Police Initiated Protection Orders and their Potential Impact on WomenA Discussion Document

Alison Towns¹
Mt Albert Psychological Services Ltd

Abstract

Access to protection orders for women experiencing domestic violence has recently been identified as a problem in New Zealand, and ways of addressing this problem are now being developed. Police-initiated protection orders, called safety orders, have been proposed to provide immediate protection to women who experience domestic violence and to overcome some of the evidential difficulties that can arise through representation in the Family Court. This discussion paper was commissioned to provide a review of the literature relating to safety orders. The paper uses a conceptual framework to identify the implicit assumptions associated with safety orders initiated by the police, and then compares these assumptions with the relevant research findings relating to police interventions, the actions women take to protect themselves and the court's responses to breaches of protection orders. Finally, conclusions are drawn about the ongoing need for education of the police and court personnel, and the need for compassionate involvement of women and their advocates in new initiatives. Research is encouraged when new initiatives are under development in this area to detect unintended consequences.

Introduction

In light of increasing international concern about domestic violence against women by their male partners (Garcia-Moreno et al. 2005), attention is being paid to new interventions that might usefully bypass some of the current problem areas. One such area is access to protection orders. Australia has been experimenting with police-initiated protection orders as a means to provide urgent protection to women who have experienced domestic violence, thereby avoiding the delays associated with protection orders initiated by the women themselves. A similar process is being developed in New Zealand (Ministry of Justice 2008).

This paper was commissioned by the New Zealand National Collective of Independent Women's Refuges to provide an independent review and discussion document on the merits and potential disadvantages of police-initiated protection orders, which will here be called safety orders. The paper begins with a description of domestic violence in the New Zealand context and follows this with a conceptual framework for the review. The paper then discusses some of the implicit assumptions of safety orders and compares these assumptions with the research findings. On the basis of this international and local research, some conclusions are drawn about ways to proceed with such innovations.

The New Zealand Context

Te Rito, the New Zealand Family Violence Prevention Strategy, defines family violence as:

… a broad range of controlling behaviours, commonly of a physical, sexual and/or psychological nature which typically involve fear, intimidation and emotional deprivation. Such violence occurs within a variety of close interpersonal relationships, such as between partners, parents and children. (Ministry of Social Development 2002:8)

Under the Domestic Violence Act 1995 (DVA), domestic violence is defined as violence against a person "by any other person with whom that person is, or has been, in a domestic relationship" (DVA s. 3[1]), where a domestic relationship is defined broadly, and includes having "a close personal relationship" (DVA s. 4). Violence is defined as physical abuse, sexual abuse and psychological abuse, "including but not limited to … intimidation … harassment, … damage to property … threats of physical abuse ... sexual abuse or psychological abuse" (DVA s. 3[2]). Also, see DVA s. 3(5), which covers the situation where a child is caused to witness such violence, or where that child is put at risk of seeing or hearing such violence (DVA s. 3[3]). Note that the clause relating to children exempts victims of domestic violence from causing or allowing children to see such violence (DVA s. 3[3]).

Typically, domestic violence requiring protection orders in New Zealand is perpetrated against women by a male partner or ex-partner, and violence towards women partners appears to be relatively common in New Zealand, although severe physical or sexual violence is less common. In a large population-based survey, Fanslow and Robinson (2004) found that of 2,674 ever-partnered New Zealand women, 33--39% had experienced physical and/or sexual violence from a male partner at some point in their lifetime, compared to Australian figures of around 20% (Grande et al. 2003, VHS 2004). Severe violence lifetime prevalence in New Zealand was 18.9--23.4% (Fanslow and Robinson 2004). Around a third of New Zealand men admitted to using physical violence against their female partner at some point in their lifetime (Leibrich et al. 1995).

In 2005 the New Zealand Police recorded 62,470 offence and non-offence family violence incidents; 62,615 children and young people aged under 17 years were involved (TAVF 2006). Between 2000 and 2004 54 women were murdered by men through family violence, and three men were murdered by women (TAVF 2006). Although the overall murder rate is declining, murders that are domestically related are not (TAVF 2006); in fact the number of deaths of women due to domestic violence has increased from an average of nine per year in the 10 years to 1987 (Fanslow et al. 1991) to an average of 15 in the years 2000 to 2004, suggesting a real increase (see TAVF 2006). In 2007, 25 of 53 murders were recorded as family violence-related (TAVF 2007). In 2005 Women's Refuge supported 17,212 women and 9,904 children (TAVF 2006). Between 90 and 95% of all applicants for protection orders in New Zealand are women, and most respondents are men (Bartlett 2006, Law Commission 2003).

Protection orders are available in New Zealand under the Domestic Violence Act 1995 (DVA) and are provided through the Family Court, whereas breaches of the Act are enforced through the Criminal Courts. Protection orders offer protection to both the applicant and the children from physical, emotional and/or sexual violence (DVA s. 16 and s.19). An application can be sought without notice if there is a risk of harm or undue hardship (DVA s. 13[1]). Without-notice applications provide urgent temporary protection orders, but these orders do not come into effect until served and the orders can be varied prior to coming into effect (DVA s. 46). Courts must take into account the applicant's and children's perceptions of the nature and seriousness of the respondent's behaviour (DVA s. 14[5]). Once served, temporary protection orders may be challenged by the respondent (DVA s. 76). With-notice applications are served if there is no immediate hardship or harm and to allow the respondent to challenge the application. Temporary protection orders are made final after three months if there is no challenge filed after they have been served, or following a defended hearing that finds in favour of the applicant (DVA ss. 77 & 78). The DVA allows for an associate of an abused person to seek a protection order on their behalf (DVA ss. 11 &12).

There are proposals underway to amend the DVA so that it will accommodate police-initiated protection orders more robustly through 72-hour safety orders (Ministry of Justice 2008). The intention is that these orders will provide short-term safety in crisis situations through police removal of the alleged offender where there is no evidence of a criminal offence. The orders will have a no-contact provision and will be available immediately as a "civil law remedy" (Ministry of Justice 2008:3), in contrast to protection orders issued under the DVA, which take some time to be processed through the Family Court. Before a safety order could be issued, the attending officer(s) would be required to conduct a risk assessment and contact an authorising senior officer, who would determine whether "the grounds for the order have been met" (Ministry of Justice  2008:3). The following discussion focuses on the issues that should be considered as part of this development. First, the principles that guide the discussion will be explained.

Conceptual Framework

The whole area of domestic violence against women is complex and filled with traps for those unfamiliar with the territory. A conceptual framework for this discussion will help to provide a structure that will allow at least some of these traps to be avoided. This paper makes use of Hudson's (2006) suggested principles of justice in guiding the discussion: discursiveness, relationalism and reflectiveness. These principles are consistent with social science knowledge in this area, which acknowledges the importance of language and power to an understanding of the dynamics of domestic violence (Adams et al. 2003, Heise 1998, Pence and Paymar 1993, Ptacek 1988, Towns and Adams 2000, Towns et al. 2003, Towns and Scott 2006).

• The principle of discursiveness refers to the importance of language and the meanings that are given, not only to the ways in which laws are constructed, but also to the ways in which actions taken to enforce those laws may be interpreted. In this review, attention will therefore be paid to the ways battered women are able to speak and be heard when police initiate safety orders, and the messages that actions relating to these orders might convey to women and others.
• The principle of relationalism refers to the importance of relationships, and how comfortable people are to express themselves in those relationships, for understanding the actions people take. In particular, this principle acknowledges that people's reactions will depend on the context they are in: for example, a woman may react differently when with a certain man from the way she may react with others, and people from a population group may react differently when their ethnic group is the minority in a community rather than the majority. This principle "recognizes individuals as embodied in a network of relationships, which include relationships with the community and with the state"(Hudson 2006:37). Hudson argues that rights are relational and are "regulatory safeguards against oppression" (p.37). The principle of relationalism recognises power relations where one party may engage in dominant and controlling practices over another. This review will therefore pay attention to relationships of power and the ways they may affect women when safety or protection orders are employed.
• The principle of reflection refers to the need for justice to be reflective and to consider the implications of actions it takes on those who may not have been considered in the formulation of its rules and regulations. Relevant to this principle are legal matters relating to the safety of women and the accountability of offenders. This principle is concerned with the need to move beyond generalised legal formulations and to pay attention to differences, in particular differences associated with race and gender, and with being in or emerging from an abusive relationship. This review will therefore attempt to examine the tensions and ethical issues relating to the implementation of safety orders. It will attend to the particular issues that arise for women in relation to the abusive relationship, and related safety and accountability issues when the justice system is involved.

Attention to these principles in a review of domestic violence against women is likely to address the criticism made of previous research in the area that it has failed to take account of context (Yllo and Bograd 1988).

Safety orders may be understood as an exercising of institutional power in the context of power exercised by a man through violence towards a woman. It is important to acknowledge the sensitivities around this power dynamic and the implications for the women who are victims of such violence in the development of any new policies.

Safety Orders -- What do we know?

With the increase in domestic violence murders in New Zealand and increasing concern about women and children's safety, the attention being paid to an alternative approach involving police-initiated safety orders is not surprising. Evidence relating to the effectiveness of such orders is necessary, but in this literature search there were few peer-reviewed published research reports found relating to such orders and their use in New Zealand, Australia or elsewhere internationally. However, an excellent review of the issues associated with their implementation has recently been released by the Government of Western Australia (Department of Attorney General 2008).

New Zealand attention to such orders was initially drawn by anecdotal reports from Australia. For example, Jack Johnson, Deputy Commissioner of Police in Tasmania, spoke of the effectiveness of police-initiated family violence orders in increasing the reports of domestic violence following their initial introduction in Tasmania (Checkpoint 2007). Burton (2003) describes a qualitative study of 60 key stakeholders -- professionals and frontline support workers -- and their opinions of third-party applications. Burton's study raised the potential for unexpected harm through further violence towards the woman survivor; the benefits to those women who cannot afford protection orders; the benefits of the validation of state responsibility to hold the perpetrator accountable; the potential for inappropriate state intervention; and issues relating to the survivors' consent and enforcement of breaches. Most participants in the study preferred police-initiated third-party applications.

Judge Boshier (2006) pointed to the possibility of using police-initiated protection orders in New Zealand to overcome limited access to the Family Courts over the weekend and to provide substantiation that there has been a violent incident:

[An] exciting issue we are considering is a wholly new way of some protection orders being obtained in the first instance. In a number of Australian States, when Police visit a scene of domestic violence, they can hand to the alleged perpetrator, an injunctive notice which has the effect of a very short term protection order. There is merit in our looking at a similar but more comprehensive process, in New Zealand. … One possibility is that as a result of a domestic violence incident, frontline Police may apply to a commissioned officer for an order with the limited duration of seven days. If the victim wished to obtain protection beyond this time an application would need to be made in the Family Court. Such an idea has much to commend it but of course like all new ideas, needs to be thoroughly thought through and debated. (Boshier 2006:11)

In New South Wales, apprehended violence orders (AVOs) may be made by the police and are heard before a magistrate in court (Lawlink 2006). The police arrange for a police prosecutor to represent the woman in court and a domestic violence liaison person is assigned to assist the woman and the police prosecutor. If the orders are undefended, the magistrate makes the AVO on the same day. If defended, the magistrate makes a time for the orders to be heard and in the meantime an interim AVO may be made to protect the woman. Once the evidence is heard the magistrate will make the order if, on the "balance of probabilities", the woman is considered to fear the defendant and there are reasonable grounds for these fears. Under federal law the state or territory courts are responsible for protection orders and applications while the family courts are responsible for parenting orders (Irwin 2006). Each court (state/territory and family) operates under different jurisdictions, with different definitions of domestic violence and family violence (Irwin 2006). In New Zealand, both protection orders and parenting orders are the responsibility of the Family Court.

The attraction of safety orders to the New Zealand Family Court is that because the police are present at the scene of the incident they can verify whether an offence occurred. The police will have documented evidence of any report of assault, and this evidence will be brought to the attention of the Family Court at any further hearings. The argument is that, as the offender is likely to be present, the application can be served immediately and the order is therefore potentially immediately enforceable. Until recently, due to restrictive legal aid provisions, protection orders in New Zealand were commonly initiated through more junior lawyers, who may not have accessed police reports of incidents and who may not be as familiar with the dynamics of domestic violence and with interviewing traumatised clients (Towns and Scott 2006). If the incident happened over the weekend, the applicant may have to wait until the court sits before receiving any protection, whereas safety orders can be implemented immediately (Boshier 2006, Ministry of Justice 2008). For the Family Court, because the police have immediate access to the offending incident, safety orders will have the advantage of countering any respondent's challenge that the action in taking out the application was vindictive and related to care of the children (see Davis 2004).

The Implicit Messages of Safety Orders

The meaning safety orders have for women needs to be considered before introducing these orders. One interpretation is that the police need to initiate such protection orders because the woman cannot or does not. Under this scenario, the woman's action in failing to initiate a protection order herself may be seen as a personal failure on her part. If she knows about the potential for a protection order yet does not seek one for herself, then her decision may be considered to be ill-informed, neglectful or simply remiss. This implicit judgement of the woman needs to be acknowledged. A second interpretation is that the woman cannot take out the order because she is too fearful of the repercussions from her partner should she do so, or because she is too traumatised, and indeed this may well be the case for some women. This interpretation suggests that the woman is helpless in the face of her partner's offending, and therefore the police will lend support to her by initiating the safety order. This action signals to the offender that the woman has the weight of the state behind her.

Both interpretations potentially fail to take into account that the woman may be acting purposefully: that she may be making considered decisions based on her current situation, including information she has gathered and knowledge of her particular partner, the potential agency response and the context of the violence. For example, Towns and Scott (2006) found that women's advocates in a large New Zealand urban context were less inclined to encourage women to apply for protection orders because they lacked confidence that the court would proceed with temporary orders and that the police would act on breaches. Women's advocates noted that applying for protection orders involved a risk of further violence for many women applicants because for some offenders the application was like a "declaration of war" (Towns and Scott 2006:159), and women who sought their help needed to be fully informed of the risk. Under the DVA, when appointing a representative, reasonable steps must have been taken "to determine the wishes of the person to whom the application relates" (DVA s. 12[3] p.13).

The Promise of Safety Orders

One appeal of police-initiated orders is the potential for preventing further domestic violence where there is repeated offending but difficulty in laying charges. If most women who experience violence from their partners have some form of protection order, then when the police are called to an incident they can at least charge the offender with a breach of the order. This presupposes, however, that the police routinely enact the law with regard to breaches of protection orders and that breach charges will be successful in preventing further violence. The latter presupposes a smooth process through the Family Court and the Criminal Court. In the following, the literature relating to the police response to protection orders and that relating to Family Court processes will be scrutinised in order to determine the validity of this presupposition.

Known Police Response to Breaches of Protection Orders

Women's experiences of police responses to domestic violence incidents in New Zealand are variable, with some women experiencing some officers as very helpful and supportive and some experiencing others as singularly judgemental and unhelpful (Hand et al. 2002, Robertson et al. 2007). One New Zealand study found that whereas some police officers were well aware of police policy in relation to domestic violence and spoke of intervening accordingly, some police officers appeared to determine their interventions in domestic violence according to their assessment of how deserving the victim appeared to be  (Towns 1999). Repeat call-outs without any apparent change contributed to a culture that labelled domestic violence incidents "just another domestic" that would be better dealt with through counselling interventions (Towns 1999). Similarly, research elsewhere indicates that some police officers consider therapy to be more appropriate than sanctions with domestic violence offenders (Harris et al. 2001, Logan et al. 2006). The New Zealand Police attempted to address this culture through a mass media campaign message that "Family violence is a crime" soon after the DVA was first introduced, but this campaign was time-limited and has not continued.

How Often do Police Lay Charges of Breaches?

International research indicates that police officers do not routinely arrest for breaches of protection orders, even when there are mandatory arrest policies (Carswell 2006, Jordan 2004). In a review of the literature on the extent of arrest due to violation of restraining orders, Kane (2000) found an arrest rate of between 20 and 40%. In his subsequent study of 818 domestic violence incidents he found an arrest rate of 76% in high-risk situations, when arrest was made to prevent further injury, and an arrest rate of only 44% in low-risk situations, when the arrest was solely for breaches of the order. In one study officers presented with vignettes describing domestic violence incidents said that they would have recommended protection orders in 61% of cases but would arrest for protection order violation in only 21% of the cases (Rigakos 1997, cited in Kane 2000).

When do Police Lay Charges of Breaches?

Kane (2000) suggested that consideration should be given to the "custody threshold" (p.565), noting that police officers appeared to apply varying standards when determining when to arrest. Police officers were found to prefer treatment rather than sanctions in domestic violence cases, particularly if the offender had a drug and alcohol problem (Logan et al. 2006). Kane noted that the police operate in a climate that de-emphasises arrest, and that they typically rely on arrest as "an emphatic last resort" (p. 564). Police officers assessed whether decisions to arrest were related to the purposes of custody and therefore actively determined "arrest worthiness" (p. 564). He found that prevention of "risk and injury to the victim were the strongest predictors of arrest, above all factors including violation of the restraining order" (p. 576). Although violation of the restraining orders was recognised by police officers as legitimate, "they appeared to have the propensity to stop short of actually enforcing the orders" (p. 576). Accordingly, Kane recommended that officers be trained to recognise different custody standards and taught to understand the importance of arresting for administrative purposes -- such as when breaches occur.

Prosecutions for Breaches and Associated Risks

 Towns and Scott (2006) found that although arrests for breaches do occur, custodial sentences for breaches were the exception in New Zealand:

In the period 2003/2004 there were 3,518 charges laid nationally for breaches of protection orders and in the period 2004/2005 there were 3,355 charges laid for such breaches. In 2003, 60% of prosecutions for breaches were successful. Around 10 to 14% of those convicted for breaches received a custodial sentence. The average custodial sentence at that time was three to four months. (p.158)

Charges for breaches may be laid for violation of the protection orders and for non-attendance at court-mandated programmes. Under the DVA programme, providers of programmes must inform the registrar of the Family Court if the man does not attend the programme and the registrar must bring the breach to the attention of the judge. Towns and Scott (2006) found that the number of breaches laid nationally had increased since the introduction of the DVA, and that the successful prosecution rate was high relative to other offences. Nevertheless, because few of the offenders received custodial sentences and most offenders who did were likely to be freed from custody within a month or two, prosecution in New Zealand does not appear to offer women much protection from the fear of their partner's further offending, although a limited respite may be a sufficient turning point for some women who experience such violence.

In their review, Logan et al. (2006) found little research relating to the extent of prosecution for breaches of protection orders. In relation to prosecution in general involving domestic violence, they found that if prosecution was not pursued there was a high risk of re-victimisation compared to cases where prosecution occurred. Consistent with findings on police action regarding arrests for breaches, prior criminal history and victim injury appeared to be associated with prosecution and conviction (Logan et al. 2006). Coulter et al. (1999) found that although 58% of 498 women who entered a shelter called the police in response to domestic violence, less than a quarter of the abusers were arrested. Poor system response has been identified by Erez and Belknap (1998) as a reason why women are not prepared to co-operate with prosecution, although 43% of the women victims of domestic violence in their study had experienced encouraging behaviour from the police.

There may well be some women who are pleased to work with the police and are relieved that the police will initiate a safety order, thereby absolving them from blame by their partners or ex-partners (Burton 2003). These women would be pleased that there is state involvement in holding the man accountable for his actions. Such women may also consider that safety orders will assist them with access to Family Court sanctions. For these women, such orders might provide a useful transition to an application for a temporary protection order.

Family Court Responses to Applications for Temporary Protection Orders

Alongside the expectation of a consistent police response, women also hold expectations that their safety will be protected by the courts. Applications for protection orders in New Zealand are made in the Family Court, and in recent years there has been concern about the ways in which the DVA is implemented. The Family Court in New Zealand and elsewhere has come under considerable pressure from fathers' rights lobby groups. These groups are reported to have argued that temporary protection orders are misused by women for tactical advantage in child care disputes, that they are too easy for women to get, and that they are fundamentally unfair to fathers whose rights are violated by without-notice applications (Laing 2003, Law Commission 2003). Lawyers and judges are particularly sensitive to any potential for natural justice rights violations. Natural justice is considered to be a fundamental civil and political right that underlies the New Zealand justice system (Towns and Scott 2006). Natural justice in this context refers to the right to all documentation, and to be informed of any legal proceedings in order to answer any charges, a right of the respondent temporarily set aside when temporary protection orders are sought by an applicant under the DVA. Others have argued that the Act has been used when violence is only a "one off" or "situational" (Doogue 2004), although the Act has provision for use in such instances when the violence arouses fear (DVA s. 3).

In response to such criticisms, the New Zealand Law Commission (2003) suggested that the threshold for the provision of temporary protection orders should be lifted to involve "substantial harm" (p. 120), and that orders should be put on-notice whenever possible. This recommendation is contrary to the existing legislative criteria for without-notice applications of harm and undue hardship under the Act. There is also concern that such enactment without a legislative change to the DVA runs counter to the New Zealand Bill of Rights Act 1990 (NZBORA), which notes that no provision of any Act can be repealed because it is assumed to be counter to any provision of the NZBORA (NZBORA s. 4). Acts can only be changed through Parliament. Davis (2004) has argued that the court's attention to fathers' rights at the cost of women's access to protection orders amounted to gender bias.

Despite the discourses of fathers' rights groups and those who support them, there is no evidence that the DVA is used vindictively in New Zealand (Law Commission 2003). There are safeguards under the Domestic Violence Rules in that lawyers must certify that the contents of the application for protection order are truthful and that the order ought to be made (Law Commission 2003). Research elsewhere has found that, despite fears that protection orders would be used vindictively, they are not used as often as was expected by the small group of women who successfully navigated their way through their provision (Romkens 2006). Regarding the DVA, Atken (1998) and Clark (2003) have argued that the intention of the Act was to lower the threshold for the provision of protection orders in order to improve the safety of women and children after some horrendous deaths due to domestic violence. The objectives of the DVA specify "more effective sanctions and enforcement in the event of the protection orders being breached" (DVA s. 5[2(e)]). Clark (2003) has argued that concerns about natural justice human rights violations due to the provisions of the Act were unfounded because the Act provides rights protection for both applicants and respondents. More recently, Chief Family Court Judge Boshier (2006) noted:

… we must balance the intention of the Domestic Violence Act to be able to deal with matters swiftly when required, against the denial of the respondent's rights when making a without notice order, and the possible impact upon the care arrangements of any children involved. This results in a trade-off. The Court grants fewer without notice applications but speeds up the process for applications with notice. (p.9)

Women's organisations' concerns are that with-notice applications, because they are served to the respondent, do not offer the same protection as without-notice applications (Hann 2004). There was concern that attention to the respondent's rights to natural justice had been balanced against a woman's right to safety and the safety of her children, in contradiction to the NZBORA²  (Towns and Scott 2006). Validation for this concern is provided by the declining number of applications for protection orders, which have gone from 7,395 in 1997, a year after the DVA came into effect, to 4,560 in 2004/2005 (Towns and Scott 2006), and this decline continued in the following year to 4,534 (Boshier 2006). In the year to June 2006, 76% of applications for temporary orders were successfully granted, compared to 84.5% in the year ended 1999 (Boshier 2006). Final orders made as a percentage of total applications filed has steadily declined, from 62% in the year to June 1999 to 52.8% in the year to June 2006 (Boshier 2006), indicating that fewer women who seek protection orders had them made final.

Perry (2000) found that 35% of applications for protection orders in the Christchurch Family Court were discharged by the court, and of those discharged within a month of the initial protection order application date, 54% of the applicants had been subjected to severe abuse. Of those discharged after one month from the initial protection order application date, 79% had been subjected to severe abuse. Perry questioned judges' ability to predict those applicants who were safe from further violence. In New Zealand, lack of Family Court resources was found to affect the time available and the information accessible for judges to make decisions concerning temporary protection orders (Towns and Scott 2006). The lack of resources and the failure to prioritise the seeking of protection orders may well affect natural justice for women applicants. Perry (2000) found that out of 73 cases, in only nine had the judges provided reasons for discharging or not discharging the applications for protection orders. When judges fail to provide reasons for not discharging the applications, the reasons are likely to be in the applicant's affidavit (Robertson et al. 2007), but when judges do not provide reasons for discharging the applications, this failure may have a direct impact on women's ability to appeal the decision³.  Proposed amendments to the DVA will address this problem (Ministry of Justice 2008).

A lack of resources or a failure to prioritise protection orders may affect the time that judges have to apply to documentation and due diligence, and ultimately women's and children's civil rights (Slote et al. 2005). A similar lack of resources has been documented elsewhere. Freedman (2003) found that Family Courts were struggling to cope with the large numbers of applications for protection orders and for child custody and access hearings. Although there had been an increase in these applications, there had been no corresponding increase in resources allocated to deal with them. Freedman argued that this lack of resources has resulted in a fact-finding gap, particularly in relation to domestic violence, and secondary traumatic stress of those in contact with such violence in the Family Court. She argued the need for compassionate witnesses, particularly through interdisciplinary communities of support:

It is important to note that the well-being of children is directly related to the well-being of the adults who care for them. Unless protective resources for adults experiencing domestic violence are widely available, children suffer. (p.571)

She noted that the lack of fact-finding resources, alongside unaddressed "judicial bias and reactivity" (p. 583), severely affected the ability of judges to make informed decisions, and this has a direct impact on the lives of women and children. In New Zealand the rights not to be deprived of life or not to be subjected to torture or cruel treatment are protected under sections 8 and 9 of the NZBORA, and the Act applies to institutional or professional practices associated with legal actions. The NZBORA puts into New Zealand legislation the United Nations International Covenant on Civil and Political Rights. New Zealand has ratified other relevant United Nations covenants that protect women and children from violence, including the United Nations Convention on the Rights of the Child in 1990, the International Convention on the Elimination of All Forms of Discrimination against Women in 1984, the Convention Against Torture and other Cruel Inhumane or Degrading Treatment or Punishment in 1989, and the Universal Declaration of Human Rights in 1948.

Court Responses to Breaches of Protection Orders

Towns and Scott (2006) found in a large urban area of New Zealand that there were significant concerns among men's programme providers and women's advocates about the follow-up of men who failed to attend court-mandated programmes -- a breach of the protection order. Estimates by men's programme providers were that only a third attend the first session of the programme, and one pointed to the lack of prioritising breaches by some court registrars, with the result that some breaches were simply not processed in the Criminal Court. Women's advocates said that this lack of accountability of the man, either through police or court lack of resources, alongside the movement in the Family Court to place applications without notice on notice wherever possible, had affected their confidence in the usefulness of the protection orders.

Messages to Battered Women

When women take out protection orders, the promise of these orders is that there will be state action: the police can be called and will offer protection should their partners breach the order, and courts will respond with accountability measures. Where there are mandatory arrest policies for breaches of the orders, the expectation of women who have experienced violence is that their partner will be arrested when they report a breach.

When the police enforce the law and act on breaches, women receive the message that their concerns about their safety are warranted and that the police will ensure their safety through law enforcement. Successful progression of their case through the courts will provide further messages to women and to others that domestic violence against women is unacceptable, and that the state has real concerns about such behaviour and will act to ensure it does not happen. On the other hand, failure of the police to enforce the Act and deal with breaches of the orders may provide a message to some women that that their concerns about their safety are trivial, unwarranted and unimportant, and that the police have more important activities than men's domestic violence against women. Failure of the courts to hold the man accountable further reinforces this message. As Erez and Belknap (1998) note, "Inappropriate or inadequate system's responses may cause battered women a deeper despair than the abuse itself" (p. 263).

The police's and courts' actions on breaches will also hold a message for the offender: he will be held accountable for his violence and his violence is not condoned. When the police or the courts fail to act, the women's male partner is likely to be supported by this lack of action and may see this as a further message that he can violate the orders with impunity. As the evidence suggests, he may actually increase his level of abuse against the woman. Safety orders will give added responsibility to police officers to ensure the messages women and offenders receive are ones of safety and accountability.

Women's Appraisal of Risk and the Risk of Re-offending

Erez and Belknap (1998) state that the single most important barrier to women's co-operation with legal action is fear of the batterer. Whether women take out protection orders or not will depend on their appraisal of risk of further violence from the man. Little is documented about women's ability to determine their risk of experiencing further violence in the New Zealand context, although new work relating to death reviews of women killed through domestic violence may assist here. Research elsewhere has found that most women were able to predict their own re-assault risk, at least with the help of a risk assessment, indicating that agencies need to attend to the woman's own assessment of risk (Campbell 2004). This means that women would need to be assessed separately from the abuser to ensure full disclosure of risk factors. In Campbell's study (2004), the majority of women who were killed were seen in the criminal justice system, the health system, the social services or a shelter in the year prior to their death, suggesting that safety planning was possible at this stage. The majority were seen in the health system. Most women who were killed had not accessed a shelter in the year prior to their death, and few women died who had been in shelters during this time. Campbell argues that this finding indicates the efficacy of shelter programmes and the need to link women's advocacy more directly with services.

Concerning lethality risk, Campbell's research on femicide suggests that at least 50% of the women who were killed did not accurately assess their partner as capable of killing them (Campbell 2004). Campbell suggests that when women assess themselves as being at low risk there is added need for a lethality assessment and safety planning, and for work in partnership with service providers that can keep women safe. Prior arrest for domestic violence was protective for those women who were most at risk, but increased the risk of murder or attempted murder for those assessed as being at lower risk. The majority of the killers had never been arrested for domestic violence, although 38% of the women killed had called the police in the year prior to their death. Protection orders were not found to be a significant risk factor for femicide relative to other factors: 24.5% of the women killed had a protection order.

International research indicates variable findings relating to the risk of further offending following the issuing of protection orders. A discussion of such risk needs to bear in mind that there will be methodological and jurisdictional differences between studies. In their review, Logan et al. (2006) found that between 23% and 70% of women with protection orders experienced a breach. In another review of the literature, Jordan (2004) stated that some studies found a reduction in violence following the issuing of protection orders, whereas a meta-analysis of stalking studies found the orders were violated around 40% of the time and that there were perceived worse events following the issuing of protection orders around 20% of the time (Spitzberg 2002). In a two-year follow-up study Klein (1996 cited in Jordan 2004) found that almost half of the offenders re-abused the victim after the order was issued. Risk of re-offending has been found to be associated with the severity and persistence in the pattern of offending, the level of resistance to the order, the presence of children biologically related to the offender, and the shorter term of the relationship (Harrell and Smith 1996, Carlson et al. 1999 cited in Jordan 2004).

Holt et al. (2003) found there were stronger decreases in risk when the orders were maintained over the full nine-month period following the incident that led to the issuing of the protection order, indicating the importance of final and long-term protection orders to reduced risk. Tellingly, in New Zealand the percentage of final orders granted annually relative to the number of applications has reduced (see above). Holt et al. (2002) also found increased likelihood of police-reported psychological abuse with temporary protection orders. A protective factor was the police arrest of the offender at the time of the incident that led to the protection order: the likelihood of severe violence decreased in the following year (Campbell 2004).

These findings suggest that research will be needed to ensure that the provision of 72-hour safety orders does not have the unintended consequence of increasing the risk to the woman of further violence from the offender. Under the new safety order provisions, the police will be expected to carry out risk assessments (Ministry of Justice 2008). Research might also determine whether or not there are benefits from the immediate removal of the offender by the police when safety orders are instituted, and from the brief no-contact period. Of note is the fact that with safety orders the police will not be expected to find accommodation for the offender (Ministry of Justice 2008), and there is the possibility that he could return to harm the woman. Consideration needs to be given to the woman's appraisal of risk and the barriers to her seeking orders when engaging in any process of police-initiated protection orders, so that her control over her situation is not undermined and eroded through institutional practices.

Women's Decision to Seek Protection Orders

Much can be learned from the research on women's decisions to take out protection orders when considering the introduction of safety orders. The international literature indicates that those women who seek protection orders have severe histories of violence (Logan et al. 2006): most have histories of severe threats, severe physical violence, economic and resource abuse, and reported injuries, and 20 to 30% have histories of sexual assault. For many of those seeking to obtain protection orders, the violence was becoming more severe and frequent and the psychological abuse was escalating (Jordan 2004). Increasing attention to the sexual assault and stalking of women who experience violence from their partners indicates that around half of men who enrolled in an intervention programme had sexually assaulted their partner at least once (Bergen 2006) and that stalking is highly correlated with domestic violence (Melton 2007). Stalking of women who had experienced domestic violence was more likely to occur if the abuser was no longer in a relationship with the woman, had an alcohol or drug problem, was more controlling, and had engaged in stalking previously (Melton 2007). Jordan (2004) found that most women who sought protection orders had experienced physical assault, beating and choking, threats of harm or death, sexual abuse, threats with a weapon, stalking, and harassment and/or assault of their children. Commonly, women who sought protection had had a lengthy exposure to abuse (Jordan 2004). Wolf et al. (2000) found that financial independence and the abuse of family or friends were important factors associated with the decision to take out protection orders. Erez and Belknap (1998) found that most women (90%) who called the police did so following a violent incident that resulted in an injury.

New Zealand research suggests that most women who seek protection orders have experienced severe violence, and that this violence has continued for years rather than months or weeks (Perry 2000). Of those who went through the Christchurch Family Court:

• the overwhelming majority (88%) had feared for their safety
• 20% considered that the physical violence was getting worse 
• 63% feared for the safety of their children and 15% believed the violence was affecting their children
• 83% were separated at the time of the application and were "attempting to break the cycle of violence that had featured in their relationships" (Perry 2000:140).

These findings suggest that police officers attending any incident should pay particular attention to the criminal actions of the offender, as is recommended in the New Zealand Cabinet paper (Ministry of Justice 2008). Safety orders are not intended for use when a criminal offence has already occurred. The authorising officer must monitor for evidence of criminal action to ensure that safety orders do not replace the treatment of an offence as a crime. In the Western Australian experience, concern was expressed about the failure of some officers to charge for a criminal offence when there was clear evidence of an injury to the woman (Department of Attorney General 2008).

Barriers to Women Seeking Orders

In their review, Logan et al. (2006) identified two types of barriers to women seeking protection orders: accessibility and acceptability. Under accessibility were eligibility criteria: meeting the statutory criteria for the provision of the orders, and bureaucracy, including repeated court hearings, limited hours for accessing the orders, and difficulties with serving the orders. Romkens (2006) found that "the cases in which the protection order was granted did not always seem to involve obviously more serious or more imminent threats than the cases in which it was turned down" (p. 170). She argues that the potential arrest of the perpetrator resulted in a rhetorical reversal in which "victims become inevitably constructed as possessing great powers that in the hands of individuals are subject to abuse" (p. 172), and it was this judicial and bureaucratic fear that precluded safe options for women. Other bureaucratic barriers included the applicant's and justice personnel's lack of knowledge of the system, the lack of 24-hour access to temporary protection orders, and the cost of the orders. Perry (2000) found that 72% of the Christchurch applicants were not in paid employment, although women who are limited financially are entitled to legal aid. Using qualitative research, Towns and Scott (2006) identified the cost of the orders in New Zealand, and the use of junior inexperienced lawyers due to inadequate legal aid, and their lack of interviewing skills, as barriers to women getting orders. New immigrant women require interpreters, which, at the time of the Towns and Scott study, were not routinely available in the Family Court.

Safety orders may offer a solution to some of the accessibility barriers women have experienced in getting protection orders, but this research emphasises the importance of educating junior and other police officers about the complexities of such violence towards women and the ways safety orders work. The proposed New Zealand safety orders will address the issue of 24-hour access to orders.

Acceptability barriers refer to the embarrassment that many women feel about the violence being brought out of the private arena and into public view (Logan et al. 2006), an issue that is relevant to safety orders. Some women may also fear the consequences of an order, which is served on the man. Perry (2000) found that, in the Christchurch Family Court, having the application changed to an on-notice application resulted in women withdrawing applications, as did the man's notice to challenge the application or defend the making of a final order. Negative perceptions of the justice system and poor prior experiences have been found to have an impact on women's determination to seek or pursue protection orders (Logan et al. 2006), and may have an impact on their determination to enforce the provisions of the order. Potential criminal charges or prosecution may also deter some women who are more concerned with reducing harm to themselves and their children than criminalising their partners through reporting breaches (Romkens 2006). There are also likely to be cultural barriers to enforcing orders (Balzer et al. 1997). For example, Māori  women were found to have sought fewer applications for orders than would be expected (Perry 2000). The impact on Māori⁴ and other indigenous cultures of colonisation and associated distrust of institutional practices should not be ignored (Cram et al. 2002, Glover 1993), and so safety orders may be an unwanted state intervention for these women.

Safety

The extent to which safety orders actually offer safety will influence women's support of these orders, as can be seen from research on their applications for protection orders. As mentioned previously, fear of retaliation affects women's determination to pursue applications for protection orders (Logan et al. 2006). Lack of timely serving of the orders has been found to be a barrier to women accessing protection orders and should be alleviated by the immediate provision of safety orders. Harrell et al. (1993 cited in Logan et al. 2006) found that, of the women who did not obtain a protection order, 35% were talked out of the order by the offender, 11% were afraid of pursuing the order due to fear of retaliation, 6% were threatened by the offender, and in 4% of the cases the offender forced his way into the woman's home. Logan et al. (2006) found that the rate of failure to serve orders ranged from 18% to 91%, depending on the jurisdiction. The lack of enforcement (discussed above) also acts as a barrier to women, who are likely to decide whether to proceed with final orders on the basis of their successful enforcement of breaches of temporary protection orders (Logan et al. 2006). Women will need to see that safety orders are enforced if they are to have confidence in them. A lack of enforcement of the orders may discourage them from reporting breaches and from applying for protection orders, which may offer them more safety in the long run.

Harm

In their review, Logan et al. (2006) found that there was a high risk of re-victimisation when prosecution was not pursued compared to when it was, although there was minimal research on the impact on women victims of violations of protection orders, and on prosecution and conviction associated with enforcement. Erez and Belknap (1998) note that when enforcing domestic violence law through the criminal courts:

… attitudes, comments, opinions or assumptions of criminal processing personnel who deal with battered women can be, and often are, harmful and demoralizing to victims. … Negative comments or discouraging attitudes by criminal processing agents underline victims' powerlessness and helplessness. The data suggest that if police officers side and identify with the abuser, if they cannot appreciate the fear or vulnerability of the victim, if they do not recognize the power imbalance between the parties (all of these situations were mentioned as problems by the victims), these agents exacerbate the situation. (p.263)

These findings emphasise the importance of an appropriate response by the police to women who experience such violence. The new safety orders will mean that the police have added powers to remove offenders and enforce breaches. With these added responsibilities come increased requirements for accountability, although the Western Australian experience has been that in some cases police officers have not acted to protect the woman despite obvious criminality (Department of Attorney General 2008). Ongoing training, requirements for action, close monitoring and enforced officer accountability remain imperatives.

Benefits of Protection Orders for Women

Although women may experience violations of the protection orders, research suggests that many women report life improvements associated with their own initiation of an order. For example, Keilitz et al. (1997) found that of the 62% of their original sample followed up six months after obtaining an order, 85% reported life improvements, 93% reported improvements in how they felt about themselves, and 81% reported feeling safer. Fischer and Rose (1995 cited in Logan et al. 2006) found that of 287 women who had obtained a protection order, 98% felt more in control of their lives and 89% more in control of their relationships. Johnson et al. (2003) argue that victims' assessment of the effectiveness of the order is complex and is not commensurate with subsequent violence: 48% of their sample considered the order to be effective, although more than half of this group had experienced violence after the order was issued. Women victims in this study found value in "building a paper trail" (p. 321) for any later arrest or prosecution. Research is required to determine whether women would experience these same benefits if the orders are police initiated.

Compellability and Dual Arrest

Compellability

Police-initiated safety orders raise the issue of compellability. When a judicial procedure is initiated without the need for victim involvement, there remains the possibility that the matter may be heard in the Family Court, although at a later date (for example, if a woman's partner challenges the safety orders). At any subsequent hearing the woman will determine whether she is able or willing to co-operate with the court proceedings. If, for example, she considers that her co-operation may result in further risk to her safety or that of her children, she may determine that she does not want to participate in the court process. The Family Court judge, however, may be unwilling to discharge the order if the woman and her children are considered to be at risk. Under this scenario, the court will need to decide whether the victim needs to be heard and should be coerced or compelled to participate.

The woman may also be expected to appear in the Criminal Court if the respondent breaches the order within the limited time of the safety order. The police prosecutor will need to determine whether the woman victim/survivor is compelled to appear as a witness in the Criminal Court in relation to any such breach. Of note is that under the new Evidence Bill 2005, New Zealand police will be able to prosecute with less reliance on the victim, but the witness "can be required by the Court to give evidence" (Boshier 2006:16). However, the victim can be excused from testifying if the court decides that "requiring them to testify would cause hardship" (Boshier 2006:16).

Ford (2003) has argued that policies of witness coercion in the context of domestic violence raise serious questions that have not yet been addressed:

Research is due on unintended consequences and policy backfires: To what extent do coercive policies dissuade women from prosecuting? What is the impact for victims of arrested batterers when prosecutors declare their cases too weak to prosecute? Above all, there is an urgent need for research to test the protective impacts of not only coercive policies but especially policies acknowledging the interests of battered women. (p. 681)

Ford (2003) has argued that there needs to be community dialogue within and outside the justice system in order to determine the best ways to proceed with prosecution in domestic violence cases:

Can all parties agree on the prosecutor's responsibility to protect the victim whose batterer is being prosecuted? Is it appropriate that she be denied a part in the process in favour of general deterrence? Can prosecution be sufficiently flexible that it complements both victim strategies for self-protection and community-based initiatives for preventing domestic violence? Should the state's interest in offender accountability obviate a victim's interests in safety on her terms? (p. 681)

While these questions were raised in the context of domestic violence prosecutions, they might also be applied to Family Court proceedings involving compellability.

Dual Arrest

Alongside these concerns about compellability there is the issue of potential unintended consequences relating to dual arrest -- the arrest of both the man and the woman -- when police initiate safety orders, which is a major concern for women's advocates. Finn and Bettis (2006), for example, found that mandatory arrest policies were used as a justification by police officers for dual arrest. They found that police officers justified dual arrest as a legal requirement and through a desire to force the woman and man into relationship counselling. Saunders (1995) found that officers who arrested victims of domestic violence tended to believe that domestic violence was justified in some circumstances and that women remained in such relationships for psychological reasons. Stalans and Finn (2006) found a more complex picture, with experienced officers more likely to arrest only the husband than rookie officers, who generally tended towards dual arrest even when the husband was understood to be the primary assailant and there was knowledge that he had engaged in violence in the past. Experienced officers were more likely to engage in dual arrest if the injured wife was in a drunken state. These researchers note that:

Laypersons and rookie officers were more likely to focus on normative issues such as who was to blame whereas experienced officers were more likely to consider pragmatic and legalistic concerns such as the future dangerousness of each disputant and the likelihood of a conviction. (p. 1150)

This finding reinforces the importance of education for inexperienced police officers, particularly in relation to dangerousness, and legal requirements for conviction in relation to domestic violence and breaches of protection orders. Carswell (2006) points to the need for police training in better evidence gathering "to identify offensive and defensive wounds" (p. 37), because research indicates that women who injure men are quite often acting in self-defence. Dual arrest was raised as a matter of concern in the Australian review (Department of Attorney General 2008:21), but more research will be required on the complex relationship between dual arrest and the provision of, or enforcement of, safety orders.

Battered Women and Control/Empowerment

Safety orders raise important matters central to women's experiences of domestic violence from their male partners or ex-partners, such as the issue of control. When women experience domestic violence they often experience a shattering loss of control over many aspects of their lives:

Violent behaviours -- be they physical, psychological, economic and social -- are not just expressive acts but also instrumental acts that coerce the actions of others. The outcome of being coerced through exposure to repeated acts of violence is inevitably the diminishment of possibilities for action: because one fears the repercussions that will follow from taking such actions … This constriction on possibilities for action sets the battering context apart from most other contexts in which people commit crimes against those they know. (V. Elizabeth, cited in Law Commission 2001:6)

Safety orders have the potential to perpetuate the lack of control women who are beaten have been experiencing through the violence they have received. Of critical importance to any implementation of these orders will be the relationship the woman has with the police and with women's advocates, the extent to which she is party to the decision to proceed in this way, and the extent to which she is empowered or disempowered by the legal process. The impact of any further erosion of her control through orders initiated on her behalf needs to be well researched, particularly if their initiation is an alternative to a criminal response. As this experience is likely to influence her perception of the police and the justice system, it may well have an impact on her willingness to contact the police during any future violent incidents. Such orders, and how they are implemented, will therefore have real implications for her future safety and that of her children. For some women such a process may mean greater safety, but for others it may not. Ongoing research is required to determine the potential benefits and harms of such orders once introduced.

Conclusion

This paper is an attempt to draw together the various issues that may arise when considering police-initiated protection orders, here referred to as safety orders. The review is not intended to be definitive, but rather to raise some of the issues that require discussion as part of any development of safety orders. In particular, consideration needs to be given to: the messages that such orders convey to women who experience violence from their partners; the current police and Family Court response to applications and enforcement of provisions under the DVA, and the messages these responses convey; the ongoing need for education of police officers, the judiciary and associated personnel; and the need for open and public debate about the ways to proceed when safety is such an integral factor for women and children. Finally, this discussion raises the need to proceed compassionately through the involvement of women victims/survivors and their advocates in any future developments. Respectful research which involves women who have experienced such violence and the associated agency responses, and that makes use of violence research expertise and inter-agency co-operation, will add to future prevention and intervention developments.

References

Adams, P.J., A. Towns and N. Gavey (1995) “Dominance and entitlement: The rhetoric men use to discuss their violence towards women” Discourse & Society,6(3):387--406. Republished as Adams, P., A. Towns and N. Gavey (2003) “Dominance and entitlement: The rhetoric men use to discuss their violence towards women” in M. Talbot, K. Atkinson and D. Atkinson (eds.) Language and Power in the Modern World (pp. 184--198), Edinburgh University Press, Edinburgh.

Atkin, B. (1998) “The Domestic Violence Act” New Zealand Law Journal, Jan:24--31.

Balzer, R., D. Haimona, M. Henare and V. Matchitt (1997) Maori Family Violence in Aotearoa, Ministry of Maori Development, Wellington.

Bartlett, E. (2006) Family Court Statistics 2004, Ministry of Justice, Wellington.

Bergen, R.K. (2006) “Men and intimate partner rape: Characteristics of men who sexually abuse their partner” Journal of Interpersonal Violence, 21(10):1375--1384.

Boshier, Judge P. (2006) “Achieving an effective response to domestic violence” speech presented to the Te Awatea Violence Research Centre, University of Canterbury, Christchurch, 20 September.

Burton, M. (2003) “Third party applications for protection orders in England and Wales: Service providers’ views on implementing Section 60 of the Family Law Act 1996” Journal of Social Welfare and Family Law, 25(2):137--150.

Campbell, J.C. (2004) “Helping women understand their risk in situations of intimate partner violence” Journal of Interpersonal Violence, 19(12):1464--1476.

Carlson, M.J., S.D. Harris and G.W. Holden (1999) “Protective orders and domestic violence: Risk factors for re-abuse” Journal of Family Violence, 14(2):205--226, cited in Jordan 2004:1423.

Carswell, S. (2006) Family Violence and the Pro-arrest Policy: A Literature Review, Ministry of Justice, Wellington.

Checkpoint  (2007) New Zealand National Radio, 19 Feb 2007.

Clark, E. (2003) “Ex parte orders in the Family Court and the New Zealand Bill of Rights Act 1990” Butterworths Family Law Journal, 4:205.

Coulter, M.L., K. Kuehnle, R. Byres and M. Alfonso (1999) “Police-reporting behavior and victim--police interactions as described by women in a domestic violence shelter” Journal of Interpersonal Violence, 14:1290--1298.

Cram, F., L. Pihama, K. Jenkins and M. Karehana (2002) Evaluation of Programmes for Maori Adult Protected Persons under the Domestic Violence Act 1995, Ministry of Justice, Wellington.

Davis, W. (2004) “Gender bias, fathers’ rights, domestic violence and the Family Court” Butterworths Family Law Journal, Dec:299--312.

Department of Attorney General (2008) A Review of Part 2 Division of the Restraining Orders Act 1997, A report for the Government of Western Australia,  http://www.department.dotag.wa.gov.au/_manifest/parliamentary_report_roa.jmf [accessed 26/8/08][link broken 21/03/2012].

Doogue, J. (2004) The Domestic Violence Act 1995 and s.16B of the Guardianship Act 1968: The Effect on Children’s Relationships with Their Non-custodial Parent, Family Court of New Zealand, Wellington.

Elizabeth, V., cited in Law Commission (2001) Some Criminal Defences with Particular Reference to Battered Defendants, Report No. 73 (pp. 6--7), Law Commission, Wellington.

Erez, E., and J. Belknap (1998) “In their own words: Battered women’s assessment of the criminal processing system’s responses” Violence and Victims, 13(3):251--266.

Fanslow, J., and E. Robinson (2004) “Violence against women in New Zealand: Prevalence and health consequences” Journal of the New Zealand Medical Association, 117(1206), www.nzma.org.nz/journal/117-1206/1173/ [accessed 11/11/05].

Fanslow, J.L., D.J. Chalmers and J.D. Langley (1991) Injury from Assault: A Public Health Problem, a report prepared for the Alcoholic Liquor Advisory Council, New Zealand.

Finn, M.A., and P. Bettis (2006) “Punitive action or gentle persuasion: Exploring police officers’ justifications for using dual arrest in domestic violence cases” Violence Against Women, 12(3):268--287.

Fischer, K., and M. Rose (1995) “When ‘enough is enough’: Battered women’s decision making around court orders of protection” Crime & Delinquency, 41(4):414--429, cited in Logan et al. 2006:193.

Ford, D.A. (2003) “Coercing victim participation in domestic violence prosecutions” Journal of Interpersonal Violence, 18(6):669--684.

Freedman, A. (2003) “Fact-finding in civil domestic violence cases: Secondary traumatic stress and the need for compassionate witnesses” Journal of Gender, Social Policy and the Law, 11(2):567--655.

Garcia-Moreno, C., H.A.F.M. Jansen, M. Ellsberg, L. Heise and C. Watts (2005) WHO Multi-country Study on Women’s Health and Domestic Violence Against Women: Initial Results on Prevalence, Health Outcomes and Women’s Responses, World Health Organization, Geneva.

Glover, M. (1993) Maori Women’s Experience of Male Partner Violence: Seven Case Studies,Te Whare Wananga o Waikato, Hamilton.

Grande, E.D., J. Hickling, A. Taylor and T. Woollacott (2003) “Domestic violence in South Australia: A population survey of males and females” Australian and New Zealand Journal of Public Health, 27:543--550.

Hand, J., V. Elizabeth, B. Martin, H. Rauwhero, M. Burton, S. Selby, et al. (2002) Free From Abuse: What Woman Say and What Can Be Done, Auckland District Health Board, Auckland.

Hann, S. (2004) The Implementation of the Domestic Violence Act 1995: A Report from the National Collective of Independent Women’s Refuges Inc., National Collective of Independent Women’s Refuges Inc., Wellington.

Harrell, A., B. Smith and L. Newmark (1993) Court Processing and the Effects of Restraining Orders for Domestic Violence Victims, Urban Institute, Washington, DC,  cited in Logan et al. 2006:185.

Harrell, A., and B.E. Smith (1996) “Effects of restraining orders on domestic violence victims” in E. Buzawa and C. Buzawa (eds.) Do Arrests and Restraining Orders Work? (pp. 214--242),  Sage, Thousand Oaks, CA, cited in Jordan 2004:1424.

Harris, S.D., K.R. Dean, G.W. Holden and M.J. Carlson (2001) “Assessing police and protective order reports of violence: What is the relation?” Journal of Interpersonal Violence, 16(6):602--609.

Heise, L.L. (1998) “Violence against women: An integrated, ecological framework” Violence Against Women, 4(3):262--290.

Holt, V.L., M.A. Kernic, T. Lumley, M.E. Wolf and F.P. Rivara (2002) “Civil protection orders and the risk of subsequent police-reported violence” Journal of the American Medical Association, 288(5):589--594.

Holt, V.L., M.A. Kernic, M.E. Wolf and F.P. Rivara (2003) “Do protection orders affect the likelihood of future partner violence and injury?” American Journal of Preventative Medicine, 24(1):16--21.

Hudson, B. (2006) “Beyond white man’s justice: Race, gender and justice in late modernity” Theoretical Criminology, 10:29--47.

Irwin, Judge M. (2006) “Domestic violence: Implications of the new legislation” paper presented at the Calabro Consulting Family Law Residential, Novotel Twin Waters, Sunshine Coast, 19 August.

Johnson, J.M., Y. Luna and J. Stein (2003) “Victim protection orders and the stake in conformity thesis” Journal of Family Violence, 18(9):317--323.

Jordan, C.E. (2004) “Intimate partner violence and the justice system: An examination of the interface” Journal of Interpersonal Violence, 19:1412--1434.

Kane, R.J. (2000) “Police responses to restraining orders in domestic violence incidents: Identifying the custody-threshold thesis” Criminal Justice and Behaviour, 27:561--580.

Keilitz S., P. Hannaford and H. Efkemann (1997) Civil Protection Orders: The Benefits and Limitations for Victims of Domestic Violence (Publ. No. R-201), National Center for State Courts Research Report, Williamsburg, VA,  cited in Logan et al. 2006:193.

Klein, A.R. (1996) “Re-abuse in a population of court-restrained male batterers: Why restraining orders don’t work” in E. Buzawa and C. Buzawa (eds.) Do Arrests and Restraining Orders Work? (pp. 192--213), Sage, Thousand Oaks, CA,  cited in Jordan 2004:1425.

Laing, L. (2003) Domestic Violence and Family Law, Australian Domestic and Family Violence Clearinghouse. http://www.austdvclearinghouse.unsw.edu.au/pdf%20files/family_law.pdf

Law Commission (2003) Dispute Resolution in the Family Court, Report No. 82, New Zealand Law Commission, Wellington.

Lawlink (2007) Information for Protected Persons (Domestic Violence), http://www.lawlink.nsw.gov.au/lawlink/local_courts/ll_localcourts.nsf/pages/lc_avo2
[accessed 17/12/08].

Leibrich, J., J. Paulin and R. Ransom (1995) Hitting Home: Men Speak about Abuse of Women Partners, Ministry of Justice, Wellington.

Logan, T., L. Shannon and R. Walker (2006) “Police attitudes toward domestic violence offenders” Journal of Interpersonal Violence, 21(10):1365--1374.

Logan, T., L. Shannon, R. Walker and T.M. Faragher (2006) “Protective orders: Questions and conundrums” Trauma, Violence & Abuse, 7(3):175--205.

Melton, H. (2007) “Predicting the occurrence of stalking in relationships characterized by domestic violence” Journal of Interpersonal Violence, 22(1):3--25.

Ministry of Justice (2008) Review of the Domestic Violence Act 1995 and Related Legislation [Cabinet papers] Paper Two: The Role of the Police and the Enforcement of Protection Orders under the Domestic Violence Act 1995,Ministry of Justice, Wellington, www.justice.govt.nz/publications/global-publications/r/a-review-of-the-domestic-violence-act-1995-and-related-legislation-a-discussion-document-december-2007/publication/  [accessed 28/7/08].

Ministry of Social Development (2002) Te Rito: New Zealand Family Violence Prevention Strategy, Government report No. ISN:0-478-25120-3, Ministry of Social Development, Wellington.

Pence, E., and M. Paymar (1993) Education Groups for Men Who Batter: The Duluth Model, Springer Publishing Company, New York.

Perry, C. (2000) “An empirical study of applications for protection orders made to the Christchurch Family Court” Butterworths Familiy Law Journal, June:139--145.

Ptacek, J. (1988) “Why do men batter their wives?” in K. Yllo and M. Bograd (eds.) Feminist Perspectives on Wife Abuse (pp. 133--157), Sage Publications, Newbury Park, London and New Delhi.

Rigakos, G. (1997) “Situational determinants of police responses to civil and criminal injunctions for battered women” Violence Against Women, 3:204--216, cited in Kane 2000:563.

Robertson, N., R. Busch, R. D’Souza, F.L. Sheung, R. Anand, R. Balzer, et al. (2007) Living at the Cutting Edge: Women’s Experiences of Protection Orders, A report for the Ministry of Women’s Affairs, University of Waikato, Hamilton.

Romkens, R. (2006) “Protecting prosecution: Exploring the powers of law in an intervention program for domestic violence” Violence Against Women, 12(2):160--186.

Saunders, D.G. (1995) “The tendency to arrest victims of domestic violence: A preliminary analysis of officer characteristics” Journal of Interpersonal Violence, 10(2):147--158.

Slote, K.Y., C. Cuthbert, C.J. Mesh, M.G. Driggers, L. Bancroft and J.G. Silverman (2005) “Battered mothers speak out” Violence Against Women, 11(11):1367--1395.

Spitzberg, B.H. (2002) “The tactical topography of stalking victimization and management” Trauma Violence Abuse, 3:261--288.

Stalans, L.J., and M.A. Finn (2006) “Public’s and police officers’ interpretation and handling of domestic violence cases: Divergent realities” Journal of Interpersonal Violence, 21(9):1129--1155.

TAVF (2006) The Taskforce for Action on Violence within Families: The First Report, The Family Violence Ministerial Team, New Zealand Government, Wellington.

TAVF (2007) Taskforce for Action on Violence within Families: The Ongoing Programme of Action, Ministry of Social Development, Wellington.

Towns, A. (1999) “Deserving and undeserving victims and ‘It's just a domestic’: The influence of social discourses on police officers’ talk of male violence against women” plenary session at the Family Violence Police Coordinators’ Conference, New Zealand Police College, Porirua, Wellington, September 30.

Towns, A., and P. Adams (2000) “If I really loved him enough he would be okay: Women’s accounts of male partner violence” Violence Against Women, 6(6):558--585.

Towns, A., P. Adams and N. Gavey (2003) “Silencing talk of men’s violence towards women” in L. Theismeyer (ed.) Discourse and Silencing: Representation and the Language of Displacement (pp. 43--77), John Benjamin, Amsterdam and Philadelphia.

Towns, A., and H. Scott (2006) “Accountability, natural justice and safety: The protection order pilot study (POPS) of the Domestic Violence Act 1995” New Zealand Family Law Journal, 7(7):157--168.

VHS (2004) The Health Costs of Violence: Measuring the Burden of Disease Caused by Intimate Partner Violence: A Summary of Findings, Victorian Health Promotion Foundation, Carlton South, Victoria.

Wolf, M.E., V.L. Holt, M.A. Kernic and F.P. Rivara (2000) “Who gets protection orders for intimate partner violence?” American Journal of Preventive Medicine, 19(4):286--291.

Yllo, K., and M. Bograd (1988) Feminist Perspectives on Wife Abuse, Sage, Newbury Park, CA.

Footnotes

¹Acknowledgements
I would like to thank the National Collective of Independent Women’s Refuges Incorporated for partially funding this discussion paper. Special thanks are given to Heather Henare, who provided encouragement and support throughout, and to the two reviewers for their helpful comments. This project was funded in part by Mt Albert Psychological Services Ltd.
Correspondence
atowns@pl.net or  P.O. Box 41124, St Lukes, Auckland 1346, New Zealand.

²See s. 8 and s. 9 of the New Zealand Bill of Rights Act 1990, which are concerned with the rights not to be deprived of life or to be subjected to torture or cruel treatment.

³This right to appeal is protected in New Zealand law under the DVA s. 91 and under the New Zealand Bill of Rights Act 1990 s. 27(2).

⁴Māori are indigenous New Zealanders.