dissenting.
This case involves a straightforward application of the rule announced in Brady v. Maryland, 373 U. S. 83 (1963), a case involving nondisclosure of material evidence by the prosecution in response to a specific request from the defense. I agree that the Court of Appeals misdescribed that rule, see ante, at 674-678, but I respectfully dissent from the Court’s unwarranted decision to rewrite the rule itself.
As the Court correctly notes at the outset of its opinion, ante, at 669, the holding in Brady was that “the suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment.” 373 U. S., at 87. We noted in United States v. Agurs, 427 U. S. 97, 103 (1976), that the rule of Brady arguably might apply in three different situations involving the discovery, after trial, of evidence that had been known prior to trial to the prosecution but not to the defense. Our holding in Agurs was that the Brady rule applies in two of the situations, but not in the third.
The two situations in which the rule applies are those demonstrating the prosecution’s knowing use of perjured testimony, exemplified by Mooney v. Holohan, 294 U. S. 103 (1935), and the prosecution’s suppression of favorable evidence specifically requested by the defendant, exemplified by Brady itself. In both situations, the prosecution’s deliberate nondisclosure constitutes constitutional error — the conviction must be set aside if the suppressed or perjured evidence was “material” and there was “any reasonable likelihood” that it “could have affected” the outcome of the trial. 427 U. S., at 103.1 See Brady, supra, at 88 (“would tend to exculpate”); *710accord, United States v. Valenzuela-Bernal, 458 U. S. 858, 874 (1982) (“reasonable likelihood”); Giglio v. United States, 405 U. S. 150, 154 (1972) (“reasonable likelihood”); Napue v. Illinois, 360 U. S. 264, 272 (1959) (“may have had an effect on the outcome”). The combination of willful prosecutorial suppression of evidence and, “more importantly,” the potential “corruption of the truth-seeking function of the trial process” requires that result. 427 U. S., at 104, 106.2
In Brady, the suppressed confession was inadmissible as to guilt and “could not have affected the outcome” on that issue. 427 U. S., at 106. However, the evidence “could have affected Brady’s punishment,” and was, therefore, “material on the latter issue but not on the former.” Ibid. Material*711ity was thus used to describe admissible evidence that “could have affected” a dispositive issue in the trial.
The question in Agurs was whether the Brady rule should be extended, to cover a case in which there had been neither perjury nor a specific request — that is, whether the prosecution has some constitutional duty to search its files and disclose automatically, or in response to a general request, all evidence that “might have helped the defense, or might have affected the outcome.” 427 U. S., at 110.3 Such evidence would, of course, be covered by the Brady formulation if it were specifically requested. We noted in Agurs, however, that because there had been no specific defense request for the later-discovered evidence, there was no notice to the prosecution that the defense did not already have that evidence or that it considered the evidence to be of particular value. 427 U. S., at 106-107. Consequently, we stated that in the absence of a request the prosecution has a constitutional duty to volunteer only “obviously exculpatory . . . evidence.” Id., at 107. Because this constitutional duty to disclose is different from the duty described in Brady, it is not surprising that we developed a different standard of materiality in the Agurs context. Necessarily describing the “inevitably imprecise” standard in terms appropriate to post-trial review, we held that no constitutional violation occurs in the absence of a specific request unless “the omitted evidence creates a reasonable doubt that did not otherwise exist.” Id., at 108, 112.4
*712What the Court ignores with regard to Agurs is that its analysis was restricted entirely to the general or no-request context.5 The “standard of materiality” we fashioned for the purpose of determining whether a prosecutor’s failure to volunteer exculpatory evidence amounted to constitutional error was and is unnecessary with regard to the two categories of prosecutorial suppression already covered by the Brady rule. The specific situation in Agurs, as well as the circumstances of United States v. Valenzuela-Bernal, 458 U. S. 858 (1982) and Strickland v. Washington, 466 U. S. 668 (1984), simply falls “outside the Brady context.” Ante, at 681.
But the Brady rule itself unquestionably applies to this case, because the Government failed to disclose favorable evidence that was clearly responsive to the defendant’s specific *713request. Bagley’s conviction therefore must be set aside if the suppressed evidence was “material” — and it obviously was, see n. 1, swpra — and if there is “any reasonable likelihood” that it could have affected the judgment of the trier of fact. Our choice, therefore, should be merely whether to affirm for the reasons stated in Part I of Justice Marshall’s dissent, or to remand to the Court of Appeals for further review under the standard stated in Brady. I would follow the latter course, not because I disagree with Justice Marshall’s analysis of the record, but because I do not believe this Court should perform the task of reviewing trial transcripts in the first instance. See United States v. Hasting, 461 U. S. 499, 516-517 (1983) (Stevens, J., concurring in judgment). I am confident that the Court of Appeals would reach the appropriate result if it applied the proper standard.
The Court, however, today sets out a reformulation of the Brady rule in which I have no such confidence. Even though the prosecution suppressed evidence that was specifically requested, apparently the Court of Appeals may now reverse only if there is a “reasonable probability” that the suppressed evidence “would” have altered “the result of the [trial].” Ante, at 682, 684. According to the Court this single rule is “sufficiently flexible” to cover specific as well as general or no-request instances of nondisclosure, ante, at 682, because, at least in the view of Justice Blackmun and Justice O’Connor, a reviewing court can “consider directly” under this standard the more threatening effect that nondisclosure in response to a specific defense request will generally have on the truth-seeking function of the adversary process. Ante, at 683 (opinion of Blackmun, J.).6
*714I cannot agree. The Court’s approach stretches the concept of “materiality” beyond any recognizable scope, transforming it from merely an evidentiary concept as used in Brady and Agurs, which required that material evidence be admissible and probative of guilt or innocence in the context of a specific request, into a result-focused standard that seems to include an independent weight in favor of affirming convictions despite evidentiary suppression. Evidence favorable to an accused and relevant to the dispositive issue of guilt apparently may still be found not “material,” and hence suppressible by prosecutors prior to trial, unless there is a reasonable probability that its use would result in an acquittal. Justice Marshall rightly criticizes the incentives such a standard creates for prosecutors “to gamble, to play the odds, and to take a chance that evidence will later turn out not to have been potentially dispositive.” Ante, at 701.
Moreover, the Court’s analysis reduces the significance of deliberate prosecutorial suppression of potentially exculpatory evidence to that merely of one of numerous factors that “may” be considered by a reviewing court. Ante, at 683 (opinion of Blackmun, J.). This is not faithful to our statement in Agurs that “[w]hen the prosecutor receives a specific and relevant request, the failure to make any response is seldom, if ever, excusable.” 427 U. S., at 106. Such suppression is far more serious than mere nondisclosure of evidence in which the defense has expressed no particular interest. A reviewing court should attach great significance to silence in the face of a specific request, when responsive evidence is later shown to have been in the Government’s possession. Such silence actively misleads in the same way as would an affirmative representation that exculpatory evidence does not exist when, in fact, it does (i e., perjury)— indeed, the two situations are-aptly described as “sides of a single coin.” Babcock, Fair Play: Evidence Favorable to *715an Accused and Effective Assistance of Counsel, 34 Stan. L. Rev. 1133, 1151 (1982).
Accordingly, although the judgment of the Court of Appeals should be vacated and the case should be remanded for further proceedings, I disagree with the Court’s statement of the correct standard to be applied. I therefore respectfully dissent from the judgment that the case be remanded for determination under the Court’s new standard.
1 do not agree with the Court’s reference to the “constitutional error, if any, in this case,” see ante, at 678 (emphasis added), because I believe a violation of the Brady rule is by definition constitutional error. Cf. United *710States v. Agurs, 427 U. S., at 112 (rejecting rule making “every nondisclosure . . . automatic error” outside the Brady specific request or perjury contexts). As written, the Brady rule states that the Due Process Clause is violated when favorable evidence is not turned over “upon request” and “the evidence is material either to guilt or punishment.” Brady v. Maryland, 373 U. S., at 87. As Justice Marshall’s explication of the record in this case demonstrates, ante, at 685-692, the suppressed evidence here was not only favorable to Bagley, but also unquestionably material to the issue of his guilt or innocence. The two witnesses who had signed the undisclosed “Contract[s] for Purchase of Information” were the only trial witnesses as to the two distribution counts on which Bagley was convicted. On cross-examination defense counsel attempted to undercut the witnesses’ credibility, obviously a central issue, but had little factual basis for so doing. When defense counsel suggested a lack of credibility during final argument in the bench trial, the trial judge demurred, because “I really did not get the impression at all that either one or both of these men were trying at least in court here to make a ease against the defendant.” A finding that evidence showing that the witnesses in fact had a “direct, personal stake in respondent’s conviction,” ante, at 683, was nevertheless not “material” would be egregiously erroneous under any standard.
“A prosecution that withholds evidence on demand of an accused which, if made available, would tend to exculpate him or reduce the penalty helps shape a trial that bears heavily on the defendant. That casts the prosecutor in the role of an architect of a proceeding that does not comport with standards of justice . . . .” Brady, supra, at 87-88.
“[W]e conclude that there is no significant difference between cases in which there has been merely a general request for exculpatory matter and eases, like the one we must now decide, in which there has been no request at all... .
“We now consider whether the prosecutor has any constitutional duty to volunteer exculpatory matter to the defense, and if so, what standard of materiality gives rise to that duty.” 427 U. S., at 107.
“The proper standard of materiality must reflect our overriding concern with the justice of the finding of guilt. Such a finding is permissible only *712if supported by evidence establishing guilt beyond a reasonable doubt. It necessarily follows that if the omitted evidence creates a reasonable doubt that did not otherwise exist, constitutional error has been committed.” Id,., at 112 (footnote omitted).
We also held in Agurs that when no request for particular information is made, post-trial determination of whether a failure voluntarily to disclose exculpatory evidence amounts to constitutional error depends on the “character of the evidence, not the character of the prosecutor.” Id., at 110. Nevertheless, implicitly acknowledging the broad discretion that trial and appellate courts must have to ensure fairness in this area, we noted that “the prudent prosecutor will resolve doubtful questions in favor of disclosure.” Id., at 108. Finally, we noted that the post-trial determination of reasonable doubt will vary even in the no-request context, depending on all the circumstances of each case. For example, “if the verdict is already of questionable validity, additional evidence of relatively minor importance might be sufficient to create a reasonable doubt.” Id., at 113.
See ante, at 678 (“Our starting point is the framework for evaluating the materiality of Brady evidence established in United States v. Agurs”), ante, at 681 (referring generally to “the Agurs standard for the materiality of undisclosed evidence”); ante, at 700 (MARSHALL, J., dissenting) (describing Agurs as stating a general rule that “there is no constitutional duty to disclose evidence unless nondisclosure would have a certain impact on the trial”). But see Babcock, Fair Play: Evidence Favorable to an Accused and Effective Assistance of Counsel, 34 Stan. L. Rev. 1133, 1148 (1982) CAgurs “distinguished” between no-request situations and the other two Brady contexts “where a pro-defense standard . . . would continue”).
I of course agree with Justice Blackmun, ante, at 679-680, n. 9, and 684, and Justice MARSHALL, ante, at 706, that our long line of precedents establishing the “reasonable likelihood” standard for use of perjured testimony remains intact. I also note that the Court plainly envisions that reversal of Bagley’s conviction would be possible on remand even under the new standard formulated today for specific-request cases. See ante, at 684.