(A Study of Filicide-Suicide Following Separation)
Carolyn Harris Johnson
The University of Western Australia
Abstract
This article is based on a qualitative study, which examined in depth, seven cases of familicide (where
disputed residency and contact was identified as being an issue between the parties) in Western Australia.
These offences typically consist of one or more young children being murdered by their father, the
homicide being followed by the perpetrator’s suicide. Common factors are identified, and suggestions
made, for screening and pro-active management of high-risk cases. Possible explanations of perpetrators’
motivation are proposed. Recommendation for future policy and practice are made.
Introduction
Familicide in families where there has been a marital separation, and subsequent dispute with
regard to custody of and/or access to children, is a significant problem. This is not only because
of the needless loss of life, but also because of the resultant trauma caused to families, officers of
the court, legal practitioners and other members of the community. This trauma persists long
after the event. At this time, due to a dearth of research in this area, professionals do not have
enough understanding of the problem to be able to say how, and why, these offences occur and,
therefore, it has been difficult to propose how they might be prevented.
Aim
The aim of the research was to determine whether there were common factors amongst a cohort
of familicide cases in Western Australia. The purpose of the research was to increase
understanding of why these offences occur. It was hoped that in undertaking the study, factors
could be identified that may assist in identifying families at risk of familicide. If this proved
possible, then strategies may be able to be established, to prevent such crimes occurring. It was
also hoped that this study would contribute towards filling some of the existing gaps in
knowledge, and provide a basis for further research.
Design
The design used for the research was exploratory and descriptive, using a collective case study
approach. A qualitative methodology was chosen which used a combination of interviews and
documentary analysis to collect the data. Interviews were included as it was believed that the
experience of survivors was critical to understanding the phenomenon of familicide.
History of the Marital Relationship
Five of the seven relationships studied, were characterized by longstanding unhappiness and
discord, but information about the quality of the other two relationships could not be located.
Further, where the instigator of the separation was known, it had been the wife in every case, and
in each case, it appeared she had difficulty in getting her partner to accept the separation. Two
women reported trying to leave previously, but finding this to be very difficult due to a range of
factors including pregnancy, fear of violence, and family pressure.
Each of the survivors interviewed described the couple’s relationship as having been
characterized by male dominance and control. They described the men as having been
possessive, both in their relationships with their partners, and with their children. Each person
interviewed reported being well aware that the man would not tolerate his partner leaving him,
even though, in several cases, it was reported the man had been involved with other women prior
to the separation.
History of Domestic Violence
In relation to domestic violence, it was found that both the police and the Family Court had
minimal information pertaining to the history of violence in some cases, and no record of it at all
in others. Violence was under-recognized, and under-reported, by respondents. Some couples had
not used the Family Court, and those that had had not fully disclosed, either to the lawyer or to
the court, the extent of the violence. In addition, there was a discrepancy found between the
knowledge held in the community about previous threats made by the perpetrator to harm, and
documented evidence held by the police and the courts.
Even in cases where information pertaining to risk was available, it was apparent that courts
did not always respond adequately to women making applications for Violence Restraining
Orders. Where applications were successful, this did not seem to help women a great deal, as
the police seemed reticent to act upon the orders granted by the court. In one case, where a
man was found hiding in his wife’s roof, he was not charged with breaching his Restraining
Order. Police responses to complaints of stalking, and other threats and harassment from
women's ex-partners, were inadequate and appeared lacking in understanding the risks
involved, and what sufficient and appropriate strategies were required to respond to this type
of complaint.
Some women, who had reported initially that their partner had not been violent, would, during
the course of the interview, disclose that he had, in fact, engaged in a variety of abusive
behaviors. It seemed that women did not necessarily view breaking objects, smashing furniture,
or the destruction of personal property, as acts of violence. Similarly, making threats, stalking,
and breaking into their homes were not always seen by them as having been violent acts. There
was no recognition of the long-term effect on them, and their children, of witnessing violent
behavior and of being subjected to other types of abuse.
It was reported that, even where physical violence had not been present in the relationship prior
to separation, the separation triggered violence in the man. There were also reports of extreme
and long-term stalking behavior by the perpetrator after separation. There were two cases of the
perpetrator hiding in his ex-partner’s roof and watching her movements through the ceiling.
History of Mental Disorder in the Perpetrator
The existence of a mental disorder in the perpetrator was difficult to determine in the absence of
any psychiatric, or psychological reports or assessments. In most cases, the wives were alienated
from the perpetrator prior to the offence, and, therefore, did not have access to knowledge about
any illness he may have had, or of any treatment that might have been undertaken. There were
features of mental disturbance reportedly evident in the behavior of most of the perpetrators in
events leading up to the offence, but some of these only came to light afterwards. For example,
one man slept outside his ex-partner's house for days, wrapped in a sheet of plastic to protect
himself from the rain.
There were reports of obsessive and irrational jealousy by the men, including jealousy of their
own children.
Most women noticed some indication of what they thought was depression in the perpetrator
prior to the offence. In some cases this had been evident for some months. In others it had been a
recurring problem for years. In addition all respondents described dysfunctional dynamics in the
families of the offender. The existence of mental disturbance, or personality disorder, in
perpetrators is considered an area worthy of further research.
History of Sexual Abuse
When asked if there had been a history of sexual abuse in the relationship, most respondents
initially answered in the negative. However, it often became clear, as the interview progressed,
that there were indications in the perpetrator of sexual deviance, and/or instances of sexually
abusive behavior. From the evidence presented by survivors, it seems that sexual violence was
associated with obsession and the need of the perpetrator to control his partner. In one case there
had been serious allegations that the man had sexually abused his three children.
Separation
In all cases, the survivors interviewed, described the break-up of the couple’s relationship as
being extremely difficult. Each of them described the reluctance of the perpetrator to let his
partner go. Often it was reported the woman’s attempts to leave the relationship had spanned
years. Each of them described the final attempt as having been characterized by a firm resolve,
on their part, that this would be the final separation. It is interesting to speculate as to what extent
their steadfast resolve communicated itself to the perpetrator, and to what extent he may have
realized that there would be no turning back this time, regardless of the woman’s previous
history of leaving and returning.
Court Process at the time of Offence
As far as could be ascertained, by document study and respondent interviews, of the seven
couples studied three couples had never made an application to the Family Court. Of these three
couples, two had never used the Court in any capacity whatsoever, and one couple had simply
used the Court Registry to file a Deed of Agreement (which would indicate superficially that
there was no dispute). Another two couples had made a legal agreement about custody and
access using a Minute of Consent Orders. One of these was an Interim Order, where the case was
scheduled to re-appear for final orders to be sought. A full trial and a judgment of the court had
resolved another. Only one case appeared to have an obvious, current, dispute before the court.
However, it is clear that to accept this analysis at face value would risk making an overly
simplistic judgment of what was a very complex set of circumstances in every case.
Threats of Harm to Self or Others
In six of the seven cases it was known that threats had been made by the perpetrator to harm
himself, or others, prior to the offence. In the remaining case no information was available about
whether or not such threats had been made leading up to the offence. It seems that often threats
were veiled, and were not recognized as genuine threats by those who heard them. Although, in
some cases, relatives or associates of the perpetrator were aware of threats being made, it did not
appear that anyone took action in response to these, either by informing the police or by warning
the intended victim. In six cases there was a history of various threats to harm should the wife
succeed in leaving the relationship. Some threats had been issued long before the final
separation, and some had been present since the commencement of the relationship.
The Victims
Of the seven families studied there were fifteen child victims of filicide. Eight girls and seven
boys were killed. Their ages ranged from seventeen months to eight years. Fourteen were killed
by their fathers and one was killed by the step-father who had raised him from infancy. There
were six suicides and one attempted suicide by the men. The man, who survived his suicide
attempt, apparently changed his mind halfway through the act, and pushed open the door of the
vehicle, to allow oxygen to enter. Additionally, two members of the extended families died as a
result of heart attacks shortly following the offences. Survivors saw these two deaths as having
been a direct result of the children’s murders.
Long Term Effect on Survivors
Interviews with survivors confirmed the trauma of familicide is enduring, and far reaching in its
effects, not simply on survivors in the immediate family, but also on the larger extended family,
and amongst friends of the families. Effects ranged from severe illness causing death, chronic
mood disturbance, depression and other mental illness, to increased use of alcohol and drugs.
Post traumatic stress symptoms such as nightmares, intrusive thoughts, trigger stimuli avoidance,
and over-protection of children were commonly reported. Several survivors stated they did not
expect to ever fully recover from the event, and one woman, who suffered from acute guilt
related to her children's deaths, said she did not believe that she deserved to recover. It is likely
that the families of the perpetrators suffer in similar ways.
Criticisms of the Court Process
Women commented on the general inability of courts to fully understand emotional disturbance
resulting from separation, and the associated risks this posed to families, as well as their
impotence in restricting the behavior of disturbed individuals. Family Court Orders when
breached rarely incurred penalties. There were criticisms of lawyers, the Family Court, and of the
Family Court Counseling Service. Respondents, who were involved in Family Court disputes,
reported that even their own lawyers did not seem to take threats of violence seriously, and in
some cases discouraged their clients from reporting it to the court. Respondents saw it as being
inappropriate that women were expected to attend Court Counseling jointly with their partners
even when there had been threats of, or actual, violence.
It was deemed inappropriate, by women, that paternal family members were seen as suitable to
supervise access, when they believed the men’s violence had been learned in their families of
origin, and the paternal families either denied or did not recognize risk.
There was evidence that neither the police nor the courts were effective in ensuring the safety of
the women, or their children, when it was known there had been a history of threats to harm or
even kill, and, as has been reported in many other studies, Restraining Orders were hard to
obtain and were ineffectual.
Dispelling Myths
Key points have emerged from the study which indicate that, although there are common themes
in these offences, there still exist some current misconceptions in the community relating to
causal factors. For example, the Family Court has often been blamed for its handling of the
dispute, but in this study there had never been a court dispute in three cases, and the existence of
a current dispute was questionable in another two cases. The press, on occasions, has depicted
these offences as acts of love, rather than as acts of extreme violence. Although there were
reports that some of the men demonstrated love for their children, and were involved in their
care, in others it was reported the men had shown little interest in child rearing, or the day-today
care of the children, prior to separation. Another misconception is that homicide in these
cases is an isolated act of violence, yet the study showed a history of the perpetrator’s need for
power and control over his family and his use of violence to ensure this, to be common. It has
also been reported that familicide in these cases is a spontaneous act, but the evidence shows
clearly that these offences are very pre-meditated. Yet another commonly held belief is that
familicide is caused by the man’s lack of contact with his children, but in six of these cases the
man had contact, and used his time with the children to kill them. In one case where the man did
not kill his children during access, but shot them while they slept, he had consistently refused to
exercise access following separation, although consistently encouraged and supported to do so by
his ex-wife.
Discussion
This study’s primary finding was that there are significant commonalities in familicide offences,
where an apparent dispute about custody, and/or access, was a factor. It was found that domestic
violence was a known factor in every case where details of the spousal relationship were known.
It was clear from reading the Family Court files and the newspaper reports, as well as listening to
respondents, that in the majority of cases the men exhibited a proprietary attitude towards both
their partners and their children. Obsessiveness, a lack of individuation and pathological jealousy
(including jealousy of their own children) were shown by the men in relation to their partners, in
the majority of cases. Egocentricity and a history in the perpetrator of threats to harm self and
others were very common. Most of these factors were universally present, where detailed
information on the nature of family relationships could be accessed. It was also found in these
cases that, prior to the offence, there had been signs of deterioration in the perpetrator's mental
health, which affected his ability to cope with the marital separation. Where this was the case,
men did not appear to have sought treatment for their mental health problem, although, due to the
suicide of most of them, this could not be verified.
The research data suggests a strong link between familicide and fear of abandonment and/or loss
of control. It seems likely from the data that a dispute about custody, and access, could possibly
have been another manifestation of the perpetrator's inability to separate from, and/or relinquish,
control of his partner and/or children.
Suggestions for improvements in services included the way that victim services are delivered,
and the kinds of education provided to legal officers, and the judiciary, about domestic violence,
risk assessment and human behavior, plus how the police respond to cases of stalking and
domestic violence.
It was clear from information obtained from respondents that, in retrospect, they could see there
had been warning signs of the impending offence, but these signs were either not recognized or
not acted upon. Collaboration, and sharing of information, between family members and
community agencies was rare, and was a major factor in the failure to safeguard the children. In
some case concerns had been raised but not acted upon.
Towards Understanding
In addition to the significant commonalities already referred to, it is suggested that all of the
offences studied incorporate at least one of the following two homicide-suicide characteristics. In
some offences, there is evidence of the characteristics of retaliatory homicide, followed by
suicide, where the prime motivation appears to be revenge on the spouse by killing the child, or
children, and where a lack of individuation from the spouse, and consequential inability to
survive the separation, leads to the subsequent suicide. In others, there is evidence more
characteristic of a depression-driven suicide, where the man appears to lack individuation from
both his partner and his children, and can see no future for himself or the children outside of the
original family unit. The homicide component in the latter type of offence appears to emanate
from the man’s inability to conceptualize his children as having an existence separate from his
own. They, therefore, become victims, by extension, of his suicide. Although all of the offences
researched appear to have characteristics of one or the other of these categories, some appear to
have characteristics of both. Until follow-up research is undertaken with surviving perpetrators,
perpetrator motivation remains a matter of conjecture.
Implications for Policy and Practice
The study brings to light significant implications for policy, practice and further research. As
with numerous other studies in the area of domestic violence it was found that community
agencies and their employees often appear to lack awareness of the many forms of domestic
violence and how it may impact on victims. Where there is awareness, there is a lack of interagency
co-operation in sharing this knowledge. It is important agencies find more effective ways
of sharing information where risk factors are identified. It was also found that there needs to be a
greater awareness of risk factors.
The legal system appears to have inadequate ways of helping women disclose the extent of
violence, and threats, to which they and their children are exposed. Where women did disclose,
they did not report the full extent and history of the violence, and they reported being advised by
their lawyers that disclosure could disadvantage them in the way the court viewed them, and their
application, especially where there were no witnesses to the violence other than the children.
They reported the focus was on obtaining agreement at the cost of safety.
It is well known that domestic violence seldom occurs in the presence of witnesses other than the
children and that part of the dynamic of family violence its secrecy. Yet women still find they are
not always believed unless they have medical evidence of injuries or “credible” witnesses to the
abuse. It is clear that lawyers, the courts, and police still need to develop more effective ways of
responding to domestic violence, especially when couples have separated. This applies from their
first contact, and how they interview the woman to elicit her story, right through to any court
appearance and beyond. Those working in the area of domestic violence are aware that women
are seldom able to tell the full extent of the abuse they have suffered at the first interview, and
this needs to be borne in mind when collecting evidence or preparing affidavits. In cases where
there are indications of deterioration in the mental state of men, following separation from their
partners, and stalking and other types of harassment are present, there is a need for particular
vigilance.
Where a threat has been made to harm, it is important this should incur consequences, and there
is a strong argument that the author of such threats should be made to undergo risk assessment,
and any recommended treatment. This would enable an accurate determination of risk to be made
and for services to be provided where appropriate. It is clear that more attention should be paid to
the safety of children who are having access with a parent who has been violent towards his
spouse, even though there may be no apparent evidence of the children having been previously
harmed. In addition, it should not be assumed that the family members of a parent with a history
of violence, and/or child abuse, will be suitable supervisors of access. As previously noted, it is
more likely that the parent learnt those behaviors in their family of origin and that violence and
abuse will be minimized, ignored or denied in that family. Therefore, their supervision is likely
to be inadequate and the children are unlikely to be protected whilst on access.
There needs to be a comprehensive service for the families of familicide perpetrators. The
researcher was advised that the Victim Support Service does not access these families. The
explanation for this is the potential for a conflict of interest. It can only be assumed they shared
the loss of the children and experienced similar grief to members of the maternal families. In
addition, they have also to cope with the loss of the children's father and the knowledge that a
member of their family was responsible for the loss of the children. It is doubtful that existing
counseling agencies are able to meet the special needs of these families.
One way for Family Courts to intervene more effectively in high risk cases is for them to accept
therapeutic jurisprudence as a model for service delivery. This would provide a strong
opportunity to develop early intervention services, with a preventative focus. Cases could be
screened at an early stage and then wherever there is evidence of family violence, child abuse,
substance abuse, or mental illness they could be referred to a diversionary program. In cases of
domestic violence or mental disturbance further screening for homicide-suicide risk factors could
be employed. In the Family Court of Western Australia, a program named “Columbus” is
currently being piloted which deals with cases where family violence and/or child abuse are
alleged. This program provides for differential case management, and is more expansive than the
“Magellan” project in Victoria and the A.C.T. which deals only with child abuse. It is hoped
such programs may eventually be expanded to incorporate other areas of family dysfunction,
substance abuse, and mental illness, as it is unlikely that a purely legal solution is adequate or
appropriate in these cases. Another way to enhance the Court’s effectiveness would be for legal
practitioners and officers of the court to be trained sufficiently about the dynamics of family
violence, so they are more able to assist women to disclose.
*The term familicide in this study is used to refer to offences where an individual murders or attempts to murder their
partner and/or child or children, and then suicides or attempts suicide.
BIBLIOGRAPHY
Alder C. & Polk K. (2001) Child Victims of Homicide Cambridge University Press, Cambridge,
U.K.
Allen, N. (1980) Homicide. Human Sciences Press, New York, London.
(1983) Homicide Followed by Suicide: Los Angeles, 1970-1979. Suicide and
Life Threatening Behavior, Vol 13(3):155-165.
Ambrose P. et al (1983) Men After Divorce. New Society, June 1983.
American Psychiatric Association. (1994) DSM -1V Washington, DC.
Ammerman R. & Hersen M. (Eds) (1992) Assessment of Family Violence. John Wiley & Sons
Inc. U.S.A.
Bailey K. (1987) Methods of Social Research. Macmillan, New York.
Baron-Cohen S. (Ed) (1997) The Maladapted Mind. Psychology Press, East Sussex, U.K.
Berman A. (1979) Dyadic Death: Murder -Suicide. Suicide 9 (1)
Biddulph S. (1994) Manhood. Finch Publishing, Sydney.
Bonney, R. (1998) Homicide II. NSW Bureau of Crime Statistics and Research, Sydney
Bowlby J. (1979) The Making and Breaking of Affectional Bonds.Tavistock Publications
Limited. London.
(1971) Attachment and Loss Volume 1: Attachment Penguin Books Ltd., Australia.
(1975) Attachment and Loss Volume 2: Separation, Anxiety and Anger. Penguin Books
Ltd., Australia.
(1980) Attachment and Loss Volume 3: Loss, Sadness and Depression. The Hogarth
Press, London.
(1988) A Secure Base Clinical Applications of Attachment Theory Routledge, London.
Chafetz J.(1988) Feminist Sociology. Peacock, Illinois
Chapman J. & Gates M. (1978) The Victimisation of Women. Sage Publications,
Beverley Hills.
Daly M.& Wilson M. 1988 Homicide. Aldine de Gruyter, New York.
Deaton W.S.& Hertica M. Growing Free Haworth Maltreatment and Trauma Press, 2001.
Denzin N. & Lincoln Y. (1994) Handbook of Qualitative Research Sage California.
9
Durkheim E. 1968 Suicide, A Study in Sociology translated by J.A.Spaulding& G.
Simpson. Routlerdge & Kegan Paul, London.
Easteal, P.W. (1993) Killing the Beloved. Australian Institute of Criminology, Canberra
Elliott & Shanahan Research (1988) Summary of Background Research for the Development of a
Campaign Against Domestic Violence. Department of the Prime Minister and Cabinet.
Ewing C. 1987 Battered Women Who Kill. Lexington Books, Lexington Mass.
Finkelhor, Gelles, Hotaling & Straus (Eds). (1983) The Dark Side of Families,Sage Publications,
Beverley Hills.
Hore E. Gibson J. & Bordow S. (1996) Domestic Homicide, Family Court of Australia,Victoria.
James M. & Carcach C. (1997) Homicide in Australia 1989-96 Australian Institute of
Criminology, Griffith A.C.T.
Jenkins A 1993 Invitations to Responsibility Dulwich Centre Publications, Adelaide.
Jordan P. (1985) The Effects of Marital Separation on Men, Family Court of Australia, Research
Report No.6.
Johnson H. & Hotton T. 2003 Losing Control, Homicide Risk in Estranged and Intact
Relationships. Homicide Studies 7(1), 58-84.
Jones A. (1996) Women Who Kill. Beacon Press, Boston.
Lansky M. (1992) Fathers Who Fail The Analytic Press, Hillsdale N.J.
Lester D. (1986) The Murderer and His Murder. AMS Press Inc, New York.
Lunde D.T. (1975) Murder and Madness, Stanford Alumni Press, Stanford, California.
Mann C.R. (1996) When Women Kill. State Uiversity of New York Press, Albany.
Minichiello V. et al (1995) In Depth Interviewing:principles,techniques,analysis. Longman,
Melbourne.
Monahan J. (111981) Predicting Violent Behaviour Sage Publications Inc, Beverley Hills.
Mouzos J. (1999) Femicide; The killing of Women in Australia 1989-1998,Australian Institute of
Criminology Canberra. (2000) Homicidal Encounters A Study of Homicide in Australia
1989-1999. Australian Institute of Criminology, Canberra.
Muller Rene J. (1994) Anatomy of a Splitting Borderline Praeger,Connecticut, London.
Naylor B. & Neale D. (1988) Domestic Homicide, Victorian Law Reform Commission Australia.
10
O'Donnell C. & Craney J. (1982) Family Violence in Australia. Longman Cheshire Pty. Ltd.
Melbourne.
Phillips B. (1971) Social Research Strategy and Tactics. McMillan, New York.
Polk Kenneth (1994) When Men Kill. Cambridge University Press, Melbourne.
Report of the Task Force on Domestic Violence to the Western Australian Government, January
1986 Breaking the Silence.
Pritchard (1995).Suicide: the ultimate rejection: a psycho-social study. Open University Press,
Pennsylvania.
Radford J. & Russell D (1992) Femicide the Politics of Woman Killing Twayne Publishers, New
York.
Resnick P.J.(1969) “Child Murder by Parents: A Psychiatric Review of Filicide”, American
Journal of Psychiatry 126:73-82.cited in Wilczynski A.1997:45.
Retterstol N. (1993) Suicide A European Perspective. Cambridge University Press, Great Britain.
Rutter M. (1972) Maternal Deprivation Reassessed. Penguin Books Australia Ltd. Australia.
Saakvitne K. & Pearlman L.A.(1996) Transforming the Pain. W.W. Norton &Co, New York.
Shneidman E. (1993) Suicide as Psychache. Jason Aronson Inc. New Jersey.
Strang Heather (1991) Homicides in Australia 1989-90. Australian Institute of
Criminology,Canberra.
(1993) Homicides in Australia 1991-92 Australian Institute of Criminology, Canberra.
Strang H.& Gerull S. (Eds) 1993 Homicide: Patterns, Prevention and Control. Australian
Institute of Criminology, Canberra.ACT.
Wallace A. (1986) Homicide: the social reality New South Wales Bureau of Crime Statistics,
Sydney.
West D.J. (1965) Murder Followed by Suicide Heineman London.
Wilczynski Ania (1997) Child Homicide. Oxford University Press, London.
Wilson Paul R. (1985) Murder of the Innocents. Rigby, Adelaide.
Wolfgang M. (1958) Patterns of Criminal Homicide John Wiley & Sons, New York.
Women's Coalition Against Family Violence (1994) Blood on Whose Hands? Victorian
Women's Trust, Australia.