Gender – Domestic Violence: ‘Shameful Statistics Exposed’

People’s Equality Network – May 1995

Domestic Violence – ‘Shameful statistics exposed’


Violence against women continues to be a significant issue in Australia. Dr Carmen Lawrence, in a press-release on ‘National Stop Violence Against Women Day’ (26 April) issued by the Office of the Status of Women, claimed that there were 5000 women and children seeking refuge from domestic violence each night in Australia, that 14000 cases of domestic violence – implied as solely against women – were reported by Victoria Police in 1992, and that 70 percent of police time in New South Wales was taken up in dealing with domestic violence. The theme of this seventh national day on domestic violence was “We Are Breaking the Silence – Your Voice will Make a Difference”; as part of that ‘breaking of the silence’, Dr Lawrence said, “these statistics are shameful and demand exposure”.

However, further analysis indicates that almost all statements in the press-release which purport to be fact are in reality either seriously misleading or seriously incorrect. Most of the figures stated in the press-release – such as those above – would seem, on detailed scrutiny of their sources, to be exaggerated by many orders of magnitude. Whilst the fact of violence against women is indeed ‘shameful’, the fact which urgently needs to be exposed is that violence against men is far more common than violence against women; and the statistic which is truly shameful, and which truly demands exposure, is that women’s violence against other women, and particularly against men, is in reality little different from that of men’s violence against women. This fact is shown by a more detailed analysis of a recent study on domestic violence by the Monash University Accident Research Centre.

The statements in the press-release, which were attributed personally to Dr Lawrence and were issued by her personal office, continue a persistent habit of exaggeration and misinformation on gender-issues by the Office of the Status of Women and other non-government organisations – such as the Women’s Electoral Lobby, of which Dr Lawrence is a prominent member – whose sole purpose appears to be to provide funding and other support for an extreme form of gender-politics which has little or no basis in fact. Such a campaign, and the concomitant misuse of government funds, would appear to be tantamount to fraud on a major scale. Detailed investigation of the activities of the Office of the Status of Women and of the recipients of its funding would appear to be urgently required. It is also particularly disturbing that a politician of such responsibility, reputation and general public respect as Dr Lawrence should lend her name to a campaign of such serious and apparently systematic misinformation: parliamentary investigation would appear to be recommended.

Data source and content

This report is an analysis of a press release issued on 26 April 1995 – ‘National Stop Violence Against Women Day’ – by the Office of the Minister for Human Services and Health (Dr Carmen Lawrence, who is also Minister Assisting the Prime Minister for the Status of Women). Additional data is derived from the Monash University Accident Research Centre (MUARC) report “Domestic Violence: Patterns and Indicators”, prepared by Virginia Routley and others (MUARC Report No. 63, November 1994); the Victoria Police Crime Statistics 1993/94 official report, and other stated sources.

This report is in part also a follow-up to the Peoples Equality Network report “Domestic Violence: recent statistics in Victoria“, published by PEN on 15 January 1995: whilst the former concentrates on statistics published in a shorter version of the MUARC report, this report is primarily concerned with public presentations of and responses to the problem of domestic violence.

Minister’s press release

The text of the press release issued by Dr Lawrence’s office was as follows:

All Australians – men and women – are being urged to speak out against violence on National Stop Violence Against Women Day.
The theme of this, the seventh national day, is “We Are Breaking the Silence – Your Voice will Make a Difference”.

“Violence against women remains one of the most under-reported of all crimes in Australia,” the Minister Assisting the Prime Minister for the Status of Women, Dr Carmen Lawrence said.

“There are no national statistics on the nature and extent of violence against women but we do know that on any night in Australia, approximately 5,000 women and children seek accommodation in refuges, most of whom are escaping violence.

“We also know that domestic violence comprises one of the largest areas of police work – some 70 per cent according to New South Wales figures.

“Victorian Police report that they received more than 14,000 calls involving domestic violence in 1992.

“And in Queensland, a 1992 study showed that one in five women admitted to the Royal Brisbane Hospital’s emergency department was a victim of domestic violence.

“These statistics are shameful and demand exposure.

“By speaking out against violence against women we can help remove the cloud of fear and secrecy that prevents many women from seeking help.
“Open debate will also enable us to accurately consider what progress we have made against violence and how much more we still have to do,” Dr Lawrence said.

National Stop Violence Against Women Day 1995 also marks the culmination of the Commonwealth Government’s three year Community Education Program.

$3.8 million has been allocated to more than 50 community projects and two national education campaigns to foster a climate of opinion intolerant of violence against women.

The press release ended by detailing some of these projects.

Peoples Equality Network would agree with Dr Lawrence that it is the responsibility of all Australians – men and women – to speak out about domestic violence against both sexes: the statistics quoted in the report, if true, would indeed be “shameful and demand exposure”. But if they are significantly incorrect, they would constitute a serious mispresentation of a significant societal problem, and would indicate that funds and other resources are likely to be used in ways which are inefficient, inappropriate, or even counter-productive. On the theme of ‘Breaking The Silence’, this report will analyse the background behind each of the key assertions made in this press release.

Monash University Accident Research Centre study on domestic violence in Victoria

An analysis of the content of the MUARC study has been published in a previous PEN report, as mentioned earlier. That report was, however, based on the shorter version of the study’s findings; the full version has now been made available by the study’s authors. Some important additional material is provided in the full version which impinges on Dr Lawrence’s statements.

The MUARC study was based on a methodology developed by Stark, Flitcraft and others in the USA, and consisted for the most part of a study of one-line injury-incident records collected by five major Victorian hospitals participating in the Victorian Injury Surveillance System (VISS) during 1991 and 1992, covering a population base of approximately one million people. Stark and Flitcraft had assumed that domestic violence was essentially a female problem – their basic model states that domestic violence is “an inherent by-product of a male-dominated patriarchy” – and had stated that male injuries were rare and were almost always the result of self-defence by a female. They therefore provided no descriptions of ‘typical’ male injuries; their own results indicated that up to 50 percent of women’s injuries presenting in the emergency department were attributable to domestic violence. The MUARC study initially used this model as a guideline, but found it seriously flawed, and needed to develop a number of corrections to the methodology.

The MUARC team, which wherever practicable excluded injuries to perpetrators, identified 239 female and 49 male cases of explicit partner-on-partner domestic violence. (The figure of 239 female cases represented approximately 1.3 percent of the total female injury caseload.) The sex-ratio of 5:1 indicates a strong disparity against women, but also indicates that, even at the explicit level, male injuries were not particularly rare. Male injuries were typically also considerably more serious than women’s – more than 70 percent of male cases were open wounds, compared to 26 percent of female cases – and resulted in a greater proportion of the hospital admissions (variously described as 28 or 40 percent); although the figure was not given in the MUARC report, Dr Joan Ozanne-Smith, another member of the team, stated on Radio National’s “Life Matters” programme that the male rate for domestic-assault injuries classed as ‘very severe’ was almost double that for females. Forty male and 37 female cases whose description consisted solely of the word ‘Domestic.’ were carried over to the ‘probable’ category.

‘Probable’ partner-inflicted assaults were defined as ‘injury resulting from assault which is neither a street assault, a robbery or a mugging’. 265 ‘probable’ female cases were identified from a detailed analysis of female injury descriptions. However, because of the absence of adequate descriptions in the Stark and Flitcraft methodology, no equivalent analysis could be undertaken of male injury descriptions; the MUARC team noted that 56 percent of assault injuries in the home were to males, and 72 percent of injuries in another’s home, but decided instead to arbitrarily apply the same 5:1 ratio derived from the ‘positive’ cases, giving 53 male cases. The exact selection criteria used for this analysis were summarised in the MUARC report as follows:

1. begin: n=18439 female injury cases
2. apply VISS codes ‘for intended violence between persons, cause of injury other person’ (VISS definition of assaults): n=932 female assaults
3. exclude ‘domestic with no further qualification’ cases, ‘positive’ cases [i.e. those already in the ‘positive’ or explicit category] and those cases for whom a partner was clearly not the assailant: n=265 probable female assaults
4. apply male:female ratio 1:5, as for positive cases. Thus add 53 males to the females: n=318 probable partner-inflicted assaults

(For reasons which were not adequately explained, neither of these figures – 265 female, 53 male – included the group of unspecified-‘Domestic.’ cases carried over from the ‘positive’ category: these cases seem to have been discarded, though they appear to be included in Table 11 in the report.) Since other evidence – even a simple ‘reality check’ – indicates that males are less likely than women to attend hospital for any injury, and far less likely to describe their injury as partner-inflicted, the arbitrary application of the 1:5 ‘positive’ ratio defeats the whole object of the ‘probable’-category analysis, which is to estimate cases not positively identified as explicit partner-inflicted domestic violence. A more consistent application of the methodology according to the site-of-injury percentages would suggest that a more likely figure for males would be somewhere in the 200-300 range, rather than the 53 arbitrarily derived by application of the previous ratio.

A further ‘suggestive’ category, where the injury is of a particular type – head and shoulder injuries for women, according to Stark and Flitcraft, and lacerations or impact wounds for men, as identified by the MUARC team – and where the explanation given for the injury may not be consistent with the injury itself, was the source of most injuries identified in the Stark and Flitcraft studies and in similar studies by Roberts and others in Brisbane. However, serious design flaws in the Stark and Flitcraft methodology – which resulted in car-accident injuries, for example, being classified as domestic violence – led to artificially high incidence figures of up to 50 percent; the MUARC team, by application of more rigorous selection criteria, identified only a further 261 ‘suggestive’ cases, bringing the female domestic-violence injury total – positive, probable and suggestive – to somewhat under 5 percent of all female injuries presenting to hospital. Application of the same methodology to male injuries resulted in 767 ‘suggestive’ male cases; since this was almost three times the female figure – and in total conflict with the Stark and Flitcraft model – the MUARC team once again arbitrarily applied the 1:5 ‘positive’ ratio, deriving only 52 male cases for this category. The selection criteria published in the report were as follows:

1. begin: n=18439 women’s injury cases
2. select cases with injuries which are common to identified [female] positive domestic violence cases, i.e. bruising to the head and face (including eyes, nose, mouth and ears), neck, forearm and trunk plus nose inflammation or fracture, face inflammation, swelling or pain and lacerations or fractures to the mouth: n=2531 cases
3. exclude unlikely groups, i.e. women aged over 50 years, injuries arising from work, horseriding, sport, violence (included in ‘positives’ or ‘probables’) and victims willing to be contacted later for research purposes and therefore needing to be consistent with their story: n=261 female suggestive cases
[note: this additional ‘filtering’ stage is missing from most Stark & Flitcraft-derived studies, resulting – as shown above – in over-estimates by a factor of ten or more]
4. apply male:female ratio of 1:5 for positives to 261 suggestive females, giving 52 males: n=313 cases suggestive of domestic violence

There is no methodological or other justification for the arbitrary application of the 1:5 ‘positive’ male:female ratio, especially as there are strong indications both from sociological studies and the report’s own data that the male injury figures are, in reality, likely to be similar to those for female injuries in this category.

The total identified female cases – 239 positive, 265 probable and 261 suggestive – were thus derived from detailed analysis; but of the male cases – given as 49 positive, 53 probable and 52 suggestive – only the first figure can be considered reliable. A consistent application of the methodology (including correction for higher overall numbers of male injuries) suggests figures of 49 positive, roughly 250 probable, and roughly 300 suggestive male cases. Although this is in conflict with Stark and Flitcraft, it conforms exactly to the predictions of most both-sex studies of domestic violence, such as those by Suzanne Steinmetz, and the National Family Violence Surveys co-ordinated by Straus and Gelles in the US.

The study’s analysis of coronial homicide records was similarly inconsistent: the MUARC team derived figures of 28 female deaths and 2 male, but only stereotypically ‘male’ forms of murder, such as shooting, strangulation, stabbing/throat cut and beating/ bashing/ assault were studied; stereotypical ‘female’ forms of murder, such as poisoning, or driving the other to suicide, were excluded.

The expanded study also provided examples of relevant descriptions. A typical male-on-female violence description read “assaulted by husband, pushed to ground, kneed in abdo, & face, pulled hair & twisted breast”. A typical female-on-male assault read “chest inflicted by girlfriend, 2 wounds, half inch wide, 25cm knife”. Stabbings also occurred in female-on-female violence – one example read “had an argument with daughter who stabbed her with a carving knife” – while another example read “had taken amphetamine injection, assaulted by 3 girls, hit with baseball bat”. All of these would be classified as ‘family incidents’ in police records.

Victorian Police Crime Statistics 1993/94

A total of 13485 ‘family incidents’ were reported to the Victoria Police during the year 1993/94. Since in some cases incidents occurred at the same household several times in the year, this figure represents significantly less than 1 percent of Victorian households (roughly consistent with the estimate of 0.7 percent derived from an interview survey by the Australian Bureau of Statistics in 1992). Some additional statistics are given in Section 10 (pp.112-118) of the Crime Statistics report. 10797 (or 80.1 percent) of the complainants were female, and 2433 (or 18.0 percent) were male; the complainant was unspecified in 255 cases. The respective figures for the ‘other party’ in the incident were 1833 female, 11208 male and 444 unspecified. Rather less than three-quarters (73.7 percent) of the incidents involved spousal or ex-spousal types of relationships; the remainder consisted primarily of parent/child relationships (of any age), with a significant proportion of ‘other family’ or house residents. ‘Violence against the person’ was identified in 1851 cases (13.7 percent) and may have occurred in a further 2890 cases (21.4 percent). Child abuse occurred in 75 cases (0.6 percent), including 28 females and 33 males under the age of ten. Well over a third of all incidents (5301 cases, or 39.3 percent) were the result of verbal abuse only, with a further 2138 (15.9 percent) apparently too trivial to specify. Access, custody and maintenance disagreements accounted for a further 1055 incidents; the remainder were ‘offensive behaviour’ (1110 cases), ‘harassment’ (512 cases) or ‘other’ (3194 cases). Alcohol and other drugs were definitely involved in 4811 cases (35.7 percent) and possibly involved in 2798 others (20.7 percent) – a total of more than half of all incidents. Firearms were actively involved in 89 cases (0.6 percent), but used in only four; the incidents included six homicides, three rapes and nine abductions.

Police applied for an intervention order in 3523 cases; breakdowns of this figure by sex, by type of incident or by type of relationship were not provided. Of 1447 charges laid, 751 were for ‘crime against the person’, comprising 51.9 percent of the charges and 5.6 percent of all incidents. Overall, 1309 males and 98 females were charged, representing 11.7 percent and 5.3 percent of the ‘other party’ figures respectively; assuming that (as is likely from hospital data) assaults were of similar severity, this indicates that men are more than twice as likely as women to be charged for the same offence – in other words, that violence by women, however severe, tended to be assumed to be inherently ‘less serious’ than violence by men.


Some of Dr Lawrence’s statements in the press release cannot be verified, since they are essentially statements of opinion; others purport to be fact, which may be verified from the named sources. (Details on sources were requested from the Office of the Status of Women – ‘OSW’ – and where stated are shown below.) Like everyone, Dr Lawrence and her staff are entitled to their own opinions; but they do have specific responsibility to represent all peoples of Australia – rather than a self-selected lobby-group – and they are certainly not entitled to mispresent facts on issues that affect all Australians. Sufficient evidence is available to indicate that the press release continues a trend in OSW and elsewhere towards a consistent mispresentation of the problem of domestic violence, resulting in a serious misplacement of limited resources.

Central to this analysis in this report is the assumption that, to quote the Peoples Equality Network ‘Philosophy Statement’, “the needs, concerns, feelings and fears of men and of women are of exactly equal value and importance”. By comparison, central to Dr Lawrence’s press release is the assumption that ‘domestic violence’ is exclusively ‘violence against women’, and that the two terms may be used interchangeably. However, as Routley and her colleagues commented in the MUARC report, “narrowing the definition exclusively to women fails to address the whole extent of the problem [of domestic violence]. That is, it denies the reality of female-to-male violence and perpetuates system abuse of males who are abused by their partners.” The following analysis demonstrates that, by over-emphasising violence to women, and concealing and exacerbating the ‘system abuse of males who are abused by their partners’, Dr Lawrence and her department are failing – if not refusing – to tackle the overall problem of domestic violence. No-one should doubt the importance of facing the issues of men’s violence against women; yet no-one should doubt that those issues will be impossible to resolve without also fully facing the issues of women’s violence against men – and also against children.

Each of the key statements in the press release may be analysed as follows:

“Violence against women remains one of the most under-reported of all crimes in Australia,” the Minister Assisting the Prime Minister for the Status of Women, Dr Carmen Lawrence said.

This statement is arguably valid, but is unverifiable since by definition it refers to what is not known. What is known is that, by comparison with men, women are already greatly protected in our society. The Victorian Inpatient Minimum Database (VIMD) data, for example, indicate that in 1992/3, 1948 men and 423 women were admitted to hospital because of assault injuries – a sex-ratio of approximately 5:1 male:female. Under-reporting of domestic violence against males is likely to be far greater than for females: although males make up approximately 11 percent of those reporting explicit domestic violence against them, in the MUARC study they accounted for 17 percent of explicit domestic violence injuries, 37 percent of the explicit domestic-violence cases with open wounds such as knife-wounds, either 28 or 40 percent of hospital admissions, and approximately 65 percent of ‘very severe’ injuries. Male socialisation suggests that in any case men would be significantly less likely than women to go to hospital for treatment of injuries of any type or cause, and far less likely than women to describe their assault as having been caused by their partner.

“There are no national statistics on the nature and extent of violence against women but we do know that on any night in Australia, approximately 5,000 women and children seek accommodation in refuges, most of whom are escaping violence.

OSW stated that the ultimate source for this statement was the Australian Census; the statement is not, however, supported by the available Census data. The only category from which the statement could have been derived is the group “hostels for the homeless, night shelters, refuges”, in which 627 females 16 and under, 987 males 16 and under, 1271 females over 16, and 3722 males over 16 were present on Census night. The figure of “approximately 5,000 women and children” would appear in fact to have been based on the adult shelter-population (total: 4993): there is therefore a strong implication that almost three-quarters of the supposed ‘women and children were in fact homeless adult males. The statement is almost certainly an exaggeration by at least four times, and – given the overall context – is likely to be an exaggeration of ten times or more. This is not acceptable.

It should also be noted that this is not ‘on any night’ but on one specific predetermined night; and by definition would have included all staff and others present at shelters and refuges on that night. The expression ‘on any one night … women and children seek accommodation’ is also misleading, since some if not many refuges provide relatively long-term accommodation. Given that a stay in a refuge is often a precondition for priority treatment in provision of housing, false allegations of abuse, in order to secure priority housing via the refuge support-system, or punitive action by the Family Court against the former partner, have unfortunately been shown to be disturbingly common during relationship breakdown.

The arbitrary inclusion of children in this statement is reprehensible, especially as it is implied that children are exclusively victims of male violence: some shelter workers have reported that often their first task in dealing with nominally ‘battered’ women in refuges is to teach them to not assault their children. Although great emphasis has been placed on women’s risk of being murdered by their partners, in reality they are, in Australia, a low-risk group: the age-group most at risk of murder are infants below the age of one year, and the majority of their murderers are women (a mother’s risk of murdering her own child is approximately ten times higher than that of being murdered by her spouse or de-facto partner). It is well-known that women are the primary physical abusers of children, and commonly also sexual abusers, yet there are as yet no available refuges for men assisting their children to escape from female violence. This issue needs urgently to be addressed.

“We also know that domestic violence comprises one of the largest areas of police work – some 70 per cent according to New South Wales figures.

This statement cannot be said to be credible, as a simple reality-check indicates. We are asked to believe that male-on-female domestic violence accounts for more than twice as much police time as all road traffic incidents and offences, all traffic duty, all patrolling, all non-domestic violence and all other police activities combined. It is entirely credible that domestic violence might account for 70 percent of the work-load of a specific sub-section of a police force: but since non-domestic violence accounts for almost ten times as many injuries as domestic violence (according to the MUARC study), and road-traffic accidents alone account for more than ten times as many deaths as those caused by assaults of any kind against any victim, the claimed 70 percent figure for domestic violence must be considered doubtful in the extreme. It is most disturbing that no newspaper or other media reported on this glaring inconsistency in the press release.

In reality, this statement appears to have been derived from a simple mis-reading of the original source given by OSW, “Domestic Violence: Research Paper No.1”, a paper read at the ACT Community Law Reform conference in 1992, and published by the Australian Institute of Criminology in March 1993: a 70 percent figure does occur in this paper, but in an entirely different context, relating to other police matters. The actual figure given in the paper (for the ACT, not NSW as incorrectly stated) for ‘family incidents’ – of which only a small minority are actual domestic violence against women – was not 70 percent, but 3.5 percent. Since this was stated as a percentage of police patrol work – not police work overall – the 70 percent figure stated by Dr Lawrence represents an error of far more even than the immediately obvious exaggeration of at least twenty times. An error of this magnitude cannot and should not be ignored.

“Victorian Police report that they received more than 14,000 calls involving domestic violence in 1992.

OSW gave as their source an article in “The Age” [Melbourne], 3 June 1993. The ultimate source for that statement would appear to be the Victorian Police Crime Statistics report for either 1991/92, or 1992/93, rather than the 1993/94 figures cited in this report, which show the slightly smaller figure of 13485 incidents. These were not, however, ‘domestic violence’ in the implied exclusive sense of violence against women, but ‘family incidents’ of all types, as shown in the previous discussion on the data content of that report. If the ‘violence against the person definite’ figure of 1851 incidents is correlated with the MUARC statement that 56 percent of violence in the home is against males, and 72 percent in another’s home – suggesting an average percentage of 64 percent for male ‘family incident’ injuries – the resultant gender breakdown would suggest injury figures of approximately 1185 male and 666 female: this would include male-on-male and female-on-female violence, and non-spousal as well as spousal forms of violence.

Since non-physical abuse such as verbal abuse is at present not generally classed as criminal activity, the true police-report figure for significant ‘domestic violence against women’ would therefore seem to be more like 650-700 cases (or approximately 0.05 percent of the state population – roughly comparable with extrapolation of the MUARC ‘positive’ female domestic-violence figure) rather than the claimed figure of 14,000 cases. Since the latter figure is presented in the context of ‘domestic violence against women’, it represents a probable exaggeration of the true figure by some twenty times. This cannot be considered acceptable.

An additional point should be noted. Assuming that the gender breakdown of family-violence injuries – ‘offences against the person’, the most serious class of family incident – is roughly correct, 2433 males requested assistance from the police, and 1185 had been injured; 10797 females had requested assistance, and 666 had been injured. This indicates that approximately 6 percent of female calls involved serious offences, compared to 49 percent of male calls. At least 7439 calls for police assistance (or 55.2 percent) were placed because of verbal abuse, or for incidents which the police regarded as too trivial to specify; most incidents of this class would probably be considered an unnecessary and inappropriate misuse of police time and resources. Given what is known of male socialisation, it is unlikely that males would account for many of these calls: this implies that as much as 50 percent of police time on ‘family incidents’ – or more than 2 percent of all police patrol-work, and hence a significant cost to the nation – would be the result of women intentionally using police as pawns in a verbal argument with their partner or ex-partner. This in turn suggests that whilst violence – especially domestic violence – against males is seriously under-reported, violence against women is, paradoxically, both under-reported (in what is not known to police) and over-reported (in false or unnecessary calls for police assistance). Since false calls often arise from exaggerated fear, which itself often arises from inflated descriptions of danger, the exaggerated domestic-violence incidence figure in the press-release is likely to increase the resultant misuse of police time rather than reduce it: which cannot be considered acceptable.

“And in Queensland, a 1992 study showed that one in five women admitted to the Royal Brisbane Hospital’s emergency department was a victim of domestic violence.

The source given by OSW was a 1992 paper by Gwen Roberts and others at the Dept of Psychiatry, University of Queensland. Several studies by ‘G. Roberts and others’ – such as G. Roberts, “Domestic violence victims in hospital emergency departments”, 2nd National Conference on Violence, Australian Institute of Criminology, Canberra, June 1993 – were cited in the MUARC study: all were stated as being similarly based on the Stark and Flitcraft methodology, and appear to suffer from the same structural problems. The ‘one in five’ figure appears to be the result of the absence of additional cross-checks such as those applied by the MUARC team, and is therefore likely to be an exaggeration by a factor of four or more: the MUARC study indicated a a total likely prevalence of 5 percent or less. It should also be noted that this represents a percentage of injuries, not population: the MUARC study indicated a likely population-wide prevalence of around 0.05 percent, roughly consistent with the 0.7 percent estimate for ‘family violence’ overall (of all types, including sibling violence) determined in a recent study by the Australian Bureau of Statistics, but certainly not consistent with a 20 percent population-wide figure implied in the statement, or the ‘one-in-three’ figure (derived from a Canadian study which was described by the ABS as ‘seriously flawed’) used by the Office of the Status of Women in one of their national advertising campaigns.

The phrasing of the statement is also misleading: the figures given by Roberts and her colleagues related to those presenting at the Brisbane emergency department, but were not necessarily ‘admitted’ – in other words required or advised to remain – in hospital. The MUARC study implied that those women admitted to hospital for domestic-violence injuries represented less than 0.2 percent of women presenting to the emergency department – a figure one hundred times less than that implied by the phrasing in the press release.

Although figures vary, often wildly, a ‘one in five’ estimate for a woman’s lifetime risk of any kind of physical domestic assault (however minor) is fairly common, and has been reported in studies from Australia, Canada and elsewhere; figures for the USA tend to be somewhat higher. However, men’s lifetime risk for assault by their partner appears to be much the same as women’s: female and male figures for domestic violence, derived from systematic both-sex surveys rather than the more common female-only ‘studies’, are generally similar, with women – not men – usually self-reporting as slightly more violent. For example, Kantor and Straus (13th World Sociology Conference, 1994) gave US figures for male-on-female minor violence as 9.9 percent, female-on-male minor violence as 9.5 percent, male-on-female severe violence as 2.0 percent, and female-on-male severe violence as 4.6 percent – in other words men and women roughly equal for minor violence, but women more than twice as likely as men to be severely violent.

“These statistics are shameful and demand exposure.

The above analysis for all the preceding statements should illustrate the problem with this statement: ‘exposing’ – in this sense, presenting as ‘fact’ – statistics that are seriously misleading, and for the most part seriously incorrect, is unlikely to help resolve the real problems of domestic violence.

“By speaking out against violence against women we can help remove the cloud of fear and secrecy that prevents many women from seeking help.

Violence against women is a serious national issue that needs to be addressed. For certain groups, there is unquestionably “[a] cloud of fear and secrecy that prevents many women from seeking help”. Unlike domestic violence against men, this is, however, an issue which has already been significantly addressed; and the consistent refusal to address the reality of female violence contributes to that ‘cloud of fear and secrecy’. No-one could reasonably describe 17 percent – the proportion of explicit domestic-violence injuries identified as male in the MUARC study – as ‘insignificant’, especially as even larger proportions of those seriously injured, and admitted to hospital, were male; yet at present there are no state-supported refuges for physically abused men, and no ‘men’s resource centres’. With the exception of the Centres Against Sexual Assault, which are required by charter to accept male as well as female victims, there are hardly any state-funded support services for abused males; even those which do provide some support for males generally state explicitly in their literature – such as is available in hospital emergency departments – that perpetrators are either ‘usually’ or ‘always’ male, and very rarely acknowledge males as victims. Conversely, support services for women almost invariably portray them as victims, and rarely acknowledge women’s power and ability to act as agents of violence – even towards children, which is reprehensible since women are by far the majority of perpetrators of physical violence against children.

Female-on-female violence is a significant proportion of the total: an ABS study in Queensland in 1993 indicated that women were the perpetrators of at least 26 percent of reported violence against women. Lesbian activists in the shelter movement, for example, now acknowledge that domestic violence in lesbian relationships is a serious problem, certainly no less of a problem than in heterosexual relationships: Wendy Kusuma, executive director of La Casa De Las Madres, a women’s shelter in San Francisco, commented “we’ve worked so hard to gain credibility for our needs as victims, insisting again and again that this happens to us and we need help. Now we’re saying we’re also perpetrators, and we need help as perpetrators too.” Melbourne’s Domestic Violence and Incest Resource Centre runs at least one programme for victims of lesbian domestic violence, but does not as yet run any programmes for any women perpetrators of violence, whether lesbian or ‘straight’. Support services which women explicitly need are simply not available, apparently because to do so would threaten the stereotypes on which so much of current gender-politics are based.

In the same way, the scale, and even the fact, of domestic violence against males has been suppressed for decades as a matter of deliberate policy on the part of some groups which appear to have considerable influence in organisations such as the Office of the Status of Women. As the respected American researchers Murray Straus and Richard Gelles indicated in their book “Intimate Violence“, this is not so much a ‘conspiracy’ as the result of simple competition for limited resources: “We have spent the better part of our professional careers watching those concerned with child abuse compete against those who wanted to provide services for spouse abuse and vice versa. An example of this kind of competition is the radical feminist argument that there is no such thing as a battered husband . This concept flies in the face of logic and empirical data. Yet radical feminists believe that if we acknowledge the existence of battered husbands, then the funding designated for programs to assist battered women will be cut further because monies will be directed at programs for battered men. Thus, many radical feminists have fought for years to keep battered husbands closeted so that the small amount of money that was available for wife abuse would not be jeopardized.” Similarly, the American journalist Katherine Dunne commented that “deliberately ignoring a full 50 percent of the domestic violence problem guarantees that it will not be solved”. Attempting to ‘remove the cloud of fear and secrecy’ over one aspect of family violence whilst redoubling efforts to increase the clouds over another part of the same problem seems likely only to make the overall problems worse.

“Open debate will also enable us to accurately consider what progress we have made against violence and how much more we still have to do,” Dr Lawrence said.

Given the comments by Straus and Gelles and by Dunne above, and also given clear evidence of structural problems in much ‘research’ on domestic violence – such as the serious inconsistencies even in the relatively well-conducted MUARC study – it is clear that ‘open debate’ is not being encouraged. The whole area is dominated by a series of circularities, of which the most prevalent is the assumption that only women are injured, therefore only women should be studied, therefore only statistics on women should be published, which is taken to ‘prove’ that only women are injured. Open debate is indeed required, in order to “enable us to accurately consider what progress we have made against violence and how much more we still have to do”: but that debate must now be fully open, and it must fully include the reality of male victims and female perpetrators.

$3.8 million has been allocated to more than 50 community projects and two national education campaigns to foster a climate of opinion intolerant of violence against women.

Given the inaccuracies and seriously misleading statements elsewhere in the press-release, this is perhaps the most disturbing statement of all. Violence is a human problem, and arises from human confusions which affect everyone and anyone, male or female. There should never be an excuse for violence against anyone, female or male; yet it can only be resolved at an individual level, through understanding and the development of awareness at that individual level. Attempts to define violence solely in gender-stereotyped sociological terms – such as that expressed in the “National Strategy on Violence Against Women” document (National Committee on Violence Against Women, October 1992) – are inherently unusable for any practical application, and appear to serve only to increase ‘the cloud of fear and secrecy’ around the whole problem. Since the empowerment of both ‘perpetrators’ and ‘victims’ is required, in order for both parties to learn how to create non-violent responses in situations of conflict, ‘a climate of intolerance’ is extremely unlikely to be useful.

From the statements in the press-release, it would appear that $3.8 million is being allocated to projects whose sole effective result will be to reinforce the existing incorrect gender-stereotyped image of domestic violence as ‘a specific women-only problem’ (to quote Anne Sherry, former head of the Office of the Status of Women), to further confuse the issues, and to stifle the urgently-needed open debate on the issues. This cannot be considered constructive, or a wise or appropriate use of public funds.


Every key statement in the press release issued by Dr Carmen Lawrence’s department for ‘National Stop Violence Against Women Day’ would appear to be either an unverifiable assertion, or seriously misleading and/or seriously incorrect in fact. Most information in the press-release which purports to be fact – “statistics [which] are shameful and demand exposure” – is either distorted or exaggerated by several orders of magnitude: it does not in any way present a true or realistic picture either of violence against women or of domestic violence in general.

Significant amounts of public money appear to have been directed towards projects which will contribute little or nothing to tackling real issues which underly the problems of violence. Central to this is a continuation of a systematic refusal of the appropriate sections of government and non-government organisations to face the reality of women’s violence, especially violence against men.

Since most of the statements in the press-release were directly attributed to Dr Lawrence, it can be seen that she has seriously and actively misled the Australian public on the realities of domestic violence. Whether this was intentional, or careless ‘economies with the truth’ as a result of an excess of zeal and her active involvement in the Women’s Electoral Lobby, it is an unacceptable political error which requires immediate political action to correct. Dr Lawrence, having personally given her name to this campaign of misinformation, would appear to be personally responsible for correcting it.


  • accurate information on the full problem of domestic violence, not just against women but also against men and children, is urgently required;
  • the Office of Status of Women should be required to issue to all addressees of the original press-release a correction-notice for urgent publication, correcting the serious errors of fact and statistics in the press-release issued by Dr Lawrence on their behalf for National Stop Violence Against Women Day;
  • recipients of the $3.8 million funds announced in the press-release should be required. to update and correct all programmes and publicity campaigns in line with the correction-notice to be issued by the Office of the Status of Women – if necessary withholding funds from those recipients refusing to do so, and/or providing further funds where incorrect material has already been issued;
  • given that the misinformation in the press-release is a continuation of a known pattern of behaviour by the Office of the Status of Women and its staff, a full investigation of the activities of the Office – including a full audit of activities of recipients of funding from that Office – is urgently required;
  • Dr Carmen Lawrence, as Minister Assisting the Prime Minister for the Status of Women, should be held personally responsible by Parliament for the full public correction of the serious misinformation on domestic violence and other matters promoted in the press release and elsewhere by the Office of the Status of Women.

“The needs, concerns, feelings and fears of men and of women are of exactly equal value and importance.”

A report by People’s Equality Network, PO Box 777, Kew, VIC 3101. (c) PEN 1995.

Related pages