with whom Justice Souter, Justice Ginsburg, and Justice Breyer join, dissenting.
In her dissenting opinion, Justice Ginsburg has cogently explained why the Court’s holding is unsound. I therefore join her opinion without reservation. I add these comments to emphasize the force of her arguments and to explain why I find the Court’s reasoning even more troubling than its holding. The limits of the Court’s holding are evident: Even if the Minnesota Lawyers Professional Responsibility Board (Board) may not sanction a judicial candidate for announcing his views on issues likely to come before him, it may surely advise the electorate that such announcements demonstrate the speaker’s unfitness for judicial office. If the solution to harmful speech must be more speech, so be it. The Court’s reasoning, however, will unfortunately endure beyond the next election cycle. By obscuring the fundamental distinction between campaigns for the judiciary and the political branches, and by failing to recognize the difference between statements made in articles or opinions and those made on the campaign trail, the Court defies any sensible notion of the judicial office and the importance of impartiality in that context.
The Court’s disposition rests on two seriously flawed premises — an inaccurate appraisal of the importance of judicial independence and impartiality, and an assumption that judicial candidates should have the same freedom “‘to express themselves on matters of current public importance’ ” as do all other elected officials. Ante, at 781-782. Elected judges, no less than appointed judges, occupy an office of trust that is fundamentally different from that occupied by policymaking officials. Although the fact that they must stand for election makes their job more difficult than that of the tenured judge, that fact does not lessen their duty to respect essential attributes of the judicial office that have been embedded in Anglo-American law for centuries.
*798There is a critical difference between the work of the judge and the work of other public officials. In a democracy, issues of policy are properly decided by majority vote; it is the business of legislators and executives to be popular. But in litigation, issues of law or fact should not be determined by popular vote; it is the business of judges to be indifferent to unpopularity. Sir Matthew Hale pointedly described this essential attribute of the judicial office in words which have retained their integrity for centuries:
“11. That popular or court applause or distaste have no influence in anything I do, in point of distribution of justice.
“12. Not to be solicitous what men will say or think, so long as I keep myself exactly according to the rule of justice.’ ”1
Consistent with that fundamental attribute of the office, countless judges in countless cases routinely make rulings that are unpopular and surely disliked by at least 50 percent of the litigants who appear before them. It is equally common for them to enforce rules that they think unwise, or that are contrary to their personal predilections. For this reason, opinions that a lawyer may have expressed before becoming a judge, or a judicial candidate, do not disqualify anyone for judicial service because every good judge is fully aware of the distinction between the law and a personal point of view. It is equally clear, however, that such expressions after a lawyer has been nominated to judicial office shed little, if any, light on his capacity for judicial service. Indeed, to the extent that such statements seek to enhance the popularity of the candidate by indicating how he would rule in specific cases if elected, they evidence a lack of fitness for the office.
*799Of course, any judge who faces reeleetion may believe that he retains his office only so long as his decisions are popular. Nevertheless, the elected judge, like the lifetime appointee, does not serve a constituency while holding that office. He has a duty to uphold the law and to follow the dictates of the Constitution. If he is not a judge on the highest court in the State, he has an obligation to follow the precedent of that court, not his personal views or public opinion polls.2 He may make common law, but judged on the merits of individual cases, not as a mandate from the voters.
By recognizing a conflict between the demands of electoral politics and the distinct characteristics of the judiciary, we *800do not have to put States to an all or nothing choice of abandoning judicial elections or having elections in which anything goes. As a practical matter, we cannot know for sure whether an elected judge’s decisions are based on his interpretation of the law or political expediency. In the absence of reliable evidence one way or the other, a State may reasonably presume that elected judges are motivated by the highest aspirations of their office. But we do know that a judicial candidate, who announces his views in the context of a campaign, is effectively telling the electorate: “Vote for me because I believe X, and I will judge cases accordingly.” Once elected, he may feel free to disregard his campaign statements, ante, at 780-781, but that does not change the fact that the judge announced his position on an issue likely to come before him as a reason to vote for him. Minnesota has a compelling interest in sanctioning such statements.
A candidate for judicial office who goes beyond the expression of “general observation about the law ... in order to obtain favorable consideration” of his candidacy, Laird v. Tatum, 409 U. S. 824, 836, n. 5 (1972) (memorandum of Rehnquist, J., on motion for recusal), demonstrates either a lack of impartiality or a lack of understanding of the importance of maintaining public confidence in the impartiality of the judiciary. It is only by failing to recognize the distinction, clearly stated by then-JuSTiCE Rehnquist, between statements made during a campaign or confirmation hearing and those made before announcing one’s candidacy, that the Court is able to conclude: “[Sjince avoiding judicial preconceptions on legal issues is neither possible nor desirable, pretending otherwise by attempting to preserve the ‘appearance’ of that type of impartiality can hardly be a compelling state interest either,” ante, at 778.
Even when “impartiality” is defined in its narrowest sense to embrace only “the lack of bias for or against either party to the proceeding,” ante, at 775, the announce clause serves that interest. Expressions that stress a candidate’s unbro*801ken record of affirming convictions for rape,3 for example, imply a bias in favor of a particular litigant (the prosecutor) and against a class of litigants (defendants in rape cases). Contrary to the Court’s reasoning in its first attempt to define impartiality, ante, at 775-776, an interpretation of the announce clause that prohibits such statements serves the State’s interest in maintaining both the appearance of this form of impartiality and its actuality.
When the Court evaluates the importance of impartiality in its broadest sense, which it describes as “the interest in openmindedness, or at least in the appearance of openmind-edness,” ante, at 778, it concludes that the announce clause is “so woefully underinclusive as to render belief in that purpose a challenge to the credulous,” ante,- at 780. It is underinclusive, in the Court’s view, because campaign statements are an infinitesimal portion of the public commitments to legal positions that candidates make during their professional careers. It is not, however, the number of legal views that a candidate may have formed or discussed in his prior career that is significant. Rather, it is the ability both to reevaluate them in the light of an adversarial presentation, and to apply the governing rule of law even when inconsistent with those views, that characterize judicial openmindedness.
The Court boldly asserts that respondents have failed to carry their burden of demonstrating “that campaign statements are uniquely destructive of openmindedness,” ante, at 781. But the very purpose of most statements prohibited by the announce clause is to convey the message that the candidate’s mind is not open on a particular issue. The lawyer who writes an article advocating harsher penalties for polluters surely does not commit to that position to the same degree as the candidate who says “vote for me because I believe all polluters deserve harsher penalties.” At the *802very least, such statements obscure the appearance of open-mindedness. More importantly, like the reasoning in the Court’s opinion, they create the false impression that the standards for the election of political candidates apply equally to candidates for judicial office.4
The Court seems to have forgotten its prior evaluation of the importance of maintaining public confidence in the “disinterestedness” of the judiciary. Commenting on the danger that participation by judges in a political assignment might erode that public confidence, we wrote: “While the problem of individual bias is usually cured through recusal, no such mechanism can overcome the appearance of institutional partiality that may arise from judiciary involvement in the making of policy. The legitimacy of the Judicial Branch ultimately depends on its reputation for impartiality and nonpartisanship. That reputation may not be borrowed by the political Branches to cloak their work in the neutral colors of judicial action.” Mistretta v. United States, 488 U. S. 361, 407 (1989).
Conversely, the judicial reputation for impartiality and openmindedness is compromised by electioneering that emphasizes the candidate’s personal predilections rather than his qualifications for judicial office. As an elected judge recently noted:
“Informed criticism of court rulings, or of the professional or personal conduct of judges, should play an *803important role in maintaining judicial accountability. However, attacking courts and judges — not because they are wrong on the law or the facts of a case, but because the decision is considered wrong simply as a matter of political judgment — maligns one of the basic tenets of judicial independence — intellectual honesty and dedication to enforcement of the rule of law regardless of popular sentiment. Dedication to the rule of law requires judges to rise above the political moment in making judicial decisions. What is so troubling about criticism of court rulings and individual judges based solely on political disagreement with the outcome is that it evidences a fundamentally misguided belief that the judicial branch should operate and be treated just like another constituency-driven political arm of government. Judges should not have ‘political constituencies.’ Rather, a judge’s fidelity must be to enforcement of the rule of law regardless of perceived popular will.” De Muniz, Politicizing State Judicial Elections: A Threat to Judicial Independence, 38 Willamette L. Rev. 367, 387 (2002).
The disposition of this case on the flawed premise that the criteria for the election to judicial office should mirror the rules applicable to political elections is profoundly misguided. I therefore respectfully dissent.
The Court largely ignores the fact that judicial elections are not limited to races for the highest court in the State. Even if announcing one’s views in the context of a campaign for the State Supreme Court might be permissible, the same statements are surely less appropriate when one is running for an intermediate or trial court judgeship. Such statements not only display a misunderstanding of the judicial role, but also mislead the voters by giving them the false impression that a candidate for the trial court will be able to and should decide cases based on his personal views rather than precedent.
Indeed, the Court’s entire analysis has a hypothetical quality to it that stems, in part, from the fact that no candidate has yet been sanctioned for violating the announce clause. The one complaint filed against petitioner Gregory Wersal for campaign materials during his 1996 election run was dismissed by the Board. App. 16-21. Moreover, when Wersal sought an advisory opinion during his 1998 campaign, the Board could not evaluate his request because he had “not specified what statement [he] would make that may or may not be a view on a disputed, legal or political issue.” Id., at 32. Since Wersal failed to provide examples of statements he wished to make, and because the Board had its own doubts about the constitutionality of the announce clause, it advised Wersal that “unless the speech at issue violates other prohibitions listed in Canon 5 or other portions of the Code of Judicial Conduct, it is our belief that this section is not, as written, constitutionally enforceable.” Ibid. Consequently, the Court is left to decide a question of great constitutional importance in a case in which either the petitioner’s statements were not subject to the prohibition in question, or he neglected to supply any concrete examples of statements he wished to make, and the Board refused to enforce the prohibition because of its own constitutional concerns.
See Buckley v. Illinois Judicial Inquiry Bd., 997 F. 2d 224, 226 (CA7 1993).
Justice Kennedy would go even further and hold that no content-based restriction of a judicial candidate’s speech is permitted under the First Amendment. Ante, at 798 (concurring opinion). While he does not say so explicitly, this extreme position would preclude even Minnesota’s prohibition against “pledges or promises” by a candidate for judicial office. Minn. Code of Judicial Conduct, Canon 5(A)(3)(d)(i) (2002). A candidate could say “vote for me because I promise to never reverse a rape conviction,” and the Board could do nothing to formally sanction that candidate. The unwisdom of this proposal illustrates why the same standards should not apply to speech in campaigns for judicial and legislative office.