The Full Court of the Family Court of Australia, by majority, upheld a decision of the primary judge, Steele J, who made a return order pursuant to the Family Law (Child Abduction Convention) Regulations 1986 (Cth) ("the Regulations"). Of particular relevance are regs 4 and 16, which are set out in the reasons of Gummow, Heydon and Crennan JJ. The basis of the Family Court's order was a finding that a child of the appellant had been wrongfully removed by the appellant from New Zealand to Australia. That finding, in turn, rested upon a conclusion that the father of the child, a resident of New Zealand, had rights of custody in relation to the child under the law of New Zealand (reg 16(1A)(c)). The existence of those claimed rights of custody is the point on which the Full Court divided. The child was born in New Zealand in 1996. At that time, and until September 2006, the appellant and the child were residents of New Zealand.
The difference between the majority (May and Thackray JJ) and the dissentient (Finn J) in the Full Court turned upon the question whether there was evidence to support Steele J's finding of fact that the appellant was "living with the father of the child as a de facto partner at the time the child was born." It is common ground that, by reason of s 17 of the Care of Children Act 2004 (NZ), if the answer to that question is in the affirmative, the appellant and the father are joint guardians of the child. In that event, the father had, and has, rights of custody within the meaning of reg 16(1A)(c).
A striking, and disconcerting, feature of the case is the absence of factual detail, on an issue that is now presented as potentially decisive, in the evidentiary material before the Family Court. Part of the explanation appears to be that, until shortly before the hearing at first instance, the issue did not emerge as a matter of serious contest. The case was dealt with as one of urgency. The application for a return order was filed on 11 October 2006. The response and the supporting affidavit of the appellant were filed on 14 November 2006. The hearing took place on Monday 18 December 2006. An ex tempore judgment was delivered. Most of the affidavit evidence was directed to matters which are irrelevant to this appeal. When, from an affidavit sworn by the appellant on the Thursday before the hearing, it became apparent that there was to be a serious dispute about the relationship between the father and the appellant at the time of the birth of the child, the evidence of the father (who was not a party to the proceedings), and other witnesses for the respondent, was not supplemented. One example of the deficiency of the material illustrates the point. The appellant was born in September 1977. The father was born in 1964. They met in October 1995. The child was born in September 1996. There was some sketchy evidence on the topic of their living arrangements at and immediately after the birth of the child. There was, however, an almost complete absence of evidence about their living arrangements during the months leading up to the birth of the child. The Family Court was left to rely upon inferences from subsequent conduct, and upon one or two broad generalisations, in order to reach a conclusion about a question of fact the answer to which was well known by both the father and the appellant. If there is one thing in the case that is certain, it is that the evidence before the Family Court did not reveal, or even attempt to reveal, the full history of the relationship in question. Since the father was not a party to the litigation, he was in a position of disadvantage. The appellant gave some evidence, not in all respects consistent, about where she was living in the weeks following the birth of the child. It is surprising that no one thought to adduce evidence about where she was living during her pregnancy.
The relationship between father and mother at the time the child was born
Before the removal of the child to Australia, in September 2006, there had been a history of disputation and litigation, between the father and the appellant, concerning the father's rights of access to the child, and alleged interference with those rights. Steele J found that, shortly after the birth of the child, the parties separated. In November 1997, a parenting agreement was reached which provided that the child would live with the appellant and spend some time with the father. In 1997, the appellant married a man described by Steele J as "a notorious criminal". Steele J said that the father "was attempting, somewhat valiantly, to build a relationship with the [c]hild". The appellant's marriage to the criminal broke down, and the husband disappeared. In 1997, following counselling, there was a counsellor's report which recorded the agreement of the appellant and the father to court orders giving custody to the appellant and rights of access to the father. The counsellor's report said that the appellant agreed to "recognise [the father's] guardianship rights." The appellant, in her evidence, denied the accuracy of that part of the report. Nevertheless, the report was in evidence before Steele J, who declined to accept the appellant's evidence on a number of matters, but made no specific finding about the report. Court orders relating to custody and access were made in 2000. There were later court proceedings over the years between 2000 and 2006.
The application initiating the present proceedings asserted, among other things, that, pursuant to s 17 of the Care of Children Act 2004, the father was a guardian of the child "as he was living with the child's mother when the child was born." Plainly, this was intended to be an elliptical assertion that the two were living as de facto partners. So much appears from the reference to s 17, which provides that the father and mother of a child are joint guardians unless the mother is the sole guardian, and further provides that, in the case of a child conceived at the relevant time, where the parents were not married or in a civil union, the mother is the sole guardian if she was not living with the father as a de facto partner at the time the child was born. By reason of s 29A of the Interpretation Act 1999 (NZ), that turned upon whether the two lived together as a couple in a relationship in the nature of marriage or civil union. A supporting affidavit of the father asserted, without elaboration, that he was the child's joint guardian by virtue of the fact that he lived with the appellant at the time of the child's birth. An affidavit sworn by the father's New Zealand lawyer repeated the same assertion, expressly relating it to s 17. The appellant's first affidavit in response, sworn on 10 November 2006, did not contradict those assertions. It alleged physical abuse, which, according to the appellant, began during her pregnancy. "This", she said, "is why I moved out with my 2-month old son and went to live with my parents." Her statement that she "moved out" is to be understood in the light of the assertions to which she was responding, and appears to confirm at least the fact of previous cohabitation.
That was the state of the evidence until 14 December 2006. It is difficult to accept that, up to that stage, the matter of the relationship of the appellant and the father at the time of the birth of the child was regarded by the parties to the litigation as a serious issue. If it had been so regarded, it was treated by both sides in a remarkably casual fashion. It is clear that the respondent alleged that the father and the appellant were living in a de facto relationship, as that term was understood in New Zealand law, at the time of the birth of the child, but the evidence tendered in support of that allegation was brief, formal, and, until shortly before the hearing, apparently not only uncontested, but consistent with the evidence of the appellant, as far as that evidence went.
In her affidavit of 14 December 2006, the appellant raised an issue about the factual basis of the father's claim to joint guardianship, and therefore custody rights. This confronted the respondent with a tactical dilemma: to seek an adjournment of the proceedings in order to obtain further evidence from the father, or to conduct the case on the existing, manifestly incomplete, evidence. The further evidence of the appellant said that since the child was born she had lived with her parents, and denied that she lived with the father when the child was born. The appellant also said that, about one and a half months after the child's birth, she stayed with the father for three nights a week for around six weeks to see if he was capable of being a father to the child. The appellant said: "During that time, I did want his home to be mine on the condition that he [was] capable of assisting in the care of [the child]." She denied that she ever lived as the father's de facto partner, and that she ever regarded herself as such. As has been noted, some of the appellant's evidence on certain matters, including cohabitation, was rejected. This piece of evidence addressed a mixed question of fact and law. To the extent to which it was bound up with the question of cohabitation at the time of the birth of the child, it was not accepted.
The primary judge preferred the evidence of the father to that of the appellant. Specifically, he rejected the appellant's evidence that she was not living with the father when the child was born. This finding of fact was re-examined by all three members of the Full Court. The majority said they agreed with it. The dissentient, Finn J, said the finding was open. She did not express disagreement with it.
The point of departure between Finn J and the majority concerned a further matter that the respondent had to prove. It has already been noted that under the New Zealand legislation defining de facto relationship, it was not enough to establish that the appellant and the father lived together; it was necessary to establish that they lived together as a couple in a relationship in the nature of marriage or civil union. Finn J held that there was no evidence from which it could be inferred that the father and the appellant had lived together as a couple in a relationship in the nature of marriage or civil union.
Finn J was correct to stress the difference between living together and living together "as a couple in a relationship in the nature of marriage or civil union". The relationship between two people who live together, even though it is a sexual relationship, may, or may not, be a relationship in the nature of marriage or civil union. One consequence of relationships of the former kind becoming commonplace is that it may now be more difficult, rather than easier, to infer that they have the nature of marriage or civil union, at least where the care and upbringing of children are not involved. (As will appear, the qualification is significant in the present case.)
When divorce, for various reasons, was more difficult, in former times, de facto relationships often existed because there was an impediment to legal marriage. A common impediment was a subsisting marriage of one of the parties. Marriage, in Australia and New Zealand, involves legal requirements of formality, publicity and exclusivity. A person may be a party to only one marriage at a time. de facto relationships, on the other hand, do not involve these elements. They are entered into, and may be dissolved, informally. In Australia, marriages are required to be entered on a public register. In New Zealand, marriages and civil unions must be registered. Parties to marriages and civil unions do not have a choice as to whether, when, and by what means they will disclose their status to the public. It goes without saying that there is no mandatory public registration of sexual relationships, even if they involve cohabitation. de facto relationships may co-exist with the marriage of one or both parties and, at least in some circumstances, people may be parties to multiple de facto relationships. Yet the law to be applied in this case acknowledges that some are, and some are not, in the nature of marriage. How is the difference to be determined? No single and comprehensive answer to that question can be given, but there is one test that is applicable to the present case.
In Stack v Dowden  2 AC 432 at 450-451 , Baroness Hale of Richmond said:
Cohabitation comes in many different shapes and sizes. People embarking on their first serious relationship more commonly cohabit than marry. Many of these relationships may be quite short-lived and childless. But most people these days cohabit before marriage .... So many couples are cohabiting with a view to marriage at some later date - as long ago as 1998 the British Household Panel Survey found that 75% of current cohabitants expected to marry, although only a third had firm plans: John Ermisch, Personal Relationships and Marriage Expectations (2000) Working Papers of the Institute of Social and Economic Research: Paper 2000-27. Cohabitation is much more likely to end in separation than is marriage, and cohabitations which end in separation tend to last for a shorter time than marriages which end in divorce. But increasing numbers of couples cohabit for long periods without marrying and their reasons for doing so vary from conscious rejection of marriage as a legal institution to regarding themselves 'as good as married' anyway: Law Commission, Consultation Paper No 179, Part 2, para 2.45.
There is no reason to doubt that the same is generally true of Australia and New Zealand. It may be added that, in Australia, what often prompts cohabiting couples to marry is a decision to have a child, and to do so within the context of a marriage. People often refer to this as "starting a family". The cohabiting parties to many relationships, especially first relationships of the "short-lived and childless" kind, may be surprised to be told that they are involved in a relationship in the nature of marriage or civil union. They may intend no such thing. The same may apply to some people in longer-term cohabitation who have chosen not to marry. It is the common intention of the parties as to what their relationship is to be, and to involve, and as to their respective roles and responsibilities, that primarily determines the nature of that relationship. The intention need not be formed in terms of legal status: to some people that is important; to others it is a matter of indifference. (By hypothesis, the parties to a relationship that satisfies the statutory description are not married, or in a civil union.) The intention may be expressed, or it may be implied. What is relevant is their intention as to matters that are characteristic of a marriage or a civil union, but that do not depend upon the formal legal status thus acquired. To describe a relationship as being in the nature of marriage implies a view about the nature of marriage. The same applies to a civil union. It is unnecessary, for present purposes, to attempt a comprehensive account of the features of a relationship that might justify such a description. Plainly, "living together" is not enough. For present purposes it is sufficient to focus upon that aspect of the relationship between the appellant and the father that gives rise to this dispute, that is to say, shared parenthood, and upon the inferences as to intention that may be drawn from that.
In Magill v Magill (2006) 226 CLR 551 at 564 , and earlier in Russell v Russell (1976) 134 CLR 495 at 548-549, reference was made to the historical role of the institution of marriage as a means of involving males in the nurture and protection of their offspring, and to the importance of the structure of marriage and the family in sustaining responsibility for, and obligations towards, children. There is a wide range of human behaviour across the spectrum between a sexual encounter and a marriage or civil union. It includes relationships which could never be described as being in the nature of marriage or civil union. Nevertheless, when a sexual union results in the birth of a child, cohabitation between the parties to the union is no longer a matter of purely personal convenience or satisfaction. The interests of a third party have intervened. Traditional concepts of marriage and the family as institutions for the protection of children, and modern concepts of shared parental responsibilities even in the absence of a formal union, may come into play in characterising the relationship. The present case provides an example.
In her December affidavit, the appellant said that after the birth of the child, for about six weeks, she stayed for three nights a week with the father to see if he was capable of being a father to the child, and that she wanted his home to be her home on condition that he was capable of assisting in the care of the child. I would infer from the fact that (contrary to her evidence) she was cohabiting with the father at the time of the birth of the child, and from her silence on the subject of her living arrangements during her pregnancy, that, before and at the time of the birth of the child, her intention was that, if possible, the father, the appellant, and the child would live together as members of a family unit. I would also infer, from his general assertions in his affidavit, and from his later conduct in vigorously asserting his claims to guardianship of the child, that the father had the same intention. These inferences receive some support from the names given to the child at birth, which reflected the father's cultural heritage.
Although other factors also may be important in deciding whether the relationship between a cohabiting couple is in the nature of marriage or civil union, where, in a given case, their union has resulted in pregnancy, and the couple are living together with the intention that, when the child is born, they and the child will form a family unit, with the parents sharing responsibility for the nurture of the child, there is at least a sound basis for characterising their relationship as having such a nature.
In the particular context of this case, and in the general context of applications for return orders under the Child Abduction Convention, the relationship is not merely one between two adults. It involves a child or children. The majority in the Full Court of the Family Court pointed out that s 29A of the Interpretation Act 1999 of New Zealand directs a court to have regard to context and purpose in interpreting a statutory expression such as "relationship in the nature of marriage or civil union". They also referred to the general purposes of the Care of Children Act 2004 of New Zealand. Both the general and the particular context are relevant to the question to be decided. If there had been no child of the union, and if all that was shown had been the fact of cohabitation of the appellant and the man in question, I would have agreed with Finn J that there was no evidence to support a finding that their relationship was in the nature of marriage or civil union. On that basis, they were simply two adults (one, the appellant, a rather young adult) cohabiting for a fairly brief time in a sexual relationship. If that were all there was to it (and, of course, in many cases of cohabitation there is a great deal more), I would not attribute to such a relationship the statutory description with which we are concerned. To do so would be to mock the institution of marriage. There was more. There was a child, or an expected child, and this was a vital part of the setting in which their relationship at the time of the birth of the child was to be characterised.
While I am not quite sure what they meant by their reference to a "relatively low threshold" of proof, I agree with the conclusion of the majority in the Full Court that the evidence, unsatisfactory as it was, sufficed to establish that the couple were living together in a de facto relationship, and not merely living together, at the time of the birth of the child. In consequence, the father is a joint guardian of the child, and therefore has rights of custody.
The majority in the Full Court decided the case adversely to the appellant on the issue considered above. May and Thackray JJ said:
Since we have found that [the child's] parents were living in a de facto relationship at the time of his birth, they are both his guardians and both have the right to determine his place of residence. The father therefore had 'rights of custody' at the time of his removal from New Zealand.
The appellant's grounds of appeal were all aimed at that conclusion.
Against the possibility that this Court might not accept the conclusion of the Full Court, the respondent, by notice of contention, advanced arguments in support of alternative grounds for deciding that the child's removal was wrongful. These arguments were based upon the father's rights of access under a court order, and the "rights" of the New Zealand courts concerning the removal of the child. Because I would reject the appellant's grounds of appeal, for the reasons given above, it is unnecessary to deal with the respondent's notice of contention. If the basis on which the Full Court decided the case is accepted, the issues which the notice of contention seeks to propound do not arise. I should add, however, that I agree with what Gummow, Heydon and Crennan JJ have said on those issues, bearing in mind, in particular, the language of the Regulations which the Family Court was to apply, and the terms of the particular order upon which the respondent relied.
The appeal should be dismissed with costs.
Gummow J, Heydon J & Crennan J
The appellant was born in Poland in 1977 and is a Polish citizen. On 20 September 1996, while she was living in New Zealand, she gave birth to a son. The father was born in 1964 in New Zealand and resides in that country. He is of Maori descent. The parties have not married, and at least since the child was a very young infant they have not lived together. The child lived in New Zealand primarily with his mother but had extensive contact with his father. Relations between the parents deteriorated. Accompanied by the child, and without advising the father of her intention to do so, the appellant travelled to Australia on 15 September 2006. She and the child reside with her parents who now live in an outer suburb of Sydney; at an earlier stage her parents had lived in New Zealand.
On 11 October 2006, the respondent ("the Authority"), as State Central Authority appointed under reg 8 of the Family Law (Child Abduction Convention) Regulations 1986 (Cth) ("the Regulations"), applied to the Family Court for orders requiring the return of the child to New Zealand. The opposition by the appellant was unsuccessful and on 18 December 2006 the primary judge (Steele J) made the orders sought by the Authority.
The appellant appeals to this Court against the dismissal on 30 April 2007 by the Full Court of the Family Court of Australia (May and Thackray JJ; Finn J dissenting) of her appeal against the orders of the primary judge.
The purpose stated in reg 1A(1) of the Regulations is to give effect to s 111B of the Family Law Act 1975 (Cth) ("the Act"). Section 111B comprises Div 2 of Pt XIIIAA of the Act and is headed "International child abduction". Section 111B(1) states:
The regulations may make such provision as is necessary or convenient to enable the performance of the obligations of Australia, or to obtain for Australia any advantage or benefit, under the Convention on the Civil Aspects of International Child Abduction signed at The Hague on 25 October 1980 (the Convention) but any such regulations shall not come into operation until the day on which that Convention enters into force for Australia.
The text of the Convention is set out as Sched 1 to the Regulations. Schedule 2 lists the Convention countries and includes New Zealand along with some 70 other countries. The Convention countries have a range of legal systems, many not based in the common law. Further, the Convention contemplates in Art 11 that proceedings for the return of children may be conducted by "[t]he judicial or administrative authorities" of the Convention country in question.
In the Full Court May and Thackray JJ observed that the ideal that an international agreement or convention should bear the same meaning in all contracting states was more easily stated than attained and their Honours continued ((2007) 211 FLR 357 at 376-377):
Not only have some countries used different words from those appearing in the Convention when enacting legislation to give effect to it, but also courts of different countries have sometimes taken different approaches in interpreting key elements of the Convention.
In some Convention countries (of which the United Kingdom is one) the text of the Convention by statute is given the force of law in the domestic law of the country. That is not the case in Australia, as attested by the Regulations, and in New Zealand. Cooke P observed in Gross v Boda  1 NZLR 569 at 570 of the New Zealand legislation as enacted in 1991 that it differed in some aspects from the text of the Convention.
This Court held in De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640 that s 111B and the Regulations are laws with respect to external affairs independently of the Convention. Regulation 1A states that the purpose of the Regulations is to give effect to s 111B and that the Regulations are intended to be construed "having regard to the principles and objects" which are mentioned in the preamble to the Convention and Art 1 thereof. However, as a consequence of De L, no provision of the Regulations will be invalid merely on the ground that it goes beyond what may be reasonably capable of being considered appropriate and adapted to implement the Convention. (This is important for the consideration later in these reasons of the comparison between Art 8 of the Convention and reg 16(1A).)
The Regulations distinguish between requests for the return of children abducted to Australia and those abducted from Australia, and requests for access to a child in Australia or in another Convention country. Parts 2 and 3 (regs 11-21) deal with the former and Pt 4 (regs 23-25) with the latter. The phrase "rights of access" is defined in reg 2(1) as including:
the right to take a child for a limited period of time to a place other than the child's habitual residence.
However, as will be further explained in these reasons, the notion of abduction is linked to that of the removal or retention of a child which is "wrongful" because it is in breach of "rights of custody".
The distinction is apparent from the statement of objects in Art 1 of the Convention as being:
Chapter III of the Convention (Arts 8-20) is titled "RETURN OF CHILDREN" and Ch IV (Art 21) is titled "RIGHTS OF ACCESS". The concern of Ch III is with the return of children whereas the more modest objective of Ch IV is the protection and observance of access rights to a child in Australia or in a Convention country. The distinction drawn by the Convention is further explained by La Forest J in Thomson v Thomson  3 SCR 551 at 580. His Lordship also stressed that despite the emphasis in the Convention upon removal or retention that is "wrongful", "the Convention is not aimed at attaching blame to the parties" (at 582).
The application for return
Regulation 13 provided for the taking of action in Australia upon receipt of a request in relation to a child removed from a Convention country to Australia. Regulation 14 empowered the Authority to apply to the "court", seeking an order under Pt 3 of the Regulations (regs 14-21) for the return under the Convention of the child.
Section 39(5)(d) of the Act conferred on the Family Court jurisdiction with respect to matters arising under the Act in respect of which proceedings were instituted under the Regulations.
The proceeding in the Family Court followed receipt of a request dated 2 October 2006 by the New Zealand Central Authority. Attached to it was an authorisation by the father to the New Zealand Central Authority to act on his behalf and supporting documents including the order of 4 December 2000 referred to below. The application to the Family Court by the Authority stated that the father had rights of custody of the child. This was said to be because
he had in his favour "an access order [dated 4 December 2000] made by the Family Court at Auckland .... which provides that the child be in his father's care every second weekend" ("the Access Order") and
he was a guardian of the child, pursuant to s 17 of the Care of Children Act 2004 (NZ) ("the NZ Children Act"), by reason of the fact that "he was living with the child's mother when the child was born"; the application also stated that pursuant to the NZ Children Act the father had "the right to determine the child's place of residence".
The Full Court divided on the question whether the Authority had crossed the threshold and established its case that the father had rights of custody which were breached when the appellant removed the child to Australia; in the absence of those rights and this breach the removal would not have been wrongful and the application by the Authority to the Family Court would not have been effectively instituted under the Regulations. In this Court, the appellant seeks to uphold the minority judgment that the Authority had failed to establish its case. For the reasons that follow the appeal to this Court should be allowed.
The evidence on the Family Court application
Questions of fact and law, including the statute law of New Zealand, were involved. In that regard some assistance is given by Pt 6 (ss 838-646) of the Evidence and Procedure (New Zealand) Act 1994 (Cth). In proceedings in a court in Australia proof is not required about the provisions and coming into operation of a New Zealand statute, or of delegated legislation thereunder, and the judge may inform himself or herself about those matters in any way thought fit (ss 38, 40).
Further, reg 29(5) of the Regulations applies generally to Convention countries and in the present litigation empowered the Family Court to take judicial notice of a law in force in New Zealand and of New Zealand decisions of a judicial or administrative character. Regulation 29(2) rendered admissible as evidence of the facts stated therein, the application and supporting documents. Regulation 29(3) rendered admissible in evidence affidavits of witnesses who resided outside Australia and did not attend for cross-examination. But reg 29(3) did not exclude any power the Family Court may have to permit cross-examination.
In the present case, when giving his ex tempore reasons Steele J said:
As is typical of these applications, the factual matters have been dealt with on affidavit evidence and have not been the subject of cross examination.
Upon various disputed events in the relationship between the parents, his Honour said it was "not possible to form a conclusive view .... without lengthy and detailed cross examination, which is not possible".
In the Full Court ((2007) 211 FLR 357 at 394), the majority, after noting that the proceedings before the primary judge "were conducted entirely on the papers", indicated that therefore it was open to the Full Court to substitute its own findings of fact.
The material before the primary judge comprises more than 200 pages of the appeal record in this Court. It includes affidavits sworn by the father in support of the application by the Authority on 3 October 2006 and thereafter on 28 November 2006, and by the appellant on 10 November 2006 and 14 December 2006. That last date was shortly before the hearing by the primary judge. The father's solicitor in Auckland also provided affidavit evidence which, among other topics, considered some of the relevant provisions of New Zealand legislation. Some affidavits by the other New Zealand deponents were filed in support of the father's case, some in support of the appellant.
Much of the affidavit evidence dealt with disputed issues of fact and alleged fact (particularly relating to domestic violence between the parents) which are not immediately relevant to the factual and legal issues which are still in dispute in this Court. These include the issue whether the appellant was living with the father as a de facto partner at the time of the birth of the child. It is unfortunate that the answer to such a question does not more readily appear on the evidence.
Both at first instance and in the Full Court much attention was given to what were said to be discrepancies in the affidavit evidence of the appellant which bore upon this issue. Perceived weaknesses in her account were relied upon to strengthen the positive case for a conclusion of wrongful removal or retention of the child which it was for the Authority to establish.
The deficiencies in the appellant's affidavit evidence would not have been left for textual analysis had one of several courses been followed. Upon application, or at the initiative of the Family Court itself, the proceedings may have been adjourned for the prompt provision of more adequate affidavit evidence. Leave may have been sought by the Authority for the cross-examination of the appellant.
Section 98 of the Act states that the Rules of Court may provide for evidence of any material matter to be given on affidavit at the hearing of proceedings other than divorce or validity of marriage proceedings. The Family Law Rules 2004 ("the Rules") are so drawn as to require evidence in chief to be given by affidavit (r 15.05). But exercise by the Family Court of its general powers expressed in Pt 1.3 of the Rules would have allowed an order permitting cross-examination of the appellant; such leave might properly have been limited by the Family Court to particular areas of dispute.
Cross-examination in interlocutory applications generally is not to be encouraged. But an application for a return order under reg 16 of the Regulations is a special type of proceeding. It is apt to achieve what in Australia is a final result upon the application for return of a child to another Convention country. To emphasise these matters is not to encourage the amplitude of the evidence to which the House of Lords referred in In re M (Children) (Abduction: Rights of Custody)  3 WLR 975 at 980. The oral evidence in that Convention application was heard over two days.
Regulation 15(2) obliged the Family Court, "so far as practicable", to give to the application by the Authority "such priority" as would "ensure that [it was] dealt with as quickly as a proper consideration of each matter relating to the application allows". If within 42 days of its filing the application had not been determined, the Authority would have been empowered by reg 15(4) to seek from the Registrar a written statement of the reasons for the absence of a determination. Regulation 15 reflects the exhortation in Art 11 of the Convention that "judicial or administrative authorities" act "expeditiously" in these matters and the reference in Art 7 to "the prompt return of children".
The judicial or administrative authorities which decide return applications in some Convention countries may not, under their legal systems, have the obligations to provide the measure of procedural fairness and to give reasons which generally apply in common law systems and which were observed here by the Family Court. Thus, in this country, the requirement of promptitude can be an onerous one.
Nevertheless, prompt decision making within 42 days is one thing, and a peremptory decision upon a patently imperfect record would be another. The references to "summary procedure" and to the dealing with applications on affidavit evidence and "in a summary manner" by the Full Court in In Marriage of Gazi (1992) 111 FLR 425 at 428 are apt to mislead. This is particularly true of the statement in that case (at 428):
The primary purpose of the Convention, the relevant legislation and regulations is to provide a summary procedure for the resolution of the proceedings and, where appropriate, a speedy return to the country of their habitual residence of children who are wrongly removed or retained in another country in breach of rights of custody or access [sic] (see Convention, Arts 7 and 11, Family Law (Child Abduction Convention) Regulations, reg 19(1)). Accordingly, whilst there may be cases in which it is appropriate to allow cross-examination of deponents of affidavits, such cases would be rare. The majority of proceedings for the return of children, pursuant to the Convention, should be dealt with in a summary manner and cross-examination of deponents of affidavits would not be appropriate.
The danger in reading such remarks too literally (and without regard to the circumstances of each particular case) is apparent in situations such as that considered in the United States by the Court of Appeals for the Third Circuit in In re Application of Adan 437 F 3d 381 (3rd Cir 2006). An application by the father for the return of his child to Argentina was resisted on the grounds that he had not established his custody rights under the law of Argentina and there was grave risk there of harm to the child. After considering the cursory treatment by the United States District Court of the application, the Court of Appeals said (at 398):
Although the Convention seeks to facilitate the prompt return of wrongfully removed children to their country of habitual residence, it does not condone deciding that a child is another country's problem and dumping her there, and nor do we.
No criticism of that degree is directed to the conduct of the present case, but In re Application of Adan provides a caution against inadequate, albeit prompt, disposition of return applications.
The policy of the Convention
The jurisdiction exercised by the Family Court in this case with respect to the international child abduction provisions of s 111B of the Act and the Regulations did not exhaust the jurisdiction which might otherwise have been attracted under other provisions of the Act. Regulation 6(2) confirmed the preservation of two other heads of jurisdiction for the return of a child to the country in which he or she habitually resided before the removal to or retention in Australia of the child. One was that with respect to parenting orders under Pt VII of the Act, to which further reference will be made. The other was jurisdiction "under any other law in force in Australia", an expression in reg 6(2) apt to include statutory adaptations of the wardship or parens patriae jurisdiction derived from the general law.
It is instructive to contrast the jurisdiction based upon s 111B of the Act and the Regulations and what, under the general law, would have been the controlling principles for this dispute in a court exercising parens patriae or wardship jurisdiction.
First, one of the grounds upon which parens patriae or wardship jurisdiction with respect to a child is asserted is the physical presence of the child within the territorial jurisdiction, even falling short of residence, if the protection of the court is needed.
Secondly, in questions of custody the paramount consideration, to which "all others yield", is the welfare of the child.
Thirdly, this remains the case even where there is an existing custody order made by a foreign court.
Thus in the joint judgment of five members of this Court in Kades v Kades (1961) 35 ALJR 251 (at 254 per Dixon CJ, McTiernan, Kitto, Taylor and Windeyer JJ) their Honours said:
The courts in Australia have complete jurisdiction over the question of the custody of the child. The order of the Supreme Court of New York is a factor which must be considered, but the responsibility lies with the courts here and the welfare of the child remains the paramount consideration: McKee v McKee  AC 352.
The nature and scope of return applications heard in the Family Court pursuant to the Regulations differ in all three of the aspects just mentioned.
Further reference should now be made to Pt VII of the Act, which is headed "Children" and comprises ss 60A-70Q. Part VII comprises 16 Divisions and provides for the making of a range of orders, of which parenting orders are but one category. Orders made under Pt VII of the Act are not formulated in terms of the grant to a person of custody of, or access to, a child. Nevertheless, the animating principle of the best interests of the child remains in Pt VII. Section 60CA states:
In deciding whether to make a particular parenting order in relation to a child, a court must regard the best interests of the child as the paramount consideration.
Pt VII is not confined to parenting orders. Section 67ZC provides:
This Court has held of a previous provision in the Act to the same effect as s 67ZC that it:
invested the Family Court with a welfare jurisdiction in respect of a child of a marriage which encompasses the substance of the traditional parens patriae jurisdiction freed from the preliminary requirement of a wardship order.
The exhaustive list in s 69E of criteria to attract jurisdiction includes the presence of the child in Australia when the application is filed in court (s 69E(1)(a)) and the satisfaction of the common law rules of private international law for the exercise of jurisdiction in the proceedings (s 69E(1)(e)). Thus while it appears that jurisdiction under Pt VII of the Act would also have been attracted in the circumstances of this case, the paramount consideration, as with the traditional wardship jurisdiction, would have been the best interests of the child in the particular circumstances of the case.
However, a different policy with respect to the best interests of the child has prevailed with return applications under the Regulations. There the focus is upon the appropriate forum. This may be detected in the statement in par (b) of reg 1A(2) of the Regulations that they were intended to be construed:
recognising, in accordance with the Convention, that the appropriate forum for resolving disputes between parents relating to a child's care, welfare and development is ordinarily the child's country of habitual residence.
A recent discussion of the provenance and mixed objectives of the Convention contains the following:
The aims of the Convention are distilled from a number of fundamental principles that featured prominently during the negotiation of the Convention and led to its wide acceptance. These are that the interests of children are paramount in cases of child abduction; that it is generally contrary to the best interests of any child to be abducted; and that it is the courts of habitual residence (normally the home environment of the child) that are generally best placed to decide on the future upbringing of the child. So the Convention seeks to restore the child's status quo in order both to reduce the incidence of international child abduction through the provision of legal rules which effectively mean there is nothing to be gained by abducting this child, and to ensure that the decision on the future of the child is taken in the forum conveniens, ie, the most appropriate jurisdiction to make such a determination.
The distinction between the exercise of jurisdiction founded in legislation adopting the Convention and the wardship jurisdiction was identified by Baroness Hale of Richmond in In re M (Children) (Abduction: Rights of Custody)  3 WLR 975 at 989-990 as follows:
In non-Convention cases the child's welfare may well be better served by a prompt return to the country from which she was wrongly removed; but that will be because of the particular circumstances of her case, understood in the light of the general understanding of the harm which wrongful removal can do ....
In Convention cases, however, there are general policy considerations which may be weighed against the interests of the child in the individual case. These policy considerations include, not only the swift return of abducted children, but also comity between the contracting states and respect for one another's judicial processes. Furthermore, the Convention is there, not only to secure the prompt return of abducted children, but also to deter abduction in the first place. The message should go out to potential abductors that there are no safe havens among the contracting states.
It was against this background that the Court held in De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640 at 658 that proceedings under the Regulations are not subject to the paramountcy principle which was then expressed in s 64 of the Act.
Nevertheless, in In re M  3 WLR 975 the House of Lords disagreed with authorities indicating that a Convention case must be "exceptional" before the court might properly decide to refuse to make an order for return. In the circumstance of the case before it, the House held that "children should not be made to suffer for the sake of general deterrence of the evil of child abduction world wide" (at 992) and dismissed the Convention application by the father of the children.
In the present litigation, it was recorded by the primary judge to be common ground that while the best interests of the child were not paramount in determining whether an order for return should be refused on one or more of the grounds in reg 16(3), those best interests were a factor in that decision making process.
The appeal to this Court does not turn upon the identification of the best interests of the child in the sense indicated by the primary judge. This is so although much of the affidavit evidence in support of the appellant's case does appear to have been prepared with a view to attracting the exercise of the discretion under reg 16(3) in favour of refusal of return of the child to New Zealand, if, as proved to be the case, she failed upon the threshold issue of jurisdiction. The appeal does turn on that threshold issue.
Whilst, as remarked above, the mere presence of the child might found the exercise of the wardship jurisdiction or that under Pt VII of the Act, the operation of the regime for which the Regulations provide is attracted by more complex criteria. These threshold matters fix upon the timing of the application, and the satisfaction of the court that the child's removal or retention was "wrongful". It is with the latter criterion that the appeal is concerned.
Paragraphs (a) and (b) of s 111B(1A) of the Act provide for the Regulations to deal with the onus of establishing that a child should not be returned and to establish rebuttable presumptions in favour of return of the child. Regulation 16 is to be read in this light.
The application to the Family Court was made well within one year of the removal of the child from New Zealand. The consequence was that reg 16(1) applied to the application. This meant that, subject to the power of the Court to refuse to make an order because the appellant, the person opposing return, established one or more of the matters in reg 16(3), the Family Court was obliged to make an order for return if "satisfied" by the Authority of a critical matter. This was that the removal to or retention in Australia of the child "was wrongful under subregulation (1A)". Was the Family Court properly "satisfied" by the Authority within the meaning of these provisions?
Sub-regulation (1A) of reg 16 should be set out. It states:
What is meant by the expression "rights of custody in relation to the child" which appears in par (c)? A detailed explanation is provided by reg 4 as follows:
The case presented by the Authority
In submissions to this Court the Authority contended that the dismissal of the appellant's appeal to the Full Court was supported upon several distinct grounds. First, reliance was placed upon the terms of the Access Order, made on 4 December 2000 and still operative when the appellant and the child left New Zealand on 15 September 2006; the Access Order was said to confer upon the father rights relating to the care of the child which were to be understood as including "the right to determine the place of residence of the child" within the meaning of reg 4(2), and it was said that the removal of the child had been in breach of the Access Order rights; these rights amounted to a "right of veto" and were to be treated as rights of custody held by the father.
Secondly, the removal was said to be in breach of rights of custody held by the New Zealand Family Court at Auckland which had made the Access Order.
Thirdly, and this was the ground which founded in this Court the debate as to the evidence, the father was said, by the operation of the statute law of New Zealand upon the facts, to have had rights which in turn answered the description in reg 16(1A) of "rights of custody".
It will be apparent that the critical provisions of the Regulations are so drawn as to involve in the determination of their operation in this case consideration of matters of New Zealand law. The Full Court correctly emphasised that the Family Court was not seeking to decide the issues under the Regulations by its assessment of the answer a New Zealand court would give upon the matters of New Zealand law: (2007) 211 FLR 357 at 376.
Insofar as matters of foreign law are involved in determination of an application under the Regulations, observations of the Court of Appeals for the Third Circuit (made respecting the role of the United States District Court in cases of alleged wrongful removal under Art 3 of the Convention) are in point. In In re Application of Adan 437 F 3d 381 at 391 (3rd Cir 2006) that Court said:
The duty of the host forum - in this case, the District Court - to make a threshold determination of custody rights under the country of origin's laws is not novel; indeed, it comports with the federal courts' frequent responsibility to examine the law and choice of law rules of another forum to determine the rights and duties of litigants. Such a determination does not, of course, bind the other forum to reach the same result in future litigation, nor does it run afoul of comity concerns. Article 3's requirement that the host country determine custody rights under the country of origin's law to ascertain whether removal was 'wrongful', and therefore whether the Convention applies, is a straightforward question of law of the sort federal courts routinely encounter, and thus presents no unusual burden on the competence of our courts.
It is convenient to consider together the first and second grounds relied upon by the Authority.
Right of veto a right of custody
The starting point must be the text of the Access Order. This comprised orders as follows:
These orders answer the description in reg 2(1) of "rights of access" because they provide for the father to take the child to a place other than that of his habitual residence with the appellant. But do they confer "rights of custody" because they confer upon the father "the right to determine the place of residence of the child" (reg 4(2))?
Some difficulty arises from the use of the term "right" in the expressions in reg 4 (and in Art 5) "rights of custody" and "right to determine". Even when used in a broad sense, to speak of a "right" in one person suggests a correlative duty, obligation, disability or liability in others. Regulation 4 is so drawn that "rights of custody", and, by inference, those of determination of place of residence, may arise by reason of a judicial decision (reg 4(3)(b)). Here, the Access Order gave rise to rights in each parent with correlative duties or obligations in the other parent to observe the requirements of the Access Order.
But unlike the court orders in some of the cases arising directly under the Convention or specific provisions made for parental rights and duties by legislation considered in other cases, the Access Order was not addressed to and imposed no prohibition on, in the absence of consent by both parents, the removal of the child from the jurisdiction of the New Zealand court which made the order.
A "right of veto" of that nature may give rise to a right in each parent to determine that there be no change in the "place of residence", using that phrase to refer to the Convention country where the child habitually resides; the right of each is attended by the correlative obligation of the other party to observe the status quo and the observance of the obligation will attract whatever remedies are given by the judicial or administrative authorities of that Convention country of habitual residence. That power of prohibition of change may answer the phrase in reg 4 (and in Art 5) "the right to determine ....". The majority of the Full Court referred to decisions in Australia and other Convention countries in which that proposition found favour and Finn J was prepared to accept them. The Authority supported this line of authority.
The appellant pointed to several difficulties in its acceptance. One fixed upon the distinction between a restriction upon change of the status quo and an active power to choose and change at will the country of residence of the child. In that respect the appellant relied upon a line of authority including the majority of the Court of Appeals for the Second Circuit in Croll v Croll 229 F 3d 133 (2nd Cir 2000).
A distinct difficulty would arise where, as in Croll v Croll itself (at 135-136), the source of the power of veto was found not in the terms of a court order or in statute but merely in an agreement between the parents. There, as the appellant correctly submitted, the "right of veto" would be at best a potential right, dependent upon a successful application to the relevant judicial or administrative authorities for its enforcement by creation of a presently imperative bar to removal. The "right to determine" spoken of in reg 4 (and Art 5) is more than an expectancy or potential right.
Putting that distinct difficulty to one side, it is unnecessary to decide on this appeal which of the above lines of authority concerning the "right of veto" should be accepted as indicative of the proper construction of reg 4(2). This is because, as we have indicated, the access rights provided for the father by the Access Order conferred no "right of veto" in any sense discussed in the authorities.
Some reliance was placed by the Authority upon s 80 of the NZ Children Act. This makes it an offence for, among other things, a person, knowing that there is in force an order of a court having jurisdiction under that statute which gives to any other person "the role of providing day-to-day care for, or contact with, the child", to take the child out of New Zealand without the leave of that court. It is not clear that the Access Order provided the father with such a role, still less apparent on the evidence that the appellant knowingly acted in breach of the Access Order. Further, in final oral submissions the Authority appeared to resile from reliance upon s 80 (and any suggestion of contempt of the Access Order) as necessary to its case.
The Authority fails in its reliance upon the Access Order as the source of custodial rights of the father because its argument does not adequately observe the distinction drawn in the Regulations and in the Convention between rights of custody and those of access. The importance of the preservation of the distinction in construing the Convention was, with respect, correctly emphasised by the House of Lords in In re D (A Child) (Abduction: Rights of Custody)  1 AC 619 at 635. Reference was made by the Authority to the "frustration" of the rights of access given the father pursuant to the Access Order, by removal of the child from New Zealand. But that description of the events that happened does not translate the rights of the father to a right of determination of the place of residence of the child and thus to a right of custody.
Moreover, the avenue which may have been open to the father to approach a New Zealand court to obtain an order barring removal of the child from New Zealand without his consent, in some of the cases dubbed a ne exeat order, did not, without more, render him a person with presently subsisting rights of custody.
The custody of the New Zealand court?
The terms of reg 4 ("a person, an institution or another body") are sufficiently broad to render what may be classed as a court the holder of rights of custody in relation to a child. This will be no curiosity at least in Convention countries which have inherited the wardship jurisdiction derived from the English Court of Chancery.
Article 8 uses broad terms and speaks of an application by a person, institution or other body claiming that a child has been removed or retained in breach of "custody rights", without limiting the identity of the applicant to the holder of the rights.
But in Australia, reg 16(1A), in defining what amounts to a "wrongful" removal, fixes (par (c)) upon the person, institution or other body seeking return and requires it to have had rights of custody immediately before the removal. Even if the New Zealand court did have such rights, it was not the Convention applicant here. The Authority moved in the present case upon the motion of the father whom it described in the application to the Family Court as "[t]he applicant under the Convention" and as having the alleged rights of custody spelled out earlier in these reasons.
The decisions of courts in other Convention countries, including those of the Supreme Court of Canada (Thomson v Thomson  3 SCR 551 at 588) and the House of Lords (In re H (A Minor) (Abduction: Rights of Custody)  2 AC 291 at 302-306) which do not link the identity of the Convention applicant to the holder of the custody rights, must be read with attention to the precise requirements in the Australian legislation. The Regulations, as Finn J indicated, do not accommodate a Convention application by a parent asserting breach of the rights of custody vested in a court.
Furthermore, the reasoning in these decisions fixes upon the engagement of the court's jurisdiction to deal with custody of the child, and the pendency of those proceedings. That court, at least in a common law system, will have an inherent power to protect its own processes once set in motion.
This reasoning is reflected in the statement by La Forest J in Thomson v Thomson  3 SCR 551 at 588:
It seems to me that when a court has before it the issue of who shall be accorded custody of a child, and awards interim custody to one of the parents in the course of dealing with that issue, it has rights relating to the care and control of the child and, in particular, the right to determine the child's place of residence.
As Finn J noted, there were no such pending proceedings in the present case. Nor can it sensibly be said that s 80 of the NZ Children Act, in creating a criminal offence to be prosecuted by the proper authorities, confers a "power of veto" with respect to the country of residence of a child which gives rise to existing custody rights vested in any court which has made an order to which the section speaks in pars (b) and (c).
It remains to consider the third ground, that upon which the majority of the Full Court upheld the primary judge.
In relation to this third ground, there was no issue between the parties in this Court that if the father was a "joint guardian" by virtue of s 17 of the NZ Children Act, then under the law of New Zealand and immediately before the removal of the child he had "rights of custody" within the meaning of reg 16(1A).
Section 15 of the NZ Children Act states that for the purpose of that statute "guardianship" of a child bears a meaning which includes "all duties, powers, rights, and responsibilities that a parent of the child has in relation to the upbringing of the child". These include the determination for the child of important matters affecting the child; changes to the child's place of residence is one of these matters (s 16(1) and (2)). Section 17(1) enacts that the father and mother of a child are guardians jointly of the child, and that this is so "unless" sub-s (2) or sub-s (3) applies. Section 17(3) provides that if a child is conceived before the commencement of the NZ Children Act (the case here), the mother of the child is the sole guardian, if the mother was neither:
Section 17 is so cast that if the facts and circumstances answer sub-s (3), then the consequence indicated in sub-s (1) applies. The mother of the child is the sole guardian. If sub-s (3) does not apply to the facts and circumstances of this case then the opening words of sub-s (1) apply, and the parents are joint guardians. There is involved here no issue of onus of proof of the kind encountered in cases dealing with the distinction between exceptions and provisos such as Vines v Djordjevitch (1955) 91 CLR 512 at 518-519. The better view is that the scheme of s 17 is to specify the various elements of the several species of right which it then establishes.
There is no issue that the child was conceived before the commencement of the NZ Children Act. The dispute has been whether within the meaning of par (b) of s 17(3) the appellant was "living with the father of the child as a de facto partner at the time the child was born". This must be read with s 29A of the Interpretation Act 1999 (NZ) ("the NZ Interpretation Act"). Reference to that provision does not appear in the reasons of the primary judge, no doubt because his Honour had not been referred to it. The Full Court located it through its own efforts during the course of argument ((2007) 211 FLR 357 at 393). Section 29A(1) of the NZ Interpretation Act states:
In an enactment, de facto relationship means a relationship between 2 people (whether a man and a woman, a man and a man, or a woman and a woman) who -
In determining whether two people "live together as a couple in a relationship in the nature of marriage or civil union" within the meaning of par (a) of s 29A(1), sub-s (3) requires the court which is determining the question to have regard to:
The purpose stated in s 3(1) of the NZ Children Act includes promotion of the welfare and best interests of children "by helping to ensure that appropriate arrangements are in place for their guardianship and care". From this provision the majority of the Full Court concluded (at 396):
it would seem appropriate to set a relatively low threshold when determining whether the parents of a child were living in a de facto relationship.
The appellant criticised this passage. She pointed to the lack of any stated comparator against which the relativity was to be assessed, and added that on any considered evaluation of the evidence in this case there could be no confident conclusion that the Authority had made out its case.
It will be observed from the New Zealand legislation that two complex and related questions emerged in this case. The first was whether the appellant and the father were living together at the time of the birth of the child. The second was whether at the time of the birth of the child the appellant and the father were in a relationship in the nature of marriage or a civil union under the law of New Zealand.
The primary judge dealt as follows with the evidence:
The evidence of the Father and the Mother is in conflict. In the circumstances of this case, I prefer the evidence of the Father, who says that he lived with the Mother at the time the Child was born. The Father did not use the phrase 'in a de facto relationship' but the evidence infers that. I am satisfied that the Father and Mother were cohabiting, so that by operation of section 17, the Father and the Mother are joint guardians of the Child. The Mother's affidavit filed 14 November 2006 in paragraph 3, sworn at a time when the issue of the parties living together was not seen as being central to the issues, records, 'I moved out with my two month old son and went to live with my parents'. Her later affidavit filed on 15 December 2006, the day before the hearing, asserts that the parties never lived together as de facto partners.
The evidence referred to in this passage was as follows. The only evidence from the father was the statement in his affidavit of 3 October 2006, in support of the Convention application:
I am [the child's] joint guardian by virtue of the fact that I lived together with the mother at the time of [the child's] birth.
Paragraph 3 of the appellant's earlier affidavit, sworn on 10 November 2006, reads:
Abuse from [the father] started during my pregnancy. This included verbal abuse, pushing, and pulling hair. After the child was born, I experienced starvation, and name-calling (such as 'Fat Cow', 'Zebra' because of my stretch marks, and 'Hungry like a Pig') when I was breast-feeding [the child]. [The father] restricted my parent's visits, the phone was disconnected, power shut off, there was no food, and strange people would visit our house demanding money for [the father's] business transactions, which I knew nothing about. This is why I moved out with my 2-month old son and went to live with my parents. My Parents were living in New Zealand at that time.
Paragraphs 2, 3, 5, 6 and 7 of the appellant's affidavit of 14 December 2006 read:
Annexed to that affidavit was a photocopy of a double sided name tag which shows the date of birth of the child as 20 September 1996 and on the reverse bears a stamp which shows the address of the appellant as that of her parents. The stamp bears a date "29.10.96".
The appellant also relied on an affidavit sworn by a family friend ("LG") who deposed:
I am aware that [the appellant] lived with her parents at [her parents' address] when [the child] was born. During that time, I visited [the appellant and the child] at her parents' house around three times a week in addition to seeing them at the local church every Sunday.
Counsel for the Authority properly pointed out that the phrase "when [the child] was born" which appears in the affidavits of the appellant and LG describes the social situation of the appellant and allows for confinement of the appellant in hospital. The same may be said of the phrase in s 17(3)(b) of the NZ Children Act, "living with the father of the child as a de facto partner at the time the child was born".
The appellant's affidavit sworn 14 December 2006 must have been received by the Authority at best only shortly before the hearing. In these circumstances it is unfortunate that one or more of the courses mentioned earlier in these reasons was not taken to supplement the record before the primary judge. The need to do so was particularly pressing given the bald statement by the father in his first affidavit. This represented all he had to say on the matter, although it was for the Authority to establish that there had been wrongful removal. Further, with respect to the evidence by the appellant there was lacking, for example, clearly expressed evidence which would base a finding as to when the parties began to cohabit and when they ceased to cohabit and whether the appellant moved into the house of the father after or before she became pregnant.
The majority in the Full Court accepted a submission by the Authority that the evidence given in par 5 of the appellant's affidavit sworn 14 December 2006 was "concocted in order to fit the provisions of the [NZ Children Act]". May and Thackray JJ said that they were drawing an "inference from the way in which the evidence unfolded". As to the address on the name tag their Honours said (at 395):
The address on the nametag might have been the mother's address at the time of the child's birth, but there could also have been some other explanation. Furthermore, no explanation was provided as to why the date shown on the tag was more than a month after the date of birth of the baby.
With respect to the affidavit provided by LG the majority said that it was "worth noting" that the deponent was "purporting to remember the wife's [sic] place of abode some ten years previously" and referred to her evidence "at least initially" that she had moved in with her parents when the child was two months old. The conclusion was that "[i]t would therefore not be surprising that the witness recalled visiting mother and baby at the home of the mother's parents" (at 395).
Their Honours continued (at 396):
Although the father provided virtually no evidence to assist the Court to determine the issue, we consider the mother's own evidence in her original affidavit was (just) sufficient to justify a finding that at the time of [the child's] birth the mother was living with the father as 'a de facto partner'.
The majority expressed as follows their conclusions upon this aspect of the case (at 396):
The [Authority] bears the onus of establishing that the child's removal from New Zealand was wrongful and therefore bears the evidentiary onus on issues such as whether or not [the child's] parents were living in a de facto marriage relationship at the time of [the child's] birth. We are satisfied that the onus has been discharged and we therefore arrive at the same conclusion as did Steele J, albeit by different route.
The satisfaction as to onus expressed in the second sentence appears to have been based on the proposition that while the father had provided virtually no evidence, that of the appellant in her first affidavit was sufficient, when turned against her by the Authority in submissions, for the Authority to discharge the onus. This also involved discounting the evidence of the name tag and that given by LG.
It was inappropriate for the Full Court to make a finding of "concoction". This was much more than a finding of an unexplained change of recollection. There had been no cross-examination of the appellant and explanation of apparent inconsistencies between two affidavits was a matter of speculation. There was, however, the other evidence which provided some support for her version of events in the later affidavit.
The proper conclusion on the record before the Full Court was that reached by Finn J in her dissenting reasons. Her Honour stressed the burden of persuasion carried by the Authority, found that the affidavit material was insufficient to found an inference that the parents had lived together as a couple in a relationship in the nature of marriage or civil union, and concluded that the Authority had "simply failed to establish its case that the father was a guardian of the child and could thus determine the child's place of residence" (at 367).
The majority of the Full Court erred in the ground upon which they upheld the decision of the primary judge. The further grounds urged by the Authority in this Court in support of the outcome in the Full Court are not made out. The consequence is that the appeal to this Court should succeed.
Counsel for the Authority properly accepted that the scheme of the Regulations (and of the Convention) was that the need for certainty and prompt disposition of wrongful removal applications presented a controversy susceptible of investigation and adjudication once only. The Authority did not seek, were the appeal to be allowed, any order for rehearing of the application.
The circumstance just mentioned emphasises the need for prompt but, so far as the circumstances permit, thorough examination on adequate evidence of the issues arising on wrongful removal applications under the Regulations.
The appeal should be allowed. Order 1 of the orders of the Full Court of 30 April 2007 should be set aside and in place thereof it should be ordered that the appeal to the Full Court be allowed, orders 1-8 made by the primary judge on 18 December 2006 be set aside and in place thereof the application filed 11 October 2006 be dismissed.
Two further points should be made. The first concerns costs. The primary judge made no order for costs, and order 2 of the orders of the Full Court, which will stand, is that each party pay their own costs of the appeal to the Full Court. Regulation 30 provides in some circumstances for an order by the Family Court that the person who has removed or retained the child pay the costs of the Authority of its successful application for return. Regulation 30 has no application here.
In this Court, the matter of costs is controlled by the general provision of s 26 of the Judiciary Act 1903 (Cth). In the circumstances, there should be no order for costs of the appeal to this Court.
The second matter concerns the stay ordered by this Court on 31 August 2007 of the orders of the primary judge. The stay was until the determination by this Court of the appeal or earlier further order of this Court. The appeal having been determined, the stay expires. In any event the relevant orders of the primary judge themselves have been set aside.
This is an appeal from a judgment entered by the Full Court of the Family Court of Australia. That Court was divided. The majority (May and Thackray JJ, in joint reasons) favoured affirming the orders of the primary judge in the Family Court (Steele J). His Honour had made orders that the Central Authority, represented by the Director-General of the Department of Community Services of New South Wales ("the State Central Authority"), "make such arrangements as are necessary for the return of the child ['K'] .... born 20 September 1996 to Auckland New Zealand, accompanied by the child's mother": SYF 4027 of 2006 ("reasons of the primary judge"). One judge in the Full Court (Finn J) favoured allowing the appeal and substituting orders dismissing the State Central Authority's application.
The proceedings, issues and disposition
The Convention and Regulations
The proceedings in the Family Court were brought by the State Central Authority in response to a request by its New Zealand counterpart ("the New Zealand Central Authority"). They were initiated on behalf of K's father. The request was made pursuant to the Hague Convention on the Civil Aspects of International Child Abduction ("the Convention"). Australia and New Zealand are Contracting States under the Convention. In Art 1, the objects of the Convention are stated to be:
In Australia, the Convention has not been incorporated directly into municipal law. Instead, its provisions are reflected in the Family Law (Child Abduction Convention) Regulations 1986 (Cth) ("the Regulations"). The Regulations are made pursuant to s 111B of the Family Law Act 1975 (Cth) ("the Act"). Substantially, the Regulations follow the language of the Convention. Their general purpose is stated to be "to enable the performance of the obligations of Australia, or to obtain for Australia any advantage or benefit, under [the Convention]". The Convention itself appears as Sched 1 to the Regulations. By reg 1A(2) it is provided that the Regulations are intended to be construed:
Requirement of wrongful removal
As Baroness Hale of Richmond has remarked, it would have been simpler if the Convention had provided that all removals or retentions of a child outside the country of habitual residence, without the consent of the other parent or the authority of a court, were wrongful. However, the Convention does not so provide. Nor do the Regulations do so. The Convention reflects compromises agreed during the negotiations. In effect, the Convention recognises, as Baroness Hale put it, that "not all parents have the right to demand the automatic return of children who have been taken away without their consent". Preconditions to that entitlement are expressed. Those preconditions must be satisfied if the somewhat drastic consequences for which the Convention provides are to be invoked. Such consequences include the intervention of designated Central Authorities; the non-voluntary return of the child, if necessary, to the country of habitual residence; likely further proceedings in that country; and a consequential impact on the lives of the child, parents and guardians concerned.
In an application under the Regulations to an Australian court, it is a precondition to the making of a "return order" that the court should be satisfied "that the child's removal or retention was wrongful". To be "wrongful", the removal or retention must be shown to be in breach of "rights of custody", as defined in the Regulations. It follows that it is necessary that the person, institution or other body requesting the return of an allegedly abducted child should be possessed of such rights, which have been breached either by the removal of the child from another Contracting State to Australia or the retention of the child in Australia. In effect, the drafting recognises that some persons (usually one parent) will have a legal entitlement unilaterally to remove a child from one Contracting State to another, however painful that course may be for other persons who assert an interest in the child and whose relationship with the child will thereby be interrupted or even terminated.
Mother's abandoned and remaining issues
A number of issues that were contested below have been abandoned following determinations adverse to the appellant. Thus, the appellant no longer contends:
that the removal of K from New Zealand was not wrongful because New Zealand was not K's country of "habitual residence";
that the father had consented to, or acquiesced in, the removal;
that a return order should be refused because there is a "grave risk" that it would expose K to physical or psychological harm or otherwise place him in an intolerable situation; or
that K had expressed a desire not to be returned to New Zealand.
The essence of the appellant's grounds of appeal is that the majority in the Full Court erred in determining that the evidence tendered in the proceedings was sufficient to sustain a finding that, at the time of the child's birth, the appellant was living with the father as a de facto partner. It was this finding that had occasioned the majority's conclusion that the father was a "guardian" of K under New Zealand law, and thus possessed of "rights of custody" in respect of K for the purposes of the Regulations.
Central Authority's contentions
By a notice of contention, the State Central Authority contends that, even if the appellant is successful on her appeal point, the orders of the Full Court should be upheld on alternative bases. First, it is contended that the father had a right to object to the removal of K from New Zealand independent of the question of "guardianship". Such a right was said to be implicit in a New Zealand court order assigning the "care" of K at specified times to the father. If such a right could be established, so it was argued, the appellant's removal of K from New Zealand without the father's consent was "wrongful" for the purposes of the Regulations. In the alternative, the Authority contended that the "New Zealand courts" had a right to object to the removal of K because, given that court order, the appellant's action in effecting such removal without the consent of the New Zealand courts put her in breach of a particular criminal prohibition applicable in that jurisdiction. This, it was submitted, also served to render the removal "wrongful".
The result: the appeal fails
The evidence relevant to the appellant's appeal point was unsatisfactory. However, having regard to the issues fought at trial, the available evidence and the conduct of the parties, the appellant has not made good her submission that the majority of the Full Court erred in reaching their conclusion. Even if this is not accepted, the State Central Authority is still entitled to succeed on the basis of the first-mentioned contention point. K's removal, in effect, rendered meaningless the father's rights under the New Zealand court order. It follows that, on either basis, the orders of the Full Court of the Family Court are sustained. The appeal to this Court should be dismissed.
In light of this conclusion, it is possible to put the remaining contention point to one side. As the judges in the Full Court acknowledged, whilst such claims have been upheld in the past, they present certain difficulties, not least because of the language and apparent assumptions of the Regulations. Because it is unnecessary to decide whether rights of custody over K existed in a New Zealand court, it is appropriate to refrain from doing so.
Birth of the child K
Many of the facts contested at trial are no longer relevant, given what the appellant correctly describes as "the limited scope of the appeal". The appellant ("the mother") is a Polish national, who had been resident in New Zealand "for at least some time" prior to the birth of K. The father was born in New Zealand, is a national of that country and is of Maori descent. The child, K, was born in Auckland, New Zealand in September 1996. It is accepted that he is a New Zealand national and he holds a New Zealand passport. The paternity of the father is acknowledged. In her own court documents, the mother identifies the names of the child, obviously derived from each parent. K's first name is Polish whilst his middle name is Maori. His surname is a hyphenated combination of the mother's and the father's surnames.
The mother and father were not married at the time of K's birth. They have not married since, nor entered a civil union under New Zealand law.
Parents' affidavit evidence
The request for the return of K drew support from an affidavit of the father dated 3 October 2006. That document contained an account of the circumstances of K's removal from New Zealand. It also set out some factual background relating to the parents' relationship with each other and with K, and to various court proceedings that had taken place in that connection. It contained the assertion: "I am [K's] joint guardian by virtue of the fact that I lived together with the mother at the time of [his] birth".
On 10 November 2006, the mother filed an affidavit of her own. It was, for the most part, non-responsive to the material contained in the father's affidavit. Instead, it seemed to be directed to establishing that the father was "violent and irresponsible" and posed a danger to her and to K. Annexed were various documents claimed to substantiate this allegation. The affidavit made no mention of the formal aspects of the father's relationship with the mother or K, or the various court proceedings as between the father and the mother. However, it did cast some incidental light on the domestic arrangements of the father and the mother around the time of K's birth, stating that:
Abuse from [the father] started during my pregnancy. This included verbal abuse, pushing, and pulling hair. After [K] was born, I experienced starvation, and name-calling .... when I was breast-feeding [K]. [The father] restricted my parent's [sic] visits, the phone was disconnected, power shut off, there was no food, and strange people would visit our house demanding money for [the father's] business transactions, which I knew nothing about. This is why I moved out with my 2-month old son and went to live with my parents.
The inference that the primary judge drew from these statements, coupled with the father's evidence, was that, at the time of K's birth, the couple were living together in what the mother had described as "our house" until the mother moved out, two months after K's birth.
On 28 November 2006, the father filed a second affidavit. It sought to respond, in a methodical fashion, to the specific allegations made by the mother. Without setting out the details, it is clear that the father's second affidavit, if accepted, would have rendered untenable the mother's claims of violence and abuse. It drew support from a number of annexures.
Then, on 14 December 2006, the mother swore a second affidavit, directing her attention belatedly to the father's assertion that she was cohabiting with him at the time of K's birth. She stated that "[s]ince [K] was born, I lived at my parents' address" and further that "[a]t no time did I ever regard myself as, or live as, [the father's] de facto partner". She stated that she had spent three nights a week at the father's house for a six week period a month and a half after the birth of K on what amounted, in effect, to a trial basis. According to the mother, "[d]uring that time, I did want his home to be mine on the condition that he [was] capable of assisting in the care of [K]". However, on her evidence, the father offered no domestic or financial assistance to her, the house was unclean, and there were "numerous" problems relating to overdue bills.
The foregoing was the extent of the direct evidence about the nature and characteristics of the relationship between the mother and father at the time of K's birth. The primary judge, and the Full Court, therefore had to resolve the conflict of testimony. No oral evidence was called by the mother or the State Central Authority. There was no cross-examination either of the father or the mother. In the result, the primary judge preferred the evidence of the father. He pointed out that the mother's second affidavit was filed only the day before the hearing. Obviously, the primary judge regarded the mother's first affidavit, sworn without apparent regard to the consequences of the admissions made, as more reliable, truthful and convincing. The same conclusion was reached by the majority in the Full Court. They considered that "the mother's own evidence in her original affidavit was (just) sufficient to justify a finding that at the time of K's birth the mother was living with the father as 'a de facto partner'": (2007) 211 FLR 357 at 396 . It will be necessary to return to these conclusions.
Early disputes in New Zealand
It was common ground that, from not long after K's birth, the mother and the father had lived separately. In 1997, the mother and father attended counselling in relation to their care of K. An agreement between them before a statutory counsellor recorded concurrence as to their future relationship and their respective contributions to the care of K. That agreement ("the parenting agreement") was in evidence. It included an acknowledgment by the mother that she would give the father two months' notice of an intention to leave the country, that she would "recognise [the father's] guardianship rights", and that the father would contribute to K's childcare costs during the mother's attendance at an educational course. It also noted that both parents agreed that it was important for K "to discover his Maori heritage and both have stated their commitment to this". In these proceedings, the mother contested having agreed to the father's guardianship rights. Nevertheless, it is apparent from the contemporaneous documents that, pursuant to the parenting agreement, the child was to live with the mother and to spend some time with the father.
When, in 1998, the father became concerned that the mother might take K from New Zealand without his consent, he applied to the District Court of New Zealand for an order forbidding such removal. Such an order was made on 24 March 1999. In June 1999, what is described in later court documents as a "custody access order" was made in respect of K, although that order was not itself in evidence in these proceedings. On 4 December 2000, it was replaced by a new order, made by consent, which in essence reflected aspects of the parenting agreement ("the NZ court order"). Under this order, K was to be in his father's care every second weekend and for half of the school holidays. The father was to provide transport to collect the child from his mother's home at 5 pm on Fridays and to return him at 5 pm on Sundays. At other times the child was to remain in his mother's care. It was common ground that the father proceeded to exercise his rights at least at the level provided for in the order.
By consent of the father, in March 2002 and December 2004 fresh orders were made suspending the non-removal order for short periods, during which intervals the mother took K to visit her parents, who had by then moved from New Zealand to Australia. Then, in June 2005, the non-removal order was discharged by order of the Family Court of New Zealand upon the joint application of the father and the mother. They recorded the basis for their application as being that "past grievances have .... been resolved and .... both parents have an excellent relationship regarding [their] son". Thereafter, the evidence shows that the mother continued to visit her parents in Australia from time to time. On one such occasion, she left K in the father's care in New Zealand.
Child's removal from New Zealand
On 14 September 2006, a dispute broke out between the mother and the father. The events of that evening, which the father and mother contest, apparently persuaded the mother to depart from New Zealand immediately and permanently with K. This she did on 15 September 2006, taking the child to Australia where he has since lived. In New Zealand, the father immediately sought the assistance of local officials. On 2 October 2006, the New Zealand Central Authority made its request on behalf of the father to the State Central Authority for the return of the child to New Zealand pursuant to Australia's obligations under the Convention. It was this request that led to the State Central Authority, on 11 October 2006, filing its application in the Family Court seeking a return order pursuant to the Regulations. It was that application that has led to these proceedings.
The Convention, Regulations and relevant legislation
The Convention and Regulations
As this Court pointed out in DP v Commonwealth Central Authority (2001) 206 CLR 401 at 411 , the task of the Family Court in a matter of this kind is "to apply the Regulations to the facts established by the evidence". It is to do so taking into account the purpose of the Regulations, being the fulfilment of Australia's obligations under the Convention. As in DP, it was not suggested in this appeal that there was any relevant discordancy between the Regulations and the Convention. In Australia, the Regulations express what is the governing law.
The critical regulation, both for the issue raised by the mother in this appeal and the issue advanced by the State Central Authority in its notice of contention, is reg 16. Relevantly, that regulation provides:
"[R]ights of custody" are defined in reg 4, which provides:
The NZ Care of Children Act
At all relevant times, K was under the age of 16 years. It is now undisputed that, before his removal to Australia in 2006, he habitually resided in New Zealand (a convention country). The question that remains is whether, immediately before such removal, the father had "rights of custody" in relation to K under the law of New Zealand which were breached by K's removal to Australia, and which he was exercising or would have exercised but for such removal.
The evidence at trial left no doubt (and indeed there was a finding) that the father continuously and sufficiently exercised such rights in relation to K as he enjoyed under New Zealand law. A measure of whether he would have continued to exercise such rights, but for the removal, can be found in the father's conduct as described by the primary judge, his repeated assertion of his rights before the courts of New Zealand, and, when K was removed to Australia, his action to invoke the assistance of the Convention through the New Zealand Central Authority.
Given that the inquiry is as to rights that existed "in relation to the child under the law of" New Zealand, it is necessary to have regard to any New Zealand statute law or common law that touches upon the content and character of the father's "rights" and assists in determining whether they are "rights of custody" within the meaning of the Regulations as understood by reference to the Convention.
It was common ground that, if the father was shown to be a legal guardian of K, he would have "rights of custody" for present purposes. For that reason, much attention was given below to the Care of Children Act 2004 (NZ) ("the NZ Act"). As the majority reasons in the Full Court indicated, if it were established that the father was a "guardian" of K under the terms of that Act, there would be no need to establish any other "rights of custody"((2007) 211 FLR 357 at 392 ). Of itself, this would be sufficient to lend the propounded legal colour to the father's rights.
Section 15 of the NZ Act provides:
For the purposes of this Act, guardianship of a child means having (and therefore a guardian of the child has), in relation to the child -
Section 16(1) of the NZ Act provides that the duties, powers, rights and responsibilities of a guardian of a child include "determining for or with the child, or helping the child to determine, questions about important matters affecting the child". Under s 16(2), "important matters affecting the child" include "changes to the child's place of residence (including, without limitation, changes of that kind arising from travel by the child) that may affect the child's relationship with his or her parents and guardians". Section 16(5) provides that, in exercising his or her duties, powers, rights and responsibilities, "a guardian .... must act jointly (in particular, by consulting wherever practicable with the aim of securing agreement) with any other guardians of the child".
Against this background, s 17 of the NZ Act (which is critical for this issue) provides, relevantly:
Section 18 of the NZ Act contains a special provision under which a father, who is not otherwise a guardian of the child, becomes a guardian if his particulars are registered on the child's birth certificate. In the present case, the father's particulars were registered on the birth certificate of K. So much is uncontested (at 392 ). However, because K was born before the commencement of the NZ Act in 2004, neither s 18 nor s 17(2) of the NZ Act is applicable.
It follows that the determination of whether or not the mother was, according to New Zealand law, the sole guardian of K is to be determined in accordance with the formula stated in s 17(3) of the NZ Act. It was for that reason (not apparently perceived by the mother at the time of making her first affidavit) that s 17(3)(b) became critical, with its reference to the character of the relationship between the mother and the father "at the time the child was born".
Two other provisions of the NZ Act must be noted in deciding whether it was shown that the father was a joint guardian of K or otherwise held "rights of custody" in relation to him. The first is s 3 of the NZ Act, which states that the purpose of that Act is to "promote children's welfare and best interests, and facilitate their development, by helping to ensure that appropriate arrangements are in place for their guardianship and care".
In addition, s 80 of the NZ Act imposes criminal responsibility on parents who remove specified children from New Zealand without leave of a relevant court. That section states:
Every person commits an offence and is liable on summary conviction to a fine not exceeding $2,500, or to imprisonment for a term not exceeding 3 months, or to both, who, without the leave of the Court, takes or attempts to take any child out of New Zealand -
The NZ Interpretation Act: The majority in the Full Court invoked s 29A of the Interpretation Act 1999 (NZ) in an attempt to shed some light on the meaning of the expression "de facto partner" in s 17(3)(b) of the NZ Act. Section 29A provides, relevantly:
The mother's complaint in this Court, as in the Full Court, was not just that the evidence before the primary judge fell short of establishing that she and the father were in fact living together at the time that the child was born, but also that their relationship at that time was not that of "de facto partners". Because a "de facto partner" in the NZ Act is necessarily a partner in a "de facto relationship", the provisions of s 29A of the Interpretation Act afford some indication of the type of relationship which the New Zealand Parliament had in contemplation when, in s 17(3)(b) of the NZ Act, it made that status a relevant criterion. It is significant that s 29A(3) directs attention, in determining whether a de facto partnership existed, to the purpose of the law expressed in ss 17(1) and 17(3)(b) and the circumstances of the relationship in question.
In any event, the expression "de facto partner" has now moved into common parlance in Australia. Permissible judicial notice affirms that New Zealand society is sufficiently similar to our own to allow Australian understandings of that expression to be applied, in a general sense, to the meaning of that phrase in the NZ Act.
Conclusion on the NZ legislation
The general rule, established by s 17(1) of the NZ Act, is that a father and mother are to be joint guardians of their child. That is the principle endorsed by the New Zealand Parliament in a statutory provision that was in force at the time of K's removal to Australia and at the time the present proceedings were heard and decided. It follows that upholding the mother's claim to be K's "sole guardian" requires the application of the exception in s 17(3) of the NZ Act. Neither the primary judge nor the Full Court was convinced that the exception in s 17(3) was applicable in this case.
Differences in the Full Court
Evidence of de facto relationship
The majority in the Full Court accepted that it was necessary first to establish precisely the rights which the father had under New Zealand law at the time of K's removal, and then to resolve the question of whether such rights amounted to "rights of custody" for the purposes of the Regulations (at 391 ).
Having examined the foregoing provisions of the Regulations (and as far as relevant the Convention), as well as the statutory provisions applicable in New Zealand, the majority concluded that there was "just" sufficient evidence to support a finding that "at the time of K's birth the mother was living with the father as 'a de facto partner'" (at 396 ). They therefore accepted that both parents were K's "guardians" at the time of his removal from New Zealand. As such, they both had the right to determine K's place of residence. The father thus had "rights of custody" at the relevant time. This rendered the mother's unilateral act of removal "wrongful". The orders of the trial judge could thus be sustained on this basis.
In her dissenting reasons, Finn J called particular attention to the limitations of the evidence of the father as to the character of his relationship with the mother "at the time the child was born". Her Honour laid emphasis on her conclusion that "it fell to the [State] Central Authority to put before [the primary judge] the necessary evidence to establish the existence of a de facto relationship": (2007) 211 FLR 357 at 367 . In the opinion of Finn J, the Authority had failed to do so (at 367 ).
Notice of contention point
In the Full Court, the issue now raised on the notice of contention was also put in the alternative by the State Central Authority (at 375 -). Having reviewed the case law, the majority concluded that, if it could be shown that the father had a right to veto K's removal from New Zealand under the law of that jurisdiction, then such removal was in breach of "rights of custody" and was therefore "wrongful". However, their Honours then turned to the question of whether a relevant right of veto arose pursuant to the father's "guardianship" of K. In light of their conclusion on that matter, they did not need to consider the argument that the father had a right of veto independent of such "guardianship".
In her reasons, Finn J accepted that the "right to determine the place of residence" of a child would include a "right of veto" (at 368 ). However, her Honour did not accept that the terms of the consent order, made by the New Zealand Family Court in December 2000 and still in force, "gave the father the right to determine the child's place of residence or to veto a decision concerning that matter (on the assumption .... that he was not a guardian of the child)" (at 368 ). In particular, Finn J was of the opinion that the expert evidence on New Zealand law, provided in an affidavit of the father's solicitor, fell short of establishing the existence of the propounded "veto".
The removal was wrongful as in breach of guardianship rights
Sufficiency of the evidence
As noted above, the mother's grounds of appeal in this Court focused on the suggested inadequacies of the evidence accepted below to support the conclusion that, under New Zealand law, the father was a joint guardian of K, and thus entitled, in law, to determine K's place of residence, evidencing rights of custody.
The evidence that was before the Family Court included the father's assertion of his guardianship rights, jointly with the mother. The history of the earlier proceedings before the New Zealand courts, and other evidence in the proceedings, demonstrated the continuous significance to the father of his relationship with K. Both before the New Zealand courts and in these proceedings in Australia, the father made it plain that he asserted such rights not merely for the enjoyment of personal contact with K but also to ensure that K would "discover his Maori heritage". In his second affidavit, the father adverted to this issue and gave emphasis to its importance for him. This is a particular feature of the present case that a New Zealand court would ordinarily be better able to evaluate, on tested evidence, than an Australian court. During argument before this Court, the possible need for sensitivity to this question by the Family Court of Australia was properly conceded for the mother:  HCATrans 795 at 27 - .
Onus and burden of proof
A question immediately arises as to which party before the Family Court bore the evidentiary onus of establishing that s 17(3)(b) of the NZ Act was, or was not, engaged.
For the mother, it was argued that the issue only arose so far as the State Central Authority was attempting to prove a breach of the father's "rights of custody" in relation to K. If the father wished to prove such "rights of custody", by reason of his status under New Zealand law as a joint guardian of K, it was for him to prove that he possessed such status. Effectively, this meant that the father, by proper evidence and argument, had to exclude the conclusion that the mother was "the sole guardian of the child" pursuant to s 17(3) of the NZ Act.
The mother further submitted that the proceedings in the Family Court comprised a "civil proceeding" and were thus subject to s 140 of the Evidence Act 1995 (Cth) ("the Evidence Act"). By that section, relevantly, the court "must find the case of a party proved if it is satisfied that the case has been proved on the balance of probabilities".
Several considerations are identified in the Evidence Act as matters that the court may take into account in deciding whether it is so satisfied. These include "the nature of the subject-matter of the proceeding". Here, that subject matter is not exclusively an inter partes civil proceeding between private individuals. It is a proceeding ostensibly brought to uphold the high purposes of the Convention and to ensure that Australia conforms to its obligations under international law. By ensuring the prompt return of a child wrongfully removed and by upholding the rights of custody and access under the law of another Contracting State, Australia gains the benefits of "reciprocity and mutual respect" between convention countries, of which New Zealand is one.
In support of her argument about the onus of proof, the mother could point to the fact that reg 16(1)(c) of the Regulations is expressed in language that appears to recognise that the applicant Central Authority must establish the element of wrongfulness required to secure a return order.
The State Central Authority did not contest that it bore the general legal obligation to satisfy the Family Court that K's removal from New Zealand to Australia was wrongful, in the sense that it breached rights of custody enjoyed by the father under the law of New Zealand. However, the Authority submitted that the primary rule established by s 17(1) of the NZ Act established the prima facie position under New Zealand law, being that both the father and the mother were joint guardians of K. In this respect, the NZ Act reflected an advance in New Zealand comparable to similar developments encouraging shared parental responsibility under Australian law. It expressed an important legislative purpose of ensuring that, in the usual case, both parents would be involved in the privileges and responsibilities of the guardianship of their child.
According to this argument, if an exception to this general rule of joint guardianship were to be proved in a particular case (at least where the other parent contested the exception and was willing to accept the duties of joint guardian), the evidentiary or forensic onus of demonstrating the application of the exception had to be borne by the party asserting it.
There are arguments both ways on the locus of the evidentiary burden of demonstrating the consequence of s 17(3) of the NZ Act or its inapplicability. Having regard to the overall purposes of the NZ Act, as declared in s 3, and its embrace of a presumption of joint guardianship signified in s 17(1), it seems appropriate to require a mother, who claims exceptionally to be the "sole guardian of [a] child", to prove a legal entitlement to that effect. In particular, this is so where the mother's object in claiming sole guardianship is to preclude the engagement of the Regulations before an Australian court, with their object of committing such decisions, by prompt determination, to the courts or authorities of the country of habitual residence from which the child was removed. To some extent it is invidious to expect an Australian court to elucidate the way in which a New Zealand court, faced with the present question, would assign the burden of persuasion.
It is possible to put this issue to one side in this appeal because, alike with the majority in the Full Court, it is my view that (assuming that the relevant evidentiary burden was borne by the State Central Authority) sufficient evidence was adduced to sustain the decision and orders of the Full Court. No error has been demonstrated to this Court that would warrant reversing that Court's decision.
Deficiencies of the evidence
I acknowledge that the available evidence addressed to the relevant questions was not entirely satisfactory. There was almost no evidence from the father illuminating, in proper detail, the nature of his relationship with the mother "at the time the child was born" (NZ Act, s 17(3)(b)). It must be remembered, however, that the father was not, as such, a party to the proceedings. It was left to the State Central Authority to advance the claim asserted by the NZ Central Authority on behalf of the father for a return order in respect of K. For this reason, it is difficult to apply to the present circumstances the ordinary expectations of adversarial litigation, viz that inferences may be drawn adverse to a party in the best position to call a witness who could have given direct evidence when that party has refrained from tendering that evidence or asking crucial questions.
It is reasonable for the mother to suggest that it would have been a "light burden" upon the State Central Authority to have procured more elaborate evidence from the father than it did. It was all but conceded in argument that, especially once the mother had filed her second affidavit, the State Central Authority should have procured additional evidence to illuminate the extent and character of the "relationship" between the father and the mother at the time the child was born. As the Full Court majority stated, if the father had deposed that he was living "as a de facto partner" of the mother at the relevant time, such evidence would arguably have involved the assertion of a conclusion of law, liable to be excluded for that reason ((2007) 211 FLR 357 at 394 ). But it would have been open to him to give and tender evidence as to relevant features of the relationship and his perceptions about its character. Had the Authority adduced such evidence, much less might have been left to judicial inference.
I therefore agree with Gummow, Heydon and Crennan JJ ("the joint reasons") that both the State Central Authority, and perhaps the Family Court, could have taken steps to supplement the factual record. In the result, however, it is arguable that their failure to do so was more disadvantageous to the father than it was to the mother. The mother was afforded an ample chance to respond to the father's claim that she had lived with him in a de facto relationship. The contradictions that emerged in the evidence which she gave were of her own making. The father, on the other hand, had no practical scope for responding to the mother's ultimate denial of the existence of such a relationship, given that the relevant affidavit was filed immediately before the hearing.
Also unsatisfactory was expert evidence tendered by the State Central Authority concerning the content and operation of New Zealand law on de facto partnerships at the relevant time. The affidavit of the father's solicitor, which addressed these matters, was defective. It was neither entirely independent nor complete. But it was the only such evidence available.
The evidence of a de facto partnership was adequate
Notwithstanding these deficiencies, which the primary judge and the majority in the Full Court acknowledged, no error has ultimately been shown to warrant reversal by this Court of the conclusion of the Full Court that, as a matter of fact, a relationship between the father and mother at the time of the child's birth was proved and that its character was shown to be that of a de facto partnership. Several factors help to sustain this conclusion.
Concurrent findings of fact
There was explicit evidence from the father that he lived together with the mother at the time of K's birth. This provided a foundation for the conclusion that the parents in fact lived together at the time identified as material by s 17(3) of the NZ Act. Acting within his powers, the trial judge accepted that evidence. The Full Court, exercising its own powers of fact-finding in an appeal by way of rehearing, affirmed such acceptance. The Authority thus has the benefit of concurrent findings of fact on this point. For this Court now to take the exceptional course of giving effect to a contrary conclusion, it would have to identify error. Simply reaching a different conclusion of its own is insufficient. The facility of appeal would be undermined, and much instability produced, if appellate courts substituted their own factual conclusions for those of trial courts without the necessity of demonstrated error. Whilst some of these considerations have less significance in the present proceedings, because all of the evidence was received in written form with no oral testimony or cross-examination, the point of principle remains. Special caution in disturbing factual conclusions reached below is required in this Court because the constitutional process of appeal has been classified as that of a strict appeal. In essence, this rule derives from the special responsibilities of this Court as a final national and constitutional tribunal concerned with correcting established error and deciding appeals, not, as such, conducting or reconducting trials.
Evidence of the mother
The mother's express contradiction of the father's evidence of cohabitation at the time of K's birth - and in particular her express denial, on the eve of the hearing, that their relationship was that of "de facto partners" and her explicit assertion that K was "conceived from [a] brief encounter with [the father]" - obliged the judges below to decide whose evidence was to be preferred. Clearly, it was open to the primary judge, having regard to all of the evidence, to prefer the father's version to that of the mother. Finn J recognised that the primary judge ((2007) 211 FLR 357 at 366 ):
was entitled for the reasons which he gave, to prefer the evidence of the father to that of the mother at least in relation to the issue of whether the parties were living together when the child was born. Indeed he might well have added that the mother's own evidence was inconsistent.
Once the fact of living together at the time made relevant by s 17(3) of the NZ Act was established (effectively by unanimous conclusions of the judges below), the characterisation of the relationship so proved must be determined on that footing.
If it is accepted that the father and the mother were living together at the time of the child's birth, the mother's later evidence to the opposite effect must be rejected. Essentially, this is because it was inconsistent with what she had originally said. Inescapably, an inference arises that such inconsistency was the result of the mother's belated discovery of the legal significance of the facts, not appreciated when the focus of her attention was on resisting a return order on other bases.
In this Court, the mother complained of a failure on the part of the courts below to particularise the "inconsistencies" alleged to compromise her evidence. However, there was a clear and obvious contradiction between the mother's first and second affidavits. The first affidavit carried an implication that the mother was living with the father both before and after the birth of K. In this Court, counsel for the mother attempted to argue that the relevant passage from the first affidavit (extracted above) could have referred to the period of intermittent cohabitation described in the second affidavit alone. Although this might be true in a technical or linguistic sense, a number of factors tell against it as a reasonable interpretation:
The mother's reference to what she described as "our house" implies a more permanent and substantial domestic arrangement than that described in the second affidavit;
The manner of expression of the mother's claim that she "moved out" of "our house" and "went to live with [her] parents" two months after the birth of K is difficult to reconcile with her later assertion that she was, in effect, "living" with them even while spending significant periods with the father on a trial basis;
The mother asserted that "[t]his is why I moved out" of "our house", and the word "this" seems to refer to all of the preceding material in the paragraph, which encompassed periods both before and after the birth of K; and
The mother herself made no attempt to reconcile the arrangements described in her second affidavit with what she had said in her first affidavit. There was no cross-referencing of events as between the two.
The mother's submission in this regard also failed to take into account the fact that the father had made a direct assertion of cohabitation at the time of K's birth. This being the case, the mother's failure to contradict it in her first affidavit, if it was indeed false as she later alleged, would have been puzzling. That she declined to do so, and indeed adduced evidence which was, on its face, consistent and reconcilable with the fact of cohabitation, strengthens the inference that the mother's allegations of abuse were premised on a factual matrix in which the parents were cohabiting at the time of the birth of K.
If the mother's evidence on the central issue concerning the fact of living together with the father is rejected as "inconsistent", the question arises: why should her evidence be accepted as it related to the character of their relationship? If she would give unreliable evidence as to the fact and existence of cohabitation at the relevant time, might she not be equally liable to give incorrect evidence about the character and features of her relationship with the father?
Confirmatory evidence in support
There are several objective considerations that tend to support the inference that a de facto partnership subsisted between the father and the mother at the time of K's birth. None is conclusive. But together, such indications call into question the mother's suggestion, in her second affidavit, that K was conceived as a result of no more than a chance event resulting from an isolated sexual encounter.
K's name, which incorporates the father's surname and includes a Maori middle name, has some significance. It is apparent that, as the father deposed, he had a significant role in naming his son. This is not easy to reconcile with the mother's evidence in her second affidavit, the implication of which is that, at the time of K's birth, the father was no more than a peripheral figure. The mother's affidavit evidence indicates that K continues to be known by his full birth name.
Moreover, the descriptions in the mother's first affidavit of alleged abuse, suffered whilst she was pregnant with and later breastfeeding K, appear to reflect a troubled but ongoing domestic relationship coinciding with the birth. The reference to the visits of strangers to what is described as "our house" is not only apparently inconsistent with the later denial of a relationship at about the time of the birth, but is also difficult to reconcile with the rejection of the existence of a de facto partnership. A very common feature of such partnerships is that the partners live together in a shared house which they describe as their own. The mother's complaints that "the phone was disconnected, power shut off, there was no food" and that her parents' visits were "restricted", reinforce the suggestion of an ongoing domestic relationship of this character.
In addition, as was noted, the mother's first affidavit "seems [to indicate that] the mother was looking to the father for financial support and considered she had some entitlement to information about his financial affairs, albeit she was aggrieved about his conduct in that regard": (2007) 211 FLR 357 at 395 .
Father's sustained assertion of rights
The State Central Authority relied on the father's long course of conduct following the birth of K to support his evidence as to the original existence of a de facto partnership. The affidavits and other documents before this Court make it clear that the father made dogged efforts, in the courts and otherwise, to maintain an ongoing relationship with K, at times, it would seem, in the face of obstructiveness on the part of the mother in this regard. The father deposed that he felt that it was important to establish links between K and his paternal relatives and their Maori traditions. The father's insistence on preserving a connection with his son tends to support the supposition that the son was the outcome of something more than the passing encounter that the mother described.
Approach of the Full Court
The mother complained about the overall approach of the majority in the Full Court to the question of establishing the existence of a de facto partnership at the time of K's birth. Specifically, she objected to the majority's statement that "it would seem appropriate to set a relatively low threshold when determining whether the parents of a child were living in a de facto relationship" (at 396 ). This complaint is connected with the question of whether the mother bore the burden of establishing the engagement of s 17(3) of the NZ Act. The impugned statement in the majority's reasons was said to follow from "the purpose of the legislation". Reflecting the object of the New Zealand Parliament, as expressed in s 17(1) of the NZ Act, the majority of the Full Court indicated that the approach they favoured "would ensure that the child has both natural parents as guardians". Their Honours pointed out that (at 396 ):
This would be consistent with the modern acceptance of the benefits children obtain from having both parents involved in their lives, regardless of whether the parents were married or not. This more contemporary approach can be seen in the New Zealand legislation itself, which has extended guardianship rights to all fathers of ex nuptial children whose name appears on their child's birth certificate.
This conclusion also reflects the modern tendency of Australian family law in this respect. In effect (at least in countries like Australia and New Zealand which share many social features in common), the law is adjusting to well-recognised features of contemporary human relationships, and specifically to accommodate the growing incidence of de facto partnerships without marriage to which children are born. Technically, the approach favoured by the majority draws support from the statement in s 17(1) of the NZ Act of the general rule of joint guardianship in New Zealand and the recognition that the sole guardian mother is an exception ("unless"). The mother's complaint about the approach of the majority of the Full Court is therefore unfounded.
Further complaints of the mother
In this Court, the mother pressed complaints also made in the Full Court about the degree of attention paid to particular items of evidence said to support her claim that she was not living with the father at the time of K's birth.
First, the mother claimed that insufficient emphasis had been placed on the evidence of what was described as a "hospital tag", which was annexed to the mother's second affidavit. The front of that tag appears to contain data relating to K, including the date ("20.9.96") and time of his birth. On the rear of the tag is affixed a computer-generated label containing the mother's name, her parents' address, and the date "29.10.96". The significance of this latter date is unclear. In her second affidavit, the mother averred that the tag "confirmed" that, on 29 October 1996, her place of residence was her parents' house. However, in this Court, it was argued for the mother that the tag comprised "at least prima facie evidence that the mother lived at her parents' address at the time the child was born" (emphasis added). Of itself, however, the tag is conclusive of nothing. The majority of the Full Court were correct to observe that (at 395 ):
The address on the nametag might have been the mother's address at the time of the child's birth, but there could also have been some other explanation. Furthermore, no explanation was provided as to why the date shown on the tag was more than a month after the date of birth of the baby.
Secondly, the mother complained about the failure of the majority in the Full Court to give proper weight to the affidavit of LG, filed on the same date as the mother's second affidavit. LG deposed that the mother had lived at her parents' house at the time of K's birth, and that she had visited the mother there. She also stated that she was "not aware that [the mother] had lived with [the father] at any time". However, as the majority in the Full Court pointed out, LG acknowledged that she was a long-term friend of the mother. In addition, her recollection of visiting the mother at her parents' house could have related to the period after she "moved out" of the father's house, not long after the birth of K (at 395 ). The majority in the Full Court were fully justified in attributing limited weight to LG's evidence.
Conclusion: no error is shown
The result is that the Full Court majority were correct (and it was certainly available to them) to prefer the version of events given by the mother herself before she was aware of the legal implications of that version. To the mother's complaint that to attribute falsehood to her without cross-examination involved procedural unfairness, there is an obvious answer. The criteria for this element of the State Central Authority's case were to be derived, substantially, from provisions of a public law of New Zealand. The need for prompt consideration of an application such as that founding these proceedings is expressly recognised both in the Convention and by the Regulations. Whilst reg 15(2) acknowledges that the priority to be given to such applications is such as "will ensure that the application is dealt with as quickly as a proper consideration of each matter relating to the application allows" and it is clear that particular care (and some delay) may be necessary where a return order is resisted on the basis of a "grave risk [of] physical or psychological harm", the overall scheme of the Convention and Regulations places a premium upon the prompt return of a child wrongfully removed.
If proceedings such as these were typically, or even regularly, to expand into fully fledged contests amounting to contested custody suits, this would operate to defeat the purposes of the Convention. It would reward abductors. It would impose very serious burdens on the parent or guardian left in the country of habitual residence. And it would effectively shift the locus of decision-making to the country of resort. But as reg 1A(2)(b) of the Regulations makes clear:
[T]he appropriate forum for resolving disputes between parents relating to a child's care, welfare and development is ordinarily the child's country of habitual residence.
In evaluating whether this Court should intervene in this appeal to correct allegedly erroneous fact-finding in the Family Court of Australia, it is appropriate that we should perform our functions in a way supportive of the stated purposes of the Regulations. Those regulations clearly envisage expeditious proceedings based on economical evidence. They do so in the knowledge that it will then be left to the courts (or other decision-makers) in the country of the child's habitual residence to resolve the substantive contest according to local law which, necessarily, such decision-makers are likely to know much better than we.
Result: Full Court's orders sustained
The result of this analysis is that no error has been demonstrated in the majority's reasons in the Full Court to warrant disturbance by this Court of the Full Court's orders. Whether or not the mother bore an evidentiary or forensic burden of establishing that the guardianship rights of the father fell to be determined under s 17(3) of the NZ Act rather than s 17(1), the evidence favoured (and certainly supported) the conclusion reached by the trial judge and by the majority in the Full Court.
Because of the unanimous finding of the Full Court that it was open to the trial judge to find that the father and mother were living together at the time of K's birth, the first element of s 17(3) was clearly satisfied. As to the second element, it was open to the majority, in an appeal by way of rehearing, to conclude that their relationship bore the character of a de facto partnership. The mother's change of her evidence between the first and second affidavits sustains the majority's preference for the mother's initial characterisation of the relationship with the father. The evidence, and especially the mother's reference to "strangers" visiting "our house", supports the inference drawn by the judges that a de facto partnership existed at that time.
Once this point is reached, the guardianship rights of the father are to be determined by the primary rule stated in s 17(1) of the NZ Act and not the exceptional rule stated in s 17(3)(b). The mother and the father were therefore joint guardians of their child, K. On that basis, the father was entitled to decide, jointly with the mother, questions concerning K's place of residence. The mother did not contest that a finding to this effect would render inescapable the conclusion that the father enjoyed "rights of custody" in relation to K.
It follows that the mother's unilateral removal of K from New Zealand was in breach of the father's rights of custody. The removal was therefore wrongful. The Full Court was right to affirm the return order made by the primary judge. This Court should dismiss the appeal from the Full Court's orders.
The removal was also wrongful as in breach of the NZ court order
Terms of the NZ court order
The alternative issue, raised by the State Central Authority's notice of contention, can be dealt with more briefly. I am prepared to consider it because of the conceded imperfections in the direct evidence relating to the appeal issue, the availability of different judicial responses to that evidence, and the importance of the matters of principle argued on the contention, not least for the future welfare of K. Also relevant, in this regard, is the essentially public character of proceedings under the Regulations, which are designed to give effect to Australia's national obligations under the Convention.
The focus of the contention point was the NZ court order dated 4 December 2000 and still in force. The precise terms of that order are extracted in the joint reasons. The NZ court order was always at the forefront of the father's complaint. It was specifically referred to as the first item in the application by the State Central Authority to found the father's "rights of custody". That application stated:
[The father] has an access order made by the Family Court at Auckland in his favour which provides that the child be in his father's care every second weekend.
The application also asserted that the close relationship between the father and the child had been protected by the NZ court order and would be renewed if compliance with the terms of that order were restored.
The father also referred to the NZ court order in his first affidavit, to which the order was annexure A.
The order, made by the Family Court of New Zealand, is expressed to replace an earlier "custody access order". It sets out a detailed regime by which K was to be "in his father's care every second weekend [and] half the school holidays". "At all other times" he was to be "in his mother's care." On the face of things, therefore, the mother and the father were to enjoy intermittent custody ("care") of K at nominated and successive times, as defined in the order. According to the record, the order was entered by consent of the mother and the father. The mother referred to it, and the fact of her consent, in her second affidavit, to which the order was annexed. With respect, I cannot agree with the suggestion that doubt might have attended the state of her knowledge about the order at the point when she left New Zealand with K.
Upon one view, the foregoing facts serve to reinforce the conclusion, already reached, that the mother and father acted for many years as K's joint "guardians". However, by its notice of contention, the State Central Authority submits that, if the father had a right to object to the removal of K from New Zealand, because of the NZ court order read with other New Zealand law, such removal would additionally be wrongful under the Regulations. On this footing, so the State Central Authority argued, the Full Court should have decided that, as a fact, the father had a right to object to K's removal from New Zealand, which right (a "right of veto") had the effect of rendering K's removal from New Zealand wrongful.
Distinguishing custody and access
The mother submitted that the Regulations drew a distinction, reflected in the Convention, between breach of rights of custody and breach of rights of access or visitation. This may be so. However, in each case, the classification of such rights depends on all of the circumstances. Here, the intermittent arrangement, under the NZ court order which expressed the father's rights, envisaged quite extensive periods at weekends and on holidays where, of necessity, the father would enjoy undisputed custody of K whilst the child was in his exclusive care, and the right during such times to decide K's place of residence.
The definition of "rights of custody" in reg 4 of the Regulations recognises that it is sufficient that such "custody" exists "in relation to" the child. It is not confined to custody "of" or "over" the child. Moreover, reg 4(2) makes it clear that "rights of custody" include "rights relating to the care of the person of the child". This the father certainly enjoyed at the times identified in the NZ court order. During the identified times, the father indisputably had the "right to determine the place of residence of the child". Moreover, that right is not the primary criterion for the type of "rights of custody" that engage the Regulations. The Regulations (and the Convention) recognise that "rights of custody" may arise by operation of law, judicial decision or agreement. Given particularly the specificity of the NZ court order relating to commitment of K to his father's care over weekends, the inference is inescapable that it was intended that the order would be complied with within New Zealand. Save for exceptional occasions where the parents agreed between themselves, or where leave was so provided by the New Zealand court, the child was to remain in New Zealand where, alone, the terms of the NZ court order could be fulfilled.
The mother's removal of K from New Zealand, and his retention thereafter in Australia, deprived the father of his rights under the NZ court order. Effecting that removal without any leave of the New Zealand court, and obviously against the father's wishes, was a breach of rights under that order. The order gave the father exclusive rights of "care" during the nominated periods. Moreover, it impliedly gave the father a right to veto the mother's unilateral alteration of the child's place of residence from New Zealand to Australia where, necessarily, the NZ court order could not be fulfilled according to its tenor.
Rights of custody by court order
As the beneficiary of the NZ court order, the father therefore had "rights of custody" of the kind referred to in reg 4(2) of the Regulations. The right under a court order to refuse consent to removal (in effect, a "right of veto" over a change of residence to another country) has been recognised in courts of high authority as amounting to "rights of custody" within the meaning of Art 5 of the Convention, and hence within the terms of reg 4(2).
In In re D (A Child) (Abduction: Rights of Custody)  1 AC 619, Lord Hope of Craighead explained why this was so. As his Lordship pointed out, the words "rights of custody" are used in the context of the Convention to "define the circumstances in which the removal or retention of a child is to be considered wrongful - 'wrongful' because the Convention proceeds on the assumption that welfare issues are best dealt with in the state where the child is habitually resident" (at 626 ). His Lordship went on at 626 :
A right to object to the child's removal to another country is as much a right of custody, for [Convention] purposes, as a right to determine where the child is to live within the country of its residence.
This analysis is clearly correct. In giving effect to the Regulations, designed to fulfil Australia's obligations under the Convention, there are strong reasons why this Court should adopt the same construction - in particular where it represents an interpretation designed to fulfil the purposes of an international treaty addressed to a major international problem rendered more urgent by the advances in the modern means of travel.
Result: the removal was "wrongful"
The foregoing reasoning does not produce a surprising result. The record shows that, over many years, the father was vigilant to assert and protect his "care" rights in the New Zealand courts. The mother's sudden removal of K from New Zealand, and her retention of him in Australia, breached the father's rights as stated in the NZ court order, to which each of the parents had agreed. Effectively, the removal of K to Australia rendered those rights nugatory. Such rights are "rights of custody" both on their face and by their operation. It follows that the mother's breach of the father's rights was also, on this basis, "wrongful". Separately, it sustains the decision and orders of the Full Court.
Conclusion and importance of the Convention
Return to New Zealand courts
It is important to recognise that all that is involved in the determination made by the Full Court is that K should be returned to New Zealand, which is accepted to have been his place of "habitual residence" before he was unilaterally removed to Australia by his mother. It would then be for the independent courts of that country, on the basis of properly tested evidence, to adjudicate the competing claims of the father and mother in respect of parenting, guardianship, care, custody and access in relation to the child.
No one is suggesting "dumping" the child on the courts of New Zealand as "another country's problem". What the Convention and Regulations envisage is that the child should be returned, following orderly consideration by the local courts, to the New Zealand courts. Those courts were earlier seised of the issues and are best placed to resolve them. The patriotic language of the United States courts, cited in the joint reasons, does not reflect the obligations stated in the Regulations or the law applicable in this country. Nor is it a fair description of what has happened in these proceedings or what the father seeks.
Reservations that might sometimes, exceptionally, arise about returning a child for decision-making in the courts or other bodies of convention countries can have no application whatever in the case of the courts of New Zealand. Indeed, the "Maori heritage" of the father and of K arguably reinforces the conclusion that, once "grave harm" and other such grounds for refusing a return order are put aside, as here they must be, the Regulations ought to be given their intended effect. The verb used in the material regulation is imperative ("must"). The proceedings in Australia have already long delayed the return of K to New Zealand in conformity with the father's custody rights and the NZ court order. Further delay should cease.
The importance of the Convention
Within the international community, specifically within knowledgeable legal circles, concern has been expressed about the non-ratification of the Convention by some states  and, where it is ratified, the effective reopening of the merits of custody disputes by foreign courts in proceedings brought to vindicate the Convention.
In DP at 449 , I attempted to explain that "[u]nless Australian courts, including this Court, uphold the spirit and the letter of the Convention as it is rendered part of Australian law by the Regulations, a large international enterprise of great importance for the welfare of children generally will be frustrated in the case of this country".
Unfortunately, in not a single case in which the Convention and Regulations have come before this Court has the Court upheld a decision of the Full Court of the Family Court of Australia ordering the return of an abducted child. This Court corrected a slip in De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640. It required reconsideration by the Full Court of the Family Court. But in JLM v Director-General NSW Department of Community Services (2001) 206 CLR 401, in the DP case and now in the present appeal, the Court has been divided. On each occasion, Gleeson CJ and I have favoured affirming the decisions of the Full Court. However, a majority has found error and set aside the Full Court's orders for return of the child to the country of habitual residence. In the result, the objective of the Convention has been defeated or delayed. Australian courts have assumed a fact-finding role which, in my view, the Convention, and the Regulations, commit to the courts of the country from which the child was taken.
With all respect to those of a different view, it is important for judicial attitudes to be adjusted in such cases or the Convention (ratified by Australia for high national and international purposes) will lose much of its efficacy so far as the courts of this country are concerned. In the three cases mentioned, it is my opinion that the approach and orders of the Full Court of the Family Court were correct. But Australian judges in the courts below will read and draw inferences from this Court's majority opinions, which have uniformly been to the contrary effect.
At least in De L, DP and JLM, the orders of this Court left open the compliance by Australian authorities with the letter and spirit of the Convention. They did so by recommitting the ultimate decision to the Family Court. In the present appeal, the result of the majority's orders is that even this will not occur. The specialist Family Court of Australia will be deprived of the function of discharging its duties in respect of the Convention and Regulations, taking into account the legal errors said to have occurred. The invocation of the Convention by the father and the New Zealand Central Authority, in the case of Australia, will simply be terminated.
I do not agree with this outcome. When mutuality between convention countries breaks down, the Convention's arrangements are likely to be defeated. Abduction is rewarded. The ultimate victims are the children.
The appeal should be dismissed with costs.
 Wenceslas v Director-General, Department of Community Services (2007) 211 FLR 357.
 Marriage Act 1961 (Cth), s 50(4) and Marriage Regulations 1963 (Cth), reg 42(2)(a); see also, for example, Births, Deaths and Marriages Registration Act 1995 (NSW), s 33.
 Births, Deaths, and Marriages Registration Act 1995 (NZ), ss 53, 62A.
 See, for example, Green v Green (1989) 17 NSWLR 343.
 Statutory Rules 1986 No 85 as amended.
 Reported as Wenceslas v Director-General, Department of Community Services (2007) 211 FLR 357.
 Child Abduction and Custody Act 1985 (UK), s 1.
 Guardianship Amendment Act 1991 (NZ).
 This is so defined in reg 2(1) as to include, by reference to s 39 of the Act, a range of federal, State and Territory courts.
 cf DP v Commonwealth Central Authority (2001) 206 CLR 401 at 426 .
 See In the Matter of M (Children)  EWCA Civ 992 at .
 The reference in the first sentence to the breach of rights of "access" appears to be per incuriam.
 See Marion's Case (1992) 175 CLR 218; P v P (1994) 181 CLR 583.
 De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640 at 657; AMS v AIF (1999) 199 CLR 160 at 168-169 ; Minister for Immigration and Multicultural and Indigenous Affairs v B (2004) 219 CLR 365 at 401-402 
 McKee v McKee  AC 352 at 365.
 P v P (1994) 181 CLR 583 at 598.
 Marion's Case (1992) 175 CLR 218 at 256, 294, 318.
 Beevers and Peréz Milla, "Child Abduction: Convention 'Rights of Custody' - Who Decides? An Anglo-Spanish Perspective", (2007) 3 Journal of Private International Law 201 at 202.
 Regulation 16(3) empowers a court to refuse to make an order for return if any one or more of pars (a)-(d) are established by a person opposing return. Paragraphs (a)-(d) state:
 The discretion may be exercised in an appropriate case by ordering return but upon attached conditions: DP v Commonwealth Central Authority (2001) 206 CLR 401 at 417 , 456 . The value of attached conditions will depend upon their enforceability and proper foundation in the evidence: DP v Commonwealth Central Authority (2001) 206 CLR 401 at 425 - .
 The appeal was conducted on the footing that the appropriate form of the Regulations was that of the compilation prepared on 1 July 2006, taking into account amendments up to SLI 2006 No 139.
 Under reg 14(1).
 Minister for Immigration and Multicultural and Indigenous Affairs v Al Khafaji (2004) 219 CLR 664 at 672 ; Western Australian Planning Commission v Temwood Holdings Pty Ltd (2004) 221 CLR 30 at 47-48 - 
 See Thomson v Thomson  3 SCR 551 at 560-561; In re D (A Child) (Abduction: Rights of Custody)  1 AC 619 at 635.
 See In re D (A Child) (Abduction: Rights of Custody)  1 AC 619 at 627-628; Furnes v Reeves 362 F 3d 702 at 714-715 (11th Cir 2004).
 Section 80 reads:
Taking child from New Zealand
Every person commits an offence and is liable on summary conviction to a fine not exceeding $2,500, or to imprisonment for a term not exceeding 3 months, or to both, who, without the leave of the Court, takes or attempts to take any child out of New Zealand -
 To obtain from the Court of Chancery the writ of ne exeat the plaintiff had to show real ground for believing that the defendant was seeking to avoid the jurisdiction or for apprehending that if allowed to depart the plaintiff would be prejudiced in his remedy: Glover v Walters (1950) 80 CLR 172 at 176. Alimony decreed by an ecclesiastical court might be enforced in Chancery by the writ of ne exeat if the husband was about to quit the realm: Vandergucht v De Blaquiere (1838) 8 Sim 315 [59 ER 125].
 CSR Ltd v Cigna Insurance Australia Ltd (1997) 189 CLR 345 at 392; Cardile v LED Builders Pty Ltd (1999) 198 CLR 380 at 393; Batistatos v Roads and Traffic Authority (NSW) (2006) 226 CLR 256 at 265 
 The pendency of proceedings within the meaning of par (a) of s 80 may give rise to considerations discussed by La Forest J in Thomson, but par (a) was not shown to apply in this case.
 Cf Australian Iron & Steel Pty Ltd v Seco (1968) 117 CLR 342 at 346.
 De L v Director-General, New South Wales Department of Community Services [No 2] (1997) 190 CLR 207.
 Wenceslas v Director-General, Department of Community Services (2007) 211 FLR 357. In the practice of the Family Court of Australia, fictitious names are assigned to case titles in cases of this kind.
 See Family Law (Child Abduction Convention) Regulations 1986 (Cth), regs 8, 9.
  ATS 2. The Convention entered into force on 1 December 1983. Australia ratified it with effect from 1 January 1987. See DP v Commonwealth Central Authority (2001) 206 CLR 401 at 411 
 The Act, s 111B(1).
 In re D (A Child) (Abduction: Rights of Custody)  1 AC 619 at 631 . See (2007) 211 FLR 357 at 379 .
 In re D  1 AC 619 at 631 .
 Regulations, reg 16(1).
 Regulations, reg 4 giving effect to Convention, Arts 3 and 5(a).
 The Explanatory Report of Professor Elisa Pérez-Vera of April 1981 made it clear that the original parties to the Convention drew a distinction between rights of custody and rights of access and did not intend "mere rights of access" to entitle a parent to demand the summary return of the child. See In re D  1 AC 619 at 631 .
 Reasons of the primary judge at -; (2007) 211 FLR 357 at 397-399 -; cf Regulations, reg 16(1A)(b).
 Reasons of the primary judge at ; (2007) 211 FLR 357 at 400-403 -; cf Regulations, reg 16(3)(a).
 Reasons of the primary judge at -; cf Regulations, reg 16(3)(b).
 Reasons of the primary judge at -; cf Regulations, reg 16(3)(c).
 See below these reasons at .
 (2007) 211 FLR 357 at 396-397 -; cf at 368-369 -.
 Reasons of the primary judge at .
 Reasons of the primary judge at .
 cf reasons of Gummow, Heydon and Crennan JJ at .
 Reasons of the primary judge at .
 Reasons of the primary judge at .
 Pursuant to the Family Proceedings Act 1980 (NZ), s 11(2).
 Reasons of the primary judge at .
 Reasons of the primary judge at .
 Reasons of the primary judge at .
 cf Convention, Art 3. See Regulations, reg 2(2).
 cf Convention, Art 5.
 See eg reasons of the primary judge at -; (2007) 211 FLR 357 at 391-392 -.
 See Evidence and Procedure (New Zealand) Act 1994 (Cth), s 40.
 The majority noted that the primary judge had apparently found that the father and mother were joint guardians on the basis of "cohabitation" at the time of K's birth. This did not reflect the relevant legal criterion. However, as the proceedings had been conducted on the papers, the majority considered that it was open to them to substitute their own finding on this issue. See (2007) 211 FLR 357 at 394 -.
 Evidence Act, s 140(2)(b).
 Regulations, reg 1A. See also Convention, Art 1.
 See above these reasons at .
 See eg the Act, s 65DAC.
 Although the locus of the general legal onus is fixed by the terms of the Regulations, it is well established that the factual or forensic onus may shift on particular issues: Richard Walter Pty Ltd v Commissioner of Taxation (1996) 67 FCR 243 at 245-246, 259; Raftland Pty Ltd v Commissioner of Taxation (2006) 227 ALR 598 at 616 .
 See eg Regulations, reg 15(2); cf Convention, Arts 1a ("prompt return"), 2 ("most expeditious procedures available").
 cf Commercial Union Assurance Company of Australia Ltd v Ferrcom Pty Ltd (1991) 22 NSWLR 389 at 418-419.
 cf DP (2001) 206 CLR 401 at 456 .
 Joint reasons at .
 cf joint reasons at .
 cf New South Wales v Fahy (2007) 81 ALJR 1021 at 1026 , 1052 , 1056 ; 236 ALR 406 at 409, 445, 449; Roads and Traffic Authority (NSW) v Dederer (2007) 81 ALJR 1773 at 1777 , 1804-1805 -, 1827 ; 238 ALR 761 at 764, 800-801, 831.
 Eastman v The Queen (2000) 203 CLR 1 at 13 , 24 , 54 , 63 ; cf at 82 , 123 . See also Mickelberg v The Queen (1989) 167 CLR 259.
 As was remarked in the course of oral argument, that assertion, given its shortness, "rather [gave] the impression that at the time the application was prepared .... it was not thought that there was likely to be a serious issue about this question":  HCATrans 795 at 42 -. See also reasons of Gleeson CJ at .
 See joint reasons at .
 Convention, Arts 1a ("prompt"), 2 ("most expeditious"), 7 ("prompt"), 9 ("without delay"), 12 ("forthwith"). See also Regulations, reg 13(3) ("as soon as practicable"), 15(2) ("as quickly as a proper consideration .... allows"), 15(4) ("42 days"), 15(4)(b) ("as soon as practicable").
 Regulations, reg 15(2).
 Regulations, reg 16(3)(b).
 See above these reasons at .
 Joint reasons at .
 cf joint reasons at .
 Regulations, reg 4(3).
 Convention, Art 3.
 C v C (Abduction: Rights of Custody)  1 WLR 654 at 663 per Lord Donaldson of Lymington MR;  2 All ER 465 at 473; In re D  1 AC 619 at 626 - per Lord Hope of Craighead, 635  per Baroness Hale of Richmond.
 cf (2007) 211 FLR 357 at 388 . See also Povey v Qantas Airways Ltd (2005) 223 CLR 189 at 202 , 246 .
 cf joint reasons at .
 cf In re M (Children) (Abduction: Rights of Custody)  3 WLR 975 (a case concerning return of children from the United Kingdom to Zimbabwe).
 Regulations, reg 16(1).
 See Ong, "Parental Child Abduction in Singapore: The Experience of a Non-Convention Country", (2007) 21 International Journal of Law, Policy and the Family 220.
 De L (1996) 187 CLR 640 at 662-663, 689-690; DP (2001) 206 CLR 401 at 424 ; JLM (2001) 206 CLR 401 at 427 .
G O'L Reynolds SC with B R Kremer (instructed by Le Vaccaro Lawyers) for the appellant.
B W Walker SC with T Tockar (instructed by Department of Community Services) for the respondent.
all rights reserved