State v. Koedatich , 112 N.J. 225 ( 1988 )

112 N.J. 225 (1988) 548 A.2d 939

STATE OF NEW JERSEY, PLAINTIFF-RESPONDENT,
v.
JAMES JEROLD KOEDATICH, DEFENDANT-APPELLANT.

The Supreme Court of New Jersey.

Argued September 28, 1987. Decided August 3, 1988.

David A. Ruhnke and Jean D. Barrett, Designated Counsel, argued the cause for appellant (Alfred A. Slocum, Public Defender, attorney).

Joseph Connor, Jr., Assistant Prosecutor, argued the cause for respondent (Lee S. Trumbull, Morris County Prosecutor, attorney).

Catherine A. Foddai, Deputy Attorney General, argued the cause for amicus curiae Attorney General of New Jersey (W. Cary Edwards, Attorney General, attorney).

Eric Neisser argued the cause for amicus curiae American Civil Liberties Union of New Jersey.

The opinion of the Court was delivered by GARIBALDI, J.

In October 1984, a Morris County jury convicted the defendant, James J. Koedatich, for the murder of Amie Hoffman and sentenced him to death. He appeals directly to this Court as of right. See R. 2:2-1(a)(3). We affirm defendant's murder conviction. Because the trial court failed to instruct the jury properly in the sentencing phase, however, we must reverse the sentence of death and remand for a new sentencing procedure.

I.

Facts

A. The Disappearance, Discovery of the Body, Investigation, and Arrest

In November 1982, Amie Hoffman was an eighteen-year-old senior and a cheerleader at Parsippany Hills High School. She was employed part-time at the Surprise Store in the Morris County Mall. She was last seen alive at aproximately 9:30 p.m. on November 23, 1982, shortly after she left work as she walked toward her car in the parking lot of the Morris County Mall. Two days later on Thanksgiving Day, November 25, 1982, the police discovered her body floating face-down in a water retention tank located in a wooded and secluded area of Randolph Township. Amie Hoffman had been stabbed several times, receiving a wound to the chest, which caused her to bleed to death. Medical evidence submitted at trial indicated that Amie had been the victim of a sexual assault. Other medical evidence established that she died approximately three to four hours after she left the mall.

Testimony at trial established that Amie's life for the few days prior to her death was relatively uneventful. On Sunday, November 21, 1982, Amie awoke at 10:30 a.m., and at 12:30 p.m. her childhood friend and fellow cheerleader, Karyn Speak, arrived and drove Amie to Timothy Day's home, where with a group of friends she watched a football game on television. At the conclusion of the football game, Day drove his friends to the Rainbow Lakes Firehouse in Intervale, where a spaghetti dinner took place in honor of the school's cheerleaders and football players. At the firehouse, there occurred an accident that forced Karyn Speak to seek emergency medical treatment at the hospital. Amie accompanied Mrs. Speak, Karyn's mother, to the hospital, and remained there for about ninety minutes while Karyn was treated. Amie then returned to Karyn's house and remained until about 11:00 p.m., when a high school classmate who had been at the Firehouse dinner drove her home.

On Monday, November 22, 1982, Amie woke early but did not go to school because of menstrual cramps. Amie and her mother spent most of the day talking and that evening they visited the Weavers, close family friends.

The next day, November 23, 1982, Amie still felt sick, so she did not go to school at the usual time. Instead, she slept late, and after chatting with her mother, Amie went to school and had lunch with her friend Karyn Speak, who had suffered only minor injury at the Firehouse dinner. Amie then accompanied Karyn Speak to a dentist's appointment in Denville. The girls returned to school at 2:30 p.m. because cheerleading practice was scheduled for that afternoon. Practice ended about 4:45 p.m. Karen Ersek, whose locker was next to Amie's, noticed that Amie was wearing purple underwear, a brownish-purple sweater, a plaid skirt, and cowboy boots. Amie dressed quickly and left at about 5:05 p.m. for her part-time job at the Surprise Store for women (there was also a Surprise Store for men) at the Morris County Mall.

According to Christene Gonzalez, who worked at the companion Surprise Store for men, Amie arrived at work shortly after 5:00 p.m. Amie took a dinner break "[a]fter 8:00, not later than 8:30" p.m.; she went to McDonald's, ordering a Coke, french fries, and Chicken McNuggets. Barbara Horwath, an employee of the neighboring Kodak Jewelers, who regularly used the Surprise Store's restroom, saw Amie eating the fries at 8:15 p.m.

Christene saw Amie again at 9:20 p.m. when Christene helped Amie close the women's store and Amie returned the favor at the men's store. Although Christene needed a ride home that evening, Amie did not offer her a ride because her mother was going to give her a permanent that night. The girls left the mall at about 9:35 p.m. with Amie carrying a knapsack bookbag. At the mall exit, Christene said goodnight to Amie and turned right while Amie turned straight back through the emptying parking lot to her car.

Barbara Horwath left the Kodak Jewelers a few minutes earlier than Amie had left the Surprise Store. Like Amie, she walked out of the main exit straight to the back lot; she walked with three other employees, gradually splitting up until she remained with one colleague, Debby McLain. They stopped to chat for five minutes. While they were talking, Barbara noticed in front of them a greenish blue vehicle, with a vinyl roof and what she described as putty marks on the driver's side. The driver's side window was down four inches, and she could see the driver in profile. His eyes were "dark," his hair curly and "light, light colored blond, curly hair," and his nose was "pointy." At one point, the driver turned toward them. Barbara could see dark markings on the sides of his nose and under his eyes. His nose was "prominent," his hair was shoulder-length, and she saw "gold around the collar."

Barbara said goodnight to Debby McLain, then walked past the car she had observed; she saw six rear lights, three on each side, and identified the car as a Chevy similar to her sister's Chevy BelAir. She got into her car and drove down toward the Mall. As she left, she saw Amie walking "up the parking lot" toward her car.

Amie's mother expected her home at about 9:45 p.m. for her permanent. When Amie had not arrived nor called, as she generally did whenever she was going to be late, Mrs. Hoffman drove to the Mall to search for her. She arrived at the Mall at 10:30 p.m. and found Amie's car sitting alone in the parking lot. The keys were in the ignition and Amie's jacket and pocketbook, with her wallet in it, were in the car. She called the police. Patrolman William Plate, a Hanover Township canine handler, responded to the Mall with his dog, Barron, to investigate a report of a missing person. Once at the mall, Plate met Mrs. Hoffman at Amie's car and attempted to get a track with Barron. After scenting, Barron tracked about two parking spaces from Amie's car but then stopped. Plate interpreted Barron's behavior as indicating that the scent ended there.

On Thanksgiving Day, November 25, 1982, the police found Amie's body in the center holding tank in the area known as "Old Mendham Water Works," located in Randolph Township. The holding tanks are made of cement and are in a very isolated area surrounded on three sides by woods. Combs Hollow Road is located about 100 yards to the west of the holding tanks with a dirt road actually leading to the tanks. A bridge separates Combs Hollow Road and the dirt road.

When discovered in the center holding tank, Amie was wearing the same panties, sweater, skirt, and cowboy boots she had worn two days earlier when she was last seen. Cut hair was found around the body as well as on the ground outside the tank. There were blood stains on the sides of the tank. Amie's ring was found on the ground near the center tank and her wristwatch was found in the holding tank. A kidney-shaped pool of blood eighteen inches long and nine inches wide was found on the sandy ground near the tank. A trail of blood led from the kidney-shaped pool to the wall of the holding tank to the right hand corner of the center tank. The police removed Amie's body from the water.

Dr. Frederick L. Roddy, First Assistant Medical Examiner of Morris County, performed the autopsy. Dr. Roddy found a long open gash on the left side of the victim's head, an L-shaped wound to the victim's right shoulder, and "short injuries" at the base of the victim's neck. The victim's left ear had been severed, leaving a deep wound that, in Dr. Roddy's view, would not have caused death but would have prevented the victim from holding her head straight; this wound extended through all of the victim's soft tissue to the spinal column. There was a short laceration at the base of the victim's nose, and two severe chest wounds, one penetrating four and a half inches, the other penetrating more than seven inches, through the victim's lungs and to her back between the ninth and tenth ribs. Dr. Roddy concluded from the structure of these wounds that they were caused by a single-edged knife held perpendicular to the victim's chest; he hypothesized that the knife was inserted, causing the four and a half inch wound, then partially withdrawn, then thrust in deeply, causing the seven inch wound. The victim had defensive wounds on her right hand, as though she had attempted to grasp the blade of the knife. There were also abrasions and bruises on the victim's left thigh and lower left arm, consistent with her having been dragged over the retention tank wall.

Dr. Roddy's internal examination revealed that the victim "had apparently eaten a short time prior to her death. In the stomach we found fragments of white meat from chicken. There was a portion of what appeared to be french fries and also a piece of what appeared to be a roll." Based on this evidence, Dr. Roddy estimated that the "time of death was probably four to five hours after she had eaten her last meal." In his internal examination of Amie Hoffman, Doctor Roddy also took swabbings of her oral cavity, vagina, and rectum, and made twelve slides from these swabs. He kept and examined six slides and sent the rest to the Medical Examiner's office in Newark. Sperm was found in the vaginal slides, which indicated that Amie had sexual intercourse before her death. Dr. Roddy found no presence of sperm on the rectal slides. Dr. Robert Goode, the State Medical Examiner, examined the six slides that Dr. Roddy sent to him. He concurred with Dr. Roddy's finding of sperm in the vagina and estimated that intercourse occurred within twenty-four hours of death. His examination revealed, however, that sperm was present in the rectal slides as well. On the basis of the autopsy, Dr. Roddy concluded that the victim "bled to death and the cause of the bleeding was the stab wound in the right chest" (i.e., the seven-inch stab wound).

As the investigation into the homicide began, the police talked to various people who had been in the vicinity of the Mall or the reservoir on the night of November 23, 1982. From Ms. Horwath's observations, the police drew composites of the car and driver she had seen in the parking lot. On December 5, 1982, because of a similarity of the vehicle and facial appearance to the Horwath composite, a Patrol Officer of the Morris County Park Police stopped the defendant, James Koedatich. The officer's notes reflect that Mr. Koedatich was 5'10 1/2" to 6' tall, with a beard and mustache. The composite sketch was "of a suspect, blond, curly hair, possible facial hair." For reasons undisclosed in the record, the Officer did not detain Mr. Koedatich at that time.

Other information obtained during the early stages of the investigation came from persons who were in the vicinity of the reservoir on the night of the murder. Police interviewed Timothy O'Grady, who had been visiting his girlfriend's mother at her home on Morris Turnpike near the reservoir on the night in question. O'Grady left at about 10:15 p.m. to get home in time to phone his girlfriend, who was away at college. O'Grady turned onto Combs Hollow Road, bearing in the direction of the dirt access road to the reservoir. He noticed a pair of headlights by the access tanks near the reservoir; the lights were heading away from the tanks down the access road. He paused before a narrow bridge on Combs Hollow Road to let the car pass; the car had come down the access road by this time. The other car had four headlights, mounted side-to-side with parking lights beneath, and a grated grill; the car was a medium to light blue. O'Grady recalled that the car's high beams were on. Once he reached home, O'Grady, an automobile aficionado, identified the car in a manual as a 1970 Chevrolet two-door. He told police that there was only one person in the car, and that this single operator was probably a male who might have had a mustache but no beard.

Police also interviewed Mrs. Mary Clark, who lived on Combs Hollow Road and who heard, at about 12:45 p.m. on November 24, 1982, "what sounded like a small sports car go by the house slowly" from the direction of the retention tanks. She did not know whether the car had come from the access road to the retention tanks. Detective Richard Longo, who interviewed Mrs. Clark, recounted that she told him that she got out of bed when she heard the engine but saw no headlights, and that the vehicle went up the road, turned around, and came by her house again; Mrs. Clark later did not remember telling the officer those things. Few other leads were available to the police during the early stages of their investigation.

Amie Hoffman's killing was followed two weeks later by the abduction and murder of twenty-five-year-old Deirdre O'Brien on December 5, 1982.^[1] As of January 15, 1983, no suspect had been arrested or charged with either the Hoffman or O'Brien murders. The defendant was a suspect in neither case.

At 11:20 p.m. on January 16, 1983, Patrolman Kevin Dormer responded to a call from 44 Harrison Street in Morristown. On entering the house, the Patrolman saw James Koedatich leaning against the kitchen table, his T-shirt rolled up; Julia Baldwin, Koedatich's mother, was washing a wound on Koedatich's back. Mr. Koedatich told police that he had been driving home on Route 24 when he was pulled over by a car with a flashing blue light and that the driver of the car had stabbed him. Mr. Koedatich was taken by ambulance — unaccompanied by police — to Morristown Memorial Hospital. Sergeant Perkalis, who also responded to the call, also instructed police officers on the scene to secure Mr. Koedatich's clothing and car as possible evidence leading to the arrest of his alleged assailant.

Due to the similarities between the alleged attack on the defendant and the O'Brien and Hoffman murders, Detective John Kinnecom of the Morris County Sheriff's Division of Criminal Investigation, who was also investigating the O'Brien murder, was called to the police garage on January 17, 1983, to supervise the collection of evidence relating to the Koedatich stabbing. When Kinnecom inspected Mr. Koedatich's car, he realized that its tread pattern was similar to the impression made at the scene of the O'Brien abduction. A comparison of the Koedatich tire tread with a sketch and photograph of the O'Brien track confirmed Kinnecom's suspicion.

Kinnecom notified the Prosecutor's Office and the State Police. A search warrant was obtained for James Koedatich's 1970 Chevrolet. Police removed the tire, and took the car to the State Police Laboratory in Little Falls for extraction of possible fiber and foam evidence. The seat cover was removed and the interior was vacuumed. Laboratory personnel also removed samples of the car's carpet. The seat cover and carpet were taken to the Federal Bureau of Investigation Laboratory in Washington, D.C. for analysis.

B. The Indictment and Pretrial Motions

On January 19, 1983, police arrested James Koedatich for the murder of Deirdre O'Brien. The State did not charge Koedatich with the murder of Amie Hoffman until December 15, 1983. During the interval between Koedatich's arrest for the O'Brien murder and his indictment for the Hoffman murder, the community questioned why an arrest in the Hoffman case was taking so long. The prosecutor's office said it was waiting for forensic evidence.

On December 15, 1983, Morris County Indictment No. 83-12-0725-I charged James Koedatich with purposeful or knowing murder by his own conduct, contrary to N.J.S.A. 2C:11-3a(1)-(2); felony murder, contrary to N.J.S.A. 2C:11-3a(3); kidnapping, contrary to N.J.S.A. 2C:13-1b; aggravated sexual assault, contrary to N.J.S.A. 2C:14-2a; unlawful possession of a weapon, contrary to N.J.S.A. 2C:39-5d; and possession of a weapon for an unlawful purpose, contrary to N.J.S.A. 2C:39-4d.

Judge Arnold denied two motions to suppress, heard in conjunction with identical motions in the O'Brien case in Warren County. A motion for a change of venue was heard and denied by Judge Muir, then Assignment Judge for Morris County, on June 27, 1984. That motion was renewed and again denied on various occasions throughout the trial before Judge Arnold Stein. In addition, pretrial motions attacking the constitutionality of the capital murder statute and requesting attorney-conducted voir dire of prospective jurors were heard and denied on July 2, 1984.

Voir dire of prospective jurors began on September 24, 1984, and concluded on October 3, 1984. With the qualification of fifteen jurors, the trial began on October 9, 1984. The State's case consisted of four basic themes: (1) the facts surrounding the disappearance and the discovery of the body; (2) Amie Hoffman's activities in the days leading up to the abduction (in order to prove that she had not had sex prior to the abduction); (3) the likeness of the defendant and/or his car to descriptions given by various witnesses; and (4) the close correlation between the fiber and foam of the interior of defendant's car and fibers and foam found on the victim's clothes and boots.

The defense planned to advance five arguments: (1) that the defendant did not match the descriptions given by the witnesses, as he was not blonde and had a beard and a mustache; (2) that one State witness was simply mistaken, and had overreacted, perhaps subconsciously, to the publicity surrounding the case and the defendant; (3) that the defendant had an alibi, in that he had been home playing video games on the night in question; (4) that the fiber and foam correlations proved nothing; and (5) that convincing evidence existed that the murder had been committed by another, namely, Kevin Sheehan, an assistant football coach at the high school Amie Hoffman attended.

C. The State's Case

Without objection, the State introduced the videotaped deposition of Mrs. Helen Cato, who testified that on a Monday in mid-November 1982, after visiting a sick friend, she stopped at the Morris County Mall at about 3:00 p.m. to do some shopping. She remembered it was a Monday because she visited that friend on that day each week. The day was sunny. She had parked her car and was smoking a cigarette and listening to the radio when she heard a loud voice. She saw a man and a young woman walking from the Mall toward her vehicle. She recognized the young woman as Amie Hoffman, a cashier at the Surprise Store. The man, she said, was James Koedatich, who was in the room with Mrs. Cato when the deposition was taken. The man had a face "like a hawk," and was wearing blue and white sneakers, blue jeans, and a denim jacket. He looked familiar. Mrs. Cato later placed him as a young man who had been to her house with her teenage daughter in the early 1960s. As the man and woman approached her car, Mrs. Cato heard him say, loudly, "Come here. Come here. Goddammit, I said come here. I'll kill you." The woman was unresponsive at first but became flustered and dropped the books she was carrying. She picked them up, stumbled, and jogged away from the man.

The State recalled Christene Gonzalez, who had earlier testified regarding Amie's activities on the evening of her disappearance. Supra at 233. She testified that she approached the prosecutor after initially testifying because she had never seen pictures of the defendant, and that while testifying she suddenly recognized the defendant as a man she had seen at the mall. She then testified that four times in September and October 1982 she had noticed a man with a very long face, high cheekbones, brownish hair, and one or two days growth of beard standing outside the Station Pub, a tavern in the Mall, and that the man was defendant Koedatich.

The State also called Dr. Isadore Mihalikis, a forensic pathologist from Pennsylvania, who studied stab wounds. Dr. Mihalikis testified, over defense counsel's objection, that the defendant's stab wound was consistent with self-infliction. Dr. Mihalikis based this opinion largely on the discovery of a cluster of three superficial cuts near the left shoulder of the jacket defendant had been wearing when he was allegedly attacked. The cuts on the jacket — none of which penetrated the defendant's skin — varied in depth from 3/16-inch to 1-1/8 inches, and had been caused by a single-edge knife. They were consistent with self-infliction, Dr. Mihalikis testified, because they were physically accessible, grouped closely together, occurred in a nonvital area, varied in depth (indicating some hesitation), and were shallow.

In connection with the fiber and forensic evidence, the State produced Special Agent Harold Deadman from the Microscopic Analysis Unit of the Federal Bureau of Investigation Laboratory and George Neighbor, principal forensic chemist with the New Jersey State Police Crime Laboratory. Special Agent Deadman testified that he specialized in the examination and comparison of hair and fibers; that he had examined and compared material removed from defendant's car with specimens from the clothes and boots of Amie Hoffman. He testified that nylon fibers removed from the heel of Amie Hoffman's right boot were consistent with fibers found in the car's carpet and that polyester fibers found on Amie's skirt matched fibers from the car's seat cover. Also, according to Deadman, nylon fibers found on Amie's skirt matched coarse fibers found in the binding of the seat cover in defendant's car. Although on cross-examination he admitted the fibers were fairly common, he nevertheless insisted that the likelihood was extremely remote that the fibers on Amie's skirt and boot came from any source other than defendant's car.

George Neighbor, whose expertise includes hair and foam comparison, examined the seat cover from defendant's 1970 Chevrolet and brushings taken from Amie's right boot. Neighbor removed a piece of foam backing from the seat cover and compared it to two pieces of foam found on the edge of the boot sole. He testified that when viewed under a microscope, the foam from the seat cover of defendant's car matched foam found on Amie's right boot.

After the testimony of Mr. Neighbor, the State's final scheduled witness, the State introduced a surprise witness, Diana Bossard. Ms. Bossard testified over defense counsel's objections that in mid-October 1982, she had stopped at the Morris County Mall to do some clothes shopping for a planned November 1 vacation. Once in the Mall, she looked for clothes in Bradlees, the Fashion Bug, and the Surprise Store. After she entered the Surprise Store at 8:45 p.m., she walked to the back of the store and saw "a young Oriental girl with long hair being talked to in a loud voice by an unkempt-looking older man." The man, who was tall, with dirty-blonde hair, a thin face with high cheekbones, and several days growth of beard, was speaking in a loud voice about dirt bikes, but the girl was not taking part in the conversation. After several minutes in the store, Bossard left without purchasing anything. Ms. Bossard identified the girl as Amie Hoffman and the man as defendant. Infra at 314-318.

Finally, the State attempted to introduce two items seized during the execution of a search warrant at defendant's residence. The first was a photo album found on defendant's nightstand, the last five pages of which consisted of pictures of "Oriental-looking" women. The second item was a letter addressed to Mr. Koedatich from Japan International; enclosed were three more pictures of Oriental women. The State's theory was that this evidence demonstrated defendant's interest in Oriental females and explained to the jury why defendant pursued Amie Hoffman over other women. Amie, who was of Oriental extraction, was adopted by the Hoffman's at age five. The Court excluded the evidence because it found there were no facts to support the State's contention that defendant wanted to have sex with Amie Hoffman or to abduct her because he was sexually attracted to Oriental females.

D. The Defense's Case

The defense opened by producing Michael Borzeka, a cousin of the defendant, and John Borzeka, the defendant's uncle, both of whom testified that they had seen the defendant on Thanksgiving Day in 1982 and that he had had a beard. According to his uncle, defendant's beard was "heavy, full." Both witnesses admitted, however, that they remembered this only because of photographs allegedly taken by Michael, and Michael testified that defendant's beard was "close to his face." The defense next produced Mrs. Clark, who testified, as recounted earlier, that she had heard a small sports car go slowly by her house on Combs Hollow Road from the direction of the retention tanks at 12:45 a.m. on the night in question. Detective Longo followed with his version of what Mrs. Clark had told him. The defense then sought to introduce the testimony of Lt. Raymond DiBiasse of the Morristown Police Department, who had received a phone call on December 14, 1982, from a self-professed psychiatrist, who said that he had a patient, a construction worker, who might have committed the murder, and that his patient had been attracted to an Oriental woman who had continually rebuked him. This evidence, the defense argued, raised the possibility that Helen Cato and Diana Bossard had seen another man — not the defendant — with Amie Hoffman. The court ruled, however, that the statement was "classic" hearsay, and that its unreliability was patent.

The defense then called David Paul Baldwin, the defendant's step-brother, who was in high school and was living at home in November 1982. Baldwin testified that defendant's car had not been running well in November. He also testified that his step-brother had had a beard that fall and that on the night in question defendant was upstairs at home playing Atari most of the evening, at least until nine o'clock. Finally, he testified that the sneakers seized in the search of the house and associated with defendant by Helen Cato were bought for defendant as a Christmas gift in December 1982. Based on the prosecutor's cross-examination of Mr. Baldwin, the defense moved for a mistrial, which was denied. Infra at 321-324.

The defense then sought to call Kevin Sheehan, the former Assistant Football Coach at Amie's high school, in order to suggest that Mr. Sheehan may have been Amie's killer. The court held evidence concerning Mr. Sheehan to be inadmissible. Infra at 298-313.

The defense called Kathleen Koedatich, defendant's sister-in-law, who testified that defendant was bearded in September 1982 and on November 21, 1982, although before the grand jury she had testified that defendant was just starting to grow a beard in November. The defense then called Julia Baldwin, the defendant's mother, who testified that her son had a beard from September to November 1982. She also testified that James Koedatich was upstairs at home playing Atari on the night in question from 8:00 p.m. to 11:00 p.m., and that she had been in the living room until eleven o'clock and would have seen the defendant if he had left. Like defendant's step-brother she stated that the sneakers the police seized were a Christmas present for defendant.

To rebut the State's fiber and foam evidence, the defense called four witnesses. Thomas Gowrie, director of sales for Falk Fibers & Fabric, a Hillsboro, North Carolina, nylon manufacturer, testified that his company sold to Western Textiles approximately 400,000 yards per year of nylon monofilament similar to that found in the defendant's seat cover. Tom Hannah, of Western Textiles, recognized the tape on the defendant's seat cover as similar to a tape that his company sells to Crest Manufacturing Co., a manufacturer of seat covers. He testified that in 1982 his company sold 2 1/2 million yards of the tape to Crest. The tape was also used in the manufacture of caps, sweatbands, and life jackets. Paul McIntyre, director of finance and administration for General Foam Corp., testified that in 1982 his company sold 340,494 linear yards of charcoal gray foam to U.S. Laminating.

The defense closed its case with its principal expert witness, Dr. John Reffner, principal research microscopist for American Cyanamid. Dr. Reffner testified that the polyester from the victim's skirt was similar, but not identical, to the polyester of the defendant's seat cover; the fabric samples could not be said to be identical, Dr. Reffner stated, because there was a "large degree of variability within the data set on the control," and because the fabric was so common. The same testimony held for the nylon found on the skirt and seat cover, for the rayon from the carpeting and the boot, and for the foam. Dr. Reffner concluded that "because of the general availability of the fibers and the variability within the data set itself ... there is no single point of match or comparison that says that these two materials were at one time joined together. In other words, there's no way to say that the fibers and the foam were part of the automobile or the seat cover." This doubt would increase, Dr. Reffner testified, if he knew that the victim worked in a clothing store. On cross-examination, Dr. Reffner acknowledged that he did not know that the rayon fibers had not been used in car carpeting since 1973. Dr. Reffner also acknowledged that he had run comparison tests with foams from hanger covers, baseball caps, and wet suits, none of which matched the foam from the seat cover. On redirect, Dr. Reffner reaffirmed his conclusion that given the common currency of the fibers, "[t]here's [sic] too many variables to allow a direct connection of these fibers from the skirt to that car." There was, in his opinion, "reasonable uncertainty" that the fibers on the victim's clothes came from the defendant's car.

The defense having rested, the State thereupon announced its intention to call, as rebuttal witnesses, two police officers who claimed to have met the defendant at his brother's service station, both of whom would testify that the defendant was clean-shaven during the fall of 1982. Over defense objections, the court allowed the rebuttal testimony. Infra at 317-320.

In conclusion, the defense called three witnesses in surrebuttal: Carol Charron and Robert Monahan, who had worked with the defendant at his brother's station that fall, and Nicholas Borzeka, defendant's uncle, who had work done at that station. Each testified that defendant had been growing a beard in September 1982, had continued growing it through the fall, and had had dark hair.

The defense summation argued that the case was based on "supposition, innuendo, nothing else." Based on the prosecutor's summation, the defense moved for a mistrial which the court denied. Infra at 324-326.

After being charged on the law, the jury retired for deliberations at 11:20 a.m. on October 25, 1984. While deliberating, the jury requested the testimony of Barbara Horwath, Diana Bossard, Christene Gonzalez, Helen Cato, and Timothy O'Grady, the identification witnesses. The jury also requested the testimony of Mrs. Hoffman regarding Amie's whereabouts on Monday, November 22, and Tuesday, November 23, and Karyn Speak's testimony about Amie's whereabouts on Tuesday, November 23. At 3:05 p.m. on October 26, 1984, the jury returned a verdict of guilty on all counts, namely, of knowing and purposeful murder, of killing in the course of committing kidnapping or aggravated sexual assault, of kidnapping, of aggravated sexual assault, and of possession of a dangerous weapon for an unlawful purpose.

E. Penalty Phase Trial

At the sentencing trial, the prosecutor introduced as aggravating factors the fact that the defendant had a prior conviction of murder, Sec. c(4)(a);^[2] that the murder was outrageously and wantonly vile, Sec. c(4)(c); that the murder was committed for the purpose of escaping detection, Sec. c(4)(f), and that the murder was committed while the defendant was engaged in committing, attempting to commit, or flight from committing kidnapping and/or aggravated sexual assault, Sec. c(4)(g).

Defense counsel commenced the penalty phase by attempting to waive the jury. The prosecutor would not consent, however, so in accordance with Sec. c(1), the application for waiver of the jury was denied. Defense counsel then presented a signed statement by defendant in which defendant made clear that he wished that no mitigating factors be presented on his behalf during the penalty phase. He also expressly requested to be executed within sixty days of being sentenced to death, if in fact he were so sentenced. By making such a request, defendant was attempting to waive his right to appeal his conviction. Defense counsel followed his client's instructions. At the sentencing trial, therefore, defense counsel made no opening statement, presented no evidence of mitigating factors, and made no closing statement to the jury. The trial court informed the jury that defendant was entitled to remain silent throughout the proceeding and that the State still was obliged to prove its case.

The trial court instructed the jury that defendant was to be sentenced to death if the jurors were "satisfied beyond a reasonable doubt that the aggravating factor or factors alleged in this case exist and that they are not outweighed by any mitigating factor." The trial court then proceeded to explain to the jury the aggravating factors that it was to consider in its deliberations, as well as the one general mitigating factor provided for by Sec. c(5)(h). The trial court also instructed the jury that it was not obliged to find beyond a reasonable doubt that a mitigating factor existed. The court did, however, instruct the jury that its decisions regarding the existence or nonexistence of aggravating or mitigating factors had to be unanimous. Elsewhere in its instructions, the trial court stated that "the defendant may be sentenced to death only if you are convinced beyond a reasonable doubt that the aggravating factor or factors either outweigh the mitigating factor or that the[y] are equal to the mitigating factors." The trial court further instructed the jury that any mitigating factors had to outweigh aggravating factors in order to permit a sentence less than death.

The jury interrupted its sentencing phase deliberations to seek clarification on whether unanimity was required in order to find a mitigating factor. The trial court instructed the jury that a mitigating factor did not exist unless it was based on the jury's unanimous finding.

The jury found unanimously that two of the proffered aggravating factors existed: the prior murder conviction factor, and the murder in the course of kidnapping/sexual assault factor. The jury was unable to agree unanimously on the more subjective factors, i.e., the vileness factor (eleven found that it did exist; one that it did not) and the murder for purpose of escaping detection factor, and thus did not find them to exist. Because the jury could not unanimously agree on the existence of a mitigating factor, none was found to exist. The court then sentenced the defendant to death.

On November 16, 1984, the Office of the Public Defender filed a Notice of Appeal on defendant's behalf over his objection. The defendant moved before this Court to have the stay of execution, which had been entered upon the filing of the Notice of Appeal, vacated. This Court denied the defendant's motion on December 5, 1984. State v. Koedatich, 98 N.J. 553 (1984).

On January 11, 1985, the trial court imposed a thirty-year prison term with a fifteen-year period of parole ineligibility on the kidnapping count and a concurrent twenty-year term with ten years of parole ineligibility for the aggravated sexual assault, the sentence to run consecutively to the sentence for murder should the death sentence be overturned. The court merged all other counts into the murder, kidnapping, and aggravated sexual assault counts. The court also levied a fine of $1,050 payable to the Violent Crimes Compensation Board. The defendant filed a motion for a limited remand for a hearing on possible taint of the verdict. This Court denied that motion on October 7, 1985.

II.

Constitutionality of Death Penalty Statute

The defendant attacks this state's capital punishment statute, N.J.S.A. 2C:11-3, on both federal and state grounds, claiming it to be violative of the prohibition against cruel and unusual punishment. U.S. Const. amends. VIII, XIV; N.J. Const. of 1947 art. I, para 12. We addressed this issue in State v. Ramseur, 106 N.J. 123, 166-82 (1987), and now reaffirm our holding in Ramseur that our state's capital punishment statute is not violative of either the federal or the state constitution's ban on cruel and unusual punishment.

Next, we consider defendant's claim that the death penalty statute is unconstitutional because it fails to control prosecutorial discretion in seeking the death sentence. Defendant maintains that the Legislature could not possibly draft a death penalty statute that would eliminate potential arbitrariness at the prosecutorial level, and consequently the death penalty is inherently unconstitutional under the federal and New Jersey constitutions. Thus, defendant asserts that the death penalty act is unconstitutional on its fact.^[3] Defendant argues that the statute gives county prosecutors unfettered discretion in seeking the death sentence in cases in which the statute allows the death sentence. Because prosecutors have unfettered discretion in determining whether to initiate capital prosecutions in those cases in which a death verdict is statutorily possible, defendant maintains that the statutory scheme fails rationally to differentiate among those for whom the State seeks the death penalty and those for whom the State seeks lesser penalties. Accordingly, plaintiff argues, the statute does not ensure a "meaningful basis for distinguishing the few cases in which [the death penalty] is imposed from the many cases in which it is not," Furman v. Georgia, 408 U.S. 238, 313, 92 S.Ct. 2726, 2764, 33 L.Ed.2d 346, 392 (1972) (White, J., concurring), thus allowing the death penalty to be "wantonly and ... freakishly imposed." Id. at 310, 92 S.Ct. at 2762, 33 L.Ed.2d at 390 (Stewart, J., concurring).

In State v. Ramseur, this Court addressed a facial challenge to the constitutionality of the death penalty act with respect to jury discretion. There, we recognized that in order to pass constitutional muster under Furman, "a capital punishment statute must achieve two objectives: limit imposition of the penalty to what is assumed to be the small group for which [it is] appropriate, see id. at 310, 92 S.Ct. at 2762, 33 L.Ed.2d at 390 (White, J., concurring), and ensure that the limited class selected for the penalty is chosen with rationality and consistency, see id. at 310, 92 S.Ct. at 2762, 33 L.Ed. 2d at 390 (Stewart, J., concurring)." State v. Ramseur, supra, 106 N.J. at 183. Both objectives, we noted, "are aimed primarily at eliminating the arbitrary nature of capital proceedings in the past and their high risk of discrimination." Ibid.

In Ramseur, this Court also looked beyond the federal constitutional requirements as formulated in Furman and later in Gregg v. Georgia, 428 U.S. 153, 96 S.Ct. 2909, 49 L.Ed.2d 859 (1976), to the New Jersey Constitution, which mandates consistency and reliability in the administration of capital punishment. State v. Ramseur, supra, 106 N.J. at 190. We observed that our state Constitution "provides an additional and, where appropriate, more expansive source of protections against the arbitrary and nonindividualized imposition of the death penalty." Ibid. At the same time, however, while we observed that our Constitution provides additional protection against the arbitrary imposition of the death penalty, we also cautioned that capital defendants are not "entitled to perfection, to totally consistent, accurate and reliable procedures." Id. at 192. We refused to invalidate the death penalty statute by construing our State Constitution to mandate requirements "that, though unspecified, could never be met." Ibid.

Thus, in Ramseur and its companion case of State v. Biegenwald, 106 N.J. 13 (1987), we did not directly address prosecutorial discretion. In Ramseur, defendant argued that the statute failed to channel jury discretion and therefore violated the constitutional requirements of Furman and Gregg. Here, by contrast, defendant argues that the statute fails to check prosecutorial discretion and therefore cannot guarantee that the State will not arbitrarily or capriciously impose the death penalty. While limiting jury sentencing discretion helps to assure that those sentenced to death are a rational subset of those actually charged of capital crimes, it does not at all assure that those convicted are a rational subset of those who could be charged with a death-eligible offense. See Note, "Discretion and the Constitutionality of the New Death Penalty Statutes," 87 Harv.L.Rev. 1690, 1713-14 (1974). As one commentator noted, "confining sentencing discretion cannot guarantee that those sentenced to death are a rationally selected subset of all capital offenders." Id. at 1714.

Nevertheless, the mere existence of several discretionary stages in the progression of a capital case does not render a death penalty scheme unconstitutional. Gregg v. Georgia, supra, 428 U.S. at 199, 96 S.Ct. at 2937, 49 L.Ed.2d at 889. At each stage of a capital case, "an actor in the criminal justice system makes a decision which may remove a defendant from consideration as a candidate for the death penalty." Ibid. To ensure a complete absence of discretion at each stage of decisionmaking would be an impossible task for the Legislature or for this Court. Indeed, to restrict discretion completely at the prosecutorial stage would be unconstitutional. Id. at 199 n. 50, 96 S.Ct. at 2937 n. 50, 49 L.Ed.2d at 889 n. 50. As the Supreme Court wrote in Gregg:

In order to repair the alleged defects pointed to by the petitioner, it would be necessary to require that prosecuting authorities charge a capital offense whenever arguably there had been a capital murder and that they refuse to plea bargain with the defendant. If a jury refused to convict even though the evidence supported the charge, its verdict would have to be reversed and a verdict of guilty entered or a new trial ordered, since the discretionary act of jury nullification would not be permitted. Finally, acts of executive clemency would have to be prohibited. Such a system, of course, would be totally alien to our notions of criminal justice.

Moreover, it would be unconstitutional. [Ibid.]

Gregg, therefore, holds that the federal Constitution does not require limits on prosecutorial discretion beyond the aggravating factors outlined in the statute. That the federal Constitution does not mandate guidelines for prosecutors in administering the death penalty statute is not dispositive, however. As this Court stated in Ramseur, we apply independent state constitutional analysis in our appellate review of death verdicts. State v. Ramseur, supra, 106 N.J. at 182. Accordingly, we examine defendant's claim of arbitrariness at the prosecutorial stage in light of the independent analysis of our own Constitution.

Defendant, unlike the defendant in State v. Smith, 202 N.J. Super. 578 (Law Div. 1985), does not claim that the prosecutor acted arbitrarily in selecting his particular case for capital prosecution. In Smith, defendant brought to the attention of the court, in support of his claim of selective and arbitrary prosecution, fifteen other cases similar to his own in which the prosecutor did not seek a death sentence. State v. Smith, supra, 202 N.J. Super. at 591. In support of defendant's claim here that the administration of the death penalty is inherently arbitrary, he relies on statistical data. Specifically, defendant submitted to this Court a portion of the preliminary report of the Public Defender's study on the New Jersey death penalty statute entitled "The Reimposition of Capital Punishment in New Jersey: Homicide Cases From 1982-1986" by Leigh B. Bienen, Assistant Deputy Public Defender, Neil Alan Weiner, Senior Research Associate at the Sellin Center for Studies in Criminology and Criminal Law at the University of Pennsylvania, and Deborah W. Denno, Professor of Sociology, University of Pennsylvania. Defendant contends that the preliminary results of this study reveal that cases are designated capital cases in a manner that is inconsistent from county to county and in a manner that is arbitrary when cases are compared by county and by identified groups of defendants and victims.

This Court anticipated in Ramseur that it would be asked to consider "concerns about possible misuse of prosecutorial discretion ... including in the review all cases in which a prosecutor had the discretion to seek the death penalty." State v. Ramseur, supra, 106 N.J. at 329; see also State v. McCrary, 97 N.J. 132, 147 (1984) (permitting limited judicial review of prosecutorial discretion to charge capital murder authorized under the Act). On the basis of the preliminary data submitted by the Public Defender, however, we find no reason to conclude at this time that prosecutorial discretion in the administration of the death penalty is being abused or that it has resulted in an unconstitutionally arbitrary scheme of selection of cases for capital prosecution.

The Public Defender's Study attempts a statistical comparison of "all cases of homicide, except vehicular manslaughter, where the homicide occurred after August 6, 1982, the effective date of the imposition of capital punishment in New Jersey." The purpose of the Study, according to its authors, is to examine the exercise of prosecutorial discretion and differences among criminal practice and procedure in all twenty-one counties in the state. The criteria for inclusion in the data base are formal charge for a homicide offense by the prosecutor's office and a final disposition of that charge at the trial court level. The Preliminary Report surveys the case characteristics of 568 cases. The authors of the study identified and assembled case characteristics based on defense counsels' responses given in interviews with researchers. The interviews incorporated 700 study variables from basic data on the defendant, the victim, the circumstances of the offense, case processing decisions and strategies, the trial, and the applicable statutory aggravating and mitigating factors.

Of the 568 cases included in the Interim Report, prosecutors served a notice of aggravating factors in cases involving 103 defendants (18.1 percent of the 568), for 71 of whom (68.9 percent of the 103) the case went to trial as a capital case before a judge or jury. Of the 71 cases that proceeded to a capital trial, 53 (74.6 percent) resulted in a capital conviction for a death eligible^[4] murder. Of the fifty-three cases that proceeded to a penalty phase trial, eighteen resulted in the imposition of the death penalty. Overall, then, of the 568 homicide cases in the Preliminary Report, prosecutors served notice of aggravating factors in 18.1 percent, 12.5 percent resulted in a capital guilt phase trial, 9.3 percent went to a penalty phase, and 3.2 percent resulted in the death sentence. The Public Defender also claims that of the 568 cases surveyed in the Preliminary Report, 325 cases — or 57.2 percent — had at least one statutory aggravating factor present and were thus death-possible. However, out of that number, the prosecutor served notice of a factor in only 103 cases. Thus, according to the Preliminary Report, 39.1 percent of the cases had an aggravating factor present but the prosecutor did not serve notice of any aggravating factors.

The Preliminary Report advances two statistical correlations indicating that the decision to pursue the death penalty in a death-possible case is dependent on (1) the policy of the prosecutor of the county in which the prosecution is to occur, which policy differs from county to county, and (2) the race of victims and defendants.

First, the Public Defender maintains that the Preliminary Report reveals wide discrepancies in the administration of the death penalty among the counties. For example, a case designated by the Preliminary Report as death-possible was likely to be prosecuted as death-eligible in Monmouth County 70 percent of the time, whereas in Hudson County death-possible cases were prosecuted as death-eligible only 19 percent of the time; in Essex and Camden Counties the rates were 23 percent and 27 percent, respectively. The rate for middle-sized counties was 41 percent, and for small-sized counties 34 percent. Conversely, the chances that a plea bargain would be offered in a given case were significantly higher in Hudson and Camden Counties than in Monmouth County. Thus, these statistics suggest that the selection of homicide cases for capital prosecution depends in part on the county in which the prosecution occurred.

The Preliminary Report offers no explanation for the discrepancies in prosecutorial selection of cases for capital prosecution. The critical question in assessing prosecutorial discretion is what standards are applied to move a case from death-possible to death-eligible status. To assist in answering this question, this Court cannot rely solely on county-by-county statistical discrepancies or on findings that are developed exclusively by defense counsels' evaluation of the case included in the data base.

That there are differences among the counties in the likelihood that a prosecutor will pursue — or the coincidence that the prosecutor has more often pursued — a capital prosecution does not, standing alone, demonstrate that the death penalty is being arbitrarily imposed. Surely, there are a myriad of reasons why a prosecutor handles different cases differently, such as the willingness of a defendant to plead guilty, the strength of the State's case, a defendant's cooperation in the State's case against a co-defendant, the relative weight of the statutory aggravating and mitigating factors, the availability and relative credibility and persuasiveness of witnesses, and the resources of the county prosecutor's office, to list only a few. As thus far presented, the Public Defender's comparisons do not persuade us that the county prosecutors are acting arbitrarily or capriciously.

The Public Defender also urges that the race of the victim^[5] affects the decision to prosecute. Cases involving white victims, which comprise 40.9 percent of cases designated death-possible by the Preliminary Report, account for 61.1 percent of all death sentences and 54.9 percent of all cases that advance to capital trial. Where the victim was white the odds that a case would advance from death-possible to death-eligible were 43 percent, whereas the odds where the victim was Hispanic were 15 percent, and where the victim was black were 26 percent. In combining race of victim with race of defendant, the preliminary data reveal that death-possible cases where both the victim and the defendant were Hispanic had a 9 percent chance of being designated death-eligible, while death-possible cases in which both the victim and the defendant were white had a 44 percent chance of being designated death-eligible, and where both the victim and defendant were black a 27 percent chance of being so designated. In cases where the defendant was black and the victim white (44 total), there was a 45 percent chance that the case would be designated death-eligible.

The authors of the Preliminary Report, however, admit that characteristics other than race may influence the progression of cases surveyed in their Report. Furthermore, they acknowledge that these data are tentative and merely suggest that impermissible or arbitrary factors may have infected the administration of the death penalty statute in this state. Moreover, although the Public Defender asks the Court to take judicial notice of this Preliminary Report, it was never part of the record of this case and was never subjected to a hearing or other proceeding at which the State had a formal opportunity to challenge it. Thus, while the submitted preliminary data may be a beginning for understanding how prosecutors are administering the capital punishment act, they are not sufficiently reliable for this Court to draw conclusions. The preliminary data before this Court in this case have been updated and expanded and submitted to this Court in other capital causes. Accordingly, it is more appropriate for the Court to analyze the Study's findings in their more comprehensive form. At this point, we state merely that this Court has considered and will continue to consider this Study.

Despite our reluctance to draw conclusions from the preliminary empirical evidence, we recognize the need for greater guidance for prosecutors as they attempt to perform their constitutional duty of enforcing this statute. Other death penalty jurisdictions have held the validity of the prosecutor's exercise of discretion depends solely on whether a factual basis exists for the charging of aggravating factors. See, e.g., People v. Free, 112 Ill.2d 154, 97 Ill.Dec. 396, 399, 492 N.E.2d 1269, 1272 (1986); Resnover v. State, 460 N.E.2d 922, 929 (Ind), cert. den., 469 U.S. 873, 105 S.Ct. 231, 83 L.Ed.2d 160 (1984); State v. Judge, 100 Wash.2d 706, 675 P.2d 219, 223 (1984). Indeed, in State v. McCrary, supra, 97 N.J. at 139, this Court acknowledged that while prosecutorial decision-making is normally beyond its purview, fundamental fairness required that this Court review the basis for the State's notice of aggravating factors in a capital case to ensure that prosecutors have cause to designate defendants as death-eligible. To be sure, we stated that our goal was "to effect only a minimal intrusion into this area of prosecutorial discretion." Id. at 142. Nevertheless, we believe there is a need to promote uniformity in the administration of the death penalty, which will be an additional safeguard against arbitrariness and an assistance to this Court in its developing proportionality review.

Accordingly, we strongly recommend that the Attorney General, and the various County Prosecutors, in consultation with the Public Defender, adopt guidelines for use throughout the state by prosecutors in determining the selection of capital cases. With the assistance of these various participants in the criminal justice system, the state can begin to develop guidelines that not only will promote uniform prosecutorial standards but also may assist the Court in its eventual proportionality review.

Finally, we reject any specific claim of arbitrariness in defendant's case. Defendant does not demonstrate any arbitrariness in the selection of his case for capital prosecution, and we will not assume that any arbitrariness existed. There is no basis for vacating defendant's sentence on this ground.

III.

Pretrial and Jury Selection Issues

A. Motion to Suppress Fiber and Foam Evidence

On January 16, 1982, approximately two months after Amie Hoffman's murder, defendant called the police to his home to report that he had been the victim of a stabbing. This bizarre event caused the police to seize and later examine defendant's car, which yielded critical evidence in both the Hoffman and O'Brien murders.

Arriving at defendant's home, the police found the defendant with a stab wound in his lower back. Defendant advised the police that while he was driving his car, a vehicle displaying a flashing blue light came up behind him. Believing the driver to be a policeman, Koedatich claimed he stopped, exited his car, and walked to the rear. The driver chastised defendant for driving too slowly. When defendant turned to reenter his car, he felt a blow to his back. In one motion he slid behind the steering wheel of his car. Once inside the car he felt blood, became frightened, drove home, and called the police. The police administered first aid and called an ambulance, which transported Koedatich to the hospital. Sergeant Perkalis drove defendant's mother to the hospital. In the meantime, Sergeant Perkalis issued a crime alert report about the stabbing to adjacent police departments.

Further interviewing of the defendant at the hospital established that the alleged attack had taken place not in Morristown but in Morris Township, and that jurisdiction over the attack belonged to Morris Township. Accordingly, the Morristown police transferred defendant's car to the Morris Township police garage in order to obtain forensic evidence. The freezing conditions in winter made the lifting of fingerprints almost impossible, so that cars were routinely placed in the garage in order to obtain fingerprints.

Because the details of the attack appeared to be similar to the O'Brien and Hoffman attacks, Detective Kinnecom, a member of the Morris County Sheriff's Criminal Investigation Division, who was responsible for the preservation and identification of physical evidence gathered at the O'Brien murder, was called to examine defendant's car. A critical piece of evidence at the scene of the O'Brien slaying was a tire track found near Ms. O'Brien's car. The police had sketched, photographed, and made a plaster cast of the tire track. From the O'Brien murder on December 5, 1982, to the date of the alleged Koedatich stabbing, a five or six week period, Detective Kinnecom had diligently examined thousands of tires to find a tire with a tread pattern similar to the impression made at the scene of the O'Brien abduction. As soon as he looked at defendant's car, he noticed that a tire on the car had a tread pattern similar to the tread pattern found near Ms. O'Brien's car. He did no further investigation, but immediately called the Prosecutor's Office and a search warrant was secured. The interior of defendant's car was searched pursuant to the search warrant.

The defense agrees that the search warrant was properly issued, but alleges that the original seizure of the car was improper and hence the fruits of the seizure, in this case the fiber and foam evidence, were improperly admitted. The defense moved to suppress the foam and fiber evidence. At the suppression hearing the defense's theory was that the defendant was really a suspect, and that the police had "secured" the car through subterfuge.

Sergeant Perkalis, the officer in charge, testified at the suppression hearing that he requested that defendant's clothes and car be taken by the police to be examined for possible evidence of the assailant's identity. He recounted the circumstances of the seizure:

Q: Once the details of the incident were related to you, did you indicate to Mr. Koedatich your desire to do anything?

A: I indicated to Mr. Koedatich we were going to secure his vehicle as possible evidence.

Q: And what was the response to that?

A: His response was, "yes, okay," he agreed.

Q: Now, the security of the evidence, was that standard police procedure?

A: Yes, sir....

Q: You were investigating this as what — this whole incident?

A: As an aggravated assault with a weapon.

Q: Upon whom?

A: Upon Mr. Koedatich.

Q: Was he a suspect at the point?

A: No, he was not.

Specifically, with respect to the seizure of the car, he testified:

Q: Did you direct Patrolmen Scott or Dormer to secure any other evidence?

A: Yes, I also indicated or I advised Mr. Koedatich at that point in time that we were going to also secure his vehicle as possible evidence also and he was very cooperative. There was [sic] no problems with that at all.

Q: Before we get to that, why did you seek to secure the automobile?

A: For possible contact between the suspect and the vehicle; that there might be some forensic evidence left on the vehicle.

Q: Such as?

A: Such as fingerprints.

Q: Why did you think that the vehicle might provide some additional forensic or trace evidence?

A: Because of the close contact between the suspect and where the assault took place.

The court found that the defendant had not been a suspect at the time of the stabbing, that the police had treated defendant like a victim of an assault, and that Detective Kinnecom reported to the garage because another officer suggested that defendant's assailant could be connected to the O'Brien and Hoffman murders. Moreover, the court noted that the defendant had initiated contact with the police, and that in the course of that investigation the police processed the car under standard procedure. Hence, the court upheld the seizure of the defendant's car as a valid consent seizure, finding "beyond a reasonable doubt that Sergeant Perkalis never ordered the defendant to turn over his car or anything of that nature."

On appeal, the Public Defender accepts the fact that the defendant had not been a suspect prior to his alleged stabbing, but argues instead that since defendant was not advised that he did not have to give his consent to the seizure of his car, his consent was not voluntarily given within the meaning of State v. Johnson, 68 N.J. 349 (1975). We disagree. In New Jersey, if the State seeks to rely in a non-custodial situation on consent as the basis for a search, it must "demonstrat[e] knowledge on the part of the person involved that he had a choice in the matter." Id. at 354 (footnote omitted). Subject to that qualification, a search conducted after a voluntary consent is clearly valid. See State v. Sugar, 108 N.J. 151, 166 (1987) (O'Hern & Stein, JJ., concurring); State v. King, 44 N.J. 346 (1965).

A consent sufficient to avoid the necessity of a warrant may be express or implied from the circumstances. "An implied consent to search is as efficacious and effective as an express consent to search." People v. Engel, 105 Cal. App.3d 489, 504, 164 Cal. Rptr. 454, 463 (1980). Consent may be "implied," because it is found to exist merely because of the person's conduct in engaging in a certain activity. W. LaFave, Search & Seizure: A Treatise on the Fourth Amendment ¶ 8.2(1), at 219 (2d ed. 1987). Moreover, courts in other states have found implied voluntary consent where, as in the instant case, the defendant has initiated police contact and has adopted a "cooperative posture in the mistaken belief that he could thereby divert or prevent police suspicion of him." Id. ¶ 8.2(g), at 204; see also Steigler v. State, 277 A.2d 662, 667 (Del. 1971) (actions of fully cooperative defendant amounted to implied consent to search and seizure: "One can hardly expect the police to get a search warrant for a house or building when the owner is obviously cooperative and gives every appearance of being the victim, rather than the perpetrator, of a crime"); State v. Fredette, 411 A.2d 65, 70 (Me. 1979) (defendant initiated police presence through urgent calls to police; invited them to enter home; and continually cooperated with police as they searched home); Lewis v. State, 285 Md. 705, 717-21, 404 A.2d 1073, 1080-1081 (Md. 1979) (defendant summoned police to house; was anxious to cooperate with investigation and willingly left house key with neighbor to give police access to premises); Commonwealth v. Harris, 387 Mass. 758, 443 N.E.2d 1287 (1982) (defendant genuinely consented in hope that cooperative attitude would deflect police suspicion); Kelly v. State, 75 Wis.2d 303, 313, 249 N.W.2d 800, 805 (Wis. 1977) (defendant called police and, under circumstances that implied that the victim had shot himself or had been shot by someone other than defendant, "there was an implied consent not only to aid the victim but to determine what had caused the death or injury and who was responsible"). Federal courts have reached similar results. See United States v. Price, 599 F.2d 494 (2d Cir.1979) (valid search where defendant told police he did not care if they searched bag because it was not his and he had picked it up by mistake); Thompson v. McManus, 512 F.2d 769 (8th Cir.) (cooperative defendant assisted police by discussing robbery as motive for brutal assault on wife: sufficient to imply consent to second search of house), cert. den., 421 U.S. 1014, 95 S.Ct. 2421, 44 L.Ed.2d 683 (1975).

In their concurring opinion in State v. Sugar, supra, 108 N.J. at 174, Justices O'Hern and Stein found, based on a close examination of all the unique circumstances of that case, that Dr. Sugar had impliedly consented to the second search of his property. In reaching this conclusion, they relied "particularly on the defendant's unabated and unqualified cooperation with police efforts to find his wife." Ibid. In Sugar, the search was not of defendant's home but rather of the curtilage surrounding his home. In this case the item searched was defendant's car, in which a person has a much lower expectation of privacy. See California v. Carney, 471 U.S. 386, 391, 105 S.Ct. 2066, 85 L.Ed.2d 406, 413 (1985) ("``the expectation of privacy with respect to one's automobile is significantly less than that relating to one's home or office'").

In New Jersey, the Appellate Division has distinguished certain cases from Johnson not only because there was a general absence of "coercion," but also because there was some form of active initiation by the party who turned over the evidence. While in Johnson the police had requested entry into the apartment, in State v. McGivern, 167 N.J. Super. 86, 87 (App.Div. 1979), the police query was limited to whether there was anything in the car's trunk; the driver opened his trunk without having been asked. Similarly, in State v. Humanik, 199 N.J. Super. 283, 304-05 (App.Div. 1985), the sister's turning over of the letter to the police "was not directly instigated by the officers"; and in State v. Anglada, 144 N.J. Super. 358, 362-63 (App.Div. 1976), the defendant "knew that [he] did not have to consent" when he allowed undercover officers into his home.

Consent is therefore a factual question to be determined from the relevant circumstances. We are persuaded by the totality of the unique circumstances that the defendant did in fact consent to the seizure of his vehicle, and therefore find the seizure valid. Defendant requested police intervention. There was no subterfuge by the police. The police acted in a reasonably objective and routine manner in examining defendant's car. He gave the appearance of cooperating fully with the police, and he led them to believe they were all working together to catch his assailant. The defendant thus orchestrated the events that led to the discovery of crucial evidence in both this case and the O'Brien case. It is not the fault of the police nor is the Constitution "at all offended when a guilty man stubs his toe." State v. McKnight, 52 N.J. 35, 52 (1968). Indeed, there can hardly be a more appropriate case in which to find that the police acted reasonably in concluding that defendant had given his consent to the seizure of his car. To rule otherwise defies common sense. Accordingly, we find the foam and fiber evidence to be admissible.

B. Publicity and Venue

Defendant asserts that pervasive pretrial publicity deprived him of his federal and state constitutional rights to a fair trial by an impartial jury. Defendant raises four publicity-related issues on appeal:

(1) Did the trial court err in denying defendant's motion for change of venue;

(2) Did the trial court err in not excusing for cause all jurors who had read or heard about the case;

(3) Was the verdict tainted because of certain jurors' knowledge; and

(4) Did the trial court err in denying defendant's motion for attorney-conducted voir dire.

All the issues are interrelated and depend essentially on the same facts and the same principles of law.

It is undisputed that the murders of Amie Hoffman and Deirdre O'Brien and the subsequent arrest of James Koedatich were attended by intense publicity. In State v. Williams, 93 N.J. 39 (1983), decided before the trial in this case, but which involved defendant as an appellant in the O'Brien murder case, this Court noted the extensive publicity surrounding these cases:

The publicity included in the record in the Koedatich case was generated primarily in early December following Ms. O'Brien's stabbing death and again in January following Koedatich's arrest. During these periods, almost-daily reports were published by several newspapers including the Morristown Daily Record, the Easton (Pennsylvania) Express and the Star-Ledger (Morris, Sussex, Warren edition). Other newspapers, including the Dover Daily Advance, the Bridgewater Courier-News, the Passaic Herald News and the New York Daily News, provided less frequent coverage. The December articles chronicled in detail the circumstances surrounding the killing, the statewide manhunt and extensive investigation that ensued and the resulting fear and anxiety felt by many area residents. These articles postulated a possible relation between the O'Brien murder and several other unsolved murders in the area. Koedatich's arrest in January was announced by banners, front page headlines. Published articles extensively reported Koedatich's background and personal history including his prior criminal involvement. In particular, the papers noted that Koedatich had just completed an 11-year prison term in Florida where he was serving a 20-year sentence for murder and armed robbery. Moreover, reports indicated that while in jail Koedatich had choked to death another inmate, although the incident was considered justifiable homicide. Coverage of subsequent legal developments in Koedatich's case reiterated the facts surrounding Ms. O'Brien's stabbing death and Koedatich's criminal history. [Id. at 50-51 n. 2.]

This publicity continued. The Public Defender's Office contacted the Star-Ledger, the Daily Advance, and the Daily Record in May 1984, and found that the Star-Ledger had by that time published more than ninety articles concerning the defendant or the "Morris County Murders," more than half of the articles published after defendant's arrest (forty-four before the arrest, forty-eight after); the Daily Advance had published about forty articles about the cases after Koedatich's arrest. The Daily Record offered no estimate. Circulation of each paper was as follows: the Star-Ledger, 45,500 daily, 62,400 Sunday in Morris County;^[6] 3,800 daily, 5,800 Sunday in Warren County, 8,100 daily, 12,200 Sunday in Sussex County; the Daily Advance, 14,359 daily, 13,020 Sunday; the Daily Record, 65,000 daily, 72,000 Sunday. The Public Defender also contacted various local radio stations and WABC-TV in New York, with inconclusive results.

1. Motion for Change of Venue

Based on the continuing publicity and on a survey purporting to show media saturation in Morris County, defense counsel brought a motion for a change of venue prior to trial. On June 27, 1984, Judge Muir, then Assignment Judge for Morris County, rejected the results of the survey because of methodological flaws; defendant does not contest this finding. The Assignment Judge ruled that there was no reason at that time to order a change of venue, and referred the issue to the discretion of the trial court. In ruling that a change of venue was not warranted at that time, the Assignment Judge relied on the standard articulated in State v. Wise, 19 N.J. 59, 73-74 (1955), namely, whether there was "clear and convincing proof that a fair and impartial trial cannot be had" in a given venue. In State v. Williams, 93 N.J. 39, 63 (1983), we articulated a new standard to be used by trial courts when considering a request for change of venue. Under the Williams standard, courts are to determine whether there exists a "realistic likelihood of prejudice" resulting from pretrial publicity. While the Assignment Judge admittedly applied the incorrect standard in passing on the change of venue request, the trial court before whom defendant was tried did not. On each of the numerous occasions on which defendant moved for a change of venue, the trial court applied the correct Williams standard. The dissent concludes that the trial court's denial of the change of venue motion was based on the "application of an erroneous and antiquated standard." That conclusion is mistaken in that the dissent's reference to the "trial court" fails to distinguish between the single venue ruling of the Assignment Judge who passed on the admissibility of the survey, and the venue rulings of the court before whom defendant was tried. The latter court never applied the incorrect standard.

On the first day of jury selection, September 24, 1984, defense counsel renewed the motion for a change of venue, which the trial court denied. Thereafter, throughout the jury selection period, defense counsel repeatedly moved for a change of venue. Finding that a fair and impartial jury could be selected, the trial court denied all the defendant's motions for change of venue.

The sixth amendment of the federal Constitution, made applicable to the states through the fourteenth amendment, and Article I, paragraph 10 of the state Constitution guarantee "to the criminally accused a fair trial by a panel of impartial, ``indifferent' jurors." Irvin v. Dowd, 366 U.S. 717, 722, 81 S.Ct. 1639, 1642, 6 L.Ed.2d 751, 755 (1961). In capital cases, it is particularly important for the trial court "to preserve the integrity of the jury and minimize the danger that prejudice will infiltrate the adjudicatory process." State v. Williams, supra, 93 N.J. at 63.

It is also well established, however, that a defendant is not entitled to jurors who are totally ignorant of the facts and issues involved in a given case:

In these days of swift, widespread and diverse methods of communication, an important case can be expected to arouse the interest of the public in the vicinity, and scarcely any of those best qualified to serve as jurors will not have formed some impression or opinion as to the merits of the case. This is particularly true in criminal cases. To hold that the mere existence of any preconceived notion as to the guilt or innocence of an accused, without more, is sufficient to rebut the presumption of a prospective juror's impartiality would be to establish an impossible standard. It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court. [Irvin v. Dowd, supra, 366 U.S. at 722-23, 81 S.Ct. at [1642-43], 6 L.Ed.2d at 756 (citations omitted, emphasis added).]

In New Jersey, we have recognized the concerns expressed in Irvin v. Dowd by holding that pervasive pretrial publicity does not necessarily preclude the likelihood of an impartial jury. State v. Biegenwald, supra, 106 N.J. at 35. "While the constitutional standard for a fair trial requires ``a panel of impartial, indifferent jurors,' the jurors actually empanelled need not be ignorant of the facts of the case." State v. Sugar, 84 N.J. 1, 23 (1980) (citation omitted).

Defendant's basic premise is that because many prospective and deliberating jurors had heard about the case as a result of the extensive pretrial publicity, he was unable to secure a fair and impartial jury. Defendant contends that the best remedy to combat this pretrial publicity was for the trial court to grant defendant's motion for a change of venue. Short of a change of venue, the defendant argues that the only other way to obtain a fair and impartial jury was for the trial court to have excluded for cause any juror who had read or heard about the case. Implicit in both these assertions is defendant's contention that the publicity in this case was so extensive and pervasive that the jurors' prejudice must be presumed. We are unpersuaded.

In recent capital cases we held that a trial court has the discretion to change venue where it is "``necessary to overcome the realistic likelihood of prejudice from pretrial publicity.'" State v. Biegenwald, supra, 106 N.J. at 33 (quoting State v. Williams, supra, 93 N.J. at 67-68 n. 13); see State v. Bey, 96 N.J. 625, 630 clarified at 97 N.J. 666 (1984). In State v. Biegenwald, we recognized, as do the federal courts, that in determining whether such a realistic likelihood of prejudice exists in a particular case, a court must first distinguish

between cases in which the trial atmosphere is so corrupted by publicity that prejudice may be presumed, Sheppard v. Maxwell, 384 U.S. 333, 352, 86 S.Ct. 1507, 1516, 16 L.Ed.2d 600, 614 (1966); Estes v. Texas, 381 U.S. 532, 542-44, 85 S.Ct. 1628, 1632-34, 14 L.Ed.2d 543, 550-51 (1965); Turner v. Louisiana, 379 U.S. 466, 472-73, 85 S.Ct. 546, 549-50, 13 L.Ed.2d 424, 429 (1965); Rideau v. Louisiana, 373 U.S. 723, 727, 83 S.Ct. 1417, 1419, 10 L.Ed.2d 663, 665-66 (1963); Marshall v. United States, 360 U.S. 310, 312-13, 79 S.Ct. 1171, 1172-73, 3 L.Ed.2d 1250, 1252 (1959) (per curiam), and cases in which pretrial publicity, while extensive, is less intrusive, making the determinative issue the actual effect of the publicity on the impartiality of the jury panel. [106 N.J. at 33.]

For instances of the latter type of case, see Patton v. Yount, 467 U.S. 1025, 1032-35, 104 S.Ct. 2885, 2889-91, 81 L.Ed.2d 847, 854-56 (1984); Dobbert v. Florida, 432 U.S. 282, 301-03, 97 S.Ct. 2290, 2302-03, 53 L.Ed.2d 344, 361-62 (1977); Murphy v. Florida, 421 U.S. 794, 800-03, 95 S.Ct. 2031, 2036-2038, 44 L.Ed.2d 589, 595-97 (1975); Irvin v. Dowd, supra, 366 U.S. at 723-28, 81 S.Ct. at 1642-46, 6 L.Ed.2d at 756-59; Simmons v. Lockhart, 814 F.2d 504 (8th Cir.1987), cert. den., ___ U.S. ___, 108 S.Ct. 1489, 99 L.Ed.2d 717 (1988); United States v. DePeri, 778 F.2d 963 (3rd Cir.1985), cert. den. sub nom. Murphy v. U.S., 457 U.S. 1110, 106 S.Ct. 1518, 89 L.Ed.2d 916 (1986).

Cases in which prejudice due to pretrial publicity may be presumed are relatively rare and arise out of the most extreme circumstances. Nebraska Press Ass'n v. Stuart, 427 U.S. 539, 554, 96 S.Ct. 2791, 2800, 49 L.Ed.2d 683 (1976) (citing case in which "trial court had failed to use even minimal efforts to insulate the trial and jurors from the ``deluge of publicity'"). In Biegenwald, we contrasted cases in which prejudice was presumed, such as Rideau v. Louisiana, supra, 373 U.S. 723, 83 S.Ct. 1417, 10 L.Ed.2d 663, and cases in which prejudice was not presumed, such as Murphy v. Florida, supra, 421 U.S. 794, 95 S.Ct. 2031, 44 L.Ed.2d 589. State v. Biegenwald, supra, 106 N.J. at 34-35.

In Rideau, during the two months preceding jury selection, the defendant's detailed confession to a murder was televised three times and seen by a substantial number of residents of the parish in which the crime took place. The Supreme Court held that a change of venue should have been granted because the televised confession in a "very real sense was Rideau's trial.... Any subsequent court proceedings in a community so pervasively exposed to such a spectacle could be but a hollow formality." 373 U.S. at 726, 83 S.Ct. at 1419, 10 L.Ed.2d at 665. In Murphy v. Florida, supra, 421 U.S. 794, 95 S.Ct. 2031, 44 L.Ed.2d 589, the defendant also received tremendous publicity because of his prior robbery of the Star of India Sapphire from a New York museum. It was undisputed that the publicity about his prior convictions was widespread. Indeed, each juror had some knowledge of defendant's past convictions. Nonetheless, the Supreme Court upheld the denial of defendant's change of venue motion, noting that publicity had largely ceased seven months before jury selection. In so ruling, the Court stated:

The voir dire in this case indicates no such hostility to petitioner by the jurors who served in his trial as to suggest a partiality that could not be laid aside. Some of the jurors had a vague recollection of the robbery with which petitioner was charged and each had some knowledge of petitioner's past crimes, but none betrayed any belief in the relevance of petitioner's past to the present case. [Id. at 800, 95 S.Ct. 2031, 2036, 44 L.Ed.2d at 595 (footnotes omitted).]

As we noted in State v. Biegenwald, the Supreme Court distinguished the intrusiveness of the publicity in Murphy from that found in Rideau v. Louisiana, supra, 373 U.S. 723, 83 S.Ct. 1417, 10 L.Ed.2d 663, and similar cases by observing:

The proceedings in [those] cases were entirely lacking in the solemnity and sobriety to which a defendant is entitled in a system that subscribes to any notion of fairness and rejects the verdict of a mob. They cannot be made to stand for the proposition that juror exposure to information about a state defendant's prior convictions or to news accounts of the crime with which he is charged alone presumptively deprives the defendant of due process. To resolve this case, we must turn, therefore, to any indications in the totality of circumstances that petitioner's trial was not fundamentally fair. [106 N.J. at 35 (quoting Murphy v. Florida, 421 U.S. at 789, 95 S.Ct. at 2036, 44 L.Ed.2d at 594).]

See also Coleman v. Kemp, 778 F.2d 1487, 1489, 1538 (11th Cir.1985) (held: totality of circumstances permitted finding of presumed prejudice due to pretrial publicity: "the manifest picture that emerges is a community that was deeply prejudiced as to both guilt and sentence"), cert. den., 476 U.S. 1164, 106 S.Ct. 2289, 90 L.Ed.2d 730 (1986); Johnson v. State, 476 So.2d 1195 (Miss. 1985) (court presumed prejudice due to media saturation and totality of circumstances, including fact that murder victim was the white mother of a police officer and defendant was black).

The Supreme Court of California has adopted a standard for granting changes of venue due to potentially prejudicial publicity that is very similar to our "realistic likelihood of prejudice" standard:

A motion for change of venue or continuance shall be granted whenever it is determined that because of the dissemination of potentially prejudicial material, there is a reasonable likelihood that in the absence of such relief, a fair trial cannot be had. ... A showing of actual prejudice shall not be required. [Maine v. Superior Court, 68 Cal.2d 375, 438 P.2d 372, 377, 66 Cal. Rptr. 724, 729 (1968) (emphasis added).]

A California court, however, will not presume prejudice without first examining certain factors that measure the potentially prejudicial effects of pretrial publicity. Martinez v. Superior Court of Placer County, 29 Cal.3d 574, 629 P.2d 502, 174 Cal. Rptr. 701 (1981). These factors include the nature and extent of the news coverage, the size of the community, the nature and gravity of the offense, and the respective standings of the victim and the accused in the community. Id. at 578, 629 P.2d at 504, 174 Cal. Rptr. at 703 (based on cited factors there was reasonable likelihood that fair trial could not be had without venue change); Williams v. Superior Court of Placer County, 34 Cal.3d 584, 668 P.2d 799, 194 Cal. Rptr. 492 (1983) (held: change of venue motion mistakenly denied; defendant was member of minority group and stranger to community; victim was white and from prominent family; and case had political overtones). But see People v. Balderas, 41 Cal.3d 144, 711 P.2d 480, 496-99, 222 Cal. Rptr. 184 (Cal. 1985) (held: pre-trial publicity not presumptively prejudicial where (1) neither victim nor defendant prominent; (2) county population equals 405,600; and (3) no evidence of unusual local hostility toward defendant's ethnic group); Odle v. Superior Court of Contra Costa County, 32 Cal.3d 932, 654 P.2d 225, 187 Cal. Rptr. 455 (1982) (held: venue motion properly denied where more than two years passed between time of intense media coverage and prospective trial date, publicity not geographically pervasive, publicity not particularly hostile, and standing of accused and victim were neutral factors in case). According to California's high court, a defendant whose pretrial venue motion is denied may renew that motion if the voir dire later reveals that there is actual prejudice against the defendant due to pretrial publicity. Id. at 947, 654 P.2d at 234, 187 Cal. Rptr. at 464.

In State v. Biegenwald, we recognized that there was extensive pretrial publicity, including numerous articles in which the prosecutor "repeatedly assumed defendant's guilt and also stated that defendant killed only for pleasure." 106 N.J. at 31. We concluded, however, that

this is not a case in which the trial court was required to presume prejudice prior to the jury voir dire. The extensive pretrial publicity was concentrated in April and May, 1983. In addition to prohibiting further comment by counsel, the trial court adjourned the trial date until mid-November, allowing nearly six months to permit the impact of the publicity to subside. [Id. at 35.]

In our view, the instant case is distinguishable from those cases in which courts have presumed prejudice as a result of pretrial publicity. The facts of this case differ in several important respects: (1) there was no evidence of extreme community hostility, as distinct from fear, directed specifically against defendant; (2) neither the victim nor the defendant was a prominent member of the community; (3) the victim was not a public servant; (4) the defendant was not an outsider; and (5) the community was not predisposed to his guilt. The absence of these factors persuades us that the pretrial publicity was not presumptively prejudicial. See Johnson v. State, supra, 476 So.2d at 1213-1215. We find further support for this conclusion in the fact that two years elapsed between the time of most intense publicity and the commencement of jury selection. See State v. Biegenwald, supra, 106 N.J. at 35. We also note that the nature of the publicity in the instant case was not as inflammatory as that found in Rideau v. Louisiana, supra, 373 U.S. 723, 83 S.Ct. 1417, 10 L.Ed.2d 663, and Coleman v. Kemp, supra, 778 F.2d 1487, nor was the defendant's trial conducted in the circus-like atmosphere of Sheppard v. Maxwell, supra, 384 U.S. 333, 86 S.Ct. 1507, 16 L.Ed.2d 600, and Estes v. Texas, supra, 381 U.S. 532, 85 S.Ct. 1628, 14 L.Ed.2d 543.

Our dissenting colleagues are of the view that the case of Fisher v. State, 481 So.2d 203 (1985), requires us to find reversible error in the trial court's refusal to grant a change of venue. We think not. That case is neither binding on this court, nor dispositive of the issue before us. In Fisher, the Mississippi Supreme Court found that the quantity and quality of pretrial publicity surrounding that case gave rise to a presumption of prejudice. As noted earlier, we find that such a presumption is not warranted in the instant case. In a very real sense, therefore, our disagreement with the dissenters is one of factual interpretation. We do not believe that the nature and quantity of pretrial publicity in this case gave rise to a presumption of prejudice; our dissenting colleagues, however, have examined the same facts and have reached a different conclusion. We do not, as the dissent suggests, "hold" that Fisher "was wrongly decided." We simply find that while the facts in Fisher might very well have warranted a presumption of prejudice, our interpretation of the facts before us does not command the same result. The publicity regarding defendant Fisher was more widespread and pervasive, and was markedly more hostile than any publicity concerning Koedatich. Moreover, the publicity in Fisher saturated a considerably smaller community, and reached its most intense level only 10 1/2 months prior to trial. Given these factual distinctions, we do not feel that a result similar to the one reached in Fisher is warranted here.

Where prejudice is not presumed, the appropriate inquiry in determining whether a change of venue is necessary to overcome the "realistic likelihood of prejudice" is whether under the totality of circumstances "the jury process resulted in a fair and impartial jury." State v. Biegenwald, supra, 106 N.J. at 35-36. Thus, the court must examine the extent to which potential jurors are biased as a result of any publicity surrounding the case. In determining the actual bias of a juror, courts place great reliance on a trial court's voir dire examination of a juror. Patton v. Yount, supra, 467 U.S. at 1038-1039, 104 S.Ct. at 2892-2893, 81 L.Ed.2d at 858. This is so because "the determination of a juror's bias is essentially one of credibility, and therefore largely one of demeanor.... [T]he trial court's resolution of such questions is entitled, even on direct appeal, to ``special deference.'" Ibid. In sum, in the absence of presumed prejudice resulting from pretrial publicity, "the Sixth Amendment inquiry turns largely on the adequacy of the voir dire." Welch v. United States, 466 A.2d 829, 835 (D.C. 1983).

In New Jersey, we give great deference to the trial court in conducting the voir dire. As this Court stated in State v. Jackson, 43 N.J. 148, (1964), cert. den. sub nom. Ravenell v. New Jersey, 379 U.S. 982, 85 S.Ct. 690, 13 L.Ed.2d 572 (1965), "the trial court is vested with broad discretionary powers in determining the qualifications of jurors.... [I]ts exercise of discretion will ordinarily not be disturbed on appeal." Id. at 160. In State v. Singletary, 80 N.J. 55 (1979), this Court discussed the rationale underlying this rule:

Decisions concerning the potential bias of prospective jurors are primarily subjective in nature. They require at bottom a judgment concerning the juror's credibility as he respond to questions designed to detect whether he is able to sit as a fair and impartial trier of fact. Consequently, such evaluations are necessarily dependent upon an observation of the juror's demeanor during the course of voir dire — observations which an appellate court is precluded from making.

* * * * * * * *

Although a juror's professions of impartiality will not always insulate him from excusal for cause, see, e.g., State v. Jackson, supra; State v. Deatore, supra [70 N.J. 100 (1976)] they will be accorded a great deal of weight, see, e.g., State v. Grillo, supra; State v. Jefferson, 131 N.J.L. 70, 72 (E & A 1943). Inasmuch as the trial judge observed the venireman's demeanor, he was in a position to accurately assess the sincerity and credibility of such statements, and we should therefore pay due deference to his evaluation.... [Id. at 63, 64.]

Thus, this Court has upheld the trial court's discretion in refusing to excuse for cause jurors whose exposure to pretrial publicity led them to form an opinion about the facts of the case and the guilt or innocence of the accused:

While the constitutional standard for a fair trial requires "a panel of impartial, ``indifferent' jurors," Irvin v. Dowd, 366 U.S. 717, 722 [81 S.Ct. 1639, 1642], 6 L.Ed.2d 751 (1961), the jurors actually empaneled need not be ignorant of the facts of the case.... ``It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court.' [State v. Sugar, supra, 84 N.J. at 23 (citations omitted).]

In State v. Williams, supra, 93 N.J. at 66-67, we set forth an "entire panoply of procedures" that a trial court may use "to determine and evaluate the character, availability, feasibility and efficacy of alternative methods of safeguarding against prejudice." In doing so, the trial court must draw upon "its own special judicial expertise and experience." Id. at 67. In addition to the options of securing foreign jurors to augment the jury pool, adjournment of the trial date and change of venue, we identified the voir dire as one of the most critical and important means of assuring jury impartiality:

[An] important, indeed critical, means for dealing with potential and latent bias is the voir dire. The court should consider the efficacy of more exhaustive and searching voir dire examinations. The court in conducting the voir dire should be particularly responsive to the requests of counsel regarding the examination of prospective jurors as to potential bias. The court could consider whether there should be a greater willingness to resolve doubts in favor of the defendant in excusing jurors for cause. Particularly in capital cases, trial judges should exercise extraordinary care in the voir dire of potential jurors * * *. [Id. at 68-69 (footnotes omitted).]

In State v. Biegenwald, supra, 106 N.J. at 35-37, and State v. Ramseur, supra, 106 N.J. at 256-57, we continued our special deference to the trial court in reviewing the voir dire on both death qualification and publicity. In Ramseur, we noted both that "[a] sensitive weighing and appraisal of a juror's entire response must be made by the trial court in its duty to resolve the question of whether the juror has shown bias or prejudgment," 106 N.J. at 257, and that this Court is "perhaps too far removed" from the realities of the voir dire to appreciate the nuances concealed by a "bloodless record"; thus, deference to the trial court is usually in order. Id. at 260 (quoting State v. Gilmore, 103 N.J. 508, 547 (1986) (Clifford, J., dissenting)).

We recognize that an appellate court faces certain difficulties when reviewing the adequacy of the voir dire. Among other things, the appellate court does not share the benefit of having made in-court observations of the potential juror. Despite an appellate court's inherently inferior vantage point, we perceive our role in reviewing the voir dire as follows:

[A]n appellate tribunal is likewise under a duty to make an independent evaluation of the facts and circumstances and of the juror's voir dire examination. It should determine for itself whether the pretrial newspaper stories are so pervasive and so prejudicial, or the juror's protestation of unaffected impartiality after reading them so unconvincing or doubtful that a new trial should be ordered.

[State v. Van Duyne, 43 N.J. 369, 386 (1964), cert. den., 380 U.S. 987, 85 S.Ct. 1359, 14 L.Ed.2d 279 (1965).]

In this case jury selection began on September 24, 1984, and ended on October 3, 1984. On the first day of jury selection, defense counsel renewed his motion for a change of venue, supplementing the record with a Sunday, September 23, 1984, front-page article from the Morris County Daily Record. The court denied the motion as premature. At the end of the first day of jury selection, defendant renewed the motion. The trial court again denied it. It noted that during the first day of jury selection, five out of twelve potential jurors had been preliminarily qualified (not excused for cause). It noted that of these five, two had no prior knowledge of the case. Based on the first day of jury selection, the trial court stated that a fair and impartial jury could be selected. Publicity continued through the first few days of jury selection.

Defense counsel renewed his motion for a change of venue on the second day of jury selection, when he introduced an article from the previous day's Daily Record depicting the defendant "shackled and with armed guards escorting him to the Court...." Defense counsel asserted that since many people read or heard about the case in the paper, he was forced to use peremptories for publicity purposes rather than for other reasons.^[7] The court then denied the motion, stating: "I do not regard as an appropriate test the fact that someone has read about this case before.... There are people who don't even remember what they read ... and there are people who say that they don't rely upon the press ... for determining guilt or innocence but will rely upon the evidence presented in the courtroom.... [T]here is no evidence ... that the citizens of Morris County cannot give to the defendant a fair and impartial trial by jury." The defendant renewed his motion again on the third day of jury selection and again on October 1. The motions were denied. The trial court held that a fair and impartial jury could be selected.

The trial court took numerous precautions in conducting the voir dire. All 105 prospective jurors were required to fill out a questionnaire. The questionnaire contained general questions concerning the juror's education, occupation, physical condition, familiarity with the defendant, the prospective witnesses, and the lawyers in the case, and whether the juror or a member of his or her family or a close friend had ever been charged with a criminal violation or had ever been the victim of a crime. The questionnaire also contained the following questions:

Do you have any opinion as to the guilt or innocence of Mr. Koedatich on the charge contained in the indictment?

Yes [] No []

Have you expressed an opinion as to the guilt or innocence of Mr. Koedatich at any prior time?

Yes [] No []

Have you read anything about this case in any newspaper?

Yes [] No []

How often do you read the following newspapers (circle one)?

a. Star Ledger      Daily
Very often
Infrequently
Never
b. Daily Record     Daily
Very often
Infrequently
Never
c. Daily Advance    Daily
Very often
Infrequently
Never
d. New York Times   Daily
Very often
Infrequently
Never

Do you read any other newspapers?

Yes [] No []

If so, list their names _______________________________________ _______________________________________________________________ _______________________________________________________________

Do you watch television news programs almost daily?

Yes [] No []

If so, list which channels and what usual time of day.

CHANNEL TIMES _______ _____ _______ _____ _______ _____

If you are selected to sit as a juror on this case, would you be willing and able to render a verdict based solely on the testimony you will hear, the other evidence presented and the law as the court will give it to you in its instructions, disregarding any other ideas, notions or beliefs about the law that you may have encountered?

Yes [] No [] _______________________________________________ (Signature of juror) _______________________________________________ (Please print name)

The trial court excused twenty-six potential jurors based solely on their written responses to the questionnaire. Of the seventy-nine prospective jurors who were asked if they had read or heard about the case, nine said that they had not; four of these nine became deliberating jurors. Seventy-seven jurors were excused by the trial court. Of these, forty were excused because of their knowledge of the case, twenty were excused for hardship reasons, five were excused because of their feelings on capital punishment, and twelve were excused for miscellaneous reasons. The court excused for cause any prospective juror who indicated knowledge of defendant's prior murder conviction or the pending charges in the O'Brien case. Additionally, the court excused for cause any juror who indicated he or she had formed an opinion about the defendant's guilt or innocence. Approximately thirteen percent of ninety-six jurors questioned by the Court indicated that they had formed such an opinion. Defense counsel exercised nine peremptory challenges. The State exercised four.

Of the fifteen jurors who sat during the trial, five had not read or heard anything about defendant: Edgar Jadwin, Herman Nadel, Joseph Fiorentino, William Festa, and Robert O'Shaughnessy. Of the remaining jurors, Judith Deelsnyder had read about defendant's arrest, Dan Zingone had not read about the defendant but had noticed his picture in a paper, Loraine Zaccaro and Glen Cosentiono had read about the case primarily two years earlier but recalled nothing of the defendant's background. Leslie Fascia, who served as an alternate, read something about the defendant "two years ago," and initially stated that she remembered reading something "about the body being found out on Route 80," but then remarked, "I don't remember." Elizabeth Criares last read about the case three months before jury selection, and associated the defendant with stabbing. Alice Budd, Deborah Brown, Margaret Andrews, and Barbara Herzig each had knowledge of the case and of defendant's arrest, but disclaimed any knowledge of defendant's background. Deborah Brown, for instance, testified that she talked about details of the case, and saw pictures in the Sunday paper but did not read the whole article and could remember nothing in particular about defendant's background. Margaret Andrews also testified to reading about the defendant in the papers, most recently "[t]his weekend" in the Daily Record, but could remember nothing in particular about defendant's background. When asked if she remembered reading or hearing anything about defendant's background, Barbara Herzig replied, "It was a while ago. I have forgotten." Loraine Zaccaro answered "No" when asked if she had "ever read or hear[d] anything about Mr. Koedatich" despite her knowledge of the circumstances of the crime.

In a recent capital case, we rejected a trial court's exclusive reliance on protective instructions in order to prevent juror exposure to prejudicial, midtrial publicity. State v. Bey (I), 112 N.J. 45, 80 (1988). In the instant case, we rely principally on the strength of the voir dire in upholding the trial court's denial of the change of venue motion. We rely on the court's use of repeated admonitions to the jury to avoid publicity only insofar as it provides additional support for our conclusion. Throughout the jury selection process and throughout trial, the trial court made every effort to preserve the jury's impartiality by reminding the jurors of their duty to avoid any publicity about the case.

Once a juror was preliminarily qualified but not yet selected as a final juror, the trial court advised the juror not to read about the case in the newspapers, not to listen to any news accounts of the case on radio or television, and not to discuss the case with anyone. Prior to swearing in the jury, the trial court asked each juror individually if he or she had followed the court's order not to read or listen to any accounts of the case or to discuss the case with anyone. Each juror stated that he or she had obeyed that order. Immediately after swearing in the jury, the trial court reminded the jurors that they had just taken the oath to discharge their obligation "faithfully, impartially and justly." Once again the trial court forcefully reminded the jurors that they could not be subject to any outside influences. The court impressed upon the jurors that the case must be decided on the evidence to be presented in the courtroom, and again instructed the jurors not to read any newspaper accounts of the case. The trial court further instructed the jurors not to watch anything on television about the case or to listen to any radio accounts of the case, and not to discuss the case with any other people or let other people discuss the case with them. Finally, the jurors were instructed not to discuss the evidence or case among themselves while it was in progress. The trial court repeated these instructions at the end of each day of trial and prior to each lunch recess.

In capital cases, we have emphasized the trial court's obligation "``to preserve the integrity of the jury and minimize the danger that prejudice will infiltrate the adjudicatory process....'" State v. Biegenwald, supra, 106 N.J. at 32 (quoting State v. Williams, supra, 93 N.J. at 63). While we recognize that a change of venue may disrupt court administration, it is imperative that the defendant in a capital case receive an impartial jury. Trial courts should not be reluctant to grant motions for change of venue in capital cases. On the contrary, courts should grant such motions liberally. While perhaps defendant's motion for change of venue could well have been granted, we do not find that the court's failure to follow that course constitutes reversible error. Instead, we conclude in this case, as we did in State v. Biegenwald, supra, that the voir dire assured the selection of an impartial jury:

Although in some instances the trial court's interrogation was more general and less searching than that requested by counsel, our independent review of the record reveals that the overall scope and quality of the voir dire was sufficiently thorough and probing to assure the selection of an impartial jury. [106 N.J. at 29.]

Based on this record, we are satisfied that the trial court took adequate measures during voir dire to insure that an impartial jury was impanelled. The trial court conducted a thorough voir dire. It discharged any juror who stated that he or she had read of the prior murder, or the O'Brien charges, or had formed an opinion about defendant's guilt or innocence. If a juror indicated he or she had read or heard about the case, the trial court carefully probed the juror about the juror's knowledge of the case and of the defendant. Finally, the court willingly accepted suggestions from defense counsel about questions to be asked and areas to be probed.

Accordingly, we conclude that the trial court did not abuse its discretion in denying the defendant's motion for a change of venue. In our view, defendant failed to show that a change of venue was necessary to overcome the "realistic likelihood of prejudice" resulting from pretrial publicity. Our independent review of the record persuades us that the trial court's voir dire assured the selection of an impartial jury.

Justice Handler in his dissent in this case and in State v. Bey (I), 112 N.J. 45, 105 (1988), urges the Court to adopt an enhanced standard of appellate review for determining reversible error in capital cases. According to Justice Handler's proposed standard, error of constitutional dimension is reversible unless the State can demonstrate that such error had "no effect" on the jury's deliberations and non-constitutional error is reversible unless the State can demonstrate that the error had "no realistic likelihood" of prejudicing the jury's deliberations. 112 N.J. at 116. The proposed standard would apply to all phases of a capital murder prosecution. Its application would not be restricted to the sentencing phase of the trial, but rather would apply with equal force to the guilt phase. Id. at 116. The standard would eliminate all other tests for determining reversible error, such as "plain error" and "harmless error." Justice Handler thus proposes a new substantive standard, one that he intends be used exclusively in capital prosecutions.

For the reasons the Court expressed in State v. Bey (I), supra, 112 N.J. at 91, we reject Justice Handler's enhanced standard of review. Cognizant of the severity of the death penalty, we shall continue to conduct a searching and scrupulously careful review of the entire record of a capital case. Nonetheless, the substantive standard of appellate review remains the same. We thus disagree with Justice Handler's finding that certain alleged errors are reversible under the enhanced standard. The dissenter relies on his enhanced standard of review to support his conclusion that reversible error was committed on any one of the following grounds: failure of the trial court to grant motion for change of venue; failure of this Court to grant limited remand to interrogate jurors in connection with a newspaper article of March 1985; prosecutorial misconduct with respect to the cross-examination of David Baldwin; and the prosecutor's summations in both the guilt and penalty phases. Finally, in an issue not raised by either the State or defendant, the dissenter uses the enhanced standard of review to conclude that the trial court's instruction to the jury that its finding of a mitigating factor had to be unanimous resulted in a coercive charge that requires that the defendant not be exposed to the death sentence on remand. We do not agree.

2. Excusal for Cause of Jurors Who Read or Heard about the Case

Defendant claims that this case is unlike State v. Biegenwald, supra, 106 N.J. 13, in that so many of the jurors here read or heard about the case. In State v. Biegenwald, the sixteen impanelled jurors "indicated that they had encountered little or no publicity regarding the case.... [A] substantial segment of the jury panel ... unequivocally and credibly demonstrated that the pretrial publicity had passed them by," and, thus, this Court was "satisfied that the jury that was impanelled was as a whole impartial." Id. at 36. The Biegenwald jury, like the jury in this case, "did not include anyone who recalled having previously read anything about other murders or a prior murder conviction." Id. at 37.

After reviewing the jurors' questionnaires and after examining them extensively on voir dire, the court did in fact qualify as final deliberating jurors some veniremen who recalled reading or hearing about defendant in the papers or television but who remembered little or nothing about the defendant. Jurors who did recall some of the details reported in media accounts did not recall anything regarding the defendant's prior murder conviction and his indictment for the O'Brien murder. Their recollection of the factual circumstances of the murder and of the defendant was slight. It is important to recall that in State v. Biegenwald one of the impanelled jurors knew of defendant's alleged prior murders but not of his prior conviction. She apparently learned of the prior murders from other veniremen. Yet, the trial court and the defense were convinced by her voir dire that she could disregard what she had heard and could serve impartially. We affirmed that conclusion by finding that the final Biegenwald jury was an impartial and fair one. Id. at 37. The fact that those jurors who admitted exposure recalled nothing regarding this defendant's prior murder convictions and his indictment in the O'Brien matter, coupled with their assurances of impartiality, persuades us to reach the same conclusion we reached in State v. Biegenwald, namely, that defendant was tried by a fair and impartial jury.

Implicit in defendant's contention that the trial court should have excluded for cause any juror who read or heard about the case is the notion that voir dire is an ineffective means of determining juror prejudice under such circumstances. Defendant thus suggests that we adopt a rule whereby any juror who has heard or read about the case would be automatically disqualified regardless of the extent of his or her recollection. We decline to adopt such a rule for which we find no support in this or any other jurisdiction. In fact, such a rule would be inconsistent with long-established precedent holding that acceptable jurors need not be entirely ignorant of the matter at hand. See Murphy v. Florida, supra, 421 U.S. at 799-800, 95 S.Ct. at 2035-36, 44 L.Ed.2d at 594-95 ("Qualified jurors need not ... be totally ignorant of the facts and issues."); Irvin v. Dowd, supra, 366 U.S. at 722-23, 81 S.Ct. at 1642, 6 L.Ed.2d at 756 (rejecting proposition that "any preconceived notion as to guilt or innocence of an accused" is sufficient to disqualify prospective juror); State v. Allen, 73 N.J. 132, 161 n. 8 (1977) (Pashman, J., concurring) (juror who has seen news reports should not automatically be excused for cause); State v. VanDuyne, supra, 43 N.J. at 385-87; see also Smith v. Phillips, 455 U.S. 209, 217, 102 S.Ct. 940, 946, 71 L.Ed.2d 78, 86 (1982) ("due process does not require a new trial every time a juror has been placed in a potentially compromising situation"). As we mentioned earlier, "[i]t is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court." Irvin v. Dowd, supra, 366 U.S. at 722-23, 81 S.Ct. at 1642, 6 L.Ed.2d at 756.

3. Certain Jurors' Knowledge

Defendant argues that the jury's verdict is invalid because a few deliberating jurors were aware of defendant's involvement in the O'Brien murder. In support of this contention, defendant relies solely on a newspaper article appearing in the Daily Record following defendant's conviction and sentencing. The article quoted statements made by two deliberating jurors and one alternate juror. See Koedatich Defense Wins Juror Appeal, Daily Record, Mar. 10, 1985, at 1, col. 1.

Although defendant was first indicted for the O'Brien murder, he was tried first for the murder of Amie Hoffman. Approximately four months after the Hoffman trial but during the O'Brien trial, the Daily Record published an article concerning jury selection in the O'Brien case. In contrasting the jury selection in the O'Brien case with that in the Hoffman case, the article noted that Judge Stein "excluded any juror who knew of Koedatich's arrest record." Ibid. The article then related that "Stein's conservative approach did not prevent some jurors from linking Koedatich to both murders once the trial began. The Daily Record has learned that some of the jurors who convicted Koedatich of the Hoffman murder knew he was also charged with killing O'Brien. This information was not allowed to be brought before the Hoffman jurors because it could prejudice Koedatich's case. But a few jurors said once the trial got under way they realized Koedatich was charged with killing both women." Ibid. Specifically, the article quotes three jurors, two of whom were deliberating jurors, Ms. Zaccaro and Ms. Herzig, and one alternate who did not deliberate, Ms. Fascia. Ms. Zaccaro stated, "I didn't know he did it in the beginning but halfway through the trial I realized he was linked to O'Brien." Ibid. Zaccaro also said that the O'Brien murder was not discussed until after the verdict was returned. From the article, it is unclear whether Ms. Herzig acknowledged that she or the other jurors knew that defendant was charged with having murdered O'Brien. However, she did state that the O'Brien murder played no part in the jury's decision to convict Koedatich of Hoffman's murder. "It had no bearing on the Hoffman case," she said. "The Hoffman case was based on evidence at the trial." Ibid. Ms. Fascia "said she thought most of the jurors were aware that Koedatich was charged with the O'Brien murder. She said jurors were surprised to learn during the penalty phase of the trial that Koedatich was convicted of murder in Florida." Ibid. Ms. Fascia was an alternate juror.^[8]

In July 1985, defendant submitted a copy of the Daily Record article of March 10, 1985, to this Court in support of his motion for a limited remand to the trial court. On remand, defendant proposed to move for a new trial and for confidential interviews of the jurors pursuant to Rule 1:16-1. In support of the remand motion before this Court, defense counsel submitted an affidavit in conjunction with the Daily Record article. The affidavit essentially repeated the allegations of the newspaper article and did not advance any other evidence or justification for a jury interrogation. The State opposed the motion arguing that defendant had not made a sufficient showing of possible jury taint to justify an interrogation pursuant to Rule 1:16-1 and, further, that the jury recall motion was not ripe for adjudication. On October 7, 1985, we denied defendant's motion.

We have long held that "[c]alling back jurors for interrogation after they have been discharged is an extraordinary procedure which should be invoked only upon a strong showing that a litigant may have been harmed by jury misconduct." State v. Athorn, 46 N.J. 247, 250 (1966). We determined that investigations into secret jury deliberations should not be readily used to invalidate verdicts because "disappointed litigants would be encouraged to tamper with jurors, to harass them and to employ fraudulent practices in an effort to induce them to repudiate their decisions. Moreover, an open invitation would be extended to any disgruntled juror who might choose to destroy a verdict to which he had previously assented." Ibid. (citations omitted). Furthermore, we have observed that "secrecy surrounding jury deliberations is necessary not only to prevent the unsettling of verdicts, but also as an aid to the deliberative process itself. Each juror must be free to discuss his or her thoughts freely." State v. LaFera, 42 N.J. 97, 106 (1964). Freedom of debate would be stifled if jurors were aware that their argument and votes would be broadcast to the world. Clark v. United States, 289 U.S. 1, 13, 53 S.Ct. 465, 469, 77 L.Ed. 993, 999 (1933). For strong policy reasons, courts "have generally refused to accept from jurors, for the purpose of impeaching a verdict, any evidence of the discussion which they may have had among themselves while considering their verdict." State v. Athorn, supra, 46 N.J. at 251 (citing cases).

We do, however, recognize two general exceptions to this rule when the "plainest principles of justice" require a new trial. State v. Athorn, supra, 46 N.J. at 251-52. First, we have held that any racial or religious bigotry manifested in jury deliberations may invalidate a verdict. State v. Levitt, 36 N.J. 266 (1961). The second exception arises when a juror informs or misinforms his or her colleagues in the jury room about the facts of the case based on his personal knowledge of facts not in evidence. In State v. Kociolek, 20 N.J. 92 (1955), a juror's affidavit revealed that the disclosure of other pending charges was the determinative factor in reaching a unanimous verdict for the imposition of the death penalty. The affidavit stated that the jury arrived at the death verdict only after some jurors mentioned, and the entire jury considered, another indictment against defendant not in evidence. Id. at 94. According to the juror's affidavit, this prejudicial information changed an eight-to-four consensus for life imprisonment into a unanimous verdict for the death penalty. Id. at 95.

There is no juror affidavit in this case. Furthermore, we strongly believe that the contents of a single newspaper article, indisputably hearsay, cannot be the sole basis for the extraordinary procedure of a post-trial jury interrogation. We have long recognized a need "to protect all the jurors against efforts of others to browse among their thoughts in search of something to invalidate their verdict." State v. LaFera, supra, 42 N.J. at 106. In LaFera, the convicted defendants engaged a private investigator to conduct a costly and extensive postconviction investigation of the jury in order to uncover some basis for impeaching the verdict. The investigation revealed some indication of one juror's prejudice and prejudgment, which that juror had expressed to former associates at his place of employment. We rejected this proffered evidence as a ground for a post-trial interrogation of the jury. We concluded that it was "unfair to jurors to permit a disappointed litigant to pick over their private associations in search of something to discredit them and their verdict." Id. at 107.

The inconclusive hearsay contents of the March 10, 1985, Daily Record article do not warrant a post-trial jury interrogation any more than the private investigation results warranted an interrogation in LaFera. This is particularly true where, as here, the newspaper interviewed the jurors five months after the conclusion of the penalty phase and during defendant's well-publicized trial for another murder. We continue to believe that post-trial jury interrogations are "an extraordinary procedure which should be invoked only upon a strong showing that a litigant may have been harmed by jury misconduct." State v. Athorn, supra, 46 N.J. at 250.

Defendant has not made the requisite strong showing of misconduct; the only supporting evidence that defendant has ever presented is defense counsel's affidavit that relies solely on the contents of the newspaper article. Defendant has never asserted that any juror impermissibly considered defendant's involvement in the O'Brien murder in reaching a decision in this case. Even assuming, as the dissenter does, the quoted juror statements that appeared in the article to be true and construing those statements most favorably to defendant, there still is no basis for concluding that defendant has made a strong showing of jury misconduct during deliberations or that any individual juror did not make an impartial determination. No juror expressed in the article that his or her decision was influenced by knowledge that defendant had been charged with the O'Brien killing; in fact, the interviewed jurors specifically stated that they had not been so influenced. Moreover, no juror stated that knowledge of defendant's involvement in the O'Brien killing had prevented him or her from reaching an impartial determination. None of the interviewed jurors suggested that the extraneous fact of Koedatich's involvement in the O'Brien murder had entered the jury deliberations. Finally, no member of the jury ever came forward to the court or counsel for the State or the defense and informed anyone of possible taint in the jury proceedings. Accordingly, there has never been any suggestion that any juror did not base his or her verdict solely on the evidence in the case. Irvin v. Dowd, supra, 366 U.S. at 722-23, 81 S.Ct. at 1642.

4. Motion for Attorney-Conducted Voir Dire

Defendant contends that the trial court's refusal to permit defense counsel to interrogate the jury during voir dire was prejudicial error. He argues that neither our decision in State v. Manley, 54 N.J. 259 (1969), nor Rule 1:8-3(a)^[9] should be construed to prohibit attorney-conducted voir dire in capital cases.

We recently decided this issue in State v. Biegenwald, supra, 106 N.J. at 27-30, where we held that our decision in State v. Manley was intended to apply in capital cases. In State v. Manley, we announced our adoption of the predecessor of Rule 1:8-3(a), which in its current form leaves the conduct of voir dire to the trial court, and grants it the "discretion to permit or restrict supplemental questioning by counsel." State v. Biegenwald, 106 N.J. at 28 (citing State v. Manley, supra, 54 N.J. at 281-83). As we stated in State v. Biegenwald, our decision to extend the decision in State v. Manley to death penalty cases was based on "the Manley opinion's ``call[] for a much more guarded discretion than previously announced in State v. Sullivan, 43 N.J. [at] 239-40; 54 N.J. at 283, a capital case, and the text of Rule 1:8-3(a).'" Id. 106 N.J. at 28-29.

Given our holding in State v. Biegenwald, the issue in the instant case is whether the trial court abused its discretion in denying the motion for attorney-conducted voir dire. As we explained in State v. Biegenwald, "[o]ur present Court rule is intended to see that voir dire is conducted to the extent reasonably possible by the court. The trial court is given discretion to permit counsel to supplement the court's interrogation of jurors by submitting questions to the court and, where the court approves, by additional personal questioning by counsel." Id. at 29 (citing R. 1:8-3(a) and State v. Manley, supra, 54 N.J. at 282-83). In the instant case, the trial court adhered to this procedure, showing itself to be very responsive to questions proposed by counsel. As in State v. Biegenwald, "[n]o instances have been cited to demonstrate that the trial court abused its discretion in refusing to allow questions to jurors." Id. 106 N.J. at 29. We also note that the trial court's voir dire was "sufficiently thorough and probing to assure the selection of an impartial jury." Ibid. We therefore hold that the trial court's refusal to permit the voir dire interrogation to be conducted by counsel was within the limits of our decision in State v. Manley, supra, 54 N.J. 259, and of Rule 1:8-3(a).

C. Exclusion for Cause of Death-Scrupled Jurors

Defendant challenges the trial court's exclusion for cause of two death-scrupled jurors, Frances Kolen and Mark Erich. Ms. Kolen indicated to the trial court, on voir dire, that she could not serve as a juror because it would be upsetting to have to deliberate in a case with gory details, and because her family had only one car. The court then asked for "[a]ny other reason that comes to ... mind"; she answered:

Whether I felt it was deserved or not I could not bring myself to say someone should be put to death.

Q: You could not?

A: I could not.

Q: Under any circumstances, regardless of the facts you couldn't vote for the death penalty?

A: I don't think so.

Q: You have pretty strong feelings about that to the extent that you just couldn't vote the death penalty regardless of what the facts were in the case?

A: I believe so.

Q: You believe what?

A: That I couldn't vote that way.

The trial court found: "I don't see how I could permit her to sit as I understand the Witherspoon case. I think I pressed that as far as I can and [have] taken her through all the ramifications that I reasonably could for the type of responses she gave.... Mrs. Kolen ... maintains she could not vote for the death penalty no matter what the circumstances.... I'm going to have to excuse her."

Defendant challenges this exclusion because the court "made no attempt to conform to the then-existing requirements of Witherspoon.... The court made no attempt to inquire whether she understood the two-stage process; whether she could weigh and evaluate aggravating and mitigating factors; or, whether she could consider the death penalty under certain exceptional circumstances...."

The State, on the other hand, points to this Court's adoption, in State v. Ramseur, supra, 106 N.J. at 256, of the Supreme Court's modification of Witherspoon v. Illinois, 391 U.S. 510, 88 S.Ct. 1770, 20 L.Ed.2d 776 (1968), in Wainwright v. Witt, 469 U.S. 412, 105 S.Ct. 844, 83 L.Ed.2d 841 (1985). Under Witt and Ramseur, to justify exclusion it is no longer necessary that a juror's opposition to the death penalty be "automatic." Rather, those cases hold that "the quest is for jurors who will conscientiously apply the law and find the facts," id. at 423, 105 S.Ct. at 852, 83 L.Ed.2d at 851 and any juror whose death penalty scruples would "``prevent or substantially impair the performance of his duties as a juror in accordance with his instructions and his oath'" is excludable. Id. at 424, 105 S.Ct. at 852, 83 L.Ed.2d at 851-52.

Furthermore, in Darden v. Wainwright, supra, 477 U.S. 168, 106 S.Ct. 2464, 91 L.Ed.2d 144, the Supreme Court, applying the Witt standard, upheld the exclusion of a juror who had merely answered "yes" to the question "Do you have any moral or religious, conscientious moral or religious principles in opposition to the death penalty so strong that you would be unable ... to vote to recommend a death penalty regardless of the facts?" Although, as the defendant points out, there were facts that distinguish this case from Darden v. Wainwright (e.g., more extensive explanation in Darden v. Wainwright of the meaning of the Wainwright v. Witt inquiry, and the absence of objection by defense counsel in Darden v. Wainwright), it is clear that jurors' bias need not be shown with "unmistakable clarity." State v. Ramseur, supra, 106 N.J. at 256. We find that Ms. Kolen's answers established, at a minimum, that her death-penalty scruples would have prevented or substantially impaired her performance of her duties in accordance with her oath. The court's excusal for cause was proper.

A closer case is presented by prospective juror Mark Erich. Mr. Erich indicated to the court that he generally could "apply the fact as to the law ... even if you don't like the principles of law or any principle...." The inquiry then turned to capital punishment.

THE COURT: Do you have any beliefs one way or the other about capital punishment?

MR. ERICH: Yes.

THE COURT: What are your beliefs?

MR. ERICH: I have trouble believing in capital punishment.

THE COURT: Do you have any fixed view about whether you have trouble or not?

MR. ERICH: No.

THE COURT: Would you be willing to listen to the facts of the case and consider all of the circumstances before you would come to a conclusion as to whether or not a person convicted of murder is to receive the death penalty?

MR. ERICH: Yes.

THE COURT: Are you telling me that there are circumstances under which you would vote for the death penalty or are you telling me that under no circumstances, whatever the facts, would you never vote for the death penalty?

MR. ERICH: Probably the latter.

THE COURT: Are you so fixed in that view that you couldn't vote for the death penalty under any circumstances?

* * * * * * * *

MR. ERICH: I would say if there was any mitigating factors, I would not look for the death penalty.

THE COURT: Any at all?

MR. ERICH: Any at all.

THE COURT: And even if you find that the aggravating circumstances outweigh the mitigating circumstances?

MR. ERICH: Yes, correct.

THE COURT: If there were no mitigating circumstances, you would, however, vote for the death penalty?

MR. ERICH: Yes.

* * * * * * * *

THE COURT: The law is this. First of all, I indicated to you if I give you the law, you have to follow it?

MR. ERICH: Right.

THE COURT: You understand that, right? You say you can do that?

MR. ERICH: Right.

THE COURT: Now, we're in a more explicit area of that question. If the aggravating factors on a scale qualitatively outweigh the mitigating factors, the penalty is death. Do you understand that?

MR. ERICH: Yes.

THE COURT: Would you nevertheless vote against the death penalty if the aggravating factors, in fact, outweigh the mitigating factors?

MR. ERICH: I would vote against the death penalty if there are any mitigating factors.

THE COURT: Any?

MR. ERICH: Any.

The prosecutor then argued that the juror should have been excused under Witherspoon, because "he used every word except automatically vote against the death penalty." The trial court deferred ruling on Mr. Erich so that the parties could research the case law. When the court did rule, it relied on the modification of Witherspoon in Adams v. Texas, 448 U.S. 38, 100 S.Ct. 2521, 65 L.Ed.2d 581 (1980), the modification adopted explicitly in Witt:

We reach the second level which I think comes into play in Adams which is that whatever side of the proposition a juror is on, the juror is not qualified if those views about capital punishment would lead him or her to ignore the law or to violate the oath to follow the law and apply the facts to the law as is charged by the Judge.

* * * * * * * *

I have gone around and around with that juror and that juror has said if there is a single mitigating circumstance, then I would not impose — I would not vote for the death penalty.

I realize that they're really findings as compared to an actual vote, but that juror seems to be troubled with the concept of capital punishment except in the most limiting circumstances.

Counsel for the State just alluded to what if the converse existed, give me one aggravating factor and I'll vote to put the person to death. I certainly wouldn't want to qualify anyone like that and this is what we're dealing with in the converse.

* * * * * * * *

He just can't follow the law. It's not just one question by me. I went around it, over it and through it and he has to be disqualified.

Defense counsel argues that the trial court's analysis was flawed because the juror simply indicated what the result of his weighing process would have been, not that he could not perform that weighing process: "There are no standards set forth in the statute as to how jurors should weigh aggravating and mitigating factors; all they must do is weigh them. The fact that a juror might have a notion of how those factors should be weighed is not a violation of the law, except that, constitutionally, mitigating factors must be considered." What the juror indicated repeatedly, however, is that if there were any mitigating factors, he could not perform the weighing process. While such a juror would not have been excludable under Witherspoon, since he did indicate that he could return a death sentence if there were no mitigating factors, under the Adams-Witt standard the test is whether his views would have prevented or substantially impaired his ability to perform his duties in accordance with his oath. The juror's responses indicated an inability to do in this case precisely what the law requires, i.e., to perform a weighing process. Since his scruples substantially impaired his ability to do what the capital punishment statute requires, the trial court's decision to exclude him was proper.

Other issues raised by the defendant's other jury selection claims were addressed and decided in State v. Ramseur, supra, 106 N.J. 123. Specifically, defendant moved before trial for implementation of the so-called "Arizona" or "struck" jury system in the exercise of his peremptory challenges. In State v. Ramseur, we held that the defendant's similar constitutional claim for a struck jury was "without merit" and failed "also as a suggested statement of desirable state policy." Our reasoning is equally applicable to defendant's claim. Id. 106 N.J. at 239-43.

Defendant also relies on the defendant's brief in State v. Ramseur to support his claim that New Jersey's process of death qualification of jurors deprives capital defendants of the right to an impartial jury under the federal and state Constitutions. In State v. Ramseur, we held that the death qualification of jurors prior to the guilt phase of a capital trial was constitutional under both the federal and state constitutions and did not offend notions of fundamental fairness. 106 N.J. at 248-54.

IV.

Alleged Trial Errors

A. Evidence of Third-Party Guilt

In order to cast doubt on his own guilt, the defendant attempted to introduce evidence regarding the possible guilt of Kevin Sheehan, a part-time football coach at the victim's high school, who made "obscene" phone calls to a telephone located outside of the cheerleaders' locker room. The trial court conducted voir dire examinations of the Sheehan-related witnesses, including Mr. Sheehan himself, in order to determine the admissibility of the proposed testimony. The court ruled that the proffered testimony was inadmissible except for the testimony of Mrs. Clark and Detective Longo. The court admitted their testimony for the limited purpose of establishing that a car other than the defendant's may have been in the vicinity of the reservoir on the night of the murder. The defendant claims that the trial court erred by refusing to admit the evidence of the possible guilt of Mr. Sheehan. We disagree.

It is well established that a defendant is entitled to prove his innocence by showing that someone else committed the crime. 1A J.H. Wigmore, Evidence §§ 139-141, at 1723-30 (1983) (hereinafter Wigmore). In Chambers v. Mississippi, 410 U.S. 284, 93 S.Ct. 1038, 35 L.Ed.2d 297 (1973), the Supreme Court recognized that an accused has a constitutional right under the due process clause of the fourteenth amendment to offer probative evidence tending to show that a third party committed the crime charged. Accord United States v. DeNoyer, 811 F.2d 436 (8th Cir.1987); United States v. Green, 786 F.2d 247 (7th Cir.1986); United States v. Crenshaw, 698 F.2d 1060 (9th Cir.1983); United States v. Armstrong, 621 F.2d 951 (9th Cir.1980); United States v. Brannon, 616 F.2d 413 (9th Cir.), cert. den. sub nom. Cox v. United States, 447 U.S. 908, 160 S.Ct. 2993, 64 L.Ed.2d 858 (1980); Pettijohn v. Hall, 599 F.2d 476 (1st Cir.1979); United States v. Morgan, 581 F.2d 933 (D.C. Cir.1978); United States v. Robinson, 544 F.2d 110 (2d Cir.1976), cert. den. sub nom. Robinson v. United States, 434 U.S. 1050, 98 S.Ct. 901, 54 L.Ed.2d 803 (1978); Holt v. United States, 342 F.2d 163 (5th Cir.1965). Likewise, this Court and many other state supreme courts recognize that a defendant may introduce evidence of a third party's guilt. State v. Sturdivant, 31 N.J. 165 (1959), cert. den., 362 U.S. 956, 80 S.Ct. 873, 4 L.Ed.2d 873 (1960); People v. Hall, 41 Cal.3d 826, 718 P.2d 99, 226 Cal. Rptr. 112 (1986); State v. Echols, 203 Conn. 385, 524 A.2d 1143 (1987); State v. Cotton, 318 N.C. 663, 351 S.E.2d 277 (1987); accord State v. Denny, 120 Wis.2d 614, 357 N.W.2d 12 (1984). Courts faced with this issue must necessarily articulate a standard by which to determine the "amount of probative force (beyond minimal relevancy) that suffices to require the admission of evidence tending to show that the crime charged was committed by another...." Wigmore, supra, § 139, at 1724-25 n. 2.

In State v. Sturdivant, supra, we set forth the standard governing the admissibility of evidence of third-party guilt:

A defendant of course may seek to prove that another agency produced the death with which he is charged. It would seem in principle to be sufficient if the proof offered has a rational tendency to engender a reasonable doubt with respect to an essential feature of the State's case.... We think it not enough to prove some hostile event and leave its connection with the case to mere conjecture. Somewhere in the total circumstances there must be some thread capable of inducing reasonable men to regard the event as bearing upon the State's case. The question of relevancy ultimately rests in a sound exercise of discretion.

[31 N.J. at 179 (citation omitted).]

Mr. Sturdivant was convicted of the sodomy-murder of his infant stepdaughter. Defendant argued on appeal that the trial court erred in refusing to admit into evidence the proffered testimony of his sister, Mrs. Campbell. She would have testified that seven days before the child's death, she observed her own child insert a can opener into the victim's vagina and rectum. Such testimony, defendant claimed, would show that the Campbell child caused the little girl's death. The evidence at trial established that the deceased and the Campbell child were not together at any time during the intervening week. We concluded that the "trial judge could not find a fair basis for a connection between [that] event and the death." Id. at 180. As a result, we found that the defendant did not meet the standard articulated above for the admissibility of such evidence.

Other courts have adopted a standard similar to that articulated in State v. Sturdivant for the admissibility of evidence of third-party guilt. Trial courts must generally make a relevancy determination, weighing the probative value of the proffered testimony against such countervailing considerations as the capacity of the evidence to mislead or confuse the jury and undue consumption of time, to name just a few. As the court stated in United States v. Armstrong, supra, 621 F.2d 951, "fundamental standards of relevancy, subject to the discretion of the court to exclude cumulative evidence and to insure orderly presentation of a case, require the admission of testimony which tends to prove that a person other than the defendant committed the crime that is charged." Id. at 953; accord United States v. Crenshaw, supra, 698 F.2d at 1066. In People v. Hall, supra, 226 Cal. Rptr. at 116, 718 P.2d at 103, the California Supreme Court recently articulated a test that closely resembles our own test in State v. Sturdivant: "To be admissible, the third party evidence need not show ``substantial proof of a probability' that the third person committed the act; it need only be capable of raising a reasonable doubt of defendant's guilt."

Other courts have established more stringent criteria for the admission of evidence of third-party culpability. In Fortson v. Indiana, 269 Ind. 161, 379 N.E.2d 147, 153 (1978), the Court held that "the mere possibility that some third person did the act is not enough. Evidence tending to incriminate another must be competent and confined to substantive facts which create more than a mere suspicion that such other person committed the particular offense in question." Elsewhere, appellate courts have recognized a trial court's discretion "to exclude such evidence if it is too speculative or conjectural or too disconnected from the facts of the case against the defendant." State v. LeClair, 425 A.2d 182, 187 (Me. 1981). In State v. Denny, supra, 120 Wis.2d 614, 357 N.W.2d 12, the Wisconsin Court of Appeals adopted the "legitimate tendency" test, which it described as follows:

The "legitimate tendency" test asks whether the proffered evidence is so remote in time, place or circumstances that a direct connection cannot be made between the third person and the crime.... as long as motive and opportunity have been shown and as long as there is also some evidence to directly connect a third person to the crime charged which is not remote in time, place or circumstances, the evidence should be admissible. [Id. 357 N.W.2d at 17 (citation omitted).]

The issue of whether the trial court abused its discretion in excluding evidence of third-party guilt is a particularly fact-sensitive one. In cases in which courts have held that the evidence was improperly excluded, some link was demonstrated between the third party and the victim or crime. As we stated in State v. Sturdivant, supra, 31 N.J. at 179, in the total circumstances there was "some thread capable of inducing reasonable men to regard the event as bearing upon the state's case." See also United States v. Green, supra, 786 F.2d 247 (evidence that defendant's co-worker, who physically resembled defendant, was convicted for similar offense improperly excluded on relevancy grounds); United States v. Armstrong, supra, 621 F.2d at 963 (evidence that another man matching the description of the robber had used bait money from the robbery to purchase a car was improperly excluded); Pettijohn v. Hall, supra, 599 F.2d at 480 (evidence that sole eyewitness to the crime, besides the victim, identified another as the criminal is critical to issue of defendant's guilt); United States v. Morgan, supra, 581 F.2d at 936 (evidence that another was selling drugs from the same house was relevant in determining whether defendant was a distributor or purchaser of drugs); United States v. Robinson, supra, 544 F.2d 110 (evidence that person shown on bank surveillance video resembled a person, other than defendant, who was a suspect in other bank robberies in the area was relevant); State v. Echols, supra, 524 A.2d at 1143 (evidence of third-party look-a-like highly relevant where identity at issue and where each victim positively identified defendant as assailant although he was in jail during the time of one assault); People v. Hall, supra, 41 Cal.3d 826, 226 Cal. Rptr. 112, 718 P.2d 99 (evidence of third-party guilt may be relevant to defendant's guilt where witness was at scene of crime and knew details of crime); State v. LeClair, supra, 425 A.2d 182 (evidence of similar fire relevant in arson prosecution); State v. Cotton, supra, 316 N.C. 663, 351 S.E.2d 277 (evidence of similar crimes committed in same manner on same night and near site of crime for which defendant charged was relevant to defendant's guilt).

On the other hand, an examination of cases in which courts have held that third-party evidence was properly excluded, the proffer of evidence did no more than "prove some hostile event and [left] its connection with the case to mere conjecture." State v. Sturdivant, supra, 31 N.J. at 179. There must be some link between the evidence and the victim or the crime. United States v. DeNoyer, supra, 811 F.2d at 440 (evidence to prove other deviate sex offenders were operating in the community was properly excluded as too remote and speculative); State v. Giguere, 184 Conn. 400, 439 A.2d 1040, 1042 (1981) (police testimony that during same period of time there were several assaults of women in the same neighborhood was properly excluded as merely affording a possible ground of possible suspicion against another person); People v. Luigs, 96 Ill. App.3d 700, 52 Ill.Dec. 98, 421 N.E.2d 961, 966 (1981) (evidence that second knife was found near the scene of the crime was properly excluded because record revealed a complete lack of testimony connecting the second knife to the crime); Fortson v. State, supra, 379 N.E.2d at 153-54 (evidence intending to show that two of victim's friends may have committed crime on the basis of their prior joint criminal activity with victim was properly excluded as not being competent and simply "cast suspicion upon others"); People v. Kent, 157 Mich. App. 780, 404 N.W.2d 668, 674 (1987) (evidence that three other fires occurred at approximately same time as the principal fire, without any facts suggesting that someone else set the other fires, was properly excluded as raising a mere suspicion that someone other than defendant set the principal fire); State v. LaRette, 648 S.W.2d 96, 103 (Mo.) (evidence of obscene telephone calls to victim received during the year prior to her slaying was excluded as disconnected and remote acts, having no other effect than to "cast a bare suspicion on another"), cert. den. sub nom Larette v. Missouri, 464 U.S. 908, 104 S.Ct. 262, 78 L.Ed.2d 246 (1983); State v. Denny, supra, 357 N.W.2d at 16-17 (evidence that murder victim may have gotten into trouble with major drug dealer, that victim owed another person $130, that victim may have had an enemy, that person gave victim shotgun in exchange for drugs, and that victim sold shotgun was properly excluded for failure to show motive, opportunity, or a direct connection with the crime).

In the instant case, the trial court concluded that the proffered testimony did not draw a sufficient link between Mr. Sheehan and the victim, thus failing to meet the liberal requirements of State v. Sturdivant, supra, for the admission of third-party-culpability evidence. Our examination of the proffered testimony leads us to the same conclusion.

The first proffer of evidence involved the testimony of Karyn Speak, Amie's childhood friend and fellow cheerleader. Ms. Speak testified on voir dire regarding "obscene" phone calls made to the pay phone located outside the cheerleaders' locker room. Ms. Speak recalled answering the phone on several occasions. On each occasion, the male voice asked for Ms. Beth Thomas, a fellow cheerleader. On none of the occasions did the male caller ask for Ms. Speak or Ms. Hoffman. Ms. Speak testified that she never actually heard the caller utter obscenities over the phone. She also testified that she did not associate the man's voice with anyone, and she did not recognize the voice as that of Mr. Sheehan. Ms. Speak testified that she had known Mr. Sheehan since the time he began working at the school as a football coach. Ms. Speak testified, however, that she never saw Mr. Sheehan observing her or the decedent. She also testified that she did not become aware that Mr. Sheehan was the person responsible for the phone calls until after Ms. Hoffman's death.

The trial court found the proffered testimony to be inadmissible hearsay and without probative value. The court found there was no causal link between the phone calls taken by Ms. Speak and the assertion that Mr. Sheehan was responsible for the murder of Amie Hoffman. According to the court, there was "no causal connection or link between the fact that she took two calls ... in a telephone booth outside the cheerleaders' locker room and the death of Amie Hoffman or her apprehension or kidnapping or anything of that nature."

We agree with the trial court that the proffered testimony regarding the obscene phone calls simply raised a possible ground of suspicion without providing any direct connection with the murder of Ms. Hoffman. See State v. Stump, 254 Iowa 1181, 119 N.W.2d 210 (evidence must raise more than a mere suspicion), cert. den., 375 U.S. 853, 84 S.Ct. 113, 11 L.Ed.2d 80 (1963). As Ms. Speak testified on voir dire, the male caller at no time requested to speak to Ms. Hoffman. In fact, her testimony showed that the calls were directed to another cheerleader altogether. Without such direct evidence of Mr. Sheehan's connection to the actual murder or to Ms. Hoffman such evidence of motive is simply too remote to be admissible.

In State v. LaRette, supra, the Supreme Court of Missouri had occasion to address the use of obscene phone calls as proof of third-party guilt:

The trial court properly excluded defendant's proffered evidence regarding obscene telephone calls received by Mary Fleming during the year prior to her slaying. Disconnected and remote acts, outside the crime itself, cannot be separately proved for the purpose of pointing up someone other than the accused; and evidence which can have no other effect other than to cast a bare suspicion on another, or to raise a conjectural inference as to the commission of the crime by another is not admissible.

[648 S.W.2d at 103]

Thus evidence of obscene phone calls without more is insufficient to cast suspicion on a third party. The admissibility of the obscene phone call evidence is even less compelling in the instant case where the obscene phone calls were not even addressed to the murder victim.

Ms. Speak also responded to questions regarding the unsupervised parties Amie is said to have hosted at her home. She testified that she attended two parties at the Hoffman house and was aware of one other party that she did not attend. Ms. Speak recalled that alcohol was served at both parties. During a lengthy colloquy between defense counsel and the trial court, defense counsel insisted that the unsupervised parties made Ms. Hoffman vulnerable to attack, an easy target to any persons attending her parties. When pushed to draw some connection between this "target theory" and Mr. Sheehan, defense counsel stated that he wished to pursue the theory that Amie and Mr. Sheehan were involved with one another on a consensual basis, until Amie threatened to disclose their relationship to her parents. According to defendant's proposed theory, Mr. Sheehan killed the victim when she made the alleged threat. The trial court concluded that the proffered testimony had nothing to do with the "events which are the subject matter of this case," and excluded the proffered testimony as irrelevant.

The proffered evidence not only was far too remote from the circumstances of Amie's murder to support any connection whatsoever between Mr. Sheehan and the crime, but also it established no link between Mr. Sheehan and Amie Hoffman. No testimony was elicited that Mr. Sheehan attended these parties or that he was a friend of Amie Hoffman. All that was established, as recognized by the trial court, was that he was a part-time coach at the high school attended by Amie and that she was a cheerleader at the school. The theories that the defense wished to pursue based on this aspect of Ms. Speak's testimony were pure conjecture and were nowhere supported by the record. This aspect of the proffered evidence regarding Mr. Sheehan's possible guilt provides the best illustration of evidence that does nothing more than cast mere suspicions on a third party. A defendant may attempt to show that another had the motive to commit a certain crime. However, he may not do so where the proffered evidence of motive simply affords a "possible ground of suspicion against another person." State v. Denny, supra, 357 N.W.2d at 17; see also People v. Mulligan, 193 Colo. 509, 568 P.2d 449, 456 (1977) (proof of motive or opportunity insufficient without proof that third party committed act directly connecting him to crime); State v. Hill, 196 Conn. 667, 495 A.2d 699, 703 (1985) (proof of motive insufficient without other proof connecting third party to offense charged).

The trial court's next Sheehan-related ruling concerned the proffered testimony of Mrs. Mary Clark and Detective Longo. The defense sought to introduce their testimony in order to show that Mrs. Clark heard a small sports car in the vicinity of the reservoir (near her home) at the approximate time of the murder. The defense also wished to present evidence that Mr. Sheehan owned a small sports car. The combined effect of this evidence would be to support the inference that the murderer was Mr. Sheehan rather than the defendant, who drove a larger Chevy. The trial court, adopting a broad view of relevance due to the circumstantial nature of the State's case, ruled that the proffered testimony of Mrs. Clark and Detective Longo was admissible for the purpose of supporting the inference that a car other than the defendant's was at the reservoir at approximately 12:40 a.m. on the morning after the murder.

We agree with the trial court's ruling admitting the testimony of Mrs. Clark and Detective Longo for the limited purpose of showing that a car other than the defendant's was in the vicinity at the time of the murder. In our view, such evidence did in fact have a legitimate tendency to prove that a third person could have committed the crime. In such a context, the proffered evidence cannot be labelled too remote from the circumstances of the crime, particularly when the State's theory was that the victim had been transported by car to the murder site. We likewise agree with the trial court's decision that evidence regarding Mr. Sheehan's possession of a small sports car at the time of the murder did not draw a sufficient link between Mr. Sheehan and the victim and hence was inadmissible.

Because it failed to draw any connection between Amie Hoffman and Mr. Sheehan, the trial court also deemed inadmissible the proffered testimony of Mr. Russell Vanderbush, an Investigator in the Major Crimes Unit of the Morris County Prosecutor's Office. According to Mr. Vanderbush, Mr. Sheehan allegedly told him that between the hours of 6:00 p.m. and 12:00 a.m. on November 23rd, he was busy preparing report cards. He claimed that after completing the cards, he deposited them in the mailbox of a coworker, Ms. Miko, who lived in Maplewood. Mr. Sheehan reportedly told Mr. Vanderbush that on his way to deposit the cards he stopped a Maplewood Police officer in a blue-and-white patrol car to ask for directions to the Miko residence. Mr. Sheehan supposedly told Mr. Vanderbush that after depositing the cards he went to Gruning's, a restaurant located in South Orange. According to Mr. Vanderbush, Mr. Sheehan claimed that he frequently went to Gruning's, where on the day in question he spoke to his friend David Butler. He also allegedly mentioned to Mr. Vanderbush that Ms. Patricia Mills was working at Gruning's on the evening of November 23rd.

Mr. Vanderbush testified that he set out to verify the veracity of Mr. Sheehan's account shortly after the November 29th interview. Mr. Vanderbush allegedly interviewed Ms. Miko, who informed him that she had no contact with Mr. Sheehan on the evening of the 23rd. According to Mr. Vanderbush, Captain Moore of the Maplewood Police Department informed him that his department was not using blue-and-white patrol cars on November 23, 1982, and that none of the Maplewood officers on duty that evening recalled giving directions to anyone. Mr. Vanderbush also testified that he spoke to Ms. Patricia Mills, who told him that she did not work at Gruning's on November 23rd. Mr. Butler also told Mr. Vanderbush that he did not see Mr. Sheehan at Grunings, and that in fact he had not seen him for approximately one year. Mr. Butler worked at Gruning's and mentioned to Mr. Vanderbush that Mr. Sheehan was not a frequent customer.

Mr. Vanderbush provided further testimony regarding information he obtained from Lt. James McLagan of the Randolph Township Police Department. According to Mr. Vanderbush, Lt. McLagan told him that Mr. Sheehan claimed when interviewed by him that he was in the company of Ms. Dorothy Cerbie between the hours of 8:00 p.m. and 12:00 a.m. on the night of the murder. Lt. McLagan allegedly told Mr. Vanderbush that on November 30th he interviewed Ms. Cerbie, who informed him that she was with Mr. Sheehan from 8:00 p.m. to 12:00 a.m. on November 24th and 25th, the day after the murder and the following morning. Ms. Cerbie reportedly told Lt. McLagan that she had no contact with Mr. Sheehan on November 23rd.

Mr. Vanderbush testified that based on Mr. Sheehan's uncorroborated and conflicting alibis, he obtained a search warrant for Mr. Sheehan's car.^[10] Among other things, investigators searched for fiber evidence that would link Mr. Sheehan to the murder. According to Mr. Vanderbush's testimony, investigators discovered nothing of evidential value as a result of the search.

In support of the proffered testimony, the defense argued that Mr. Vanderbush's testimony tended to prove that Mr. Sheehan displayed a consciousness of guilt during the investigation. The defense offered to present further proof of Mr. Sheehan's consciousness of guilt by producing the testimony of defense investigators Charlie Johnson and Walter Root. The defense investigators would testify that when approached in September 1984 in connection with the Hoffman murder, Mr. Sheehan began to sweat and informed the investigators that they would have to contact his attorney. The investigators would testify further that when they told Mr. Sheehan that they simply wanted to know where he had placed the knife that he had told his mother to hide in the garage, Mr. Sheehan "began to sweat profusely and ran off into his house." The defendant argued that Mr. Vanderbush's testimony tended to show that Mr. Sheehan demonstrated consciousness of guilt while under investigation for the murder of Amie Hoffman. This and his uncorroborated and inconsistent alibis, according to defendant, created an inference that he rather than defendant was guilty of Amie's murder.

After hearing the testimony of Mr. Vanderbush, the trial court summed up the problem with the defense's theory:

Listen, I want you to think of two words tonight, Amie Hoffman. Why am I asking you? What have we heard at all linking Mr. Sheehan with Amie Hoffman? I haven't heard a thing linking Mr. Sheehan with Amie Hoffman. That's what I'm concerned about.

Finding no such link, and finding Mr. Vanderbush's testimony to be hearsay within hearsay, the trial court declared the testimony inadmissible. We agree with that ruling.

The defense also sought to introduce evidence of an incriminating statement that Mr. Sheehan allegedly made to Mr. Dale Jones, then First Assistant in Charge of Homicide with the Essex County Public Defender's Office. Defendant's attorney, Mr. Kenny, claimed that he had been told by Mr. Jones that Mr. Sheehan told him (Mr. Jones) that he (Mr. Sheehan) told his mother (Mrs. Sheehan) to "hide the knife in the garage" after being called in for questioning. On voir dire, Mr. Jones testified that he in fact spoke to Mr. Sheehan on November 29, 1982, in his capacity as an attorney, and that he spoke to Mr. Kenny in April 1982 when Mr. Kenny visited Mr. Jones for strategy meetings in his (Mr. Kenny's) preparation for the Koedatich trial, but Mr. Jones denied that he ever told Mr. Kenny anything regarding Mr. Sheehan's alleged statement.

Because of Mr. Jones' denial, Mr. Kenny called himself to the stand. Mr. Kenny's co-counsel conducted the voir dire examination of Mr. Kenny, who testified that he went to see Mr. Jones in April 1984 in preparation for the instant case, at which time Mr. Jones informed him that he was at the Livingston Police Headquarters on the night the police brought Sheehan in for questioning. Mr. Sheehan's mother reportedly called the Public Defender's Office, requesting that an attorney be sent to meet her son, who had just been called in for questioning. According to Mr. Kenny's testimony, Mr. Jones told him that he spoke briefly to Mr. Sheehan at the station, until Mr. Sheehan's attorney arrived, and that Mr. Jones then told him that Mr. Sheehan had informed him that he had instructed his mother via telephone to hide the knife in the garage.

While recognizing that the evidence it sought to introduce was hearsay, the defense argued that Mr. Sheehan's statement to Mr. Jones should be admitted as a declaration against interest. The defense insisted that Mr. Jones's statement to Mr. Kenny should be admitted on similar grounds since it was obviously against his own professional interest to reveal his client's confidences to a third party. According to the defense's position, Mr. Jones's statement to Mr. Kenny was particularly damaging to his own interests because he allegedly disclosed the information to another attorney who would use it to exculpate his own client. The defense claimed that the proffered evidence of Mr. Sheehan's statement to Mr. Jones, combined with the earlier testimony of Mr. Vanderbush, supported the inference that Mr. Sheehan might possibly be responsible for the murder of Amie Hoffman. The trial court considered the defense's position and determined that although the alleged statements of Mr. Sheehan and Mr. Jones could properly be admitted as declarations against interest, the proffered evidence was inadmissible nonetheless because it did not draw a sufficient link between Mr. Sheehan and the victim, thus failing to satisfy the requirements of State v. Sturdivant, supra, 39 N.J. 165.

The defense also conducted a voir dire examination of Mr. Sheehan. On the advice of his attorney, Mr. Sheehan invoked the fifth amendment privilege against self-incrimination in response to virtually every question. Relying on In re Boiardo, 34 N.J. 599 (1961), for authority, the trial court demanded on several occasions to know Mr. Sheehan's basis for invoking the privilege. For example, the trial court refused to recognize the privilege when Mr. Sheehan refused to answer whether he was a coach at the Parsippany Hills High School. Questions in response to which Mr. Sheehan was able successfully to invoke the privilege included: whether he owned a Datsun 280-Z; whether he knew Amie Hoffman; whether he knew about obscene phone calls made outside the cheerleaders' locker room; whether he gave false information to the Livingston Police regarding his whereabouts on the night of the murder; where he was on the night of the murder; whether he ever spoke to Dale Jones; whether he had met with Dale Jones since the night of their alleged interview; and whether he abducted and killed Ms. Hoffman. The trial court concluded that "the fact that a person has a right to plead the privilege against self-incrimination for an offense for which another is on trial, does not give rise to any reasonable inference that the person who invokes such a privilege is guilty of the crime for which another person is charged."

In ruling on the admissibility of the Sheehan-related evidence, the trial court found that the alleged statements of Mr. Sheehan and Mr. Jones could properly be admitted as declarations against interest. Nevertheless, it held that the evidence was inadmissible because it did not draw a sufficient link between Mr. Sheehan and the victim. In Chambers v. Mississippi, supra, 410 U.S. 284, 93 S.Ct. 1038, 35 L.Ed.2d 297, the Court considered the admission of hearsay statements of third parties where those statements had been made against penal interest. In that case, the Court held that the third-party declaration was admissible. There was, however, a direct link between the statement and the crime. Indeed in Chambers, the statement was a confession to the murder. The third party had signed a sworn written confession to the murder, and on three separate occasions orally declared to three different persons that he had committed the murder. Again, in State v. Larsen, 91 Idaho 42, 415 P.2d 685 (1966), the offer of proof consisted of an alleged oral confession made by a third party to three disinterested persons shortly after the body was found. The third party told them that he killed the deceased, and then proceeded to relate certain other details concerning the crime. Although the Court in State v. Larsen recognized that such out-of-court third-party confessions are generally admissible, it found this proffered evidence inadmissible because there was no "other substantial evidence which tend[ed] to show clearly that the declarant [was] in fact the person guilty of the crime for which the accused [was] on trial." 415 P.2d at 692. These cases are illustrative of the type of proof courts have considered in determining the admissibility of third-party statements against penal interest. In these cases, the proffered evidence drew a direct connection between the third party and the commission of the crime. Here, there was no factual basis in the record to support such a finding.

After our review of the record, we are satisfied that the trial court did not abuse its discretion in ruling that the proffered testimony of Mr. Sheehan's possible guilt be excluded. As the trial court stated numerous times throughout the voir dire of the Sheehan-related witnesses:

I don't understand how that could be evidence that he committed the offense or gives rationale — rises to a rational belief that he did it. I mean going to give you probably the most generous standard that I could think of.... How are you going to tie this man up to Amie Hoffman at all?

And in its final ruling on the admissibility of the evidence, the court applied most liberally the State v. Sturdivant standard, "giving the benefit, every favorable benefit to the defendant, [and still could not] find this testimony to be admissible." The trial court concluded:

In any event, I've indicated for the past two days that I've been looking for some evidence that there's a link with Amie Hoffman in this case. There seems to be conjecture, surmise, speculation and suspicion offered — in an attempt to show that Mr. Sheehan either did or may have had something to do with her death — either did or may have had something to do with Amie Hoffman's killing.... There's nothing linking Mr. Sheehan to Amie Hoffman at all.... I think I've made sufficient rulings, conclusions and findings as a basis for my decision not to permit any of this testimony in evidence.... [I] do not find that this testimony could be submitted to a jury to create — to permit the jury to conclude or draw any inference that another person — that a person other than James Koedatich committed this crime or that there is reasonable doubt as to whether a person other than James Koedatich committed this crime.

In sum, we find that the trial court did not abuse its discretion in ruling inadmissible the proffered evidence of Mr. Sheehan's possible guilt.

B. Evidence of Defendant's Self-Inflicted Stab Wounds

Defendant challenges the trial court's admission into evidence of the circumstances surrounding defendant's arrest, i.e., the self-infliction of the stab wounds and Mr. Koedatich's subsequent call to the police. The evidence to which defendant objected involved the testimony of an expert witness regarding the self-inflicted nature of the stab wounds, as well as the testimony of the patrolman who went to the wounded defendant's home shortly after his stabbing. Supra at 241. In admitting this evidence, the court found that the prosecutor "has a right to give some background evidence as to some of the circumstances leading up to [the arrest]. He has an obligation to exclude certain things in that chain of evidence which would be so prejudicial as to deny the Defendant a fair trial. But I cannot put an incident in which the Defendant described a self-inflicted wound in that category; I simply cannot."

Defendant claims that the court should have invoked Evid. Rule. 4 to exclude this evidence as prejudicial. The court's failure to do so, the defendant contends, constituted a denial of due process of law under Pulley v. Harris, 465 U.S. 37, 104 S.Ct. 871, 79 L.Ed.2d 29 (1984). Under Evid. Rule. 4, it is within the discretion of the trial court to exclude evidence if "its probative value is substantially outweighed by the risk that its admission will either (a) necessitate undue consumption of time or (b) create substantial danger of undue prejudice or of confusing the issues or of misleading the jury." See State v. Sands, 76 N.J. 127, 144 (1978). "In making these evaluations, the trial court has been entrusted with a wide latitude of judgment." State v. Balthrop, 92 N.J. 542, 548 (Schreiber, J., dissenting) (citing State v. Sands, supra, 76 N.J. at 144). A trial court's ruling will not be upset unless there has been an abuse of that discretion, i.e., there has been a clear error of judgment. Ibid.

Defendant argues that the evidence surrounding his arrest was prejudicial in that it contained information that could have caused the jury to associate him with the O'Brien murder. Patrolman Dormer testified that defendant reported being pulled over on the road by a vehicle with a flashing blue light on top. According to the defense, this information had the tendency to connect defendant with the O'Brien murder because Ms. O'Brien was reportedly forced off the road by another vehicle. Furthermore, there had been numerous reports following the O'Brien murder of other cars being forced off the road, in some cases by a vehicle with a flashing blue light. Since the vehicle with a falshing light was exclusively associated with the O'Brien murder, defendant argues, admission of such evidence was unduly prejudicial. We disagree.

In admitting the proffered evidence, the trial court was sensitive to the potentially prejudicial effect of any connection to the O'Brien murder: "We do not permit direct evidence on the State's case as to the O'Brien matter. But that does not mean that any sensible link between the reporting of this incident and the ultimate charge against Mr. Koedatich for this offense is obliterated." We believe that the trial court complied with this Court's majority opinion in State v. Balthrop, supra, 92 N.J. at 546. There, we established a rule requiring a trial court to articulate on the record the basis for its finding of "substantial danger of undue prejudice, or the absence thereof, that would accrue to the objecting party if the proffered evidence were introduced." Ibid. The trial court assured defendant that admission of the evidence would not be prejudicial to him because no connection with the O'Brien murder would be permitted. Our review of the record satisfies us that the trial court complied with those assurances. We therefore find that the trial court's judgment in admitting the evidence was sound.

C. Identification Testimony of Diana Bossard

Defendant challenges the trial court's decision to admit the testimony of Diana Bossard, a surprise witness who came forward with identification testimony just prior to the close of the State's case. Ms. Bossard's story was that she had been at the Surprise Store in the Morris County Mall during the second or third week in October 1982 at about 8:45 p.m. While there, she saw an unkempt-looking older man talking in a loud voice to a young Oriental girl with long hair. The man and the young girl were on opposite sides of a clothing rack; the girl was staring straight ahead into the Mall while the man "was trying to impress this girl" by talking of "dirt bikes, motorcycles, racing, something of that nature." The man, she recollected, "had on a tan jacket, vinyl or thick leather, and jeans," and was "tall, had light brown to dirty blond," and sharp features ("his cheekbones were high, like his cheeks were sunken. He was balding around the forehead"). Ms. Bossard identified the girl as Amie, and the man as the defendant. On cross-examination, Ms. Bossard acknowledged having seen pictures of both the victim and defendant in the newspaper, having read about the case in the newspapers, and having only recently come forward. She also admitted that her observation of the man and girl was "a quick look" or "a glance," and stated, on voir dire, that she "would be surprised [if] he had a beard" and that she had learned the previous winter that "he was in prison for murder."

Before the trial court permitted Ms. Bossard to testify, defense counsel argued that her testimony should be inadmissible because (1) she admitted that the observation was a quick glance, and (2) she admitted having been exposed to the "mass hysteria" surrounding the death of Amie Hoffman and to the publicity surrounding defendant's arrest. In fact, defendant argued that but for Ms. Bossard's exposure to the publicity, she would not have identified defendant as the man in the Surprise Store.

The trial court found that there was no improper conduct on behalf of the State in producing the surprise witness. Defendant does not now challenge that finding. The trial court stated that "just because a witness shows up with information late in the proceedings" does not mean "that that witness must automatically be barred." The court then ruled on whether there was anything "suggestible" about Ms. Bossard's testimony:

The second [issue] is whether or not there is anything suggestible about her testimony. It appears to me, by any measure of proof beyond a reasonable doubt, clear and convincing; any lesser burden of proof, that there has been nothing which has been done externally that would be suggestible, that would be — that would make her testimony influenced or changed by outside sources or forces.

* * * * * * * *

Now, that is not anything which approaches suggestibility which would tend to taint any evidence presented as to identification....

It is certainly rare, in the absence of a finding of suggestibility, the violation of a person's constitutional rights, that identification and the value of that evidence should not be submitted to the jury, would not be submitted to the jury for their ultimate evaluation.

Defendant now claims that the admission of this testimony violated his due-process rights because Ms. Bossard's admitted exposure to pretrial publicity — including photographs of the victim and defendant — was so suggestive as to taint her recollection. The standards for assessing whether pretrial encounters with a defendant will taint the testimony of an eyewitness are set forth in Manson v. Brathwaite, 432 U.S. 98, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977):

[R]eliability is the linchpin in determining the admissibility of identification testimony.... The factors to be considered ... include the opportunity of the witness to view the criminal ... the witness' degree of attention, the accuracy of his prior description of the criminal, the level of certainly demonstrated at the confrontation, and the time between the crime and the confrontation. Against these factors is to be weighed the corrupting effect of the suggestive identification itself.

[Id. at 114, 97 S.Ct. at 2253, 53 L.Ed.2d at 154.]

Ms. Bossard's in-court identification of the defendant was not the product of any impermissibly suggestive law-enforcement procedures. Rather, the issue before us is whether the pictures she saw of the defendant in the newspaper tainted her recollection of the incident in the Surprise Store. In other words, could seeing the defendant's picture in the newspaper have caused Ms. Bossard mistakenly to identify defendant as the man she saw interacting with Amie Hoffman at the Surprise Store? An affirmative response to that question would certainly have affected the reliability of Ms. Bossard's testimony. Also of potential harm to her credibility was the fact that she took months to come forward with her identification. We believe, however, that it was reasonable under the totality of the circumstances to commit the issue of Ms. Bossard's credibility to the jury:

We are content to rely upon the good sense and judgment of American juries, for evidence with some element of untrustworthiness is customary grist for the jury mill. Juries are not so susceptible that they cannot measure intelligently the weight of identification testimony that has some questionable feature. [Manson v. Brathwaite, supra, 432 U.S. at 116, 97 S.Ct. at 2254, 53 L.Ed.2d at 155.]

Thus it was proper for the trial court to treat the issue as one of weight rather than admissibility. We find that the trial court did not abuse its discretion in making that evidentiary ruling.

D. Remedy Pursuant to Rule 3:13-3(f)

At the close of defendant's case, the State announced its intention to call, as rebuttal witnesses, two police officers who claimed to have met defendant at his brother's service station. Both officers would testify that defendant was clean shaven during the fall of 1982. Whether defendant had facial hair was a question in dispute throughout the trial. Defense counsel objected to the proffer on various grounds: (1) the defense was once again forced to confront surprise witnesses; (2) the State knew defendant's facial characteristics were in dispute and should have named the officers earlier; and (3) the defense would suddenly be required to call surrebuttal witnesses on short notice. The prosecutor responded by stating that he did not learn of the officers' willingness to testify regarding defendant's facial hair until after the defense case, at which time the officers responded to his request for such information. The prosecutor also argued that he supplied the defense with a report of the officers' testimony the previous night at 5:00 p.m.

The trial court did not disagree with the State's purpose in seeking to admit the officers' testimony. Rather, in its view, the admissibility of the proffered evidence hinged on whether or not the State should have disclosed the names of the witnesses at an earlier date. After hearing the officers' testimony on voir dire, the court ruled that the testimony was admissible:

Now, it seems to me from what I've heard so far that there's ample opportunity for that testimony to be met. There's nobody being put off guard by it, and I don't see why it shouldn't be permitted in the presence of the jury. And I don't see why we can't go on with the case.

MR. KENNY: I, of course, object, but you've heard my arguments before. My position now is, Judge, is how much time am I going to have after this order to rebut their testimony?

THE COURT: Tomorrow morning we're hearing summations. No prejudice at all in terms of surprise or anything that can't be met by the defense, from what I've been able to hear today in this hearing out of the presence of the jury.

Lloyd Henderson, a police officer for Randolph Township, testified that he had been doing business in September 1982 with defendant's brother's service station, and had met defendant there. He described the defendant as clean-shaven. The officer testified that by November defendant's hair was light colored, and that he may have been unshaven. On cross-examination, defense counsel showed the officer defendant's photo driver's license dated November 8, 1982, and asked: "so, therefore, sir, either you're lying, or this picture is not truthful. Isn't that correct?" Defense counsel concluded:

Q: Now, on October 27, 1982, the defendant's hair was lighter than yours, and the defendant did not have a beard or a mustache. Isn't that correct, sir?

A: Yes, sir.

Q: And on November 11th or 12th, in that area, his hair was lighter than yours and he didn't have a beard or a mustache.

A: Yes, sir.

Q: And on November 8, 1982, his license shows a beard and a mustache. Isn't that correct, sir?

A: According to this, yes.

Q: And, his — And it also shows his hair being much darker than yours; as a matter of fact, being the same color as his hair is today. Isn't that correct, sir?

A: It appears that way, yes.

Q: Isn't it a fact, sir, that you're not telling the truth here today about this, because you were asked not to by your Chief McLagan; Isn't that a fact, sir?

A: Absolutely not.

MR. KENNEY: I have no further questions.

Similar testimony and cross-examination was elicited from Gary Gouck, another Randolph Township Police Officer ("His hair was ... lighter than mine, somewhat lighter than mine, cleanshaven").

In conclusion, the defense called three witnesses in surrebuttal, Carol Charron and Robert Monahan, who had worked with defendant at his brother's station that fall, and Nicholas Borzeka, defendant's uncle, who had had work done at defendant's brother's station. Each testified that defendant had been growing a beard in September, had continued growing it through the fall, and had had dark hair. The State elicited from Ms. Charron the admission that she had visited defendant at the jail the night before her testimony. The prosecutor also read to the witness her grand jury testimony that she could not remember if he had had a beard, and that "a couple of times he started growing it." Lastly, the prosecution elicited from Mr. Monahan the fact that he remembered seeing Officers Henderson and Gouck at the station.

Defendant now claims on appeal that this testimony violated his rights under Rule 3:13-3. Rule 3:13-3(f) provides that if a party discovers additional evidence subject to discovery,

he shall promptly notify the other party or his attorney of the existence thereof. If at any time during the course of the proceedings it is brought to the attention of the court that a party has failed to comply ... it may order such party to permit the discovery or inspection of materials ... grant a continuance, or prohibit the party from introducing in evidence the material not disclosed, or it may enter such order as it deems appropriate.

The State urges that the notification of the defense was timely, since the prosecutor had interviewed the officers hours before the defense was notified. The State also points out that Rule 3:13-3(a)(7) requires the prosecutor to disclose the names and addresses of "any persons whom the prosecuting attorney knows to have relevant evidence." (Emphasis added).

In our view, it is not necessary to reach the question of whether the prosecutor was in fact unaware that the two police officers knew defendant as a result of their department's active investigation of the case. Even assuming such a violation, it is difficult to see how defendant was harmed by admission of the officers' testimony. The defense was able to produce all of the surrebuttal witnesses it deemed necessary. Moreover, defense counsel's cross-examination of the officers was vigorous and thorough, a result the trial court anticipated in its ruling by observing that there was "ample opportunity for that testimony to be met." Therefore, even if we were to assume arguendo that the trial court erred in admitting the testimony, we would be compelled to disregard such error as harmless. See State v. Macon, 57 N.J. 325, 337-38 (1971).

E. Prosecutorial Misconduct

Defendant argues that various instances of prosecutorial misconduct during both the guilt and penalty phases warrant reversal of his conviction and sentence. He cites three instances: (1) the prosecutor violated his duty to disclose the names and addresses of potential witnesses; (2) the prosecutor's cross-examination of defendant's step-brother constituted misconduct not cured by the court's instruction; and (3) the prosecutor's summations at the guilt and penalty phases exceeded the scope of the evidence and were inflammatory. Because we have elsewhere addressed the prosecutor's alleged violation of Rule 3:13-3, supra at 319, we limit our present discussion to the prosecutor's alleged misconduct during the cross-examination of the defendant's step-brother and during the State's guilt phase summation.^[11]

1. Cross-examination of Defendant's Step-brother

The defense called David Paul Baldwin, the defendant's step-brother, who attended high school and lived at home in November 1982. Baldwin testified that his brother's car had not been running well in November 1982; that his brother had a beard that fall; that on the night in question the defendant was upstairs playing Atari most of the evening, at least until 9 o'clock; and that the sneakers seized in the search of the house and associated with the defendant by Helen Cato were bought for the defendant as a Christmas gift in December 1982.

During the course of his cross-examination, the prosecutor asked Baldwin if his brother had ever worked in Mendham. As he asked this question, the prosecutor pointed to Mendham on a map introduced into evidence to show the area near the retention tanks in Randolph Township, in an obvious attempt to show an association between defendant and the place where the victim's body was found. He did this in spite of the fact that he knew that Koedatich had not worked in Mendham until after the murder.

The prosecutor went on to elicit testimony that Baldwin himself had been to the Morris County Mall on numerous occasions and that he played video games there. He developed the fact that defendant had a great interest in video games, and then proceeded to ask Baldwin whether the defendant was in the Morris County Mall on the afternoon of the crime, allegedly knowing that that was not the fact. The prosecutor's cross-examination of Mr. Baldwin caused a motion for a mistrial. The trial court ruled on that motion as follows:

THE COURT: ... Nobody is going to have a perfect trial, but they're going to get a fair trial in here, I'm going to make sure of that.

There are four items brought as the motion as the basis for the motion for mistrial.

One of the detectives, when asked a question yesterday, said as soon as he got out — and stopped himself. Then immediately the subject moved on to something else which was totally unrelated to that kind of an answer.

Second, Mr. Rubbinaccio asked today an open-ended question, do you know where he lived. That's a dangerous question because you never know what the witness is going to answer. Do you know where he lived before Thanksgiving. He asked David Baldwin, the brother of the Defendant. Dangerous question, it should not have been asked. But it was cured and then in the context of the way it was dealt with, certainly was cured.

Item No. 3 is asking this young man when, if he knew had his brother worked in Mendham. Something well beyond the scope of the direct, I might add, but more importantly, having nothing to do with the facts of this case that he started working there in December, long after the events which give rise to this indictment.

I'll have to give a curative instruction as to that.

And fourth, the question is, quite frankly, I can say was absolutely outrageous to ask a witness, did he know that his brother was in Morris County Mall on the day of, the time that the victim disappeared in a case where his presence is so critical to be proven beyond a reasonable doubt, with absolutely no factual basis for that question is totally improper.

I'll have to cure it, but I can cure it.

I am satisfied that given the appropriate curative instruction, forcefully, strongly and I hope in a clear and articulate manner, I'm satisfied that there would be no need to grant a mistrial in this case.

The motion is accordingly denied.

Having denied the motion, the court instructed the jury to disregard the questions relating to defendant's employment in Mendham and to defendant's possible presence in the Morris County Mall on November 23rd:

THE COURT: Before the witness is brought back, I have a couple of things to discuss with you concerning some questions that were asked this morning because they should not have been asked this morning. I'll deal with two areas.

One. Mr. Baldwin was asked a question about whether or not he knew if his brother, the Defendant in this case, had worked in Mendham. There was not as far as I can tell, will not be any factual basis in any evidence which has been presented or will be presented to support that question being asked.

It is to be disregarded, nothing to do with the evidence in this case.

The second question asked, was to the effect or the substance of the question was, Mr. Baldwin was asked if he knew that his brother, the Defendant, was in the Morris County Mall on that particular afternoon. That particular afternoon I presume being the afternoon of the 23rd of November. There is not nor will there be as far as any one knows to this date, any basis in facts of that question having been asked. You have not heard or it is expected that you will have any reason to hear or will hear any evidence of the presence that afternoon of Mr. Koedatich in the Morris County Mall playing video games or his presence that afternoon at all for that matter.

The question defendant raises is whether the curative instructions were sufficient. Defendant argues that "[t]o tell the jurors that there was no factual basis to support asking whether the defendant worked in Mendham and ``it is to be disregarded ...' is tantalizing rather than curative and probably highlighted the testimony better than the prosecutor could have on his summation." State v. Farrell, 61 N.J. 99, 107 (1972); cf. State v. Kelly, 97 N.J. 178, 218 (1984) (prosecutorial misconduct is grounds for reversal only if it was so egregious as to deprive defendant of a fair trial); State v. Tirone, 64 N.J. 222, 229 (1974) ("In the context of the summation as a whole, we cannot say that the prosecutor's comments were so inflammatory as to deny defendant a fair trial."); State v. Bucanis, 26 N.J. 45, 56, cert. denied, 357 U.S. 910, 78 S.Ct. 1157, 2 L.Ed.2d 1160 (1958). In State v. Ramseur, supra, 106 N.J. at 320-24, this Court held that a prosecutor's cross-examination of a witness, during which he made "several statements that can be construed as providing his personal opinion about defendant's guilt," was improper. The Court's reasoning was that "[s]tatements such as those ... are improper because they divert the jurors' attention from the facts of the case before them." Id. at 322. Similarly, in this case the prosecutor's inquiries regarding "facts" that he knew had no basis, from which inferences could be drawn that would be wholly untrue, arguably diverted the jurors' attention from the relevant facts of the case in a manner that was necessarily prejudicial.

As we pointed out in State v. Ramseur, supra, 106 N.J. 123, the determination that a prosecutor has been guilty of misconduct "does not end our inquiry. Prosecutorial misconduct is not ground for reversal ... unless the conduct was so egregious that it deprived defendant of a fair trial." Ibid. The Court then set forth the standards for determining whether the misconduct denied defendant a fair trial: "we consider whether defense counsel made a timely and proper objection, whether the remark was withdrawn promptly, and whether the court ordered the remarks stricken from the record and instructed the jury to disregard them. See State v. Bogen, 13 N.J. 137, 141-42 (1953)." State v. Ramseur, supra, 106 N.J. at 322-23. The Court in State v. Ramseur found that the defendant was not deprived of a fair trial; it concluded, however, with the following admonition:

A prosecutor willing to engage in proscribed conduct to obtain a conviction in a capital case betrays his oath.... Because death is a uniquely harsh sanction, the Court of necessity will more readily find prejudice resulting from prosecutorial misconduct in a capital case than in other criminal matters; prosecutors who fail to take seriously their particularly stringent ethical obligations in capital cases thus strongly risk postponing, and even jeopardizing, the enforcement of the law. [Id. at 324.]

The procedure the trial court followed in this case comported with the curative procedures outlined in Ramseur: the defense objected, the objections were sustained, and the court issued curative instructions. We therefore conclude that defendant's right to a fair trial was honored.

2. State's Guilt Phase Summation

The effect of the misconduct that occurred during the cross-examination of defendant's step-brother was exacerbated, defendant argues, by misconduct during the State's guilt phase summation. During his summation, the prosecutor stated the following regarding defendant's arrest: "If he had walked out of there, he would have been given a license to kill." Before defense counsel could object (he had risen and was starting to object), the court dismissed the jury and left the bench. When the jury returned, the court issued a curative instruction, telling the jurors that "such phrases as ``a license to kill' have no part of the language which is appropriate for your consideration in this case.... I'm instructing you at this time to disregard the last comment during the last summation."

Defense counsel moved for a mistrial on the grounds that the prosecutor was insinuating that defendant was a serial killer, responsible for other murders in Morris County. The trial court agreed with that interpretation of the prosecutor's remarks, but denied the motion: "It's the prosecutorial equivalent of soiling the floor of the courtroom.... It was very close to a suggestion, but yet not a suggestion that you were going to let a serial murderer go ... and I frankly, I think that's probably what it was intended to do.... I'm satisfied that, because I took a recess and gave a strong cautionary instruction ... any damage done by that statement was ameliorated...." Defendant submits that this improper conduct was incurable.

The State argues that the prosecutor's comments in summation were proper, noting that prosecutors are entitled to wide latitude in summation. State v. Mayberry, 52 N.J. 413, 437 (1968), cert. den., 393 U.S. 1043, 89 S.Ct. 673, 21 L.Ed.2d 593 (1969). The State also argues that the content of the prosecutor's statements could reasonably be inferred from the evidence. State v. Carter, 91 N.J. 86, 125 (1982). Lastly, the State argues that even if the prosecutor's statements were objectionable, the trial court's "curative charge removed any possibility of prejudice to the defendant." In fact, the State argues, the judge's curative instruction "focused the jury's attention on its obligation to decide this case fairly, impartially, and objectively without bias, prejudice or sympathy."

After a careful review of the record, we are satisfied that the trial court's quick action in immediately leaving the bench and its curative instruction prevented the prosecutor's statements from "substantially prejudic[ing] the defendant's fundamental right to have a jury fairly evaluate the merits of his defense." State v. Bucanis, supra, 26 N.J. at 56. Here, as in the cross-examination context, supra, at 321, there was "a timely and proper objection ... the remark was withdrawn promptly ... and the court ordered the remarks stricken from the record and instructed the jury to disregard them." State v. Ramseur, supra, 106 N.J. at 323. As we stated in State v. Ramseur, however, "the fact that the prosecutor's misconduct in this case cannot be said to have prejudiced defendant in no way excuses it." Ibid. We therefore reiterate our warning to prosecutors in capital cases "that this Court will not hesitate to refer on its own motion possible violations of the special ethical rules governing prosecutors to the appropriate district ethics committee for disciplinary action." Ibid.

V.

Penalty Phase

Because the trial court failed to instruct the jury properly in the sentencing phase, we must reverse the sentence of death and remand for a new sentencing proceeding. Specifically, the trial court erred in requiring that the mitigating factors must outweigh the aggravating factors in order for a sentence less than death to be imposed, and in requiring that the jury must agree unanimously on the existence of a mitigating factor.

A. Weighing Aggravating and Mitigating Factors

In State v. Biegenwald, supra, 106 N.J. at 63, we held "that as a matter of fundamental fairness the jury must find that the aggravating factors outweigh mitigating factors, and that this balance must be found beyond a reasonable doubt." In that case, we held that the trial court's failure properly to instruct the jury necessitated, as a matter of plain error, that defendant's death sentence be vacated and the case remanded for a new sentencing hearing. Id. at 62-69.

In this case, the trial court instructed the jury, in pertinent part, as follows:

The defendant must be sentenced to death if you are satisfied beyond a reasonable doubt that the aggravating factor or factors alleged in this case exist and that they are not outweighed by any mitigating factor.

* * * * * * * *

The defendant will be sentenced to death if you find beyond a reasonable doubt the existence of an aggravating factor or factors and that the mitigating factor does not outweigh the aggravating factor or factors which exist. Put another way, the defendant may be sentenced to death only if you are convinced beyond a reasonable doubt that the aggravating factor or factors either outweigh the mitigating factor or that they are equal to the mitigating factors. (emphasis added.)

Thus the trial court erred by instructing the jury that death could be imposed only if the aggravating factors were "not outweighed by any mitigating factor." We vacate defendant's sentence and remand for resentencing.

B. Jury Unanimity Unnecessary to Find a Mitigating Factor

The trial court instructed the jury that it had to agree unanimously about the existence or non-existence of a mitigating factor.^[12] The verdict sheet likewise indicated that the jury had to be unanimous on the question of the existence of mitigating factors.

In State v. Bey II, 112 N.J. 123 (1988), we held that a jury need not agree unanimously regarding the existence of a mitigating factor. See also Mills v. Maryland, ___ U.S. ___, 108 S.Ct. 1860, 100 L.Ed.2d 384 (1988) (trial court erred in penalty phase charge by requiring jury be unanimous in order to find mitigating factor). For the reasons expressed in Bey II, the trial court's mitigating-factor instruction compels us to vacate the defendant's death penalty and remand the case for a new sentence proceeding. However, we do not find these instructions to have the same effect as the supplemental instructions in State v. Ramseur, supra, 106 N.J. at 304-15, where we found they deprived defendant of a permissible deadlocked jury and, by statute, a verdict other than death. Instead, on remand the trial court should instruct the jury in accordance with State v. Bey II.

C. Must the Jury Hear Evidence of Mitigating Factor

The penalty phase opened with defense counsel attempting to waive the jury during the penalty phase. The prosecutor would not consent, so the trial court, in accordance with N.J.S.A. 2C:11-3c(1), denied the application. Defense counsel then presented to the trial court the following statement, which the defendant wished to sign before the court:

My attorney has explained to me my rights for a motion to a new trial to present evidence of mitigating factors at the death penalty phase in my case concerning the death of Amie Hoffman and also my right to appeal my conviction and the guilty verdict. I have told my attorney that I do not want anything done on my behalf and that I want to be executed within 60 days of my being sentenced to death.

Defendant signed the statement in the trial court's presence. Defense counsel then informed the court that the defendant had instructed him to do absolutely nothing concerning the mitigating or sentencing phase of this case, including making no opening statement, not presenting evidence concerning mitigating factors, and making no summation.

Pursuant to defendant's instructions, defense counsel presented no evidence during the penalty phase proceeding. He made no opening statement to the jury, nor did he offer any summation to the jury. Indeed, defendant even sought to stipulate to the aggravating factors. However, the trial court refused to accept that stipulation and decided that no stipulations would be accepted other than defendant's previous conviction of murder, and the State would have to prove the existence of the alleged aggravating factors. Defendant maintained his position throughout the sentencing phase and presented no mitigating evidence. Indeed, even after he was sentenced, defendant continued to resist further trial proceedings. He urged this Court to vacate the stay of execution that the court issued after the Office of the Public Defender filed a Notice of Appeal on defendant's behalf. We denied his motion. State v. Koedatich, supra, 98 N.J. 553.

Defendant now alleges that his death sentence must be vacated because the jury heard no evidence of mitigating factors during the penalty phase.^[13] Defendant makes two claims with respect to this issue. First, he argues that his waiver of his right to present mitigating evidence was invalid because it did not comply with Rule 3:9-2, which governs the acceptance of guilty pleas. Secondly, he argues that defense counsel's failure to present mitigating evidence, notwithstanding his own express instructions in that regard, constituted ineffective assistance of counsel.

We are unpersuaded by defendant's first contention that the waiver was invalid. Specifically, we disagree with defendant's contention that the procedures employed in the instant case were insufficient to assure voluntariness. Defendant concedes that one may make a voluntary waiver of a constitutional right. However, the waiver must be knowing and voluntary, and whether such a waiver is made depends upon whether the totality of the circumstances supports that conclusion. Johnson v. Zerbst, 304 U.S. 458, 464-65, 58 S.Ct. 1019, 1023, 82 L.Ed. 1461, 1466-67 (1938); Fitzpatrick v. Wainwright, 800 F.2d 1057, 1065 (11th Cir.1986); United States v. Ellison, 798 F.2d 1102, 1108 (7th Cir.1986), cert. den., 479 U.S. 1038, 107 S.Ct. 893, 93 L.Ed.2d 845 (1987); State v. McCombs, 81 N.J. 373, 378 (1979); State v. Johnson, 68 N.J. 349, 353-54 (1975). As the United States Supreme Court has recently stated in the context of a waiver of the right against self-incrimination:

First the relinquishment of the right must have been voluntary in the sense that it was the product of a free and deliberate choice rather than intimidation, coercion, or deception. Second, the waiver must have been made with a full awareness both of the nature of the right being abandoned and the consequences of the decision to abandon it.

[Colorado v. Spring, 479 U.S. 564, 107 S.Ct. 851, 93 L.Ed.2d 954, 965 (1987) (quoting Moran v. Burbine, 475 U.S. 412, 106 S.Ct. 1135, 89 L.Ed.2d 410, 421 (1986)).]

Defendant claims that the waiver was not knowingly or voluntarily made. He analogizes this situation to the entering of a guilty plea, and argues that the court, at a minimum, should establish standards as stringent as those set forth in Rule 3:9-2, which requires completion of a detailed Administrative Office of the Courts form "explaining the nature of the offense, the possible penalties and the rights he or she is waiving when entering a guilty plea." We find defendant's analogy to the guilty-plea context to be imperfect at best, since the trial court rejected the idea of a stipulation and required the State to prove its case during the penalty phase. In any event, our review of the record satisfies us that defendant's decision to forgo the presentation of mitigating evidence during the penalty phase was not the result of coercion. Rather, the record indicates that his waiver of that right was knowing and voluntary. The fact that defendant unequivocally expressed his wishes to the trial court, combined with his later attempt to waive his appeal rights, convinces us that he was aware of the consequences of his actions.

Although we are unpersuaded by defendant's legal reasoning, we find persuasive policy reasons exist for not allowing a defendant in a capital case to execute even a knowing and voluntary waiver of his right to present mitigating evidence during the penalty phase. These policy reasons are based substantially on the State's "interest in a reliable penalty determination." People v. Deere, 41 Cal.3d 353, 710 P.2d 925, 931, 222 Cal. Rptr. 13 (1985); State v. Koedatich, supra, 98 N.J. at 554 (O'Hern, J., concurring in part, dissenting in part); State v. Hightower, 214 N.J. Super. 43, 44 (App.Div. 1986).

In State v. Hightower, the Appellate Division held that a defendant may not prevent his attorneys from presenting mitigating evidence. The Appellate Division stated its reasons for allowing a defense attorney to present mitigating evidence even over his client's express order not to contest the imposition of the death sentence:

Certainly tension exists between the desires of the client as expressed to his lawyer and the constitutional necessity to insure that the ultimate penalty is not extracted in a "wanton and freakish manner." In normal circumstances, the lawyer is required by the Rules of Professional Conduct to "abide by a client's decisions concerning the objectives of representation."

* * * * * * * *

Under our statutory scheme, a jury may impose the death penalty only if the aggravating factors outweigh the mitigating factors beyond a reasonable doubt. If the jury did not hear the evidence allegedly in mitigation, it could have difficulty discharging its statutory, and indeed moral, duty. Our conclusion is reinforced by a recent amendment to the death penalty statute which requires that an appeal must be taken even if defendant does not want to appeal and that our State Supreme Court must review the issue of proportionality of the sentence on defendant's request.

* * * * * * * *

The defendant now says he only wants to appeal the guilt phase and not challenge the penalty phase, and inferentially also not challenge the proportionality of the death penalty as applied to him. But this is no guarantee that he will still think this way if the death penalty is imposed. Defendant's present desire thus may eventually thwart effective proportionality review or require a new trial on the penalty phase. We conclude that the jury should hear all relevant testimony. [Id. at 44-46 (citations omitted).]

See also State v. Koedatich, supra, 98 N.J. at 554 (O'Hern, J., concurring in part, dissenting in part) ("What is required at the capital sentencing stage is an individualized determination on the basis of the character of the individual and the circumstances of the crime. The record before us does not disclose how or whether the jury was informed of the essential information concerning the character of the defendant that should precede the jury's judgment.")

We find the reasoning in State v. Hightower, supra, 214 N.J. Super. 43, and in Justice O'Hern's concurring and dissenting opinion in State v. Koedatich to be persuasive. Essential to our statute is that its application cannot result in death sentences that are "wantonly and ... freakishly imposed." Furman v. Georgia, supra, 408 U.S. at 310, 92 S.Ct. at 2762-2763, 33 L.Ed.2d at 390 (Stewart, J., concurring). In Gregg v. Georgia, supra, 428 U.S. at 188-95, 96 S.Ct. at 2934, 49 L.Ed.2d at 883-87, the Supreme Court held that a sentencing jury procedure that gave the jury individualized information about the defendant and consistent guidelines for applying that information avoided the random and disproportionate imposition of the death penalty and thus was constitutional. Thus, "[w]hat is required at the capital sentencing stage is an individualized determination on the basis of the character of the individual and circumstances of the crime." State v. Koedatich, supra, 98 N.J. at 554 (O'Hern, J., concurring in part, dissenting in part) (citing Eddings v. Oklahoma, 455 U.S. 104, 110-12, 102 S.Ct. 869, 874-75, 71 L.Ed.2d 1, 8-9 (1982)).

These procedures are established not only to protect the interests of the accused, but also to enable a state to enact a constitutional death penalty statute. Under our death penalty statute, before a defendant may be sentenced to death, the jury must find beyond a reasonable doubt that the proffered aggravating factors exist. The jury then must weigh the aggravating factors against the mitigating before it imposes the death penalty; the death sentence is imposed only if the jury finds beyond a reasonable doubt that the aggravating factors outweigh the mitigating factors. A defendant who prevents the presentation of mitigating evidence "withholds from the trier of fact potentially crucial information bearing on the penalty decision no less than if the defendant was himself prevented from introducing such evidence by statute or judicial ruling." People v. Deere, supra, 710 P.2d at 931.

Courts have recognized that the qualitative difference between death and any other penalty gives rise to "a corresponding difference in the need for reliability in the determination that death is the appropriate punishment in a specific case." Woodson v. North Carolina, supra, 428 U.S. 280, 305, 96 S.Ct. 2978, 2991, 49 L.Ed.2d 944, 961; accord State v. Ramseur, supra, 106 N.J. at 326. It is self-evident that the state and its citizens have an overwhelming interest in insuring that there is no mistake in the imposition of the death penalty. Accordingly, we have the constitutional and statutory duty to review every judgment of death. Without any evidence in the record of mitigating factors we are missing a significant portion of the evidence that enables us to determine if the imposition of the death penalty was appropriate. Hence, we would be unable to discharge our constitutional and statutory requirement to review a judgment, and, therefore, we would fail to safeguard the state's interest in insuring the reliability of death-penalty decisions. On this ground alone, there must be a new penalty trial.

We turn nevertheless to the issue of whether defense counsel's failure to present mitigating evidence, despite his or her client's instructions, constitutes "ineffective assistance of counsel" in violation of the sixth amendment to the United States Constitution. The problem arises from defense counsel's untenable position. The Model Rules of Professional Conduct provide that under normal circumstances an attorney is to "abide by a client's decisions concerning the objectives of representation." MRPC 1.2(a). In this context, however, following the client's instructions may result in the client's execution. The defendant argues that such representation can only be characterized as ineffective assistance of counsel. We think not.

The standards for ineffective assistance of counsel were set forth by the Supreme Court in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Under the Strickland two-part test, the court must consider first whether the challenged conduct (in this case, the failure to present mitigating evidence at the penalty trial) was deficient, and, if so, whether "there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. at 694, 104 S.Ct. at 2068, 80 L.Ed.2d at 698. In State v. Fritz, 105 N.J. 42, 58 (1987), we adopted for state constitutional purposes the Strickland standard for ineffective assistance of counsel.

Courts have reached varying conclusions when faced with a claim of ineffective assistance of counsel arising from counsel's accession to a defendant's wish not to present mitigating evidence. In People v. Deere, supra, 41 Cal.3d 353, 710 P.2d 925, 222 Cal. Rptr. 13, the California Supreme Court held that counsel's failure to introduce mitigating evidence — and his endorsement of his client's decision to seek death — constituted ineffective assistance. The court held that "the defense attorney's honest but mistaken belief that he had ``no right whatsoever to infringe upon his [client's] decisions about his own life' operated to deny defendant his right to the effective assistance of counsel." 41 Cal.3d at 364, 710 P.2d at 931-32, 222 Cal. Rptr. at 20. The court endorsed the view, consistent with the Appellate Division's view in State v. Hightower, supra, 214 N.J. Super. at 46, that the lawyer's duty to the client is altered in a capital setting:

[A]t the penalty phase defense counsel has a critical obligation to present a case in mitigation, first, ``to assure reliability in capital sentencing,' and second, ``to attempt to convince the sentencer that, notwithstanding the defendant's guilt, he or she is a person who should not die.' [Id. at 364, 710 P.2d at 931, 222 Cal. Rptr. at 20.]

See also People v. Bloyd, 43 Cal.3d 333, 729 P.2d 802, 821-22, 233 Cal. Rptr. 368, 386 (1987) (following Deere; failure of defense counsel to offer mitigating evidence, because defendant "expressly informed me that he didn't want this kind of evidence to be presented," constitutes ineffective assistance of counsel); People v. Burgener, 41 Cal.3d 505, 714 P.2d 1251, 1274-76, 224 Cal. Rptr. 112, 137 (1986) (following Deere; defendant instructed counsel not to call psychiatrist, parents, former cellmate, and to stipulate to aggravating factors: "The fact that defense counsel deliberately refrained from introducing any evidence in support of a lesser penalty than death, though such evidence was available, in itself required reversal").

There is, however, authority for the contrary proposition. In Autry v. McKaskle, 727 F.2d 358, 361-62 (5th Cir.), cert. den., 465 U.S. 1085, 104 S.Ct. 1458, 79 L.Ed.2d 906 (1984), the Fifth Circuit held that "[b]y no measure [including Strickland] can Autry block his lawyer's effort and later claim the resulting performance was constitutionally deficient.... If Autry knowingly made the choices, Carver was ethically bound to allow Autry's wishes...." See also Felde v. Blackburn, 795 F.2d 400, 401 (5th Cir.1986) (following Autry; attorney followed instructions in requesting that the jury return a sentence of death; case remanded for competency hearing).

We do not believe that the doctrine of "ineffective assistance of counsel" is applicable in this case. In Trimble v. State, 693 S.W.2d 267 (Mo. Ct. App. 1985), the Missouri Court of Appeals pointed out the dilemma:

The difficulty in the case arises from the manner in which the issue is posed. Here, we have no inadvertence or neglect of counsel. Trial counsel was prepared to offer the evidence and argument, and urged his client, movant, to permit that effort. Movant was obdurate. He did not want that course pursued. Even the trial court's warning that movant's chosen course of action would, in all probability, result in the death penalty, failed to persuade him. The issue is thus to determine trial counsel's duty in this unusual situation. The state urges that counsel had no duty to act, as the client has essential control of the case.... [T]he decision was for the client, and the lawyer must accept the client's determination, even though contrary to the lawyer's advice. [Id. at 278-79.]

See also Larette v. State, 703 S.W.2d 37, 40 (Mo. Ct. App. 1985) (where counsel had prepared evidence regarding mitigation, so that failure to introduce it "was a consequence of movant's recalcitrance," counsel's conduct was reasonable under Strickland, and no prejudice was proven since the evidence of guilt was overwhelming).

Applying Strickland-Fritz to this case, we do not find that defense counsel's conduct was anything but reasonable; he was, after all, adhering to ethical canons by following his client's instructions. Application of the Strickland standard to this situation, in short, results in an anomaly: defense counsel's conduct was reasonable, but the result of that conduct was prejudicial. In our view, it is simply the wrong standard to apply. As Justice Broussard of the California Supreme Court so aptly stated in his concurring opinion in People v. Deere:

I hesitate ... to describe this case as one involving "the ineffective assistance of counsel." The constitutional right to the effective assistance of counsel belongs to defendant personally. A man facing the awful alternatives of execution or life imprisonment without possibility of parole could rationally prefer execution, or at least feel that the comparative advantage of life imprisonment was not worth the humiliation and loss of dignity he believes entailed in the presentation of mitigating evidence. Here counsel satisfied himself that his client was making a rational, knowing, and intelligent decision, and then acted in accord with his client's wishes. I do not believe his conduct violated any constitutional right of defendant.

Although counsel in this case fulfilled his obligations to his client, he failed to perform a role assigned to him by the state, that of presenting the mitigating evidence necessary to assure the reliability of the penalty determination. But the fact that the state assigns defense counsel a role which may require him to act contrary to his client's wishes on a matter of such vital importance to the client presents a troubling picture. The defense of a capital case often requires a close and trusting relationship between counsel and client; yet our decision requires counsel to violate that trust, to take a position against his client, and perhaps to present evidence revealed to him in confidence by his client. Trial courts should explore methods of alleviating this conflict.^[14] In some cases it might be desirable for counsel, in addition to presenting mitigating evidence, to inform the jury of defendant's personal position. In other cases, the court might permit the defendant himself to address the jury. Alternatively, the court could call persons with mitigating evidence as its own witnesses, or appoint new counsel to call them, and thereby place on the record the mitigating evidence essential to a careful, balanced penalty determination.

In sum, both the state's need to assure the fairness and reliability of the penalty determination, and defendant's rights to personal choice and dignity, command respect. It is essential that the penalty trial constitute a balanced presentation of aggravating and mitigating evidence, but this goal should be achieved, as far as possible, with respect and accommodation for defendant's personal values and for his relationship with counsel. [41 Cal.3d at 369, 710 P.2d at 934-35, 222 Cal. Rptr. at 23-24 (Broussard, J., concurring).]

Justice Broussard thus concurred "not on the ground of incompetency of counsel, but because no steps were taken to assure a fair and balanced penalty trial." Id. 41 Cal.3d at 369-70, 710 P.2d at 935, 222 Cal. Rptr. at 24.

In sum, it follows that the mitigating factors must be introduced, regardless of the defendant's position. As Justice Broussard stated, trial courts should explore methods of handling this situation that are sensitive to the need to preserve the close and trusting attorney-client relationship. Justice Broussard's proposed solutions merit serious consideration, although we do not adopt or reject any of them. Nor do we foreclose the possibility that cases may arise in which the trial court may have to compel defense counsel to present the mitigating evidence despite a client's instructions to the contrary.

D. Prosecutorial Misconduct: Penalty Phase Summation

Defendant claims that prosecutorial misconduct occurred during the penalty phase summation. In defendant's view, the prosecutor's remarks in summation infected the deliberations with inappropriate considerations regarding the victim and with the prosecutor's personal beliefs about the case. Defendant argues that by telling the jury that "this part of the trial ... is when we think of the victim" and that "[t]his is how I want to know Amie Hoffman, a cheerleader, someone who had a life, who had a future," the prosecutor improperly shifted the focus of deliberations from the background and record of the accused and the particular circumstances of the crime to the character of the victim. See Booth v. Maryland, 482 U.S. ___, 107 S.Ct. 2529, 96 L.Ed.2d 440 (1987) (Maryland statute that provided that Victim Impact Statement be introduced at sentencing phase of capital cases held violative of eighth amendment because it focused on personal characteristics of victim and effect of murder on victim's family, rather than on defendant). Defendant also claims that when the prosecutor remarked to the jury that "if this isn't the right [circumstance in which to vote for death], I don't know if there is one," he impermissibly injected his personal beliefs into the deliberations. This was particularly improper, defendant argues, because there was nothing in the record that would allow jurors to compare this crime and this defendant with other crimes committed by other defendants. According to defendant, the jury was left to rely on the prosecutor's expertise or to believe the prosecutor was either exaggerating or fabricating. The Court cannot presume, defendant maintains, that any juror would believe the prosecutor was exaggerating or fabricating, and thus must find that these comments were prejudicial. See State v. Farrell, supra, 61 N.J. at 106; see also Darden v. Wainwright, supra, 477 U.S. 168, 106 S.Ct. 2464, 91 L.Ed.2d 144 (condemning summation in which prosecutor called defendant "an animal" and otherwise injected personal views); Tucker v. Zant, 724 F.2d 882, 889 (11th Cir.1984).

The State argues in response that the prosecutor's statements were reasonably inferrable from the evidence and were made in order to establish the requisite aggravating factors. However, the State argues that even if this Court finds some prosecutorial misconduct, reversal is not required. This is so, the State argues, because defendant's failure to object during the State's summation imposes on him the burden of proving plain error. In other words, the State would have defendant prove that the prosecutor's summation was clearly capable of producing an unjust result. State v. Hock, 54 N.J. 526, 528 (1969), cert. den., 399 U.S. 930, 90 S.Ct. 2254, 26 L.Ed.2d 797 (1970). The dissent, on the other hand, argues that the alleged prosecutorial misconduct must be examined under the "enhanced standard" of review it has proposed. Post at 344. Application of this new standard to the current allegations of misconduct, the dissent contends, requires reversal of the death penalty. We do not agree.

In the recent case of Darden v. Wainwright, supra, 477 U.S. 168, 106 S.Ct. 2464, 91 L.Ed.2d 144, the United States Supreme Court articulated a test for prosecutorial misconduct similar to the test that we articulated in State v. Bucanis, supra, 26 N.J. 45, namely, "whether the prosecutor's comments ``so infected the trial with unfairness as to make the resulting conviction a denial of due process.'" 477 U.S. at 169, 106 S.Ct. at 2466, 91 L.Ed.2d at 157 (quoting Donnelly v. DeChristoforo, 416 U.S. 637, 94 S.Ct. 1868, 40 L.Ed.2d 431 (1974)). To warrant reversal under our own standard a prosecutor's statements in summation must "substantially prejudice the defendant's fundamental right to have a jury fairly evaluate the merits of his defense." State v. Bucanis, supra, 26 N.J. at 56; accord State v. Biegenwald, supra, 106 N.J. at 40; State v. Ramseur, supra, 106 N.J. at 322; see also State v. Kelly, 97 N.J. 178, 218 (1984) ("While not condoning all aspects of the prosecutor's conduct, we conclude that, in the context of the entire trial, it did not cause defendant to be denied a fair trial"); State v. Tirone, supra, 64 N.J. at 229 ("In the context of the summation as a whole, we cannot say that the prosecutor's comments were so inflammatory as to deny defendant a fair trial"). We do not find that the prosecutor's remarks in this case necessitate reversal under this standard.

E. By His Own Conduct

Defendant did not object to the jury charge nor to the verdict sheet used, but now claims that because the verdict sheet did not include a finding that defendant committed the murder "by his own conduct," the sentence of death must be vacated.

We find little merit in defendant's assertion that the death penalty must be vacated because there was no express jury finding that defendant committed the homicidal act by his own conduct. N.J.S.A. 2C:11-3c provides that only a person convicted of murder under N.J.S.A. 2C:11-3a(1) or (2) and "who committed the homicidal act by his own conduct or ... as an accomplice" shall be death-eligible.

There is no question that the only proof submitted to the jury by the State was that defendant "by his own conduct" committed the murder of Amie Hoffman. There was no allegation that Amie's murder involved an accomplice. Defendant's defense was based on denial and alibi. The State and the defense recognized that this case involved a single murderer. As the trial court reviewed the proposed verdict sheet with counsel, the prosecutor asked if the "by his own conduct" language was going to be added to the verdict sheet. The trial court insisted that to include such language would be redundant. The trial court added:

There's no suggestion in this case that the jury would consider anything other than that this defendant did it or didn't he do it. Not like there's a third person helping along here; any evidence presented like that. So I don't think it's necessary to add language.

Defense counsel did not object to the trial court's failure to include this language nor did he request that this language be added. Given the facts and the trial court's instructions, the jury could not have returned a guilty verdict unless it found that defendant committed the homicidal act "by his own conduct."

In its instructions to the jury, the court preliminarily charged the jury on presumption of innocence, burden of proof, and reasonable doubt. Noting that identification was an issue in the case, the court stressed that

the burden of proof on the State extends to each and every element of each offense charged. This specifically, of course, includes the burden of proving beyond a reasonable doubt the identity of the defendant as the perpetrator of the crimes with which he stands charged. (Emphasis added).

After these preliminary instructions, the court charged the jury on the elements to be proven by the State on each count. With respect to N.J.S.A. 2C:11-3a(1) and (2), the trial court noted that defendant was charged with the purposeful or knowing murder of Amie Hoffman, and thereafter explained the terms "purposely" and "knowingly." In addition, the court informed the jury that the circumstances surrounding the crime could be considered in determining whether defendant acted knowingly or purposely in committing the murder. In summing up the elements of the offense, the trial court emphasized that

the essential determination for you to make in regard to the charge of murder in this case is whether the defendant committed the killing purposely or knowingly as I have defined these terms for you.

What the State says is that the defendant kidnapped Amie Hoffman, had forced sexual intercourse with her and stabbed her to death. In order for you to find the defendant guilty of murder, the State must first establish beyond a reasonable doubt that the killing of Amie Hoffman was committed by the defendant, James Jerold Koedatich, and that the killing was done by him purposely or knowingly as I have defined those terms for you.

If, after a consideration of all the evidence, you are convinced beyond a reasonable doubt that the defendant caused the death of Amie Hoffman either purposely or knowingly, then your verdict should be guilty. (Emphasis added).

The only fair inference to be drawn from the proof adduced at trial was that defendant alone murdered Amie Hoffman. Moreover, the jury was repeatedly instructed that it could find defendant guilty only if it found that he "by his own conduct" murdered Amie Hoffman. It had to find that defendant was the "perpetrator" of the murder in order to convict him. Accordingly, the jury's verdict, by definition, reflected a finding that defendant "by his own conduct" committed the homicidal act of murdering Amie Hoffman. Certainly where there is any question at all on this issue ("by his own conduct"), it must be submitted to the jury explicitly in the guilt phase. Here, there was no question raised in this regard. Furthermore, the court's charge, in view of the evidence and contentions of the State and defense, left no alternative to the jury but to find that either defendant killed Amie by his own conduct, or he did not kill her at all.

We affirm defendant's convictions for murder, kidnapping, and aggravated sexual assault, but reverse the imposition of the death penalty and remand the matter for a new sentencing proceeding.

CLIFFORD, J., dissenting.

I agree with Justice Handler that under any recognized standard of review, the conclusion is inescapable that the prosecutor's egregious conduct resulted in reversible error, given the weakness of the State's case and the implications of our holding in State v. Ramseur, 106 N.J. 123, 324 (1987). Post at 365-370. Twenty-five years ago Justice Brennan declared for this Court:

The weaker the State's case against the defendant, the more, not the less, is it the prosecuting attorney's obligation to stay within bounds, that no man be convicted unjustly. A prosecuting attorney's standing rests upon his reputation for prosecuting fairly, "that guilt shall not escape or innocence suffer," not upon the number of convictions he obtains. In the heat of advocacy he may with propriety employ to the full his talent for forceful expression, but he must confine himself to fair comment upon the facts in evidence.

[State v. Bogen, 13 N.J. 137, 141 (1953).]

Everyone seems to recognize the outrageous nature of the prosecutor's conduct in this case. The majority concludes, however, that the error was harmless, given the prompt curative instructions of the trial court. See ante at 323 and 325-326. I would hasten to acknowledge that the trial court's efforts to corral the prosecutor, to secure this loose cannon in the courtroom, bordered on the heroic. I cannot imagine how the court could have done more in its attempt to fashion some semblance of order out of the chaos left by the prosecutor's mangling of the evidence and by his taking of liberties with the record in his summation to the jury. But here, as in the view of three members of the Court in State v. DiPaglia, 64 N.J. 288 (1974), "the circumstances created by the tactics of the prosecutor did not lend themselves to remedial instruction * * * sufficient to avoid the potential for harm." Id. at 306 (dissenting opinion).

It is worth recalling that even though "the evidence for conviction may be deemed legally sufficient," a defendant is "nonetheless entitled to a trial * * * fair and free from prejudicial error." State v. Jackson, 43 N.J. 148, 156 (1964).

[T]he sound administration of justice dictates that the means as well as the ends be just; where serious trial error has undermined the proceeding, there must be reversal without regard to our own views as to guilt. This is particularly true where, as here, lives are at stake and stricter appellate approaches are warranted. [Ibid. (citations omitted).]

See also State v. Ramseur, supra, 106 N.J. at 322-23 (reversal of a criminal conviction is warranted if prosecutorial misconduct is "so egregious that it deprived defendant of a fair trial").

A word of caution. This Court took note, in State v. Spano, 64 N.J. 566 (1974), of "a plethora of cases in which prosecutors have been admonished," and of the phenomenon of "improper comments * * * fast becoming too prevalent, thereby vitiating defendant's right to a fair trial." Id. at 568. We there announced that "[h]enceforth, an expression of displeasure may not suffice. Further and more severe action may be necessary." Id. at 569. Consistent with that salutary declaration I voted, as a majority of the Court did not, to reverse convictions on the basis of prosecutorial misconduct in State v. DiPaglia, supra, 64 N.J. at 298 (dissenting opinion); State v. Perry, 65 N.J. 45, 55 (1974) (dissenting opinion); and State v. Kenny, 68 N.J. 17, 32 (1975) (concurring opinion).

Here, too, I would vote to reverse based on prosecutorial excess, as so vividly recounted by Justice Handler. I see little hope of avoiding repetition of the deprivation of a fundamental constitutional right to a fair trial if we do no more than "reiterate our warning to prosecutors in capital cases" that dire consequences may flow from their violations of the "special ethical rules governing prosecutors." Ante at 325.

HANDLER, J., dissenting.

The majority upholds James J. Koedatich's capital murder conviction despite strong evidence that the jury's determination was prejudiced by the deluge of pretrial publicity and by the prosecutor's misconduct. In so doing it not only fails to honor its previous professions that "death is different," thus requiring searching appellate review, but also frustrates the settled jurisprudence of this Court safeguarding a defendant's right to a fair trial before an impartial jury. The majority's review of the record is perfunctory and deferential rather than searching; its conclusion that the trial judge's rulings on publicity and venue should be sustained, and that the prosecutorial misconduct does not constitute reversible error, fail to meet the standards for the review of capital prosecutions exemplified by this Court in State v. Bey (I), 112 N.J. 45 (1988).

The Court upholds the trial court's rulings on publicity and venue despite the unprecedented local saturation of prejudicial pre-trial publicity and despite the trial court's application of an erroneous and antiquated standard in denying a change of venue. In the same vein, the majority disregards the stricter standard for adjudging prosecutorial excesses in capital cases, acquiescing in ordinary curative instructions that fail to overcome the effect of outrageous prosecutorial misconduct. The majority fails in the process to appreciate that the prejudicial effect of the publicity and prosecutorial misconduct was exacerbated because of the rather tentative circumstantial evidence of defendant's guilt. Finally, the Court also breaks with the implications of its precedent in State v. Ramseur, 106 N.J. 123 (1987), by requiring that defendant again be subjected to the death penalty despite indisputable evidence that the jury was deadlocked on the existence of mitigating factors effectively depriving defendant of the opportunity for an "acquittal" of the death penalty through a non-unanimous determination, see State v. Bey (II), 112 N.J. 123 (1988), and by indicating that checks on prosecutorial discretion may not be needed.

I dissent from these determinations. The case undeniably is riddled with significant errors and is fraught with prejudice to the defendant. Because I believe that defendant was denied a fair trial before an impartial jury, I would reverse defendant's conviction and sentence. Furthermore, because defendant's death sentence was coerced, I believe that he cannot, consistent with State v. Ramseur, supra, be exposed again to a death sentence on remand.

I.

I have argued, concurring in State v. Bey (I), supra, 112 N.J. at 106-20, that an enhanced standard of appellate review is warranted in capital cases, and that this standard should be defined expressly by this Court. In my view, such a standard entails a two-step analysis. First, as a consequence of the direct appeal posture of capital appeals and the eventual requirement of a proportionality review, this Court should conduct an independent and heightened review of the record below to verify the fact-finding made in the trial court. Second, because of the risk of inconsistent results exemplified by federal death-penalty jurisprudence, this searching review of the record below should be complemented by a clearly defined standard of appellate reversibility that eschews normal standards of review such as "harmless error." I have suggested that where a given error is of constitutional magnitude, it must be shown to have had no effect on the jury's deliberations; other errors must be shown to have had no realistic likelihood of prejudicing the jury's deliberations. See State v. Bey (I), supra, 112 N.J. at 116-17 (Handler, J., concurring). A "principal virtue" of enhanced appellate review, I have argued, "is that ... it requires the appellate court to see the case as a whole. Thus, each error ... is critically evaluated ... both for its individual effect on deliberations and for its effect on the structure of the entire case." Id. at 117. This is nowhere more true than in the context of this case, in which the underlying conviction was based entirely on rather tenuous circumstantial evidence.

An appropriate appellate standard to be used in a case such as this, i.e., a capital case in which the State's proofs are tenuous, was discussed by the Mississippi Supreme Court in Fisher v. State, 481 So.2d 203 (Miss. 1985). The Fisher Court was confronted with a record with striking parallels to that of this case in terms of the circumstantial nature of the proofs and the weakness of the State's case against defendant.

In contrast to the Court in this case, the Mississippi Supreme Court began its review of the record by acknowledging the "principle that on appeal convictions of capital murder and sentences of death will be subjected to heightened scrutiny.... ``What may be harmless error in a case with less at stake becomes reversible error when the penalty is death.'" Id. at 211 (quoting Irving v. State, 361 So.2d 1360, 1363 (Miss. 1978), cert. denied, 441 U.S. 913, 99 S.Ct. 2014, 60 L.Ed.2d 386 (1979)). Such scrutiny, the court observed, "requires that trial errors be considered not necessarily individually but for their cumulative effect." Id. at 211. The court then addressed the application of heightened scrutiny in the context of a case based on circumstantial evidence:

While there may be legitimate differences of opinion as to just when and how "heightened scrutiny" works in death penalty cases, it would seem clear that this approach is most needed and most applicable in cases resting upon circumstantial evidence and where the matter of whether the defendant is guilty at all is by no means free of all doubt. As the facts suggest, this is an appropriate case for heightened scrutiny. [Fisher v. State, supra, 481 So.2d at 211.]

This reasoned approach contrasts sharply with the review undertaken by the majority. Although in New Jersey, we treat circumstantial and direct evidence similarly in terms of the burden of proof on the state, see State v. Fiorello, 36 N.J. 80 (1961), the central concern of the Fisher court was the weakness of the state's evidence, which underscored the need for enhanced appellate review. Thus, to the extent that an enhanced standard of review creates a distinctive criminal jurisprudence applicable to defendants in capital cases, this does not so much argue against an enhanced standard as justify it. See Comment, "Deadly Mistakes: Harmless Error in Capital Sentencing," 54 U.Chi.L.Rev. 740, 747-49 (1987).

Moreover, adoption of such a standard of review will not prove to be unworkable in terms of the trial of cases because it governs only appellate courts. I would emphasize, further, that an express enhanced standard of review for capital murder appeals has been applied in other jurisdictions, without engendering procedural chaos or deleterious consequences and without making capital prosecutions impossible. See, e.g., Fisher v. State, 481 So.2d 203 (Miss. 1985); Weeks v. State, 456 So.2d 395 (Ala.Cr.App. 1983), aff'd 456 So.2d 404 (Ala. 1984), cert. denied, 471 U.S. 1030, 105 S.Ct. 2051, 85 L.Ed.2d 324 (1985); Commonwealth v. Zettlemoyer, 500 Pa. 16, 454 A.2d 937 (1982), cert. denied, 463 U.S. 1236, 104 S.Ct. 31, 77 L.Ed.2d 1452 (1983); State v. Wood, 648 P.2d 71 (Utah), cert. denied, 459 U.S. 988, 103 S.Ct. 341, 74 L.Ed.2d 383 (1982).

The majority's comments concerning the enhanced standard of review are problematic because even under conventional standards of review the guilt conviction here should be reversed. The failure to grant a change of venue, initially because the trial court applied the wrong test for such a motion, is reversible error under the standard established in State v. Williams, 93 N.J. 39 (1983). Further, the prosecutor's egregious misconduct should be reversible error under the dictates of the majority's opinion in State v. Ramseur, supra, 106 N.J. at 320-24. Thus, I am strongly of the view that under a conventional standard of review the errors mentioned herein were sufficiently prejudicial to warrant reversal.

Nevertheless, because this case is a capital murder appeal I firmly believe that this Court should apply an enhanced standard of review as part of the needed effort to establish a consistent standard of review for these unique appeals. I commend the standard proffered in State v. Bey (I), supra (Handler, J., concurring). The force of the principle that "[w]hat may be error in a case with less at stake becomes reversible error when the penalty is death," Irving v. State, supra, 361 So.2d at 1363 (citations omitted), is redoubled in a context where not only is a life at stake, but the proofs used to secure the sentence placing that life in jeopardy are highly circumstantial and tenuous.

That urgency attends this case. As can be inferred from the majority's painstaking account of the facts, the State's case was extremely proof-sensitive. No one saw the defendant kill Amie Hoffman. There were no fingerprints or tire prints to tie the defendant to the scene of the crime. No incidents of the victim — jewelry, purse, wallet — were found on the defendant, in the defendant's car, or in the defendant's home. No microscopic trace of the victim was found in the defendant's car. None of the debris observed and recovered at the scene of the crime, few of the fibers (and only the most common of these), and none of the scrapings taken from the victim's leg, mini-pad, clothing, and footwear matched anything of the defendant. Pubic and other human and animal hairs found on the victim did not match the defendant. The spermatozoa type was inconclusive, and thus could not be matched with defendant's. The State's case was based on contested and inconsistent identifications linking the defendant to the victim, evidence of the circumstances of defendant's arrest (which evidence had more to do with the circumstances of defendant's other murder charge, and thus had the capacity to trigger juror recollection of that charge), and on a general correlation between fibers found on the victim and the common carpet and foam fibers of defendant's car. The State's experts acknowledged that nothing conclusive could be said based on the fiber comparisons, but argued that the combination of correlations resulted in a "strong association." The strength of the evidence against defendant was augmented, regrettably, by repeated insinuations by the prosecutor of facts that were not in evidence, of "facts" that the prosecutor knew were not true and, as the trial court acknowledged, by prosecutorial allusion to the serial killings that had terrorized Morris County and, indirectly, to the defendant's status as the accused in another of the serial killings.

I do not contend that the evidence was insufficient; to the contrary, I believe that a rational, impartial jury, on the factual record below, could have found the defendant guilty. The question on this appeal, however, is whether the jury that deliberated defendant's fate can in good conscience be considered to have been impartial, and whether that jury was influenced by factors that corrupted its verdict. I emphasize the tentative nature of the State's proofs, however, because as the certainty of the proofs diminishes, the likelihood that error will be prejudicial increases correspondingly. This fact should inform our appellate review of this capital conviction.

II.

The murders of Amie Hoffman and Deirdre O'Brien and the subsequent arrest of James Koedatich were attended by massive and intense adverse publicity. The Court itself acknowledges the nature and duration of this publicity. Ante at 265-66. See State v. Williams, supra, 93 N.J. at 50-51 n. 2.

Based on this publicity, defense counsel brought a motion for a change of venue prior to trial. The trial court, in June 1984, denied the motion believing that a change of venue was not required under the standards of State v. Williams, supra. Defense counsel renewed the motion for a change of venue on the first day of jury selection, September 24, 1984. The court denied the motion as "premature." The publicity continued through the first few days of jury selection. On the third day of jury selection, the defendant again renewed his motion for a change of venue; once more, it was denied.

The trial court attempted to counter the adverse publicity by giving potential jurors cautionary instructions as to exposure to the publicity.^[1] Significantly, the court excused for cause any prospective juror who indicated knowledge of the defendant's prior murder conviction or his indictment for the murder of Deirdre O'Brien. Of the fifteen jurors who sat during the trial, only five had not read or heard anything about the defendant; the remaining jurors had read something of the case.^[2]

Defendant raises a host of issues deriving from the court's efforts to manage the effect of this massive pretrial publicity. In particular, defendant alleges that the court's conduct of the voir dire was inadequate, that the court erred in failing to excuse for cause jurors who had read or heard about the case, and that the court erred in excluding two death-scrupled jurors. However, it is the majority's treatment of the two critical issues relating to the prejudicial effect of the pretrial publicity — propriety of the court's denial of the defendant's motion to change venue, and the effect of the post-trial disclosure by several jurors that they had realized during the trial that defendant had been charged with another murder — that require a response.

A.

The Court's decision sustaining the lower court's denial of a change of venue reflects not merely the failure to apply an enhanced standard of review, but also the frustration of the principle underlying our jurisprudence safeguarding jury impartiality: "[p]reservation of the jury's independence from extraneous — even judicial — influences." State v. Corsaro, 107 N.J. 339, 350 (1987). The law respecting motions for a change of venue in capital cases was initially presented in State v. Williams, supra, 93 N.J. 39. It was further expounded in State v. Biegenwald, 106 N.J. 13 (1987). See State v. Bey (I), supra, 112 N.J. 45. The Court observed in Biegenwald that under the old test as set forth in State v. Wise, 19 N.J. 59, 73-74 (1955), few defendants succeeded in obtaining a change of venue. "Accordingly, in 1983 in a capital case we modified the defendant's burden, conferring on trial courts the discretion to change venue where it is ``necessary to overcome the realistic likelihood of prejudice from pretrial publicity.'" 106 N.J. at 33 (quoting State v. Williams, 93 N.J. at 67-68 n. 13). It should be noted at the outset, therefore — and all parties agree — that the trial court erred initially in applying State v. Wise; as this Court held in Biegenwald, the effect of Williams was to "[modify] the defendant's burden" from clear and convincing proof that an impartial jury cannot be obtained to a "realistic likelihood of prejudice from pretrial publicity." 106 N.J. at 33 (quoting State v. Williams, supra, 93 N.J. at 67-68 n. 13). The question before us, therefore, is not whether the defendant could show with clear and convincing proof that an impartial jury could not be obtained, but whether, given the saturation of publicity, the defendant could show that there was a "realistic likelihood" of prejudice.

In assessing whether such a "realistic likelihood" exists, the Biegenwald Court held, courts must first distinguish "between cases in which the trial atmosphere is so corrupted by publicity that prejudice may be presumed ... and cases in which pretrial publicity, while extensive, is less intrusive, making the determinative issue the actual effect of the publicity on the impartiality of the jury panel." Biegenwald, supra, 106 N.J. at 33 (citations omitted). The pretrial publicity in this case meets, I believe, the standard for finding that prejudice should be presumed, and, moreover, that the publicity actually had an effect on the jury's impartiality thus meeting either of the two tests in Biegenwald.

According to the Court, the publicity surrounding the case was intense, but its prejudicial effect dissipated over time and so prejudice cannot be presumed. The defendant was arrested in January 1983, was not indicted until December of 1983, and was not tried until September 24, 1984 — nearly two years after the most intense publicity. Hence, the Court believes that it was confronted with publicity that was not "presumptively prejudicial" and therefore the consequences of the publicity could properly be assessed in terms of its actual effect on juror impartiality. Ante at 273-74. I strongly disagree. Further, the Court concludes that the jury that was impaneled was impartial, thus sustaining the trial court's decision to deny the motion for venue, a decision based in part on the trial court's initial application of the wrong standard. I strongly disagree with the majority's analysis of this pretrial publicity, an analysis that is not, I believe, supported by the record.

1.

As the majority observes, cases in which appellate courts have held that a trial court should have presumed prejudice and granted a change of venue "are relatively rare and arise out of the most extreme circumstances." Ante at 269. It was, of course, in recognition of that fact that this Court in Williams, and supreme courts in states such as California and Mississippi, reformulated the standard for granting changes of venue in terms of the existence of a realistic likelihood of prejudice. The case law from jurisdictions whose standard comports with the standard modified by Williams is, therefore, irrelevant. See, e.g., State v. Jenkins, 508 So.2d 191, 193 (La. App. 3 Cir.), writ denied, 512 So.2d 438 (La. 1987) ("defendant must prove more than the mere knowledge of the public facts surrounding the offense," and so must show "that there exists such prejudice in the collective mind of the community that a fair trial is impossible"); State v. Lanscak, 404 N.W.2d 192, 193 (Iowa App. 1987) (defendant must show a "substantial likelihood that the defendant cannot receive a fair and impartial jury trial in the county"); Slone v. State, 496 N.E.2d 401, 404 (Ind. 1986) (to obtain a change of venue, defendant must adduce "evidence of community bias or prejudice sufficient to convince the trial court that he could not obtain a fair trial in that county;" further, "[i]t is insufficient to establish local prejudice warranting a change of venue unless there is a demonstration that jurors were unable to deliberate fairly") (citations omitted); State v. Bobo, 727 S.W.2d 945, 948 (Tenn.), cert. denied, ___ U.S. ___, 108 S.Ct. 204, 98 L.Ed.2d 155 (1987) (issue of change of venue "addresses itself to the sound discretion of the trial judge; his decision will be respected absent an affirmative and clear abuse of that discretion") (citation omitted).

The relevant case law, as the majority acknowledges, is that from jurisdictions sharing New Jersey's "realistic likelihood" standard. Ante at 270-71. Thus, the majority quotes with approval Maine v. Superior Court, 68 Cal.2d 375, 383, 66 Cal. Rptr. 724, 729, 438 P.2d 372, 377 (1968) (quoting American Bar Association Project on Minimum Standards for Criminal Justice, Standards Relating to Fair Trial and Free Press, § 3.2(c) (1966)), in which the California Supreme Court states that venue should be changed where "there is a reasonable likelihood that in the absence of such relief, a fair trial cannot be had.... A showing of actual prejudice shall not be required." (quoted ante at 271).

The California Court explained the standard as follows: "The phrase ``reasonable likelihood' denotes a lesser standard of proof than ``more probable than not.' ... Further, when the issue is raised before trial, any doubt as to the necessity of removal to another county should be resolved in favor of a venue change." Martinez v. Superior Court of Placer County, 29 Cal.3d 574, 578, 629 P.2d 502, 503-04, 174 Cal. Rptr. 701, 702-03 (1981) (citations omitted) (emphasis added). In determining that a change of venue should have been granted, the California Court has relied on such factors as (1) "[t]he element of sensationalism always present in reporting of events concerning a capital case," id. at 581, 629 P.2d at 505, 174 Cal. Rptr. at 704; (2) the fact that "[i]n a small town, in contrast to a large metropolitan area, a major crime is likely to be embedded in the public consciousness with greater effect and for a longer time," id. at 581-82, 629 P.2d at 506, 174 Cal. Rptr. at 705 (citing Steffen v. Municipal Court, 80 Cal. App.3d 623, 145 Cal. Rptr. 782 (1978)) (ordering change in venue on this ground in county with population of 600,000 — almost 200,000 greater than Morris County's in 1984); and (3) the relative "status of the victim and the accused in the community," by which the court contrasts descriptions of victims as a "popular teenage couple," or "popular athlete," with descriptions of defendants such as "anonymous transients," or "an alleged heroin addict." Id. at 584-85, 629 P.2d at 507-08, 174 Cal. Rptr. at 707 (citations omitted).

The majority cites these factors with approval, then fails to apply them. The "sensationalism always present in reporting of events concerning a capital case" was not only present but heightened in this case by the understandable atmosphere of community terror created by the fact that there were multiple murders and, for months, no arrest; and the additional fact that the suspect who was ultimately arrested had served eleven years in Florida for murder. The publicity of the case was inherently pervasive and sensational. Although Morris County is hardly a "small town," the atmosphere of terror was closer to what one might expect of a smaller community, and thus establishes the California Court's second indication that venue should be changed. Moreover, the relative status of the victim and the accused supports a change of venue. Residents of Morris County were saturated with the contrast between Amie Hoffman, a young, beautiful woman, well-liked in school where she earned high grades and was a cheerleader, and the defendant, essentially a stranger to the community who had returned to live with his mother after having served eleven years in Florida for murder. In short, all of the California factors cited with approval by the majority were present here; prejudice should have been presumed.

An approach similar to California's has been adopted in another state that embraces the "reasonable likelihood" standard, Mississippi. As in New Jersey, Mississippi for years made it extremely difficult to secure a change of venue. Recognizing that its past "discussions of the point have been cryptic, superficial," and that "[i]f an unbiased jury is not impaneled, it does not matter how fair the remainder of the proceedings may be," Fisher v. State, supra, 481 So.2d at 215-16, the Mississippi Supreme Court has, in a series of decisions beginning with Johnson v. State, 476 So.2d 1195 (Miss. 1985), refined its test of when venue should be changed to require that venue be changed "where, under the totality of the circumstances, it appears reasonably likely that, in the absence of such relief, the accused's right to a fair trial may be lost." Fisher, supra, 481 So.2d at 220 (citing Maine v. Superior Court, supra, 68 Cal.2d at 383, 438 P.2d at 377, 66 Cal. Rptr. at 729). The Mississippi Court, pursuant to this new standard, has posited certain nonexclusive factors that serve to establish an irrebutable presumption of prejudice, mandating a change of venue. State v. White, 495 So.2d 1346, 1349 (Miss. 1986). This test, to be applied under a heightened standard of review, includes as a factor the community's "supposition that the appellant was guilty of other crimes." Id.

The most significant Mississippi case applying these elements is Fisher v. State, supra, 481 So.3d 203, a prosecution that bears great factual similarity to this case. Fisher, like this case, involved multiple rape-murders in which the community was terrorized by both the nature of the crimes and the length of time for which the assailant remained at large. Id. at 206-07. Both cases were founded on circumstantial evidence; indeed, the case against Fisher seems, on balance, to be the stronger of the two, as the police found jewelry of the victim in Fisher's truck.

The similarity does not end, however, with the circumstances of the crimes; rather, it extends to the nature and extent of the publicity and to the character of the accused. The Mississippi Supreme Court focused on the extensive pretrial publicity in light of that trial court's denial of a motion for a change of venue, stating, "Our dominant concern is the saturation media coverage — television, radio and newspaper — given the prosecution of Larry Fisher." 481 So.2d at 217. Every factor cited by the court in Fisher as "of great importance" to the venue issue is also present in this case. First, there were multiple crimes by an unknown assailant which generated community awareness and attendant fear that there was a multiple offender on the loose in the area. Second, there was extensive reportage that the defendant was charged not with "a single sex-related murder" but also with another murder. Third, it was reported repeatedly that the defendant, like the defendant in Fisher, had a previous conviction. Finally, similar to the publicity in Fisher, it is undisputed that the details of the evidence against defendant — including facts inadmissible at trial, specifically evidence concerning the O'Brien murder — were reported widely.

The degree of saturation in this case also parallels that of Fisher. Although Morris County's population — 405,000 in 1984 — is much larger than the Lauderdale County's, 80,000, the degree of saturation is almost identical. The in-county circulation of the three largest newspapers serving Morris County — approximately 130,000 — is comparable to the newspaper's circulation in Fisher relative to the size of its county. Morris County is serviced by numerous radio and television stations, as is Lauderdale County. Not surprisingly, therefore, the evidence of the infection of the prospective jurors is very strong in in both cases; while every prospective juror had heard of the Fisher case, approximately 90% of the potential jurors had heard of this case.

The Fisher court expressly discounted, moreover, the jurors' professions of their own impartiality. The court acknowledged, in fact, that "[a]ll twelve of those seated so proclaimed," but the Fisher court, although it assumed that those professions and the trial court's acceptance of them were in good faith, did not rely upon these statements by the jurors. Id. at 220-22. The court insisted that "[t]he saturation pretrial publicity described above, however, suggests that there was and remains substantial doubt that Larry Fisher could then or ever get a fair trial in Lauderdale County." Id. at 221-22. Prejudice, in other words, should have been presumed. The court concluded:

We are convinced that the trial judge scrupulously respected Fisher's right to a fair trial in all matters save one, one without which the rest are rendered meaningless. The facts recited above are substantially convincing to our minds that no jury could be impaneled in Lauderdale County which could bring to Larry Fisher's trial that degree of fairness and impartiality which is his right. Certainly on this record there was ... a reasonable likelihood that, without a change of venue to a county outside the coverage of the Meridian news media, Fisher could not be afforded his constitutionally assured fair trial.... [Id. at 223.]

In my view, Fisher is both indistinguishable and compelling authority for reversal in this case. The standards for granting a change of venue are identical. The type of crime involved is identical. The proofs adduced in both cases are tenuous. The backgrounds of the defendants are similar. The type of publicity involved is identical, and the saturation depth is so close to render any distinction unpersuasive.

The majority relies much too heavily on the passage of time to argue that prejudice should not be presumed. Such an argument was persuasively rejected by the Mississippi Court in Fisher, which pointed out that regardless of the interval of media restraint, "the damage had already been done." Id. at 219. In addition, as the Fisher court noted even skillful voir dire may not be effective in this area because: "Many whose views may be substantially affected by pretrial publicity may not know that they are incapable of sitting as fair and impartial jurors." Id. at 221 (citations omitted). Moreover, given the sensational nature of this coverage and the fairly brief interval between the intense publicity and trial, the majority's finding that this interval was sufficient is not credible. Finally, the evidence that most members of the actual jury were aware of the defendant through the media coverage and the subsequent indication that jurors were aware of the highly damaging and non-admissible evidence concerning the O'Brien murder at the time they were deliberating further undercuts the majority's conclusion that prejudice should not be presumed.

To hold as the majority does today is, in short, to hold that Fisher was wrongly decided. This, in my opinion, is not a tenable position. Writing recently of the salutary changes that have resulted since Mississippi's adoption of the "realistic likelihood" standard, Justice Robertson noted that since Johnson and Fisher adopted the realistic likelihood test, "there have been at least three capital murder trials in changed venues which resulted in not guilty verdicts.... In addition," he added, "we all know what happened to Fisher ... on remand. In Lauderdale County, Fisher had been convicted of capital murder and sentenced to death. On retrial in Rankin County, Fisher was found not guilty." Lutes v. State, 517 So.2d 541, 550-51 (Miss. 1987) (Robertson, J., dissenting).

Under the standards cited approvingly by the majority, the level of pretrial publicity was such that the likelihood of prejudice in this case was "realistic," to say the least. Prejudice should have been presumed. Venue should have been changed.

2.

Having upheld the trial court's refusal, a decision made initially under an erroneous standard, to presume prejudice, the Court then turns to the question of whether there is a showing of "actual effect" of prejudice from the publicity. Relying on assumed parallels with State v. Biegenwald, supra, 106 N.J. 13, and on the strength of the voir dire, and ignoring evidence to the contrary, the Court determines that there was no such prejudicial effect.^[3] Again, I disagree.

First, Biegenwald is readily distinguishable from this case. In this case, only five of the fifteen jurors asserted no prior knowledge of the case; two-thirds of the impanelled jury indicated that they had encountered publicity regarding the case. It simply cannot be said in this case, as it could in Biegenwald,^[4] that "a significant portion of the jury array was relatively unexposed to pretrial publicity" or that the impanelled jurors "indicated that they had encountered little or no publicity regarding the case." Biegenwald, supra, 106 N.J. at 36-37. As in Biegenwald, some members of the jury did recollect knowledge of prejudicial information; unlike in Biegenwald, such jurors never satisfied any court that they could remain impartial despite that knowledge. Indeed, the trial court here excused for cause all jurors with such knowledge. This case, in other words, is sufficiently unlike Biegenwald to justify a presumption of a realistic likelihood of prejudice.

Even if the Court is correct to require a showing of "actual effect," that burden is more than met under the circumstances. As this Court stated in State v. Van Duyne, 43 N.J. 369, 386 (1964), cert. denied, 380 U.S. 987, 85 S.Ct. 1359, 14 L.Ed.2d 279 (1965), "an appellate tribunal ... should determine for itself whether the pretrial newspaper stories are so persuasive and so prejudicial, or the jury's protestation of unaffected impartiality after reading them so unconvincing or doubtful that a new trial should be ordered." See also State v. Biegenwald, supra, 106 N.J. at 36 (purporting to follow Van Duyne). All fifteen jurors, it is true, affirmed as a matter of course that they could decide the case solely on the basis of the evidence presented at trial. The Court, in effect, acquiesces in the trial court's acceptance of such an affirmation as sufficient to overcome any indication of prejudice, relying on "the strength of the voir dire in upholding the ... denial of the change of venue motion." Ante at 281. The Court makes much of the laudable prophylactic measures undertaken by the trial court to assure that the jury remained insulated from the publicity. Id. I agree that the measures undertaken were both necessary and proper; had an impartial jury been obtainable from that community, the trial court's management of the case would have ensured one. The majority ignores completely, however, the fact that the trial court's measures simply did not work. They could not work because the saturation of publicity ran too deep. See Fisher v. State, supra, 481 So.2d at 220-23.

The inefficacy of the court's efforts is clearly indicated by evidence that several jurors recalled in the course of the trial that the defendant was implicated in the O'Brien murder.^[5] It is a peculiar irony of the court's refusal to recognize this "actual effect" that those jurors with knowledge of the O'Brien murder charge — "most of the jurors," according to Juror Fascia — would have been excluded as presumptively prejudiced if they had recalled such knowledge on voir dire, notwithstanding any affirmations of impartiality they might have been willing to make. This Court's disregard of evidence of "actual effect" in assessing the "realistic likelihood of prejudice" is difficult to understand. Its reliance, moreover, on the "strength" of a voir dire that obviously failed even to identify those jurors who were aware of highly prejudicial information — let alone to assess their ability to be impartial — is frankly inexplicable. The Court, instead of independently assessing the gravity of error (see, e.g., State v. Van Duyne, supra), purports to defer to the trial court; then it disregards the trial court's own estimation that this knowledge would presumptively disqualify a venireperson.

The Court's result is also difficult to square with its decision in State v. Bey (I), supra, 112 N.J. 45. In Bey (I), this Court reversed defendant's murder conviction because the trial court failed to ensure that the jury was not exposed to mid-trial publicity respecting defendant's other alleged offenses, including another murder. The Court held that the test for such publicity was, "[i]f the court is satisfied that the published information has the capacity to prejudice the defendant, it should determine if there is a realistic possibility that such information may have reached one or more of the jurors." Id. at 86. If there is such a realistic possibility, a voir dire must be conducted to ascertain whether any juror has been exposed. Id. at 86-87. The Court then held that the midtrial publicity had a "[great] capacity to prejudice a defendant's case" (id. at 90), and that where the publicity had such a strong potential for prejudice, the usual assumptions about jurors following their oaths and adhering to the judge's instructions are not warranted. Id. at 81-83. The evil sought to be redressed in Bey (I), moreover — that jurors might proceed to deliberations having been exposed to prejudicial information, and without assuring the court, in light of that exposure, of their capacity to deliberate impartially — undeniably occurred in this case. The Court's determination that there is no remedy for this is thus inconsistent with its holding in Bey (I).

We have ruled — in non-capital cases — that "where pretrial publicity is not highly prejudicial or inflammatory in nature, a trial court should accept the word of a juror who says ... she can set aside previous knowledge of the facts of the case...." See State v. Singletary, 80 N.J. 55, 64 (1979). This is, however, a capital case. Moreover, unlike Singletary, the publicity in the instant case was highly prejudicial and inflammatory. Where, as here, the pretrial publicity has been both extensive and laden with prejudicial information, the normal presumption that jurors who assert their impartiality should be believed should not apply; in such a situation, a juror who has been exposed to the pretrial publicity should be presumptively disqualified. See State v. Williams, supra, 93 N.J. at 68-69; Fisher v. State, supra. It follows that the court should not have accepted "the word" of jurors that they could disregard that publicity; the publicity was too great.

When the Court does discuss the jurors' knowledge of prejudicial information, it contradicts itself. First, the Court holds that post-verdict investigation of the jury would be improper in this case because there is no allegation of injection of improper facts into the deliberations or a manifestation of racial or religious bigotry. Ante at 288-90. The Court then turns to the article and equivocates, concluding that "[e]ven assuming ... the quoted juror statements appearing in the article to be true ... there still is no basis for concluding that ... Koedatich's involvement in the O'Brien murder had entered the jury deliberations." Ante at 290-91. At a minimum, the fact that several jurors entered deliberations with knowledge that would have justified excusing them for cause indicates that this knowledge was a factor in the deliberations of some jurors.

The Court is willing to conclude that the jurors' post-verdict comments do not indicate an erosion of impartiality. It accepts at face value that Ms. Zaccaro indicated that the other indictment was not discussed until after the verdict, Ms. Herzig insisted that the deliberations had been impartial, and Ms. Fascia stated that the jurors had actually been surprised to learn of the defendant's prior murder conviction. Ante at 287-91. The Court assumes that so long as the jurors did not discuss during deliberations the prejudicial information, their independent knowledge of that information did not affect deliberations. I reject this assumption; who can believe that a juror's knowledge of such information will not color his or her contributions to the deliberations? Furthermore, in a capital case, the ultimate verdict of life or death rests with each individual juror's evaluation of aggravating versus mitigating factors. State v. Bey (II), 112 N.J. 123 (1988). The majority's assumption allows the weighing process of the individual juror with knowledge of prejudicial information to be corrupted by that prejudicial knowledge. In my judgment, the undeniable fact that some jurors deliberated with knowledge of defendant's other murder charge — particularly given the tenuous nature of the proofs — is more than sufficient to taint the jury's verdict.

In State v. Kociolek, 20 N.J. 92 (1955), this Court reversed a death sentence because some of the jurors had learned, during the trial, that the defendant had been indicted for robbery and assault with intent to kill, and had informed the other jurors of the charges during deliberations. Justice Brennan wrote:

It suffices that the intrusion of the Glenn incident into the jurors' deliberations, standing alone, introduced illegal and extraneous evidence fraught with peril for the defendant, an action the more grievous because taken in disregard, actually in defiance, of the explicit instructions of the trial judge. [Id. at 103.]

The facts of this case are as egregious as the facts of Kociolek. As argued by defendant: "[h]ere ... the press linked the two murders, along with a third murder with which the defendant was never charged, and conveyed the theory that one person was responsible for all three."^[6] Ms. Zaccaro's statement, moreover, that she "didn't know that he did it in the beginning but halfway through the trial I realized he was linked to O'Brien," can be read to mean that the connection with O'Brien was the decisive factor in her determination that "he did it;" the impermissible information may have significantly influenced her deliberations. In any event, this is precisely the context in which deference to juror protestations of impartiality and presumptions that jurors follow instructions should no longer obtain. See State v. Bey (I), supra, 112 N.J. at 81-83. In line with our cases safeguarding the integrity of jury deliberations, and with the enhanced standard of review applicable in capital cases, the Court's inquiry should be whether the State can show that the awareness on the part of several jurors during trial of the flagrantly detrimental information that defendant had been charged with another murder had no realistic likelihood of affecting prejudicially their determination of his guilt. Given the entirely circumstantial and relatively tenuous nature of the proofs, I believe that the answer is clear: there was a "realistic likelihood" of prejudice.

B.

The Court's determinations, in short, turn the enhanced standard of review applicable in capital cases not only upside down, but inside out. To the extent that the Court's review is influenced by the fact that this is a capital case, that influence seems to have lessened the scrutiny of the record below; to the extent that the Court's holdings today apply to criminal defendants generally, those holdings have corrupted the principle underlying our jurisprudence safeguarding the impartiality of juries: "[p]reservation of the jury's independence from extraneous — even judicial — influences." State v. Corsaro, supra, 107 N.J. at 350 (and cases cited). See also In re Kozlov, supra, 79 N.J. at 239 (courts must "seek out and expose outside factors impinging upon the jury's freedom of action and thus its impartiality and essential integrity"); State v. Van Duyne, supra, 43 N.J. 369.

As a result, this trial was allowed to occur in a venue that had been terrorized by a series of brutal murders, and saturated by publicity concerning those crimes. This publicity linked the defendant to other crimes and related the circumstances of his prior conviction for murder. The passage of time failed to diminish the extent to which the jury panel was exposed, and the trial court's conduct of the voir dire failed to discover the true extent of that exposure. Several jurors entered deliberations in this case, in which the evidence was entirely circumstantial and tenuous, with knowledge that the defendant had been accused of another murder, similar in method, close in time and near in location to the charge before them. The Court engages in a hypertechnical shell game with these facts; it first rules out the presumption of prejudice from the saturation publicity, it then discounts evidence concerning actual prejudice resulting from the publicity, and, finally, it obviates any further inquiry into jury taint by eliminating both presumed and actual prejudice. This legerdemain cannot disguise the Court's refusal to do what fairness in capital cases demands: to see the case as a whole.^[7]See State v. Bey (I), supra 112 N.J. at 64 (Handler, J., concurring). Legal categories and niceties aside, it is difficult, if not impossible to believe that defendant was not prejudiced, and was tried by an impartial jury, much less that the evidence does not support finding a "realistic likelihood of prejudice." I can only wonder what it will take in future cases, in light of this Court's holding, to hold that the likelihood of prejudice was "realistic."

III.

The erroneous refusal to change the trial venue, resulting in a jury that had known about the facts of the case, compounded the impact of the comments and actions by the prosecutor that are claimed to constitute reversible error. These claimed errors are directly attributable to the calculated misconduct of the prosecutor, misconduct that was recognized by the trial judge who tried unsuccessfully to cure these repeated instances of outrageous behavior. Because these curative instructions could not be sufficient, in light of the circumstances, I dissent from the majority's treatment of these trial incidents.

Prosecutorial misconduct occurred during the cross-examination of defendant's step-brother, David Paul Baldwin, who had been in high school, and was living at home in November 1982. On direct he had testified that defendant's car was not running well in November 1982, that defendant was at home playing Atari games at least until 9:00 p.m. on the night in question, that defendant had a beard at that time, and that the sneakers seized from the house were a Christmas 1982 gift and so could not have been worn by defendant in November 1982. The cross examination was replete with questions for which there was no factual basis and which were clearly beyond the scope of the direct examination.

The prosecutor attempted to suggest, from the fact that the step-brother played video games at the Mall, that the defendant had been at the Mall the afternoon of the murder playing video games; this suggestion was made allegedly "with full knowledge that that was completely untrue." The prosecutor then, while pointing to a map showing the proximity of the retention tanks to Mendham, asked Baldwin if the defendant ever worked in Mendham; he asked this despite the knowledge that the defendant had not worked in Mendham until after the date of the murder. The prosecutor also brought out, over an objection, that defendant had lived in Florida.

There can be no doubt that the prosecutor's misconduct was egregious. The trial court itself acknowledged that these areas of inquiry were flagrantly improper; with respect to the suggestion that defendant may have been in the Mall on the afternoon of the murder, the court stated: "I think it's a dreadful avenue, that's another day that he's in the Mall with no basis in fact...." Nevertheless, the court denied defense counsel's motion for a mistrial and, instead, instructed the jury to disregard the questions relating to the defendant's employment in Mendham, and to the defendant's possible presence in the Morris County Mall on November 23rd.^[8]

The Court now rules that the curative instructions were sufficient to render any error harmless. In State v. Ramseur, supra, 106 N.J. at 320-24, this Court held that a prosecutor's cross-examination of a witness, during which he made "several statements that can be construed as providing his personal opinion about defendant's guilt," was improper. The Court reasoned that "[s]tatements such as those ... are improper because they divert the jurors' attention from the facts of the case before them." Id. at 322. Similarly, in this case the prosecutor's inquiries regarding "facts" that he knew had no basis, from which inferences could be drawn that would be wholly untrue, diverted the jurors' attention from the relevant facts of the case in a manner that was necessarily prejudicial.

The Court in Ramseur found that the defendant was not deprived of a fair trial because curative instructions were given following prompt objections. It concluded, however, with the following admonition:

A prosecutor willing to engage in proscribed conduct to obtain a conviction in a capital case betrays his oath.... Because death is a uniquely harsh sanction, the Court of necessity will more readily find prejudice resulting from prosecutorial misconduct in a capital case than in other criminal matters; prosecutors who fail to take seriously their particularly stringent ethical obligations in capital cases thus strongly risk postponing, and even jeopardizing, the enforcement of the law. [Id. at 324.]

In my opinion, it is not possible to conclude that defendant's right to a fair trial was preserved. Ramseur does not address whether, given the uniqueness of a capital proceeding, a prosecutor's misconduct may be so outrageous, that this Court should "more readily find prejudice" despite the use of curative procedures. However, the clear implication, I believe, of the Court's admonition in Ramseur is its recognition that a prosecutor is obliged to conduct the trial in a fair and proper manner. The Court, while nominally adhering to the expressions of Ramseur, fails to apply this more stringent standard of review.

Here, the prosecutor insinuated, in his cross-examination, that the defendant was familiar with the area of the murder because he worked in Mendham; this the prosecutor knew to be misleading. He also suggested that the defendant may have been at the Mall playing video games on the afternoon of the murder, encouraging the jury to consider a possibility with no basis in fact. These questions were not only outside the scope of direct examination, but also beyond the factual scope of the entire case.

Moreover, the prejudicial effect of this misconduct was exacerbated by misconduct during the prosecutor's summation at the guilt phase of the trial. During his summation at the end of the guilt phase, in discussing defendant's arrest, the prosecutor stated: "If he had walked out of there, he would have been given a license to kill." The trial court issued a curative instruction, telling the jurors that "such phrases as ``a license to kill' have no part of the language which is appropriate for your consideration in this case.... I am instructing you at this time to disregard the last comment during the last summation." Defense counsel moved for a mistrial, on the grounds that the prosecutor was insinuating that the defendant was a serial killer, responsible for other killings in Morris County. The trial court agreed with that interpretation of the prosecutor's remarks: "It's the prosecutorial equivalent of soiling the floor of the courtroom.... It was very close to a suggestion, but yet not a suggestion that you were going to let a serial murderer go ... and I frankly, I think that's probably what it was intended to do...." The court, however, denied the motion: "I'm satisfied that, because I took a recess and gave a strong cautionary instruction ... any damage done by that statement was ameliorated...."

I firmly believe that, under this Court's standard of review for prosecutorial misconduct in a capital case (id. at 324), the prosecutor's highly improper questions during cross-examination, in which he suggested the existence of incriminating facts unsupported by evidence, and his references to other murders, insinuating that defendant had added culpability, require reversal. Under an appropriate enhanced standard of review there was a realistic likelihood, particularly given the weakness of the prosecution's case, that these errors had a prejudicial effect on the jury's deliberations; indeed, the more tenuous are the proofs, the greater are both the incentive for a prosecutor to resort to improper tactics to gain a conviction and the likelihood of a prejudicial effect resulting from those tactics. However, even if the majority were to apply conventional harmless error analysis, this egregious conduct should be reversible error given the weakness of the State's case and the implications of its holding in Ramseur. Finally, the evidence before us indicates that something that occurred during the trial triggered the jurors' recollection of prejudicial information; the State's intentional allusion to that information, in the context of an entirely circumstantial case, may well have been decisive in prejudicing the jury.

This Court should follow its opinion in Ramseur, "more readily find prejudice," and reverse the conviction.

IV.

The Court reverses defendant's death sentence. I agree with this result for the reasons relied on by the Court. I think it important to emphasize, however, that the type of prosecutorial misconduct that occurred during the sentencing phase would itself warrant a reversal.

This misconduct occurred during the penalty phase summation, when the prosecutor made repeated references to the victim.^[9] These remarks constituted misconduct by infecting the deliberations with inappropriate considerations regarding the victim and with the prosecutor's personal beliefs about the case.

In Booth v. Maryland, ___ U.S. ___, 107 S.Ct. 2529, 96 L.Ed.2d 440 (1987), the Supreme Court disapproved, by a 5-to-4 vote, the introduction of a Victim Impact Statement (VIS) during the penalty phase of a capital case. The Victim Impact Statement describes "the effect of the crime on the victim and his family." The Statement in question described "the personal characteristics of the victims" (they were a "close couple," "married for fifty-three years," "loving parents and grandparents"). Id. at ___, n. 3, 107 S.Ct. at 2531 n. 3, 96 L.Ed.2d at 446 n. 3. The Court held that "[w]hen carrying out [its penalty-phase] task the jury is required to focus on the defendant as a ``uniquely individual human bein[g].' ... The focus of a VIS, however, is not on the defendant, but on the character and reputation of the victim and the effect on his family. These factors may be wholly unrelated to the blameworthiness of a particular defendant." Id. at ___, 107 S.Ct. at 2533-34, 96 L.Ed.2d at 449. The Court also noted:

Nor is there any justification for permitting such a decision to turn on the perception that the victim was a sterling member of the community rather than someone of questionable character. This type of information does not provide a "principled way to distinguish [cases] in which the death penalty was imposed, from the many cases in which it was not." Godfrey v. Georgia, 446 U.S. 420, 433, 64 L.Ed.2d 398, 100 S.Ct. 1759 [1767] (1980) (opinion of Stewart, J.). See also Skipper v. South Carolina, 476 U.S. 1, ___, 90 L.Ed.2d 1, 106 S.Ct. 1669 [___] (1986) (Powell, J. concurring in judgment).

We also note that it would be difficult — if not impossible — to provide a fair opportunity to rebut such evidence without shifting the focus of sentencing hearing away from the defendants.... The prospect of a "mini-trial" on the victim's character is more than simply unappealing; it could well distract the sentencing jury from its constitutionally required task — determining whether the death penalty is appropriate in light of the background and record of the accused and the particular circumstances of the crime. We thus reject the contention that the presence or absence of emotional distress of the victim's family, or the victim's personal characteristics, are proper sentencing considerations in a capital case. [Id. at ___, 107 S.Ct. at 2534-35, 96 L.Ed.2d at 450-51.]

By telling the jury that "this part of the trial ... is when we think of the victim ..." and that "[t]his is how I want to know Amie Hoffman, a cheerleader, someone who had a life, who had a future," the prosecutor improperly shifted the focus of deliberations from the "background and record of the accused and the particular circumstances of the crime" to the character of the victim. In my opinion, the thrust of the prosecutor's argument is the same as the thrust of the VIS in Booth. Moreover, as a matter of state constitutional law, we should apply an enhanced standard of review and conclude that there was a realistic likelihood that the prosecutor's remarks did affect the jury's deliberations.

Defendant also claims that by telling the jury "if this isn't the right [circumstance in which to vote for death], I don't know if there is one," the prosecutor committed misconduct by injecting his personal beliefs into the deliberations. The jury was induced to rely on the prosecutor's expertise or to "believe the prosecutor was either exaggerating or fabricating. We do not think any juror can be expected to believe the prosecutor would do so and thus we cannot say the defendant was not prejudiced by the prosecutor's comments as to his private belief." State v. Farrell, 61 N.J. 99, 106 (1982); Tucker v. Zant, 724 F.2d 882, 889 (11th Cir.1984).

Under the enhanced standard of review, the Court must conclude that this form of prosecutorial misconduct taints the sentencing determinations. It must in all fairness lead to the reversal of the death penalty.

V.

Defendant argues that the death penalty is constitutionally infirm because the wide latitude given to prosecutors' decisions as to whether to charge a defendant with capital murder has resulted in an unfettered discretion that leads to an arbitrary and capricious imposition of the death penalty. The majority proposes to follow this Court's recent decision in State v. Ramseur, supra, in which the Court had side-stepped the issue, finding that there was not enough evidence presented and that such an analysis would be more properly made as part of proportionality review. 106 N.J. at 324-31. I dissented from this Court's treatment of prosecutorial discretion in Ramseur and continue to do so. Id. at 404-08. (Handler, J., dissenting). In addition, I dissent strongly from the majority's intimations in this case that only certain discrepancies are of constitutional import. Ante at 255-56.

The lodestar principle identified by modern federal death penalty jurisprudence beginning with Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972), has been the prevention of the standardless imposition of the death penalty. The Supreme Court's apparent unwillingness to address the problem of prosecutorial discretion, that may lead to the arbitrary imposition of the death penalty, is undermined by the Court's acknowledgement of the need for guided sentencing discretion. I believe strongly that this Court should not and cannot legitimately follow the path of analysis taken by the Supreme Court and that some guidance of prosecutorial discretion should be mandated.

The United States Supreme Court, in its post-Furman v. Georgia death-penalty jurisprudence, has endorsed systems of guided sentencing discretion. As Justices Stewart, Powell, and Stevens stated in Gregg v. Georgia, 428 U.S. 153, 199, 96 S.Ct. 2909, 2937, 49 L.Ed.2d 859, 889 (1976) (Stewart, J., concurring):

in order to minimize the risk that the death penalty would be imposed on a capriciously selected group of offenders, the decision to impose it had to be guided by standards so that the sentencing authority would focus on the particularized circumstances of the crime and the defendant.

However, in Gregg v. Georgia, supra, the Court rejected the claim that the possibility of discretion pervasive in the system, from the original decision to charge through executive commutation, rendered the statute's operation inevitably arbitrary. The Court reasoned that "[t]he existence of these discretionary stages is not determinative.... At each of these stages an actor in the criminal justice system makes a decision which may remove a defendant from consideration as a candidate for the death penalty.... Nothing in any of our cases suggests that the decision to afford an individual defendant mercy violates the Constitution." Id. at 199, 96 S.Ct. at 2937, 49 L.Ed.2d at 889.^[10]

In McCleskey v. Kemp, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987), the Supreme Court again rejected the contention that unguided prosecutorial discretion rendered a death-penalty statute unconstitutional. In McCleskey, the Court was presented with extensive data showing that the likelihood of receiving a death sentence in Georgia was strongly influenced by the race of the victim. Noting that the Georgia Supreme Court had found McCleskey's sentence proportionate to other death sentences, and that proportionality review is not constitutionally required where statutory guidance is adequate (id. at ___, 107 S.Ct. at 1774, 95 L.Ed.2d at 288), the Court rejected McClesky's contention that his sentence was excessive. Moreover, the Court opined that prosecutorial discretion was a necessary and entrenched part of the judicial system providing leniency and that it should not be condemned concluding: "[A]bsent a showing that the Georgia capital punishment system operates in an arbitrary and capricious manner, McCleskey cannot prove a constitutional violation by demonstrating that other defendants who may be similarly situated did not receive the death penalty." Id. The Court dismissed the statistical evidence of race-induced discrepancies in sentencing as "an inevitable part of our criminal justice system;" such disparities, when "unexplained," are not to be assumed to be "invidious." Id. at ___, 107 S.Ct. at 1777, 95 L.Ed.2d at 291-92.

Justices Stevens, Blackmun, and Marshall endorsed Justice Brennan's dissenting view that Supreme Court precedents "reflect a realization of the myriad of opportunities for racial considerations to influence criminal proceedings.... The discretion afforded prosecutors and jurors ... creates such opportunities. No guidelines govern prosecutorial decisions to seek the death penalty...." Id. at ___, 107 S.Ct. at 1788, 95 L.Ed.2d at 305 (Brennan, J., dissenting). Justice Blackmun, in particular, reviewed the record evidence regarding prosecutorial discretion and stressed that some standards were necessary, concluding that the defendant was entitled to relief under the Equal Protection Clause of the fourteenth amendment. Id. at ___, 107 S.Ct. 1805, 95 L.Ed.2d at 325 (Blackmun, J., dissenting).

The course of federal jurisprudence should not distract state courts from an independent evaluation of the issue, an evaluation that the majority recognizes is required. Ante at 252-53. In District Attorney v. Watson, 381 Mass. 648, 666-68, 411 N.E.2d 1274, 1284-85 (1980), the availability of unguided prosecutorial discretion led the Supreme Judicial Court of Massachusetts to declare the Massachusetts capital statute unconstitutional under the Massachusetts constitution:

It may be that c. 488 would meet the Federal constitutional requirements, if tested, but here we appraise the statute under ... the State Constitution. We accept the Furman principle that untrammeled discretion in imposing the death penalty is intolerable; we find unacceptable under the State Constitution the premise of the post-Furman cases that statutory guidelines and standards may be entirely curative.... Furman and subsequent cases do not address the discretionary powers exercised at other points in the criminal justice process.... Furman stands indifferent to the exercise of the prosecutor's "untrammeled discretion." For reasons which may be valid in the context of his duties, but which do not assist evenhandedness, the prosecutor ... may forgo a first degree murder indictment and seek an indictment for ... a lesser charge. Also, in a first degree murder case, perhaps pursuant to plea bargaining, the prosecutor may in his uncurbed discretion nol prosse that part of the indictment which charges murder in the first degree. [Id. at 666-68, 411 N.E.2d at 1284-85.]^[11]

Most courts, however, have followed the approach of the Illinois court in People v. Free, 112 Ill.2d 154, 162-63, 97 Ill.Dec. 396, 399, 492 N.E.2d 1269, 1272, cert. denied, 479 U.S. 871, 107 S.Ct. 246, 93 L.Ed.2d 170 (1986) (and cases cited), in which the validity of the prosecutor's exercise of discretion depends on whether a factual basis exists for the charging of aggravating factors. See also Resnover v. State, 460 N.E.2d 922, 929 (Ind.), cert. denied, 469 U.S. 873, 105 S.Ct. 231, 83 L.Ed.2d 160 (1984); State v. Judge, 100 Wash.2d 706, 713, 675 P.2d 219, 223 (1984); State v. Harris, 48 Ohio St.2d 351, 358-59, 359 N.E.2d 67, 71 (1976), judgment vacated on other grounds, 438 U.S. 911, 98 S.Ct. 3148, 57 L.Ed.2d 1155 (1978).

The Illinois court's reasoning is insufficient for this State's statute. As I intimated in dissent in State v. Ramseur, the breadth of the definition of capital murder, and the absence of meaningful narrowing through the consideration of aggravating factors, renders the jury's discretion standardless. The absence of standards for prosecutors in such a scheme redoubles the infirmity of unguided discretion because the flaws that plague the jury's function in the statute — the failure of the guilt-phase definition and the penalty-phase aggravating factors to guide discretion — are replicated in the absence of prosecutorial standards. 106 N.J. at 405-06 (Handler, J., dissenting). Moreover, the lack of guidance with respect to the prosecutor's decision to charge a defendant with capital murder unacceptably increases the danger that the death penalty will be imposed arbitrarily because the needed narrowing function is not provided at this crucial initial stage of a prosecution.

Not surprisingly, given the latitude for decision-making allowed by the statute, arbitrary results are emerging. The Public Defender has compiled an instructive study that indicates that unguided prosecutorial discretion leads to or has contributed to disparity and disproportionate imposition of the death penalty. L. Bienen, N. Weiner, D. Denno, P. Allison, D. Mills, "The Reimposition of Capital Punishment in New Jersey: Interim Report, Parts I and II" (1988). The Court acknowledges the relevance of this study but then fails to give any credit to its conclusions or implications. Ante at 253-58. I do not insist that the Public Defender's study must be accepted in all its aspects and would acknowledge that it is flawed in some respects, chiefly its use of defense counsel to assess the strength of the cases studied. However, the study lends credence to the Public Defender's claims that equally serious crimes are not prosecuted to equal degrees by the varying counties, indicating that differing county policies and budgets, not to mention intense political pressure in given cases, may well contribute to these disparities. It may be that the data are not strongly indicative of different treatment related to the races of the victim and/or defendant but the statistics tentatively show that racial influence may be present. The question for this Court, I believe, is whether these disparate results are indicative of the arbitrary consequences of unfettered prosecutorial discretion. Minimally, it must give the Court pause; realistically, it calls for further study and for the imposition of a mandatory system to guide prosecutors' judgments as to whether to charge on capital murder.

The majority dismisses the importance of the discrepancies suggested by the Public Defender's study arguing that the data are not indicative of any racial bias in charging capital murder. However, the majority leaves open this question stating that it would find racial bias to be intolerably arbitrary. I am troubled by the majority's insistence that the Study must prove bias because I believe that the Study's preliminary evidence is sufficiently strong to warrant a showing by the State that no bias in charging exists.

I am at a loss, however, to explain how the majority can possibly conclude that county by county discrepancies are not indicative of arbitrary imposition of the death penalty. Ante at 255-57. The majority recognizes the evidence supporting the defendant's claim of such a disparity but then conclusively dismisses this evidence, stating:

That there are differences among the counties in the likelihood that a prosecutor will pursue — or the coincidence that the prosecutor has more often pursued — a capital prosecution does not, standing alone, demonstrate that the death penalty is being arbitrarily imposed. Surely, there are a myriad of reasons why a prosecutor handles different cases differently, such as the willingness of a defendant to plead guilty, the strength of the State's case, a defendant's cooperation in the State's case against a co-defendant, the relative weight of the statutory aggravating and mitigating factors, the availability and relative credibility and persuasiveness of witnesses, and the resources of the county prosecutor's office, to list only a few. [Ibid.]

The majority's holding that county by county discrepancies are of no constitutional significance stands in contradiction to Ramseur in which the Court held that the universe for assuring proportionality was statewide, stating:

We believe that statewide uniformity is the more appropriate measure, and therefore anticipate that comparisons will be made to "similar" cases throughout the State. [106 N.J. at 329.]

The standard of statewide uniformity is now being emptied of content by the majority's determination that county to county discrepancies do not indicate the possible arbitrary and capricious imposition of the death penalty. I do not see how this can be squared with Ramseur.

The county by county discrepancies, contrary to the majority's view, are indicative that defendants in different areas of the state are likely being treated differently. To avoid this potential for the arbitrary imposition of the death penalty, the majority should require the development of a statewide standard for deciding whether to charge capital murder. The majority's suggestion that it strongly recommends that guidelines be developed by the State Attorney General, county prosecutors, and the Public Defender does not, in light of its reluctance to find any constitutional difficulties in the data presented by the Study, suffice to ensure that such standards will be promptly developed. Such standards will lessen the likelihood of arbitrary death penalty prosecutions, and, as the majority notes, promote uniformity and assist in the Court's eventual proportionality review. Ante at 258-59. I would require the development of these standards.

Because the majority fails to require such a limit on prosecutorial discretion and because the death penalty statute itself does not adequately narrow the class of death-eligible persons, I continue to hold the view that the capital murder statute is constitutionally invalid because it does not overcome the genuine risk of arbitrary and capricious application. 106 N.J. at 405-06. The indications suggested by the data are that such arbitrary results may be occurring and accordingly I strongly disagree with the majority's reluctance to face the implications of this data and to adequately guide the discretion given to county prosecutors.

VI.

A final issue remains, an issue neither raised nor addressed by either party, but one that warrants this Court's attention under an enhanced standard of appellate review. See ante at 344-45. This Court has decided that if any one juror believes that a mitigating factor exists, that juror is entitled to carry his belief through to the weighing process. See State v. Bey (II), supra, 112 N.J. 123. The reasoning behind this rule is that if a juror believes that a mitigating factor exists, he might well determine that the aggravating factor(s) do not outweigh the mitigating factor; since a non-unanimous verdict is a verdict (of 30 years to life) in a penalty-phase proceeding, any individual juror who decides that a mitigating factor exists and is not outweighed beyond a reasonable doubt by the aggravating factors has, in effect, spared a defendant's life. On the other hand, if that individual juror believes that the mitigating factor exists and would find, if he or she weighed them, that the mitigating factor was not outweighed, but is unable to perform the weighing process because unanimity is required to find mitigating factors, that juror is disenfranchised.

It is clear in this case that the trial court erred in requiring unanimity in order to find that a mitigating factor exists. The Court, therefore, correctly reverses defendant's death sentence. Ante at 326-27. I submit, however, that the error runs much deeper, and that the dispositive principle of State v. Ramseur requires that this defendant not face a possible death sentence on retrial. I therefore disagree with the Court, which rules that defendant may again be exposed to the death sentence. Ante at 327.

The trial court, as noted, instructed the jury that their findings had to be unanimous. This instruction was reflected in the following interrogatory on the special verdict form: "Do you unanimously agree that the following exists as a mitigating factor?" In attempting to clarify for the jury, during deliberations, whether a "no" answer meant "no, we are not unanimous" or "we are unanimous in saying ``no,'" the court contradicted itself. The court told Juror Number 5: "Anything that you have to decide, you have to decide by unanimous vote. If you can't reach a unanimous vote, you have not reached a decision on the issue." The court thus implied that all had to agree one way or another. When Juror Number 15 voiced confusion over that standard, the court read the question aloud, then said: "If the answer is ``no,' it means that you can't reach a unanimous vote that it exists. Okay?" The court thus instructed the jury first that they had to reach a unanimous verdict to say "no," and then that they had to say "no" if they could not reach a unanimous verdict.

In addition, it was evident that there were jurors who thought that the mitigating factor existed. When the jury returned, the court stated: "Everybody agrees there's hopeless deadlock on that last [mitigating factor] question, is that correct? THE FOREMAN: Yes."^[12]

In State v. Ramseur, supra, 106 N.J. at 312-15, this Court held that because of coercive supplemental jury charges, which broke the deadlock that could have saved the defendant's life, the defendant would not face a possible death sentence at resentencing. The Court stated:

We hold that where a trial court in a capital case has erroneously given coercive supplemental instructions ... to a jury that has expressed its inability to agree, the law must afford defendant the benefit of the final non-unanimous verdict that might have been returned absent the coercion. Having erroneously been deprived of a substantial opportunity to a jury verdict resulting in imprisonment rather than death, the defendant may not be subject to another capital sentencing proceeding.

We recognize that any reversible error in a capital case may be said in some sense to have deprived a defendant of the opportunity to a jury verdict resulting in imprisonment, and that nevertheless the usual and proper remedy for such errors is reversal of the death sentence and a retrial of the sentencing proceeding in which the defendant may again face the death penalty.... But a Czachor error is critically different from other prejudicial errors in that by definition it occurs after the jury has clearly demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict for the death sentence. The erroneous coercive charge has, not simply as a possible consequence but as its sole purpose, ending the disagreement that would save defendant's life. Even if the jury's disagreement has not reached the point at which further deliberations would be improper, those deliberations must take place in an atmosphere free of coercion. If such coercion occurs, the defendant has irrevocably lost not merely a theoretical possibility but a substantial likelihood that, absent the error, the jury would have reached a verdict resulting in imprisonment rather than death. [Id. at 313-14 (emphasis added).]

The Ramseur Court thus took pains to emphasize the nature of the coercive supplemental charge; the critical element, the Court held, is that "by definition it occurs after the jury has clearly demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict for the death sentence." Id. at 313.

The primary vice of the coercive supplemental charge, then, is that it compels jurors who believe that a mitigating factor is not outweighed beyond a reasonable doubt by aggravating factors into abandoning that position for the sake of unanimity. That was precisely the effect of the trial court's charge in this case; any juror who believed that the mitigating factor existed was coerced into abandoning that belief prior to the weighing process by the requirement of unanimity. Because all agreed, moreover, that two aggravating factors existed, the inability to find a mitigating factor unanimously meant that the verdict of death was automatic. Thus, any jurors who believed that the mitigating factor existed were literally coerced into voting for death; indeed, the coercion was so powerful that their belief one way or another about the mitigating factor's weight was not even allowed to be registered. In Ramseur, by contrast, the jurors reached the weighing process, at least, before the coercion forced jurors to abandon their beliefs.

As in Ramseur, moreover, the jury in this case demonstrated an inability or unwillingness to bring in an uncoerced unanimous verdict. There was obvious resistance by the jury to abandoning the position that the mitigating factor existed; the jury sought clarification of the need for unanimity, and ended deliberations deadlocked on the existence of the mitigating factor. This, coupled with the facts that no mitigating evidence was presented, and that the jury also deadlocked on the existence of the only aggravating factors that called for substantive judgment, raises "a substantial likelihood that, absent the error, the jury would have reached a verdict of imprisonment rather than death." 106 N.J. at 314. Since "[t]he remedy must be commensurate with the wrong" (id. at 313), and since the wrong here is indistinguishable in principle from the wrong in Ramseur, the remedy should be the same; defendant should not face another death sentence.

I note, finally, that the Ramseur Court buttressed its holding on the coercive charge by reference to other errors, e.g., the jury was not properly apprised of its responsibility. Id. at 315. In this case, the coercion of the jurors culminated a trial that featured blatant prosecutorial misconduct and infection of the jury with prejudicial publicity, among other infirmities, and in which the court erroneously acquiesced in the defendant's desire not to present any mitigating evidence whatsoever. Under the enhanced standard of appellate review applicable in capital cases, these facts only strengthen, in my view, the case for not allowing this defendant to face another death sentence.

VII.

I state, in conclusion, that the unconstitutionality of the capital murder-death penalty statute is reconfirmed by this case. It is not, I suggest, simply judicial inexperience that has produced incongruous and inconsistent results. The statute, lacking cohesive and understandable standards, is insusceptible of fair and impartial management. The absence of guided prosecutorial discretion in this statutory framework, coupled with no current means of determining the proportionality of sentences, inflicts constitutional wounds too deep to close. These chronic constitutional infirmities as well as the failure to develop and apply an enhanced standard of review explain, in large measure, the anomalies that are arising in our capital murder jurisprudence.

James Koedatich may well have murdered Amie Hoffman; if so, he should be punished severely as provided by law. The State did not prove that he had, however, before a jury that was untainted by highly prejudicial publicity. The State's weak case was in reality buttressed by this deluge of publicity, which, to some extent must have conditioned the jurors to a finding of guilt. Finally, the State did not prove defendant's guilt without attempting to influence the jury's deliberations in flagrantly improper ways.

The prejudice deriving from these errors requires that the conviction and sentence must be reversed.

Justice O'HERN joins in Part II of this dissent.

For affirmance of conviction, reverse, sentence and remand — Chief Justice WILENTZ, and Justices POLLOCK, GARIBALDI and STEIN — 4.

For reversal — Justices CLIFFORD, HANDLER and O'HERN — 3.

NOTES

[1] Mr. Koedatich was ultimately convicted for the murder of Ms. O'Brien. On January 5, 1983, there was a shotgun murder of twenty-seven-year-old Judy Brown in Parsippany. No one has been tried for the murder of Ms. Brown.

[2] N.J.S.A. 2C:11-3, containing the Code's murder provisions, consisted of five subsections, (a) to (e), at the time of these crimes and their trials. The death penalty provisions are found in subsections (c) to (e). For convenience, in referring to these provisions we shall, for instance, use Sec. c(1) to designate N.J.S.A. 2C:11-3c(1).

[3] Amicus Curiae The American Civil Liberties Union, takes the position that the issue of prosecutorial discretion is inseparable from the mandate of proportionality review. Furthermore, defendant does not advance a proportionality argument independent of his challenge to prosecutorial discretion.

[4] The Interim Report distinguishes between death-possible homicides and death-eligible homicide. Our discussion of the Report will employ that distinction as well. A death-possible homicide is defined as a homicide in which a factual basis exists for serving notice of an aggravating factor regardless of whether the prosecution actually served notice of a factor. A case becomes death-eligible when the prosecutor serves notice of an aggravating factor.

[5] The Preliminary Report does not exhibit that a defendant's race impermissibly influences the selection of a death-possible case as death-eligible. White defendant cases advanced at the highest rate (.44) relative to blacks (.30) and Hispanics (.18); white defendants were nearly two and one-half times more likely to progress to the death-eligible stage than were Hispanic defendants. At the final processing stage, the imposition of the death sentence, white-defendant cases still outpaced blacks and Hispanics. Based, then, on the race of the defendant, whites appear to have had the greatest probabilities of advancing to death in New Jersey's judicial system and Hispanics the least probabilities, at least at this early stage of tracking the imposition of the death sentence.

[6] In 1983 and in 1984, respectively, Morris County had a population of 413,900 and 417,800; Warren County a population of 84,900 and 85,400; and Sussex County a population of 119,100 and 119,700. See N.J. Dept. of Labor, Office of Demographic and Economic Analysis, Official State Estimate, Population Estimates for New Jersey: Revised Estimates, July 1, 1983 and July 1, 1984.

[7] In certain cases, courts have placed reliance on the fact that the defense failed to exercise all its peremptory challenges. Welch v. United States, supra, 466 A.2d 829; Simmons v. Lockhart, supra, 814 F.2d 504; People v. Balderas, supra, 711 P.2d 480. In this case, the defense exercised only nine of its twenty peremptory challenges. The defense's position is that since so many prospective jurors had read or heard of the case, it was useless for him to exercise his challenges. Our review of the voir dire of the final jury, as it relates to the publicity, indicates that there are significant enough differences in the responses of the potential jurors for the defense counsel in certain instances to have exercised one or more of his peremptory challenges.

The fact that challenges were not used to excuse jurors who heard or read about the Hoffman case somewhat weakens the argument that the jury was tainted by pretrial publicity. State v. Belton, 60 N.J. 103, 107-08 (1972). In State v. LaRocca, 81 N.J. Super. 40, 43 (App.Div. 1963), a juror stated in voir dire that she had read prejudicial information about defendant. Defense counsel did not exercise a peremptory challenge. Id. at 44. There was no evidence that prejudicial information was injected into the jury's deliberations. Id. The Court held good cause to question the jurors was not shown. Id. at 45.

Moreover, as noted in State v. Ramseur, supra, 106 N.J. at 239-40 n. 52, Rule 1:8-3(d) authorizes a trial court upon request to grant defendant in a capital case additional peremptory challenges. No such request was made in this case. Judging from the trial court's willingness to allow supplementary attorney questions on voir dire, we believe such a request would have been granted.

[8] We find little merit in defendant's argument that the failure of these jurors to state their knowledge of the O'Brien charges in voir dire deprived him of his right to exercise peremptory challenges. Our examination of these jurors' questionnaires and voir dire discloses that they made full disclosure. Nor is there any evidence regarding deception during voir dire. See, e.g., State v. Thompson, 142 N.J. Super. 274 (App.Div. 1976) (juror did not advise court on voir dire that he worked as a jail guard but injected into deliberations knowledge of criminal law work).

[9] Rule 1:8-3(a) provides:

For the purpose of determining whether a challenge should be interposed, the court shall interrogate the prospective jurors in the box after the required number are drawn without placing them under oath. The parties or their attorneys may supplement the court's interrogation in its discretion. At trials of crimes punishable by death, the examination shall be made of each juror individually, as his name is drawn, and under oath.

[10] The Prosecutor alleges there were twenty or thirty other suspects in this case whose cars also were searched.

[11] Discussion of prosecutorial misconduct during the penalty phase summation appears infra at 336.

[12] Your verdict in this case must be unanimous, members of the jury. You must agree as to the existence or non-existence of any aggravating factor or factors. You must agree as to the existence or non-existence of the mitigating factor.

(emphasis added.)

[13] Defendant in the subsequent O'Brien trial, in which he received a life sentence, allowed defense counsel to present mitigating evidence in the penalty phase, and took the stand himself to beg for his life. We presume therefore that at the new sentencing hearing in this case, defendant will not continue his refusal to allow evidence of mitigating factors to be introduced. However, there is no reason to be confident on this question. Hence, the issue must be addressed.

[14] The majority in People v. Deere, supra, 710 P.2d at 933 n. 5, sets forth certain perceived difficulties with Justice Broussard's suggested solutions to the problem.

[1] All 105 prospective jurors were required to fill out a questionnaire, in addition to answering questions from the trial court. Twenty-six jurors were excused by the court before questioning on the basis of the questionnaires. Of the 79 prospective jurors who were asked if they had read or heard about the case, nine said that they had not; four of these nine became deliberating jurors. Forty of the 70 who indicated some knowledge of the case were excused for cause by the court.

[2] Judith Deelsnyder had read about defendant's arrest; Loraine Zaccaro and Glenn Cosentiono had read about the case primarily two years earlier, but recalled nothing of the defendant's background; Elizabeth Criares had last read about the case three months before jury selection, and associated the defendant with stabbing; Alice Budd, Deborah Brown, Margaret Andrews, and Barbara Herzig each had knowledge of the case and of the defendant's arrest, but disclaimed any knowledge of the defendant's background; Leslie Fascia, an alternate, remembered reading something about the defendant "two years ago," and remembered reading something "about the body being found, out on Route 80. I don't remember."

[3] In Biegenwald, the court acknowledged that "there was extensive pretrial publicity," including numerous articles in which the prosecutor "repeatedly assumed defendant's guilt and also stated that defendant killed only for pleasure." 106 N.J. at 30-31. The Court concluded, however, that the trial court was not required to "presume prejudice" because the trial court adjourned the trial date, "allowing nearly six months to permit the impact of the publicity to subside." Id. at 35. The Court then concluded that the proper test was whether the jury selection process actually resulted in a fair and impartial jury, and that the trial judge's voir dire resulted in an impartial jury panel. Id. at 36-37. I strongly disagreed. Id. at 81-98 (Handler, J., dissenting).

[4] In Biegenwald, the Court found that the voir dire had effectively established an impartial jury, specifically because "[s]everal of the trial jurors stated that they had never heard of the defendant before coming to court" and all sixteen jurors "indicated that they had encountered little or no publicity regarding the case." Id. at 36. Only one juror expressed knowledge of alleged prior murders — and not of the prior conviction — and she "convinced the trial court," moreover, "that she could disregard what she heard and serve impartially." Id. at 37.

[5] An article on March 10, 1985, in the Daily Record, notes that Judge Stein [the trial court] "excluded any juror who knew of Koedatich's arrest record," but then relates that "Stein's conservative approach did not prevent some jurors from linking Koedatich to both murders once the trial began. The Daily Record has learned that some of the jurors who convicted Koedatich of the Hoffman murder knew he was also charged with killing O'Brien. This information was not allowed to be brought before the Hoffman jurors because it could prejudice Koedatich's case. But a few jurors said once the trial got underway they realized Koedatich was charged with killing both women." Specifically, Ms. Zaccaro stated, "I didn't know he did it in the beginning but halfway through the trial I realized he was linked to O'Brien." Ms. Zaccaro also said that the O'Brien murder was not discussed until after the verdict was returned. Leslie Fascia "said she thought most of the jurors were aware that Koedatich was charged with the O'Brien murder. She said jurors were surprised to learn during the penalty phase ... that Koedatich was convicted of murder in Florida." The majority's emphasis on the fact that we rejected an interlocutory appeal by the defendant requesting an examination of all jurors based on this article is not determinative. Our rejection was influenced by the uncertain and incomplete nature of the appeal process; it did not purport to rule on the materiality of the report as probative evidence of jury taint.

[6] In addition, defendant argues that the silence of the jurors on voir dire — their failure to recollect — on their exposure to prejudicial information "deprived the defendant ``of the opportunity to have excused a juror by appropriate challenge.'" (citing In re Kozlov, 79 N.J. 232, 235 (1979); State v. Thompson, 142 N.J. Super. 274, 280-82 (App.Div. 1976)). Given the nature of the prejudicial information, defendant argues, it is reasonable to assume that a peremptory challenge would have been exercised had the jurors been qualified by the court. I agree. See State v. Bey (II), 112 N.J. 123, 188 (1988) (Handler, J., dissenting). The majority's answer to this argument — that there were many instances where defense counsel would have been justified in exercising a challenge based on publicity but failed to (ante at 277 n. 7), lends inadvertent support to my general view: that defendant was denied a fair trial before an impartial jury.

[7] The Court's failure to safeguard the impartiality of this jury is the more egregious because several of its evidentiary rulings are expressly dependent on that impartiality. Thus, the Court upholds the trial court's admission of testimony relating to the circumstances of the defendant's arrest — including references to flashing blue lights and being forced off the road — despite the fact that this information related closely to the circumstances of the O'Brien murder. Ante at 314-15. I agree that in a neutral venue, a venue unfamiliar with the facts of these cases, the ruling would have been proper. Under the circumstances in which the trial was conducted, and in view of the fact that something did in fact trigger juror recollection, I cannot agree that this venue was neutral. The atmosphere was fraught with the potential for prejudice. Similarly, I agree that in a neutral venue, it would have been "reasonable under the totality of the circumstances to commit the issue of [Diana] Bossard's credibility to the jury." Ante at 317. This was, however, a jury which may have been as subject to suggestible publicity as the witness herself, and was thus not in a position to judge a predisposition it may well have shared.

[8] The prejudice attributable to these events was exacerbated because of the circumstantial nature of the State's case. The predominantly circumstantial nature of the incriminating evidence also arguably supports the defendant's contention that the trial court wrongly excluded evidence implicating a third party in the murder. These rulings, discussed by the majority (ante at 297-312), excluded evidence that a third party, Kevin Sheehan, owned a sports car (ostensibly connecting with testimony that a witness heard a small sports car at the vicinity and time of the murder), that Sheehan gave an inconsistent account of his whereabouts on the night in question, that Sheehan implicated himself in front of his own attorney, that Sheehan would invoke the privilege against self-incrimination if questioned, that Amie Hoffman held unsupervised parties, and that obscene phone calls were received in the cheerleader's locker room. Despite the fact that some of this evidence could be excluded on hearsay and Evidence Rule 4 grounds, it might have established a possible, albeit tenuous, connection between Sheehan and the murder. It is not at all clear that the trial court fully appreciated that greater latitude toward admissibility is appropriate in this kind of case, precisely for the reason that "the evidence against defendant was completely circumstantial, and because the jury fixed the punishment, it was essential that no proof which might conceivably affect the finding of guilt or innocence or the punishment imposed, be excluded if properly admissible." People v. Davis, 29 Ill.2d 127, 129, 193 N.E.2d 841, 842-43 (1963); see also State v. LeClair, 425 A.2d 182, 186 (Me. 1981) ("Especially where the state's case is based upon circumstantial evidence, the court should allow the defendant ``wide latitude' to present all the evidence relevant to his defense, unhampered by piecemeal rulings on admissibility." (Citation omitted.)) Given the uncertain and highly circumstantial nature of the state's case against defendant, and the blatant prosecutorial misconduct discussed infra, I believe the trial court should have been more willing to admit this evidence of third party guilt.

[9] The prosecutor stated on summation:

This part of the trial, folks, is when we think about Amie Hoffman because as the judge said before, Mr. Kenny has his client sitting here and I don't have a client here and that's hard, it's hard to get across (sic) to you during the course of a trial that there once was a human being named Amie Hoffman who I think we've come to know to some extent and we have pictures that were introduced. This is how I want to know Amie Hoffman, a cheerleader, someone who had a life, who had a future. That's the Amie Hoffman that I want to know....

* * * * * * * *

Let's think a little bit about what this girl went through, what was going on in her head when she was taken from the mall and taken up to that retention tank in Randolph because what was she thinking about when that knife was to her nose and to her throat and she was being raped. Was she thinking about was she going to get out of there? Did she have any idea that she wouldn't make it out of that area?

This whole case protected the rights of the defendant. The judge was here ruling on all the evidence, making sure that admissible evidence went in, making sure that this trial went according to our due process and our constitution. Well, where was Amie Hoffman's due process. Where was the victim's judge? There he is right there because he is the one that sentenced Amie Hoffman to die. She didn't have a jury of the community to decide whether or not she lives....

* * * * * * * *

You know, it's not easy for me to have to come in here on Monday morning and ask twelve people to sentence someone to die. That's not easy. I share that burden with you but there comes a point in time when you have to say it's over. When a person like the victim is killed, there has to be some justice and no matter how many times you turn those pages and how many times you think about can I sentence someone to die, each of you told us at the beginning that depending on the facts and circumstances, you could consider the death penalty and I took that to mean that if I presented the right case to you, a case strong, a case of cold-bloodedness, you could sentence someone to die and you knew at that time like I'm standing here right now that I have to ask you that.

I've proved to you the aggravating factors. You have all the reason in the world to sentence him to death. I'm asking you to do what you said you could to in the right set of circumstances and, folks, if this isn't the right set, I don't know if there is one.

[10] Justice White, concurring, addressed the prosecutorial discretion argument separately "because the cases in which no capital offense is charged escape the view of the Georgia Supreme Court and are not considered by it in determining whether a particular sentence is ... disproportionate." Id. 428 U.S. at 225, 96 S.Ct. at 2949, 49 L.Ed.2d at 903.

[11] This decision was later overturned by a referendum amending the State Constitution.

[12] When the court came to the question in the verdict sheet, utter confusion reigned:

THE COURT: Okay. Do you unanimously find that the following exists as a mitigating factor: Any factor which is relevant to the defendant's character or record or to the circumstances of the offense of murder?

THE FOREMAN: No.

THE COURT: The answer is no?

THE FOREMAN: No.

THE COURT: And is that — that, of course, is unanimous, is that correct?

THE FOREMAN: Yes, your Honor.