dissenting.
The majority upholds James J. Koedatich’s capital murder conviction despite strong evidence that the jury’s determination was prejudiced by the deluge of pretrial publicity and by the prosecutor’s misconduct. In so doing it not only fails to honor its previous professions that “death is different,” thus requiring searching appellate review, but also frustrates the settled jurisprudence of this Court safeguarding a defendant’s right to a fair trial before an impartial jury. The majority’s review of *343the record is perfunctory and deferential rather than searching; its conclusion that the trial judge’s rulings on publicity and venue should be sustained, and that the prosecutorial misconduct does not constitute reversible error, fail to meet the standards for the review of capital prosecutions exemplified by this Court in State v. Bey (I), 112 N.J. 45 (1988).
The Court upholds the trial court’s rulings on publicity and venue despite the unprecedented local saturation of prejudicial pre-trial publicity and despite the trial court’s application of an erroneous and antiquated standard in denying a change of venue. In the same vein, the majority disregards the stricter standard for adjudging prosecutorial excesses in capital cases, acquiescing in ordinary curative instructions that fail to overcome the effect of outrageous prosecutorial misconduct. The majority fails in the process to appreciate that the prejudicial effect of the publicity and prosecutorial misconduct was exacerbated because of the rather tentative circumstantial evidence of defendant’s guilt. Finally, the Court also breaks with the implications of its precedent in State v. Ramseur, 106 N.J. 123 (1987), by requiring that defendant again be subjected to the death penalty despite indisputable evidence that the jury was deadlocked on the existence of mitigating factors effectively depriving defendant of the opportunity for an “acquittal” of the death penalty through a non-unanimous determination, see State v. Bey (II), 112 N.J. 123 (1988), and by indicating that checks on prosecutorial discretion may not be needed.
I dissent from these determinations. The case undeniably is riddled with significant errors and is fraught with prejudice to the defendant. Because I believe that defendant was denied a fair trial before an impartial jury, I would reverse defendant’s conviction and sentence. Furthermore, because defendant’s death sentence was coerced, I believe that he cannot, consistent with State v. Ramseur, supra, be exposed again to a death sentence on remand.
*344I.
I have argued, concurring in State v. Bey (I), supra, 112 N.J. at 106-20, that an enhanced standard of appellate review is warranted in capital cases, and that this standard should be defined expressly by this Court. In my view, such a standard entails a two-step analysis. First, as a consequence of the direct appeal posture of capital appeals and the eventual requirement of a proportionality review, this Court should conduct an independent and heightened review of the record below to verify the fact-finding made in the trial court. Second, because of the risk of inconsistent results exemplified by federal death-penalty jurisprudence, this searching review of the record below should be complemented by a clearly defined standard of appellate reversibility that eschews normal standards of review such as “harmless error.” I have suggested that where a given error is of constitutional magnitude, it must be shown to have had no effect on the jury’s deliberations; other errors must be shown to have had no realistic likelihood of prejudicing the jury’s deliberations. See State v. Bey (I), supra, 112 N.J. at 116-17 (Handler, J., concurring). A “principal virtue” of enhanced appellate review, I have argued, “is that ... it requires the appellate court to see the case as a whole. Thus, each error ... is critically evaluated ... both for its individual effect on deliberations and for its effect on the structure of the entire case.” Id. at 117. This is nowhere more true than in the context of this case, in which the underlying conviction was based entirely on rather tenuous circumstantial evidence.
An appropriate appellate standard to be used in a case such as this, i.e., a capital case in which the State's proofs are tenuous, was discussed by the Mississippi Supreme Court in Fisher v. State, 481 So.2d 203 (Miss.1985). The Fisher Court was confronted with a record with striking parallels to that of this case in terms of the circumstantial nature of the proofs and the weakness of the State’s case against defendant.
*345In contrast to the Court in this case, the Mississippi Supreme Court began its review of the record by acknowledging the “principle that on appeal convictions of capital murder and sentences of death will be subjected to heightened scrutiny____ ‘What may be harmless error in a case with less at stake becomes reversible error when the penalty is death.’ ” Id. at 211 (quoting Irving v. State, 361 So.2d 1360, 1363 (Miss.1978), cert. denied, 441 U.S. 913, 99 S.Ct. 2014, 60 L.Ed.2d 386 (1979)). Such scrutiny, the court observed, “requires that trial errors be considered not necessarily individually but for their cumulative effect.” Id. at 211. The court then addressed the application of heightened scrutiny in the context of a case based on circumstantial evidence:
While there may be legitimate differences of opinion as to just when and how "heightened scrutiny” works in death penalty cases, it would seem clear that this approach is most needed and most applicable in cases resting upon circumstantial evidence and where the matter of whether the defendant is guilty at all is by no means free of all doubt. As the facts suggest, this is an appropriate case for heightened scrutiny. [Fisher v. State, supra, 481 So. 2d at 211.]
This reasoned approach contrasts sharply with the review undertaken by the majority. Although in New Jersey, we treat circumstantial and direct evidence similarly in terms of the burden of proof on the state, see State v. Fiorello, 36 N.J. 80 (1961), the central concern of the Fisher court was the weakness of the state’s evidence, which underscored the need for enhanced appellate review. Thus, to the extent that an enhanced standard of review creates a distinctive criminal jurisprudence applicable to defendants in capital cases, this does not so much argue against an enhanced standard as justify it. See Comment, “Deadly Mistakes: Harmless Error in Capital Sentencing,” 54 U.Chi.L.Rev. 740, 747-49 (1987).
Moreover, adoption of such a standard of review will not prove to be unworkable in terms of the trial of cases because it governs only appellate courts. I would emphasize, further, that an express enhanced standard of review for capital murder appeals has been applied in other jurisdictions, without engen*346dering procedural chaos or deleterious consequences and without making capital prosecutions impossible. See, e.g., Fisher v. State, 481 No.2d 203 (Miss.1985); Weeks v. State, 456 So.2d 395 (Ala.Cr.App.1983), aff'd 456 So.2d 404 (Ala.1984), cert. denied, 471 U.S. 1030, 105 S.Ct. 2051, 85 L.Ed.2d 324 (1985); Commonwealth v. Zettlemoyer, 500 Pa. 16, 454 A.2d 937 (1982), cert. denied, 463 U.S. 1236, 104 S.Ct. 31, 77 L.Ed.2d 1452 (1983); State v. Wood, 648 P.2d 71 (Utah), cert. denied, 459 U.S. 988, 103 S.Ct. 341, 74 L.Ed.2d 383 (1982).
The majority’s comments concerning the enhanced standard of review are problematic because even under conventional standards of review the guilt conviction here should be reversed. The failure to grant a change of venue, initially because the trial court applied the wrong test for such a motion, is reversible error under the standard established in State v. Williams, 93 N.J. 39 (1983). Further, the prosecutor’s egregious misconduct should be reversible error under the dictates of the majority’s opinion in State v. Ramseur, supra, 106 N.J. at 320-24. Thus, I am strongly of the view that under a conventional standard of review the errors mentioned herein were sufficiently prejudicial to warrant reversal.
Nevertheless, because this case is a capital murder appeal I firmly believe that this Court should apply an enhanced standard of review as part of the needed effort to establish a consistent standard of review for these unique appeals. I commend the standard proffered in State v. Bey (I), supra (Handler, J., concurring). The force of the principle that “[w]hat may be error in a case- with less at stake becomes reversible error when the penalty is death,” Irving v. State, supra, 361 So. 2d at 1363 (citations omitted), is redoubled in a context where not only is a life at stake, but the proofs used to secure the sentence placing that life in jeopardy are highly circumstantial and tenuous.
That urgency attends this case. As can be inferred from the majority’s painstaking account of the facts, the State’s case *347was extremely proof-sensitive. No one saw the defendant kill Amie Hoffman. There were no fingerprints or tire prints to tie the defendant to the scene of the crime. No incidents of the victim — jewelry, purse, wallet — were found on the defendant, in the defendant’s car, or in the defendant’s home. No microscopic trace of the victim was found in the defendant’s car. None of the debris observed and recovered at the scene of the crime, few of the fibers (and only the most common of these), and none of the scrapings taken from the victim’s leg, mini-pad, clothing, and footwear matched anything of the defendant. Pubic and other human and animal hairs found on the victim did not match the defendant. The spermatozoa type was inconclusive, and thus could not be matched with defendant’s. The State’s case was based on contested and inconsistent identifications linking the defendant to the victim, evidence of the circumstances of defendant’s arrest (which evidence had more to do with the circumstances of defendant’s other murder charge, and thus had the capacity to trigger juror recollection of that charge), and on a general correlation between fibers found on the victim and the common carpet and foam fibers of defendant’s car. The State’s experts acknowledged that nothing conclusive could be said based on the fiber comparisons, but argued that the combination of correlations resulted in a “strong association.” The strength of the evidence against defendant was augmented, regrettably, by repeated insinuations by the prosecutor of facts that were not in evidence, of “facts” that the prosecutor knew were not true and, as the trial court acknowledged, by prosecutorial allusion to the serial killings that had terrorized Morris County and, indirectly, to the defendant’s status as the accused in another of the serial killings.
I do not contend that the evidence was insufficient; to the contrary, I believe that a rational, impartial jury, on the factual record below, could have found the defendant guilty. The question on this appeal, however, is whether the jury that deliberated defendant’s fate can in good conscience be con*348sidered to have been impartial, and whether that jury was influenced by factors that corrupted its verdict. I emphasize the tentative nature of the State’s proofs, however, because as the certainty of the proofs diminishes, the likelihood that error will be prejudicial increases correspondingly. This fact should inform our appellate review of this capital conviction.
II.
The murders of Amie Hoffman and Deirdre O’Brien and the subsequent arrest of James Koedatich were attended by massive and intense adverse publicity. The Court itself acknowledges the nature and duration of this publicity. Ante at 265-66. See State v. Williams, supra, 93 N.J. at 50-51 n. 2.
Based on this publicity, defense counsel brought a motion for a change of venue prior to trial. The trial court, in June 1984, denied the motion believing that a change of venue was not required under the standards of State v. Williams, supra. Defense counsel renewed the motion for a change of venue on the first day of jury selection, September 24, 1984. The court denied the motion as “premature.” The publicity continued through the first few days of jury selection. On the third day of jury selection, the defendant again renewed his motion for a change of venue; once more, it was denied.
The trial court attempted to counter the adverse publicity by giving potential jurors cautionary instructions as to exposure to the publicity.1 Significantly, the court excused for cause any prospective juror who indicated knowledge of the defendant’s prior murder conviction or his indictment for the murder of *349Deirdre O’Brien. Of the fifteen jurors who sat during the trial, only five had not read or heard anything about the defendant; the remaining jurors had read something of the case.2
Defendant raises a host of issues deriving from the court’s efforts to manage the effect of this massive pretrial publicity. In particular, defendant alleges that the court’s conduct of the voir dire was inadequate, that the court erred in failing to excuse for cause jurors who had read or heard about the case, and that the court erred in excluding two death-scrupled jurors. However, it is the majority’s treatment of the two critical issues relating to the prejudicial effect of the pretrial publicity — propriety of the court’s denial of the defendant’s motion to change venue, and the effect of the post-trial disclosure by several jurors that they had realized during the trial that defendant had been charged with another murder — that require a response.
A.
The Court’s decision sustaining the lower court’s denial of a change of venue reflects not merely the failure to apply an enhanced standard of review, but also the frustration of the principle underlying our jurisprudence safeguarding jury impartiality: “Preservation of the jury’s independence from extraneous — even judicial — influences.” State v. Corsaro, 107 N.J. 339, 350 (1987). The law respecting motions for a change of venue in capital cases was initially presented in State v. Williams, supra, 93 N.J. 39. It was further expounded in *350State v. Biegenwald, 106 N.J. 13 (1987). See State v. Bey (I), supra, 112 N.J. 45. The Court observed in Biegenwald that under the old test as set forth in State v. Wise, 19 N.J. 59, 73-74 (1955), few defendants succeeded in obtaining a change of venue. “Accordingly, in 1983 in a capital case we modified the defendant’s burden, conferring on trial courts the discretion to change venue where it is ‘necessary to overcome the realistic likelihood of prejudice from pretrial publicity.’ ” 106 N.J. at 33 (quoting State v. Williams, 93 N.J. at 67-68 n. 13). It should be noted at the outset, therefore — and all parties agree — that the trial court erred initially in applying State v. Wise; as this Court held in Biegenwald, the effect of Williams was to “[modify] the defendant’s burden” from clear and convincing proof that an impartial jury cannot be obtained to a “realistic likelihood of prejudice from pretrial publicity.” 106 N.J. at 33 (quoting State v. Williams, supra, 93 N.J. at 67-68 n. 13). The question before us, therefore, is not whether the defendant could show with clear and convincing proof that an impartial jury could not be obtained, but whether, given the saturation of publicity, the defendant could show that there was a “realistic likelihood” of prejudice.
In assessing whether such a “realistic likelihood” exists, the Biegenwald Court held, courts must first distinguish “between cases in which the trial atmosphere is so corrupted by publicity that prejudice may be presumed ... and cases in which pretrial publicity, while extensive, is less intrusive, making the determinative issue the actual effect of the publicity on the impartiality of the jury panel.” Biegenwald, supra, 106 N.J. at 33 (citations omitted). The pretrial publicity in this case meets, I believe, the standard for finding that prejudice should be presumed, and, moreover, that the publicity actually had an effect on the jury’s impartiality thus meeting either of the two tests in Biegenwald.
According to the Court, the publicity surrounding the case was intense, but its prejudicial effect dissipated over time and so prejudice cannot be presumed. The defendant was arrested in January 1983, was not indicted until December of 1983, and *351was not tried until September 24, 1984 — nearly two years after the most intense publicity. Hence, the Court believes that it was confronted with publicity that was not “presumptively prejudicial” and therefore the consequences of the publicity could properly be assessed in terms of its actual effect on juror impartiality. Ante at 273-74. I strongly disagree. Further, the Court concludes that the jury that was impaneled was impartial, thus sustaining the trial court’s decision to deny the motion for venue, a decision based in part on the trial court’s initial application of the wrong standard. I strongly disagree with the majority’s analysis of this pretrial publicity, an analysis that is not, I believe, supported by the record.
1.
As the majority observes, cases in which appellate courts have held that a trial court should have presumed prejudice and granted a change of venue “are relatively rare and arise out of the most extreme circumstances.” Ante at 269. It was, of course, in recognition of that fact that this Court in Williams, and supreme courts in states such as California and Mississippi, reformulated the standard for granting changes of venue in terms of the existence of a realistic likelihood of prejudice. The case law from jurisdictions whose standard comports with the standard modified by Williams is, therefore, irrelevant. See, e.g., State v. Jenkins, 508 So.2d 191, 193 (La.App. 3 Cir.), writ denied, 512 So.2d 438 (La.1987) (“defendant must prove more than the mere knowledge of the public facts surrounding the offense,” and so must show “that there exists such prejudice in the collective mind of the community that a fair trial is impossible”); State v. Lanscak, 404 N.W.2d 192, 193 (Iowa App.1987) (defendant must show a “substantial likelihood that the defendant cannot receive a fair and impartial jury trial in the county”); Slone v. State, 496 N.E.2d 401, 404 (Ind.1986) (to obtain a change of venue, defendant must adduce “evidence of community bias or prejudice sufficient to convince the trial court that he could not obtain a fair trial in that county;” further, “[i]t is insufficient to establish local prejudice warranting a change of *352venue unless there is a demonstration that jurors were unable to deliberate fairly”) (citations omitted); State v. Bobo, 727 S.W.2d 945, 948 (Tenn.), cert. denied, — U.S.-, 108 S.Ct. 204, 98 L.Ed.2d 155 (1987) (issue of change of venue “addresses itself to the sound discretion of the trial judge; his decision will be respected absent an affirmative and clear abuse of that discretion”) (citation omitted).
The relevant case law, as the majority acknowledges, is that from jurisdictions sharing New Jersey’s “realistic likelihood” standard. Ante at 270-71. Thus, the majority quotes with approval Maine v. Superior Court, 68 Cal.2d 375, 383, 66 Cal.Rptr. 724, 729, 438 P.2d 372, 377 (1968) (quoting American Bar Association Project on Minimum Standards for Criminal Justice, Standards Relating to Fair Trial and Free Press, § 3.2(c) (1966)), in which the California Supreme Court states that venue should be changed where “there is a reasonable likelihood that in the absence of such relief, a fair trial cannot be had____ A showing of actual prejudice shall not be required.” (quoted ante at 271).
The California Court explained the standard as follows: “The phrase ‘reasonable likelihood’ denotes a lesser standard of proof than ‘more probable than not.’ ... Further, when the issue is raised before trial, any doubt as to the necessity of removal to another county should be resolved in favor of a venue change.” Martinez v. Superior Court of Placer County, 29 Cal.3d 574, 578, 629 P.2d 502, 503-04, 174 Cal.Rptr. 701, 702-03 (1981) (citations omitted) (emphasis added). In determining that a change of venue should have been granted, the California Court has relied on such factors as (1) “[t]he element of sensationalism always present in reporting of events concerning a capital case,” id. at 581, 629 P.2d at 505, 174 Cal.Rptr. at 704; (2) the fact that “[i]n a small town, in contrast to a large metropolitan area, a major crime is likely to be embedded in the public consciousness with greater effect and for a longer time,” id. at 581-82, 629 P.2d at 506, 174 Cal.Rptr. at 705 (citing Steffen v. Municipal Court, 80 Cal.App.3d 623, 145 *353Cal.Rptr. 782 (1978)) (ordering change in venue on this ground in county with population of 600,000 — almost 200,000 greater than Morris County’s in 1984); and (3) the relative “status of the victim and the accused in the community,” by which the court contrasts descriptions of victims as a “popular teenage couple,” or “popular athlete,” with descriptions of defendants such as “anonymous transients,” or “an alleged heroin addict.” Id. at 584-85, 629 P.2d at 507-08, 174 Cal.Rptr. at 707 (citations omitted).
The majority cites these factors with approval, then fails to apply them. The “sensationalism always present in reporting of events concerning a capital case” was not only present but heightened in this ease by the understandable atmosphere of community terror created by the fact that there were multiple murders and, for months, no arrest; and the additional fact that the suspect who was ultimately arrested had served eleven years in Florida for murder. The publicity of the case was inherently pervasive and sensational. Although Morris County is hardly a “small town,” the atmosphere of terror was closer to what one might expect of a smaller community, and thus establishes the California Court’s second indication that venue should be changed. Moreover, the relative status of the victim and the accused supports a change of venue. Residents of Morris County were saturated with the contrast between Amie Hoffman, a young, beautiful woman, well-liked in school where she earned high grades and was a cheerleader, and the defendant, essentially a stranger to the community who had returned to live with his mother after having served eleven years in Florida for murder. In short, all of the California factors cited with approval by the majority were present here; prejudice should have been presumed.
An approach similar to California’s has been adopted in another state that embraces the “reasonable likelihood” standard, Mississippi. As in New Jersey, Mississippi for years made it extremely difficult to secure a change of venue. Recognizing that its past “discussions of the point have been *354cryptic, superficial,” and that “[i]f an unbiased jury is not impaneled, it does not matter how fair the remainder of the proceedings may be,” Fisher v. State, supra, 481 So.2d at 215-16, the Mississippi Supreme Court has, in a series of decisions beginning with Johnson v. State, 476 So.2d 1195 (Miss.1985), refined its test of when venue should be changed to require that venue be changed “where, under the totality of the circumstances, it appears reasonably likely that, in the absence of such relief, the accused’s right to a fair trial may be lost.” Fisher, supra, 481 So.2d at 220 (citing Maine v. Superior Court, supra, 68 Cal.2d at 383, 438 P.2d at 377, 66 Cal.Rptr. at 729). The Mississippi Court, pursuant to this new standard, has posited certain nonexclusive factors that serve to establish an irrebutable presumption of prejudice, mandating a change of venue. State v. White, 495 So.2d 1346, 1349 (Miss. 1986). This test, to be applied under a heightened standard of review, includes as a factor the community’s “supposition that the appellant was guilty of other crimes.” Id.
The most significant Mississippi case applying these elements is Fisher v. State, supra, 481 So.3d 203, a prosecution that bears great factual similarity to this case. Fisher, like this case, involved multiple rape-murders in which the community was terrorized by both the nature of the crimes and the length of time for which the assailant remained at large. Id. at 206-07. Both cases were founded on circumstantial evidence; indeed, the case against Fisher seems, on balance, to be the stronger of the two, as the police found jewelry of the victim in Fisher’s truck.
The similarity does not end, however, with the circumstances of the crimes; rather, it extends to the nature and extent of the publicity and to the character of the accused. The Mississippi Supreme Court focused on the extensive pretrial publicity in light of that trial court’s denial of a motion for a change of venue, stating, “Our dominant concern is the saturation media coverage — television, radio and newspaper — given the prosecution of Larry Fisher.” 481 So.2d at 217. Every factor cited by *355the court in Fisher as “of great importance” to the venue issue is also present in this case. First, there were multiple crimes by an unknown assailant which generated community awareness and attendant fear that there was a multiple offender on the loose in the area. Second, there was extensive reportage that the defendant was charged not with “a single sex-related murder” but also with another murder. Third, it was reported repeatedly that the defendant, like the defendant in Fisher, had a previous conviction. Finally, similar to the publicity in Fisher, it is undisputed that the details of the evidence against defendant — including facts inadmissible at trial, specifically evidence concerning the O’Brien murder — were reported widely.
The degree of saturation in this case also parallels that of Fisher. Although Morris County’s population — 405,000 in 1984 —is much larger than the Lauderdale County’s, 80,000, the degree of saturation is almost identical. The in-county circulation of the three largest newspapers serving Morris County— approximately 130,000 — is comparable to the newspaper’s circulation in Fisher relative to the size of its county. Morris County is serviced by numerous radio and television stations, as is Lauderdale County. Not surprisingly, therefore, the evidence of the infection of the prospective jurors is very strong in in both cases; while every prospective juror had heard of the Fisher case, approximately 90% of the potential jurors had heard of this case.
The Fisher court expressly discounted, moreover, the jurors’ professions of their own impartiality. The court acknowledged, in fact, that “[a]ll twelve of those seated so proclaimed,” but the Fisher court, although it assumed that those professions and the trial court’s acceptance of them were in good faith, did not rely upon these statements by the jurors. Id. at 220-22. The court insisted that “[t]he saturation pretrial publicity described above, however, suggests that there was and remains substantial doubt that Larry Fisher could then or ever get a fair trial in Lauderdale County.” Id. at 221-22. Prejudice, in other words, should have been presumed. The court concluded:
*356We are convinced that the trial judge scrupulously respected Fisher’s right to a fair trial in all matters save one, one without which the rest are rendered meaningless. The facts recited above are substantially convincing to our minds that no jury could be impaneled in Lauderdale County which could bring to Larry Fisher’s trial that degree of fairness and impartiality which is his right. Certainly on this record there was ... a reasonable likelihood that, without a change of venue to a county outside the coverage of the Meridian news media, Fisher could not be afforded his constitutionally assured fair trial____ [Id. at 223.]
In my view, Fisher is both indistinguishable and compelling authority for reversal in this case. The standards for granting a change of venue are identical. The type of crime involved is identical. The proofs adduced in both cases are tenuous. The backgrounds of the defendants are similar. The type of publicity involved is identical, and the saturation depth is so close to render any distinction unpersuasive.
The majority relies much too heavily on the passage of time to argue that prejudice should not be presumed. Such an argument was persuasively rejected by the Mississippi Court in Fisher, which pointed out that regardless of the interval of media restraint, “the damage had already been done.” Id. at 219. In addition, as the Fisher court noted even skillful voir dire may not be effective in this area because: “Many whose views may be substantially affected by pretrial publicity may not know that they are incapable of sitting as fair and impartial jurors.” Id. at 221 (citations omitted). Moreover, given the sensational nature of this coverage and the fairly brief interval between the intense publicity and trial, the majority’s finding that this interval was sufficient is not credible. Finally, the evidence that most members of the actual jury were aware of the defendant through the media coverage and the subsequent indication that jurors were aware of the highly damaging and non-admissible evidence concerning the O’Brien murder at the time they were deliberating further undercuts the majority’s conclusion that prejudice should not be presumed.
To hold as the majority does today is, in short, to hold that Fisher was wrongly decided. This, in my opinion, is not a *357tenable position. Writing recently of the salutary changes that have resulted since Mississippi’s adoption of the “realistic likelihood” standard, Justice Robertson noted that since Johnson and Fisher adopted the realistic likelihood test, “there have been at least three capital murder trials in changed venues which resulted in not guilty verdicts____ In addition,” he added, “we all know what happened to Fisher ... on remand. In Lauderdale County, Fisher had been convicted of capital murder and sentenced to death. On retrial in Rankin County, Fisher was found not guilty.” Lutes v. State, 517 So.2d 541, 550-51 (Miss.1987) (Robertson, J., dissenting).
Under the standards cited approvingly by the majority, the level of pretrial publicity was such that the likelihood of prejudice in this case was “realistic,” to say the least. Prejudice should have been presumed. Venue should have been changed.
2.
Having upheld the trial court’s refusal, a decision made initially under an erroneous standard, to presume prejudice, the Court then turns to the question of whether there is a showing of “actual effect” of prejudice from the publicity. Relying on assumed parallels with State v. Biegenwald, supra, 106 N.J. 13, and on the strength of the voir dire, and ignoring evidence to the contrary, the Court determines that there was no such prejudicial effect.3 Again, I disagree.
First, Biegenwald is readily distinguishable from this case. In this case, only five of the fifteen jurors asserted no prior *358knowledge of the case; two-thirds of the impanelled jury indicated that they had encountered publicity regarding the case. It simply cannot be said in this case, as it could in Biegenwald,4 that “a significant portion of the jury array was relatively unexposed to pretrial publicity” or that the impanelled jurors “indicated that they had encountered little or no publicity regarding the case.” Biegenwald, supra, 106 N.J. at 36-37. As in Biegenwald, some members of the jury did recollect knowledge of prejudicial information; unlike in Biegenwald, such jurors never satisfied any court that they could remain impartial despite that knowledge. Indeed, the trial court here excused for cause all jurors with such knowledge. This case, in other words, is sufficiently unlike Biegenwald to justify a presumption of a realistic likelihood of prejudice.
Even if the Court is correct to require a showing of “actual effect,” that burden is more than met under the circumstances. As this Court stated in State v. Van Duyne, 43 N.J. 369, 386 (1964) , cert. denied, 380 U.S. 987, 85 S.Ct. 1359, 14 L.Ed.2d 279 (1965) , “an appellate tribunal ... should determine for itself whether the pretrial newspaper stories are so persuasive and so prejudicial, or the jury’s protestation of unaffected impartiality after reading them so unconvincing or doubtful that a new trial should be ordered.” See also State v. Biegenwald, supra, 106 N.J. at 36 (purporting to follow Van Duyne). All fifteen jurors, it is true, affirmed as a matter of course that they could decide the case solely on the basis of the evidence presented at trial. The Court, in effect, acquiesces in the trial court’s acceptance of such an affirmation as sufficient to overcome any *359indication of prejudice, relying on “the strength of the voir dire in upholding the ... denial of the change of venue motion.” Ante at 281. The Court makes much of the laudable prophylactic measures undertaken by the trial court to assure that the jury remained insulated from the publicity. Id. I agree that the measures undertaken were both necessary and proper; had an impartial jury been obtainable from that community, the trial court’s management of the case would have ensured one. The majority ignores completely, however, the fact that the trial court’s measures simply did not work. They could not work because the saturation of publicity ran too deep. See Fisher v. State, supra, 481 So.2d at 220-23.
The inefficacy of the court’s efforts is clearly indicated by evidence that several jurors recalled in the course of the trial that the defendant was implicated in the O’Brien murder.5 It is a peculiar irony of the court’s refusal to recognize this “actual effect” that those jurors with knowledge of the O’Brien murder charge — “most of the jurors,” according to Juror Fascia— would have been excluded as presumptively prejudiced if they *360had recalled such knowledge on voir dire, notwithstanding any affirmations of impartiality they might have been willing to make. This Court’s disregard of evidence of “actual effect” in assessing the “realistic likelihood of prejudice” is difficult to understand. Its reliance, moreover, on the “strength” of a voir dire that obviously failed even to identify those jurors who were aware of highly prejudicial information — let alone to assess their ability to be impartial — is frankly inexplicable. The Court, instead of independently assessing the gravity of error (see, e.g., State v. Van Duyne, supra), purports to defer to the trial court; then it disregards the trial court’s own estimation that this knowledge would presumptively disqualify a venireperson.
The Court’s result is also difficult to square with its decision in State v. Bey (I), supra, 112 N.J 45. In Bey (I), this Court reversed defendant’s murder conviction because the trial court failed to ensure that the jury was not exposed to mid-trial publicity respecting defendant’s other alleged offenses, including another murder. The Court held that the test for such publicity was, “[i]f the court is satisfied that the published information has the capacity to prejudice the defendant, it should determine if there is a realistic possibility that such information may have reached one or more of the jurors.” Id. at 86. If there is such a realistic possibility, a voir dire must be conducted to ascertain whether any juror has been exposed. Id. at 86-87. The Court then held that the midtrial publicity had a “[great] capacity to prejudice a defendant’s case” (id. at 90), and that where the publicity had such a strong potential for prejudice, the usual assumptions about jurors following their oaths and adhering to the judge’s instructions are not warranted. Id. at 81-83. The evil sought to be redressed in Bey (I), moreover — that jurors might proceed to deliberations having been exposed to prejudicial information, and without assuring the court, in light of that exposure, of their capacity to deliberate impartially — undeniably occurred in this case. The *361Court’s determination that there is no remedy for this is thus inconsistent with its holding in Bey (I).
We have ruled — in non-capital cases — that “where pretrial publicity is not highly prejudicial or inflammatory in nature, a trial court should accept the word of a juror who says ... she can set aside previous knowledge of the facts of the case — ” See State v. Singletary, 80 N.J. 55, 64 (1979). This is, however, a capital case. Moreover, unlike Singletary, the publicity in the instant case was highly prejudicial and inflammatory. Where, as here, the pretrial publicity has been both extensive and laden with prejudicial information, the normal presumption that jurors who assert their impartiality should be believed should not apply; in such a situation, a juror who has been exposed to the pretrial publicity should be presumptively disqualified. See State v. Williams, supra, 93 N.J. at 68-69; Fisher v. State, supra. It follows that the court should not have accepted “the word” of jurors that they could disregard that publicity; the publicity was too great.
When the Court does discuss the jurors’ knowledge of prejudicial information, it contradicts itself. First, the Court holds that post-verdict investigation of the jury would be improper in this case because there is no allegation of injection of improper facts into the deliberations or a manifestation of racial or religious bigotry. Ante at 288-90. The Court then turns to the article and equivocates, concluding that “[e]ven assuming ... the quoted juror statements appearing in the article to be true ... there still is no basis for concluding that ... Koedatich’s involvement in the O’Brien murder had entered the jury deliberations.” Ante at 290-91. At a minimum, the fact that several jurors entered deliberations with knowledge that would have justified excusing them for cause indicates that this knowledge was a factor in the deliberations of some jurors.
The Court is willing to conclude that the jurors’ post-verdict comments do not indicate an erosion of impartiality. It accepts at face value that Ms. Zaccaro indicated that the other indict*362ment was not discussed until after the verdict, Ms. Herzig insisted that the deliberations had been impartial, and Ms. Fascia stated that the jurors had actually been surprised to learn of the defendant’s prior murder conviction. Ante at 287-91. The Court assumes that so long as the jurors did not discuss during deliberations the prejudicial information, their independent knowledge of that information did not affect deliberations. I reject this assumption; who can believe that a juror’s knowledge of such information will not color his or her contributions to the deliberations? Furthermore, in a capital case, the ultimate verdict of life or death rests with each individual juror’s evaluation of aggravating versus mitigating factors. State v. Bey (II), 112 N.J. 123 (1988). The majority’s assumption allows the weighing process of the individual juror with knowledge of prejudicial information to be corrupted by that prejudicial knowledge. In my judgment, the undeniable fact that some jurors deliberated with knowledge of defendant’s other murder charge — particularly given the tenuous nature of the proofs — is more than sufficient to taint the jury’s verdict.
In State v. Kociolek, 20 N.J. 92 (1955), this Court reversed a death sentence because some of the jurors had learned, during the trial, that the defendant had been indicted for robbery and assault with intent to kill, and had informed the other jurors of the charges during deliberations. Justice Brennan wrote:
It suffices that the intrusion of the Glenn incident into the jurors’ deliberations, standing alone, introduced illegal and extraneous evidence fraught with peril for the defendant, an action the more grievous because taken in disregard, actually in defiance, of the explicit instructions of the trial judge. [Id. at 103.]
The facts of this case are as egregious as the facts of Kociolek. As argued by defendant: “[h]ere ... the press linked the two murders, along with a third murder with which the defendant was never charged, and conveyed the theory that one person was responsible for all three.”6 Ms. Zaccaro’s *363statement, moreover, that she “didn’t know that he did it in the beginning but halfway through the trial I realized he was linked to O’Brien,” can be read to mean that the connection with O’Brien was the decisive factor in her determination that “he did it; ” the impermissible information may have significantly influenced her deliberations. In any event, this is precisely the context in which deference to juror protestations of impartiality and presumptions that jurors follow instructions should no longer obtain. See State v. Bey (I), supra, 112 N.J. at 81-83. In line with our cases safeguarding the integrity of jury deliberations, and with the enhanced standard of review applicable in capital cases, the Court’s inquiry should be whether the State can show that the awareness on the part of several jurors during trial of the flagrantly detrimental information that defendant had been charged with another murder had no realistic likelihood of affecting prejudicially their determination of his guilt. Given the entirely circumstantial and relatively tenuous nature of the proofs, I believe that the answer is clear: there was a “realistic likelihood” of prejudice.
B.
The Court’s determinations, in short, turn the enhanced standard of review applicable in capital cases not only upside down, but inside out. To the extent that the Court’s review is influenced by the fact that this is a capital case, that influence seems to have lessened the scrutiny of the record below; to the *364extent that the Court’s holdings today apply to criminal defendants generally, those holdings have corrupted the principle underlying our jurisprudence safeguarding the impartiality of juries: “[preservation of the jury’s independence from extraneous — even judicial — influences.” State v. Corsaro, supra, 107 N.J. at 350 (and cases cited). See also In re Kozlov, supra, 79 N.J. at 239 (courts must “seek out and expose outside factors impinging upon the jury’s freedom of action and thus its impartiality and essential integrity”); State v. Van Duyne, supra, 43 N.J. 369.
As a result, this trial was allowed to occur in a venue that had been terrorized by a series of brutal murders, and saturated by publicity concerning those crimes. This publicity linked the defendant to other crimes and related the circumstances of his prior conviction for murder. The passage of time failed to diminish the extent to which the jury panel was exposed, and the trial court’s conduct of the voir dire failed to discover the true extent of that exposure. Several jurors entered deliberations in this case, in which the evidence was entirely circumstantial and tenuous, with knowledge that the defendant had been accused of another murder, similar in method, close in time and near in location to the charge before them. The Court engages in a hypertechnical shell game with these facts; it first rules out the presumption of prejudice from the saturation publicity, it then discounts evidence concerning actual prejudice resulting from the publicity, and, finally, it obviates any further inquiry into jury taint by eliminating both presumed and actual prejudice. This legerdemain cannot disguise the Court’s refusal to do what fairness in capital cases demands: to see the case as a whole.7 See State v. Bey (I), supra 112 N.J. at 64 *365(Handler, J., concurring). Legal categories and niceties aside, it is difficult, if not impossible to believe that defendant was not prejudiced, and was tried by an impartial jury, much less that the evidence does not support finding a “realistic likelihood of prejudice.” I can only wonder what it will take in future cases, in light of this Court’s holding, to hold that the likelihood of prejudice was “realistic.”
III.
The erroneous refusal to change the trial venue, resulting in a jury that had known about the facts of the case, compounded the impact of the comments and actions by the prosecutor that are claimed to constitute reversible error. These claimed errors are directly attributable to the calculated misconduct of the prosecutor, misconduct that was recognized by the trial judge who tried unsuccessfully to cure these repeated instances of outrageous behavior. Because these curative instructions could not be sufficient, in light of the circumstances, I dissent from the majority’s treatment of these trial incidents.
Prosecutorial misconduct occurred during the cross-examination of defendant’s step-brother, David Paul Baldwin, who had been in high school, and was living at home in November 1982. On direct he had testified that defendant’s car was not running well in November 1982, that defendant was at home playing Atari games at least until 9:00 p.m. on the night in question, *366that defendant had a beard at that time, and that the sneakers seized from the house were a Christmas 1982 gift and so could not have been worn by defendant in November 1982. The cross examination was replete with questions for which there was no factual basis and which were clearly beyond the scope of the direct examination.
The prosecutor attempted to suggest, from the fact that the step-brother played video games at the Mall, that the defendant had been at the Mall the afternoon of the murder playing video games; this suggestion was made allegedly “with full knowledge that that was completely untrue.” The prosecutor then, while pointing to a map showing the proximity of the retention tanks to Mendham, asked Baldwin if the defendant ever worked in Mendham; he asked this despite the knowledge that the defendant had not worked in Mendham until after the date of the murder. The prosecutor also brought out, over an objection, that defendant had lived in Florida.
There can be no doubt that the prosecutor’s misconduct was egregious. The trial court itself acknowledged that these areas of inquiry were flagrantly improper; with respect to the suggestion that defendant may have been in the Mall on the afternoon of the murder, the court stated: “I think it’s a dreadful avenue, that’s another day that he’s in the Mall with no basis in fact____” Nevertheless, the court denied defense counsel’s motion for a mistrial and, instead, instructed the jury to disregard the questions relating to the defendant’s employment in Mendham, and to the defendant’s possible presence in the Morris County Mall on November 23rd.8
*367The Court now rules that the curative instructions were sufficient to render any error harmless. In State v. Ramseur, supra, 106 N.J. at 320-24, this Court held that a prosecutor’s cross-examination of a witness, during which he made “several statements that can be construed as providing his personal opinion about defendant’s guilt,” was improper. The Court reasoned that “[statements such as those ... are improper because they divert the jurors’ attention from the facts of the case before them.” Id. at 322. Similarly, in this case the prosecutor’s inquiries regarding “facts” that he knew had no basis, from which inferences could be drawn that would be wholly untrue, diverted the jurors’ attention from the relevant facts of the case in a manner that was necessarily prejudicial.
The Court in Ramseur found that the defendant was not deprived of a fair trial because curative instructions were given following prompt objections. It concluded, however, with the following admonition:
*368A prosecutor willing to engage in proscribed conduct to obtain a conviction in a capital case betrays his oath____ Because death is a uniquely harsh sanction, the Court of necessity will more readily find prejudice resulting from prosecutorial misconduct in a capital case than in other criminal matters; prosecutors who fail to take seriously their particularly stringent ethical obligations in capital cases thus strongly risk postponing, and even jeopardizing, the enforcement of the law. [Id. at 324.]
In my opinion, it is not possible to conclude that defendant’s right to a fair trial was preserved. Ramseur does not address whether, given the uniqueness of a capital proceeding, a prosecutor’s misconduct may be so outrageous, that this Court should “more readily find prejudice” despite the use of curative procedures. However, the clear implication, I believe, of the Court’s admonition in Ramseur is its recognition that a prosecutor is obliged to conduct the trial in a fair and proper manner. The Court, while nominally adhering to the expressions of Ramseur, fails to apply this more stringent standard of review.
Here, the prosecutor insinuated, in his cross-examination, that the defendant was familiar with the area of the murder because he worked in Mendham; this the prosecutor knew to be misleading. He also suggested that the defendant may have been at the Mall playing video games on the afternoon of the murder, encouraging the jury to consider a possibility with no basis in fact. These questions were not only outside the scope of direct examination, but also beyond the factual scope of the entire case.
Moreover, the prejudicial effect of this misconduct was exacerbated by misconduct during the prosecutor’s summation at the guilt phase of the trial. During his summation at the end of the guilt phase, in discussing defendant’s arrest, the prosecutor stated: “If he had walked out of there, he would have been given a license to kill.” The trial court issued a curative instruction, telling the jurors that “such phrases as ‘a license to kill’ have no part of the language which is appropriate for your consideration in this case____ lam instructing you at this time to disregard the last comment during the last summation.” Defense counsel moved for a mistrial, on the grounds that the *369prosecutor was insinuating that the defendant was a serial killer, responsible for other killings in Morris County. The trial court agreed with that interpretation of the prosecutor’s remarks: “It’s the prosecutorial equivalent of soiling the floor of the courtroom____ It was very close to a suggestion, but yet not a suggestion that you were going to let a serial murderer go ... and I frankly, I think that’s probably what it was intended to do____” The court, however, denied the motion: “I’m satisfied that, because I took a recess and gave a strong cautionary instruction ... any damage done by that statement was ameliorated____”
I firmly believe that, under this Court’s standard of review for prosecutorial misconduct in a capital case (id. at 324), the prosecutor’s highly improper questions during cross-examination, in which he suggested the existence of incriminating facts unsupported by evidence, and his references to other murders, insinuating that defendant had added culpability, require reversal. Under an appropriate enhanced standard of review there was a realistic likelihood, particularly given the weakness of the prosecution’s case, that these errors had a prejudicial effect on the jury’s deliberations; indeed, the more tenuous are the proofs, the greater are both the incentive for a prosecutor to resort to improper tactics to gain a conviction and the likelihood of a prejudicial effect resulting from those tactics. However, even if the majority were to apply conventional harmless error analysis, this egregious conduct should be reversible error given the weakness of the State’s case and the implications of its holding in Ramseur. Finally, the evidence before us indicates that something that occurred during the trial triggered the jurors’ recollection of prejudicial information; the State’s intentional allusion to that information, in the context of an entirely circumstantial case, may well have been decisive in prejudicing the jury.
This Court should follow its opinion in Ramseur, “more readily find prejudice,” and reverse the conviction.
*370IV.
The Court reverses defendant’s death sentence. I agree with this result for the reasons relied on by the Court. I think it important to emphasize, however, that the type of prosecutorial misconduct that occurred during the sentencing phase would itself warrant a reversal.
This misconduct occurred during the penalty phase summation, when the prosecutor made repeated references to the victim.9 These remarks constituted misconduct by infecting the *371deliberations with inappropriate considerations regarding the victim and with the prosecutor’s personal beliefs about the case.
In Booth v. Maryland, — U.S. -, 107 S.Ct. 2529, 96 L.Ed.2d 440 (1987), the Supreme Court disapproved, by a 5-to-4 vote, the introduction of a Victim Impact Statement (VIS) during the penalty phase of a capital case. The Victim Impact Statement describes “the effect of the crime on the victim and his family.” The Statement in question described “the personal characteristics of the victims” (they were a “close couple,” “married for fifty-three years,” “loving parents and grandparents”). Id. at-, n. 3, 107 S.Ct. at 2531 n. 3, 96 L.Ed.2d at 446 n. 3. The Court held that “[w]hen carrying out [its penalty-phase] task the jury is required to focus on the defendant as a ‘uniquely individual human bein[g].’ ... The focus of a VIS, however, is not on the defendant, but on the character and reputation of the victim and the effect on his family. These factors may be wholly unrelated to the blameworthiness of a particular defendant.” Id. at-, 107 S.Ct. at 2533-34, 96 L.Ed.2d at 449. The Court also noted:
Nor is there any justification for permitting such a decision to turn on the perception that the victim was a sterling member of the community rather than someone of questionable character. This type of information does not provide a “principled way to distinguish [cases] in which the death penalty was imposed, from the many cases in which it was not.” Godfrey v. Georgia, 446 U.S. 420, 433, 64 L.Ed.2d 398, 100 S.Ct. 1759 [1767] (1980) (opinion of Stewart, J.). See also Skipper v. South Carolina, 476 U.S. 1,-, 90 L.Ed.2d 1, 106 S.Ct. 1669 [-] (1986) (Powell, J. concurring in judgment).
We also note that it would be difficult — if not impossible — to provide a fair opportunity to rebut such evidence without shifting the focus of sentencing hearing away from the defendants____ The prospect of a “mini-trial” on the victim’s character is more than simply unappealing; it could well distract the sentencing jury from its constitutionally required task — determining whether *372the death penalty is appropriate in light of the background and record of the accused and the particular circumstances of the crime. We thus reject the contention that the presence or absence of emotional distress of the victim’s family, or the victim’s personal characteristics, are proper sentencing considerations in a capital case. [Id. at -, 107 S. Ct. at 2534-35, 96 L.Ed.2d at 450-51.]
By telling the jury that “this part of the trial ... is when we think of the victim ...” and that “[tjhis is how I want to know Amie Hoffman, a cheerleader, someone who had a life, who had a future,” the prosecutor improperly shifted the focus of deliberations from the “background and record of the accused and the particular circumstances of the crime” to the character of the victim. In my opinion, the thrust of the prosecutor’s argument is the same as the thrust of the YIS in Booth. Moreover, as a matter of state constitutional law, we should apply an enhanced standard of review and conclude that there was a realistic likelihood that the prosecutor’s remarks did affect the jury’s deliberations.
Defendant also claims that by telling the jury “if this isn’t the right [circumstance in which to vote for death], I don't know if there is one,” the prosecutor committed misconduct by injecting his personal beliefs into the deliberations. The jury was induced to rely on the prosecutor’s expertise or to “believe the prosecutor was either exaggerating or fabricating. We do not think any juror can be expected to believe the prosecutor would do so and thus we cannot say the defendant was not prejudiced by the prosecutor’s comments as to his private belief.” State v. Farrell, 61 N.J. 99, 106 (1982); Tucker v. Zant, 724 F.2d 882, 889 (11th Cir.1984).
Under the enhanced standard of review, the Court must conclude that this form of prosecutorial misconduct taints the sentencing determinations. It must in all fairness lead to the reversal of the death penalty.
V.
Defendant argues that the death penalty is constitutionally infirm because the wide latitude given to prosecutors’ decisions *373as to whether to charge a defendant with capital murder has resulted in an unfettered discretion that leads to an arbitrary and capricious imposition of the death penalty. The majority proposes to follow this Court’s recent decision in State v. Ramseur, supra, in which the Court had side-stepped the issue, finding that there was not enough evidence presented and that such an analysis would be more properly made as part of proportionality review. 106 N.J. at 324-31. I dissented from this Court’s treatment of prosecutorial discretion in Ramseur and continue to do so. Id. at 404-08. (Handler, J., dissenting). In addition, I dissent strongly from the majority’s intimations in this case that only certain discrepancies are of constitutional import. Ante at 255-56.
The lodestar principle identified by modern federal death penalty jurisprudence beginning with Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed. 2d 346 (1972), has been the prevention of the standardless imposition of the death penalty. The Supreme Court’s apparent unwillingness to address the problem of prosecutorial discretion, that may lead to the arbitrary imposition of the death penalty, is undermined by the Court’s acknowledgement of the need for guided sentencing discretion. I believe strongly that this Court should not and cannot legitimately follow the path of analysis taken by the Supreme Court and that some guidance of prosecutorial discretion should be mandated.
The United States Supreme Court, in its post-Furman v. Georgia death-penalty jurisprudence, has endorsed systems of guided sentencing discretion. As Justices Stewart, Powell, and Stevens stated in Gregg v. Georgia, 428 U.S. 153, 199, 96 S.Ct. 2909, 2937, 49 L.Ed.2d 859, 889 (1976) (Stewart, J., concurring):
in order to minimize the risk that the death penalty would be imposed on a capriciously selected group of offenders, the decision to impose it had to be guided by standards so that the sentencing authority would focus on the particularized circumstances of the crime and the defendant.
However, in Gregg v. Georgia, supra, the Court rejected the claim that the possibility of discretion pervasive in the system, *374from the original decision to charge through executive commutation, rendered the statute’s operation inevitably arbitrary. The Court reasoned that “[t]he existence of these discretionary stages is not determinative____ At each of these stages an actor in the criminal justice system makes a decision which may remove a defendant from consideration as a candidate for the death penalty____ Nothing in any of our cases suggests that the decision to afford an individual defendant mercy violates the Constitution.” Id. at 199, 96 S.Ct. at 2937, 49 L.Ed.2d at 889.10
In McCleskey v. Kemp, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed. 2d 262 (1987), the Supreme Court again rejected the contention that unguided prosecutorial discretion rendered a death-penalty statute unconstitutional. In McCleskey, the Court was presented with extensive data showing that the likelihood of receiving a death sentence in Georgia was strongly influenced by the race of the victim. Noting that the Georgia Supreme Court had found McCleskey’s sentence proportionate to other death sentences, and that proportionality review is not constitutionally required where statutory guidance is adequate (id. at -, 107 S.Ct. at 1774, 95 L.Ed.2d at 288), the Court rejected McClesky’s contention that his sentence was excessive. Moreover, the Court opined that prosecutorial discretion was a necessary and entrenched part of the judicial system providing leniency and that it should not be condemned concluding: “[A]b-sent a showing that the Georgia capital punishment system operates in an arbitrary and capricious manner, McCleskey cannot prove a constitutional violation by demonstrating that other defendants who may be similarly situated did not receive the death penalty.” Id. The Court dismissed the statistical *375evidence of race-induced discrepancies in sentencing as “an inevitable part of our criminal justice system;” such disparities, when “unexplained,” are not to be assumed to be “invidious.” Id. at-, 107 S.Ct. at 1777, 95 L.Ed.2d at 291-92.
Justices Stevens, Blackmun, and Marshall endorsed Justice Brennan’s dissenting view that Supreme Court precedents “reflect a realization of the myriad of opportunities for racial considerations to influence criminal proceedings____ The discretion afforded prosecutors and jurors ... creates such opportunities. No guidelines govern prosecutorial decisions to seek the death penalty____” Id. at -, 107 S.Ct. at 1788, 95 L.Ed.2d at 305 (Brennan, J., dissenting). Justice Blackmun, in particular, reviewed the record evidence regarding prosecutorial discretion and stressed that some standards were necessary, concluding that the defendant was entitled to relief under the Equal Protection Clause of the fourteenth amendment. Id. at -, 107 S.Ct. 1805, 95 L.Ed.2d at 325 (Blackmun, J., dissenting).
The course of federal jurisprudence should not distract state courts from an independent evaluation of the issue, an evaluation that the majority recognizes is required. Ante at 252-53. In District Attorney v. Watson, 381 Mass. 648, 666-68, 411 N.E.2d 1274, 1284-85 (1980), the availability of unguided prosecutorial discretion led the Supreme Judicial Court of Massachusetts to declare the Massachusetts capital statute unconstitutional under the Massachusetts constitution:
It may be that c. 488 would meet the Federal constitutional requirements, if tested, but here we appraise the statute under ... the State Constitution. We accept the Furman principle that untrammeled discretion in imposing the death penalty is intolerable; we find unacceptable under the State Constitution the premise of the post-Furman cases that statutory guidelines and standards may be entirely curative____ Furman and subsequent cases do not address the discretionary powers exercised at other points in the criminal justice process____ Furman stands indifferent to the exercise of the prosecutor’s "untrammeled discretion.” For reasons which may be valid in the context of his duties, but which do not assist evenhandedness, the prosecutor ... may forgo a first degree murder indictment and seek an indictment for ... a lesser charge. Also, in a first degree murder case, perhaps pursuant to plea bargaining, the *376prosecutor may in his uncurbed discretion nol prosse that part of the indictment which charges murder in the first degree. [Id. at 666-68, 411 N.E.2d at 1284-85.] 11
Most courts, however, have followed the approach of the Illinois court in People v. Free, 112 Ill.2d 154, 162-63, 97 Ill.Dec. 396, 399, 492 N.E.2d 1269, 1272, cert. denied, 479 U.S. 871, 107 S.Ct. 246, 93 L.Ed.2d 170 (1986) (and cases cited), in which the validity of the prosecutor’s exercise of discretion depends on whether a factual basis exists for the charging of aggravating factors. See also Resnover v. State, 460 N.E.2d 922, 929 (Ind.), cert. denied, 469 U.S. 873, 105 S.Ct. 231, 83 L.Ed.2d 160 (1984); State v. Judge, 100 Wash.2d 706, 713, 675 P.2d 219, 223 (1984); State v. Harris, 48 Ohio St.2d 351, 358-59, 359 N.E.2d 67, 71 (1976), judgment vacated on other grounds, 438 U.S. 911, 98 S.Ct. 3148, 57 L.Ed.2d 1155 (1978).
The Illinois court’s reasoning is insufficient for this State’s statute. As I intimated in dissent in State v. Ramseur, the breadth of the definition of capital murder, and the absence of meaningful narrowing through the consideration of aggravating factors, renders the jury’s discretion standardless. The absence of standards for prosecutors in such a scheme redoubles the infirmity of unguided discretion because the flaws that plague the jury’s function in the statute — the failure of the guilt-phase definition and the penalty-phase aggravating factors to guide discretion — are replicated in the absence of prosecutorial standards. 106 N.J. at 405-06 (Handler, J., dissenting). Moreover, the lack of guidance with respect to the prosecutor’s decision to charge a defendant with capital murder unacceptably increases the danger that the death penalty will be imposed arbitrarily because the needed narrowing function is not provided at this crucial initial stage of a prosecution.
Not surprisingly, given the latitude for decision-making allowed by the statute, arbitrary results are emerging. The *377Public Defender has compiled an instructive study that indicates that unguided prosecutorial discretion leads to or has contributed to disparity and disproportionate imposition of the death penalty. L. Bienen, N. Weiner, D. Denno, P. Allison, D. Mills, “The Reimposition of Capital Punishment in New Jersey: Interim Report, Parts I and II” (1988). The Court acknowledges the relevance of this study but then fails to give any credit to its conclusions or implications. Ante at 253-58. I do not insist that the Public Defender’s study must be accepted in all its aspects and would acknowledge that it is flawed in some respects, chiefly its use of defense counsel to assess the strength of the cases studied. However, the study lends credence to the Public Defender’s claims that equally serious crimes are not prosecuted to equal degrees by the varying counties, indicating that differing county policies and budgets, not to mention intense political pressure in given cases, may well contribute to these disparities. It may be that the data are not strongly indicative of different treatment related to the races of the victim and/or defendant but the statistics tentatively show that racial influence may be present. The question for this Court, I believe, is whether these disparate results are indicative of the arbitrary consequences of unfettered prosecutorial discretion. Minimally, it must give the Court pause; realistically, it calls for further study and for the imposition of a mandatory system to guide prosecutors’ judgments as to whether to charge on capital murder.
The majority dismisses the importance of the discrepancies suggested by the Public Defender’s study arguing that the data are not indicative of any racial bias in charging capital murder. However, the majority leaves open this question stating that it would find racial bias to be intolerably arbitrary. I am troubled by the majority’s insistence that the Study must prove bias because I believe that the Study’s preliminary evidence is sufficiently strong to warrant a showing by the State that no bias in charging exists.
*378I am at a loss, however, to explain how the majority can possibly conclude that county by county discrepancies are not indicative of arbitrary imposition of the death penalty. Ante at 255-57. The majority recognizes the evidence supporting the defendant’s claim of such a disparity but then conclusively dismisses this evidence, stating:
That there are differences among the counties in the likelihood that a prosecutor will pursue — or the coincidence that the prosecutor has more often pursued — a capital prosecution does not, standing alone, demonstrate that the death penalty is being arbitrarily imposed. Surely, there are a myriad of reasons why a prosecutor handles different cases differently, such as the willingness of a defendant to plead guilty, the strength of the State’s case, a defendant’s cooperation in the State’s case against a co-defendant, the relative weight of the statutory aggravating and mitigating factors, the availability and relative credibility and persuasiveness of witnesses, and the resources of the county prosecutor’s office, to list only a few. [Ibid. ]
The majority’s holding that county by county discrepancies are of no constitutional significance stands in contradiction to Ramseur in which the Court held that the universe for assuring proportionality was statewide, stating:
We believe that statewide uniformity is the more appropriate measure, and therefore anticipate that comparisons will be made to “similar” cases throughout the State. [106 N.J. at 329.]
The standard of statewide uniformity is now being emptied of content by the majority’s determination that county to county discrepancies do not indicate the possible arbitrary and capricious imposition of the death penalty. I do not see how this can be squared with Ramseur.
The county by county discrepancies, contrary to the majority’s view, are indicative that defendants in different areas of the state are likely being treated differently. To avoid this potential for the arbitrary imposition of the death penalty, the majority should require the development of a statewide standard for deciding whether to charge capital murder. The majority’s suggestion that it strongly recommends that guidelines be developed by the State Attorney General, county prosecutors, and the Public Defender does not, in light of its reluctance to find any constitutional difficulties in the data present*379ed by the Study, suffice to ensure that such standards will be promptly developed. Such standards will lessen the likelihood of arbitrary death penalty prosecutions, and, as the majority notes, promote uniformity and assist in the Court’s eventual proportionality review. Ante at 258-59. I would require the development of these standards.
Because the majority fails to require such a limit on prosecutorial discretion and because the death penalty statute itself does not adequately narrow the class of death-eligible persons, I continue to hold the view that the capital murder statute is constitutionally invalid because it does not overcome the genuine risk of arbitrary and capricious application. 106 N.J at 405-06. The indications suggested by the data are that such arbitrary results may be occurring and accordingly I strongly disagree with the majority’s reluctance to face the implications of this data and to adequately guide the discretion given to county prosecutors.
VI.
A final issue remains, an issue neither raised nor addressed by either party, but one that warrants this Court’s attention under an enhanced standard of appellate review. See ante at 344-45. This Court has decided that if any one juror believes that a mitigating factor exists, that juror is entitled to carry his belief through to the weighing process. See State v. Bey (II), supra, 112 N.J. 123. The reasoning behind this rule is that if a juror believes that a mitigating factor exists, he might well determine that the aggravating factor(s) do not outweigh the mitigating factor; since a non-unanimous verdict is a verdict (of 30 years to life) in a penalty-phase proceeding, any individual juror who decides that a mitigating factor exists and is not outweighed beyond a reasonable doubt by the aggravating factors has, in effect, spared a defendant’s life. On the other hand, if that individual juror believes that the mitigating factor exists and would find, if he or she *380weighed them, that the mitigating factor was not outweighed, but is unable to perform the weighing process because unanimity is required to find mitigating factors, that juror is disenfranchised.
It is clear in this case that the trial court erred in requiring unanimity in order to find that a mitigating factor exists. The Court, therefore, correctly reverses defendant’s death sentence. Ante at 326-27. I submit, however, that the error runs much deeper, and that the dispositive principle of State v. Ramseur requires that this defendant not face a possible death sentence on retrial. I therefore disagree with the Court, which rules that defendant may again be exposed to the death sentence. Ante at 327.
The trial court, as noted, instructed the jury that their findings had to be unanimous. This instruction was reflected in the following interrogatory on the special verdict form: “Do you unanimously agree that the following exists as a mitigating factor?” In attempting to clarify for the jury, during deliberations, whether a “no” answer meant “no, we are not unanimous” or “we are unanimous in saying ‘no,’ ” the court contradicted itself. The court told Juror Number 5: “Anything that you have to decide, you have to decide by unanimous vote. If you can’t reach a unanimous vote, you have not reached a decision on the issue.” The court thus implied that all had to agree one way or another. When Juror Number 15 voiced confusion over that standard, the court read the question aloud, then said: “If the answer is ‘no,’ it means that you can’t reach a unanimous vote that it exists. Okay?” The court thus instructed the jury first that they had to reach a unanimous verdict to say “no,” and then that they had to say “no” if they could not reach a unanimous verdict.
In addition, it was evident that there were jurors who thought that the mitigating factor existed. When the jury returned, the court stated: “Everybody agrees there’s hopeless *381deadlock on that last [mitigating factor] question, is that correct? THE FOREMAN: Yes.” 12
In State v. Ramseur, supra, 106 N.J. at 312-15, this Court held that because of coercive supplemental jury charges, which broke the deadlock that could have saved the defendant’s life, the defendant would not face a possible death sentence at resentencing. The Court stated:
We hold that where a trial court in a capital case has erroneously given coercive supplemental instructions ... to a jury that has expressed its inability to agree, the law must afford defendant the benefit of the final non-unanimous verdict that might have been returned absent the coercion. Having erroneously been deprived of a substantial opportunity to a jury verdict resulting in imprisonment rather than death, the defendant may not be subject to another capital sentencing proceeding.
We recognize that any reversible error in a capital case may be said in some sense to have deprived a defendant of the opportunity to a jury verdict resulting in imprisonment, and that nevertheless the usual and proper remedy for such errors is reversal of the death sentence and a retrial of the sentencing proceeding in which the defendant may again face the death penalty____ But a Czachor error is critically different from other prejudicial errors in that by definition it occurs after the jury has clearly demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict for the death sentence. The erroneous coercive charge has, not simply as a possible consequence but as its sole purpose, ending the disagreement that would save defendant’s life. Even if the jury’s disagreement has not reached the point at which further deliberations would be improper, those deliberations must take place in an atmosphere free of coercion. If such coercion occurs, the defendant has irrevocably lost not merely a theoretical possibility but a substantial likelihood that, absent the error, the jury would have reached a verdict resulting in imprisonment rather than death, [/d. at 313-14 (emphasis added).]
The Ramseur Court thus took pains to emphasize the nature of the coercive supplemental charge; the critical element, the *382Court held, is that “by definition it occurs after the jury has clearly demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict for the death sentence.” Id. at 313.
The primary vice of the coercive supplemental charge, then, is that it compels jurors who believe that a mitigating factor is not outweighed beyond a reasonable doubt by aggravating factors into abandoning that position for the sake of unanimity. That was precisely the effect of the trial court’s charge in this case; any juror who believed that the mitigating factor existed was coerced into abandoning that belief prior to the weighing process by the requirement of unanimity. Because all agreed, moreover, that two aggravating factors existed, the inability to find a mitigating factor unanimously meant that the verdict of death was automatic. Thus, any jurors who believed that the mitigating factor existed were literally coerced into voting for death; indeed, the coercion was so powerful that their belief one way or another about the mitigating factor’s weight was not even allowed to be registered. In Ramseur, by contrast, the jurors reached the weighing process, at least, before the coercion forced jurors to abandon their beliefs.
As in Ramseur, moreover, the jury in this case demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict. There was obvious resistance by the jury to abandoning the position that the mitigating factor existed; the jury sought clarification of the need for unanimity, and ended deliberations deadlocked on the existence of the mitigating factor. This, coupled with the facts that no mitigating evidence was presented, and that the jury also deadlocked on the existence of the only aggravating factors that called for substantive judgment, raises “a substantial likelihood that, absent the error, the jury would have reached a verdict of imprisonment rather than death.” 106 N.J. at 314. Since “[t]he remedy must be commensurate with the wrong” (id. at 313), and since the wrong here is indistinguishable in principle from the wrong in *383Ramseur, the remedy should be the same; defendant should not face another death sentence.
I note, finally, that the Ramseur Court buttressed its holding on the coercive charge by reference to other errors, e.g., the jury was not properly apprised of its responsibility. Id. at 315. In this case, the coercion of the jurors culminated a trial that featured blatant prosecutorial misconduct and infection of the jury with prejudicial publicity, among other infirmities, and in which the court erroneously acquiesced in the defendant’s desire not to present any mitigating evidence whatsoever. Under the enhanced standard of appellate review applicable in capital cases, these facts only strengthen, in my view, the case for not allowing this defendant to face another death sentence.
VII.
I state, in conclusion, that the unconstitutionality of the capital murder-death penalty statute is reconfirmed by this case. It is not, I suggest, simply judicial inexperience that has produced incongruous and inconsistent results. The statute, lacking cohesive and understandable standards, is insusceptible of fair and impartial management. The absence of guided prosecutorial discretion in this statutory framework, coupled with no current means of determining the proportionality of sentences, inflicts constitutional wounds too deep to close. These chronic constitutional infirmities as well as the failure to develop and apply an enhanced standard of review explain, in large measure, the anomalies that are arising in our capital murder jurisprudence.
James Koedatich may well have murdered Amie Hoffman; if so, he should be punished severely as provided by law. The State did not prove that he had, however, before a jury that was untainted by highly prejudicial publicity. The State’s weak case was in reality buttressed by this deluge of publicity, which, to some extent must have conditioned the jurors to a finding of guilt. Finally, the State did not prove defendant’s guilt without *384attempting to influence the jury’s deliberations in flagrantly improper ways.
The prejudice deriving from these errors requires that the conviction and sentence must be reversed.
Justice O’HERN joins in Part II of this dissent.
For affirmance of conviction, reverse, sentence and remand — Chief Justice WILENTZ, and Justices POLLOCK, GARIBALDI and STEIN — 4.
For reversal — Justices CLIFFORD, HANDLER and O’HERN — 3.
A11 105 prospective jurors were required to fill out a questionnaire, in addition to answering questions from the trial court. Twenty-six jurors were excused by the court before questioning on the basis of the questionnaires. Of the 79 prospective jurors who were asked if they had read or heard about the case, nine said that they had not; four of these nine became deliberating jurors. Forty of the 70 who indicated some knowledge of the case were excused for cause by the court.
Judith Deelsnyder had read about defendant's arrest; Loraine Zaccaro and Glenn Cosentiono had read about the case primarily two years earlier, but recalled nothing of the defendant’s background; Elizabeth Criares had last read about the case three months before jury selection, and associated the defendant with stabbing; Alice Budd, Deborah Brown, Margaret Andrews, and Barbara Herzig each had knowledge of the case and of the defendant’s arrest, but disclaimed any knowledge of the defendant’s background; Leslie Fascia, an alternate, remembered reading something about the defendant "two years ago,” and remembered reading something "about the body being found, out on Route 80. I don’t remember.”
In Biegenwald, the court acknowledged that "there was extensive pretrial publicity,” including numerous articles in which the prosecutor "repeatedly assumed defendant's guilt and also stated that defendant killed only for pleasure.” 106 N.J. at 30-31. The Court concluded, however, that the trial court was not required to "presume prejudice" because the trial court adjourned the trial date, "allowing nearly six months to permit the impact of the publicity to subside." Id. at 35. The Court then concluded that the proper test was whether the jury selection process actually resulted in a fair and impartial jury, and that the trial judge’s voir dire resulted in an impartial jury panel. Id. at 36-37. I strongly disagreed. Id. at 81-98 (Handler, J., dissenting).
In Biegenwald, the Court found that the voir dire had effectively established an impartial jury, specifically because "[s]everal of the trial jurors stated that they had never heard of the defendant before coming to court” and all sixteen jurors “indicated that they had encountered little or no publicity regarding the case." Id. at 36. Only one juror expressed knowledge of alleged prior murders —and not of the prior conviction — and she "convinced the trial court," moreover, "that she could disregard what she heard and serve impartially." Id. at 37.
An article on March 10, 1985, in the Daily Record, notes that Judge Stein [the trial court] "excluded any juror who knew of Koedatich’s arrest record,” but then relates that "Stein’s conservative approach did not prevent some jurors from linking Koedatich to both murders once the trial began. The Daily Record has learned that some of the jurors who convicted Koedatich of the Hoffman murder knew he was also charged with killing O’Brien. This information was not allowed to be brought before the Hoffman jurors because it could prejudice Koedatich’s case. But a few jurors said once the trial got underway they realized Koedatich was charged with killing both women." Specifically, Ms. Zaccaro stated, "I didn't know he did it in the beginning but halfway through the trial I realized he was linked to O'Brien.” Ms. Zaccaro also said that the O’Brien murder was not discussed until after the verdict was returned. Leslie Fascia "said she thought most of the jurors were aware that Koedatich was charged with the O’Brien murder. She said jurors were surprised to learn during the penalty phase ... that Koedatich was convicted of murder in Florida.” The majority’s emphasis on the fact that we rejected an interlocutory appeal by the defendant requesting an examination of all jurors based on this article is not determinative. Our rejection was influenced by the uncertain and incomplete nature of the appeal process; it did not purport to rule on the materiality of the report as probative evidence of jury taint.
In addition, defendant argues that the silence of the jurors on voir dire— their failure to recollect — on their exposure to prejudicial information "de*363prived the defendant of the opportunity to have excused a juror by appropriate challenge.’ ’’ (citing In re Kozlov 79 N.J. 232, 235 (1979); State v. Thompson, 142 N.J.Super. 274, 280-82 (App.Div.1976)). Given the nature of the prejudicial information, defendant argues, it is reasonable to assume that a peremptory challenge would have been exercised had the jurors been qualified by the court. I agree. See State v. Bey (II), 112 N.J. 123, 188 (1988) (Handler, J., dissenting). The majority's answer to this argument — that there were many instances where defense counsel would have been justified in exercising a challenge based on publicity but failed to (ante at 277 n. 7), lends inadvertent support to my general view: that defendant was denied a fair trial before an impartial jury.
The Court’s failure to safeguard the impartiality of this jury is the more egregious because several of its evidentiary rulings are expressly dependent on that impartiality. Thus, the Court upholds the trial court's admission of testimony relating to the circumstances of the defendant’s arrest — including references to flashing blue lights and being forced off the road — despite the *365fact that this information related closely to the circumstances of the O’Brien murder. Ante at 314-15. I agree that in a neutral venue, a venue unfamiliar with the facts of these cases, the ruling would have been proper. Under the circumstances in which the trial was conducted, and in view of the fact that something did in fact trigger juror recollection, I cannot agree that this venue was neutral. The atmosphere was fraught with the potential for prejudice. Similarly, I agree that in a neutral venue, it would have been "reasonable under the totality of the circumstances to commit the issue of [Diana] Bossard’s credibility to the jury." Ante at 317. This was, however, a jury which may have been as subject to suggestible publicity as the witness herself, and was thus not in a position to judge a predisposition it may well have shared.
The prejudice attributable to these events was exacerbated because of the circumstantial nature of the State’s case. The predominantly circumstantial nature of the incriminating evidence also arguably supports the defendant’s contention that the trial court wrongly excluded evidence implicating a third party in the murder. These rulings, discussed by the majority (ante at 297-312), excluded evidence that a third party, Kevin Sheehan, owned a sports car (ostensibly connecting with testimony that a witness heard a small sports car at the vicinity and time of the murder), that Sheehan gave an inconsistent *367account of his whereabouts on the night in question, that Sheehan implicated himself in front of his own attorney, that Sheehan would invoke the privilege against self-incrimination if questioned, that Amie Hoffman held unsupervised parties, and that obscene phone calls were received in the cheerleader’s locker room. Despite the fact that some of this evidence could be excluded on hearsay and Evidence Rule 4 grounds, it might have established a possible, albeit tenuous, connection between Sheehan and the murder. It is not at all clear that the trial court fully appreciated that greater latitude toward admissibility is appropriate in this kind of case, precisely for the reason that "the evidence against defendant was completely circumstantial, and because the jury fixed the punishment, it was essential that no proof which might conceivably affect the finding of guilt or innocence or the punishment imposed, be excluded if properly admissible." People v. Davis, 29 Ill.2d 127, 129, 193 N.E.2d 841, 842-43 (1963); see also State v. LeClair, 425 A.2d 182, 186 (Me.1981) ("Especially where the state’s case is based upon circumstantial evidence, the court should allow the defendant ‘wide latitude’ to present all the evidence relevant to his defense, unhampered by piecemeal rulings on admissibility." (Citation omitted.)) Given the uncertain and highly circumstantial nature of the state’s case against defendant, and the blatant prosecutorial misconduct discussed infra, I believe the trial court should have been more willing to admit this evidence of third party guilt.
The prosecutor stated on summation:
This part of the trial, folks, is when we think about Amie Hoffman because as the judge said before, Mr. Kenny has his client sitting here and I don’t have a client here and that’s hard, it's hard to get across (sic) to you during the course of a trial that there once was a human being named Amie Hoffman who I think we've come to know to some extent and we have pictures that were introduced. This is how I want to know Amie Hoffman, a cheerleader, someone who had a life, who had a future. That’s the Amie Hoffman that I want to know____
********
Let’s think a little bit about what this girl went through, what was going on in her head when she was taken from the mall and taken up to that retention tank in Randolph because what was she thinking about when that knife was to her nose and to her throat and she was being raped. Was she thinking about was she going to get out of there? Did she have any idea that she wouldn't make it out of that area?
This whole case protected the rights of the defendant. The judge was here ruling on all the evidence, making sure that admissible evidence went in, making sure that this trial went according to our due process and our constitution. Well, where was Amie Hoffman's due process. Where was the victim's judge? There he is right there because he is the one that sentenced Amie Hoffman to die. She didn't have a jury of the community to decide whether or not she lives____
********
You know, it's not easy for me to have to come in here on Monday morning and ask twelve people to sentence someone to die. That’s not easy. I share that burden with you but there comes a point in time when you have to say it’s over. When a person like the victim is killed, there has to be some justice and no matter how many times you turn those pages and how many times you think about can I sentence someone to die, each of you told us at the beginning that depending on the facts and circumstances, you could consider the death penalty and I took that to mean that *371if I presented the right case to you, a case strong, a case of cold-bloodedness, you could sentence someone to die and you knew at that time like I’m standing here right now that I have to ask you that.
I’ve proved to you the aggravating factors. You have all the reason in the world to sentence him to death. I’m asking you to do what you said you could to in the right set of circumstances and, folks, if this isn’t the right set, I don’t know if there is one.
Justice White, concurring, addressed the prosecutorial discretion argument separately "because the cases in which no capital offense is charged escape the view of the Georgia Supreme Court and are not considered by it in determining whether a particular sentence is ... disproportionate.” Id. 428 U.S. at 225, 96 S.Ct. at 2949, 49 L.Ed.2d at 903.
This decision was later overturned by a referendum amending the State Constitution.
When the court came to the question in the verdict sheet, utter confusion reigned:
THE COURT: Okay. Do you unanimously find that the following exists as a mitigating factor: Any factor which is relevant to the defendant’s character or record or to the circumstances of the offense of murder?
THE FOREMAN: No.
THE COURT: The answer is no?
THE FOREMAN: No.
THE COURT: And is that — that, of course, is unanimous, is that correct? THE FOREMAN: Yes, your Honor.