Anyone who has followed Australian family law over the last few years will be acutely aware of the level of conflict there is over the text of the Family Law Act 1975, especially in relation to parenting after separation. Family law is a field full of advocates. Views are often passionately held, and debate can too often resemble a form of trench warfare in which the goal is to capture territory rather than finding the common ground between different views and concerns. This adversarial approach to the issues also affects research. In this field, there is too much policy-based "evidence", and too little evidence-based policy. The outcomes of this approach to public policy have been unsatisfactory. The Family Law Act 1975 reflects various compromises between advocacy groups, and lacks coherence as a result.

This article seeks to suggest where the middle ground might be found in public policy, by placing Australian developments in the law of parenting after separation in a historical and comparative perspective. The thesis of this article (and a book, Parkinson, 2011) is that family law around the Western world has shifted fundamentally and irreversibly. The model on which divorce reform was predicated in the late 1960s and early 1970s has irretrievably broken down. Jurisdictions across the Western world have come to the sometimes painful conclusion that while marriage may be dissoluble, parenthood is not. Whereas once family law was premised on the indissolubility of marriage, now a defining feature of family law in Western societies is the notion that parenthood is indissoluble.

There has been considerable resistance to this transition from advocates, for many reasons. A major argument has been that the involvement of both parents in children's lives increases the risk of violence against women. That is an issue of great importance. But the middle ground is to be found in articulating more clearly the circumstances when parenthood ought to be dissoluble, rather than resisting the historic transformation in the law of parenting after separation.

The indissolubility of parenthood

In the last thirty years, profound changes have occurred in family law all around the Western world. There has come to be a recognition that children generally benefit from the involvement of both parents in their lives - in the absence of serious violence, abuse or high conflict between parents - and therefore children's relationship with both parents ought to be supported after separation.

Divorce as the dissolution of the family

This is a revolution. The model on which divorce reform was predicated in the late 1960s and early 1970s was that dead marriages should be given a decent burial and that it should be possible for the parties to get on with their lives and start afresh once decisions had been made about financial matters and custody. In the divorce law at that time, issues about property and custody were dealt with by a once-and-for-all process of allocation. If the parties could not reach their own agreement, then the court allocated the property. The aim in some jurisdictions was to achieve a clean break in terms of the financial affairs of the parties, apart from child support.

The court also allocated the children (Schepard, 2004, pp. 3-4). Typically, the courts would award "custody" to one parent, usually the mother, and grant "access" or "visitation" to the other. There was little difference in this respect between common law countries and the civil law countries of Western Europe. "Custody" included virtually all the rights and powers that an adult needed to bring up a child, including the right to make decisions about a child's education and religion. Both parents were legal guardians at common law, but this meant little, because the powers that were classified as powers of "guardianship" included only such matters as consent to marriage of a minor and inheritance rights in the event of his or her untimely death. Since maternal custody was the predominant pattern, fathers were frequently relegated to a peripheral role in their children's lives.

Custody law was thus binary in character. The assumption that was universally held at that time was that custody decisions involved a definitive choice between one home and another. In this traditional conceptualisation of what was involved in custody decision-making, "access" was simply a "legal concession to the loser" (Halem, 1980, pp. 213-14). Once this allocation had occurred, then people could get on with their lives with the past behind them. The old marriage was dead and they could begin anew, repartner, and build a new family life, with only residual ties to their former spouses. Those ties were through child support obligations - which were poorly enforced - spousal maintenance where ordered, and ongoing access time with the children.

The consequence of this view of custody decision-making was that divorce involved a clean break in terms of parental responsibility once the issue of custody allocation was decided. In a perceptive article written in 1986, Irène Théry, a French sociologist, characterised the original divorce reform model as the substitution model of post-divorce parenting. Under the substitute family model, the parents' legal divorce necessarily required a divorce between them not only as partners but also as parents. Only one of the two parents could continue in that role after the divorce. It followed that the marriage breakdown marked the dissolution of the nuclear family, and its substitution with a "new" family constellation for the child. Parental authority was awarded to the sole custodial parent, and there was a strong differentiation between the role of the custodial parent and that of the non-custodial parent. This way of seeing divorce was expressed pithily by the New York Court of Appeals in 1978: "Divorce dissolves the family as well as the marriage" (Braiman v. Braiman, 378 NE 2d 1019, 1022 [NY 1978]).

The emergence of the enduring family

It was not long after the first flush of the divorce revolution that this idea of post-separation parenting began to change. Théry argued, in her 1986 article, that the substitution model of the post-separation family was gradually being displaced and that a new concept of post-separation parenting was emerging. This she called the idea of the "enduring family". In this conceptualisation, divorce is a "transition between the original family unit and the re-organisation of the family which remains a unit, but a bipolar one" (Théry, 1986, p. 356). She noted that this conceptualisation of post-separation parenting implies the refusal of having to make a choice between parents, in favour of joint parental authority.

Change has occurred only very gradually in family law around the Western world, but the relentless march of progress has been in the direction that Théry anticipated. The history of family law reform in the last 30 years has seen the abandonment of the assumption that divorce could dissolve the family as well as the marriage when there are children involved. As Professor Margo Melli wrote: "Today, divorce is not the end of a relationship but a restructuring of a continuing relationship" (Melli, 2000, p. 638). Marriage may be freely dissoluble, but parenthood is not.

The transformation in custody law

The indissolubility of parenthood is seen in many different ways in modern family law. One aspect of it is financial. Child support is now vigorously enforced in Australia, and in many other countries (Oldham & Melli, 2000). In some jurisdictions, spousal maintenance is experiencing a revival; however, the main way in which the indissolubility of parenthood is being expressed is in terms of the law of parenting after separation.

Reforms began in a relatively mild and largely semantic way, with the shift in the USA in particular from the notion of sole custody to joint legal custody in the early 1980s (Schepard, 1985).

In Europe, the law reform process took a different form. Rather than making joint custody (in the sense of joint legal responsibility) an option, or even establishing a presumption in favour of this, European countries made joint parental responsibility the default position in the absence of a court order to the contrary.

In England and Wales, for example, a radical reconceptualisation of post-separation parent­ing occurred in 1989. The Children Act 1989 provided that each parent has "parental responsibility" and retains that responsibility after the marriage breakdown. Instead of making a custody order that gives to one parent, to the exclusion of the other, a bundle of rights and powers to make decisions about the welfare of the child, the new law provided that court orders should focus on the practical issues. Where will the child live? What contact arrangements need to be put in place? These orders are known as residence and contact orders. They say nothing about parental responsibility; that is, they do not carry with them a bundle of parental powers and responsibilities to the exclusion of the other parent, except to the practical extent required in the terms of the order. The philosophy of the Children Act 1989 is that parental responsibility continues after separation as it existed before the relationship breakdown, subject to any orders to the contrary by the court (Smart, 1997).

Similar developments also occurred in France, where the law is based upon a principle of coparentalité (Fulchiron, 2002; Vauvillé, 2002).By legislation passed in 1993 (Loi 93-22), the Civil Code was amended to remove the language of "custody". It was replaced with the language of "parental authority". The legislation provided that parental authority was to be exercised in common and that parental separation did not change this.

In many other jurisdictions, the law has also been amended to encourage or provide for continuing joint parental responsibility after divorce. This was how joint custody became the norm in Scandinavia (Parkinson, 2011). A similar approach was adopted in Germany, which amended its Civil Code to provide that parents have joint parental responsibility during marriage and unmarried parents may agree to joint parental responsibility by formal declaration. As in other European countries, this joint responsibility continues after separation unless the court orders otherwise.

In all these jurisdictions, the effect of the legislative reforms has been that legal divorce ends relationships as spouses but not as parents. Indeed, it is now irrelevant in most jurisdictions whether the parents had been married at all. Biological parenthood, rather than marriage, is what gives rise to enduring rights and obligations.

Whether or not parenthood is, in practice, indissoluble for primary caregivers (pre­dominantly women) under these statutes depends to a great extent on the attitude of the non-resident parent. If a non-resident father desires to remain closely involved with his children, then the modern ideas on post-separation parenting give him much leverage.

The consequence of this major shift in the focus of family law is that the promise of freedom to begin afresh that was held out as the meaning of divorce in the divorce reform movements of the late 1960s and 1970s has proved to be somewhat empty where children are involved.

Encouraging the involvement of both parents

The demise of the concept of sole custody was, however, only the beginning of the transition that has occurred in the law of parenting after separation. Increasingly, legislation around the Western world is emphasising the importance of both parents being involved in children's lives. Whereas previously there had been a choice between the mother and the father as the custodial parent, now a spectrum of choices is on offer to the courts. In most cases, there will still be a primary custodian, a parent with whom the child lives for the majority of the time. However, the significance of that allocation to one parent or the other is not as great as it once was. The question has changed from being about which parent the child will live with to being about how the child's time will be shared between the parents.

In most jurisdictions, legislatures have resisted the temptation to be too prescriptive about what time allocation between the parents will promote meaningful involvement. Courts have retained the flexibility to try to discern what will be in the best interests of the child in each case. Nonetheless, a common thread in legislation across the Western world has been towards the encouragement of shared parenting after divorce. A number of jurisdictions now have legislation that gives some encouragement to considering shared parenting arrangements, and the trend in terms of law reform is strongly in that direction, in situations where there are no issues of violence or abuse.

France offers one example. The principle of coparentalité, established in 1993, was strengthened by legislation enacted in 2002. Article 373-2-9 of the Civil Code now provides that the residence of a child may be fixed alternately at the domicile of each of the parents or at the domicile of one of them. The listing of alternating residence first, before sole residence, was intended to indicate encouragement of this option (Fulchiron, 2002).

In Belgium, the law of 18 July 2006 provides encouragement for alternating residence - indeed that emphasis was expressed in the title of the legislation ("Loi tendant à privilégier l'hébergement égalitaire de l'enfant dont les parents sont séparés et réglementant l'exécution forcée en matière d'hébergement d'enfant.").¹ This law provides that when parents are in dispute about residency, the court is required to examine "as a matter of priority", the possibility of ordering equal residency if one of the parents requests it to do so. If the court considers that equal residency is not the most appropriate arrangement, it may decide to order unequal residency. An equal time arrangement is not presumed to be in the best interests of the child; nonetheless it is the first option that ought to be considered when parents cannot agree on the arrangements, and this has led to a significant increase in shared care arrangements in that country (Sodermans, Matthijs, & Swicegood, 2013).

The message of such legislative directions in these different jurisdictions is clear. Contact, visitation or access - howsoever it is described - is no longer the order a parent receives as a consolation if he or she loses the prize of custody. Nor is it to be the right only of a visitor, as the language of "visitation" might suggest. Rather, the assumption is that the time that the non-resident parent has with the child will be such as to allow him or her a meaningful, continuing involvement in the life of the child. Fathers, in particular, are no longer to be marginalised by post-separation parenting arrangements.

The Australian reforms in comparative perspective

Although hotly debated at the time and subsequently, the reforms to the Family Law Act 1975 in Australia, first in 1995 and then subsequently in 2006 and 2011, are broadly consistent with these international trends. As in other countries, all parents have parental responsibility under Australian law, irrespective of whether they have ever married or lived together. As a result of the reforms in 1995, parental responsibility is deemed to continue after relationship breakdown, subject to the effect of any court order to the contrary (Family Law Act 1975, s 61C).

The position evolved further with the 2006 amendments to the Act. The 2006 amendments created a presumption of equal shared parental responsibility, but the presumption does not apply if there are reasonable grounds to believe that a parent of the child (or a person who lives with a parent of the child) has engaged in abuse of the child or family violence (Family Law Act 1975, s 61DA).

The 1995 legislation also changed the language of parenting orders in a way that is consistent with developments elsewhere. Like the Children Act 1989 in England and Wales, the court could make orders about residence and contact, as well as specific issue orders. In 2006, the language was changed again. Courts now may make orders concerning where children will live and how much time the other parent will spend with them. The resulting terminology is less pithy than "residence" and "contact"; the price of greater sensitivity in the language has been greater prolixity.

There was a lot of resistance to the 1995 reforms at the time they were made (Armstrong, 2001), and controversy continued afterward (Graycar, 2000), but these reforms were quite unremarkable in international terms. They also represented only a modest evolution from the pre-existing law (Chisholm, 1996).

The 2006 legislation in Australia also reflected international trends towards encouraging the involvement of both parents in children's lives after separation. One of the objectives of the Family Law Act, as amended by the Family Law Amendment (Shared Parental Responsibility) Act 2006, is to ensure that "children have the benefit of both of their parents having a meaningful involvement in their lives, to the maximum extent consistent with the best interests of the child" (s 60B(1)(a)). This is, importantly, balanced by another object of the legislation, the need to protect children from physical or psychological harm from being subjected to, or exposed to, abuse, neglect or family violence which may necessitate restraints on contact by one parent (s 60B(1)(b)). These objects are translated into primary considerations in determining what is in the best interests of the child. Further amendments in 2011 have clarified the prioritisation to be made between these two considerations.

Although the effect of the 2006 reforms was not always understood by the general public, there is no presumption of shared parenting in Australian law, and still less, equal time. The most that the legislation imposes by way of presumed outcome is a presumption in favour of equal shared parental responsibility in the absence of violence or abuse. However, if parental responsibility is to continue to be shared, then there is at least strong encouragement in the legislation to consider shared parenting, and to do so positively (Goode & Goode (2006) FLC 93-286).

The 2011 amendments to the Act modify this emphasis only a little. The requirement to consider equal time and substantial and significant time remains, but in the evaluation of what arrangements are in the best interests of the child, greater weight is to be given to the need to protect children from harm than to the benefit to the child of having a meaningful relationship with both parents.

Resistance to reform

This transformation in the law of parenting after separation around the Western world is all the more remarkable because it has not occurred without serious resistance. In the main, that opposition has come from women's groups and feminist advocates for whom the sole custody model represented an optimal post-separation parenting arrangement.

Over the years, various arguments have been made against laws that encourage joint custody and shared parenting time (Boyd, 2003; Cohen & Gershbain, 2001; Smart & Sevenhuijsen, 1989). One of the main arguments has been that the more that legislation supports and encourages the involvement of non-resident parents, the more it exposes women to the risk of violence and abuse (Jaffe & Crooks, 2004). The problem of domestic violence has thus taken centre stage in campaigns against changes to the law that promote joint custody and greater contact between non-resident parents and children.

However, the issue of protecting women and children from violence has not proved effective as an argument against having any provisions in legislation that encourage non-resident parent involvement. That has been a source of frustration for some advocates (Graycar, 2012), but there are reasons why these arguments have not achieved much traction.

One reason is that the evidence base for the supposed connection between laws that encourage the involvement of non-resident parents in their children's lives, and an increased risk of violence, is very weak. There is simply no evidence for a linear relationship between the time that non-resident parents spend with their children, and a greater incidence of post-separation violence towards the primary caregiver. Another reason is that politicians have responded to concerns about violence and abuse - logically enough - by strengthening the provisions in the legislation addressing those issues. That happened with the Family Law Reform Act 1995 in Australia. Alongside various changes to the law that placed an emphasis on the importance of both parents in children's lives, the Parliament enacted a substantial number of provisions concerning family violence. No such provisions had been in place prior to 1995.

A similar legislative approach was adopted in 2006. The protection of children from harm was made one of the objects of Part VII of the Act dealing with children, and one of two primary considerations in determining the best interests of the child.

This approach was adopted again in 2011, with the government further strengthening those parts of the legislation that emphasise the need to protect children from harm, and deleting a relatively small number of provisions that had caused anxiety, while leaving the substance of the 2006 reforms intact. Specifically, the government resisted the pressure to make fundamental changes to the shared parenting emphasis of the 2006 legislation.

When parenthood should be dissoluble

Recognition of the notion that families endure beyond the separation of the parents does not necessarily involve an assumption that all families can or should endure. Nor does it mean that the goal of interventions in all cases ought to be to try to build a cooperative co-parenting relationship.

Protection from family violence

As is now well understood, many women and children are at risk from male violence and abuse before, during and after separation. It is appropriate that an absolute priority be given to the safety of victims of violence and their children when there is a risk of serious harm.

How can the protection of victims of violence be given an absolute priority when the law in general promotes the indissolubility of parenthood? The new definition of family violence introduced by the 2011 amendments offers the potential for an improved level of differentiation between types of violence, and therefore a more nuanced understanding of the dynamics of the violence that has occurred within a particular family (Parkinson, 2012). However, there still remains a need for greater clarity in the law about when family violence ought to lead to orders for sole parental responsibility and restrictions on contact.

Current safety concerns

One issue with the Australian legislation is that in so many places it focuses on a history of family violence or abuse at any time in the course of the relationship, rather than on current safety concerns, and without sufficient clarity about how that history needs to be taken into account in decision-making. For example, any incident that can be characterised as family violence - however far in the past - is sufficient to rebut the presumption of equal shared parental responsibility (s 61DA); any history of family violence is sufficient to provide grounds for exemption from the requirement to produce an s 60I certificate before filing an application for parenting orders in court;² any history of family violence triggers a requirement on the court to take certain steps promptly,³ irrespective of whether there are any current safety concerns (s 67ZBB), as long as the person alleging the violence sees it as an issue for the court's decision-making. The legislation also requires consideration of family violence orders that are no longer current in determining what is in the best interests of the child, as a factor separate to a history of family violence (s 60CC(3)).

This demonstrates that Australian law takes the issue of domestic violence very seriously - and so it should. However, it also casts the net very wide, given the prevalence of histories of family violence among parents who have separated (Kaspiew et al., 2009), and the exponential growth in the number of family violence orders made in certain parts of the country over the last ten years (Parkinson, Cashmore, & Single, 2011). The Australian Institute of Family Studies (AIFS), in its evaluation of the 2006 family law reforms, found that 26% of mothers and 17% of fathers reported being physically hurt by their partners. A further 39% of mothers and 36% of fathers reported emotional abuse (Kaspiew et al., 2009, p. 26). The sheer volume of cases where there are allegations of violence or abuse in parenting disputes (Moloney et al., 2007), and/or where there is a previous family violence order, makes it difficult to deal with all relevant cases promptly (as required by s 67ZBB). When everything is urgent, nothing is urgent. A focus on current safety concerns rather than a history of violence during the course of the relationship per se, is important to allow the concentration of resources on the parents and children who are at most risk as a result of post-separation parenting arrangements.

The AIFS evaluation found - in interviews with about 10,000 parents, conducted on average fifteen months after separation - that a smaller number of parents had current safety concerns either for themselves or their children than had reported a history of violence or emotional abuse. Four per cent of fathers and 12% of mothers were concerned about their personal safety; and 15% of fathers and 18% of mothers expressed concerns about the safety of their child - either alone or in addition to concerns about personal safety (Kaspiew et al., 2009, p. 28).

The AIFS team also found that a history of family violence did not necessarily impede friendly or cooperative relationships between the parents. Sixteen per cent of mothers who reported being physically hurt by their ex-partner during the course of the relationship reported friendly relationships at the time of the interview, and a further 24% reported having a cooperative relationship. While others reported distant or conflictual relationships, only 19% reported a continuing fearful relationship. Fifty-five per cent of mothers and 50% of fathers who reported emotional abuse by their ex-partner during the course of the relationship reported friendly or cooperative relationships by the time of interview (Kaspiew et al., 2009, pp. 31-32). By way of contrast, where a parent had current safety concerns either for themselves or for their child, it was much more likely that they would report difficult relationships with the other parent (Kaspiew et al., 2009, pp. 32-33).

Parents who had concerns about the safety of their children reported that their children had a significantly lower level of wellbeing than the children of parents who did not have such concerns, while a history of family violence was no longer statistically significant in terms of child wellbeing according to mothers' reports once socio-demographic characteristics and family dynamics were controlled for (Kaspiew et al., 2009, p. 269).

Section 61 of the Care of Children Act 2004 in New Zealand offers a clearer focus to the inquiry regarding a history of violence because it focuses on current safety concerns and levels of risk. In New Zealand, the question that has to be asked is "whether a child will be safe if a violent party provides day-to-day care for, or has contact (other than supervised contact) with, the child". Various considerations are listed to assist the court in assessing that question.

Although the Family Law Act is not so well focused, the 2011 amendments to the Act, if properly understood, may well assist in making clearer how family violence needs to be taken into account in parenting disputes.

The two safety priorities

The 2011 amendments, together with reforms made at earlier times, identify two priorities in cases where safety is an issue.

First, in assessing what is in the best interests of the child, greater weight is to be given to the need to protect the child from physical or psychological harm due to being subjected to, or exposed to, abuse, neglect or family violence than to the benefit to the child of having a meaningful relationship with both of their parents. The first priority, then, is child safety if there are current concerns that the child may be at risk.

Secondly, there is s 60CG (first introduced in 1995), which provides:

In considering what order to make, the court must, to the extent that it is possible to do so consistently with the child's best interests being the paramount consideration, ensure that the order … does not expose a person to an unacceptable risk of family violence.

The second priority, then, is parental safety. Where there is a present risk of violence towards the primary caregiver, but the child will nonetheless benefit from spending time with the non-resident parent, measures need to be put in place as far as possible to ensure that the parents do not meet, or meet only in a public place where the risk of violence is lessened. The use of contact centres to facilitate handovers is one way in which this can occur.

Focusing on the two priorities for protection will, it is to be hoped, provide adequate guidance on when a sole parental responsibility order, together with restrictions on contact, would be appropriate. However, the importance of these two priorities might be missed in the way the legislation is currently drafted because they are placed in different sections of the Act and the court is required to consider so many other aspects of a history of violence in an unfocused way.

Of course a history of family violence remains important, even where there are no current safety concerns. The problem is that the Act gives little clarity about why a history of violence is important and how it should be weighed against other factors. A history of violence is, for example, an important issue to explore in terms of the children's attitudes towards living with, or going on visits to, a violent parent. A child's fear of the violent parent, or concern about the parent's unpredictability, are relevant matters to examine, as are the ways in which witnessing the violence has affected the children's love for, and trust in, the parent (Holt, Buckley, & Whelan, 2008; Wolfe, Crooks, Lee, McIntyre-Smith, & Jaffe, 2003).

A history of coercive controlling violence is particularly relevant in determining parenting arrangements after separation. It is important, for example, in assessing the mother's capacity for parenting and her attitude towards contact between the child and the other parent. For many women who experience this kind of subjugation and control, the psychological effects may have a greater lasting impact than the physical abuse. These effects include fear and anxiety, loss of self-esteem, depression and post-traumatic stress (Kelly & Johnson, 2008, pp. 483-484). They may significantly affect a mother's capacity to parent (Erikson, 2005), particularly in the context of coping with the stresses of the relationship breakup and the litigation about parenting arrangements. Mothers may be misdiagnosed as suffering from various psychopathologies, even though their deficiencies and problems are situational and reactive to the experience of abuse.

The experience of coercive controlling violence may also explain a parent's resistance to regular contact between the children and the father, even if it can be made safe through contact handovers, or her desire to relocate a long way from the other parent when there is not another convincing rationale for the move other than to get away.

In particular, coercive, controlling violence against an intimate partner is a window to the soul. It reveals much about the character of a person. It may be indicative of a tendency to dominate and control the children rather than to nurture and empower them (Kelly & Johnson, 2008). There is also a likelihood of there being ongoing issues about the safety of the mother and high levels of conflict between the parents.

It is also important when considering the history of violence, that optimism should not be allowed to triumph over experience. Because there is such a reluctance to sever face-to-face contact between a parent and a child, the use of contact centres, where available, is often an attractive compromise position. Contact centres allow for supervised handovers of children in order to avoid the parents meeting, and may provide supervised contact in cases where there is concern about abuse of a child.

Nevertheless, in contact centres, there can be a conflict between an institutional imperative to help the parents to "self-manage" to the extent that they no longer need the services of the centre, and the need for ongoing protection from violence or abuse. Services that have high levels of demand will want to move people off their books in order to place others on them (Sheehan, Dewar, & Carson, 2007). In some cases, therapeutic work with parents may be helpful where, by improving levels of cooperation and trust between the parents, the primary carer can build enough trust and confidence in the other parent that she feels safe to move beyond the security of using the contact handover service.

Where, however, the reason for the use of the centre is because of ongoing concerns about safety, the notion that the parents can be assisted towards a healthy enough co-parental relationship is, for the most part, likely to be unrealistic (Harrison, 2008; Parker, Rogers, Collins, & Edleson, 2008). Services should provide life support to a parent-child relationship only for a relatively limited period. After that, if serious safety issues have not been and cannot be resolved, then the hard decisions need to be taken, with the priority being the safety and the wellbeing of the primary caregiver.

Intractable conflict

Orders for sole parental responsibility are also appropriate in situations of intractable conflict. This was recognised by the House of Representatives Family and Community Affairs Committee (2003) in its landmark report, which formed the basis for the 2006 reforms. Recommendation 2 in the Committee's report was that "Part VII of the Family Law Act 1975 be amended to create a clear presumption against shared parental responsibility with respect to cases where there is entrenched conflict, family violence, substance abuse or established child abuse, including sexual abuse" (p. 41).

It proved difficult to translate that into legislation, not least because of the problem of providing a legislative definition of entrenched conflict. However, the recommendation should not be forgotten. While high levels of conflict are particularly problematic in shared care arrangements, given the degree of interaction between parents that is typically needed, the issues about entrenched conflict do not arise only in relation to shared care. In any situation where the track record suggests that it will be very difficult indeed for the parents to agree on aspects of parental decision-making, it is likely to be better to make an order for sole parental responsibility.

Nonetheless, it is important to seek to understand the reasons for entrenched conflict. In some families, hostility arising from unresolved anger about the separation or issues concerning new partners (Smart & May, 2004) might be reduced with therapeutic support. There are also cases that are high conflict for a reason, not least that significant child protection concerns have not been resolved (Birnbaum & Bala, 2010; Cashmore & Parkinson, 2011). There are others where the issues arise from alienation of the child.

In one sense, any litigated case is one of high conflict, but the conflict may well diminish once a decision has been made. Cases where there is a lot of re-litigation suggest that the conflict may be entrenched, and that more radical surgery is needed than simply restructuring or adjusting the parenting arrangements.

When the parents have never been a family

One of the most difficult issues to address is when the parents who are in dispute about their child have never actually lived together. In such cases, typically the child has been conceived in the context of a relatively short-term relationship.

In Australia, about 11-12% of all births are to single mothers, according to data from 2000-01 (De Vaus & Gray, 2004). Since 1975, the rate at which children are born into lone-mother families has increased more rapidly than the rate at which children experience parental separation (De Vaus & Gray, 2004). Judges and practitioners report from their experience that many disputes in the family courts in Australia concern infant children of parents who have never lived together.

The question arises, then, whether it is realistic to expect parents to develop a cooperative joint parenting relationship when they have never known what it is to live together and raise the child as a couple. Is it to be presumed that children will benefit from a relationship with a parent whom they do not know through the intimacy of the daily child care tasks that occur naturally in most intact families?

This is an area where Australian law fails to make adequate distinctions. In 1995, principles were introduced that "children have the right to know and be cared for by both their parents, regardless of whether their parents are married, separated, have never married or have never lived together", and that they "have a right to spend time on a regular basis with both their parents" (Family Law Act 1975 s 60B(2)). These principles apply except when it would be contrary to a child's best interests.

Cases where the parents have never lived together are quite varied. Biological fathers of children born to single mothers may in many cases represent potentially important social capital to children if a relationship can be established and maintained, but the sociological evidence - at least from the US - would appear to indicate that the long-term prognosis for relationships between many of these fathers and their children is one of increasingly tenuous connection (Carlson, McLanahan, & Brooks-Gunn, 2008; Cheadle, Amato, & King, 2010). Such fathers ought to pay child support, of course, however tenuous their connection with the child may be. The question is whether any differentiation should be made in other respects (as a matter of law) between fathers in this situation and those who have been involved in caring for the child in the context of a marriage or a domestic partnership, at least for a period of time.

When parents have never lived together, arguably a legal presumption of equal shared parental responsibility is inappropriate. There have to be limits on the extent to which those who have never formed families as two biological parents should be treated as if they had done. This is particularly the case where the relationship has been characterised by conflict from the beginning, or near the beginning, of the child's life.

Avoiding a presumption of equal shared parental responsibility in cases where there are disputes before the courts, leaves the position as a neutral one. The parents may agree to share equal parental responsibility or the judge may order that it be so; but the law should not presume a particular outcome.

Conclusion

Cases where there is a risk of serious harm to the child or a primary caregiver, relationships that have intractable conflict, or where the child has begun life with a single mother, are all situations where the old sole custody norm is likely to be more appropriate than trying to preserve an ongoing co-parenting relationship. These are not the only situations of course, and other cases will depend on their facts.

Sometimes, perhaps, family law systems around the world try too hard to keep alive relationships that aren't sufficiently healthy to survive without intensive care. By no means all father-child relationships survive parental separation, nor should they, and family law systems need to come to terms with that. As the poet Arthur Clough (1974) once wrote:

Thou shalt not kill;
but need'st not strive
officiously to keep alive.

Accepting both the indissolubility of parenthood for most separated parents and appropriate limitations on joint parental responsibility offers the best way forward for consensus, and for an appropriately balanced family law system. Researchers and policy-makers need to move beyond the gender wars, even if those in the middle of post-separation conflict are stuck in the trenches.