concurring:
I concur in the opinion of the court and note that the foundation of the dissent is that the Hawaii Supreme Court “is, as a matter of law, biased against Partington.” I do not find it so easy to reach the dissent’s conclusion in the light of what, in terms of current law and history, has constituted a biased tribunal. The ideal has always been an absolutely impartial judge. In practice, the ideal has been tempered by the qualification put by the old Constitution of Massachusetts which guaranteed a judge “as impartial as the common lot of humanity will admit.” Constitution of Massachusetts, Part 1, Declaration of the Rights of the Inhabitants (1780). What the common lot of humanity will admit has varied.
A leading recent case from this circuit is United States v. Conforte, 624 F.2d 869 (9th Cir.), cert. denied, 449 U.S. 1012, 101 S.Ct. 568, 66 L.Ed.2d 470 (1980). Here a panel consisting of Circuit Judges Kennedy and Tang and District Judge Palmieri considered a criminal appeal by Joseph Con-forte, the operator of a legalized house of prostitution in Nevada. The district judge who had presided at his trial had, as president of the Reno Unit of the American Contract Bridge League, participated in the decision to reject Conforte’s application to join the League because of his prior convictions and proprietorship of the house of prostitution. At a cocktail party in 1973 the judge had also advised against accepting a contribution from Conforte for the University of Nevada football team. The judge had remarked that Conforte ran a house of prostitution and “was not good for Reno.” Id. at 879.
District Judge Warren J. Ferguson heard Conforte’s motion for a new trial on the grounds of the district judge’s bias and found that the district judge had not been disqualified. United States v. Conforte, 457 F.Supp. 641 (D.Nev.1978). Upholding this ruling, Judge Kennedy for the panel stated: “We find no reasonable grounds for questioning the judge’s impartiality because of bias or prejudice_ [T]he negative bias or prejudice of the kind alleged here will disqualify only if it is an attitude or state of mind — [that evinces] an aversion or hostility of a kind or degree that a fair minded person could not entirely set aside *132when judging certain persons or causes.” Conforte, 624 F.2d at 881.
This standard, it is apparent, is such that a previous legal ruling would not constitute disqualifying bias because it is not the kind of aversion or hostility that a fair-minded person would be unable to set aside. Specifically, at a later point in the opinion, the panel held: “A judge’s view on legal issues may not serve as the basis for motions to disqualify.” Id. at 882 (citations omitted).
The history that lies behind this treatment of disqualifying prejudice as a term of art is worth reviewing. The Judiciary Act of 1789 provided that district judges should not review their own decisions on appeal to the circuit courts, but at the same time provided that Supreme Court Justices should regularly act as circuit judges and made no provision for disqualifying them from reviewing their own decisions. In 1790 Chief Justice John Jay composed a letter to be sent by the entire Supreme Court to President Washington on this problem. Jay pointed out that having once decided an issue, a justice would not be
secured against the influence of those predilections for individual opinions, and of those reluctances to relinquish sentiments publicly, though, perhaps, too hastily given, which insensibly and not unfrequently infuse into the minds of the most upright men some degree of partiality for their official and public acts.
Letter from John Jay to James Iredell (Sept. 15, 1790) (enclosing draft), reprinted in 2 G. McRee, Life and Correspondence of James Iredell (1857). Jay also alluded to the case where a justice might disqualify himself from sitting on his own decision but the public would still believe that in the decision of the Supreme Court reviewing his decision, “mutual interest had generated mutual civilities and tendernesses injurious to right.” Id. Chief Judge Jay observed that appellate cases were usually doubtful with “much appearance of reason on both sides”; it would shake the confidence of the losing litigant and the public if the court in reaching such difficult questions appeared to be biased by the pre-judgment made by one or more of its members.
It is unclear whether any other justice signed Chief Justice Jay’s letter. It was never sent. 1 J. Goebel, Jr., History of the Supreme Court of the United States, 556 (1971). The Supreme Court did not hesitate to act as a court of appeals from decisions of circuit courts made by the Justices as circuit judges. When in 1803 the practice was finally challenged by a litigant, the Court held that acquiescence in the practice prevented change. Stuart v. Laird, 1 Cranch 299, 309 (1803). The practice of judging appeals from one’s own decisions continued in the Marshall court. E.g., The Antelope, 23 U.S. (10 Wheat.) 66, 6 L.Ed. 268 (1825).
It could not be much clearer that the Founders and the early Supreme Court did not believe that a judge was biased as a matter of law because he had previously ruled on the question being presented to him. Eventually by statute the practice ceased in 1891. Judiciary Act of March 3, 1891, ch. 517, 26 Stat. 826, § 3. But the problem has scarcely disappeared. It is the regular practice of the circuit courts of appeal to schedule sessions en banc in which panel members, who have previously heard the case, sit as members. The session is called a “rehearing” but the reality is that the reargument before the en banc court is a fresh hearing. Often there is new briefing and always there is new oral argument. There are always judges sitting who have never heard the ease before. At the same time the judges on the en banc panel who have heard the case the first time and decided it are not regarded as biased as a matter of law in sitting to judge this fresh appeal.
One difference exists between these practices and the case Partington presents. At least in the practices mentioned a party has had an opportunity to defend himself before the judge who ruled against him the first time, whereas Partington suffered an adverse ruling in a case where he had no possibility of defending himself. This distinction, however, is not material if the question is one of prejudice on the part of the judge. In the federal practices mentioned and in the Hawaii case the judges at *133one point have made up their minds. In the most literal sense they are prejudiced judges: they have prejudged the case now before them. In the federal practices it was and is assumed that they can unmake their minds and consider the case afresh. Why should this assumption not be the case in Hawaii?
When a litigant was, for the first time, permitted to challenge a federal judge on the grounds of bias, it was on the basis of a statute enacted in 1911. The challenge could only be for “personal bias or prejudice.” Act of March 3, 1911, ch. 231, 36 Stat. 1090, § 21. Interpreting this statute, the Supreme Court said that the party seeking to disqualify the judge must “state facts which tend to show not merely adverse rulings already made, which may be right or wrong, but facts and reasons which tend to show personal bias or prejudice.” Ex Parte American Steel Barrel, Co. and Seaman, 230 U.S. 35, 43-44, 33 S.Ct. 1007, 1009-10, 57 L.Ed. 1379 (1913). Applying the statute to disqualify District Judge Keneshaw Mountain Landis in 1921, the Court observed that the bias “must be based upon something other than rulings in the case.” Berger v. United States, 255 U.S. 22, 31, 41 S.Ct. 230, 232, 65 L.Ed. 481 (1921). In Denis v. Perfect Parts, 142 F.Supp. 263 (D.Mass.1956), Judge Bailey Aldrich had ruled in a previous case that the patent in question was valid. When the same patentee brought a ease against a different defendant, the defendant objected that Judge Aldrich was prejudiced because he had already ruled on the validity of the patent. Judge Aldrich refused to disqualify himself, ruling, “The statute is directed to personal bias, not to previous judicial exposure to the same or similar questions.” Perfect Parts, 142 F.Supp. at 263. The general standard that the bias must be personal and also arise from an extrajudicial source has been stated as the law of this circuit. United States v. Carignan, 600 F.2d 762, 763-64 (9th Cir.1979). A judge’s participation in a “related or prior proceeding” is not sufficient even though he ruled against one of the present parties in that earlier proceeding. Davis v. Fendler, 650 F.2d 1154, 1163 (9th Cir.1981). The dissent appears unwilling to acknowledge that as the law now stands a prior adverse ruling by a judge does not show the judge to be biased.
The point could be raised that the Justices of the Hawaii Supreme Court might be prejudiced because of the publicity attending the reversal in the Clarke murder case. The fact of publicity generated by a judges’s prior ruling was considered at length in the Watergate case where the defendants included John Ehrlichman, Bob Haldeman and John Mitchell. The Court of Appeals for the District of Columbia, sitting en banc, rejected the defendants’ contention that Judge John Sirica had been biased against them because his “public image and reputation” had “become inextricably intertwined with the prosecution of the Watergate matter.” United States v. Haldeman, 559 F.2d 31, 137 (D.C.Cir.1976), cert. denied sub. nom., Ehrlichman v. U.S., 431 U.S. 933, 97 S.Ct. 2641, 53 L.Ed.2d 250 (1977). The court held that this kind of “interest” on the part of the judge was not disqualifying. Interest in order to disqualify must emanate from extrajudicial sources, otherwise “the prohibition would have had a reach far beyond anything Congress could rationally have contemplated.” Id. The court refused to hold that Judge Sirica had been impermissibly biased. Haldeman was favorably cited by Judge Kennedy in Conforte, 624 F.2d at 882.
Sensibilities no doubt are changing on the question of judicial bias. In 1765 Blackstone could write: “... the law will not suppose a possibility of bias or favor in a judge, who is already sworn to administer impartial justice, and whose authority greatly depends upon that presumption and idea.” 3 W. Blackstone, Commentaries on the Laws of England *361. While recusal on the basis of challenge by a litigant had existed in Roman law, canon law, and early English law, it had disappeared by Blackstone’s time, id., and was permitted in federal law only in 1911. Since that turning point there have been other signs that judicial disqualification may be looked at more sharply. For example, Judge Aldrich’s decision in Perfect Parts was criticized by a *134student writer. Note, “Disqualification of Judges for Bias in the Federal Courts,” 79 Harv.L.Rev. 1435, 1451 (1966). In 1974 Congress enacted the statute requiring judges to disqualify themselves if a reasonable person could doubt their impartiality. 28 U.S.C. § 455(a). The practice of the Supreme Court in its first century in relation to circuit decisions now looks strange to us. It may be that the time will come when the participation of panel judges in en banc circuit hearings will appear an impermissible breach of impartiality. It may happen even that prior ruling on a matter will disqualify a judge from sitting on another case involving the same issue. But it is still generally recognized that it is difficult to lay down an absolute rule disqualifying a judge because he has ruled against a litigant in an earlier context. Note, supra at 1452. What due process requires is being worked out case by case. Given the history and the current state of law, the federal courts should give the Hawaii Supreme Court a chance to decide what due process requires.
In making that decision, a Justice of the Hawaii Supreme Court will be asking, “Have I so made up my mind by my earlier ruling in the criminal case that I cannot give a fair hearing to Partington’s explanation of why he thought his strategy was appropriate?” and “Will public confidence in the courts be diminished if I, a justice who’s already found Partington to have given ineffective assistance, now sit on a case where he is defending himself against this charge?” A federal court is not in a position to anticipate how a justice of the Hawaii Supreme Court will answer those questions.
The dissent places weight on a statement in Withrow, upholding a procedure in which a later acquittal by the administrative body concerned would not be “logically inconsistent” with a prior finding of probable cause. Withrow v. Larkin, 421 U.S. 35, 57, 95 S.Ct. 1456, 1469, 43 L.Ed.2d 712 (1975). There would, however, be no logical inconsistency between the Supreme Court reversing the murder conviction in Clarke because of ineffective assistance of counsel and then finding that in the disciplinary proceeding against Partington he had explained his strategy so that the court was now satisfied that he had not breached his responsibilities as a lawyer. The question of sixth amendment deprivation, decided in Clarke, is logically distinct from the disciplinary standards which Partington must meet to satisfy the Bar.
The dissent also places emphasis on the “practical” probability of prejudice. But the dissent admits that a judge may charge an attorney with violating Fed.R.Civ.P. 11, order him to show cause, conduct a subsequent hearing, and find against him. The dissent also admits that a court may cite an attorney for contempt of court, order him to show cause, and then preside over the contempt trial and find against him. As a practical matter, the probability of a judge in these circumstances having in mind a vivid impression of the professional breach of the lawyer, and being influenced by this impression, is much greater than the probability of an appeals court, months or years after the initial decision that occasioned the reference of the lawyer to disciplinary proceedings, being still prejudiced against the lawyer. The dissent is on treacherous grounds when it argues that practical prejudice is more likely in Partington’s case than in the cases it accepts as routine judicial practice.
Judges have not been particularly good at recognizing their own biases. The major federal reforms have been made by the legislature — the elimination of Supreme Court Justices reviewing their own decisions in 1891; the introduction of recusal motions by litigants in 1911; the reasonable person test of objectivity in 1974. This history suggests the need for heightened self-scrutiny. The impartiality that “the common lot of humanity will admit” is now closer to the ideal.