dissenting.
This case presents questions of public concern. Following protracted proceedings, the Parole Board concluded that Trantino’s release from prison would pose an unjustifiable risk of danger to the public. In vacating this determination, the Appellate Division decided that the evidence did not warrant that conclusion. The issue is now before us. The stakes are high. If the Appellate Division is correct, Trantino deserves his freedom and is entitled to parole. But if the Parole Board is right, the consequences of Trantino’s release will be felt, not by some pain-free public entity, nor by some penitent psychologist, social worker or public official, but by tomorrow’s next victim for whose protection and welfare we *199hold office. In my view, the Parole Board’s denial of release is supported by substantial credible evidence in the record. I would not wager the safety of the public on the odds that Trantino is a changed man.
I.
The questions presented can be articulated with disarming ease. We are to determine whether the Parole Board: (1) followed express and implied legislative policies, (2) based its decision on substantial credible evidence present in the record, and (3) reached a reasonable conclusion grounded in the relevant facts. Trantino v. New Jersey State Parole Bd., 154 N.J. 19, 24, 711 A.2d 260 (1998). Because of the meandering course this case has taken, our earlier decisions defined with precision the correct legal standard for determining parole in this case. The parole determination hinges upon whether there is a substantial likelihood Trantino will commit another crime if released. Id. at 39, 711 A.2d 260.
Our inquiry focuses upon whether the factual findings made by the Parole Board could reasonably have been reached on sufficient credible evidence in the record. Id. at 24, 711 A.2d 260 (citing State Parole Bd. v. Cestari, 224 N.J.Super. 534, 547, 540 A.2d 1334 (App.Div.), certif. denied, 111 N.J. 649, 546 A.2d 558 (1988)). The principles that guide us are well settled. We are not to weigh the evidence anew, but instead merely determine whether the record supports the conclusion reached by the Parole Board. State v. Johnson, 42 N.J. 146, 164, 199 A.2d 809 (1964). We must review the record in light of the contentions advanced, but not initially from the point of view of how we would decide the matter if we were the administrative agency. Ibid. That is not to say that we are mechanistically to rubber-stamp the actions of administrative tribunals. We must nonetheless remember that the Parole Board is vitally concerned with recidivism and is expressly charged by the Legislature to withhold release when there is a substantial likelihood the inmate will commit new offenses if paroled.
*200This limited scope of review is grounded in strong public policy concerns and practical realities. First, we recognize that our commitment to fairness is shared with the other branches of government. The judicial branch should be the quiet, but not necessarily the quiescent, branch. We should defer to the expertise of the administrative agency charged with the responsibility of resolving these questions, and intervene only if we are thoroughly convinced that a miscarriage of justice has occurred. We must remember that appellate judges have no monopoly on justice.
Second, the Parole Board is better situated than we to determine factual issues. We traditionally review cases based on the transcripts, documents and briefs submitted. The voluminous records do not always accurately depict what occurred below. For example, a person’s manner may negate a barb his printed word seems to hold, just as it may supply a sting the transcript will not show. We are thus required to accord deference to the findings of the administrative agency that are substantially influenced by its opportunity to hear and see the witnesses and to have the “feel of the case,” an opportunity which a reviewing court cannot enjoy. Id. at 161, 199 A.2d 809.
These principles have universally been applied in our review of the factual findings of trial judges, see, e.g., State v. Locurto, 157 N.J. 463, 474, 724 A.2d 234 (1999); State v. Barone, 147 N.J. 599, 615, 689 A.2d 132 (1997); Meshinsky v. Nichols Yacht Sales, 110 N.J. 464, 475, 541 A.2d 1063 (1988); Rova Farms Resort v. Investors Ins. Co., 65 N.J. 474, 483-84, 323 A.2d 495 (1974); State v. Johnson, 42 N.J. at 161-62, 199 A.2d 809, and administrative agencies, see, e.g., In re Taylor, 158 N.J. 644, 655, 731 A.2d 35 (1999); Brady v. Board of Review, 152 N.J. 197, 210, 704 A.2d 547 (1997); Clowes v. Terminix Int'l, Inc., 109 N.J. 575, 587, 538 A.2d 794 (1988); Goodman v. London Metals Exch., Inc., 86 N.J. 19, 28-29, 429 A.2d 341 (1981); Close v. Kordulak Bros., 44 N.J. 589, 599, 210 A.2d 753 (1965). They have particular efficacy in the context of our review of the Parole Board’s actions.
*201The decision of a parole board involves “discretionary assessment[s] of a multiplicity of imponderables entailing primarily what a man is and what he may become rather than simply what he has done.” Greenholtz v. Inmates of Nebraska Penal and Corr. Complex, 442 U.S. 1, 10, 99 S.Ct. 2100, 2105, 60 L.Ed.2d 668, 677 (1979) (citing Radish, The Advocate and the Expert — Counsel in the Peno-Correctional Process, 45 Minn. L.Rev. 803, 813 (1961)). New certainties exist. The decision differs from the “traditional mold of judicial decisionmaking in that the choice involves a synthesis of record facts and personal observation filtered through the experience of the decisionmaker and leading to a predictive judgment as to what is best both for the individual inmate and for the community.” Id. at 8, 99 S.Ct. at 2104, 60 L.EdZd 668. Parole decisions often involve no more than a value judgment upon a factual complex rather than an evident application of a precise rule of law. This value judgment inevitably reflects the seasoning and the experience of the one who judges.
Stripped to its essentials, a parole board’s decision concerns a prediction as to an inmate’s future behavior, a prognostication necessarily fraught with subjectivity. Puchalski v. New Jersey State Parole Board, 104 N.J.Super. 294, 300, 250 A.2d 19 (App. Div.1969). To a greater degree than is the case with other administrative agencies, the Parole Board’s decision-making function involves individualized discretionary appraisals. Beckworth v. New Jersey State Parole Board, 62 N.J. 348, 358-59, 301 A.2d 727 (1973). In reviewing the Parole Board’s determinations, we must take care not to arrogate to ourselves a power expressly granted to that administrative agency. We may overturn the Parole Board’s decisions only if they are arbitrary and capricious. “Arbitrary and capricious action of administrative bodies means willful and unreasoning action, without consideration and in disregard of circumstances.” Worthington v. Fauver, 88 N.J. 183, 204-05, 440 A.2d 1128 (1982) (quoting Bayshore Sewerage Co. v. Department of Environ. Protection, 122 N.J.Super. 184, 199, 299 A.2d 751 (Ch.Div.1973)). “Where there is room for two opinions,” administrative agency decisions must be considered valid “when exercised *202honestly and upon due consideration, even though it may be believed that an erroneous conclusion has been reached.” Ibid.
II.
Against this backdrop, I find ample credible evidence in the record supporting the Parole Board’s conclusion that there is a substantial risk Trantino will commit another crime if released. This evidence consists of the opinions of the experts and corroborative factual data.
A. Trantino’s Psychological Profile
The Parole Board heavily relied upon the testimony of its Chief Psychologist, Dr. Glen Ferguson, in finding that Trantino posed a substantial risk of recidivism. Both the majority and the Appellate Division discount Ferguson’s ultimate conclusion that Trantino should be denied parole. While it is true that Ferguson underwent a metamorphosis in his opinion respecting Trantino, I am satisfied that the Parole Board acted reasonably in accepting the witness’s ultimate conclusion that release of the inmate would pose an unjustifiable risk to the public.
Ferguson examined Trantino in August 1995 and August 1998. In his earlier evaluation, Ferguson was highly supportive of Trantino’s release, finding that a “combination of well established social supports, motivation for ongoing treatment, and age [would] most likely lead to a successful parole outcome.” In reaching this conclusion, Ferguson apparently perceived that Trantino possessed “a high level of moral standards” at the time he murdered Officers Peter Voto and Gary Tedesco. Ferguson observed that it was “quite conceivable [Trantino] was experiencing a drug induced psychotic state when he committed the offense,” which could explain how he was able to commit a brutal crime “so contrary to his moral character.” I digress to note the aura of unreality that surrounds this enigmatic statement. It is undisputed that at the time the crimes were committed Trantino was a philandering wife abuser who was harboring two wanted killers and who had committed a string of burglaries and robberies. In light of these *203circumstances, Ferguson’s statement that Trantino’s crimes were “so contrary to his moral character” is utterly confounding.
In any event, Ferguson was less supportive of Trantino’s release in his August 1998 evaluation. While continuing to view parole as a reasonable possibility, Ferguson observed that psychological testing revealed antisocial personality traits suggestive of “deeply ingrained character pathology.” Ferguson reported that “Rorschach results revealfed] a self-absorbed quality ... tending] to dominate [Trantino’s] view of the world.” Trantino’s thinking process was said to be “marked by flawed logic or faulty judgment.” Significantly, Ferguson noted that Trantino was “very lax about controlling his emotional displays,” and that he was “a rather negative and probably angry person.” Ferguson concluded that “[t]he antisocial aspects of [Trantino’s] personality [were] undeniable” and “difficult to treat.”
Ferguson ultimately recommended against releasing Trantino in his testimony before the Parole Board on June 2,1999 and June 9, 1999. In his presentation, Ferguson summarized the history of Trantino’s psychological evaluations, noting that of the fourteen in-depth examinations, only four were supportive of parole. Of the forty-two cursory evaluations, thirty-five were supportive of parole. Ferguson attributed this difference to Trantino’s ingratiating personality which tended to mask the inmate’s antisocial personality traits.
Ferguson testified that Trantino’s mental condition was marked by “narcissistic” and “antisocial personality disorders.” In addition, Ferguson asserted that Trantino demonstrated many of the traits common to a “borderline” personality, a far more severe pathology. Specifically, Ferguson found that Trantino’s personality exhibited (1) unstable interpersonal relationships, (2) unstable self-image, (3) impulsiveness, (4) moodiness, (5) chronic feelings of emptiness, (6) difficulty in controlling anger, and (7) paranoia. It is to be noted that, while most of the experts who had examined Trantino over the years agreed that he suffered from narcissistic and antisocial behavior disorders, only Ferguson suggested the *204additional diagnosis of borderline personality. According to Ferguson, the three illnesses harbor “overlapping” symptoms, thus accounting for the failure of the other experts to render a similar finding as to borderline personality disorder.
While perhaps it may be reasonable to disregard Ferguson’s diagnosis concerning borderline personality disorder, the thrust of the expert’s conclusion that the amalgam of Trantino’s personality traits make the inmate a potentially “explosive” and “violent” individual cannot similarly be ignored. Ferguson concluded that Trantino was a “particularly dangerous individual” because of his “unpredictability” and his “unwillingness to clearly examine potential problems in the future.” Ferguson noted that Trantino was bound to be confronted with “blows to his ego” if released, and that absent “external controls,” his narcissism would pose serious behavioral problems.
Ferguson was not alone in recognizing Trantino’s potential for violence. Dr. Naftali Berrill, a nationally recognized expert in forensic and clinical psychology, diagnosed Trantino as suffering from a narcissistic and antisocial personality disorder.1 While *205discounting the possibility of a borderline personality, Berrill found that Trantino’s narcissistic traits made him “disruptive,” “manipulative,” and “confrontational.” Berrill testified that Trantino had been enveloped “into a culture of psycho-babble” because of his involvement in psychological testing over the years. Because of his intelligence, Trantino had been able to enlist the support of others by masking his antisocial traits. Berrill explained that people afflicted with Trantino’s degree of narcissism “lack empathy for other[s]” and cannot “appreciate anyone else’s needs or wants.” If their needs are not met, “they feel somewhat insulted or diminished,” and there is a “tremendous emotional response to the anger” they feel. According to Berrill, people like Trantino “stand outside of humanity.” They cannot “make a connection [with] other people’s pain or ... hurt .... They cannot connect [with] ... the full ramifications of the pain, destruction [and] the chaos they cause in other people’s lives.”
Berrill stressed that Trantino’s narcissism and antisocial personality disorders were extremely difficult to treat. According to Berrill, people harboring these traits cannot fully appreciate “the notion” that they act in “[a] predatory fashion.” Berrill explained that Trantino’s condition was “nearly impossible to treat” because he “stand[s] so far outside the common definition of humanity.” As poignantly phrased by Berrill, “[t]here is no end point to this game; it is a lifetime [affliction].”
Berrill considered Trantino’s alleged inability to recall the crime a major stumbling block to treatment. By clinging to the fiction *206that the killing of the victims was the result of a drug-induced psychotic episode, Trantino could avoid “having other people see[ ] him as a monster.” Berrill explained that while Trantino accepted responsibility for the crime in a superficial sense, “there [was] no genuine sense of being remorseful.” Berrill added that by “us[ing] the notion of amnesia,” Trantino could avoid confronting his culpability.
The testimony of the Parole Board’s Assistant Chief Psychologist, Mark O’Sullivan, essentially mirrored that of Ferguson and Berrill in various particulars. O’Sullivan administered a PCL-R test on Trantino shortly before the parole hearing. The PCL-R test is a widely used method to measure psychopathic personality traits. The PCL-R scale examines twenty issues based upon the subject’s entire life experience, with a score of thirty as the cutoff “at which an inmate may be reliably classified as a psychopath.” A typical inmate generally scores a twenty-three.
The test has been used on numerous occasions to determine Trantino’s personality traits. In general, Trantino scored relatively low on the scale. However, the validity of the earlier test results depended largely on the amount of information known to the tester concerning the subject’s life experiences. Unlike the prior PCL-R testing, on the most recent occasion, O’Sullivan had available to him all of the records of the New York parole authorities. O’Sullivan thus reviewed the “forty-plus years’ worth of psychological evaluations, court transcripts and decisions, crime scene photographs, transcripts of parole hearings, parole decisions, newspaper articles, media appearances” and other source information. Trantino’s final score was thirty-three. O’Sullivan concluded, “based on clinical information, research, and the PCL-R score, indicators ... suggest that [Trantino] would be a poor risk [if released].” O’Sullivan noted that “psychopaths have tended to recidivate at a higher rate and with a higher rate of violence,” and that Trantino’s score of thirty-three placed him in the ninety-first percentile of all inmates with reference to their potential to recidivate if released on parole.
*207Much of the evidence presented to the Parole Board pertained to Trantino’s excellent prison record. Dr. Mario Papparozzi, the Associate Director of the Criminal Justice Policy Center at the College of New Jersey, presented testimony concerning the “role that long term infraction-free behavior in an institutional setting may have on predicting future criminal behavior.” Papparozzi explained that “just because someone is infraction-free in an institutional setting for a period of time” does not mean that “if the individual is released [he] will remain infraction free ____” Papparozzi noted that inmates have numerous “deterrent factors” in an institution, but that such constraining influences are substantially diluted when these individuals are released on parole. The witness added that in terms of “actuarial risk assessment,” “institutional adjustment is [not] in the top three items ... accounting for [parole] outcomes.”
Papparozzi’s testimony concerning risk assessment was supported by that of Dr. Timothy Brennan, a criminologist and statistician. Brennan testified that “institutional adjustment tends to be overwhelmed by other factors in predicting future behavior.” He explained that inmates adept at manipulating others tend to behave well in prison. In the absence of the external constraints provided by the structured prison setting, this good behavior may not be replicated on parole. Brennan noted that “prior criminal history and length of prior criminal history tend to be the more dominant factors in predicting criminal behavior.”2
*208Social science literature and legal commentary support the conclusions reached by Papparozzi and Brennan. See, e.g., Lawrence F. Travis III and Vincent O’Leary, A History of Parole in Probation, Parole, and Community Corrections 109 (1976) (effectiveness of correctional treatment programs in reducing recidivism has been disappointing); Norval Morris, The Future of Imprisonment 35 (1974) (“neither the prisoner’s avoidance of prison disciplinary offenses nor his involvement in prison training programs is correlated with later successful completion of parole or with later avoidance of a criminal conviction”); The National Conference on Parole, Parole in Principle and Practice, A Manual and Report 104 (1957) (“experience has shown that there is not necessarily a one-to-one relationship between good conduct in the institution and success on parole”); James B. Jacobs, Sentencing by Prison Personnel: Good Time, 30 UCLA L.Rev. 217, 264 (1982) (prisoners successful in the structured setting of a prison may not adhere to society’s rules when released); Louis B. Schwartz, Options in Constructing a Sentencing System: Sentencing Guidelines Under Legislative or Judicial Hegemony, 67 Va. L.Rev. 637, 659 (1981) (there is a “lack of correlation between good behavior in prison and good behavior in the community”). These conclusions also comport with common sense. Prison is a quintessentially abnormal environment “dominated by ... a subculture with its own norms and values.” James B. Jacobs, Sentencing by Prison Personnel: Good Time, 30 UCLA L.Rev. at 264. “Individuals incapable of coping with the extraordinary pressures of prison life may cope well enough with the stresses of everyday life on the streets.” Ibid. Conversely, “there are many individuals who have learned to survive and even ‘prosper’ in [the] prison [setting] who cannot or will not adhere to the rules imposed by our larger society.”3 Ibid.
*209By this brief summary, I do not mean to suggest that the experts were uniform in their views concerning the dismal prospect of success if Trantino is released.4 The experts’ respective opinions wildly diverged on the subject. For example, Trantino’s expert, Dr. Barry Rosenfeld, the Senior Psychologist for the New *210York City Criminal and Supreme Courts, concluded, “the majority of predictive' factors suggest that Trantino will be capable of successfully adjusting to parole and does not present a high risk of re-offending.” Among the factors noted were Trantino’s excellent prison record, clinical testing results, and the views expressed by other psychologists over the years.
Rosenfeld’s findings, however, were not uniformly positive. He noted, for example, that some of the clinical tests such as the Millan Clinical Multiaxal Inventory and the Minnesota Multiphasic Personality Inventory “generated highly questionable personality profiles.” These tests indicated that Trantino’s “actual self-perception” disclosed a “marked lack of insight.” Rosenfeld conceded that if these test results are “genuine,” Trantino’s ability to cope with difficulties is likely somewhat “compromised.” According to Rosenfeld, “individuals with such naive self-perceptions typically lack insight into their own limitations!!,] and are therefore vulnerable to developing psychological difficulties when severe stresses arise and cannot be avoided.” Rosenfeld also acknowledged that Trantino’s attempt to present himself as well-adjusted and psychologically fit “suggested some degree of deception.” While finding that Trantino’s pre-incarceration diagnosis of antisocial personality disorder was no longer accurate, Rosenfeld observed that the inmate’s “residual symptoms of psychopathy such as grandiosity, a need for attention, and limited empathy” were consistent with a “narcissistic personality disorder.”
The Parole Board rejected Rosenfeld’s opinion finding that he was biased in favor of Trantino. In my view, the record amply supports the Parole Board’s conclusion. For example, Rosenfeld blindly accepted Trantino’s statement that he voluntarily surrendered because he did not recall having killed the victims and thus believed himself to be completely innocent. Perhaps that is so. A more sinister explanation, one that I believe is better supported by the evidence, is that Trantino knew his compatriot in crime had been killed by the police in a hail of bullets, and he, Trantino, wished to avoid that fate. So too, Rosenfeld’s scoring of Tranti*211no’s PCL-R test reveals the psychologist’s blind faith in Trantino’s convoluted explanation of the manner in which the crime occurred and Trantino’s background and psychological history. Rosenfeld scored Trantino in the range of fifteen based on the limited information given to him by Trantino’s lawyer. Later, additional information, some of it highly derogatory with respect to Trantino’s truthfulness, was disclosed to Rosenfeld. Rosenfeld nevertheless maintained that none of the information caused him to change the PCL-R test result. Moreover, it appears that Rosenfeld considered only Trantino’s current status. It is undisputed that the PCL-R checklist should be based on the subject’s entire life experience.
I do not mean to denigrate Dr. Rosenfeld’s analysis. Nor do I ignore the findings of other psychologists who, at various times, have recommended that Trantino should be paroled. But it is not our function to weigh the opinions of the experts and decide which of these conclusions is the most credible.5 That is the role of the Parole Board. The Appellate Division’s aim was simply “to determine whether the [Parole Board’s] findings could reasonably have been reached on sufficient credible evidence present in the record.” State v. Johnson, 42 N.J. at 162, 199 A.2d 809. While the Appellate Division purported to apply the correct legal standard, it *212exceeded its power and improperly intruded upon the fact-finding authority of the Parole Board.
Because the Appellate Division found that the Parole Board’s conclusion was incorrect, “we almost instinctively scrutinize the record more searchingly.” Id. at 163,199 A.2d 809. I note in that respect that the Appellate Division’s decision makes no reference whatsoever to the analyses provided by Berrill, O’Sullivan, Papparozzi or Brennan. Beyond this, the court improperly rejected Ferguson’s opinion, apparently because the witness’s current conclusion differed from his earlier views and because he phrased his ultimate determination in terms of Trantino’s “potential” for violence rather than the “substantial likelihood” of recidivism. It is true that the Parole Board, the Appellate Division, and ultimately the majority, could properly consider the fact that Ferguson initially was strongly supportive of Trantino’s release. However, the witness was not bound in perpetuity to his earlier opinion. As Ferguson candidly explained, his 1995 evaluation was one of the first he made as a psychologist. Moreover, Ferguson’s subsequent opinion that Trantino posed an unjustifiable risk of danger if released was based on information that was not available to the witness when he prepared his 1995 report. Finally, the fact that Ferguson did not couch his opinion in the statutory language is not a sound basis for rejecting the expert’s conclusion. Ferguson is.a psychologist, not a lawyer. The Appellate Division’s parsing of the words he used to express his opinion adds little to the ease. To put the matter at rest, I have read and reread Ferguson’s testimony, and I have no doubt whatsoever that he considered Trantino a substantial danger to the public in terms of the likelihood of recidivism. The witness was not required to parrot the statutory language in order to be believed.
I thus conclude that the Appellate Division itself was manifestly mistaken in its failure to accept the Parole Board’s conclusion concerning Trantino’s psychological pathology. The majority repeats that mistake. In my view, the Parole Board could reasonably have found, based on the evidence I have described, that *213Trantino is manipulative, deceptive and explosive, and that these traits make him a poor candidate for parole.
B. Corroborative Facts
The final determination concerning a prisoner’s risk of recidivism rests with the Parole Board, not psychiatrists, social workers or psychologists. Trantino v. New Jersey State Parole Bd., 154 N.J. at 36, 711 A.2d 260. The ultimate decision is a legal one, not a social science one, even though it is guided by expert testimony. Cf. In re D.C., 146 N.J. 31, 59, 679 A.2d 634 (1996) (role of medical testimony in commitment proceedings). We have said the courts should take care not “to abdicate [their] decision-making responsibility to experts.” In re Registrant G.B., 147 N.J. 62, 87, 685 A.2d 1252 (1996). This principle applies with equal force to administrative agencies. Administrative agencies bear the ultimate responsibility for the decisions they render. In that context, it is at least arguable that the expert testimony I have described may not, standing alone, serve as a sufficient basis to support the Parole Board’s denial of Trantino’s release.
The record, however, contains facts tending to corroborate the experts’ opinion that Trantino is a dangerous person likely to recidivate if granted parole, because he is deceptive, manipulative and lacks empathy for others. It is true that many of these incidents are dredged from Trantino’s dark past and perhaps were inflated in importance in the Parole Board’s written determinations. Consideration of these various incidents separately, which is Trantino’s thesis here, may show no one in itself is sufficiently meaningful. But in combination, and coupled with Trantino’s psychological history and current mental state, they tend to corroborate the experts’ prognostication of danger.
First, there is Trantino’s convenient amnesia concerning his role in the commission of the crimes. In our earlier opinion, we noted the presence in the record of evidence that “Trantino’s memory loss is consistent, long-standing and genuine.” Trantino v. New Jersey State Parole Board, 154 N.J. at 35, 711 A.2d 260. We also observed that the record contained evidence tending to show that *214Trantino’s acceptance of responsibility for his crimes was “sincere and legitimate.” Ibid. I am satisfied that these statements did not foreclose the Parole Board from revisiting the subject, as the majority asserts. In a variety of circumstances and factual settings, appellate courts are often presented with records containing substantial evidence supporting inconsistent theses. As I pointed out earlier, it is not our job to pick and choose which thesis is accurate. That role is ordinarily performed by trial judges and administrative agencies, and we are bound to defer to their conclusions if they are reasonable.
The Parole Board found that Trantino’s inability to recall the circumstances of the crime was feigned, and that his abstract, obtuse and convoluted acknowledgment of guilt was not meaningful. There is substantial support for that conclusion in the record.
It is to be recalled that Trantino’s defense at trial was voluntary intoxication. State v. Trantino, 44 N.J. 358, 362-63, 209 A.2d 117 (1965). Trantino testified that he took two Dexedrine pills and consumed a considerable quantity of liquor on the day of the homicides. Id. at 362, 209 A.2d 117. He denied any recollection of the slaying of the officers, saying he recalled only a loud explosion followed by a confusion of wild sound and light within which his accomplice appeared to be a devil. Id. at 362-63, 209 A.2d 117. The prosecution presented strong evidence debunking Trantino’s claim that he was under the influence of alcohol and drugs when he killed the victims and that he could not recall the events in question. For example, although Trantino disavowed awareness of the homicides, the getaway driver, Patricia MacPhail, who helped him flee to New York City, testified that he told her he had murdered the policemen to help his accomplice, Frank Falco, who was wanted for murder. Id. at 363, 209 A.2d 117. At the time of the trial, voluntary intoxication could be considered in determining whether the defendant in fact performed the mental operations which the prosecution was required to prove to elevate murder in the second degree to murder in the first degree. Id. at 369, 209 A.2d 117. Also at the time of trial, the proof necessary to *215convict a defendant of first degree murder was far more substantial than that necessary to prove first degree murder under current laws. The State then was required to establish “premeditation, deliberation and wilfulness in the execution of the design to kill.” State v. Hudson, 38 N.J. 364, 369, 185 A.2d 1 (1962). In any event, the jury was instructed accordingly, and the verdict of murder in the first degree “express[ed][its] rejection of [Trantino’s] testimony and the expert opinion resting upon it.” State v. Trantino, 44 N.J. at 369, 209 A.2d 117.
In the resulting appeal, Trantino’s “principal emphasis” was on his claim that the killings were the result of a drug-induced psychotic episode that he could not precisely recall. Id. at 370, 209 A.2d 117. In rejecting that argument, Chief Justice Weintraub wrote:
There was ample evidence from which the jury could conclude that Trantino knew what was going on and what he was doing. He did not menace Sergeant Voto until the officer discovered the gun. From what defendant told Patricia MacPhail, the jury could find he feared that arrests would follow discovery of the gun and that Falco would have to face a murder charge in New York. Trantino’s own testimony as to events immediately before and immediately after the killings could be found to be inconsistent with the claim of alcoholic stupor at the time of the killings. He knew there was something to run from, and although he says he does not know what it was, the jury could conclude, as Mrs. MacPhail’s testimony revealed, that defendant knew he had killed the policemen. In fact none of the witnesses who were at the tavern, nor a milkman from whom defendant sought a ride shortly after the murders, supported the claim of alcoholic prostration.
Id. at 371, 209 A.2d 117.
The point to be stressed here is that a jury, having the opportunity to see and hear Trantino testify, rejected the claim that he could not recall the events in question. To put it more bluntly, the jury viewed Trantino’s claim of a drug-induced psychotic episode, and saw nothing but the heart of darkness. Trantino’s later attempts to accept responsibility for the crimes while denying any recollection of the shootings merely resurrected a claim earlier considered and rejected by a well-informed jury. No amount of historical revisionism can change that basic fact.
Beyond this, the Parole Board could reasonably have questioned the accuracy of our statement that “Trantino’s memory loss is *216consistent, long-standing and genuine.” Trantino v. New Jersey State Parole Board, 154 N.J. at 35, 711 A.2d 260. In point of fact, Trantino has wavered in his claim of amnesia. At his postconviction relief hearing in 1967, for example, Trantino testified unequivocally that he did not kill the victims. At his parole hearing in 1980, Trantino recounted that he was “innocent” of killing the policemen, noting that he had left the tavern before the crimes were committed. Trantino repeated that version when he testified at his parole hearing in 1982.
Trantino’s failure to come to terms with the brutal crimes he committed is important for several reasons. I do not doubt that a sentence should not be enlarged because a prisoner continues to insist he is innocent. But the greater demand is that the Parole Board take into account the man and his prospect for recidivism. The aims of punishment are several, State v. Ivan, 33 N.J. 197, 199-202, 162 A.2d 851 (1960), but the hope is that the sentence, mild or severe, will reshape the offender. In assaying the prospects for recidivism, the Parole Board must look to the man as well as to the offense, and it is here that the inmate’s attitude toward the truth is highly relevant. In a variety of contexts, we have recognized that a candid acknowledgment of guilt is the first sign of redemption. See State v. Poteet, 61 N.J. 493, 497, 295 A.2d 857 (1972); State v. Forcella, 52 N.J. 263, 275, 245 A.2d 181 (1968), rev’d in part, Funicello v. New Jersey, 403 U.S. 948, 91 S.Ct. 2278, 29 L.Ed.2d 859 (1971); State v. De Stasio, 49 N.J. 247, 260, 229 A.2d 636, cert. denied, 389 U.S. 830, 88 S.Ct. 96, 19 L.Ed.2d 89 (1967).
Trantino’s bland acceptance of responsibility while claiming amnesia is neither candid nor redemptive. It is instead both manipulative and deceptive. I leave the question of moral blameworthiness to the philosophers. I, instead, point to Dr. Berrill’s opinion that Trantino’s alleged inability to recall the crimes constitutes strong evidence of continued pathology.
Trantino was also disingenuous in his testimony regarding his relationship with his first wife, Helene. Trantino admitted that he *217was less than an ideal husband, having had sexual intercourse with another woman during the couple’s honeymoon and thereafter engaging in a series of extramarital affairs. He testified that he struck Helene “three or four times” in the course of the marriage, but he confidently assured the Parole Board that she would not characterize his conduct toward her as “abusive.” He was wrong. In her statement to New York parole authorities, Helene maintained that Trantino often beat her. Helene recounted that she submitted to Trantino sexually because of her fear as to what he would do if she were to refuse his demands.
Trantino also either minimized or lied about the significance of his 1956 strong-armed robbery of the dentist’s payroll and the injuries inflicted upon the victim. This incident again indicates Trantino’s penchant to mitigate his involvement in crime. It also suggests his inability to appreciate the impact of his conduct on others.
Coneededly, these events took place many years ago. They nevertheless are relevant because they contribute to the composite picture of the “whole man,” his current mental state, and his prospects for the future. They have resonance not in the fact that they occurred or in the prospect that they will be repeated if Trantino is released. Rather, this evidence confirms Dr. BerrilTs opinion that Trantino is dangerous because he cannot appreciate how his conduct affects others. It also confirms the fact that Trantino was less than truthful in his testimony before the Parole Board.
Perhaps too much emphasis has been placed on the Talbot Hall incident. The simple fact remains, however, that those present at the scene were able to discern a crack in Trantino’s veneer, a side of his personality that he otherwise kept hidden. Trantino must have known that he was under a microscope. Yet he refused to complete the psychological testing. Moreover, Trantino either minimized or lied about the incident in his testimony at the Parole Board hearing.
*218Finally, Trantino exhibited a significant lack of empathy in his preliminary decision to write another book pertaining to the killings. The majority’s interpretation of this incident as disclosing Trantino’s reformation is belied by the record. Obviously Trantino has a First Amendment right to publish another book. However, the concern lies not with the publication itself, but with Trantino’s failure to recognize that such action may cause further emotional trauma to the victims’ families. As noted by the Parole Board, Trantino’s disregard for the feelings of the victims’ families tended to corroborate Dr. Berill’s findings that Trantino lacks the capacity to understand how his actions affect others.
In light of this evidence, I do not believe that the Parole Board went wide of the mark in finding that there is a substantial likelihood of recidivism if Trantino is released. Resolution of these issues by the Parole Board required a reconciliation of competing social values. The interest of public security was at war with the concern for Trantino’s personal freedom. Believing that the interest of public safety was paramount, the Parole Board sought to alleviate the tension between these values by denying Trantino’s release. I cannot fairly say that the Parole Board was wrong.
III.
The difference between my view and that of my colleagues is not one of good intentions. We all wish to accommodate the interests of public safety and those of our parole system. Our commitment to the demands of public security is equally shared. Nor do we differ in our devotion to the rule of law. The question is not whether the rule of law shall prevail, but where that principle takes us. Our government is one of laws, but it is run by men and women with different experiences and viewpoints. My disagreement with the majority is grounded in how we approach the evidence. In my experience, claims of rehabilitation by convicted offenders come to the court with numbing frequency and rubber-stamp regularity. The story of sin and redemption is as *219old as humankind. It is all too easy to succumb to the chimeric vision that good ultimately triumphs over evil in the human soul. But reality bites, and too often the truth lies elsewhere. The Parole Board viewed Trantino’s claims of rehabilitation with a healthy dose of skepticism. So do 1.1 would reverse the Appellate Division’s decision and reinstate the Parole Board’s denial of release. I would remand the matter to the Appellate Division for the limited purpose of reviewing the Department of Corrections’ actions relating to Trantino’s placement in the penal system.
For modification and affirmance — Justices STEIN, COLEMAN, LONG and Judge HAVEY, temporarily assigned — 4.
For reversal — Judge BAIME, temporarily assigned — 1.
The majority notes its concern that Berrill along with Brennan and Papparozzi, did not personally examine Trantino. The record discloses that these witnesses carefully reviewed Trantino's psychological history, including approximately thirty years worth of probation reports, psychological and psychiatric reports, and psychological testing. This issue must be considered within its historical context In 1982, the Supreme Court revised our Rules of Evidence to mandate that in formulating his or her opinion, an expert may rely upon reliable documentary evidence, such as reports of others. Biunno, Current N.J. Rules of Evidence, comment on N.J.R.E. 703 (2000). That rule, which carried over into our new rules of evidence and appears in the Federal Rules of Evidence, has been consistently applied in New Jersey, Costantino v. Ventriglia, 324 N.J.Super. 437, 450, 735 A.2d 1180 (App.Div. 1999); State v. Dishon, 297 N.J.Super. 254, 280-81, 687 A.2d 1074 (App.Div.), certif. denied, 149 N.J. 144, 693 A.2d 112 (1997); Glowacki v. Underwood Memorial Hosp., 270 N.J.Super. 1, 17-18, 636 A.2d 527 (App.Div. 1994); State v. Smith, 262 N.J.Super. 487, 521, 621 A.2d 493 (App.Div.), certif. denied, 134 N.J. 476, 634 A.2d 523 (1993); Industrial Dev. Assoc. v. Commercial Union Surplus Lines Ins. Co., 222 N.J.Super. 281, 296-97, 536 A.2d 787 (App.Div.1988); Correa v. Maggiore, 196 N.J.Super. 273, 283, 482 A.2d 192 (App.Div.1984); State v. Stevens, 136 N.J.Super. 262, 264, 345 A.2d 804 *205(App.Div.1975), and by the federal circuits. Southland Sod Farms v. Stover Seed Co., 108 F.3d 1134, 1142 (9th Cir.1997); In James Wilson Assoc., 965 F.2d 160, 172-73 (7th Cir. 1992); Durflinger v. Artiles, 727 F.2d 888, 892-93 (10th Cir. 1984); United States v. Lawson, 653 F.2d 299, 301-302 (7th Cir.1981); United States v. Hill, 655 F.2d 512, 515-16 (3d Cir.1981); United States v. Morrison, 531 F.2d 1089, 1094-95 (1st Cir.1976); United States v. Partin, 493 F.2d 750, 764-65 (5th Cir.1974); United States v. Harper, 460 F.2d 705, 706-707 (5th Cir.1972); Birdsell v. United States, 346 F.2d 775, 779-80 (5th Cir.), cert. denied, 382 U.S. 963, 86 S.Ct. 449, 15 L.Ed.22d 366 (1965), reh’g denied, 383 U.S. 923, 86 S.Ct. 900, 15 L.Ed.2d 680 (1966); Carrington v. Civil Aeronautics Board, 337 F.2d 913, 916 (4th Cir. 1964); Jenkins v. United States, 307 F.2d 637, 641-42 (D.C.Cir.1962).
The majority discounts the opinions of Berrill, O'Sullivan, Papparozzi, and Brennan because their reports were given after the denial of parole and in the course of setting a future eligibility date. The longstanding practice of the Parole Board is to bifurcate the question of whether parole is granted and the determination of a future eligibility date. However, they form constituent parts of a single decision. See N.J.A.C. 10A:71-3.18; N.J.A.C. 10A:71-3.21. If evidence of exoneration had come to us in the course of the proceedings setting a future eligibility date, we would have undoubtedly considered it. The compelling nature of the evidence presented by Berrill, O'Sullivan, Papparozzi, and Brennan is not lessened merely because it was presented during determination of the future eligibility date.
In 1979 our Legislature adopted the statutory framework necessary to implement a “contract parole” program. See N.J.S.A. 30:4—123.67. Contract parole allows for the pre-determination of a release date for an inmate provided the inmate complies with a program of behavioral, educational and rehabilitative *209goals designed to reduce prison misconduct and enhance the inmate’s post-release community adjustment. William Parker, Highlights of the Parole Process, in Probation, Parole, and Community Correction 144 (3rd. Ed.1984); James O. Finckenauer and Carol Rauh, Contract Parole. Some Legal and Rehabilitative Issues of Mutual Agreement Programming for Parole Release, 5 Cap. U.L.Rev. 175, 179 (1976). Although the Legislature provided the requisite guidelines for establishing a contract parole program, the Department of Corrections and the Parole Board chose not to implement the necessary regulations to bring such a program to fruition. See Raymond v. N.J. State Parole Board, 221 N.J.Super. 381, 384, 534 A.2d 741 (App.Div.1987). In any event, the standard is not whether Trantino has earned his release by good conduct in prison. It is whether he presents an unjustifiable risk of recidivism if released.
The majority asserts that the Parole Board erred by ignoring Dr. Michael Weiner’s report. In my view, this was a report well worth rejecting. As noted by the Parole Board, the report pertained to the psychological adaptation of Trantino to the structured setting of a halfway house, rather than the likelihood that Trantino would commit another crime if released into the community. Although Weiner ultimately concluded that Trantino's likelihood of recidivism was relatively low, I note that the conclusion is wholly disembodied from, and inconsistent with, the psychologist’s specific findings.
In his report, Weiner noted that his examination of Trantino "found more impressive evidence for lack of empathy, manipulativeness, grandiosity, lack or remorse or guilt, lying, [and] failure to accept responsibility for his own actions----’’ The report detailed Trantino’s continuing narcissistic personality, finding him to be "intensely egocentric” and "likely to be exploitive in the future.” Weiner observed that the "degree of his self-centeredness is such that he is a considerable risk to erode his support system and to relapse into exploiting others,” which would “eventually graduate into criminality.”
Weiner further stated that Trantino’s incessant "revision of history” to "promote the notion of his innocence” allowed him to avoid all responsibility for the murders he committed. Weiner found that Trantino remains "unremorseful to the array of people he has victimized directly and indirectly" and exhibits only "superficial empathy.” In fact, Weiner noted that Trantino, "speaks of the murdered Detective Sgt. Voto and police trainee Tedesco with utterly no emotion or compassion for either of them.” Weiner concluded that, ”[p]eeling away his exterior, [Trantino] remains a controlling, calculating individual with no real sense of attachment. He maintains the tendency to explain away, deny, or minimize his meaningful sources of shame or guilt.”
The majority makes repeated reference to the "selectivity" of the Parole Board in focusing only on the psychological evidence supportive of its denial of parole. This criticism is unwarranted. The Parole Board clearly considered every scrap of available evidence in assessing the likelihood of recidivism. The record does not in any way suggest that the Parole Board ignored, overlooked, or undervalued crucial evidence. As I pointed out earlier, the opinions of the experts wildly diverged. It was thus incumbent on the Parole Board to determine which opinions to accept and which opinions to reject. The Parole Board was not required to make this determination with surgical precision or to discuss the testimony, reports and statements of every witness, describing in detail why it found some more credible than others. See State v. Locurto, 157 N.J. at 471-74, 724 A.2d 234. In that context, the Parole Board was not "selective” in its consideration of the evidence, but instead was necessarily selective in its determination concerning what evidence was the most compelling.