dissenting.

Petitioner was convicted in California state court of first-degree murder and sentenced to death. On appeal, he argued that application of the death penalty in this case was arbitrary because it was excessive when compared with penalties imposed in similar cases. The California Supreme Court noted that petitioner “presented] an elaborate survey of published [California] Court of Appeal decisions to demonstrate the hypothesis that many first de[*1054] gree murderers of equal or greater culpability have received sentences less than death.” 50 Cal. 3d 668, 718, 789 P. 2d 887, 916 (1990). However, the State Supreme Court refused to review petitioner’s submissions, declaring that “[cjomparative proportionality review is not constitutionally required.” Ibid. Although the court cited its own prior decisions for that conclusion, ibid., those precedents ultimately derive from this Court’s opinion in Pulley v. Harris, 465 U. S. 37 (1984). See People v. Rodriguez, 42 Cal. 3d 730, 778, 726 P. 2d 113, 143-144 (1986) (relying on Pulley in rejecting proportionality review). In Pulley, the Court sustained California’s capital punishment statute against Eighth Amendment attack, rejecting the claim that the “Eighth Amendment . . . requires a state appellate court, before it affirms a death sentence, to compare the sentence in the case before it with the penalties imposed in similar cases if requested to do so by the prisoner.” 465 U. S., at 43-44.

I dissented from the decision in Pulley, and I continue to believe that it was wrongly decided. The singling out of particular defendants for the death penalty when their crimes are no more aggravated than those committed by numerous other defendants given lesser sentences is unacceptable. As Justice Brennan pointed out in his dissent in Pulley, comparative proportionality review, at the very least, “serves to eliminate some of the irrationality that currently surrounds imposition of a death sentence” and “can be administered without much difficulty by a court of statewide jurisdiction.” 465 U. S., at 71. In the present case, petitioner has not merely “requested” review for comparative proportionality, cf. id., at 44, but has (in the lower court’s own words) “present[ed] an elaborate survey of published Court of Appeal decisions,” allegedly showing that “many first degree murderers of equal and greater culpability have received sentences less than death.” 50 Cal. 3d, at 718, 789 P. 2d, at 916. I cannot understand how this Court can reconcile a refusal to review such evidence with our capital jurisprudence.

As we have often recognized, “[bjecause of the uniqueness of the death penalty, ... it [cannot] be imposed under sentencing procedures that creat[e] a substantial risk that it would be inflicted in an arbitrary and capricious manner.” Gregg v. Georgia, 428 U. S. 153, 188 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.). Indeed, we have “insiste[d] that capital punishment be imposed fairly, and with reasonable consistency, or not at[*1055] all.” Eddings v. Oklahoma, 455 U. S. 104, 112 (1982). The allegation in this petition, accompanied by a proffer of significant evidence, is that the death sentence imposed upon petitioner was not “fair” precisely because it was not “consistent.” The refusal even to consider petitioner’s evidence surely “creates a substantial risk” that “arbitrary and capricious” capital punishment will result. I would hope that the Court would reexamine its views on this matter. This petition should be granted and the case remanded for an examination of petitioner’s submissions.

Even if I did not believe that failure to consider petitioner’s evidence on the issue of proportionality violated the Eighth Amendment, I would grant the petition and vacate the sentence below, adhering to my view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. Gregg v. Georgia, supra, at 231 (Marshall, J., dissenting).