¶ 50. (concurring). I agree with the majority that this is not a true "anonymous jury" case. Nevertheless, the majority treats it as an anonymous jury case and concludes, inevitably, that the error is harmless.
¶ 51. I write separately because I disagree with the majority's application of the harmless error rule in this case. Even if this were an anonymous jury, and *511even if harmless error applies to anonymous jury cases, the majority incorrectly sets forth the circumstances in which the rule can be applied. Additionally, I write to emphasize that measures that shield juror information not only implicate the defendant's rights, but also contradict the presumption of openness that defines the American judicial system.
( — I
¶ 52. By treating this numbers jury as an anonymous jury, the majority dodges the threshold certified issue presented to us by the court of appeals. The certified issue inquires: "Whether voir dire by number constitutes an anonymous jury under State v. Britt, 203 Wis. 2d 25, 553 N.W.2d 528 (Ct. App. 1996), when the restriction applies only to using the jurors' names in open court but does not prohibit inquiry into any other identifying information."
¶ 53. The State argues, and I agree, that the jury empanelled in this case is not an anonymous jury. No information was withheld from the parties and no significant restriction was placed on questions asked at voir dire. The only restriction here is how jurors are addressed.
¶ 54. As the court of appeals notes in its certification, the restriction on Tucker is less than the restrictions placed on the parties in Britt. In that case, the circuit court restricted references on the record to the jurors' names, addresses, and employers. Here the only restriction curtailed the use of the jurors' names on the record. The parties were advised to use numbers only, rather than names, in referring to the jurors.
¶ 55. In Britt, there were indications of witness intimidation and harassment, factors that the court considered relevant to jury safety. Here, no such factors *512existed. Instead, the circuit court expressed a policy of conducting voir dire by number without making any specific finding that anonymity was necessary to protect the jury.
¶ 56. Not only are the factors that existed in Britt absent from this case, but the record also fails to reveal any other factors that would justify empanelling an anonymous jury. The majority sets forth the Thomas test to determine whether an anonymous jury is warranted. United States v. Thomas, 757 F.2d 1359 (2d Cir. 1985). This two-prong test provides that an anonymous jury is warranted only if there is a "strong reason" to believe that the jury needs protection and if the court takes "reasonable precaution[s]" to minimize the impact of anonymity on jurors' views of the defendant. Majority op., ¶ 13. It then notes that this approach has been widely adopted by both federal and state courts, including the Seventh Circuit.
¶ 57. What the majority fails to acknowledge, however, is that the Seventh Circuit and others set a standard requiring something more than just assertions or beliefs in order to justify anonymity. The standard requires actual evidence indicating that intimidation is likely:
In short, that the defendants had the ability and incentive to threaten jurors, without additional evidence indicating that they were likely to act on that ability and incentive, was not enough to justify the unusual step of juror anonymity.
United States v. Mansoori, 304 F.3d 635, 651 (7th Cir. 2002). See, e.g., United States v. Darden, 70 F.3d 1507, 1532-1533 (8th Cir. 1995); United States v. Ross, 33 F.3d 1507, 1520-1521 (11th Cir. 1994); United States v. Vario, 943 F.2d 236, 240-241 (2nd Cir. 1991).
*513¶ 58. If this were an anonymous jury, then I agree with the majority that the circuit court was required to make an individualized determination of the need for such a jury. I part ways with the majority because it fails to draw a distinction between a numbers jury as here, and an anonymous jury The result of this failure is that in concluding that harmless error applies in this "numbers only" situation, it incorrectly extends harmless error as a remedy in all truly anonymous jury cases. Such a widespread extension is contrary to precedent.
A
¶ 59. The majority opinion correctly notes the import of the right to an impartial jury Yet, after acknowledging the import of the right, it diminishes the right by subjecting it to the harmless error rule. In failing to consider whether there may be structural error, the majority commits its own error.
¶ 60. The majority departs from well-established precedent which recognizes that certain constitutional deprivations are structural and therefore defy harmless error analysis. Chapman v. California, 386 U.S. 18, 23 (1967) (describing the right to counsel and the right to an impartial judge as examples of constitutional rights so basic to a fair trial that their infraction can never be treated as harmless error); Arizona v. Fulminante, 499 U.S. 279, 310 (1991) (describing a structural error as a "defect affecting the framework within which the trial proceeds, rather than simply an error in the trial process itself' and citing the following cases as involving constitutional errors that are not subject to the harmless error rule: Vasquez v. Hillery, 474 U.S. 254 (1986) (unlawful exclusion of members of the defendant's race from a grand jury); McKaskle v. Wiggins, 465 U.S. 168 (1984) (the right to self-*514representation at trial); Waller v. Georgia, 467 U.S. 39 (1984) (right to public trial)).
¶ 61. Similar to the rights described above, the improper use of an anonymous jury is not simply an error in the trial process. Rather, the improper use of an anonymous jury effectively denies an accused of the right to a trial by an impartial jury, thereby tainting the entire framework within which the trial process occurs. As such, it is a structural error that is not subject to harmless error review.
B
¶ 62. I recognize, however, that federal courts of appeals that have addressed the issue have not foreclosed the possibility that in very limited circumstances the harmless error rule could be applied to anonymous jury violations. The federal courts of appeals have imposed a much stricter standard in applying the harmless error rule than the majority sets forth here.
¶ 63. In United States v. Sanchez, 74 F.3d 562 (5th Cir. 1996), the defendant did not know the jurors' names and addresses or their spouses' or employers' names. The Fifth Circuit focused on the factors set forth in United States v. Krout, 66 F.3d 1420 (5th Cir. 1995), that justify an anonymous jury Those factors include: "(1) the defendants' involvement in organized crime; (2) the defendants' participation in a group with the capacity to harm jurors; (3) the defendants' past attempts to interfere with the judicial process or witnesses; (4) the potential that, if convicted, the defendants will suffer a lengthy incarceration and substantial monetary penalties; and (5) extensive publicity that could enhance the possibility that jurors' names would become public and expose them to intimidation and harassment." Sanchez, 74 F.3d at 564.
*515¶ 64. The court noted that virtually none of the factors listed in Krout justified the empanelling of an anonymous jury in the case before it. No one could demonstrate that the defendant was involved in organized crime or participated in a group that would attempt to harm jurors. Although the defendant was a renegade policeman, there was no evidence that he had attempted to interfere with the judicial process. Finally, the court noted that there was no indication that the jurors would be subjected to the type of publicity that would result in intimidation and harassment.
¶ 65. In finding error, the Sanchez court concluded that more than mere speculation or inferences of potential risk was needed to justify an anonymous jury. There must be actual evidence that anonymity was warranted before the defendant's right to be tried before a panel of identified jurors was required to be sacrificed. Id. at 565.
¶ 66. The government argued that even if the Krout criteria were lacking, the error was harmless because extensive voir dire was conducted that enabled the defendant to pick an unbiased jury. The Sanchez court disagreed. It concluded that the harmless error rule could not apply in anonymous jury cases except in very limited circumstances. It stated:
Unless the type of circumstances listed in Krout exist, where the defendant has essentially compromised his right, he should receive a verdict, not from anonymous decisionmakers, but from people he can name as responsible for their actions. In closer cases on the merits of requiring anonymity, there might be room for a harmless error analysis, but this is not such a case. The conviction must be reversed and remanded for retrial.
Id.
*516¶ 67. A recent Seventh Circuit case, United States v. Mansoori, 304 F.3d 635 (7th Cir. 2002), discussed Sanchez and endorsed its conclusion that an erroneous decision to empanel an anonymous jury is not harmless where almost none of the pertinent factors supported the decision. It noted, however, that the Sanchez court left open the possibility that harmless error analysis might be appropriate in limited circumstances. It concluded that the case before it constituted such limited circumstances.
¶ 68. The Mansoori court described the case as a narcotics conspiracy that embraced "a large-scale, gang-related operation with ready access to firearms." Id. at 651. It noted that the defendants and their unindicted co-conspirators used weapons and violence in furtherance of the conspiracy, that the defendants faced very long prison terms, and that there had been some pretrial publicity. The court observed that although the circumstances were not, by themselves, sufficient to justify empanelling an anonymous jury, the record did support a basis for the concern of juror safety. That basis in the record, together with a three and one-half day "searching and thorough" voir dire which protected the defendants' right to an unbiased jury, justified a harmless error analysis. Id. at 652.
¶ 69. Here we note that none of the pertinent factors supports a decision to empanel an anonymous jury. There is no inference of organized crime or suggestion that the defendant is involved with a group that has the capacity to harm jurors. Likewise, there is no evidence to suggest that the defendant previously attempted to interfere with the judicial process or that this case received any publicity whatsoever. It was unlikely that the defendant would suffer a lengthy *517incarceration or substantial monetary penalties. Ip-deed, as a result of her conviction, the defendant was placed on probation.
¶ 70. Regardless of whether use of the harmless error rule is precluded because an improper anonymous jury constitutes structural error, or whether it is precluded here because none of the pertinent factors supports the use of an anonymous jury, one thing is clear. The majority's general application of harmless error as a remedy in an anonymous jury case is error, and it is hardly harmless.
¶ 71. Ultimately, I concur in the mandate of the majority because I conclude that there was no error. The defendant had access to all juror information, including their names. Given that the only limitation here was how the jurors were addressed and that the judge advised the jury that it was her practice to use numbers rather than names, I do not find that the defendant's rights were violated.
II.
¶ 72. Measures to shield juror information not only implicate the defendant's rights, but also contradict the presumption of openness that defines the American judicial system. The selection of jurors has always presumptively been a public process.
¶ 73. Since the sixteenth century jurors have been selected in public. Press-Enterprise Co. v. Superior Court, 464 U.S. 501, 507 (1984). It is not surprising that trials in colonial America adopted the presumptive openness of the jury selection process that developed in England. Id. at 508. This openness "enhances both the basic fairness of the criminal trial and the appearance of fairness so essential to public confidence in the system." Id. at 508.
*518¶ 74. As the majority noted, the empanelling of an anonymous jury is a relatively recent phenomenon. Majority op., ¶ 12. Although courts have recognized that the use of an anonymous jury may be appropriate in limited circumstances, they have been understandably narrow in defining those circumstances. "The use of an anonymous jury will remain a device of last resort. . . ." Krout, 66 F.3d at 1427. It is considered to be a "drastic measure." Sanchez, 74 F.3d at 564.
¶ 75. A few years ago this court unanimously rejected a petition for an administrative rule governing juror confidentiality. The petition provided that jurors be referred to only by number and that no personal juror identifying information could be elicited during voir dire. It allowed that a party may, after the trial, petition the court for access to personal juror identifying information for purposes of developing a motion for a new trial.
¶ 76. The breadth of such a proposal and its effect on our tradition of public trials were apparent to many who appeared in opposition to the proposal at the public hearing. The State Bar of Wisconsin was one of the groups that appeared in opposition to the petition.1 It cautioned that an anonymous jury should be used only *519in "an extremely rare circumstance." (Statement of State Bar of Wisconsin filed Nov. 16, 1998, at 8.)
¶ 77. A trial is a public event and a public trial lies at the foundation of our legal tradition. The public trial is rooted in the "principle that justice cannot survive behind walls of silence." Sheppard v. Maxwell, 384 U.S. 333, 349 (1966).
¶ 78. The majority's failure to distinguish between a numbers jury and an anonymous jury serves to dilute the jurisprudence on anonymous juries. It compounds this problem by applying an across-the-board "no harm, no foul" analysis of the harmless error rule thus serving to make the use of anonymous juries more commonplace.
¶ 79. The result of the majority's decision will stand in stark contrast to the federal courts of appeals decisions that have emphasized that use of an anonymous jury is the "last resort" and "a drastic measure."
¶ 80. It stands in stark contrast to the position of many who appeared at our public administrative hearing opposing the proposed rule, noting that if anonymous juries are to be used, they should be used only in extremely rare circumstances.
¶ 81. But, most importantly, the result of the majority's decision lies in stark contrast to the presumption of openness that defines our Anglo-American judicial tradition. The promise of a public trial by an impartial jury is a cornerstone of that tradition. The most common rationale for an anonymous jury is the protection of the jurors. A jury that sits in fear may not fill the expectation of impartiality. Yet, the use of an anonymous jury is a double-edged sword — it can give a sense of security or it can breed and feed fear. When fear is the result, then jury anonymity is a solution that exacerbates the problem it was intended to solve.
Petition No. 98-08, In the Matter of the Amendment of Supreme Court Rules: (Proposed) SCR 73.04: Juror Confidentiality, Public Hearing November 17, 1998; Order Denying Petition January 19, 1999. Groups that appeared or registered in opposition to the proposed rule include the Wisconsin District Attorneys Association, the Judges of Milwaukee County, the Milwaukee Bar Association, the Wisconsin Association of Trial Lawyers, the Wisconsin Newspaper Association, the Wisconsin Broadcasters Association, and the Freedom of Information Council.