OWENS, J.

¶ 1 Rene P. Paumier appeals his conviction for residential burglary and third degree theft. This case requires us to determine if Paumier's right to a public trial was violated when the trial court individually questioned potential jurors in chambers. We have previously held that a court may close a courtroom to the public only after considering the factors established in State v. Bone-Club, 128 Wn.2d 254, 258-59, 906 P.2d 325 (1995). Moreover, we have held in State v. Wise, ___ Wash.2d ___, ___, 288 P.3d 1113 (2012), that individual questioning of potential jurors in chambers without first considering the Bone-Club factors is a closure creating a presumption of prejudice. Therefore, Paumier is entitled to a new trial because the trial court closed the courtroom without first considering the Bone-Club factors. Because we affirm the Court of Appeals on this issue, there is no reason to address whether the trial court also violated Paumier's right to self-representation. We affirm the Court of Appeals reversal of the trial court on the public trial right grounds alone.

FACTS

¶ 2 Paumier was convicted of residential burglary and third degree theft in Mason County Superior Court. He was sentenced to 25 months for the burglary and 365 days for the theft. Paumier is now appealing the jury selection process and his right to represent himself.

¶ 3 During voir dire, the trial judge individually questioned four potential jurors in her chambers. The trial judge, sua sponte, offered to privately question any juror on sensitive matters if a juror so chose. Specifically, the judge said:

Suppl. Report of Proceedings (RP) at 9-10. The private matters discussed included personal health issues, criminal history, and familiarity with the defendant or the crime. The prosecution, defense counsel, and Paumier were all present for the questioning and offered no objections. Further, the in-chambers questioning was recorded and transcribed by the court. But the trial judge never conducted a Bone-Club analysis1 prior to privately questioning the potential jurors. Such an analysis would have, among other things, required the judge to consider alternatives to closure and to mention Paumier's right to a public trial. Finally, of the four privately questioned, two jurors were excused.

¶ 4 After two days of jury selection, Paumier requested to represent himself. The trial judge denied Paumier's request stating that "the request comes too late" as the jury had already been selected (although not sworn in). 1 Partial RP at 9. The jury ultimately convicted Paumier of both residential burglary and second degree theft.

¶ 5 Paumier then appealed his convictions, claiming that the trial court violated both his right to a public trial and his right to self-representation. The Court of Appeals reversed the trial court on both grounds. State v. Paumier, 155 Wn.App. 673, 685, 687, 230 P.3d 212 (2010). The State petitioned for review by this court on both issues, which we granted. State v. Paumier, 169 Wn.2d 1017, 236 P.3d 206 (2010).

ISSUE

¶ 6 Did the trial court err in failing to conduct a Bone-Club analysis prior to individually questioning jurors in chambers?

ANALYSIS

Failing To Conduct a Bone-Club Analysis before Privately Questioning Potential Jurors in Chambers Is Structural Error

¶ 7 Paumier claims the private questioning of four potential jurors violated his right to a public trial. Whether a defendant's constitutional right to a public trial has been violated is reviewed de novo on direct appeal. Wise, ___ Wash.2d at ___, 288 P.3d 1113 (quoting State v. Easterling, 157 Wn.2d 167, 173-74, 137 P.3d 825 (2006)).

¶ 8 It is well established that a criminal defendant has a right to a public trial as guaranteed by our state and federal constitutions. U.S. Const. amend. VI; Wash. Const. art. I, § 22 ("the accused shall have the right ... to have a speedy public trial"); State v. Momah, 167 Wn.2d 140, 147, 217 P.3d 321 (2009), cert. denied, ___ U.S. ___, 131 S.Ct. 160, 178 L.Ed.2d 40 (2010). "This presumption of openness extends to voir dire." Id. at 148, 217 P.3d 321. However, as "[t]he right to public trial is not absolute," the presumption may be overcome. Wise, ___ Wash.2d at ___, 288 P.3d 1113; see also Bone-Club, 128 Wash.2d at 259, 906 P.2d 325. A trial court may close the courtroom, so long as it considers the five criteria outlined in Bone-Club, 128 Wash.2d at 258-59, 906 P.2d 325. As part of the Bone-Club analysis, the trial judge must consider alternatives to closure to ensure the least restrictive means of closure is adopted. Wise, ___ Wash.2d at ___, 288 P.3d 1113; Bone-Club, 128 Wash.2d at 259-60, 906 P.2d 325. Even the United States Supreme Court requires a trial court to consider alternatives before closing the courtroom. Presley v. Georgia, 558 U.S. 209, 130 S.Ct. 721, 725, 175 L.Ed.2d 675 (2010).

¶ 9 We addressed the same issue — whether private questioning of potential jurors in chambers without conducting a Bone-Club analysis violates a defendant's public trial right — in Wise. Because the issue is identical and the facts are similar, we rely on and incorporate the reasoning from that case here. The following rules summarize part of our holding in Wise. To begin, individually questioning potential jurors is a courtroom closure requiring a Bone-Club analysis. Wise, ___ Wash.2d at ___, 288 P.3d 1113. Failure to conduct the Bone-Club analysis is structural error warranting a new trial because voir dire is an inseparable part of trial. Id. at ___, ___-___, 288 P.3d 1113.

¶ 10 Applying those rules here, the trial court erroneously closed the courtroom when it privately questioned potential jurors during voir dire without first conducting a Bone-Club analysis. Such an error is structural and warrants a new trial just as it did in Wise. "[W]e cannot reasonably order a `redo' of voir dire to remedy the public trial right violation that occurred here." Id. at ___-___, 288 P.3d 1113. Accordingly, we are left with no other choice but to order a new trial.

¶ 11 Today's holding may seem in conflict with our previous decision in Momah, but it is not. As we made clear in Wise, Momah relied on unique facts to conclude that no public trial right violation occurred when the jurors were individually questioned. Id. at ___, 288 P.3d 1113. Specifically, the defendant in Momah "affirmatively assented to the closure of voir dire and actively participated in designing the trial closure and [] though it was not explicit, the trial court ... effectively considered the Bone-Club factors." Wise, ___ Wash.2d at ___, 288 P.3d 1113. In stark contrast, these facts do not exist here. Paumier's mere presence in the courtroom does not qualify as active participation. Further, the trial court gave no indication it considered any of the Bone-Club factors. Thus, our holding is not in conflict with Momah.

Structural Error, Like Violation of the Public Trial Right, Presumes Prejudice

¶ 12 The next concerns we must address are whether Paumier had to contemporaneously object to the individual questioning to preserve the error and if he must show prejudice on appeal. Ordinarily, a party must contemporaneously object to preserve an error. RAP 2.5. However, RAP 2.5(a) allows an unobjected to error to be raised on appeal if it is a "manifest error affecting a constitutional right." This court has previously interpreted "manifest error" as requiring a defendant to show actual prejudice. State v. O'Hara, 167 Wn.2d 91, 99, 217 P.3d 756 (2009). Here, that would mean Paumier must show actual prejudice because he failed to object to the closure during trial.2 But RAP 2.5(a) does not apply in its typical manner here because the improper courtroom closure was structural error. As noted in Wise, "[n]othing in our rules or our precedent precludes different treatment of structural error as a special category of `manifest error affecting a constitutional right.'" Wise, ___ Wash.2d at ___ n. 11, 288 P.3d 1113 (quoting RAP 2.5(a)(3)).

¶ 13 In fact, there is good reason to treat structural errors, like violation of a defendant's public trial right, differently.3 A structural error "affect[s] the framework within which the trial proceeds" and renders a criminal trial an improper "`vehicle for determin[ing] guilt or innocence.'" Arizona v. Fulminante, 499 U.S. 279, 310, 111 S.Ct. 1246, 113 L.Ed.2d 302 (1991) (quoting Rose v. Clark, 478 U.S. 570, 578, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986)). The right to a public trial is a unique right that is important to both the defendant and the public. Wise, ___ Wash.2d at ___, 288 P.3d 1113; Momah, 167 Wash.2d at 148, 217 P.3d 321. Moreover, assessing the effects of a violation of the public trial right is often difficult. Wise, ___ Wash.2d at ___, 288 P.3d 1113 (quoting United States v. Marcus, ___ U.S. ___, 130 S.Ct. 2159, 2165, 176 L.Ed.2d 1012 (2010)). Requiring a showing of prejudice would effectively create a wrong without a remedy. Therefore, we do not require a defendant to prove prejudice when his right to a public trial has been violated. Wise, ___ Wash.2d at ___, 288 P.3d 1113.

CONCLUSION

¶ 14 Following the rule enunciated in Wise, we find that Paumier need not prove that violation of his public trial right prejudiced him. The trial court's failure to conduct a Bone-Club analysis was structural error that warrants reversal on appeal, with or without a contemporaneous objection.4 To be clear, our holding does not preclude a trial judge from closing a courtroom for individual questioning. Rather, our holding merely requires a trial court to conduct a Bone-Club analysis first. Because that analysis was not conducted here, Paumier is entitled to a new trial. We affirm the Court of Appeals.

WE CONCUR: TOM CHAMBERS, MARY E. FAIRHURST, DEBRA L. STEPHENS, Justices and GERRY L. ALEXANDER, Justice Pro Tem.

MADSEN, C.J. (dissenting).

¶ 15 Several cases concerning the right to a public trial have come before the court, raising a number of questions about a defendant's right to a public trial, including when a violation of this right occurs and what remedies are available if the right is violated. State v. Sublett, ___ Wash.2d ___, ___ P.3d ___ (2012) (plurality opinion); State v. Paumier, ___ Wash.2d ___, 288 P.3d 1126 (2012); State v. Wise, ___ Wash.2d ___, 288 P.3d 1113 (2012); In re Pers. Restraint of Morris, ___ Wash.2d ___, 288 P.3d 1140 (2012) (plurality opinion). In Sublett, I have written an extensive concurrence placing these multiple issues in context and explaining why I believe the court's jurisprudence in this area is erroneous in many respects.

¶ 16 Unfortunately, the court has adopted a series of such inflexible rules that Mr. Rene Paumier's conviction in the present case must be reversed — not on the ground that the closure of the proceedings for private, limited, in-chambers questioning of potential jurors was unjustified and a violation of the right to a public trial, but instead because the trial court did not inquire into whether the closure was justified.

¶ 17 I agree that a trial court errs when, before closing the courtroom, it fails to make an on-the-record inquiry into whether closure is justified under article I, section 22 of the Washington State Constitution. State v. Bone-Club, 128 Wn.2d 254, 906 P.2d 325 (1995). The "Bone-Club" inquiry must be made to determine whether the interest claimed to justify closure of the proceedings is a compelling interest that overrides the defendant's right to a public trial and whether the proposed closure is essential to preserve that interest, and the court must ensure that the closure is narrowly tailored. Id. at 258-59, 906 P.2d 325 (other requirements exist, including that anyone present must be given the opportunity to object to closure). A nearly identical inquiry is required under the Sixth Amendment to the United States Constitution before closing the proceedings in a criminal trial. Waller v. Georgia, 467 U.S. 39, 45, 104 S.Ct. 2210, 81 L.Ed.2d 31 (1984).

¶ 18 But contrary to the majorities here and in Wise and a majority of the court in Morris, I do not agree that the error in failing to conduct the on-the-record inquiry and enter written findings must be deemed structural error requiring reversal of the defendant's conviction and a new trial. It is highly likely that if the required inquiry and findings had been made, the result would be that closure was justified and not a violation of article I, section 22 or the Sixth Amendment. Thus, the error in these cases is the failure to conduct the inquiry, not an unjustifiable closure that necessarily violates the defendant's right to a public trial.

¶ 19 But in each of these three cases, the failure to conduct the inquiry, alone, is deemed to be the equivalent of an unconstitutional, impermissible, unjustifiable closure that constitutes structural error — the most egregious form of constitutional error, for which no harmless error standard can be applied. Thus, the majorities in these cases equate the failure to conduct the inquiry — which is, without question, a serious error — with a violation of the right to a public trial, which is a far more serious error.1

¶ 20 The result is a rule that says in effect that the defendant has a constitutional right to the inquiry into whether his right to a public trial would be violated by closure, and if that inquiry is not conducted it is a constitutional violation of the very worst sort, i.e., structural error. And this is true regardless of whether the inquiry, if made, would show that the closure was perfectly constitutional.

¶ 21 It must be remembered when considering these cases that the majority in Wise has also virtually distinguished out of existence the one case where this court examined the record on review to determine whether a violation of the right to a public trial occurred and whether there was structural error requiring reversal, in circumstances where the trial court failed to engage in the Bone-Club inquiry prior to closing the proceeding for limited voir dire of potential jurors. See Wise, ___ Wash.2d at ___, 288 P.3d 1113 (in effect overruling State v. Momah, 167 Wn.2d 140, 217 P.3d 321 (2009), cert. denied, ___ U.S. ___, 131 S.Ct. 160, 178 L.Ed.2d 40 (2010)). The court in Momah held that reversal was not required.

¶ 22 I believe that posttrial examinations of the records in this case, Wise, and Morris, should be made. It is highly likely such review would show that the closures in these cases were not unconstitutional. Each of these cases involves the question whether limited, private, individual questioning of a few potential jurors on sensitive matters violates the right to a public trial, as I explain in my concurrence in Sublett. Importantly, the public nature of the proceedings is protected to a large degree by the fact that the proceedings were recorded, transcribed, and made part of the public record.

¶ 23 But because there was no Bone-Club or Waller inquiry before the private questioning of the venire members occurred, the defendants each obtain an entirely new trial, no matter the costs in delay, likely loss of evidence, costs in terms of time and effort of everyone involved (trial court, attorneys, victims, witnesses, etc.), and the added financial burden placed on the criminal justice system. They obtain this trial not because their right to a public trial was violated, but because in the absence of the appropriate inquiry we do not know at this stage of the proceedings whether their right to a public trial was violated. It makes no difference to the majorities whether posttrial appellate review or remand for fact findings or a hearing could show that the closures satisfied Bone-Club.

¶ 24 As I also show in my concurrence in Sublett, appellate courts in other jurisdictions routinely engage in posttrial inquiries into whether a closure was justified. In fact, in Waller, the United States Supreme Court itself examined the record of the consolidated cases before it to determine if the closure of the suppression hearing that occurred was justified under the Waller factors. Waller, 467 U.S. at 48-49, 104 S.Ct. 2210 ("[a]pplying these tests to the cases at bar"). In the Sublett concurrence, I also cite a number of cases where courts have found no public trial violations in connection with limited in-chambers questioning of potential jurors.

¶ 25 I would not assume that every closure in the absence of a Bone-Club inquiry is an unconstitutional violation of the defendant's right to a public trial. Rather than automatically granting new trials in these cases, this court should examine the records to determine whether the closures were justified. If the record does not resolve the question, then the cases should be remanded for factual determinations of whether the closure was justified under the Bone-Club factors. This remedy can resolve the question whether the closure actually constituted a closure of the trial in violation of the right to a public trial. If either on the appellate record or on remand (for entry of factual findings or a hearing followed by factual findings) a determination can be made through a posttrial Bone-Club inquiry that the closure did not violate the defendant's article I, section 22 right to a public trial, then the matter is at an end.

¶ 26 There is nothing in United States Supreme Court precedent that prevents this approach. Any constraints are of this court's own doing, and they can be traced to Bone-Club. But in Bone-Club, there is nothing that explains why there cannot be a posttrial inquiry into whether an unconstitutional closure in fact occurred.

¶ 27 Bearing in mind that the proponent has the burden of justifying closure, if, after a posttrial evaluation, it turns out that either a closure is found to be unjustified or if the question cannot be resolved to show a constitutional closure, then the conclusion would have to be that the defendant's right to a public trial was violated. Then, and only then, would it be necessary to decide whether the violation was structural error requiring reversal and a new trial.

¶ 28 And if it turns out that an unconstitutional closure occurred, then, as Justice Wiggins correctly explains, the rules of appellate procedure should apply in public trial right cases just as they do in any appellate case involving a claimed constitutional violation. Moreover, as he also explains, just because structural error is found in a particular context involving a particular constitutional right, this does not necessarily mean that any and all violations of that particular constitutional right will be structural error in all contexts. As I point out in my concurrence in Sublett, many courts have held, as this court did in Momah, that public trial violations are not always structural error.

¶ 29 If the issue is properly reached, the court should conclude that, as in Momah, no structural error occurred here. Then, because Mr. Paumier failed to object to the closure, he should be required to satisfy the strict requirements to prevail when claimed constitutional error was not preserved. Under the rules of appellate procedure he is not entitled to any relief, as Justice Wiggins' dissent shows.

¶ 30 In summary, in this case, as in Wise and Morris, the trial court's error was the failure to engage in the Bone-Club inquiry on the record prior to closing the court for private, limited questioning of a few potential jurors on sensitive matters. The failure to make the Bone-Club inquiry on the record prior to closing the proceeding is a serious error implicating the important right of the defendant to a public trial. However, this error is not itself a closure of the courtroom. The simple fact is that no determination has ever been made about whether the closure in this case, or in Wise or Morris, was justified, and so no determination has ever been made about the constitutionality of these closures.

¶ 31 As many appellate courts have either done themselves, or have directed lower courts to do, a posttrial inquiry into the propriety of the closure should be conducted. If this can be done on the appellate record, it should be done. If not, these cases should be remanded for entry of findings on the matter or a hearing followed by findings, whichever is appropriate in the circumstances.

¶ 32 But instead, the majorities in these cases have unfortunately perpetuated a theory of public trial cases that equates (a) the required inquiry into whether closure is justified to (b) an unjustified or unjustifiable closure, which is an unconstitutional closure. With this theory of public trial cases, the error in failing to conduct the Bone-Club inquiry automatically transforms any closure into an unconstitutional closure that is structural error, with the defendant automatically obtaining the windfall of reversal of his conviction and a new trial. I cannot agree with this approach.

¶ 33 As should be apparent, I believe the court should overrule our cases to the extent they require reversal of convictions and new trials solely because the trial court failed to engage in the Bone-Club inquiry before trial. I also believe that the court should overrule any case to the extent it concludes that the failure to engage in the Bone-Club inquiry, alone, is structural error.

¶ 34 For the reasons stated here and in more detail in my concurrence in Sublett, I dissent.

WIGGINS, J. (dissenting).

¶ 35 It is not given to us to have perfect trials. See Brown v. United States, 411 U.S. 223, 231-32, 93 S.Ct. 1565, 36 L.Ed.2d 208 (1973) ("[T]here are no perfect trials."). Nevertheless, the fruitless search for a perfect trial is reflected in the majority opinions in this case, State v. Wise, ___ Wash.2d ___, 288 P.3d 1113 (2012) and In re Personal Restraint of Morris, ___ Wash.2d ___, 288 P.3d 1140 (2012) (plurality opinion). The approach advocated for in these cases belies a platonic conception of a trial as something that has the potential to be wholly without flaws. But a trial is a uniquely human affair and can only be as flawless as the judges and lawyers who conduct it. We strive for perfection but rarely attain it. Humans are imperfect.

¶ 36 That is why, on review, our task is not to determine whether the defendant received a trial completely free of defects, but to determine whether the defendant received a fair trial — a trial that does credit to our justice system and to the concept of due process. See Lutwak v. United States, 344 U.S. 604, 619, 73 S.Ct. 481, 97 L.Ed. 593 (1953) ("A defendant is entitled to a fair trial but not a perfect one."). This means, in all instances, that before reversing a conviction we must inquire whether a claimed error actually made the trial less fair, which ordinarily means asking whether it caused prejudice or was harmless.1

¶ 37 In Rene Paumier's case, the claimed public trial error is entirely theoretical; that is, it is premised solely on notions of policy and judicial administration that have nothing to do with the fairness of the underlying trial or whether Paumier committed the crime of which he is accused.

¶ 38 The majority reverses Paumier's conviction for an error to which Paumier never objected at trial, an error from which neither Paumier nor the majority can identify any prejudice whatsoever. Indeed, the limited in-chambers voir dire probably helped Paumier's case by encouraging potential jurors to be more forthcoming in responding to voir dire. Lacking any indication of real prejudice, the majority extends to this case a presumption of prejudice that neither we nor the United States Supreme Court has ever applied to limited unobjected-to in-chambers voir dire, the presumption of "structural error."

¶ 39 The structural error doctrine should be limited to extraordinary circumstances that render a criminal trial fundamentally unfair. We should instead apply the well-developed and more precise rules we have incorporated into RAP 2.5, which we adopted for cases exactly like this. RAP 2.5 inexorably points to the conclusion that Rene Paumier's conviction must be affirmed. This conclusion is consistent with our prior cases, including State v. Bone-Club, 128 Wn.2d 254, 906 P.2d 325 (1995); In re Personal Restraint of Orange, 152 Wn.2d 795, 100 P.3d 291 (2004); State v. Brightman, 155 Wn.2d 506, 122 P.3d 150 (2005); and State v. Easterling, 157 Wn.2d 167, 137 P.3d 825 (2006) because the error here is different than in those cases. I respectfully dissent.

I. The Public Trial Violation in This Case Is Not a Structural Error

¶ 40 The term "structural error" has an established meaning, and we have already grappled with how to apply it in the context of the public trial right. By labeling the error in this case a structural error, the majority opinion defies that established meaning and sends this court down a hazardous detour we would do better to avoid.

A. Structural Error

¶ 41 A structural error is an error that "`"necessarily render[s] a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence."'" State v. Momah, 167 Wn.2d 140, 149, 217 P.3d 321 (2009) (alteration in original) (quoting Washington v. Recuenco, 548 U.S. 212, 218-19, 126 S.Ct. 2546, 165 L.Ed.2d 466 (2006) (quoting Neder v. United States, 527 U.S. 1, 9, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999))), cert. denied, ___ U.S. ___, 131 S.Ct. 160, 178 L.Ed.2d 40 (2010). Structural errors "`infect the entire trial process'" and deprive the defendant of "`basic protections,'" without which "`no criminal punishment may be regarded as fundamentally fair.'" Neder, 527 U.S. at 8-9, 119 S.Ct. 1827 (quoting Brecht v. Abrahamson, 507 U.S. 619, 630, 113 S.Ct. 1710, 123 L.Ed.2d 353 (1993) and Rose v. Clark, 478 U.S. 570, 577, 106 S.Ct. 3101, 92 L.Ed.2d 460 (1986)).

¶ 42 The remedy for structural error is automatic reversal and remand for a new trial. Id. This remedy is truly automatic because, unlike most constitutional errors, structural errors are not subject to harmless error review. Id.

¶ 43 Structural errors are rare and encompass only the most egregious constitutional violations. There is a "`strong presumption'" that errors are not structural, id. at 8, 119 S.Ct. 1827 (quoting Rose, 478 U.S. at 579, 106 S.Ct. 3101), and structural errors comprise a "`very limited class of cases.'" Id. (quoting Johnson v. United States, 520 U.S. 461, 468, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997)). Examples include complete denial of counsel, a biased trial judge, racial discrimination in the selection of a grand jury, denial of right to self-representation, and a defective reasonable-doubt instruction. Id.; see also State v. Vreen, 143 Wn.2d 923, 930, 26 P.3d 236 (2001) (denial of peremptory challenge is structural error). In Washington, we have been hesitant to classify errors as structural. See, e.g., In re Pers. Restraint of Benn, 134 Wn.2d 868, 921, 952 P.2d 116 (1998) (rejecting argument that violation of the right to be present is a structural error).

¶ 44 We have already grappled with how to apply structural error principles in the context of the public trial right. We have found that it is a structural error for a judge to close a courtroom for a significant portion of a criminal trial without conducting a Bone-Club analysis. We made this determination in Bone-Club, Orange, Brightman, and Easterling. See Momah, 167 Wash.2d at 150-51, 217 P.3d 321. In each of those cases, we concluded that the closure error made the trial less fair, and prejudice resulting from the error was so clear that a new trial was required. See id. For example, in Easterling we ordered a new trial because the court excluded the defendant from a portion of his own trial, during which his codefendant struck a deal with the State to testify against him. 157 Wash.2d at 172-73, 181, 137 P.3d 825. In Orange, we ordered a new trial because the trial judge excluded the defendant's family from most of voir dire even after defense counsel specifically requested the family be allowed to attend. The closure prevented the family from "`contribut[ing] their knowledge or insight to the jury selection'" and prevented venirepersons from seeing "`interested individuals.'" 152 Wash.2d at 812, 100 P.3d 291 (emphasis omitted) (quoting Watters v. State, 328 Md. 38, 48, 612 A.2d 1288 (1992)).

¶ 45 However, more recently, we held in Momah that not every public trial violation is a structural error. 167 Wash.2d at 150-51, 217 P.3d 321. In Momah, we listed several criteria for determining when a public trial error is structural and when it is not: (1) whether the trial court closed the courtroom based on interests other than the defendant's or to safeguard the defendant's constitutional rights (such as the right to a fair trial); (2) whether the closure impacted the fairness of the defendant's proceedings; (3) whether the defendant was consulted or given the opportunity to object, and whether the defendant assented to or actively participated in the closure; and finally (4) whether the record suggests that the court considered the defendant's right to a public trial when it closed the courtroom. Id. at 151-52, 217 P.3d 321.

¶ 46 Our holding in Momah is consistent with United States Supreme Court precedent. Despite what the majority implies, our Supreme Court has never held that any public trial violation, no matter how small, is a structural error. See Waller v. Georgia, 467 U.S. 39, 104 S.Ct. 2210, 81 L.Ed.2d 31 (1984); Presley v. Georgia, 558 U.S. 209, 130 S.Ct. 721, 175 L.Ed.2d 675 (2010). Indeed, it would be preposterous to conclude that any time a category of errors has been deemed structural, every single error within that category must also be structural. The Second Circuit Court of Appeals explained this exact point in Gibbons v. Savage, 555 F.3d 112, 119-20 (2d Cir.2009) (citations omitted):

Even if public trial violations constitute a category of errors susceptible to structural error analysis, the Supreme Court has never said categorically that there can be no nonstructural public trial errors. And indeed, the Supreme Court is unlikely to do so given its hesitance to classify errors as structural, Neder, 527 U.S. at 8-9, 119 S.Ct. 1827. The approach we adopted in Momah, 167 Wash.2d at 150-51, 217 P.3d 321, is fully consistent with United States Supreme Court precedent, and we should have no hesitation about applying it here.

B. This Is Not a Case of Structural Error

¶ 47 Turning to the specific question presented by this case, we have never held that partial in-chambers voir dire without a Bone-Club analysis is a structural error. We have considered this question in two cases: Momah, 167 Wn.2d 140, 217 P.3d 321, and State v. Strode, 167 Wn.2d 222, 217 P.3d 310 (2009) (plurality opinion).

¶ 48 In Momah, we found that this error was not structural. 167 Wash.2d at 156, 217 P.3d 321. There, the trial court questioned several prospective jurors in chambers because there was a danger that the jury pool would be tainted by prior knowledge of pretrial publicity. On appeal, we held that there was no structural error, relying on certain key facts that distinguished Momah from other public trial cases, namely, that the defendant affirmatively assented to closure, the trial judge consulted with the defendant about the closure, and the trial judge's express purpose in closing the courtroom was to safeguard the defendant's right to a fair trial. Id. at 151-52, 217 P.3d 321.

¶ 49 In Strode, we also did not hold that the closure error was structural even though the facts that distinguished Momah from an ordinary public trial case were absent. Strode was a split decision consisting of a four-vote plurality authored by Justice Alexander,2 a two-vote concurrence authored by Justice Fairhurst,3 and a dissent authored by Justice Charles Johnson.4 The plurality opinion labeled in-chambers voir dire without a Bone-Club analysis as structural error. However, neither of the other two opinions mentioned structural error at all, nor did they discuss harmless error review. Thus, even in Strode, there were only four votes for structural error, and a majority of this court did not find that the error was structural.

¶ 50 To determine if the error in this case is structural, we must ask whether the error "`"necessarily render[s] a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence."'" Momah, 167 Wash.2d at 149, 217 P.3d 321 (alteration in original) (quoting Recuenco, 548 U.S. at 218-19, 126 S.Ct. 2546 (quoting Neder, 527 U.S. at 19, 119 S.Ct. 1827)).

¶ 51 The improper in-chambers voir dire that occurred here did not constitute structural error because it did not render the trial unfair, nor did it convert an otherwise sound trial into an unreliable vehicle for determining guilt or innocence. An error like this fails to meet the high standard for structural error and does not belong in the same class of errors as complete denial of counsel, a biased trial judge, or racial discrimination in the selection of a grand jury. Certainly, the closure here violates the public trial right. And in some instances, closures of this kind may warrant reversal. But the closure here does not rise to the level of a structural error that warrants automatic reversal.

¶ 52 We must begin our structural error analysis with a straightforward inquiry into whether improper in-camera voir dire renders a criminal trial fundamentally unfair. This requires analyzing what impact, if any, in-camera voir dire may have had on the fairness of jury selection. I fail to see how interviewing jurors in chambers had any adverse impact on these proceedings. The defendant still had the opportunity to question jurors and challenge them for cause or peremptorily. This process occurred on the record and in the presence of counsel, the judge, and the defendant. Furthermore, voir dire is extensively governed by statute, see chapter 2.36 RCW, jurors take an oath to tell the truth, State v. Tharp, 42 Wn.2d 494, 499, 256 P.2d 482 (1953), and we presume jurors follow the judge's instructions, State v. Johnson, 124 Wn.2d 57, 77, 873 P.2d 514 (1994). These procedural safeguards are more than enough to ensure that jury selection is fair even if a small portion of it occurs in chambers.

¶ 53 If anything, in-chambers voir dire protects the defendant's right to a fair and unbiased trial. Empirical studies have shown that prospective jurors often do not reveal sensitive information if required to do so in open court. See Paula L. Hannaford, Safeguarding Juror Privacy: A New Framework for Court Policies and Procedures, 85 Judicature 18, 23 (2001). Interviewing certain jurors in-chambers encourages a fair trial by eliciting this information and allowing counsel to root out potential bias and prejudice. This is true even where there has been no Bone-Club analysis prior to closure. Questioning jurors in chambers on sensitive topics simply does not render a trial fundamentally unfair in the same way as, for example, complete denial of counsel or a biased trial judge.

¶ 54 Further, the criteria set forth in Momah, 167 Wash.2d at 150-51, 217 P.3d 321, weigh against finding structural error. The criteria are (1) the interests on which closure was based, (2) whether the closure impacted the fairness of the proceedings, (3) whether the defendant objected or assented to the closure, and (4) whether the court considered the defendant's right to a public trial. Id. The first factor suggests the error is not structural. The closure here appears to have been based on the defendant's fair trial right: by encouraging jurors to be more forthcoming about sensitive topics in-chambers, counsel can better eliminate bias and prejudice and ensure a fair trial. Turning to the second factor, the closure here appeared to have no negative impact on the fairness of the proceedings. Unlike in Easterling or Orange, there is no readily detectable prejudice, nor indeed any hint of adverse impact at all. As to the third factor, Paumier did not object to the closure and appeared to go along with it, although there is no evidence of affirmative assent in this record as there was in Momah, 167 Wn.2d 140, 217 P.3d 321. This factor is inconclusive at best but, if anything, indicates the error is not structural. Last, the fourth factor counsels in favor of finding that the error is structural: there appears to be no evidence in the record that the judge considered Paumier's right to a public trial before closing the courtroom. Taken as a whole, this is not a structural error under the criteria set forth in Momah. Id.

¶ 55 Finally, on a practical level, the majority opinion creates a disturbing win-win for the defendant. The majority would allow defense counsel to lie in the weeds, silently consent to private questioning (and reap the benefits of increased candor), while secretly nursing a public trial issue that would virtually guarantee success on appeal.5 This would allow any defense counsel who notices a public trial error like this one to remain quiet and gamble on a jury verdict knowing that the public trial issue will allow a do-over once it is raised on appeal.

¶ 56 I fail to see how the partial chambers voir dire in this case rendered Paumier's trial fundamentally unfair or made it an unreliable vehicle for determining guilt or innocence. Accordingly, I would hold that there is no structural error here.

II. When Error Is Not Structural and the Defendant Does Not Object, RAP 2.5 Is a Procedural Bar to Appeal

¶ 57 A proper determination that the error here is not structural requires abandoning the majority's conclusion that RAP 2.5 does not bar this appeal.

¶ 58 Paumier asserts his public trial claim for the first time on appeal. He did not object to in-chambers voir dire at the time of the closure or at any time during the trial.

¶ 59 It is a fundamental principle of appellate litigation that a party may not assert on appeal a claim that was not first raised at trial. Yakus v. United States, 321 U.S. 414, 444, 64 S.Ct. 660, 88 L.Ed. 834 (1944); State v. Davis, 41 Wn.2d 535, 250 P.2d 548 (1953). This rule is grounded in notions of fundamental fairness and judicial economy. See 2A Karl B. Tegland, Washington Practice: Rules Practice RAP 2.5(1), at 192 (6th ed.2004); Smith v. Shannon, 100 Wn.2d 26, 37, 666 P.2d 351 (1983). A trial court should be given the opportunity to respond to and correct mistakes at the time they are made to avoid unnecessary retrials and appeals.

¶ 60 In Washington, this principle is enshrined in RAP 2.5, which states that an appellate court need not review errors raised for the first time on appeal. There is an exception for any "manifest error affecting a constitutional right." RAP 2.5(a)(3). If an error is constitutional in nature, it can be reviewed for the first time on appeal only if it is "manifest," meaning it "had practical and identifiable consequences in the trial of the case" and can survive harmless error review. State v. O'Hara, 167 Wn.2d 91, 98-100, 217 P.3d 756 (2009). In other words, a defendant who does not object must show actual prejudice resulting from the error. Id. Ordinarily, constitutional errors are presumed prejudicial and the burden is on the State to show the error is harmless beyond a reasonable doubt. State v. Guloy, 104 Wn.2d 412, 425, 705 P.2d 1182 (1985). But where the defendant fails to preserve a constitutional issue by objecting, the burden shifts under the clear parameters of RAP 2.5 and the defendant must affirmatively show prejudice. O'Hara, 167 Wash.2d at 98-100, 217 P.3d 756.

¶ 61 It is wholly appropriate to apply RAP 2.5 to public trial errors. In Waller, the Supreme Court noted that state procedural bars apply in full force where the right to a public trial has been violated. Waller, 467 U.S. at 42 n. 2, 104 S.Ct. 2210. In that case, one of the defendants, Cole, did not object to closure at trial. The Supreme Court remanded his case so that the state court could determine "whether Cole is procedurally barred from seeking relief as a matter of state law." Id. Moreover, the Supreme Court has held that the federal plain error rule, Fed.R.Crim.P. 52(b) (which is similar to our RAP 2.5), applies to structural errors. Johnson v. United States, 520 U.S. 461, 466, 117 S.Ct. 1544, 137 L.Ed.2d 718 (1997). There is no basis in federal law, nor in RAP 2.5 itself, nor in our case law,6 for not applying RAP 2.5 to public trial violations.

¶ 62 In the past, RAP 2.5 has not been a major feature of our public trial cases because where error is structural, our RAP 2.5 analysis is straightforward. See Easterling, 157 Wash.2d at 173 n. 2, 137 P.3d 825. In our previous cases, we have nearly always held that the closure error was structural and have also presumed prejudice even where there was no contemporaneous objection. We did so in Bone-Club, Orange, Brightman, and Easterling. If an error is labeled structural and presumed prejudicial, like in these cases, it will always be a "manifest error affecting a constitutional right"; in other words, RAP 2.5 will apply, but it will always be satisfied because prejudice has been presumed and structural errors defy harmless error analysis. See Easterling, 157 Wash.2d at 173 n. 2, 137 P.3d 825. Moreover, it makes sense to presume prejudice despite the lack of objection when an error is structural because by the time we have decided an error is structural, we have already determined that it is of such an egregious nature that it has rendered the underlying trial unfair and deprived the defendant of "`basic protections'" without which "no criminal punishment may be regarded as fundamentally fair." Neder, 527 U.S. at 8-9, 119 S.Ct. 1827.

¶ 63 However, in my view, this case is different from our previous cases because the closure error here is not structural. Where a public trial error is not structural, we must conduct a more thorough analysis under RAP 2.5.

¶ 64 It also does not make sense to presume prejudice in a case like this where the error is not structural and the defendant did not contemporaneously object. This is so for four reasons. First, it does not make sense to presume prejudice where, had the trial judge simply performed a Bone-Club analysis, there is every reason to believe the closure would have occurred in exactly the same manner. It is hard to imagine how not doing a Bone-Club analysis prejudiced the defendant. This is particularly pertinent in light of the fact that the trial court likely would have performed a Bone-Club analysis had the defendant simply objected in a timely manner. Second, prejudice is unlikely to result from in-chambers voir dire because the statutory schemes that govern voir dire and juries, such as chapter 2.36 RCW, provide ample protection to prevent prejudice. There are extensive procedures in place that give the parties opportunity to examine jurors and evaluate whether they are objective and can follow the law. See, e.g., Tharp, 42 Wn.2d 494, 256 P.2d 482 (holding that prospective jurors must take an oath before voir dire begins). Further, we presume jurors will follow the instructions given to them by the court. Johnson, 124 Wash.2d at 77, 873 P.2d 514. To presume prejudice in a case like this is tantamount to presuming that at least one of the jurors questioned in chambers concealed facts relevant to that juror's ability to follow the law and be fair, which conflicts with our presumption that jurors follow the court's instructions. Third, if there is any prejudice resulting from the in-chambers voir dire, it is prejudice to the public's right to observe proceedings in open court, not prejudice to Paumier. It is not at all clear that a defendant can assert the public's right to open courts, let alone rely on prejudice to the public's right in order to satisfy RAP 2.5. See Strode, 167 Wash.2d at 236, 217 P.3d 310 (Fairhurst, J., concurring). Finally, we do not need to presume prejudice given that in-chambers voir dire was done on the record and, having reviewed the transcript, we fail to detect any hint of prejudice. Given all of this, it simply does not make sense to presume prejudice from partial in-camera voir dire where the defendant did not object at trial.

¶ 65 Where a closure error like this one is not structural and the defendant did not object at trial, RAP 2.5 is a procedural bar to appeal. I would hold that before we will hear a claim of nonstructural public trial error not objected to below, a criminal defendant must satisfy RAP 2.5 by showing that the closure error had practical and identifiable consequences in the trial of their case.

III. Paumier Should Not Be Awarded a New Trial

¶ 66 Applying these principles to this case, I would hold that Paumier is not entitled to a new trial.7 Since the public trial error here was not structural and Paumier did not object, we must conduct a thorough RAP 2.5 analysis. This means we will review the issue only if it is "manifest," meaning it had "practical and identifiable consequences in the trial of the case" and can survive harmless error review. O'Hara, 167 Wash.2d at 98-100, 217 P.3d 756. Paumier has not made this showing, nor does the record suggest that any such showing can be made. Paumier gives us no reason to believe that the in-chambers questioning of several jurors on sensitive topics had any practical and identifiable consequences in the trial of his case. He is not entitled to relief, so I respectfully dissent.

IV. Conclusion

¶ 67 Everyone accused of a crime deserves a fair trial, but no one is entitled to a perfect trial. The trial in this case was by all indications a fair and just vehicle for determining Paumier's guilt or innocence. And while it is true that the trial had a constitutional defect (failure to conduct a Bone-Club hearing), there has been no showing whatsoever that this defect impacted the fairness of the trial in any way. I would affirm Paumier's conviction.

WE CONCUR: CHARLES W. JOHNSON and JAMES M. JOHNSON, Justices.