Concurring and Dissenting.—I concur in the judgment except insofar as it orders vacation of petitioner Sakarias’s death judgment. As to Sakarias, I respectfully dissent.
Sakarias conspired with petitioner Waidla to rob and kill Avo and Yiivi Piirisild, their former benefactors. They invaded the Piirisilds’ home, lay in wait, and ambushed and murdered Yiivi. During a prolonged and horrific assault, numerous blows and wounds were inflicted with a knife and a hatchet. Sakarias admitted he used both weapons on the victim. As planned, the killers carried away property from the Piirisild residence. When later apprehended, Sakarias denied remorse and insisted he and Waidla wanted to kill Avo as well. The majority concede, as they must, that Sakarias’s guilt of capital murder is conclusive, and that ample evidence supports his jury’s decision to sentence him to death.
Yet the majority say Sakarias’s due process rights were prejudicially violated as to penalty when, in their separate trials, Prosecutor Ipsen, acting *172in “bad faith,” sought to enhance each petitioner’s culpability by attributing the same single antemortem hatchet-chopping wound to each killer. Ipsen’s bad faith is demonstrated, the majority assert, because (1) the available evidence pointed strongly to Waidla, not Sakarias, as the perpetrator of this act, and (2) Ipsen “manipulated” the evidence at Sakarias’s trial by failing to elicit certain medical testimony he had earlier used to attribute the same act to Waidla. The majority insist that the misimpression thus conveyed to Sakarias’s jury may have influenced its penalty decision. I disagree.
At the outset, I discern no bad faith in Ipsen’s conduct. Our referee was never asked to make such a finding, and he did not do so. In my view, there is no basis for a bad faith determination. Ipsen adhered to the well-established rule against the knowing presentation of false evidence. (Napue v. Illinois (1959) 360 U.S. 264, 269 [3 L.Ed.2d 1217, 79 S.Ct. 1173]; In re Jackson (1992) 3 Cal.4th 578, 595-596 [11 Cal.Rptr.2d 531, 835 P.2d 371]; see also Brown v. Borg (9th Cir. 1991) 951 F.2d 1011, 1014—1017.) Moreover, he presumably discharged his obligation to give the defense any otherwise unavailable evidence he possessed that materially undermined the prosecution’s case against Sakarias. (E.g., Brady v. Maryland (1963) 373 U.S. 83 [10 L.Ed.2d 215, 83 S.Ct. 1194] (Brady).) Nor was there any secret, at the time of Sakarias’s trial, about Ipsen’s case against Waidla. Because Sakarias’s trial followed Waidla’s, this information was by then a matter of public record.1
Under such circumstances, the People would not generally be required to introduce, in their own case, evidence helpful to the defense. Instead, the prosecution could properly rely on the defense to expose the gaps and weaknesses in its proof.2 I see no reason why a different rule should apply in Sakarias’s case simply because the omitted evidence was earlier presented against Waidla.
Further, I believe Ipsen demonstrated no bad faith by theorizing, in each defendant’s trial, that the antemortem hatchet chop was inflicted by that defendant. I have two reasons for this conclusion. First, the law governing inconsistent prosecutorial arguments is complex and unsettled; indeed, a case presenting such issues is currently under review by the United States Supreme Court. (Stumpf v. Mitchell (6th Cir. 2004) 367 F.3d 594 (Stumpf), *173cert. granted sub nom. Mitchell v. Stumpf (2005) 543 U.S. 1042 [160 L.Ed.2d 610, 125 S.Ct. 824].) Because appellate judges significantly disagree about what a prosecutor may and may not do in this regard, there seems little basis to conclude that Ipsen’s strategy constituted bad faith. Second, the evidence of who delivered the antemortem chopping blow is hardly as clear as the majority suggest. Our referee found that Ipsen himself was sincerely uncertain which of the two murderers had committed this particular act. Under these circumstances, I cannot find bad faith in Ipsen’s efforts to make a plausible case against each petitioner.
Federal courts of appeals have reached mixed results when faced with claims that prosecutors violated defendants’ due process rights by presenting inconsistent theories in separate trials. For example, in Drake v. Francis (11th Cir. 1984) 727 F.2d 990, reversed in part sub nomine Drake v. Kemp (11th Cir. 1985) 762 F.2d 1449, the prosecution won separate convictions against two men for robbing, and fatally beating and stabbing, a Georgia barber in his shop. In defendant Campbell’s trial, the prosecution urged that Campbell was the actual murderer, while in Drake’s trial, the state claimed that, given Campbell’s physical disabilities, Drake must have joined the attack on the victim. A three-judge panel rejected Drake’s “novel” (id., at p. 994) claim of a due process violation, observing that the state’s theories in the two trials were “fairly consistent.” (Ibid.) In each trial, the panel noted, the state made clear its belief that both men were involved in the criminal episode, and merely stressed the degree of participation of the defendant then on trial. (Ibid.)3
Several other cases have found due process implications in the state’s use of inconsistent theories against separately tried codefendants. But these decisions are characterized by stark facts, and by sharp divisions within the reviewing courts.
For example, in Smith v. Groose (8th Cir. 2000) 205 F.3d 1045, the prosecutor, in separate trials, used contradictory statements by the same witness—one of which was necessarily false—to convict, on conflicting factual theories, two unrelated defendants (members of separate burglary parties) for felony murder of the same victim. In these circumstances, the three-judge panel held that “the Due Process Clause forbids a state from using inconsistent, irreconcilable theories to secure convictions against two or more defendants in prosecutions for the same offenses arising out of the same event.” (Id., at p. 1049, italics added.)
*174In Thompson v. Calderon (9th Cir. 1997) 120 F.3d 1045 (Thompson), reversed on other grounds sub nomine Calderon v. Thompson (1998) 523 U.S. 538 [140 L.Ed.2d 728, 118 S.Ct. 1489], the prosecution won a murder conviction against defendant Leitch on the theory that Leitch enlisted Thompson’s help to kill victim Fleischli because Fleischli was interfering in Leitch’s marriage. Later, though no new evidence had surfaced, the prosecution switched gears entirely and urged in Thompson’s trial, through different witnesses, that Thompson, acting alone, had raped and then killed Fleischli. On habeas corpus, Thompson argued that his capital rape and murder convictions, obtained on this inconsistent ground, violated his due process rights.
Because the case involved a threshold procedural dispute, and because relief was granted on other grounds, not all 11 en banc judges addressed the “inconsistent theories” claim. Four judges found Thompson had suffered a prejudicial due process violation, reasoning that, in his case, the prosecution, using unreliable informants, had deviated from the theory of joint culpability it had otherwise consistently advanced. (Thompson, supra, 120 F.3d 1045, 1047, 1055-1059 (plur. opn. of Fletcher, J.).) Two more judges advocated a remand to determine prejudice. (Id., at pp. 1063-1064 (conc. opn. of Tashima, J.).)
However, three other members of the Thompson court expressed serious reservations about the notion that inconsistent prosecutorial theories in separate trials are constitutionally improper, at least where the evidence is ambiguous and the prosecution introduces no knowing falsehoods.
Judge Kozinski, for example, noted a line of cases holding that judicial estoppel will not apply against the government. (Thompson, supra, 120 F.3d 1045, 1070 (dis. opn. of Kozinski, J.), citing Nichols v. Scott (5th Cir. 1995) 69 F.3d 1255, 1272, United States v. McCaskey (5th Cir. 1993) 9 F.3d 368, 378 & United States v. Kattar (1st Cir. 1988) 840 F.2d 118, 129-130, fin. 7.) Moreover, Judge Kozinski observed, “[pjrosecutors are not omniscient. [Citation.] They may be confronted with witnesses who present mutually inconsistent versions of what happened, and there may be no way of knowing which version—if any—is true. Is the prosecutor then precluded from presenting either case to the jury? Must he pick one based on his intuition? I believe not. A prosecutor, like any other lawyer, is entitled to retain skepticism about the evidence he presents and trust the jury to make the right judgment. After all, the guarantee of due process encompasses a fair trial before a fair judge and jury; the right to a lawyer and to exculpatory evidence available to the prosecution; and the right not to have the prosecutor lie to the jury. But I cannot see that it encompasses the right to have a prosecutor who is convinced of the defendant’s guilt. We trust the adversary process, the good *175sense of jurors, the presumption of innocence and the prosecution’s heavy burden of proof to ensure a verdict that is fair to the defendant.” (Thompson, supra, 120 F.3d 1045, 1071 (dis. opn. of Kozinski, J.).)4
In a separate dissenting opinion, Judge Kleinfeld put it another way: “We, not prosecutors, are supposed to adhere to stare decisis. That is because equal justice for all requires that we decide like cases alike. Prosecutors are not bound by the principle of stare decisis, because they do not decide cases. A prosecutor cannot present evidence or a case he knows to be false. But there is no reason to think the prosecutor [of Leitch and Thompson] knew either theory he presented was false, when he presented it or at any time. [][] . . . [f] There is no reason why a prosecutor’s change of theory at a later time should be treated as a due process violation. The standard boilerplate instruction tells juries that ‘arguments of counsel are not evidence.’ [Citation.] The jury is supposed to decide the case based on the evidence and the judge’s instructions. The lawyers offer the jury theories to help them make sense of the evidence. But . . . [t]he lawyers were not at the scene of the crime, and can only, like the jurors, draw inferences. It is up to the jury, not the prosecutor, to decide what happened amidst a lot of lies.” (Thompson, supra, 120 F.3d 1045, 1074-1075 (dis. opn. of Kleinfeld, J.).)
In Stumpf, supra, 367 F.3d 594—the case currently under high court review—the crucial issue for capital murder, which required intent to kill, was which of two accomplices, Wesley or Stumpf, fired shots that killed Mary Jane Stout during a robbery of the Stout home. Criminal proceedings against Stumpf proceeded while Wesley’s extradition from outside the state was pending. Stumpf pled guilty to capital murder charges. At the required evidentiary hearing to establish a factual basis for the plea, the state sought to establish Stumpf s homicidal intent by showing he was the actual killer. To this end, the state presented evidence that the fatal shots were fired from the gun Stumpf had earlier used to wound Norman Stout. The three-judge plea court found that Stumpf had personally killed Mary Jane. At the time of sentencing, the same court made this an express factor in its decision to impose the death penalty.
Later, at Wesley’s capital murder trial, the prosecution argued that Wesley was the actual killer. For this purpose, the prosecution adduced the testimony *176of Wesley’s cellmate, not available when Stumpf was convicted and sentenced. The cellmate testified Wesley had admitted that, after shooting Norman, Stumpf panicked, dropped his weapon, and fled, whereupon Wesley picked up the gun and used it to murder Mary Jane. On this evidence, Wesley’s jury convicted him of capital murder (though it later recommended against the death penalty). Stumpf thereupon sought to withdraw his own guilty plea, but the state opposed the motion, urging that Wesley’s cellmate was not a credible witness.
On federal habeas corpus, a majority of the three-judge court of appeals panel held that the state had prejudicially violated Stumpf’s due process rights as to both guilt and penalty. The majority reasoned that the state had cast doubt upon the fundamental fairness and reliability of Stumpf’s death judgment by later pressing, against Wesley, an inconsistent, irreconcilable factual theory of the same capital murder. (Stumpf, supra, 367 F.3d 594, 610-618; but see id., at pp. 618-623 (dis. opn. of Boggs, C. J.).) As noted, it appears the United States Supreme Court will decide whether the court of appeals majority was correct.
Here, Ipsen did not use inconsistent theories to obtain capital murder judgments against two defendants, where only one could be guilty or death-eligible. There is no doubt that Sakarias and Waidla together committed the first degree murder of Viivi Piirisild with special circumstances, and that both men were enthusiastic participants in the gruesome attack. The only dispute in this case relates to a particular detail which a jury could consider in deciding whether death was the appropriate penalty for each killer. Under these circumstances, and given the uncertain state of the law, I cannot find in Ipsen’s conduct an act of bad faith amounting to a constitutional violation.
The majority insist Ipsen lacked good faith justification for attributing the antemortem chopping blow to Sakarias, because the evidence clearly pointed to Waidla as the culprit. Again, I disagree. Ipsen was not an eyewitness to Viivi’s murder. Neither Waidla nor Sakarias confessed to delivering the disputed blow. Evidence on that point was entirely circumstantial. In my view, it was not so clear as to preclude Ipsen from reserving personal judgment—as he did—and presenting the plausible case against each man.
As the majority notes, the case against Waidla is as follows: When the attack began in the Piirisilds’ living room, Sakarias was using a knife, while Waidla was using the blunt edge of the hatchet to bludgeon the victim. At some point, she was moved from the living room to the bedroom. There is evidence she was already dead by that time, and the two hatchet chops to which Sakarias admitted were, he said, inflicted in the bedroom. Hence, the earlier, antemortem chopping blow must have been inflicted in the living room, and by Waidla.
*177This is a plausible scenario, but a substantial case could be made against Sakarias as well. After all, neither Sakarias nor Waidla ever attributed any hatchet chops to Waidla. On the other hand, Sakarias conceded that he did use the hatchet to deliver such blows. Indeed, Sakarias admitted inflicting two of the three hatchet chops disclosed by the evidence, all of which were in relatively close proximity on the victim’s head. As Ipsen implied in Sakarias’s trial, one could infer that Sakarias had simply understated by one the number of chops he delivered. Moreover, while most of the bloodstains and spatters were found in the living room, there were significant spatters in the bedroom as well. In one place in the bedroom, a detective testified at Sakarias’s trial, there was enough blood “to actually start to trickle down the wall.” This was some, if not conclusive, evidence that the victim bled in the bedroom, and was thus still alive.
The majority claim Waidla must have inflicted the antemortem wound because the medical examiner testified at Waidla’s trial that an abrasion on Viivi’s back was probably caused by dragging, and that the “nonhemorrhagic” nature of the abrasion suggested it was sustained postmortem. Thus, the majority infer, the abrasion must have occurred when the victim, already dead, was dragged to the bedroom. Because Sakarias admitted only two chop wounds in the bedroom, the majority reason he must have inflicted the two postmortem wounds, and none other.
But this analysis is hardly conclusive. Expert opinions are often subject to debate and interpretation.5 Moreover, even if we credit an inference that the victim was already dead when she was moved to the bedroom, that does not prove Sakarias wielded the hatchet only thereafter. We know this weapon was transferred from Waidla to Sakarias at some point, but we do not know exactly when. Given the uncertain evidence, I conclude that so long as Ipsen did not introduce false evidence, he acted in good faith by presenting alternative theories about this detail of the attack. (See, e.g., Parker v. Singletary (11th Cir. 1992) 974 F.2d 1562, 1578.) Having satisfied his discovery obligations, Ipsen could properly rely, in each case, on the due diligence of the defense to expose weaknesses in the People’s proof.
By concluding that Ipsen could only present the case they think is stronger, the majority intrude much too far into matters which, for good reason, have traditionally been left to prosecutorial discretion. I cannot join the majority’s attempt to second-guess the prosecution’s strategy in this way.
*178In any event, I do not accept the majority’s conclusion that Sakarias suffered prejudice on the issue of penalty. The undisputed details of Sakarias’s role in this brutal murder are aggravated in the extreme. Though the Piirisilds had bestowed many kindnesses on Sakarias and Waidla, the two men decided they had been slighted by their benefactors. Filled with hate and greed, Sakarias and Waidla hatched a plan to burglarize and rob the couple. But once in the Piirisild home, they did not simply take property and leave. Instead, they “started waiting for Viivi” with murderous intent. (People v. Sakarias (2000) 22 Cal.4th 596, 613 [94 Cal.Rptr.2d 17, 995 P.2d 152].)
As the majority recite, Sakarias personally used his knife to stab Viivi at least four times in the chest. Two of these wounds were potentially fatal. Sakarias ceased his attack with the knife only when its handle broke off. He also inflicted at least two chopping wounds to Viivi’s head with the hatchet. His jury must have understood that, regardless of whether Viivi was then still alive, Sakarias administered these blows for the purpose of ensuring her death.
Far from horrified at their bloody work, the murderers stayed for a snack, and Sakarias calmly ate liverwurst from the Piirisilds’ refrigerator. Later, he made clear to the police his only regret—that they had not killed Avo Piirisild as well.
Thus, the uncontroverted evidence demonstrates Sakarias’s full, remorseless involvement in the murder plot, and details the many grievous blows and wounds he administered during the joint attack on the victim. Under these circumstances, it stretches credulity to suggest that the issue whether he inflicted a single additional blow—gruesome as it was—could alone have tipped the jury’s penalty determination.
The majority note that Ipsen made Sakarias’s responsibility for the antemortem hatchet chop a significant theme of his argument. But such references occurred, for the most part, at the guilt phase of Sakarias’s trial, where they could have caused no prejudice. As the majority concede, Ipsen gave this subject only two brief references at the penalty phase. The bulk of Ipsen’s penalty argument was devoted to rebutting the defense case in mitigation, including Sakarias’s claims of extreme mental disorder. In my view, this further reduces any chance that the penalty outcome was affected.6
*179I do not mean to imply that I would never find prejudicial misconduct in a prosecutor’s use of irreconcilable theories and evidence against separately tried defendants. As the cases have suggested, difficult questions arise where, for example, such tactics lead to the convictions of two persons for a crime only one could have committed. But such issues are not presented here. Nothing that happened in this case persuades me that the penalty judgment against Sakarias is unfair.
I would discharge the order to show cause as to both Waidla and Sakarias.
Thus, Ipsen did not even have a Brady duty to alert Sakarias’s counsel to the evidence presented in Waidla’s trial. There is no Brady claim of prosecutorial suppression of evidence “ ‘when information is fully available to a defendant at the time of trial and his only reason for not obtaining and presenting the evidence to the Court is his lack of reasonable diligence. . . .’” (People v. Morrison (2004) 34 Cal.4th 698, 715 [21 Cal.Rptr.3d 682, 101 P.3d 568], quoting United States v. Brown (5th Cir. 1980) 628 F.2d 471, 473.) Surely reasonable diligence in Sakarias’s case included reading the public transcript of his codefendant’s trial.
As we recently noted, “the adversary system [remains] the primary means by which truth is uncovered. [Citation.]” (People v. Morrison, supra, 34 Cal.4th 698, 715.)
Reviewing, en banc, the three-judge panel’s decision, the court of appeals granted Drake relief on other grounds. A majority of the en banc court thus declined to discuss Drake’s “inconsistent theories” claim. (Drake v. Kemp, supra, 762 F.2d 1449, 1451.) Only one member would have addressed and granted relief on this claim. (Id., at pp. 1470-1479 (conc. opn. of Clark, J.).)
Judge Kozinski conceded that, if only one defendant could have committed a capital crime, but two were convicted and sentenced to death in separate trials on irreconcilable theories by mutually inconsistent verdicts, the state might be required to determine which judgment was false, and to take steps to set it aside, because “the better view seems to be that the state has no right to execute an innocent man, [no matter how] fairly it has obtained the conviction.” (Thompson, supra, 120 F.3d 1045, 1071 (dis. opn. of Kozinski, J.), italics added.) Such is hardly the situation here.
1 realize the evidence that the victim “bled out” on the living room carpet may also support an inference that she was dead or dying when later moved to the bedroom. Of course, this evidence was presented in Sakarias’s trial. Thus, the inference was there for Sakarias’s jury to draw. This further diminishes the possibility that, because the “back abrasion” testimony was omitted from Sakarias’s trial, Sakarias’s jury got a materially misleading picture of the chronology of events.
The majority note that Sakarias’s penalty jury deliberated extensively, and announced at one point they were unable to reach a verdict. But we expect no less than careful deliberation in a capital case. Any difficulties the jury experienced in reaching a penalty consensus most likely arose from the case in mitigation, which included evidence of Sakarias’s youth, the harsh life he had endured as an Estonian conscript in the Soviet Army, and his mental and emotional problems.