(concurring).
I concur with the majority that an objective examination of the facts and circumstances of this case mandates a holding that the Chief Judge of the Hennepin County District Court did not err when she denied the state’s motion, to remove the trial judge because he lacked impartiality. Nevertheless, I write separately for two reasons. First, I disagree with comments the majority makes with respect to judge “shopping.” The majority implies that the state might be trying to use its motion to have the trial judge removed for cause as a way to find another judge who would rule more favorably on the issue of what expert gang testimony should be admitted. Second, I disagree with the majority’s conclusions as to the status of the judge’s pretrial rulings.
In response to the state’s motion to remove the trial judge, Burrell asserted that the state’s real reason for trying to remove the judge is not a genuine concern about the judge’s impartiality, but is instead a desire to get a new judge assigned to the case who would issue more favorable rulings on what expert gang testimony should be allowed. In commenting on this argument, the majority concludes that “[i]n the context of the larger battle over expert gang testimony, it is conceivable that the State is seeking a judge who will rule more favorably as it seeks to challenge DeShay and Lopez-Rios.” DeShay and Lopez-Rios are recent opinions we issued on the admission of expert gang testimony in criminal trials. State v. DeShay, 669 *604N.W.2d 878 (Minn.2003); State v. Lopez-Rios, 669 N.W.2d 603 (Minn.2003).
I do not conclude that the state is judge “shopping” or that it filed its motion to remove the trial judge for cause for the purpose of getting a new judge who might rule more favorably on what expert gang testimony the state could offer. The state asked the judge to remove himself for cause immediately after learning that Bur-rell was considering waiving his right to a jury trial. Here, it is important to maintain a focus on the sequence of events. The state made its motion to remove the judge in March 2007 based upon comments the judge allegedly made in January 2007. The record shows that in January there was no indication that Burrell was considering waiving his right to a jury trial. Therefore, this new fact that the judge was going to be the trier of fact was a significant change in the status of the case. Because of this change, it is understandable that the state would have heightened concerns about comments it believed the judge made in January about the strength of the state’s case. Legitimate concern about whether the judge had already made up his mind about the case provides a proper basis to move to remove the judge because of impartiality. This concern, while not sufficient to warrant removal under the facts and circumstances of this particular case, has merit and the state’s motion surely was not frivolous.
I am also aware of the particular circumstances of the party making this motion. Unlike a private litigant who may easily think that he will never again appear before a specific trial judge in a matter, or even in Hennepin County District Court again, the state will have many more cases before this judge and his colleagues. The state appears on a daily basis before several judges in Hennepin County District Court. As a result, I believe that the state would be cautious in bringing a motion that, in essence, asserts that a particular judge will not be impartial, and that it would not bring such a motion lightly at this stage of a criminal proceeding as a means to disqualify a judge with whom it was merely unhappy. Therefore, I disagree with the majority’s implication that the state’s motion in this case is about judge shopping. I conclude that the reasons are more substantive and profound; and are based on legitimate concerns about obtaining a fair trial. Thus, while I agree that those concerns do not rise to a level that warrants removal of this judge, we should acknowledge that the state’s reasons for bringing its motion are legitimate and should not be dismissed as merely judge shopping.
I also disagree with the majority’s position with respect to the trial judge’s pretrial rulings. In discussing the state’s motives for bringing the motion to remove the judge, the majority notes that, “[ejven if the State were successful in replacing him, * * * the current judge’s pretrial rulings would stand unless the State could demonstrate ‘extraordinary circumstances.’ ” Unlike the majority, I do not believe that the judge’s prior rulings are binding on this judge or a new judge. These rulings regarding what a gang expert can base his or her opinion on and what specific expert gang testimony could be admitted in this case were made in October and November 2005. I conclude that a new judge, as well as the current judge, is free to reconsider the prior rulings relating to expert gang testimony because the law in this area is more fully developed since those rulings were made in late 2005. Further, I conclude that the judge’s prior rulings were premature and based on an inadequate record.
Not counting our decision reversing Burrell’s convictions, we addressed the ad*605mission of gang-expert testimony in six recent cases, three of which were decided after we issued State v. Burrell, 697 N.W.2d 579 (Minn.2005) and after the trial judge made his rulings in this case. See State v. Mahkuk, 736 N.W.2d 675 (Minn.2007); State v. Martinez, 725 N.W.2d 733 (Minn.2007); State v. Jackson, 714 N.W.2d 681 (Minn.2006); State v. Blanche, 696 N.W.2d 351 (Minn.2005); State v. DeShay, 669 N.W.2d 878 (Minn.2003); State v. Lopez-Rios, 669 N.W.2d 603 (Minn.2003).
In these cases, we cautioned against the use of gang-expert testimony in criminal cases. Blanche, 696 N.W.2d at 374; Lopez-Rios, 669 N.W.2d at 612. We said, “gang-expert testimony should be admitted only if it is helpful to the jury in making the specific factual determinations that jurors are required to make.” Blanche, 696 N.W.2d at 373. In Mahkuk, we said, “[t]o be admissible, gang expert testimony must add precision or depth to the jury’s ability to reach conclusions about matters that are not within its experience.” 736 N.W.2d at 686 (internal quotation marks omitted). While we.have not stated that a gang expert could not base his or her opinion on hearsay, we have stated that “the state should not be permitted to launder inadmissible hearsay evidence, turning it into admissible evidence by the simple expedient of passing it through the conduit of purportedly ‘expert opinion.’ ” DeShay, 669 N.W.2d at 886.
Our comments in DeShay about the impropriety of using an expert as a means to have inadmissible hearsay evidence admitted are consistent with the Minnesota Rules of Evidence regarding expert testimony generally. See Minn. R. Evid. 703 (“If a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data [upon which the expert relied in forming his or her opinion] need not be admissible in evidence.”). As the Committee Comment to Rule 703 notes, this rule “is aimed at permitting experts to base opinions on reliable hearsay and other facts that might not be admissible under these rules of evidence.” Minn. R. Evid. 703 comm. cmt.-1989. However, “[a]lthough an expert may rely on inadmissible facts or data in forming an opinion, the inadmissible foundation should not be admitted into evidence simply because it forms the basis for an expert opinion.” Id. Rule 703(b) requires that “[u]nderlying expert data * * * be independently admissible in order to be received upon direct examination.”
In some of our cases, we have concluded that the admission of gang-expert testimony was proper, while in others we have held it was in error, albeit harmless error. For example, in Mahkuk we held that in a trial for first-degree premeditated murder and first-degree premeditated murder for the benefit of a gang, it was within the “permissible scope of gang expert testimony” for an expert to testify about the defendant’s membership in a specific gang, the rivalry between the defendant’s gang and another gang, and the specific reasons why the victims may have been targeted in this case. 736 N.W.2d at 686. On the other hand, in Blanche, a trial for first-degree murder, conspiring to commit first-degree murder, and crime committed for the benefit of a gang, we held that it was error to allow a gang expert to testify about the conduct of gang members in general, including that gang members retaliate against each other by shooting at each other, because we concluded that under the facts in Blanche, “there was a risk that the jury would improperly use this evidence to conclude that Blanche was the shooter simply because he is a member of a gang.” 696 N.W.2d at 374. We also said that it was error to allow a gang expert to testify that gang members in *606general do not falsely accuse their own gang members of crime, but we held that the erroneous admission of this testimony did not warrant a new trial. Id. In DeShay, we held that gang-expert testimony about crimes that gangs in general commit or about drive-by shootings and criminal sexual conduct committed by the gang that defendant was a member of were “neither helpful nor relevant to the fact issues before the jury,” where the defendant was charged with conspiracy to commit a controlled-substance crime. 669 N.W.2d at 886. Nevertheless, we held that the admission of the expert’s testimony was harmless error. Id. at 888. Finally, in Lopez-Rios, we held that it was error to admit much of a gang expert’s testimony, but that the error was harmless. 669 N.W.2d at 612-13.
We have issued several opinions in the area of gang-expert testimony that were not directly referenced in State v. Burrell when we instructed the district court, on remand, to determine what gang-expert testimony should be admitted. See 697 N.W.2d 579, 601-02 (Minn.2005). It does not appear that these more recent opinions were considered by the trial judge when he first made his evidentiary rulings on this issue. When our holdings in these newer cases and our earlier cases are properly considered, it becomes evident that whether gang-expert testimony is admissible in a particular case often depends on the trial record and the trial testimony of lay witnesses. For example, in Mah-lcuk, we noted that “[w]hen feasible,” it is best to prove the “for-the-benefit-of-a-gang element” through the testimony of lay witnesses who have firsthand knowledge of gang related issues. 736 N.W.2d at 686. In Blanche, we said that to be admissible, “the gang expert’s testimony must have been helpful to the jury in making factual determinations” and that it was error to admit the testimony in that case, in part, because it was “largely duplicative” of testimony given by fact witnesses. 696 N.W.2d at 374; see also Lopez-Rios, 669 N.W.2d at 612. In DeShay, we noted that in order to be admissible, a gang-expert’s testimony must be helpful to the jury, but that “[g]ang expert testimony in this non-complex drug conspiracy, to the extent relevant, was largely duplicative, giving little assistance to the jury in evaluating the evidence.” 669 N.W.2d at 886.
In its opinion dismissing the trial judge’s certified questions regarding the admissibility of gang-expert testimony, the court of appeals aptly noted that it could not answer several of the certified questions because the answers were largely “contingent upon testimony that may or may not be presented, or may be ruled inadmissible, at trial.” State v. Burrell, No. A06-149, 2006 WL 2807166, at *4 (Minn.App. Oct.3, 2006), rev. denied (Minn. Dec. 20, 2006). I agree. I conclude that it was premature for the trial judge to issue rulings on what, if any, expert testimony should be allowed in this case until there was a more developed trial record, and it became clear what lay witness testimony the state had actually presented on the homicide charge in general and the for-the-benefit-of-a-gang element in particular. I conclude that this need for a more fully developed record, combined with our additional case law on the issue of gang-expert testimony, case law which has developed since the time of the trial judge’s initial rulings, unquestionably points to a conclusion that the trial judge not only may but must reevaluate his evidentiary rulings, and if necessary, change those rulings. To conclude otherwise would unduly elevate form over substance.