Report on sentencing for manslaughter in cases involving intimate relationships

International Data


The National Homicide Monitoring Program at the Australian Institute of Criminology, established in 1989, routinely collects data on some 47 variables relating to each incident of homicide that comes to police attention in Australia. It includes data relating to the victim, the suspect or offender (where one has been identified) and the setting or context within which the incident occurred - including time, location and weapon used. Data are derived exclusively from police records.

A total of 2226 homicide incidents were recorded from July 1, 1989, to June 30, 1996. The offender was known in 2024 of these incidents, and 543, just over 25%, involved intimate partners. The annual number of intimate homicide incidents in Australia remained at a relatively constant level from 1989 to 1996, in contrast to the numbers in the United States and Canada, which appear to be declining.

Most of the intimate homicides (77%) were committed by males against females; 21 % involved females killing males, and the remaining 2% involved other types of relationships, such as same-sex couples. In 63% of the cases when a male killed a female, the persons involved were in a current spousal relationship, and 67% of intimate homicide incidents were the result of altercations of a domestic nature occurring at the victim's or the offender's home. A sharp instrument such as a knife was the most common weapon used. According to the data, 36% of offenders were under the influence of alcohol at the time of the incident and 29% had a previous criminal record; however, the report cautions that one-quarter of the records do not provide this information, and it was therefore assumed that neither alcohol nor previous criminal record was a factor in those cases. One in three intimate-partner homicides resulted from conflict associated with jealousy or the termination of a relationship. The remaining incidents were classified as "general domestic altercations."

Earlier research indicates that intimate homicides are often the unplanned consequences of events that escalate out of control. The report concludes that the private nature of intimate-partner homicide, which makes it difficult to obtain reliable and comprehensive data, also makes any policy intervention difficult. 38

In the early 1990s, an Australian study looked at sentencing patterns in specific cases of intimate partner homicide. The study analyzed 10 cases where women who were battered were charged with killing a partner and 20 cases of men who either pleaded guilty to or were found guilty of either murder or manslaughter.

Concerning the female offenders, the study concluded that there was no clear-cut relationship between most of the background variables and the types of sentence that the battered woman received. If the woman was Aboriginal, had an alcohol problem, or committed the homicide without intent to kill and showed remorse, she received a non-custodial sentence. The two women who received the longest sentences were found guilty of manslaughter rather than murder on the grounds of either diminished responsibility or provocation, but not because of a lack of intent to kill. In the provocation case, the offender had been subject to extreme violence from the deceased, yet that factor did not appear to weigh in her favour; possibly, it was negated by the degree of violence that she used, her lack of remorse, and the intent to kill. A traumatic childhood and alcoholism seemed to act in the defendant's favour only if accompanied by one or more of these crucial variables. Neither self-defence nor the battered woman syndrome was mentioned in any of the ten cases analyzed. The report noted that although battered woman syndrome has been a viable defence in the U.S. since 1979, it was not until 1992 that it was recognized in an Australian case.39

The review of the male offender cases revealed what it described as "a rather large range of sentences without any obvious thread of consistency." The report suggests that this may show a glaring lack of a guiding principle, though it also acknowledged that such a principle might simply not be clear from the limited information found in sentencing decisions. The report states as well that there does not seem to be any trend when contrasting those guilty of murder with those guilty of manslaughter. The study looked at such variables as premeditation, means of killing, intoxication and mental illness, though the various judges and states had different definitions of each of these factors and their importance and impact.

One of the few consistent elements in mitigating sentencing and/or determining whether to allow the defendant to plead to manslaughter was the perception of the nature of the victim and her degree of compatibility with societal norms The judges' remarks clearly indicate what they believed the ideal wife should - or shouldn't - be. Some of the variables identified as not associated with the ideal woman were leaving one's husband, having an affair, not taking care of the children, nagging one's husband, and failing to appreciate the husband's work on behalf of the family. For example, if a male offender had been left by his female partner, even in response to a history of being battered by him, the court would often accept that this man had been provoked by jealousy and the loss of his children. At the same time, if a male had met the cultural values for male behaviour (e.g. worked hard to support his family), he was likely to be sentenced to less time in prison than someone who was unemployed or had left his family.

The three Aboriginal cases were also informative, as they proved to be exceptions to the general trends cited above. The court seemed to define what was culturally normal or ideal in a substantially different way for Aboriginal males and females than for non-Aboriginals. For example, alcohol abuse was accepted as a mitigating factor in the Aboriginal cases, but not in the non-Aboriginal cases.

The report concludes that the number of what it calls "lenient sentences" perpetuates the notion that domestic violence is not treated as a serious matter.40


Sweden has an international reputation as a peaceful country; for instance, at the United Nation's Conference on Women in Beijing in 1995, Sweden was celebrated as the nation with the highest rate of equality in the world. But according to an article by an employee of Sweden's Crime Victim Compensation and Support Authority, many women in Sweden have historically believed that although the situation may be better than in some other countries, Swedish women still do not have equality vis-à-vis their male counterparts. The author points out that official statistics for salaries and corporate power structures confirm that equality in Sweden is a myth. Men have enjoyed a privileged position in Swedish society, dominating the legal system as legislators, judges and law professors, with the article suggesting that this has resulted in the development of legal concepts from a male perspective, which has led in turn to gender biases when disputes involve intimate relationships.

In 1993, the Swedish government convened a Commission on Violence Against Women, which was seen as quite unique. The directive stated that the Commission should work "from a women's perspective," and it stressed that violence against women had to be seen in relation to "an unbalanced distribution of power between men and women."

The Commission presented its main report, Kvinnofrid ("Woman's Peace"), in June 1995. The report considered what the UN, the European Council, the Nordic Council, Canada and Norway had done with respect to the issue of gendered power and violence, and concluded that violence in intimate relationships is very different from other situations. For instance, the report pointed out that abused women are totally vulnerable in all aspects of their lives, not only in terms of physical or sexual abuse. It gave examples of partners hiding telephones or keys, forbidding contact with family or friends, and subjecting them to constant verbal abuse, and concluded that reality for women in such situations was "dominated by violence and threats".

As a result of the Commission's work, a new crime of "Violation of Woman's Peace" was proposed. It was intended to address violence and other abuses directed at women in close relationships with men. In particular, the proposal targeted certain types of behavior, which alone would not necessarily constitute criminal offences, but in the context of an intimate relationship, should be treated as criminal, in the Commission's view. Another goal was to make other criminal sanctions (e.g. assault, threats) easier to prove. The proposed text was to have been included in Chapter Four of the Penal Code, "On Crimes against Liberty and Peace," but it was not enacted. It would have read:

A man who against a close-related or former close-related woman uses violence or threat of violence or subjects her to other physical or psychological influence, aimed to lasting insult of the woman's integrity and hurt of her self-respect, shall be judged for violation of woman's peace, to imprisonment for not less than one year and not more than six years.

The same penalties would also have applied in relationships between two men or two women.

Responses to the proposals were mixed. Woman's organizations and the Equal Opportunities Ombudsman were positive. However, the leading law faculty (Uppsala University), Sweden's Lawyers' Association, and the judiciary expressed serious concern about the proposed legislation.

The legislation eventually enacted in 1998 was described as being softened from what was originally contemplated by the Commission, as described above. In order for an act to be criminal, it must involve an intimate partner and be part of a repeated pattern that harms the person's well being. The legislation states that:

A person who commits criminal acts as defined in Chapters 3, 4 or 6 against another person having, or have had, a close relationship to the perpetrator shall, if the acts form a part of an element in a repeated violation of that person's integrity and suited to severely damage that person's self-confidence, be sentenced for gross violation of integrity to imprisonment for at least six months and at most six years.

If the acts described in the first paragraph were committed by a man against a woman to whom he is, or has been, married or with whom he is, or has been cohabiting under circumstances comparable to marriage, he shall be sentenced for gross violation of a woman's integrity to the same punishment. (Law, 1998: 393)41

Although the Commission originally suggested in its report that the minimum penalty should be one year, the government believed that six months would be a sufficient deterrent.

Feedback on the new legislation has been mixed. On the one hand, statistics indicate that a number of charges have been laid and offenders have been sentenced to prison. On the other hand, some critics have expressed concern about the quality of police investigations.42

United States

The Bureau of Justice Statistics (BJS), established in 1979 under the Justice Systems Improvement Act, is a component of the Office of Justice Programs in the U.S. Department of Justice. The BJS mission is:

To collect, analyze, publish, and disseminate information on crime, criminal offenders, victims of crime, and the operation of justice systems at all levels of government. These data are critical to Federal, State, and local policymakers in combating crime and ensuring that justice is both efficient and evenhanded.

The BJS published two reports in 2001 that examined intimate partner violence from 1993 to 1999 and Intimate Homicide Trends in the U.S. between 1976 and 1999. What follows are the highlights of those reports.


Females were more likely than males to be murdered by an intimate partner in 1999, as they have been historically - 74% or 1,218 of the 1,642 persons murdered that year by an intimate partner were female. Moreover, between 1976 and 1999, about one-third of all female murder victims were killed by an intimate partner. Nonetheless, overall numbers and rates of intimate partner homicide have fallen significantly. The number of women killed by intimates was stable for two decades and declined after 1993. In 1976, 1,600 women were killed by an intimate partner, and by1999, this number had dropped to 1,218 women.

Although the intimate homicide rate has fallen for blacks in every gender and relationship category, the rate for white females has not declined in all categories. For example, the rate for white girlfriends was higher in 1999 than it was in 1976. Furthermore, despite the decline starting in 1993, intimate partner homicide rates increased between 1997 and 1999 for females aged 35-49 (the most vulnerable age category) and 65 or older.


Male murder victims were substantially less likely than female murder victims to have been killed by an intimate partner - it accounted for about 4 % of all male murder victims from 1976 to 1999. Of the 1,642 persons murdered by an intimate partner in 1999, 424 or 26% were men. In 1976, 1357 men were killed by an intimate partner, but this number dropped to 424 in 1999.43

Finally, it is worth recalling Marvin Wolfgang's 1958 classic study of 588 homicides in Philadelphia,44 which revealed substantial differences in the ways men and women are killed in cases of intimate partner homicide. In the 47 cases in which wives killed husbands, Wolfgang concluded that 38 of the husbands had "strongly provoked" the killing, and in 28 cases precipitated their own deaths by striking a blow against the woman or showing and using a deadly weapon. This compared with only 9% of wife killings that Wolfgang deemed "victim precipitated." These findings on the gendered nature of intimate partner homicide have been replicated in numerous other studies.45

Great Britain

The "Homicide in Britain Study," completed in early 2002,46 was new research that examined the killing of women and children within the family context. The study attempted to develop a typology of different types of murder and to find out the murderers' motivations and intentions. Various types of risk factors were examined, including childhood and background characteristics of the offender, the relationship between the offender and the victim, and circumstances in which the murders occurred.

The main findings relating to intimate-partner murders were as follows:

  • 75% of men who kill women in intimate relationships are likely to have problems in relationships with women in general.
  • 44% had a previous conviction for violence, more than half of which involved violence against women.
  • 60% had used violence against a previous female partner, and 59% had committed non-lethal violence against the woman they eventually killed.
  • The murders occurred when the couples were married (41%), cohabiting (32%), dating (25%) and divorced (2%).
  • Possessiveness and jealousy were clear factors in 26% of the cases that ended in partner murders.

Date modified: