Teague v. Lane

Justice Brennan,

with whom Justice Marshall joins, dissenting.

Today a plurality of this Court, without benefit of briefing and oral argument, adopts a novel threshold test for federal review of state criminal convictions on habeas corpus. It does so without regard for — indeed, without even mentioning — our contrary decisions over the past 35 years delineating the broad scope of habeas relief. The plurality further appears oblivious to the importance we have consistently accorded the principle of stare decisis in nonconstitutional cases. Out of an exaggerated concern for treating similarly situated habeas petitioners the same, the plurality would for the first time preclude the federal courts from considering on collateral review a vast range of important constitutional *327 challenges; where those challenges have merit, it would bar the vindication of personal constitutional rights and deny society a check against further violations until the same claim is presented on direct review. In my view, the plurality’s “blind adherence to the principle of treating like cases alike” amounts to “letting the tail wag the dog” when it stymies the resolution of substantial and unheralded constitutional questions. Griffith v. Kentucky, 479 U. S. 314, 332 (1987) (White, J., dissenting). Because I cannot acquiesce in this unprecedented curtailment of the reach of the Great Writ, particularly in the absence of any discussion of these momentous changes by the parties or the lower courts, I dissent.

hH

The federal habeas corpus statute provides that a federal court “shall entertain an application for a writ of habeas corpus in behalf of a person in custody pursuant to the judgment of a State court only on the ground that he is in custody in violation of the Constitution or laws or treaties of the United States.” 28 U. S. C. §2254 . 1 For well over a century, we have read this statute and its forbears to authorize federal courts to grant writs of habeas corpus whenever a person’s liberty is unconstitutionally restrained. Shortly after the Habeas Corpus Act of 1867, ch. 27, 14 Stat. 385 , empowered federal courts to issue writs of habeas corpus to state authorities, we noted: “This legislation is of the most comprehensive character. It brings within the habeas corpus jurisdiction of every court and of every judge every possible case of privation of liberty contrary to the National Constitution, treaties, *328 or laws. It is impossible to widen this jurisdiction.” Ex parte McCardle, 6 Wall. 318, 325-326 (1868). See also Fay v. Noia, 372 U. S. 391, 426 (1963) (“Congress in 1867 sought to provide a federal forum for state prisoners having constitutional defenses by extending the habeas corpus powers of the federal courts to their constitutional maximum”). Nothing has happened since to persuade us to alter that judgment. Our thorough survey in Fay v. Noia of the history of habeas corpus at common law and in its federal statutory embodiment led us to conclude that “conventional notions of finality in criminal litigation cannot be permitted to defeat the manifest federal policy that federal constitutional rights of personal liberty shall not be denied without the fullest opportunity for plenary federal judicial review.” Id., at 424 . In Noia we therefore held that federal courts have the power to inquire into any constitutional defect in a state criminal trial, provided that the petitioner remains “in custody” by virtue of the judgment rendered at that trial. Our subsequent rulings have not departed from that teaching in cases where the presentation of a petitioner’s claim on collateral review is not barred by a procedural default. See, e. g., Rose v. Mitchell, 443 U. S. 545, 550-565 (1979); Jackson v. Virginia, 443 U. S. 307, 320-324 (1979).

In particular, our decisions have made plain that the federal courts may collaterally review claims such as Teague’s once state remedies have been exhausted. In Brown v. Allen, 344 U. S. 443 (1953), for example, we held that state prisoners alleging discrimination in the selection of members of the grand jury that indicted them and the petit jury that tried them were entitled to reconsideration of those allegations in federal court. “Discriminations against a race by barring or limiting citizens of that race from participation in jury service,” we noted, “are odious to our thought and our Constitution. This has long been accepted as the law.” Id., at 470 (citations omitted). See also Vasquez v. Hillery, 474 U. S. 254 (1986); Rose v. Mitchell, supra.

*329 Our precedents thus supply no support for the plurality’s curtailment of habeas relief. 2 Just as it was “a fortuity that we overruled Swain v. Alabama, 380 U. S. 202 (1965) [which set forth an unduly strict standard for proving that a prosecutor’s use of peremptory challenges was racially discriminatory in violation of the Equal Protection Clause], in a case that came to us on direct review” when “[w]e could as easily *330 have granted certiorari and decided the matter in a case on collateral review,” Griffith v. Kentucky, 479 U. S., at 332 (White, J., dissenting), so too there is no reason why we cannot decide Teague’s almost identical claim under the Sixth Amendment on collateral review rather than in a case on direct review. Because there is no basis for extending the Court’s rationale in Stone v. Powell, 428 U. S. 465 (1976), to preclude review of Teague’s challenge to the composition of the jury that convicted him, and because I perceive no other ground consistent with our precedents for limiting the cogni-zability of constitutional claims on federal habeas corpus, I would reach the merits of Teague’s Sixth Amendment argument and hold in his favor.

II

Unfortunately, the plurality turns its back on established case law and would erect a formidable new barrier to relief. Any time a federal habeas petitioner’s claim, if successful, would result in the announcement of a new rule of law, the plurality says, it may only be adjudicated if that rule would “plac[e] ‘certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe,’” ante, at 307, quoting Mackey v. United States, 401 U. S. 667, 692 (1971) (Harlan, J., concurring in judgments in part and dissenting in part), or if it would mandate “new procedures without which the likelihood of an accurate conviction is seriously diminished.” Ante, at 313.

A

Astonishingly, the plurality adopts this novel precondition to habeas review without benefit of oral argument on the question and with no more guidance from the litigants than a three-page discussion in an amicus brief. See Brief for Criminal Justice Legal Foundation as Amicus Curiae 22-24. 3 *331 Although the plurality’s approach builds upon two opinions written by Justice Harlan some years ago, see Mackey v. United States, supra, at 675 (opinion concurring in judgments in part and dissenting in part); Desist v. United States, 394 U. S. 244, 256 (1969) (dissenting opinion), it declines fully to embrace his views. No briefing or argument at all was devoted to the points at which the plurality departs from his proposals. It is indeed ironic that in endorsing the bulk of Justice Harlan’s approach to the provision of federal habeas relief, the Court ignores his reminder that our “obligation of orderly adherence to our own processes would demand that we seek that aid which adequate briefing and argument lends to the determination of an important issue.” Mapp v. Ohio, 367 U. S. 643, 677 (1961) (dissenting opinion). Before breaking so sharply with precedent, the plurality would have done well, I think, to recall what we said in Ladner v. United States, 358 U. S. 169, 173 (1958): “The question of the scope of collateral attack upon criminal sentences is an important and complex one .... We think that we should have the benefit of a full argument before dealing with the question.”

B

Equally disturbing, in my view, is the plurality’s infidelity to the doctrine of stare decisis. That doctrine “demands respect in a society governed by the rule of law,” Akron v. *332 Akron Center for Reproductive Health, Inc., 462 U. S. 416, 419-420 (1983), because it enhances the efficiency of judicial decisionmaking, allowing judges to rely on settled law without having to reconsider the wisdom of prior decisions in every case they confront, and because it fosters predictability in the law, permitting litigants and potential litigants to act in the knowledge that precedent will not be overturned lightly and ensuring that they will not be treated unfairly as a result of frequent or unanticipated changes in the law. We have therefore routinely imposed on those asking us to overrule established lines of cases “the heavy burden of persuading the Court that changes in society or in the law dictate that the values served by stare decisis yield in favor of a greater objective.” Vasquez v. Hillery, 474 U. S., at 266 .

In this case, as when we considered the reviewability of grand jury discrimination on habeas corpus, “we have been offered no reason to believe that any such metamorphosis has rendered the Court’s long commitment to a rule of reversal outdated, ill-founded, unworkable, or otherwise legitimately vulnerable to serious reconsideration.” Vasquez v. Hillery, supra, at 266 . None of the reasons we have hitherto deemed necessary for departing from the doctrine of stare decisis are present. Our interpretations of the reach of federal habeas corpus have not proceeded from inadequate briefing or argumentation, nor have they taken the form of assertion unaccompanied by detailed justification. See, e. g., Copperweld Corp. v. Independence Tube Corp., 467 U. S. 752, 766 (1984). No new facts or arguments have come to light suggesting that our reading of the federal habeas statute or our divination of congressional intent was plainly mistaken. See, e. g., Monell v. New York City Dept. of Social Services, 436 U. S. 658 (1978). In addition, Congress has done nothing to shrink the set of claims cognizable on habeas since it passed the Habeas Corpus Act of 1867, despite our consistent interpretation of the federal habeas statute to permit adjudication of *333 cases like Teague’s. Finally, the rationale for our decisions has not been undermined by subsequent congressional or judicial action. See, e. g., Braden v. 30th Judicial Circuit Court of Kentucky, 410 U. S. 484, 497-499 (1973). None of the exceptions to the doctrine of stare decisis we have recognized apply. I therefore remain mystified at where the plurality finds warrant to upset, sua sponte, our time-honored precedents.

C

The plurality does not so much as mention stare decisis. Indeed, from the plurality’s exposition of its new rule, one might infer that its novel fabrication will work no great change in the availability of federal collateral review of state convictions. Nothing could be further from the truth. Although the plurality declines to “define the spectrum of what may or may not constitute a new rule for retroactivity purposes,” it does say that generally “a case announces a new rule when it breaks new ground or imposes a new obligation on the States or the Federal Government.” Ante, at 301. Otherwise phrased, “a case announces a new rule if the result was not dictated by precedent existing at the time the defendant’s conviction became final.” Ibid. This account is extremely broad. 4 Few decisions on appeal or collateral review are “dictated” by what came before. Most such cases involve a question of law that is at least debatable, permitting a rational judge to resolve the case in more than one way. Virtually no case that prompts a dissent on the relevant legal point, for example, could be said to be “dictated” by prior decisions. By the plurality’s test, therefore, *334 a great many cases could only be heard on habeas if the rule urged by the petitioner fell within one of the two exceptions the plurality has sketched. Those exceptions, however, are narrow. Rules that place “ ‘certain kinds of primary, private individual conduct beyond the power of the criminal lawmaking authority to proscribe,’” ante, at 307, quoting Mackey v. United States, 401 U. S., at 692 (Harlan, J., concurring in judgments in part and dissenting in part), are rare. And rules that would require “new procedures without which the likelihood of an accurate conviction is seriously diminished,” ante, at 313, are not appreciably more common. The plurality admits, in fact, that it “believe[s] it unlikely that many such components of basic due process have yet to emerge.” Ibid. The plurality’s approach today can thus be expected to contract substantially the Great Writ’s sweep.

Its impact is perhaps best illustrated by noting the abundance and variety of habeas cases we have decided in recent years that could never have been adjudicated had the plurality’s new rule been in effect. Although “history reveals no exact tie of the writ of habeas corpus to a constitutional claim relating to innocence or guilt,” Schneckloth v. Bustamonte, 412 U. S. 218, 257 (1973) (Powell, J., concurring), the plurality’s decision to ignore history and to link the availability of relief to guilt or innocence when the outcome of a case is not “dictated” by precedent would apparently prevent a great many Fifth, Sixth, and Fourteenth Amendment cases from being brought on federal habeas.

For example, in Nix v. Whiteside, 475 U. S. 157 (1986), the Court ruled that a defendant’s right to counsel under the Sixth Amendment is not violated when a defense attorney refuses to cooperate with him in presenting perjured testimony at trial. Clearly, the opposite result sought by the petitioner could not have been dictated by prior cases, nor would the introduction of perjured testimony have improved the accuracy of factfinding at trial. The claim presented on habeas was therefore novel yet well outside the plurality’s *335 exceptions. Were the claim raised tomorrow on federal collateral review, a court could not reach the merits, as did we. The same is true of numerous right-to-counsel and representation claims we have decided where the wrong alleged by the habeas petitioner was unlikely to have produced an erroneous conviction. See, e. g., Moran v. Burbine, 475 U. S. 412 (1986) (failure of police to inform defendant that attorney retained for him by somebody else sought to reach him does not violate Sixth Amendment); McKaskle v. Wiggins, 465 U. S. 168 (1984) (pro se defendant’s right to conduct own defense not violated by unsolicited participation of standby counsel); Jones v. Barnes, 463 U. S. 745 (1983) (appellate defense counsel does not have Sixth Amendment duty to raise every nonfrivolous issue requested by defendant); Morris v. Slappy, 461 U. S. 1 (1983) (state court’s denial of continuance until public defender initially assigned to represent defendant became available does not violate Sixth Amendment); Wainwright v. Torna, 455 U. S. 586 (1982) (per curiam) (no deprivation of right to counsel when defense attorney failed to make timely filing of application for certiorari in state court); Moore v. Illinois, 434 U. S. 220 (1977) (Sixth Amendment violated by corporeal identification conducted after initiation of adversary criminal proceedings in the absence of counsel); Ross v. Moffitt, 417 U. S. 600 (1974) (States need not provide indigent defendants with counsel on discretionary appeals).

Likewise, because “the Fifth Amendment’s privilege against self-incrimination is not an adjunct to the ascertainment of truth,” Tehan v. Shott, 382 U. S. 406, 416 (1966), claims that a petitioner’s right to remain silent was violated would, if not dictated by earlier decisions, ordinarily fail to qualify under the plurality’s second exception. In Estelle v. Smith, 451 U. S. 454 (1981), for example, we held that a psychiatrist who examined the defendant before trial without warning him that what he said could be used against him in a capital sentencing proceeding could not testify against him at such a proceeding. Under the plurality’s newly fashioned *336 rule, however, we could not have decided that case on the merits. The result can hardly be said to have been compelled by existing case law, see id., at 475 (Rehnquist, J., concurring in judgment), and the exclusion of such testimony at sentencing cannot have influenced the jury’s determination of the defendant’s guilt or enhanced the likely accuracy of his sentence. 5 Nor is Estelle v. Smith unique in that respect. See, e. g., Greer v. Miller, 483 U. S. 756 (1987) (single question by prosecutor during cross-examination concerning defendant’s postarrest silence does not violate Fifth Amendment); Moran v. Burbine, supra (failure of police to inform defendant of efforts of attorney to reach him does not vitiate waiver of Miranda rights); Fletcher v. Weir, 455 U. S. 603 (1982) (per curiam) (prosecutor’s use of defendant’s postar-rest silence for impeachment purposes does not constitute due process violation when defendant did not receive Miranda warnings during the period of his postarrest silence); Jenkins v. Anderson, 447 U. S. 231 (1980) (Fifth Amendment not violated by prosecutor’s use of prearrest silence to impeach defendant’s credibility).

Habeas claims under the Double Jeopardy Clause will also be barred under the plurality’s approach if the rules they seek to establish would “brea[k] new ground or imposte] a new obligation on the States or the Federal Government,” ante, at 301, because they bear no relation to the petitioner’s *337 guilt or innocence. See, e. g., Crist v. Bretz, 437 U. S. 28 (1978) (state law providing that jeopardy does not attach until first juror is sworn is unconstitutional); Chaffin v. Stynchcombe, 412 U. S. 17 (1973) (rendition of higher sentence by jury upon retrial does not violate Double Jeopardy Clause). So, too, will miscellaneous due process and Sixth Amendment claims that relate only tangentially to a defendant’s guilt or innocence. See, e. g., Bordenkircher v. Hayes, 434 U. S. 357 (1978) (no due process violation when prosecutor carries out threat to reindict on stiffer charge); Barker v. Wingo, 407 U. S. 514 (1972) (5-year delay does not violate right to speedy trial). And of course eases closely related to Teague’s, such as Lockhart v. McCree, 476 U. S. 162 (1986), where we held that the removal for cause of so-called “Witherspoon-exdudables” does not violate the Sixth Amendment’s fair cross section requirement, would be beyond the purview of this Court when they arrived on habeas.

D

These are massive changes, unsupported by precedent. 6 They also lack a reasonable foundation. By exaggerating the importance of treating like cases alike and granting relief to all identically positioned habeas petitioners or none, “the Court acts as if it has no choice but to follow a mechanical notion of fairness without pausing to consider ‘sound principles *338 of decisionmaking.’” Griffith v. Kentucky, 479 U. S., at 332-333 (White, J., dissenting), quoting Stovall v. Denno, 388 U. S. 293, 301 (1967). Certainly it is desirable, in the interest of fairness, to accord the same treatment to all ha-beas petitioners with the same claims. Given a choice between deciding an issue on direct or collateral review that might result in a new rule of law that would not warrant retroactive application to persons on collateral review other than the petitioner who brought the claim, we should ordinarily grant certiorari and decide the question on direct review. Following our decision in Griffith v. Kentucky, supra, a new rule would apply equally to all persons whose convictions had not become final before the rule was announced, whereas habeas petitioners other than the one whose case we decided might not benefit from such a rule if we adopted it on collateral review. Taking cases on direct review ahead of those on habeas is especially attractive because the retrial of habeas petitioners usually places a heavier burden on the States than the retrial of persons on direct review. Other things being equal, our concern for fairness and finality ought to therefore lead us to render our decision in a case that comes to us on direct review.

Other things are not always equal, however. Sometimes a claim which, if successful, would create a new rule not appropriate for retroactive application on collateral review is better presented by a habeas case than by one on direct review. In fact, sometimes the claim is only presented on collateral review. In that case, while we could forgo deciding the issue in the hope that it would eventually be presented squarely on direct review, that hope might be misplaced, and even if it were in time fulfilled, the opportunity to check constitutional violations and to further the evolution of our thinking in some area of the law would in the meanwhile have been lost. In addition, by preserving our right and that of the lower federal courts to hear such claims on collateral review, we would not discourage their litigation on federal habeas corpus and *339 thus not deprive ourselves and society of the benefit of decisions by the lower federal courts when we must resolve these issues ourselves.

The plurality appears oblivious to these advantages of our settled approach to collateral review. Instead, it would deny itself these benefits because adherence to precedent would occasionally result in one habeas petitioner’s obtaining redress while another petitioner with an identical claim could not qualify for relief. 7 In my view, the uniform treatment of habeas petitioners is not worth the price the plurality is willing to pay. Permitting the federal courts to decide novel ha-beas claims not substantially related to guilt or innocence has profited our society immensely. Congress has not seen fit to withdraw those benefits by amending the statute that provides for them. And although a favorable decision for a petitioner might not extend to another prisoner whose identical claim has become final, it is at least arguably better that the wrong done to one person be righted than that none of the injuries inflicted on those whose convictions have become final be redressed, despite the resulting inequality in treatment. I therefore adhere to what we said in Stovall v. Denno, supra, where we held that the rules we laid down in United States v. Wade, 388 U. S. 218 (1967), and Gilbert v. *340 California, 388 U. S. 263 (1967), should not be applied retroactively:

“We recognize that Wade and Gilbert are, therefore, the only victims of pretrial confrontations in the absence of their counsel to have the benefit of the rules established in their cases. That they must be given that benefit is, however, an unavoidable consequence of the necessity that constitutional adjudications not stand as mere dictum. Sound policies of decision-making, rooted in the command of Article III of the Constitution that we resolve issues solely in concrete cases or controversies, and in the possible effect upon the incentive of counsel to advance contentions requiring a change in the law, militate against denying Wade and Gilbert the benefit of today’s decisions. Inequity arguably results from according the benefit of a new rule to the parties in the case in which it is announced but not to other litigants similarly situated in the trial or appellate process who have raised the same issue. But we regard the fact that the parties involved are chance beneficiaries as an insignificant cost for adherence to sound principles of decision-making.” Id., at 301 (footnotes omitted).

I see no reason to abandon these views. Perfectly evenhanded treatment of habeas petitioners can by no means justify the plurality’s sua sponte renunciation of the ample benefits of adjudicating novel constitutional claims on habeas corpus that do not bear substantially on guilt or innocence.

rH I

Even if one accepts the plurality’s account of the appropriate limits to habeas relief, its conclusion that Teague’s claim may not be heard is dubious. The plurality seeks to give its decision a less startling aspect than it wears by repeatedly mischaracterizing Teague’s Sixth Amendment claim. As the plurality would have it, Teague contends “ ‘that petit juries actually chosen must mirror the community and reflect the *341 various distinctive groups in the population,’” ante, at 292, quoting Taylor v. Louisiana, 419 U. S. 522, 538 (1975), and that fairness in jury selection “ ‘require[s] proportional representation of races upon a jury.’” Ante, at 301, quoting Akins v. Texas, 325 U. S. 398, 403 (1945). Teague, however, makes no such claim — which is presumably why the plurality quotes dicta from other cases rather than Teague’s brief. He submits, rather, that “the Sixth Amendment guarantees the accused a jury selected in accordance with procedures that allow a fair possibility for the jury to reflect a cross section of the community.” Brief for Petitioner 4 (emphasis added). Indeed, Teague specifically disavows the position attributed to him by the plurality: “The defendant is not entitled to a jury of any particular composition and no requirement exists that the petit jury mirror the distinctive groups in the population . . . .” Ibid. Teague’s claim is simply that the Sixth Amendment’s command that no distinctive groups be systematically excluded from jury pools, Taylor v. Louisiana, supra, or from venires drawn from them, Duren v. Missouri, 439 U. S. 357 (1979), applies with equal force to the selection of petit juries. He maintains that this firmly established principle prohibits the prosecution from using its peremptory challenges discriminatorily to prevent venirepersons from sitting on the jury merely because they belong to some racial, ethnic, or other group cognizable for Sixth Amendment purposes. Teague’s claim is therefore closely akin to that which prevailed in Batson v. Kentucky, 476 U. S. 79 (1986), where we held that the Equal Protection Clause prohibits the prosecution from using its peremptory challenges to exclude venirepersons from the jury solely because they share the defendant’s race. The only potentially significant difference is that Teague’s claim, if valid, would bar the prosecution from excluding venirepersons from the petit jury on account of their membership in some cognizable group even when the defendant is not himself a member of that group, whereas the Equal Protection Clause might not *342 provide a basis for relief unless the defendant himself belonged to the group whose members were improperly excluded. 8

Once Teague’s claim is characterized correctly, the plurality’s assertions that on its new standard his claim is too novel to be recognized on habeas corpus, ante, at 301, and that the right he invokes is “a far cry from the kind of absolute prerequisite to fundamental fairness that is ‘implicit in the concept of ordered liberty,”’ ante, at 314, are dubious. The requirement Teague asks us to impose does not go far beyond our mandates in Taylor, Duren, and Batson ; indeed, it flows quite naturally from those decisions. The fact that the Sixth Amendment would permit a challenge by a defendant who did not belong to a cognizable group whose members were discriminatorily excluded from the jury does not alter that conclusion. As we said in Rose v. Mitchell, 443 U. S., at 555 -556:

“Discrimination on the basis of race, odious in all aspects, is especially pernicious in the administration of justice. Selection of members of a grand jury because they are of one race and not another destroys the appearance of justice and thereby casts doubt on the integrity of the judicial process. The exclusion from grand jury service of Negroes, or any group otherwise qualified to serve, impairs the confidence of the public in the administration of justice. As this Court repeatedly has *343 emphasized, such discrimination ‘not only violates our Constitution and the laws enacted under it but is at war with our basic concepts of a democratic society and a representative government.’ Smith v. Texas, 311 U. S. 128, 130 (1940) (footnote omitted). The harm is not only to the accused, indicted as he is by a jury from which a segment of the community has been excluded. It is to society as a whole. ‘The injury is not limited to the defendant — there is injury to the jury system, to the law as an institution, to the community at large, and to the democratic ideal reflected in the processes of our courts.’ Ballard v. United States, 329 U. S. 187, 195 (1946).” (Emphasis added.)

The plurality’s assertion that Teague’s claim fails to fit within Justice Harlan’s second exception is also questionable. It bears noting that Justice Powell, long a staunch advocate of Justice Harlan’s views on the scope of collateral review, leaned to the opposite opinion: “Whenever the fairness of the petit jury is brought into question doubts are raised as to the integrity of the process that found the prisoner guilty. Collateral relief therefore may be justified even though it entails some damages to our federal fabric.” Rose v. Mitchell, supra, at 584, n. 6 (Powell, J., concurring) (citation omitted). Justice Jackson rightly observed:

“It is obvious that discriminatory exclusion of Negroes from a trial jury does, or at least may, prejudice a Negro’s right to a fair trial, and that a conviction so obtained should not stand. The trial jury hears the evidence of both sides and chooses what it will believe. In so deciding, it is influenced by imponderables — unconscious and conscious prejudices and preferences — and a thousand things we cannot detect or isolate in its verdict and whose influence we cannot weigh. A single juror’s dissent is generally enough to prevent conviction. A trial jury on which one of the defendant’s race has no chance to sit may not have the substance, and cannot *344 have the appearance, of impartiality, especially when the accused is a Negro and the alleged victim is not.” Cassell v. Texas, 339 U. S. 282, 301-302 (1950) (dissenting opinion).

More recently, in Vasquez v. Hillery, 474 U. S., at 263 , we expressly rejected the claim that “discrimination in the grand jury has no effect on the fairness of the criminal trials that result from that grand jury’s actions.” Because “intentional discrimination in the selection of grand jurors is a grave constitutional trespass, possible only under color of state authority, and wholly within the power of the State to prevent,” id., at 262 , we reaffirmed our decision in Rose v. Mitchell, supra, and held that a prisoner may seek relief on federal habeas for racial discrimination in the selection of the grand jury that indicted him and that such claims are not subject to harmless-error review. Compelling the State to indict and try him a second time, we said, despite the heavy burdens it imposes, “is not disproportionate to the evil that it seeks to deter.” 474 U. S., at 262 . The plurality’s assertion that an allegation, like Teague’s, of discrimination in the selection of the petit jury — with far graver impact on the fundamental fairness of a petitioner’s trial than the discrimination we condemned in Hillery — is too tangentially connected with truth finding to warrant retroactive application on habeas corpus under its new approach therefore strains credibility.

C

A majority of this Court’s Members now share the view that cases on direct and collateral review should be handled differently for retroactivity purposes. See Griffith v. Kentucky, 479 U. S. 314 (1987); Allen v. Hardy, 478 U. S. 255 (1986) (per curiam); Williams v. United States, 401 U. S. 646, 665 (1971) (opinion of Marshall, J.). In Griffith , the Court adopted Justice Harlan’s proposal that a new rule be applied retroactively to all convictions not yet final when the rule was announced. If we had adhered to our precedents, *345 reached Teague’s Sixth Amendment claim, and ruled in his favor, we would ultimately have had to decide whether we should continue to apply to habeas cases the three-factor approach outlined in Stovall v. Denno, 388 U. S., at 297 , or whether we should embrace most of the other half of Justice Harlan’s proposal and ordinarily refuse to apply new rules retroactively to cases on collateral review, except in the cases where they are announced.

In my view, that is not a question we should decide here. The better course would have been to grant certiorari in another case on collateral review raising the same issue and to resolve the question after full briefing and oral argument. Justices Blackmun and Stevens, ante, pp. 319-320, disagree. They concur in the Court’s judgment on this point because they find further discussion unnecessary and because they believe that, although Teague’s Sixth Amendment claim is meritorious, neither he nor other habeas petitioners may benefit from a favorable ruling. As I said in Stovall v. Denno, supra, at 301 , according a petitioner relief when his claim prevails seems to me “an unavoidable consequence of the necessity that constitutional adjudications not stand as mere dictum. ” But I share the view of Justices Blackmun and Stevens that the retroactivity question is one we need not address until Teague’s claim has been found meritorious. Certainly it is not one the Court need decide before it considers the merits of Teague’s claim because, as the plurality mistakenly contends, its resolution properly determines whether the merits should be reached. By repudiating our familiar approach without regard for the doctrine of stare decisis, the plurality would deprive us of the manifold advantages of deciding important constitutional questions when they come to us first or most cleanly on collateral review. I dissent.

Prisoners sentenced by a federal court may seek to have their sentences vacated, corrected, or set aside “upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States, or that the court was without jurisdiction to impose such sentence, or that the sentence was in excess of the maximum authorized by law, or is otherwise subject to collateral attack.” 28 U. S. C. § 2255 . The plurality does not address the question whether the rule it announces today extends to claims brought by federal, as well as state, prisoners.

Until today, this Court has imposed but one substantive limitation on the cognizability of habeas claims. In Stone v. Powell, 428 U. S. 465 (1976), the Court held that where a State has provided a defendant with an opportunity for full and fair litigation of a claim that evidence used against him was obtained through an unlawful search or seizure in violation of the Fourth Amendment, he may not relitigate that claim on federal habeas. The Court noted, however, that “Fourth Amendment violations are different in kind from denials of Fifth or Sixth Amendment rights,” id,., at 479, and it expressly stated that its decision was “not concerned with the scope of the habeas corpus statute as authority for litigating constitutional claims generally,” in substantial part because “the exclusionary rule is a judicially created remedy rather than a personal constitutional right.” Id., at 495, n. 37 . None of the Court’s reasoning in Stone v. Powell supports the plurality’s present decision not to adjudicate Teague’s claim, because Teague is attempting to vindicate what he alleges is a fundamental personal right, rather than trying to invoke a prophylactic rule devised by this Court to deter violations of personal constitutional rights by law enforcement officials. In cases of this kind, our reluctance to allow federal courts to interfere with state criminal processes has never been deemed paramount. See Vasquez v. Hillery, 474 U. S. 254, 262 (1986); Rose v. Mitchell, 443 U. S. 545, 584, n. 6 (1979) (Powell, J., concurring in judgment).

Our ruling in Rose v. Mitchell, supra, confirms this conclusion. We there rejected the argument that our holding in Stone v. Powell should be extended to preclude federal habeas review of claims of racial discrimination in the selection of members of a state grand jury, notwithstanding the fact that the selection of petit jurors was free from constitutional infirmity and that guilt was established beyond a reasonable doubt at a trial devoid of constitutional error. Teague’s challenge to the composition of the petit jury is perforce on even firmer ground. See also Kimmelman v. Morrison, All U. S. 365 (1986) (counsel’s failure to litigate competently petitioner’s Fourth Amendment claim cognizable on habeas); Jackson v. Virginia, 443 U. S. 307, 320-324 (1979) (sufficiency of the evidence claims may be brought on habeas).

As the plurality points out, ante, at 300, our decision in Allen v. Hardy, 478 U. S. 255 (1986) (per curiam), addressed the retroactive application of our holding in Batson v. Kentucky, 476 U. S. 79 (1986), even *331 though the petition for certiorari in that case did not discuss that issue. Our decision in Allen, however, applied settled retroactivity doctrine; unlike the plurality’s opinion today, it did not announce a sharp break with past practice. And although the course we followed in Mapp v. Ohio, 367 U. S. 643 (1961), was urged on us by amicus rather than by the parties themselves, incorporation of the protections of the Bill of Rights through the Fourteenth Amendment was by no means a novel step at that time, and the relevant issues were familiar from our prior cases. Nor does the fact that the parties here debated the extent to which Batson should be applied retroactively diminish the startling abruptness of the plurality’s action, for the adoption of a version of Justice Harlan’s approach to retro-activity to bar habeas review of most claims that would result in new rules of law if they prevailed was not even mentioned by the parties.

Compare Justice Stewart’s much more restrained approach in Milton v. Wainwright, 407 U. S. 371 (1972): “An issue of the ‘retroactivity’ of a decision of this Court is not even presented unless the decision in question marks a sharp break in ‘he web of the law. The issue is presented only when the decision overrules clear past precedent, or disrupts a practice long accepted and widely relied upon.” Id., at 381, n. 2 (dissenting opinion) (citations omitted).

In “limiting the scope of the second exception to those new procedures without which the likelihood of an accurate conviction is seriously diminished,” ante, at 313, the plurality presumably intends the exception to cover claims that involve the accuracy of the defendant’s sentence as well as the accuracy of a court’s determination of his guilt. See Smith v. Murray, 477 U. S. 527, 538 (1986) (no “fundamental miscarriage of justice” where introduction of testimony at sentencing phase of capital case “neither precluded the development of true facts nor resulted in the admission of false ones”). Thus, the plurality’s new rule apparently would not prevent capital defendants, for example, from raising Eighth Amendment, due process, and equal protection challenges to capital sentencing procedures on habeas corpus.

The plurality’s claim that “our cases have moved in the direction of reaffirming the relevance of the likely accuracy of convictions in determining the available scope of habeas review,” ante, at 313, has little force. Two of the cases it cites —Kuhlmann v. Wilson, 477 U. S. 436, 454 (1986) (plurality opinion), and Murray v. Carrier, 477 U. S. 478 (1986) — discuss the conditions under which a habeas petitioner may obtain review even though his claim would otherwise be procedurally barred. They do not hold that a petitioner’s likely guilt or innocence bears on the cognizability of habeas claims in the absence of procedural default. And the Court has limited Stone v. Powell, 428 U. S. 465 (1976), as noted above, see supra, at 328-330, and n. 2, to Fourth Amendment exclusionary rule claims, passing up several opportunities to extend it.

The plurality’s complaint that prior retroactivity decisions have sometimes led to more than one habeas petitioner’s reaping the benefit of a new rule while most habeas petitioners obtained no relief because of “our failure to treat retroactivity as a threshold question,” ante, at 305, is misguided. The disparity resulting from our deciding three years later, in Solem v. Stumes, 465 U. S. 638 (1984), not to apply retroactively the rule of Edwards v. Arizona, 451 U. S. 477, 484-487 (1981), should not be ascribed to our failure to make retroactivity a threshold question, but rather to our failure to decide the retroactivity question at the same time that we decided the merits issue. If both decisions are made contemporaneously, see, e. g., Witherspoon v. Illinois, 391 U. S. 510, 523, n. 22 (1968); Stovall v. Denno, 388 U. S. 293 (1967), then only one exception need be made to the rule of equal treatment. The plurality may find even this slight inequality unacceptable, but the magnitude of the disparity is not, and need not be, as large as its example suggests.

The plurality’s persistent misreading of Teague’s claim, ante, at 301-302, n. 1, is puzzling. To be sure, Teague does argue that the principles informing our decision in Duren v. Missouri, 439 U. S. 357 (1979), should be extended to the selection of the petit jury. But Duren does not require that every venire provide a microcosm of the community; it demands, instead, that no group be systematically excluded from venires unless a significant state interest would thereby be manifestly and primarily advanced. Lack of proportional representation of a cognizable group on a given petit jury, in Teague’s view, helps to establish a prima facie Sixth Amendment violation; contrary to the plurality’s suggestion, he does not contend that it is itself a per se violation.