WILLIE DAVIS BROWN, A/K/A WILL BROWN, PETITIONER V. UNITED STATES OF AMERICA No. 85-5731 In the Supreme Court of the United States October Term, 1986 On Writ of Certiorari to the United States Court of Appeals for the Tenth Circuit Brief for the United States TABLE OF CONTENTS Opinion below Jurisdiction Question Presented Statement Introduction and summary of argument Argument: In cases pending on direct appeal, the decision in Batson v. Kentucky should not be given retroactive effect A. Because Batson overruled Swain v. Alabama, it should be held nonretroactive B. Traditional retroactivity analysis favors holding Batson nonretroactive 1. The new rule does not have such a fundamental impact on the integrity of factfinding as to compel retroactive application 2. Prosecutors, trial judges, and appellant courts justifiably have relied on the standard of Swain 3. Retroactive application of the Batson rule, even on direct review, would seriously disrupt the administration of justice Conclusion OPINION BELOW The opinion of the court of appeals (J.A. 53-61) is reported at 770 F.2d 912. JURISDICTION The judgment of the court of appeals was entered on August 20, 1985. A petition for rehearing was denied on September 12, 1985 (J.A. 66). The petition for a writ of certiorari was filed on October 10, 1985, and granted on June 2, 1986 (J.A. 67). The jurisdiction of this Court is invoked under 28 U.S.C. 1254(1). QUESTION PRESENTED QUESTION PRESENTED Whether the holding in Batson v. Kentucky, No. 84-6263 (Apr. 30, 1986), should be given retroactive effect to cases pending on direct appeal. STATEMENT After a jury trial in the United States District Court for the Western District of Oklahoma, petitioner was convicted of conspiracy to possess marijuana and cocaine with intent to distribute them, in violation of 21 U.S.C. 846. He was sentenced to five years' imprisonment. The court of appeals affirmed. J.A. 53-61. 1. The evidence at trial showed that petitioner supplied drugs from his home in Florida for distribution in Oklahoma. Several co-conspirators testified about petitioner's drug dealing activities. In particular, the co-conspirators testified that petitioner had supplied a large quantity of marijuana that was seized in Florida while in transit to Oklahoma. J.A. 54. The evidence also included testimony from a Florida deputy sheriff who described his previous arrest of petitioner in 1980. In that incident, petitioner was arrested at a Florida airport after giving a suitcase containing 17 pounds of marijuana to another person. J.A. 54-55. 2. Petitioner and co-defendant Sam Ponder, both black, were tried for the drug conspiracy. Ponder was represented by E. Melvin Porter, a prominent black member of the Oklahoma State Senate. During the jury voir dire the prosecutor used two of his peremptory challenges to excuse black jurors (J.A. 14). No black jurors served on petitioner's jury (J.A. 55). During the second day of the jury selection, before additional veniremen were summoned, the prosecutor telephoned the jury clerk. Somewhere later during the trial, the district court questioned the jury clerk in chambers and later held a hearing concerning the incident (J.A. 2-52). The prosecutor testified that he had asked the jury clerk how many blacks would be called as additional veniremen, but he denied that he had intended to influence the jury clerk's selection of the potential jurors (J.A. 14-15). The prosecutor recalled that he had said to the jury clerk, "(w)e would like to have as few black jurors as possible" (J.A. 51). The jury clerk remembered the conversation differently. She recalled the prosecutor's comment as having been, "Don't get any blacks on this jury" (J.A. 38). The jury clerk testified that she took no action in response to the prosecutor's remarks (J.A. 11, 55). The prosecutor explained at the hearing that, based on personal experience, he considered Senator Porter a gifted attorney who had a special appeal to blacks in the community. The prosecutor thus planned to excuse blacks who knew or had heard of Senator Porter. J.A. 15-16, 19. The district court concluded that the prosecutor's contract with the jury clerk, while raising serious questions of professional impropriety, did not affect the integrity of the jury selection process (J.A. 11, 45-46). 3. The court of appeals affirmed the conviction and rejected petitioner's contention that the prosecutor's use of peremptory challenges or his contact with the jury clerk violated petitioner's right to an impartial jury (J.A. 56-57). Relying on Swain v. Alabama, 380 U.S. 202 (1965), the court reasoned that the prosecutor's use of peremptory challenges in this case did not establish a systematic practice of excluding blacks in "case after case" (J.A. 56). The court condemned the prosecutor's contact with the jury clerk. Because the clerk disregarded the prosecutor's statement, however, the court concluded that the contact could not have influenced the jury selection process (J.A. 60). /1/ Thus, the prosecutor's conduct "had no impact on the possibility that the jury would not represent a cross-section of the community" (J.A. 60). /2/ 4. The petition for a writ of certiorari was filed after this Court granted certiorari in Batson v. Kentucky, No. 84-6263. After Batson was decided on April 30, 1986, the Court granted the petition limited to the question of the retroactivity of Batson in cases pending on direct review. The case was set for argument in tandem with Griffth v. Kentucky, No. 85-5221, in which certiorari was granted limited to the same issue. J.A. 67. INTRODUCTION AND SUMMARY OF ARGUMENT It bears emphasis at the outset that "the choice between retroactivity and nonretroactivity in no way turns on the value of the constitutional guarantee involved." Johnson v. New Jersey, 384 U.S. 719, 728 (1966); accord, Gosa v. Mayden, 413 U.S. 665, 675-676 (1973) (plurality opinion). When a decision, like Batson v. Kentucky, No. 84-6263 (Apr. 30, 1986), touches on such sensitive areas as race relations, there is a temptation to assert, since it is rooted in profound moral and constitutional values, that it has always been the "true" law. The retroactivity jurisprudence of this Court, however, has avoided the fiction that overruled decisions never existed and the mistaken notion that constitutional rights are somehow cheapened by holding them nonretroactive. The Court has "firmly rejected the idea that all new interpretations of the Constitution must be considered always to have been the law and that prior constructions to the contrary must always be ignored." Williams v. United States, 401 U.S. 646, 651 (1971) (plurality opinion) (citing Linkletter v. Walker, 381 U.S. 618 (1965)); accord, Gosa v. Mayden, 413 U.S. at 676. And the Court "do(es) not disparage a constitutional guarantee in any manner by declining to apply it retroactively." Johnson v. New Jersey, 384 U.S. at 728; accord, Michigan v. Payne, 412 U.S. 47, 55 n.10 (1973). A. Although the retroactivity of new constitutional decisions once depended on a three-factor test, this Court has recently adopted a new approach to retroactivity questions in cases pending on direct review. In such cases, the Court has stated that it will hold new decisions automatically retroactive, unless the decisions constitute a clear break with the past. This Court's decision in Batson v. Kentucky, No. 84-6263 (Apr. 30, 1986), overruled Swain v. Alabama, 380 U.S. 202 (1965), and therefore announced a clear break with the past. Batson is no less a clear break simply because it may have been foreshadowed by other cases. This Court has invariably recognized that decisions that overrule prior precedent are clear breaks with the past, even if it may have been possible to foresee the overruling. For that reason, this Court's new approach to retroactivity requires that Batson be held nonretroactive in cases pending on direct review. B. Even under the three-factor test that the Court previously applied to resolve all retroactivity questions, it would still be appropriate to hold that Batson is not retroactive. The Court has already determined in Allen v. Hardy, No. 85-6593 (June 30, 1986) (per curiam), that none of the three factors favors retroactivity, and thus that Batson should not applied to cases that had become final by the time Batson was decided. That holding was correct, and it supports nonretroactivity with regard to cases on direct as well as collateral review. In Allen v. Hardy, the Court assumed that Batson has some positive impact on the truthfinding function of juries. The Court nonetheless held that that factor was not sufficient to require that the decision be held retroactive. The Court has consistently denied retroactive effect to decisions that have some positive impact on the truthfinding function when those decisions, like Batson, serve a mixture of constitutional values and complement other safeguards of reliability. In this regard, it is particularly noteworthy that the Court denied retroactive effect to the decisions requiring jury trials in serious criminal cases and outlawing the systematic exclusion of women from state juries. In Allen v. Hardy, the Court also found that prosecutors and lower courts justifiably had relied on Swain. The good-faith reliance of lower courts is unquestioned, and it was certainly reasonable for prosecutors to read Swain to permit the exercise of peremptory challenges against blacks on the basis of race in particular cases. The reasonableness of that interpretation is strongly confirmed by the language of Swain, no matter how wrong the Court may now feel that such an interpretation of the Constitution was. Finally, in Allen v. Hardy, the Court found that the retroactive application of Batson on collateral review would disrupt the administration of justice, because hearings would be necessary and it would be impossible in virtually every case for the prosecutor to recall the reasons for his peremptory challenges. Those considerations apply equally on direct review. The number of cases pending on direct review, of course, is smaller than the number on collateral and direct review combined, but there are nonetheless a significant number of pending cases. The factor, like the others, therefore counsels against holding Batson retroactive to cases pending on direct review at the time Batson was decided. ARGUMENT IN CASES PENDING ON DIRECT APPEAL, THE DECISION IN BATSON v. KENTUCKY SHOULD NOT BE GIVEN RETROACTIVE EFFECT In Swain v. Alabama, 380 U.S. 202 (1965), this Court concluded that it could not "hold that the striking of Negroes in a particular case is a denial of equal protection of the laws" (380 U.S. at 221). The Court confirmed that peremptory challenges are "exercised on grounds normally thought irrelevant to legal proceedings or official action, namely, the race, religion, nationality, occupation or affiliations of people summoned for jury duty" (id. at 220). The Court added that "(t)o subject the prosecutor's challenge in any particular case to the demands and traditional standards of the Equal Protection Clause would entail a radical change in the nature and operation of the challenge" (id. at 221-222). Swain was subjected to some academic criticism, and some courts held that the use of peremptory challenges in an individual case was subject to challenge under provisions of state constitutions or principles of federal law other than constitutional equal protection (see Batson v. Kentucky, No. 84-6263 (Apr. 30, 1986), slip op. 1 n.1, 9 n.14). Before Batson, however, no court had held that federal equal protection principles were violated by the prosecutor's exercise of peremptory challenges to exclude members of the defendant's race from the petit jury at the defendant's trial. On April 30, 1986, the Court brought about the "radical change" that it had eschewed in Swain. Explicitly overruling everything in Swain that was contrary to the new decision, that Court expressly contradicted Swain and held that "the Equal Protection Clause forbids the prosecutor" in a particular case "to challenge potential jurors solely on account of their race" (Batson v. Kentucky, slip op. 9, 20 n.25). Thereafter, this Court recognized in Allen v. Hardy, No. 85-6593 (June 30, 1986) (per curiam), slip op. 3, that the rule in the Batson was "an explicit and substantial break with prior precedent." As we show below, under the approach to retroactivity questions adopted by a majority of this Court in United States v. Johnson, 457 U.S. 537 (1982), and Shea v. Louisiana, No. 85-5920 (Feb. 20, 1985), Batson should be held nonretroactive because it overruled Swain. The Court in Johnson abandoned the three-factor approach of Stovall v. Denno, 388 U.S. 293 (1967), for cases pending on direct appeal, but preserved certain principles from the Court's earlier retroactivity decisions. Under one of those principles -- the "clear break" principle -- the nonretroactivity of Batson is "preordained" (id. at 558). Even under the analysis previously employed by the Court in making retroactivity determinations, Batson would still be held nonretroactive, because none of the traditional factors set forth in Stovall v. Denno favors retroactivity. A. Because Batson Overruled Swain v. Alabama, It Should Be Held Nonretroactive 1. Until 1965, the Court gave full retroactive effect to each interpretation of the Constitution prescribing rules for the conduct of criminal trials. The Court changed its approach in Linkletter v. Walker, 381 U.S. 618 (1965). After Linkletter, when faced with a question of retroactivity the Court typically inquired into three issues: (1) the purpose to be served by the new standards, (2) the extent of reliance by law enforcement authorities on the old standards, and (3) the effect on the administration of justice of a retroactive application of the new standards. Stovall v. Denno, 388 U.S. 293, 297 (1967). At all times, however, there were Members of the Court who found application of the Stovall factors an unsatisfactory approach (see generally United States v. Johnson, 457 U.S. at 545 & n.9, 547-548 & n.10). Justice Harlan, in particular, would have applied all new rules of constitutional law to all cases on direct review (see Desist v. United States, 394 U.S. 244, 258-259 (1969) (dissenting opinion)). /3/ In United States v Johnson, supra, the Court revised its approach to retroactivity. In place of the three-factor Stovall test, the Court concluded, as Justice Harlan had suggested, that new decisions should ordinarily be held retroactive in cases pending on direct review. See also Shea v. Louisiana, No. 82-5920 (Feb. 20, 1985), slip op. 8. As an exception to that rule, the Court held that a new decision would not be given retroactive effect, even to cases on direct review, if the new decision represented "a clear break with the past" (457 U.S. at 549). The Court noted that it had always found the necessary "clear break" whenever "a decision explicitly overrules a past precedent of this Court" (id. at 551). The Johnson Court made it clear that, notwithstanding the prior Stovall approach, a determination that a decision is a clear break with the past ends that inquiry into retroactivity on direct appeal of a criminal case, in contrast to civil cases, in which that determination begins the inquiry (457 U.S. at 549-550 n.12). Applying its new analysis, the Court held that Payton v. New York, 445 U.S. 573 (1980), was retroactive on direct appeal. The Court did so, however, only after noting that Payton did not overturn "any consistent judicial history," so that its nonretroactivity was not "preordained under the 'clear break' principle stated above" (457 U.S. at 558). By preserving this category of decisions nonretroactive on direct appeal, the Court partially rejected Justice Harlan's approach. Justice Harlan would have held new constitutional decisions retroactive on direct appeal in all criminal cases, including those in which precedent had been overruled. /4/ The Johnson Court stated explicity that it was "preserving, rather than overruling," the cases endorsing nonretroactivity of decisions that overruled prior precedent (457 U.S. at 563 n.21). The Court thus agreed with the view earlier expressed by Justice Brennan that, notwithstanding the general merit of Justice Harlan's approach, that approach goes too far when it is used to justify giving retroactive effect to a decision that "overrule(d) clear past precedent of this Court." United States v. Peltier, 422 U.S. 531, 548-549 & n.7 (1975) (dissenting opinion). Although the holding in Johnson was confined to the retroactivity of Fourth Amendment decisions (457 U.S. at 562), the Court's extensive reexamination of the post-Linkletter decisions covered retroactivity decisions involving a broad range of constitutional provisions. In addition, the Court has subsequently extended Johnson's reasoning beyond the Fourth Amendment context (see Shea v. Louisiana, slip op. 8 (applying Johnson in case involving Fifth Amendment privilege against compulsory self-incrimination); Solem v. Stumes, 465 U.S. 638, 646 (1984) (same)). The Court has thus confirmed that as a general principle, a new decision may be "so clearly a break with the past that prior precedents mandate nonretroactivity" (Shea v. Louisiana, slip op. 6; see also Stumes, 465 U.S. at 646). /5/ 2. Although neither petitioner nor amici curiae even acknowledge the "clear break" exception recognized in Johnson, they do assert that Batson was not a clear break with the past. Instead, they argue, Batson was foreshadowed by earlier decisions that should have put prosecutors on notice that it was improvident to rely on Swain's advice (380 U.S. at 220-221) that a peremptory challenge might be exercised on the basis of race. /6/ As this Court has made clear several times, the fact that a decision has been foreshadowed does not in any way make it less of a "clear break," so long as it overrules a prior decision. Few decisions have had their overruling more clearly foreshadowed than Olmstead v. United States, 277 U.S. 438 (1928), which held that a physical trespass is necessary before the Fourth Amendment has any application to electronic surveillance. /7/ Yet the Court denied retroactive effect to the decision that overruled it. Desist v. United States, supra (denying retroactive effect to Katz v. United States, 389 U.S. 347 (1967)). As the Court noted in Desist, "However clearly our holding in Katz may have been foreshadowed, it was a clear break with the past * * *" (349 U.S. at 248). /8/ Just three Terms ago, the Court reinterated the same point -- that a decision can be both foreshadowed and a clear break with the past. The Court wrote: "A decision that overrules much-criticized precedent may well have been clearly foreshadowed. * * * Our cases indicate that even in this situation authorities are generally entitled to rely on existing case law, whatever its disrepute" (Solem v. Stumes, 465 U.S. at 646 n.6). In sum, any "foreshadowing" of Batson is simply irrelevant. /9/ No matter what the state of the law before Batson, this Court's precedents require that it be denied retroactive effect simply because it overruled Swain. /10/ B. Traditional Retroactivity Analysis Favors Holding Batson Nonretoractive Johnson and Shea applied the Court's new retroactivity analysis to decisions construing the Fourth Amendment and the principles of Miranda v. Arizona, 384 U.S. 436 (1966). We do not believe that there is any persuasive reason that the analysis of Johnson should be applied only in those two contexts. Nevertheless, if the Court declines to apply the Johnson analysis in other settings, it should look to the traditional three-part retroactivity test set forth in Stovall v. Denno, supra. The Stovall test is also useful to demonstrate that, even apart from the analysis of Johnson, the rule in Batson would not have been held retroactive under any of this Court's prior precedents. Petitioner and amici curiae assert that the analysis in Stovall v. Denno requires that Batson be held retroactive. The Court, of course, has already applied the three-part Stovall test in assessing the retroactive application of Batson to cases on collateral review, and it has concluded that none of the three factors favors retroactivity. Allen v. Hardy, No. 85-6593 (June 30, 1986) (per curiam). Petitioner (Br. 17, 18, 22) and amici (Br. 10 & n.2) disagree with the Courts determination. We submit that the Court correctly analyzed the Stovall factors in Allen v. Hardy. The Stovall approach to retroactivity is therefore no more helpful to petitioner than the approach adopted in Johnson. 1. The New Rule Does Not Have Such a Fundamental Impact on the Integrity of Factfinding as to Compel Retroactive Application Under the first Stovall factor, retroactive effect is most appropriate when the purpose of the new standard "goes to the heart of the truthfinding function" (Solem v. Stumes, 465 U.S. at 645). "Where the major purpose of a new constitutional doctrine is to overcome an aspect of the criminal trial that substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts in past trials, the new rule has been given complete retroactive effect" (Williams v. United States, 401 U.S. at 653 (plurality opinion)). /11/ This Court already has decided that Batson does not so fundamentally implicate the integrity of the truthfinding process that the first Stovall factor favors retroactivity (Allen v. Hardy, slip op. 4). The Court did so even on the explicit assumption that "the rule in Batson may have some bearing on the truthfinding function of a criminal trial" (ibid.). For two reasons, that assumption did not require a determination that the first Stovall factor favors retroactivity. First, the Batson rule was "designed 'to serve multiple ends'" (ibid. (quoting Brown v. Louisiana, 447 U.S. 323, 329 (1980) (plurality opinion)). Those ends include "ensur(ing) that States do not discriminate against citizens who are summoned to sit in judgment against a member of their own race," and "strengthen(ing) public confidence in the administration of justice" (ibid,; see Batson, slip op. 6-7). Second, "the new rule joins other procedures that protect a defendant's interest in a neutral factfinder," including voir dire and instructions channeling the jurors' role away from impermissible factors such as passion or prejudice (Allen v. Hardy, slip op. 4 & n.2). Therefore, the Court reasoned, there is a "high probability that the individual jurors seated in a particular case were free from bias" (id. at 4). Although petitioner (Br. 13-18) and amici (Br. 11-25) urge the Court to reverse its determination on this issue, they offer nothing inconsistent with the Court's analysis. The major effort of the briefs is to show exactly what this Court has already assumed: that the Batson rule has some bearing on the truthfinding function in a criminal trial. /12/ Amici admit (Br. 12) that the Batson rule serves multiple ends (accord, Allen v. Hardy, slip op. 2 (Marshall, J., dissenting)), and they do not dispute that, even in a pre-Batson case, other jury selection procedures make it highly likely that the individual jurors seated in a particular case were free from bias. /13/ In these circumstances, as the Court noted in Allen v. Hardy, "the fact that a rule may have some impact on the accuracy of a trial does not compel a finding of retroactivity" (slip op. 4). /14/ That proposition finds ample support in this Court's cases. In one of its earliest retroactivity cases, the Court denied retroactive effect to Escobedoe v. Illinois, 378 U.S. 478 (1964), and Miranda v. Arizona, 384 U.S. 436 (1966), despite recognizing that they "guard against the possibility of unreliable statements in every instance of in-custody interrogation." Johnson v. New Jersey, 384 U.S. 719, 730 (1966). The Court was aware that it thereby denied retroactive effect to decisions with some impact on the truthfinding process, /15/ but it "emphasize(d) that the question whether a constitutional rule of criminal procedure does or does not enhance the reliability of the fact-finding process at trial is necessarily a matter of degree. * * * We are thus concerned with a question of probabilities and must take account, among other factors, of the extent to which other safeguards are available to protect the integrity of the truth-determining process at trial." 384 U.S. at 728-729. In Stovall itself, the Court held that the new rules requiring the presence of counsel at pretrial lineups were nonretroactive even though "we should * * * assume there have been injustices in the past which could have been averted by having counsel present at the confrontation for identification" (388 U.S. at 299; see United States v. Wade, 388 U.S. 218 (1967); Gilvert v. California, 388 U.S. U.S. 263 (1967)). The Court conceded that the new rules "are aimed at avoiding unfairness at trial by enhancing the reliability of the fact-finding process in the area of identification evidence" (Stovall, 388 U.S. at 298). Yet that was not enough to require that the new rules be applied retroactively. The Court noted that cases requiring the assistance of counsel at trial or on appeal had been held retroactive, but it held that the "certainty and frequency" of injustice in lineup cases "differs greatly enough from the cases involving absence of counsel at trial or on appeal to justify treating the situations as different in kind for the purpose of retroactive application" (id. at 299). As in Stovall, there can be no claim that Batson corrects a defect in the truthfinding process comparable to the absence of counsel at trial or on appeal. The possibility that the outcome of some cases might have been different, if Batson had been applied earlier, is "different in kind" from the type of impact on truthfinding that the Stovall Court had in mind in formulating this part of the retroactivity test. See also Desist, 394 U.S. at 249 n.14. /16/ A number of cases decided after Stovall are particularly instructive because they involve the right to an impartial jury, and it is the impartiality of past juries that is called into question by the assertion that Batson has a fundamental impact on the truthfinding function. In Destefano v. Woods, 392 U.S. 631 (1968) (per curiam), the Court declined to give retroactive effect to Duncan v. Louisiana, 391 U.S. 145 (1968), which held that the states cannot deny a request for a jury trial in serious criminal cases, and Bloom v. Illinois, 391 U.S. 194 (1968), which extended the right to jury trials for serious criminal contempts. It certainly cannot be said that those decisions lacked impact on the truthfinding function. As petitioner notes (Br. 15), the decision in Duncan was intended as "an inestimable safeguard against the complaint, biased, or eccentric judge" (391 U.S. at 156); and the purpose of the jury trial guarantee generally is to "prevent() miscarriages of justice" and "assur(e) that fair trials are provided" (id. at 158). Nonetheless, the Court held that Stovall did not require retroactive application of the new rules. Because there was no suggestion that trials without juries were inevitably unfair, the Court concluded that it was not necessary to require new trials for all persons who had been convicted under procedures inconsistent with the Sixth Amendment right to jury trial. DeStefano, 392 U.S. at 633-634. Several years after DeStefano, the Court decided in Gosa v. Mayden, 413 U.S. 665 (1973), not to give retroactive effect to O'Callahan v. Parker, 395 U.S. 258 (1969). In O'Callahan the Court found that a person in military service charged with a crime that is not "service connected" is entitled to the constitutional protections of indictment by a grand jury and trial by jury in a civilian court. The plurality recognized that the constitutional rights at issue had an impact on truthfinding. /17/ The plurality explained, however, that an impact on truthfinding that was an improvement at the margin, rather than correction of some fundamental flaw, did not rise to the level needed to satisfy the first Stovall factor (413 U.S. at 680): (r)etroactivity is not required by a determination that the old standard was not the most effective vehicle for ascertaining the truth, or that the truth-determining process has been aided somewhat by the new standard, or that one of several purposes in formulating the new standard was to prevent distortion in the process. In language as applicable to Batson as to O'Callahan, the plurality held that the first Stovall factor did not favor retroactivity because "(n)othing said in O'Callahan indicates that the major purpose of that decision was to remedy a defect in the truthfinding process" (413 U.S. at 682). /18/ Subsequently, in Daniel v. Louisiana, 420 U.S. 31 (1975) (per curiam), the Court declined to accord retroactive application to Taylor v. Louisiana, 419 U.S. 522 (1975). In Taylor, this Court held that the exclusion of women from jury venires violates the Sixth and Fourteenth Amendment requirement that petit juries be selected from a source fairly representative of the community. Although it was a Sixth Amendment decision, Taylor was very similar to Batson in purpose and effect. /19/ Indeed, Taylor is the case on which amici rely for their theory that representation of diverse viewpoints is necessary to a "political" component of the jury's truthfinding function (Br. 13-14). The Court observed that Taylor, like Duncan, was "concerned generally with the function played by the jury in our system of criminal justice, more specifically the function of preventing arbitrariness and repression" (Daniel, 420 U.S. at 32). But Taylor, like Duncan, "did not rest on the premise that every criminal trial, or any particular trial, was necessarily unfair because it was not conducted in accordance with what we determined to be the requirements of the Sixth Amendment" (ibid). Therefore, despite the reliability-enhancing aspect of Taylor, the Court held that the first Stovall factor did not favor retroactivity. These principles apply directly to this case. Even on the assumption that Batson has increased the reliability of jury verdicts, the Court has not suggested (and petitioner and amici do not contend) that trials conducted under the Swain rules necessarily resulted in a jury that could not fairly judge the case. /20/ In sum, it is difficult to see how the right announced in Batson could have more of an effect on the truthfinding function than the rights announced in Duncan, O'Callahan, and Taylor. The case on which petitioner and amici principally rely is entirely consistent with this analysis and does not support retroactivity for Batson. In Brown v. Louisiana, 447 U.S. 323 (1980), the Court gave retroactive effect to the rule in Burch v. Louisiana, 441 U.S. 130 (1979), that a conviction by only five members of a six-person jury violates the Sixth Amendment jury trial guarantee. The plurality confirmed that rules that "marginally implicate the reliability and integrity of the factfinding process" or "'tend() incidentally' to avoid unfairness" need not be given retroactive effect, especially if they are "designed primarily to foster other, equally fundamental values in our system of jurisprudence" (447 U.S. at 329 (quoting Gosa v. Mayden, 413 U.S. at 680 (plurality opinion)). /21/ But the rule in Burch did not fit that mold. The rule in Burch resulted from a line-drawing process in which the Court had to choose some minimum number of jurors whose concurrence was necessary in order for a conviction to be deemed a fair result of the proper functioning of the jury. See Burch, 441 U.S. at 136-138. No other constitutional values were at stake. Because "Burch established that the concurrence of six jurors was constitutionally required to preserve the substance of the jury trial right and assure the reliability of its verdict" (Brown, 447 U.S. at 334), and imposed the constitutional requirement on no other basis, the requirement "fundamentally implicate(d) 'the fairness of the trial -- the very integrity of the fact-finding process'" (ibid. (quoting Linkletter, 381 U.S. at 639)). /22/ Batson is not at all like Burch. It is not a case in which the Court recognized that a line must be drawn somewhere between that which can be deemed reliable and fair and that which cannot. As the Court noted in Allen v. Hardy, supra, Batson is a case in which the new rule may be said to have some incidental, marginal impact on the truthfinding process but was primarily designed to foster other, equally fundamental values, such as the right of citizens to participate as jurors without exclusion on the basis of race. Brown v. Louisiana specifically preserves the rules that make such a decision nonretroactive. Therefore, under the analysis employed in Brown, as well as in DeStefano, Gosa and Daniel, the first Stovall factor does not favor retroactivity. 2. Prosecutors, Trial Judges, and Appellate Courts justifiably Have Relied on the Standard of Swain This Court has already recognized in Allen v. Hardy that "(t)he rule in Batson v. Kentucky is an explicit and substantial break with prior precedent" (slip op. 3) and that "(t)here is no question that prosecutors, trial judges, and appellate courts throughout our state and federal systems justifiably have relied on the standard of Swain" (id. at 5 (footnote omitted)). Nothing more need be said in order to decide that the second Stovall factor strongly disfavors retroactivity. Petitioner and amici, however, invite the Court to revisit this holding. Their argument suggests that Swain condemned as a violation of equal protection the use of peremptory challenges on the basis of race even in the individual case, but that it created a crippling burden of proof for those who sought to prove such a violation. Therefore, they argue, a prosecutor who took race into account in the exercise of his peremptory challenges in any case was unreasonably and in bad faith violating a constitutional right, but his conduct was insulated from review under Swain's evidentiary burden (Pet. Br. 18-19; Br. of Amici Curiae 26-28). For that reason, they contend that any reliance on Swain as a basis to consider race in the exercise of peremptory challenges was unjustifiable. This argument is gravely flawed. But even if there were merit to the argument, it would be insufficient to overcome the legitimate reliance on Swain by lower courts, whose good faith has not been called into question. It is those courts, not just prosecutors, who would have to face the burdens of hearings into prosecutors' motivation for the exercise of peremptory challenges. And, in the typical case where "the prosecutor, relying on Swain, would have had no reason to think * * * an explanation would someday be necessary" (Allen v. Hardy, slip op. 5), prosecutors will be unable to rebut any prima facie case made out by defendants, and a new trial will be necessary. The burdens of new trials will fall on the lower courts that justifiably relied on Swain, not just on the prosecutors whose good faith petitioner and amici question. Apart from the reliance on Swain by the lower courts, we submit that it was reasonable for prosecutors to read Swain to permit the exercise of peremptory challenges on the basis of race as long as the prosecutor did not challenge jurors of a particular race "in case after case, whatever the circumstances, whatever the crime and whoever the defendant or victim may be" (Swain, 380 U.S. at 223). It may now appear that the reading of Swain was incorrect, but that is not the point. All that is required to satisfy the second Stovall criterion is that, before Batson, it was reasonable to read Swain in that manner. See, e.g., Reed v. Ross, 468 U.S. 1, 17 (1984) (quoting Johnson, 457 U.S. at 551) (clear break occurs when Court "'disapprov(es) a practice this Court arguably has santioned in prior cases'"); Adams v. Illinois, 405 U.S. 278, 282 (1972) (plurality opinion) (reliance on inference of law that is "not unreasonable in light of our decisions" suffices to meet the second Stovall criterion). The Court in Swain rejected the claim that the use of peremptory challenges to strike blacks in the case before the Court violated the Equal Protection Clause. In rejecting the equal protection argument, the Court explicitly found "merit in th(e) position" that the peremptory challenge system, "in and of itself, provides justification for striking any group of otherwise qualified jurors in any given case, whether they be Negores, Catholics, accountants, or those with blue eyes" (380 U.S. at 212). In language that hardly connoted disapproval, the Court observed the frequent exercise of the peremptory challenge on the basis of "race, religion, nationality, occupation, or affiliations of people summoned for jury duty" (id. at 220 (footnote omitted)) and stated that jurors "are challenged in light of the limited knowledge cousel has of them, which may include their group affiliations, in the context of the case to be tried" (id at 221). The Court explained that "veniremen are not always judged solely as individuals," because "the question a prosecutor or defence counsel must decide is not whether a juror of a patricular race or nationality is in fact partial, but whether one from a different group is less likely to be" (id. at 220-221 (footnote omitted)). In a conclusion that could not have been more unequivocal, the Court stated, "we cannot hold that the striking of Negroes in particular case is a denial of equal protection of the laws" (id at 221). The reader of that portion of the opinion in Swain could only be left with the impression that the Court did not consider it purposeful discrimination to use peremptory challenges on the basis of "group affiliations," such as race, in the context of the case to be tried. /23/ In a separate section of the opinion, the Court in Swain then considered "a broader claim" (380 U.S. at 222) that "rais(ed) a different issue and (might) well require a different answer" (id. at 223). That claim was that it violated the Equal Protection Clause for the prosecutor to strike black jurors "in case after case, whatever the circumstances, whatever the crime, and whoever the defendant or the victim may be" (ibid). The Court declined to decide whether there was an equal protection violation even in that case because "even if a State's systematic striking of Negroes in the selection of petit juries raises a prima facie case under the Fourteenth Amendment, * * * the record in this case is not sufficient to demonstrate that the rule has been violated" (id. at 224). /24/ As one basis for its holding that the record was insufficient, the Court noted the prosecutor's statement that "striking is done differently depending on the race of the defendant and the victim of the crime" and indicated that that statement did not "support an inference that the prosecutor was bent on striking Negroes, regardless of trial-related considerations" (id. at 225-226). Batson stands for the proposition that the Court was wrong in Swain to treat the race of a juror as a legitimate trial-related consideration in the exercise of peremptory challenges in individual cases. /25/ But nothing in Batson makes it unreasonable for a reader of Swain to have concluded that peremptory challenges could be exercised on the basis of race in individual cases. Despite the language of Swain, petitioner and amici read Swain to mean only that the use of peremptory challenges was shielded from judicial inquiry by a presumption of regularity, not that the use of peremptory challenges to strike jurors of a particular race in an individual case did not violate equal protection in principles. They would have the Court believe that only an unreasonable prosecutor acting in bad faith would read Swain to hold that the exercise of peremptory challenges on the basis of race does not violate equal protection principles. That claim is easily met. Members of the Court itself have described the holding of Swain in the manner we suggest, and throughout the past 20 years commentators and courts whose good faith is not open to question have read Swain to hold that the exercise of peremptory challenges on the basis of race does not violate equal protection principles. In McCray v. New York, 461 U.S. 961 (1983) (denial of certiorari) the dissenting Justices characterized Swain as holding "that the prosecutor's use of peremptory challenges to strike Negroes from the jury panel in one particular case did not deny the defendant the equal protection of the laws" (461 U.S. at 964). In Gilliard v. Mississippi, 464 U.S. 867 (1983) (denial of certiorari), the dissenting Justices recited the holding of Swain that "the prosecution is free to use peremptory challenges to remove Negroes from the jury in any given case so long as the prosecution does not remove Negroes from juries 'in case after case, whatever the circumstances, whatever the crime and whoever the defendant or victim'" (464 U.S. at 870 (quoting Swain, 380 U.S. at 223)). The dissenters noted the view of "at least 19 jurisdictions" that "the exclustion of Negroes by peremptory challenges is constitutional in the absence of evidence of systematic exclusion" (464 U.S. at 871 (footnote omitted)). And in Thompson v. United States, No. 83-6809 (nov. 13, 1984) (denial of certiorari), the dissenters stated that Swain "authorized the presumption that a Negro juror will be partial to a Negro defendant simply because both belong to the same race" (slip op. 3-4) and added that Swain's "rule if without other virtue, was at least clear" (id. at 5). Surely no prosecutor can be faulted for agreeing with the statements of these Justices that Swain adopted a clear, substantive Fourteenth Amendment rule that permitted consideration of race in the exercise of peremptory challenges and that Swain did not simply establish a burden of proof for finding an equal protection violation. /26/ Lower courts and commentators, including both proponents and critics, have consistently read Swain to permit consideration of race in the exercise of peremptory challenges, as a matter of substantive constitutional law. One of the earliest law review commentaries on Swain declared that it "held that the use of peremptory challenges to exclude Negroes from the jury in a particular case was permissible regardless of the prosecutor's motives." The Supreme Court, 1964 Term, 79 Harv. L. Rev. 56, 135 (1965); see also Note, Fair Jury Selection Procedures, 75 Yale L.J. 322, 322-323, 325 (1965); Finkelstein, The Application of Statistical Decision Theory to the Jury Descrimination Cases, 80 Harv. L. Rev. 338, 347 (1966). Another thoughtful and highly critical commentary explored the possibility that Swain had merely set a burden of proof insulating peremptory challenges from review, but concluded that "the majority was willing to accept the respective races of the defendant and the victim of the crime as constituting, without more, (acceptable) considerations (related to the cases)" (Comment, Swain v. Alabama: A Constitutional Blueprint for the Perpetuation of the All-White Jury, 52 Va. L. Rev. 1157, 1169 n.42 (1966)). An author who was sharply critical of Swain in every respect expressed a similar understanding stating that the Swain Court "adhered to the view that * * * (use of the peremptory challenge) against members of groups suspected of sympathy with the adverse party was common, legitimate, and beyond the reach of equal protection" (Kuhn, supra note 24, 41 S. Cal. Rev. at 285 (footnote omitted)). "For example, where the defendant is Negro, the prosecution may remove Negro veniremen out of fear for their sympathy for the defendant" (id. at 284 (footnote omitted)). Swain was seen to permit removal of black jurors through peremptory challenge "in cases where, as a group, they may be considered potentially sympathetic to the defense" (id. at 293). A 1982 article supporting Swain reached the same conclusion about the proper interpretation of the Court's equal protection analysis. Saltzburg & Powers, Peremptory Challenges And The Clash Between Impartiality And Group Representation, 41 Md. L. Rev. 337, 345 (1982) (The Swain Court "recognized two possible motives for exercising challenges against black jurors. The first -- the use of race as a proxy by which to identify probable prejudice in a particular case -- was explicity approved by all the (J)ustices, except Justice Black who concurred in the result without opinion.") /27/ In 1977, a "black-letter" commentary on Swain and the federal and state cases that followed it declared that the Court had "held that a prosecutor's use of the peremptory challenge in a single case to remove blacks from the venire was not a violation of the Constitution" (Annot., 79 A.l.R.3d 14, 18 (1977)). The annotation listed Swain itself, 23 federal cases, and 101 state cases (id. at 27-32) said "to support the proposition that the use of peremptory challenge against members of a race or class in a single case does not violate the defendant's constitutional rights" (id. at 27). /28/ In practice suggestions for defense counsel, the annotation advised that "it is essential to remember that the Supreme Court, in the Swain decision, has recognized that under the appropriate circumstances the race or class of prospective jurors is a trial-related consideration and thus an appropriate basis for the exercise of the peremptory challenge" (id. at 24-25). Finally, several federal courts of appeals in the last two years have examined Swain closely, and none of them has read Swain as petitioner and amici do. /29/ Several, including at least one court that upheld a Sixth Amendment claim, have agreed that Swain permitted the exercise of peremptory challenges on the basis of race in an individual case. The most thorough examination of Swain may be found in a decision rendered just two months before Batson by the en banc Fifth Circuit (United States v. Leslie, 783 F.2d 541 (1986), petition for cert. pending, No. 85-1961). With 11 judges concurring, the fifth Circuit stated, after exhaustive quotation and analysis of Swain, that the Court had made it clear that race or other group affiliation is a legitimate ground for challenge in an individual case (783 F.2d at 548). Five judges dissented from the majority's decision not to require inquiry into peremptory challenges under the supervisory power, but none disputed the majority's interpretation of Swain. To like effect are Booker v. Jabe, 775 F.2d 762, 766 (6th Cir. 1985) (quoting Swain, 380 U.S. at 221) ("under Swain jurors may be excused on the basis of 'their group affiliations,' including race, religion and sex"), and United States v. Clark, 737 F.2d 679, 682 (7th Cir. 1984) (Swain held "that it is not a denial of the equal protection of the laws for a prosecutor to base peremptory challenges on racial grounds, provided that he is not doing so in pursuance of a systematic policy of racial exclusion from juries"). See also McCray v. Abrams, 750 F.2d 1113, 1123-1124 (2d Cir. 1984), vacated and remanded, No. 84-1426 (June 30, 1986); id. at 1137, 1138 (Meskill J., dissenting); United States v. Thompson, 730 F.2d 82, 85 (8th Cir. 1984), cert. denied, No. 83-6809 (Nov. 13, 1984). In sum, as the Court noted in Allen v. Hardy, prosecutors and lower courts have reasonably interpreted Swain to permit the exercise of peremptory challenges on the basis of race in an individual case. In light of the reasonableness of that interpretation and the consequent reliance on that reading of Swain, the second Stovall factor -- "the extent of the reliance by law enforcement authorities on the old standards" (388 U.S. at 297) -- strongly favors holding Batson nonretroactive. 3. Retroactive Application of the Batson Rule, Even on Direct Review, Would Seriously Disrupt the Administration of Justice In Allen v. Hardy, this Court applied the third Stovall factor and concluded that retroactive application of the Batson rule on collateral review of final convictions would seriously disrupt the administration of justice" (slip op. 5). The concerns expressed there would likewise be applicable, albeit not to the same degree, to the administration of justice if the rule in Batson should be applied retroactively to cases pending on direct appeal at the time it was decided. It would obviously be highly disruptive to remand every case not yet final in which the defendant has alleged that the prosecutor exercised peremptory challenges in a discriminatory manner. On remand, the district court would be required first to reconstruct the surrounding circumstances to determine whether the defendant's proof made out a prime facie case. And, even if the lower courts were able to reconstruct events to that extent, the prosecutor would have to recall why he made a decision, sometimes several years earlier, that most probably was intuitive in the first place. The Court expressed this concern in Allen v. Hardy (slip op. 5): Where a defendant made out a prima facie case, the court then would be required to ask the prosecutor to explain his reasons for the challenges, a task that would be impossible in virtually every case since the prosecutor, relying on Swain, would have had no reason to think such an explanation would someday be necessary. As a result, many convictions would have to be vacated, and retrials would be hampered by the passage of time (ibid). The need for an evidentiary hearing in every case (except those in which the prosecutors' unexplained actions are so clearly "nondiscriminatory" as to call for no explanation) is extremely important under this Court's retroactivity cases. The administration-of-justice criterion was conceived in Stovall as a qualitative factor at least as much as a quantitative one. The Court relied (388 U.S. at 300) on the fact that hearings would have to be held, with problems of "unavailability of witnesses and dim memories," in holding that the lineup cases were not retroactive. On a quantitative level, the truism that retroactivity on direct review will affect fewerk cases than retroactivity on collateral review cannot suffice to reverse this Court's conclusion in Allen v. Hardy, unless the Court is prepared to hold all new constitutional decisions retroactive on direct review. In any event, we note that there are at least seven federal cases pending in this Court on direct review raising Batson issues. /30/ Those cases demonstrate the fallacy of amici's suggestion (Br. 31) that the impact of retroactivity will be significantly mitigated because of the failure of litigants to preserve a Batson issue, for the preservation vel non of a Batson issue has already become a subject of dispute in many of those cases, /31/ and extensive litigation over that question may be expected if Batson is held retroactive. There are undoubtedly many more such cases pending in the courts of appeals, which, like this Court, presumably have held such cases pending the decisions in Batson and in this case. We have not attempted to ascertain the number of state cases in which convictions could be upset if Batson were held retroactive, but that number is surely much larger than the number of federal cases. For those reasons, retroactivity will have a significant negative impact on the administration of justice - the third Stovall factor - even if retroactivity is limited to cases on direct review. CONCLUSION The judgment of the court of appeals should be affirmed. Respectfully submitted. CHARLES FRIED Solicitor General STEPHEN S. TROTT Assistant Attorney General WILLIAM C. BRYSON Deputy Solicitor General ROY T. ENGLERT, JR. Assistant to the Solicitor General RANGELEY WALLACE Attorney AUGUST 1986 /1/ In rejecting petitioner's argument concerning the prosecutor's contact with the jury clerk, the court of appeals stated that the district court thoroughly explained the matter to the jury (J.A. 60). The statement of the court of appeals is incorrect. The court apparently meant that the district court had explained this issue to counsel. There was, of course, no reason to discuss the matter with the jury. /2/ We do not defend the prosecutor's telephone contact with the jury clerk. No matter what the prosecutor actually said, he should not have made the call in the first place. Nonetheless, the court of appeals was correct to observe that the prosecutor's call to the jury clerk had no effect on the composition of the jury and therefore provided no basis for reversal of the conviction. In any event, even if it gave rise to an independent ground for reversal, the prosecutor's indefensible conduct in this case would have no direct impact on the analysis of the more general question on which this Court granted certiorari -- whether Batson should be given retroactive effect in cases pending on direct review. /3/ By contrast, in cases on collateral review Justice Harlan would have followed an approach of automatic nonretroactivity with certain limited exceptions (see Mackey v. United States, 401 U.S. 667, 681-695 (1971) (separate opinion)). The Court has not adopted this aspect of Justice Harlan's theory but instead has continued to apply the Stovall factors on collateral review (see Allen v. Hardy, slip op. 3; Shea v. Louisiana, slip op. 7; id. at 1 (Rehniquist, J., dissenting); Solem v. Stumes, 465 U.S. 638, 643 (1984); Johnson, 456 U.S. at 563-564 (Brennan, J., concurring)). /4/ Justice Harlan dissented on this ground in Desist v. United States, 394 U.S. 244 (1969); Williams v. United States, 401 U.S. 667 (1971). Desist held nonretroactive on direct appeal the decision in Katz v. United States, 389 U.S. 347 (1967), which specifically rejected Goldman v. United States, 316 U.S. 129 (1942), and Olmstead v. United States, 277 U.S. 438 (1928). Williams held nonretroactive on direct appeal the decision in Chimel v. California, 395 U.S. 752 (1969), which overruled United States v. Rabinowitz, 339 U.S. 56 (1950), and Harris v. United States, 331 U.S. 145 (1947). Mackey held nonretroactive on direct appeal the decisions in Marchetti v. United States, 390 U.S. 39 (1968), and Grosso v. United States, 390 U.S. 62 (1968), which overruled United States v. Kahriger, 345 U.S. 22 (1953), and Lewis v. United States, 348 U.S. 419 (1955). Commentators have been quick to point out this divergence between Justice Harlan's approach and that taken in Johnson. See, e.g., Haddad, The Finality Distinction in Supreme Court Retroactivity Analysis: An Inadequate Surrogate for Modification of the Scope of Federal Habeas Corpus, 79 Nw. U.L. Rev. 1062, 1075 (1985); Note, Retroactivity and the Exclusionary Rule: A Unifying Approach, 97 Harv. L. Rev. 961, 970 & n.72 (1984). /5/ In Shea the Court held that the rule of Edwards v. Arizona, 451 U.S. 477 (1981), should be applied retroactively to cases pending on direct appeal. Earlier, the Court had held in Stumes that Edwards would not be applied retroactively on collateral review but had noted that because Edwards "did not overrule any prior decision or transform standard practice* * *it (was) not the sort of 'clear break' case that is almost automatically nonretroactive" (465 U.S. at 647). /6/ In a related argument, amici deny that Swain contained any such advice. Amici suggest that any prosecutor who relied on Swain was not engaging in legitimate conduct but was knowingly engaged in constitutionally prohibited discrimination with the knowledge that Swain imposed an evidentiary burden that insulated that conduct from review. They assert that any reliance on Swain was therefore illegitimate. Br. 25-28. This argument is also featured in the brief for petitioner in No. 85-5221. We show at pp. 29-40, infra, that this argument misconstrues Swain. For present purposes, it suffices to note that there is no basis in the Johnson analysis to conclude that there is a relevant distinction between cases overruling an evidentiary burden imposed in a prior case and cases overruling a norm of permitted conduct stated in a prior case. /7/ Although Olmstead is the clearest example, there are certainly other decisions whose overruling was predictable. For example, Justice Brennan noted that by the time of Chimel v. California, 395 U.S. 752 (1969), the doctrinal underpinnings of Harris v. United States, 331 U.S. 145 (1947), and United States v. Rabinowitz, 339 U.S. 56 (1950), permitting broad searches incident to lawful arrests, "had long since been rejected" (Williams v. United States, 401 U.S. 646, 661 (1971) (concurring in the result)). Neither Justice Brennan nor the Williams plurality, however, found the erosion of the doctrinal underpinnings of Harris and Rabinowitz a sufficient basis to give Chimel retroactive effect. See also, Haddad, "Retroactivity Should Be Rethought": A Call for the End of the Linkletter Doctrine, 60 J. Crom. L., Criminology & Police Sci. 417, 435-436 & n.174 (1969) (noting other precedents that were held sufficient to defeat retroactivity of the decisions overruling them even though the precedents were "under attack" and inevitably to be overruled). /8/ The foreshadowing of the overruling of Olmstead stands in marked contrast to this Court's pre-Batson treatment of Swain. Dissenting in Desist, Justice Fortas asserted that Olmstead was "moribund" long before its overruling and that "there (could) be no doubt whatever that if the present cases had been presented to this Court a day, a year, or a number of years before Katz, we would have held that the petitioner's constitutional rights had been violated" (394 U.S. at 277, 279). Swain, however, was only 20 years old at the time Batson arose, and Members of the Court had referred to Swain with apparent approval in fairly recent times. See Regents of the University of California v. Bakke, 438 U.S. 265, 319 n.53 (1978) (opinion of Powell, J.) ("Universities, like the prosecutor in Swain, may make individual decisions, in which ethnic background plays a part * * *."); Apodaca v. Oregon, 406 U.S. 404, 413 (1972) (plurality opinion) (relying on Swain). /9/ If any foreshadowing of Batson were relevant, we would by no means concede that the outcome in Batson was foreshadowed in a meaningful way. Only two Justices had stated that they would overrule Swain. See McCray v. New York, 461 U.S. 961, 963-970 (1983) (Marshall, J., dissenting from denial of certiorari). The three other Justices who had recently spoken to the issue had indicated no view on the merits, saying only that they considered the issue sufficiently important to give plenary consideration at some future date after "further consideration of the substantive and procedural ramifications of the problem by other courts" (id. at 962 (opinion of Stevens, J.)). Moreover, the suggestion that Batson was foreshadowed by lower courts' decision resting on bases other than the Equal Protection Clause misses the point. Those courts based their rulings on grounds other than equal protection precisely because they believed that Swain was not open to challenge. For example, the Supreme Court of California "assume(d) that if the present question were before the high court it would reaffirm Swain." People v. Wheeler, 22 Cal. 3d 258, 285, 583 P.2d 748, 767, 148 Cal. Rptr. 890, 908 (1978). See also Booker v. Jabe, 775 F.2d 762, 766 (6th Cir. 1985), vacated and remanded sub nom. Michigan v. Booker, No. 85-1028 (June 30, 1986); McCray v. Abrams, 750 F.2d 1113, 1119, 1124 (2d Cir. 1984), vacated and remanded, No. 84-1426 (June 30, 1986). /10/ The lower courts that have addressed the question whether Batson should be afforded retroactive application have held that is should not, in part because Batson overruled Swain. In Esquival v. McCotter, 791 F.2d 350 (1986), the Fifth Circuit, declining to apply the rule in Batson retroactively to a collateral federal proceeding, observed that Batson "overrules Swain" and "qualifies as 'a clear break with the past' which should not be applied retroactively" (791 F.2d at 352). In Simpson v. Massachusetts, no. 85-1931 (June 26, 1986), the First Circuit declined to apply Batson retroactively to a collateral attack on a criminal conviction. The court of appeals quoted with approval from the dissent of the Chief Justice in Batson, to the effect that Batson "clearly 'overrule(s) (a) prior decision' and drastically 'transform(s) standard practice'" (slip op. 5 (quoting Batson, slip op. 21 (Burger, C.J., dissenting) (quoting Stumes, 465 U.S. at 647))). Finally, the Supreme Court of North Carolina held, in State v. Jackson, 343 S.E.2d 814 (1986), that Batson did not apply retroactively, even on direct appeal, in part because Batson was a "clear break" with past precedent in which the Court "explicitly rejected the prior Swain requirement and unequivocally overruled Swain" (343 S.E.2d at 825). The Eleventh Circuit also recently noted that Batson "substantially altered the evidentiary burden, formerly prescribed by Swain v. Alabama, * * * that a criminal defendant must meet to make out an equal protection claim." Fleming v. Kemp, 794 pf.2d 1478, 1483 (11th Cir. 1986). /11/ The Williams plurality took the view that this factor is sufficient to compel retroactive application even in the face of "good-faith reliance by state or federal authorities on prior constitutional law" (401 U.S. at 653). According to the Johnson majority, however, the purpose of the rule is "less critical than the Court's express threshold determination" that the new rule so changes the law that it should not be applied retroactively (457 U.S. at 549). /12/ Petitioner and amici do suggest that the impact of Batson on the integrity of the truthfinding process is more fundamental than this Court recognized in Allen, but it is only the adjectives in their briefs, not the research that they cite, that goes beyond what the Court said. Moreover, to reach the conclusion that the rule of Batson would have changed the outcome of a signigicant number of trials, petitioner and amici must indulge the gross generalization -- not supported by their empirical data -- that minority jurors will regularly be the most vehement advocates to acquit a defendant of the same race. But the rule of Batson is based on the unwillingness of courts to accept that kind of gross generalization (slip op. 17; see also Saltzburg & Powers, Peremptory Challenges And The Clash Between Impartiality And Group Representation, 41 Md. L. Rev. 337, 371 (1982)). Finally, petitioner and amici take no account of the fact that the prosecutor who is forbidden to use a peremptory challenge on a minority juror still has that peremptory challenge available to strike another juror he regards as likely to favor the defendant. Thus, the marginal impact of the minority juror will be limited not just because he is one of a number of jurors, but also because he replaces the person who seems most predisposed to vote the same way that -- by petitioner's and amici's assumption -- the minority juror is predisposed to vote. /13/ Petitioner (Br. 17-18) does suggest that group voir dire conducted by a judge is not as great a safeguard as voir dire conducted by counsel, and he asserts that a jury instruction "may not be sufficient." The first point is merely a quibble with the method used by particular courts to conduct voir dire, and the second point does nothing to undercut the Court's reasoning in Allen v. Hardy. The Court did not suggest that jury instructions, by themselves, are always sufficient to do away with potential bias on every jury. But the Court did hold that the jury instructions, combined with the other safeguards, create a high probability of freedom from bias. /14/ We do not challenge the assumption that the Batson rule has some bearing on the truthfinding function in a criminal trial. Nonetheless, some comment on the literature cited in the brief of petitioner in No. 85-5221 is appropriate, lest the Court be led to believe that there is an overwhelming and unanimous body of social science research showing that a jury with minority representation will reach singificantly different results from jury without minority representation. The 15 sources cited in petitioner's brief in No. 85-5221 (at 22 n.7) were all cited to the Court in an appendix to a brief in Batson under the heading "Sociological Literature Concerning Blacks and the Jury System" (Br. of the Lawyers Committee for Civil Rights Under Law as Amicus Curiae at A2-A3, Batson). As that heading implies, the studies cover a large range of subjects related to blacks and the jury system. Some of the literature empirically compares the decisionmaking of white and black jurors (e.g., Bernard, Interaction Between the Race of the Defendant and That of Jurors in Determining Verdicts, 5 Law and & Psychology Rev. 103 (1979); Ugwuegbu, Racial and Evidental Factors in Jurors Attribution of Legal Responsibility, 15 J. Experimental Soc. Psychology 133 (1979)). Much of it does not (e.g., H. Kalven & H. Zeisel, The American Jury 196-198 (1966) (profiling criminal defendants without any reference to jury decisionmaking); id. at 210-213 (discussing jurors' perception of categories of criminal defendants, including blacks, without regard to race of jurors); Gerard & Terry, Discrimination Against Negroes in the Administration of Criminal Law in Missouri, 1970 Wash. U.L.Q. 415 (discussing race of defendant but not race of jurors)). Only one article even arguable supports the assertion (85-5221 Pet. Br. 22 n.7) that a "pro-prosecution effect" resulting from the exclusion of blacks from juries is especially pronounced when "the government's evidence is insubstantial and the defendant is black." See Ugwuegbu, supra, 15 J. Experimental Soc. Psychology at 139-140. /15/ The Court noted that it had already "denied retroactive application to Griffin v. California, (380 U.S. 609 (1965), despite the fact that comment on the failure to testify may sometimes mislead the jury concerning the reasons why the defendant has refused to take the witness stand" (384 U.S. at 729). /16/ The Supreme Court of North Carolina, unanimous on this issue, analyzed this Court's cases and reached the same conclusion in State v. Jackson, 343 S.E.2D 814, 825-826 (1986) ("Those rulings which can be said to substantially affect the basic truth-finding process have generally been those which either gave an accused the ability to effectively present his case * * or those which have placed restrictions on the prosecution's ability to present improper evidence against an accused. * * * We conclude that while the ruling in Batson may tend to incidentally avoid unfairness in the trial, it is not one which bears substantially on the truth-finding process.") /17/ "It * * * would demean the constitutional rights to indictment and trial by a jury to assert that those guarantees do not play some role in assuring the integrity of the truthdetermining process" (413 U.S. at 681 (plurality opinion)). /18/ See also Hankerson v. North Carolina, 432 U.S. 233, 243 (1977) (emphasis added by the Court) ("'"(w)here the major purpose of new constitutional doctrine is to overcome an aspect of the criminal trial that substantially impairs its truthfinding function and so raises serious questions about the accuracy of guilty verdicts in past trials, the new rule (is) given complete retroactive effect"'") (quoting Ivan V. v. City of New York, 407 U.S. 203, 204 (1972) (per curiam) (quoting Williams, 401 U.S. at 653 (plurality opinion))); Reed v. Ross, 468 U.S. 1, 5 (1984) (same). /19/ First, in both Taylor and Batson, the Court observed that the jury safeguards an accused against the arbitrary exercise of power by prosecutor or judge (Taylor, 419 U.S. at 530; Batson, slip op. 6). Second, in both Taylor and Batson the Court relied on the fact that community participation is critical to public confidence in our system of justice (Taylor, 419 U.S. at 530; Batson, slip op. 6). Third, despite its moorings in the Sixth Amendment, Taylor overruled a prior decision upholding exclusion of women from juries against an Equal Protection Clause challenge (Hoyt v. Florida, 368 U.S. 57 (1961)), and the Court plainly brought traditional equal protection principles to bear on the problem in the course of holding that more than a rational basis was required for systematic exclusion on the basis of gender of 53% of the citizens eligible for jury service (see generally 419 U.S. at 531-535 & nn.15-17). Batson, of course, was rooted in the proposition that the Equal Protection Clause does not permit the state to make judgments about jurors on the basis of their race (see slip op. 17). /20/ Just five days after Batson, the Court observed in another context that "an impartial jury consists of nothing more than 'jurors who will conscientiously apply the law and find the facts.'" (Lockhart v. McCree, No. 84-1865 (May 5, 1986), slip op. 14 (quoting Wainwright v. Witt, No. 83-1427 (Jan. 21, 1085, slip op. 10). There is no reason to presume that pre-Batson juries failed this test of impartiality. Batson did not correct procedures that could result in the inclusion of jurors who could not fairly judge a defendant. Cf. Turner v. Murray, No. 84-6646 (Apr. 30, 1986). The procedures changed in Batson were altered because they might result in the exclusion through peremptory challenges, on the basis of race, of jurors who could judge a defendant just as fairly as other jurors. As in Lockhart, the jurors selected under the Swain procedures in any particular case "could have ended up on the jury through the 'luck of the draw'" (Lockhart, slip op. 14), and there would have been no basis to question their ability to adjudge guilt and innocence impartially. And the impartiality of past juries is not called into question just because empirical studies might show that some decisions might have been different if there had been minority representation on those juries (see Br. of Amici Curiae 19-24). Such studies fall far short of demonstrating that juries without minority representation decide cases unfairly, just as it was insufficient in Lockhart to show through empirical studies that one type of jury was more "conviction-prone" than another (Lockhart, slip op. 9). /21/ In a footnote, the plurality quoted some of the formulations from past cases of the standard for determining whether a rule had a sufficiently fundamental impact on truthfinding to meet the first Stovall test (Brown, 447 U.S. at 329 n.6). Despite amici's claim that the jury's judgement is not "limited to the binary alternatives of 'guilt' or 'innocence'" (Br. 13), the cases quoted in the Brown footnote make it abundantly clear that whether or not the jury's function is so limited, it is the determination of guilt and innocence that concerns this Court when it decides whether to give retroactive effect to a new constitutional decision. See, e.g., Johnson v. New Jersey, 384 U.S. at 728 (referring to "'clear danger of convicting the innocent'" (citation omitted)); Roberts v. Russell, 392 U.S. 293, 295 (1968) (referring to "serious risk that the issue of guilt or innocence may not have been reiably determined"); Williams, 401 U.S. at 655 n.7 (plurality opinion) (referring to "likelihood that the results of a number of those trials were factually incorrect"); United States v. United States Coin & Currency, 401 U.S. 715, 724 (1971) (referring to "significant chance that innocent men had been wrongfully punished in the past"); Michigan v. Payne, 412 U.S. 47, 61-62 (1973) (Marshall, J., dissenting) (referring to "a rule that was central to the process of determining guilt or innocence"). /22/ The plurality distinguished DeStefano and Daniel because Duncan v. Louisiana, supra, and Taylor v. Louisiana, supra, did not rest on the premise that trials held before a judge alone or a jury from which women were systematically excluded are necessarily unfair (Brown, 447 U.S. at 335 n.13). Because Burch decided the number of jurors whose concurrence was constitutionally required to make a verdict fair, it necessarily did rest on the premise that every criminal verdict rendered by a nonunanimous six-person jury was unfair. The Brown plurality also distinguished DeStefano because DeStefano involved the existence of other safeguards to the factfinding process and because states denying jury trials before Duncan, unlike states permitting nonunanimous six-person jury verdicts before Burch, had acted in justifiable reliance on earlier opinions of the Court (ibid). In these respects as well, Batson resembles Duncan and Taylor far more than Burch. /23/ Amici conceded the reasonableness of this interpretation of Swain in their brief in Batson (see Br. Amici Curiae of the NAACP Legal Defense and Eductional Fund, Inc., the American-Jewish Committee, and the American Jewish Congress at 48-49 ("the prosecutor who is conscientious in his desire to obey the Constitution nevertheless sees nothing unconstitutional about peremptorily challenging blacks qua blacks in particular cases, so long as he does not do it in all cases")). Other amici in Batson conceded not only that that reading of Swain was reasonable, but that is was correct interpretation of the case (Br. of Michael McCray, the New York Civil Liberties Union, and the American Civil Liberties Union as Amici Curiae at 36 ("Swain * * * hold(s) that the disqualification of blacks from criminal juries on racial grounds is not discrimination")). /24/ Several analyses of Swain have focused on its failure to hold even that systematic exlcusion of blacks was an equal protection violation. See e.g., United States v. Newman, 549 F.2d 240, 248 (2nd Cir. 1977); United States v. Pearson, 448 F.2d 1207, 1215 & n.17 (5th Cir. 1971); People v. Wheeler, 22 Cal. 3d 258, 284, 583 P.2d 748, 766-767, 148 Cal. Rptr. 890, 907-908 (1978); Kuhn, Jury Discrimination: The Next Phase, 41 S. Cal. L. Rev. 235, 289 & n.221 (1968). /25/ The Court in Baston held that "the prosecutor may not rebut the defendant's prima facie case of discrimination by stating merely that he challenged jurors of the defendant's race on the assumption or his intuitive judgment -- that they would be partial to the defendant because of their shared race" (slip op. 17). As the Court observed in Swain, however, a prosecutor's exercise of peremptory challenges against jurors of the same race as the defendant is not based on an assumption or intuitive judgment that "a juror of a particular race or nationality" would be partial, but on the assumption or intuitive judgment that "one from a different group is less likely to be" (380 U.S. at 221 (emphasis added)). We take Batson to mean that a prosecutor's verbatim recitation of the Swain explanation for the use of peremptory challenges on the basis of race would not be sufficient to rebut a prima facie case. If we are right, then Batson overruled a critical holding of Swain that had nothing to do with burdens of proof. /26/ Nor could a prosecutor be faulted for believing that an interpretation of Swain that the Court declined to review was still good law, even in the face of "flag(s) * * * (that) warned that (Swain) was in stormy weather" (Linkletter, 381 U.S. at 637; see also pp. 12-15, supra). Like the admission of illegally seized evidence (see Linkletter), prosecutorial comment on defendant's exercise of Fifth Amendment rights (see Tehan v. United States ex rel. Shott, 382 U.S. 406 (1966)), wiretaps conducted without probable cause (see Desist), denial of the right to jury trial in serious criminal cases (see DeStefano), and systematic exclusion of women from petit juries (see Daniel), the practice permitted by Swain may now seem profoundly wrong to a majority of the Court. That is true of any decision that the Court overrules. See Haddad, supra note 7, 60 J. Crim. L. Criminology & Police Sci. at 429 ("the Supreme Court must damn with vehemence an old rule when it raises a new safeguard to the level of a constitutional right"). But the Court's retroactivity decisions have never departed from the principle that government officials may legitimately rely on a decision of the Court, however "wrong," until it is overruled. /27/ Numerous other law review commentators have read Swain to permit the exercise of peremptory challenges in individual cases on the basis of race. See, e.g., Imlay, Federal Jury Reformation: Saving a Democratic Institution, 6 Loy. L.A.L. Rev. 247, 269-270 (1973); Johnson, Black Innocence and the White Jury, 83 Mich. L. Rev. 1611, 1667 (1985); Note, Systematic Exclustion of Cognizable Groups by Use of Peremptory Challenges, 11 Fordham Urban L.J. 927, 934 (1938); Note, The Jury: A Reflection of the Prejudices of the Community, 20 Hastings L.J. 1417, 1431 (1969); Comment, The Sixth Amendment: Limiting the Use of Peremptory Challenges, 16 J. Mar. L. Rev. 349, 350 (1983); Comment, The Prohibition of Group-Based Steroetypes in Jury Selection Procedures, 25 Vill. L. Rev. 339, 340 (1980); Note, Limiting the Peremptory Challenge: Representation of Groups on Petit Juries, 86 Yale L.J. 1715, 1721-1722, 1724 (1977); Case Comment, 21 Am. Crim. L. Rev. 477, 483 (1984); Casenote, 21 B.C.L. Rev. 1197, 1201, 1202-1203 (1980). /28/ The 1986 supplement to the annotation lists another seven federal and 68 state cases in support of the same proposition ( 9 A.L.R.3d 3-4 (Supp. 1986)). /29/ Before Batson, there were to our knowledge no federal judges and only one state-court judge who had actually adopted petitioner's reading of Swain. State v. Wiley, 144 Ariz. 525, 545-546, 698 P.2d 1244, 1263-1264 (1985) (Feldman, J., specially concurring). /30/ Orji v. United States, No. 84-6732; Tolliver v. United States, No. 85-1663; Leslie v. United States, No. 85-1961; Evans v. United States, No. 85-6549; Grandison v. United States, No. 85-6645; Kelly v. United States, No. 86-5095; Allen v. United States, No. 86-5255. /31/ For example, in Nos. 85-6549 and 85-6645, the petitions raise only a Sixth Amendment issue, but petitioners claimed entitlement to the benefit of Batson in supplemental filings. In No. 85-1961, petitioner claims (Pet. 7 n.1) that he "reserve(d)" constitutional issues for this Court's review, but the only judges in the court of appeals to comment on the matter had suggested the opposite (United States v. Leslie, 783 F.2d at 566 (Williams, J., dissenting) ("the constitutional issue of Swain was not even raised by the defendant in this case")).