Domestic Violence Fatality Reviews: From a Culture of Blame to a Culture of Safety

By Neil Websdale, Ph.D., Judge Michael Town and Byron Johnson, Ph.D.
Spring, 1999 |  Juvenile and Family Court Journal (p.61-74)


As courts and communities try to confront domestic violence, the question of what to do about domestic violence fatalities continually resurfaces.  Normally, these fatalities are handled by the criminal justice system, which investigates the deaths and identifies and charges the perpetrators, when appropriate.  Such criminal justice handling, however, does little to review the effectiveness of the various systems charged with serving and protecting those vulnerable to domestic violence and death.  This shortcoming is all the more significant given that most communities have experienced a high profile domestic violence homicide.

Traditionally, these tragedies have resulted in finger pointing, anger, fear, frustration and distrust.  Sometimes, this finger pointing has found voice in the form of editorials, lawsuits, and legislative hearings.   These forms of finger pointing, sometimes referred to as “tombstone technology” in fields such as aviation and nuclear power, have not been productive.  [footnote 1]  They can result in accusations of stonewalling and cover-ups.  Consequently, many community members, including judges, court administrators, elected officials, prosecutors, law enforcement officials, and battered women’s advocates are looking for workable and fair models to review domestic violence fatalities, whit a view to preventing future details.

This search is not for the fainthearted since it requires a paradigm shift from a culture of blame to a culture of safety in which domestic violence deaths are reviewed through the lens of preventive accountability.  Fortunately, there are workable models in the fields of medicine and aviation upon which to draw.  These models teach courts and communities that, with vigor, honesty, and candor, they can build reliable systems that value accountability and help prevent future death and injury from domestic violence. Because domestic violence deaths exhibit predictable patterns and etiologies, they are preventable.

We argue that the establishment of domestic violence fatality review teams is one effective way of reducing domestic violence homicides.  After briefly outlining the scope and extent of domestic violence related deaths, this article discusses the history of domestic violence fatality reviews and presents several models that appear to be both effective and fair.  In particular, we emphasize that these models form part of an emerging process that will take years to unfold.  We especially recommend judicial leadership in promoting and establishing local review processes.  This is particularly so in jurisdictions where unified family court or closely coordinated juvenile/family court exists.

It is not our intent to present a formula for conducting such reviews.  Rather, this article presents a variety of apparently effective models, since the authors believe communities will review “domestic violence deaths” in their own unique ways.  By raising key questions and presenting workable models, the authors hope to contribute to the discussion about domestic violence fatality reviews within the framework of a culture of safety rather than a culture of blame.

Scope and Extent of Domestic Violence Related Deaths:  An Outline 

Each year deaths attributable to domestic violence constitute a significant proportion of the total number of homicides.  These homicides take a number of forms.  The largest sub-category of domestic violence deaths is “intimate partner homicide.”  This form of domestic homicide involves the killing of a person by his/her intimate or former intimate partner [footnote 2].   The other major sub-category consists of “family homicide,” which involves the killing of a victim by that person’s relative by blood or marriage.  Examples of family homicide include cases in which parents or guardians kill children (filicide), brothers kill brothers (fratricide), sisters kill sisters (sorroricide) or children kill parents (patricide). [footnote 3]

Intimate partner and family homicides are no the only deaths attributable to domestic violence.  A number of researchers have argued that many women who commit suicide do so within the context of battering relationships, thus making it possible to argue that domestic violence was a prime casual factor in their self-killing. [footnote 4]  Similarly, because battered women appear more vulnerable to HIV infections than non-battered women, [footnote 5] one might argue that some deaths of women attributed to HIV, or some complication thereof, might be traceable to their status as battered women.

The Emergence of Domestic Violence Fatality Reviews

Medical Fatality Reviews

The emergency of child and adult domestic violence fatality reviews is traceable to death reviews in the medical profession.  Often, these reviews are called morbidity and mortality reviews.  The medical fatality review model is based on the internal reviews of deaths that occur in hospital settings.  Personnel involved with patients who die in questionable circumstances present information to the review team.  The team gathers the information together and reaches a conclusion about the reasons for the fatality.

One of the initial problems identified in implementing effective medical reviews was their emphasis on “catching rascals, rather than on improving hospital wide performance” (Rosen and Susman, 1983).  This blaming approach was highlighted in an October 1998 editorial in the Journal of the American Medical Association (JAMA).  The editorial notes that the health care system continues to rely upon “requiring individual error-free performance enforced by punishment, a strategy abandoned long ago by safer industries such as aviation and nuclear power.”  [footnote 6]

Child Fatality Reviews

Unfortunately, many child fatality review teams also emerged with a similarly punitive ethos. [footnote 7]  In some cases, review teams inappropriately blamed battered mothers for failing to protect children killed by abusive male partners. [footnote 8]  Other child death review teams appropriately sought to identify breakdowns in the system of service delivery, focusing less on individual accountability and more on system-wide service coordination.

The most progressive child review teams currently recognize the need to blend multiple systems-wide accountability with a non-punitive ethos.  This does not mean there is no accountability.  Rather, there is a recognition that risk and error are inevitable aspects of the coordinated delivery of complex services.  Errors, therefore, should be identified and rectified within an open climate of honesty and healing.

As the Colorado Child Death Review Committee points out, if cases are handled improperly, or if a crime is committed, agencies with the greatest involvement and clearest responsibility are asked to put things right.  In especially egregious situations, matters can be submitted to a grand jury.  [footnote 9]  This philosophy seems to have permeated through to the review of adult domestic violence deaths in this jurisdiction.  For example, the mission statement of Project Safeguard (Denver) recognizes that “perpetrators of domestic violence are ultimately responsible for the death of victims.  Thus, the goal of this committee is not to place blame but rather to better understand the dynamics of domestic violence with death is involved and thereby diminish the possibilities of future fatalities.  [footnote 10]

Most child death reviews involve child fatalities caused by abuse and/or neglect and as such constitute a form of domestic violence death review.  However, child fatality review teams have not always been quick to recognize the key links between adult domestic violence and the killing of children.  There are notable exceptions and some teams have begun to work on identifying these links. [footnote 11]  For example, Detective Linda Burton, who heads up the Child Death Review Team in Hillsborough County,Florida, reports that at least two-thirds of children killed in homicides in Hillsborough County from 1994-1993 had mothers or other female caretakers who had been beaten by their intimate male partners. [footnote 12]

The Charan Investigation

One of the first domestic violence reviews, the “Charan investigation” involved the very detailed public analysis of a domestic homicide-suicide in San Francisco.  The Charan investigation pinpointed the widespread breakdown of several systems.

The Facts. On January 15, 1990, Joseph Charan killed his wife, Veena Charan, then took his own life.  For a period of 15 months prior to her death, Veena Charan sought the support of various government agencies.  Veena had been separated from Joseph and was awarded custody of their nine-year-old son.  During the 15 months preceding her death she made numerous reports to the police.  Immediately prior to her death, Joseph was arrested on felony wife beating and malicious mischief.  As a result of his conviction for this offense, Joseph received a 12-month suspended jail sentence.  He was put on probation through the Adult Probation Department with the following three conditions:  (1) mandated domestic violence counseling; (2) a stay away order, and (3) 30 days jail time, of which he was ordered to serve four days, with the remainder to be served in the Sheriff’s Work Alternative Program.  Veena Charan had obtained a restraining order, which Mr. Charan violated on several occasions.  He also attempted to kidnap his son at the son’s school.  It was at the school that Mr. Charan killed his wife in front of school teachers and school children, before committing suicide.

Questions raised by the investigation.  The San Francisco Domestic Violence Consortium which commissioned the Charan investigation, requested answers to three clusters of questions:

  1. Do the departments of the City and County of San Francisco Have policies and procedures relating to domestic violence?  If so, what are they and how adequate are they?
  2. Is there sufficient information-sharing among the departments in these particular types of cases?
  3. Are there sufficient data to evaluate the effectiveness of the system?  If not, what additional data need to be collected?  What changes, if any, to current procedures can be adopted to avert future tragedies?

Gaps in service delivery identified.  As a result of this investigation, the cases files and public testimony identified four essential gaps in service delivery in the Charan case:

  1. 1.      Communication and Coordination.

Aside from the communication between the San Francisco Police Department and the District Attorney’s Office, there was little communication among the multiple agencies that had contact with Veena Charan.  These multiple agencies included the municipal court, adult probation, family court services, and social services.  The review committee called of centralization of information and better coordination of service delivery.

  1. 2.      Data Collection.

The commission recognized the need for systematic information about domestic violence cases.  The investigation noted, “Data on the number of domestic violence cases handled by the departments ranged from very limited to none at all.” [footnote 13]  The Commission deemed the data to be of central importance in the identification of the level of need for services and the subsequent delivery of services.

  1. 3.      Access to Services.

The Commission pointed out that the lack of sensitivity to and an understanding of the multicultural and gay/lesbian issues in the city departments increases the numbers of those suffering from domestic violence.

  1. 4.      Training.

Most of the training recommendations concerned multicultural awareness.  [footnote 14]  Translation services were lacking.  Specifically, there was a lack of translators in the Superior Court, Civil Division, and a limited number of translators in the Criminal Division.  This problem created delays and misunderstandings of the agreements/court orders and proceedings.  Specifically, the investigation called for the development of domestic violence advisory committees in each city department working with domestic violence cases.



Additional Highlights from the Charan Investigation

  • Based on the incident repots involving Joseph Charan, the San Francisco Police Department did not deem serious the injuries Veena Charan and other family members received at the hands of Joseph Charan.  Specifically, the report indicated that “had the investigator looked at the pattern of violence established by Mr. Charan, and presented that information to the District Attorney’s Office, stronger measures and responses to the situation may have prevented Joseph Charan from continuing the escalation of violence that lead to the murder-suicide.” [footnote 15]
  • According to the felony protocol of the District Attorney’s Office, prior history was one of the factors to be taken into account regarding re-booking.  If the Assistant District Attorney had access to the same information the Commission did, the re-booking charges may have been different.
  • Probation officers were not trained adequately in the dynamics of domestic violence.
  • The Commission called for greater domestic violence training of Municipal Court, Criminal Division.  In particular, it stated a “need for training judges on the interpretation of restraining orders.” [footnote 16]
  • Family Court Services refused to answer questions posed by the Commission, citing their need to maintain confidentiality.  The Commission described this failure as “intransigence.” [footnote 17]  The report stated the resistance of Family Court Services “is indicative of the lack of the department’s efforts to improve the City’s responses to battered women and their children.” [footnote 18]  The Commission also criticized the mediation strategies of the Family Court.


Contemporary Domestic Violence Death Reviews:  Some Effective Emerging Models

Informal and semi-formal adult domestic violence death reviews have been conducted in a number of states for the past decade.  More recently some states, such as California,Nevada, and Delaware, have introduced legislation to regulate the review process and protect review participants from liability.  However, adult domestic violence fatality reviews are a relatively recent phenomenon and it is accurate to state that most domestic violence related deaths currently are not subject to systematic and multi-agency review.

The principal purpose of domestic violence fatality review is to reduce domestic violence related deaths and injuries through the identification and subsequent rectification of problems in the civil and criminal justice systems, including the delivery of multiple services to families.  One way of doing this is through reviews such as the Charan investigation, which, as the authors have shown, scrutinized one case very closely yielding various concrete recommendations regarding service delivery.

Other review practices aim to do the same, although they use different approaches.  These approaches are influenced by the availability of resources to fund reviews, the commitment of different agencies and jurisdictions, and their experience of domestic violence deaths.  Some review teams examine large numbers of deaths whit a view to identifying just how many are the result of domestic violence.  This type of wide-angle approach tends to reveal the extensive role played by domestic violence in loss of life in general.  An exemplar of this wide-angle approach can be found in Philadelphia.

Philadelphia:  A Multidisciplinary Model

The Philadelphia Women’s Death Review Team is a multi-agency, multidisciplinary group convened as a public-private collaboration.  It seeks to reduce the number of domestic violence deaths by examining the roll of violence in the lives of Philadelphia women killed by an intimate partner, as well as the effects of the killing on their children.

Scope of review.  Without substantial funding or any legislation, the Philadelphia Department of Public Health, with support from the District Attorney’s Office, is conducing reviews.  This multidisciplinary team goes down to medical examiners’ offices to review all deaths of women from 15 to 60 years of age, not just domestic violence cases.  [footnote 19]  These deaths either could be related directly to domestic violence or related indirectly due to the women’s inability to access health care.  Roughly three thousand women die in Philadelphia every year and the team expects to look at 400 to 500 deaths.  The team’s central objective is to be able to identify any domestic violence directed at decedents in the 12 months prior to the fatality.  The meetings are quarterly and each review takes about 30 minutes.

Observations.  To date, the Philadelphia Team [footnote 20] has made the following important observations about deaths of women:

  • It is difficult to locate information on many of the female decedents, especially psychological data.  Many women led invisible lives and their deaths often went unnoticed.  Many women who died prematurely were not known to any community/legal systems.
  • Often, perpetrators of domestic homicide are known within their communities, and not only in their role as offenders.  Some were known to mental health providers.  The team has asked whether it is possible to flag or track offenders who need but refuse psychiatric help.
  • Gun merchants do not always refuse to sell firearms to individuals with Protection from Abuse Orders against them.  Additionally, judges do not always order domestic violence perpetrators to relinquish previously acquired weapons.  The team has raised a number of questions about the use of the judiciary to remove or manage access to weaponry.
  • Women who die from HIV/AIDS are often connected to lifestyles involving drug use and prostitution.  It is well documented that prostitutes suffer inordinate amounts of abuse at the hands of men.

Santa Clara County,California:  A County Based Domestic Violence Coordinating Council Model

The Santa Clara County Death Review Committee began work in 1994 and appears to have been among the first domestic violence review teams in the country.  It defines “domestic violence related death” as one where the perpetrator and victim were “romantically linked,” either at the time of the death or prior to the death.

The final report [footnote 21] of he committee, published in October 1997, contained information on 51 domestic violence homicides.  The report included data on types of deaths (homicide, homicide-suicide, accidental death, or police shooting); police agencies involved in the case; age, race, sex and substance abuse history of the parties; presence of children; weaponry used; status of the relationship (divorced, cohabiting, separated, etc.); existence of prior restraining orders; prior police involvement; and location of residence.



Highlights of the 1997Santa Clara Death Review Final Report.

  • The average adult age of perpetrators and victims was 33 years (females 32, males 35).
  • Of the 51 perpetrators, 44 were male, seven female.
  • Firearms were used in 29 of the 51 homicides.  The report stressed that “as a community we must advocate for handgun control.” [footnote 22]
  • In 26 of the 51 cases, the parties were separated or divorced at the time of death.
  • Police had prior domestic violence contacts with the parties in 11 cases.
  • In six cases, restraining orders were either active (four) or in the process of being issued (two).
  • Of the 51 victims, 17 were Asian, [footnote 23] 14 white, 12 Hispanic, five African American, two mixed-race and one Indian (not Native American).
  • The report noted a need for educating the public through agencies such as schools and the media.  For example, the Santa County Report recommended all school districts develop a curriculum that addresses domestic violence.  [footnote 24]
  • The report noted a need for creating a greater awareness of the links between workplace violence and domestic violence.  The Santa County Team noted that seven of its 51 deaths occurred in the work place. [footnote 25]


Asian victims were over-represented among victims, although only one of the Asian cases came to the attention of the community agencies prior to the killing.  The report noted, “This made members feel that we were not getting the word out about the dangers of domestic violence to the Asian community.” [footnote 26]  This led to calls for greater Asian representation on the death review committee.  The report noted three Asian members on the team.  One committee member helped form the Asian Community Against Domestic Violence Coalition, which organized a domestic violence conference for the Vietnamese community in September of 1997.

The suggestion that more Asian women need to be accessed through support services should not be taken to mean that those women who did not utilize services were somehow culpable for their own deaths.   Karin Wang (1996) points to the way the cultural background of Asian women makes it difficult for them to utilize the support services offered by a predominately white-run domestic violence movement.  [footnote 27]  In addressing this issue, Wang argues that battered Asian-American women have not been well understood by the domestic violence movement. [footnote 28]

Although California legislation does not address the issue of domestic violence shelters turning over their records for purposes of death review, informally the team seems to have worked around the issue of client confidentiality and it appears a mechanism has emerged so shelter team members do share information.

The Florida Fatality Review Project:
Governor’s Research and Policy Model

The Florida Department of Law Enforcement reported 230 domestic homicides for 1994.  As a result of this disturbing statistic, the Florida Governor’s Task Force on Domestic and Sexual Violence funded a study of domestic fatalities in order to better understand, intervene in, and prevent these crimes.  Unlike reviews in Philadelphia and Santa Clary County, a team of researchers, hired specifically for that purpose, conducted the Florida reviews.  These researchers did identify system failures, although this was not the prime focus of their research.  Rather, they gathered evidence on the overall dynamics of cases prior to death.

Dynamics examined.  For each domestic fatality in 1994 the researchers examined the following dynamics:

1.  The perpetrator-victim dyad.

In particular, the researchers examined the multiple dynamics of these murders, paying particular attention to the sex, race, ethnicity, sexual orientation, geo-cultural background (rural, suburban, urban) socioeconomic status, and marital status/family relationship between perpetrators and victims.

2.  The situational antecedents to fatality.

Researchers explored the following:

  • A prior history of domestic violence in the relationship;
  • The presence or absence of injunctions (restraining or protection orders) both prior to the fatality or when the fatality occurred;
  • Whether a divorce was pending at the time of death (with married couples);
  • Whether there was any sign of relationship breakdown (variously measured);
  • Whether there was any sign of acknowledged conflict in the relationship;
  • History of drug/alcohol abuse;
  • The residential origins of the perpetrator and victim;
  • Whether the victim or perpetrator had any history of emotional problems or mental illness and the specific forms of these problems.  (See Table 3)

3.  The lethal incident.

Here researchers documented:

  • The specific mode of killing;
  • The types of weaponry used (handgun, rifle, shotgun, other firearm, knife or cutting instrument, blunt object, motor vehicle, poison, explosives, fire or incendiary device, personal weapons such as fist, feet, teeth, etc.)
  • The availability of weapons;
  • The involvement of drugs or alcohol during or immediately preceding the fatal episode;
  • The presence of other parties at the scene (e.g., children, police, other professionals);
  • The non-fatal wounding of others at the scene;
  • The involvement of professionals at the scene; and
  • The location of the fatal incident

Data sources.  Researchers drew information from the following data sources:  police records; social service reports; court documents; newspaper accounts; autopsy reports; mental health records; hospital and public health/medical data; and, other information that may have had a bearing on the decedent and her/his family.  They also interviewed professionals, including, but not limited to, police, court personnel, mental health workers, social service providers and advocates for battered women.

Disparity in numbers.  Although the Florida Department of Law Enforcement (FDLE) documented 230 domestic fatalities in Florida during 1994, the research revealed a total of 328 domestic fatalities that year.  [footnote 29]

The disparity stemmed from four major issues:

  1. Police departments often do not include child deaths due to abuse and neglect as part of their official domestic homicide count.  The researchers included these deaths.
  2. Police Departments often do not include the suicide victims in domestic homicides-suicides in their official count.  The researchers included these deaths.  (However, the researchers did not include deaths from suicide related to domestic violence.  This unknown figure represents a huge number of potential deaths stemming from domestic victimization and is an area in urgent need of systematic research and policy initiatives.)
  3. Police sometimes did not code domestic deaths as such.
  4. Police departments did not include boyfriend/girlfriend deaths as domestic homicides because those deaths did not strictly meet the terms of the statute.

Adopting a broader definition of domestic homicide than law enforcement sources, the researchers showed that in 1994 approximately one-third of all Florida homicides were related to domestic-violence.  [footnote 30]  This ratio contrasted sharply with official police data, which identified only one-fifth of all homicides in Florida in 1994 as being caused by domestic violence.

Essential findings from the Florida Fatality Review Project.

  • The analysis indicated that 294 of the 328 fatalities were consistent with the Florida Domestic Violence Statute.  [footnote 31]  The 34 remaining domestic fatalities either fell outside the statute criteria (e.g. victim and perpetrator were not married, lived at different addresses, and had no children together) or the researchers simply did not have enough information to determine if they met all the criteria of the statute.
  • Men perpetrated nearly all cases with multiple victims.  In only six cases did a woman kill more than one victim, or murder her partner and then commit suicide.  In no case did a woman murder her husband, her children, and then herself.
  • Many of the factors present in the multiple domestic killings also appear in the killing of individual women.  Men killed most o the individual women.  Nearly all of these cases involved women who had an extensive history of violent victimization prior to being killed.  As the statistical analysis reveals, other important factors include prior threats to kill, escalating abuse, and obsessive possessiveness and jealousy on the part of perpetrators.  In fewer cases, there was prior documented involvement of police or other criminal justice agencies.  Of all adult women victims, only three were killed by other women.  Five adult female fatalities resulted from women killing themselves as a part of multiple killing scenarios.
  • When women are killed in either multiple or single-victim domestic fatalities, it is usually the final event in an abusive relationship of long standing.  When men are killed by other men or by women, it is rarely, if ever, the end product of a battering relationship in which the men are the victims of abuse.  When other men in domestic situations kill men, it is often because the two men are competing for a woman who has, in many cases, been victimized by one of the men.  Men perpetrated three-quarters of all adult male domestic fatalities.  Only one-quarter of the men who died were killed by women.
  • Women who killed men nearly always did so out of self-defense, or less often, the defense of their children.  These women have always, or nearly always, been pushed to the brink of human endurance by the batterers they eventually kill. While the killing of batterers by the long-standing victims of battering may not qualify as self-defense in a court of law, the act of defensive or preemptive violence by women is qualitatively different from the offensive acts of violence perpetrated by men against women.

The commission of intimate partner homicide by women varied considerably by race and ethnicity.  Although African-American women constitute roughly one-eight of Florida’s female population, they comprised 16 of the 24 women who killed their male intimates.  Black women who killed, like their Caucasian and Latino counterparts, were essentially backed into a corner with nowhere to go.  Like white women, some had children to care for, nearly all, if not all, had been brutalized by their intimate partners; and most had, for whatever reasons, not sought out or received support from criminal justice and other state agencies.  Their partners were often obsessively possessive, and a good number of these violent men had threatened to kill them.  From initial field interviews in Florida, it appears that black women are less likely to use shelter and criminal justice services than white or Latina women are, therefore, more likely to be entrapped to the point of committing lethal violence.  [footnote 32]

  • Missing data hamper the statistical analysis of child fatalities.  Nevertheless, there are certain themes that seem to pervade these tragedies.  The most common correlate is that the death of children resulting from abuse or neglect occurs in homes where caretakers tend not to be married.  About one-third of the perpetrators were mothers’ boyfriends, one-third were biological fathers, and approximately a quarter were biological mothers.  These men sometimes had criminal records, including a history of violence.  It is clear from multiple sources of data that child fatalities normally occur within the context of poverty, often abject poverty.

Research findings also revealed that 50 percent of the children about whom there is reliable data have been physically abused before, often for a long period of time.  However, it is not necessarily the cases that this prior abuse has come to the attention of authorities.  For example, very few of the families in which child fatalities occurred had prior documented contact with the police.

Children under five years of age are clearly the most vulnerable to violence.  Over half the child victims in the sample were under two years of age.  Those who were older were often killed with easily obtained firearms.

As part of the grant from the Violence Against Women Grants Office, the Florida Governor’s Task Force has provided technical assistance to four pilot Florida fatality review teams that are just beginning to review cases at this time of writing.  Using the Florida fatality review project and its methodological approach as a touchstone, teams in Metro-Dade (Miami), Volusia-Putnam (Daytona Beach), Palm Beach, and Hillsborough (Tampa) counties have formed, created operational guidelines, and constituted various subcommittees regarding matters such as ethics of death reviews, data collection, working with family members o the decedents, effecting policy changes, and introducing confidentiality legislation.

The issue of confidentiality and immunity from suits remains the biggest stumbling block in Florida, as teams negotiate their concerns about liability.  Some state statutes have already dealt with this matter, providing their teams with immunity from various legal actions stemming from the review process.  [footnote 33]  In Florida, initial reviews have begun with homicide-suicide cases where there is no pending criminal prosecution and cases effectively are closed and the risks of liability limited.

The Use of Witnesses:  An Emerging Issue for Fatality Review Teams

In Florida, as elsewhere, review teams are discussing many important issues.  One mater concerns the feasibility of bringing in witnesses to improve understanding of domestic deaths.  Other states have empowered teams formally in the area.  For example, the Delaware review team has the power and authority to administer oaths and to compel the attendance of witnesses whose testimony is related to the death under review.  It also can compel the production of records related to the death by filing a praecipe [footnote 34] (request) for a subpoena, through the office of the Attorney General, with the Prothonotary [footnote 35] (clerk) of any county.



“Red flags” identified with the 1994 fatalities in Florida

The researchers also identified “red flags” or situational antecedents to the fatalities presented in order of their documented frequency in the 106 cases where men killed intimate female partners:

  • Prior history of domestic violence (approximately 85 percent of cases).  Among these cases battered women often reported an increasing entrapment.
  • Obsessively possessive believes on the part of the perpetrator (approximately 70 percent of cases).  Stalking behavior, close surveillance, inability to sleep on the part of the perpetrator, acute depression, perhaps a history of medication use, history of suicidal ideation, or less commonly, documented suicide attempts often accompanied this.
  • Attempts to break away from the perpetrator, including divorce, separation, and estrangement (approximately 70 percent of cases).  In a number of cases of breaking away, researchers identified accompanying relationship difficulties regarding such matters as child custody/visitation.
  • Prior police involvement in the case (approximately 50 percent of cases involving lone women killed and 30 percent of cases where women died in the course of homicide-suicides or familicides).
  • Prior criminal history on the part of the perpetrator.  In 43 percent of those cases where men killed their intimate female partners in non-multiple episodes (67 cases), the men had prior histories of criminal behavior, nearly always involving violence.  In about a fifth of the multiple killings, the male perpetrators had prior criminal histories of violence.
  • Threats to kill the eventual victim (documented in approximately 30 percent of cases).  These often were communicated to family friends, relatives, neighbors, and others prior to the homicide.
  • Issuance of restraining orders (approximately 20 percent of cases).
  • Alcohol or drug use that often escalated prior to the fatal episode (approximately 20 percent of cases).


This is a thorny issue and some advocates for battered women have argued for retaining certain informal elements of fatality review [footnote 36].  Some particularly important questions relating to subpoena power are noted below:

  • Is the resort to subpoena overly inquisitorial and punitive?
  • Do teams themselves want the ability to subpoena witnesses?
  • Could family members of victims be subpoenaed?  Given the involvement of families of victims in homicide trials and their need for information about the death, closure, and input, how should teams work with grieving families?
  • What about the conflicts between confidentiality/immunity of teams from suits, and the need to be open and honest with families about death?
  • How might teams work with domestic violence shelters?  For example, ought shelter staff and records be available for review, or should they somehow be immune?  One can envisage many reasons for not allowing teams to access shelter contacts with women.  Would not this problem, already identified by a small number of African-American women in Florida, be something that an assessment of systems delivery would address?  Can women entering domestic violence programs be asked to waive access to their records in the vent of their subsequent deaths?  Is such a waiver not insensitive?  Realistic?  Or both?  Or, do a victim’s rights evaporate upon death?

The authors pose these questions and issues in the spirit of inquiry to encourage discussion and debate.  One specific area that needs to be explored further is the role of the judiciary in domestic violence fatality reviews.

The Role of the Judiciary in Domestic Violence

Fatality Reviews

What role should the judiciary and judges play in reviewing cases or situations known to the court where there is fatal domestic violence?  Should judges simply defer to others when such tragic circumstances arise or should they participate in some process to determine whether the system could be improved so future fatalities might be prevented?

At a practical level, what should judges do when legislators, the media, and others in their community raise a hue and cry regarding a highly publicized domestic violence fatality?  How should judges deal with the professional and ethical constraints, which limit and guide them as part of the third branch of government?  These questions inevitably surface when the issue of court involvement in domestic violence fatality reviews arises.  Fortunately, other disciplines, such as medicine, have addressed the same issues carefully and it is now part of their ongoing training, protocol, discussion and literature.  [footnote 37]  Perhaps it is time for the justice system to recognize it has a rol to respond in similar fashion.

It is accepted that judges may provide leadership in their courts and in their communities with respect to the fair, prompt and effective management of domestic violence cases. [footnote 38]  This role is consistent with their judicial duties as laid down by the ABA Code of Judicial Conduct, which specifically acknowledges that judges are in a unique position to administration of justice.  [footnote 39]  Under the ABA Code, judges are required to maintain professional competence in judicial administration and cooperate with other judges and court officials in the administration of court business.  [footnote 40]

As part of their leadership role and administrative responsibilities, some judges and courts have decided to play a key role in domestic violence fatality reviews.  [footnote 41]  While court and judicial participation in such reviews is still evolving, several models have emerged.   Some courts conduct internal (or in-house) reviews, while other courts participate in an external and formal team review.  [footnote 42]  Some courts participate in both kinds of reviews.  While some courts and judges have been active in convening such teams, other courts are more passive participants.  Each model seems to depend in part on the local judicial and legal culture, as well as the judges’ professional views of what role they and their courts can and should play.

One may ask, why include the judge or a member of the court in a domestic violence fatality review, however structured?  There are good reasons for doing so.  Judges and their staff, as well as attorneys and others who appear before the court, (including intake officers, social workers, probation officers, and others) usually follow established procedures in processing, presenting, and deciding cases.  If the case has the potential for harm to any litigant or family member, court procedures and risk assessments should be reviewed carefully from time to time for fairness and effectiveness.  Such review is from a systemic standpoint and should not be focused on blame or finger pointing.  Error recognition, accountability, honesty and systemic improvement should be the focus rather than denial, blame, and personalizing the review.

For example, in a criminal case at arraignment, the court must determine bail or release conditions.  If a defendant is ultimately convicted, the court must determine whether a defendant requires prison or a term of probation.  If probation is ordered, it comes with a wide array of services and/or sanctions, such as a jail term, a stay away order, domestic violence intervention counseling, substance abuse treatment, mental health counseling, and more.  If a judge sentences someone to probation and that person later kills a family member, the system might be reviewed to assure that the court had all the information before it by way of pre-sentence reports, related cases, criminal abstracts, and risk assessments, to fashion a fair and appropriate sentence that is cognizant of community safety, accountability and treatment issues.

Another good example lies in family court, particularly where he court is a unified family court or a closely coordinated juvenile and family court, as the judges are often well acquainted with the families under the one judge, one family, one service team concept.  [footnote 43]  Those calendars require a keen sense of which cases are risk laden.  They are often high volume calendars where the judge may make ten, twenty or more custody, visitation, or restraining order decisions daily.  The obvious question is whether the judge is privy to all relevant information and is utilizing a valid checklist or risk assessment when rendering these key decisions.

Historically, there has been very limited participation by the judiciary in domestic violence fatality reviews.  Because the judiciary is an integral part of the criminal justice system and the family court system, participation by the judiciary is increasing.  As discussed above, in the Charan investigation the final report recommended system improvements in a variety of areas that required court involvement, ranging from translators, to training for probation officers and judges, to opening up case records in family court services.  The Charan case is no exceptional.  Almost every community can recite a widely publicized killing, often accompanied by widespread media and legislative concern about whether the system is responsive to and protective of battered women and children.  It is far better to have a constructive review process in place rather than simply to react in ad hoc fashion, as has been the case historically.  Indeed, with the advent of these review teams across the nation, those jurisdictions without such teams will come under even greater scrutiny and pressure to create a domestic violence fatality review team.

What can judges and the judiciary do to address the concerns raised by those in the community who ask whether the problem of domestic violence is being addressed fairly, promptly and effectively?  Judges and the judiciary can learn from the track record of current review teams and from other disciplines, such as medicine, about how they deal with fatalities.  They can participate in forming their own domestic violence fatality review processes and start to discern how and where the various justice systems succeed and where they need improvement in protecting victims of domestic violence.  Additionally, they can make a real effort to assure their communities that the justice system fairly and effectively addresses the litigant’s legal problems and at the same time adequately protects them.  Ultimately, judges must create a legal and judicial culture of safety, which prevents future harm whenever possible.


When a SwissAir flight went down off the coast of Newfoundland in 1998, claiming the lives of over 200 passengers and crew, a very high profile investigation ensured.  Millions of dollars were spent and no stone left unturned in an effort to understand how such a crash could have occurred.  These efforts were designed to identify the cause of the crash, with a view of preventing similar tragedies in the future.  Given that many people use the airways, such efforts are entirely understandable.

Domestic violence claims the lives of several thousand people per year.  Because intimate family relationships are a part of most people’s lives, much more time should be devoted to exploring domestic violence deaths, with a view to preventing them.  This article has suggested that reviewing domestic violence fatalities is a laudable development, one that will help prevent future deaths in families, reduce domestic violence in general, improve the delivery of multiple services to families experiencing such violence, and make an unequivocal statement about the undesirability of this illegal and highly injurious behavior.  As is happening in the field of medicine, this article recommends creating a culture of safety in order to review domestic violence deaths effectively, honestly and openly.

We have highlighted several approaches to death reviews.  The results of these reviews have been used to implement, or at least suggest, greater system coordination in dealing with people victimized by family violence.  We have acknowledged the need for communities to develop their own review processes and the authors hope that some of the highlighted information will contribute to those developments.

Of especial importance, for the purposes of this article, is the recommendation that judges adopt leadership roles I working toward establishing local domestic violence fatality reviews.  Such leadership is entirely in keeping with the role of judges in communities and government.


  1. See Leape, L.L., 1994, “Error in Medicine,” JAMA, 272:1851-1857.
  2. Such a definition may exclude boyfriend-girlfriend relationships where there has never been cohabitation or a child in common even where the dynamics of the relationship and eventual killing precisely parallel those between couples covered by statute.  To qualify as an “intimate partner,” states often require the couple to have lived together at some point or to share a child in common.  For example, Florida Statute (1994 s. 741.28) defines “domestic violence” as “any assault, battery, sexual assault, sexual battery, or any criminal offense resulting in the physical injury or death of one family or household member by another who is or was residing in the same single dwelling unit.”  A “family or household member” refers to “spouses, former spouses, persons related by blood or marriage, persons who are presently residing together as if a family or who have resided together in the past as if a family, and persons who have a child in common regardless of whether they have been married or have resided together at any time.”According to the Bureau of Justice Statistics (BJS, 1998; Bureau of Justice Statistics,U.S. Department of Justice.  1998.  “Violence by Intimates:  Analysis of Data on Crimes by Current or Former Spouses, Boyfriends, and Girlfriends,” NCJ-167237 March 1998), in the United States during the 1976-1996 period, intimate partner murder fell by 36 percent from 3,000 (1976) to 1,800 (1996).  The number of U.S. women murdered by intimates fell from 1,600 in 1976 to 1,326 in 1996.  During the same period the number of men murdered by intimates decreased from 1,357 (1976) to 516 (1996).  This overall decline in intimate murder is most marked in the black community.  The per capita rate of intimate murders among blacks was 11 times that among whites in 1976, but only four times that among whites in 1996.
  3. For a discussion of these different categories, see Charles Patrick Ewing, 1997, Fatal Families:  The Dynamics of Intrafamilial Homicide.  Sage. Thousand Oaks,CA; Neil Websdale, 1999, Understanding Domestic Homicide, Northeastern Press. Boston,MA; for a discussion of parricide see Kathleen Heide, 1995, Why kids kill parents:  child abuse and adolescent homicide.  Sage. Thousand Oaks,CA.
  4. Evan Stark and Ann Flitcraft.  1995, “Killing the Beast Within:  Woman Battering and the Female Suicidality.”  International Journal of Health Services 25 (1): 43-64.) addresses the links between suicide, suicidal behavior, and domestic violencesWebsdale (1999) explores some of these links.
  5. Websdale, N & Johnson, B. 1997.  “Battered Women’s Vulnerability to HIV Infection,”  Justice Professional, vol. 10, no. 4, pp. 183-198.
  6. Editorial, “Promoting Patient Safety by Preventing Medical Error,” JAMA, October 28, 1998, vol. 280, no. 16, p. 1444-1447.  Quoted p. 1444.
  7. Maria Stone, 1995, “Domestic Violence Fatality Reviews.”Boalt Law School, p. 13.
  8. See for example the case of Leonard Morrow who murdered his wife Latonya and two young children before committing suicide.  The Hopkins County child death review team defensively concluded that “it does not appear that the system failed Mrs. Morrow,” but rather that “Mrs. Morrow failed to allow the system to protect her.”  (Cited in Websdale, N. 1998, Rural Woman Battering and the Justice System:  An Ethnography.  Sage. Thousand Oaks,CA.  p. 149.  Full discussion in Websdale, 1998, p. 147-150).
  9. See Stone, 1995, pp. 15-17 and especially note 59.
  10. Project Safeguard, 815 E. 22nd Avenue,Denver,Colorado,80205, 303/863-7606.
  11. This growing recognition of the links between adult parental domestic violence and child abuse, neglect, and death is reflected in recent legislation in Delaware.  Delaware Statute Title 13 s 2105 empowers a domestic violence coordinating council to investigate and review, through a review panel, the facts and circumstances in all deaths occurring in Delaware resulting from domestic violence.  Child deaths are to be reviewed jointly by the Child Death Review Commission and the domestic violence fatal incident review panel.  The death of a minor will be reviewed by the domestic violence fatal incident review panel only if the child’s parents or guardians were involved in an abusive relationship and where the minor’s death is directly related to that abuse.
  12. See Websdale, 1991, Chapter 6, Table 6.6.  Websdale shows that in just over half the 57 cases of domestic child homicides where two parents/caretakers were involved in the care of the child, the female parent/caretaker was being beaten by the male parent/caretaker prior to the homicide.
  13. Investigation, p.5.
  14. For a good recent discussion of these issues see Karen Wang, 1996, “Battered Asian American Women:  Community Responses fro the Battered Women’s Movement and the Asian American Community.”  Asian Law Journal 3:151-185.
  15. Report p. 7.
  16. Report p. 11.
  17. Report p. 12.
  18. Report p. 13.
  19. This includes deaths classified as homicides, suicides, unintentional injury, undetermined cause, those with inadequate certificates, and peculiar circumstances (asthma, AIDS).  This is not to suggest that the deaths of women aged over 60 are not due to domestic violence.  For example, “suicide pacts” where elderly men kill their female partners and then themselves cannot be assumed to be free of a history of domestic violence.  Indeed, gerontologist Donna Cohen found that homicide-suicides involving elderly women in West Central Florida from 1988-1994 doubled.  In all, such homicides accounted for 20 percent of the total homicides of people aged over 55.  Cohen also notes that while 50 percent of the women’s health had deteriorated, two-thirds had expressed “no desire to die.”  Evidence that women killed in so-called suicide pacts had expressed “no desire to die” may suggest they were being battered prior to their demise (Cited in Charles Ewing, 1997 p. 143).
  20. At time of writing the Philadelphia Team is in the process of producing preliminary systematic data on the deaths of women.  Contact Dawn Berney for details, 215-985-2500.
  21. By Rolanda Pierre-Dixon, Chair.
  22. Santa County Death Review Committee Final Report, October 1993-September 1997. p 14.
  23. The term “Asian” is not defined in the report.
  24. Santa Clara County Death Review Committee Final Report, October 1993-September 1997. p 15.
  25. Santa Clara County Death Review Committee Final Report, October 1993-September 1997. p 5.
  26. Report, p. 13.
  27. Wang, 1996.  She defines “Asian American” broadly to include “all persons of Asian ancestry living in the United States” (1996:152, n3).  This includes people from East Asia (including China, Japan and Korea), Southeast Asia (including Burma,Cambodia,Laos,Thailand,Vietnam), South Asian (India) and the Philippines.
  28. Asian women differ from white women in at least three ways.  First, Wang points to the fact that the majority of Asian women are immigrants and therefore experience numerous language problems.  These problems make it difficult for them to obtain help from police, social services, or immigration services.  For example, if police officers attending domestic disputes at Asian homes can understand the man and not the woman, it is likely that without special translator services, the Asian woman’s story will be marginalized or go unheard.  Second, the Asian culture emphasis on saving face and valuing the family above the individual, makes Asian women more hesitant when it comes to breaking up the family.  Such a pronounced belief in the sanctity of the family in the face of violent victimization, combined with a cultural antipathy toward divorce, may make it more difficult for white shelter workers and advocates to offer support and understanding to groups like Korean women.  Third, the traditional Asian gender roles of male provider and female homemaker are often disrupted by an American economy that requires both partners to work outside the home.  This may be seen as liberating by Asian women, but it may, as Wang points out, be very threatening to the partners of Asian women.  See Wang, 1996, p. 171.
  29. For a very detailed case study analysis of these deaths, see Neil Websdale, Understanding Domestic Homicide, Northeastern University Press. Boston,MA. 1999.
  30. Preliminary findings from 1995 reveal similar discrepancies between FDLE data and that number of domestic violence fatalities identified by the broader definition used by Drs. Websdale and Johnson.  Although FDLE identified 195 domestic homicides in 1995, as of October 1, 1998, Drs. Websdale and Johnson have confirmed at least 285 domestic violence related deaths.
  31. See footnote 1 for Florida Statute (1994) s 741.28.
  32. For further discussion of these and related matters, see Beth Richie, 1996, Compelled to Crime, Routledge,New York; Websdale, supra note 29, 1999.
  33. InNevada, information can be shared among team members regarding the decedent or any person who was in contact with the victim and any other information deemed by the team to be pertinent to the review.  This information is to remain confidential (N.R.S. 217.475 ss4).  In addition, each member of the team is immune from civil or criminal liability for an activity related to the review of the death (N.R.S. 217.475 ss 8).  Subsection 9 states that the “results of the review … are not admissible in any civil action or proceeding.”  In Delaware, the review process, and any records created by it, shall be exempt from the provisions of the Freedom of Information Act in Chapter 100 of Title 29.  All records and documents contributing to the formulation of reviews are deemed confidential.  Such records and documents are not subject to subpoena or discovery.  Team members will not be required to make any statements regarding review deliberations (Delaware Statute Title 13 s 2105(h)).  Likewise members and their agents will be immune from claims and not be subject to any suits, liability, damages or other recourse, civil or criminal, arising from any act, proceeding, decision or determination undertaken or performed or recommendation made, provided such persons acted in good faith and without malice in carrying out their responsibilities; good faith and lack of malice are presumed and the burden of proving otherwise falls upon the complainant (Delaware Statute Title 13 s 2105 (i)).
  34. A praecipe is an original writ drawn up in the alternative.
  35. A prothonotary is an officer who officiates as principal clerk of court in states such as Pennsylvania (Delaware Statute Title 13 s 1205 (d)).
  36. For example, Barbara J. Hart made this point in her recent speech on social justice and fatality reviews in a national summit at Key West,Florida, October 26-29, 1998.
  37. See for example, Editorial, “Promoting Patient Safety by Preventing Medical Error,” JAMA, October 28, 1998, vol. 280, no 16 p. 1444-1447.  The authors suggest that error in medicine is real and common and must be reduced.  They argue that if this error is met with blame and distrust, then suppression, stonewalling, and cover-up follow; all of which fail to reduce future harm.  Alternatively, they suggest that system changes can prevent harm to patients and led to the transition from a culture of blame to a culture of safety.  The editorial is well written and has generated much discussion nationally.  See also, Gawande, Atul, “When Doctors Make Mistakes,” The New Yorker, (Feb. 1, 1999) at p. 40.  This excellent and readable article details a difficult medical procedure and sets forth how a hospital morbidity and mortality conference addresses physician error and the procedures needed to overcome it. Query if courts and the justice system could adapt such conferences when preventable domestic violence takes place?
  38. “Judges must provide leadership in their courts and in their communities to ensure that family violence cases are effectively managed and that adequate resources are available.”  Family Violence:  Improving Court Practice, p. 15 (National Council of Juvenile and Family Court Judges, 1990).
  39. Canon 4.B of the revised ABA Model Code of Judicial Conduct states “A judge may speak, write, lecture, teach and participate in other extrajudicial activities concerning the law, the legal system, the administration of justice and non-legal subjects, subject to the requirements of this code.”  The commentary discusses how the judge is in a unique position to contribute to the improvement of the administration of justice including the criminal justice system.
  40. See Canon 3. C.1 of the revised ABA Model Code of Judicial Conduct (1990).  Most states have adopted some version of the American Bar Association’s Model Code of Judicial Conduct, which was revised in 1990.  See Shaman, Lubet and Alfini, Judicial Conduct and Ethics pp. 3-6 (Michie, 2nd ed. 1995).  The preamble to the 1990 version states that Judges “must respect and honor the judicial office as a public trust and strive to enhance and maintain confidence in our legal system.
  41. See Websdale, N., Sheenan, M., and Johnson, B. 1998.  Domestic Violence Fatality Reviews:  Summarizing National Developments.  National Council of Juvenile and Family Court Judges. Reno,Nevada.  This publication was part of the National Summit on Domestic Violence Fatality Reviews held in Key West, Florida In October 1998.
  42. See for example the Philadelphia Women’s Death Review Team:  An Interdisciplinary Team to Reduce the Number of Violence-Related Deaths of Philadelphia Women.  A paper summarizing the work of the team is available from the Family Violence Department of the National Council of Juvenile and Family Court Judges or from the Philadelphia Health Management Corporation,260 South Broad Street,Philadelphia,PA19102-5085.
  43. Typically a unified family court’s jurisdiction includes all juvenile cases (delinquency, status, detention, waiver and child abuse), divorce, paternity, adoption, nonsupport, guardianship of adults and children, civil restraining orders, civil commitment in mental health cases, and in some jurisdictions crimes within the family ranging from domestic violence to intra-family murder.  Traditionally, one judge or one judicial team hears all cases affecting one family and the judge has a broad array of services to assist these families.  See, e.g. materials on ABA Summit on Unified Family Courts:  Exploring Solutions for Families, Women and Children in Crisis (held May 14-16, 1998,Philadelphia,PA).


Neil S. Websale is an Associate Professor of Criminal Justice at Northern Arizona University completing a research sabbatical at Vanderbilt University.

Judge Michael A. Town is a Circuit Court Judge in Honolulu, Hawaii.

Byron R. Johnson is the Director of the Center for Crime and Justice Policy at Vanderbilt University.

Upcoming Events

  • No upcoming events
AEC v1.0.4