I concur in the opinion to the extent that it affirms the judgment of guilt, and the finding of special circumstances. I dissent to the imposition of the death penalty.
I.
In Lockett v. Ohio (1978) 438 U.S. 586, 604 [57 L.Ed.2d 973, 990, 98 S.Ct. 2954], the United States Supreme Court said that “in all but the rarest kind of capital case,” the jury cannot “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.” In People v. Harris (1984) 36 Cal.3d 36, 71 [201 *977Cal.Rptr. 782, 679 P.2d 433], we held, citing Lockett, that a trial court erred in excluding evidence of Harris’s poetry. That poetry, we noted, could show Harris as a unique, sensitive, feeling person (pp. 69-70); it could also show that he possessed a talent which he could use for the benefit of others if he were allowed to live.
Defendant in the present case is an accomplished artist who has sold over 90 paintings and won numerous prizes. If a jury is entitled to hear and consider Harris’s poetry as a mitigating factor, certainly it would be entitled to consider defendant’s far greater artistic achievement.
Under the instructions given by the trial court, however, defendant’s artistic talent could be considered, if at all, only under Penal Code section 190.3, factor (k), which speaks of matters extenuating the gravity of the crime. We have recognized, however, that those instructions are ambiguous, and could be misunderstood to permit jury consideration only of matters which somehow lessened the seriousness of the crime itself, not of other character and background evidence. (People v. Easley (1983) 34 Cal.3d 858, 878 at fn. 10 [196 Cal.Rptr. 309, 671 P.2d 813].)
The prosecutor here resolved that ambiguity, informing the jury clearly and unequivocally—but erroneously—that it could not consider defendant’s artistry as a mitigating consideration under factor (k). He said: “Were there any other extenuating circumstances that somehow lessened the gravity of the crime? No. The only thing that the defendant has brought forward here that he testified to, he says he testified to it in behalf of his argument for death, is his ability to paint, [if] Well, we can see that. These paintings are high quality. He does fine work for someone who certainly learned on his own, took the initiative to learn it in prison. He should be commended for that. That’s fine. But it certainly is not a factor in mitigation, [if] When you hear the law as I anticipate the Court will instruct you, you won't hear anything about whether or not someone is a painter. If this defendant happened to be a carpenter, happened to be a plumber, happened to be a lawyer and he went out and did these deeds that he did, just because he was any of those professions wouldn’t make it a mitigating factor. In fact, some people might think if he was a lawyer who did it, it would be a factor in aggravation; but be that as it may, the occupation of a man is not a factor in mitigation.” (Italics added.)
The majority opinion acknowledges that the prosecutor’s statement was erroneous, but asserts that he “later retreated somewhat” (p. 957) from this view when he later conceded that defendant’s character and background were a proper consideration but insufficient to warrant a life sentence. “In *978so arguing,” they assert, “he implicitly retreated from any prior erroneously narrow interpretation of factor (k).” (P. 957.)
These carefully qualified words—“later retreated somewhat,” and “implicitly retreated”—betray the majority’s uneasiness with their conclusion. (The majority’s addition of a discussion of prejudice, after finding no error, further indicates their insecurity with that conclusion.) What the prosecutor did was argue that defendant’s brutalized childhood did not justify giving him only a life sentence. From that discussion one could infer, perhaps, a concession that under some circumstances a defendant’s childhood might be mitigating.1 But it would take a careful and sophisticated legal analysis to infer from that implied concession a further implied concession that defendant’s artistic talent could be considered.2
What the jury heard was a clear, unequivocal statement from the prosecutor that defendant’s artistry could not be considered as a mitigating factor... It is sophistry to suggest that the jurors may have taken facially unrelated statements by the prosecutor and, by a process of complex legal reasoning, found a latent ambiguity in the prosecutor’s comments.
The majority assert, however, that the prosecutor’s misstatement was “cured” by defense counsel. (P. 957.) I can understand how a judicial admonition may cure improper argument, but how can a comment by defense counsel do the job? At best, all defense counsel can do is raise a conflict, leaving the jury with two contradictory interpretations of the law, one correct and one erroneous.
When the majority assert that defense counsel “cured” the prosecutor’s error, they assume that the jury followed the defense interpretation. I see no basis for making that assumption. If we must decide which view the jury *979followed, the better bet is that they followed the prosecutor. As pointed out by Justice Peters in People v. Talle (1952) 111 Cal.App.2d 650, 677 [245 P.2d 633]: “Defense counsel and the prosecuting officials do not stand as equals before the jury. Defense counsel are known to be advocates for the defense. The prosecuting attorneys are government officials and clothed with the dignity and prestige of their office. What they say to the jury is necessarily weighted with that prestige.”3
It is not necessary, however, for us to decide whom the jury believed. The issue we must decide is that stated by Justice O’Connor in her concurring opinion in California v. Brown (1987) 479 U.S. 538, 546 [93 L.Ed 2d 934, 943, 107 S.Ct. 837]]: whether “[i]n combination with the instructions, the comments of the prosecutor”—note that she does not mention defense counsel—“may create a ‘legitimate basis for finding ambiguity concerning the factors actually considered by the’ jury.” Here the instruction was ambiguous, the prosecutor said the jury could not consider defendant’s artistic talent, and the defense said it could. On these facts, how can anyone say there is no legitimate basis for ambiguity as to whether the jury thought it could consider that evidence?
II.
I join in the concurring and dissenting opinion of Justice Mosk, which concludes that the prosecutor’s argument misled the jurors with respect to both the nature of the penalty determination and the responsibility which they bear. I write to add my views on some aspects of this issue.
This is the third case in recent months in which we consider a prosecutor’s argument designed to persuade a jury that the law in some fashion relieves jurors of full responsibility to determine whether a defendant should live or die. In People v. Hendricks (1988) 44 Cal.3d 635 [244 *980Cal.Rptr. 181, 749 P.2d 836], the prosecutor told the penalty jury that they were finders of fact, that while they had a very important decision to make, the law “ ‘takes a little bit of sting out in the sense that you have to decide facts. Once you decide, if you do, that the aggravating circumstances outweigh the mitigating circumstances, its automatic. You shall impose death.’ ” (P. 659.) Despite this obviously erroneous advice given the jury—a penalty jury does not find facts, but functions as the conscience of the community to render a normative judgment—this court affirmed the death penalty.
In People v. Milner (1988) 45 Cal.3d 227 [246 Cal.Rptr. 713, 753 P.2d 669], the prosecutor told the jurors that the law did not require them to shoulder a burden of determining what “is just or true in this particular case.” (P. 254.) The jury should return the death penalty, he said, because “ ‘when you add up the circumstances in this case, there is no other conclusion that you can come to, if you stick within the protection of the law . . . . [1f] All juries facing the same set of facts, would come up with the same decision, and this law, hiding under the veil of that law, will protect you from shouldering the weight of your decision.’ ” (P. 255.) The court reversed the penalty judgment, without citing Hendricks, supra, 44 Cal.3d 635.
In the present case the prosecutor told the jury: “It’s never easy for someone to ask for another man’s life. But your burden is lightened in this case because of the law. [1{] The way the law is set up, as I’ll explain it to you, the weighing process that you go through and the fact that if the aggravating circumstances outweigh the mitigating circumstances, you shall return a verdict of death, flj] It’s just that simple. The law lightens your burden in that regard .... [Y]our duty as a juror is simply to apply the facts. . . . You simply apply those facts to the law, to reach the determination ... as to what is the appropriate punishment. ... If the circumstances in aggravation outweigh the circumstances in mitigation, you shall return a verdict of death.” The majority affirm the death judgment, without citing Milner, supra, 45 Cal.3d 227.
The prosecutors in each of these three cases are saying basically the same thing. They are telling the jurors that they do not have the ultimate responsibility to determine whether the defendant should live or die. The law has taken the sting out of that decision, protected them from that responsibility, lightened their burden. Their role is a limited one. According to the prosecutors’ argument, the jurors are finders of fact, who determine what aggravating and mitigating factors exist, place them in a scale, and report how the balance tips.
*981This argument radically misstates the role of the penalty jury. In People v. Brown (1985) 40 Cal.3d 512 [220 Cal.Rptr. 637, 709 P.2d 440] and People v. Allen (1986) 42 Cal.3d 1222 [232 Cal.Rptr. 849, 729 P.2d 115], we recognized that jury instructions in the language of the 1987 law could reasonably be interpreted “to require a juror (i) to determine whether ‘the aggravating circumstances outweighed the mitigating circumstances’ without regard to the juror’s personal view as to the appropriate sentence, and then (ii) to impose a sentence of death if aggravation outweighs mitigation even if the juror does not personally believe death is the appropriate sentence under all the circumstances . . . .” (Allen, supra, 42 Cal.3d at p. 1277.) As Allen emphasized, however, “we concluded in Brown that the statute was not intended to, and should not, be interpreted in that fashion.” (Ibid.) The prosecutor’s argument here interpreted the instructions in exactly the manner we condemned in Brown and Allen. To tell the jurors that their function is one of finding facts, that once they have found the facts the law dictates the result, and that the law shields them from the full responsibility of determining whether the defendant should live or die, is error—in this case, in Milner, supra, 45 Cal.3d 227, and in Hendricks, supra, 44 Cal.3d 635. If there is a principled ground to distinguish the three decisions, it does not appear in our opinions.
The majority argues that the prosecutor in this case also spoke of the “appropriate punishment,” a term which may imply a moral determination, and told the jurors that defendant deserved death. Was the prosecutor equivocating—telling the jury with one breath to render an objective, factual determination, and with the next to decide what result is morally appropriate? I think not. In context, it is clear to me that when the prosecutor said death was the “appropriate” punishment,.he meant only that it was the punishment which would result from an objective weighing of aggravating and mitigating factors by a fact-finding jury. And while he told the jury that defendant deserved death, he did not say that the jury’s task was to make that determination. To the contrary, his message was clear and consistent: if you follow the law and carry out an objective, non-normative weighing of the factors you must return a death penalty—and do not feel guilty about it, because fortunately this defendant really deserves death. Thus he concluded his argument: “The law provides for it, the law provides a weighing process and the law provides an answer here. It is a just answer and justice should be done according to the law and a death returned.”
Since our opinions to date do not provide a method for analyzing issues such as that raised in the present case, I take this opportunity to set out my view of the matter. The 1978 death penalty law requires that jurors decide whether aggravating factors outweigh mitigating factors; the Constitution requires that they decide whether death is the appropriate penalty for this *982defendant. There are two logical ways in which the jurors can relate these decisions. First, what seems to me the more natural way, is to weigh the factors objectively and use that analysis as a guide to deciding whether death is appropriate. This is apparently what we contemplated in People v. Brown, supra, 40 Cal.3d 512, 542, footnote 19, when we approved a jury instruction to the effect that to return a death verdict the jurors must be persuaded, not that the aggravating factors marginally outweigh the mitigating factors, but that they are “so substantial in comparison with the mitigating circumstances that it warrants death instead of life without parole.” The other way is for the jury to take moral and normative considerations into account during the weighing process, adjusting the weights so that the decision whether aggravating factors outweigh mitigating reflects a normative decision whether death is appropriate. The majority opinion in People v. Hendricks, supra, 44 Cal.3d 635, 654, endorses this method.
When a prosecutor argues, as in this case, that a jury must return a death verdict if aggravating factors outweigh mitigating factors, his words by necessary implication bar the jury from deciding the case by one method we approved in Brown, supra, 40 Cal.3d 512, that of weighing the factors objectively, then determining whether the aggravating factors so substantially outweigh the mitigating as to warrant death. This leaves only the second method of deciding appropriateness—one of adjusting the weights themselves to reflect a moral assessment of the appropriateness of death.4 If we are to affirm a judgment of death under such circumstances, we must satisfy ourselves that, even though many persons (such as the Attorney General arguing this case) would interpret the court’s instructions to preclude a normative assessment during the weighing process, this particular jury in fact understood it had the power and duty to make such an assessment. We probably should insist upon some affirmative indication in the record that the jury understood that it must decide appropriateness during the weighing process; we should certainly not affirm when the record affirmatively shows that the jury was advised to the contrary.
In the present case the prosecutor told the jurors that the weighing process is akin to fact-finding—an objective, non-normative process with which the jurors, having sat through a guilt trial, were already familiar. A fact-finding jury does not make a moral assessment of the facts. If an attorney were to argue to a guilt phase jury that even though the evidence shows a fact to be true, the jury should find to the contrary in order to give *983defendant the appropriate punishment, courts would without hesitation brand the argument as improper. Thus to tell a jury that it is a fact-finding body is to tell them that they should not concern themselves with the penal consequences of their findings—that it is not their function to decide what penalty is morally appropriate. A jury which believes it must determine factually whether aggravating factors outweigh mitigating and that the law then dictates the result is one which does not understand its function.
I would reverse the judgment of death.
Appellant’s petition for a rehearing was denied August 18, 1988. Mosk, J., was of the opinion that the petition should be granted.
The majority also note that the prosecutor reminded the jury that despite defendant’s prison-acquired vocations, defendant committed crimes after his release from prison. Nothing in this statement implies that the jury could consider defendant’s artistic accomplishment as a mitigating factor in itself. And no reasonable juror could think that was what the prosecutor intended, not when that prosecutor had told him, minutes earlier, that defendant’s artistic ability “certainly is not a factor in mitigation.”
The jury would have to go through the following reasoning process: The prosecutor said that factor (k) includes only those matters which lessen the gravity of the crime. But he also said that defendant’s traumatic childhood did not justify a life sentence. The way he said, that implied that we could consider defendant’s childhood, but that it does not weigh enough to call for a life sentence. If we can weigh the events of his childhood, even though they do not weigh enough to tip the balance, we should also be able to weigh the events of his adult life. This means we should be able to consider his past artistic achievements and, by further implication, his present talent and possible future achievements. So when the prosecutor said such defendant’s artistry “ ‘certainly is not a factor in mitigation,’ ” he really did not mean it, because that statement might be inconsistent with the way he analyzed defendant’s childhood.
“In many situations, defense and prosecution arguments that seem to be equivalent will not be equivalent in the eyes of the jury. . . . [Especially in the penalty trial context. . .it [the prosecutor’s argument] is likely to have a greater impact on the jury because the jury is much more likely to be swayed by the prosecutor than the defense counsel. [j[] In any criminal case, jurors will be likely to view the prosecutor in a more favorable light than defense counsel. As Alschuler has said, ‘The assistant district attorney is the representative of an elected, presumably popular public official, and the mere fact that he is a state employee may create a sense of trust and an expectation of fairness that a defense counsel would find difficult to match through the most strenuous exertion of his charm.’ [Quoting Alschuler, Courtroom Misconduct by Prosecutors and Judges (1972) 50 Tex. L. Rev. 629, 632.] Moreover, the fact that the penalty stage jury has already found the defendant guilty of a capital crime places defense counsel at a special disadvantage. Because the jury has accepted the prosecutor’s position at the guilt stage, they are especially likely to view the prosecutor as more credible than defense counsel and to give greater weight to any comments he may make during closing argument.” (White, The Death Penalty in the Eighties (1987) p. 97.)
One problem with the prosecutor’s argument in People v. Myers (1987) 43 Cal.3d 250 [233 Cal.Rptr. 264, 729 P.2d 698], which led us to reverse the penalty judgment in that case, is that the prosecutor erroneously told the jurors that “ ‘It would not be appropriate for you to determine what result you want to obtain and then seek to shade the factors or the weight to give to the various factors.’ ” (P. 275.)