concurring in part and dissenting in part.
Defendant was convicted of third-degree possession of cocaine, third-degree possession of cocaine with intent to distribute, and fourth-degree hindering apprehension or prosecution, as well as a disorderly persons offense. In 1987, a year prior to the events that gave rise to the instant charges against him, defendant had pled guilty to possession of cocaine, possession of cocaine with intent to distribute, and theft. Ante at 381, 625 A.2d at 1087.
The trial court, relying on State v. Sands, 76 N.J. 127, 386 A.2d 378 (1978), ruled that the prosecution would be allowed to use defendant’s prior convictions to impeach defendant on cross-examination. Paced with the likelihood that if he exercised his right to testify on his own behalf the jury would hear evidence that he had committed identical crimes just one year earlier, defendant declined to testify.
Today, the Court reverses defendant’s convictions, finding that the trial court erred in agreeing to admit unsanitized evidence of defendant’s prior convictions for crimes similar to those for which defendant was being tried. Ante at 392-393, 625 A.2d at 1092-1093. The Court establishes a limited sanitization rule applicable to prior convictions that are similar to the crime for which defendant is being tried, which will specify only the “date, degree, and number of [a defendant’s] prior similar convictions.” Ante at 392, 625 A.2d at 1093.
I write separately to express my view of the standards for sanitization, which differ somewhat from the rule announced by the Court. In so doing, I reiterate the analysis I proposed in State v. Pennington, 119 N.J. 547, 604-09, 575 A.2d 816 (1990) (Handler, J., concurring in part and dissenting in part), in support of the “Connecticut Rule,” in which all prior crimes, not *396directly reflecting on credibility, are sanitized before admission into evidence for the purpose of impeachment.
I
The relevance of prior crimes to the credibility and competence of witnesses has developed in New Jersey law for almost two centuries. Sands, supra, 76 N.J. 127, 386 A.2d 378. Under the common law, criminal defendants were not competent to testify on their own behalf. Id. at 139, 386 A.2d 378. Under a statute enacted by the Legislature in 1799, and in effect until 1871, no person convicted of any one of a number of specified crimes could testify as a witness “unless he or she be first pardoned * * *.” State v. Hawthorne, 49 N.J. 130, 133, 228 A.2d 682 (1967) (overruled by Sands, supra, 76 N.J. at 127, 386 A.2d 378). The list of crimes provided in the 1799 statute, however, did not include many crimes of “great turpitude,” some of which had direct bearing on the veracity of a witness, including counterfeiting, bribery, and extortion by judges. Sands, supra, 76 N.J. at 134, 386 A.2d 378 (citing State v. Henson, 66 N.J.L. 601, 605-06, 50 A. 468 (E. & A.1901)).
The Legislature, in 1871, modified the that rule to allow previously-convicted defendants to testify on their own behalf. Ibid. Three years later, in 1874, previously convicted persons were allowed to testify on another’s behalf, provided that the opposing party could, for the purpose of impeachment, introduce evidence of the witness’ prior convictions. Id., 76 N.J. at 135, 386 A.2d 378. As the Court noted in Sands, however, the 1874 act — although limiting the disability to testify imposed on people with previous convictions — also vested “a general power in the prosecutor to use any prior conviction to impeach a witness.” Ibid, (citing Hawthorne, supra, 49 N.J. at 133-35, 228 A.2d 682).
That rule is currently codified at N.J.S.A. 2A:81-12. The rule, however, is phrased permissively, i.e., prior convictions “may be shown * * (emphasis added). Accordingly, con*397troversy arose over whether trial courts possessed the discretion to prevent admission of prior-convictions evidence on the basis of undue prejudice to the defendant. This Court, in 1967, addressed that issue in Hawthorne, supra, 49 N.J. at 130, 228 A.2d 682. There, the Court found that trial courts did not possess the “discretion to admit or exclude evidence of a defendant’s previous convictions of a crime.” Id. at 134-35, 228 A.2d 682.
Eleven years later, the effect of the rule was ameliorated when the Court overturned Hawthorne in Sands, supra, 76 N.J. at 127, 386 A.2d 378. The Court held, in Sands, that trial courts were empowered to decide whether evidence of prior convictions would be admitted for the purpose of impeachment. Id., 76 N.J. at 144, 386 A.2d 378. The general standard governing the determination of admissibility was the “remoteness” of the prior convictions to the crime charged. Although “[ojrdinarily evidence of prior convictions should be admitted,” ibid., trial courts were obliged to “balance the lapse of time and the nature of the crime to determine whether the relevance with respect to credibility outweighs-the prejudicial effect to the defendant.” Id. at 144-45, 386 A.2d 378. Accordingly, Sands may be seen as bringing Hawthorne’s sweeping and unconditional interpretation of the language of N.J.S.A. 2A:81-12 under the restraining influence of Evidence Rule 4. Accord State v. Whitehead, 104 N.J. 353, 358, 517 A.2d 373 (1986) (holding “[i]n New Jersey, the admissibility of prior convictions for impeachment purposes is governed by the interplay between N.J.S.A. 2A:81-12 and Evidence Rule 4”). That rule provides, inter alia, that where the probative value of evidence “is substantially outweighed by the risk that its admission will * * * create substantial danger of undue prejudice,” the trial court may, in its discretion, exclude the evidence. Evid.R. 4. See Sands, supra, 76 N.J. at 144, 386 A.2d 378.
The Court, in its decision today, recognizes that the rule authorizing impeachment based on prior convictions implicates virtually the same concerns that arise from the rules governing *398-the use of other-crimes evidence. Drawing on its prior analysis of Evidence Rule 55, the Court notes that “other crimes evidence may be simultaneously probative and prejudicial,” ante at 384, 625 A.2d at 1089, and that other-crimes evidence has a unique tendency to turn a jury against the defendant. State v. Stevens, 115 N.J. 289, 302, 558 A.2d 833 (1989); ante at 384, 625 A.2d at 1089. As the Court stated in Stevens:
It is this inflammatory characteristic of other-crime evidence that mandates a careful and pragmatic evaluation by trial courts, based on the specific context in which the evidence is offered, to determine whether the probative worth of the evidence outweighs its potential for undue prejudice.
[115 N.J. at 303, 558 A.2d 833.]
The uniquely prejudicial nature of other-crimes evidence, the Court observes, may also manifest itself when such evidence, in the form of prior convictions, is introduced for the purpose of impeachment. Ante at 384, 625 A.2d at 1089. A concern over balancing the probative worth of prior-crimes evidence against its prejudicial effect was clearly involved in the impeachment rule formulated by the Court in Sands. Nevertheless, that rule did not provide an effective mechanism for achieving the “careful and pragmatic” evaluation of evidence of prior crimes by trial courts. The exclusive reliance of the Sands rule on a “remoteness” standard failed to foreclose the admission of prior-crimes’ evidence that generated undue prejudice to defendants. Today the Court confronts the highly prejudicial effect of prior-crimes evidence and attempts to rectify the patent shortcomings of the existing rule and practices. Ante at 391, 625 A.2d at 1092.
II
Faced with the competing goals of protecting criminal defendants from undue prejudice and respecting the State’s legitimate interest in “impeach[ing] the credibility of a criminal defendant who previously has been convicted of a similar crime,” ante at 391, 625 A.2d at 1092, the Court acknowledges that the Sands approach is inadequate and that only the saniti*399zation of similar prior offenses can be effective. I fully concur in the reasons that bring the Court to that conclusion. Three years ago in Pennington, I expressed the view that sanitization would provide a means by which “a defendant could be impeached without being subjected to the extraordinary prejudice that follows if the prior crime was specifically named or described.” 119 N.J. at 607, 575 A.2d 816 (Handler, J., dissenting).
The dissenting opinion argues that sanitization is inconsistent with Sands and the statute, N.J.S.A. 2A:81-12, as well as the proposed Evidence Rule 609. Post at 409, 625 A.2d at 1101. I would maintain that that criticism is unfounded and that, in recognizing the need for imposing some form of sanitization of prior convictions for impeachment purposes, the Court has effectively answered the criticism of the dissenting opinion. Moreover, I would argue that the criticism of the dissenting opinion begs completely the central question before the Court: does assuring a fair trial for criminal defendants require the revision of the rule of Sands, the limitation of N.J.S.A. 2A:81-12 by the imperative of Evidence Rule 4, and the prospective modification of proposed Evidence Rule 609, which “follows the existing rule admitting any prior conviction”? Post at 409, 625 A.2d at 1101 (emphasis added).
Interestingly, while admitting that courts, in the admission of prior-crimes evidence, should be cognizant of the “danger of undue prejudice,” post at 409, 625 A.2d at 1101, the dissent ignores entirely the ever-growing body of empirical evidence and scholarly consensus that juries use prior-crimes evidence not to assess credibility but to infer guilt based on bad character. See ante at 385-387, 625 A.2d at 1089-1090. Because the dissent offers no argument or evidence to refute the Court’s contention that unsanitized prior crimes — particularly when those crimes are similar to the offense for which the defendant is being tried — have an unduly prejudicial effect on juries, I take it that the dissent accepts that contention. Consequently, the insistence of the dissent on uncritical adherence to the rule *400of Sands, which provides minimal constraints on the sweeping language of N.J.S.A. 2A:81-12, is all the more perplexing.
Although the Legislature may, where a rational basis exists, establish a statutory presumption of the probative worth of certain types of evidence, State v. Sinclair, 57 N.J. 56, 64, 269 A.2d 161 (1970), it may not relieve trial courts of their duty to exclude evidence the prejudicial effect of which seriously compromises a defendant’s right to a fair trial. Accordingly, I disagree with the dissent and concur in the Court’s decision to overrule its contrary holdings in Sands and Pennington.
Ill
The Court establishes a limited-sanitization rule. It would require sanitization of only those prior convictions that are similar to the offense for which the defendant is being tried. Ante at 391, 625 A.2d at 1092. Dissimilar prior convictions could still be admitted unsanitized. For cases in which multiple prior convictions existed — some similar, some dissimilar — the State would have the option of admitting all the prior convictions in a sanitized form, or excluding the similar prior convictions while admitting the dissimilar prior convictions in an unsanitized form. Ibid.
The Court’s rule, in my opinion, is much too limited and will not, I submit, overcome the almost ineradicable undue prejudice that inheres in unsanitized prior-crimes evidence. My differences with the Court’s decision are reflected to some degree in my disagreement with its interpretation of the Pennington case. My understanding of Pennington and my reading of the studies cited by the Court on behalf of the rule it enunciates today lead me to take issue with the mechanism of sanitization that the Court has established, ante at 391, 393, 625 A.2d at 1092, 1093.
In Pennington, the trial court permitted the admission of evidence of the defendant’s non vult plea to a first degree murder charge. 119 N.J. at 604, 575 A.2d 816. The defendant *401had received second-degree sentences for that offense. Ibid. The defendant invited this Court to adopt the “Connecticut Rule,” which provided for the sanitization of all prior convictions that do not bear directly on veracity. The Court declined that invitation, because it “would mark a pronounced departure from the Sands rule, which has served well for twelve years.” Id. at 586, 575 A.2d 816.
I dissented from that part of the Court’s holding in Pennington because I believed that, particularly in the context of a capital-murder prosecution, the effect on the jury of admitting evidence of a prior murder conviction would be so inflammatory as to deny the defendant a fair trial. Id. at 606, 575 A.2d 816 (Handler, J., dissenting). At that time I believed that limiting instructions simply would not avail to undo the extreme prejudice caused by admission of a prior murder conviction. Ibid. Additionally, I was convinced that the refusal of the trial court to sanitize the defendant’s prior murder conviction deprived the defendant of due process of law by unduly burdening the defendant’s right to testify on his own behalf. Id. at 608-09, 575 A.2d 816.
Intent on minimizing the inconsistency between its decision today and its determination of Pennington, ante at 388-390, 625 A.2d at 1091-1092, the Court eschews a broader form of sanitization. The Court does so apparently because it believes that the introduction of Pennington’s prior murder conviction “would not have been influential on the question of Pennington’s responsibility for the victim’s death.” Ante at 390, 625 A.2d at 1092.
In its description of the Pennington case, the Court states that “[bjecause Pennington admitted the robbery and the shooting of the victim in the course of the robbery, Pennington’s responsibility for the victim’s death was not disputed.” Ante at 389, 625 A.2d at 1091. By that statement, the Court apparently means that no real question existed about the fact that Pennington had caused the victim’s death. That says nothing, *402however, about the impact of the prior murder conviction on the critical issue in contention at Pennington’s trial, namely, his state of mind.
The Court itself acknowledges that “the issue to be resolved by the jury was the specific form of homicide that Pennington had committed, which depended largely on the jury’s conclusions about his state-of-mind.” That acknowledgment cannot be reconciled with the Court’s subsequent assertion that the introduction of Pennington’s prior murder conviction “might have had only a marginal impact on the jury’s assessment of Pennington’s credibility with respect to his state-of-mind.” Ante at 392, 625 A.2d at 1093. In Pennington, to paraphrase a statement quoted by the Court, the difference between lack of credibility as a repetitive felon and lack of credibility as one who had murdered before, “ ‘was negligible to the prosecution, catastrophic to the accused.’ ” Ante at 392, 625 A.2d at 1092 (quoting Bendelow v. United States, 418 F.2d 42, 53 (5th Cir.1969) (Godbold, J., concurring in part, dissenting in part), cert, denied, 400 U.S. 967, 91 S.Ct. 379, 27 L.Ed.M 387 (1970)).
The defendant in Pennington was charged with capital murder, which then required that the State prove the defendant “purposely or knowingly” had caused the death of his victim. State v. Gerald, 113 N.J. 40, 49, 549 A.2d 792 (1988). The defendant, although granting that he had shot the victim, disputed that he had intended to cause the victim’s death. As the Court acknowledges, and correctly observes, “The critical issue at trial was Pennington’s mental state.” Ante at 389, 625 A.2d at 1091.
The central task of the jury in Pennington was to determine whether the defendant had intentionally committed murder. The inference is compelling that the defendant, in choosing not to testify on his own behalf during the guilt phase of the trial, believed that the admission of evidence of his prior murder conviction would so inflame the jury that a conviction for intentional murder was virtually assured. That conclusion *403seems inescapable and is in fact well supported by the studies that the Court marshals for its holding today. Ante at 386-387, 625 A.2d at 1090-1091.
The prejudicial impact of a prior conviction for murder introduced only for impeachment is distinctive and can be devastating. In that regard, I would observe that although prior convictions for similar crimes have the strongest prejudicial effect on jurors, the prejudicial impact of prior convictions for murder is only slightly less. The Court cites the Wissler and Saks study for the proposition that “the highest conviction rate resulted] when the prior crime was the same as the charged offense,” ante at 386-387, 625 A.2d at 1090. The study demonstrates that when compared to a prosecution in which no prior-crimes evidence was introduced, in cases in which the jury heard impeachment evidence of prior convictions for the same crime, the conviction rate rose dramatically by an average of 75%. Roselle L. Wissler & Michael J. Saks, On the Inefficacy of Limiting Instructions, 9 J.L. Hum. Behavior 37, 43 (1985) (See Table 2 “Per cent of Juries Voting to Convict”). When the prior crime was different from that charged, the increase in conviction rate was lower. Nevertheless, when murder was introduced as the dissimilar prior crime, the conviction rate increased to 70%. Ibid.
The defendant in Pennington was faced, then, with the worst of all possible worlds: a prior conviction that was both similar to the crime for which he was being tried and was also for murder. I am still convinced of what I wrote in dissent in Pennington: “It is naive to believe that a jury struggling to decide whether [the] defendant intentionally committed murder would not be influenced by the knowledge that [the] defendant had killed before, or realistically would consider the prior murder as indicative only of defendant’s lack of truthfulness.” 119 N.J. at 606, 575 A.2d 816.
The Court, in its reading of Pennington, ante at 392, 625 A.2d at 1093, seems to be making two assertions: (1) because *404the fact that Pennington was the causal agent of the victim’s death was virtually undisputed, the evidence of his prior murder conviction could not have had a measurable impact on the jury’s determination that Pennington had caused (i.e., “was responsible for”) the victim’s death; and (2) in the light of that evidence, the introduction of Pennington’s prior murder had no significant impact on the jury’s determination of Pennington’s state of mind.
Two comments must be made about the second assertion: it does not follow from the first assertion and it is contradicted by the studies the majority cites in support of its holding today. The point at issue in Pennington was not the defendant’s responsibility as the cause of the victim’s death. The point at issue was the defendant’s state of mind when he took the actions that caused that death. The Court’s assertion, that evidence of Pennington’s prior murder conviction “might have had only a marginal impact on the jury’s assessment” of his credibility on his state-of-mind at the time he shot his victim, is wrong. Ante at 392, 625 A.2d at 1093. Even more troublesome and less defensible is the apodictic certitude of the Court’s belief that the admission of Pennington’s prior murder did not justify the reversal of his capital murder conviction.
Finally, the premise on which the majority’s treatment of Pennington rests — that the prior conviction did not have a determinative impact on the jury’s resolution of the state-of-mind question — is contrary to common sense, scholarly consensus, empirical evidence, and the rationale of the Court’s own holding.
It seems clear to me that the Court can find no refuge in Pennington. Its ineluctable lesson is that defendants confronted with prior convictions for identical crimes are most at peril from the misuse of that evidence. Defendants placed in the same position as Pennington are prejudiced, and that “prejudice is self-evident, inescapable, and destructive.” Ibid. Pennington also illustrates the obvious point that when prior *405crimes involve egregious charges or circumstances, their prejudicial potential and capacity for misuse become extreme. That analysis and understanding of Pennington demonstrate, I believe, the need for a broad and effective sanitization rule in dealing with prior-crimes convictions as a basis for impeaching a criminal defendant.
IV
I am not convinced that the Court’s mechanism is adequate for ameliorating the undue prejudice to defendants that accrues from the admission of prior convictions. As earlier noted, certain crimes, like murder, have an undue prejudicial effect nearly equal to that of similar prior convictions. Supra at 385, 625 A.2d at 1089. As one of the authors cited by the Court has observed, “The risk of prejudice is acute when the impeaching conviction is for a particularly reprehensible offense, such as murder or sexual assault.” James H. Gold, Sanitizing Prior Conviction Impeachment Evidence to Reduce Its Prejudicial Effects, 27 Ariz.L.Rev. 691, 697 (1985) (cited ante at 385, 625 A. 2d at 1089). The author ultimately concludes:
There is also no sound reason for employing sanitization only when the impeaching crime is similar to the charged offense. * * * Although the need for sanitization may be greatest in this situation, sanitization also alleviates the risk of prejudice which results when the defendant is impeached with a dissimilar, but serious prior offense.
{Id. at 703.]
Accordingly, I would argue for the rule I proposed in Pennington: sanitization of all prior crimes not directly reflecting on credibility. 119 N.J. at 607, 575 A.2d 816.
This Court, in the context of Evidence Rule 55, offered the guidance that “[i]n weighing the probative worth of other-crime evidence, a court should consider not only its relevance but whether its proffered use in the case can adequately be served by other evidence.” Stevens, supra, 115 N.J. at 303, 558 A.2d 833. In the case of prior convictions for impeachment purposes, I would contend that “[t]he introduction of a sanitized prior conviction carries enough of a negative connotation to *406effectively impeach a defendant, and this effectiveness is unaffected by the degree of similarity between the impeaching crime and the charged offense.” Gold, supra, 27 Ariz.L.Rev. at 701.
Sanitizing all prior convictions that do not reflect directly on veracity, the so-called “non-crimen falsi” crimes, will better protect a defendant’s right to a fair trial without significantly diminishing the State’s interest in impeachment by use of prior convictions.
The use of prior convictions for impeachment purposes is founded on the belief that “ ‘one may place more credence in the testimony of one who has lived within the rules of society and the discipline of the law than in that of one who has so demonstrated antisocial tendency as to be involved in and convicted of serious crime.’ ” State v. Sinclair, 57 N.J. 56, 64, 269 A.2d 161 (1970) (quoting State v. Harless, 23 Utah 2d 128, 130, 459 P.2d 210 (1969)). The intuitive appeal of that idea is difficult to deny. Perhaps that intuitive appeal comes from the fact that “[i]t is much easier to conclude that a person is bad than that he did something bad.” State v. Allen, 429 Mich. 558, 420 N.W.2d 499, 504 (1988).'
Juries in our system of criminal justice, however, are not charged with determining a defendant’s guilt based on the defendant’s propensity to commit crime. In fact, they are expressly prohibited from doing so. Nevertheless, the overwhelming consensus of empirical studies on the use of prior-crimes evidence for impeachment purposes confirms a point made by one of the researchers in Kalvin and Zeisel’s landmark study, The American Jury: that juries exhibit an
“almost universal inability and/or unwillingness either to understand or follow the court’s instructions on the use of defendant’s prior criminal record for impeachment purposes. The jurors almost universally used defendant’s record to conclude that he was a bad man and hence was more likely than not guilty of the crime for which he was then standing trial.”
[Id. at 508 (quoting a letter from Dale W. Broeder, Associate Professor, University of Nebraska College of Law, who conducted intensive jury inter*407views for The American Jury study, to Yale Law Journal, dated March 14, I960).]
Today’s decision by the Court makes a significant advance toward eliminating the undue prejudice that results from the admission of prior convictions. Regrettably, I do not believe that the mechanism for remedying that prejudice, which the Court proposes today, provides sufficient protection for defendant’s rights.
By way of an aside — albeit one extremely important and necessary to express — I would remark on the still more disturbing fact that the Court has come to an awareness of the devastating impact of similar prior-crimes evidence on the fairness of a criminal trial in a non-capital case, when it was resolutely oblivious to that awareness in a capital case. When read in the light of Pennington, the Court’s decision today is further confirmation of a proposition for which I have argued before: that “[although the Court professes to grant capital-murder defendants enhanced protections under the Constitution and statute, it fails to deliver such protections.” State v. Marshall, 123 N.J. 1, 256, 586 A.2d 85 (1991) (Handler, J., dissenting). It is as if the Court has, however indeliberately, given the phrase “death is different,” Gregg v. Georgia, 428 U.S. 153, 188, 96 S.Ct. 2909, 2932, 49 L.Ed.2d 859, 883 (1976); Furman v. Georgia, 408 U.S. 238, 306, 92 S.Ct. 2726, 2760, 33 L.Ed.2d 346, 388 (1972) (Stewart, J., concurring); State v. Ramseur, 106 N.J. 123, 524 A.2d 188 (1987); State v. Williams, 93 N.J. 39, 459 A.2d 641 (1983), an inverted meaning.
Nevertheless, although I believe the decision of the Court today further exposes infirmities in the Court’s capital-punishment jurisprudence, I commend the Court for its willingness to repudiate aspects of that jurisprudence in a non-capital context and to enunciate a rule that promises — even with the deficiencies I have identified — to provide greater fairness in the trial of all criminal eases.
*408V
For those reasons, I concur in part and dissent in part from the judgment of the Court.