¶ 100. {concurring). I concur with the decision and the mandate of the court. I agree with the court's interpretation and analysis of the Confrontation Clause under the facts of this case. While I disagree with the majority's statement of the harmless error test, I agree with its application of the harmless error analysis in this case. I also conclude that the State met its burden beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained. I write separately because of the majority's decision not to discuss whether the prior testimony in this case was admissible under a recognized hearsay exception. I also write separately to discuss this court's misstatement of the harmless error rule.
HH
¶ 101. While the confrontation issue may indeed be easy to resolve in this case, the hearsay question is equally easy to resolve. The evidence in this case was not admissible under either the former testimony exception or the residual exception to the hearsay rule.
¶ 102. Wisconsin Stat. § 908.045(1) excludes from the hearsay rule, provided the declarant is unavailable as a witness:
Testimony given as a witness at another hearing of the same or a different proceeding, or in a deposition taken in compliance with law in the course of another proceeding, at the instance of or against a party with an opportunity to develop the testimony by direct, cross-, or redirect examination, with motive and interest similar to those of the party against whom now offered.
¶ 103. David Sullivan testified at the trial of co-defendant Robert Jones. Sullivan testified that he had given a gun to Hale, not Jones, shortly before the *627murders. During Jones' trial, Sullivan expressed fear in testifying and guilt in providing the murder weapon to Hale. Later, when subpoenaed for the Hale trial, Sullivan simply disappeared. Yet the trial court admitted Sullivan's testimony during the Jones' trial against Hale in Hale's trial, under the theory that Jones' cross-examination in his trial could be admitted against Hale because Jones possessed a motive and interest similar to Hale.
¶ 104. It is important to note here that Jones' cross-examination of Sullivan took the gun out of Jones' possession and placed it in Hale's possession. Jones was minimizing his involvement at the expense of Hale through his cross-examination. Jones identified Hale as the shooter in a statement to the police prior to the trial. Hale, according to the trial court, made statements implicating Jones. Indeed, Jones and Hale had antagonistic defenses. Compare State v. Nutley, 28 Wis. 2d 527, 543, 129 N.W.2d 155 (1964).
¶ 105. A motion to sever was filed by Hale. The State conceded that severance was required. The court agreed, ordering separate trials for Hale and Jones. Under these facts, and given this procedural posture, there can be no misstating the interests and motives of these two individuals as being similar. Jones was clearly putting the blame on Hale and used Sullivan's testimony to further that purpose. Jones and Hale were pointing fingers toward each other. The trial court should not have admitted that testimony as former testimony under Wis. Stat. § 908.045(1), as their motives and interests were certainly not similar.
¶ 106. I also disagree with the court of appeals that Sullivan's testimony was admissible under the "residual hearsay exception" of Wis. Stat. § 908.045(6). Section 908.045(6) excludes from the hearsay rule, *628provided the declarant is unavailable as a witness, "[a] statement not specifically covered by any of the foregoing exceptions but having comparable circumstantial guarantees of trustworthiness."
¶ 107. This exception "is for the novel or unanticipated category of hearsay that does not fall under one of the named categories, but which is as reliable as one of those categories." State v. Stevens, 171 Wis. 2d 106, 120, 490 N.W.2d 753 (Ct. App. 1992). Accordingly, "[i]t is intended that the residual hearsay exception rule will be used very rarely, and only in exceptional circumstances." Id. The State has failed to identify how this case is novel or exceptional. Nor has the State clearly established comparable circumstantial guarantees of trustworthiness, particularly in light of the antagonistic defenses between Jones and Hale. The shooter was not positively identified at the scene, and the gun can be traced back to Sullivan. Thus, Sullivan had ample reason to place the gun in someone else's hands. Consequently, there was no sufficient showing of particularized guarantees of trustworthiness in this case.
I — I HH
¶ 108. In ¶¶ 60-61 of its opinion, the majority properly states the Chapman1 harmless error test. In footnote 9, however, the majority notes that while adhering to the Chapman test in recent years, the United States Supreme Court and this court have articulated alternative wording for the test. While the majority is correct that there are alternative wordings for the harmless error test, the different wordings make all the difference. The alternative wordings, although *629all falling under the umbrella of harmless error, are fundamentally different tests that depend on the nature of the error and are not interchangeable.
¶ 109. The Court in Chapman made it clear that "before a federal constitutional error can be held harmless, the court must be able to declare a belief that it was harmless beyond a reasonable doubt." Chapman v. California, 386 U.S. 18, 24 (1967), reh'g denied, 386 U.S. 987 (1967). An error is harmless if the beneficiary of the error proves "beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained." Id. This is the basic test that is applied to most constitutional violations that occur during a criminal trial, but not all of them.
¶ 110. Certain types of errors are "structural" in nature, and are considered so fundamental and pervasive that they require reversal without regard to the facts or circumstances of the particular case. Id. at 23 n. 8; Delaware v. Van Arsdall, 475 U.S. 673, 681 (1986); Neder v. United States, 527 U.S. 1, 8 (1999). These errors include a complete denial of counsel,2 a biased trial judge,3 racial discrimination in the selection of a grand jury,4 denial of self-representation at trial,5 denial of a public trial,6 and a defective reasonable-doubt instruction.7 Neder, 527 U.S. at 8.
¶ 111. Most constitutional errors are analyzed using the basic harmless error test set forth in Chap*630man. Whether such an error is harmless depends on a number of factors, all accessibly to reviewing courts. Van Arsdall, 475 U.S. at 684 (confrontation violation). Compare, e.g., Fahy v. Connecticut, 375 U.S. 85 (1963) (illegally seized evidence); Gilbert v. California, 388 U.S. 263 (1967) (illegally seized evidence); Satterwhite v. Texas, 486 U.S. 249 (1988) (right to consult with counsel); Arizona v. Fulminante, 499 U.S. 279 (1991) (involuntary confessions); and Chapman, 386 U.S. 18 (comments on defendant's silence). For these types of errors, the analysis begins with an evaluation of the nature of the error and the harm it is alleged to have caused in order to determine whether the error did not contribute to the verdict obtained beyond a reasonable doubt. State v. Weed, 2003 WI 85, ¶ 30, 263 Wis. 2d 434, 666 N.W.2d 485; State v. Carlson, 2003 WI 40, ¶ 87, 261 Wis. 2d 97, 661 N.W.2d 51 (Sykes, J., dissenting). The appropriate standard is not whether there is sufficient evidence, absent the error, to support the verdict. Weed, 163 Wis. 2d, ¶¶ 28-32. Nor does the defendant have to show "outcome determinative" prejudice in order to state a violation. Van Arsdall, 475 U.S. at 679-80.
¶ 112. Certain types of constitutional errors by their very nature lend themselves to a form of an "outcome determinative" approach. One such error involves ineffective assistance of counsel. Strickland v. Washington, 466 U.S. 668 (1984). The Court in Strickland held that in order to challenge the conviction on the grounds of ineffective assistance of counsel, a defendant would have to show that counsel's performance was deficient and that the deficient performance prejudiced the defense. Id. at 687. In discussing the nature of the error, the Court determined that it is not enough to show that the error had some conceivable outcome of the proceeding, as virtually every act or omission of *631counsel would meet that test. Id. at 693. But the Court refused to adopt a strict "outcome determinative" approach that the defendant would have to show that counsel's deficient conduct more likely than not altered the outcome of the case. Id. Instead, the Court adopted a modified outcome approach, that the defendant must show that there is a reasonable probability that, but for counsel's errors, the result of the proceeding would have been different. Id. at 694. A reasonable probability is a probability sufficient to undermine confidence in the outcome. Id. This test shifts the burden to the defendant and is considerably different than the reasonable possibility standard set forth in Chapman.
¶ 113. Similarly, the United States Supreme Court has, on more than one occasion, adopted a form of "outcome determinative" harmless error standard in matters involving errors in jury instructions. In Neder, the Court examined whether the failure to instruct the jury on an uncontested element of the offense could be harmless error. The Court concluded that where a defendant did not and could not bring forth facts contesting the omitted element, "answering the question whether the jury verdict would have been the same absent the error does not fundamentally undermine the purposes of the jury trial guarantee." Neder, 527 U.S. at 19. See also, Pope v. Illinois, 481 U.S. 497 (1987); Yates v. Evatt, 500 U.S. 391 (1991). The Court explained in Yates that when dealing with presumptions in jury instructions, one cannot look subjectively into the minds of the jurors. Id. at 404-05. A court must approach the inquiry by asking whether the force of the evidence presumably considered by the jury in accordance with the instructions is so overwhelming as to leave it beyond a reasonable doubt that the verdict resting on that evidence would have been the same in *632the absence of the presumption. Id. The Court once again focused on the nature of the error in determining whether the error was harmless.
¶ 114. The proper harmless error test for a confrontation violation was set forth in Van Arsdall. The Court specifically rejected an outcome determinative test for this type of violation, distinguishing its approach in Strickland. Van Arsdall, 466 U.S. at 679-80. As the focus of the Confrontation Clause centers on an individual witness, the focus of the prejudice inquiry must be on the particular witness, not the outcome of the entire trial. Id. at 680. Factors to consider include the importance of the testimony in the prosecution's case, whether the testimony was cumulative, the presence or absence of evidence corroborating or contradicting the testimony of the witness on material points, the extent of cross-examination otherwise permitted, and the overall strength of the prosecution's case. Id. at 684. While the majority here does not cite Van Arsdall, it is nevertheless clear that it essentially applies Van Arsdall. See majority op., ¶ 61.
¶ 115. This court has merged the harmless error analysis for trial constitutional errors into one standard. Compare State v. Dyess, 124 Wis. 2d 525, 540-47, 370 N.W.2d 222 (1984) (concluding that "reasonable probability" means essentially the same thing as "reasonable possibility"); with State v. Harvey, 2002 WI 93, ¶ 46, 254 Wis. 2d 442, 647 N.W.2d 189 (concluding that the Court in Neder applies the harmless error test in the same manner, regardless of the nature of the error complained of). See also, id., ¶¶ 50-52 (Crooks, J., concurring). This construction of the harmless error *633test is flawed.8 The harmless error analysis must depend upon the nature of the error. Some errors can never be harmless, some will have to be evaluated under the Chapman "reasonable possibility" test, and some will have to be viewed in terms of the outcome absent any error.
¶ 116. We should not try to fit a "square peg" into a "round hole." This court will necessarily have to struggle with the application of the harmless error analysis in light of the error. Because the majority in this matter has essentially, but appropriately, adopted the factors set forth in Van Arsdall,9 and because the *634error in this case was harmless beyond a reasonable doubt in light of those factors, I agree with and join the decision rendered here today.
HH HH 1 — I
¶ 117. I join the decision and the mandate of the court because it correctly interprets the Confrontation Clause of the United States Constitution while protecting the face-to-face requirements of the Wisconsin Constitution. While the majority misidentifies the appropriate harmless error test to be applied in a confrontation violation, it correctly applies the harmless error analysis to facts of this case. Let there be no doubt, however, regarding the admissibility of Sullivan's "former testimony" during Hale's trial — that testimony clearly did not fall within a hearsay exception and should not have been admitted into evidence against Hale.
¶ 118. For the foregoing reasons, I respectfully concur.
Chapman v. California, 386 U.S. 18 (1967), reh'g denied, 386 U.S. 987 (1967).
Johnson v. United States, 520 U.S. 461, 468 (1997)(citing Gideon v. Wainwright, 372 U.S. 335 (1963)).
Turney v. Ohio, 273 U.S. 510 (1927).
Vasquez v. Hillery, 474 U.S. 254 (1986).
McKaskle v. Wiggins, 465 U.S. 168 (1984).
Waller v. Georgia, 467 U.S. 39 (1984).
Sullivan v. Louisiana, 508 U.S. 275 (1993).
Some of the confusion may stem from language in Neder, where the Court suggests that the harmless error inquiry for failure to instruct the jury on an uncontested element "must be essentially the same" as the inquiry for the erroneous exclusion of evidence in violation of the Fifth Amendment guarantee against self-incrimination and the Sixth Amendment right to confront witnesses. Neder v. United States, 527 U.S. 1, 18 (1999). One writer suggests that the Court is split 5-4 on whether to apply an "overwhelming evidence" standard as opposed to looking at the effect of the error on the jury. Jeffrey 0. Cooper, Searching for Harmlessness: Method and Madness in the Supreme Court's Harmless Constitutional Error Doctrine, 50 Kan. L. Rev. 309, 324 (2002). While the Court may he divided, and may be moving in the direction of looking at the strength of the evidence in evaluating harmless error, it appears to have used an outcome approach only in cases involving ineffective assistance of counsel or jury instruction errors, the language in Neder notwithstanding.
The Seventh Circuit Court of Appeals apparently has the same difficulty that the majority has in applying harmless error with a confrontation violation. In United States v. Gilbert, _ F.3d _, No. 03-3365-CR (7th Cir. 2004), the court frames the harmless error analysis as "whether it is clear beyond a reasonable doubt that a rational jury would have found Gilbert guilty even absent the admission of Sherese's statement." Id. at_. *634Yet, without saying so, the court also appears to apply a Van Arsdall-type analysis. The court reasoned that the erroneously admitted evidence was the most probative. Id. at_. Absent that evidence, the court could not determine whether the jury would credit other testimony, or whether other evidence, if credited, would be sufficient. Id. The prosecution argued that this was the most and perhaps the only probative evidence that was offered on the issue of possession. Id. The court concluded that, in light of the evidence as a whole, it could not determine that the jury would have returned a guilty verdict absent the error. Id. The focus was clearly on the nature of the error and its impact on the jury