STATE OF ARIZONA v. VICTOR ALFONZO LUCAS (2025)

OPINION

¶1 Picking a jury is more art than science. But trial courts must take certain steps when doing so. They must “conduct voir dire orally,” “ask any necessary questions about the prospective jurors’ qualifications to serve,” and allow counsel to question prospective jurors as a group. Ariz. R. Crim. P. 18.5(d)–(f). Here, the superior court used only a written jury questionnaire and individual questioning to pick the jury. The court did not verbally question the prospective panel as a group or allow counsel to do so. By not doing so, the court erred, which the State concedes. And because we lack confidence the errors did not impact the verdict, we vacate and remand.

FACTS AND PROCEDURAL HISTORY

¶2 On November 13, 2022, after dropping a patient off, two emergency medical technicians (“EMTs”) drove back to their station. When near their station, the EMT in the passenger seat noticed a car one lane over. He recognized the driver, Victor Lucas (“Lucas”), as a cashier at a convenience store nearby their station. He told the other EMT that “the guy from the gas station” was in the next car over. Lucas waved; they waved back.

¶3 What happened next “surprised” and “utterly terrified” the EMTs. Lucas, with his window down, pulled out a gun, pointed it at the EMTs, and fired four shots over their ambulance. Lucas then pulled into the convenience store and the EMTs pulled into a nearby parking lot. The EMTs called their supervisor, who contacted dispatch to send the police.

¶4 The ambulance did not have any bullet holes, and no one was harmed. Officers located a gun in Lucas's car matching a description the EMTs gave; officers also found fired cartridges.

¶5 The State charged Lucas with three crimes: aggravated assault, disorderly conduct, and unlawfully discharging a firearm. Lucas pled not guilty, and the case went to trial.

¶6 On the morning trial started, the court had prospective jurors complete a written questionnaire at the courthouse. Lucas claims the ten selected trial jurors took 15.5 minutes on average to do so.

¶7 The questionnaire asked about 45 questions, with most providing space for the jurors to supplement their answers. After asking for background information, the questionnaire asked whether the jurors knew anyone involved in the case, had read about the case, or had been involved in a similar case. It also attempted to confirm the jurors had no bias. For example, it asked whether they ever had contact with the Peoria Police Department or the Maricopa County Attorney's Office and whether they, their family members, or their close friends were crime victims. And the questionnaire tried to confirm that the prospective jurors would follow the court's instructions.

¶8 The questionnaire also asked about various principles the jurors would have to apply if chosen. For example, it asked two questions about the presumption of innocence. Question 35 asked:

The law requires the State to prove Mr. Lucas guilty beyond a reasonable doubt. The law also presumes that Mr. Lucas is innocent until proven guilty. Do you believe that just because Mr. Lucas has been charged with a crime that Mr. Lucas must have done something wrong or else he would not have been charged?

Question 39 asked, “Do you believe Mr. Lucas is guilty simply because he has been charged with an offense?”

¶9 Prospective Juror 17 answered “Yes” to Question 35, adding, “I believe if he had a fire arm [sic] and intent to hurt someone that is wrong.” She also answered “Yes” to Question 39, adding, “I believe if he was intending to hurt someone that could be considered guilty.”

¶10 Prospective Juror 25 also answered “Yes” to Question 35, explaining, “I think if he used or threatened anyone with a deadly weapon he is guilty.” She also answered “Yes” to Question 39, explaining, “I think if he was charged for using or threatening someone with a deadly weapon he is guilty.”

¶11 After reviewing the responses, the court used them to excuse jurors for cause and hardship. When the court asked counsel about Prospective Juror 17, the State responded, “We don't have anything for 17 for cause or hardship, Your Honor.” Defense counsel responded, “Neither do we.” So the court kept Prospective Juror 17.

¶12 But when the court asked about Prospective Juror 25, the State moved to strike her based on “the presumption of innocence question.” It explained, “they said that I think if he used or threatened anyone with a deadly weapon, he is guilty. It sounds like they've already prejudged this case based off the questionnaire.” When defense counsel agreed, the court excused Prospective Juror 25.

¶13 The court used the questionnaire to excuse, with both parties’ consent, eighteen prospective jurors for cause and six for hardship. After doing so, fourteen prospective jurors remained.

¶14 Near the end of selection, defense counsel said he was “under the impression that I have a chance to speak to the remaining jurors as a group.” The court responded, “That wasn't our plan. We generally won't bring them in and ․ address them as a group.” Instead, the parties could question the remaining jurors “individually and [then] typically go off their questionnaires.” Defense counsel clarified he “would like to voir dire every remaining person on the panel.” The court denied that request.

¶15 The State suggested defense counsel “make a record as to what he wanted to talk to the jurors about specifically so that we can have that issue preserved for appellate purposes[.]” Defense counsel agreed to “make a proffer.” He explained that, if given a chance, he would ask “the jurors today to indulge me in a hypothetical” and then pose this scenario:

[I]magine, what I like to call old-school burglary, common law burglary. The breaking and entering into the house of another with the intent to commit a felony in the nighttime. And I would walk through the jury and say it was on high-definition video the person broke in and entered the house, and it wasn't their house, and they wanted to steal a TV. But I make it clear to the panel that this would have happened twelve o'clock high noon. And I would ask the jury, knowing that the State only proved four out of the five parts of the crime, who would cut them a break and who would still vote to convict.

In past experiences, when I pose this to the juries, a fair number -- I would estimate the majority would raise their hand, and to me that would -- sort of that hypothetical helps clarify that they -- the State has to prove and meet their burden beyond a reasonable doubt as to each element of the crime. Um, I then sort of did my own self-rehabilitation.

Defense counsel again stated he wanted “to engage in a broader discussion with everybody.” The court did not verbally question the panel or allow counsel to do so. Instead, it responded, “we'll go ahead as I indicated.”

¶16 The court let the parties individually question two prospective jurors, and it excused one based on that questioning (leaving 13 prospective jurors). The parties did not individually question Prospective Juror 17.

¶17 The court then selected the ten remaining jurors with the lowest numbers. Using that process, the court seated Prospective Juror 17 on the jury, designating her Trial Juror 5.

¶18 After a four-day trial, the jury found Lucas guilty on all three counts. The court sentenced him to serve six years for aggravated assault, two years for disorderly conduct, and two years for unlawful discharge of a firearm, all to run concurrently.

¶19 Lucas timely appealed. We have jurisdiction. A.R.S. § 13-4033(A)(1).

DISCUSSION

¶20 Lucas argues the court violated Arizona Rule of Criminal Procedure 18.5 by not allowing the parties to question the prospective jurors as a group after striking some based on the questionnaire. Lucas argues that error was structural and requires reversal even without prejudice. But even if not structural, Lucas argues the State has not shown the error was harmless. The State concedes error. But it asks us to affirm because, in its view, Lucas has not shown that the chosen jury was unfair or partial.

I.

¶21 We determine first whether the court erred. Lucas argues it did by refusing to allow him to question the remaining jurors as a group. The State concedes error, but that concession does not bind us. See State v. Solis, 236 Ariz. 242, 249 ¶ 23 (App. 2014). We independently decide whether the court erred; we conclude it did.

A.

¶22 “A criminal defendant has a right to a fair and impartial jury.” State v. Montoya, 258 Ariz. 128, 155 ¶ 74 (2024) (citing Ariz. Const. art. 2, §§ 23, 24; U.S. Const. amends. XI, XIV). To implement that right, “[t]he court ․ must excuse a prospective juror ․ if there is a reasonable ground to believe” they “cannot render a fair and impartial verdict.” Ariz. R. Crim. P. 18.4(b). “A juror is not impartial and should be excused for cause when the juror indicates a predisposition for or against a party or witness.” State v. Bingham, 176 Ariz. 146, 147 (App. 1993).

¶23 Jury selection is “the primary means by which a court” enforces “a defendant's right to be tried by” an impartial jury. Gomez v. United States, 490 U.S. 858, 873 (1989). As used here, “voir dire” (meaning “to speak the truth”) describes the process of questioning prospective jurors to determine whether they are qualified to serve. In other words, voir dire's primary purpose “is to identify those who should be excluded from the jury.” State v. Wooten, 193 Ariz. 357, 363 ¶ 28 (App. 1998).

¶24 Verbal voir dire is critical—it lets the trial judge evaluate the jurors’ demeanor as well as words. See Gomez, 490 U.S. at 874-75 (during voir dire, the court “must scrutinize not only spoken words but also gestures and attitudes of all participants to ensure the jury's impartiality”); State v. Colorado, 256 Ariz. 97, 102 ¶ 22 (App. 2024) (“If the juror's demeanor, conduct, or other factors give the court a reason to disbelieve his or her personal assurances of serving fairly and impartially, the court should strike the juror for cause.”). What is more, a prospective juror's verbal answers often prompt other jurors to raise their hands and share their views. Finally, group voir dire allows prospective jurors to indicate whether they know any other member of the prospective panel. These benefits are lost without verbal questioning.

B.

¶25 Arizona Rule of Criminal Procedure 18.5 establishes the procedures for voir dire in criminal cases. Recognizing the importance of verbal voir dire, Rule 18.5 requires the court to “conduct voir dire orally” by calling a subset of the prospective jurors to the jury box or examining “all members of the panel[.]” Ariz. R. Crim. P. 18.5(d). “The court must then ask any necessary questions about the prospective jurors’ qualifications to serve in the case.” Ariz. R. Crim. P. 18.5(e). Verbal “examination of the prospective jurors” must be “thorough,” and the court must “control” the examination. Ariz. R. Crim. P. 18.5(f). During verbal voir dire, the court must “elicit some detail” about problematic written or verbal responses from prospective jurors “to determine the absence or presence of any prejudice[.]” State v. Rose, 121 Ariz. 131, 139–40 (1978). These requirements remain despite our supreme court eliminating peremptory strikes. See Colorado, 256 Ariz. at 101 ¶ 18.

¶26 The court, though, does not have a monopoly on voir dire. Instead, “[u]pon request, the court must allow the parties sufficient time, with other reasonable limitations, to conduct a further oral examination of the prospective jurors.” Ariz. R. Crim. P. 18.5(f); State v. Shone, 190 Ariz. 113, 115 (App. 1997). “When feasible, the court should permit liberal and comprehensive examination by the parties[.]” Ariz. R. Crim. P. 18.5 (2022 cmt. to R. 18.5(f)). True, the court can “manage voir dire, including to preclude improper, excessive, or abusive questioning.” Ariz. R. Crim. P. 18.5(f). But, “[u]nder existing Arizona law,” the court cannot stop counsel from questioning prospective jurors as a group. See State v. Anderson, 197 Ariz. 314, 320-21 (2000).

C.

¶27 Courts cannot substitute a written questionnaire for verbal questioning. As our supreme court explained, “Rule 18.5 cannot rationally be read to permit the trial judge to use written questionnaires” to “dispense with the thorough oral voir dire the rule requires the judge to make and to allow counsel.” Anderson, 197 Ariz. at 321. The superior court did not “conduct voir dire orally” or “conduct a thorough oral examination of the prospective jurors[.]” Ariz. R. Crim. P. 18.5(d), (f). It conducted no group examination. And the court did not allow Lucas “to conduct a further oral examination of the prospective jurors.” Ariz. R Crim. P. 18.5(f). The court erred twice: once in not conducting its own voir dire and again in not allowing the parties to do so.

¶28 This is not to say the court cannot use written questionnaires to winnow out prospective jurors. To the contrary, the court can dismiss jurors based on questionnaire responses, particularly if neither party objects nor seeks further questioning. See State v. Jones, 197 Ariz. 290, 303 (2000); Anderson, 197 Ariz. at 324 ¶ 24. So the court did not err by working with counsel to narrow the jury pool to a subset of prospective jurors to whom neither side objected. But after doing so, the court had to “ask any necessary questions about the prospective jurors’ qualifications to serve in the case,” Ariz. R. Crim. P. 18.5(e), elicit some detail about any problematic questionnaire responses, Rose, 121 Ariz. at 139–40, and allow the parties to examine the prospective jurors as a group, Ariz. R. Crim. P. 18.5(f).

II.

¶29 Because the superior court erred, we next decide whether to vacate Lucas's convictions.

A.

¶30 Ordinarily, “[f]or objected-to trial error, we conduct harmless error review, which ‘places the burden on the state to prove beyond a reasonable doubt that the error did not contribute to or affect the verdict or sentence.’ ” State v. Strong, 258 Ariz. 184, 200 ¶ 45 (2024) (quoting State v. Henderson, 210 Ariz. 561, 567 ¶ 18 (2005)). “The inquiry ․ is not whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered, but whether the guilty verdict actually rendered in this trial was surely unattributable to the error.” State v. Bible, 175 Ariz. 549, 588 (1993) (quoting Sullivan v. Louisiana, 508 U.S. 275, 279 (1993)). “We must be confident beyond a reasonable doubt that the error had no influence on the jury's judgment.” Id.

¶31 The State instead urges us to apply a modified harmless error standard—one requiring Lucas to show prejudice. And it argues Lucas has made no showing “that the impaneled jurors were unable to render a fair verdict.”

¶32 The State is correct that, in State v. Moody, our supreme court said “[w]e will not disturb the trial court's selection of the jury in the absence of a showing that a jury of fair and impartial jurors was not chosen.” 208 Ariz. 424, 451 ¶ 93 (2004) (citations omitted). But that standard pre-dated Moody. It first appeared in State v. Tison, 129 Ariz. 546, 551 (1981). There, the defendant contended “the trial court erred in failing to conduct individual voir dire.” Id. At that time, Rule 18.5 provided that “if good cause appears, the court may permit counsel to examine an individual juror.” Id. (emphasis added). Our supreme court rejected the defendant's claim because the superior court offered to let defense counsel question individual jurors; he just chose not to. Id. The court also said “[w]e will not disturb the trial court's selection of the jury in the absence of a showing that a jury of fair and impartial jurors was not chosen.” Id. With that, the Tison standard was born.

¶33 The Tison standard next appeared in State v. Walden, 183 Ariz. 595, 607 (1995). The defendant in Walden argued the trial court violated Rule 18.5 by not asking “potential jurors each of the questions he proposed[.]” Id. The supreme court disagreed, concluding “[t]he court conducted extensive voir dire and asked many questions designed to uncover any potential bias.” Id. at 608. The court, quoting Tison, also said a defendant must show “that a jury of fair and impartial jurors was not chosen.” See id. at 607 (quoting 129 Ariz. 546, 551 (1981)).

¶34 The Tison standard appeared next in Moody, a death-penalty appeal. Moody, 208 Ariz. at 435 ¶ 1. Moody argued “the trial court erred both by refusing to administer his proposed jury questionnaire and by giving a factually incorrect reason for doing so.” Id. at 451 ¶ 93. Rejecting that argument, the supreme court explained that the trial court had discretion whether to use jury questionnaires. Id. The court then said, “We will not disturb the trial court's selection of the jury in the absence of a showing that a jury of fair and impartial jurors was not chosen.” Id.

¶35 There is a common thread running through Tison, Walden, and Moody. In each, the court exercised discretion when controlling the scope and content of voir dire. In Moody, the court exercised discretion about whether to use jury questionnaires. 208 Ariz. at 451 ¶ 93. In Walden, the court exercised discretion about what questions to ask prospective jurors. 183 Ariz. at 608. And, in Tison, the court exercised discretion about whether to question individual jurors. 129 Ariz. at 551.

¶36 That common characteristic—discretion—also appears in subsequent opinions applying the Tison standard. See State v. Escalante-Orozco, 241 Ariz. 254, 271 ¶ 33 (2017) (applying Tison to an argument “that the trial court erred by imposing a five-minute time limit for counsel's questions to individual prospective jurors”); State v. Sanders, 245 Ariz. 113, 126 ¶¶ 46-47 (2018) (same); State v. Bush, 244 Ariz. 575, 582–83 ¶¶ 19-20 (2018) (applying Tison to an argument about the use and content of jury questionnaires); State v. Riley, 248 Ariz. 154, 173 ¶¶ 43-44 (2020) (applying Tison to an argument that “the trial court erred by failing to question, sua sponte” two jurors).

¶37 The bottom line: the Tison standard applies when the court exercises discretion to control voir dire. But that key characteristic is missing. The superior court did not exercise discretion. It failed to conduct mandated voir dire and refused to allow the parties to do so. See Ariz. R. Crim. P. 18.5(d), (f); supra ¶ 27. Without the court exercising discretion, we do not apply the Tison standard.

B.

¶38 The errors here are more like those in Shone. There, we applied a traditional harmless error review. The State admits the similarity—it says, “As in Shone, the trial court erred when it denied counsel's request to voir dire the panel.” Despite that admission, the State asks us to treat the errors here differently than we did in Shone. We will not deviate.

¶39 In Shone, the court violated Rule 18.5 when it “denied defense counsel's two direct requests to voir dire the panel.” 190 Ariz. at 115. We explained that, under Rule 18.5, “a party requesting voir dire must be permitted a reasonable time to conduct an examination, subject to certain limitations.” Id. The superior court violated that mandate, and so we applied traditional harmless error review. Id. (“Here, a harmless error analysis answers the question presented.”). Because the errors here are more akin to those in Shone, that opinion guides us, not Tison or Moody.

C.

¶40 Defense counsel repeatedly asked to question the prospective jurors as a group. With those requests, counsel notified the superior court that it needed to comply with Rule 18.5. See Shone, 190 Ariz. at 116 (“[A]n objection made during voir dire ․ preserve[s] the error, and the error is not waived simply by agreeing to pass the panel.”). And, because Lucas objected, the State must show the court's errors did not affect the verdict. Strong, 258 Ariz. at 200 ¶ 45.

¶41 The State has not done so. In fact, the State makes no effort to do so; it relies only on the Tison standard. The State's failure to argue harmlessness is, alone, sufficient to vacate the judgment.

¶42 But, had the State tried to show harmlessness, we still would lack confidence that the errors were harmless. This is “because the record discloses that the trial court seated a juror who may have been biased.” Shone, 190 Ariz. at 116 (emphasis added). In responding to the questionnaire, Prospective Juror 17 indicated “that Mr. Lucas must have done something wrong or else he would not have been charged[.]” And she indicated she “believe[d] Mr. Lucas is guilty simply because he has been charged with an offense[.]” Those responses are like responses from Prospective Juror 25, who all agreed should be dismissed for prejudging Lucas's guilt.

¶43 We acknowledge Prospective Juror 17 elaborated on her “yes” answers. But, even accounting for that additional information, the most we can say is that maybe, as the State claims, Prospective Juror 17 would find Lucas guilty only if the State proved certain facts. Or maybe, as the State theorizes, she “did not fully understand some element of the initial question.” Or maybe she could have been rehabilitated with additional questioning. See Bingham, 176 Ariz. at 147. We will never know. All we know is that her “yes” answers raised serious doubt about her “ability to be fair and impartial[.]” Shone, 190 Ariz. at 116; see also State v. Sexton, 163 Ariz. 301, 302-03 (App. 1989). And that doubt remains because the court did not conduct its own voir dire or allow the parties to do so. “We therefore cannot say that the guilty verdict actually returned in this trial was surely unattributable to the error[s].” Shone, 190 Ariz. at 116.

CONCLUSION

¶44 We vacate Lucas's convictions and remand for further proceedings. Because we do not find the court's errors harmless, we need not (and do not) decide whether they were structural.

¶45 I join in the Majority's holding reversing Victor Lucas's convictions and sentences because the jury that convicted him was not selected in the manner prescribed by Arizona Rule of Criminal Procedure (“Rule”) 18.5. I write separately to express my view that failing to conduct voir dire in person as required by Rule 18.5 constitutes structural error, and is “thus subject to automatic reversal on appeal” without a showing of prejudice. See Greer v. United States, 593 U.S. 503, 513 (2021) (citation modified).

¶46 Generally, a defendant challenging his conviction on appeal cannot obtain relief merely by showing that an error occurred at trial. On the contrary, a defendant's conviction may be affirmed despite, for example, the trial court's error in instructing the jury, or in admitting evidence, if the error was harmless. See State v. Moore, 222 Ariz. 1, 14-15, ¶ 69 (2009) (affirming convictions despite error in jury instruction); State v. Davolt, 207 Ariz. 191, 205, ¶¶ 39-43 (2004) (affirming convictions despite error in admission of evidence). Trial error is harmless if the State proves beyond a reasonable doubt that the error did not contribute to or affect the verdict. State v. Strong, 258 Ariz. 184, 200, ¶ 45 (2024) (citation omitted).

¶47 Most trial error is subject to harmless error analysis, and for good reason: a defendant has the right “to a fair trial,” but “not a perfect one,” Rose v. Clark, 478 U.S. 570, 579 (1986) (citations omitted), and “the virtually inevitable presence” of error does not necessarily vitiate the “underlying fairness of the trial,” Delaware v. Van Arsdall, 475 U.S. 673, 681 (1986) (citations omitted). Error that does not impair the fundamental fairness of the trial gives no cause to doubt the integrity of the outcome. Greer, 593 U.S. at 513 (“Discrete defects in the criminal process ․ do not necessarily render a criminal trial fundamentally unfair or an unreliable vehicle for determining guilt or innocence.” (citation modified)). “Where a reviewing court can find that the record developed at trial establishes guilt beyond a reasonable doubt, the interest in fairness has been satisfied and the judgment should be affirmed.” Rose, 478 U.S. at 579.

¶48 Harmless error analysis “presupposes a trial ․ at which the defendant, represented by counsel, may present evidence and argument before an impartial judge and jury.” Id. at 578 (footnote omitted) (citation omitted). Because a trial without those basic protections is neither fair nor a reliable vehicle for ascertaining the truth, any verdict reached at such a trial is inherently suspect. See State v. Ring, 204 Ariz. 534, 552, ¶ 45 (2003) (noting that “structural errors deprive defendants of basic protections without which a criminal trial cannot reliably serve its function as a vehicle for determination of guilt or innocence and no criminal punishment may be regarded as fundamentally fair.” (citation modified)). Moreover, an error that “undermines the structural integrity of the criminal tribunal” leaves the reviewing court unable to “evaluate the resulting harm.” Vasquez v. Hillery, 474 U.S. 254, 263-64 (1986) (footnote omitted). The harm, for example, from the wrongful denial of a defendant's request to substitute one retained attorney for another cannot be measured because “[i]t is impossible to know what different choices [the defendant's preferred counsel] would have made, and then to quantify the impact of those different choices on the outcome of the proceedings.” United States v. Gonzalez-Lopez, 548 U.S. 140, 150 (2006). Prejudice from structural error need not be shown, in other words, because it generally cannot be shown; a reviewing court cannot possibly determine whether the verdict would have been the same if the trial had not been infused with structural error. State v. Garcia-Contreras, 191 Ariz. 144, 148, ¶ 15 (1998) (“Because structural error is impossible to quantify, it defies analysis by harmless-error standards.” (citation modified)).

¶49 The denial of the right to an impartial jury is, indisputably, structural error. See Gomez v. United States, 490 U.S. 858, 876 (1989) (“Among those basic fair trial rights that can never be treated as harmless is a defendant's right to an impartial adjudicator, be it judge or jury.” (citation modified)). And because, in my view, the in-person voir dire process mandated by Rule 18.5 is essential to the selection of an impartial jury, the failure to conduct voir dire in person is structural error, too. Cf. Garcia-Contreras, 191 Ariz. at 148-49, ¶¶ 17-20 (holding that error in “involuntarily exclud[ing]” defendant “from the entire jury selection proceeding[,]” which denied him the right “to be brought face to face with the jurors at the time challenges are made[,]” was “not amenable to harmless error review.” (citation modified)).

¶50 “Jury selection is the primary means by which a court may enforce a defendant's right to be tried by a jury free from ethnic, racial, or political prejudice, or predisposition about the defendant's culpability.” Gomez, 490 U.S. at 873 (citations omitted). In-person voir dire safeguards the right to an impartial jury by allowing the judge, the attorneys, and the parties to evaluate each prospective juror's manner and demeanor as well as his or her words. Id. at 874-75 (“To detect prejudices, ․ [t]he court ․ must scrutinize not only spoken words but also gestures and attitudes of all participants to ensure the jury's impartiality.” (citation omitted)); Garcia-Contreras, 191 Ariz. at 148, ¶ 17 (reversing defendant's convictions due to his improper exclusion from courtroom during jury selection, and noting that defendants may perceive “impressions” and “prejudices” from jurors’ “looks and gestures” (citation omitted)). Perhaps equally important, in-person voir dire allows the prospective jurors to see the defendant, the attorneys, and the victim (if any), which may, in turn, expose biases of which the jurors themselves may not have been aware. See, e.g., People v. Mataele, 513 P.3d 190, 210 (Cal. 2022) (affirming dismissal for cause of prospective juror who ”acknowledged” that she might be “biased toward the defense” “after seeing how young the defendants were”); State v. Anderson, 603 So.2d 780, 783 (La. Ct. App. 1992) (affirming dismissal for cause of prospective juror who indicated “she would have a problem passing judgment on the defendant” because, inter alia, “she had seen the defendant numerous times at her husband's store”), sentence set aside 609 So.2d 831 (La. 1992).

¶51 In State v. Anderson, 197 Ariz. 314 (2000), three prospective jurors in a capital case gave written responses to a jury questionnaire indicating strong opposition to the death penalty. Denying defense counsel's request to question them in person, the trial court dismissed all three prospective jurors for cause. Id. at 318, ¶ 5. On appeal, our supreme court held that the dismissal of the prospective jurors, who “were not seen or heard” by the judge or the parties and who had not been given a chance “to clear up the uncertainties arising from their responses to the questionnaire,” created a “defect[ ] in the trial mechanism” that was “not subject to harmless error analysis.” Id. at 323-24, ¶¶ 22, 23 n. 5 (citation modified). Accordingly, the Anderson court concluded, the dismissal of the prospective jurors for cause without allowing defense counsel a chance to conduct oral voir dire was structural error. Id. at 324, ¶ 23.

¶52 In my view, the same is true here. Other than Trial Juror 5, see supra ¶ 16, neither the judge nor the parties personally saw or heard from any of the jurors who decided Lucas's fate before they were empaneled. We will never know what answers any of those jurors might have given in response to oral questioning, or what the judge or the parties may have gleaned from a particular juror's tone or demeanor while answering questions. We will never know, moreover, if a prospective juror's responses to oral questioning may have jogged the memories of others in the room, prompting them to raise their hands to share additional relevant information they had neglected to provide when answering the questionnaire. I would hold, therefore, that the denial of Lucas's request to conduct oral voir dire precluded a meaningful assessment of the prospective jurors’ fitness to serve, and so created a “defect in the trial mechanism” that is “not subject to harmless error analysis,” see id. at 323-24, ¶ 22.

¶53 Written juror questionnaires undoubtedly play a valuable role in the jury selection process by promoting the efficient use of time while encouraging disclosure of candid responses to inquiries of a sensitive nature. But written questionnaires can never replace the in-person voir dire mandated by Rule 18.5. See id. at 321, ¶ 15 (“Rule 18.5 cannot rationally be read to permit the trial judge to use written questionnaires in order to dispense with the thorough oral voir dire the rule requires[.]”). Only by conducting voir dire in the manner required by Rule 18.5 can the court, counsel, and the parties meaningfully assess prospective jurors’ impartiality. The failure to comply with Rule 18.5 undermined the fundamental fairness of the trial here, and therefore constitutes structural error requiring reversal without a determination of prejudice.

CATLETT, Judge:

Presiding Judge Michael S. Catlett delivered the opinion of the Court, in which Judge Daniel J. Kiley and Vice Chief Judge David D. Weinzweig joined. Judge Kiley filed a special concurrence.