SAWYER v. WHITLEY, WARDEN
No. 91-6382
Supreme Court of the United States
Argued February 25, 1992-Decided June 22, 1992
505 U.S. 333
R. Neal Walker argued the cause for petitioner. With him on the briefs were Nicholas J. Trenticosta and Sarah L. Ottinger.
Dorothy A. Pendergast argued the cause for respondent. With her on the brief was John M. Mamoulides.
Paul J. Larkin, Jr., argued the cause for the United States as amicus curiae urging affirmance. With him on the brief were Solicitor General Starr, Assistant Attorney General Mueller, Deputy Solicitor General Roberts, and Associate Deputy Attorney General McBride.*
CHIEF JUSTICE REHNQUIST delivered the opinion of the Court.
The issue before the Court is the standard for determining whether a petitioner bringing a successive, abusive, or defaulted federal habeas claim has shown he is “actually innocent” of the death penalty to which he has been sentenced so that the court may reach the merits of the claim. Robert Wayne Sawyer, the petitioner in this case, filed a second
*Douglas G. Robinson, Julius L. Chambers, George H. Kendall, and Larry W. Yackle filed a brief for the NAACP Legal Defense and Educational Fund, Inc., as amicus curiae urging reversal.
Kent Scheidegger and Charles L. Hobson filed a brief for the Criminal Justice Legal Foundation as amicus curiae urging affirmance.
In 1979-13 years ago-petitioner and his accomplice, Charles Lane, brutally murdered Frances Arwood, who was a guest in the home petitioner shared with his girlfriend, Cynthia Shano, and Shano‘s two young children. As we recounted in our earlier review of this case, Sawyer v. Smith, 497 U. S. 227 (1990), petitioner and Lane returned to petitioner‘s home after a night of drinking and argued with Arwood, accusing her of drugging one of the children. Petitioner and Lane then attacked Arwood, beat her with their fists, kicked her repeatedly, submerged her in the bathtub, and poured scalding water on her before dragging her back into the living room, pouring lighter fluid on her body and igniting it. Arwood lost consciousness sometime during the attack and remained in a coma until she died of her injuries approximately two months later. Shano and her children were in the home during the attack, and Shano testified that petitioner prevented them from leaving.1
At trial, the jury failed to credit petitioner‘s “toxic psychosis” defense, and convicted petitioner of first-degree murder. At the sentencing phase, petitioner testified that he was intoxicated at the time of the murder and remembered
Sawyer‘s conviction and sentence were affirmed on appeal by the Louisiana Supreme Court. State v. Sawyer, 422 So. 2d 95 (1982). We granted certiorari, and vacated and remanded with instructions to reconsider in light of Zant v. Stephens, 462 U. S. 862 (1983). Sawyer v. Louisiana, 463 U. S. 1223 (1983). On remand, the Louisiana Supreme Court reaffirmed the sentence. Sawyer v. State, 442 So. 2d 1136 (1983), cert. denied, 466 U. S. 931 (1984). Petitioner‘s first petition for state postconviction relief was denied. Louisiana ex rel. Sawyer v. Maggio, 479 So. 2d 360, reconsideration denied, 480 So. 2d 313 (La. 1985).3 In 1986, Sawyer filed his first federal habeas petition, raising 18 claims, all of which were denied on the merits. See Sawyer v. Butler, 848 F. 2d 582 (CA5 1988), aff‘d on rehearing en banc, 881 F. 2d 1273 (CA5 1989). We again granted certiorari and affirmed the Court of Appeals’ denial of relief. Sawyer v. Smith, supra.4
The present petition before this Court arises out of Sawyer‘s second petition for federal habeas relief. After granting a stay and holding an evidentiary hearing, the District Court denied one of Sawyer‘s claims on the merits and held that the others were barred as either abusive or successive. 772 F. Supp. 297 (ED La. 1991). The Court of Appeals granted a certificate of probable cause on the issue whether petitioner had shown that he is actually “innocent of the death penalty” such that a court should reach the merits of the claims contained in this successive petition. 945 F. 2d, at 814. The Court of Appeals held that petitioner had failed to show that he was actually innocent of the death penalty because the evidence he argued had been unconstitutionally kept from the jury failed to show that Sawyer was ineligible for the death penalty under Louisiana law. For the third time we granted Sawyer‘s petition for certiorari, 502 U. S. 965 (1991), and we now affirm.
Unless a habeas petitioner shows cause and prejudice, see Wainwright v. Sykes, 433 U. S. 72 (1977), a court may not reach the merits of: (a) successive claims that raise grounds identical to grounds heard and decided on the merits in a previous petition, Kuhlmann v. Wilson, 477 U. S. 436 (1986); (b) new claims, not previously raised, which constitute an abuse of the writ, McCleskey v. Zant, 499 U. S. 467 (1991); or (c) procedurally defaulted claims in which the petitioner failed to follow applicable state procedural rules in raising the claims, Murray v. Carrier, 477 U. S. 478 (1986). These cases are premised on our concerns for the finality of state judgments of conviction and the “significant costs of federal habeas review.” McCleskey, supra, at 490-491; see, e. g., Engle v. Isaac, 456 U. S. 107, 126-128 (1982).
In Smith v. Murray, 477 U. S. 527 (1986), we found no misсarriage of justice in the failure to examine the merits of procedurally defaulted claims in the capital sentencing context. We emphasized that the miscarriage of justice exception is concerned with actual as compared to legal innocence,
In subsequent cases, we have emphasized the narrow scope of the fundamental miscarriage of justice exception. In Dugger v. Adams, 489 U. S. 401 (1989), we rejected the petitioner‘s claim that his procedural default should be excused because he had shown that he was actually innocent. Without endeavoring to define what it meant to be actually innocent of the death penalty, we stated that “[d]emonstrating that an error is by its nature the kind of error that might have affected the accuracy of a death sentence is far from demonstrating that an individual defendant probably is ‘actually innocent’ of the sentence he or she received.” Id., at 412, n. 6. Just last Term in McCleskey v. Zant, supra, at 502, we held that the “narrow exception” for miscarriage of justice was of no avail to the petitioner because the constitutional violation, if it оccurred, “resulted in the admission at trial of truthful inculpatory evidence which did not affect the reliability of the guilt determination.”
The present case requires us to further amplify the meaning of “actual innocence” in the setting of capital punishment. A prototypical example of “actual innocence” in a colloquial sense is the case where the State has convicted the wrong person of the crime. Such claims are of course regularly made on motions for new trial after conviction in both state and federal courts, and quite regularly denied because the evidence adduced in support of them fails to meet the rigorous standards for granting such motions. But in rare instances it may turn out later, for example, that another person has credibly confessed to the crime, and it is evident
It is more difficult to develop an analogous framework when dealing with a defendant who has been sentenced to death. The phrase “innocent of death” is not a natural usage of those words, but we must strive to construct an analog to the simpler situation represented by the case of a noncapital defendant. In defining this analog, we bear in mind that the exception for “actual innocence” is a very narrow exception, and that to make it workable it must be subject to determination by relatively objective standards. In the every day context of capital penalty proceedings, a federal district judge typically will be presented with a successive or abusive habeas petition a few days before, or even on the day of, a scheduled execution, and will have only a limited time to determine whether a petitioner has shown that his case falls within the “actual innocence” exception if such a claim is made.7
Since our decision in Furman v. Georgia, 408 U. S. 238 (1972), our Eighth Amendment jurisprudence has required those States imposing capital punishment to adopt procedural safeguards protecting against arbitrary and capricious impositions of the death sentence. See, e. g., Gregg v. Georgia, 428 U. S. 153 (1976); Proffitt v. Florida, 428 U. S. 242 (1976); Jurek v. Texas, 428 U. S. 262 (1976). In response, the States have adopted various narrowing factors that limit the
But once eligibility for the death penalty has been established to the satisfaction of the jury, its deliberations assume a different tenor. In a series of cases beginning with Lockett v. Ohio, 438 U. S. 586, 604 (1978), we have held that the
Considering Louisiana law as an example, then, there are three possible ways in which “actual innocence” might be defined. The strictest definition would be to limit any showing to the elements of the crime which the State has made a capital offense. The showing would have to negate an essential element of that offense. The Solicitor General, filing as amicus curiae in support of respondent, urges the Court to adopt this standard. We reject this submission as too narrow, because it is contrary to the statement in Smith that the concept of “actual innocence” could be applied to mean “innocent” of the death penalty. 477 U. S., at 537. This statement suggested a more expansive meaning to the term of “actual innocence” in a capital case than simply innocence of the capital offense itself.
The most lenient of the three possibilities would be to allow the showing of “actual innocence” to extend not only to the elements of the crime, but also to the existence of aggravating factors, and to mitigating evidence that bore not on the defendant‘s eligibility to receive the death penalty, but only on the ultimate discretionary decision between the death penalty and life imprisonment. This, in effect, is what petitioner urges upon us. He contends that actual innocence of the death penalty exists where “there is a ‘fair probability’ that the admission of false evidence, or the preclusion of true mitigating evidence, [caused by a constitutional error] resulted in a sentence of death.” Brief for Petitioner 18 (cita-
Insofar as petitioner‘s standard would include not merely the elements of the crime itself, but the existence of aggravating circumstances, it broadens the extent of the inquiry but not the type of inquiry. Both the elements of the crime and statutory aggravating circumstances in Louisiana are
But we reject petitioner‘s submission that the showing should еxtend beyond these elements of the capital sentence to the existence of additional mitigating evidence. In the first place, such an extension would mean that “actual innocence” amounts to little more than what is already required to show “prejudice,” a necessary showing for habeas relief for many constitutional errors. See, e. g., United States v. Bagley, 473 U. S. 667, 682 (1985); Strickland v. Washington, 466 U. S. 668, 694 (1984). If federal habeas review of capital sentences is to be at all rational, petitioner must show something more in order for a court to reach the merits of his claims on a successive habeas petition than he would have had to show to obtain relief on his first habeas petition.13
But, more importantly, petitioner‘s standard would so broaden the inquiry as to make it anything but a “narrow” exception to the principle of finality that we have previously described it to be. A federal district judge confronted with
The Court of Appeals in this case took the middle ground among these three possibilities for defining “actual innocence” of the death penalty, and adopted this test:
“[W]e must require the petitioner to show, based on the evidence proffered plus all rеcord evidence, a fair probability that a rational trier of fact would have entertained a reasonable doubt as to the existence of those facts which are prerequisites under state or federal law for the imposition of the death penalty.” 945 F. 2d, at 820 (footnotes omitted).
In the present petition, Sawyer advances two claims, arising from two distinct groups of evidentiary facts that were not considered by the jury that convicted and sentenced Sawyer. The first group of evidence relates to petitioner‘s role in the offense and consists of affidavits attacking the credibility of Cynthia Shano and an affidavit claiming that one of Shano‘s sons told a police officer that Sawyer was not responsible for pouring lighter fluid on Arwood and lighting it, and that in fact Sawyer tried to prevent Charles Lane from lighting Arwood on fire. Sawyer claims that the police failed to produce this exculpatory evidence in violation of his due process rights under Brady v. Maryland, 373 U. S. 83 (1963). The second grouр consists of medical records from Sawyer‘s stays as a teenager in two different mental health
The Court of Appeals held that petitioner‘s failure to assert his Brady claim in his first petition constituted an abuse of the writ, and that he had not shown cause for failing to raise the claim earlier under McCleskey. 945 F. 2d, at 824. The ineffective-assistance claim was held by the Court of Appeals to be a successive claim because it was rejected on the merits in Sawyer‘s first petition, and petitioner failed to show cause for not bringing all the evidence in support of this claim earlier. Id., at 823. Petitioner does not contest these findings of the Court of Appeals. Tr. of Oral Arg. 7. Therefore, we must determine if petitioner has shown by clear and convincing evidence that but for constitutional error, no reasonable juror would find him eligible for the death penalty under Louisiana law.
Under Louisiana law, petitioner is eligible for the death penalty because he was convicted of first-degree murder-that is, an intentional killing while in the process of committing an aggravated arson-and because at the sentencing phase the jury found two valid aggravating circumstances: that the murder was committed in the course of an aggravated arson, and that the murder was especially cruel, atrocious, and heinous. The psychological evidence petitioner alleges was kept from the jury due to the ineffective assistance of counsel does nоt relate to petitioner‘s guilt or innocence of the crime.16 Neither does it relate to either of the aggravating factors found by the jury that made petitioner eligible for the death penalty. Even if this evidence had been before the jury, it cannot be said that a reasonable juror would not have found both of the aggravating factors that
We are convinced that the evidence allegedly kept from the jury due to an alleged Brady violation also fails to show that the petitioner is actually innocent of the death penalty to which he has been sentenced. Much of the evidence goes to the credibility of Shano, suggesting, e. g., that contrary to her testimony at trial she knew Charles Lane prior to the day of the murder; that she was drinking the day before the murder; and that she testified under a grant of immunity from the prosecutor. 2 App. 589-608. This sort of latter-day evidence brought forward to impeach a prosecution witness will seldom, if ever, make a clear and convincing showing that no reasonable juror would have believed the heart of Shano‘s account of petitioner‘s actions.
The final bit of evidence petitioner alleges was unconstitutionally kept from the jury due to a Brady violation was a statement made by Shano‘s then 4-year-old son, Wayne, to a police officer the day after the murder. Petitioner has submitted an affidavit from one Diane Thibodeaux stating that she was present when Wayne told a police detective who asked who had lit Arwood on fire that “Daddy [Sawyer] tried to help the lady” and that the “other man” had pushed Sawyer back into a chair. 2 App. 587. The affidavit also states that Wayne showed the officer where to find a cigarette lighter and a can of lighter fluid in the trash. Ibid. Because this evidence goes to the jury‘s finding of aggravated arson, it goes both to petitioner‘s guilt or innocence of the crime of first-degree murder and the aggravating circumstance of a murder committed in the course of an aggravated arson. However, we conclude that this affidavit, in view of
We therefore hold that petitioner has failed to show by clear and convincing evidence that but for constitutional error at his sentencing hearing, no reasonable juror would have found him eligible for the death penalty under Louisiana law. The judgment of the Court of Appеals is therefore
Affirmed.
JUSTICE BLACKMUN, concurring in the judgment.
I cannot agree with the majority that a federal court is absolutely barred from reviewing a capital defendant‘s abu-
I
The Court repeatedly has recognized that principles of fundamental fairness underlie the writ of habeas corpus. See Engle v. Isaac, 456 U. S. 107, 126 (1982); Sanders v. United States, 373 U. S. 1, 17-18 (1963). Even as the Court has erected unprecedented and unwarranted barriers to the federal judiciary‘s review of the merits of claims that state prisoners failed properly to presеnt to the state courts, or failed to raise in their first federal habeas petitions, or previously presented to the federal courts for resolution, it consistently has acknowledged that exceptions to these rules of unreviewability must exist to prevent violations of fundamental fairness. See Engle, 456 U. S., at 135 (principles of finality and comity “must yield to the imperative of correcting a fundamentally unjust incarceration“). Thus, the Court has held, federal courts may review procedurally defaulted, abusive, or successive claims absent a showing of cause and
By the traditional understanding of habeas corpus, a “fundamental miscarriage of justice” occurs whenever a conviction or sentence is secured in violation of a federal constitutional right. See
In a trio of 1986 decisions, however, the Court ignored these traditional teachings and, out of a purported concern for state sovereignty, for the preservation of state resources, and for the finality of state-court judgments, shifted the focus of federal habeas review of procedurally defaulted, successive, or abusive claims away from the preservation of constitutional rights to a fact-based inquiry into the petitioner‘s innocence or guilt. See Wilson, 477 U. S., at 454 (plurality opinion) (“[T]he ‘ends of justice’ require federal courts to entertain [successive] petitions only where the prisoner supplements his constitutional claim with a colorable showing of factual innocence“); Carrier, 477 U. S., at 496 (“[I]n an extraordinary case, where a constitutional violation has probably resulted in the conviction of one who is actually innocent, a federal habeas court may grant the writ even in the absence of a showing of cause for the procedural default“);
The Court itself has acknowledged that “the concept of ‘actual,’ as distinct from ‘legal,’ innocence does not translate easily into the context of an alleged error at the sentencing phase of a trial on a capital offense.” Smith, 477 U. S., at 537. Undaunted by its own illogic, however, the Court adopted just such an approach in Smith. There, the Court was confronted with a claim that the introduction at sentencing of inculpatory statements made by Smith to a court-appointed psychiatrist violated the
Just last Term, in McCleskey v. Zant, the Court again described the “fundamental miscarriage of justice” exception as a “safeguard against compelling an innocent man to suffer an unconstitutional loss of liberty,” 499 U. S., at 495 (quoting Stone v. Powell, 428 U. S. 465, 491-492, n. 31 (1976)). Although the District Court granted relief to McCleskey on his claim that state authorities deliberately had elicited inculpatory admissions from him in violation of his
The Court today takes for granted that the fоregoing decisions correctly limited the concept of a “fundamental miscarriage of justice” to “actual innocence,” even as it struggles, by ignoring the “natural usage of those words” and resorting to “analog[s],” see ante, at 341, to make sense of “actual innocence” in the capital context. I continue to believe, however, that the Court‘s “exaltation of accuracy as the only characteristic of ‘fundamental fairness’ is deeply flawed.” Smith, 477 U. S., at 545 (STEVENS, J., dissenting).
As an initial matter, the Court‘s focus on factual innocence is inconsistent with Congress’ grant of habeas corpus jurisdiction, pursuant to which federal courts are instructed to entertain petitions from state prisoners who allege that they are held “in custody in violation of the Constitution or laws or treaties of the United States.”
Most important, however, the focus on innocence assumes, erroneously, that the only value worth protecting through federal habeas review is the accuracy and reliability of the guilt determination. But “[o]ur criminal justice system, and our Constitution, protect other values in addition to the reliability of the guilt or innocence determination, and the statutory duty to serve ‘law and justice’ should similarly reflect those values.” Smith, 477 U. S., at 545 (STEVENS, J., dissenting). The accusatorial system of justice adopted by the Founders affords a defendant certain process-based protections that do not have accuracy of truth finding as their primary goal. These protections—including the
Nowhere is this single-minded focus on actual innocence more misguided than in a case where a defendant alleges a constitutional error in the sentencing phase of a capital trial. The Court‘s ongoing struggle to give meaning to “innocence of death” simply reflects the inappropriateness of the inquiry. See Smith, 477 U. S., at 537; Adams, 489 U. S., at 412, n. 6; ante, at 340. “Guilt or innocence is irrelevant in that context; rather, there is only a decision made by representatives of the community whether the prisoner shall live or die.” Wilson, 477 U. S., at 471-472, n. 7 (Brennan, J., dissenting).
Only by returning to the federal courts’ central and traditional function on habeas review, evaluating claims of constitutional error, can the Court ensure that the ends of justice are served and that fundamental miscarriages of justice do not go unremedied. The Court would do well to heed Justice Black‘s admonition: “[I]t is never too late for сourts in habeas corpus proceedings to look straight through procedural screens in order to prevent forfeiture of life or liberty in flagrant defiance of the Constitution.” Brown v. Allen, 344 U. S. 443, 554 (1953) (dissenting opinion).2
II
A
When I was on the United States Court of Appeals for the Eighth Circuit, I once observed, in the course of reviewing a death sentence on a writ of habeas corpus, that the decisional process in a capital case is “particularly excruciating” for someone “who is not personally convinced of the rightness of capital punishment and who questions it as an effective deterrent.” Maxwell v. Bishop, 398 F. 2d 138, 153-154 (1968), vacated, 398 U. S. 262 (1970). At the same time, however, I stated my then belief that “the advisability of capital punishment is a policy matter ordinarily to be resolved by the legislature.” Id., at 154. Four years later, as a Member of this Court, I echoed those sentiments in my separate dissenting opinion in Furman v. Georgia, 408 U. S. 238, 405 (1972). Although I reiterated my personal distaste for the
My ability in Maxwell, Furman, and the many other capital cases I have reviewed during my tenure on the federal bench to enforce, notwithstanding my own deep moral reservations, a legislature‘s considered judgment that capital punishment is an appropriate sanction, has always rested on an understanding that certain procedural safeguards, chief among them the Federal Judiciary‘s powеr to reach and correct claims of constitutional error on federal habeas review, would ensure that death sentences are fairly imposed. Today, more than 20 years later, I wonder what is left of that premise underlying my acceptance of the death penalty.
B
Only last Term I had occasion to lament the Court‘s continuing “crusade to erect petty procedural barriers in the path of any state prisoner seeking review of his federal constitutional claims” and its transformation of “the duty to protect federal rights into a self-fashioned abdication.” Coleman v. Thompson, 501 U. S. 722, 759, 761 (1991) (dissenting opinion). This Term has witnessed the continued narrowing of the avenues of relief available to federal habeas petitioners seeking redress of their constitutional claims. See, e. g., Keeney v. Tamayo-Reyes, 504 U. S. 1 (1992) (overruling in part Townsend v. Sain, 372 U. S. 293 (1963)). It has witnessed, as well, the execution of two victims of the “new habeas,” Warren McCleskey and Roger Keith Coleman.
Warren McCleskey‘s case seemed the archetypal “fundamental miscarriage of justice” that the federal courts are charged with remedying. As noted above, McCleskey dem
The execution of Roger Keith Coleman is no less an affront to principles of fundamental fairness. Last Term, the Court refused to review the merits of Coleman‘s claims by effectively overruling, at Coleman‘s expense, precedents holding that state-court decisions are presumed to be based on the merits (and therefore, are subject to federal habeas review) unless they explicitly reveal that they were based on state procedural grounds. See Coleman, 501 U. S., at 762-764 (dissenting opinion). Moreover, the Court‘s refusal last month to grant a temporary stay of execution so that the lower courts could conduct a hearing into Coleman‘s well-supported claim that he was innocent of the underlying offense demonstrates the resounding hollowness of the Court‘s professed commitment to employ the “fundamental miscarriage of justice exception” as a “safeguard against compelling an innocent man to suffer an unconstitutional loss of liberty.” McCleskey v. Zant, 499 U. S., at 495 (internal quotation marks omitted). See Coleman v. Thompson, 504 U. S. 188, 189 (1992) (opinion dissenting from denial of stay of execution).
As I review the state of this Court‘s capital jurisprudence, I thus am left to wonder how the ever-shrinking authority of
JUSTICE STEVENS, with whom JUSTICE BLACKMUN and JUSTICE O‘CONNOR join, concurring in the judgment.
Only 10 years ago, the Court reemphasized that “[t]he writ of habeas corpus indisputably holds an honored position in our jurisprudence. Tracing its roots deep into English common law, it claims a place in Art. I of our Constitution. Today, as in prior centuries, the writ is a bulwark against convictions that violate ‘fundamental fairness.’ Wainwright v. Sykes, 433 U. S. 72, 97 (1977) (STEVENS, J., concurring).” Engle v. Isaac, 456 U. S. 107, 126 (1982). It is this centrality of “fundamental fairness” that has led the Court to hold that habeas review of a defaulted, successive, or abusive claim is available, even absent a showing of cause, if failure to consider the claim would result in a fundamental miscarriage of justice. See Sanders v. United States, 373 U. S. 1, 17-18 (1963); Engle, 456 U. S., at 135.
In Murray v. Carrier, 477 U. S. 478, 495, 496 (1986), the Court ruled that the concept of “fundamental miscarriage of justice” applies to those cases in which the defendant was
While the conviction of an innocent person may be the archetypal case of a manifest miscarriage of justice, it is not the only case. There is no reason why “actual innocence” must be both an animating and the limiting principle of the work of federal courts in furthering the “ends of justice.” As Judge Friendly emphasized, there are contexts in which, irrespective of guilt or innocence, constitutional errors violate fundamental fairness. Friendly, Is Innocence Irrelevant? Collateral Attack on Criminal Judgments, 38 U. Chi. L. Rev. 142, 151-154 (1970). Fundamental fairness is more than accuracy at trial; justice is more than guilt or innocence.
Nowhere is this more true than in capital sentencing proceedings. Because the death penalty is qualitatively and morally different from any other penalty, “[i]t is of vital importance to the defendant and to the community that any decision to impose the death sentence be, and appear to be, the consequence of scrupulously fair procedures.” Smith v. Murray, 477 U. S. 527, 545-546 (1986) (STEVENS, J., dissenting). Accordingly, the ends of justice dictate that “[w]hen a condemned prisoner raises a substantial, colorable
Thus the Court‘s first and most basic error today is that it asks the wrong question. Charged with averting manifest miscarriages of justice, the Court instead narrowly recasts
I
The Court today holds that, absent a showing of cause, a federal court may not review a capital defendant‘s defaulted, successive, or abusive claims unless the defendant “show[s] by clear and convincing evidence that, but for a constitutional error, no reasonable juror would have found [him] eligible for the death penalty.” Ante, at 336.
This definition of “innocence of the death sentence” deviates from our established jurisprudence in two ways. First, the “clear and convincing evidence” standard departs from a line of decisions defining the “actual innocence” exception to the cause-and-prejudice requirement. Second, and more fundamentally, the Court‘s focus on eligibility for the death penalty conflicts with the very structure of the constitutional law of capital punishment.
As noted above, in Murray v. Carrier, the Court held that in those cases in which “a constitutional violation has probably resulted in the conviction of one who is actually innocent, a federal habeas court may grant the writ even in the absence of a showing of cause for the procedural default.” 477 U. S., at 496 (emphasis added). The Court has since frequently cоnfirmed this standard. See, e. g., Coleman v. Thompson, 501 U. S. 722, 748 (1991); Dugger v. Adams, 489 U. S. 401, 412, n. 6 (1989); Teague v. Lane, 489 U. S. 288, 313 (1989). In subsequent decisions, both those involving “innocence of the offense” and those involving “innocence of the death sentence,” the Court has employed the same standard of proof. For example, in Smith v. Murray, 477 U. S. 527
As we noted in another context, “[t]his outcome-determinative standard has several strengths. It defines the relevant inquiry in a way familiar to courts, though the inquiry, as is inevitable, is anything but precise. The standard also reflects the profound importance of finality in criminal proceedings. Moreover, it comports with the widely used standard for assessing motions for new trial based on newly discovered evidence.” Strickland v. Washington, 466 U. S. 668, 693-694 (1984).
Equally significant, this “probably resulted” standard is well calibrated to the manifest miscarriage of justice exception. Not only does the standard respect the competing demands of finality and fundamental fairness, it also fits squarely within our habeas jurisprudenсe. In general, a federal court may entertain a defaulted, successive, or abusive claim if a prisoner demonstrates cause and prejudice. See generally McCleskey v. Zant, 499 U. S. 467, 493-495 (1991). To show “prejudice,” a defendant must demonstrate “a reasonable probability that, but for [the alleged] erro[r], the result of the proceeding would have been different.” Strickland, 466 U. S., at 694; see also United States v. Bag-
The Court today repudiates this established standard of proof and replaces it with a requirement that a defendant “show by clear and convincing evidence that . . . no reasonable juror would have found [him] eligible for the death penalty.” Ante, at 336 (emphasis supplied). I see no reason to reject the established and well-functioning “probably resulted” standard and impose such a severe burden on the capital defendant. Although we have frequently recognized the State‘s strong interest in finality, we have never suggested that that interest is sufficient to outweigh the individual‘s claim to innoсence. To the contrary, the “actual innocence” exception itself manifests our recognition that the criminal justice system occasionally errs and that, when it does, finality must yield to justice.
“The function of a standard of proof . . . is to ‘instruct the factfinder concerning the degree of confidence our society thinks he should have in the correctness of factual conclusions for a particular type of adjudication.’ . . . The standard serves to allocate the risk of error between the litigants and to indicate the relative importance attached to the ultimate decision.” Addington v. Texas, 441 U. S. 418, 423 (1979) (citation omitted). Neither of these considerations
First, there is no basis for requiring a federal court to be virtually certain that the defendant is actually ineligible for the death penalty before merely entertaining his claim. We have required a showing by clear and convincing evidence in several contexts: For example, the medical facts underlying a civil commitment must be established by this standard, Addington v. Texas, 441 U. S. 418 (1979), as must “actual malice” in a libel suit brought by a public official. New York Times Co. v. Sullivan, 376 U. S. 254, 279-280 (1964); see also Anderson v. Liberty Lobby, Inc., 477 U. S. 242 (1986). And we have required a related showing in cases involving deportation, Woodby v. INS, 385 U. S. 276, 285-286 (1966), and denaturalization, Schneiderman v. United States, 320 U. S. 118, 125 (1943). In each of these contexts, the interests of the nonmoving party were truly substantial: personal liberty in Addington, freedom of expression in New York Times, residence in Woodby, and citizenship in Schneiderman. In my opinion, the State‘s interest in finality in a capital prosecution is not nearly as great as any of these interests. Indeed, it is important to remember that “innocence of the death sentence” is not a standard for staying or vacating a death sentence, but merely a standard for determining whether or not a court should reach the merits of a defaulted claim. The State‘s interest in “finality” in this context certainly does not warrant a “clear and convincing” evidentiary standard.
Nor is there any justification for allocating the risk of error to fall so severely upon the capital defendant or attaching greater importance to the initial sentence than to the issue of whether that sentence is appropriate. The States themselves have declined to attach such weight to capital sentences: Most States provide plain-error review for defaulted claims in capital cases. See Smith v. Murray, 477 U. S., at 548-550, n. 20 (collecting authorities). In this regard, the
Indeed, the Court‘s ruling creates a perverse double standard. While a defendant raising defaulted claims in a noncapital case must show that constitutional error “probably resulted” in a miscarriage of justice, a capital defendant must present “clear and convincing evidence” that no reasonable juror would find him eligible for the death penalty. It is heartlessly perverse to impose a more stringent standard of proof to avoid a miscarriage of justice in a capital case than in a noncapital case.
In sum, I see no reason to depart from settled law, which clearly requires a defendant pressing a defaulted, successive, or abusive claim to show that a failure to hear his claim will “probably result” in a fundamental miscarriage of justice. In my opinion, a corresponding standard governs a defaulted, successive, or abusive challenge to a capital sentence: The defendant must show that he is probably—that is, more likely than not—“innocent of the death sentence.”
II
The Court recognizes that the proper definition of “innocence of the death sentence” must involve a reweighing of the evidence and must focus on the sentencer‘s likely evaluation of that evidence. Thus, the Court directs federal courts to look to whether a “reasonable juror would have found the petitioner eligible for the death penalty.” Ante, at 336 (emphasis added). Nevertheless, the Court inexplicably limits this inquiry in two ways. First, the Court holds that courts should consider only evidence concerning aggravating factors. As demonstrated below, this limitation is wholly without foundation and neglects the central role of mitigating evidence in capital sentencing proceedings. Second, the
It is well established that, “in capital cases, the sentencer may not refuse to consider or be precluded from considering any relevant mitigating evidence.” Hitchcock v. Dugger, 481 U. S. 393, 394 (1987) (internal quotation marks and citations omitted). Yet in ascribing a narrow, eligibility-based meaning to “innocence of the death sentence” the Court neglects this rudimentary principle.
As the Court recognizes, a single general directive animates and informs our capital-punishment jurisprudence: “[T]he death penalty [may not] be imposed under sentencing procedures that creat[e] a substantial risk that [the death penalty] would be inflicted in an arbitrary and capricious manner.” Gregg v. Georgia, 428 U. S. 153, 188 (1976) (opinion of Stewart, Powell, and STEVENS, JJ.). As applied and developed over the years, this constitutional requirement has yielded two central principles. First, a sentencing scheme must “genuinely narrow the class of persons eligible for the death penalty.” Zant v. Stephens, 462 U. S. 862, 877 (1983). Second, the sentencer must “not be precluded from considering, as a mitigating factor, any aspect of a defendant‘s character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death.” Lockett v. Ohio, 438 U. S. 586, 604 (1978) (opinion of Burger, C. J.) (emphasis in original). Although these principles—one narrowing the relevant class, the other broadening the scope of considered evidence—seemingly point in opposite directions, in fact both serve the same end: ensuring that a capital sentence is the product of individualized and reasoned moral decisionmaking.
First, the Court implicitly repudiates the requirement that the sentencer be allowed to consider all relevant mitigating evidence, a constitutive element of our
Moreover, the Court‘s holding also clashes with the theory underlying our capital-punishment jurisprudence. The nonarbitrariness—and therefore the constitutionality—of the death penalty rests on individualized sentencing determinations. See generally California v. Brown, 479 U. S. 538, 544-546 (1987) (O‘CONNOR, J., concurring). This is the dif-
The Court‘s definition of “innocent of the death sentence” is flawed in a second, related way. The Court‘s analysis not only neglects errors that preclude a sentencer‘s consideration of mitigating factors; it also focuses too narrowly on eligibility. The Court requires a defendant to call into question all of the aggravating factors found by the sentencer and thereby show himself ineligible for the death penalty.
In sum, the Court‘s “innocent of the death sentence” standard is flawed both in its failure to consider constitutional errors implicating mitigating factors and in its unduly harsh requirement that a defendant‘s eligibility for the death penalty be disproved.
III
In my opiniоn, the “innocence of the death sentence” standard must take into account several factors. First, such a standard must reflect both of the basic principles of our capital-punishment jurisprudence. The standard must recognize both the need to define narrowly the class of “death-eligible” defendants and the need to define broadly the scope of mitigating evidence permitted the capital sentencer. Second, the “innocence of the death sentence” standard should also recognize the distinctive character of the capital sentencing decision. While the question of innocence or guilt of the offense is essentially a question of fact, the choice between life imprisonment and capital punishment is both a question of underlying fact and a matter of reasoned moral judgment. Thus, there may be some situations in which, although the defendant remains technically “eligible” for the death sentence, nonetheless, in light of all of the evidence,
These requirements are best met by a standard that provides that a defendant is “innocent of the death sentence” only if his capital sentence is clearly erroneous. This standard encompasses several types of error. A death sentence is clearly errоneous if, taking into account all of the available evidence, the sentencer lacked the legal authority to impose such a sentence because, under state law, the defendant was not eligible for the death penalty. Similarly, in the case of a “jury override,” a death sentence is clearly erroneous if, taking into account all of the evidence, the evidentiary prerequisites for that override (as established by state law) were not met. See, e. g., Johnson v. Singletary, 938 F. 2d 1166, 1194-1195 (CA11 1991) (Tjoflat, C. J., concurring in part and dissenting in part) (concluding that the sentencing “judge, as a matter of law, could not have sentenced the petitioner to death” because there was insufficient evidence to meet the jury-override standard established in Tedder v. State, 322 So. 2d 908, 910 (Fla. 1975)). A death sentence is also clearly erroneous under a “balancing” regime if, in view of all of the evidence, mitigating circumstances so far outweighed aggravating circumstances that no reasonable sentencer would have imposed the death penalty. Cf. Jackson v. Virginia, 443 U. S. 307, 316-318 (1979). Such a case might arise if constitutional error either precluded the defendant from demonstrating that aggravating circumstances did not obtain or precluded the sentencer‘s consideration of important mitigating evidence.
Moreover, the “clearly-erroneous” standard is duly protective of the State‘s legitimate interests in finality and respectful of the systemic and institutional costs of successive habeas litigation. The standard is stringent: If the sentence “is plausible in light of the record viewed in its entirety” it is not clearly erroneous “even though [the court is] convinced that had it been sitting as the [sentencer], it would have weighed the evidence differently.” Anderson v. Bessemer City, 470 U. S. 564, 574 (1985). At the same time, “clearly-erroneous” review allows a federal court to entertain a defaulted claim in the rare case in which the “court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” United States v. United States Gypsum Co., 333 U. S. 364, 395 (1948).
Finally, the “clearly-erroneous” standard is workable. As was true of the cause-and-prejudice standard adopted in McCleskey v. Zant, the clear-error standard is “[w]ell-defined in the case law [and] familiar to federal courts. . . . The standard is an objective one, and can be applied in a manner that comports with the threshold nature of the abuse of the writ inquiry.” 499 U. S., at 496. Federal courts have long applied the “clearly-erroneous” standard pursuant to Rule 52 of the
In my opinion, then, the “clearly-erroneous” standard is the core of the “innocence of the death sentence” exception. Just as a defendant who presses a defaulted, successive, or abusive claim and who cannot show cause must demonstrate that it is more likely than not that he is actually innocent of the offense, so a capital defendant who presses such a claim and cannot show cause must demonstrate that it is more likely than not that his death sentence was clearly erroneous. Absent such a showing, a federal court may not reach the merits of the defendant‘s defaulted, successive, or abusive claim.
IV
It remains to apply this standard to the case at hand. As the majority indicates, Sawyer alleges two constitutional errors. First, he contends that the State withheld certain exculpatory evidence, in violation of Sawyer‘s due process rights as recognized in Brady v. Maryland, 373 U. S. 83 (1963). Second, Sawyer argues that his trial counsel‘s failure to uncover and present records from Sawyer‘s earlier treatments in psychiatric institutions deprived him of effective assistance of counsel as guaranteed by the
As Sawyer failed to assert his Brady claim in an earlier habeas petition and as he cannot show cause for that failure, the court may only reach the merits of that “abusive” claim if Sawyer demonstrates that he is probably actually innocent of the offense or that it is more likely than not that his death sentence was clearly erroneous. As Sawyer‘s ineffective-assistance claim was cоnsidered and rejected in an earlier
Sawyer points to two pieces of exculpatory evidence allegedly withheld by the State. First, he offers the affidavit of a woman (Diane Thibodeaux) who, on occasion, took care of the small child who witnessed the crime. That account appears to conflict with contemporaneous police reports. While police records indicate that the child implicated Sawyer in the cruel burning of the victim, Thibodeaux avers that the child stated to her that Sawyer‘s codefendant, Charles Lane, set the victim afire. Second, he offers other affidavits casting doubt on the credibility of Cindy Shano, the State‘s principal witness. Sawyer emphasizes that Shano testified under a grant of immunity and highlights inaccuracies in her trial testimony. Finally, as part of his
Viewed as a whole, the record does not demonstrate that failure to reach the merits of Sawyer‘s claims would constitute a fundamental miscarriage of justice. First, in view of the other evidence in the record, the Thibodeaux affidavit and questions concerning Shano‘s testimony do not establish that Sawyer is “probably . . . actually innocent” of the crime of first-degree murder. At most, Thibodeaux‘s hearsay statements cast slight doubt on the facts underlying the burning of the victim. Similаrly, although the challenges to Shano‘s testimony raise questions, these affidavits do not demonstrate that Sawyer probably did not commit first-degree murder. Thus, Sawyer has not met the standard “actual innocence” exception.
Second, the affidavits and the new medical records do not convince me that Sawyer‘s death sentence is clearly erroneous. The jury found two statutory aggravating factors—that the murder was committed in the course of an aggra
In sum, in my opinion Sawyer has failed to demonstrate that it is more likely than not that his death sentence was clearly erroneous. Accordingly, I conclude that the court below was correct in declining to reach the merits of Sawyer‘s successive and abusive claims.
V
The Court rejects an “innocence of death” standard that recognizes constitutional errors affecting mitigating evidence because such a standard “would so broadеn the inquiry as to make it anything but a ‘narrow’ exception to the principle of finality.” Ante, at 345. As the foregoing analysis indicates, however, the Court‘s concerns are unfounded. Indeed, even when federal courts have applied a less restrictive standard than the standard I propose, those courts have rarely found “innocence of death” and reached the merits of a defaulted, successive, or abusive claim. See Deutscher v. Whitley, 946 F. 2d 1443 (CA9 1991); Stokes v.
Similarly, I do not share the Court‘s concern that a standard broader than the eligibility standard creates “a far more difficult task” for federal courts. Ante, at 346. As noted above, both the “probably resulted” standard and the “clearly-erroneous” standard have long been applied by federal courts in a variety of contexts. Moreover, to the extent that the “clearly-erroneous” standard is more difficult to apply than the Court‘s “eligibility” test, I believe that that cost is far outweighed by the importance of making just decisions in the few cases that fit within this narrow exception. To my mind, any added administrative burden is surely justified by the overriding interest in minimizing the risk of error in implementing the sovereign‘s decision to take the life of one of its citizens. As we observed in Gardner v. Florida, 430 U. S. 349, 360 (1977), “if the disputed matter is of critical importance, the time invested in ascertaining the truth would surely be well spent if it makes the difference between life and death.”
Notes
“First degree murder is the killing of a human being:
“(1) When the offender has specific intent to kill or to inflict great bodily harm and is engaged in the perpetration or attempted perpetration of aggravated kidnapping, second degree kidnapping, aggravated escape, aggravated arson, aggravated rape, forcible rape, aggravated burglary, armed robbery, first degree robbery or simple robbery;
“(2) When the offender has specific intent to kill or to inflict great bodily harm upon a fireman or peace officer engaged in the performance of his lawful duties;
“(3) When the offender has specific intent to kill or to inflict great bodily harm upon more than one person; or
“(4) When the offender has specific intent to kill or inflict great bodily harm and has offered, has been offered, has given, or has received anything of value for the killing.
“Whoever commits the crime of first degree murder shall be punished by death or life imprisonment at hard labor without benefit of parole, probation, or suspension of sentence in accordance with the determination of the jury.”
The Eighth Circuit has adopted a similar test: “‘In the penalty-phase context, this exception will be available if the federal constitutional error alleged probably resulted in a verdict of death against one whom the jury would otherwise have sentenced to life imprisonment.‘” Stokes v. Armontrout, 893 F. 2d 152, 156 (1989), quoting Smith v. Armontrout, 888 F. 2d 530, 545 (1989).
“Thus, a petitioner may make a colorable showing that he is actually innocent of the death penalty by presenting evidence that an alleged constitutional error implicates all of the aggravating factors found to be present by the sentencing body. That is, but for the alleged constitutional error, the sentencing body could not have found any aggravating factors and thus the petitioner was ineligible for the death penalty. In other words, the petitioner must show that absent the alleged constitutional error, the jury would have lacked the discretion to impose the death penalty; that is, that he is ineligible for the death penalty.” Johnson v. Singletary, 938 F. 2d, at 1183 (emphasis in original).
Even considering the affidavit of Wayne Shano, it cannot be said that no reasonable juror would have found that petitioner committed the aggravated arson, given Cynthia Shano‘s testimony as to petitioner‘s statements to Lane on the day of the murder and petitioner‘s fingerprints on the can of lighter fluid.
