concurring.

I join the Court’s opinion, and consider it an important step toward ending the confusion that has resulted from applying Linkletter v. Walker, 381 U. S. 618 (1965), on a case-by-case basis. I concluded in 1977 that the Court would be well advised to adopt Justice Harlan’s view as to the retroactive application of our decisions both with respect to cases pending at the time on direct appeal and with respect to cases pending on habeas corpus petitions. See Hankerson v. North Carolina, 432 U. S. 233, 246 (1977) (concurring in judgment). The Harlan view is stated in Mackey v. United States, 401 U. S. 667, 675 (1971) (opinion concurring in judgment in Mackey and dissenting from judgment in Williams v. United States, 401 U. S. 646 (1971)); and Desist v. United States, 394 U. S. 244, 256 (1969) (dissenting opinion). I was persuaded by Justice Harlan’s reasoning then, and have followed it since. See Hankerson v. North Carolina, supra; Harlin v. Missouri, 439 U. S. 459, 460 (1979) (concurring in judgment); Brown v. Louisiana, 447 U. S. 323, 337 (1980) (con*329curring in judgment); Solem v. Stumes, 465 U. S. 638, 651 (1984) (concurring in judgment).

As the cases we decide today involve only the retroactivity of decisions pending on direct review, it was not necessary for the Court to express an opinion with respect to habeas corpus petitions. As I read the Court’s opinion, this question is carefully left open until it is squarely presented. It is to be hoped that the Court then will adopt the Harlan view of retroactivity in cases seeking relief on habeas petitions. See Mackey v. United States, supra, at 681-695. Under that view, habeas petitions generally should be judged according to the constitutional standards existing at the time of conviction.

dissenting.

As I stated in my dissenting opinion in Shea v. Louisiana, 470 U. S. 51, 61 (1985), I am willing to adopt both aspects of the approach to retroactivity propounded by Justice Harlan in his opinion in Mackey v. United States, 401 U. S. 667, 675 (1971). In Justice Harlan’s view, new constitutional rules governing criminal prosecutions should apply retroactively for cases pending on direct appeal when the rule is announced, and, with narrow exceptions, should not apply in collateral proceedings challenging convictions that become final before the rule is announced. The majority today adopts only a portion of this approach. I therefore join Justice White’s dissent, agreeing with him that, under the present state of our retroactivity jurisprudence, the majority erred in rejecting the reasons cited in Allen v. Hardy, 478 U. S. 255 (1986), for making Batson v. Kentucky, 476 U. S. 79 (1986), nonretroactive.

delivered the opinion of the Court.

These cases, one state and one federal, concern the retrospective application of Batson v. Kentucky, 476 U. S. 79 (1986).

In Batson, 476 U. S., at 96-98, this Court ruled that a defendant in a state criminal trial could establish a prima facie case of racial discrimination violative of the Fourteenth Amendment, based on the prosecution’s use of peremptory challenges to strike members of the defendant’s race from the jury venire, and that, once the defendant had made the prima facie showing, the burden shifted to the prosecution to come forward with a neutral explanation for those challenges. In the present cases we consider whether that ruling is applicable to litigation pending on direct state or federal review or not yet final when .Batson was decided. We answer that question in the affirmative.

I

A. No. 85-5221. Petitioner Randall Lamont Griffith, a black person, was indicted in 1982 in the Circuit Court of Jefferson County, Ky. (the same court where Batson was tried), on charges of first-degree robbery, theft by unlawful taking, *317and being a persistent felony offender in the second degree. App. 2. On the first day of trial, the prosecution and defense attorneys conducted voir dire examination of the jury venire and exercised their peremptory challenges.1 The prosecution used four of its five allotted challenges to strike four of the five prospective black jurors. The defense used eight of its allotted nine challenges to strike prospective white jurors. There were two duplicate strikes. The two extra jurors who remained because of the duplicate strikes, one of whom was a black person, then were removed by random draw.2 Thus, no black person remained on the jury. Id., at 5, 12-13.

Defense counsel expressed concern that Griffith was to be tried by an all-white jury. He asked the court to request the prosecutor to state his reasons for exercising peremptory challenges against the four prospective black jurors. The request was refused. Id., at 13. Counsel then moved for discharge of the panel, alleging that the prosecutor’s use of peremptory challenges to remove all but one of the prospective black jurors constituted a violation of Griffith’s Sixth and Fourteenth Amendment rights. Id., at 15. The court denied the motion. The jury returned a verdict of guilty on the charge of first-degree robbery and fixed petitioner’s punishment at 10 years’ imprisonment.3 The jury then found petitioner guilty of being a persistent felony offender, and, pur*318suant to Ky. Rev. Stat. §532.080 (1985), enhanced his sentence to 20 years’ imprisonment.

The Supreme Court of Kentucky, with an unpublished memorandum opinion, affirmed the judgment of conviction. App. 17. The court rejected petitioner’s claim that the prosecutor’s use of peremptory challenges deprived him of guaranteed equal protection. It relied on Swain v. Alabama, 380 U. S. 202 (1965), where this Court ruled that a black defendant did not establish a violation of the Equal Protection Clause solely on proof of the prosecutor’s use of peremptory challenges to strike black jurors at the defendant’s own trial. Id., at 221-222. The Court noted, however, that an inference of purposeful discrimination could be raised where a prosecutor had engaged in a pattern of challenging black jurors in a series of cases. See id., at 223-224. The Kentucky court concluded that Swain disposed of petitioner’s claim and it “decline[d] to go further than the Swain court.” App. 18.

Griffith timely filed here a petition for a writ of certiorari, While his petition was pending, this Court decided Batson v. Kentucky, supra, where it rejected a portion of the reasoning of Swain v. Alabama on which the Kentucky court had relied. 476 U. S., at 89-96. Two months later, in Allen v. Hardy, 478 U. S. 255 (1986) (per curiam), we held that the ruling in Batson was not to be applied retroactively to a case on federal habeas review. We granted certiorari in Griffith’s case, 476 U. S. 1157 (1986), limited to the question whether the ruling in Batson applies retroactively to a state conviction pending on direct review at the time of the Batson decision.

B. No. 85-5731. In 1984, petitioner Willie Davis Brown, a black person, was convicted by a jury in the United States District Court for the Western District of Oklahoma on narcotics charges. During jury selection, two venire panels were assembled. 6 Record 2-10.4 There were six prospec*319tive black jurors in the total venire. Four were excused for cause by the court and the other two were excused by the prosecutor’s use of peremptory challenges. Id., at 20.5 Defense counsel objected to the prosecutor’s use of peremptory challenges to strike the black persons from the jury, claiming that petitioner was thereby denied a jury representative of the community. Id., at 20-21. No action was taken in response to that objection.

As prospective jurors were being assembled for the second venire panel, the prosecutor called the jury clerk to inquire about the racial composition of the additional venire. At a hearing held later while the jury was deliberating, there was evidence that the prosecutor said to the clerk: “We would like to have as few black jurors as possible.” App. 51. The clerk testified, however, that she remembered the prosecutor’s comment to be: “Don’t get any blacks on this jury.” Id., at 38-39. The clerk went on to say that she did not alter the jury selection in any way in response to the prosecutor’s comment. Id., at 44-45. The District Court concluded that the prosecutor’s contact with the jury clerk “would have to be looked at and dealt with by someone,” id., at 44, inasmuch as it fell “into the category of possible prosecutorial misconduct,” id., at 46, but that it did not affect the integrity of the selection of the jury. Id., at 45. The court therefore concluded that a new trial would not be necessary if the jury convicted petitioner. Id., at 46.

The United States Court of Appeals for the Tenth Circuit affirmed the judgment of conviction. 770 F. 2d 912 (1985). It rejected Brown’s claim that the prosecutor’s use of peremptory challenges to exclude prospective black jurors, *320combined with his call to the jury clerk, violated petitioner’s right to an impartial jury. The court concluded that Brown had not met Swain’s threshold requirement that petitioner must show a systematic and intentional course of conduct by the prosecutor calculated to exclude black jurors in “case after case.” 770 F. 2d, at 914. It further concluded that the communication by the prosecutor to the jury clerk did not suggest a pattern of systematic exclusion of black jurors. Although the court observed that the prosecutor’s action was “improper” and “must be condemned,” ibid., it concluded, as had the District Court, that the prosecutor’s request had no effect on the selection of Brown’s jury.

Prior to our Batson decision, petitioner timely filed with this Court a petition for a writ of certiorari. We granted certiorari, 476 U. S. 1157 (1986), again limited to the question whether the ruling in Batson applies retroactively to a federal conviction then pending on direct review. The case was set for argument in tandem with Griffith’s case.

I — I HH

Twenty-one years ago, this Court adopted a three-pronged analysis for claims of retroactivity of new constitutional rules of criminal procedure. See Linkletter v. Walker, 381 U. S. 618 (1965). In Linkletter, the Court held that Map-p v. Ohio, 367 U. S. 643 (1961), which extended the Fourth Amendment exclusionary rule to the States, would not be applied retroactively to a state conviction that had become final before Mapp was decided. The Court explained that “the Constitution neither prohibits nor requires retrospective effect” of a new constitutional rule, and that a determination of retro-activity must depend on “weighting] the merits and demerits in each case.” 381 U. S., at 629. The Court’s decision not to apply Mapp retroactively was based on “the purpose of the Mapp rule; the reliance placed upon the [previous] doctrine; and the effect on the administration of justice of a retrospective application of Mapp” 381 U. S., at 636. See also *321Stovall v. Denno, 388 U. S. 293, 297 (1967) (retroactivity depends on “(a) the purpose to be served by the new standards, (b) the extent of the reliance by law enforcement authorities on the old standards, and (c) the effect on the administration of justice of a retroactive application of the new standards”).

Shortly after the decision in Linkletter, the Court held that the three-pronged analysis applied both to convictions that were final6 and to convictions pending on direct review. See Johnson v. New Jersey, 384 U. S. 719, 732 (1966); Stovall v. Denno, 388 U. S., at 300. In the latter case, the Court concluded that, for purposes of applying the three factors of the analysis, “no distinction is justified between convictions now final . . . and convictions at various stages of trial and direct review.” Ibid. Thus, a number of new rules of criminal procedure were held not to apply retroactively either to final cases or to cases pending on direct review. See, e. g., Stovall v. Denno, supra; DeStefano v. Woods, 392 U. S. 631, 635, n. 2 (1968); Desist v. United States, 394 U. S. 244, 253-254 (1969); Daniel v. Louisiana, 420 U. S. 31 (1975) (per curiam).

In United States v. Johnson, 457 U. S. 537 (1982), however, the Court shifted course.7 In that case, we reviewed at some length the history of the Court’s decisions in the area of retroactivity and concluded, in the words of Justice Harlan: “ ‘ “[RJetroactivity” must be rethought. ’ ” Id., at 548 (quoting Desist v. United States, 394 U. S., at 258 (dissenting opinion)). Specifically, we concluded that the retroactiv*322ity analysis for convictions that have become final must be different from the analysis for convictions that are not final at the time the new decision is issued.8 We observed that, in a number of separate opinions since Linkletter, various Members of the Court “have asserted that, at a minimum, all defendants whose cases were still pending on direct appeal at the time of the law-changing decision should be entitled to invoke the new rule.” 457 U. S., at 545, and n. 9 (collecting opinions).9 The rationale for distinguishing between cases that have become final and those that have not, and for applying new rules retroactively to cases in the latter category, was explained at length by Justice Harlan in Desist v. United States, 394 U. S., at 256 (dissenting opinion), and in Mackey v. United States, 401 U. S. 667, 675 (1971) (opinion concurring in judgment). In United States v. Johnson, we embraced to a significant extent the comprehensive analysis presented by Justice Harlan in those opinions.

In Justice Harlan’s view, and now in ours, failure to apply a newly declared constitutional rule to criminal cases pending on direct review violates basic norms of constitutional adjudi-| cation. First, it is a settled principle that this Court adjudi-* cates only “cases” and “controversies.” See U. S. Const., Art. III, §2. Unlike a legislature, we do not promulgate new rules of constitutional criminal procedure on a broad basis. Rather, the nature of judicial review requires that we adjudicate specific cases, and each case usually becomes the vehicle for announcement of a new rule. But after we have *323decided a new rule in the case selected, the integrity of judicial review requires that we apply that rule to all similar cases pending on direct review. Justice Harlan observed:

“If we do not resolve all cases before us on direct review in light of our best understanding of governing constitutional principles, it is difficult to see why we should so adjudicate any case at all. ... In truth, the Court’s assertion of power to disregard current law in adjudicating cases before us that have not already run the full course of appellate review, is quite simply an assertion that our constitutional function is not one of adjudication but in effect of legislation.” Mackey v. United States, 401 U. S., at 679 (opinion concurring in judgment).

As a practical matter, of course, we cannot hear each case pending on direct review and apply the new rule. But we fulfill our judicial responsibility by instructing the lower courts to apply the new rule retroactively to cases not yet final. Thus, it is the nature of judicial review that precludes us from “[s]imply fishing one case from the stream of appellate review, using it as a vehicle for pronouncing new constitutional standards, and then permitting a stream of similar cases subsequently to flow by unaffected by that new rule.” Ibid. See United States v. Johnson, 457 U. S., at 546-547, 555.

Second, selective application of new rules violates the principle of treating similarly situated defendants the same. See Desist v. United States, 394 U. S., at 258-259 (Harlan, J., dissenting). As we pointed out in United States v. Johnson, the problem with not applying new rules to cases pending on direct review is “the actual inequity that results when the Court chooses which of many similarly situated defendants should be the chance beneficiary” of a new rule. 457 U. S., at 556, n. 16 (emphasis in original). Although the Court had tolerated this inequity for a time by not applying new rules retroactively to cases on direct review, we noted: “The time for toleration has come to an end.” Ibid.

*324In United States v. Johnson, our acceptance of Justice Harlan’s views led to the holding that “subject to [certain exceptions], a decision of this Court construing the Fourth Amendment is to be applied retroactively to all convictions that were not yet final at the time the decision was rendered.” Id., at 562. The exceptions to which we referred related to three categories in which we concluded that existing precedent established threshold tests for the retro-activity analysis. In two of these categories, the new rule already was retroactively applied: (1) when a decision of this Court did nothing more than apply settled precedent to different factual situations, see id., at 549, and (2) when the new ruling was that a trial court lacked authority to convict a criminal defendant in the first place. See id., at 550.10

The third category — where a new rule is a “clear break” with past precedent — is the one at issue in these cases. We described it in United States v. Johnson, 457 U. S., at 549-550:

“[WJhere the Court has expressly declared a rule of criminal procedure to be ‘a clear break with the past,’ Desist v. United States, 394 U. S., at 248, it almost invariably has gone on to find such a newly minted principle nonretroactive. See United States v. Peltier, 422 U. S. 531, 547, n. 5 (1975) (Brennan, J., dissenting) (collecting cases). In this . . . type of case, the traits of the particular constitutional rule have been less critical than the Court’s express threshold determination that the ‘“new” constitutional interpretatio[n] ... so change[s] the law that prospectivity is arguably the proper course,’ Williams v. United States, 401 U. S., at 659 (plurality opinion). Once the Court has found that the new rule was unanticipated, the second and third Stovall factors —reliance by law enforcement authorities *325on the old standards and effect on the administration of justice of a retroactive application of the new rule — have virtually compelled a finding of nonretroactivity. See, e. g., Gosa v. Mayden, 413 U. S., at 672-673, 682-685 (plurality opinion); Michigan v. Payne, 412 U. S., at 55-57.”

Thus, we recognized what may be termed a “clear break exception.” Under this exception, a new constitutional rule was not applied retroactively, even to cases on direct review, if the new rule explicitly overruled a past precedent of this Court, or disapproved a practice this Court had arguably sanctioned in prior cases, or overturned a longstanding practice that lower courts had uniformly approved. Id., at 551. The Fourth Amendment ruling in Payton v. New York, 445 U. S. 573 (1980), with which United States v. Johnson was concerned, was not a clear break in any of these senses, and thus its retroactivity status was not “effectively preordained” by falling within the “clear break” exception. 457 U. S., at 553-554.

In Shea v. Louisiana, 470 U. S. 51 (1985), we applied United States v. Johnson and held that the Fifth Amendment rule announced in Edwards v. Arizona, 451 U. S. 477 (1981), which prohibited the use, after a suspect had requested counsel, of a confession obtained by police-instigated interrogation without the suspect’s attorney’s being present, was retroactive to cases on direct review when Edwards was decided. Using Johnson’s rationale, we concluded there was nothing about a Fourth Amendment rule that suggested it should be given greater retroactive effect than a Fifth Amendment rule. 470 U. S., at 59. In addition, as in United States v. Johnson, we concluded that the new rule did not fall within the “clear break” exception. The previous Term, in Solem v. Stumes, 465 U. S. 638, 647 (1984), the Court had explicitly recognized that Edwards was “not the sort of ‘clear break’ case that is almost automatically non-*326retroactive.”11 Although, in Shea, we expressed some doubt' as to “the merits of a different retroactivity rule for cases” in which a new rule is a clear break with the past, we explained that “we have no need to be concerned with the question here.” 470 U. S., at 59, n. 5.

h — I I — I I — I

The question whether a different retroactivity rule should apply when a new rule is a “clear break” with the past, however, is squarely before us in the present cases. In Allen v. Hardy, 478 U. S. 255 (1986), a case which was here on federal habeas, we said that the rule in Batson “is an explicit and substantial break with prior precedent” because it “overruled [a] portion of Swain.” 478 U. S., at 258.12 We therefore now reexamine the rationale for maintaining a “clear break” exception to the general proposition that new rules governing criminal procedure should be retroactive to cases pending on direct review. For the same reasons that persuaded us in United States v. Johnson to adopt different conclusions as to convictions on direct review from those that already hadj become final, we conclude that an engrafted exception based| solely upon the particular characteristics of the new rule adopted by the Court is inappropriate.

First, the principle that this Court does not disregard current law, when it adjudicates a case pending before it on direct review, applies regardless of the specific characteristics of the particular new rule announced. The Court recognized in United States v. Johnson that the fact that a new rule is a clear break with the past is relevant primarily because it implicates the second and third Stovall factors of reliance by law enforcement officials and the burden on the administra*327tion of justice imposed by retroactive application. But even if these factors may be useful in deciding whether convictions that already have become final should receive the benefit of a new rule, the “clear break” exception, derived from the Stovall factors, reintroduces precisely the type of case-specific analysis that Justice Harlan rejected as inappropriate for cases pending on direct review.

Second, the use of a “clear break” exception creates the same problem of not treating similarly situated defendants the same. James Kirkland Batson, the petitioner in Batson v. Kentucky, and Randall Lamont Griffith, the petitioner in the present Kentucky case, were tried in Jefferson Circuit Court approximately three months apart.13 The same prosecutor exercised peremptory challenges at the trials. It was solely the fortuities of the judicial process that determined the case this Court chose initially to hear on plenary review. Justice Powell has pointed out that it “hardly comports with the ideal of ‘administration of justice with an even hand,’” when “one chance beneficiary — the lucky individual whose case was chosen as the occasion for announcing the new principle — enjoys retroactive application, while others similarly situated have their claims adjudicated under the old doctrine.” Hankerson v. North Carolina, 432 U. S. 233, 247 (1977) (opinion concurring in judgment), quoting Desist v. United States, 394 U. S., at 255 (Douglas, J., dissenting). See also Michigan v. Payne, 412 U. S. 47, 60 (1973) (Marshall, J., dissenting) (“Different treatment of two cases is justified under our Constitution only when the cases differ in some respect relevant to the different treatment”). The fact that the new rule may constitute a clear break with the past has no bearing on the “actual inequity that results” when only *328one of many similarly situated defendants receives the benefit of the new rule. United States v. Johnson, 457 U. S., at 556, n. 16 (emphasis omitted).

We therefore hold that a new rule for the conduct of criminal prosecutions is to be applied retroactively to all cases, state or federal, pending on direct review or not yet final, with no exception for cases in which the new rule constitutes a “clear break” with the past. Accordingly, in No. 85-5221, the judgment of the Supreme Court of Kentucky is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion. In No. 85-5731, the judgment of the United States Court of Appeals for the Tenth Circuit is reversed, and the case is remanded to that court for further proceedings consistent with this opinion.

It is so ordered.

with whom The Chief Justice and Justice O’Connor join, dissenting.

Last Term this Court decided that the rule announced in Batson v. Kentucky, 476 U. S. 79 (1986), should not apply on collateral review of convictions that became final before the decision in Batson was announced. Allen v. Hardy, 478 *330U. S. 255 (1986). In reaching this judgment, the Court weighed the three factors that it has traditionally considered in deciding the retroactivity of a new rule of criminal procedure: “ ‘ “(a) the purpose to be served by the new standards, (b) the extent of the reliance by law enforcement authorities on the old standards, and (c) the effect on the administration of justice of a retroactive application of the new standards.””’ Id., at 258 (quoting Solem v. Stumes, 465 U. S. 638, 643 (1984), in turn quoting Stovall v. Denno, 388 U. S. 293, 297 (1967)). No Justice suggested that this test is unworkable. The question, then, is why the Court feels constrained to fashion a different rule for cases on direct review. The reasons the Court offers are not new, and I find them as unpersuasive today as I have in the past:

“Two concerns purportedly underlie the majority’s decision. The first is that retroactivity is somehow an essential attribute of judicial decisionmaking, and that when the Court announces a new rule and declines to give it retroactive effect, it has abandoned the judicial role and assumed the function of a legislature — or, to use the term Justice Harlan employed in describing the problem, a ‘super-legislature.’ Desist v. United States, 394 U. S. 244, 259 (1969) (Harlan, J., dissenting). The second (and not completely unrelated) concern is fairness. It is the business of a court, the majority reasons, to treat like cases alike; accordingly, it is unfair for one litigant to receive the benefit of a new decision when another, identically situated, is denied the same benefit. The majority’s concerns are no doubt laudable, but I cannot escape the conclusion that the rule they have spawned makes no sense.

“As a means of avoiding what has come to be known as the super-legislature problem, the rule announced by the majority is wholly inadequate. True, the Court is not and cannot be a legislature, super or otherwise. But I should think that concerns about the supposed usurpa*331tion of legislative authority by this Court generally go more to the substance of the Court’s decisions than to whether or not they are retroactive. Surely those who believe that the Court has overstepped the bounds of its legitimate authority in announcing a new rule of constitutional law will find little solace in a decision holding the new rule retroactive. If a decision is in some sense illegitimate, making it retroactive is a useless gesture-that will fool no one. If, on the other hand, the decision is a salutary one, but one whose purposes are ill-served by retroactive application, retroactivity may be worse than useless, imposing costs on the criminal justice system that will likely be uncompensated for by any perceptible gains in ‘judicial legitimacy.’

[[Image here]]

“The claim that the majority’s rule serves the interest of fairness is equally hollow. Although the majority finds it intolerable to apply a new rule to one case on direct appeal but not to another, it is perfectly willing to tolerate disparate treatment of defendants seeking direct review of their convictions and prisoners attacking their convictions in collateral proceedings. As I have stated before, see [United States v.] Johnson, [457 U. S. 537, 566-568 (1982)] (White, J., dissenting); Williams v. United States, 401 U. S. 646, 656-659 (1971) (plurality opinion), it seems to me that the attempt to distinguish between direct and collateral challenges for purposes of retroactivity is misguided. Under the majority’s rule, otherwise identically situated defendants may be subject to different constitutional rules, depending on just how long ago now-unconstitutional conduct occurred and how quickly cases proceed through the criminal justice system. The disparity is no different in kind from that which occurs when the benefit of a new constitutional rule is retroactively afforded to the defendant in whose case it is announced but to no others; the Court’s new *332approach equalizes nothing except the numbers of defendants within the disparately treated classes.” Shea v. Louisiana, 470 U. S. 51, 62-64 (1985) (White, J., dissenting).1

The Court’s invocation of fairness also overlooks the fact that it is a fortuity that we overruled Swain v. Alabama, 380 U. S. 202 (1965), in a case that came to us on direct review. We could as easily have granted certiorari and decided the matter in a case on collateral review, such as Allen v. Hardy. In that case, the principle of treating like cases alike would dictate that all cases on collateral review receive the benefit of the new rule. I trust that the Court would not go that far in letting the tail wag the dog; good judgment would — I hope — win out over blind adherence to the principle of treating like cáses alike. Yet today the Court acts as if *333it has no choice but to follow a mechanical notion of fairness without pausing to consider “sound principles of decision-making,” Stovall v. Denno, 388 U. S., at 301.

For the foregoing reasons, I would adhere to the approach set out in Stovall v. Denno, supra, at 300, and recognize no distinction for retroactivity purposes between cases on direct and collateral review. But even if I saw some merit in applying the Harlan approach to cases on direct appeal, I would nonetheless preserve the exception for “clear breaks” recognized in United States v. Johnson, 457 U. S. 537 (1982). Under our precedent, “a decision announcing a new standard ‘is almost automatically nonretroactive’ where the decision ‘has explicitly overruled past precedent.’” Allen v. Hardy, 478 U. S., at 258 (quoting Solem v. Stumes, 465 U. S., at 646, 647). As the majority in Johnson explained:

“Once the Court has found that [a] new rule was unanticipated, the second and third Stovall factors — reliance by law enforcement authorities on the old standards and effect on the administration of justice of a retroactive application of the new rule — have virtually compelled a finding of non-retroactivity.” 457 U. S., at 549-550 (citations omitted).

The Court has already recognized that Batson constitutes “an explicit and substantial break with prior precedent,” and that “prosecutors, trial judges, and appellate courts throughout our state and federal systems justifiably have relied on the standard of Swain. ” Allen v. Hardy, supra, at 258, 260. The reasons that the Court gave in Allen v. Hardy for concluding that “retroactive application of the Batson rule on collateral review of final convictions would seriously disrupt the administration of justice,” 478 U. S., at 260, apply equally to retroactive application of the Batson rule on direct review.2

*334The majority knows that it is penalizing justifiable reliance on Swain, and in doing so causing substantial disruption in the administration of justice; yet the majority acts as if it has no principled alternative. This is not true; it would be a far sounder rule, and no less principled, to apply the Stovall test to determine retroactivity on both direct and collateral review. I respectfully dissent.