STATE OF NEW JERSEY, PLAINTIFF-RESPONDENT,
GLADYS KELLY, DEFENDANT-APPELLANT.
The Supreme Court of New Jersey.
 Sheri Woliver, Assistant Deputy Public Defender, argued the cause for appellant (Joseph H. Rodriguez, Public Defender, attorney).
 Hilary L. Brunell, Assistant Prosecutor, argued the cause for respondent (George L. Schneider, Essex County Prosecutor, attorney).
Elizabeth M. Schneider, a member of the New York bar, argued the cause for amici curiae American Civil Liberties Union of New Jersey and New Jersey Coalition for Battered Women (Frank Askin and Stephen M. Latimer, attorneys).
Nadine Taub submitted a brief on behalf of amicus curiae American Psychological Association (Nadine Taub, attorney; Kit Kinports and Bruce J. Ennis, members of the District of Columbia bar, and Donald N. Bersoff, a member of the Maryland bar, of counsel).
The opinion of the Court was delivered by WILENTZ, C.J.
The central issue before us is whether expert testimony about the battered-woman's syndrome is admissible to help establish a claim of self-defense in a homicide case. The question is one of first impression in this state. We hold, based on the limited record before us (the State not having had a full opportunity to prove the contrary), that the battered-woman's syndrome is an appropriate subject for expert testimony; that the experts' conclusions, despite the relative newness of the field, are sufficiently reliable under New Jersey's standards for scientific testimony; and that defendant's expert was sufficiently qualified. Accordingly, we reverse and remand for a new trial. If on retrial after a full examination of these issues the evidence continues to support these conclusions, the expert's testimony on the battered-woman's syndrome shall be admitted as relevant to the honesty and reasonableness of defendant's belief that deadly force was necessary to protect her against death or serious bodily harm.
On May 24, 1980, defendant, Gladys Kelly, stabbed her husband, Ernest, with a pair of scissors. He died shortly thereafter at a nearby hospital. The couple had been married  for seven years, during which time Ernest had periodically attacked Gladys. According to Ms. Kelly, he assaulted her that afternoon, and she stabbed him in self-defense, fearing that he would kill her if she did not act.
Ms. Kelly was indicted for murder. At trial, she did not deny stabbing her husband, but asserted that her action was in self-defense. To establish the requisite state of mind for her self-defense claim, Ms. Kelly called Dr. Lois Veronen as an expert witness to testify about the battered-woman's syndrome. After hearing a lengthy voir dire examination of Dr. Veronen, the trial court ruled that expert testimony concerning the syndrome was inadmissible on the self-defense issue under State v. Bess, 53 N.J. 10 (1968). Apparently the court believed that the sole purpose of this testimony was to explain and justify defendant's perception of the danger rather than to show the objective reasonableness of that perception.
Ms. Kelly was convicted of reckless manslaughter. In an unreported decision relying in part on Bess, the Appellate Division affirmed the conviction. We granted certification, 91 N.J. 539 (1983), and now reverse.
Defendant raises six issues on appeal. She claims: (1) that the trial court erred in excluding expert testimony on the battered-woman's syndrome; (2) that the trial court's charge on provocation was erroneous; (3) that the trial court erred in excluding testimony that Mr. Kelly had sexually assaulted one of Ms. Kelly's daughters; (4) that improper prosecutorial conduct caused her to be denied a fair trial; (5) that the trial court erred in admitting testimony about her earlier conspiracy conviction; and (6) that her sentence was excessive.
The Kellys had a stormy marriage. Some of the details of their relationship, especially the stabbing, are disputed. The following is Ms. Kelly's version of what happened — a version that the jury could have accepted and, if they had, a version  that would make the proffered expert testimony not only relevant, but critical.
The day after the marriage, Mr. Kelly got drunk and knocked Ms. Kelly down. Although a period of calm followed the initial attack, the next seven years were accompanied by periodic and frequent beatings, sometimes as often as once a week. During the attacks, which generally occurred when Mr. Kelly was drunk, he threatened to kill Ms. Kelly and to cut off parts of her body if she tried to leave him. Mr. Kelly often moved out of the house after an attack, later returning with a promise that he would change his ways. Until the day of the homicide, only one of the attacks had taken place in public.
The day before the stabbing, Gladys and Ernest went shopping. They did not have enough money to buy food for the entire week, so Ernest said he would give his wife more money the next day.
The following morning he left for work. Ms. Kelly next saw her husband late that afternoon at a friend's house. She had gone there with her daughter, Annette, to ask Ernest for money to buy food. He told her to wait until they got home, and shortly thereafter the Kellys left. After walking past several houses, Mr. Kelly, who was drunk, angrily asked "What the hell did you come around here for?" He then grabbed the collar of her dress, and the two fell to the ground. He choked her by pushing his fingers against her throat, punched or hit her face, and bit her leg.
A crowd gathered on the street. Two men from the crowd separated them, just as Gladys felt that she was "passing out" from being choked. Fearing that Annette had been pushed around in the crowd, Gladys then left to look for her. Upon finding Annette, defendant noticed that Annette had defendant's pocketbook. Gladys had dropped it during the fight. Annette had retrieved it and gave her mother the pocketbook.
After finding her daughter, Ms. Kelly then observed Mr. Kelly running toward her with his hands raised. Within seconds  he was right next to her. Unsure of whether he had armed himself while she was looking for their daughter, and thinking that he had come back to kill her, she grabbed a pair of scissors from her pocketbook. She tried to scare him away, but instead stabbed him.
The central question in this case is whether the trial court erred in its exclusion of expert testimony on the battered-woman's syndrome. That testimony was intended to explain defendant's state of mind and bolster her claim of self-defense. We shall first examine the nature of the battered-woman's syndrome and then consider the expert testimony proffered in this case and its relevancy.
In the past decade social scientists and the legal community began to examine the forces that generate and perpetuate wife beating and violence in the family. What has been revealed is  that the problem affects many more people than had been thought and that the victims of the violence are not only the battered family members (almost always either the wife or the children). There are also many other strangers to the family who feel the devastating impact, often in the form of violence, of the psychological damage suffered by the victims.
Due to the high incidence of unreported abuse (the FBI and other law enforcement experts believe that wife abuse is the most unreported crime in the United States), estimates vary of the number of American women who are beaten regularly by their husband, boyfriend, or the dominant male figure in their lives. One recent estimate puts the number of women beaten yearly at over one million. See California Advisory Comm'n on Family Law, Domestic Violence app. F at 119 (1st report 1978). The state police statistics show more than 18,000 reported cases of domestic violence in New Jersey during the first nine months of 1983, in 83% of which the victim was female. It is clear that the American home, once assumed to be the cornerstone of our society, is often a violent place.
While common law notions that assigned an inferior status to women, and to wives in particular, no longer represent the state  of the law as reflected in statutes and cases, many commentators assert that a bias against battered women still exists, institutionalized in the attitudes of law enforcement agencies unwilling to pursue or uninterested in pursuing wife beating cases. See Comment, The Battered Wife's Dilemma: Kill or be Killed, 32 Hastings L.J., 895, 897-911 (1981).
Another problem is the currency enjoyed by stereotypes and myths concerning the characteristics of battered women and their reasons for staying in battering relationships. Some popular misconceptions about battered women include the beliefs that they are masochistic and actually enjoy their beatings, that they purposely provoke their husbands into violent behavior, and, most critically, as we shall soon see, that women who remain in battering relationships are free to leave their abusers at any time. See L. Walker, The Battered Woman at 19-31 (1979).
As these cases so tragically suggest, not only do many women suffer physical abuse at the hands of their mates, but a significant number of women kill (or are killed by) their husbands. In 1978, murders between husband and wife or girlfriend and boyfriend constituted 13% of all murders committed in the United States. Undoubtedly some of these arose from battering incidents. Federal Bureau of Investigation, Crime in the United States 1978 (1978). Men were the victims in 48% of these killings. Id.
As the problem of battered women has begun to receive more attention, sociologists and psychologists have begun to focus on the effects a sustained pattern of physical and psychological  abuse can have on a woman. The effects of such abuse are what some scientific observers have termed "the battered-woman's syndrome," a series of common characteristics that appear in women who are abused physically and psychologically over an extended period of time by the dominant male figure in their lives. Dr. Lenore Walker, a prominent writer on the battered-woman's syndrome, defines the battered woman as one
who is repeatedly subjected to any forceful physical or psychological behavior by a man in order to coerce her to do something he wants her to do without concern for her rights. Battered women include wives or women in any form of intimate relationships with men. Furthermore, in order to be classified as a battered woman, the couple must go through the battering cycle at least twice. Any woman may find herself in an abusive relationship with a man once. If it occurs a second time, and she remains in the situation, she is defined as a battered woman. [L. Walker, supra, at xv].
According to Dr. Walker, relationships characterized by physical abuse tend to develop battering cycles. Violent behavior directed at the woman occurs in three distinct and repetitive stages that vary both in duration and intensity depending on the individuals involved. L. Walker, supra, at 55-70.
Phase one of the battering cycle is referred to as the "tension-building stage," during which the battering male engages in minor battering incidents and verbal abuse while the woman, beset by fear and tension, attempts to be as placating and passive as possible in order to stave off more serious violence. Id. at 56-59.
Phase two of the battering cycle is the "acute battering incident." At some point during phase one, the tension between the battered woman and the batterer becomes intolerable and more serious violence inevitable. The triggering event that initiates phase two is most often an internal or external event in the life of the battering male, but provocation for more severe violence is sometimes provided by the woman who can no longer tolerate or control her phase-one anger and anxiety. Id. at 59-65.
Phase three of the battering cycle is characterized by extreme contrition and loving behavior on the part of the battering  male. During this period the man will often mix his pleas for forgiveness and protestations of devotion with promises to seek professional help, to stop drinking, and to refrain from further violence. For some couples, this period of relative calm may last as long as several months, but in a battering relationship the affection and contrition of the man will eventually fade and phase one of the cycle will start anew. Id. at 65-70.
The cyclical nature of battering behavior helps explain why more women simply do not leave their abusers. The loving behavior demonstrated by the batterer during phase three reinforces whatever hopes these women might have for their mate's reform and keeps them bound to the relationship. R. Langley & R. Levy, Wife Beating: The Silent Crisis 112-14 (1977).
Some women may even perceive the battering cycle as normal, especially if they grew up in a violent household. Battered Women, A Psychosociological Study of Domestic Violence 60 (M. Roy ed. 1977); D. Martin, Battered Wives, 60 (1981). Or they may simply not wish to acknowledge the reality of their situation. T. Davidson, Conjugal Crime, at 50 (1978) ("The middle-class battered wife's response to her situation tends to be withdrawal, silence and denial ...").
Other women, however, become so demoralized and degraded by the fact that they cannot predict or control the violence that they sink into a state of psychological paralysis and become unable to take any action at all to improve or alter the situation. There is a tendency in battered women to believe in the omnipotence  or strength of their battering husbands and thus to feel that any attempt to resist them is hopeless. L. Walker, supra, at 75.
In addition to these psychological impacts, external social and economic factors often make it difficult for some women to extricate themselves from battering relationships. A woman without independent financial resources who wishes to leave her husband often finds it difficult to do so because of a lack of material and social resources.
Even with the progress of the last decade, women typically make less money and hold less prestigious jobs than men, and are more responsible for child care. Thus, in a violent confrontation where the first reaction might be to flee, women realize soon that there may be no place to go. Moreover, the stigma that attaches to a woman who leaves the family unit without her children undoubtedly acts as a further deterrent to moving out.
In addition, battered women, when they want to leave the relationship, are typically unwilling to reach out and confide in their friends, family, or the police, either out of shame and humiliation, fear of reprisal by their husband, or the feeling they will not be believed.
Dr. Walker and other commentators have identified several common personality traits of the battered woman: low self-esteem, traditional beliefs about the home, the family, and the female sex role, tremendous feelings of guilt that their marriages are failing, and the tendency to accept responsibility for the batterer's actions. L. Walker, supra, at 35-36.
Finally, battered women are often hesitant to leave a battering relationship because, in addition to their hope of reform on the part of their spouse, they harbor a deep concern about the possible response leaving might provoke in their mates. They literally become trapped by their own fear. Case histories are replete with instances in which a battered wife left her husband  only to have him pursue her and subject her to an even more brutal attack. D. Martin, supra, at 76-79.
The combination of all these symptoms — resulting from sustained psychological and physical trauma compounded by aggravating social and economic factors — constitutes the battered-woman's syndrome. Only by understanding these unique pressures that force battered women to remain with their mates, despite their long-standing and reasonable fear of severe bodily harm and the isolation that being a battered woman creates, can a battered woman's state of mind be accurately and fairly understood.
The voir dire testimony of Dr. Veronen, sought to be introduced by defendant Gladys Kelly, conformed essentially to this outline of the battered-woman's syndrome. Dr. Vernonen, after establishing her credentials, described in general terms the component parts of the battered-woman's syndrome and its effects on a woman's physical and mental health. The witness then documented, based on her own considerable experience in counseling, treating, and studying battered women, and her familiarity with the work of others in the field, the feelings of anxiety, self-blame, isolation, and, above all, fear that plagues these women and leaves them prey to a psychological paralysis that hinders their ability to break free or seek help.
Dr. Veronen stated that the problems of battered women are aggravated by a lack of understanding among the general public concerning both the prevalence of violence against women and the nature of battering relationships. She cited several myths concerning battered women that enjoy popular acceptance — primarily that such women are masochistic and enjoy the abuse they receive and that they are free to leave their husbands but choose not to.
Dr. Veronen described the various psychological tests and examinations she had performed in connection with her independent research. These tests and their methodology, including their interpretation, are, according to Dr. Veronen, widely  accepted by clinical psychologists. Applying this methodology to defendant (who was subjected to all of the tests, including a five-hour interview), Dr. Veronen concluded that defendant was a battered woman and subject to the battered-woman's syndrome.
In addition, Dr. Veronen was prepared to testify as to how, as a battered woman, Gladys Kelly perceived her situation at the time of the stabbing, and why, in her opinion, defendant did not leave her husband despite the constant beatings she endured.
Whether expert testimony on the battered-woman's syndrome should be admitted in this case depends on whether it is relevant to defendant's claim of self-defense, and, in any event, on whether the proffer meets the standards for admission of expert testimony in this state. We examine first the law of self-defense and consider whether the expert testimony is relevant.
The present rules governing the use of force in self-defense are set out in the justification section of the Code of Criminal Justice. The use of force against another in self-defense is justifiable "when the actor reasonably believes that such force is immediately necessary for the purpose of protecting himself against the use of unlawful force by such other person on the present occasion." N.J.S.A. 2C:3-4(a). Further limitations exist when deadly force is used in self-defense. The use of such deadly force is not justifiable
unless the actor reasonably believes that such force is necessary to protect himself against death or serious bodily harm.... [N.J.S.A. 2C:3-4(b)(2)].
These principles codify decades of prior case law development of the elements of self-defense. We focus here on the critical  requirement that the actor reasonably believe deadly force to be necessary to prevent death or serious bodily harm, for the proffer of expert testimony was argued to be relevant on this point.
Self-defense exonerates a person who kills in the reasonable belief that such action was necessary to prevent his or her death or serious injury, even though this belief was later proven mistaken. "Detached reflection cannot be demanded in the presence of an uplifted knife," Justice Holmes aptly said, Brown v. United States, 256 U.S. 335, 343, 41 S.Ct. 501, 502, 65 L.Ed. 961, 963 (1921); and the law accordingly requires only a reasonable, not necessarily a correct, judgment. See State v. Hipplewith, 33 N.J. 300, 316-17 (1960); State v. Mount, 73 N.J.L. 582, 583 (E. & A. 1905); State v. Lionetti, 93 N.J.L. 24 (Sup.Ct. 1919).
While it is not imperative that actual necessity exist, a valid plea of self-defense will not lie absent an actual (that is, honest) belief on the part of the defendant in the necessity of using force. While no case in New Jersey has addressed the point directly, the privilege of self-defense does not exist where the defendant's action is not prompted by a belief in its necessity: "He has no defense when he intentionally kills his enemy in complete ignorance of the fact that his enemy, when killed, was about to launch a deadly attack upon him." W. LaFave & A. Scott, Criminal Law § 53, at 394 (1972). The intent of the  drafters of the present Code was that a necessity to act should not give rise to a meritorious plea of self-defense where the defendant was unaware of that necessity. Final Report of the New Jersey Criminal Law Revision Commission, Vol. II: Commentary, at 83 (1971) [hereinafter cited as Commission Report]. Ultimately, of course, it is for the jury to determine if the defendant actually did believe in the necessity of acting with deadly force to prevent an imminent, grave attack. See, e.g., State v. Fair, 45 N.J. 77, 93 (1965).
Honesty alone, however, does not suffice. A defendant claiming the privilege of self-defense must also establish that her belief in the necessity to use force was reasonable. See, e.g., State v. Mellillo, 77 N.J.L. 505 (E. & A. 1908); State v. Mark Len, 108 N.J.L. 439, 440 (Sup.Ct. 1932). As originally proposed, the new Code of Criminal Justice would have eliminated the reasonableness requirement, allowing self-defense whenever the defendant honestly believed in the imminent need to act. See Commission Report, supra, Vol. I, at 26-27 (proposed Section 2C:3-4), and Vol. II: Commentary, at 82-83. This proposed change in the law was not accepted by the Legislature. N.J.S.A. 2C:3-4 as finally enacted retains the requirement that the defendant's belief be reasonable.
Thus, even when the defendant's belief in the need to kill in self-defense is conceded to be sincere, if it is found to have been unreasonable under the circumstances, such a belief cannot be held to constitute complete justification for a homicide. As  with the determination of the existence of the defendant's belief, the question of the reasonableness of this belief "is to be determined by the jury, not the defendant, in light of the circumstances existing at the time of the homicide." State v. Hipplewith, supra, 33 N.J. at 316; see State v. Bess, supra, 53 N.J. at 16; State v. Fair, supra, 45 N.J. at 93; State v. Jayson, 94 N.J.L. 467, 471 (E. & A. 1920). It is perhaps worth emphasizing here that for defendant to prevail, the jury need not find beyond a reasonable doubt that the defendant's belief was honest and reasonable. Rather, if any evidence raising the issue of self-defense is adduced, either in the State's or the defendant's case, then the jury must be instructed that the State is required to prove beyond a reasonable doubt that the self-defense claim does not accord with the facts; acquittal is required if there remains a reasonable doubt whether the defendant acted in self-defense. State v. Abbott, 36 N.J. 63, 72 (1961). See generally State v. Chiarello, 69 N.J. Super. 479 (App.Div. 1961).
With the foregoing standards in mind, we turn to an examination of the relevance of the proffered expert testimony to Gladys Kelly's claim of self-defense.
Gladys Kelly claims that she stabbed her husband in self-defense, believing he was about to kill her. The gist of the State's case was that Gladys Kelly was the aggressor, that she consciously intended to kill her husband, and that she certainly was not acting in self-defense.
The credibility of Gladys Kelly is a critical issue in this case. If the jury does not believe Gladys Kelly's account, it  cannot find she acted in self-defense. The expert testimony offered was directly relevant to one of the critical elements of that account, namely, what Gladys Kelly believed at the time of the stabbing, and was thus material to establish the honesty of her stated belief that she was in imminent danger of death.
The State argues that there is no need to bolster defendant's credibility with expert testimony concerning the battering because the State did not attempt to undermine defendant's testimony concerning her prior mistreatment at the hands of her husband. The State's claim is simply untrue. In her summation, the prosecutor suggested that had Ernest Kelly lived, he might have told a different story from the one Gladys told. (In its brief, the State argues that evidence in the case suggests that Gladys Kelly's claims of abuse could have been contradicted by her husband.) This is obviously a direct attempt to undermine defendant's testimony about her prior mistreatment.
Moreover, defendant's credibility was also attacked in other ways. Gladys Kelly's prior conviction for conspiracy to commit robbery was admitted into evidence for the express purpose of impeachment, even though this conviction had occurred nine years before the stabbing. Other questions, about Gladys Kelly's use of alcohol and drugs and about her premarital sexual conduct, were clearly efforts to impeach credibility.
As can be seen from our discussion of the expert testimony, Dr. Veronen would have bolstered Gladys Kelly's credibility.  Specifically, by showing that her experience, although concededly difficult to comprehend, was common to that of other women who had been in similarly abusive relationships, Dr. Veronen would have helped the jury understand that Gladys Kelly could have honestly feared that she would suffer serious bodily harm from her husband's attacks, yet still remain with him. This, in turn, would support Ms. Kelly's testimony about her state of mind (that is, that she honestly feared serious bodily harm) at the time of the stabbing.
On the facts in this case, we find that the expert testimony was relevant to Gladys Kelly's state of mind, namely, it was admissible to show she honestly believed she was in imminent danger of death. Ibn-Tamas v. United States, 407 A.2d 626 (D.C. 1979) (on remand, trial court excluded expert testimony on battered-woman's syndrome; the Court of Appeals affirmed the exclusion of the testimony, holding that the trial court was not compelled to admit the evidence; 455 A.2d 893 (D.C. 1983)); Hawthorne v. State, 408 So.2d 801 (Fla. Dist. Ct. App. 1982), petition for review denied, 415 So.2d 1361 (Fla. 1982); Smith v. State, 247 Ga. 612, 277 S.E.2d 678 (1981); State v. Anaya, 438 A.2d 892 (Me. 1981); State v. Allery, 101 Wash.2d 591, 682 P.2d 312 (Wash.Sup.Ct. 1984); see also People v. Minnis, 118 Ill. App.3d 345, 74 Ill.Dec. 179, 455 N.E.2d 209 (1983) (expert testimony on battered-woman's syndrome admissible to explain reasons why defendant dismembered body of victim/husband where prosecution introduced fact of dismemberment as substantive evidence of guilt). But see State v. Thomas, 66 Ohio St.2d 518, 423 N.E.2d 137 (1981). Moreover, we  find that because this testimony was central to the defendant's claim of self-defense, its exclusion, if otherwise admissible, cannot be held to be harmless error.
 We also find the expert testimony relevant to the reasonableness of defendant's belief that she was in imminent danger of death or serious injury. We do not mean that the expert's testimony could be used to show that it was understandable that a battered woman might believe that her life was in danger when indeed it was not and when a reasonable person would not have so believed, for admission for that purpose would clearly violate the rule set forth in State v. Bess, supra, 53 N.J. 10. Expert testimony in that direction would be relevant solely to the honesty of defendant's belief, not its objective reasonableness. Rather, our conclusion is that the expert's testimony, if accepted by the jury, would have aided it in determining whether, under the circumstances, a reasonable person would have believed there was imminent danger to her life.
At the heart of the claim of self-defense was defendant's story that she had been repeatedly subjected to "beatings" over the course of her marriage. While defendant's testimony was somewhat lacking in detail, a juror could infer from the use of the word "beatings," as well as the detail given concerning some of these events (the choking, the biting, the use of fists),  that these physical assaults posed a risk of serious injury or death. When that regular pattern of serious physical abuse is combined with defendant's claim that the decedent sometimes threatened to kill her, defendant's statement that on this occasion she thought she might be killed when she saw Mr. Kelly running toward her could be found to reflect a reasonable fear; that is, it could so be found if the jury believed Gladys Kelly's story of the prior beatings, if it believed her story of the prior threats, and, of course, if it believed her story of the events of that particular day.
The crucial issue of fact on which this expert's testimony would bear is why, given such allegedly severe and constant beatings, combined with threats to kill, defendant had not long ago left decedent. Whether raised by the prosecutor as a factual issue or not, our own common knowledge tells us that most of us, including the ordinary juror, would ask himself or herself just such a question. And our knowledge is bolstered by the experts' knowledge, for the experts point out that one of the common myths, apparently believed by most people, is that battered wives are free to leave. To some, this misconception is followed by the observation that the battered wife is masochistic, proven by her refusal to leave despite the severe beatings; to others, however, the fact that the battered wife stays on unquestionably suggests that the "beatings" could not have been too bad for if they had been, she certainly would have left. The expert could clear up these myths, by explaining that one of the common characteristics of a battered wife is her inability to leave despite such constant beatings; her "learned helplessness"; her lack of anywhere to go; her feeling that if she tried to leave, she would be subjected to even more merciless treatment; her belief in the omnipotence of her battering husband; and sometimes her hope that her husband will change his ways.
Unfortunately, in this case the State reinforced the myths about battered women. On cross-examination, when discussing an occasion when Mr. Kelly temporarily moved out of the  house, the State repeatedly asked Ms. Kelly: "You wanted him back, didn't you?" The implication was clear: domestic life could not have been too bad if she wanted him back. In its closing argument, the State trivialized the severity of the beatings, saying:
I'm not going to say they happened or they didn't happen, but life isn't pretty. Life is not a bowl of cherries. We each and every person who takes a breath has problems. Defense counsel says bruised and battered. Is there any one of us who hasn't been battered by life in some manner or means?
Even had the State not taken this approach, however, expert testimony would be essential to rebut the general misconceptions regarding battered women.
The difficulty with the expert's testimony is that it sounds as if an expert is giving knowledge to a jury about something the jury knows as well as anyone else, namely, the reasonableness of a person's fear of imminent serious danger. That is not at all, however, what this testimony is directly aimed at. It is aimed at an area where the purported common knowledge of the jury may be very much mistaken, an area where jurors' logic, drawn from their own experience, may lead to a wholly incorrect conclusion, an area where expert knowledge would enable the jurors to disregard their prior conclusions as being common myths rather than common knowledge. After hearing the expert, instead of saying Gladys Kelly could not have been beaten up so badly for if she had, she certainly would have left, the jury could conclude that her failure to leave was very much part and parcel of her life as a battered wife. The jury could conclude that instead of casting doubt on the accuracy of her testimony about the severity and frequency of prior beatings, her failure to leave actually reinforced her credibility.
Since a retrial is necessary, we think it advisable to indicate the limit of the expert's testimony on this issue of reasonableness. It would not be proper for the expert to express the opinion that defendant's belief on that day was reasonable, not because this is the ultimate issue, but because the area of expert knowledge relates, in this regard, to the reasons for  defendant's failure to leave her husband. Either the jury accepts or rejects that explanation and, based on that, credits defendant's stories about the beatings she suffered. No expert is needed, however, once the jury has made up its mind on those issues, to tell the jury the logical conclusion, namely, that a person who has in fact been severely and continuously beaten might very well reasonably fear that the imminent beating she was about to suffer could be either life-threatening or pose a risk of serious injury. What the expert could state was that defendant had the battered-woman's syndrome, and could explain that syndrome in detail, relating its characteristics to defendant, but only to enable the jury better to determine the honesty and reasonableness of defendant's belief. Depending on its content, the expert's testimony might also enable the jury to find that the battered wife, because of the prior beatings, numerous beatings, as often as once a week, for seven years, from the day they were married to the day he died, is particularly able to predict accurately the likely extent of violence in any attack on her. That conclusion could significantly affect the jury's evaluation of the reasonableness of defendant's fear for her life.
Having determined that testimony about the battered-woman's syndrome is relevant, we now consider whether Dr. Veronen's testimony satisfies the limitations placed on expert testimony by Evidence Rule 56(2) and by applicable case law. See State v. Cavallo, 88 N.J. 508, 516 (1982). Evidence Rule 56(2) provides that an expert may testify "as to matters requiring scientific, technical or other specialized knowledge if such testimony will assist the trier of fact to understand the evidence or determine a fact in issue." In effect, this Rule imposes three basic requirements for the admission of expert testimony: (1) the intended testimony must concern a subject matter that is beyond the ken of the average juror; (2) the field testified to must be at a state of the art such that an expert's testimony could be sufficiently reliable; and (3) the witness must have sufficient expertise to offer the intended testimony. See N.J. Rules of Evidence (Anno. 1984), Comment 5 to Evid.R. 56.
 The primary justification for permitting expert testimony is that the average juror is relatively helpless in dealing with a subject that is not a matter of common knowledge. Angel v. Rand Express Lines, Inc., 66 N.J. Super. 77, 85 (App.Div. 1961). Thus, the proponent of expert testimony must demonstrate that testimony would "enhance the knowledge and understanding of lay jurors with respect to other testimony of a special nature normally outside of the usual lay sphere." State v. Griffin, 120 N.J. Super. 13, 20 (App.Div. 1972).
As previously discussed, a battering relationship embodies psychological and societal features that are not well understood by lay observers. Indeed, these features are subject to a large group of myths and stereotypes. It is clear that this subject is beyond the ken of the average juror and thus is suitable for explanation through expert testimony.
The second requirement that must be met before expert testimony is permitted is a showing that the proposed expert's testimony would be reliable. The rationale for this requirement is that expert testimony seeks to assist the trier of fact. An expert opinion that is not reliable is of no assistance to anyone.
To meet the requirement that the expert's testimony be sufficiently reliable, defense counsel must show that the testimony satisfies New Jersey's standard of acceptability for scientific  evidence. State v. Cavallo, supra, 88 N.J. at 516-17 (citing State v. Hurd, 86 N.J. 525, 536 (1981)). The technique or mode of analysis used by the expert must have a sufficient scientific basis to produce uniform and reasonably reliable results so as to contribute materially to the ascertainment of the truth. Id. 88 N.J. at 517 (citing State v. Cary, 49 N.J. 343, 352 (1967)); State v. Hurd, supra, 86 N.J. at 536.
In a relatively new field of research, such as that of the battered-woman's syndrome, there are three ways a proponent of scientific evidence can prove its general acceptance and thereby its reliability: (1) by expert testimony as to the general acceptance, among those in the profession, of the premises on which the proffered expert witness based his or her analysis; (2) by authoritative scientific and legal writings indicating that the scientific community accepts the premises underlying the proffered testimony; and (3) by judicial opinions that indicate the expert's premises have gained general acceptance. State v. Cavallo, 88 N.J. at 521. Applying those methods to the case at bar, we note that judicial opinions thus far have been split concerning the scientific acceptability of the syndrome and the methodology used by the researchers in this area. On the other hand, Dr. Veronen, the proffered expert, testified that the battered-woman's syndrome is acknowledged and accepted by practitioners and professors in the fields of psychology and psychiatry. Dr. Veronen also brought to the court's attention the findings of several researchers who have published reports confirming the presence of the battered-woman's syndrome. She further noted that the battered-woman's syndrome has  been discussed at several symposia since 1977, sponsored by such organizations as the Association for the Advancement of Behavior Therapy and the American Sociological Association. Briefs submitted to this Court indicate that there are at least five books and almost seventy scientific articles and papers about the battered-woman's syndrome.
Thus, the record before us reveals that the battered woman's syndrome has a sufficient scientific basis to produce uniform and reasonably reliable results as required by State v. Cavallo, and Evid.R. 56(2). The numerous books, articles and papers referred to earlier indicate the presence of a growing field of study and research about the battered woman's syndrome and recognition of the syndrome in the scientific field. However, while the record before us could require such a ruling, we refrain from conclusively ruling that Dr. Veronen's proffered testimony about the battered-woman's syndrome would satisfy New Jersey's standard of acceptability for scientific evidence. This is because the State was not given a full opportunity in the trial court to question Dr. Veronen's methodology in studying battered women or her implicit assertion that the battered-woman's syndrome has been accepted by the relevant scientific community.
Finally, before expert testimony may be presented, there must be a showing that the proffered expert witness has sufficient expertise to offer the intended testimony. State v. Cavallo, supra, 88 N.J. at 516. In this case, it appears that Dr. Veronen is qualified to testify as an expert. She has a Ph.D. in clinical psychology, as well as an M.A. from North Texas State. She is a member of four professional associations. As of 1980, when she was offered as a witness at Ms. Kelly's trial, Dr.  Veronen had been an assistant professor at the medical school at the University of South Carolina for three years. Twenty percent of her time at the Universty was spent teaching, some of it on topics related to the battered-woman's syndrome, and 80% of her time was spent conducting research, most of it on the psychological reaction of women who are victims of violent assaults. She had spent two years studying the battered-woman's syndrome, with the goal of changing the patterns of fear and anxiety of battered women. Dr. Veronen is a clinical psychologist, licensed to practice in two states, and in that capacity had, by 1980, treated approximately thirty battered women and seen seventy others. Because these thirty women have several important characteristics in common with Ms. Kelly (the thirty women had all been in battering relationships for more than two years, were beaten more than six times, and were within the same age group as Ms. Kelly), Dr. Veronen is familiar with battered women who share Ms. Kelly's background.
We have concluded that the appropriate disposal of this appeal is to reverse and remand for a new trial. On the record before us, although the trial court did not rule on the matter, it appears that Dr. Veronen qualified as an expert, and that the degree of reliability of the conclusions in this field of expertise was sufficient to allow their admission. Alternatively we could retain jurisdiction and remand, solely for the purpose of allowing the prosecutor to continue cross-examination of Dr.  Veronen as well as to introduce such contrary testimony as the prosecutor sees fit. The transcript discloses that the prosecutor had concluded her cross-examination on Dr. Veronen's qualifications but had never been given the opportunity fully to cross-examine the expert on the reliability of this developing field of scientific knowledge. The possibility of such further cross-examination was foreclosed by the trial court when it ruled evidence of the syndrome was inadmissible because irrelevant. Furthermore, as noted above, the trial court never actually ruled whether Dr. Veronen qualified as an expert, finding this unnecessary because of his holding that the testimony would not be admissible under State v. Bess, supra, 53 N.J. 10, even if she was an expert.
 Our conclusion, reversing and ordering a new trial, is based on the apparent unfairness in this case of the kind of limited remand that we ordered in State v. Sikora, supra, 44 N.J. at 465-66, 474 (Weintraub, C.J., concurring). Here a limited remand would be to the trial court to exercise its discretion, a very broad discretion, on the issue of the expert's qualifications and the reliability of the knowledge proffered. We do not know what conflicting expert testimony the prosecution would offer, but the entire scenario of a limited remand when the defendant has already been convicted and when the court whose discretion will largely determine the outcome of the limited remand has already excluded the evidence, with prosecution experts who might not have been called at the original trial, seems an artificial trial setting, and significantly less favorable to defendant than what might have occurred if the trial court had had the benefit of the views expressed herein at the time. Obviously there is no way to recreate the precise situation of the trial, but all things considered, we think fairness requires a new trial where all of these matters may be reconsidered.
Apart from her claims concerning the exclusion of the expert testimony, the defendant raises five additional issues on appeal. Although our disposition of this case makes it unnecessary to  consider these issues, we dispose of them briefly to assist the trial court in the event they surface again at the new trial.
During trial, defendant sought to introduce testimony from Edith Cannon, defendant's 17-year-old daughter by another marriage, to the effect that shortly before the fatal encounter she had told her mother that Ernest Kelly had been subjecting her to physical and sexual abuse since age 13. The defense asserted that this evidence of Glady Kelly's knowledge of the victim's prior aggressive behavior demonstrated that her fear of the decedent was justifiable and that her subsequent behavior was reasonable. See McCormick on Evidence § 249, at 588-89 (E. Cleary ed., 2d Ed. 1972); VI J. Wigmore Evidence § 1789, at 314 (Chad.Rev.Ed. 1972).
The trial court, however, excluded this evidence in reliance upon Evidence Rule 4, stating:
We will get involved with trials within trials — trying cases of sexual aggression. That daughter was not present at the time of the alleged stabbing by her mother of her stepfather. There has been no evidence indicating that the safety of the daughter was threatened on May 24.
Whether the probative value of a particular piece of evidence is outweighed by its potential prejudice is a decision normally left to the discretion of the trial court; and this "discretion is a broad one." State v. Sands, 76 N.J. 127 (1978); see also Evid.R. 4, Comment 1.
If the only relevance of this testimony was to reinforce the proof that defendant feared the decedent for good reason, its limited added force might very well be outweighed  by the obvious prejudice injected into the case in the form of proof that decedent sexually abused his daughter. The testimony, however, has further relevance in that it very strongly supports the conclusion that the Kelly household was the scene of the batterings that would produce the battered-woman's syndrome. As our Legislature noted in its findings included in the Prevention of Domestic Violence Act, "there is a positive correlation between spouse abuse and child abuse...." N.J.S.A. 2C:25-2. Given the critical importance of the proof of the battered-woman's syndrome in this case, we are inclined to believe that, on balance, such testimony should have been admitted. We are aware that in the context of an appellate review, a decision of a trial court must stand unless it can be shown that the trial court palpably abused its discretion, that is, that its finding was so wide of the mark that a manifest denial of justice resulted. State v. Carter, 91 N.J. 86, 106 (1982); State v. Boratto, 80 N.J. 506 (1979); State v. Rogers, 19 N.J. 218 (1955); Hill v. Newman, 126 N.J. Super. 557 (App.Div. 1973), certif. den., 64 N.J. 508 (1974); Evid.R. 4, Comment 2. Nevertheless, absent any significant new factor bearing on this issue, the trial court on remand should allow the testimony, giving such appropriate instruction to the jury as will minimize the possibility of its prejudicial impact.
Defense counsel also contends that the trial court erred in allowing the State to question defendant about her earlier conviction. Counsel asserts that the trial court "lost sight" of the grounds for admitting defendant's prior record. This claim is without merit.
Ms. Kelly was convicted of conspiracy to commit robbery in 1971, and over defense counsel's objection the trial court ruled that evidence of the earlier conviction was admissible. During cross-examination, the prosecution questioned Ms. Kelly about her earlier conviction:
 Q. Mrs. Kelly, have you ever been convicted of a crime?
A. Yes, once.
Q. What were you convicted of?
A. Conspiracy to robbery with some — two other peoples was involved —
Q. You were convicted of conspiracy to commit robbery?
Q. When was that?
A. Nine years ago, I think.
A. Something like that. I was given three years probation....
That was the only time during the two week trial that evidence as to Ms. Kelly's prior conviction was elicited or referred to.
Prior convictions ordinarily may be used to impeach the defendant's credibility. State v. Sands, 76 N.J. 127, 146 (1978); N.J.S.A. 2A:81-12. The trial court, recognizing that, instructed the jury as to the limited purpose for which it could consider Ms. Kelly's conspiracy conviction:
The only reason you heard that testimony was not because if you find that she committed a crime in 1971, therefore she must have committed this crime with which she is charged. The only reason you may use that if you wish to is to affect her believability as a witness. That is the sole and exclusive purpose of hearing that and using that evidence.
There was no error on this point.
We reject defendant's contention that the prosecutor's conduct denied the defendant her right to receive a fair trial. The defense claims that the prosecutor improperly used closing arguments to glorify her function as a prosecutor and make an inflammatory appeal to the jury, and used her opening statement  to suggest that Ms. Kelly's indictment was evidence of guilt. These complaints were not raised at trial, and thus need not be dealt with in the same way as those raised by a timely challenge. State v. Macon, 57 N.J. 325, 333 (1971). We note, however, that the trial court properly instructed the jury that the indictment is not proof of guilt, and our review of the closing statement does not reveal plain error. R.2:10-2. The prosecutor neither exalted her role at length, nor disparaged the role of defense counsel. See State v. Thornton, 38 N.J. 380 (1962), cert. denied, 374 U.S. 816, 83 S.Ct. 1710, 10 L.Ed.2d 1039 (1963). Nor did the prosecutor's closing remarks exceed the wide latitude permitted counsel during summation. See State v. Mayberry, 52 N.J. 413, 437 (1968), cert. denied, 393 U.S. 1043, 89 S.Ct. 673, 21 L.Ed.2d 593 (1969).
Defendant also claims that the prosecutor was too aggressive, asked improper questions about Ms. Kelly's personal life in an attempt to cast aspersions on defendant's moral character, and made too many objections, most of which were overruled. 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. See State v. Tirone, 64 N.J. 222, 229 (1974). There were sufficient facts on which the jury could base its finding of guilt on the reckless manslaughter charge. In light of the entire record, any impropriety that did occur was harmless and incapable of producing an unjust result. See State v. LaPorte, 62 N.J. 312 (1973); R. 2:10-2.
The defendant argues that the charge to the jury regarding provocation as an element of manslaughter was in error because it did not state that reasonable and sufficient provocation may arise from a course of ill treatment. We agree that the instructions on provocation were deficient. It is well settled that when there is evidence of prior physical abuse  of defendant by the decedent, the jury must be told that a finding of provocation may be premised on "a course of ill treatment which can induce a homicidal response in a person of ordinary firmness and which the accused reasonably believes is likely to continue." State v. Guido, 40 N.J. 191, 211 (1963). The jury must be instructed "to consider not only decedent's conduct and threats that night, but also his prior mistreatment of defendant." State v. Lamb, 71 N.J. 545, 551 (1976). On retrial, this aspect of the trial court's instruction should be changed.
Ms. Kelly also contends that the sentence imposed — five years in state prison — was excessive. She asserts that imprisonment would result in a serious injustice that overrides the need to deter such conduct by others, N.J.S.A. 2C:44-1(d), and that she should instead be granted probation or entry into a release program. She cites several mitigating factors, including her abuse at the hands of Mr. Kelly and her children's need to have their mother at home.
The presumptive sentence for a second degree crime is seven years. N.J.S.A. 2C:44-1(f)(1). In ordering a sentence of five years, the trial court agreed with defendant that there was a preponderance of mitigating factors, allowing it to sentence  her to a minimum term for a second degree crime. N.J.S.A. 2C:43-6(a)(2); 2C:44-1(f)(1). See State v. Roth, 95 N.J. 334, 359, 471 A.2d 370 (1984). Although we appreciate the hardship that would result from defendant's incarceration, she is not the truly extraordinary defendant whose imprisonment would represent the "serious injustice" envisioned by the Criminal Code. Roth, supra, 95 N.J. at 358.
HANDLER, J., concurring in part and dissenting in part.
The record in this case persuasively establishes the professional acceptance and scientific reliability of the clinical psychological condition referred to as the "battered women's syndrome." Therefore, I would rule that expert evidence of the battered women's syndrome is both competent and relevant as related to the defense of self-defense. Consequently, no further expert testimony or evidence concerning the admissibility  of this doctrine should be required on a retrial of this case. I would also allow into evidence on the retrial the testimony of defendant's expert that defendant was suffering battered women's syndrome when she killed her husband. That testimony was unquestionably relevant to defendant's claim of self-defense. In addition, the evidence in this case indicates that repeated sexual and physical victimization of a woman's children may, in conjunction with her own abused treatment, contribute to the development of battered women's syndrome. I therefore concur in the majority's determination to allow on a retrial evidence of the decedent's sexual assaults upon defendant's daughter as related to the issue of the battered women's syndrome and defendant's defense of self-defense.
The Court in this case takes a major stride in recognizing the scientific authenticity of the battered women's syndrome and its legal and factual significance in the trial of certain criminal cases. My difference with the Court is quite narrow. I believe that defendant Gladys Kelly has demonstrated at her trial by sufficient expert evidence her entitlement to the use of the battered women's syndrome in connection with her defense of self-defense. I would therefore not require this issue — the admissibility of the battered women's syndrome — to be tried again.
This Court's opinion presents a cogent and thorough explanation of the perplexing and tragic condition of the battered women's syndrome. This condition refers to a congeries of common traits in women who are subjected to prolonged physical and psychological abuse by their mates. Women suffering battered women's syndrome have low self-esteem, strong feelings of personal guilt over their failing marriages, and self-blame for the violence that their mates inflict upon them. Ante at 195-196, citing L. Walker, The Battered Woman 35-36 (1979) (Walker). Typically, such battered women are dominated  by unshakeable fear, which often traps them into remaining with their battering mates. Id., citing D. Martin, Battered Wives 76-79 (1981) (Martin). Victims of battered women's syndrome frequently become so demoralized and degraded that they lapse into a psychological torpor, a state of "learned helplessness." Ante at 194-195, citing Walker, supra, at 75.
The relationships that typify the syndrome usually involve cyclical behavior. One recurrent phase of the cycle includes a period of contrite behavior by the batterer, which reinforces the illusion of these victimized women that their mates will change and reform, further binding them to the relationship. Ante at 193, citing Walker, supra, at 55-70; R. Langley & R. Levy, Wife Beating: The Silent Crisis 112-14 (1977). Many battered women perceive the battering cycle as commonplace, and refuse to acknowledge the abnormality of their plight. Ante at 194, citing T. Davidson, Conjugal Crime (1978); Battered Women, A Psychosociological Study of Domestic Violence 60 (M. Roy ed. 1977); Martin, supra, at 60.
The Court's opinion explains that the abusive pattern that characterizes this syndrome is a phenomenon that puzzles and confuses the untutored lay person. The violence common to the syndrome is the subject of widespread ignorance and misinformation. It has spawned myths as to its causes and distorted stereotypes of its victims. Ante at 192. Some common misconceptions about battered women include the beliefs that they are masochistic and actually enjoy their physical and psychological suffering, that they purposely provoke their mates into violent behavior and, most critically, that women who remain in battering relationships are free to leave their abusers at any time. Id., citing Walker, supra, at 19-31.
This Court's enlightened exposition of the battered women's syndrome, drawn from the record in this case lays a firm foundation for a determination of the admissibility of expert testimony relating to the syndrome in the trial of particular  criminal causes under the Code of Criminal Justice, N.J.S.A. 2C:1-1 et seq., and our rules of evidence.
Evidence Rule 56(2) provides that an expert may testify "as to matters requiring scientific, technical or other specialized knowledge if such testimony will assist the trier of fact to understand the evidence or determine a fact in issue." In effect, this rule imposes three basic requirements for the admission of expert testimony: (1) the intended testimony must concern a subject matter that is beyond the ken of the average juror; (2) the field testified to must be at a state of the art such that an expert's testimony could be sufficiently reliable; and (3) the witness must have sufficient expertise to offer the intended testimony. See N.J. Rules of Evidence (Anno. 1984), Comment 5 to Evid.R. 56; State v. Cavallo, 88 N.J. 508 (1982); State v. Hurd, 86 N.J. 525 (1981).
The first criterion for the admission of expert testimony under Evidence Rule 56(2) is that the subject matter is fully comprehended primarily by experts, persons who have special training and education in the particular field. Correlatively, the subject matter ordinarily falls beyond the common understanding of persons of average intelligence and education. In this case, it has been firmly established that the battered women's syndrome is a subject that is properly within the special comprehension of experts. Also, as the record abundantly demonstrates, the battered women's syndrome is a subject that is not fully understood by the average person. Consequently, it is an appropriate matter for elucidation through expert testimony. State v. Griffin, 120 N.J. Super. 13, 29 (App. Div. 1972), certif. den. 62 N.J. 73 (1972); Nesta v. Meyer, 100 N.J. Super. 434 (App.Div. 1968), cited in State v. Cavallo, supra, 88 N.J. at 518; Angel v. Rand Express Lines, Inc., 66 N.J. Super. 77, 85 (App.Div. 1961).
The second requirement of Evidence Rule 56(2) that must be met before expert testimony on a particular subject is permitted  is a showing that the proposed testimony would be reliable. State v. Cavallo, supra, 88 N.J. at 516-17 (1982). There must be a sufficient scientific basis for the expert testimony. The asserted scientific body of knowledge must be considered reliable by those who have professional training and responsibility in the field. Romano v. Kimmelman, 96 N.J. 66, 80 (1984); State v. Hurd, supra, 86 N.J. at 536; State v. Cary, 49 N.J. 343, 352 (1967).
There are generally three ways a proponent of expert testimony can prove its reliability in terms of its general acceptance within the professional community. First, such general acceptance can be established by the testimony of knowledgeable experts. Second, authoritative scientific literature can be used to establish professional acceptance. Finally, persuasive judicial decisions that acknowledge such general acceptance of expert testimony can be followed. State v. Cavallo, supra, 88 N.J. at 521.
These criteria for the admissibility of expert testimony relative to the battered women's syndrome have been met in this case. Because the battered women's syndrome is a relatively new field of research, only a few courts have had the opportunity to consider its evidential admissibility. Some courts have already acknowledged the scientific acceptability of the syndrome and the reliability of the methodology used by practitioners and researchers in this field. See, e.g., State v. Allery, 101 Wash.2d 591, 596, 682 P.2d 312, 315 (1984) (en banc) (battered women's syndrome sufficiently accepted in scientific community and sufficiently outside lay competence so as to be appropriate subject of expert testimony in criminal trial); State v. Anaya, 438 A.2d 892 (Me. 1981); Smith v. State, 247 Ga. 612, 277 S.E.2d 678 (1981). Other courts have not yet done so. Compare Hawthorne v. State, 408 So.2d 801 (Fla. Dist. Ct. App. 1982), petition for review denied, 415 So.2d 1361 (Fla. 1982) and Ibn-Tamas v. United States, 407 A.2d 626 (D.C.Ct.App. 1979) (remanding to trial court for further consideration of scientific acceptability) with Buhrle v. State, 627 P.2d 1374 (Wyo. 1981)  and State v. Thomas, 66 Ohio St.2d 518, 423 N.E.2d 137 (1981) (holding that subject was not sufficiently established as a matter of scientific expertise). In light of the compelling record that has been established in this case, I am persuaded of the soundness of those decisions that have concluded that the battered women's syndrome constitutes a valid subject of expert testimony. I am satisfied that these decisions are correct and will emerge as the authoritative position on this issue.
The record before us, based on expert testimony, including scientific writings, further reveals that the battered women's syndrome has gained general acceptance as a scientific doctrine within the professional community. Dr. Lois Veronen, a highly qualified expert in the field, testified that the battered woman's syndrome is acknowledged and accepted by practitioners and researchers in the fields of psychology and psychiatry. In addition, Dr. Veronen testified to the existence of numerous authoritative books, articles and papers evidencing the scientifically recognized, expanding field of study and research about the battered woman's syndrome. See Buckelew v. Grossbard, 87 N.J. 512 (1981); Calabrese v. Trenton State College, 82 N.J. 321 (1980). The abundance of this authoritative literature was also made evident on this appeal — over 70 scientific articles and several books have been published on the subject. Dr. Veronen further testified that, since 1977, the battered women's syndrome has been recognized at several symposiums sponsored by such organizations as the Association for the Advancement of Behavior Therapy and the American Sociological Association. See Giannelli, "The Admissibility of Novel Scientific Evidence: Frye v. United States, a Half-Century Later," 80 Colum.L.Rev. 1197 (1980) (under appropriate circumstances, speeches, addresses, and other non-written sources may be used to demonstrate the acceptance of a premise by the scientific community).
Public policy considerations complement these traditional modes for determining whether a particular subject matter is reliable and within the purview of expert knowledge. An  emerging public policy acknowledges the battered women's syndrome. Psychiatrists, psychologists, and social scientists, as well as the legal and law enforcement community, have begun to come to grips with the forces that generate and perpetuate familial and domestic violence. See, e.g., R. Langley & R. Levy, Wife Beating: The Silent Crisis (1979); Martin, supra; Walker, supra; R. Gelles, The Violent Home: A Study of Physical Aggression between Husbands and Wives (1971); Battered Women: A Psychosociological Study of Domestic Violence (M. Roy, ed. 1977). The New Jersey Legislature has recognized the pervasiveness and gravity of domestic violence, which in so many cases forms the backdrop against which the battered women's syndrome appears. See Prevention of Domestic Violence Act, L. 1981, c. 426, N.J.S.A. 2C:25-1 to -16; Shelters for Victims of Domestic Violence Act, L. 1979, c. 337, N.J.S.A. 30:14-1 to -17; New Jersey Supreme Court Task Force on Women in the Courts, Summary Report at 5-6 (Nov. 21, 1983). The Legislature was presumably aware of the burgeoning expert opinion and literature that recognized the battered women's syndrome as both a contributing cause and devastating consequence of domestic and familial violence. This growing awareness extends to the national level as well, as evidenced, for example, by the U.S. Attorney General's formation, in September 1983, of a task force on family violence "to review [the] basic assumptions that underpin the handling of [domestic] violence cases." Statement of Attorney General William French Smith, September 19, 1983.
The final requirement of Evidence Rule 56(2) for the admission of expert testimony is the showing that the proffered expert witness has sufficient expertise to testify. State v. Cavallo, supra, 88 N.J. at 516. In this case, as recognized by the Court, Dr. Veronen was clearly highly qualified to testify as an expert with respect to the psychological condition of battered women's syndrome. Ante at 211. Furthermore, her proffered testimony fully met the standards for the receipt of expert testimony concerning the battered women's syndrome.
 In addition to her general knowledge of the battered women's syndrome, Dr. Veronen was familiar with the facts in this case and competent to testify in that regard. Dr. Veronen described the various psychological tests and examinations she had performed in connection with her independent research and the application of this methodology to defendant. Dr. Veronen was prepared to express her professional opinion that Gladys Kelly was an abused woman suffering from battered women's syndrome when she fatally stabbed her husband.
In sum, the record fashioned in this case convincingly demonstrates, through the testimony of an eminently qualified expert witness, that expert testimony concerning the battered women's syndrome is now generally accepted and regarded as reliable within the professional community. Its competence and relevance as evidence in the trial of particular criminal cases has been shown. The battered women's syndrome is sufficiently reliable to authorize its admissibility as a proper subject of expert testimony. In my view, this evidence should have been allowed in the trial of this case.
I concur in the majority's determination that the testimony of defendant's seventeen-year-old daughter, Edith Cannon, concerning the decedent's beatings of Gladys and her children, should have been admitted into evidence at the trial. Ante at 215-216. Defendant's daughter was also prepared to testify that she had been sexually abused by decedent since she was 12 years of age and had related this to her mother. However, the trial judge, on the basis of Evidence Rule 4, excluded Edith's testimony that she had told her mother about the decedent's sexual assaults upon her.
The expert evidence fairly shows that such circumstances — the physical and sexual abuse of battered women's children — cannot be separated from all of the factors that contribute to the syndrome. Such child abuse occurs in 75% of the battering  relationships that eventuate in homicide, and frequently constitutes a "critical factor in the tension * * * before some lethal incidents." Walker, supra, at 11. Consequently, such evidence of child abuse is relevant in a case in which the battered women's syndrome is a material issue.
To reiterate, expert testimony on the battered women's syndrome and the applicability of this syndrome to the defendant's claim of self-defense should be allowed on the retrial of this case. Evidence of the victim's abuse of the defendant's children, including sexual assaults on her daughter, are part of the dismal composite that constitutes the battered women's syndrome. Such evidence is highly probative of the issue of self-defense in the context of the battered women's syndrome and its evidential worth clearly outweighs its potential for prejudice or confusion.
In sum, I believe the Court acts without sufficient warrant in remanding this case to permit the issue of the general admissibility of expert testimony on the battered women's syndrome to be tried anew. The record reveals that the issue of admissibility was fairly presented at trial. That record has generated an evidential base sufficiently solid to permit, if not mandate, our acceptance of the battered women's syndrome as expert doctrine. While it is arguable that the State did not fully challenge the evidence below, its position on appeal is essentially that the evidence proffered at the trial was not adequate to establish the scientific reliability of the battered women's syndrome. The Court now unanimously rejects that position. I think it pointless and unfair to encourage the State to renew its attacks upon the authenticity of the battered women's syndrome doctrine.
For the reasons expressed, I dissent in part from the Court's decision.
 For reversal — Chief Justice WILENTZ, and Justices CLIFFORD, SCHREIBER, POLLOCK, O'HERN and GARIBALDI — 6.
Concurring in part and dissenting in part — Justice HANDLER — 1.
 This version of the homicide — with a drunk Mr. Kelly as the aggresor both in pushing Ms. Kelly to the ground and again in rushing at her with his hands in a threatening position after the two had been separated — is sharply disputed by the State. The prosecution presented testimony intended to show that the initial scuffle was started by Gladys; that upon disentanglement, while she was restrained by bystanders, she stated that she intended to kill Ernest; that she then chased after him, and upon catching up with him stabbed him with a pair of scissors taken from her pocketbook.
 The works that comprise the basic study of the problem of battered women are all relatively recent. See, e.g., R. Langley & R. Levy, Wife Beating: The Silent Crisis (1979); D. Martin, Battered Wives (1976); L. Walker, The Battered Woman (1979); R. Gelles, The Violent Home: A Study of Physical Aggression between Husbands and Wives (1971); Battered Women: A Psychosociological Study of Domestic Violence (M. Roy ed. 1977).
Similarly, legislative activity in this field is relatively new; for example, New Jersey's Prevention of Domestic Violence Act, L. 1981, c. 426, N.J.S.A. 2C:25-1 to -16 and the Shelters for Victims of Domestic Violence Act, L. 1979, c. 337, N.J.S.A. 30:14-1 to-17.
In enacting the Prevention of Domestic Violence Act, the New Jersey Legislature recognized the pervasiveness and seriousness of domestic violence:
The Legislature finds and declares that domestic violence is a serious crime against society; that there are thousands of persons in this State who are regularly beaten, tortured and in some cases even killed by their spouses or cohabitants; that a significant number of women who are assaulted are pregnant; that victims of domestic violence come from all societal and economic backgrounds and ethnic groups; that there is a positive correlation between spouse abuse and child abuse; and that children, even when they are not themselves physically assaulted, suffer deep and lasting emotional effects from exposure to domestic violence. It is therefore, the intent of the Legislature to assure the victims of domestic violence the maximum protection from abuse the law can provide. [N.J.S.A. 2C:25-2].
 In her book, The Battered Woman, Dr. Lenore Walker cites research by sociologists Straus, Gelles, and Steinmetz finding that in 1976 at least one assault between family members occurred in 28% of all American homes. Id. at 70.
 In 1976, for example, battered women in California and New York instituted class actions alleging that the police customarily denied women legal protection by refusing to assist battered women or arrest their abusing husbands. The cases were settled by consent judgment. Scott v. Hart, No. C-76-2395 (N.D.Cal., filed Oct. 28, 1976); Bruno v. Codd, 90 Misc.2d 1047, 396 N.Y.S.2d 974 (Sup.Ct. 1977), aff'd, 47 N.Y.2d 582, 393 N.E.2d 976, 419 N.Y.S.2d 901 (1979).
 Alcohol is often an important component of violence toward women. Evidence points to a correlation between alcohol and violent acts between family members. In one British study, 44 of 100 cases of wife abuse occurred when the husband was drunk. Gayford, "Wife Battering: A Preliminary Survey of 100 Cases," British Medical Journal 1:194-197 (1975). Gelles, in The Violent Home: A Study of Physical Aggression between Husbands and Wives (1979), found that in 44 families where violence had occurred, drinking accompanied the violence in 21 of the cases. He also posited that alcohol and family violence are more closely related than alcohol and other types of violence.
 Prior to the enactment of the Code, former N.J.S.A. 2A:113-6 provided a statutory basis for self-defense claims specifically and justification defenses generally. However, as noted by the New Jersey Criminal Law Revision Commission, the law concerning justification was that found in the cases, since the literal wording of 2A:113-6 was not followed. Final Report of the New Jersey Criminal Law Revision Commission Vol. II: Commentary, at 78-79 (1971).
 See also Restatement of Torts 2d § 63 (1965) at 101. Under principles of self-defense as a justification for the torts of assault and battery — which closely parallel criminal self-defense principles — no privilege of self-defense exists for one acting in ignorance of another's intent to inflict harm on him. Cf. Perkins, "Self-Defense Re-examined," 1 U.C.L.A.L.Rev. 133, 134 (1954).
 The rejected form of § 2C:3-4 was patterned after § 3.04 of the Model Penal Code. The purpose of the proposed Code and M.P.C. provisions was to prevent one who killed in the honest but mistaken and unreasonable belief in the necessity of the action from being convicted of a crime like murder, which is premised on an act motivated by unlawful purpose. See Model Penal Code § 3.04 commentary at 14-15 (Tent. Draft No. 8 1958); Commission Report, supra, Vol. II: Commentary, at 83-84.
 In State v. Powell, 84 N.J. 305 (1980), we explicitly recognized that before enactment of the Code the doctrine of imperfect self-defense could reduce murder to manslaughter when the defendant honestly but unreasonably perceived himself in such danger as to require the use of deadly force. However, we expressed no opinion on whether imperfect self-defense was available under the new Code for the purpose of reducing murder to manslaughter. The resolution of that issue is immaterial to the case at bar.
 The factual contentions of the parties eliminated any issue concerning the duty to retreat. If the State's version is accepted, defendant is the aggressor; if defendant's version is accepted, the possibility of retreat is excluded by virtue of the nature of the attack that defendant claims took place. We do not understand that the State claims defendant breached that duty under any version of the facts. If, however, the duty becomes an issue on retrial, the trial court will have to determine the relevancy of the battered-woman's syndrome to that issue. Without passing on that question, it appears to us to be a different question from whether the syndrome is relevant to defendant's failure to leave her husband in the past.
 The State may not bar the introduction of expert testimony about the battered-woman's syndrome by stipulating that the defendant's fear of serious bodily harm was honestly held. In State v. Laws, 50 N.J. 159 (1967), we rejected the suggestion that the State should be compelled to stipulate to — and not introduce evidence on — those facts that the defendant did not dispute. We held that subject to the trial court's overriding control of the proceedings, the State "should have the right to make a full showing before the jury whenever it considers such course necessary for the proper presentation of its case." Id. at 184. Similar considerations compel the same result here, should the defendant seek to introduce testimony on a fact — the honesty of defendant's fear of serious bodily harm — that the State does not contest. This holding protects the defendant's due process rights by allowing her to offer testimony to establish a defense. See Webb v. Texas, 409 U.S. 95, 98, 93 S.Ct. 351, 353, 34 L.Ed.2d 330, 333 (1972) (citing Washington v. Texas, 388 U.S. 14, 19, 87 S.Ct. 1920, 1923, 18 L.Ed.2d 1019 (1967)). Without the introduction of expert testimony to dispel common misconceptions about battered women, a jury might well question the stipulation of honesty.
 The State contends that even if the expert testimony is held admissible, its exclusion should be considered harmless error because of defendant's conviction for reckless manslaughter. The State's position is that under N.J.S.A. 2C:3-9(b) as it existed at the time of Gladys Kelly's conviction, self-defense was not available as a defense for any offense for which recklessness or negligence suffices to establish culpability, including, presumably, offenses where the defendant was reckless or negligent in believing the use of force to be necessary, or in acquiring or failing to acquire any knowledge that is material to the justifiability of the use of force. The argument, therefore, is that the expert testimony could not have saved defendant from the reckless manslaughter verdict.
N.J.S.A. 2C:3-9(b) was never intended to serve the function ascribed to it by the State. In fact, inclusion of the provision in the Code appears to have been an error, which has since been corrected by its repeal. See L. 1981, c. 290. The reasons for the inclusion and repeal of this provision are concisely stated in the legislative history of the repealer:
As originally drafted, justification defenses (i.e. self-defense) under the code were available to a defendant if his belief in the necessity of the use of force was honestly held. In conjunction with this provision, the code also provided in 2C:3-9b that if the defendant was reckless or negligent in forming that belief, he could be convicted of a crime for which recklessness or negligence was the required mental element. As enacted, however, the code requires not only that a defendant's belief be honestly held but also that his belief in the necessity to use force be reasonable. This requirement that a defendant's belief be both honest and reasonable vis a vis a justification defense obviates the necessity for the provision in 2C:3-9b that the reckless or negligent use of force can establish criminal liability. Therefore, the amendment in section 7 would delete this provision. [Senate Judiciary Committee, Statement to Committee Substitute for S. 2537 at 2 (1982)].
In other words, when the original draft of the Code provided that an honest belief in the need for deadly force sufficed to establish self-defense, the Code had to deal with the situation in which that belief, though honest, had been recklessly formed. The subsequently repealed section, N.J.S.A. 2C:3-9(b), performed that function by providing that such an honest belief, recklessly formed, was no justification for offenses when culpability was based on that very same recklessness. The Code as passed, however, defined self-defense as requiring a reasonable belief, thereby rendering section 9(b) unnecessary since, under that definition, self-defense could not be established as a justification for any offense if the actor's belief in the need for force, though honest, was recklessly formed, i.e., was unreasonable. The repealer simply clarified the legislative intent that existed when the Code first became law, which was that self-defense based on a reasonable belief in the need for deadly force would constitute justification — a complete defense — to the charge of reckless manslaughter. If the jury here found defendant's belief was both honest and reasonable, it would be required to acquit her of all charges.
 At least two other courts agree that expert testimony about the battered-woman's syndrome is relevant to show the reasonableness as well as the honesty of defendant's fear of serious bodily harm. Ibn-Tamas v. United States, 407 A.2d 626, 634-35 (D.C. 1979) (expert testimony "would have enhanced Mrs. Ibn-Tamas' general credibility in responding to cross-examination designed to show that the testimony about the relationship with her husband was implausible," and also "would have supplied an interpretation of the facts which differed from the ordinary lay perception"); Hawthorne v. State, 408 So.2d 801, 806-07 (Fla. Dist. Ct. App. 1982) (expert testimony would "aid the jury in interpreting the surrounding circumstances as they affected the reasonableness of [defendant's] belief," because "a jury would not understand why [defendant] would remain [with her husband]"); State v. Allery, 101 Wash.2d 591, 682 P.2d 312, 316 (Wash.Sup.Ct. 1984) (court approved use of expert testimony "[t]o effectively present the situation as perceived by the defendant, and the reasonableness of her fear ... to enable the jury to overcome stereotyped impressions about women who remain in abusive relationships"). But see Commonwealth v. Light, 458 Pa. 328, 326 A.2d 288 (1974) (psychiatric testimony held to be of no help in determining whether a fear of serious bodily harm was reasonable).
Defendant's counsel at oral argument made it clear that defendant's basic contention was that her belief in the immediate need to use deadly force was both honest and reasonable; and that the evidence concerning the battered-woman's syndrome was being offered solely on that issue. We therefore are not faced with any claim that a battered woman's honest belief in the need to use deadly force, even if objectively unreasonable, constitutes justification so long as its unreasonableness results from the psychological impact of the beatings. The effect of cases like State v. Sikora, 44 N.J. (1965) (opinion of psychiatrist that acts of defendant, admittedly sane, were predetermined by interaction of events and his abnormal character held inadmissible on issue of premeditation), and State v. Bess, 53 N.J. 10 (1968) (reasonableness of belief in need for deadly force not measured by what would appear "reasonable" to abnormal defendant) is not before us. Nor is there any claim that the battering provocation might have some legal effect beyond the potential reduction of defendant's culpability to manslaughter, or something other than an "immediate" need for deadly force will suffice. See State v. Felton, 110 Wis.2d 485, 329 N.W.2d 161 (1983), (battered wife stabs sleeping husband).
 Of course, expert testimony that meets these three criteria is still subject to other rules of evidence. For example, the probative value of the testimony must not be substantially outweighed by the risk that its admission would necessitate undue consumption of time or create substantial danger of undue prejudice or of confusing the issues or of misleading the jury. Evid. R. 4. The danger of undue prejudice would be only slightly greater if expert testimony on the battered-woman's syndrome is introduced than without it, however, because the jury, even without it, will certainly hear about the past beatings from lay witnesses.
 The following courts agree that the battered-woman's syndrome is beyond the understanding of the average person: Ibn Tamas v. United States, supra, 407 A.2d 626; Smith v. State, supra, 247 Ga. 612, 277 S.E.2d 678; Hawthorne v. State, supra, 408 So.2d 801; State v. Anaya, supra, 438 A.2d 892. But see State v. Thomas, supra, 66 Ohio St.2d 518, 423 N.E.2d 137.
 Compare State v. Anaya, supra, 438 A.2d 892, and Smith v. State, supra, 247 Ga. 612, 277 S.E.2d 678 (both cases accepting expert testimony without reservation), with Hawthorne v. State, supra, 408 So.2d 801, and Ibn-Tamas v. United States, supra, 407 A.2d 626 (both cases remanding to trial court for further consideration of scientific acceptability), and with Buhrle v. State, 627 P.2d 1374 (Wyo. 1981), and State v. Thomas, supra, 66 Ohio St.2d 518, 423 N.E.2d 137 (both cases holding that subject was not sufficiently established as a matter of scientific expertise).
 Under appropriate circumstances, speeches, addresses, and other similar sources may be used to demonstrate the acceptance of a premise by the scientific community. See Giannelli, "The Admissibility of Novel Scientific Evidence: Frye v. United States, a Half-Century Later," 80 Colum.L.Rev. 1197, 1217 (1980).
 In addition to her general knowledge of the battered-woman's syndrome, Dr. Veronen is quite familiar with the facts in this case. Dr. Veronen interviewed Ms. Kelly for approximately five hours, during which time the two spoke about Ms. Kelly's background, Ms. Kelly's first meeting with Mr. Kelly, Ms. Kelly's relationship with her children and Mr. Kelly, the history of the physical abuse she suffered, and her stabbing of Mr. Kelly. Dr. Veronen also reviewed several psychological tests that were administered to Ms. Kelly, and from those concluded that Ms. Kelly was a battered woman. In addition, Dr. Veronen reviewed statements of eyewitnesses to the stabbing, police reports, and Ms. Kelly's hospital reports following the stabbing.
 It is not contended by anyone that the battered woman's syndrome has been so well established in the scientific community and is so well known by the public as to authorize the Court to take judicial notice of it. Therefore, unlike some expert evidence (radar, for example, State v. Dantonio, 18 N.J. 570 (1955)) where all that is required is to show that the accepted body of scientific knowledge is being correctly applied, here the very existence and reliability of such scientific knowledge has to be established. As a matter of fact, the literature suggests that while there is a growing body of research concerning the syndrome, it is still in a relatively uncertain stage, there remaining some doubt about its validity. It is, therefore, necessary for this Court to be sure that on remand the State has an adequate opportunity to present such proofs as might persuade the trial court that the syndrome has not yet achieved sufficient acceptance in the scientific community to warrant its admissibility. While our dissenting colleague is apparently convinced both from the record and his own research that as a matter of law the syndrome has achieved that level of acceptability to warrant its admission, that procedure, leading to that conclusion, seems to us manifestly unfair to the State. Even if we were inclined to agree with our dissenting colleague on this issue, that would be beside the point, for what is involved here is not the correctness of the conclusion concerning the general acceptability within the scientific community of the battered-woman's syndrome, but the fundamental fairness of the proceedings in the trial court that might lead to such a conclusion. It is absolutely clear that the only proceedings concerning the syndrome before the trial court was the voir dire testimony of Dr. Veronen, that the State was permitted cross-examination only as to her qualifications, and that the court repeatedly assured the assistant prosecutor that "ample time" would be given on all issues concerning the syndrome. Not only was "ample time" not given, but no time was allowed, for the trial court, apparently believing that the proposed use of this testimony had been made clear, decided that the testimony would be inadmissible as a matter of law even if the witness were ruled to be an expert and even if the body of knowledge were ruled to be beyond the ken of jurors and generally accepted within the scientific community. The court's ruling that the expert's testimony was inadmissible was prefaced by the following statement: "I fully appreciate you have not had another opportunity to examine the Witness, Mrs. Cooper," the remark of the court being addressed to the assistant prosecutor. Throughout the transcript there were repeated references by the court and Mrs. Cooper to the fact that her role, up to that point, had been confined to cross-examination only on the witness's qualifications.
 Evidence Rule 4 provides:
The judge may in his discretion exclude evidence if he finds that 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.
 The trial court has discretion to bar the admission of prior convictions if it finds that their probative value is outweighed by their prejudice to the defendant. State v. Sands, supra, 78 N.J. at 146. The burden of proving that the prior convictions should be excluded, however, falls on the defendant. Ibid. We do not find that the trial court abused its discretion in allowing the limited testimony cited earlier, even though the remoteness of Ms. Kelly's earlier conviction would also have supported a ruling that her conspiracy conviction could not be brought out.
 We note that given defendant's conviction for reckless manslaughter under N.J.S.A. 2C:11-4(b)(1), these instructions would not constitute harmful error. Both reckless homicide and homicide committed in the heat of passion resulting from a reasonable provocation constitute manslaughter, and both are crimes of the second degree. N.J.S.A. 2C:11-4(b), (c). Defendant's conviction for reckless manslaughter instead of manslaughter by provocation, therefore, did not prejudice her in terms of sentencing. Neither did it produce a compromise verdict of the type referred to in State v. Christener, 71 N.J. 55 (1976) where the concern was with the prejudicial effect of overcharging the jury by giving instructions on first degree murder that were not sufficiently supported by the evidence. This is easily distinguishable from the problem here, which involves only a deficient instruction for an alternate theory of the offense for which the defendant actually was convicted.
 We note that under the Code even if it is certain that the actor's life will soon be threatened, the actor may not use deadly defensive force until that threat is imminent. If he or she does, the crime in most cases would presumably be murder or manslaughter (see N.J.S.A. 2C:3-4a & 4b(2); 2C:11-3; 2C:11-4b), the last exposing the actor to a sentence of ten years in prison with a five-year discretionary parole ineligibility term or, if a firearm is used, a three-year mandatory parole ineligibility term. N.J.S.A. 2C:43-6a, b & c. The requirement that the use of deadly force, in order to be justifiable, must be immediately necessary, has as its purpose the preservation of life by preventing the use of deadly force except when its need is beyond debate. The rule's presumed effect on an actor who reasonably fears that her life will soon be endangered by an imminent threat is to cause her to leave the danger zone, especially if, because of the circumstances, she knows she will be defenseless when that threat becomes imminent. The rule, in effect, tends to protect the life of both the potential aggressor and victim. If, however, the actor is unable to remove herself from the zone of danger (a psychological phenomenon common to battered women, according to the literature), the effect of the rule may be to prevent her from exercising the right of self-defense at the only time it would be effective. Instead she is required by the rule to wait until the threat against her life is imminent before she responds, at which time she may be completely defenseless.
There is, of course, some danger that any attempt to mitigate what may be undeserved punishment in these cases (by some further statutory differentiation of criminal responsibility) might weaken the general deterrent effect of our homicide laws. That is a matter the Legislature might wish to examine.