dissenting.
The defendant in this case, James I. Hunt, was convicted of the murder of Edward Lawson and thereafter sentenced to death. He filed a direct appeal challenging both the conviction and the sentence. This Court affirms defendant’s conviction for Lawson’s murder and related offenses but reverses the imposition of the death penalty. The Court further concludes that because of circumstances relating to jury coercion, the defendant cannot again be exposed to the death sentence.
I concur in the Court’s reversal of the death sentence and its determination that the defendant cannot once again face the *390death penalty. However, I disagree with the Court’s refusal to set aside the defendant’s murder conviction.
I again express my fundamental differences with the Court with respect to the validity of the New Jersey capital murder-death penalty statute on state constitutional and fundamental fairness grounds. See, e.g., State v. Williams (II), 113 N.J. 393, 458-61 (1988) (Handler, J., dissenting); State v. Rose, 112 N.J. 454, 562-67 (1988) (Handler, J., dissenting). I need not repeat those positions. Nevertheless, in this case, I feel that it is important to address three issues, namely, the use of death-qualified jurors in the trial of the issue of guilt in capital causes; the failure to give the jury proper instructions concerning the offense of serious-bodily-injury murder as required under State v. Gerald, 113 N.J. 40 (1988); and the validity and application of the standards that govern one of the statute’s aggravating factors that is invoked to determine death eligibility: N.J.S.A. 2C:11-3c(4)(c) (“c(4)(c)”).
I.
The defendant challenges the empanelling of the jury in this case on a number of grounds. These contentions and the Court’s disposition of them are presented in its opinion. Ante at 347-364. Among the contentions raised by the defendant is that use of a death-qualified jury during the guilt phase of his bifurcated trial and the death-qualification process itself are unconstitutional. Ante at 355-356. The majority, without discussion, summarily rejects these contentions. Ante at 356. I believe the state of the record as well as evolving authority demand a fuller consideration of these contentions. There are, in my view, strong reasons for invalidating or modifying the death-qualification process in the prosecution of capital cases and for developing alternative jury and trial procedures to overcome the prejudice inherent in the current use of death-qualified juries.
*391A.
The defendant contends that the death-qualification process, which requires potential jurors to express, prior to the guilt phase, their ability to return a death sentence, deprived him of the right to an impartial jury. The death-qualification process itself, according to defendant, improperly conditions jurors to a predisposition, prior to the actual guilt-phase trial, that the defendant is guilty of murder.
There is in this case some factual substance to the defendant’s complaint that the death-qualified jury was, indeed, conviction-prone. The Court acknowledges that of the eighty-eight potential jurors who were not excused for reasons unrelated to their views on the death penalty, twenty-two, or 25%, were excused because of their opposition to the death penalty, while only six, or 6.8%, were excused because they would automatically impose the death penalty on all persons convicted of murder. Ante at 356. These statistics graphically underscore the pervasive awareness on the part of jurors of the realistic need to deal with the imposition of the death penalty at some time during the trial. This awareness saturated the selection process and indicates that a jury thus selected will be unduly sensitized and conditioned to the defendant’s guilt.
In meeting defendant’s contention, the Court contents itself with its rather conclusory determination in State v. Williams (II), supra, 113 N.J. 393. There it observed that as long as the trial court ascertains that each juror will apply the law, and that each juror understands both the presumption of innocence and the State’s burden to prove the defendant guilty beyond a reasonable doubt, allegations of conviction-proneness may be regarded as exaggerated or illusory. Id. at 414 n. 6. Reiterating Williams (II), the Court states that “[sjince a properly instructed jury can understand that death qualification is based on a hypothetical finding of guilt, and nothing more, we believe that the risk of prejudice to the guilt-innocence phase is minimal.” Ante at 356.
*392This cavalier conclusion fails to give any weight to the indications that, in addition to selecting for service jurors whose attitudes are more punitive than those who are death-scrupled, the process of death qualification itself engenders partiality among the jurors. The selection process, I believe, underscores the issue of the imposition of the death penalty itself. State v. Bey (II), 112 N.J. 123, 192 (1988) (Handler, J., dissenting) (citing Haney, “On the Selection of Capital Juries: The Biasing Effects of the Death Qualification Process,” 8 Law & Hum. Behav. 121 (1984)). The entire process has a strong gravitational pull in the direction of guilt. It confronts potential jurors with an atmosphere charged with the likelihood of guilt, rather than “a hypothetical finding of guilt, and nothing more.” As Justice O’Hern observed, one cannot read the transcript of a death-qualifying voir dire proceeding without a sense of “impending doom.” See State v. Ramseur, 106 N.J. 123, 341 (1987) (O’Hern, J., concurring) (quoting Hovey v. Superior Court of Alameda County, 28 Cal. 3d 1, 70-71, 616 P. 2d 1301, 1348, 168 Cal.Rptr. 128, 175 (1980)). Minimally, therefore, the jury in the trial of a capital cause should be death-qualified only if and when it becomes necessary to determine whether the death penalty should be imposed.
B.
I continue to disagree with the position that the use of a death-qualified jury in a bifurcated trial of a capital case is valid under state-constitutional and fundamental-fairness standards. See State v. Bey (II), supra, 112 N.J. at 191-98 (Handler, J., dissenting). This Court, in Ramseur, without extended reasoning, chose to follow the decision in Lockhart v. McCree, 476 U.S. 162, 106 S.Ct. 1758, 90 L.Ed.2d 137 (1986), where the Supreme Court ruled that it was not unconstitutional for a death-qualified jury to determine a defendant’s guilt in a capital case. Ramseur, supra, 106 N.J. at 248-54. I disagreed, noting the continuing social-science research that strongly suggests that persons who favor the imposition of the death *393penalty may have more punitive attitudes than those who disfavor the death penalty. Id. at 431 (Handler, J., dissenting). I have pointed out more recently in State v. Bey (II), supra, 112 N.J. at 191, that the literature appearing since McCree buttresses the conclusion that death-penalty proponents are conviction-prone, citing Seltzer, Lopes, Dayan, and Canan, “The Effect of Death Qualification on the Propensity of Jurors to Convict: The Maryland Example,” 29 How. L.J. 571 (1986). There is, in my view, enough evidence to justify abandonment or, at the very least, modification of the current death-qualified jury methodology in the trial of capital cases.
I struggle in vain to identify cogent reasons to explain the Court’s position on this fundamental issue. In Ramseur, in Bey (II), in Williams (II), and now in this case, the Court refuses to come to grips with the clearly disturbing implications created by using a death-qualified jury. Its passive acceptance of the Supreme Court’s lead on this fundamental issue is both baffling and unsettling for several reasons.
First, the Court gravely misunderstands the weight that a state should attribute to the federal constitution with respect to the criminal law of capital-murder. The Supreme Court itself has repeatedly adjured states to look to their own constitutions and laws. See California v. Ramos, 463 U.S. 992, 1013-14, 103 S.Ct. 3446, 3459-60, 77 L.Ed.2d 1171, 1188-89 (1983). Consequently, this Court’s frequent attempts to clone the federal constitution to determine and define critical capital-murder issues and rights is more than a doctrinal distraction. It has become a major barrier to the development of a cohesive body of substantive and procedural law to govern the prosecution of these complex and unique causes.
Second, in terms of capital-murder jurisprudence, the Supreme Court’s approach boarders on the chaotic. For example, we have only recently rejected the federal court’s own shifting constitutional analysis of what culpable state of mind is necessary in order for a murder to rise to the level of a capital *394offense. E.g., Gerald, supra, 113 N.J. at 69-91. We most emphatically pronounced that protections under the New Jersey Constitution are different from and greater than those under the federal Constitution. Id. at 76. Yet, in Ramseur, on the equally important issue of fair and impartial juries, the Court said that “the protections regarding death qualification afforded under the New Jersey Constitution are no different from or greater than those under the federal Constitution.” 106 N.J. at 251. It should be apparent that the Court has become engaged in the random selection of constitutional protections, sometimes federal, sometimes state; its approach to capital-murder jurisprudence is becoming indistinguishable from the federal approach in its lack of consistency.
Third, and most importantly, because of the Court’s reflexive recourse to the federal experience, the protections that the Court ultimately chooses are ironically drained of vigor. The Court’s approach fails woefully to appreciate just how minimal are the protections of the federal constitution in capital-murder law. This is true, for example, with respect to a defendant’s constitutional interest in jury impartiality, an important issue in this case. Thus, in Buchanan v. Kentucky, 483 U.S. 402, 107 S.Ct. 2906, 97 L.Ed.2d 336 (1987), the Supreme Court, in the trial of a defendant not charged with capital murder, upheld the use of a death-qualified jury because he was being tried jointly with a defendant charged with capital murder. The Court approved this practice for reasons of prosecutorial convenience, despite its perception that “ ‘death qualification’ in fact produces juries somewhat more ‘conviction-prone’ than ‘non-death-qualified’ juries’...” 483 U.S. at 415 n. 16, 107 S.Ct. at 2913 n. 16, 97 L.Ed.2d at 350 n. 16. This attitude departs radically from our own insistence on jury impartiality. See, e.g., State v. Corsaro, 107 N.J. 339, 352-53 (1987); State v. Ragland, 105 N.J. 189, 195-96 (1986); State v. Ingram, 98 N.J. 489, 500 (1985); State v. Collier, 90 N.J. 117, 124 (1982); State v. Ingenito, 87 N.J. 204, 216-17 (1981); State v. Czachor, 82 N.J. 392 (1980); State v. Simon, 79 N.J. 191 (1979). Our commit*395ment to jury impartiality is not a recent development. It is deeply embedded in our tradition of criminal justice, appropriately heightened in capital-murder prosecutions. Historically, this Court has understood that “[i]t is vital in capital cases that the jury should pass upon the case free from external causes tending to disturb the exercise of deliberate and unbiased judgment.” State v. Kociolek, 20 N.J. 92, 105 (1955) (quoting Mattox v. United States, 146 U.S. 140, 149, 13 S. Ct. 50, 53, 36 L.Ed. 917, 921 (1892)); see State v. Jackson, 43 N.J. 148, 158 (1964), cert. den. sub nom. Ravenell v. New Jersey, 379 U.S. 982, 85 S.Ct. 690, 13 L.Ed.2d 572 (1965); State v. Mount, 30 N.J. 195, 212-13 (1959). It is difficult to imagine an “external cause” that would tend more to “disturb the deliberate and impartial judgment” of the jury than death qualification, which both selects a more punitive jury and conditions that jury to believe that the defendant is guilty.
In matters surrounding the fairness and impartiality of jurors, perhaps more so than with respect to other fact-sensitive and evidentiary aspects of a criminal trial, we are entitled to draw on common sense grounded in ordinary human experience. In the selection of a jury, we are not dealing with threshold questions entailing the competence or admissibility of evidence. Courts are not, therefore, remitted to determinations that must be informed only by the kind of knowledge that satisfies rigorous scientific standards. Thus, on the issue of death qualification, we should be able to rely on our common sense and knowledge concerning the fairness of ordinary persons of average experience, particularly when this knowledge is further informed by studies demonstrating the conviction-proneness of death-qualified juries. Under our jurisprudence, that risk of jury unfairness should be enough to compel the State to develop alternatives. There is, in my estimation, a strong constitutional impetus to explore and develop a better way.
*396c.
There are alternatives to the current death-qualification methodology of trying capital causes. See, e.g., Berry, “Remedies to the Dilemma of Death-Qualified Juries,” 8 U. Ark. Little Rock L.J. 479 (1985) (hereafter “Remedies”); Comment, “Inequities and Abuses of Death Qualification: Causes and Cures,” 23 S.D.L. Rev. 281 (1987). One realistic alternative is the use of dual juries, which can be adapted from current experience.
The multiple-jury procedure is not a concept wholly alien to our system of jurisprudence. It has been developed primarily as a solution to the dilemma posed by the United States Supreme Court’s decision in Bruton v. United States, 391 U.S. 123, 88 S.Ct. 1620, 20 L.Ed. 2d 476 (1968).1 See State v. Corsi, 86 N.J. 172, 176 (1981). Under this procedure, two juries are impanelled separately, each with a separate responsibility to determine the guilt or innocence of a particular defendant. The juries, however, are seated together for the presentation of all matters, including the introduction of evidence or testimony, that are relevant to both defendants. Conversely, each jury is excused from the presentation of such matters that would be irrelevant or inadmissible in the trial of the particular defendant for which it is responsible. Thus, common or overlapping matters, that is, matters relevant to both trials, need be presented only once, resulting in a substantial savings of time and enhancing the likelihood of consistent results, without prejudicing either party. Gaynes, “Two Juries/One Trial: Panacea of Judicial Economy or Personification of Murphy’s Law,” 5 Am. J. Trial Advoc. 285 (1981).
*397While the Bruton mandate is the major reason for employing multiple juries, the procedure has also been used to minimize prejudice at a joint trial of codefendants who assert antagonistic or inconsistent defenses. See People v. Kramer, 103 Mich.App. 747, 303 N.W.2d 880 (1981); People v. Brooks, 92 Mich.App. 393, 285 N.W.2d 307 (1979).2 It has also been proposed for use in complex litigation where the laws of two or more states must be applied. See Martin v. Bell Helicopter, No. 77-F-533 (U.S.Dist.Ct., Colo., Feb. 21, 1980) (see 66 A.B.A. J. 787 (1980)) (case settled prior to actual use of multiple juries).
The advantages of the multiple-jury procedure have spurred many courts to approve it. See, e.g., United States v. Hayes, 676 F.2d 1359, 1366-67 (11th Cir.), cert. den., 459 U.S. 1040, 103 S.Ct. 455, 74 L.Ed.2d 608 (1982); United States v. Rimar, 558 F.2d 1271 (6th Cir.1977), cert. den., 435 U.S. 922, 98 S.Ct. 1484, 55 L.Ed.2d 515 (1978); United States v. Rowan, 518 F.2d 685, 689-90 (6th Cir.), cert. den. sub nom. Jackson v. United States, 423 U.S. 949, 96 S.Ct. 368, 46 L.Ed.2d 284 (1975); United States v. Sidman, 470 F.2d 1158 (9th Cir.1972), cert. den., 409 U.S. 1127, 93 S.Ct. 948, 35 L.Ed.2d 260 (1973); People v. Wardlow, 118 Cal.App.3d 375, 382-387, 173 Cal.Rptr. 500, 502-05 (1981); People v. Brooks, 92 Mich.App. 393, 285 N.W.2d 307 (1979). The multiple-jury procedure has even been approved for use in a capital-murder prosecution of codefendants posing a Bruton-type problem. State v. Beam, 109 Idaho 616, 710 P.2d 526 (1985), cert. den., 476 U.S. 1153, 106 S.Ct. 2260, 90 L.Ed.2d 704 (1986).3
*398This procedure can be adapted to the trial of a capital-murder case. The prosecution of the case would remain bifurcated; the guilt and penalty issues to be tried sequentially. Under such a procedure, a defendant’s guilt or innocence would be determined by a jury that is empanelled under the standards controlling the selection of juries in ordinary criminal cases; such jurors would not be death-qualified. A second jury, duly death-qualified, would also be empaneled to determine the issue of the imposition of the death penalty. Because evidence probative of guilt is also probative of penalty, the guilt phase of the prosecution would be tried before both juries. The former jurors would determine guilt or innocence of murder and death-eligibility. In the event this jury determined the defendant not guilty of capital murder, i.e., that the defendant was guilty of a lesser form of murder or another form of homicide, the second jury would be discharged. The trial court would then determine the sentence as in an ordinary non-capital case. If the first jury determined that the defendant was guilty of capital murder, the sentencing phase of the prosecution would then be tried only before the second, death-qualified jury. This jury would determine the appropriate sentence based on the evidence adduced during the guilt-phase trial, including the defendant’s conviction, as well as the evidence presented during sentencing-phase trial. In sum, under such a procedure, the issue of defendant’s guilt would be determined by a conviction-neutral jury uninfluenced by the death-qualification process. The defendant’s sentence would be determined by a death-qualified jury.
*399There are, of course, differences in the use of a multiple-jury procedure in a capital prosecution from its use in other settings. In a Bruton-type case, the concern is that a single jury cannot constitutionally hear and review evidence presented at a joint trial of codefendants when such evidence implicates a non-testifying defendant. Thus, in these cases, each jury is excused with respect to evidence or other matters that is not admissible in the trial in which it must determine guilt or innocence. In contrast, the rationale for employing multiple jurors in capital cases would be to avoid the determination of a defendant’s guilt or innocence by a jury programmed through the process of death qualification to return a verdict of guilty. This can be achieved by delegating the responsibility for determining guilt only to a non-death-qualified jury. Moreover, only after the non-death-qualified jury rendered a verdict of guilty of capital murder would the death-qualified sentencing jury exercise its responsibility for determining the penalty. There would be no need to continue the participation of the guilt-phase jury in the sentencing trial.
There undoubtedly are complicating considerations in such a procedure. Some critics have maintained that the use of two juries in a capital case, one to determine guilt and one punishment, results in a division of responsibility that dilutes accountability. See Grigsby v. Mabry, 758 F.2d 226, 247 (1985) (Gibson, J., dissenting), rev’d, Lockhart v. McCree, 476 U.S. 162, 106 S.Ct. 1758, 90 L.Ed.2d 137 (1986). Because the guilt-phase jury does not actually render a sentence of death, arguably some jury members may feel less restrained in finding guilt because the responsibility for deciding the ultimate sentence is passed on to the sentencing jury. See Crump, “Lockhart v. McCree: The ‘Biased But Unbiased Juror,’ What are the State’s Legitimate Interests?,” 65 Den. U.L. Rev. 1, 19 (1988); Berry, Remedies, supra, 8 U.Ark. Little Rock L.J. at 499-500.
The point is esoteric. In the overwhelming majority of criminal cases in this country sentencing is undertaken by the court; jury sentencing is the exception to the rule. In fact, in non-cap*400ital cases, jury sentencing occurs in only seven states. III American Bar Association Standards for Criminal Justice Ch. 18, § 1.1, at 20 (2d ed.1980). Hence, in the typical case, the jury is often admonished not to concern itself with the sentencing consequences of its determination of guilt. Moreover, even in capital cases, where the penal stakes are the highest, a number of state statutory schemes provide that after a jury trial on guilt, the court alone is to impose the sentence. Also, some states permit the trial court to override the sentencing verdict of the jury.4 In our own state, we permit the court “for good cause” in a capital case to discharge the jury that presided at trial and empanel a new jury for the penalty phase. N.J.S.A. 2C:ll-3(c)(l).
An equally tenuous objection relates to juror nullification; arguably, jurors in capita] cases can intentionally, albeit wrongfully, acquit a defendant. It is surmised that such jurors, because of their aversion to the death penalty, would automatically acquit any capital defendant despite the compelling nature of the evidence of guilt. Note, “Lockhart v. McCree: Death Qualification as a Detriment of the Impartiality and Representativeness of a Jury in Death Penalty Cases,” 72 Cornell L.Rev. 1075, 1106-07 (1987).
This argument exaggerates the significance of “jury nullification” in the context of capital-murder prosecutions. It is a phenomenon that attends all criminal causes. “[A] jury has the prerogative of returning a verdict of innocence in the face of overwhelming evidence of guilt.” State v. Ingenito, supra, 87 *401N.J. at 212. Such a verdict in favor of the criminal defendant, even in a death-penalty case, is subject to neither review nor reversal. See McCleskey v. Kemp, 481 U.S. 279, 311, 107 S. Ct. 1756, 1777, 95 L.Ed.2d 262, 291 (1987). Acquiescence in this jury prerogative is influenced by the belief that in a criminal proceeding the jury serves as the conscience of the community. See, e.g., United States v. Quarles, 350 U.S. 11, 18-19, 76 S. Ct. 1, 5-6, 100 L.Ed. 8, 15 (1955); Ingenito, 87 N.J. at 212; Gobert, “In Search of the Impartial Jury,” 79 J.Crim.L. & Criminology 269, 301-02 (1988). Notions of jury nullification are appropriately implicated in our recognition that a jury in a capital-murder prosecution may indeed be allowed to consider sympathy for the defendant. See Ramseur, supra, 106 N.J. at 296-97.
Another argument is based on so-called “residual doubts.” A juror may harbor lingering doubt concerning guilt that could be carried into the sentencing phase of the trial and form the basis of a refusal to impose the death penalty. L. Black, Capital Punishment: The Inevitability of Caprice and Mistake in Administering the Death Penalty (part I) (1981); see Heiney v. Florida, 469 U.S. 920, 921-22, 105 S.Ct. 303, 304, 83 L.Ed.2d 237, 238 (1984) (Marshall, J., dissenting). Such “residual doubt” has been advanced as a protection for defendants that would be lost if the use of a single death-qualified jury in the trial of both guilt and penalty in capital causes were abandoned. See, e.g., Grigsby v. Mabry, supra, 758 F.2d at 247-48 (Gibson, J. , dissenting) (State’s interest in a unitary jury system based, in part, on the theory that defendant may benefit from jury’s residual doubts.); Smith v. Balkcom, 660 F.2d 573, 580-81 (5th Cir.1981), modified, 671 F.2d 858 (5th Cir.1982) cert. den., 459 U.S. 882, 103 S.Ct. 181, 74 L.Ed.2d 148 (1982).
Nevertheless, a capital defendant does not have a constitutional right to require capital juries to consider their residual doubts about guilt in the sentencing phase of the trial. Franklin v. Lynaugh, — U.S. -,-, 108 S.Ct. 2320, 2326-27, 101 L.Ed.2d 155, 165-66 (1988). Moreover, as already noted, there are numerous instances where the guilt-phase jury does *402not decide the sentence of a capital defendant, and consequently the defendant does not benefit from residual doubt.
In sum, these several objections expressed in terms of juror accountability, nullification, and residual doubt do not severally or collectively overcome the prejudice of conviction-proneness; they do not justify the retention of death-qualified juries to determine guilt. Conversely, the use of death-qualified jurors to determine criminal guilt generates a material risk of jury partiality. It is a consideration that weighs heavily in favor of a multiple-jury procedure for the trial of capital causes.
D.
Another reason that is advanced to justify the current use of the death-qualified jury is prosecutorial convenience; the State has an interest in a single-jury, sequential or bifurcated trial because it is simpler and easier to try a defendant for capital murder. See Bey (II), supra, 112 N.J. at 197 (Handler, J., dissenting). Prosecutorial preference for the death-qualified, single-jury bifurcated trial methodology cannot, however, outweigh the defendant’s right to an impartial jury. See id. at 197-98 (Handler, J., dissenting). Yet, the Court now appears to agree with the Supreme Court’s view that prosecutorial convenience can outweigh the defendant’s interest in a fair trial by an impartial jury. See, e.g., Buchanan, supra, 483 U.S. at 417, 107 S.Ct. at 2915, 97 L.Ed.2d at 352. Rather, the Court should heed Justice O’Hern’s earlier observation that “the inconvenience entailed in providing for a non-death-qualified jury — one that is fair and impartial — in the trial of guilt is not too high a price to pay to vindicate constitutional interests.” Ramseur, supra, 106 N.J. at 339-43 (O'Hern, J., concurring).
The convenience that attends the single-jury bifurcated-trial approach simply is not important enough to be continued. The asserted inconvenience of the multiple-jury bifurcated system has to do more with physical and financial demands than procedural complexity. A multiple-jury system may involve *403physical adjustments and financial cost. It is a rare situation, however, where these considerations outweigh constitutional interests. See, e.g., Ballew v. Georgia, 435 U.S. 223, 98 S.Ct. 1029, 55 L.Ed.2d 234 (1978). As Justice O’Hern so eloquently stated, “[ojurs is a civilization uniquely committed to the value of human life. Our nation does not view the lives of its citizens as easily expendable. We do not measure the value of human life in. dollars.” Ramseur, 106 N.J. 123, 342 (O’Hern, J., concurring).
I do not proffer the multiple-jury bifurcated-trial procedure as an alternative to the single, death-qualified jury bifurcated procedure to suggest that under the State Constitution it is required and must be implemented. Rather, the proffer is made to demonstrate that there are viable alternatives that should be explored and can be developed. I would again urge the Court to address Justice O’Hern’s concern that “[t]he real question for us is not what the State can do, but rather what we should do in the just exercise of our common law supervisory power over criminal practice within our jurisdiction.” Id. at 333-34.
II.
In State v. Gerald, we held as a matter of state-constitutional law that before an accused can be convicted of capital murder, the State must prove beyond a reasonable doubt that the defendant purposely or knowingly intended to cause the victim’s death. 113 N.J. at 69-91. We ruled further that the crime of homicide in which the defendant intended to inflict only serious bodily injury on another — but not death — is not capital murder. Ibid. It is critical, then, that the jury be given a charge that draws the distinction between one who intends the death of another, and is therefore death-eligible, and one who intends to cause serious bodily injury to, but not the death of, the victim. See id. at 91-92.
*404The majority acknowledges that such a charge was not given in this case. Instead, “the charge instructed the jury that the defendant would be guilty of capital murder if he contemplated causing only serious bodily injury.” Ante at 376. This charge was reinforced by the verdict sheet, which “did not require the jury to find that the defendant contemplated causing death, rather than serious bodily injury that results in death,” in order to reach a verdict of guilty of capital murder. Ibid. Nevertheless, the majority concludes that this was “harmless error.” I disagree.
Although “not essential to [its] ... disposition,” the majority finds that “the evidence is overwhelming that defendant knowingly killed” Edward Lawson. Ante at 376. This determination leads the Court to conclude further, albeit tentatively, that the defendant was not entitled to a number of jury charges, including a charge on serious-bodily-injury murder. Thus the Court states, “[o]nly in the most tenuous sense would the evidence have supported a charge of aggravated or reckless manslaughter or of intending to inflict serious bodily injury, but not death, on another.” Ibid. In my view, the Court errs in both its approach to the question of whether the evidence would support a charge for a lesser offense and in its reliance on misleading and irrelevant facts as a basis to conclude that the evidence is overwhelming that defendant knowingly killed his victim.
The Court’s analysis in this case is similar to the tack it took in State v. Rose, supra, 112 N.J. at 479-85. In that case, the defendant was convicted of capital murder and sentenced to death for the fatal shooting of a police officer. On appeal, the defendant alleged that in light of evidence arguably indicating that his actions resulted from panic, the trial court’s refusal to charge the jury on aggravated manslaughter constituted reversible error. A majority of the Court, however, relied on the weight of the evidence that the defendant shot the officer in the stomach from very close range with a sawed-off shotgun, that the shooting was witnessed, and that defendant had confessed *405to it to conclude that defendant’s firing of the shotgun was intentional rather than accidental, and thus found no rational basis for allowing the jury to consider aggravated manslaughter as a possible verdict. Id. at 482-83.
As I pointed out in dissent in Rose, the Court’s approach was indeed novel in that it emphasized the quantum of evidence supporting the greater crime rather than an examination of the record for evidence that would rationally support a jury verdict convicting the defendant of the lesser offense as required by our decision in State v. Crisantos (Arriagas), 102 N.J. 265, 276 (1986). In essence, the Court constructed a new formulation of the law according to which a “rational basis is not established unless a defendant has not only pointed to evidence that could support a conviction on the lesser charge, but also, and quite apart from this, explained why the jury should have credited that evidence by assailing the strength of the evidence of the greater charge.” Rose, supra, 112 N.J. at 560 (Handler, J., dissenting). I noted further that because the degree to which a jury will believe and weigh evidence is inscrutable, the question of what weight to afford evidence that could be read to support a lesser charge had previously been left to the jury. Rational basis had been established in our prior cases by an existence of slightly more than “a scintilla” of evidence of the lesser charge, not by any assessment of its weight as against the evidence supporting conviction of the greater charge. Ibid. I thus concluded in Rose that to the extent the majority’s approach required courts both to hypothesize the weight a jury would have ascribed to a lesser charge and then assess the weight of that evidence as against the weight of the greater charge in order to determine whether a rational basis exists to charge the lesser offense, “it departs from all precedent, imposes an unrealistic standard, and usurps the function of the jury.” Ibid.
We see in this case such radical departure from established precedent, as well as the usurpation of the jury function, in the Court’s inexplicable failure to examine the record for evidence *406sufficient to enable a jury to conclude rationally that the defendant intended to cause serious bodily harm to, but not the death of, his victim. Instead, as it did in Rose, the Court concentrates on the “overwhelming” evidence that the defendant intended to kill Edward Lawson. By conducting a weighing of the evidence more appropriate to a jury’s deliberation rather than the ascertainment of “rational basis” that is required by Crisantos, the Court’s analysis illustrates the serious errors that can occur when the Court usurps the jury function.
The damage of such analysis is particularly evident when it is noted that the facts proffered by the majority as “overwhelming evidence” that defendant intended to kill are equally supportive of an intent to inflict serious bodily injury rather than death. Specifically, the fact that after hearing that Lawson had abused his sister, defendant grabbed a knife and said, “I told this motherfucker about fucking with my sister” and “I know what I got to do,” is no more indicative of an intent to kill than of an intent to inflict serious bodily injury, particularly when “serious bodily injury” is defined for the jury, as it was in this case, as “bodily injury which creates a substantial risk of death.” Moreover, defendant’s post-killing statement that “I killed him. I broke the knife in him,” is totally irrelevant to defendant’s state of mind at the time he committed the homicide; while it does indicate defendant’s recognition of the consequences of his actions, it does not even tend to establish that defendant intended to kill at the time of the attack. It could well have been that defendant, having meant to hurt, even seriously hurt Lawson, was merely describing or expressing surprise at the result of his attack. Finally, the wounds inflicted are no more indicative of an intent to kill than of an intent to commit serious bodily injury that creates a substantial risk of death to the victim and that actually results in death.
In this case, the evidence was sufficient to enable the jury to determine that defendant intended to kill his victim. However, there also existed a rational basis for a jury to conclude that defendant intended to cause only serious bodily injury to, rather *407than the death of, Edward Lawson. There are no “smoking-gun” statements by the defendant in which he expresses an intent to kill the deceased; nor are the wounds inflicted clearly indicative of anything more than that defendant intended to commit grievous bodily injury on his victim. Hence, there was a sufficient basis to charge serious-bodily-injury murder as this Court defined the offense in State v. Gerald, supra, 113 N.J. at 69-92.
Nevertheless, the majority professes that it need not base its decision on the perceived insufficiency of the evidence. It concludes rather that “even if the evidence were sufficient, defendant was not prejudiced by the absence of the charges.” Ante at 377. It observes, simply, that because the defendant is no longer death-eligible due to sentencing errors below, “[t]he net result is that defendant stands convicted of murder, not capital murder.” Ante at 376.
The court’s rationale thus appears to be based on the assumption that the defendant would have benefited from the appropriate charge only to the extent that the jury would have convicted him of non-capital murder, a benefit he has now already secured. But the prejudice is not simply that defendant was deprived of a conviction of ordinary or non-capital murder, prejudice neatly neutralized by rendering him non-death-eligible. The real harm in failing to instruct the jury on serious-bodily-injury murder is that the jury was deprived of the full spectrum of choices on which to base its weighty determination of criminal liability, denying the defendant the opportunity to secure a conviction on a less serious offense, namely, a form of manslaughter.5
*408We have held that at the very core of the guarantee of a fair trial in a criminal case is the judicial obligation to assure the jury’s impartial deliberations based solely on the evidence and in accordance with proper and adequate instructions. State v. Simon, 79 N.J. 191, 206 (1979). When a defendant requests that the jury be instructed on a lesser-included offense, the trial court is obligated, in view of defendant’s interest, to examine the record thoroughly to determine if there exists a rational-basis for a verdict convicting the defendant of the included offense. Crisantos, supra, 102 N.J. at 278. Indeed, “so paramount is the duty to insure a fair trial that a jury must deliberate in accordance with correct instructions even when such instructions are not requested by counsel.” State v. Grunow, 102 N.J. 133, 148 (1986); see State v. Moore, 113 N.J. 239, 288 (1988) (trial court’s failure to charge on diminished capacity constitutes reversible error although charge was never requested by defense counsel). Thus, “the court ordinarily has a supervening responsibility to charge the jury concerning any version of the offense ‘clearly indicate[d]’ by the evidence to require proper consideration.” Grunow, supra, 102 N.J. at 148 (citing State v. Choice, 98 N.J. 295, 299 (1985)); see Ramseur, 106 N.J. at 270-71 n. 62 (“the trial court has a duty ‘to charge the applicable law to the jury based upon facts regardless of what requests counsel may make____’”). This duty stems from the belief “that enforcement of the criminal law is too important to be controlled completely by the contentions of the adversaries; and that the court has an obligation to see to it that the jury, as the representative of the public, is given all of the facts and all of the possible offenses that might reasonably be found from such facts.” Choice, supra, 98 N.J. at 298-99.
*409We have held in an anologous situation that charges unsupported by the evidence should not be given to the jury in order to prevent the manipulation of the deliberative process and unwarranted compromise verdicts. Thus, in State v. Christener, 71 N.J. 55, 69-74 (1976), we held that it is reversible error to instruct the jury on a degree of a crime that is more serious than the evidence will support. We concluded that an instruction on the unsupported charge could subtly influence the jury to infer that the evidence would support the more serious charge and further induce the jury artificially to inflate the weight of the evidence to correspond with the greater crime charged. Ibid. Although in Christener the jury did not ultimately convict the defendant of the more serious, evidentially unsound, crime of first-degree-murder, we acknowledged the very real possibility that without the greater charge the jury could have found defendant not guilty. Id. at 69. The strong, gravitational pull of the unwarranted, first-degree-murder charge may have drawn the jury to a compromise verdict, causing it to settle on the middle ground of manslaughter rather than an acquittal of the defendant. The Christener decision, therefore, stands for the proposition that the jury’s deliberative process may not be manipulated by the intrusion of unsupported charges. This concern soundly concurs with the overriding principle that the jury’s deliberations remain free of extraneous influences and that the jury be fully informed of all significant facts and the full spectrum of charges in order to safeguard the integrity of that deliberative process and the ultimate verdict.
The case before us presents an analgous problem to the one identified in Christener. In Christener, the erroneous instruction on the greater crime of first-degree murder unduly influenced the jury to inflate the value of the evidence and ultimately increase the seriousness of the crime of which it convicted the defendant. So in this case, the failure to charge a less serious offense that is supported by the evidence could subtly induce a jury to believe that the evidence would support the *410more serious offenses; the failure to charge could prevent the jury from deflating the weight of the evidence in order to conform to a lesser charge. What the Court conveniently overlooks here is that the jury might not have so readily rejected the evidence of a form of manslaughter had it been given the appropriate, full range of choices.
It should now be better understood that the lines demarcating certain kinds of homicides are both critically important and extremely fine. The difference between (1) knowing capital murder, i.e., murder committed when one is practically certain of causing death, and (2) knowing serious-bodily-injury murder, i.e., murder committed with practical certainty of causing only serious bodily injury that happens to result in death, and (3) aggravated manslaughter, i.e., a killing committed with a conscious disregard of a substantial risk of death that manifests extreme indifference to the value of human life, is exquisite, and, indeed, may in a given case be so attenuated that a court should rarely if ever withhold from a jury the ultimate responsibility of determining which offense has been committed in light of all of the evidence.6 In my view, the failure of the trial court to allow the jury to resolve the subtle distinctions between knowing capital murder, knowing serious-bodily-injury murder, *411and the various forms of manslaughter constitutes reversible error.
We have said in ordinary criminal cases that “it is speculative to forecast what verdict a jury would have returned if properly instructed on the basis of the verdict that a jury returned after an incomplete instruction.” Crisantos, supra, 102 N.J. at 273. This conclusion stems, in part, from our recognition that a criminal jury has the power to convict on a lesser charge although the evidence establishes the greater charge beyond a reasonable doubt. Id. at 272 (citations omitted), see McCleskey v. Kemp, supra, 481 U.S. at 311, 107 S.Ct. at 1777, 95 L.Ed.2d at 291. Accordingly, the fact that the jury in this case convicted the defendant of capital murder does not necessarily mean that the jury would have convicted the defendant of serious-bodily-injury murder rather than a lessor form of manslaughter if given the full range of available choices. In truth, we do not know what the jury would have done given the appropriate charge, nor should we guess.
I would therefore reverse the defendant’s conviction for capital murder and remand for a re-trial on the charge of murder, including both accomplice murder, which was charged below, and serious-bodily-injury murder. In addition, defendant should be given the full opportunity on remand to assert that the evidence presented would support only a form of manslaughter.
III.
In the penalty phase, the State relied on a solitary aggravating factor, c(4)(c). With respect to this factor the State resorted to the evidence produced during the guilt phase; it did not introduce any additional evidence at the penalty-phase trial. The defendant offered expert testimony to support his contention that he was under extreme mental or emotional disturbance, N.J.S.A. 2C:ll-3c(5)(a), as well as his own testimony and that of members of his family to establish his devotion to them *412and his reliability as a worker under the “catch-all” factor, N.J.S.A. 2C:ll-3c(5)(h). In his summation, the prosecutor argued that the evidence of twenty-four stab wounds alone was sufficient to support the existence of aggravating factor c(4)(c). However, these remarks failed to point out any evidence from which the jury could infer and determine beyond a reasonable doubt that the defendant intended to inflict gratuitous or added pain and suffering other than that which occurred in the purposeful killing of the victim.
At the close of the penalty phase, the jury found the existence of aggravating factor c(4)(c) in the case as well as four mitigating factors; it nevertheless concluded that the solitary aggravating factor was not outweighed by the mitigating factors.
Our narrowing construction of c(4)(c) in Ramseur excised the first part of the statutory definition — that the murder was outrageously or wantonly vile, horrible, or inhuman — and focused exclusively on the second part — that the murder “involved torture, depravity of mind, or an aggravated assault on the victim.” 106 N.J. at 198-99. We also concluded that the terms “torture” and “aggravated battery” were identical in purpose; both of these factors exist if the defendant “intended to, and did in fact, cause extreme physical or mental suffering — in addition to death.” Id. at 208. In other words, “the extreme physical or mental suffering must be precisely what defendant wanted to occur in addition to death.” Id. at 208-09.
The trial court here focused preliminarily on whether the murder was “outrageously or wantonly vile, horrible or inhuman,” rather than on whether the defendant intended to inflict extreme physical or mental pain on Lawson in addition to death. In so doing, the trial court used the definition that we subsequently found unconstitutional in Ramseur. I thus concur in the Court’s determination that the erroneous instruction warrants a reversal of the defendant’s death penalty.
*413The Court, however, does not address defendant’s contention that the State failed to proffer sufficient evidence that the murder of Lawson involved either “torture” or an “aggravated battery.” The Court eschews further consideration of this claim because it rules, for other reasons, that the defendant cannot be again tried for capital murder and exposed to the death penalty. I believe the issue should be addressed. The circumstances of purposeful murder involving multiple wounds and non-instantaneous death are recurrent. See Bienen, Weiner, Denno, Allison & Mills, “The Reimposition of Capital Punishment in New Jersey: The Role of Prosecutorial Discretion,” 41 Rutgers L. Rev. 27,356 (1988) (Study indicating that murder by stabbing is one of the most frequently-employed methods of committing homicide). It behooves the Court to explain whether such evidence is sufficient to enable a jury to conclude beyond a reasonable doubt that the murder is to be regarded as capital murder under c(4)(c).
In my opinion such evidence alone, without additional evidence demonstrating a torturous or sadistic intentional state of mind, cannot suffice as a matter of law to establish the aggravating factor of c(4)(c). We addressed a similar issue in Ramseur. There the defendant, after having stabbed the victim and walked away from her, returned within a couple of minutes to inform the bleeding victim that he was going to kill her grandchildren before inflicting the remaining fatal stab wounds. 106 N.J. at 288. From that conduct, we found that a jury could infer that “Ramseur, in addition to purposely killing the victim, also purposely inflicted severe mental pain prior to her death.” Id. at 288.
By contrast, here the only evidence is that Lawson was stabbed twenty-four times, was shocked by the attack, and bled for twenty minutes before dying. The State relies solely on the multiple stab wounds, which the prosecutor described on summation as “24 aggravated batteries,” to establish aggravating factor c(4)(c). Nothing, however, suggests that defendant intended anything but to kill Lawson; there is no evidence that *414defendant intended to torture Lawson or needlessly to prolong his death in order to add to his pain and suffering or to inflict unnecessary or gratuitous severe pain in addition to death. Although those wounds doubtless caused the victim to suffer extreme psychological and physical pain, nothing inferable from the manner in which the wounds were inflicted or the nature of the wounds themselves or the defendant’s state of mind indicates that defendant intended to inflict any pain except that which occurs inevitably when murder is accomplished by stabbing and death is not instantaneous or sudden.
If the c(4)(c) factor could be sustained in this case, where the only evidence is several non-fatal wounds and death was not instantaneous, there would be “no principled way to distinguish this case, in which the death penalty was imposed, from the many cases in which it was not.” Godfrey v. Georgia, 446 U.S. 420, 433, 100 S.Ct. 1759, 1767, 64 L.Ed.2d 398, 409 (1980). We said as much in Ramseur.
[T]he essence of the Legislative concern is the defendant’s state of mind. We do not believe that the Legislature intended to distinguish between two murderers each of whom intended to inflict immediate death upon the victim without any additional suffering whatsoever, when one victim dies immediately and the other lives for a long period of time and experiences excruciating pain. That capricious event alone would be perceived as an insufficient basis on which to inflict death on that defendant while imposing imprisonment on the other. [106 N.J. at 207.]
As reprehensible as it is, a multiple stabbing in itself, without anything more, is insufficient to sustain a finding beyond a reasonable doubt that a defendant intended to inflict, in addition to death, extreme physical or mental pain on his victim.
I would therefore hold that there was insufficient evidence to support submission of factor c(4)(c) to the jury. See Rose, supra, 112 N.J. at 531 (evidence was insufficient to support the submission of factor e(4)(c) to the jury because “[ajside from evidence that defendant had fired the shotgun and was knowledgeable about its capacity to inflict devastating injury, there was no proof that defendant’s intention was to cause [the victim] pain and suffering, rather than to kill him.”).
*415IV.
In conclusion, I find the continued use of death-qualified juries to try the issue of criminal guilt in a capital trial to be a fatal infirmity in our capital-murder statutory scheme. The Court at the very least should conscientiously undertake to modify this procedure and to explore alternatives. Further, I find that reversible error was committed in this case in the failure of the trial court to charge serious-bodily-injury murder as defined by State v. Gerald. The defendant’s murder conviction, in my opinion, should therefore be reversed. Finally, I believe the Court should determine in this case that the evidence was insufficient to establish beyond a reasonable doubt the aggravating factor of c(4)(c).
Therefore, I write separately to express these reasons and to record my dissent.
For affirmance in part; reversal in part; remandment —Chief Justice WILENTZ and Justices POLLOCK, CLIFFORD, O’HERN, GARIBALDI, and STEIN — 6.
For reversal — Justice HANDLER — 1.
That case involved a joint trial of Bruton and a codefendant at which the government introduced into evidence the oral confession of the nontestifying codefendant implicating Bruton. The Supreme Court held that because Bruton could not effectively cross-examine the confession of the nontestifying codefendant, its admission violated Bruton's sixth amendment right of confrontation.
The difficulty that arises in a joint trial in these cases is that the State has effectively pitted one defendant against the other with each trying to save himself at the other’s expense; the result is that the prosecutor is relieved of his burden of proof. People v. Hurst, 396 Mich. 1, 7, 238 N.W.2d 6, 10 (1976). The multiple-jury procedure resolves this inequity because each defendant’s jury hears none of the evidence presented by his codefendant.
Some State courts have discouraged the use of the multiple-jury procedure. See, e.g., State v. Lambright, 138 Ariz. 63, 673 P.2d 1 (1983), cert. den., 469 U.S. *398892, 105 S.Ct. 267, 83 L.Ed.2d 203 (1984); People v. Williams, 93 Ill.2d 309, 67 Ill.Dec. 97, 444 N.E.2d 136 (1982), cert. den., 467 U.S. 1218, 104 S.Ct. 2666, 81 L.Ed.2d 371 (1984); Scarborough v. State, 50 Md.App. 276, 278-81, 437 A.2d 672, 674-76 (1981). However, the primary reason for such disapproval appears to stem from the risk that either counsel or the court would unintentionally disclose inadmissible information to the "wrong" jury, which could constitute reversible error. See, e.g., People v. Rainge, 112 Ill.App.3d 396, 68 Ill.Dec. 87, 92, 445 N.E.2d 535, 550 (1983). This, however, is not a problem in the context of a capital-murder prosecution involving one defendant.
In Nevada, the jury is given responsibility for imposing the sentence in a capital case, but if the jury cannot agree, a panel of three judges may impose the sentence. Nev.Rev.Stat. § 175.556 (1987). In Arizona, Idaho, Montana, and Nebraska, the court alone imposes the sentence. Ariz.Rev.Stat.Ann. § 13-703 (Supp.1987); Idaho Code § 19-2515 (1987); Mont.Code Ann. § 46-18-301 (1987); Neb.Rev.Stat. § 29-2522 (1985). Alabama, Florida, and Indiana actually allow a judge to override a jury's recommendation of a life sentence. Ala.Code § 13A-5-46 (1988); Fla.Stat. § 921.141(3) (1985); Ind. Code § 35-50-2-9(e)(2) (Supp.1985).
The defendant sought to have aggravated manslaughter charged in the case but the trial court denied this request, permitting only passion/provocation manslaughter to be charged. The majority appears overly concerned with defense counsel’s inarticulate request for an aggravated manslaughter charge as well as with other procedural errors. See ante at 375. Although such concern would be implicated in ordinary cases where "plain error" analysis is *408appropriate, such a tack is inconsistent with this Court’s "meticulous and comprehensive procedure for reviewing capital cases” and its established belief that "in death penalty cases an appellate court must subject the record to intense scrutiny. The stark fact that a litigant’s life is at stake intensifies the obligation of judicial review." Bey (I), 112 N.J. 45, 93 (1988).
Even the Commentary to the proposed Code revision of 1971, which introduced the concepts ultimately codified as "knowing” murder and "aggravated manslaughter,” states that the demarcating line between the two is so fine that whether one charge is implicated over the other in a particular case must be left to the jury’s determination. Rose, supra, 112 N.J. at 563 (Handler, J., dissenting) (quoting New Jersey Penal Code, Volume, II: Commentary, Final Report of the New Jersey Criminal Law Revision Commission, 1971).
It is precisely this blending of knowing murder and aggravated manslaughter that led me to conclude in Gerald, as a matter of state constitutional law, that knowing murder should not be included in the class of death-eligible murders because of the intolerable risk that reckless murder will also becme capital murder. 113 N.J. at 150-53 (Handler, J., dissenting). At the very least, since the two are so similar in definition and mode of proof yet so drastically different in consequence, the jury should be given the opportunity to fit the evidence to the appropriate crime and intelligently choose between them.