mistrial. Counsel was called as a witness to testify during the post-conviction trial, but he was not asked any questions about this incident or to explain why he did not move for a mistrial.

The post-conviction court found that counsel's testimony was credible, and that petitioner's testimony was also credible “except to the extent that he testified that his trial attorney did nothing” after he reported the “possible contact with jurors while restrained in the courthouse hallway.” Rather, the post-conviction court found that, after petitioner reported that “possible contact” to his attorney, counsel discussed it with the deputies and then the trial judge and the prosecuting attorney in chambers, off the record, but did not move for a mistrial.

The post-conviction court concluded that petitioner “did not prove” that his trial attorney failed to exercise reasonable professional skill and judgment by not requesting a mistrial. In support of that conclusion, the post-conviction court stated that trial counsel “did investigate the matter” and discussed it with the judge and prosecutor but took no action on the record. The post-conviction court also noted the absence of evidence in the record in several respects. Specifically, it pointed out that petitioner had “offered only his perception of the hallway encounter” and that there was “no evidence from the viewpoint of the jurors.” The court also noted that there was “no evidence that the jurors recognized [petitioner] or that [the encounter] had any impact on them.” And although there was no evidence in the record about why counsel chose not to move for a mistrial, the post-conviction court observed that counsel “may not have wanted a mistrial or may have concluded that a mistrial [motion] would not have been successful.”⁴

As the majority opinion points out, the post-conviction court's findings of historical fact are binding on this court if there is evidence to support them. ––– Or at –––– (citing Lichau, 333 Or at 359) (slip op at 2:10-12). If the post-conviction court has not expressly made findings and there is evidence from which the facts could be decided in more than one way, we must presume that the court decided the facts in a manner that is consistent with its conclusions of law. ––– Or at –––– (citing Lichau, 333 Or at 359; Ball v. Gladden, 250 Or 485, 487, 443 P2d 621 (1968)) (slip op at 2:12-16).

The majority opinion cites those standards and acknowledges that petitioner has the burden of proof, but, as I will explain, its analysis nevertheless gives petitioner the benefit of the absence of evidence in some areas. The majority opinion also fails to account for plausible inferences that could be made from the evidence that would support the post-conviction court's conclusions.

For example, as noted, there was no evidence of what the jurors saw from their viewpoint; whether any jurors had recognized the person being escorted in shackles as the defendant in the case they were deliberating; what trial counsel discussed off the record with the deputies, the trial court, and the prosecuting attorney; and what trial counsel concluded from his investigation. There was also no evidence in the record about why counsel did not move for a mistrial. That absence of evidence cannot weigh in petitioner's favor, because he has the burden of proof.

The post-conviction court found that counsel “did investigate” this incident. If the three jurors who “looked at” petitioner saw him, it is plausible to infer that they may not have recognized the person in shackles wearing jail clothes emerging from a doorway at the end of the hall escorted by deputies as the defendant in their case. The last time the jurors had seen the petitioner was earlier that day, when they were sent to deliberate. At that time, he was in the courtroom dressed in street clothes. When the jurors were heading down the stairs after concluding their deliberations at the end of the day, it is certainly plausible that, if they did see a person escorted by deputies emerging from a doorway at the end of the hall in shackles wearing jail clothes, they may not have recognized him as the same person they had seen in street clothes in the courtroom earlier that day.⁵ That would explain why counsel told petitioner that “there's no reason” to request a mistrial. If trial counsel had concluded from his investigation that jurors had not recognized the person being transported, then his decision to not request a mistrial was not constitutionally inadequate.

The majority opinion declines to “account for the possibility” that defense counsel did not move for a mistrial because he had concluded that the jurors had not recognized petitioner in the hallway. ––– Or at –––– (slip op at 32:7-9). The majority opinion ignores that possibility, because, in its view, there was no “disputed factual matter” in the post-conviction court record as to counsel's perception and what he concluded after completing his investigation. ––– Or at –––– (slip op at 31:2-9). But there was a factual dispute about what counsel learned and concluded from his investigation, because that undoubtedly affected his determination that there was “no reason” to request a mistrial. It is true that there is no direct evidence of what counsel had learned and concluded from his investigation—not surprisingly, given the time between those events and the post-conviction trial— but the factual dispute is reflected in counsel's statement to petitioner that he thought at the time that there was “no reason” to request a mistrial. The majority opinion's conclusion that the record “contains no evidence either way as to what counsel concluded” after his investigation, ––– Or at –––– (emphasis in majority opinion) (slip op at 31:14-15), overlooks the inference that can be made from counsel's statement.⁶

The majority opinion also states that the applicable standard of review does not permit it “to endorse that level of speculation,” because the post-conviction court's ultimate legal conclusion “does not appear to have relied on what counsel did or did not conclude about the jurors’ actual perceptions[.]” ––– Or at –––– (slip op at 31:22 - 32:1-6). But, as noted, the standard of review requires this court to presume that the post-conviction court would have found the facts in a way that supports the post-conviction court's conclusion, and it requires this court to make that presumption in the absence of any specific factual finding or reliance by the post-conviction court. See Lichau, 333 Or at 359 (“[I]f findings are not made on all [historic] facts, and there is evidence from which such facts could be decided more than one way, we will presume that the facts were decided in a manner consistent with the post-conviction court's ultimate conclusion.” (Emphasis added; internal quotation marks and alterations omitted)).

As discussed above, although there is no direct evidence of what counsel learned and concluded from his investigation, there is evidence from which the facts could be decided in more than one way. Instead of presuming that the post-conviction court would have decided the facts in a manner consistent with its conclusion, the majority opinion declines to “account for” that possibility, ––– Or at –––– (slip op at 32:7-9) choosing instead to view the facts in a manner that is consistent with the majority opinion's own conclusions. That, in my view, does not correctly apply the standard of review and the burden of proof.

As for the post-conviction court's suggestion that counsel “may not have wanted” a mistrial—and the state's argument on review that counsel may have had “strategic reasons” for choosing not to move for a mistrial—the majority opinion states that the argument “falls short” because “nothing in the record suggests that counsel considered moving for a mistrial” and “then made a strategic decision” not to do so. ––– Or at –––– (slip op at 29:5-8). Again, the record reflects that counsel told petitioner that there was “no reason” to request a mistrial, and the absence of additional evidence in the record cannot favor petitioner, because he has the burden of proof.

The majority opinion also concludes that requesting a mistrial was so “obviously beneficial” to petitioner that there would be no reason to not request one, as if there were no potential downsides to requesting a mistrial. ––– Or at –––– (slip op 29:13-17). However, the benefits to a criminal defendant of moving for a mistrial are not so obvious that reasonable defense counsel must always request a mistrial if there are any arguable grounds for doing so, as this court has recognized. In State v. Moore, 361 Or 205, 390 P3d 1010 (2017), we issued a peremptory writ of mandamus when a trial court sua sponte declared a mistrial over a criminal defendant's objection. The defense counsel in that case told the court: “ ‘We don't want a mistrial. We like this jury; we like how we're doing.’ ” Id. at 220. We concluded that “[t]he fact that [the] defendant and his counsel believed that the jury was favorably disposed to his case and objected to a mistrial is an important consideration” in determining whether to declare a mistrial. Id.

Thus, depending on the case, one potential downside to moving for a mistrial is the possibility that the court may grant the request, leading to a retrial that the defendant may not want. As in Moore, a criminal defendant and his attorney may not want a mistrial and a retrial if they “like” the jury or “liked” how the trial had gone. Although there is no direct evidence in this case that petitioner or his defense counsel “liked” the jury or how the trial had gone, there is some evidence that would support that inference. During the first day of deliberations, the jury asked the court what it should do if there were not ten votes on a particular charge, noting that the verdict form did not have a way to express a hung jury.⁷ Under those circumstances, it is certainly plausible that a reasonable defense attorney might not want a mistrial, as the post-conviction court noted.⁸

Another possible downside to requesting a mistrial is that, if the court declared a mistrial, the victim would have to testify again at the resulting retrial. As the Court of Appeals noted, petitioner “was twice tried for charges arising out of a domestic violence incident in which he punched his domestic partner in the face, had sex with her against her will, and took her vehicle.” Nelson v. Cain, 334 Or App 326, 327, 556 P3d 1 (2024). Testifying about a sexual assault can be traumatizing. See People v. Cogswell, 48 Cal 4th 467, 478, 227 P3d 409, 414, 106 Cal Rptr 3d 850, 856 (2010) (noting that sexual assault victims “are particularly likely to be traumatized because of the nature of the offense” and requiring victims to “relive and to recount in a public courtroom the often personally embarrassing intimate details of a sexual assault far overshadows the usual discomforts of giving testimony as a witness”); Vivian Berger, Man's Trial, Woman's Tribulation: Rape Cases in the Courtroom, 77 Colum L Rev. 1, 23 (1977) (discussing the possibility that a rape victim's “ordeal during the trial unjustly compounds the trauma of the event and its aftermath”). Although there is no evidence in this case about how traumatizing it was to the victim—petitioner's former domestic partner—to testify, twice, about this incident, it is possible that, at the time, petitioner might not have wanted to traumatize her again by forcing her to relive the incident for a third time on retrial, especially if he liked how his second trial had gone.

Turning to the majority opinion's treatment of the law, it acknowledges that “no Oregon appellate court had previously held that a mistrial is warranted when jurors briefly see a defendant in shackles outside the courtroom,” ––– Or at –––– (slip op at 50:22 - 51:1-2), but it concludes that the state of the law was such that that an attorney exercising reasonable skill and judgment “would have seen an opportunity to argue for a mistrial” under the circumstances of this case, ––– Or at –––– (slip op at 51:2-4). The majority opinion relies on Burdge v. Palmateer, 338 Or 490, 112 P3d 320 (2005), for the proposition that a constitutionally adequate defense counsel must make an “obviously beneficial” argument even if the law is unsettled.⁹

Although I agree with that proposition as a general principle, I do not agree that it should be applied in this case. In Burdge, this court concluded that the defense attorney's assistance in the petitioner's trial was not constitutionally inadequate. 338 Or at 500.¹⁰ In fact, this court has never—before today—applied Burdge to hold that a defense lawyer's failure to advance a plausible argument on an unsettled legal issue amounts to constitutionally inadequate assistance of counsel. See Jackson v. Franke, 369 Or 422, 433-34, 507 P3d 222 (2022) (noting that, “[t]o date, * * * this court has not been persuaded by a claim * * * that any lawyer exercising reasonable skill and judgment would have raised a particular unsettled legal issue”).¹¹

The majority opinion states that this court applied Burdge in the context of unsettled case law in Antoine v. Taylor, 368 Or 760, 499 P3d 48 (2021), ––– Or at –––– (slip op at 27:2-4). Although we cited and applied Burdge in Antoine, we again held in that case that the petitioner's criminal defense lawyer's assistance was not constitutionally inadequate. 368 Or at 780. Moreover, as I will explain, our analysis in Antoine demonstrates how this court should analyze the issue considering the petitioner's burden of proof. Accordingly, I discuss Antoine in some detail.

The petitioner in Antoine had been charged with multiple counts of first-degree sexual abuse and first-degree sodomy. Id. at 762-63. His defense counsel filed a demurrer, arguing that the indictment failed to provide adequate notice of the basis of the charges. Id. at 763. The trial court denied the demurrer and the petitioner was convicted. Id. at 764-65. On direct appeal, the Court of Appeals held that, under State v. Hale, 335 Or 612, 621, 75 P3d 448 (2003), cert. den., 541 US 942 (2004), the petitioner's inadequate notice claim should have been raised in a motion for a pretrial election, not in a demurrer. State v. Antoine, 269 Or App 66, 78, 344 P3d 69, rev. den., 357 Or 324 (2015). The petitioner later sought post-conviction relief, alleging that his trial counsel provided inadequate assistance by failing to move for a pretrial election. The post-conviction court granted relief, but the Court of Appeals reversed, holding that trial counsel had not performed deficiently given the state of the law at the time of the petitioner's trial. Antoine v. Taylor, 303 Or App 485, 499, 465 P3d 238 (2020). We allowed review and affirmed the Court of Appeals. Antoine, 368 Or at 780.

We stated that our decision in Burdge was “the most pertinent” to the issue presented on review, id. at 768, noting that Burdge’s “analytical approach remains applicable” even though Antoine did not involve a pure issue of statutory interpretation, id. at 770. “[T]o address whether [the] petitioner's trial counsel performed deficiently,” we explained, the court must “examine the state of the law at the time of petitioner's trial to evaluate whether trial counsel's decision to litigate the notice issue primarily through a demurrer was unreasonable.” Id. at 771.

In analyzing that issue, we noted that, under the case law decided before our decision in Hale, a demurrer had been considered an appropriate procedural way to challenge the sufficiency of notice. Id. at 772. The petitioner in Antoine argued, however, that his trial counsel was constitutionally inadequate, because Hale had subsequently “foreclosed one of the two possible ways to obtain pretrial notice—a demurrer.” Id. at 775. We rejected that argument, concluding that “petitioner's reading of Hale is not so obvious ‘that any lawyer exercising reasonable professional skill and judgment necessarily would have seen it.’ ” Id. at 777 (quoting Burdge, 338 Or at 497-98). In reaching that conclusion, we also noted “what was absent from Hale.” Id. at 778. Specifically, we noted that Hale did not address earlier cases or otherwise explain why, “under the statutory scheme concerning demurrers,” a defendant could not establish that an indictment failed to provide sufficient notice through a demurrer. Id. at 779. We also noted that the Court of Appeals had considered pretrial notice issues raised through a demurrer in “at least three decisions issued after Hale” and before the petitioner's underlying criminal trial. Id. (emphasis in original).

In addition, we observed that the petitioner “has identified no decision available to trial counsel that clearly held that a demurrer was not an appropriate mechanism” to challenge the sufficiency of pretrial notice, and that the petitioner “has not presented any evidence that his reading of Hale was widely shared or even that any other attorney had read Hale in that manner[.]” Id. at 780. We also noted that the petitioner's “theory of deficient performance fails to account for the longstanding body of authority holding that a demurrer is an appropriate way to raise a challenge” to the adequacy of pretrial notice. Id. Thus, we determined in Antoine, “as in Burdge,” that, “when we take into account the information available to trial counsel at the time,” we could not conclude “that any lawyer exercising reasonable professional skill and judgment necessarily would have seen the arguable ambiguity in the case law subsequently identified by the Court of Appeals” on direct appeal in the petitioner's case. Id. (citation and internal quotation marks omitted).

Notably, in Antoine, the absence of evidence in the record showing what other lawyers knew or understood at the time—and the petitioner's failure to identify any case that “clearly held” that his lawyer's decision to challenge the adequacy of notice through a demurrer and not a pretrial motion to elect—weighed against the petitioner, because he had the burden of proof. Id.

Where—as here—there was no case that “clearly held” that counsel must take an action that a post-conviction petitioner contends was required for a constitutionally adequate defense, our cases have examined whether the existing state of the law “obviously foreshadowed” the legal rule that the post-conviction petitioner is advocating. The majority opinion does not engage with that required analysis.

For example, the petitioner in Jackson, alleged that his defense counsel was constitutionally inadequate at the time of his 2001 trial, because he did not object to an expert witness's diagnosis of child sexual abuse in the absence of corroborating evidence. 369 Or at 424. This court later held in State v. Southard, 347 Or 127, 142, 218 P3d 104 (2009), that such expert testimony is inadmissible. We stated that the issue in post-conviction in Jackson was whether the law at the time of the petitioner's criminal trial “so obviously foreshadowed” the rule subsequently adopted in Southard that any lawyer exercising reasonable professional skill and judgment would have objected to the diagnosis of child sexual abuse in the absence of physical evidence. Jackson, 369 Or at 435-36. We concluded that “our ultimate answer in Southard was not so obviously foreseeable that an examination of the existing case law” established that all counsel exercising reasonable professional skill and judgment would have objected to the testimony. Id. at 438.¹²

Thus, our decisions in Antoine and Jackson analyzed the issue in the way that the majority opinion should have analyzed it in this case. Because petitioner has the burden of proof, the court should analyze whether petitioner presented facts establishing that the only acceptable course of action was to move for a mistrial; whether petitioner cited a pre-2013 case that “clearly held” that jurors seeing a criminal defendant in shackles outside the courtroom required a mistrial; and, if not, whether petitioner established that the existing law “so obviously foreshadowed” such a ruling that all defense attorneys exercising reasonable professional skill and judgment would request a mistrial upon being told that that had happened. The majority opinion falls short of that required analysis.

As explained above, there are many important facts that are missing from this record and there are plausible inferences that can be made from the facts that were presented that would make trial counsel's decision to not move for a mistrial reasonable. And, as I will explain next, petitioner did not establish that existing law so obviously foreshadowed the application of cases addressing jurors observing a defendant shackled in the courtroom to jurors seeing a defendant in shackles outside the courtroom. Instead, petitioner's theory of inadequate assistance, like the petitioner's theory in Antoine, “fails to account” for a “longstanding body of authority” holding that jurors briefly observing a criminal defendant in shackles outside the courtroom does not necessarily provide grounds for a mistrial. Antoine, 368 Or at 780. This court has never held—until today's ruling—that jurors seeing a criminal defendant in shackles outside the courtroom requires a mistrial, and, as I will explain, the state of the law in 2013 did not “so obviously foreshadow” such a ruling that all reasonable counsel would have requested one.

The majority opinion states that this court “has never limited” a defendant's right to avoid having jurors see him shackles “to in-court observations.” ––– Or at –––– (emphasis in majority opinion) (slip op at 27:8-11). That is true, but petitioner has the burden of establishing that existing law so obviously foreshadowed the application of in-court shackling principles to an out-of-court observation; the state does not have to establish that existing law was limited to in-court observations. The majority opinion suggests that petitioner met his burden, because “at least some of the concerns” arising from in-court shackling “could be reasonably understood to arise in the out-of-court context as well[.]” ––– Or at –––– (citing Guinn v. Cupp, 304 Or 488, 497, 747 P2d 984 (1987); State v. Kessler, 57 Or App 469, 473, 645 P2d 1070 (1982)) (slip op at 27:11-13).¹³ From those observations, the majority opinion concludes that a defense lawyer exercising reasonable skill and judgment would have known that jurors seeing a defendant in shackles outside the courtroom required a request for a mistrial.

In reaching that conclusion, the majority opinion states that the only Oregon case to address jurors seeing a defendant in shackles outside the court room—State v.Walton, 311 Or 223, 809 P2d 81 (1991)—“in no way foreclosed” an argument that “the circumstances in this case warranted a mistrial.” ––– Or –––– (slip op at 35:6-8). The question, though, is not whether the Walton decision foreclosed an argument that the circumstances warranted a mistrial. Rather, the question here is “whether petitioner demonstrated by a preponderance of the evidence that [his lawyer] failed to exercise reasonable professional skill and judgment[,]” Montez, 355 Or at 7, when he did not move for a mistrial. As in Antoine, it was petitioner's burden to establish that, under the state of the law in 2013, all reasonable counsel would have requested a mistrial; it was not the state's burden to establish that a mistrial was “foreclosed” by the law.

Walton thus provides relevant guidance on the issue as we framed it in Jackson—whether the law that existed in 2013 “so obviously foreshadowed” today's ruling that reasonable criminal defense lawyers would have known then that they must request a mistrial upon learning that jurors had seen the defendant in shackles outside the courtroom. As I will explain, the result in Walton and other cases involving jurors viewing a criminal defendant in shackles outside the courtroom did not “obviously foreshadow” the rule of law that the majority opinion announces today.

The defendant in Walton contended on direct appeal that “the trial court erred in denying his motion for a mistrial after a juror and an alternate juror observed him in handcuffs with a group of other prisoners.” 311 Or at 250.¹⁴ After that incident, the trial court questioned the two jurors individually outside the presence of other jurors. Id. Both jurors told the court that, “although they had seen [the] defendant in handcuffs, they had not discussed their observations with any other juror and that their observations would not affect their impartiality.” Id. The trial court then “instructed both jurors that they were to disregard their observations, that they were not to discuss their observations with anyone, and that [the] defendant's custody status had no bearing on his guilt or innocence.” Id. The alternate juror “did not participate further in [the] defendant's trial[,]” and the trial court “offered to replace the juror” who had seen the defendant in handcuffs with “an ‘untainted’ alternate.” Id. The defendant rejected that offer, and he was ultimately convicted. Id.

On appeal, the defendant contended, as noted, that the trial court had erred in denying his motion for a mistrial. Id. This court “f[ou]nd no reversible error.” Id. We explained that the trial court had given a curative instruction, that jurors “are presumed to follow a trial court's instructions[,]” that the defendant “ha[d] not shown that the juror who served in his case [who had seen him in handcuffs] did not follow the court's instructions[,]” and that the defendant had rejected the court's offer to replace the juror who had seen him in handcuffs with an “untainted” juror. Id.

Accordingly, the only Oregon case addressing jurors observing a defendant in shackles outside the courtroom does not “obviously foreshadow” that such an observation would by itself be grounds for a mistrial. Cases from other jurisdictions prior to 2013 confirm that that was not the governing rule. Those cases generally make a distinction between jurors seeing a defendant in shackles in the courtroom, and jurors inadvertently seeing a defendant in restraints while being transported to or from the courtroom. See, e.g., U.S. v. Jones, 468 F3d 704, 709 (10th Cir 2006), cert. den., 565 US 1252 (2012); Williams v. Woodford, 384 F3d 567, 592-93 (9th Cir 2004), cert. den., 546 US 934 (2005); Schiro v. Clark, 963 F2d 962, 976 (7th Cir 1992), aff'd on other grounds sub nom Schiro v. Farley, 510 US 222, 114 S Ct 783, 127 L Ed. 2d 47 (1994); U.S. v. Moreno, 933 F2d 362, 368 (6th Cir), cert. den., 502 US 895 (1991); U.S. v. Pina, 844 F2d 1, 8 (1st Cir 1988); Clark v. Wood, 823 F2d 1241, 1245 (8th Cir), cert. den., 484 US 945 (1987); United States v. Diecidue, 603 F2d 535, 549-50 (5th Cir 1979), cert. den., 445 US 946 (1980); United States v. Taylor, 562 F2d 1345, 1359 (2d Cir), cert. den., 432 US 909 (1977); see also Sheldon R. Shapiro, Annotation, Propriety and Prejudicial Effect of Gagging, Shackling, or Otherwise Physically Restraining Accused During Course of State Criminal Trial, 90 ALR 3d 17 § 23[a] (1979) (collecting state cases standing for that principle).

For example, the Indiana Supreme Court concluded in one case, that, “[b]eing seen by jurors while being transported in handcuffs is not grounds for reversal, absent a showing of actual harm.” Sweet v. State, 498 NE2d 924, 929 (Ind 1986). That court concluded in another case that a trial court did not abuse its discretion when it denied a motion for a mistrial after members of the jury “viewed [the defendant] handcuffed to a police officer in the hallway” during a recess. Jenkins v. State, 492 NE2d 666, 669 (Ind 1986). The court explained that “reasonable jurors could expect [the defendant] to be in police custody while in the hallway of the courthouse.” Id.

Similarly, the Arizona Supreme Court acknowledged that a defendant in a criminal case “has a right to appear in civilian clothing and be free from visible restraints in the courtroom during trial.” State v. Hardy, 230 Ariz 281, 292, 283 P3d 12, 23 (2012). But the court concluded that “[a] juror's inadvertent exposure to the defendant in handcuffs outside the courtroom * * * is not inherently prejudicial, and a defendant is not entitled to a new trial absent a showing of actual prejudice.” Id., 283 P3d at 23 (internal quotation marks omitted). That reasoning “also applies to a juror's inadvertent exposure to the defendant in jail garb.” Id. Thus, the court held that “[t]he trial court did not abuse its discretion in denying the [defendant's] motion for mistrial.” Id.

In federal court, “[e]xposure of the jury to a defendant in shackles requires a mistrial only when the exposure is so ‘inherently prejudicial’ as to deny the defendant's constitutional right to a fair trial.” Moreno, 933 F2d at 368 (quoting Pina, 844 F2d at 8)). The Sixth Circuit has “distinguished the inherent prejudice to a defendant who is shackled while in the courtroom from a defendant who has been observed in shackles for a brief period elsewhere in the courthouse.” Id. The Ninth Circuit has held that “a jury's brief or inadvertent glimpse of a defendant in physical restraints outside of the courtroom does not warrant habeas relief unless the petitioner makes an affirmative showing of prejudice.” Williams, 384 F3d at 593 (collecting Ninth Circuit cases). The Eighth Circuit has explained that “it is a normal and regular as well as highly desirable and necessary practice to handcuff prisoners when they are being taken from one place to another, and the jury is aware of this.” Clark, 823 F2d at 1245 (emphasis added; brackets and internal quotation marks omitted).

The majority opinion cites no case holding otherwise in these or any analogous circumstances. Instead, it is “not persuaded” by those cases, ––– Or at –––– (slip op at 35:9-11), concluding that that “nonbinding case law discussing the prejudice standard for out-of-courtroom encounters fails to provide even persuasive authority that counsel's decision not to seek relief at all constituted adequate performance in this case[,]” ––– Or at –––– (slip op at 36:4-7). Again, that frames the wrong question given the burden of proof and the standard of review. The issue is not whether there was “persuasive authority” supporting the decision to not seek a mistrial; the question is whether the post-conviction court erred in concluding that petitioner did not meet his burden of establishing that, given the state of the law in 2013, all reasonable defense lawyers would have moved for a mistrial based on petitioner's report that jurors had seen him in shackles outside the courtroom.

The state of the law in 2013 regarding the effect of jurors seeing a criminal defendant in shackles was as follows: (1) it was well-established that jurors seeing a defendant in shackles in the courtroom provided grounds for a mistrial; (2) the only Oregon case addressing jurors seeing a defendant in shackles during transport outside the courtroom—Walton—held that the trial court did not err in denying the defendant's motion for a mistrial based on that incident; and (3) cases from other jurisdictions had consistently recognized that jurors seeing a criminal defendant in shackles during transport outside the courtroom was not so inherently prejudicial that a mistrial was required, because such an observation is fundamentally different from seeing the defendant in shackles in the courtroom. I cannot conclude that that state of the law “obviously foreshadowed” a ruling that no Oregon court has made or even suggested until today—that, to provide a constitutionally adequate defense, a lawyer must request a mistrial upon learning that jurors had seen the defendant in shackles outside the courtroom.

Given that state of the law, the factual record in this case, the burden of proof and persuasion, and the standard of review, I cannot agree that the post-conviction court erred when it concluded that petitioner had not met his burden of demonstrating by a preponderance of the evidence that his lawyer failed to exercise reasonable professional skill and judgment when he did not move for a mistrial. There may be a case where a petitioner for post-conviction relief establishes that the benefits of requesting a mistrial so obviously outweighed any possible downside that every reasonable counsel would request a mistrial under the circumstances and applicable law. On this factual record and the state of the law in 2013, I am not convinced that this is that case.

II. PREJUDICE TO PETITIONER

We have stated that, to establish prejudice resulting from a defense lawyer's failure to exercise reasonable professional skill and judgment, a petitioner “must demonstrate a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different.” Montez, 355 Or at 8 (internal quotation marks omitted). A reasonable probability “is a probability sufficient to undermine confidence in the outcome.” Id. (internal quotation marks omitted). Showing that the error “had some conceivable effect on the outcome of the proceeding[ ]” is not enough. Id. (internal quotation marks omitted). That is because “ ‘only those acts or omissions by counsel which have a tendency to affect the result of the prosecution can be regarded as of constitutional magnitude.’ ” Lichau, 333 Or at 363 (quoting Krummacher, 290 at 875 (brackets omitted)); see also Johnson v. Premo, 361 Or 688, 710, 399 P3d 431 (2017) (quoting Lichau, 333 Or at 359 (“[T]o obtain post-conviction relief, a petitioner must show that counsel's inadequacy had ‘a tendency to affect the result of his trial.’ ”)).

The majority opinion states that the tendency to affect the outcome standard “demands more than mere possibility, but less than probability[,]” citing Green v. Franke, 357 Or 301, 322, 350 P3d 188 (2015). ––– Or at –––– (slip op at 36:18-19). Applying that articulation of the standard, the majority opinion concludes that petitioner has met the “tendency to affect the outcome standard” because “there is more than a mere possibility that the trial court, in its exercise of discretion, would have granted a mistrial had counsel requested one.” ––– Or at –––– (slip op at 36:22 - 37:1-3). In my view, the majority opinion errs when it applies Green’s “more than a mere possibility” standard in this case.

As I will explain, Green expressly limited that articulation of the standard to post-conviction cases involving a claim that counsel's inadequate assistance affected “the result of the prosecution,” Lichau, 333 Or at 363, that is, the jury's guilty verdict. This court has not applied the “more than a mere possibility” standard outside that context, and, in my view, the majority opinion errs when it applies that standard in assessing whether counsel's failure to move for a mistrial was unconstitutionally prejudicial.

The issue in Green was whether the petitioner was entitled to post-conviction relief based on his criminal defense lawyer's failure to request an instruction “directing the jury to consider the evidence concerning each alleged victim separately and only as that evidence pertained to a specific charge or charges relating to that victim.” Green, 357 Or at 303. The post-conviction court denied relief, but the Court of Appeals reversed. Id. We allowed the state's petition for review, reversed the Court of Appeals’ decision, and remanded the case to the post-conviction court. Id.¹⁵

We also addressed the state's argument that “the Court of Appeals applied the wrong legal standard for prejudice[,]” noting that, if we agreed with the state's argument, “it would obviate the need for a remand.” Id. at 321. According to the state, the Court of Appeals had erred when it evaluated the effect a limiting instruction “could have had” on the jury's verdict instead of using “the probability-based formulation of the prejudice standard” that the post-conviction court had used. Id. at 321 (emphasis in original). We concluded that “the post-conviction court was mistaken” in applying its “probability-based formulation” of the standard, but we expressly limited that holding to cases involving the effect trial counsel's inadequacy had on the outcome of a jury trial. Id.

We stated that, under our prior cases, “where the effect of inadequate assistance of counsel on the outcome of a jury trial is at issue, it is inappropriate to use a ‘probability’ standard for assessing prejudice.” Id. at 322 (emphasis added). As we explained, “because many different factors can affect the outcome of a jury trial, in that setting, the tendency to affect the outcome standard demands more than mere possibility, but less than probability.” Id. (emphasis added). We then distinguished the prejudice standard in other settings that did not involve an assessment of how counsel's inadequacy may have affected the outcome of a jury trial. Specifically, we stated that, “where the prejudice inquiry focuses on a different type of outcome, for example, whether in the absence of inadequate assistance of counsel the petitioner would have entered a guilty plea, this court has described the tendency to affect the outcome test in terms of actual probability.” Id. at 323 n 13 (emphasis added).

The majority opinion cites no case applying its articulation of the prejudice standard outside the context of assessing the effect of inadequate assistance of counsel on the outcome of a jury trial. Instead, the majority opinion states that the Green court “did not purport to create a brightline rule,” and no other appellate decision “has considered whether such a limitation should exist[.]” ––– Or at –––– (slip op at 39:19-21 - 40:1). It is true that no court has expressly held that Green’s “more than a mere possibility” standard is limited to post-conviction cases addressing how trial counsel's constitutional inadequacy may have affected the jury's outcome, as stated in Green. But no court has applied that standard outside the context of assessing how counsel's inadequate assistance affected the jury's verdict, either. Again, because petitioner has the burden of persuasion on the prejudice issue, the question is whether he established that a court has applied that articulation of the standard in other contexts, not whether a court has limited it to jury verdicts.

Moreover, this court made the same observation that it made in Green—that the “mere possibility” standard applies where the disputed issue is the effect of inadequate assistance of counsel “on the outcome of a jury trial[,]” 357 Or at 322—in Richardson v. Belleque, 362 Or 236, 406 P3d 1074 (2017). The issue in Richardson was whether the petitioner's defense counsel had rendered constitutionally inadequate representation during the sentencing phase of the petitioner's trial. 362 Or at 238. That phase focused on whether the petitioner had a “severe personality disorder” indicating a propensity toward crimes that seriously endanger the life or safety of others, because ORS 161.725(1)(a) required the jury to find that the defendant had such a disorder before the court could sentence him as a dangerous offender. Id. We concluded that the petitioner's lawyer had rendered constitutionally inadequate assistance “by failing to adequately investigate and to consult an expert in preparation for the dangerous-offender hearing.” Id. at 239. We noted that whether the petitioner had the type of “severe personality disorder” that would support a dangerous offender sentence was a question for the jury. Id. at 247-48 (describing the jury instructions). At the dangerous-offender hearing, the jury “unanimously found” that the petitioner suffered from such a disorder. Id. at 248.

In assessing whether counsel's inadequate assistance at that hearing was prejudicial for purposes of post-conviction relief, we reiterated that, in Green, “this court explained that when a jury determination is involved, it is ‘inappropriate to use a “probability” standard for assessing prejudice.’ ” Id. at 265 (quoting Green, 357 Or at 322 (emphasis added)). We held that defense counsel's failure to call an expert witness or otherwise investigate his client's mental condition at the sentencing phase of a dangerous-offender proceeding was prejudicial, because “there was more than a mere possibility that the jury could have rejected the state's contention that petitioner suffered from a ‘severe personality disorder.’ ” Id. at 267-68 (emphasis added).

Other cases have applied a different analysis when the alleged constitutional inadequacy may have affected something other than a jury verdict. For example, in Peiffer v. Hoyt, 339 Or 649, 652, 125 P3d 734 (2005), the petitioner in a post-conviction case alleged that her criminal defense lawyer was constitutionally inadequate because he had failed to file a motion to suppress evidence that, according to the petitioner, had been “seized illegally” from her residence. The petitioner alleged that, if counsel had filed a motion to suppress, that “would have changed the outcome of [the petitioner's] case.” Id. at 652-53. The post-conviction court denied relief, concluding that, if counsel had filed a motion to suppress, it “would have been denied.” Id. On review in this court, the petitioner contended that she would have prevailed on a motion to suppress, because the officers had “conducted two separate searches” and the search warrant issued by the court “clearly did not authorize the second search * * *.” Id. at 660. We disagreed and affirmed the judgment of the post-conviction court because we could not say “that the evidence required the post-conviction court to find that the officers had exceeded the scope of the warrant” when they seized the evidence in dispute. Id.

Thus, under Peiffer, a post-conviction court addressing whether a petitioner was prejudiced by trial counsel's failure to file a motion to suppress would need to first determine whether the trial court would have granted the motion if counsel had filed it, leading to the suppression of the challenged evidence. If so, the post-conviction court would then need to determine whether the suppression of that evidence would have had a tendency to affect the outcome at trial, because the court cannot presume that suppression of evidence would always affect the outcome. That same general approach applies when counsel's alleged constitutional inadequacy was his failure to move for a mistrial, as the Court of Appeals has recognized.¹⁶

To establish prejudice from counsel's failure to move for a mistrial, a petitioner must show that the trial court would have granted the motion and declared a mistrial, or that it would have granted relief other than a mistrial, and that such relief would have had a tendency to affect the outcome in the case. That is how the Court of Appeals approached the issue in Monahan v. Belleque, 234 Or App 93, 227 P3d 777, rev. den., 348 Or 669 (2010). There, the court stated that a petitioner for post-conviction relief could show prejudice from trial counsel's failure to move for a mistrial by establishing that, if counsel had moved for a mistrial, “that motion would have had a tendency to cause the court to either grant the motion or to give a curative instruction that would have had a tendency to affect the prosecution in the underlying criminal case.” Id. at 99.

Under that approach, the prejudice analysis requires the court to assess what would have happened if trial counsel had moved for a mistrial. By 2013, this court had “ ‘consistently held that a motion for a mistrial is addressed to the sound discretion of the trial judge.’ ” State v. Smith, 310 Or 1, 24, 791 P2d 836 (1990) (quoting State v. Jones, 242 Or 427, 433, 410 P2d 219 (1966)). The reason for that rule is that the trial judge “is in the best position to ‘assess and rectify the potential prejudice to the defendant.’ ” Id. (quoting State v. Farrar, 309 Or 132, 164, 786 P2d 161 (1990)). The majority opinion acknowledges the trial court's discretion, ––– Or at –––– (slip op at 41:14-22), but it does not fully apply the mistrial standards that had been established by the Court of Appeals in evaluating a trial court's discretionary decision when a party requests a mistrial.

For example, in State v. Evans, 211 Or App 162, 166, 154 P3d 166 (2007), aff'd, 344 Or 358, 182 P3d 175 (2008), the Court of Appeals stated that, “[i]n ruling on a motion for a mistrial, a trial court must decide whether to grant the motion, to cure the effect of inappropriate conduct or testimony by giving a proper instruction instead, or to do nothing at all.” In State v. Woodall, 259 Or App 67, 313 P3d 298 (2013), rev. den., 354 Or 735 (2014), the Court of Appeals stated that, “of all the options a court has at its disposal * * * granting a motion for a ‘mistrial is a drastic remedy to be avoided if possible.’ ” Id. at 74-75 (quoting State v. Embry, 19 Or App 934, 941, 530 P2d 99 (1974)). In Embry, the Court of Appeals stated that it had “consistently held” that, before granting a mistrial based on “a potentially prejudicial impropriety,” it would be proper for the trial court, in its sound discretion, “to take no action at all if it determines that the impropriety is not sufficiently prejudicial as to require the intervention of the judge, or to take curative action in order to avoid the granting of a mistrial[.]” Embry, 19 Or App at 941 (internal citations omitted).

In this case, if petitioner's trial counsel had moved for a mistrial in 2013, then, under established Oregon law, the trial court would have been required to avoid the “drastic remedy” of declaring a mistrial if possible. Id. That would require the court to consider the alternatives. One obvious alternative would have been to simply ask the jurors if any of them had seen petitioner when they were leaving the courthouse on Friday. If none of the jurors acknowledged seeing him, that meant that the jurors either did not see the person being escorted in shackles from a doorway down the hall, or that they did not recognize that person as the defendant in their case. Under either circumstance, it would have been appropriate for the trial court to deny the motion for a mistrial without taking any further action. In that scenario, trial counsel's failure to move for a mistrial could not have been either constitutionally inadequate or prejudicial to petitioner.

If any jurors stated that they had seen and recognized petitioner, then the court could have considered a curative instruction, as in Walton, or it could have offered to excuse those jurors and replace them with alternate jurors, as the trial court had offered to do in Walton. If the court had excused those jurors, and petitioner was then convicted by jurors who had not seen him in shackles in the hallway, there would be no prejudice to the petitioner. If the trial court's inquiry revealed that one or more jurors had seen and recognized the person in shackles, and the court had concluded that it was not possible to excuse those jurors and replace them with alternate jurors, it is still not clear given the state of the law in 2013 that the trial court would have exercised its discretion to declare a mistrial. The court might have declined to declare a mistrial and given a cautionary instruction, and, under Walton, such a decision would have been within the trial court's discretion.

In my view, the existence of multiple alternatives that could have addressed the situation short of declaring a mistrial means that petitioner established, at best, that counsel's failure to move for a mistrial might have “had some conceivable effect on the outcome of the proceeding[,]” Montez, 355 Or at 8 (internal quotation marks omitted). As we stated in Montez, that “is not enough” to establish prejudice. Id.

Accordingly, for all the foregoing reasons, I respectfully dissent.

FOOTNOTES

1.   Before the post-conviction court and the Court of Appeals, petitioner alleged multiple claims that, over the course of two different trials, counsel provided him with inadequate assistance of counsel in various ways. Before this court, only counsel's conduct in connection with the jurors seeing him shackled during the second trial is at issue; we leave undisturbed the Court of Appeals’ resolution of petitioner's other claims on appeal.

2.   Throughout these proceedings, petitioner has maintained that counsel's alleged deficiencies deprived him of his right to counsel under both Article I, section 11, of the Oregon Constitution and the Sixth Amendment to the United States Constitution. Although the terminology used in connection with each provision sometimes differs, compare State v. Smith, 339 Or 515, 526, 123 P3d 261 (2005) (describing right under both constitutions as a right to “adequate assistance” (internal quotation marks omitted)) with Strickland v. Washington, 466 US 668, 686, 104 S Ct 2052, 80 L Ed. 2d 674 (1984) (Sixth Amendment guarantees not just counsel, but “effective” counsel), this court “has * * * recognized that the standards for determining the adequacy of legal counsel under the state constitution are functionally equivalent to those for determining the effectiveness of counsel under the federal constitution[,]” Montez v. Czerniak, 355 Or 1, 6-7, 322 P3d 487 (2014) (acknowledging court's ongoing commitment to “interpreting and applying Article I, section 11, independently of the United States Supreme Court's interpretation of the Sixth Amendment”). Unless specifically indicated otherwise, references to “inadequate” assistance in this opinion encompass both terms.

3.   Although handcuffs, belly chains, and ankle cuffs or “leg irons” are distinct forms of restraints, we use the words “shackled,” “shackles,” and “shackling” hereafter to describe petitioner's general disposition when the jury viewed him. See Webster's Third New Int'l Dictionary 2083 (unabridged ed. 2002) (shackled means “to confine the limbs of so as to prevent free motion.”).

4.   As noted above, petitioner alleged that counsel provided him with inadequate assistance of counsel in both trials in various ways. Before this court, only counsel's conduct in connection with the jurors seeing him shackled during the second trial is at issue.

5.   Contrary to the post-conviction court's implicit understanding, petitioner never “testified that his trial attorney did nothing after he repo[r]ted the possible contact with jurors[.]” As indicated above, petitioner testified that, upon being told that the jurors had seen petitioner in shackles, counsel “let the * * * judge know and the DA know and that's about—that's all that happened.” Conceivably, the court was referring to the operative petition, in which petitioner asserted that “trial counsel was not concerned and took no steps to petitioner's knowledge, [and] did not request a mistrial.” In any event, there is no contention that counsel did anything beyond passing petitioner's report on to the judge and prosecuting attorney.

6.   Although the state argues that the post-conviction court properly denied relief in petitioner's case, it perceives the post-conviction court's ruling as misstating the applicable prejudice standard. We will clarify the applicable standard below when we address whether counsel's performance was prejudicial. ––– Or at –––– (slip op at 36:9 - 50:20)

7.   Petitioner's reference to a “reasonable probability” invokes the standard for prejudice under the Sixth Amendment. See, e.g., Strickland, 466 US at 694 (stating that standard). We have explained, however, that the federal and state constitutional tests for prejudice in this context are “functionally equivalent.” Montez, 355 Or at 6-8 (comparing “reasonable probability” standard under Sixth Amendment with the “tendency to affect the outcome” standard applicable under Article I, section 11).

8.   In Long, defense counsel assigned error to the trial court's denial of a motion to dismiss the jury venire before voir dire “on the grounds that [the] defendant had been shackled in the presence of the panel.” Id. at 89. Ultimately, this court affirmed the trial court's ruling, reasoning that, other than bringing the motion itself, the defendant had offered no support for his assertion that he had been shackled in the presence of the venire. Id.

9.   In State v. Bates, the Court of Appeals consolidated these four considerations into three, collapsing the first—impinging the presumption of innocence— and second—detracting from the dignity of the judicial proceedings, into a single form of prejudice. 203 Or App 245, 251, 125 P3d 42 (2005), rev. den., 340 Or 483 (2006). We do not understand Bates as having in any way diminished the significance of either type of prejudice or as suggesting that each of those harms can only occur when accompanied by the other.

10.   The Court of Appeals has subsequently applied Kessler in a number of shackling cases, including before 2013, when petitioner's trial took place. See, e.g., State ex rel Juv. Dept. v. Millican, 138 Or App 142, 146, 906 P2d 857 (1995), rev. den., 323 Or 114 (1996) (holding that juveniles have the same rights as adult defendants with regards to shackling); State v. Merrell, 170 Or App 400, 403, 12 P3d 556 (2000), rev. den., 331 Or 674 (2001) (observing that Oregon courts have treated shackling issues similarly under Article 1, section 11, “impartial jury” clause and federal Due Process Clause); Bates, 203 Or App at 249-50 (holding that any error in the trial court denying defense counsel's objection to wearing a nonvisible stun belt was harmless under Article 1, Section 11, and the federal Due Process Clause); Wall, 252 Or App at 443 (holding that the trial court erred in denying defense counsel's motion to remove nonvisible leg restraint, also under state and federal constitutions).

11.   This court has since discussed the right to appear free of unnecessary restraints during trial in other cases decided after the date of petitioner's trial. See, e.g., State v. Washington, 355 Or 612, 628, 330 P3d 596, cert. den., 574 US 1016 (2014) (announcing, for the first time in our court, that the right to appear free from shackles is grounded in the “impartial jury” clause, Article 1, section 11); see also State v. Guzek, 358 Or 251, 270-71, 363 P3d 480 (2015) (holding that evidence supported trial court's decision to require defendant to a wear stun belt during penalty phase of capital prosecution).

12.   In the post-conviction court, petitioner alleged that counsel's performance had been deficient because, in part, counsel had not “inquire[d] of the court staff and jury what was observed.” That allegation was made in conjunction with the allegation that counsel had performed inadequately by not moving for a mistrial, and petitioner's briefing does not argue that counsel was independently obligated to conduct a further factual inquiry, with or without the trial court's involvement or, with respect to contacting any of the jurors, the trial court's authorization.

13.   The post-conviction court cited cases from jurisdictions outside Oregon stating that viewing a defendant in handcuffs outside the courtroom is distinguishable from trying the defendant in restraints and therefore not inherently prejudicial. We discuss some of that case law when we address the prejudice prong of petitioner's post-conviction case. For now, we simply note that neither the post-conviction court nor the state has explained how those nonbinding decisions—or the force of their reasoning—would somehow foreclose the argument that, under Oregon law, a mistrial was warranted here.

14.   This court addressed a shackling incident that appears to have occurred outside the courtroom in State v. Walton, 311 Or 223, 250, 809 P2d 81 (1991), where a defendant on trial for aggravated murder was seen by a juror and alternate juror while wearing handcuffs and grouped with other prisoners. The trial court had denied the defendant's motion for mistrial, but, after questioning those jurors individually and outside the presence of the other jurors, the court gave a curative instruction that told them to disregard what they had seen. Id. Although the viewing in Walton appears to have occurred outside the courtroom, this court did not address that fact or whether it had any bearing on the defendant's right not to be shackled in the presence of the jury. We discuss that case further below.

15.   The state may believe that, by bringing the issue to the trial court's attention, counsel satisfied his obligation to advance a favorable shackling argument on petitioner's behalf. But whatever argument, if any, counsel advanced to the court in chambers, that approach fell short of counsel's obligation to protect petitioner's interests, as it neither resulted in a mistrial or other corrective action nor preserved any related argument for appeal. See State v. Williams, 322 Or 620, 624 n 7, 912 P2d 364 (1996) (“Discussion off the record of matters as to which issues on appeal could arise is ill-advised, either because no official record is made of the matters or because whatever record that is made often is summary in nature.”).

16.   In Walton, for example, the trial court had declined the defendant's motion for mistrial but, in response to that motion, had taken various protective measures and offered to take more. 311 Or at 250 (noting that trial court had, among other things, inquired of the potentially affected jurors, instructed them against drawing adverse inferences based on what they might have seen, and offered to replace one juror with an alternate who had not seen the defendant in shackles).

17.   Although the post-conviction court observed that the record disclosed only petitioner's perception, not the perception of any of the jurors, there is no indication in the court's written opinion that counsel's understanding of the event played any role in its conclusion that counsel's performance had not been inadequate. The opinion's entire discussion of performance relates to the fact that petitioner's encounter with the jurors had been brief and outside the courtroom, and that the available case law discussing such encounters had distinguished them from in-court shackling cases.

18.   The post-conviction court expressly found that petitioner had reported to counsel that “three of [his] jurors looked at him and saw him” and that counsel recalled petitioner telling him about that encounter.

19.   As previously noted, we agree with the dissent that, although the opinion in Walton does not state where that observation took place, it is most reasonable to assume that it took place outside the courtroom, particularly given that the defendant was seen in the company of other prisoners, something that would be unlikely to occur in the courtroom during a trial.

20.   We discuss the significance of the discretion afforded trial courts in evaluating mistrial motions below, when we address the applicable prejudice standard.

21.   The state makes the same argument, which we also reject.

22.   In the state's view, because what petitioner must show is that there was a “ ‘tendency to affect the outcome of the prosecution’ ” (quoting Green, 357 Or at 322 (emphasis in state's brief)), that assessment applies to the jury's decision, not the trial court's decision whether to grant a mistrial in the first instance. Thus, the state argues, petitioner must first establish that the trial court was legally obligated to declare a mistrial if requested, only after which the “tendency to affect the outcome” analysis applies. The state reasons that, as with cases in which the forgone objection relates to evidence that the jury ultimately hears, the “tendency” analysis only applies once an appellate court determines that, as a matter of law, the jury should not have heard certain evidence or should have been instructed in a specific way. See Peiffer v. Hoyt, 339 Or 649, 660, 125 P3d 734 (2005) (holding that post-conviction petitioner could not establish that counsel's failure to file a suppression motion had been prejudicial without first showing that trial court would have been legally required to suppress the evidence). The state does not explain, however, why a mistrial motion is comparable to an evidentiary motion—a mistrial motion does not determine what the jury considers in rendering its verdict; it determines whether the jury renders a verdict at all. Thus, we disagree with the state's proposed application of Green and proceed to discuss the dissent's proposition.

23.   The dissent points to Richardson v. Belleque, 362 Or 236, 406 P3d 1074 (2017), as additional support, ––– Or at –––– (Bushong, J., dissenting) (slip op at 29:1 - 30-4), but, because that case, like Green, involved a jury-determination issue, its citation to Green lends no support to the dissent's argument that the “more than mere possibility” standard applies only to jury determinations.

24.   Further, we do not understand Walton, which dealt with an encounter that presumably took place outside the courtroom, as drawing a distinction with respect to where the encounter took place, because nothing about the court's conclusion was dependent on, or even referenced, the location of the defendant. 311 Or at 250.

25.   Petitioner was charged with and ultimately convicted of first-degree rape and sexual abuse, crimes that include the element of forcible compulsion, ORS 163.375, which requires the jury to find that a person used physical force or threatened the victim with immediate or future death or physical injury, ORS 163.305.

26.   The dissent likewise relies on case law from other jurisdictions. ––– Or at –––– (Bushong, J., dissenting) (slip op at 21:18 - 23:21). From our review of that case law, it is similarly sparse and conclusory with regard to this issue, and the dissent does not elaborate further as to any other justifications for distinguishing in-court from out-of-court shackling.

1.   Under the Sixth Amendment to the United States Constitution, the petitioner must show that counsel's performance was deficient, and that the deficient performance caused prejudice to the defense. Strickland v. Washington, 466 US 668, 687, 104 S Ct 2052, 80 L Ed. 2d 674 (1984). We have stated that the standards for adequacy of counsel under the state and federal constitutions are “functionally equivalent.” Montez, 355 Or at 6.

2.   Petitioner's reference to the possibility of a “hung jury” was likely based on the jury's question to the court during the first day of deliberations, which asked what it should do if there were not enough votes to reach a verdict, noting that the verdict form did not have a space to indicate a “hung jury.”

3.   Petitioner's counsel was admitted to the Oregon bar in 1992 and the Arizona bar in 1985. He practiced continually in criminal defense throughout his career.

4.   The post-conviction court also made a factual finding that the deputies escorting petitioner out of the holding cell into the courthouse hallway “saw jurors in the hallway and pulled [petitioner] back into the holding cell.” That finding as to what the deputies saw does not mean that the jurors who were going down the stairs necessarily saw and recognized the person in shackles and cuffs as the defendant in their case, and, as the post-conviction court noted, there was “no evidence” from the viewpoint of the jurors and “no evidence that the jurors recognized” petitioner.

5.   The Court of Appeals stated that “the post-conviction court found that there was no evidence that the jurors recognized petitioner in the hallway.” Nelson v. Cain, 334 Or App 326, 330, 556 P3d 1 (2024).

6.   The majority opinion states that I am “wrong” in concluding that “there is evidence from which [the] facts could be decided [in] more than one way,” Ball, 250 Or at 487, because “for a fact to be in dispute, there must be evidence of that fact in the record.” ––– Or at –––– (slip op at 31:9-11). But Ball does not state that there must be evidence in the record of the fact that is in dispute. It suffices if there is evidence “from which [the] facts could be decided [in] more than one way,” Ball, 250 Or at 487, which means that evidence of counsel's statement to petitioner from which the court could infer what counsel concluded from his investigation is enough.

7.   At the time of petitioner's criminal trial, ten votes were required for a conviction for felonies other than first-degree murder. Or. Const., Art. I, § 11; ORS 136.450 (2013), amended by Or. Laws 2019, ch. 635, § 8, amended again by Or. Laws 2021, ch. 478, § 11. Seven years later, the United States Supreme Court held that a unanimous jury verdict was required by the Sixth Amendment to the United States Constitution for all felony convictions. Ramos v. Louisiana, 590 US 83, 93, 140 S Ct 1390, 206 L Ed. 2d 583 (2020).

8.   If counsel chose not to move for a mistrial because he liked petitioner's chances of prevailing on some or all charges better than his chances of prevailing on retrial if the court declared a mistrial, the actual verdict in this case provided at least some support for that optimism. The jury acquitted petitioner of one of the three counts of first-degree sexual abuse—a Measure 11 crime that carried a mandatory 75-month prison sentence—finding him guilty of the lesser-included offense of third-degree sexual abuse (which does not carry a mandatory prison sentence).

9.   The majority opinion concludes that, under Burdge, a failure to make a legal argument constitutes constitutionally inadequate assistance of counsel when (1) a legal issue arises in the course of representing a defendant; (2) the legal issue “has not yet been decided under controlling case law”; (3) the issue “is susceptible of an interpretation that so obviously would offer potential benefits to the defendant that all reasonable counsel would identify it”; and (4) “all such counsel would conclude that, under the circumstances, the potential benefits of advancing that interpretation exceeded any risks[.]” ––– Or at –––– (slip op at 26:9-15).

10.   The issue in Burdge was whether a post-conviction petitioner's criminal defense lawyer provided constitutionally inadequate assistance when he failed to object to the application of a statute to increase the petitioner's sentence. 338 Or at 494. The Court of Appeals concluded that a reasonable lawyer would have objected, because the argument was plausible, the petitioner “faced no tactical risks for making” the argument, and the petitioner “would have reaped a substantial benefit if the argument had succeeded.” Id. (summarizing Court of Appeals’ decision). We reversed, concluding that the Court of Appeals had failed to apply the “reasonable professional skill and judgment” test. Id. at 497. Even if the statute at issue in that case was ambiguous, we explained, it was “not so obviously ambiguous that any lawyer exercising reasonable professional skill and judgment necessarily would have seen it.” Id. at 497-98.

11.   The Court of Appeals has been persuaded that a defense lawyer exercising reasonable skill and judgment would have raised an unsettled legal issue only in limited circumstances, such as when the unsettled issue was whether the defendant's convictions would merge. See Bumgarner v. Nooth, 254 Or App 86, 97, 295 P3d 52 (2012) (post-conviction court did not err in granting relief when trial counsel failed to argue that multiple convictions for first-degree rape and other sexual offenses should merge given “the obvious possible benefits to [the] petitioner of raising the merger issue”); Ross v. Hill, 235 Or App 340, 349-50, 231 P3d 1185, rev. den., 349 Or 56 (2010) (post-conviction court erred in concluding that the petitioner's trial counsel was not deficient in failing to argue that two kidnapping convictions should merge). The Court of Appeals was persuaded in another case that a defense lawyer exercising reasonable skill and judgment would have sought dismissal of an attempted rape charge on statute of limitations grounds, “given the obvious possible benefits” of raising the statute of limitations—“dismissal of the charge, with no negative strategic impact on the defense of the other charges[.]” Lamb v. Coursey, 238 Or App 647, 652, 243 P3d 130 (2010).

12.   Similarly, the issue in Miller v. Lampert, 340 Or 1, 5-6, 125 P3d 1260 (2006), was whether a post-conviction petitioner's trial counsel was constitutionally inadequate by failing to argue that the petitioner could not be punished with an enhanced sentence based on factual findings made by the court, not the jury. Two years after the petitioner's criminal trial, the Supreme Court held that practice to be unconstitutional in Apprendi v. New Jersey, 530 US 466, 490, 120 S Ct 2348, 147 L Ed. 2d 435 (2000). This court rejected the petitioner's claim for post-conviction relief, concluding that, under the state of the law that existed at the time of the petitioner's trial, counsel “was not required to anticipate” the Supreme Court's subsequent ruling in Apprendi and object to the enhanced sentence. Miller, 340 Or at 16.

13.   The majority opinion identifies four types of “manifest prejudice” that result when a person is tried in shackles: (1) shackles impinge on the presumption of innocence; (2) they detract from the dignity of the judicial proceedings; (3) they may inhibit a defendant's ability to consult with defense counsel; and (4) they may inhibit the defendant's decision on whether to take the stand as a witness. ––– Or at –––– (slip op at 19:17-21 - 20:1). Three of those concerns apply only to in-court shackling.

14.   Although this court's opinion in Walton does not expressly state that the jurors’ observation of the defendant in handcuffs occurred outside the courtroom, that is the only inference that makes any sense. There would be no reason for a defendant in a criminal trial to be “in handcuffs with a group of other prisoners” inside the courtroom. The majority opinion agrees that it is “reasonable to assume” that the jurors’ viewing in Walton occurred outside the courtroom, ––– Or at –––– n 19 (slip op at 34 n 19).

15.   We noted on review that this court had recently decided Pereida-Alba v. Coursey, 356 Or 654, 342 P3d 70 (2015), a post-conviction case that involved a trial counsel's failure to request a jury instruction. Green, 357 Or at 312-13. We concluded in Green that “the proper course of action” under those circumstances would be to reverse the decision of the Court of Appeals, reverse the post-conviction court's judgment, and remand the petitioner's claim to the post-conviction court “for further consideration in light of this court's decision in Pereida-Alba.” Id. at 320.

16.   The majority opinion states that it is “hesitant” to recognize the limitation “that this court might someday” confront, and questions whether this case “is of the sort in which to make such a distinction[,]” in part because no party “besides the author of the dissent” has suggested that the “more than a mere possibility” articulation does not apply to petitioner's claim. ––– Or at –––– (slip op at 40:2-8). With respect, I disagree. First, this court has an obligation, in my view, to determine the prejudice standard that governs this case and then apply that standard, regardless of whether any party has raised the specific issue. Second, the state asserted in its brief that petitioner “misstates the prejudice standard” in his opening brief. Thus, the prejudice standard that applies here is an issue that is in dispute in this case.

DeHOOG, J.

Bushong, J., dissented and filed an opinion.