State v. Fortin
This text of 843 A.2d 974 (State v. Fortin) is published on Counsel Stack Legal Research, covering Supreme Court of New Jersey primary law. Counsel Stack provides free access to over 12 million legal documents including statutes, case law, regulations, and constitutions.
Opinions
Justice ALBIN
delivered the opinion of the Court.
Defendant Steven Fortin was convicted by a jury of capital murder and sentenced to death. He claims that he was denied a fair trial as a result of various rulings of the trial court in the guilt and penalty phases of the trial. We conclude that the trial errors were sufficiently egregious so as to deny defendant a fair trial and, therefore, we are constrained to reverse.
[559]*559I.
Facts and Procedural History
On August 11,1994, twenty-five-year-old Melissa Padilla resided at the Gem Motel in Woodbridge with her four young children, ages two through five, and her boyfriend, Hector Fernandez.1 Padilla and her children were receiving public assistance and had been placed at the motel by a social service agency. That day, at approximately 11:00 p.m., she left the motel to buy some food for her family and walked to a nearby Quick Chek convenience store located at the intersection of U.S. Highway Route 1 North and Avenel Street. Padilla’s walk took her along a 1000-foot dirt path that ran parallel to Route 1 North. At 11:29 p.m., she purchased iced tea, pizza, bread, three sandwiches, a candy bar, and coffee with a $20 bill, receiving $0.66 in change. She left the Quick Chek and began on her way back to the motel.
When Padilla did not return, Fernandez became concerned and went to look for her, accompanied by the sons of the motel desk clerk, eleven-year-old Christopher and five-year-old Antoine, and a friend, Trent Eubanks, who was staying at the Gem Motel that evening. Five-year-old Antoine discovered Padilla’s body 500 feet from the motel inside one of four concrete thirty-inch pipes, which lay on the path Padilla had taken to and from the store.
By the time the first police officer arrived on the scene at approximately 1:00 a.m., Fernandez had pulled Padilla’s body out of the pipe. The officer found Fernandez and Eubanks standing near Padilla’s badly battered body, which was naked from the waist down. Dr. Marvin Schuster, the eounty medical examiner, was called to the scene and pronounced Padilla dead at 1:06 a.m.
County investigators also arrived and canvassed the area for evidence. They observed Padilla’s bloodied face, her blood-soaked shirt, and blood on her arms and hands. A pool of blood had [560]*560collected in the concrete pipe and a bloody trail marked the distance Padilla’s body had been dragged from the pipe. Blood spattering evidence indicated that the assault had taken place inside the construction pipe.
The police found no money or jewelry on or near Padilla’s body. They did discover, however, a bloody dollar bill and receipt from the convenience store in the vicinity. Padilla’s groceries were strewn about the area, including three sandwich containers, one of which was empty. The sandwich was later discovered, partially eaten, on a nearby street. The police found Padilla’s shorts, with her underpants inside, in a tree a short distance from the sandwich. The investigation uncovered no identifiable fingerprints other than those of the victim.
An autopsy was conducted the next day. Dr. Schuster determined that Padilla had suffered numerous injuries, including a broken nose and bruises to her face and chest; lacerations to her chin and left breast that were possibly bite marks; and lacerations to the anus inflicted shortly before death that were consistent with forceful penetration by an object, possibly a finger or penis. Padilla suffered no observable injuries to her vagina. The evidence and absence of evidence — the presence of “few and scattered spermatozoa” in the vagina and no semen found on the body — suggested that Padilla was not vaginally assaulted. The bruises to Padilla’s face and forehead were caused by a combination of blunt force trauma and the scraping of her skin against the concrete surface of the pipe. A fracture in Padilla’s hyoid bone, hemorrhaging of her epiglottis (the upper portion of the windpipe), and abrasions to her neck revealed that she had been manually strangled. The medical examiner concluded that Padilla died from asphyxiation and that her anal injuries were the result of a sexual assault at or near the time of her death.
On August 11, 1994, the day of the murder, Steven Fortin lived with Dawn Archer, his former girlfriend, in the Douglas Motel located on Route 1, north of the Quick Chek, in Woodbridge. That evening, Archer and Fortin left their motel to visit a friend, [561]*561Charles Bennett, who lived in the Five Oaks Apartments on the northbound side of Route 1, less than a mile south of the Douglas Motel. The Five Oaks Apartments were located to the south of the Gem Motel. Fortin and Archer walked south along the northbound side of Route 1 and stopped at the Quick Chek to buy some cigarettes.2 After their arrival at Bennett’s apartment around 9:00 p.m., the three drank alcohol. Archer remembered arguing with Fortin and Bennett asking them to leave. Bennett denied witnessing an argument between Fortin and Archer.
At approximately 10:30 p.m., Fortin and Archer left Bennett’s apartment and continued to argue. After walking a short distance, Fortin turned violent and began to choke and kick Archer. He threw her to the ground, and Archer ran into a nearby restaurant, Bud’s Hut, screaming for help.
At 10:32 p.m., a Woodbridge police officer responded to the Bud’s Hut parking lot, where he found Archer intoxicated, red-faced and nose-bloodied. Archer told the police officer that her boyfriend, Steven Fortin, had attacked her. She, however, declined to sign a complaint against him. The police took Archer to the hospital, but she refused treatment.
At approximately 11:15 p.m., Fortin returned to Bennett’s apartment looking for Archer. Bennett observed a few scratches on Fortin’s leg but none on his face. Fortin explained that he had gotten the scratches while in a fight with Archer. He left Bennett’s apartment a few minutes later. Fortin’s walk back to the Douglas Motel would have set him in the direction of the brutal assault of Padilla. Archer’s and Bennett’s accounts placed Fortin near the Gem Motel and the Quick Chek around 11:30 p.m. on the evening of August 11,1994.
On August 12, Fortin called a friend, Ron Celis, and met him at a diner. Fortin appeared upset and told Celis he was having “woman problems.” Celis observed scratches on Fortin’s face. [562]*562Fortin explained that he got the scratches while traversing through a wooded area. Archer next saw Fortin on August 13, when the two reconciled. She noticed he had scratches, several inches long, on his face, chest, and arms.
Fortin and Archer moved out of the Douglas Motel and became itinerant travelers, mostly staying with friends. Eventually, they went to Maine to stay with Fortin’s parents. During a trip to visit Archer’s father in Massachusetts, Fortin struck Archer and the two ¡larted for good.
At the time of the investigation of the Padilla murder, the police had not gathered those facts and they had few leads to go on. There were no witnesses to the crime. Although the police took fingerprint exemplars from several local men who had been convicted of sexual assaults, there were no matches to fingerprints lifted from the scene. Fortin was not on their radar screen. Without leads or suspects, the investigation was stalled.
Free access — add to your briefcase to read the full text and ask questions with AI
Justice ALBIN
delivered the opinion of the Court.
Defendant Steven Fortin was convicted by a jury of capital murder and sentenced to death. He claims that he was denied a fair trial as a result of various rulings of the trial court in the guilt and penalty phases of the trial. We conclude that the trial errors were sufficiently egregious so as to deny defendant a fair trial and, therefore, we are constrained to reverse.
[559]*559I.
Facts and Procedural History
On August 11,1994, twenty-five-year-old Melissa Padilla resided at the Gem Motel in Woodbridge with her four young children, ages two through five, and her boyfriend, Hector Fernandez.1 Padilla and her children were receiving public assistance and had been placed at the motel by a social service agency. That day, at approximately 11:00 p.m., she left the motel to buy some food for her family and walked to a nearby Quick Chek convenience store located at the intersection of U.S. Highway Route 1 North and Avenel Street. Padilla’s walk took her along a 1000-foot dirt path that ran parallel to Route 1 North. At 11:29 p.m., she purchased iced tea, pizza, bread, three sandwiches, a candy bar, and coffee with a $20 bill, receiving $0.66 in change. She left the Quick Chek and began on her way back to the motel.
When Padilla did not return, Fernandez became concerned and went to look for her, accompanied by the sons of the motel desk clerk, eleven-year-old Christopher and five-year-old Antoine, and a friend, Trent Eubanks, who was staying at the Gem Motel that evening. Five-year-old Antoine discovered Padilla’s body 500 feet from the motel inside one of four concrete thirty-inch pipes, which lay on the path Padilla had taken to and from the store.
By the time the first police officer arrived on the scene at approximately 1:00 a.m., Fernandez had pulled Padilla’s body out of the pipe. The officer found Fernandez and Eubanks standing near Padilla’s badly battered body, which was naked from the waist down. Dr. Marvin Schuster, the eounty medical examiner, was called to the scene and pronounced Padilla dead at 1:06 a.m.
County investigators also arrived and canvassed the area for evidence. They observed Padilla’s bloodied face, her blood-soaked shirt, and blood on her arms and hands. A pool of blood had [560]*560collected in the concrete pipe and a bloody trail marked the distance Padilla’s body had been dragged from the pipe. Blood spattering evidence indicated that the assault had taken place inside the construction pipe.
The police found no money or jewelry on or near Padilla’s body. They did discover, however, a bloody dollar bill and receipt from the convenience store in the vicinity. Padilla’s groceries were strewn about the area, including three sandwich containers, one of which was empty. The sandwich was later discovered, partially eaten, on a nearby street. The police found Padilla’s shorts, with her underpants inside, in a tree a short distance from the sandwich. The investigation uncovered no identifiable fingerprints other than those of the victim.
An autopsy was conducted the next day. Dr. Schuster determined that Padilla had suffered numerous injuries, including a broken nose and bruises to her face and chest; lacerations to her chin and left breast that were possibly bite marks; and lacerations to the anus inflicted shortly before death that were consistent with forceful penetration by an object, possibly a finger or penis. Padilla suffered no observable injuries to her vagina. The evidence and absence of evidence — the presence of “few and scattered spermatozoa” in the vagina and no semen found on the body — suggested that Padilla was not vaginally assaulted. The bruises to Padilla’s face and forehead were caused by a combination of blunt force trauma and the scraping of her skin against the concrete surface of the pipe. A fracture in Padilla’s hyoid bone, hemorrhaging of her epiglottis (the upper portion of the windpipe), and abrasions to her neck revealed that she had been manually strangled. The medical examiner concluded that Padilla died from asphyxiation and that her anal injuries were the result of a sexual assault at or near the time of her death.
On August 11, 1994, the day of the murder, Steven Fortin lived with Dawn Archer, his former girlfriend, in the Douglas Motel located on Route 1, north of the Quick Chek, in Woodbridge. That evening, Archer and Fortin left their motel to visit a friend, [561]*561Charles Bennett, who lived in the Five Oaks Apartments on the northbound side of Route 1, less than a mile south of the Douglas Motel. The Five Oaks Apartments were located to the south of the Gem Motel. Fortin and Archer walked south along the northbound side of Route 1 and stopped at the Quick Chek to buy some cigarettes.2 After their arrival at Bennett’s apartment around 9:00 p.m., the three drank alcohol. Archer remembered arguing with Fortin and Bennett asking them to leave. Bennett denied witnessing an argument between Fortin and Archer.
At approximately 10:30 p.m., Fortin and Archer left Bennett’s apartment and continued to argue. After walking a short distance, Fortin turned violent and began to choke and kick Archer. He threw her to the ground, and Archer ran into a nearby restaurant, Bud’s Hut, screaming for help.
At 10:32 p.m., a Woodbridge police officer responded to the Bud’s Hut parking lot, where he found Archer intoxicated, red-faced and nose-bloodied. Archer told the police officer that her boyfriend, Steven Fortin, had attacked her. She, however, declined to sign a complaint against him. The police took Archer to the hospital, but she refused treatment.
At approximately 11:15 p.m., Fortin returned to Bennett’s apartment looking for Archer. Bennett observed a few scratches on Fortin’s leg but none on his face. Fortin explained that he had gotten the scratches while in a fight with Archer. He left Bennett’s apartment a few minutes later. Fortin’s walk back to the Douglas Motel would have set him in the direction of the brutal assault of Padilla. Archer’s and Bennett’s accounts placed Fortin near the Gem Motel and the Quick Chek around 11:30 p.m. on the evening of August 11,1994.
On August 12, Fortin called a friend, Ron Celis, and met him at a diner. Fortin appeared upset and told Celis he was having “woman problems.” Celis observed scratches on Fortin’s face. [562]*562Fortin explained that he got the scratches while traversing through a wooded area. Archer next saw Fortin on August 13, when the two reconciled. She noticed he had scratches, several inches long, on his face, chest, and arms.
Fortin and Archer moved out of the Douglas Motel and became itinerant travelers, mostly staying with friends. Eventually, they went to Maine to stay with Fortin’s parents. During a trip to visit Archer’s father in Massachusetts, Fortin struck Archer and the two ¡larted for good.
At the time of the investigation of the Padilla murder, the police had not gathered those facts and they had few leads to go on. There were no witnesses to the crime. Although the police took fingerprint exemplars from several local men who had been convicted of sexual assaults, there were no matches to fingerprints lifted from the scene. Fortin was not on their radar screen. Without leads or suspects, the investigation was stalled. In April 1995, the Maine State Police communicated with the Woodbridge police department about Fortin, who had been arrested for sexually assaulting Maine State Trooper Vicki Gardner. Fortin’s savage assault on Trooper Gardner became the centerpiece of the State’s case. The State would argue that the unique similarities between the sexual assaults against Gardner and Padilla led to only one conclusion — that the same man committed both crimes.
On the evening of April 3,1995, Trooper Gardner was driving in a marked patrol car on Interstate Highway 95 in Maine, returning home from a visit to her parents. She was off-duty and not in uniform. Around 8:30 p.m., Gardner encountered Fortin in a car parked on the shoulder of Interstate 95, facing in the opposite direction of traffic. She stopped to investigate. Fortin explained to Trooper Gardner that he was lost. In response to a request for his credentials, Fortin could produce only a New Jersey driver’s permit; he had no registration or insurance card. After detecting the smell of alcohol, Trooper Gardner asked Fortin to step from his ear and to take a seat in the front passenger side of her patrol car. Walking slightly off-balance, Fortin entered the patrol car, [563]*563where the trooper administered field sobriety tests. As a result of the tests, the trooper concluded that Fortin was legally intoxicated and decided to issue him several summonses, including one for driving under the influence.
Trooper Gardner radioed headquarters for back-up assistance from an on-duty officer. With Fortin seated beside her, and sensing no danger, the trooper completed the paperwork on the charges as she waited for a back-up officer. After forty-five minutes to an hour had passed without the arrival of assistance, Fortin told the trooper he had a proposition for her. Before he spoke any further, Trooper Gardner explained to him that she would listen, but that she was issuing him summonses and that he would have to post bail because he was an out-of-state resident. Fortin then proposed that she let him “get in the car and drive away and [she] could pretend that nothing had ever happened.” The trooper repeated to him that the charges were serious and that he would have to be placed under arrest.
While she proceeded with her paperwork, Fortin suddenly grabbed her around the neck and slammed her head against the doorpost, knocking her unconscious. When Gardner awoke, For-tin was strangling her. She struggled to fight him off, but again lost consciousness.
When she regained consciousness, Gardner found herself lying across the front seat of the car with her head against the passenger door, naked from the waist down with her shirt pulled up and her breasts exposed. Her face was so bloody and bruised that one eye was swollen shut and she could barely see out of the other. Her nose was so badly broken that it had to be surgically “stitch[ed]” back to her face. Fortin also had bitten her on the chin, her left breast nipple, and on the lateral side of her left breast. Despite her injuries, Trooper Gardner managed to escape. Fortin later was arrested.3
[564]*564Dr. Lawrence Ricci, who examined Trooper Gardner, found significant vaginal and anal injuries. Those injuries included bruising of the vaginal area and a complete tear of the hymen, and a large laceration from the anal opening to the rectum with extensive bruising of the anus. Dr. Ricci concluded that the anal injuries were consistent with penetrating trauma from fingers or a fist. When Trooper Gardner later spoke with investigators from her department, she recalled that Fortin had forced his finger into her vagina, causing her pain. She had no recollection of Fortin penetrating her anally.
One detail of Fortin’s assault on the trooper caught the attention of the Woodbridge police: Fortin had bitten Gardner on the breast during the assault. On April 24, 1995, two Woodbridge detectives traveled to Skowhegan, Maine to interview Fortin. Fortin waived his Miranda rights and spoke with the detectives for two and one-half hours. He indicated that he had lived in Woodbridge at the time of the Padilla murder and acknowledged reading about it in the newspaper. When confronted with the evidence of similar bite marks in the two cases, Fortin said, “If the evidence shows that I did it it would probably be the reason and I must have been involved.” He continued, “I’m not admitting anything. If the proof shows I did then I must have done it. I don’t recall.” In response to repeated questions concerning details of the Padilla murder, he stated that he had no recollection.
The physical evidence gathered did not rule in or out Fortin’s involvement in the murder. No identifiable fingerprints, other than Padilla’s, were found. Fortin was not the source of any hairs recovered, and a pubic hair found on Padilla did not match Fortin, Fernandez, or Padilla. Testing also excluded Fortin as a source of the blood found on the dollar bill and was inconclusive as to saliva found on a cigarette butt. DNA testing of specimens at the scene either excluded Fortin as a source or proved inconclusive. The spermatozoa recovered from Padilla’s vagina could not be interpreted reliably because of insufficient DNA for the necessary [565]*565controls in the testing procedures; the uncertified results, too, were inconclusive.
Dr. Lowell Levine, the State’s forensic expert in odontology, compared photographs of the marks on Padilla’s chin and breast to molds of Fortin’s teeth. Levine concluded to a “high degree of probability” that Fortin made the bite marks found on Padilla’s chest. Levine, however, conjectured that Fortin “could have” been responsible for the bite mark on Padilla’s chin. Dr. Norman Sperber, the defense’s forensic odontologist, stated that bite-mark comparison is an imprecise science, far less reliable than DNA analysis and identification through dental records. Sperber opined that the injuries to Padilla’s breast and chin probably were not bite marks and, if they were, they could not be attributed to Fortin.
The State introduced Robert R. Hazelwood, a retired FBI agent and expert in violent sexual crimes, to catalogue the similarities between the crimes committed against Trooper Gardner and Padilla. The purpose of Hazelwood’s testimony was to show that the manner in which the two crimes were committed was so unique that only one person committed both crimes. That Fortin had sexually assaulted Trooper Gardner was not disputed.
At trial, Hazelwood focused on motive, modus operandi, and signs of ritual, finding unique similarities between the two crimes on all three grounds. First, Hazelwood concluded that both crimes were motivated by anger. In support of that conclusion, Hazelwood cited the evidence that both Padilla and Gardner were severely beaten, both were bitten and manually strangled, and both suffered serious anal injuries.
Second, Hazelwood found seventeen similarities in the modus operandi of the two crimes. The similarities were: (1) both crimes were “high risk” for detection, (2) committed impulsively, (3) against female victims, (4) of the same age range (25-34); (5) both were crimes of opportunity against victims who crossed the offender’s path, (6) adjacent to or on well-traveled roadways, (7) at night, (8) while the victims were alone, and (9) the attacks [566]*566occurred at the same location as the initial confrontation; (10) both crimes involved the use of blunt force consistent with blows from fists, (11) without weapons, (12) that caused primarily facial trauma, and (13) broken noses; (14) in both crimes the victims were undressed from the waist down, (15) their undergarments were found inside their pants or shorts, (16) their shuts were left on but their bras removed, and (17) there was the absence of any fresh seminal fluid in or on their bodies. Hazelwood testified that he had never before seen all seventeen of these characteristics present in any crime other than those committed against Padilla and Gardner.
Finally, Hazelwood testified about ritualistic behaviors present in both crimes. He defined a ritual as a “repeated pattern of behavior” “comprised of those acts unnecessary to the commission of the crime” that “complement! ] the underlying motivation of the crime.” According to Hazelwood, rituals are “designed for one single purpose, psychosexual gratification.” Hazelwood found “five ritualistic behaviors that were similar between the two crimes”: (1) bite marks to the chins, (2) bite marks to the left breast, (3) injurious anal penetration, (4) facial battering, and (5) manual, frontal strangulation. Hazelwood concluded that he had not seen the same combination of ritualistic behaviors in his work over the course of his thirty-year career. He also stated that he had never seen the particular combination of modus opemndi and ritualistic behaviors “in any other crime and I’ve never heard of it and I’ve never read of it.”
On September 6,1995, a Middlesex County Grand Jury indicted Fortin for the murder of Melissa Padilla by his own conduct, N.J.S.A. 2C:ll-3c(a)(l); felony murder, N.J.S.A. 2C:ll-3a (two counts); first-degree robbery, N.J.S.A. 2C:15-1; and first-degree aggravated sexual assault, N.J.S.A. 2C:14-2a. (Fortin App. Div. at 581). On March 18, 1997, the State announced its intention to seek the death penalty by filing a Notice of Aggravating Factors, alleging three such factors: the murder involved aggravated assault or torture, N.J.S.A. 2C:ll-3c(4)(c); the murder was commit[567]*567ted to escape apprehension for another offense, N.J.S.A. 2C:11-3c(4)(f); and the murder was committed during the commission of, or attempt to commit, or flight after committing robbery and sexual assault, N.J.S.A. 2C:ll-3c(4)(g).
In a pre-trial hearing, the trial court ruled admissible pursuant to N.J.R.E. 404(b) the other-crime evidence of Fortin’s sexual assault of Trooper Gardner and Hazelwood’s expert testimony. State v. Fortin, 162 N.J. 517, 523-24, 745 A.2d 509, 512 (2000) (Fortin I)- In view of that ruling, the court granted defendant’s motion to empanel separate juries for the guilt and penalty phases. On interlocutory appeal, the Appellate Division affirmed the 404(b) ruling, but concluded that Hazelwood’s analysis was not sufficiently reliable to be admitted as expert testimony. Id. at 524-25, 745 A.2d at 512-13. This Court granted leave to appeal and affirmed the Appellate Division with one significant modification. We concluded that Hazelwood could testify as “an expert in criminal investigative techniques,” but could not “testify on the ultimate issue of whether the person that assaulted Trooper Gardner [was] the same person that murdered Melissa Padilla.” Id. at 528-29, 745 A.2d at 515.
The guilt phase was tried to a jury between November 2 and December 7, 2000. The jury convicted Fortin of capital murder, aggravated sexual assault, first-degree robbery, and both counts of felony murder.
A separate jury was impaneled for the penalty trial, which was conducted between February 14 and February 26, 2001. That jury unanimously found two aggravating factors: aggravated assault and torture, and a murder committed during the commission of, or attempt to commit, or flight after committing robbery and sexual assault. The jury rejected the aggravating factor that the murder was committed for the purpose of escaping detection. The jury unanimously found the mitigating factors that Fortin was an abuser of drugs, including cocaine, marijuana, and heroin, and an abuser of alcohol for a substantial period of time, and other factors relevant to his record, his character, or the offense. The [568]*568jury concluded that the two aggravating factors outweighed the mitigating factors beyond a reasonable doubt. Consequently, the trial court filed a judgment of capital conviction on the date of the jury’s verdict and sentenced Fortin to the penalty of death.
On April 24, 2001, the trial court sentenced defendant on the non-capital convictions. The court merged the two felony murder counts into the capital murder conviction. Fortin was sentenced to twenty years imprisonment with ten years of parole ineligibility on the aggravated sexual assault conviction and to a consecutive twenty years imprisonment on the first-degree robbery conviction. Both of those sentences were made to run consecutive to the twenty-year sentence Fortin received in Maine for the crimes he committed against Trooper Gardner.
Defendant appeals to this Court as of right under Rule 2:2-1 (a).
GUILT PHASE
II.
Voir Dire
Defendant did not contest that he had committed the savage sexual assault on Trooper Gardner in Maine. With defendant clearly identified as the perpetrator of that brutal crime, the State’s central theory was that whoever attacked Trooper Gardner also attacked Melissa Padilla because the distinctive characteristics of the sexual assaults against Gardner and Padilla were so “bizarre” and “unique” as to constitute the signature of a single individual.
In Fortin I, supra, this Court recognized that the N.J.R.E. 404(b) “other-crime” evidence — the depraved attack on Trooper Gardner — had the clear capacity to inflame the passions of the jury and cause irreparable prejudice if not used for the limited purpose of establishing identity. 162 N.J. at 534, 745 A.2d at 518-19. In that regard, we proposed a limiting instruction to be given to the guilt-phase jury to ensure the proper use of that evidence. [569]*569Id. at 534-35, 745 A.2d at 518-19. The purpose of the instruction was to make certain that the jury did not convict defendant of the Padilla murder solely because he “committed another crime elsewhere” and, was, therefore, “a bad man with a propensity for crime.” Id. at 535, 745 A.2d at 519. The assault on Trooper Gardner was not relevant to any aggravating factor in the penalty phase. The trial court considered evidence of the assault on Gardner so potentially prejudicial that it bifurcated the penalty phase — at which that evidence was not admissible — from the guilt phase. Even a limiting instruction was not considered an adequate safeguard to protect penalty-phase jurors from the taint of such powerful and irrelevant evidence.
Against that background, the selection of the jury in the guilt phase of this capital murder trial proceeded. The trial court turned aside defendant’s repeated requests that the court advise the juror panelists that they would hear evidence of a sexual assault committed by defendant against a Maine State Trooper, and that it instruct them on the consideration to be given to that evidence. Defendant contends that the trial court’s failure to ask prospective jurors whether that evidence would impair their ability to remain fair and impartial and abide by the court’s instruction on the limited use to be given to that evidence denied him the opportunity to “intelligently exercise” his for-cause and peremptory challenges. He claims the court’s dereliction violated his right to a fair and impartial trial under the Federal and State Constitutions. U.S. Const, amends. VI, XIV; N.J. Const, art. I, ¶ 10.
A.
In reviewing that claim, we first look to the three-day jury selection process. The trial court used a struck jury system in which peremptory challenges were made only when sixteen potential jurors had been qualified to sit in the jury box. In all, three panels of prospective jurors were required to select sixteen jurors. When one panel of jurors was exhausted by removals either for [570]*570cause or by peremptory challenges, another was brought to the courtroom, until the process was concluded.
At the outset, the court read to each panel of prospective jurors the indictment, which alleged that defendant had robbed, sexually assaulted, and murdered Melissa Padilla. The court then gave an opening statement, explaining the voir dire process, and encouraged the juror panelists to speak their minds freely. The court asked a number of general questions, including whether the jurors had prior knowledge of the case, whether any of them had a law enforcement background or a close relationship with law enforcement agents, and whether the nature of the charges would prevent them from making a fair and impartial determination based solely on the evidence. The court also explained that the jurors might “hear evidence of some other offense being committed,” but that the evidence would not be introduced to “inflame” them or to demonstrate that the “person is a bad person.” That abstract, hypothetical reference to other-crime evidence did not refer in any way to defendant. At no time did the court tell the jurors that they would hear evidence of defendant’s brutal attack on a female state trooper in Maine.
Three times before voir dire began, defendant requested that the trial court question potential jurors about their ability to remain impartial and follow limiting instructions in light of the evidence of defendant’s crime in Maine. Defense counsel proposed a seven-paragraph question for that purpose, which we repeat in pertinent part:
This case will involve allegations by the State that [defendant] murdered Melissa Padilla ... in Woodbridge, New Jersey on August the 11th, 1994.
A fundamental issue in this case will be did [defendant] commit this crime and will the State prove he committed the crime beyond a reasonable doubt.
As part of its proofs, the State will produce evidence of a crime the defendant committed in the State of Maine. You will be instructed that the evidence you will hear' about the incident in Maine can be considered by you for a limited purpose. ...
I will ask you to think about the Maine evidence in conjunction with the limiting instruction. I will then ask if you can follow this limiting instruction given the nature of the Maine evidence. And just as importantly, whether you can be a fan- [571]*571and impartial juror in this case and decide whether the State has proven [defendant] is guilty beyond a reasonable doubt of killing Melissa Padilla, not based on the fact he might be a bad person who committed a serious crime in Maine.
The evidence you will hear is that on April the 3rd, 1995, about eight months after Padilla was killed, Trooper Vicki[ ] Gardner, while acting as a State Trooper was assaulted by [defendant]. He hit Gardner a number of times in the face, fracturing her nose. He bit and sexually assaulted Trooper Gardner by digitally penetrating her both vaginally and anally. Trooper Gardner was also strangled. She survived the incident with a fractured nose, numerous bumps, bruises, and scrapes to basically her entire body.
I’m going to ask all of you individually what your reaction was to the evidence and the instructions, whether you feel you can follow the instructions given the nature of the Maine evidence, and whether you can give [defendant] a fair trial despite all you will hear about the Maine incident.
The two prosecutors representing the State, one of whom was a seasoned capital litigator, raised no objection to the proposed instruction, except to insist on the expansion of paragraph five to include language that defendant had bitten Trooper Gardner on the chin and breast, and that the trooper had suffered those bite marks, as well as injuries to her anus and vagina.
The trial court clearly was vexed that the attorneys had agreed to inquire about a subject that would consume additional time in selecting a jury, asking: “Gentlemen, do we really want to get a jury in this case?” The court stated that generally it would abide by an agreement of the parties on the need for a particular juror inquiry, but that its interpretation of State v. Manley, 54 N.J. 259, 255 A.2d 193 (1969), led it to foreclose any disclosure of defendant’s sexual assault on Trooper Gardner. The court focused on Manley’s call for “ ‘an expedient selection of a fair and impartial jury,’ ” and its disapproval of the improper use of voir dire to give a favorable spin to a party’s preferred view of legal principles and the facts. (Quoting id. at 280, 255 A.2d at 205). The trial court expressed its concern that potential jurors might not be able to “keep an open mind” after hearing about “one isolated incident,” the assault on Trooper Gardner, and feared that some jurors might say to themselves, “oh, my goodness.” Evidently, the court believed that asking jurors, after disclosure of such a prior crime, “now can you be absolutely fair in this case?,” was a pointless [572]*572exercise. The trial court considered defendant’s voir dire instruction to be “loading the deck,” and refused to “put a layperson through having to in seconds make that determination, to assimilate all of that information in a vacuum and then say, well, that’s not going to have any effect on me.”
Throughout the jury selection process, defendant continued to press the court to give the proposed instruction, fearing that the nature of his crime against a female trooper in Maine would render some jurors incapable of returning a fair verdict, particularly those with law enforcement ties. After the trial court excused for cause four prospective jurors who could not remain fair and impartial given the nature of the charges in the Padilla case, the court denied another request by defense counsel for voir dire on the Maine crime. One potential juror, L.D., informed the court at sidebar that she had learned “all about the case,” including the “female State Trooper in Maine,” from newspaper reports. The court excused L.D. for cause sua sponte, because L.D. had formulated a “pretty good” opinion about the case. Afterwards, the court excused two prospective jurors who could not maintain their impartiality in light of the murder charges, and seven jurors whose law enforcement ties affected their ability to be fair.
Defendant renewed his voir dire request and expressed concern about shielding prospective jurors from evidence of the sexual assault against a “female State Trooper.” Defendant inferred that if a number of jurors could not keep an open mind after hearing the nature of the charges regarding the Padilla murder, then others might similarly be affected if they knew of defendant’s crime in Maine. Moreover, defendant was uncertain whether “anybody with a close connection to law enforcement” could sit impartially as a juror. Defense counsel urged the court to reconsider its position:
Once, again, I’m asking the court to please inform the jurors what they’re going to hear from the State of Maine. So[ ] we have the ability to judge the impact that this evidence is going to have on any potential juror____So, at least, the exercise of our peremptory challenges are done with a knowledge of ... whether anybody is going to be biased by .. hearing [about] the Maine case.
[573]*573The State did not resist defendant’s request. The trial court, however, was not persuaded:
I hope that this is the last time I have to address this business of whether or not I should set forth, with particularity, the details of the 404[(b) evidence]____There is no way that I could present that in a vacuum. It would be presented without this jury having heard the first word of evidence.
To hit them with what the defendant’s alleged conduct was in Maine, would be totally unfair. It would truly affect the ability of any juror to be fail- and impartial. When this evidence is presented in the proper context, with the proper instructions, when they already have heard evidence concerning the charge against the defendant, ... they will know how to put it in that prospective [sic].
The court excused another juror, C.F., for cause sua sponte because he too remembered from reading the newspapers “something about” defendant’s Maine crime against a state trooper, and admitted that he would “probably have some prejudice.” Undeterred, defendant renewed his voir dire request, only to have it denied again.
The court also denied defendant’s request that it question J.B., a panelist with “numerous friends” in local law enforcement, as to whether the Maine-crime evidence would undermine his ability to remain impartial. The court reasoned that J.B., like the other panelists with close ties to law enforcement, had not indicated that he had “close/mate friends ... [in] law enforcement,” and that it did not “follow” that a juror with such ties would be “more aggrieved at any law enforcement person being assaulted.” (Emphasis added). The court also denied defendant’s request to excuse for cause R.S., an East Jersey State Prison plumber who had twice been assaulted by inmates and whose daughter had been dating a Carteret police officer for ten years, and C.M., a New Jersey State Prison corrections officer in Trenton with an uncle on the Sayreville police force.
Defendant exercised the last of his twenty peremptory challenges to remove R.S. from the panel. Although defendant requested three additional peremptory challenges, the court granted only one to allow the removal of C.M. Defendant was not permitted the additional two challenges requested to “make up for the one ... used for [R.S.],” and to remove M.C., who was related by [574]*574marriage to a state trooper, and was “nervous” about the ease because she frequented the area where Padilla had been killed.
In all, the trial court questioned 154 potential jurors. The court removed many jurors for cause, including twenty-seven who had ties to law enforcement, and twenty-two who admitted that they could not be impartial after hearing the nature of the charges in the Padilla case. Of the eighteen jurors with law enforcement connections not disqualified by the court for cause, the defense removed ten, and the State two, by peremptory challenges. Thus, six individuals with law enforcement ties became sworn jurors, and five of those six became deliberating jurors.4
After the jury was sworn and impaneled, the prosecutor gave his opening remarks, in which he predictably and properly discoursed on the Maine crime and its relationship to the State’s other proofs. The prosecutor described both Padilla’s murder and defendant’s attack on Trooper Gardner in Maine:
He attacked that female State Trooper. He beat her. He sexually assaulted her. He strangled her into unconsciousness. And, in a bizarre and unique kind of attack, he bit her on the left breast. And he bit her on the chin. And ... he forced something into her anus that caused the flesh to lacerate.
After the State’s opening, the court gave a limiting instruction to the effect that the Maine evidence could only be considered in determining the identity of Padilla’s killer, and that it could not be considered as evidence that defendant was “a bad person, with a propensity for committing bad acts.” The late timing of the disclosure and the limiting instruction, however, already had denied the court and the parties the opportunity to learn whether the sixteen sworn jurors would have answered any of the voir dire [575]*575questions differently had they known that they were to receive evidence of defendant’s sexual assault on Trooper Gardner.
B.
Our State and Federal Constitutions guarantee the right to trial by an impartial jury. U.S. Const, amends. VI, XIV; N.J. Const, art. I, ¶ 10. “[A]n impartial jury is a necessary condition to a fair trial” in our constitutional framework. State v. Williams, 113 N.J. 393, 409, 550 A.2d 1172, 1179 (1988) (Williams II) (citing Sheppard v. Maxwell, 384 U.S. 333, 362-63, 86 S.Ct. 1507, 1522-23, 16 L.Ed.2d 600, 620 (1966)). Jurors, therefore, must be “as nearly impartial as the lot of humanity will admit.” State v. Williams, 93 N.J. 39, 60, 459 A.2d 641, 652 (1983) (Williams I) (internal quotation marks omitted). The “requirement of fairness — and particularly jury impartiality — is heightened in eases in which the defendant faces death.” Id. at 61, 459 A.2d at 652. A trial is poisoned at its inception if the jurors deciding the case cannot review the evidence dispassionately, through the light of reason.
The trial court’s duty “to take all appropriate measures to ensure the fair and proper administration of a criminal trial” must begin with voir dire. Id. at 62, 459 A.2d at 653. A “vital aspect” of that responsibility is to ensure the impaneling of only impartial jurors by ferreting out potential and latent juror biases. Id. at 62-63, 68, 459 A.2d at 653. To carry out that task, a thorough voir dire “should probe the minds of the prospective jurors to ascertain whether they hold biases that would interfere with their ability to decide the case fairly and impartially.” State v. Erazo, 126 N.J. 112, 129, 594 A.2d 232, 241 (1991). Although a trial court’s exercise of its broad discretionary powers in conducting voir dire “will ordinarily not be disturbed on appeal,” Williams II, supra, 113 N.J. at 410, 550 A.2d at 1180 (internal quotation marks omitted), we have not hesitated to correct mistakes that undermine the very foundation of a fair trial — the selection of an impartial jury.
[576]*576In this case, the other-crime evidence was not only the most critical component of the State’s case, but the evidence most likely to inflame a jury and render it incapable of reasoned analysis. It may very well be that some jurors, given the shocking nature of the attack on Trooper Gardner, would have been incapable of honoring the court’s limiting instruction and would have presumed guilt based on that crime alone. That concern was more than a theoretical possibility. Two prospective jurors who read accounts of defendant’s crime in Maine confided to the trial court that they would be incapable of rendering a fair verdict.
The trial court claimed that it was carrying out the mandate of Manley, supra, by rejecting defendant’s proposed voir dire questions. 54 N.J. 259, 255 A.2d 193. We, therefore, first look to Manley to determine whether the court, in exercising its discretion, was true to the holding of that case. In Manley, the defendant, who was on trial for non-capital murder, sought to query prospective jurors about their reactions to his prior conviction for atrocious assault and battery. Id. at 263-64, 255 A.2d at 195-96. The court precluded that questioning because the prior criminal conviction would become evidential for impeachment purposes only if the defendant took the stand to testify, and the defendant refused to commit to testifying at so early a stage in the proceedings. Id. at 264-65, 255 A.2d at 196. Accordingly, at the time of jury selection, the admissibility of the prior conviction was contingent on an uncertain event — whether the defendant would elect to testify. The trial court, “by refusing to allow defendant to admit the previous conviction while questioning prospective jurors,” was protecting the defendant from irremediable prejudice in the event that he did not testify. Id. at 270-71, 255 A.2d at 199. Under those circumstances, this Court held that the trial court did not abuse its discretion by refusing to permit voir dire regarding the defendant’s prior conviction, because the defendant had not indicated a present intention to take the stand at trial. Ibid.
Manley emphasized that trial courts are vested with wide discretion to determine the line of inquiry at voir dire. Id. at 269, [577]*577255 A.2d at 198-99. For example, had the defendant indicated an intention to testify, it would have been within the discretion of the trial court to allow the lawyers to question potential jurors about the defendant’s prior conviction. Ibid. Preserving the trial court’s discretion was critical because even had the defendant indicated an intention to testify during jury selection, he could not be bound by that decision.
In addition to deciding the precise issue in Manley, this Court took the opportunity to address endemic abuses in the manner in which lawyers conducted voir dire at that time. Id. at 271-83, 255 A.2d at 199-206. The Court’s objective was “to draw some restrictive guidelines for future control of the jury-drawing process.” Id. at 263, 255 A.2d at 195. Given the prevalence of “voir dire examination of jurors at inordinate length and on improper subjects,” ibid., this Court intended to eliminate counsels’ efforts at partisan persuasion and indoctrination of prospective jurors, and their use of “the hypothetical question intended and so framed as to commit or to pledge jurors to a point of view.” Id. at 281, 255 A.2d at 205.
In recent years, we have taken occasion to correct the misapplication of Manley by trial courts in capital cases. See, e.g., State v. Biegenwald, 126 N.J. 1, 33, 594 A.2d 172, 188 (1991) (Biegenwald IV) (“Regrettably, we perceive from the records in many of the cases coming before us that trial courts have read Manley ... to limit voir dire to the bare minimum necessary to qualify a juror.”); State v. Moore, 122 N.J. 420, 455, 585 A.2d 864, 882 (1991) (“Although Manley may be read as discouraging [the questioning of prospective jurors concerning their understanding of the burden of proof and presumption of innocence] ... capital cases require a thorough and searching inquiry in regard to voir dire.”) (internal quotation marks omitted). Once again, we do so here. In capital cases, “[c]ounsel must be afforded the opportunity for a thorough voir dire to evaluate and assess jurors’ attitudes in order to effectively participate in jury selection. If counsel is unable to screen out prejudice and bias, that inevitably leads to [578]*578unfair juries.” Williams II, supra, 113 N.J. at 409, 550 A.2d at 1179. We are unwilling to undermine the integrity of the trial process, even where the evidence of guilt is compelling. Ibid. The right to a fair trial does not depend on the nature of the crime charged or the quantum of evidence produced against a defendant. Ibid.
In Biegenwald IV, supra, we held that, in the penalty phase of a capital ease, the trial court’s refusal to permit voir dire of the defendant’s other murder convictions denied the defendant his right to a fair and impartial jury. 126 N.J. at 34-35, 594 A.2d at 188-89. By prohibiting inquiry about the defendant’s prior murder convictions that were certain to be introduced in support of the N.J.S.A. 2C:11-3c(4)(a) aggravating factor, the court denied both itself and counsel the opportunity to search for impartial jurors who were willing and able to follow the law and the court’s instructions. 126 N.J. at 32, 594 A.2d at 187-88. Absent that inquiry, neither court nor counsel could exercise for-cause and peremptory strikes “intelligently and effectively.” Ibid.
In Moore, supra, another capital case, the defendant was convicted of murdering his pregnant wife and her eighteen-month-old son with a hammer. 122 N.J. at 427, 430, 585 A.2d at 867-68. At times during jury selection, the trial court resisted the questioning of prospective jurors concerning the victims’ status. Id. at 447-48, 585 A.2d at 877-78. For example, the court, on occasion, would not permit a response to the question, “Would the fact that one of the victims here was a child influence you so that it would be more likely that you would impose the death penalty?” Id. at 448, 585 A.2d at 878. We found nothing improper in asking such a question, and also found that a response indicating a juror was more likely to convict on the basis of a victim’s status, if not providing grounds to excuse for cause, might provide a reason for counsel to exercise a peremptory challenge. Ibid. Although we reversed the defendant’s convictions on other grounds, we noted that the trial court had “overread” Manley as prohibiting such inquiries, and offered guidance for future capital trial voir dires [579]*579concerning the impact a victim’s status might have on prospective guilt- or penalty-phase jurors. Id. at 446-51, 585 A.2d at 877-80. A prospective juror’s bias in favor of conviction or the death penalty based on a victim’s status, as noted, is a legitimate ground for the exercise of a peremptory challenge. Id. at 448, 585 A.2d at 878. We instructed that “open-ended questioning” should be permitted on the issue of victim status “as it relates to any prejudice or predisposition affecting the juror’s ability to adjudge fairly in the guilt phase or the ability to consider mitigating evidence in any penalty phase.” Id. at 451, 585 A.2d at 880. Those inquiries need not lead to removal of a juror for cause, but may impel one of the parties to exercise a peremptory challenge. Ibid.
In this case, the holding in Manley does not support the trial court’s categorical rejection of inquiry into the jurors’ ability to remain fair and impartial and to follow the court’s limiting instructions in light of the evidence of defendant’s sexual assault of Trooper Gardner. Unlike Manley, in this case, the introduction at trial of that other crime was not a possibility, but a certainty. Indeed, the other-crime evidence was central to the State’s case. The success of the prosecution hinged on whether the similarities between the attack on Trooper Gardner and the attack on Padilla established the identity of Padilla’s killer. Although we do not approve of all of the language in defendant’s proposed voir dire instruction — e.g., “I’m going to ask all of you individually what your reaction was to the evidence and the instructions” (emphasis added) — the purpose of the requested inquiry was to ferret out juror bias.
Defendant’s proposed voir dire was the antithesis of the “hypothetical question” we intended to foreclose in Manley. There was nothing contingent about the admissibility of the other-crime evidence, and nothing conjectural about its power to evoke a visceral and emotional response from jurors. The trial court’s refusal to .make any inquiry, much less a searching one, of the other-crime evidence is similar to the constitutionally-flawed pro[580]*580cess we condemned in Biegenwald IV. In that case, the trial court precluded questioning of prospective penalty jurors concerning the defendant’s prior murder convictions that the State introduced as an aggravating factor in support of a sentence of death. Biegenwald IV, supra, 126 N.J. at 32, 594 A.2d at 187-88.
Reason and experience tell us that prospective jurors have varying thresholds for processing and reacting to evidence. Most prospective jurors, even when confronted with shocking evidence related to a brutal crime, presumably will be able to follow the court’s instructions and render a fair and impartial verdict. Some jurors, however, will be so disturbed or repulsed by the gruesome details of a crime that they will lose their ability to be objective and will be incapable of dispassionate consideration of the evidence. For the most part, those jurors will be honest and forthcoming in response to direct questions by the court. Our courts must not be fearful of asking those questions out of concern that jury selection will be protracted. The disclosures by the two jurors who read accounts of defendant’s crime in Maine and who freely admitted their inability to remain impartial based on that information should have suggested to the trial court that, given the explosive nature of the 404(b) evidence, not all jurors would be capable of following the limiting instructions on the use of that evidence.
The court and the parties needed to know whether the jurors could resist the temptation to consider the heinous assault on Trooper Gardner as proof of propensity to commit a crime, rather than solely as proof of the identity of Padilla’s killer. That was reason enough to permit the voir dire on the subject, given the number of jurors who were excused for cause because they could not remain impartial based on the nature of the crimes committed against Padilla. But here, the potential prejudice was compounded incalculably by the fact that the other-crime evidence was that defendant savagely, sexually assaulted a law enforcement officer. Six jurors who sat on the case, five of whom became deliberating jurors, had ties to law enforcement personnel, but were never [581]*581asked whether, in light of those relationships, they could remain fair and impartial and follow the limiting instructions for the use of the other-crime evidence. The absence of any voir dire on that subject deprived the trial court of the opportunity to strike for cause, and defendant of the opportunity to challenge peremptorily those jurors with law enforcement connections who may have revealed latent biases had they known the nature of the other-crime evidence to be introduced at trial.
In Moore, supra, we approved of inquiring about a juror’s ability to remain impartial in view of a victim’s condition or status. 122 N.J. at 451, 585 A.2d at 880. We acknowledged the importance of the trial court and parties learning whether jurors harbored any prejudice in such circumstances that would predispose them toward rendering a particular verdict. Id. at 448, 585 A.2d at 878. Such reasoning applies with equal force to the victim-status of Trooper Gardner. There was a need to know whether jurors, particularly those with law enforcement ties, could remain open-minded in a case in which evidence of defendant’s sexual assault on a female state trooper was certain to be introduced.
Expedience can never trump the considered and thoughtful selection of jurors whose impartiality and fairness must be beyond reproach. The extra time necessary to impanel twelve dispassionate jurors in this case would have been a small price to pay for the assurance of a fair trial.
C.
We conclude that the trial court improperly limited the scope of voir dire. The court denied defendant his right to a fair trial by depriving him of the opportunity to discover which jurors would be unwilling or unable to remain impartial and follow the court’s limiting instructions after hearing the highly inflammatory evidence about defendant’s crime against Trooper Gardner. We, therefore, vacate defendant’s convictions and remand to the Law Division for a new trial consistent with this opinion.
[582]*582III.
Expert Testimony of Robert Hazelwood
Robert Hazelwood testified as the State’s expert witness on the subject of “violent sexual crimes” to aid the prosecution in identifying defendant as the murderer of Melissa Padilla. Hazel-wood presented impressive credentials. He served thirty-three years in law enforcement, including twenty-three years at the Federal Bureau of Investigation. During his tenure with the FBI, Hazelwood was a member of the Behavioral Science Unit, where he studied serial sex offenders and killers. By his own account, Hazelwood had participated in more than 7,000 violent crime investigations, the majority of which were sex-related. He had authored forty articles in peer-reviewed journals on violent crime and five books on criminal and sexually-deviant behavior, and had lectured on such subjects to law enforcement agencies and as an adjunct faculty member at several universities. Upon being qualified as an expert, Hazelwood compared what he claimed were the unique and distinctive characteristics of defendant’s sexual assault on Trooper Gardner to those of Padilla’s murder for the purpose of allowing the jury to infer that only one person committed both crimes. Hazelwood stated that he had never seen, heard, or read of “this combination of behaviors in any other crime.”
Relying on Fortin I, supra, 162 N.J. 517, 745 A.2d 509, defendant contends that Hazelwood never produced a database of cases from which he made his comparisons and derived his conclusions, as ordered by this Court as a pre-condition to his testimony. Accordingly, defendant argues that the trial court should not have permitted Hazelwood to testify in light of his failure to comply with this Court’s discovery order. Without the database, defendant concludes that he was denied, in essence, his constitutional right to confront Hazelwood on the terms required by this Court and, therefore, his right to a fair trial.
In assessing defendant’s claim, our review begins with Fortin I. In State v. Fortin, supra, the Appellate Division held that the [583]*583sexual assault of Trooper Gardner was admissible to prove identity under N.J.R.E. 404(b) and that Hazelwood was not qualified to give expert testimony through the use of linkage analysis. 318 N.J.Super. at 580, 724 A.2d at 820. This Court agreed with the Appellate Division that the “proposed expert testimony of Hazel-wood concerning linkage analysis lack[ed] sufficient scientific reliability to establish that the same perpetrator committed the Maine and New Jersey crimes.” Fortin I, supra, 162 N.J. at 525, 745 A.2d at 513. “Linkage analysis” is the comparison of two or more crimes for common characteristics of a unique or distinctive nature that permits a trained investigator to conclude that the same perpetrator committed the crimes. Id. at 522-23, 745 A.2d at 511— 12. This Court concluded that linkage analysis had not attained a sufficient level of scientific reliability to warrant its admissibility at trial. Id. at 525, 745 A.2d at 513. This Court found that it “is a field in which only Hazelwood and a few of his close associates are involved.” Id. at 527, 745 A.2d at 514. As such, “there are no peers to test his theories and no way in which to duplicate his results.” Ibid.
On the other hand, this Court concluded that Hazelwood, based on his experience, could testify as an expert in criminal investigative techniques and, as such, could discuss similarities between the crimes, provided he did not “draw[ ] conclusions about the guilt or innocence of the defendant.” Id. at 528, 745 A.2d at 515.5 Hazel-wood specifically was enjoined from “testify[ing] on the ultimate issue of whether the person that assaulted Trooper Gardner [was] the same person that murdered Melissa Padilla.” Id. at 528-29, 745 A.2d at 515. This Court recognized the potential for reflexive acceptance and misuse of Hazelwood’s testimony given his authoritative credentials and the seeming application of scientific-like analysis that undergirded his opinions. Id. at 533, 745 A.2d at 518. Nevertheless, this Court reasoned that Hazelwood’s testimo[584]*584ny could be helpful to a jury in showing that the evidence established an “unusual pattern,” provided “the witness can from a reliable database offer evidence that a combination of bite marks on the breast, bite marks on the chin, and rectal tearing inflicted during a sexual attack is unique in his experience of investigating sexual assault crimes.” Id. at 532, 745 A.2d at 517-18 (emphasis added). This Court emphasized that Hazelwood’s testimony required a foundation: “If there is such a database of cases, the witness’s premise can be fairly tested and the use of the testimony invokes none of the concerns that we have expressed about the improper use of expert testimony.” Id. at 533, 745 A.2d at 518 (emphasis added).
This Court clearly set the production of a reliable database as an essential qualifier to Hazelwood’s testimony. The approach outlined by the Fortin I Court would ensure that his crime-scene comparison techniques would be subject to verification, allowing the defense a fair opportunity to test his methods and credibility in the crucible of cross-examination. The issue now to be resolved is whether Hazelwood provided to the defense the database that was contemplated in Fortin I.
After the decision in Fortin I, the defense requested a “comprehensive listing” of the 4,000 cases referred to in Hazelwood’s motion testimony, including the names of the cases, their locations, copies of police reports, the evidence reviewed by Hazelwood, and copies of his interviews. The defense also requested a listing of the crime scenes Hazelwood had visited and any database he had relied on in formulating his opinion on the unique characteristics between the Gardner and Padilla crimes.
Hazelwood responded, through the prosecutor, that he neither had a list of the files of those cases that he had investigated during his years in law enforcement, nor access to them, and that “[n]o database, evidence or scientific studies were reviewed in forming [his] opinion.” He professed to have “relied upon [his] experience, education and training in arriving at [his] opinion.” The prosecutor explained that Hazelwood’s opinion would be “based on his [585]*585life’s work as a military policeman, FBI special agent, violent crime analyst, sex crimes specialist, behavioral scientist, author, consultant on violent sex crimes around the world and distinguished national and international lecture[r] on violent sex crimes.” Defendant responded by moving for either the production of the database or the preclusion of Hazelwood’s testimony.
At a hearing addressing the defense’s discovery request, the trial court expressed the view that it was “entirely reasonable” that Hazelwood did not have a list of cases, just as any lawyer “would be hard pressed to go back” and provide a list of cases in which he had been counsel from the beginning of his career. The court noted that experts often testify from their experience without having to provide lists or details of their previous cases. The court stated that it interpreted Fortin I’s reference to “reliable database,” to mean a trustworthy database, and that defendant’s argument addressed the weight to be accorded Hazelwood’s testimony, not its admissibility. The trial court emphasized that defense counsel could cross-examine Hazelwood about the number of cases he had reviewed that involved strangulation and bite marks.
In his trial testimony, Hazelwood identified seventeen similarities in the modus operandi and five ritualistic behaviors similar between the Gardner and Padilla crimes. See supra Part I. In his trial testimony Hazelwood defined modus operandi as “the behaviors necessary to commit the crime and the consequences of those behaviors.” Modus operandi is a “learned behavior” that develops based on the criminal’s ongoing “experience,” “education,” “maturity,” and “the demands of the crime.” Hazelwood claimed that he had never “seen all 17 of these [modus operandi] characteristics together” other than in the Gardner and Padilla crimes.
According to Hazelwood, ritualistic behaviors are “static” and “repeated patterns of behaviors ... designed for one single purpose, psychosexual gratification.” Hazelwood found “five ritualistic behaviors that were similar between the two crimes.” See supra Part I. Hazelwood had never encountered that combination [586]*586of ritualistic behaviors before or a bite mark to the chin “in all of the 7,000 eases that [he had] worked on over more than 30 years.” Hazelwood had never seen, read, or heard of such similarities in modus opemndi and ritualistic behaviors in any other crime.
Although that conclusion came perilously close to a prohibited ultimate-issue opinion — an opinion as to the guilt of defendant — • Fortin I permitted such testimony provided that Hazelwood was able “from a reliable database” to say that the combination of factors were “unique in his experience of investigating sexual assault crimes.” See Fortin I, supra, 162 N.J. at 532, 745 A.2d 509. The missing link here was the failure of Hazelwood to produce any database before or during his testimony.
This Court clearly stated in Fortin I that a “database of cases” would permit Hazelwood’s assumptions to be “fairly tested.” Id. at 533, 745 A.2d at 518. We cannot accept the State’s argument that Hazelwood provided a “reliable database” to support his opinion by reference to his expert report, his curriculum vitae, his publications, and his pretrial testimony, all of which were known to this Court at the time of Fortin I. See id. at 521, 745 A.2d at 511 (discussing Hazelwood’s credentials). Had any of those items qualified as the “reliable database,” this Court undoubtedly would have said so. Defendant was not required, as suggested by the State, to assemble Hazelwood’s database by researching his publications and tracking down all or some portion of the relevant 7,000 cases that he investigated over the course of his law enforcement career. Surely, Hazelwood — an author of five books and scores of articles, a university adjunct faculty member, a frequent lecturer, a former FBI agent, and former member of the FBI Behavioral Science Unit — could have compiled some manner of database of eases on which he had based his conclusions. We cannot agree with the trial court that Hazelwood’s reference to his experience, training, and education was a substitute for a “database of cases” or that the failure to provide such case information only went to the weight to be given to his opinion, rather than its admissibility.
[587]*587Hazelwood’s testimony, although presented as the application of criminal investigative techniques, was couched in the aura of science, more particularly, behavioral science. He was permitted to testify to his understanding of the state of mind of the perpetrator, who he described as impulsive, motivated by anger, and driven by the need for “psychosexual gratification.” A very thin line demarcated the boundary between linkage analysis, which this Court found not to have achieved an acceptable level of scientific reliability, and the uniqueness analysis that this Court permitted as a subject of expert testimony.
In Fortin I, this Court was concerned about providing some means by which to test the validity of Hazelwood’s conclusions. The database was to serve as the means by which those conclusions would be subject to verification on cross-examination. We cannot ignore the fact that Hazelwood’s testimony — that he had never seen, read, or heard of two crimes with the same characteristics in common — came close to identifying defendant as the murderer without uttering the words. That powerful testimony was presented through the quasi-science of uniqueness analysis. Hazelwood was offered as a respected authority. The danger that the Trooper Gardner assault could be used as propensity evidence, despite the best-crafted limiting instruction, was a grave concern. In Fortin I, this Court also recognized the danger that a jury could be swept along by testimony that drew even a facile connection between the two gruesome crimes. Id. at 534, 745 A.2d at 518 (citing State v. Stevens, 115 N.J. 289, 309, 558 A.2d 833, 844 (1989)).
Although Hazelwood’s uniqueness analysis may not sound strictly in scientific method, there are enough common elements to invoke the principles and, therefore, the protections of State v. Kelly, 97 N.J. 178, 478 A.2d 364 (1984). In Kelly, this Court cautioned that “[t]he technique or mode of analysis used by the expert must have a sufficient scientific basis to produce uniform and reasonably reliable results so as to contribute materially to the ascertainment of the truth.” Id. at 210, 478 A.2d at 380; see [588]*588also State v. Cavallo, 88 N.J. 508, 516, 443 A.2d 1020, 1024 (1982) (“[E]xpert testimony is admissible only if the expert has sufficient expertise to offer the intended testimony and the testimony itself is sufficiently reliable.”).
The Rules of Evidence, contrary to the State’s assertions, do provide the basis for the production of a database before an expert testifies. N.J.R.E. 705 states, “The expert may testify in terms of opinion or inference and give reasons therefor without prior disclosure of the underlying facts or data, unless the court requires otherwise. The expert may in any event be required to disclose the underlying facts or data on cross-examination.” (Emphasis added). Fortin I, supra, unmistakably required prior disclosure of a reliable database to ensure the validity of Hazel-wood’s testimony and to allay this Court’s concerns about its improper use. 162 N.J. at 533, 745 A.2d at 518.
The Fortin I Court provided two examples of a reliable database to validate an expert’s testimony. Id. at 532, 745 A.2d at 517-18. In State v. Zola, the State sought to prove that the defendant had sexually assaulted the victim by presenting the testimony of a forensic chemist, who opined that his tests on a vaginal sample from the victim indicated the presence of saliva. 112 N.J. 384, 408-09, 548 A.2d 1022, 1034-35 (1988), cert. denied, 489 U.S. 1022, 109 S.Ct. 1146, 103 L.Ed.2d 205 (1989). To support his opinion, the expert referred to his logbook of test results from approximately 1,644 other vaginal samples. The Court held that the State’s failure to make the database of those samples available to the defense before trial was harmless error. Id. at 410-12, 548 A.2d at 1035-36.
As a second example of a reliable database, this Court cited a table entitled “Incident Characteristics by Homicide Type, 1976-95 (in Percent),” in James Alan Fox and Jack Levin, Multiple Homicide: Patterns of Serial and Mass Murder, 23 Crime & Just. 407, 435 (1998). As the name suggests, that table tabulates thousands of homicides and displays incident characteristics— weapon used, victim-offender relationship, and circumstances-— [589]*589broken down by type of homicide. The database supported the authors’ conclusion that the firearm is the most effective means of mass destruction and the “weapon of choice” in both mass-murder incidents and single victim crimes. Id. at 434-35. Surely if thousands of murder eases and hundreds of tests performed on bodily fluids can be tabulated in a database, the basic information for a database in this case can be compiled as well.
Hazelwood’s database should have consisted of violent sexual assault cases that he had investigated, studied, or analyzed during his professional career, and the peculiar modus operandi and ritualistic characteristics of those crimes. Such a database would have provided some basis for verifying the frequency of sexual assaults in which perpetrators bite the faces or breasts of their victims, or manually strangle them, or engage in high risk attacks, to name but a few of the characteristics Hazelwood found distinctive in this case. If Hazelwood was correct about the unique combination of characteristics that the Gardner and Padilla assaults had in common, the database would have strengthened and validated his conclusions. The jury also was entitled to know if there were any flaws in his analysis.
We do not suggest that the database had to be comprised of all of the eases investigated, studied, or analyzed by Hazelwood, or even a majority of them. We understand that it might be overly burdensome or impossible to construct such a record if he were not keeping such records on a running basis and if he truly were denied access to the records by other law enforcement authorities. Hazelwood, however, holds himself out as an expert in this field and presumably has kept records for the purpose of conducting research, publishing articles and books, and presenting lectures. We believe that if he had the will to do so, he could provide some credible database for submission to the trial court.
The database, at a minimum, must permit an acceptable basis for comparison. We are not prepared on the present record to say what number of cases would constitute a sufficient database. That determination we leave to the trial court, which must conduct [590]*590a N.J.R.E. 104 hearing. At that hearing, the trial court must determine what number of cases can be reconfigured within reason and what number of case comparisons are necessary to give the opinion validity.
Accordingly, we conclude that the trial court committed reversible error in permitting Hazelwood to testify absent the production of a reliable database. On remand, the trial court will conduct a N.J.R.E. 104 hearing consistent with this opinion.
IV.
Third-Party Guilt
Defendant contends that the trial court improperly barred him from introducing evidence that would have supported an argument that someone else — a third party — killed Melissa Padilla. Defendant sought to introduce evidence that Padilla had sold crack cocaine earlier on the day of her murder for the purpose of drawing the inference that her death was drug-related. He also sought to show that unidentified sperm found in a vaginal swab from Padilla’s body could not have come from defendant and, therefore, must have come from her killer. In both instances, the trial court found that there was no rational basis to support a theory of third-party guilt. We agree.
The constitutional right to present a defense confers on the defendant the right to argue that someone else committed the crime. State v. Jimenez, 175 N.J. 475, 486, 815 A.2d 976, 982 (2003); State v. Koedatich, 112 N.J. 225, 297, 548 A.2d 939, 976 (1988) (Koedatich II), cert. denied, 488 U.S. 1017, 109 S.Ct. 813, 102 L.Ed.2d 803 (1989). In this case, defendant’s general denial by the entry of a not-guilty plea implicitly raised the defense of third-party guilt. In fact, during summation defendant argued that Hector Fernandez, Padilla’s live-in boyfriend, was the third party who killed Padilla. It appears that the third-party evidence [591]*591he now alleges was improperly excluded would have taken him in another direction.
The right to argue third-party guilt — that someone in general, or in particular, other than the defendant committed the crime — does not address whether specific evidence is admissible in support of such a defense. Evidence in support of third-party guilt, or any theory offered by the prosecution or defense, must satisfy the standards of the New Jersey Rules of Evidence. Scientific evidence offered through expert testimony must not only be relevant but also must meet the threshold of reliability.
There is no burden on the defendant to prove his innocence through the introduction of evidence of third-party guilt. The defendant does not have to show that the evidence supports a probability that another person committed the crime. Third-party guilt evidence “need only be capable of raising a reasonable doubt of defendant’s guilt” to warrant its admissibility. Koedatich II, supra, 112 N.J. at 299, 548 A.2d at 977 (quoting People v. Hall, 41 Cal.3d 826, 226 Cal.Rptr. 112, 718 P.2d 99, 103 (1986) (internal quotation marks omitted)). Such evidence cannot be withheld from the jury “if the proof offered has a rational tendency to engender a reasonable doubt with respect to an essential feature of the State’s case.” State v. Sturdivant, 31 N.J. 165, 179, 155 A.2d 771, 778 (1959), cert. denied, 362 U.S. 956, 80 S.Ct. 873, 4 L.Ed.2d 873 (1960). Stated more concretely, there must be “some link ... between the third party and the victim or crime,” Koedatich II, supra, 112 N.J. at 300, 548 A.2d at 978, “capable of inducing reasonable” people to regard the evidence “as bearing upon the State’s case,” Sturdivant, supra, 31 N.J. at 179, 155 A.2d at 778. The connection between the third party and the crime cannot be left to conjecture. Ibid. We recognize that a trial court’s decision regarding the admissibility of evidence is fact sensitive and, therefore, our review is deferential and limited to whether there has been an abuse of discretion. See Koedatich II, supra, 112 N.J. at 300, 548 A.2d at 977-78.
[592]*592B.
Defendant claims that the trial court erred in not permitting him to introduce undisputed evidence that Hector Fernandez had purchased $100 worth of crack cocaine for himself and Padilla on the day of the murder, and that Padilla had sold the drugs later that day. Defendant’s objective was to argue that Padilla was killed in a drug deal gone awry, or in the course of a robbery for drugs or drug money. Investigators learned of Padilla’s drug dealing from witnesses who were interviewed during the homicide investigation. However, no evidence was ever presented that Padilla had drugs in her possession when she left the Gem Motel to purchase food for her family and encountered her killer. No one testified that she was selling drugs at or about the time of her murder. Moreover, the scene of the crime failed to reveal any evidence of drugs or evidence suggestive of a drug sale. All that was found near the bloodied corpse were the remnants of Padilla’s purchases from the Quick Chek and a bloody dollar bill. The jewelry that she ordinarily wore was missing. Padilla’s killer had battered, sexually assaulted, and strangled her, inflicting traumatic injuries to her anus, but none to her vaginal area.
From this tableau, defendant argued that some unidentified denizen from that drug-infested area of Woodbridge Township near the murder scene killed Padilla. The State countered that no rational basis in the record supported the theory that Padilla’s death was in any way related to drug dealing. Moreover, the State indicated that if the defense were permitted to pursue a theory that a drug fiend killed Padilla, it would introduce defendant’s own significant history of drug addiction and abuse because witnesses placed him in close proximity to the scene at the approximate time of the murder.
The court found that there was no link between the evidence of the victim’s drug sales and the murder. Even if some connection had been made, the court concluded that the drug-dealing evidence would not tend to raise a reasonable doubt about the State’s case because defendant, based on his drug history, would have had [593]*593the same motivation and opportunity to rob and attack Padilla as the generic “crazed drug abuser.” In other words, assuming the crime was somehow related to Padilla’s drug dealing, defendant was squarely in the class of people — those with drug habits — who would have been suspected of committing the offense.
We agree with the trial court that the evidence did not suggest, even inferentially, that Padilla’s drug dealing was connected in any way to her murder. Even if we were to assume that there was evidential support for that theory, defendant would have fit the profile of the prototypical suspect, and the door would have been opened to his own extensive drug history. If Padilla’s drug dealing had any probative value, it was substantially outweighed by the risk of undue prejudice and confusion of the issues. See N.J.R.E. 403. We find that the drug-dealing evidence bore no relevance to any material issue in the case and, accordingly, did not have the capacity to raise a reasonable doubt of defendant’s guilt.
Defendant contends that the trial court improperly precluded him from introducing evidence concerning sperm found in the victim’s vagina that would have suggested a third party as the killer. We do not find that the trial court abused its discretion on the record before us. We note that the record presented by defendant is far from a model of clarity and the scientific issues were not elaborated with precision. On remand, defendant may request a new N.J.R.E. 104 hearing if he can produce evidence that will satisfy the standards for admissibility of evidence in accordance with the discussion that follows.
The trial court clearly understood and articulated the issues. First, the court queried whether it would be “logical for a jury to conclude that if someone else’s [sperm] was found in the victim’s body besides her boyfriend and the defendant,” left there “within a reasonable time of death,” that a “third person committed the crime.” The answer to that question appears clear enough: such [594]*594evidence would be relevant for the purpose of raising a reasonable doubt of defendant’s guilt.
First, we address whether the sperm recovered from Padilla’s body had been ejaculated at or near the time of the victim’s death. Dr. Marvin Shuster, the Middlesex County Chief Medical Examiner at the time of the murder, testified to the presence of sperm in the victim’s vagina. When asked whether he could determine the age of the sperm found in the victim’s vagina, he responded, “It’s highly variable.” He stated that “[w]ith fresh ejaculates” one would expect to see intact sperm, but he found “few and scattered sperm[ ], some of which [were] intact, and many of which [were] represented only by sperm heads.” Dr. Shuster was unable to give an opinion as to the age of the sperm and, therefore, could not determine whether or not the recovered sperm were ejaculated into the victim’s vagina at or near the time of her death.
On the other hand, Dr. Geetha Natarajan, who succeeded Dr. Shuster as Middlesex County medical examiner, had no reservation reaching a firm conclusion with respect of the age of the sperm. She reasoned that had there been intercourse “immediately prior to death,” there would have been “numerous sperms,” rather than “a few sperm heads or an intact sperm.” She concluded that “there was no ejaculation into [the victim’s] vagina at the time or immediately prior to her death” and, therefore, the recovered ejaculate “had nothing to do with her murder.”
Although the defense was not bound by that conclusion, Dr. Shuster’s testimony left the age of the sperm, at best, as a matter of conjecture. Dr. Shuster could not discount the possibility that the recovered sperm entered into the victim’s vagina at or near the time of her death and, therefore did not render an opinion within a reasonable degree of scientific certainty. Had defendant been able to pinpoint the entry of the sperm to the time of Padilla’s death, then, even in the absence of identifying the donor, that evidence would have been admissible to challenge the State’s theory that the Padilla murder and the Trooper Gardner assault [595]*595both involved digital penetration only. But the defense could not mount such an attack on pure speculation.
The second issue is whether the vaginal swab submitted to Cellmark Laboratories for DNA (deoxyribonucleic acid) typing provided some reasonable basis to argue third-party guilt. At a N.J.R.E. 104 hearing, Paula Yates, a specialist in forensic science at Cellmark, testified that she performed a DNA analysis of the vaginal swab. She stated that she was unable to determine within a reasonable degree of scientific certainty or probability whether the sample she examined excluded the DNA of defendant, Fernandez, or even Padilla. She was unable to render such an opinion because of the absence of “control dots” that are visible when there is a sufficient amount of DNA to be tested and typed reliably for the presence of particular alleles.6 According to Yates, the absénce of the control dots meant that she had no way of knowing if she was observing all of the DNA in the vaginal swab. Yates testified that, without the control dots, the analysis could not exclude Padilla, Fernandez, or Fortin as the source of the DNA.
Despite those limitations, the prosecutor and defense counsel prodded Yates to analyze the available data and to assume it to be “interpretable.” Working under that assumption, Yates testified that the data indicated that there were sperm from at least two different donors, a primary donor who contributed a number 1.2 allele and a secondary donor who contributed a number 2 allele. The analysis could not exclude defendant as the primary donor [596]*596because his DNA typing had similarities to that of the primary profile, in particular the number 1.2 allele. Yates stated that defendant could not have been the secondary donor because he could not have contributed the number 2 allele. Yates also stated that Fernandez could not have contributed the number 2 allele, but defense counsel did not explore the import of that answer, although the answer suggested Fernandez’s exclusion as a secondary donor. Defense counsel did not inquire of Yates whether Fernandez’s alleles included or excluded him as a primary donor.
Based on the N.J.R.E. 104 hearing testimony, the court entered an order barring any evidence or argument regarding “penile penetration resulting in an ejaculation by an unknown third-party at the approximate time of [Padilla’s] murder.” The court reasoned that Cellmark, the laboratory that analyzed the vaginal swab for DNA, “could not within a reasonable degree of scientific certainty or probability exclude any of the three known persons [defendant, Fernandez, or Padilla] from being the source of the sperm fraction analyzed.” On that basis, the court found that “defendant could not rationally argue that an unknown third-party [engaged in penile penetration of] the victim at the time of her death.”
During the guilt phase of the trial, Dr. Charlotte Word of Cellmark testified about the DNA analysis of the dollar bill and the cigarette butt found at the murder scene. On cross-examination, defense counsel questioned Dr. Word concerning the vaginal swab. As when Paula Yates testified, Dr. Word qualified her answers by assuming that there were “appropriate controls” and that the test had produced an “interpretable result.” Dr. Word stated the number 2 allele could not have come from defendant, Fernandez, or Padilla. There was no follow-up question asking Dr. Word to explain the meaning of her response regarding the number 2 allele. Dr. Word noted that Cellmark’s report did not provide that conclusion because of the absence of “the appropriate controls” and presumably that is why Dr. Word did not render an opinion within a reasonable degree of scientific probability on that [597]*597subject. After Dr. Word’s testimony, defendant did not seek reconsideration of the court’s earlier ruling barring evidence or argument concerning the DNA analysis of the vaginal swab.
Based on that record, we conclude that the trial court did not abuse its discretion in keeping non-competent and irrelevant evidence before the jury. The sine qua non for the admissibility of scientific evidence is reliability. Expert testimony must comply with the requirements of N.J.R.E. 702. That rule provides:
If scientific, technical, or other specialized knowledge will assist the tiler of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify thereto in the form of an opinion or otherwise.
[N.J.R.E. 702.]
We have recognized three basic requirements for the admission of expert testimony pursuant to N.J.R.E. 702:
(1) the intended testimony must concern a subject matter that is beyond the ken of the average juror; (2) the field testified to must be at a state of the art such that an expert’s testimony could be sufficiently reliable; and (3) the witness must have sufficient expertise to offer the intended testimony.
[State v. Kelly, supra, 97 N.J. at 208, 478 A.2d at 379.]
Not only must the methodology be valid and the procedures applied correctly, but the results achieved also must be reliable. “In addition to showing its general acceptance in the scientific community, a party offering scientific evidence must show that the technique, methodology or procedure was correctly used to produce that evidence.” State v. Marcus, 294 N.J.Super. 267, 275, 683 A.2d 221, 225 (App.Div.1996), certif. denied, 157 N.J. 543, 724 A.2d 803 (1998). An expert offering scientific opinion testimony must do so within a reasonable degree of certainty or probability. See State v. Freeman, 223 N.J.Super. 92, 116, 538 A.2d 371, 384 (App.Div.1988) (holding that medical opinion testimony “must be couched in terms of reasonable medical certainty or probability; opinions as to possibility are inadmissible”) (internal citations omitted), certif. denied, 114 N.J. 525, 555 A.2d 637 (1989).
[598]*598In this case, both Yates and Dr. Word testified that Cellmark could not reach conclusions within a reasonable degree of scientific certainty or probability because of the absence of appropriate controls. Defendant presented no expert testimony to refute that assertion. Thus, we are left with a record in which the test results of the vaginal swab were deemed scientifically unreliable. That unchallenged testimony ultimately led the trial court correctly to exclude the evidence. Even were we to accept the testimony of Yates and Dr. Word in which they assumed interpretable results, defendant would fare no better because he was not eliminated as a potential source of the sperm found on the victim’s body.
Moreover, this Court is reluctant to come to scientific conclusions based on some of the disjointed testimony adduced from the experts. We decline to speculate as to the import of the testimony that Fernandez did not possess the number 2 allele. The record must provide those answers. The most liberal interpretation of the record in favor of defendant reveals that Padilla had sexual relations with someone other than her boyfriend or defendant at some unknown time before her death. That hardly counts as evidence that that unidentified person killed Padilla. The expert testimony failed to provide any relevant connection between the sperm analyzed and the murder. Reasonable doubt cannot arise from pure conjecture.
Accordingly, based on the record before us, we find that the trial court properly exercised its discretion in barring evidence and argument that some unknown third party engaged in penile penetration at the time of the victim’s death.
PENALTY PHASE
V.
Mitigating Factors and Mitigating Evidence
In the penalty phase of capital litigation, the jury determines whether the State has presented proof beyond a [599]*599reasonable doubt of the existence of aggravating factors, and, if there is such proof, whether those aggravating factors outweigh any mitigating factors beyond a reasonable doubt. N.J.S.A. 2C:ll-3c(2)(a), (3). If the State satisfies both standards, the defendant is sentenced to death; if not, he is sentenced to a minimum term of imprisonment of thirty years without parole eligibility. N.J.S.A. 2C:ll-3e(3)(a)-(e).
In this case, the State alleged three aggravating factors: the murder involved aggravated assault or torture, N.J.S.A. 2C:11-3e(4)(c); the murder was committed to escape apprehension for another offense, N.J.S.A. 2C:ll-3c(4)(f); and the murder was committed during commission of, or an attempt to commit, or flight after committing robbery or sexual assault, N.J.S.A. 2C:11-3c(4)(g).
Defendant may prove mitigation in the penalty phase by two different means, through the introduction of mitigating factors and through the introduction of mitigating evidence. The mitigating factors are set forth in N.J.S.A 2C:ll-3c(5)(a)-(h). The jury weighs the mitigating factors against the aggravating factors in determining whether to impose a capital or non-capital penalty. N.J.S.A. 2C:ll-3c(3). Mitigating evidence is any evidence that supports a mitigating factor or undermines an aggravating factor. State v. Feaster, 156 N.J. 1, 86, 716 A.2d 395, 437-38 (1998), cerf. denied, 532 U.S. 932, 121 S.Ct 1380, 149 L.Ed.2d 306 (2001).
In addition to the mitigating factors specifically enumerated in N.J.S.A. 2C:11-3c(5)(a)-(g), the catch-all provision allows the jury to consider “[a]ny other factor which is relevant to the defendant’s character or record or to the circumstances of the offense.” N.J.S.A. 2C:11-3c(5)(h). Defendant proposed eleven mitigating factors under the catch-all category. The State challenged two of those factors:
10. 5(h) The victim sold drugs for her boyfriend on the night she was killed. Individuals gave statements they bought drugs from the victim the night she was killed, and the victim and her boyfriend were overheard arguing about drugs the [600]*600same night, thereby refuting the State’s allegation that Steve Fortin robbed the victim.
11. 5(h) The victim’s live-in boyfriend admitted engaging in consensual anal sex with the victim thus refitting the State’s allegation that Steve Fortin sexually assaulted the victim.
The State argued that those two mitigating factors did not relate to defendant’s “character” or “record,” or to the “circumstances of the offense.” Defendant contended that the proposed mitigating factors rebutted the 4(g) felony murder-aggravating factor — that Fortin killed Padilla in the course of a robbery and sexual assault. The trial court rejected the proposed mitigating factors because they were not relevant to the statutory language of N.J.S.A. 2C:11-3c(5)(h). We agree. We also find no support for defendant’s contention that he was barred from introducing relevant mitigation evidence to undermine the existence of the State’s aggravating factors.
First, we address the basis for the trial court’s rejection of the two mitigating factors. Our capital sentencing jurisprudence requires that courts grant defendants “wide leeway in presenting evidence in mitigation of the death penalty.” State v. Bey, 112 N.J. 123, 157, 548 A.2d 887, 904 (1988) (Bey II), cert. denied, 513 U.S. 1164, 115 S.Ct. 1131, 130 L.Ed.2d 1093 (1995). Although the scope of N.J.S.A. 2C:11-3c(5)(h) is broad, it is not unlimited. State v. Gerald, 113 N.J. 40, 103, 549 A.2d 792, 824 (1988). Defendant does not claim that the proposed mitigating factors were relevant to his character or record. He does claim, however, that they were relevant to the “circumstances of the offense.” In Gerald, we construed the phrase “circumstances of the offense” to encompass only those “circumstances surrounding the commission of the crime itself.” Id. at 104, 549 A.2d at 825. For example, the relative role of the “defendant’s participation in the crime” meets that standard. Ibid. The jury may weigh whether the defendant was a “ringleader” or a lesser player in comparison to codefendants. Ibid. In contrast, the sentences received by codefendants are not a relevant mitigating factor. Ibid. In State v. Timmendequas, we rejected as a factor bearing [601]*601on “circumstances of the offense” and “character” the defendant’s offer to plead guilty in exchange for a life sentence. 161 N.J. 515, 626-27, 737 A.2d 55, 115 (1999) (Timmendequas I), cert. denied, 534 U.S. 858, 122 S.Ct. 136, 151 L.Ed.2d 89 (2001). For similar reasons, the period of parole ineligibility a defendant would serve if he received a non-capital sentence is not a mitigating factor. State v. Morton, 155 N.J. 383, 466, 715 A.2d 228, 270 (1998) (Morton I), cert. denied, 532 U.S. 931, 121 S.Ct. 1380, 149 L.Ed.2d 306 (2001). On the other hand, a defendant’s claim-of-right defense in the penalty phase — his honest belief that he was entitled to recover money from the person he murdered — is a permissible mitigating factor that could have a countervailing effect on the 4(g) aggravating factor, “murder committed during a felony.” State v. Mejia, 141 N.J. 475, 500, 662 A.2d 308, 321 (1995).
Applying those principles to this case, the trial court properly rejected the two proposed catch-all mitigating factors. That Padilla sold drugs earlier in the evening bore no relationship to any of the circumstances of the offense. The proposed drug-dealing mitigating factor was a thinly disguised effort to demean the character of the victim while in no way lessening defendant’s personal involvement or moral culpability in the murder. Contrary to defendant’s assertion, Padilla’s involvement in drugs did not “refute” the State’s allegation that he committed a felony-robbery-murder or suggest that someone else robbed her. Her drug dealing was not germane to any legitimate defense or mitigating factor.
In addition, the victim’s prior sexual conduct was.not relevant as a mitigating factor. That Padilla’s boyfriend had anal sex with her one to two weeks before her murder did not qualify as a mitigating factor because it too bore no relationship to the crime. Such evidence did not refute the evidence that Padilla was the victim of a sexual crime at the time of the murder. Defendant was unable to present any evidence that the “fresh” wounds to Padilla’s anus were caused by consensual sexual relations with her boyfriend.
[602]*602We next consider defendant’s argument that he was denied the opportunity to present mitigating evidence. This Court has recognized that mitigating evidence serves multiple purposes:
(1) to weaken the State’s proofs concerning the existence of aggravating factors;
(2) to establish the existence of mitigating factors; and (3) to bolster the weight of those mitigating factors found to exist in an attempt to have those factors outweigh the aggravating factors found to exist during the jurors’ ultimate deliberation.
[State v. Martini 131 N.J. 176, 316, 619 A.2d 1208, 1282 (1993) (Martini I), cert. denied, 516 U.S. 875, 116 S.Ct. 203, 133 L.Ed.2d 137 (1995).]
The defendant has a right to both criticize the evidence supporting the aggravating factors as well as to present his own evidence supporting mitigating factors and undermining aggravating factors. State v. Josephs, 174 N.J. 44, 116, 803 A.2d 1074, 1117 (2002); see also N.J.S.A 2C:11-3c(2)(b) (providing that defendant “may offer, without regard to the rules governing the admission of evidence at criminal trials, reliable evidence relevant to any of the mitigating factors”); N.J.S.A. 2C:11-3c(2)(d) (providing that defendant and State may rebut any evidence presented during penalty trial and may argue adequacy of evidence establishing any aggravating or mitigating factor).
In State v. Davis, this Court emphasized the different evidentiary standards applicable to the guilt and penalty phases. 96 N.J. 611, 619-20, 477 A.2d 308, 312 (1984). Given this “most sensitive and critical aspect of the capital punishment statute,” the Court found that the Legislature intended “a broad understanding of evidential relevance” during the penalty phase. Id. at 619, 477 A.2d at 312. The Court noted “doubts must be resolved in favor of admission when evidence of a mitigating factor is offered by the defendant.” Id. at 620, 477 A.2d at 312. Additionally, evidence inadmissible during the guilt phase, such as a elaim-of-right defense to robbery, may be admissible during the penalty phase to lessen the weight that a jury might give to the felony-robbery-murder aggravating factor. Mejia, supra, 141 N.J. at 500, 662 A.2d at 320-21. However, the Court cautioned in Davis, supra, that a relaxation of the standards of admissibility is not the equivalent of “automatic admissibility.” 96 N.J. at 623, 477 A.2d [603]*603at 314. A court may exclude mitigation evidence if its “probative value is substantially outweighed by its unfounded or speculative character or by the risk it would cause confusion.” Morton I, supra, 155 N.J. at 461, 715 A.2d at 267.
The drug sales evidence did not discredit the felony-robbery-murder aggravating factor because there was no evidence that connected drug dealing to the murder. We found the drug-dealing evidence irrelevant in the guilt phase; we find it irrelevant in the penalty phase as well. See supra Part IV.B. The trial court, nevertheless, gave defense counsel considerable latitude to explore this subject. Defense counsel elicited through cross-examination of the State’s witnesses the drug milieu of the area in which the crime was committed. For instance, he established that Padilla was killed in an area spotted with welfare motels and known for drug activity. He also proved that a civilian witness had smoked crack cocaine with the victim in the room of a local motel and that a law enforcement officer had questioned several witnesses about the use of cocaine. Defendant was not inhibited in his presentation of mitigating evidence, even that of questionable relevance.
Moreover, defendant’s repeated assertions that he was not allowed to present evidence in rebuttal of the sexual assault aggravating factor is not supported by the record. At the penalty-phase hearing, in cross-examining Dr. Lawrence Ricci, the State’s sexual assault expert, defense counsel relied on the statement given by Padilla’s boyfriend to the police that he and the victim engaged in consensual anal sex one to two weeks before her death. Defense counsel suggested through his questioning that the injuries to Padilla’s anus were caused by consensual sexual activity with the boyfriend. Dr. Ricci dashed that theory by testifying that the anal injuries, whether caused by consensual or non-eonsensual activity, were “fresh injuries” that occurred within twenty-four to forty-eight hours of death. Accordingly, there was no connection between the victim’s sexual activity with her boyfriend and her injuries discovered at the time of her death.
[604]*604In sum, the trial court properly struck the two mitigating factors in question and gave wide latitude to the admission of mitigating evidence at the penalty hearing.
VI.
Waiver of Ex Post Facto Challenges in Sentencing
On August 22, 2000, six months before the penalty-phase proceedings in this case, the Legislature enacted a law that allowed a sentence of life without parole in certain capital cases. N.J.S.A. 2C:ll-3b(4) provides that whenever a jury in a capital case finds the presence of one or more aggravating factors, but does not return a death verdict, “the defendant shall be sentenced by the court to a term of life imprisonment during which the defendant shall not be eligible for parole.” The statute also provides that: “[t]his act shall take effect immediately.” L. 2000, c. 88. As of the time of the Padilla murder, a capital defendant faced a thirty-year to life sentence with a minimum thirty-year parole disqualifier if the jury found at least one aggravating factor but did not return a death verdict. N.J.S.A. 2C:ll-3b(l). Before the penalty phase began in this case, defendant announced that he would subject himself to the life-without-parole alternative and asked the court to instruct the jury in accordance with the newly-enacted provision.
Defendant undoubtedly believed that if the jury were instructed on the life-without-parole option, it might be less inclined to return a death verdict. Such reasoning not only has roots in our common intuition, but empirical support from statistical surveys. The United States Supreme Court, in Simmons v. South Carolina, noted a survey conducted by the University of South Carolina’s Institute for Public Affairs in which more than seventy-five percent of those questioned “indicated that if they were called upon to make a capital sentencing decision as jurors, the amount of time the convicted murderer actually would have to spend in prison would be an ‘extremely important’ or a ‘very important’ factor in choosing between life and death.” 512 U.S. 154, 159, 114 S.Ct. [605]*6052187, 2191, 129 L.Ed.2d 133, 140 (1994); see also Taylor v. State, 672 So.2d 1246, 1273 (Miss.) (reversing death sentence because jury might have opted for life sentence had it been told that defendant faced life without parole), cert. denied, 519 U.S. 994, 117 S.Ct. 486, 136 L.Ed.2d 379 (1996); William J. Bowers & Benjamin D. Steiner, Death by Default: An Empirical Demonstration of False and Forced Choices in Capital Sentencing, 77 Tex. L. Rev. 605, 650, 671 (1999) (finding most jurors “are more likely to vote for death” because of belief that convicted killers will serve prison terms below mandatory minimum for parole and will get out of prison “far too soon”).
No one disputed that the life-without-parole provision of N.J.S.A. 2C:ll-3b(4) imposes a more severe sentence than the previous law of life with a minimum parole disqualifier of thirty years. The State and defendant agreed that the Ex Post Facto Clause barred the application of the life-without-parole provision unless waived by defendant. Defendant informed the court that he would waive any ex post facto challenge to the application of the new provision. The State, however, declined to consent to defendant’s application. The trial court ruled that defendant could not waive an ex post facto objection to N.J.S.A. 2C:ll-3b(4) and, therefore, refused to instruct the jury in accordance with the new statute. The trial court never squarely addressed whether the Legislature intended N.J.S.A. 2C:ll-3b(4) to apply to defendant’s penalty-phase proceedings.
At a hearing prior to the penalty-phase trial, the court “recognize[d] the tactical advantage to [defendant] to present to a jury ... that at the minimum he is going to serve a life imprisonment with no possibility of parole.” The court denied defendant the benefit of the new statute for two reasons. First, the court stated that despite defendant’s willingness to waive application of the Ex Post Facto Clause and to waive the right to any appeal arising from giving a life-without-parole instruction to the jury, future appellate counsel might not abide by defendant’s decision. Second, the trial court voiced apprehension that the Commissioner of [606]*606the Department of Corrections might not carry out the jury’s mandate of a life-without-parole sentence if the Commissioner disagreed with the court’s decision to permit waiver of the protections of the Ex Post Facto Clause. Despite defendant’s willingness to subject himself to life imprisonment with no possibility of parole, the court instructed the jury that defendant potentially could be released from prison in thirty years.7
In deciding whether the trial court properly exercised its discretion, we must determine whether the Legislature intended N.J.S.A. 2C:ll-3b(4) to extend only to those capital murders committed after the statute went into effect or to all capital murders reaching the penalty phase after its effective date. If the statute was not intended to apply to capital murders that occurred before the enactment, then the inquiry ends. On the other hand, if the statute was intended to apply to that category of cases, its reach would be limited to those in which the defendant waived the constitutional protection of the Ex Post Facto Clause. That raises the additional question of whether the defendant is empowered, without the consent of the State or the court, to waive that constitutional right. We first examine whether N.J.S.A. 2C:11-3b(4), after its effective date, applies to all penalty-phase trials, including those capital murder defendants willing to waive protection of the Ex Post Facto Clause.
Whenever interpreting legislation, we begin by looking at the statutory language to divine its meaning. Alan J. Cornblatt, P.A. v. Barow, 153 N.J. 218, 231, 708 A.2d 401, 407 (1998). Here, the language of the statute does not provide an answer because the legislative direction that the “act shall take effect immediately” is insolubly ambiguous. Because the words of the statute are susceptible to different interpretations, we must look to other aids of statutory construction that shed light on legislative [607]*607intent, “such as the statute’s purpose, legislative history, and statutory context.” Township of Pennsauken v. Schad, 160 N.J. 156, 170, 733 A.2d 1159, 1166 (1999). The legislative history is scant and does not provide much insight into the meaning of the statute. However, in this case, the policy considerations that gave rise to the statute inform our interpretation of the Legislature’s intent. We recognize that applying the statute to murders that occurred both before and after the enactment of N.J.S.A. 2C:11-3b(4), with due regard to the right of defendants to invoke or waive protection of the Ex Post Facto Clause, would give the most wide-ranging effect to the legislative mandate. The State counters that the Legislature was aware of the Ex Post Facto Clause in passing N.J.S.A. 2C:ll-3b(4) and would not have intended the statute to apply to cases beyond its constitutional reach by leaving to individual defendants the opportunity to opt in by waiving the clause. If the statute were not intended to apply to defendant’s case, then the Ex Post Facto Clause would not come into play and the issue of waiver would be moot.
We conclude that the policy objectives of the legislation clearly support the most far-reaching application of the statute. That means the statute embraces all capital murder cases proceeding to the penalty phase after the legislation’s effective date, provided that, in cases in which the murder preceded the effective date, the defendant must be willing to waive the protection of the Ex Post Facto Clause. We reach this result for several reasons. One of the obvious statutory objectives was to eliminate recidivism for those who commit society’s most abhorrent crimes. In those cases in which the penalty-phase jury finds beyond a reasonable doubt an aggravating factor but does not return a death verdict, the Legislature upgraded the punishment from thirty years to life with a thirty-year parole disqualifier to life without parole, thus removing any chance of the convicted murderer’s release. We cannot discern any reason why the Legislature would not have wanted to extend that punishment to the greatest number of cases constitutionally permissible. Additionally, assuming that the odds of a death verdict increase if the jury does not have the option of a [608]*608life-without-parole sentence, we cannot discern any reason why the Legislature would have wanted to divide the fates of defendants, who had yet to proceed to the penalty phase, between those whose crimes occurred before and after enactment of N.J.S.A. 2C:ll-3b(4). The legislative goal of proportionality in capital sentencing favors similar treatment for similarly situated defendants, to the extent possible. N.J.S.A. 2C:ll-3(e). Applying N.J.S.A. 2C:ll-3b(4) to the greatest number of cases constitutionally permissible furthers the goal of proportionate sentencing. We also are mindful that when a criminal statute is at issue, “we are guided by the rule of lenity” and “interpret ambiguous language in favor of a criminal defendant.” State v. Livingston, 172 N.J. 209, 218, 797 A.2d 153, 158 (2002). In light of the purpose of the new law, in cases such as this one, we cannot find a legislative intent to have a jury return a death verdict only because it did not have the option of returning a life-without-parole sentence.
We next address whether the defendant can waive the protection of the Ex Post Facto Clause. The United States and New Jersey Constitutions prohibit ex post facto legislation.8 The Ex Post Facto Clause was intended to interdict the retroactive application of criminal laws that harm the accused. Miller v. Florida, 482 U.S. 423, 430, 107 S.Ct. 2446, 2451, 96 L.Ed.2d 351, 359-60 (1987) (quoting Weaver v. Graham, 450 U.S. 24, 29, 101 S.Ct. 960, 964, 67 L.Ed.2d 17, 23 (1981)); State v. T.P.M., 189 N.J.Super. 360, 366-67, 460 A.2d 167, 170-71 (App.Div.1983). The drafters of that clause understood that it would be unjust to prosecute a person for a crime or to impose a harsher sentence based on a law that was not on the books at the time of the commission of the act covered by the subsequent legislation. The [609]*609Ex Post Facto Clause provides assurance that “federal and state legislatures [are] restrained from enacting arbitrary or vindictive legislation.” Miller, supra, 482 U.S. at 429, 107 S.Ct. at 2451, 96 L.Ed.2d at 359. It also ensures that legislative enactments “give fair warning of their effect and permit individuals to rely on their meaning until explicitly changed.” State v. Muhammad, 145 N.J. 23, 56, 678 A.2d 164, 181 (1996) (quoting Weaver, supra, 450 U.S. at 28-29, 101 S.Ct. at 964, 67 L.Ed.2d at 23).
It is well established that a defendant may waive a constitutional right. Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 1023, 82 L.Ed. 1461, 1466 (1938); see also Colorado v. Spring, 479 U.S. 564, 572, 107 S.Ct. 851, 856, 93 L.Ed.2d 954, 965 (1987) (noting that “a suspect may waive his Fifth Amendment privilege” against self-incrimination); Patton v. United States, 281 U.S. 276, 290, 50 S.Ct. 253, 255, 74 L.Ed. 854, 859 (1930) (holding that “a person charged with a crime punishable by imprisonment for a term of years may, consistent ] with the [C]onstitution[ ] ..., waive trial by a jury”); State v. Crisafi, 128 N.J. 499, 509, 608 A.2d 317, 323 (1992) (“Defendants possess not only the right to counsel, but the right to dispense with counsel and to proceed pro se.”); State v. Hill, 115 N.J. 169, 173, 557 A.2d 322, 323-24 (1989) (noting that defendant can waive constitutional protection against search and seizure through consent). In a capital case, the defendant is permitted to waive his right to a bifurcated jury, State v. Parker, 256 N.J.Super. 336, 341, 606 A.2d 1154, 1157 (Law Div.1992), and may plead guilty to capital murder, see, e.g., State v. Simon, 161 N.J. 416, 433-34, 737 A.2d 1, 9-10 (1999); State v. Nelson, 155 N.J. 487, 494, 715 A.2d 281, 284 (1998) (Nelson I).
We cannot accept the trial court’s reasons for not honoring defendant’s willingness to waive the protection of the Ex Post Facto Clause. Defendant was represented by two attorneys, one a private practitioner and the other a Deputy Public Defender, both of whom were certified criminal trial attorneys. They represented to the court that defendant wished to waive application of the Ex Post Facto Clause and, in response to the court’s concerns, [610]*610to waive his right to appeal that issue. Not satisfied with defendant’s proffer, the court made inquiry through a representative of the Administrative Office of the Courts with a high level administrator of the Office of the Public Defender to learn whether that Office would honor defendant’s waiver on appeal. The Public Defender administrator responded that his office gives “free rein” to its appellate counsel to raise all appropriate issues. Without a guarantee that its decision would be free from appellate review, the court declined to give the life-without-parole instruction to the jury.
We disapprove of the inquiry made by the court. Defendant was represented by counsel who were responsible for his ease. All inquiries should have been directed to them. The court should have gone no further. Instead, the court sought and received information from an administrator with no connection to the case and then charted a course based on that irrelevant information. The decision whether to waive a constitutional right is personal to the defendant; it belongs to him, not the attorney, and certainly not to an administrative officer of the Office of the Public Defender. See State v. Dunne, 124 N.J. 303, 312, 590 A.2d 1144, 1148 (1991) (stating that “New Jersey’s consistent tradition has been that every right and privilege secured by our State Constitution belongs to each citizen, ‘as a personal right’ ”) (quoting State v. Stevens, 84 N.J.L. 561, 563, 87 A. 118, 119 (Sup.Ct. 1913)); see also Rock v. Arkansas, 483 U.S. 44, 51-53, 107 S.Ct. 2704, 2708-10, 97 L.Ed.2d 37, 46-47 (1987) (holding that constitutional right to testify on one’s own behalf belongs to defendant personally); Faretta v. California, 422 U.S. 806, 819, 95 S.Ct. 2525, 2533, 45 L.Ed.2d 562, 572 (1975) (noting that Sixth Amendment grants accused personal right to waive constitutional right to counsel in favor of self-representation). Indeed, an attorney is ethically bound to “abide by a client’s decisions concerning the objectives of representation.” Model Rules of Prof l Conduct R. 1.2(a) (2003).
[611]*611There was no need for the court to extract from defendant a waiver of a right to appeal from his waiver of the ex post facto bar. So long as the waiver procedure demonstrates that a defendant’s decision was knowing, intelligent, and voluntary, the waiver is impervious to challenge on appeal. Koedatich II, supra, 112 N.J. at 328-29, 548 A.2d at 992-93. A waiver is valid and binding on a defendant whenever, under the totality of the circumstances, it is given freely and with full knowledge of the nature of the right abandoned and the consequences of abandoning it. Spring, supra, 479 U.S. at 571-75, 107 S.Ct. at 856-58, 93 L.Ed.2d at 964-66; Koedatich II, supra, 112 N.J. at 328-29, 548 A.2d at 992-93. If the ex post facto waiver procedure were not constitutionally permissible, then even a waiver of the right of appeal likely would have had little validity. Additionally, the court should not have been concerned with whether the Commissioner of Corrections would have abided by the imposition of a life-without-parole sentence. The Commissioner cannot defy a sentence that is lawfully imposed. The court should have been concerned only with a proper interpretation of N.J.S.A. 2C:11-3b(4) and with following the appropriate constitutional principles.
New Jersey trial courts have accepted a capital defendant’s offer to waive his ex post facto right to challenge statutes mandating life imprisonment without parole. See, e.g., Timmendequas I, supra, 161 N.J. at 637, 737 A.2d at 121-22 (noting capital defendant waived ex post facto challenge to N.J.S.A. 2C:11-3b(3), which mandates life without parole for defendants convicted of murdering child less than fourteen-years-old in course of sexual assault); State v. Nelson, 173 N.J. 417, 433-34, 482, 803 A.2d 1, 10 (2002) (Nelson II) (illustrating defendant’s waiver of ex post facto challenge to application of N.J.S.A. 2C:11-3b(2), which mandates life without parole for conviction of murdering police officer). Other jurisdictions also have allowed defendants to waive ex post facto challenges to the application of life imprisonment without the possibility of parole sentencing options that took effect after the commission of a capital murder. See, e.g., Furnish v. Commonwealth, 95 S.W.3d 34, 50-51 (Ky.2002) (holding reversible error not [612]*612to instruct jury of sentencing option of life without parole in case of defendant “ “willing to make a knowing, intelligent and voluntare ]y waiver of any right to attack this statute as a violation of the ex post facto prohibition of the U.S. and Kentucky Constitutions’ ”) (citation omitted), cert. denied, — U.S. -, 124 S.Ct. 115, 157 L.Ed.2d 80, 72 U.S.L.W. 3238 (2003); Willie v. State, 738 So.2d 217, 220 (Miss.1999) (holding that on remand for resentencing “jury should be instructed on all three options available under the amended statute” and if defendant “agrees to a sentence of life in prison without parole, the trial judge should take care to ascertain that [defendant] has validly waived his ex post facto rights before accepting the plea agreement”); Wade v. State, 825 P.2d 1357, 1363 (Okla.Crim.App.1992) (reversing death sentence because trial court erred in refusing defendant’s request to instruct jury based on sentencing scheme enacted subsequent to commission of underlying offense and after he waived ex post facto challenges); State v. McDonnell, 329 Or. 375, 987 P.2d 486, 494-96 (1999) (holding that life-without-parole provision was effective as of date of enactment in capital case and defendants convicted of murder could waive ex post facto challenge).
We understand that the issues confronting the trial court were difficult and somewhat novel and did not lend themselves to easy resolution. On remand, in any new penalty-phase trial, the court must give defendant the option of a life-without-parole jury instruction consistent with N.J.S.A. 2C:ll-3b(4), provided defendant is willing to give a knowing, intelligent, and voluntary waiver of the application of the Ex Post Facto Clause.
VII.
Social History Evidence
In the penalty phase, defendant was prepared to present, through the testimony of two expert witnesses, a complete thirty-year social history of his life, from birth to the August 11, 1994 murder of Melissa Padilla, with two significant exceptions. Defendant intended to carve out of the presentation of his extensive [613]*613criminal history any reference to his 1983 manslaughter conviction for the killing of his brother and his 1995 conviction for the sexual assault of Trooper Gardner. The trial court determined that the State was entitled to rebut that incomplete portrait of defendant’s life with the introduction of those two prior convictions in a sanitized form. The court recognized that the State could not introduce prior bad conduct, other than a murder conviction, in support of a statutory aggravating factor and that such specific past conduct was admissible only to rebut defendant’s mitigation evidence. See N.J.S.A. 2C:11-3c(4)(a); State v. Rose, 112 N.J. 454, 503, 548 A.2d 1058, 1082-83 (1988). To minimize potential prejudice to defendant, the court stated that it would allow the use of the convictions in rebuttal provided that graphic and inflammatory details were removed. The court ruled that the 1983 manslaughter conviction would be admissible, but only as a conviction for a second-degree assault against his brother, and the 1995 sexual assault conviction would be admissible, but only as a first-degree assault conviction. Defendant claims that to avoid the introduction of the 1983 and 1995 sanitized convictions, he was forced to forgo the presentation of any social history. We find that defendant made a tactical decision to abandon the presentation of his mitigation evidence to forestall the State from introducing relevant and damaging rebuttal evidence. We perceive no error in the court’s ruling.
In the penalty phase of a capital trial, if the jury unanimously determines that the State has proven the existence of one or more statutory aggravating factors beyond a reasonable doubt and that the aggravating factor or factors outweigh beyond a reasonable doubt all of the mitigating factors presented by the defendant, the court must impose the punishment of death. N.J.S.A. 2C:11-3e(3)(a). If the State falls short of persuading even a single juror, the punishment is a term of imprisonment, ranging from a minimum period of thirty years without parole eligibility to life. [614]*614N.J.S.A. 2C:ll-3c(3)(c). Defendant proffered three mitigating factors: he was under the influence of an extreme mental or emotional disturbance, N.J.S.A. 2C:ll-3c(5)(a); he suffered from a mental disease or defect, or intoxication that sufficiently impaired his capacity to appreciate the wrongfulness of his conduct, N.J.S.A. 2C:ll-3c(5)(d); and his character and record, N.J.S.A. 2C:ll-3c(5)(h). In support of those mitigating factors, defendant submitted the report of Lois Nardone, M.S.W. That report gave a detailed thirty-year portrait of defendant’s troubled life, a life that included physical and emotional abuse by a brother; severe beatings by his father; lack of nurture or support from his family throughout childhood and adolescence; sexual abuse by an adult male when he was a teenager; and a substantial criminal past, ending with the Padilla murder. Defendant also submitted the report of Dr. Maureen R. Santina, a forensic psychologist, who relied on Nardone’s “complete psychosocial history.” Dr. Santina diagnosed defendant’s mental-health disorders, which included drug and alcohol dependence, attention deficit/hyperactivity disorder, borderline personality disorder, and avoidant personality disorder with mixed narcissistic and antisocial personality.
Nardone’s report gave a sweeping review of defendant’s prior criminal history. As a juvenile, defendant was adjudicated a delinquent for attempted theft of a motor vehicle in 1978, theft of a motor vehicle by an unlicensed driver in 1980, criminal trespass in 1981, a disorderly persons offense in 1981, and shooting an eighteen-year-old with a BB gun in 1982. With regard to that last offense, defendant reported that he was “tired of being pushed around” and wanted to “scar” the young man by shooting him. As an adult, he was convicted of criminal mischief and possession of a machete for an unlawful purpose in 1991. For the machete offense, defendant was sentenced to prison for three years. His parole was later revoked in 1992 for failure to refrain from using drugs. He was arrested on a charge of possession of controlled dangerous substances in 1993, which was later dismissed. In 1994, defendant was arrested on counts of criminal trespass, [615]*615burglary, and criminal mischief.9 In addition to noting defendant’s assault on his girlfriend, Dawn Archer, on the night of the murder, Nardone’s report is replete with descriptions of defendant’s drug and alcohol abuse, his failures at keeping a job, and his inability to adjust positively to probation and counseling.
The only convictions missing from Nardone’s report were defendant’s 1983 theft and manslaughter convictions and his 1995 sexual assault conviction. Defense counsel had “eliminated” the April 3, 1995, sexual assault of Trooper Gardner from Nardone’s report by instructing Nardone to cut off her analysis of defendant’s social history at Padilla’s 1994 murder, and counsel redacted from Nardone’s report reference to defendant’s 1983 second-degree manslaughter conviction.10
The State held in reserve for rebuttal Dr. Michael Weiner, a psychiatrist, who prepared a forty-nine page report in which he considered defendant’s entire criminal record along with information that painted a starkly different picture of defendant’s formative years than the one that appeared in Nardone’s report. Dr. Weiner was prepared to testify about the “normal,” “close,” and “caring” family that defendant had described to him, the omitted convictions, and evidence that “raise[d] the serious likelihood that” defendant was not the victim of sexual abuse in his adolescence as depicted in Nardone’s report. Dr. Weiner also was dismissive of Dr. Santina’s diagnoses of attention deficit/hyperactivity disorder, borderline personality disorder, and avoidant personality disorder. He found that defendant suffered from alcohol and drug dependence, sexual sadism, and antisocial personality disorder. Moreover, he concluded that defendant met the criteria for the diagnosis of psychopathy, which, in part, is evidenced by a “selfish and parasitic exploitation” of others. In short, Dr. Weiner’s rebuttal [616]*616was aimed at attacking the accuracy and reliability of the defense experts’ reports and at providing an alternate picture of defendant’s background and mental health status.
Defendant argued that both the 1983 manslaughter conviction and the 1995 sexual assault conviction were outside the scope of cross-examination because neither defense expert “mentioned” those incidents in their reports. The State maintained that it was entitled to cross-examine defendant’s experts on their failure to consider those convictions in rendering their “social history” and “mental health” reports.
The court granted the State permission to examine the defense experts on the incomplete presentation of defendant’s prior criminal record in their reports. At defense counsel’s request, the court agreed that counsel could “restore” Nardone’s original report, so that her credibility would not be tarnished on the stand by counsel’s redaction of the 1983 manslaughter conviction that had made her report “incomplete.” The court also ruled that the State could adduce on cross-examination that Nardone had “stopped her social history at the killing of Ms. Padilla ... and was specifically requested not to go further.” As noted, the court required the State to refer to defendant’s 1983 conviction of manslaughter as a “second-degree assault” on defendant’s brother and to his 1995 sexual assault conviction as a “first-degree assault.” The trial court followed the model in State v. Brunson, in which this Court held that when a defendant testifies and the State seeks to impeach his credibility with a prior criminal conviction that is similar in kind to the charged offense, the prior criminal conviction should be sanitized to minimize the potential improper use of the evidence to show propensity to commit the charged offense. 132 N.J. 377, 391-92, 625 A.2d 1085, 1092-93 (1993). For example, a testifying defendant on trial for third-degree distribution of drugs, who has a prior conviction of third-degree distribution of drugs, could be impeached only by reference to a conviction of a third-degree crime. The trial court recognized that the Brunson analogy was not a perfect fit to the circumstances of this case because [617]*617defendant’s credibility was not subject to attack as a testifying witness, but the court was moved by similar concerns about the undue prejudicial impact of unredacted convictions.
In an apparent effort to limit the damaging rebuttal testimony that would have come from Dr. Weiner, defendant abandoned the presentation of psychological testimony from Dr. Santina and withdrew mitigating factors N.J.S.A 2C:ll-3c(5)(a) and (d). As a result, the defense removed the threat of Dr. Weiner offering a psychiatric opinion that defendant was a psychopath as well as other testimony regarding his mental state. Dr. Weiner, however, was available to rebut Nardone’s narrative of defendant’s social history.
Defendant ultimately elected not to call Nardone at the penalty trial. Defendant’s only mitigating evidence was presented under the “catch-all” factor, N.J.S.A. 2C:ll-3c(5)(h), through cross-examination of the State’s witnesses. The jury thus learned nothing of defendant’s social history or about the alleged physical, emotional, and sexual abuse that he suffered as a child.
In the penalty phase of a capital case, the jury must be allowed to consider under the catch-all mitigating factor, N.J.S.A. 2C:ll-3c(5)(h), any evidence that is “relevant to the defendant’s character or record or to the circumstances of the offense.” N.J.S.A. 2C:ll-3e(5)(h) closely tracks the language of Lockett v. Ohio, which required individualized consideration of the defendant in a capital case as a constitutional imperative under the Eighth Amendment. 438 U.S. 586, 604, 98 S.Ct 2954, 2964, 57 L.Ed.2d 973, 990 (1978); see also Timmendequas I, supra, 161 N.J. at 625-26, 737 A.2d at 114-15. The general purpose of mitigating factor 3c(5)(h) is “to present extenuating facts regarding the defendant’s life or character or the circumstances surrounding the murder that would justify a sentence less than death.” Bey II, supra, 112 N.J. at 170, 548 A.2d at 911.
[618]*618Character evidence within the context of N.J.S.A. 2C:ll-3e(5)(h) “embracefs] those individual qualities that distinguish a particular person.” State v. Davis, supra, 96 N.J. at 618, 477 A.2d at 311. A capital defendant’s “background” or “social history” falls squarely within the realm of character evidence to be considered under that mitigating factor. See, e.g., Timmendequas I, supra, 161 N.J. at 628-32, 737 A.2d at 116-19 (discussing forensic social worker’s presentation of evidence of defendant’s “childhood” and “social history” under catch-all factor); State v. Bey, 129 N.J. 557, 615-16, 610 A.2d 814, 844 (1992) (Bey III) (noting that “all the evidence presented at trial about defendant’s background could be viewed as part of the catch-all mitigating factor”), cert. denied, 513 U.S. 1164, 115 S.Ct. 1131, 130 L.Ed.2d 1093 (1995). In this case, defendant’s “social history” evidence implicated both his “character” and his “record” under N.J.S.A. 2C:11-3c(5)(h).
Defendant claims that the “Hobson’s choice” foisted upon him by having to choose between introducing Nardone’s social history testimony along with sanitized evidence of his 1983 and 1995 convictions, or presenting none of that testimony, effectively deprived him of all mitigating evidence, and thus a fair sentencing trial. Defendant, however, did not intend to confíne the proposed mitigating “social history” evidence to his childhood, as is made clear by Nardone’s report. Cf., e.g., State v. Harris, 165 N.J. 303, 323, 757 A.2d 221, 232 (2000) (Harris II) (noting that defense experts’ review of mitigating evidence was “limited to defendant’s mental condition as a young child presumably because the defense did not want the jury to have specific knowledge of his criminal record”); Timmendequas I, supra, 161 N.J. at 545, 737 A.2d at 70 (noting that defense expert’s review was limited to defendant’s social history up to age seventeen); State v. Cooper, 151 N.J. 326, 345, 700 A.2d 306, 314 (1997) (Cooper I) (same), cert. denied, 528 U.S. 1084, 120 S.Ct. 809, 145 L.Ed.2d 681 (2000). Instead, defendant sought to present his “complete” social history and criminal record up to the time of Padilla’s murder, which occurred exactly [619]*619twenty days after his thirtieth birthday, and approximately eight months before his assault of Trooper Gardner. He did so, apparently, in the hope of demonstrating those facets of his life experience that contributed to the man he had become.
The reliability of Nardone’s report and proposed testimony was therefore directly at issue, and subject to cross-examination and rebuttal as to its accuracy and completeness in light of the glaring omissions of defendant’s 1983 and 1995 convictions from Nardone’s otherwise comprehensive discussion of defendant’s extensive juvenile and adult record. See State v. Josephs, supra, 174 N.J. at 127-28, 803 A.2d at 1123-24 (holding that State was entitled to expose bias, inaccuracy, and incompleteness of mitigating expert’s social history testimony); Timmendequas I, supra, 161 N.J. at 592-96, 737 A.2d at 96-99 (same).
The trial court’s sensible solution of sanitizing the omitted 1983 and 1995 convictions to first- and second-degree assault convictions was both fair and reasonable.11 A capital defendant is not entitled to present an incomplete picture of his social history and criminal record by carving out and omitting those offenses he deems unfavorable to his mitigation defense. Although he may limit his social history evidence to a particular subject area or certain aspects of his life, such as childhood, he may not omit sanitized adult convictions from an otherwise “complete” social and criminal history by excising a conviction that falls within the relevant time period or by terminating his adult social history at a date immediately preceding the offense he wishes to exclude from the jury’s consideration. See Harris II, supra, 165 N.J. at 323, [620]*620757 A.2d at 232 (allowing defendant to limit social history evidence to mental condition as young child).
We also note that the sanitized convictions were no more prejudicial than the criminal history or other unflattering details described in Nardone’s report. For example, Nardone would have testified that defendant had shot an eighteen-year-old with a BB-gun with the intent of scarring the victim and had committed theft offenses, which were the subject of juvenile adjudications. She would have testified that defendant was convicted as an adult of possession of a machete for an unlawful purpose, that he was arrested for drug offenses, and that he had beaten Dawn Archer on the night of Melissa Padilla’s murder. She also would have testified that defendant did not respond well to probation or counseling. In other words, the two sanitized convictions would not have impugned the spotless reputation of a pillar of the community, but would have been additional pieces of a picture of a dysfunctional person who did not conform his conduct to the law.
Nardone’s report presented itself as “a complete psychosocial history” of defendant from birth to age thirty, yet the omitted 1983 assault conviction occurred when defendant was nineteen years old. The 1995 “first-degree assault” occurred just eight months after the Padilla murder, the point at which Nardone abruptly concluded her social history. In light of the temporal proximity of the “first-degree assault” conviction to defendant’s thirty-year social history, the conviction was clearly relevant to show that the social history was less than “complete.”
Defendant misconstrues Rose, supra, 112 N.J. 454, 548 A.2d 1058, and State v. Long, 119 N.J. 439, 575 A.2d 435 (1990), as standing for the proposition that a capital defendant’s prior and subsequent convictions are not fair game for cross-examination or rebuttal to refute the existence of a mitigating factor unless the defendant adduces “good-character” evidence through his mitigation witnesses. In Rose, supra, we held that the use of specific instances of prior misconduct to impeach a capital defendant’s character evidence was permitted under Evid. R. 46, the predeces[621]*621sor rule to N.J.R.E. 404(c) and N.J.R.E. 405. 112 N.J. at 503, 548 A.2d at 1082-83. In that case, the defendant’s mitigation witnesses had testified about his good character, including his kindness to children and his thoughtful, earing nature. Id. at 498, 503, 548 A.2d at 1082-83. This Court found no fault with the prosecution introducing evidence of “specific instances of the person’s conduct” to rebut that characterization. Id. at 502-03, 548 A.2d at 1082-83 (quoting Evid. R. 46). N.J.R.E. 405(b) provides that “[w]hen character or a trait of character of a person is an essential element of a charge, claim, or defense, evidence of specific instances of conduct may also be admitted.” (Emphasis added). As in Rose, defendant placed at issue a trait of his character through the introduction of mitigation evidence. Unlike Rose, this defendant was not introducing evidence to prove that he was of good character, but rather evidence to show the experiences, forces, and circumstances that shaped his personality. The State, nevertheless, was not obliged to sit back silently while defendant picked only those pieces of evidence that suited his purposes and that painted an incomplete picture. As the commentary to N.J.R.E. 405 observes: “In the penalty phase of a capital case, N.J.S. 2C:ll-3c(2)(b) permits the State to ignore the prohibition against using prior specific instances of conduct, if such evidence serves to rebut evidence offered by the defendant.” Biunno, Current N.J. Rules of Evidence, comment 4 on N.J.R.E. 405 (2003).
In Long, supra, we held that the defendant’s 1985 simple-assault conviction was properly introduced to rebut his assertion under mitigating factor N.J.S.A. 2C:11-3c(5)(f) that he lacked a significant history of prior criminal activity. 119 N.J. at 459, 502, 575 A.2d 435, 444. We rejected the defendant’s claim that the State’s rebuttal evidence was limited to the timeframe of mitigating character evidence that he had introduced up to the subject 1982 murder, which he committed at the age of twenty-four. Id. at 451, 459, 502, 575 A.2d at 444, 466-67. Instead, we implicitly held that “[njeither case law nor Court Rules requires such limitation” on the rebuttal of mitigating evidence that a defendant has artificially confined to a period of time closely preceding a [622]*622relevant conviction he wishes to exclude. Id. at 502, 575 A.2d at 466.
Defendant’s 1983 and 1995 convictions were relevant and admissible at his penalty phase because he intended to give a distorted view of his troubled youth and adulthood by including all of his convictions, but two. Had defendant chosen to limit his social history to his childhood, Nardone could have testified to defendant’s alleged troubled home- and school-life, his repeated sexual abuse by an adult male from ages thirteen to sixteen, his many contacts with the juvenile justice system, and his struggles with drug and alcohol abuse. Instead, defendant gambled on an all-or-nothing strategy of either presenting a social history up to the night of Padilla’s murder without the 1983 and 1995 sanitized convictions or presenting no social history at all. In light of the powerful and damaging rebuttal testimony expected from Dr. Weiner, defendant was in the best position to know whether the presentation of a mitigation defense would have made more likely the return of a death sentence. Defendant made a calculated decision to deny the State the opportunity to call Dr. Weiner in rebuttal. That strategy did not achieve its desired result. The failure of that strategy, however, does not give rise to a valid claim that the trial court abused its discretion by giving the State permission to rely on relevant rebuttal evidence.
VIII.
Jurors’ Consideration of Rejected Aggravating Factor
In the penalty phase, the trial court charged the jury on basic capital sentencing concepts. The jury was instructed to determine whether the State had proven the existence of each aggravating factor beyond a reasonable doubt. The jury was told that unless it unanimously found an aggravating factor defendant would be sentenced to life imprisonment. The jury also was told that if it found one or more aggravating factors, then it would have to decide whether there was support for any of the claimed mitigating factors. The court further instructed the jury that if even a [623]*623single juror found evidential support for a mitigating factor, all jurors would have to consider that factor in the final weighing process. Last, the jury was charged that if all twelve jurors concluded that one or more aggravating factors outweighed the mitigating factors beyond a reasonable doubt, defendant would be sentenced to death, and if not, he would be sentenced to a term of life imprisonment.
The instructions omitted one critical point: if the jury did not unanimously find the existence of an aggravating factor, that factor was to be given no further consideration by any juror in the sentencing process. The poorly crafted special verdict sheet did nothing to ameliorate the absence of that instruction. The verdict sheet returned by the jury indicated that at least one juror balanced a rejected aggravating factor against the mitigating factors. The verdict sheet also suggested that the forbidden subject of the rejected aggravating factor may have been discussed in the final phase of the penalty deliberations. Defendant’s trial ended before this Court decided Nelson II, supra, 173 N.J. at 442, 803 A.2d at 15, and State v. Koskovich, 168 N.J. 448, 524, 776 A.2d 144, 192-93 (2001), which squarely held that a rejected aggravating factor is entitled to no further consideration by any juror in the sentencing process. In light of our decision to reverse, we take this occasion to reaffirm that principle and repeat the guidance we have given to the trial courts for future cases.
In the penalty phase, the State presented three aggravating factors: murder involved aggravated assault and/or torture, N.J.S.A. 2C:ll-3c(4)(c); murder committed for the purpose of escaping detection, apprehension, trial, punishment or confinement for another offense committed by the defendant, N.J.S.A. 2C:11-3c(4)(f); and murder committed while defendant engaged in the commission of, or an attempt to commit, or flight after committing or attempting to commit robbery and/or sexual assault, N.J.S.A. 2C:ll-3c(4)(g). The jury unanimously found that the State had proven the existence of the 4(c) and 4(g) aggravating factors beyond a reasonable doubt. The jury, by a vote of eleven-to-one, [624]*624rejected the existence of the 4(f) aggravating factor. Thus, the jury was authorized to weigh only the (4)(c) and (4)(g) aggravating factors against the mitigating factors submitted by defendant. See Nelson II, supra, 173 N.J. at 442-43, 803 A.2d at 15-16; Koskovich, supra, 168 N.J. at 519, 776 A.2d at 188-89 (citing Bey II, supra, 112 N.J. at 159, 548 A.2d at 905). As noted, the court failed to instruct the jury explicitly that all jurors must disregard any aggravating factor not unanimously found by the entire jury.
The lack of clarity in the verdict sheet underscored the importance of such an instruction. The language of the verdict sheet in this case was identical to the “ambiguous wording” of the one discussed in Nelson II, supra, 173 N.J. at 445-46, 803 A.2d at 17. The verdict sheet read: “If you have unanimously found more than one aggravating factor present, then indicate as to each such factor whether it by itself outweighs the mitigating factors beyond a reasonable doubt.” After deliberations, the jury returned the completed verdict sheet:
Aggravating Factor “a” NO (0) YES (12) [(4)(c)]
Aggravating Factor “b” NO (11) YES (1) K4)(f)3
Aggravating Factor “c” NO (0) Yes (12) C(4)(g)]
Defense counsel did not object at any time to the court’s instructions on the weighing process or to the verdict sheet. Upon the return of the special verdict sheet, the trial court did not ask the jury for a clarification of its answers.
The verdict sheet reveals that all twelve jurors found that the 4(c) and 4(g) aggravating factors outweighed the mitigating factors beyond a reasonable doubt. But it also shows that one juror found that the rejected 4(f) aggravating factor outweighed the mitigating factors beyond a reasonable doubt. That aggravating factor should not have received any consideration in the final [625]*625weighing process. In view of the fact that a rejected aggravating factor played some role in those final deliberations, we do not know to what extent it may have infected the weighing process of those factors properly considered by the jury.
This Court has emphasized that “ ‘clear and correct jury instructions are essential for a fair trial,’ ” Koskovich, supra, 168 N.J. at 507, 776 A.2d at 180 (quoting State v. Brown, 138 N.J. 481, 522, 651 A.2d 19, 39 (1994)), and are “ ‘even more crucial in a capital case because of the jury’s responsibility to decide whether a defendant shall live or die.’ ” Id. at 524, 776 A.2d at 192 (quoting Bey II, supra, 112 N.J. at 162, 548 A.2d at 906). “[W]e are especially sensitive to ensuring the correctness of jury instructions concerning the jury’s balancing of aggravating and mitigating faetors[] because that balancing represents the core of the jury’s function in the penalty phase.” Ibid. “The need for clear verdict sheet directions is no less important” because “[i]f verbal instructions are unclear, or if jurors do not fully comprehend verbal instructions, the typewritten verdict sheet is likely the primary road map they will use to direct their deliberative path.” Nelson II, supra, 173 N.J. at 449, 803 A.2d at 20. Given the requirement for “enhanced ‘reliability in the determination that death is the appropriate punishment in a specific case,’ ” Bey II, supra, 112 N.J. at 162, 548 A.2d at 906 (quoting Caldwell v. Mississippi, 472 U.S. 320, 330, 105 S.Ct. 2633, 2640, 86 L.Ed.2d 231, 240 (1985)), “there must be little chance that the jury as a whole, or even an individual juror, is confused about the process.” Koskovich, supra, 168 N.J. at 526, 776 A.2d at 193.
Defendant did not object to the jury charge at the time it was given; therefore, any challenge to the charge must come under the plain error standard. State v. Josephs, supra, 174 N.J. at 98, 803 A.2d at 1106; see State v. Chew, 150 N.J. 30, 82, 695 A.2d 1301, 1328 (1997) (Chew I) (noting that defendant must establish “ ‘clear’ ” or “ ‘obvious’ ” error affecting “ ‘substantial rights’ ” to establish plain error) (quoting United States v. Olano, 507 U.S. 725, 734, 113 S.Ct. 1770, 1777, 123 L.Ed.2d 508, 519 [626]*626(1993)), cert. denied, 528 U.S. 1052, 120 S.Ct. 593, 145 L.Ed.2d 493 (1999). Even under that standard, however, “[a]s an indication of the paramount importance of accurate jury instructions, we have held that erroneous instructions on material issues are presumed to be reversible error.” State v. Marshall, 173 N.J. 343, 359, 801 A.2d 1142, 1154 (2002) (Marshall IV) (citing State v. Martin, 119 N.J. 2, 15, 573 A.2d 1359,1366 (1990)).
We conclude that the jury instructions and verdict sheet suffered from many of the infirmities discussed in Koskovich and Nelson II. As in Koskovich, supra, the court below did not “explicitly inform the jurors that if the jury as a group rejected an aggravating factor, no one juror could consider that factor during the weighing process, even if he or she voted for that factor individually.” 168 N.J. at 524, 776 A.2d at 193. In addition to the lack of a “clear and explicit instruction” on the non-role of a rejected aggravating factor, id. at 525, 776 A.2d at 193, here “the verdict sheet did not fail simply to clarify the law; it exacerbated the ambiguity by suggesting the consideration of aggravating factors not unanimously found.” Nelson II, supra, 173 N.J. at 454, 803 A.2d at 23. One juror, presumably the single juror who found the 4(f) aggravating factor to be established beyond a reasonable doubt, proceeded to erroneously balance that rejected factor against defendant’s mitigating factors.
In view of our reversal of defendant’s conviction, we need not determine whether the imperfect charge and the ambiguous verdict sheet combined to meet the plain error standard. The Model Jury Charge has been modified to encompass our holdings in Nelson II and Koskovich:
In order for the jury to find an aggravating factor, all jurors must unanimously agree, beyond a reasonable doubt, that the aggravating factor is present. If one or more of you do not agree that the aggravating factor is present, the jury is deemed not to have found that aggravating factor and no juror is permitted to consider that factor or any evidence introduced solely in suppoH of that aggravating factor during the process of weighing aggravating against mitigating factors. In other words, if eleven jurors found that an aggravating factor exists but one juror did not find that the aggravating factor exists, no juror shall consider that aggravating factor in his or her deliberations.
[627]*627[Judges Bench Manual for Capital Causes, Appendix J-34 (Oct. 22, 2002) (emphasis added).]
The trial court in this case did not have the benefit of our holdings in Nelson II and Koskovich or the amended Model Jury-Charge. We trust that trial courts in all future penalty-phase proceedings will follow the Model Jury Charge.
IX.
During the jury selection process in the penalty phase, defendant moved to dismiss M.T. for cause on the basis of that juror’s limited involvement with Melissa Padilla’s children. At the time of the murder, M.T. was friendly with the Padilla children’s babysitter, Jasmine. M.T. had seen the Padilla children four or five times, and on one of those occasions, when Jasmine was babysitting, M.T. was present with the children for three or four hours. M.T. learned of Padilla’s murder from Jasmine. She lost touch with Jasmine after the murder and, by the time of her jury service, had not seen her or the Padilla children for five or six years. M.T. had never met Padilla or been to her home, and could not recall the children’s names or Jasmine’s last name. She expressed her sympathy for the children who “wouldn’t grow up with a mother.” In light of M.T.’s assurances that she was capable of remaining fair and impartial, the court rejected defendant’s challenge for cause, finding no “reason particularly to disqualify [M.T.].”
Defendant contends that the trial court abused its discretion under Rule l:8-3(d) by refusing to grant him one extra peremptory challenge to remove M.T. We need not decide the issue because we are remanding for a new trial. Trial courts generally are given wide discretion to make voir dire decisions based on their first-hand opportunity to observe prospective jurors; they are in the best position to make judgments about a juror’s ability to be fair and dispassionate. However, we have serious misgivings about leaving a juror on the panel with a [628]*628connection to the victim’s family in a capital case and, therefore, we offer some observations for future guidance.
As in the guilt phase, the court used a struck jury system, in which peremptory challenges were made only after a sufficient number of potential jurors had been qualified. The selection of the fifteen-member-jury began with a pool of forty-seven prequalified panelists. Twenty peremptory challenges were allotted to defendant, and twelve to the State. See R. 1:8 — 3(d). Defendant exercised two peremptory challenges before M.T. was called to the jury box, and another seventeen after M.T. was qualified as a juror. When each party had one peremptory challenge remaining, M.T. was still in the box, and only two of the forty-seven qualified panelists remained in the selection pool. At that point, defendant requested an extra peremptory challenge to remove M.T. See R. l:8-3(d).
It is difficult to discern from the cold record why defendant chose not to exercise a peremptory challenge to remove M.T. earlier in the jury selection process. Presumably, he believed those jurors he removed by peremptory challenge were more likely than M.T. to return a death verdict. In any event, the trial court denied defendant the extra peremptory challenge. The court expressed concern that there were only two pre-qualified panelists remaining for selection, with both the prosecution and defense holding one remaining peremptory challenge apiece. The court clearly was disinclined to pre-qualify any more prospective jurors, and the granting of an additional challenge might have required it to do so. As in the guilt-phase voir dire, the court appeared excessively concerned with saving time in the jury selection process. The court stated that the prosecution could transfer its remaining peremptory to defendant if it wished, but that it would not grant an additional challenge. Not surprisingly, the State did not part with its last peremptory challenge. Rather than excuse M.T., defendant exercised his last peremptory challenge to remove W.B., a prospective juror who had expressed his belief that life sentences are more expensive than executions, and [629]*629“raise[ ] the taxes up.” M.T. thus became a deliberating juror and returned a death verdict.
Trial courts conducting voir dire are best able to evaluate the credibility and sincerity ofta prospective juror in a way that the record will never allow. For that reason, trial courts are granted broad discretion to assess the qualifications of jurors, and the exercise of that discretion “will ordinarily not be disturbed on appeal.” State v. Josephs, supra, 174 N.J. at 104, 803 A.2d at 1109 (quoting State v. Jackson, 43 N.J. 148, 160, 203 A.2d 1, 7 (1964), cert. denied, 379 U.S. 982, 85 S.Ct. 690, 13 L.Ed.2d 572 (1965) ). Nevertheless, a defendant may be entitled to a new trial if he has been forced to use one of his allotted peremptory challenges in order to excuse a juror who should have been excused for cause. State v. Deatore, 70 N.J. 100, 104-06, 358 A.2d 163, 165-67 (1976); see also State v. Pereira, 202 N.J.Super. 434, 439, 495 A.2d 431, 434 (App.Div.1985) (reversing conviction “where defendant was obliged to use a peremptory challenge to excuse a venireman whom he had a right to challenge for cause and to proceed to trial with an unsatisfactory jury”).
Expedience must never override the painstaking process of selecting impartial jurors in a capital case. The prequalification of additional jurors in the penalty phase, however inconvenient or time-consuming, should not matter if the objectivity of a juror seated in the box is subject to some reasonable doubt. We think it ill-advised, as a general rule, to seat any juror who is acquainted with a murder victim’s loved ones, no matter how convincingly that individual proclaims his or her ability to remain impartial. Those good-faith assurances do not assuage our common-sense concern that someone who had met and interacted with the young children of the victim of a heinous sexual assault and murder might be incapable of impartiality. See State v. Koskovich, supra, 168 N.J. at 511, 776 A.2d at 183 (noting unlikelihood that capital jurors can “divest themselves of ordinary human emotions and disregard all feelings they may have for the victim’s family”); see also Deatore, supra, 70 N.J. at 104-05, 358 A.2d at [630]*630165-66 (reversing robbery convictions because trial court erroneously refused, after all defense peremptory challenges were exhausted, to grant extra peremptory challenge or permit further voir dire of juror who insisted she would be fair and impartial despite acquaintance with victim both personally and professionally); Jackson, supra, 43 N.J. at 157, 160, 203 A.2d at 7 (reversing capital convictions where trial court erroneously refused, after all defense peremptory challenges were exhausted, to dismiss for cause challenged replacement juror who insisted he would be fair and impartial despite close ties to local law enforcement witness).
“In any sound judicial system it is essential not only that justice be done but also that it appear to be done.” Jackson, supra, 43 N.J. at 160-61, 203 A.2d at 8 (emphasis added). This is true particularly in capital cases, where life itself hangs in the balance. Id. at 156, 203 A.2d at 5. A prospective juror whose obvious connection with a party might affect that juror’s impartiality should be excused by the court, even if counsel does not make the request. Deatore, supra, 70 N.J. at 106, 358 A.2d at 167. It is better to err in favor of removing a juror where there is evidence of potential partiality or bias, than to permit that juror to sit in judgment, leaving the fairness of a capital trial in doubt.
We, therefore, caution the trial courts that in any capital case in which a prospective juror is acquainted with the victim’s loved ones, the most prudent course is to remove that juror to forestall both potential injustice and the appeai*ance of injustice.
X.
Jury Instmction on Future Dangerousness
At the penalty-phase charge conference, the trial court reviewed with counsel the instructions it intended to give the jury. Defense counsel specifically requested that the court instruct the jury on the model charge that directs the jury not to consider the defendant’s future dangerousness in determining whether to impose a capital or non-capital penalty. Judges Bench Manual for Capital [631]*631Causes, Appendix J-54 to -55. The court charged the jury in the very language endorsed by defense counsel:
If any of you find that the State has failed to prove beyond a reasonable doubt that the aggravating factor or factors outweigh the mitigating factor or factors, or if you cannot reach a unanimous verdict on the question of punishment, then the punishment shall be imprisonment for the defendant____You must not speculate as to the precise sentence. This defendant might receive from me between 30 yeai-s and life. You must not speculate as to whether or not this defendant would or would not be released after 30 years. As far as you are concerned if you have not decided upon death for the defendant, you must then assume that his possible future release would not endanger society.
[(Emphasis added).]
Defendant now takes issue with the word “assume” and argues that the court should have given a more forceful instruction that future dangerousness was to be completely disregarded and given no consideration. Defendant contends that the trial court erroneously injected the concept of future dangerousness into the jury deliberations in violation of his federal due process rights and this State’s capital jurisprudence. We find that this argument lacks merit.
The State may not argue to a jury that a reason to return a capital verdict is the threat defendant’s future release will pose to the public if he were to be sentenced to a term of imprisonment. State v. Loftin, 146 N.J. 295, 371, 680 A.2d 677, 714 (1996) (Loftin I); State v. Rose, supra, 112 N.J. at 521, 548 A.2d at 1093. A defendant’s future dangerousness is not a statutory aggravating factor and must play no part in a jury’s deliberations. Loftin I, supra, 146 N.J. at 371, 680 A.2d at 714; see N.J.S.A. 2C:11-3c(4)(a)-(1) (listing aggravating factors). The trial court’s penalty-phase instructions correctly informed the jury of the possible sentences that defendant faced were it to return a verdict other than death. See Loftin I, supra, 146 N.J. at 370, 680 A.2d at 713-14 (requiring that capital sentencing jury be fully informed of full range of sentencing options). While describing the minimum thirty-year period of parole ineligibility of a non-capital sentence, the court advised the jury that it “must then assume that [the defendant’s] possible future release would not [632]*632endanger society.” Although there would have been no harm in including in the charge the additional language defendant now suggests, to the effect that the defendant’s future dangerousness is not to be discussed or given any consideration in the jury’s deliberations, defendant made no such request. The court properly charged the jury on the law, admonishing the jury not to take future dangerousness into account. Defense counsel approved that instruction. We do not find that the court’s instruction prompted the jury to consider the very subject that it was expressly forbidden to discuss. See Nelson II, supra, 173 N.J. at 447, 803 A.2d at 18 (“[Tjhere can be no assumption that the jury did not faithfully follow the [court’s] admonition.”) (quoting State v. Manley, supra, 54 N.J. at 271, 255 A.2d at 200). We conclude that there was no error in the charge.
XI.
Submission of Aggravating Factors to Grand Jury
Defendant urges this Court to reconsider its decision in Martini I, supra, which held that the New Jersey Constitution does not require aggravating factors to be submitted to a grand jury in a capital case. 131 N.J. at 228, 619 A.2d at 1233-34. In Martini I, the Court rejected the theory that “aggravating factors are the junctional equivalent of elements of the crime of capital murder, thereby necessitating their presentment to a grand jury and inclusion in an indictment.” Id. at 222-23, 619 A.2d at 1231 (emphasis added). Defendant argues that the legal landscape upon which Martini I rests has been so altered that the rationale supporting the decision is no longer viable.
Since our decision in Martini I, the United States Supreme Court in Ring v. Arizona declared unconstitutional Arizona’s capital sentencing law that delegated to the judge rather than the jury the determination whether to impose the death penalty based on the finding of aggravating factors. 536 U.S. 584, 588-89, 122 S.Ct. 2428, 2432, 153 L.Ed.2d 556, 564 (2002). The Court in Ring held that “[bjecause Arizona’s enumerated aggravating factors [633]*633operate as ‘the junctional equivalent of an element of a greater offense,' the Sixth Amendment requires that they be found by a jury.” Id. at 609, 122 S.Ct. at 2443, 153 L.Ed.2d at 577 (quoting Apprendi v. New Jersey, 530 U.S. 466, 494 n. 19, 120 S.Ct. 2348, 2365 n. 19, 147 L.Ed.2d 435, 457 n. 19 (2000)) (emphasis added). Ring did not address the question whether aggravating factors in a capital case must be considered by a grand jury pursuant to the Fifth Amendment. Id. at 597 n. 4, 122 S.Ct. at 2437 n. 4, 153 L.Ed.2d at 569 n. 4. Under Article I, Paragraph 8 of our State Constitution, however, the State must present proof of every element of an offense to the grand jury and specify those elements in the indictment.12 Martini I’s conclusion that a capital aggravating factor does not constitute an essential element of capital murder no longer has validity in the wake of Ring. In view of this new reality, we can see no principled reason for our continued adherence to the notion that aggravating factors are not elements of capital murder. In all future cases, in accord with our State Constitution, aggravating factors in a capital case must be submitted to a grand jury.
Our State Constitution guarantees that “[n]o person shall be held to answer for a criminal offense” unless charged by a grand jury in the form of an indictment. N.J. Const, art. 1, ¶ 8. That constitutional provision requires that the State present to the grand jury proof to support every element of the offense before the return of an indictment and that every element must be alleged in the indictment. State v. Hogan, 144 N.J. 216, 227, 676 A.2d 533, 538 (1996); State v. Wein, 80 N.J. 491, 497, 404 A.2d 302, 305 (1979). In the absence of such proof, the indictment is subject [634]*634to dismissal. See State v. La Fera, 35 N.J. 75, 81, 171 A.2d 311, 314 (1961).
Although the legislative determination of what constitutes the elements of an offense is entitled to considerable weight, “a State cannot through mere characterization change the nature of the conduct actually targeted.” Apprendi, supra, 530 U.S. at 493 n. 18, 120 S.Ct. at 2364 n. 18, 147 L.Ed.2d at 457 n. 18. Stated differently, the State cannot clothe an element as a sentencing factor in an effort to elude the right to trial by jury. See Ring, supra, 536 U.S. at 605, 122 S.Ct. at 2441, 153 L.Ed.2d at 574 (noting that “[m]erely because the state legislature placed its hate crime sentence enhancer within the sentencing provisions of the criminal code does not mean that the finding of a biased purpose to intimidate is not an essential element of the offense”) (quoting Apprendi, supra, 530 U.S. at 495, 120 S.Ct. at 2365-66, 147 L.Ed.2d at 458). By that same reasoning, the State cannot mask an element as a sentencing factor to deny a defendant his right to a grand jury presentation. We now must revisit whether aggravating factors are elements of capital murder.
Under New Jersey’s death penalty statute, N.J.S.A. 2C:ll-3c, the State must prove specific facts in each step of a three-tiered process before a defendant may be sentenced to death. Fust, the State must prove beyond a reasonable doubt that the defendant purposefully or knowingly caused death or serious bodily injury resulting in death. N.J.S.A. 2C:ll-3a(l), (2). Second, the State must prove beyond a reasonable doubt one of the capital “triggers” in order to advance the defendant to the penalty-phase trial.13 N.J.S.A. 2C:11-3e. The State must establish that the defendant,
[635]*635committed the homicidal act by his own conduct; or who as an accomplice procured the commission of the offense by payment or promise of payment of anything of pecuniary value; or who, as a leader of a narcotics trafficking network ... commanded or by threat or promise solicited the commission of the offense, or, if the murder occurred during the commission of the crime of terrorism, any person who committed the crime of terrorism____
[N.J.S.A. 2C:ll-3e.]
The jury considers the capital “triggers” only after finding a defendant guilty of knowing or purposeful murder. See State v. Feaster, 156 N.J. 1, 42-43, 716 A.2d 395, 414-15 (1998), cert. denied, 532 U.S. 932, 121 S.Ct. 1380, 149 L.Ed.2d 306 (2001); State v. Gerald, 113 N.J. 40, 100, 549 A.2d 792, 823 (1988). Third, in the penalty-phase trial, the State must prove beyond a reasonable doubt the existence of any alleged statutory aggravating factors. N.J.S.A. 2C:ll-3c(2)(a). If the jury finds one or more aggravating factors, it must then determine whether those outweigh all of the mitigating factors beyond a reasonable doubt. N.J.S.A. 2C:ll-3c(3)(a).
The capital sentencing procedure in New Jersey, therefore, requires three different findings beyond a reasonable doubt by the jury before an accused receives a death sentence: (1) knowing and purposeful killing, (2) one of the capital triggers, and (3) one or more aggravating factors. The jury’s failure to find any one of those facts renders the defendant ineligible for the death penalty. Additionally, before a defendant may be sentenced to death, the jury must then find that the aggravating factors outweigh beyond a reasonable doubt the mitigating factors. N.J.S.A 2C:11-3c(3)(a). The following chart summarizes the role of the different factual findings made by the jury in a capital case.
[636]*636[[Image here]]
The finding of a purposeful or knowing killing by the jury only exposes the defendant to a sentence between thirty years and life imprisonment. Without any other finding by the jury, life imprisonment is the maximum allowable sentence under the capital-murder statute. N.J.S.A 2C:ll-3b. The jury must find a capital trigger and an aggravating factor to elevate the punishment from a term of imprisonment to death.
In Martini I, supra, this Court stated that the Capital Penalty Act defined criminal homicide as having only two constituent elements that “must be presented to a grand jury and included in the murder indictment”: a purposeful or knowing state of mind, and causation. 131 N.J at 224, 619 A.2d at 1232. Although not defined as elements of capital murder under our decisional law, see, e.g., State v. Cruz, 171 N.J. 419, 427, 794 A.2d 165, 171 (2002) (Cruz II), we also require that the capital “triggers” defined in N.J.S.A. 2C:ll-3c be presented to a grand jury and specified in an indictment. R. 3:7-3(b).14
On the other hand, under the Act, the prosecutor alone, not a grand jury, determines whether to charge the defendant with capital murder based on a finding that there is evidence to support the existence of an aggravating factor. If the prosecutor makes [637]*637an affirmative finding of an aggravating factor, he must serve on the defendant a Notice of Aggravating Factors. The statute requires:
Prior to the commencement of the sentencing proceeding, or at such time as he has knowledge of the existence of an aggravating factor, the prosecuting attorney shall give notice to the defendant of the aggravating factors which he intends to prove in the proceeding.
[N.J.S.A. 2C:ll-3c(2)(e).]
The Martini I Court reasoned that the State Constitution does not require aggravating factors to be submitted to the grand jury in capital cases because those factors are not elements; they “merely enlarge the available sentencing options for certain defendants.” 131 N.J. at 224, 619 A.2d at 1232. The Court found that because aggravating factors “affect only the severity of the sentence that a convicted defendant faces,” the concerns underlying Paragraphs 8 and 1015 of Article 1, did not come into play. Id. at 223, 619 A.2d at 1231. Those concerns are that a defendant must be “adequately informed of the alleged elements of the crimes” in order to “mount an informed defense” and that the defendant must not be subject to “prosecution of unfounded charges.” Ibid.
The Martini I Court noted that in capital cases the defendant has the opportunity to challenge the sufficiency of the evidence supporting an aggravating factor in a summary proceeding before the trial court. Id. at 227, 619 A.2d at 1233 (citing State v. McCrary, 97 N.J. 132, 142, 478 A.2d 339, 344-45 (1984)). The Court was satisfied that the notice provisions of N.J.S.A. 2C:11-3c(2)(e) and a McCrary hearing adequately provided capital defendants “the protections contemplated by the Constitution: adequate notice and well-founded prosecutions.” Id. at 227, 619 A.2d at 1233.
Clearly, however, those assurances of notice and a well-founded prosecution are not adequate substitutes for the constitutional [638]*638right of a grand jury presentation if aggravating factors are elements of the crime of capital murder. The grand jury serves an important and historic purpose in standing between the defendant and the power of the State, protecting the defendant from unfounded prosecutions. Hogan, supra, 144 N.J. at 227, 676 A.2d at 538; State v. Del Fino, 100 N.J. 154, 164, 495 A.2d 60, 64-65 (1985). The grand jury, which is comprised of members of a cross-section of the community, represents a democratic safeguard to our judicial system. State v. Gilmore, 103 N.J. 508, 526, 511 A.2d 1150, 1159 (1986); State v. Zavala, 259 N.J.Super. 235, 241, 611 A.2d 1169, 1172 (Law Div.1992) (citing Ballard v. United States, 329 U.S. 187, 195, 67 S.Ct. 261, 265, 91 L.Ed. 181, 187 (1946)). We rely on the common wisdom of ordinary citizens to make the important fact-finding decisions that must precede the filing of an indictment, even in murder cases.
A majority of States have abandoned the indictment requirement and permit prosecutions to proceed on the basis of an information filed by the prosecutor. See generally Wayne R. LaFave et al., Criminal Procedure § 1.3(m) (2d ed. 1999 & Supp. 2002). In those States, typically, there is a judicial hearing to determine whether probable cause supports the charges. Id. § 1.3(n). The filing of a Notice of Aggravating Factors may be consonant with the procedures in those States that commence prosecution by information. However, the drafters of our 1947 Constitution kept them faith in the institution of the grand jury. In New Jersey, the grand jury remains a constitutional bulwark against hasty and ill-founded prosecutions and continues to lend legitimacy to our system of justice by infusing it with a democratic ethos. Article 1, Paragraph 8 embodies a fundamental right that has continuing vitality today.
Martini I, supra, suggested that “factors that affect only the severity of a sentence” and “merely enlarge the available sentencing options for certain defendants” are not elements of an offense and, therefore, do not implicate the grand jury protections of Article I, Paragraph 8. 131 N.J. at 223-24, 619 A.2d at 1231. [639]*639However, those statements may be read too broadly given that many of the crimes defined in the New Jersey Code of Criminal Justice grade the severity of the crime, and accordingly the severity of the sentence, based on the finding of an element of a greater crime. Those elements have very much the look of aggravating factors. For example, robbery is a second-degree crime, but the use of a weapon in the course of the robbery elevates the robbery to a first-degree crime. N.J.S.A 2C:15-lb. A defendant commits a third-degree aggravated assault when he purposely or knowingly causes significant bodily injury and a second-degree aggravated assault when he causes serious bodily injury. N.J.S.A 2C:12-lb(l), (7). In both of the above examples, an aggravating factor that the Legislature has designated as an element of the offense affects the “severity of a sentence” or “enlarge[s] the sentencing options.”
We note that the line demarcating elements and aggravating factors has proven elusive. Apprendi, supra, 530 U.S. at 494, 120 S.Ct. at 2365, 147 L.Ed.2d at 457. As a matter of constitutional law, aggravating factors, other than a prior criminal conviction, that raise a sentence beyond the statutory maximum are deemed elements that must be tried to a jury and proven beyond a reasonable doubt. Id. at 490, 120 S.Ct. at 2362-63, 147 L.Ed.2d at 455; State v. Stanton, 176 N.J. 75, 95, 820 A2d 637, 649 (2003) (quoting Harris v. United States, 536 U.S. 545, 556, 122 S.Ct. 2406, 2414, 153 L.Ed.2d 524, 537 (2002)). We, therefore, turn to the recent line of United States Supreme Court cases that provide a constitutional framework for defining elements of a criminal offense.
In Apprendi, supra, the Supreme Court invalidated a statutory scheme that allowed a sentencing judge to make a factual finding by a preponderance of the evidence that raised the defendant’s sentence from a second-degree crime, punishable by a term of between five to ten years in prison, to a first-degree crime, punishable by a term of between ten and twenty years. 530 U.S. at 474, 120 S.Ct. at 2354, 147 L.Ed.2d at 445. The Court declared [640]*640unconstitutional New Jersey’s hate crime law because that statute required a jury to decide by proof beyond a reasonable doubt whether the defendant was guilty of the second-degree offense of possession of a weapon for an unlawful purpose, but left to a judge to decide by the preponderance of evidence standard whether the defendant possessed the weapon for the purpose of intimidating the victim because of his or her “race, color, gender, handicap, religion, sexual orientation or ethnicity.” Id. at 468-69, 491-92, 120 S.Ct. at 2351, 2363, 147 L.Ed.2d at 442, 455. The judge’s finding exposed the defendant to an “extended term” of double the sentence permitted by the jury’s finding. Id. at 469, 120 S.Ct. at 2351, 147 L.Ed.2d at 442.
In assessing whether the legislative scheme comported with the federal constitution, the Court engaged in a historical analysis under the common law.
Any possible distinction between an “element” of a felony offense and a “sentencing factor” was unknown to the practice of criminal indictment, trial by jury, and judgment by court as it existed during the years surrounding our Nation’s founding. As a general rule, criminal proceedings were submitted to a jury after being initiated by an indictment containing “all the facts and circumstances which constitute the offence, ... stated with such certainty and precision, that the defendant ... may be enabled to determine the species of offence they constitute, in order that he may prepare his defense accordingly ... and that there may be tw doubt as the judgment which should be given, if the defendant be convicted.”
[Id. at 478, 120 S.Ct. at 2356, 147 L.Ed.2d at 448 (quoting J. Archbold, Pleading and Evidence in Criminal Cases 44 (15th ed. 1862)) (footnote omitted).]
It also noted:
“Where a statute annexes a higher degree of punishment to a common-law felony, if committed under particular circumstances, an indictment for the offence, in order to bring the defendant within that higher degree of punishment, must expressly charge it to have been committed under those circumstances, and must state the circumstances with certainty and precision.”
[Id. at 480, 120 S.Ct. at 2357, 147 L.Ed.2d at 449 (quoting Archbold, Pleading and Evidence in Criminal Cases, supra, at 51 (internal citation omitted)).]
As can be seen, the specificity we require in indictments has antecedents in the traditions of our law.
The Apprendi Court held that due process requires that “[o]ther than the fact of a prior conviction, any fact that increases the penalty for a crime beyond the prescribed statutory maximum [641]*641must be submitted to a jury, and proved beyond a reasonable doubt.” Apprendi, supra, 530 U.S. at 490, 120 S.Ct. at 2362-63, 147 L.Ed.2d at 455. The Court concluded that “[i]t is unconstitutional for a legislature to remove from the jury the assessment of facts that increase the prescribed range of penalties to which a criminal defendant is exposed.” Id. at 490, 120 S.Ct. at 2363, 147 L.Ed. 2d at 455 (quoting Jones v. United States, 526 U.S. 227, 252-53, 119 S.Ct. 1215, 1228, 143 L.Ed.2d 311, 332 (1999) (Stevens, J., concurring)). The Court emphasized that “the relevant inquiry is one not of form, but of effect does the required finding expose the defendant to a greater punishment than that authorized by the jury’s guilty verdict?” Id. at 494, 120 S.Ct. at 2365, 147 L.Ed.2d at 457.
In Ring, supra, the Supreme Court logically extended its Apprendi holding to strike down Arizona’s capital sentencing scheme that allowed a judge alone to make the factual findings required to sentence a defendant to death. 536 U.S. at 588-89, 122 S.Ct. at 2432, 153 L.Ed.2A at 564. Under Arizona law, the maximum punishment that a defendant could receive based solely on a jury’s finding of guilt of first-degree murder was life imprisonment. Id. at 597, 122 S.Ct. at 2437, 153 L.Ed.2d at 569. That defendant could not be sentenced to death, the statutory maximum penalty for first-degree murder, unless at a hearing the judge found beyond a reasonable doubt at least one aggravating circumstance and then found no mitigating circumstances sufficiently substantial to call for leniency. Id. at 592-93, 122 S.Ct. at 2434-35, 153 L.Ed.2d at 566.
The Court, in Ring, concluded that the Sixth Amendment’s jury trial guarantee made applicable to the States entrusted to the jury the ultimate finding of the aggravating factor that exposed the defendant to the penalty of death. The Court reasoned that “[cjapital defendants, no less than noncapital defendants, ... are entitled to a jury determination of any fact on which the legislature conditions an increase in their maximum punishment.” Id. at 589, 122 S.Ct. at 2432, 153 L.Ed.2d at 564. The Court rejected [642]*642Arizona’s argument that, because the statute at issue specified “death or life imprisonment” as the only sentencing options, Ring’s death sentence was within the range of punishment authorized by the jury verdict. The Court noted:
[Arizona's] argument overlooks Apprmdi’s instruction that “the relevant inquiry is one not of form, but of effect.” In effect, “the required finding [of an aggravated circumstance] expose[d] [Ring] to a greater punishment than that authorized by the jury’s guilty verdict.” The Arizona first-degree murder statute “authorizes a maximum penalty of death only in a formal sense” for it explicitly cross-references the statutory provision requiring the finding of an aggravating circumstance before imposition of the death penalty. If Arizona prevailed on its opening argument, Apprendi would be reduced to a “meaningless and formalistic” rule of statutory drafting.
[Id. at 604, 122 S.Ct. at 2440-41, 153 L.Ed.2d at 573-74 (citations omitted).]
Where “enumerated aggravating factors operate as ‘the functional equivalent of an element of a greater offense,’ the Sixth Amendment requires that they be found by a jury.” Id. at 609, 122 S.Ct. at 2443, 153 L.Ed.2d at 577 (citations omitted).
Ring thus fundamentally altered the constitutional significance of aggravating factors that expose a defendant to the death penalty. In the context of deciding when the Fifth Amendment’s double jeopardy protection applied to a penalty-phase hearing in a capital case, the Supreme Court in Sattazahn v. Pennsylvania reasserted the principles established in Ring:
[Ring] held that aggravating circumstances that make a defendant eligible for the death penalty “operate as ‘the functional equivalent of an element of a greater offense.’” That is to say, for purposes of the Sixth Amendment’s jury-trial guarantee, the underlying offense of “murder” is a distinct, lesser included offense of “murder plus one or more aggravating circumstances”: Whereas the former exposes a defendant to a maximum penalty of life imprisonment, the latter increases the maximum permissible sentence to death. Accordingly, we held that the Sixth Amendment requires that a jury, and not a judge, find the existence of any aggravating circumstances, and that they be found, not by a mere preponderance of the evidence, but beyond a reasonable doubt.
[ 537 U.S. 101, 111, 537 U.S. 101, 123 S.Ct. 732, 739, 154 L.Ed.2d 588, 598-99 (2003) (plurality opinion) (internal citations omitted).]
See also Sattazahn, supra, 537 U.S. at 126 n. 6,123 S.Ct. at 747 n. 6, 154 L.Ed.2d at 608 n. 6 (Ginsburg, J., dissenting) (noting that “capital sentencing proceedings involving proof of one or more [643]*643aggravating factors are to be treated as trials of separate offenses, not mere sentencing proceedings”).
Unlike Arizona, New Jersey always has required that capital “triggers” and aggravating factors go to the jury and that the facts in issue be proven beyond a reasonable doubt. On more than one occasion, this Court has acknowledged that, in important ways, aggravating factors are functionally indistinguishable from the elements of a crime. State v. Biegenwald, 106 N.J. 13, 59, 524 A2d 130, 154 (1987) (Biegenwald II) (noting “the functional similarity of aggravating factors and the weighing process itself to the traditional proof of ‘elements of an offense’ ”). As we observed in State v. Ramseur:
It is cleai- to us, however, that functionally, the aggravating factors in the Act are indistinguishable, for this purpose, from the elements of a crime. For example, no more or less than premeditation under our prior law, proof of an aggravating factor could mark the difference between imprisonment and death.
[ 106 N.J. 123, 201 n. 27, 524 A.2d 188, 226 n. 27 (1987).]
Before Apprendi and Ring, New Jersey treated aggravating factors as though they were elements of the offense by submitting them to a petit jury and subjecting them to proof beyond a reasonable doubt. Following Ring, today, those procedures are a constitutional imperative. In a constitutional sense, the Arizona capital aggravating factors are indistinguishable from those in New Jersey in the manner in which they operate to elevate a term of imprisonment to a death sentence.16 For [644]*644purposes of the Sixth Amendment’s jury-trial guarantee, aggravating factors “operate as the functional equivalent of an element of the greater offense .... [o]f murder plus one or more aggravating circumstances.” Sattazahn, supra, 537 U.S. at 111, 123 S.Ct. at 739, 154 L.Ed.2d at 599 (internal citations and quotation marks omitted). Thus, murder is a distinct lesser-included offense of the greater offense of capital murder. Ibid. After Ring, federal prosecutors in capital prosecutions under the Federal Death Penalty Act, 18 U.S.C.A. §§ 3591-3598, have submitted to federal grand juries those aggravating factors that make a defendant eligible for the imposition of a sentence of death. They are doing so by obtaining superseding indictments that contain a “Notice of Special Findings” alleging the aggravating factors. See, e.g., United States v. Haynes, 269 F.Supp.2d 970, 973 (W.D.Tenn.2003) (noting that, post-iümgr, government obtained a superseding indictment alleging aggravating factors); United States v. Sampson, 245 F.Supp.2d 327, 329 (D.Mass.2003) (same); United States v. Church, 218 F.Supp.2d 813, 814 (W.D.Va.2002) (same); United States v. Matthews, 246 F.Supp.2d 137, 140 (N.D.N.Y.2002) (same); United States v. Lentz, 225 F.Supp.2d 672, 675 (E.D.Va. 2002) (same); United States v. Regan, 221 F.Supp.2d 672, 677 (E.D.Va.2002) (same). Federal prosecutors apparently have assumed that Ring requires a grand jury to conclude that the government’s decision to seek the death penalty is supported by sufficient evidence. The federal experience shows no sign that submission of aggravating factors to the grand jury has impaired [645]*645law enforcement’s ability to prosecute capital cases. The aggravating factors in the federal capital sentencing scheme serve a substantially similar purpose to the aggravating factors under our Capital Penalty Act.17
Although we recognize that the Fifth Amendment right to indictment by a grand jury does not apply to the States, Hurtado v. California, 110 U.S. 516, 538, 4 S.Ct. 111, 122, 28 L.Ed. 232, 239 (1884), we have never construed our grand jury provision under Article I, Paragraph 8 as providing lesser protection than its federal analogue. Ramseur, supra, 106 N.J. at 215 n. 42, 524 A.2d at 233 n. 42 (1987); State v. Porro, 152 N.J.Super. 259, 265, 377 A.2d 950, 953 (Law Div.1977), aff'd, 158 N.J.Super. 269, 385 A.2d 1258 (App.Div.), cert. denied, 439 U.S. 1047, 99 S.Ct. 724, 58 L.Ed.2d 706 (1978). If aggravating factors and capital triggers are the functional equivalent of elements of capital murder pursuant to the Sixth Amendment’s right to trial by jury, we see no [646]*646reason to define them as something other than elements for purposes of the state constitutional right to a grand jury presentation. We, therefore, hold that our State Constitution requires that aggravating factors be submitted to the grand jury and returned in an indictment. In a criminal justice system in which all of the elements of a crime must be submitted to the grand jury it would be odd to make capital murder the one exception. Our analysis of the development of federal law does not compel this result; logic and fairness and the historical importance we attach to our grand jury system do.
In light of Apprendi and Ring and our recognition that “functionally, the aggravating factors in [N.J.SA. 2C:ll-3c] are indistinguishable ... from the elements of a crime,” Ramseur, supra, 106 N.J. at 201 n. 27, 524 A.2d at 226 n. 27, we conclude that Article I, Paragraph 8 requires the submission of the aggravating factors and capital “triggers” to the grand jury. Courts do not lightly overrule their own precedents. The United States Supreme Court, by declaring unconstitutional one part of Arizona’s capital penalty law in Ring v. Arizona, overruled its decision in Walton v. Arizona 497 U.S. 639, 110 S.Ct. 3047, 111 L.Ed.2d 511 (1990), “to the extent that it allow[ed] a sentencing judge, sitting without a jury, to find an aggravating circumstance necessary for imposition of the death penalty.” 536 U.S. at 609, 122 S.Ct. at 2443, 153 L.Ed.2d at 576-77. The Court took note of its break with precedent: “Although the doctrine of stare decisis is of fundamental importance to the rule of law[,] ... [o]ur precedents are not sacrosanct. [W]e have overruled prior decisions where the necessity and propriety of doing so has been established. We are satisfied that this is such a case.” Id. at 608, 122 S.Ct. at 2442-43, 153 L.Ed.2d at 576 (internal quotation marks and citations omitted). In light of Ring, federal constitutional law now clearly defines the elements of capital murder in a way that is fatally at odds with Martini I. We are left with no alternative but to part ways with that portion of the Martini I decision inconsistent with our holding today.
[647]*647We are mindful that the rule we announce overrules Martini I’s interpretation of Article I, Paragraph 8 and thereby is “ ‘a new rule of law.’ ” State v. Afanador, 151 N.J. 41, 57, 697 A.2d 529, 536 (1997) (quoting State v. Cupe, 289 N.J.Super. 1, 11, 672 A.2d 1233, 1238 (App.Div.), certif. denied, 144 N.J. 589, 677 A.2d 761 (1996)). We limit this rule to purely prospective application for several reasons.18 In determining whether to apply a new rule retroactively or prospectively, we consider the following three factors: “ ‘(1) the purpose of the rule and whether it would be furthered by a retroactive application, (2) the degree of reliance placed on the old rule by those who administered it, and (3) the effect a retroactive application would have on the administration of justice.’ ”19 Knight, supra, 145 N.J. at 251, 678 A.2d at 651-52 (quoting State v. Nash, 64 N.J. 464, 471, 317 A.2d 689, 692 (1974)).
“The first factor, the purpose of the new rule, is often the pivotal consideration.” State v. Burstein, 85 N.J. 394, 406, 427 A.2d 525, 531 (1981). New rules are given complete retroactive effect “where the purpose of the new rule ‘is to overcome an aspect of the criminal trial that substantially impairs its truth-[648]*648finding function’ and which raises ‘serious questions about the accuracy of guilty verdicts in past trials.’ ” Id. at 406-07, 427 A.2d at 531 (quoting Williams v. United States, 401 U.S. 646, 653, 91 S.Ct. 1148, 1152, 28 L.Ed.2d 388, 395 (1971)). We have given a new rule retroactive effect where it “deals with the ultimate fairness and soundness of the jury’s verdict.” State v. Czachor, 82 N.J. 392, 408, 413 A.2d 593, 601 (1980). On the other hand, in eases where “the new rule is designed to enhance the reliability of the factfinding process but the old rule did not ‘substantially’ impair the accuracy of that process,” we have declined to grant retroactive effect to the new rule if “the countervailing interests of the State reliance on the old rule and the undisrupted administration of justice were held to outweigh the negligible effect that the old rule had on the integrity of the truth-finding process.” Burstein, supra, 85 N.J. at 408-09, 427 A.2d at 532.
The previous rule, enunciated in Martini I, had little impact on the truth-finding process. Before our decision today, we had in place safeguards that served as a substitute for the indictment process. Prosecutors were required to give timely notice of the aggravating factors and defendants were able to contest the validity of any alleged aggravating factor through the process established in McCrary, supra, 97 N.J. at 142, 478 A.2d at 344 — 45. In this case, at a McCrary hearing the trial court rejected a defense motion to dismiss aggravating factors 4(c) (murder involving aggravated assault and/or torture) and 4(f) (murder committed for the purpose of escaping detection), finding an adequate factual basis to support those factors. Furthermore, a defendant who advances to the penalty phase is not sentenced to death unless the jury finds an aggravating factor beyond a reasonable doubt and that all of the aggravating factors outweigh beyond a reasonable doubt all of the mitigating factors. N.J.S.A. 2C:11-3c(3)(a). Together, those procedures ensured the accuracy of the truth-finding process endorsed in Martini I, and we find no evidence that the failure to submit the aggravating factors to the grand jury in this case impaired the accuracy of that process. Given our resolution of this issue, we need not reach the question whether the finding [649]*649of the sentencing jury rendered moot the failure of the indictment to allege the aggravating factors. See United States v. Mechanik, 475 U.S. 66, 70, 106 S.Ct. 938, 942, 89 L.Ed.2d 50, 56 (1986); State v. Murphy, 110 N.J. 20, 28, 538 A.2d 1235, 1238-39 (1988) (limiting Mechanik to violations of the “ ‘technical variety’ ” that raised no issue of “ ‘fundamental fairness’ ”) (quoting United States v. Taylor, 798 F.2d 1337, 1340 (10th Cir.1986)). In light of the reliance by the State on Martini I and the opportunity for defendants to avail themselves of a McCrary hearing, we apply this new rule prospectively.
In all future capital cases, the State is required to submit the aggravating factors to the grand jury and specify in the indictment those factors it intends to prove at a penalty hearing.20 In any case in which there are separate juries empanelled in the guilt phase and penalty phase of a capital case, the trial court must not read the aggravating factors in the indictment to the guilt-phase jury except when one or more of the aggravating factors are also elements of the crimes charged. In such a case, the trial court, as always, will read all of the elements of the [650]*650crimes charged in the indictment. The purpose is to prevent irrelevant and prejudicial information from tainting the guilt-phase trial. This simple rule follows from our precedent in cases in which counts in an indictment are bifurcated into separate trials. See, e.g., State v. Ragland, 105 N.J. 189, 193, 519 A.2d 1361, 1363 (1986). In any case in which the same jury is empanelled in the guilt and penalty phases of a capital case, the trial court — without reference to the indictment — must advise the prospective jurors of the alleged aggravating factors during jury selection to permit voir dire on those factors.
Our holding today applies only to those cases that have yet to reach the penalty-phase. In those capital cases yet to be tried, the State must present the aggravating factors to the grand jury, which then may return a supplemental indictment specifying those factors the defendant will face at a penalty trial. By this opinion, we in no way intend to limit the flexibility of the Attorney General or prosecutors as they go about the difficult task of deciding which murder cases merit a capital prosecution. Prosecutors have the discretion to present a criminal homicide case to the grand jury at one hearing while continuing to review the circumstances that bear on classifying the case as capital murder. In such cases in which there is a later decision to prosecute the ease as capital murder, the State may submit the aggravating factors to the grand jury and seek a supplemental indictment. We leave these matters to the judgment of the executive branch. Nevertheless, we expect that any such decision will be made at an early stage after indictment so as not to cause prejudicial delay.
XII.
Conclusion
In light of our decision to reverse, we find no need to address the other issues raised by defendant. We remand for proceedings consistent with this opinion.
Related
Cite This Page — Counsel Stack
843 A.2d 974, 178 N.J. 540, Counsel Stack Legal Research, https://law.counselstack.com/opinion/state-v-fortin-nj-2004.