Douglas, Heather; Fitzgerald, Robin --- "Legal Processes and Gendered Violence: Cross-Applications for Domestic Violence Protection Orders" [2013] UNSWLawJl 3; (2013) 36(1) UNSW Law Journal 56

LEGAL PROCESSES AND GENDERED VIOLENCE:

CROSS-APPLICATIONS FOR DOMESTIC VIOLENCE PROTECTION ORDERS

HEATHER DOUGLAS^* AND ROBIN FITZGERALD^**

I INTRODUCTION

Official statistics consistently demonstrate the gendered nature of domestic violence (‘DV’). A recent report states that violence against women affected one in three Australian women and cost the economy around $13.6 billion in 2009 with women being most harmed.^[1] Over the past two decades, the legal response to DV has been increasingly focused on civil domestic violence protection order legislation in Australia, Canada, the United Kingdom and the United States.^[2] Domestic violence protection orders (‘DVPOs’) are now the most common legal remedy sought by, or on behalf of, women experiencing DV.^[3] In all Australian states^[4] a civil DVPO can be made by the lower courts to restrict and prohibit a perpetrator of DV (a respondent) from committing further acts of violence against a person (an aggrieved).^[5] While in the vast majority of these cases, applications are lodged by or on behalf of one partner (typically a female) against the other partner (typically a male),^[6] in a smaller proportion of cases both partners seek protection orders against each other. In some cases these ‘cross-applications’ will result in ‘cross-orders’, or mutual protection orders being made by the court resulting in a DVPO against both parties. In the event of a cross-order, there are conditions attached to each partner’s DVPO. In Queensland, all DVPOs will include a condition that the party be of good behaviour toward the aggrieved and individual DVPOs may also include other conditions, for example, a person may be prohibited from making contact with the aggrieved and from entering specified premises.

In Queensland, domestic violence support workers have claimed that the number of cross-applications (made by both private parties and police) and cross-orders has been steadily increasing in recent years.^[7] Recently the New South Wales and Australian Law Reform Commissions (‘ALRC report’) pointed to the potential misuse of cross-orders and the consequent need for some legislative reform.^[8] There is very limited research available about the use of cross-applications and resulting orders in any Australian jurisdiction.^[9] Wangmann has produced the only study focused on cross-applications.^[10] Based on a sample from New South Wales (‘NSW’) local courts, Wangmann’s study found that cross-applications are used by some men as a tactic to bring about mutual withdrawal and that in some cases cross-applications ‘must also be seen as a possible extension of the violence and abuse itself’.^[11]

Drawing on the findings from a recent study, conducted by the authors, of cross-applications and resulting orders in two Queensland magistrates courts, this article explores the context in which cross-applications are made and the typical court outcomes in such cases to consider why cross-applications are being increasingly made in Queensland courts.^[12] In this article we begin with a discussion of the influence of feminist jurisprudence on the development of the civil protection order system in Australia and elsewhere and an examination of how this literature has understood the ‘problem’ of cross-applications and cross-orders. We next propose three interlinked explanations for the rise of cross-applications by both private applicants and the police in Queensland. Based on a review of the literature, possible explanations include increased acceptance, in some circles, of research that finds there is gender symmetry in the way violence is perpetrated, developments in Australian family law, and the role conflict experienced by police in responding to domestic violence. The article then considers these explanations in light of the Queensland case study.

II DOMESTIC VIOLENCE PROTECTION ORDER SCHEMES AND THE USE OF CROSS-APPLICATIONS AND ORDERS

Given the modern focus on legal remedies as a response to domestic violence,^[13] it is difficult to imagine a time when law did not intervene in this context. However, as Goldfarb argues, it was not very long ago that law was, in any practical sense, excluded in matters of violence in the home because what happened there was a private affair and a woman’s identity was subsumed into that of her husband.^[14] In the 1970s and 1980s many feminist activists, fighting for the rights of women, engaged with the law in an effort to ‘end the marginalisation’ of domestic violence from both public and legal recognition.^[15] This activism led to a proliferation of rules and regulations in Australian states throughout the 1980s, including DVPO legislation.^[16] Recent feminist scholarship has questioned the role and relevance of criminal justice responses to DV, particularly with respect to the development of ‘no-drop policies’ and presumptive or mandatory arrest policies.^[17] Although these policies were sometimes promoted by feminist academics, critical discussion has emerged from the perceptions of those working in the DV field that such criminal justice responses may result in a loss of autonomy for victims of DV and other unintended consequences such as increased arrest of women.^[18] The problems with the criminal justice system’s engagement with domestic violence have, in practical terms, shifted attention to the civil response of DVPOs^[19] and yet the civil response has received much less scrutiny.^[20]

DVPOs were introduced as a legal mechanism that aimed to acknowledge the public nature of DV.^[21] DVPOs also held out the promise of protecting women from future harm by offering another option for women reluctant to prosecute their partners or unsatisfied with the backward-looking focus of criminal law responses.^[22] DVPOs also provided a number of advantages, regardless of whether criminal responses were endorsed and implemented. For instance, DVPOs can be obtained fairly cheaply^[23] and quickly, while criminal matters often take some months to finalise.^[24] The burden of proof is also lower for DVPOs precisely because they are ‘civil’ orders and generally the court must be ‘satisfied on the balance of probabilities’ that there has been DV or a threat of DV and it is likely that the violence will reoccur or the threat will be carried out.^[25] Conditions of orders can be tailored to the needs of the individual parties.^[26] Victims of DV can maintain their autonomy by helping to design the most appropriate conditions for their circumstances. As a result, conditions can be made flexible enough to allow the aggrieved to continue to live with the respondent.^[27] At the same time, breach of a DVPO condition can lead to criminal prosecution and some suggest that this mechanism discourages further violence.^[28] Finally, where it otherwise may not have occurred, DVPOs can provide the opportunity to bring victims of DV into contact with the legal system which can trigger other support, such as housing, financial advice and counselling.^[29]

However, arguments to the contrary have regularly been raised. For example, some argue that DVPOs might inflame an already tense situation,^[30] that they are often ignored by respondents as being ‘merely a piece of paper’ and ultimately that they do not have the same level of legitimacy as criminal prosecution and sentencing.^[31] Studies of the effectiveness of DVPOs in the United States generally indicate high levels of re-abuse even while a DVPO is in place.^[32] In Australia some have questioned whether DVPOs are appropriately enforced by police, pointing out that police failure to respond reduces the effectiveness of DVPOs as a measure of discouragement and may reduce a victim’s willingness to report incidents of abuse.^[33] Muller, Desmaris and Hamel observe that ‘[i]deally, restraining orders should be available to victims without them having to overcome unnecessary obstacles or putting themselves at risk of further harm, yet not be so freely granted that they can be manipulated by a vindictive partner.’^[34] However, research shows that, in some cases, DVPOs are used as a strategy on the part of the abuser to further abuse.^[35] Thus DVPOs may not in fact provide any real protection.

Referring specifically to heterosexual spousal relationships, Fitzpatrick has reported that the occurrence of cross-applications in Queensland is rising, both in absolute number and as a proportion of total DVPOs.^[36] While some cross-applications may be genuine, in that each party faces a threat of continued violence and both parties are equally in need of protection from each other, concerns have been raised that this may not be the case in a significant proportion of incidents.^[37] Wangmann has suggested that respondents to a DVPO application might make their own application (a cross-application) as a form of intimidation and as an extension of their abusive behaviour.^[38] Support workers in the domestic violence field have argued that reactive cross-applications may disproportionately affect female DV victims whose earlier claims for protection can be trivialised or even silenced.^[39] Cross-orders have also been criticised by others who suggest that they do not promote responsibility and accountability among offenders.^[40] Moreover, cross-orders can be difficult for police to enforce, which ultimately stands in the way of the aim of maintaining safety within families.^[41]

The ALRC report expressed concern about the use of cross-applications as a tactic or bargaining tool that might be used by men to bring about the result of mutual withdrawal of an application.^[42] It also found that cross-orders put both parties at risk of prosecution of a breach charge and therefore increased the risk of parties becoming enmeshed in the criminal justice system.^[43] This risk may provide a disincentive to the vulnerable party to alert police of any breach.^[44] DVPOs are often consensually made to avoid violent reactions from the perpetrator, expedite the process, and cooperate with busy lawyers and magistrates and, in some cases, police.^[45] Topliffe’s research suggests that victims often consent to cross-orders being made for these reasons.^[46] In 2006, the Victorian Law Reform Commission identified similar concerns about cross-applications and recommended that cross-orders should not be made by consent and that magistrates should be satisfied there are sufficient grounds for making cross-orders on the basis that both parties have committed domestic violence.^[47]

III THE INFLUENCE OF THE DEBATE OVER GENDER (A)SYMMETRY AND VIOLENCE

A large body of empirical research underscores the asymmetrical nature of intimate partner violence.^[48] Lloyd has observed that ‘overwhelmingly ... domestic and family violence is perpetrated by men against women’, and put simply, ‘the biggest risk factor for becoming a victim of ... domestic and family violence is being a woman’.^[49] In fact, a presumption of the gendered nature of domestic violence underlies most research on the topic, including the recent ALRC report on domestic and family violence.^[50] Overwhelmingly, it is women who are the victims of intimate partner homicide and serious assault.^[51] Nonetheless, a second body of research has questioned the asymmetrical nature of domestic violence and suggested that women are as violent as men.^[52] Dobash and Dobash have attempted to explain the puzzle of contradictory research findings related to gender symmetry in domestic violence.^[53] They suggest that the contradictory findings in the American context result from the different methods and analytical tools that researchers use in approaching the problem. Specifically, they point to a divergence between domestic violence researchers and violence-against-women researchers.^[54]

In trying to understand the divergence, Kimmel points out that many domestic violence researchers, who argue that there is gender symmetry in violence between heterosexual partners, generally apply the conflict tactics scale (‘CTS’), or other behaviour-based instruments, in large-scale surveys to assess the level of violence in a population.^[55] He explains that one of the problems with CTS is that it measures discrete acts of violence without taking into account the context of the violence, including how the violence is initiated.^[56] CTS is based in conflict theory and ranks acts of violence in a hierarchical fashion, and some argue that this leads to the devaluation of psychological abuse.^[57] CTS based studies are generally concerned with acts that took place in the past year^[58] and CTS assumes that accounts of violence given by parties are unbiased and reliable, a view that some researchers have questioned.^[59] Coates and Wade have shown how language can be used to conceal violence, mitigate responsibility and blame victims on the one hand, and to expose violence, clarify offender’s responsibility and honour victim’s resistance on the other hand.^[60] Kimmel argues that violence-against-women researchers, who find significant gender asymmetry in violence, are more likely to rely on victimisation studies. He explains that such studies tend to include questions relating to a wide range of assaults, including sexual assaults, and ask about the violence perpetrated by both current and former partners.^[61] Dobash and Dobash claim that these kinds of studies aim to obtain a deeper understanding of the context of violence.^[62]

In a 2004 study Dobash and Dobash interviewed 122 men and 134 women and compared their accounts of violence. They found differences between men and women and their use of violence in terms of the nature, frequency, underlying intention, intensity, level of physical injury and emotional impact.^[63] Dobash and Dobash found that women did not use intimidating or coercive forms of controlling behaviour and that men usually reported women’s violence as ‘inconsequential’.^[64] Hester’s recent research has also considered whether, and if so how, men and women use violence differently. In a recent study Hester tracked 96 men and women who police had identified as perpetrating a DV ‘incident’.^[65] Her research findings were similar to those of the 2004 Dobash and Dobash study. Hester found that, while cases were varied, there were distinct patterns by gender with significant differences between male and female perpetrators of DV. For example, she found that a greater number of incidents of violence were reportedly perpetrated by men than by women. Overall, men’s violence was more severe – a greater number of men were arrested – and was more likely to cause fear and control the victim. On the other hand, women were more likely to use weapons, but this was most often to protect themselves.^[66]

Similarly, in their study of 2090 police cases involving intimate partner violence, Buzawa and Hirschel found that in the context of the domestic sphere, men’s violence was more serious than women’s violence. They also found that men were more likely than women to have a criminal history, a prior record of violence, and be under the influence of alcohol or drugs.^[67] Moracco and colleagues examined the profile of male DVPO respondents in 731 cases in the United States and found that most male perpetrators had extensive criminal histories of violence.^[68] In light of their results, Moracco and colleagues suggested that when men’s DV came to the court’s attention as a result of a DVPO application, the alleged violence was likely to reflect chronic behaviour and not an isolated initial assault.^[69]

DVPO cross-applications provide another vantage point from which to consider the symmetry–asymmetry debate. Wangmann’s research shows differences between men’s and women’s complaints of violence among couples lodging cross-applications.^[70] In her study both men and women most commonly alleged assault, but, notably, allegations of sexual assault were virtually absent.^[71] The absence of sexual assault allegations is not surprising. Despite the fact that sexual assault by an intimate partner is frequently reported by women who are interviewed by researchers in a confidential setting,^[72] it is well known that there are very low reporting rates of sexual violence to officials in the public sphere.^[73] Wangmann’s study also found that women were more likely than men to allege attempted strangling and choking.^[74]

In the context of these debates, the work of Johnson and colleagues towards developing a typology of violence reframed the conversation, as it provided a framework that could reflect both gender symmetry and gender asymmetry.^[75] Johnson argued that it was not just the measuring tools that led to different outcomes – rather, that there are different types of violence being measured.^[76] Initially he identified two categories of abuse. Johnson called the first type ‘situational couple’ violence; this was ‘violence that enters a relationship when a disagreement that turns into an angry argument escalates into violence.’^[77] According to Johnson this type of violence can be mild or severe and is often an isolated incident. It is perpetrated by men and women, although violence carried out by men in this context is likely to be more severe and, in some cases, can be dangerous. Second, Johnson identified ‘intimate terrorism,’^[78] or coercive controlling violence, as the form of violence that most reflects the type of violence that many associate with DV.^[79] According to Johnson, coercive controlling violence is almost always carried out by men against women and, although it is much less common than situational couple violence, it is very dangerous. He observes that victims of coercive controlling violence are attacked more often, and the violence is less likely to stop.^[80]

In Johnson’s recent work with Kelly, three additional categories are identified.^[81] One of the additional categories is ‘separation-instigated violence’.^[82] This reflects a situation where violence is perpetrated, most frequently by the male, on the partner who is leaving, who is most frequently the female. This is usually a one-time situation and can be very dangerous.^[83] The second additional category, ‘violent resistance’,^[84] reflects a situation where a victim of intimate terrorism, usually a woman, fights back with violence, but is not seeking to control the other party. Lastly, a category called ‘mutual violent control’^[85] has been added. This reflects a situation where two parties use violence to control each other. Ansara and Hindin recently examined the application of Johnson’s typology to the general population in Canada and their research supports the gendered nature of the distribution. Ansara and Hindin show that, in their national sample, women experienced a greater range and much more serious types of victimisation than men.^[86] The results showed that although men and women were equally likely to experience the least severe types of physical aggression, ‘not embedded in a pattern of control’, only women experienced the most ‘severe and chronic pattern of violence and control involving high levels of fear and injury’.^[87] In fact, Ansara and Hindin found that a large proportion (19 per cent) of all women in the population with ex-partners reported that they experienced these most severe types of violence.^[88]

Stark has argued that the theory developed by Johnson resembles the child in the story of the emperor’s new clothes – a reality that was obvious as soon as he said it – and that it ‘crystallised observations’ being made since the 1980s that there are in fact different kinds of violence.^[89] However, a number of researchers have warned against the use of typologies in understanding DV on the basis that the typology framework is as yet under-developed, and inaccurate categorisation is potentially serious and risks minimisation of serious violence.^[90] Other commentators have questioned the value of Johnson’s typology specifically within the justice system. For example, Goodmark and Stark both raise concerns about the risks that police, judicial officers and lawyers will ‘mis-categorise’ DV leading to inappropriate and insufficient responses.^[91] Moreover, in Australia, Wangmann has voiced concerns that a ‘formalised typology’ suggests a level of scientific validity that has not necessarily been demonstrated, likely misses the nuances of intimate partner violence, and is at risk of too easily becoming a solution to manage justice system caseloads.^[92]

Despite these concerns there is evidence that since its publication, Johnson’s DV typology has become influential in Australian policy discussions.^[93] Recently the ALRC report discussed the role of typologies in some detail. While the report cautiously found that the typologies should not, at this stage, be reflected in legislation, it suggested that in some cases it may be useful for Family Courts to hear evidence about typologies of violence.^[94] In addition, a recent report by the Family Court of Australia makes explicit reference to Johnson’s work.^[95] In light of the above, it is interesting to note that recent statistics published by the NSW Standing Committee on Social Issues showed an increase in police proceedings against women for DV in NSW.^[96] From 2001 to 2010 the average yearly increase in the number of female DV offenders was 10 per cent compared to a 2 per cent increase for male DV offenders.^[97]

The question of whether these increases in police proceedings against women reflect ‘the true patterns’ of female offending or merely shifts in police policy and practice was ‘one of the more controversial aspects’ of the NSW Standing Committee’s inquiry into DV.^[98] Nonetheless, NSW inquiry participants raised concerns about supporting the application of Johnson’s typology within police practice in an environment where, increasingly, police are faced with very complex DV situations. Inquiry participants were concerned that police often do not have the ‘tools and training to sort out what is happening’ and may look for a ‘conceptual shorthand’ in order to make quick judgements about situational couple violence rather than investigating the context in more depth.^[99] In the context of the DVPO system the risk of using the typology framework is that coercive controlling violence may be miscategorised as situational couple violence by police and judicial officers, potentially leading to more cross-applications and ultimately more cross-orders being made.

IV FAMILY LAW

The rise in state-issued DVPOs in Queensland may be explained in part by the 2006 changes to the Family Law Act 1975 (Cth) (‘FLA’). In 2006, as part of a package of reforms to the FLA, section 51DA introduced a presumption of ‘equal shared parental responsibility’. The presumption does not apply where a parent has engaged in family violence.^[100] Alongside the presumption, section 60CC(3)(k) of the FLA stated that a judge, making a decision in a family law dispute about what order supports the best interests of the child, should consider any DVPO that involves ‘the child or a member of the child’s family’ including whether the order is final and whether the order was contested by a person.^[101] The rise in cross-applications from 2006 as a proportion of all DVPOs in Queensland may, in part, be attributable to the introduction of the presumption of the shared care provision in the FLA and the role of DV applications in rebutting that presumption. The 2006 amendment to the FLA made it clear that the existence of a DVPO was a relevant consideration in Family Court orders about the care of children. This may have encouraged those who were not actually fearful of DV to obtain a cross-order to neutralise the effect of any DVPO on the equal shared parental responsibility presumption and therefore on the child placement and contact decision.^[102]

Research in the United States has explored the manipulation of the use of DVPOs as ‘strategic ploys’ to gain advantage in the family court^[103] and similar concerns have been raised in Australian research.^[104] For example, in their recent study, Parkinson, Cashmore and Webster interviewed family lawyers in NSW about their experience with DVPOs.^[105] They found that there was a general belief among the lawyers they interviewed that in some situations DVPOs were used for tactical advantage in family court proceedings.^[106] According to their interviewees, perceived advantages included avoiding child contact arrangements, obtaining occupancy orders in relation to housing and various advantages in relation to immigration issues.^[107]

V POLICE AND ROLE CONFLICT IN RESPONSES TO DOMESTIC VIOLENCE

Police play a key role in responding to DV: they are often the first to attend a situation involving DV and may be crucial in referring parties to services and support.^[108] Importantly, it is usually police who advise parties about, and often apply on behalf of the aggrieved person for, DVPOs.^[109] In Queensland, similar to other jurisdictions, police can apply for a DVPO even where the aggrieved does not consent to the order being made,^[110] although it is not clear how often this occurs. In her study of cross-applications, Wangmann found that police frequently lodged cross-applications on behalf of women and men; however, this was particularly the case when the woman’s application was lodged first.^[111] Research has sometimes characterised police responses to DV as biased, inconsistent and inadequate,^[112] although there is also recognition that police alone cannot prevent DV.^[113] However, a failure to discharge the role properly can affect the way victims of DV engage with police in the future.^[114] Role conflict in the context of policing DV may explain some of the problems faced by police in this context.^[115]

Balenovich and colleagues observe that, unlike other crimes, DV assumes an interdisciplinary role for police: there is an expectation that police will both prosecute the offender and connect the victim to social services.^[116] They note that ‘in this new social context of proactive police policy towards DV, the conflict between officers’ crime control mentality and social service provider mentality is a major concern.’^[117] Nonetheless, research suggests that there is variation in police views about their role with respect to DV, and many police view DV matters as essentially a social service concern. In a recent American survey of 315 police officers, most agreed with the statement ‘domestic violence should be handled by treatment’.^[118] Typically Queensland police officers surveyed in 2005 for research conducted by the Crime and Misconduct Commission (‘the CMC study’) reflected this view; for example, one said ‘police are becoming arbiters of the family. We are often talking about normal family disagreements that suddenly need regulating by police. These are a social worker’s tasks not ours.’^[119] Commonly, police interviewed for the CMC study reported that DV is a difficult to handle ‘grey area’, or a matter that should be dealt with civilly by government departments, not the police.^[120] Similarly, a police officer writing about his experience of policing DV in the United States observed that DV is a ‘family problem’^[121] and that:

[w]hen a police officer arrives to investigate a domestic violence call, their first view of the family unit is one of dysfunction. Something has gone wrong beyond the participant’s ability to work problems out calmly. The police must face the conflict and attempt to prevent future violence from occurring.^[122]

The connection between policing roles and strategies and DVPO cross-applications has been raised by the National Council to Reduce Violence Against Women and their Children (‘NCRVWC’). Their report ‘Time for Action’,^[123] underlined the important role of police in addressing DV, but also identified weaknesses in the police approach that, they argued, contributed to the increase in dual arrests (where both parties are arrested). The NCRVWC reported that proper evidence-gathering processes undertaken by police^[124] and better methods of police accountability could help to eliminate dual arrests.^[125] Such approaches may also reduce police cross-applications. The NCRVWC report also suggested that criminal charges, where relevant, should be laid by the police, and that cross-orders should be avoided.^[126]

Some evidence suggests that the use of cross-applications by the police may result from an institutional tendency for police to apply a criminal law approach which focuses on discrete incidents – in this way the approach may be similar to that taken by DV researchers.^[127] This approach may lead to a more gender-neutral approach opening the door to the greater identification of situational couple violence – that is, gender symmetry – as a description when officers handle DV incidents. For example, various brochures available via the Queensland Police Service website^[128] explain that domestic and family violence ‘occurs when one person in a relationship uses violent or abusive behaviour to control another.’^[129] It provides an inclusive list of types of behaviours and acts that might form DV, which include: physical abuse, damage to property, sexual abuse, verbal abuse, harassment or intimidation, financial abuse, or threatening any of these. The definition is focused on controlling violence but makes no statement about the over-representation of women as victims of DV. Presumably police are being trained to recognise one-sided coercive controlling violence as the type of DV situation where one of the parties will need protection. However if, as the studies quoted above suggest, police regularly interpret DV as family dysfunction, more akin to situational couple violence, there is a risk that dangerous coercive controlling violence is missed and approaches like cross-applications may be deemed appropriate by police.

DV workers in the field have also suggested that police often support both parties in applying for protection orders (ie, cross-applications) as a simple and quick way to finalise their involvement in complex DV cases.^[130] Similar reasons to those associated with police cross-applications have been identified in studies of DV situations where dual arrest occurs.^[131] According to Wangmann, police may justify a cross-application on the basis that the court is better placed to work it out, alternatively Wangmann suggests that sometimes cross-applications may reflect sloppy police work.^[132] Other alternative approaches may require a much more complex assessment of the situation including whether there is a history of violence perpetrated by one party against the other, the nature of the injuries sustained by both parties, the likelihood of violence in the future, and whether one person was acting in self-defence. While dual arrest is a concern, some studies suggest that in practice it is not particularly common,^[133] especially where children are involved.^[134] Evidence suggests this may in part be because of the timing of many DV-related police interventions, which are often late at night when it may be difficult for police to find a carer for the children.^[135] Similarly it may be difficult to arrange alternative accommodation for the parties and police may determine that a cross-application is the most appropriate approach. Police are increasingly involved in applications for cross-orders, whether they have applied on behalf of one or both parties. These do not necessarily require either party to be removed and may be perceived by police as an expedient way of dealing with a complex issue. Further, evidence indicates that police may not see any real problem with cross-applications or cross-orders, on the basis that the victim is still protected by the order.^[136]

VI THE QUEENSLAND CASE STUDY

OF CROSS-APPLICATIONS

Limited data is collected by the Queensland Department of Justice and Attorney General (‘JAG’). However, JAG was able to provide figures about numbers of cross-orders made in Queensland from 2004–2011. Table 1 and Figure 1 illustrate that in the year following the implementation of FLA reforms, there was a roughly 33 per cent increase in the proportion of cross-applications – from a generally stable rate of about 12 per cent from 2004–05 to 2006–07, to a rate of about 16 per cent beginning in 2007–08 and continuing until 2010–11.

Table 1: Number and Percentage of Cross-applications, Queensland, 2004–05 to 2010–11

Figure 1: Cross-applications as a Percentage of Total Originating Applications Lodged, Queensland, 2002–03 to 2010–11

The results of a study of cross-applications lodged in Queensland magistrates courts are presented in this section of the article. The study was undertaken with the aim of filling a clear gap with respect to information about cross-applications in the state. Information was collected from 328 pairs of magistrates court files (656 files in total) representing a census of all cross-orders involving current or former heterosexual intimate partners lodged in two Queensland Magistrates Courts in the financial years 2008–09 and 2009–10. Respectively, the two selected Courts – Beenleigh and Brisbane – have the second and third largest case volumes related to DVPO applications in the state.^[137] Data were drawn directly from the Protection Order Application (‘DV1’),^[138] the form required to lodge a DVPO in Queensland, and from additional details of the incident(s) and allegation(s) contained in the court file. As is the case with any analysis of administrative data sources, the results presented here should be read as an assessment of data captured for the legal process, rather than, necessarily, the actual and nuanced nature of DV incidents. Nonetheless, with few exceptions the broad range of elements available from the DV1 form were completed by the aggrieved or a representative on behalf of the aggrieved person. In the following section a descriptive analysis of the nature of cross-applications in the Queensland sample is presented.^[139]

A Characteristics of the Aggrieved

Over the two-year period of this study, cross-applications among heterosexual couples accounted for 15 per cent (n = 392) of all DVPO applications in Brisbane (n = 2562) and about 9 per cent (n = 264) of applications in Beenleigh (n = 3103). All couples were in heterosexual intimate partner relationships – defined for the purposes of this study as current or former partners including spouses, boyfriends or girlfriends.^[140] A majority of couples (90 per cent, n = 295) categorised their relationship as ‘spousal’– including current and former married and de facto relationships – and the remaining 10 per cent of couples (n = 33) categorised their relationship as an ‘intimate personal’ – including current and former engaged, betrothed or dating partnerships. Information about the total number of aggrieved men and women with children is not available through the files; however, over one quarter (27 per cent) of aggrieved women and one fifth (21 per cent) of aggrieved men named children as victims of DV on their application.

B Allegations and Previous Orders

Allegations were extracted from the narratives provided by the applicants. A list of allegations was prepared based on the forms of violence identified in section 11 of the Domestic and Family Violence Protection Act 1989 (Qld), and on commonly occurring allegations described in the narratives – for example, attempted strangulation. We examined the frequencies of these allegations within each gender, and also tested for statistical differences between genders using chi-square. Differences between men and women discussed in the text are statistically significant at the p < 0.05 level, however all comparisons are presented in Tables.

Table 2 indicates that applications lodged by aggrieved women and men show some similarities in patterns of allegations. For example, assault, verbal harassment and damage to property, respectively, were the most common allegations made by both women and men. Nonetheless, significantly greater proportions of women than men made each of these allegations. Higher proportions of women than men also made allegations of intimidation, threats and stalking. Thirteen per cent of women alleged that they were the victim of attempted murder (compared to 1 per cent of men) – and the vast majority (90 per cent) of these women alleged that the method of attempted murder was strangulation, a finding that is consistent with Wangmann’s findings in NSW.^[141] Only women made allegations of sexual assault victimisation (2 per cent, n = 7), and the relative infrequency of this allegation by women is also consistent with other research.^[142]

Roughly equal proportions of aggrieved men (7.3 per cent, n = 24) and aggrieved women (6.7 per cent, n = 22) reported that their partner had access to a weapon (Table 2).^{[143 ]}Aggrieved men were proportionately more likely to make allegations of weapons use (as opposed to mere access) than were aggrieved women. Six per cent (n = 21) of men alleged some type of weapons use by their partner compared to 3 per cent (n = 10) of aggrieved women. It was not possible to discern the specific type of weapon used with the available data.

Applicants also provided information about any other pending, current or former court orders involving themselves and the respondent. For 40 per cent of couples (n = 131) some type of current or former order was reported. In most cases this was a prior Queensland DVO (35 per cent, n = 116), followed by Children’s Court orders (17 per cent, n = 54), Family Court orders (16 per cent, n = 53), and other Interstate DV orders, including from New Zealand (15 per cent, n = 48). Table 2 shows that reports of alleged breaches of previous DVPOs occurred relatively infrequently. However, a significantly greater proportion of women (3 per cent, n = 10) than men (1.5 per cent, n = 5) reported an alleged breach.

Table 2: Distribution of Allegations Made by Aggrieved Women and Men in Connection to the DVPO Application

C Who Lodges the Application?

As is the case in other jurisdictions, in Queensland aggrieved parties are able to lodge DVPO applications privately, on their own behalf,^[144] or have the application lodged for them by the police.^{[145 ]}Police were involved in the largest proportion of cross-applications at the Beenleigh and Brisbane Magistrates Courts, lodging the application on behalf of one or both aggrieved partners for 80 per cent of cross-application couples (n = 261 couples). When police were involved, for the vast majority of couples (80 per cent, n = 210), police lodged the DVPO application on behalf of both partners, rather than the male aggrieved alone (7 per cent, n = 19 couples) or the female partner alone (12 per cent, n = 32 couples). Across all cross-applications in the study, both partners lodged the DVPO privately in only 20 per cent (n = 67 couples) of cases.

Table 3: Differences Between DVPO Applicant-types, by Allegation and Gender of the Aggrieved

Findings in Table 3 show a consistent pattern of allegations when police are involved either on behalf of one or both partners – specifically, assault followed by damage and verbal harassment predominate. In contrast, when both partners lodge privately the pattern is significantly different with proportionately more allegations of intimidation, threats, stalking and sexual offences particularly on the part of the female aggrieved.

D Outcomes of Cross-Application Orders

There are a number of possible outcomes for DVPO cross-applications. In many cases orders are made by consent, which occurs when both parties indicate to the court that the requested order should be made.^[146] In other cases orders can be made after a hearing: in this context the magistrate will most often hear evidence. Alternatively, parties can indicate to the court that they wish to withdraw their application, which results in no order being made. Finally, the court can dismiss an application; this usually occurs when the aggrieved does not attend court or, very occasionally, where the magistrate finds there is insufficient evidence to support an order being made. Table 4 (see over) shows that a majority of aggrieved women and men obtained orders, rather than having their orders dismissed or withdrawn. However, there were statistically significant gender differences in the distribution of these outcomes. Greater proportions of women than men had orders made by consent (42 per cent and 38 per cent, respectively) and by the court (40 per cent and 37 per cent, respectively). In contrast, greater proportions of men than women had orders that were dismissed (16 per cent and 12 per cent, respectively) or withdrawn (9 per cent and 7 per cent, respectively).

We also examined whether partners in a couple tended to have the same or different outcomes. Overall, our results showed that a majority (73 per cent) of couples had the same outcome (these results not shown in a Table). In an additional 12 per cent of cases the outcome was the same, but arrived at through different means – eg, dismissed versus withdrawal, or consent versus order by the court. The outcomes were different for men and women for only 15 per cent of couples – ie, an order was granted to one partner but not the other. In most (74 per cent) of these cases (n = 37 couples) the woman was granted an order while the man’s application was dismissed or withdrawn.

Table 4: Outcome of the DVPO for Aggrieved Women and Men¹

In this study, police involvement in the application increased the chances that an order would be made either by consent or after a hearing. Table 5 shows that for over 92 per cent of couples (n = 192) applications made by police on behalf of both partners resulted in a DVPO being made for both parties by consent or after a hearing. Where police lodged applications for the aggrieved woman only, the greatest proportion of outcomes resulted in the order being made for the woman (59 per cent) but not the man followed by the order being made for both partners (31 per cent). As mentioned previously, police lodged on behalf of only one partner in relatively few cross-applications in the study. Where the police applied for the man only, over one-half (56 per cent) resulted in orders being made for both partners, followed by orders being made for the man, but not the woman (28 per cent). Finally, where both partners applied privately, the most common outcome was for the application to be dismissed or withdrawn for both partners (40 per cent).

Table 5: Distribution of Order Outcomes by Lodger of the Application¹

VII DISCUSSION

Queensland Magistrates Court data show a sharp rise in cross-applications as a proportion of all DVPOs since 2006. As previously noted there was a roughly 33 per cent increase from a generally stable rate of about 12 per cent from 2004–05 to 2006–07, to a rate of about 16 per cent beginning in 2007–08 and continuing until 2010–11 (Table 1/Figure 1). The increase in the proportion of cross-applications is also demonstrated when Wangmann’s 2002–03 rates (5–11 per cent) from a study of three Sydney courts are compared to the 2008–09 and 2009–10 Brisbane and Beenleigh rates found in this study (9–15 per cent).^[147] While great caution must be exercised in making the comparison and jurisdictional differences may account, at least partially, for the NSW–Queensland divergence, the observed increase coincides with the 2006 changes in the FLA, and may also coincide with an emergence of typology-like thinking that may have begun to influence responses to DV. In addition to the marked increase in the rate of cross-applications as a proportion of all DVPOs, it is possible to discern particular features of the incidents and parties involved in these cases.

A major finding in this study was that police were extensively involved in cross-applications for DVPOs. Police were involved in lodging an application for a DVPO on behalf of at least one partner in 80 per cent (n = 261) of couples and for 64 per cent (n = 210) of couples police were involved in lodging an application for both partners.^[148] This has implications for whether a fair hearing of both applications can be ensured where both parties are, in effect, represented by the same advocate.

Police involvement in the application also considerably increased the chances that the application would be successful rather than dismissed or withdrawn. Where police applied on behalf of both parties, the order was made for both for the vast majority (92 per cent) of couples (n = 192). The rate of success was similarly high for parties whose applications were lodged by the police when their partners’ applications were lodged privately. However, orders were made only about 27 per cent of the time when both parties lodged privately. It is likely that police applications are less likely to be withdrawn, at least in part, because applicants feel supported in the process and are more willing to continue with their application. Research about aggrieved women and DVPOs has identified a number of reasons why they may not persist with the process to obtain a final DVPO. One of the most common reasons cited is fear of retaliation.^[149] Such fears may be alleviated if police make the application on behalf of the aggrieved.

Some research suggests that withdrawal or dismissal of DVPO applications may occur because the applicant does not attend court as a result of perceived positive changes in the respondent’s behaviour.^[150] However, a study on help-seeking by victims of DV found that the aggrieved woman’s perception of threat and her attachment to the respondent was the best predictor of whether she would persist with the application procedure – that is, the study found that the higher the threat and the closer the attachment, the less likely she was to complete the application process.^[151] Such studies suggest that many applications are withdrawn not because danger has subsided but that the aggrieved perceives increased risk in continuing with the application. This help-seeking study also found that the main reason aggrieved women’s applications for DVPOs were not pursued related to action or inaction by the aggrieved, for example through failing to attend court or withdrawing the application, and not through a court’s assessment that there was insufficient merit to grant the DVPO.^[152] While the Queensland study results do not provide an explanation for withdrawal or dismissal of applications, it is possible that a number of the applications where there was no police involvement were withdrawn or dismissed because aggrieved applicants feared retaliation rather than necessarily suggesting a lack of merit.

Certainly, the results of this study suggest that, in the context of cross-applications, the chances of obtaining a DVPO are significantly increased where there is police support. While it may be that police apply on behalf of either one or both the female and male aggrieved in only the most complex and dangerous cases, it is difficult to assess the context of police selection when looking at the allegations recorded in our administrative data.^[153] An important concern is that most often both parties are not equally at risk. When police support both parties’ applications for a DVPO, the usefulness of the order to an at-risk party is neutralised since the police assumption is that both parties are equally at fault (as well as equally at risk).^[154] This issue usually arises at the application stage but may also arise later when police are called to investigate a breach. If there is a cross-order in place a breach action may be less likely or dual arrest may be more likely. Further, a party with a cross-order may not report a breach because of a concern that she too may be charged.

There were some similarities in the allegations made by women and men against each other in this study. For example, across all couples, assault was the most common allegation. For 61 per cent (n = 199) of couples both partners made an allegation of assault against each other. Importantly, however, these data do not indicate the relative seriousness of alleged assaults made by aggrieved parties. Damage and verbal harassment allegations were the second most common category for both genders, and were made by roughly one-quarter of all women and men against each other.

Beyond these similarities, the Queensland results were also consistent with Wangmann’s NSW results in showing that more women than men alleged the most serious offences. For example, many more women (12 per cent, n = 42) than men (1 per cent, n = 4) alleged attempted murder, and a large majority of these cases were by strangulation. This is an important concern. Block and colleagues’ research, conducted in the United States, found that in 68 per cent of cases where a woman alleged attempted strangulation by her intimate partner the incident was followed quickly with a severe incident of DV.^[155] Allegations of attempted strangulation or choking are not uncommon in DV cases, and they are usually alleged to be perpetrated by men against women – this is considered in risk assessment tools used by DV workers as a ‘red flag’ for future serious abuse.^[156] In Queensland police documents, strangulation is also identified as a risk factor for DV.^[157] Intimidation was much more likely to be alleged by women (14 per cent, n = 45) than men (6 per cent, n= 19). In his research Stark notes that intimidation is a type of coercive and controlling violence. He employs the term ‘intimidation’ to encapsulate a range of ‘tactics that supplement violence’ and are used to induce fear and humiliation.^[158] Threats may also be thought of as a form of intimidation, and thus, as a form of coercive and controlling violence.^[159]

It is notable that in our study female aggrieved (13 per cent, n = 41) were also more likely than male aggrieved (9 per cent, n = 31) to allege threats. However, despite the connections made in the literature between attempted strangulation and the high risk of future injury, and between coercive controlling violence, intimidation and threats, in this study allegations of attempted murder by strangulation, intimidation, or threats generally did not make it more likely that police would apply for DVPO on behalf of the aggrieved person. While the involvement of police in DVPO applications is extremely important, it may be that in deciding to apply for a DVPO police are often not sufficiently assessing coercive and controlling behaviour in the alleged violence, and therefore whether there might be one party who is more in need of protection than the other.

While female and male aggrieved equally alleged that the respondent had access to weapons, it is notable that male aggrieved (6 per cent, n = 21) were twice as likely to allege that their female partner had actually used a weapon, as compared with the female aggrieved (3 per cent, n = 19). These findings are consistent with previous research. Some have argued that women’s disproportionate resort to using weapons in DV matters may be seen as an attempt at ‘levelling the playing field’.^[160] Hester’s study found that women often resorted to the use of weapons to protect themselves from abuse.^[161] She found that women who used a weapon did so in cases where men were also identified as a perpetrator, while when men were alleged to have used a weapon they were more often recorded as the sole perpetrator.^[162]

VIII WILL RECENT LEGAL REFORM AFFECT CURRENT PRACTICES AROUND CROSS-ORDERS?

In response to concerns about the way DV was being ‘managed’ under the family law system and particularly in response to concerns about the way the provisions introduced into the FLA in 2006 were being interpreted and implemented,^[163] new provisions of the FLA commenced operation in June 2012 and are relevant to this research. While the presumption of equal shared parental responsibility remains in the FLA,^[164] section 60CC(3)(k) FLA has been re-examined and now allows ‘inferences’ to be drawn from the existence of a DVPO, taking account of the nature of the order, the circumstances in which the order was made, any evidence admitted in proceedings for the order, any findings made by the court or in proceedings for the order, and any other relevant matter. Clearly these changes do not reduce the focus on DVPOs in Family Court proceedings – indeed they may increase the importance of DVPO applications.

The 2012 changes to the FLA may encourage parties to present as evidence the primary applications for DVPOs and the transcripts of hearings associated with them as these may provide relevant information to family law decision-makers considering DV in child proceedings. It will arguably be as important as ever under the new provisions for parties to make a reactive cross-application for a DVPO in an attempt to neutralise or rebut the primary application for a DVPO and where possible to have the police applying for a DVPO on one’s behalf to maximise the chances of success. While orders made in response to an application for DVPO will be important, other results such as a withdrawal or dismissal of an application will require particular consideration, especially where a police officer has not applied on behalf of the aggrieved. Our findings show that applications for DVPOs are much more likely to be withdrawn or dismissed where police have not applied on behalf of the aggrieved. Although this is true for both men and women, the concern is that such a ‘result’ may reflect something other than a cessation of violence: it may actually be a symptom of increased danger and possibly protective behaviour on the part of the aggrieved. Bagshaw and colleagues have noted that a disengagement or failure to engage with services does not necessarily suggest there is no violence.^[165]

Other changes introduced into the FLA include a change to the definition of family violence. It is now defined as ‘violent, threatening or other behaviour by a person that coerces or controls a member of the person’s family ... or causes the family member to be fearful’.^[166] Certainly consideration of this definition may influence the inferences drawn by Family Court decision-makers about specific DVPOs, for example cross-orders may be interpreted to suggest situational couple violence. While this new definition should encourage Family Court decision-makers to carefully examine the type and context of the violence alleged in DVPO applications, there is also a risk that we will see more DVPO applications that consciously speak in the language of coercive control (‘fear’, ‘threat’, ‘intimidation’, ‘strangulation’) as a tactic to obtain desired results from the Family Court or as a way to neutralise the other party’s application in the Family Court. Alternatively it is possible that this will have a beneficial effect, ensuring that more than physical single incidents are taken into account.

In response to recent government reports,^{[167 ]}the Domestic and Family Violence Protection Act 2012 (Qld) (‘new DV Act’) has been dramatically revised and new provisions came into effect in Queensland in September 2012. Some of the changes to the legislation were introduced precisely to reduce the use of cross-orders.^[168] The Explanatory Notes identified the disproportionate use of cross-orders and cross-applications and that this

is inconsistent with the notion that domestic violence is characterised by one person being subjected to an ongoing pattern of abuse by another person who is motivated by the desire to dominate and control them. Both people in a relationship can not be a victim and perpetrator of this type of violence at the same time.^[169]

Police now have power to issue a Police Protection Notice^[170] which will operate as a short-term response to ‘low to medium level violence’.^[171] In effect it will operate as a short-term protection order and a breach could result in the respondent being charged with an offence making them liable to a maximum of two years imprisonment.^[172] Police Protection Notices also operate as an application for a DVPO that is filed with the Magistrates Court for its consideration at a later date.^[173] Cross-notices are not permitted,^[174] but presumably police could still apply for DVPOs (rather than Police Protection Notices) on behalf of both partners. As a consequence of the reforms if this occurred it would be up to the magistrate to decide whether he or she should hear both applications or should refer one of the applications to another court for hearing by a different magistrate.^[175] Should the magistrate decide to refer one of the applications for hearing to another court, both of the parties could be required to attend at both hearings; in some cases this could potentially double the victimisation experienced by an aggrieved person. Further, in deciding whether to make an order the magistrate must have regard to the principles in section 4 of the new DV Act;^[176] in the context of cross-applications, section 4(2)(d) is particularly relevant – it states:

in circumstances in which there are conflicting allegations of domestic violence or indications that both persons in a relationship are committing acts of violence, including for their self-protection, the person who is most in need of protection should be identified.

This provision effectively recommends that in most cases magistrates will need to make a choice about which applicant in a cross-application situation should obtain an order. Presumably, however a significant proportion of cross-applications will continue to result in orders being made by consent.^[177]

An expanded definition of DV has been included in the new DV Act. The definition includes physical, economic and emotional abuse and it also includes behaviour that is ‘coercive’, or behaviour that ‘in any other way controls or dominates the ... person and causes the ... person to fear for [their] safety or wellbeing’.^[178] A consideration of the alleged violence will, as in family law decisions, be important in deciding which party is most in need of protection when cross-applications are made. The ALRC Report recommended further education and training for police on this issue.^[179] The risk is that there is potential for misinterpreting dangerous behaviours, leaving an aggrieved unprotected, and decision-makers need to be wary of the co-opting of the language and allegations associated with coercive and controlling behaviours by those who are not actually fearful or in need of protection.

IX CONCLUSION

The rising use of cross-applications in Queensland, together with initial descriptive findings from the Queensland case study, underline the difficulties of using the law to protect vulnerable people from DV. From the 1970s the feminist movement attempted to identify the special characteristics of DV, specifically that it employed coercive controlling behaviours.^[180] Johnson has taken this concept and incorporated it into a complex model of typologies of violence and in recent times his analysis has been considered by law reform agencies and in legislative responses as a way to understand the complexities of and differences in violence in couple relationships.^[181] While coercive controlling violence has been at the heart of DV legislative reforms since they were introduced in Australia throughout the 1980s,^[182] the legislative responses to DV have become increasingly complex, multi-layered and contingent. Cross-applications and cross-orders are a symptom of this complexity – a cross-order means much more than that the parties simply have a protective order. It has implications for the residence of children, engagement with the criminal justice system and most importantly victim safety. As it has been noted elsewhere, more than statute reform is required to perfect the legal response to DV as the impact of legal change is dependent on wider social and cultural contexts.^[183] Current reforms are unlikely to change the behaviours and decisions of police officers and judicial decision-makers in the short-term and they will continue to be required to consider the complex and contingent contexts of the lives of the parties before them, as noted by Wangmann:

This call for ‘context’ is not new but it is challenging to work out how this might be done, particularly in practice settings where a professional may have only one encounter with a victim or a perpetrator. Here, the skill of the person making the assessment, the skill at listening, probing further, building trust and rapport, have a significant influence on the nature and quality of information that might be revealed in that encounter and, therefore, the assessments that might be made.^[184]

To be sure, this study highlights the need for closer scrutiny of the nature and extent of cross-applications. Our results were limited by a lack of comparative qualitative information which would have provided context to the descriptive results. Wangmann highlights the need for using both qualitative and quantitative approaches to more fully understand both the nature of DV and the subsequent criminal justice processes resulting from DVPOs.^[185] This initial descriptive study was limited to the quantitative data available from the Queensland DV1 application form. While providing some scope to understand the distribution and type of cross-applications, this study shows that the numbers of cross-applications and orders is rising in Queensland and underlines the extensive involvement of police in this increase. The study also shows that while both men and women commonly allege assault, damage to property and verbal harassment, women do tend to allege particularly serious matters, including attempted strangulation, much more regularly. This study also provides important baseline information for assessing the impact of the Domestic and Family Violence Act 2012 (Qld); however, further qualitative research needs to examine the context of incidents leading to cross-applications.

* Professor, T C Beirne School of Law, The University of Queensland.

** PhD (Carleton), Lecturer, School of Social Science, The University of Queensland. The authors wish to thank Danielle le Poidevin, Jim McNicol and Abraham O’Neill for research assistance and Heather Nancarrow and Zoe Rathus and the anonymous reviewers for helpful comments on earlier drafts. This research was assisted by two small grants from the T C Beirne School of Law and The University of Queensland.

[1] The National Council to Reduce Violence against Women and their Children, The Cost of Violence Against Women and Their Children (Commonwealth of Australia, 2009) 4.

[2] Heather Douglas and Lee Godden, ‘The Decriminalisation of Domestic Violence: Examining the Interaction between the Criminal Law and Domestic Violence’ (2003) 27 Criminal Law Journal 32; T K Logan et al, The Kentucky Civil Protective Order Study: A Rural and Urban Multiple Perspective Study of Protective Order Violation Consequences, Responses, & Costs (University of Kentucky, 2009).

[3] Sally F Goldfarb, ‘Reconceiving Civil Protection Orders for Domestic Violence: Can Law Help End the Abuse without Ending the Relationship?’ (2008) 29 Cardozo Law Review 1487, 1489, 1504. In Queensland in 2010–2011, 22 346 applications were issued: Magistrates Court of Queensland, ‘Annual Report 2010–2011’ (Queensland Courts, Brisbane, 2011) 25; in New South Wales in 2010, 22 691 final domestic violence orders (‘DVOs’) were made: Local Court of New South Wales, ‘Annual Review 2010’ (Local Court of New South Wales, Sydney, 2010) 17; in Victoria in 2010–2011, 28 141 DVOs were finalised: Magistrates’ Court of Victoria, ‘Annual Report’ (Magistrates’ Court of Victoria, Melbourne, 2011) 115. In the US, the domestic violence order is similarly the most common legal intervention in DV cases with over one million applications being issued every year: Kathryn E Moracco et al, ‘Who Are the Defendants in Domestic Violence Protection Orders Cases?’ (2010) 16 Violence Against Women 1201, 1203.

[4] See generally Karen Wilcox, ‘Recent Innovations in Australian Protection Order Law – A Comparative Discussion: Topic Paper 19’ (Australian Domestic and Family Violence Clearinghouse, Sydney, 2010); see also Domestic Violence (Family Protection) Act 1989 (Qld); Domestic Violence (Family Protection) Act 2012 (Qld); Domestic Violence and Protection Orders Act 2008 (ACT); Crimes (Domestic and Personal Violence) Act 2006 (NSW); Domestic Violence Act 2007 (NT); Domestic Violence Act 1994 (SA); Family Violence Act 2004 (Tas); Crimes (Family Violence) Act 1987 (Vic); Restraining Orders Act 1997 (WA). See also Alexandra Harland et al, Family Law Principles (Thomson Reuters, 2011): ch 10 contains a very good overview of Australian protection order schemes.

[5] ‘Respondent’ and ‘aggrieved’ are the terms used in the Domestic and Family Violence Protection Act 1989 (Qld) s 21 and in the Domestic and Family Violence Act 2012 (Qld) s 21.

[6] The majority of applicants for DVPOs in Queensland are women in heterosexual relations: see Crime and Misconduct Commission, ‘Policing Domestic Violence in Queensland: Meeting the Challenges’ (2005) 36. However, the previous Domestic Violence (Family Protection) Act 1989 (Qld) s 11A and current Domestic Violence (Family Protection) Act 2012 (Qld) s 13 allow for people in other kinds of relationships to apply for a DVPO – for example, family members and informal carers – so it is difficult to identify clear figures: see Australian Law Reform Commission and New South Wales Law Reform Commission, Family Violence – A National Legal Response: Final Report, Report No 114 (2010) 162.

[7] Caroline Fitzpatrick, ‘Cross-applications, No Applications and Applications Used to Control’ (Paper presented at Court Network Conference, Brisbane, 21 April 2010).

[8] Australian Law Reform Commission and New South Wales Law Reform Commission, Family Violence, above n 6, 879.

[9] Victorian Law Reform Commission, Review of Family Violence Laws: Report (2006) [8.91]; Australian Law Reform Commission and New South Wales Law Reform Commission, Family Violence, above n 6, 30.

[10] Jane Wangmann, ‘She Said ...’ ‘He Said ...’: Cross Applications in NSW Apprehended Domestic Violence Order Proceedings (March 2009) Sydney eScholarship Repository <http://ses.library.usyd.edu.au/handle/2123/5819> Jane Wangmann, ‘Gender and Intimate Partner Violence: A Case Study from NSW’ [2010] UNSWLawJl 42; (2010) 33 University of New South Wales Law Journal 945.

[11] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 967.

[12] This study focuses on cross-orders among heterosexual couples; however, since intimate partner violence also occurs among same-sex couples, future work will examine cross-applications in this population.

[13] See generally Australian Law Reform Commission and New South Wales Law Reform Commission, Family Violence, above n 6.

[14] Goldfarb, above n 3, 1487, 1489.

[15] Evan Stark, ‘Reconsidering State Intervention in Domestic Violence Cases’ (2006) 5 Social Policy and Society 149, 157; Leigh Goodmark, A Troubled Marriage: Domestic Violence and the Legal System (New York University Press, 2012) 9–10.

[16] DVPOs were introduced in NSW in 1982 under Crimes Act 1900 (NSW) s 547AA. DVPOs were introduced in Queensland in 1989 in response to the report: Domestic Violence Taskforce, ‘Beyond These Walls: Report of the Queensland Domestic Violence Taskforce to the Minister for Family Services and Welfare Housing’ (1988). For an early overview of these developments see Nicholas Seddon, Domestic Violence in Australia: The Legal Response (Federation Press, 2^nd ed, 1993).

[17] Sabina Crawley, ‘Reviewing the NT Government “No Drop” Policy: Moving from a Punitive Approach to Victim Support’ [2004] IndigLawB 63; (2004) 6(6) Indigenous Law Bulletin 14–16; Goodmark, above n 15, ch 5.

[18] Kimberly D Bailey, ‘Lost in Translation: Domestic Violence, “The Personal is Political” and the Criminal Justice System’ (2010) 100 Journal of Criminal Law and Criminology 1255, 1255.

[19] See Douglas and Godden, above n 2, 32.

[20] Sesha Kethineni and Dawn Beichner, ‘A Comparison of Civil and Criminal Orders of Protection as Remedies for Domestic Violence Victims in a Midwestern County’ (2009) 24 Journal of Family Violence 311; Douglas and Godden, above n 2, 32–43; Wangmann, ‘She Said ...’ ‘He said ...’, above n 10; Clare Connelly and Kate Cavanagh, ‘Domestic Abuse, Civil Protection Orders and the “New Criminologies”: Is there any Value in Engaging with the Law?’ (2007) 15 Feminist Legal Studies 259, 281; Chilla Bulbeck et al, Characteristics of Parties Involved in Domestic Violence Protection Orders: An Analysis of Court and Police Data (August 1997) Criminology Research Council <http://www.criminologyresearchcouncil.gov.au/reports/29-94-5.pdf.> .

[21] See essays in Martha Fineman (ed), The Public Nature of Private Violence: Women and the Discovery of Abuse (Routledge, 1995); Carol Jordan et al, ‘The Denial of Emergency Protection: Factors Associated With Court Decision Making’ (2008) 23 Violence and Victims 603. Peace and good behaviour orders informed the development of DVPOs; for a discussion of this history, see Queensland Law Reform Commission, A Review of the Peace and Good Behaviour Act, Report No 63 (2007) ch 2.

[22] Jordan et al, above n 21, 603.

[23] Logan et al, above n 2, 8.

[24] Kethineni and Beichner, above n 20, 311–12. For an analysis of the way breach of DVO charges are dealt with in Queensland courts, see Heather Douglas, ‘The Criminal Law's Response to Domestic Violence: What's Going On?’ [2008] SydLawRw 22; (2008) 30 Sydney Law Review 439.

[25] In this sense the burden is much lower than for criminal law where the standard is ‘beyond reasonable doubt’; see Domestic and Family Violence Protection Act 1989 (Qld) ss 9, 20; see also Elizabeth Topliffe, ‘Why Civil Protection Orders are Effective Remedies for Domestic Violence but Mutual Protection Orders are Not’ (1991– 1992) 67 Indiana Law Journal 1039, 1048.

[26] Kethineni and Beichner, above n 20, 320; The National Council to Reduce Violence Against Women and their Children, ‘Time for Action: The National Council’s Plan for Australia to Reduce Violence against Women and their Children 2009–2021’ (2009) 98; Jane Wangmann, ‘Incidents v Context: How Does the NSW Protection Order System Understand Intimate Partner Violence?’ [2012] SydLawRw 32; (2012) 34 Sydney Law Review 695, 698.

[27] Goldfarb, above n 3, 1487–551, 1490; Silke Meyer, ‘Why Women Stay: A Theoretical Examination of Rational Choice and Moral Reasoning in the Context of Intimate Partner Violence’ (2012) 45 Australia and New Zealand Journal of Criminology 179.

[28] Kethineni and Beichner, above n 20, 311–12; see Heather Douglas and Tanya Stark, Stories from Survivors: Domestic Violence and Criminal Justice Interventions (2010) The University of Queensland <http://www.law.uq.edu.au/documents/publications/douglas/Stories-From-Survivors-Douglas-Stark.pdf> 32–9.

[29] Goldfarb, above n 3, 1487, 1509; see Heather Douglas, ‘Battered Women’s Experiences of the Criminal Justice System: Decentring the Law’ (2012) 20 Feminist Legal Studies 121.

[30] Patrick Parkinson, Judy Cashmore and Atlanta Webster, ‘The Views of Family Lawyers on Apprehended Violence Orders After Parental Separation’ (2010) 24 Australian Journal of Family Law 313, 330.

[31] Kethineni and Beichner, above n 20, 311–12.

[32] Ibid; cf Logan et al, above n 2, 8; Victoria L Holt et al, ‘Civil Protection Orders and Risk of Subsequent Police-Reported Violence’ (2002) 288 Journal of the American Medical Association 589.

[33] Australian Law Reform Commission and New South Wales Law Reform Commission, above n 6, 74, 537.

[34] Henry J Muller, Sarah L Desmarais and John M Hamel, ‘Do Judicial Responses to Restraining Order Requests Discriminate Against Male Victims of Domestic Violence?’ (2009) 24 Journal of Family Violence 625, 627.

[35] Ibid.

[36] Fitzpatrick, above n 7.

[37] Australian Law Reform Commission and New South Wales Law Reform Commission, above n 6, 877–82.

[38] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 967.

[39] Fitzpatrick, above n 7, 2; Douglas and Godden, above n 2, 28–9.

[40] Victorian Law Reform Commission, above n 9, 282–3.

[41] Ibid 282.

[42] Australian Law Reform Commission and New South Wales Law Reform Commission, above n 6, 877, [18.208].

[43] Ibid 371. While a breach charge reflects contempt for a court order, in sentencing such offences judges have focused more on contempt for the rights of the aggrieved, see CCR v Queensland Police Service [2010] QDC 486.

[44] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 967.

[45] Topliffe, above n 25, 1055.

[46] Ibid.

[47] Victorian Law Reform Commission, above n 9, 284.

[48] Australian Law Reform Commission and New South Wales Law Reform Commission, above n 6, 51.

[49] Libby Lloyd, ‘Recent Developments and Best Practice in the Area of Family Violence at Both State and Federal Level’ (Paper presented at Australian Institute of Judicial Administration Family Violence Conference, Brisbane, 1 October 2009) <http://aija.org.au/Family%20Violence%2009/Papers/

Lloyd.pdf>. In 2005 the Queensland Crime and Misconduct Commission reported that 78 per cent of domestic violence victims are female: Crime and Misconduct Commission, above n 6, 35.

[50] Australian Law Reform Commission and New South Wales Law Reform Commission, above n 6, 50; see also Clare Murphy and Janet Fanslow, ‘Building Collaborations to Eliminate Family Violence: Facilitators, Barriers and Good Practice’ (Issues Paper No 1, New Zealand Family Violence Clearinghouse, March 2012); Amanda L Robinson and Emma Howarth, ‘Judging Risk: Key Determinants in British Domestic Violence Cases’ (2012) 27 Journal of Interpersonal Violence 1489, 1491.

[51] Connelly and Cavanagh, above n 20, 281; Australian Institute of Criminology, Homicide Victim–Offender Relationship Statistics (2006–2007) <http://www.aic.gov.au/en/statistics/homicide/victim-offender.aspx> .

[52] For an overview of this literature, see Russell P Dobash and R Emerson Dobash, ‘Women’s Violence to Men in Intimate Relationships: Working on a Puzzle’ (2004) 44 British Journal of Criminology 324. See also Kris Henning and Brian Renauer, ‘Prosecution of Women Arrested for Intimate Partner Abuse’ (2005) 20 Violence and Victims 361, who argue that women are as aggressive as men and that greater social attention to domestic violence issues has encouraged male victims of abuse to make a complaint. See also Zeev Winstok, ‘The Paradigmatic Cleavage on Gender Differences in Partner Violence Perpetration and Victimization’ (2011) 16 Aggression and Violent Behaviour 303; Julie People, ‘Trends and Patterns in Domestic Violence Assaults’ (2005) 89 Crime and Justice Bulletin 1.

[53] Dobash and Dobash, above n 52, 324, 349.

[54] Ibid 326–7; Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 946–7.

[55] Michael S Kimmel, ‘“Gender Symmetry” in Domestic Violence: A Substantive and Methodological Research Review’ (2002) 8 Violence Against Women 1332, 1333, 1335. For an overview of the literature that points to concerns with CTS see Wangman, ‘Gender and Intimate Partner Violence’, above n 10, 948–50.

[56] Kimmel, above n 55, 1332, 1342. See also Evan Stark, ‘Commentary on Johnson’s “Conflict and Control: Gender Symmetry and Asymmetry in Domestic Violence”’ (2006) 12 Violence Against Women 1019, 1020. See also Molly Dragiewicz and Walter S DeKeseredy, ‘Claims About Women’s Use of Non-Fatal Force in Intimate Relationships: A Contextual Review of Canadian Research’ (2012) 18 Violence Against Women 1008, 1012.

[57] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 949.

[58] Kimmel, above n 55, 1332, 1334.

[59] Dobash and Dobash, above n 52, 324–49, 329–30, 333; Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 949.

[60] Linda Coates and Allan Wade, ‘Language and Violence: Analysis of Four Discursive Operations’ (2007) 22 Journal of Family Violence 511, 513.

[61] Kimmel, above n 55, 1332, 1337.

[62] Dobash and Dobash, above n 52, 327.

[63] Ibid 343.

[64] Ibid.

[65] Marianne Hester, ‘Who Does What to Whom? Gender and Domestic Violence Perpetrators’ (Report, University of Bristol in association with the Northern Rock Foundation, Bristol, 2009), 5; see also Marianne Hester, ‘Portrayal of Women as Intimate Partner Domestic Violence Perpetrators’ (2012) 18 Violence Against Women 1067, 1072.

[66] Hester, ‘Who Does What to Whom?’, above n 65. See also Heather C Melton and Joanne Belknap, ‘He Hits, She Hits: Assessing Gender Difference and Similarities in Officially Reported Intimate Partner Violence’ (2003) 30 Criminal Justice and Behaviour 328, 344, where the authors suggested women’s greater use of weapons was a ‘means of levelling the playing field’.

[67] Eve S Buzawa and David Hirschel, ‘Domestic Violence: The Beginning, Continuation, or Final Act in a Criminal Career?’ (2008) 3 Victims and Offenders 391, 408.

[68] Moracco et al, above n 3, 1217.

[69] Ibid.

[70] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 947.

[71] Ibid 958–9.

[72] See, eg, Douglas and Stark, above n 28, 22.

[73] Australian Institute of Family Studies ‘Statistical Information’ (Commonwealth Government, Canberra, 2010) <http://www.aifs.gov.au/acssa/statistics.html> indicates only around 14 per cent of women sexually assaulted by their intimate report the matter to police.

[74] Wangmann, ‘Gender and Intimate Partner Violence’, above n 10, 959.

[75] Michael P Johnson, Typology of Domestic Violence: Intimate Terrorism, Violent Resistance, and Situational Couple Violence (Northeastern University Press, 2008); Michael P Johnson, ‘Conflict and Control: Gender Symmetry and Asymmetry in Domestic Violence’ (2006) 12 Violence Against Women 1003; Michael P Johnson, ‘Patriarchal Terrorism and Common Couple Violence: Two Forms of Violence Against Women’ (1995) 57 Journal of Marriage and Family 283; Joan B Kelly and Michael P Johnson, ‘Differentiation Among Types of Intimate Partner Violence: Research Update and Implications for Interventions’ (2008) 46 Family Court Review 476. See also Evan Stark, Coercive Control: How Men Entrap Women in Personal Life (Oxford University Press, 2007) 104.

[76] Johnson, ‘Conflict and Control’, above n 75, 1014.

[77] Johnson, Typology of Domestic Violence, above n 75, 11.

[78] Ibid 7.

[79] See, eg, the Duluth model: Domestic Abuse Intervention Programs, The Duluth Model (2011) <http://www.theduluthmodel.org/> .