The opinion of the Court was delivered by
HANDLER, J.
The defendants, James Brown and Ronald Emm, became involved in an apparent driving contest that resulted in Brown’s car striking a third vehicle, killing its driver. Both defendants were indicted for death by auto, tried jointly, and convicted. On a post-verdict motion, the trial court held that the defendants had been prejudiced by their joint trial because their defenses were in conflict. The court found further that this prejudice was exacerbated by the erroneous admission of Emm’s pre-arrest silence, the jury’s inability to evaluate the credibility of two eyewitnesses, and the court’s failure to instruct the jury on lesser-included motor vehicle offenses. The court concluded that the cumulative prejudice generated by the joint trial justified the grant of new and separate trials for defendants. The State appealed that determination. In a reported decision, the Appellate Division affirmed. 228
N.J.Super.
211, 549
A.
2d 462 (1988). We granted the State’s petition for certification. 114
N.J.
497, 555 A.2d 618 (1989).
I.
On the night of September 23, 1985, the defendants were both driving in the northbound lane of Route 34, apparently seeking to pass one another. Brown’s vehicle suddenly swerved into the southbound lane, striking Frank Dimitri’s oncoming car and killing him. The State contended that both defendants were at fault in causing the fatal collision. According to the State, the evidence demonstrated that Brown and Emm had been engaged in a “cat-and-mouse” game, alternately tailgating and passing each other over a two-mile stretch down Morristown Road and onto Route 34, when Brown’s vehicle swerved into the opposite lane of travel and crashed into Dimitri’s car. Witnesses to this episode included Emm’s girlfriend Marilyn Decker, who was a passenger in Emm’s car at the time, and several other motorists, Roger Martin, Christopher Mosley, Nicholas LaConte, and Harry Masked.
Defendants offered different versions of the events leading up to the collision, each implicating the other as the sole or primary guilty party. The sixty-six-year-old Brown testified that he was slow to accelerate his silver Plymouth Volare when a traffic light on Morristown Road, a two-lane roadway that intersects Route 34, turned green and that the eighteen-year-old Emm, driving a green Oldsmobile Cutlass, tried to “beat” him across the intersection by passing him on the right. Unable to pass, Emm tailgated and flashed his headlights at Brown’s vehicle down Morristown Road and onto Route 34, causing Brown to have a near-collision with an oncoming New Jersey Transit bus. Brown was too frightened to stop, and he drove past his own driveway, as well as a local gas station and an autobody shop, heading for the Old Bridge police station further up Route 34. Emm, then driving on the right side of Brown’s car, began to push Brown’s car into the oncoming traffic. The fatal collision with Dimitri’s car occurred when Emm, on the righthand shoulder of the highway, forced Brown into the opposite lane.
Emm acknowledged that after the light turned green Brown was slow to accelerate and that he then sought to pass Brown on the right, even though traffic in that lane was required to turn right. Emm, however, claimed that he actually succeeded in passing Brown. This maneuver angered Brown, who started to follow him, driving erratically, tailgating and flashing his high beams along the stretch of Morristown Road. Brown finally passed Emm on the right shoulder of Route 34 and began to move in front of him. The fatal accident occurred when, as Brown’s car passed Emm’s, it glanced off a curb, appeared to lose control, and veered diagonally across the road into the southbound lane, colliding with Dimitri’s car.
Witnesses Nicholas LaConte and Christopher Mosley were riding together on Morristown Road in a Chevrolet pick-up truck driven by LaConte, when Brown and Emm both passed them. Mosley testified that a gray or green Volare, driven by a “bald head, older man,” passed first, followed by a “tan,
mustard color” Oldsmobile. The cars appeared to be “fighting for position around [a] bend” in Morristown Road. Mosley testified further that after the two vehicles passed, they forced an oncoming car off the road because their vehicles were traveling side-by-side, taking up both lanes. In Mosley’s words, they were “like fighting for land both back and forth back and forth into each other’s lane.” On cross-examination, Mosley testified that the Oldsmobile passed them first, and the Volare passed second. This inconsistency regarding whether the Vo-lare or the Oldsmobile passed first was highlighted on re-direct- and re-cross-examination.
LaConte testified that the two cars were traveling “side-by-side ’til they got to the end of the road [Morristown Road],” with the Volare in the southbound lane and the Oldsmobile in the northbound lane. Both Mosley and LaConte testified that Brown’s and Emm’s vehicles appeared to bump each other, although, an examination of the cars after the accident proved there was no such contact.
Harry Maskell, the New Jersey Transit bus driver, testified that he clearly saw Emm riding in the northbound lane of Route 34 and turning his wheel towards Brown’s car, forcing Brown into the southbound lane. Maskell had to swerve his bus onto the shouldér of the southbound lane to avoid hitting Brown. Roger Martin, another motorist, testified that he saw the vehicles appear to bump each other as they drove side-by-side past the intersection of Cottrell Road and Route 34. Martin observed Brown’s car half in the southbound lane and half in the northbound lane, and Emm’s car half in the northbound lane and half on the shoulder. He then saw Emm’s car veer to the left and force Brown’s car further into the southbound lane.
Emm did not stop after the collision, but proceeded to the Cheesequake Fire Department where he was a volunteer firefighter. After reporting the collision, Emm returned to the accident scene with other firefighters to help extricate Dimitri and Brown. During this time, Emm did not disclose to investí-
gating police officers that he had been involved in the incident. Further police investigation revealed the role of a third car in the accident. Two days later, Emm reported to the police that he had been driving the “other” car that witnesses had observed and mentioned.
Both defendants were indicted in February 1986 for death by auto, contrary to
N.J.S.A.
2C:ll-5. In March, defendants moved for separate trials. The court denied the motion for severance.
State v. Brown (Brown I),
219
N.J.Super.
412, 530
A.
2d 402 (Law Div.1987). The court ruled that although the respective defenses of the defendants were antagonistic, they were “not necessarily irreconcilable or mutually exclusive.”
Id.
219
N.J.Super.
at 419, 530
A.
2d 402. The Appellate Division denied defendants’ motions for leave to appeal from that ruling. In January 1988, defendants renewed their motions for severance before another judge. This application was again denied and, after a joint trial, both defendants were convicted. However, following their convictions, the trial court granted their motions for new and separate trials. The trial court based its decision on four main factors: (1) that Brown’s and Emm’s defenses were mutually exclusive and antagonistic; (2) that the court had improperly admitted evidence of Emm's pre-arrest silence; (3) that the jury was not able to evaluate properly the credibility of witnesses Mosley and LaConte; and (4) that the court should have instructed the jury on the lesser-included offenses of reckless and careless driving. The Appellate Division affirmed, substantially deferring to the trial court’s determination that the prejudice resulting from the joint trial justified new and separate trials.
We consider each of the grounds of the decisions of the lower courts in order to determine whether, separately or cumulatively, they justify the severance of the trial and the grant of new trials. The decision whether to grant severance rests within the trial court’s sound discretion and is entitled to great deference on appeal.
State v. Laws,
50
N.J.
159, 175, 233 A.2d 633 (1967),
cert. denied,
393
U.S.
971, 89
S.Ct.
408, 21
L.Ed.2d 384 (1968);
State v. Sanchez,
224
N.J.Super.
231, 245, 540 A.2d 201 (App.Div.),
certif. denied,
111
N.J.
653, 546 A.2d 561 (1988). In reviewing a trial court’s decision to grant a new trial following a jury verdict, an appellate court must be “guided by essentially the same standard as that controlling the trial judge’s review of a jury verdict” and must “weigh heavily” the trial court’s views on “credibility of witnesses, their demeanor, and [the trial court’s] general ‘feel of the case.’ ”
State v. Sims,
65
N.J.
359, 373, 322 A.2d 809 (1974). If the trial court acts under a misconception of the applicable law, however, the appellate court need not give such deference.
See State v. Steele,
92
N.J.Super.
498, 507, 224 A.2d 132 (App.Div.1966);
see also Baxter v. Fairmont Food Co.,
74
N.J.
588, 598-600, 379 A.2d 225 (1977) (judge must have pervading sense of “wrongness” in overturning jury verdict; “feel of the case” factor does not control appellate review when trial court’s “subjective conclusions” are not supported by the record). Moreover, to the extent that the weight of the evidence is germane to its decision, the trial court ought not to engage in a weighing of the persuasiveness of that evidence but only of its sufficiency.
Kulbacki v. Sobchinsky,
38
N.J.
435, 445, 185 A.2d 835 (1962);
see Battista v. Olson,
213
N.J.Super.
137, 142, 516 A.2d 1117 (App.Div.1986). We have also observed that the trial judge’s decision is not “entitled to any special deference where it rests upon a determination as to worth, plausibility, consistency, or other tangible considerations apparent from the face of the record with respect to which [s]he is no more peculiarly situated to decide than the appellate court.”
Dolson v. Anastasia,
55
N.J.
2, 7, 258 A.2d 706 (1969). Moreover, even intangible considerations must be supported by some factual basis or other objective elements.
Baxter, supra,
74
N.J.
at 601, 379 A.2d 225.
II.
The dominant issue in this case is whether defendants were entitled to be tried separately. Resolution of that question will
strongly influence, if not dictate, our determination of the remaining issues bearing on whether to grant new and separate trials to defendants.
Our
Rule
of practice governing the severance of joint trials,
Rule
3:15-2(b), prescribes a general standard focusing on the existence of prejudice:
If for any other reason it appears that a defendant or the State is prejudiced by a permissible or mandatory joinder of offenses or of defendants in an indictment or accusation the court may order an election or separate trials of counts, grant a severance of defendants, or direct other appropriate relief.
Despite a concern for prejudice, in considering a motion for severance, the trial court should “balance the potential prejudice to defendant’s due process rights against the State’s interest in judicial efficiency.”
State v. Coleman,
46
N.J.
16, 24, 214 A.2d 393 (1965),
cert. denied,
383
U.S.
950, 86
S.Ct.
1210, 16
L.Ed.2d
212 (1966). Joint trials also offer advantages to our criminal justice system other than judicial economy. “Joint trials generally serve the interests of justice by avoiding inconsistent verdicts and enabling more accurate assessment of relative culpability — advantages which sometimes operate to the defendant’s benefit.”
Richardson v. Marsh,
481
U.S.
200, 210, 107
S.Ct.
1702, 1708, 95 L.Ed.2d 176, 187 (1987). When the crimes charged arise from the same series of acts, and when much of the same evidence is needed to prosecute each defendant, a joint trial is preferable.
Ibid.; see State v. Moore,
113
N.J.
239, 273, 550 A.2d 117 (1988);
State v. Briley,
53
N.J.
498, 503, 251 A.2d 442 (1969). The danger by association that inheres in all joint trials is not in itself sufficient to justify a severance, provided that by proper instructions to the jury, the separate status of co-defendants can be preserved.
State v. Freeman,
64
N.J.
66, 68, 312 A.2d 143 (1973). Thus, the quantum of real prejudice is critical in any determination to grant a severance.
Courts have generally held that defendants cannot be tried together fairly when their defenses are antagonistic and mutually exclusive or irreconcilable. The test for granting
severance, however, is a rigorous one. Separate trials are required only when defendants “present defenses that are antagonistic at their core.”
United States v. Berkowitz,
662 F.2d 1127, 1134 (5th Cir.1981). The mere existence of hostility, conflict, or antagonism between defendants is not enough.
Determination of central or core antagonism necessarily focuses on the mutual exclusivity of defenses. “Mutual exclusivity” demands that the jury’s universe of choices be limited to two: the jury can believe
only
either one defendant or the other. The prosecutor’s theory of the case, and the defenses themselves, must force the jury to choose between the defendants’ conflicting accounts and to find only one defendant guilty. “When ... the jury can reasonably accept the core of the defense offered by either defendant only if it rejects the core of the defense offered by his co-defendant, the defenses are sufficiently antagonistic to mandate separate trials.”
State v. Vinal,
198
Conn.
644, 652, 504 A.2d 1364, 1368 (1986). If the jury can return a verdict against one or both defendants by believing neither, or believing portions of both, or, indeed, believing both completely, the defenses are not mutually exclusive. Defenses that do not demand that the jury choose one or the other in order to return a verdict, though clearly in conflict and antagonistic, are not mutually exclusive.
See State v. Sanchez, supra,
224
N.J.Super,
at 247-48, 540 A.2d 201.
The fact that one defendant seeks to escape conviction by placing guilt on his or her co-defendant has not been considered sufficient grounds for severance.
See Lewis v. State,
220
Ark.
914, 251
S.W.2d
490, 494 (1952) (fact that each defendant claims other alone is at fault does not make their defenses mutually exclusive if it is possible to hold both defendants liable);
Simcic v. United States,
86 A.2d 98 (D.C.) (where defendants’ cars collided, resulting in one or both of them striking a third car, no right to severance because even though defendants’ defenses were antagonistic, both defendants’ actions could have caused the death),
aff'd sub nom. Miciotto v.
United States,
198
F.2d
951 (D.C.Cir.1952);
Commonwealth v. Davis,
388
Pa.Super.
224, 565 A.2d 458 (1989) (where co-defendants each claimed other was driving at time of fatal collision, defenses not “mutually exclusive” and no right to severance because three people were in car, and jury could have found that neither defendant was driving);
see also State v. Varricchio,
176
Conn.
445, 408 A.2d 239 (1979) (where two defendants were racing along highway and collided with another vehicle, causing the death of its driver, trial court did not abuse its discretion by denying defendants’ severance motion because defenses were neither antagonistic nor mutually exclusive).
As noted, the trial court in this case concluded that defendants should have been tried separately because their defenses were antagonistic and mutually exclusive or irreconcilable. It reasoned that “Mr. Brown’s testimony, if accepted, would have led to the conviction of Mr. Emm and Mr. Emm’s testimony ... would have led to the conviction of Mr. Brown.” The trial court failed to recognize, however, that although defendants’ versions were in conflict, their defenses did not compel the jury to believe one defendant at the expense of the other in order to reach a verdict.
See, e.g., United States v. Berkowitz, supra,
662 F.2d 1127;
State v. Vinal, supra,
198
Conn.
644, 504 A.2d 1364. We thus agree with the analysis of the motion court: the jury could find
both
defendants at fault.
Brown I, supra,
219
N.J.Super.
at 419. Hence, their defenses were not “mutually exclusive.”
The trial court also stressed the high degree of antagonism between defendants. That antagonism, however, did not itself create impermissible prejudice, nor was it compounded by other factors that would militate against a joint trial.
See, e.g., State v. Young,
46
N.J.
152, 215 A.2d 352 (1969);
State v. Pickles,
46
N.J.
542, 218 A.2d 609 (1966); see discussion,
infra,
relating to other alleged errors arising from joint trial. The trial court thus noted that “we had a very messy and aggressive trial being fought between Emm and Brown.” The Appellate Divi
sion agreed, observing that the trial interrogation “was to a large extent adversarial between defendants rather than between the State and each defendant.” 228
N.J.Super.
at 221, 549 A.2d 462.
It is possible that the level of antagonism between co-defendants, despite the absence of mutually exclusive defenses, can become so intense as to justify a severance. Some courts have recognized that the presence of extreme conflict between co-defendants may alone convince the jury that at least one must be guilty, or that both are guilty.
See Commonwealth v. Moran,
387
Mass.
644, 658-660, 442
N.E.2d
399, 408 (1982). In this case, however, the courts below focused on the effects of the antagonism and seemingly concluded that by taking advantage of this antagonism, the prosecutor defaulted on his own prosecutorial responsibilities, thus inducing the jury to find both defendants guilty. Vigorous and, indeed, relentless cross-examination of each defendant by the other may be “adversarial” or “prosecutorial,” but the fervor of such examination does not mean that there was an abandonment of prosecutorial responsibility. What the courts below apparently viewed as a failure to prosecute was evidenced essentially by minimal cross-examination of certain witnesses and a somewhat tempered summation. This did not, under the circumstances, equate with prosecutorial dereliction. Hence, this record, even as explained by the trial court, does not reveal that the confrontation between defendants, coupled with any impermissible prosecutorial tactics, was the factor that precipitated the jury’s guilty verdicts.
The trial court also gave added weight to the presence of antagonism between the defendants because it “felt that the State’s case standing alone was rather weak.” The court concluded that the weakness of the State’s case, coupled with the hostility between defense counsel, may have confused the jury about the proper burden of proof. The court supported this conclusion by pointing to the jury’s request for additional instructions concerning the allocation of the burden of proof. The court, however, rectified the problem with clarifying in
structions. The State obviously bore the burden of proof, the defendants did not, and the jury was emphatically so instructed.
We have recognized often that a jury can be adequately instructed with respect to conflicts among parties and can understand and apply the correct burdens of proof.
See, e.g., State v. Freeman, supra,
64
N.J.
at 68-69, 312 A.2d 143;
State v. Gardner,
54
N.J.
37, 46, 252 A.2d 726 (1969). This occurred here. We are thus not persuaded that there was jury confusion that deprived defendants of a fair trial.
In sum, we determine that the defenses posed by defendants were neither irreconcilable nor mutually exclusive. Further, we do not find that the antagonism between the parties was caused by, or resulted in, prosecutorial misconduct in the form of an abandonment of prosecutorial responsibility. Finally, although the conflict between the defendants made difficult the jury’s determination of the facts in the case, we conclude that the jury did not suffer irremediable confusion that impaired its deliberations or impugned its verdicts.
III.
Emm waited two days before reporting his involvement in the accident to the police. The trial court believed that Brown had a constitutional right to confront Emm and impeach his credibility by inquiring into Emm’s pre-arrest silence. The court stated that it could not ignore the prejudice that Brown would
suffer if evidence of Emm’s silence was not admitted, notwithstanding misgivings over its probative worth. Accordingly, the court allowed both the State and Brown to question Emm about his pre-arrest silence but refused to allow the prosecutor or Brown’s attorney to comment to the jury on the fact of Emm’s pre-arrest silence.
At the post-trial hearing for a new trial, the trial court reconsidered this issue. It noted that it probably would not have allowed the inquiry by the State into Emm’s pre-arrest silence had Emm been tried separately. The court also focused on the conflict between Brown’s right to confront witnesses against him and the prejudice to Emm caused by admission of his pre-arrest silence. These reasons were mentioned by the court as additional support for its decision to grant new and separate trials. The Appellate Division agreed that admission of Emm’s pre-arrest silence was improper, but based its decision on the fact that Emm’s silence was not probative of guilt.
Brown II, supra,
228
N.J.Super.
at 222-23, 549 A.2d 462.
It is by no means clear that had Emm been tried separately, his pre-arrest silence was so lacking in probative worth that it would not have been admissible. In
State v. Deatore,
70
N.J.
100, 108-09, 358 A.2d 163 (1976), we held that New Jersey common law provides a defendant with the right to remain silent when in police custody or under interrogation, ruling that defendant’s silence “at or near” the time of arrest may not be introduced to impeach his credibility.
Deatore,
however, did not deal specifically with “silence” that significantly precedes an arrest.
The issue has been addressed by the Supreme Court in
Jenkins v. Anderson,
447
U.S.
231, 100
S.Ct.
2124, 65
L.Ed.2d
86 (1980). In
Jenkins,
the defendant waited two weeks before reporting to police that he had , stabbed the victim in self-defense. At trial, the prosecutor attempted to impeach defendant’s credibility by suggesting that the defendant would have spoken sooner had his defense been true.
Id.
447
U.S.
at 235,
100
S.Ct.
at 2127, 65
L.Ed.2d
at 92. The Supreme Court concluded that once a defendant testifies, the prosecution is allowed to “advance[ ] the truth-finding function to the criminal trial” by using defendant’s silence for impeachment purposes.
Id.
447
U.S.
at 238, 100
S.Ct.
at 2129, 65
L.Ed.2d
at 95.
In the wake of the Supreme Court’s invitation in
Jenkins
to “formulate evidentiary rules defining the situations to which silence is viewed as more probative than prejudicial,”
id.
447
U.S.
at 240, 100
S.Ct.
at 2130, 65 L.Ed.2d at 96, courts have approached the issue of pre-arrest silence in different ways. Some have followed
Jenkins,
generally admitting pre-arrest silence for the purposes of impeaching a defendant’s credibility.
See People v. Redmond,
29 Cal.3d 904, 176
Cal.Rptr.
780, 633 P.2d 976 (1981) (once defendant elects to testify, scope of cross-examination is wide, and jury is free to consider why defendant failed to explain or deny any evidence or fact against him that it would be reasonable for him to deny or explain). Others have resolved the admissibility of pre-arrest silence in terms of surrounding circumstances.
Compare People v. Conyers,
52 N.Y.2d 454, 459, 438 N.Y.2d 741, 744, 420 N.E.2d 933, 935-36 (1981) (probative value of “pretrial” silence is nearly always outweighed by the risk of undue prejudice: “the use of such evidence for impeachment purposes cannot be justified in the absence of unusual circumstances”)
with People v. Collier,
426
Mich.
23, 393 N.Y.2d 346 (1986) (pre-arrest silence found admissible because it would have been natural for defendant to report to the police that he had been robbed by the person he was charged with assaulting).
The lower courts in this case relied heavily on the Appellate Division decision in
State v. Merola,
214
N.J.Super.
108, 518 A.2d 518 (1986),
certif. denied,
107
N.J.
91, 526 A.2d 168 (1987). In
Merola,
the State alleged that the defendant had killed one person and wounded another during a drug deal; the defendant claimed that the two victims had shot each other during the drug deal. The defendant did not surrender to the police until he read about the incident in a local paper. At trial, the
prosecutor questioned the defendant about his failure to have communicated with the police immediately. On appeal, defendant argued that this questioning violated his privilege against self-incrimination. In explaining why the defendant’s pre-arrest silence was not probative, the court considered whether silence was natural or unnatural under the circumstances. It found that “[a] likely explanation [for defendant’s silence] was [his] conscious reliance upon his right to remain silent.”
Id.
214
N.J.Super,
at 118, 518
A.
2d 518. Thus, two primary factors were considered in determining, whether pre-arrest silence can be admitted for impeachment purposes: one relating to whether there is a right on the part of the defendant
not
to speak that would be violated if his or her pre-arrest silence were used for impeachment purposes, the other relating to whether pre-arrest silence is probative at all on the issue of credibility or culpability.
The right to remain silent is now codified at
N.J.S.A.
2A:84A-19.
Rule
25 of the New Jersey Rules of Evidence tracks the language of
N.J.S.A.
2A:84A-19.
Evidence Rule
25 suggests that the right to remain silent might exist only in the face of a compulsion to speak.
Cf. State v. Ripa,
45
N.J.
199, 212
A.
2d 22 (1965) (error to permit on State’s case, as evidence of-guilt, testimony that defendant had refused to answer police questions about a homicide). It describes “a right to refuse to disclose” incriminating matter “in an action, or to a police officer or other official.” Moreover, the text of the rule does
not indicate a
per se
bar to use of pre-arrest silence for the purposes of impeachment. Our courts have not considered any testimonial privilege as a
per se
obstacle to the evidentiary use of such pre-arrest silence or conduct.
See, e.g., State v. Merola, supra,
214
N.J.Super.
at 118, 518 A.2d 518;
State v. Bonet,
132
N.J.Super.
186, 333 A.2d 267 (App.Div.1975).
We conclude, in general conformity with
Jenkins,
that under our own law a defendant has no right
not
to speak prior to arrest. We are also of the view, however, that a defendant has no duty
to
speak prior to arrest. Simply, there is no legal constraint one way or the other — either to speak or not to speak — prior to an arrest. Consequently, evidence of pre-arrest silence, particularly in the absence of official interrogation, does not violate any right of the defendant involving self-incrimination.
See State v. Ripa, supra,
45
N.J.
199, 212 A.2d 22; see
also State v. Burt,
59
N.J.
156, 165, 279 A.2d 850 (1971) (Hall, J., concurring in part and dissenting in part) (reference to whether “defendant ever told anyone about an accidental shooting before his trial testimony” does not “constitute any infringement upon the Fifth Amendment right to remain silent.”),
cert. denied,
404
U.S.
1047, 92
S.Ct.
728, 30 L.Ed.2d 735 (1972). Thus, in effect, the probative worth of such pre-arrest silence should be considered objectively and neutrally, without added coloration attributable to any legal right in such silence.
We now hold that pre-arrest silence may be admitted for impeachment purposes provided no governmental compulsion is involved. In determining admissibility, the probative worth of pre-arrest silence as bearing on credibility must be assessed in light of all the surrounding circumstances. If it can be inferred by the fact-finder that a reasonable person situated as the defendant, prior to arrest, would naturally have come forward and mentioned his or her involvement in the criminal episode, particularly when this is assessed against the defendant’s apparent exculpatory testimony, then the failure to have
done so has sufficient probative worth bearing on defendant’s credibility for purposes of impeachment.
As mentioned, in concluding that Emm’s pre-arrest silence should have been excluded, the trial court was strongly influenced by the reasoning of the court in
State v. Merola, supra,
214
N.J.Super.
108, 518 A.2d 518. The court in
Merola
determined that a defendant’s pre-arrest silence could be excluded from evidence if it reflected a “conscious reliance upon his right to remain silent.”
Id.
214
N.J.Super.
at 118, 518 A.2d 518. To the extent this reasoning suggests, incorrectly, that in this case Emm possessed and relied on a right to be silent, it may have misled the trial court to conclude that evidence of Emm’s silence should have been excluded.
The Appellate Division affirmed the trial court’s ruling, however, because it found that Emm’s pre-arrest silence did not have probative worth for impeachment purposes. It reasoned that “if Emm truly believed his version of the facts,” he might also have believed he had no “civil or moral duty” to speak out about his involvement before he did. 228
N.J.Super.
at 222-23, 549 A.2d 462.
The flaw in this analysis is that the court considered only one possible inference that could be drawn from Emm’s pre-arrest silence. Arguably, if Emm really believed he was not responsible for the fatality, it may have been “unnatural” for him to have reported his own involvement in the accident because it had incriminatory potential. It is equally possible, however, that it would have been “natural” for Emm to have spoken to the police about his involvement sooner than he did had his testimonial version of the episode been true, and he believed that it exonerated him. Moreover, according to his own version, Emm possessed important, perhaps unique, information relating to the accident. It would appear natural for a reasonable person in Emm’s place to have related to the police — who were then investigating the accident and obviously seeking to learn how the accident had occurred — that he was an eye
witness. That inference seems particularly strong in light of the fact that Emm, himself a volunteer fireman, had come forward and actually returned to the scene with his fellow firefighters to assist the police officers at the accident. Pre-ar-rest silence in these circumstances bears on the defendant’s veracity.
See State v. Burt, supra,
59
N.J.
at 164, 279
A.2d
850 (Hall, J., concurring in part and dissenting in part) (evidence of “a failure to previously volunteer an exculpatory explanation testified to at trial [can be used] as a means of challenging the credibility of the testimony”).
We conclude, therefore, that evidence regarding pre-arrest silence is admissible if, when viewed objectively and neutrally in light of all circumstances, it generates an inference of consciousness of guilt that bears on the credibility of the defendant when measured against the defendant’s apparent exculpatory testimony. Here, the probative worth of Emm’s failure to mention to the police that he had witnessed the events — whether that entailed a consciousness of guilt, a desire not to become involved, a feeling that it was simply unnecessary, or a belief that he had already fulfilled whatever duty he had — was a matter, ultimately, for the jury in assessing Emm’s credibility.
At the motion for a new trial, the court expressed additional concerns over its treatment of this issue. It observed that Brown’s attorney, rather than the State, had “pursued” the matter of Emm’s pre-arrest silence. We find nothing amiss in the prosecutor’s decision to add only minimal cross-examination of Emm to that by Brown’s attorney. That decision was a matter of tactics, not an abuse of prosecutorial discretion.
Supra
at 607-608, 573 A.2d at 891-892. The court also noted that despite its instructions, the jury came back with questions concerning whether the burden of proof was on the State and whether it should consider only the evidence presented by the State, or that presented by the defendants as well. As the trial court itself observed, it answered these questions
“in a very emphatic manner to the jury.” As we earlier noted, we see no reason to conclude that these forceful instructions clearly affixing the State’s burden of proof were not fully understood and followed by the jury.
Supra
at 607-608, 573 A.2d 892.
We conclude, in sum, that Emm’s pre-arrest silence was properly admitted for impeachment purposes and, further, that it was carefully limited only to impeachment.
Under any version of the facts, the jury could have inferred that Emm’s silence was indicative of prevarication when measured against his testimonial version of the fatal accident. Further, in concurrence with the Appellate Division, we do not find that Emm’s impeachment confused the jury, or relieved the State of its prosecutorial responsibilities or ultimate burden of proof.
IV.
The trial court believed that if defendants had been tried separately, the jury probably would not have found witnesses LaConte and Mosley to be credible. That perception contributed to its decision that defendants suffered prejudice by their joint trial.
Both Mosley and LaConte testified that they believed defendants’ cars had bumped each other several times as they drove down the road and that they expected some real damage to
each car as a result of this contact, but no dents or scratches were subsequently found. Their description of the colors of the cars was also incorrect. Mosley testified that the Volare was gray and green and that the Oldsmobile was a “tan mustard color.” In fact, Brown’s Volare was silver and Emm’s Oldsmobile was green. In addition, Mosley and LaConte gave inconsistent testimony regarding their own actions prior to and following the accident. These factors led the trial court at the motion for a new trial to question their credibility and conclude that “their recollection of what they said they saw the [defendants’] cars doing was at least exaggeration and unreliable.” The trial court did not find, however, that either witness was “lying” or that their testimony was so lacking in credibility that the State failed to satisfy its burden of proof or that the verdict was against the weight of the evidence. Rather, the court explained that because the jury was “presented with the spectacle of [the defendants] slugging it out against each other ... [this] probably led [the jury] to disregard” the testimonial infirmities of these witnesses.
This conclusion — that a jury trying each defendant separately might have disbelieved these witnesses — does not impugn the trial court’s original decision to conduct a joint trial or justify its later decision to grant new and separate trials. The miscellaneous discrepancies and inaccuracies did not render the testimony of LaConte and Mosley incredible or unworthy of belief. The Appellate Division so found, 228
N.J.Super.
at 219, 549
A.
2d 462, and the trial court itself surmised only that the testimony was “exaggeration and unreliable,” not useless.
Further, the Appellate Division, disagreeing with the trial court, found that although Mosley’s and LaConte’s recollection of events unrelated to the accident was not credible, they testified credibly and in a “straightforward manner concerning their observations as the cars passed their truck.”
Brown II,
228
N.J.Super.
at 219, 549
A.
2d 462. Mosley and LaConte clearly described the operations of defendants’ cars while they
were in view. That critical evidence indicated an on-going, cat-and-mouse driving episode.
Moreover, witnesses Martin and Marshall also testified that defendants were involved in some kind of dangerous driving confrontation. In addition, the testimony of Mosley that defendants’ cars appeared to bump each other was indirectly supported by Martin, who also believed that defendants’ cars, occupying both sides of the roadway, came into contact with one another. Other mistakes or inaccuracies, such as the color of defendants’ vehicles, were made by other witnesses as well, and were explained to the jury in terms of the low visibility that night and the rapidity of the unfolding events.
Even though the testimony of these other witnesses considered separately might not equate with a mutual cat-and-mouse game, it was not inconsistent with that given by Mosley and LaConte. Hence, as the Appellate Division itself determined, the trial court’s evaluation of witness credibility was “largely an irrelevant issue.” 228
N.J.Super,
at 219, 549
A.
2d 462. Although the observations of Mosley and LaConte in some particulars were wrong, the truthfulness and weight of their testimony were for the jury to evaluate.
See State v. Ernst,
32
N.J.
567, 583, 161
A.
2d 511 (1960) (jury may believe some, all, or none of a witness’s testimony);
see also State v. Ingenito,
87
N.J.
204, 212, 432 A.2d 912 (1981) (credibility of witnesses is usually assessed by jury). It was not for the trial court to engage in the assessment of the persuasiveness of the evidence.
See Kulbacki v. Sobchinsky, supra,
38
N.J.
at 445, 185 A.2d 835.
Despite its reservations, the Appellate Division yielded to the trial court’s determination of the witnesses’.credibility. “[T]he judge’s forceful observations as to the credibility of these four witnesses must be given some credence by us.”
Id.
87
N.J.
at 220, 432 A.2d 912. We need not be so deferential. In this case the appellate courts are not confronted with a trial court’s “feel of the case” in the sense that the lower court was presented
with a verdict that is inexplicable under any version of the evidence or with truly incredible witnesses who were bizarrely accredited by the jury.
See State v. Johnson,
42
N.J.
146, 161-62, 199
A.2d
809 (1964);
see also Carrino v. Novotny,
78
N.J.
355, 360-61, 396 A.2d 561 (1979) (jury verdict should be reversed only if “so distorted and wrong” as to constitute a manifest denial of justice);
Dolson v. Anastasia, supra,
55
N.J.
at 8-9, 258 A.2d 706 (jury verdict should be reversed only if jury could not have reasonably and fairly reached its verdict).
The trial court, as earlier noted, believed that in separate trials a jury probably “would have rejected LaConte and Mosley’s description of what the drivers were doing.” Although the trial court did not conclude that the jury’s verdict was against the weight of the evidence under
Rule
3:20-1, we surmise from its explanation that it thought that the jury did not have the
opportunity
to assess the credibility of these witnesses because of the joint trial.
See R.
3:20-1. We fail to perceive how the joint trial of these defendants prevented the jury from having a fair and adequate opportunity to assess the credibility of all the witnesses. That opportunity was afforded the jury through the opening remarks of all counsel, through the full examination and cross-examination of these witnesses by all counsel, including the prosecutor, by relevant comments of all counsel during summation and, of course, by the careful and complete instructions of the trial court concerning how jurors should consider and determine witness credibility.
In sum, the trial court’s feel of the case relating to the testimony of Mosley and LaConte cannot on this record support its ultimate determination that new trials are required in the interest of justice.
R.
3:20-1. That a jury in a separate trial might assess credibility differently is not the test to be applied with respect to the entitlement to a separate trial. “[A] separate trial should not be granted merely because it would offer defendant a better chance of acquittal.”
State v. Sanchez, supra,
224
N.J.Super.
at 245, 540 A.2d 201.
V.
At the hearing on defendants’ motions for new and separate trials, the trial court raised
sua sponte
the issue of the propriety of its jury instructions, particularly whether it should have charged the jury with careless or reckless driving as lesser-included offenses of the death-by-auto charge. The Appellate Division, finding no reversible error, observed that it would have been appropriate for the trial court to discuss the lesser-included offenses of reckless and careless driving in order to give the jury a fuller understanding of the offense of death by auto under the general guidance of
State v. Concepcion,
111
N.J
373, 545 A.2d 119 (1988).
Brown II, supra,
228
N.J.Super.
at 224, 549 A.2d 462.
We have determined that motor vehicle infractions can be lesser-included offenses of the crime of death by auto.
State v. Muniz,
118
N.J.
319, 571 A.2d 948 (1990). In such circumstances, in order to lessen the risk of a so-called “all-or-nothing” verdict, the court should consider giving the jury an instruction, alerting it to the availability of other lesser-included offenses indicated by the defendant’s conduct. The court should make the jury aware that a defendant’s conduct could implicate lesser-included motor vehicle offenses, but should not permit the jury to determine the defendant’s guilt or innocence of these offenses.
Id.
at 609-610, 573 A.2d at 893.
The trial court apparently felt that with a more ample charge, the jury might have found that defendant Brown was guilty only of careless driving. The basis for that conclusion was the court’s belief that it could have charged the jury to consider and determine whether defendants were guilty of the lesser-included motor vehicle offenses. As noted, however, a jury may not be so instructed.
Ibid.
In this case, neither the State nor defendants requested that reckless and careless driving be included in the court’s instructions. Because this alleged error was not raised at trial, we can
reverse only if plain error exists, that is, if the error “is clearly capable of producing an unjust result.”
R.
2:10-2.
We are unconvinced that under the plain error standard the charge as given was clearly capable of bringing about an unjust result or, indeed, that the jury’s verdict would have been different had the charge been amplified. In light of all the testimony actually presented to the jury, and given the circumstances that it was a dark and rainy night, calling for greater caution as a matter of ordinary common sense, there is little likelihood that the jury convicted both defendants of the death-by-auto charge because it harbored the feeling that either was guilty only of careless driving or reckless driving unrelated to the death of Dimitri, and that the convictions for the indictable offenses were compromise verdicts actuated by a desire to avoid acquitting defendants entirely.
The trial court’s belief that with a better instruction the result might have been different and more favorable for Brown was not expressed in terms of the stringent plain error standard. Its reasoning seemingly was influenced more by its ultimate decision that the defendants should have been tried separately than by any genuine and serious deficiency in the charge itself. The instructions as given do not convince us that the omission of an amplified charge reached the level of plain error.
VI.
The trial court based its grant of new and separate trials on the irreconcilability of defenses and the consequent prejudice attributable to the joint trial of defendants. The court’s decision was supported and explained in terms of its “feel of the case.” The trial court’s feelings about the case, however, must reflect a clear and correct determination that such prejudice outweighs the benefits accruing from a joint trial. We fully respect the trial court’s conscientious assessment of this case and the exceptionally lucid and complete articulation of its
perceptions. We also understand and appreciate the deference given by the Appellate Division, despite its own contrary reservations, to the judgment of the trial court. We find, however, that severance and new trials were not required here because of the irreconcilability of defenses or any resultant prejudice considered separately or cumulatively.
The trial court exercised sound judgment, great skill and firm control in presiding over a difficult trial. It need not, in the final analysis, have doubted its success in navigating the litigants, the attorneys and the jury to a result that was sufficiently free of error and prejudice to withstand post-trial and appellate attack.
Accordingly, we conclude that the jury’s verdict under all the circumstances should not be supplanted. The judgment of the Appellate Division is reversed and the judgments of conviction are reinstated.
For reversal and reinstatement
— Justices CLIFFORD, HANDLER, POLLOCK, O’HERN, GARIBALDI and STEIN — 6.
Opposed
— None.