MAJOR US SUPREME CT CASES

KANSAS, PETITIONER v. MICHAEL LEE MARSH, II. 

No. 04-1170

JUSTICE SCALIA, concurring.

I join the opinion of the Court. I write separately to clarify briefly the import of my joinder, and to respond at somewhat greater length first to JUSTICE STEVENS' contention that this case, and cases like it, do not merit our attention, and second to JUSTICE SOUTER's claims about risks inherent in capital punishment.

I

Part III of the Court's opinion -- which makes plain why Walton v. Arizona, 497 U.S. 639, 110 S. Ct. 3047, 111 L. Ed. 2d 511 (1990), controls this case -- would be sufficient to reverse the judgment below. I nonetheless join Part IV as well, which describes why Kansas's death penalty statute easily satisfies even a capital jurisprudence as incoherent as ours has become. In doing so, I do not endorse that incoherence, but adhere to my previous statement that "I will not . . . vote to uphold an Eighth Amendment claim that the sentencer's discretion has been unlawfully restricted." Id., at 673, 110 S. Ct. 3047, 111 L. Ed. 2d 511 (concurring in part and concurring in judgment).

II

JUSTICE STEVENS' dissent gives several reasons why this case, and any criminal case in which the State is the petitioner,  does not deserve our attention. "'No rule of law,'" he says, "'commanded the Court to grant certiorari.'" Post, at 3 (quoting California v. Ramos, 463 U.S. 992, 1031, 103 S. Ct. 3446, 77 L. Ed. 2d 1171 (1983) (STEVENS, J., dissenting)). But that is true, of course, of almost our entire docket; it is in the very nature of certiorari jurisdiction. Also self-evident, since the jurisdiction of the Kansas Supreme Court ends at the borders of that State, is the fact that "'no other State would have been required to follow the [Kansas] precedent if it had been permitted to stand.'" Post, at 3 (STEVENS, J., dissenting) (quoting Ramos, supra, at 1031, 103 S. Ct. 3446, 77 L. Ed. 2d 1171 (STEVENS, J., dissenting)). But if this signaled the impropriety of granting certiorari, we would never review state-court determinations of federal law, even though they patently contradict (as the determination below does) the holdings of other state courts and Federal Courts of Appeals, compare 278 Kan. 520, 534-537, 102 P.3d 445, 457-459 (2004) (case below), and State v. Kleypas, 272 Kan. 894, 1005-1007, 40 P.3d 139, 225-226 (2001), with, e.g., State v. Hoffman, 123 Idaho 638, 646-647, 851 P.2d 934, 942-943 (1993), and Jones v. Dugger, 928 F.2d 1020, 1029 (CA11 1991) -- and indeed, even when they patently contradict our own decisions. Our principal responsibility under current practice, however, and a primary basis for the Constitution's allowing us to be accorded jurisdiction to review state-court decisions, see Art. III, §  2, cls. 1 and 2, is to ensure the integrity and uniformity of federal law. n1 See this Court's Rule 10(b), (c). Fulfillment of this responsibility is, to put it mildly, an adequate answer to the charge that "'nothing more than an interest in facilitating the imposition of the death penalty in [Kansas] justified this Court's exercise of its discretion to review the judgment of the [Kansas] Supreme Court.'" Post, at 3 (STEVENS, J., dissenting) (quoting Ramos, supra, at 1031, 103 S. Ct. 3446, 77 L. Ed. 2d 1171 (STEVENS, J., dissenting)). 

n1 The dissent observes that Congress did not initially grant us the full jurisdiction that the Constitution authorizes, but only allowed us to review cases rejecting the assertion of governing federal law. See post, at 3-4, n. (opinion of STEVENS, J.). That is unsurprising and immaterial. The original Constitution contained few guarantees of individual rights against the States, and in clashes of governmental authority there was small risk that the state courts would erroneously side with the new Federal Government. (In 1789, when the first Judiciary Act was passed, the Bill of Rights had not yet been adopted, and once it was, it did not apply against the States, see Barron ex rel. Tiernan v. Mayor of Baltimore, 32 U.S. 243, 7 Pet. 243, 8 L. Ed. 672 (1833).) Congress would have been most unlikely to contemplate that state courts would erroneously invalidate state actions on federal grounds. The early history of our jurisdiction assuredly does not support the dissent's awarding of special preference to the constitutional rights of criminal defendants. Even with respect to federal defendants (who did enjoy the protections of the Bill of Rights), "during the first 100 years of the Court's existence there was no provision made by Congress for Supreme Court review of federal criminal convictions, an omission that Congress did not remedy until 1889 and beyond." R. Stern, E. Gressman, S. Shapiro, & K. Geller, Supreme Court Practice 66 (8th ed. 2002). In any case, present law is plain. The 1988 statute cited by the dissent and forming the basis of our current certiorari jurisdiction places States and defendants in precisely the same position. They are both entitled to petition for our review. 

The dissent's assertion that our holding in Ramos was "ironic," post, at 2 (opinion of STEVENS, J.), rests on a misguided view of federalism and, worse still, of a republican form of government. Only that can explain the dissent's suggestion that Ramos's reversal of a state-court determination somehow undermined state authority. The California Supreme Court had ruled that a jury instruction inserted into the state penal code by voter initiative, see 463 U.S., at 995, n. 4, 103 S. Ct. 3446, 77 L. Ed. 2d 1171, was invalid as a matter of federal constitutional law. See id., at 996, 997, n. 7, 103 S. Ct. 3446, 77 L. Ed. 2d 1171. When state courts erroneously invalidate actions taken by the people of a State (through initiative or through normal operation of the political branches of their state government) on state-law grounds, it is generally none of our business; and our displacing of those judgments would indeed be an intrusion upon state autonomy. But when state courts erroneously invalidate such actions because they believe federal law requires it -- and especially when they do so because they believe the Federal Constitution requires it -- review by this Court, far from undermining state autonomy, is the only possible way to vindicate it. When a federal constitutional interdict against the duly expressed will of the people of a State is erroneously pronounced by a State's highest court, no authority in the State -- not even a referendum agreed to by all its citizens -- can undo the error. Thus, a general presumption against such review displays not respect for the States, but a complacent willingness to allow judges to strip the people of the power to govern themselves. When we correct a state court's federal errors, we return power to the State, and to its people.

That is why our decision in Ramos was necessary. Our solemn responsibility is not merely to determine whether a State Supreme Court "has adequately protected [a defendant's] rights under the Federal Constitution," post, at 2 (STEVENS, J., dissenting). It is to ensure that when courts speak in the name of the Federal Constitution, they disregard none of its guarantees -- neither those that assure the rights of criminal defendants, nor those that assure what Justice Black, in his famous dissent in In re Winship, 397 U.S. 358, 385, 90 S. Ct. 1068, 25 L. Ed. 2d 368 (1970), called "the most fundamental individual liberty of our people -- the right of each man to participate in the self-government of his society." Turning a blind eye to federal constitutional error that benefits criminal defendants, allowing it to permeate in varying fashion each state Supreme Court's jurisprudence, would change the uniform "law of the land" into a crazy quilt. And on top of it all, of course, what the dissent proposes avowedly favors one party to the case: When a criminal defendant loses a questionable constitutional point, we may grant review; when the State loses, we must deny it. While it might be appropriate for Congress to place such a thumb upon the scales of our power to review, it seems to me a peculiar mode of decisionmaking for judges sworn to "impartially discharge . . . all the duties" of their office, 28 U.S.C. §  453.

Our decision to grant certiorari is guided by the considerations set forth in Rule 10. None of them turns on the identity of the party that the asserted misapplication of federal law has harmed. When state legislation is thwarted -- not on the basis of state law, but on the basis of a questionable application of the Federal Constitution or laws -- I shall continue to vote to grant the resulting petition for certiorari.

III

Finally, I must say a few words (indeed, more than a few) in response to Part III of JUSTICE SOUTER's dissent. This contains the disclaimer that the dissenters are not (yet) ready to "generalize about the soundness of capital sentencing across the country," post, at 9; but that is in fact precisely what they do. The dissent essentially argues that capital punishment is such an undesirable institution -- it results in the condemnation of such a large number of innocents -- that any legal rule which eliminates its pronouncement, including the one favored by the dissenters in the present case, should be embraced. See ibid.

As a general rule, I do not think it appropriate for judges to heap either praise or censure upon a legislative measure that comes before them, lest it be thought that their validation, invalidation, or interpretation of it is driven by their desire to expand or constrict what they personally approve or disapprove as a matter of policy. In the present case, for example, people might leap to the conclusion that the dissenters' views on whether Kansas's equipoise rule is constitutional are determined by their personal disapproval of an institution that has been democratically adopted by 38 States and the United States. But of course that requires no leap; just a willingness to take the dissenters at their word. For as I have described, the dissenters' very argument is that imposition of the death penalty should be minimized by invalidation of the equipoise rule because it is a bad, "risky," and "hazardous" idea, ibid. A broader conclusion that people should derive, however (and I would not consider this much of a leap either), is that the dissenters' encumbering of the death penalty in other cases, with unwarranted restrictions neither contained in the text of the Constitution nor reflected in two centuries of practice under it, will be the product of their policy views -- views not shared by the vast majority of the American people. The dissenters' proclamation of their policy agenda in the present case is especially striking because it is nailed to the door of the wrong church -- that is, set forth in a case litigating a rule that has nothing to do with the evaluation of guilt or innocence. There are, of course, many cases in which the rule at issue does serve that function, see, e.g., House v. Bell, 547 U.S. ___, 126 S. Ct. 2064, 165 L. Ed. 2d 1, 2006 U.S. LEXIS 4675 (2006). (Marsh himself has earned a remand by application of one such rule, see ante, at 2-3.) But as the Court observes, see ante, at 16-17, guilt or innocence is logically disconnected to the challenge in this case to sentencing standards. The only time the equipoise provision is relevant is when the State has proved a defendant guilty of a capital crime. n2  

N2 Not only are the dissent's views on the erroneous imposition of the death penalty irrelevant to the present case, but the dissent's proposed holding on the equipoise issue will not necessarily work to defendants' advantage. The equipoise provision of the Kansas statute imposes the death penalty only when the State proves beyond a reasonable doubt that mitigating factors do not outweigh the aggravators. See ante, at 2. If we were to disallow Kansas's scheme, the State could, as Marsh freely admits, replace it with a scheme requiring the State to prove by a mere preponderance of the evidence that the aggravators outweigh the mitigators. See Tr. of Oral Rearg. 36. I doubt that any defense counsel would accept this trade. The "preponderance" rule, while it sounds better, would almost surely produce more death sentences than an "equipoise beyond a reasonable doubt" requirement. 

There exists in some parts of the world sanctimonious criticism of America's death penalty, as somehow unworthy of a civilized society. (I say sanctimonious, because most of the countries to which these finger-waggers belong had the death penalty themselves until recently -- and indeed, many of them would still have it if the democratic will prevailed. n3 ) It is a certainty that the opinion of a near-majority of the United States Supreme Court to the effect that our system condemns many innocent defendants to death will be trumpeted abroad as vindication of these criticisms. For that reason, I take the trouble to point out that the dissenting opinion has nothing substantial to support it. 

n3 It is commonly recognized that "many European countries . . . abolished the death penalty in spite of public opinion rather than because of it." Bibas, Transparency and Participation in Criminal Procedure, 81 N. Y. U. L. Rev. 911, 931-932 (2006). See also id., at 932, n. 88. Abolishing the death penalty has been made a condition of joining the Council of Europe, which is in turn a condition of obtaining the economic benefits of joining the European Union. See Waters, Mediating Norms and Identity: The Role of Transnational Judicial Dialogue in Creating and Enforcing International Law, 93 Geo. L. J. 487, 525 (2005); Demleitner, Is There a Future for Leniency in the U.S. Criminal Justice System? 103 Mich. L. Rev. 1231, 1256, and n. 88 (2005). The European Union advocates against the death-penalty even in America; there is a separate death-penalty page on the website of the Delegation of the European Commission to the U.S.A. See http://www.eurunion.org/legislat/deathpenalty/deathpenhome.htm (as visited June 17, 2006, and available in Clerk of Court's case file). The views of the European Union have been relied upon by Justices of this Court (including all four dissenters today) in narrowing the power of the American people to impose capital punishment. See, e.g., Atkins v. Virginia, 536 U.S. 304, 317, n. 21, 122 S. Ct. 2242, 153 L. Ed. 2d 335 (2002) (citing, for the views of "the world community," the Brief for the European Union as Amicus Curiae). 

It should be noted at the outset that the dissent does not discuss a single case -- not one -- in which it is clear that a person was executed for a crime he did not commit. If such an event had occurred in recent years, we would not have to hunt for it; the innocent's name would be shouted from the rooftops by the abolition lobby. The dissent makes much of the new-found capacity of DNA testing to establish innocence. But in every case of an executed defendant of which I am aware, that technology has confirmed guilt.

This happened, for instance, only a few months ago in the case of Roger Coleman. Coleman was convicted of the gruesome rape and murder of his sister-in-law, but he persuaded many that he was actually innocent and became the poster-child for the abolitionist lobby. See Glod & Shear, DNA Tests Confirm Guilt of Man Executed by Va., Washington Post, Jan. 13, 2006, p. A1; Dao, DNA Ties Man Executed in '92 to the Murder He Denied, N. Y. Times, Jan. 13, 2006, p. A14. Around the time of his eventual execution, "his picture was on the cover of Time magazine ('This Man Might Be Innocent. This Man Is Due to Die'). He was interviewed from death row on 'Larry King Live,' the 'Today' show, 'Primetime Live,' 'Good Morning America' and 'The Phil Donahue Show.'" Frankel, Burden of Proof, Washington Post, May 14, 2006, pp. W8, W11. Even one Justice of this Court, in an opinion filed shortly before the execution, cautioned that "Coleman has now produced substantial evidence that he may be innocent of the crime for which he was sentenced to die." Coleman v. Thompson, 504 U.S. 188, 189, 112 S. Ct. 1845, 119 L. Ed. 2d 1 (1992) (Blackmun, J., dissenting). Coleman ultimately failed a lie-detector test offered by the Governor of Virginia as a condition of a possible stay; he was executed on May 20, 1992. Frankel, supra, at W23; Glod & Shear, Warner Orders DNA Testing in Case of Man Executed in '92, Washington Post, Jan. 6, 2006, pp. A1, A6.

In the years since then, Coleman's case became a rallying point for abolitionists, who hoped it would offer what they consider the "Holy Grail: proof from a test tube that an innocent person had been executed." Frankel, supra, at W24. But earlier this year, a DNA test ordered by a later Governor of Virginia proved that Coleman was guilty, see, e.g., Glod & Shear, DNA Tests Confirm Guilt of Man Executed by Va., supra, at A1; Dao, supra, at A14, even though his defense team had "proved" his innocence and had even identified "the real killer" (with whom they eventually settled a defamation suit). See Frankel, supra, at W23. And Coleman's case is not unique. See Truth and Consequences: The Penalty of Death, in Debating the Death Penalty: Should America Have Capital Punishment? The Experts on Both Sides Make Their Best Case, 128-129 (H. Bedau & P. Cassell eds. 2004) (discussing the cases of supposed innocents Rick McGinn and Derek Barnabei, whose guilt was also confirmed by DNA tests).

Instead of identifying and discussing any particular case or cases of mistaken execution, the dissent simply cites a handful of studies that bemoan the alleged prevalence of wrongful death sentences. One study (by Lanier and Acker) is quoted by the dissent as claiming that "'more than 110' death row prisoners have been released since 1973 upon findings that they were innocent of the crimes charged, and 'hundreds of additional wrongful convictions in potentially capital cases have been documented over the past century.'" Post, at 8 (opinion of SOUTER, J.). For the first point,  Lanier and Acker cite the work of the Death Penalty Information Center (more about that below) and an article in a law review jointly authored by Radelet, Lofquist, and Bedau (two professors of sociology and a professor of philosophy). For the second point, they cite only a 1987 article by Bedau and Radelet. See Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21. In the very same paragraph which the dissent quotes, Lanier and Acker also refer to that 1987 article as "having identified 23 individuals who, in their judgment, were convicted and executed in this country during the 20th century notwithstanding their innocence." Lanier & Acker, Capital Punishment, the Moratorium Movement, and Empirical Questions, 10 Psychology, Public Policy & Law 577, 593 (2004). This 1987 article has been highly influential in the abolitionist world. Hundreds of academic articles, including those relied on by today's dissent, have cited it. It also makes its appearance in judicial decisions -- cited recently in a six-judge dissent in House v. Bell, 386 F.3d 668, 708 (CA6 2004) (en banc) (Merritt, J., dissenting), for the proposition that "the system is allowing some innocent defendants to be executed." The article therefore warrants some further observations.

The 1987 article's obsolescence began at the moment of publication. The most recent executions it considered were in 1984, 1964, and 1951; the rest predate the Allied victory in World War II. (Two of the supposed innocents are Sacco and Vanzetti.) Bedau & Radelet, supra, at 73. Even if the innocence claims made in this study were true, all except (perhaps) the 1984 example would cast no light upon the functioning of our current system of capital adjudication. The legal community's general attitude toward criminal defendants, the legal protections States afford, the constitutional guarantees this Court enforces, and the scope of federal habeas review, are all vastly different from what they were in 1961. So are the scientific means of establishing guilt, and hence innocence -- which are now so striking in their operation and effect that they are the subject of more than one popular TV series. (One of these new means, of course, is DNA testing -- which the dissent seems to think is primarily a way to identify defendants erroneously convicted, rather than a highly effective way to avoid conviction of the innocent.)

But their current relevance aside, this study's conclusions are unverified. And if the support for its most significant conclusion -- the execution of 23 innocents in the 20th century -- is any indication of its accuracy, neither it, nor any study so careless as to rely upon it, is worthy of credence. The only execution of an innocent man it alleges to have occurred after the restoration of the death penalty in 1976 -- the Florida execution of James Adams in 1984 -- is the easiest case to verify. As evidence of Adams' innocence, it describes a hair that could not have been his as being "clutched in the victim's hand," Bedau & Radelet, supra, at 91. The hair was not in the victim's hand; "it was a remnant of a sweeping of the ambulance and so could have come from another source." Markman & Cassell, Protecting the Innocent: A Response to the Bedau-Radelet Study, 41 Stan. L. Rev. 121, 131 (1988). The study also claims that a witness who "heard a voice inside the victim's home at the time of the crime" testified that the "voice was a woman's," Bedau & Radelet, supra, at 91. The witness's actual testimony was that the voice, which said "' "In the name of God, don't do it"' " (and was hence unlikely to have been the voice of anyone but the male victim), "'sounded "kind of like a woman's voice, kind of like strangling or something . . . ."' " Markman & Cassell, Protecting the Innocent, at 130. Bedau and Radelet failed to mention that upon arrest on the afternoon of the murder Adams was found with some $ 200 in his pocket -- one bill of which "was stained with type O blood. When Adams was asked about the blood on the money, he said that it came from a cut on his finger. His blood was type AB, however, while the victim's was type O." Id., at 132. Among the other unmentioned, incriminating details: that the victim's eyeglasses were found in Adams' car, along with jewelry belonging to the victim, and clothing of Adams' stained with type O blood. Ibid. This is just a sample of the evidence arrayed against this "innocent." See id., at 128-133, 148-150.

Critics have questioned the study's findings with regard to all its other cases of execution of alleged innocents for which "appellate opinions . . . set forth the facts proved at trial in detail sufficient to permit a neutral observer to assess the validity of the authors' conclusions." Id., at 134. (For the rest, there was not "a reasonably complete account of the facts . . . readily available," id., at 145.) As to those cases, the only readily verifiable ones, the authors of the 1987 study later acknowledged, "We agree with our critics that we have not 'proved' these executed defendants to be innocent; we never claimed that we had." Bedau & Radelet, The Myth of Infallibility: A Reply to Markman and Cassell, 41 Stan. L. Rev. 161, 164 (1988). One would have hoped that this disclaimer of the study's most striking conclusion, if not the study's dubious methodology, would have prevented it from being cited as authority in the pages of the United States Reports. But alas, it is too late for that. Although today's dissent relies on the study only indirectly, the two dissenters who were on the Court in January 1993 have already embraced it. "One impressive study," they noted (referring to the 1987 study), "has concluded that 23 innocent people have been executed in the United States in this century, including one as recently as 1984." Herrera v. Collins, 506 U.S. 390, 430, n. 1, 113 S. Ct. 853, 122 L. Ed. 2d 203 (1993) (Blackmun, J., joined by STEVENS and SOUTER, JJ., dissenting). n4  

n4 See also Callins v. Collins, 510 U.S. 1141, 1158, n. 8, 114 S. Ct. 1127, 127 L. Ed. 2d 435 (1994) (Blackmun, J., dissenting from denial of certiorari) ("Innocent persons have been executed, see Bedau & Radelet, Miscarriages of Justice in Potentially Capital Cases, 40 Stan. L. Rev. 21, 36, 173-179 (1987), perhaps recently, see Herrera v. Collins, 506 U.S. 390, 113 S. Ct. 853, 122 L. Ed. 2d 203 (1993), and will continue to be executed under our death penalty scheme"). 

Remarkably avoiding any claim of erroneous executions, the dissent focuses on the large numbers of non-executed "exonerees" paraded by various professors. It speaks as though exoneration came about through the operation of some outside force to correct the mistakes of our legal system, rather than as a consequence of the functioning of our legal system. Reversal of an erroneous conviction on appeal or on habeas, or the pardoning of an innocent condemnee through executive clemency, demonstrates not the failure of the system but its success. Those devices are part and parcel of the multiple assurances that are applied before a death sentence is carried out.

Of course even in identifying exonerees, the dissent is willing to accept anybody's say-so. It engages in no critical review, but merely parrots articles or reports that support its attack on the American criminal justice system. The dissent places significant weight, for instance, on the Illinois Report (compiled by the appointees of an Illinois Governor who had declared a moratorium upon the death penalty and who eventually commuted all death sentences in the State, see Warden, Illinois Death Penalty Reform: How It Happened, What It Promises, 95 J. Crim. L. & Criminology 381, 406-407, 410 (2006)), which it claims shows that "false verdicts" are "remarkable in number." Post, at 9 (opinion of SOUTER, J.). The dissent claims that this Report identifies 13 inmates released from death row after they were determined to be innocent. To take one of these cases, discussed by the dissent as an example of a judgment "as close to innocence as any judgments courts normally render," post, at 7, n. 2: In People v. Smith, 185 Ill. 2d 532, 708 N.E.2d 365, 236 Ill. Dec. 779 (1999) the defendant was twice convicted of murder. After his first trial, the Supreme Court of Illinois "reversed [his] conviction based upon certain evidentiary errors" and remanded his case for a new trial. Id., at 534, 708 N.E.2d, at 366. The second jury convicted Smith again. The Supreme Court of Illinois again reversed the conviction because it found that the evidence was insufficient to establish guilt beyond a reasonable doubt. Id., at 542-543, 708 N.E.2d, at 370-371. The court explained: 

"While a not guilty finding is sometimes equated with a finding of innocence, that conclusion is erroneous. Courts do not find people guilty or innocent. . . . A not guilty verdict expresses no view as to a defendant's innocence. Rather, [a reversal of conviction] indicates simply that the prosecution has failed to meet its burden of proof." Id., at 545, 708 N.E.2d, at 371. 

This case alone suffices to refute the dissent's claim that the Illinois Report distinguishes between "exoneration of a convict because of actual innocence, and reversal of a judgment because of legal error affecting conviction or sentence but not inconsistent with guilt in fact," post, at 7, n. 2. The broader point, however, is that it is utterly impossible to regard "exoneration" -- however casually defined -- as a failure of the capital justice system, rather than as a vindication of its effectiveness in releasing not only defendants who are innocent, but those whose guilt has not been established beyond a reasonable doubt.

Another of the dissent's leading authorities on exoneration of the innocent is Gross, Jacoby, Matheson, Montgomery, & Patil, Exonerations in the United States 1989 Through 2003, 95 J. Crim. L. & Criminology 523 (2006) (hereinafter Gross). The dissent quotes that study's self-congratulatory "criteria" of exoneration -- seemingly so rigorous that no one could doubt the study's reliability. See post, at 8, n. 3 (opinion of SOUTER, J.). But in fact that article, like the others cited, is notable not for its rigorous investigation and analysis, but for the fervor of its belief that the American justice system is condemning the innocent "in numbers," as the dissent puts it, "never imagined before the development of DNA tests." Post, at 6 (opinion of SOUTER, J.). Among the article's list of 74 "exonerees," Gross 529, is Jay Smith of Pennsylvania. Smith -- a school principal -- earned three death sentences for slaying one of his teachers and her two young children. See Smith v. Holtz, 210 F.3d 186, 188 (CA3 2000). His retrial for triple murder was barred on double jeopardy grounds because of prosecutorial misconduct during the first trial. Id., at 194. But Smith could not leave well enough alone. He had the gall to sue, under 42 U.S.C. §  1983, for false imprisonment. The Court of Appeals for the Third Circuit affirmed the jury verdict for the defendants, observing along the way that "our confidence in Smith's convictions is not diminished in the least. We remain firmly convinced of the integrity of those guilty verdicts." 210 F.3d at 198.

Another "exonerated" murderer in the Gross study is Jeremy Sheets, convicted in Nebraska. His accomplice in the rape and murder of a girl had been secretly tape recorded; he "admitted that he drove the car used in the murder . . ., and implicated Sheets in the murder." Sheets v. Butera, 389 F.3d 772, 775 (CA8 2004). The accomplice was arrested and eventually described the murder in greater detail, after which a plea agreement was arranged, conditioned on the accomplice's full cooperation. Ibid. The resulting taped confession, which implicated Sheets, was "the crucial portion of the State's case," State v. Sheets, 260 Neb. 325, 327, 618 N.W.2d 117, 122 (2000). But the accomplice committed suicide in jail, depriving Sheets of the opportunity to cross-examine him. This, the Nebraska Supreme Court held, rendered the evidence inadmissible under the Sixth Amendment. Id., at 328, 335-351, 618 N.W.2d, at 123, 127-136. After the central evidence was excluded, the State did not retry Sheets. Sheets v. Butera, 389 F.3d at 776. Sheets brought a §  1983 claim; the U.S. Court of Appeals for the Eighth Circuit affirmed the District Court's grant of summary judgment against him. Id., at 780. Sheets also sought the $ 1,000 he had been required to pay to the Nebraska Victim's Compensation Fund; the State Attorney General -- far from concluding that Sheets had been "exonerated" and was entitled to the money -- refused to return it. The court action left open the possibility that Sheets could be retried, and the Attorney General did "not believe the reversal on the ground of improper admission of evidence . . . is a favorable disposition of charges," Neb. Op. Atty. Gen. No. 01036 (Nov. 9), 2001 Neb. AG LEXIS 35, 2001 WL 1503144, *3.

In its inflation of the word "exoneration," the Gross article hardly stands alone; mischaracterization of reversible error as actual innocence is endemic in abolitionist rhetoric, and other prominent catalogues of "innocence" in the death-penalty context suffer from the same defect. Perhaps the best-known of them is the List of Those Freed From Death Row, maintained by the Death Penalty Information Center. See http://www.deathpenaltyinfo.org/article.php?scid=6&did=110. This includes the cases from the Gross article described above, but also enters some dubious candidates of its own. Delbert Tibbs is one of them. We considered his case in Tibbs v. Florida, 457 U.S. 31, 102 S. Ct. 2211, 72 L. Ed. 2d 652 (1982), concluding that the Double Jeopardy Clause does not bar a retrial when a conviction is "reversed based on the weight, rather than the sufficiency, of the evidence," id., at 32, 102 S. Ct. 2211, 72 L. Ed. 2d 652. The case involved a man and a woman hitchhiking together in Florida. A driver who picked them up sodomized and raped the woman, and killed her boyfriend. She eventually escaped and positively identified Tibbs. See id., at 32-33, 102 S. Ct. 2211, 72 L. Ed. 2d 652. The Florida Supreme Court reversed the conviction on a 4-to-3 vote. 337 So.2d 788 (1976). The Florida courts then grappled with whether Tibbs could be retried without violating the Double Jeopardy Clause. The Florida Supreme Court determined not only that there was no double-jeopardy problem, 397 So.2d 1120, 1127 (1981) (per curiam), but that the very basis on which it had reversed the conviction was no longer valid law, id., at 1125, and that its action in "reweighing the evidence" in Tibbs' case had been "clearly improper," id., at 1126. After we affirmed the Florida Supreme Court, however, the State felt compelled to drop the charges. The State Attorney explained this to the Florida Commission on Capital Cases: "'By the time of the retrial, [the] witness/victim . . . had progressed from a marijuana smoker to a crack user and I could not put her up on the stand, so I declined to prosecute. Tibbs, in my opinion, was never an innocent man wrongfully accused. He was a lucky human being. He was guilty, he was lucky and now he is free. His 1974 conviction was not a miscarriage of justice.'" Florida Commission on Capital Cases, Case Histories: A Review of 24 Individuals Released From Death Row 136-137 (rev. Sept. 10, 2002) http://www.floridacapitalcases.state.fl.us/Publications/innocentsproject.pdf. Other state officials involved made similar points. Id., at 137.

Of course, even with its distorted concept of what constitutes "exoneration," the claims of the Gross article are fairly modest: Between 1989 and 2003, the authors identify 340 "exonerations" nationwide -- not just for capital cases, mind you, nor even just for murder convictions, but for various felonies. Gross 529. Joshua Marquis, a district attorney in Oregon, recently responded to this article as follows:  

"Let's give the professor the benefit of the doubt: let's assume that he understated the number of innocents by roughly a factor of 10, that instead of 340 there were 4,000 people in prison who weren't involved in the crime in any way. During that same 15 years, there were more than 15 million felony convictions across the country. That would make the error rate .027 percent -- or, to put it another way, a success rate of 99.973 percent." The Innocent and the Shammed, N.Y. Times, Jan. 26, 2006, p. A23. 

The dissent's suggestion that capital defendants are especially liable to suffer from the lack of 100% perfection in our criminal justice system is implausible. Capital cases are given especially close scrutiny at every level, which is why in most cases many years elapse before the sentence is executed. And of course capital cases receive special attention in the application of executive clemency. Indeed, one of the arguments made by abolitionists is that the process of finally completing all the appeals and reexaminations of capital sentences is so lengthy, and thus so expensive for the State, that the game is not worth the candle. The proof of the pudding, of course, is that as far as anyone can determine (and many are looking), none of cases included in the .027% error rate for American verdicts involved a capital defendant erroneously executed.

Since 1976 there have been approximately a half million murders in the United States. In that time, 7,000 murderers have been sentenced to death; about 950 of them have been executed; and about 3,700 inmates are currently on death row. See Marquis, The Myth of Innocence, 95 J. Crim. L. & Criminology 501, 518 (2006). As a consequence of the sensitivity of the criminal justice system to the due-process rights of defendants sentenced to death, almost two-thirds of all death sentences are overturned. See ibid. "Virtually none" of these reversals, however, are attributable to a defendant's "'actual innocence.'" Ibid. Most are based on legal errors that have little or nothing to do with guilt. See id., at 519-520. The studies cited by the dissent demonstrate nothing more.

Like other human institutions, courts and juries are not perfect. One cannot have a system of criminal punishment without accepting the possibility that someone will be punished mistakenly. That is a truism, not a revelation. But with regard to the punishment of death in the current American system, that possibility has been reduced to an insignificant minimum. This explains why those ideologically driven to ferret out and proclaim a mistaken modern execution have not a single verifiable case to point to, whereas it is easy as pie to identify plainly guilty murderers who have been set free. The American people have determined that the good to be derived from capital punishment -- in deterrence, and perhaps most of all in the meting out of condign justice for horrible crimes -- outweighs the risk of error. It is no proper part of the business of this Court, or of its Justices, to second-guess that judgment, much less to impugn it before the world, and less still to frustrate it by imposing judicially invented obstacles to its execution.