Bozin, Danielle --- "The Hague Child Abduction Convention's Grave Risk of Harm Exception: Traversing the Tightrope and Maintaining Balance between Comity and the Best Interests of the Child" [2016] UTasLawRw 3; (2016) 35(1) University of Tasmania Law Review 24

---

The Hague Child Abduction Convention’s Grave Risk of Harm Exception: Traversing the Tightrope and Maintaining Balance Between Comity and the Best Interests of the Child

DANIELLE BOZIN*

Abstract

This article provides a critical analysis of divergent judicial opinions about how Australian courts should interpret the ‘grave risk of harm’ exception in the Hague Convention on Civil Aspects of International Child Abduction. Conflicting views about the extent to which the exception to a child’s return should permit consideration of a child’s best interests may be a manifestation of the balancing act that must be performed during return proceedings. The Convention seeks to protect children from the harmful effects of international parental child abduction through prompt return, whilst also accommodating situations where non-return is justified on welfare considerations. Perhaps the Family Court’s restrictive interpretation of the exception is a display of the tightrope becoming unsteady; by swaying towards comity, the individual child’s welfare is sacrificed. The High Court has arguably managed to master the art of maintaining balance whilst traversing the tightrope, by expeditiously examining the child’s welfare and the potential consequences awaiting them if returned to their habitual residence.

I INTRODUCTION

This article critiques the Australian judiciary’s interpretation of the ‘grave risk of harm’ exception to a child’s return under the Hague Convention on Civil Aspects of International Child Abduction (‘the Convention’).^[1] When a child is abducted from another Convention country to Australia, a left-behind parent may utilise the Convention to seek their child’s return through the initiation of return proceedings in an Australian court. During return proceedings the abducting parent may raise several limited exceptions in an attempt to prevent the child’s return. One such exception provides that the court may refuse to order a child’s return if there is a grave risk that returning them to the place that was their habitual residence immediately prior to their abduction would expose them to physical or psychological harm, or otherwise place them in an intolerable situation.^[2] Australian case law reveals divergent judicial opinions about the extent to which the grave risk of harm exception should permit consideration of the individual child’s welfare. This article critiques the breadth of this exception’s application and suggests that a true state of balance, where the best interests of all children are attained, may realistically be difficult to achieve. However, the High Court’s broad approach to interpreting the exception is arguably justified on the basis that it goes some way to remedy the social implications arising from the feminisation of international parental child abduction.

The Family Court of Australia, along with dissenting High Court judges, have demonstrated a tendency to interpret the exception restrictively.^[3] In P v Commonwealth Central Authority,^[4] the Full Court of the Family Court said that there is a long line of authority both in Australia and internationally that the grave risk of harm exception is to be construed narrowly. Taking a narrow approach to the exception increases the ambit of the Convention’s reach by expanding the types of situations in which children will be returned under the Convention. It is also an outward manifestation of the value placed upon promoting comity between Convention countries, and a desire to protect the best interests of children generally through the action of prompt return.^[5] Accordingly, under this approach, consideration of the best interests of the individual child is limited during return proceedings.^[6] The Family Court’s approach has been supported by dissenting judges in the High Court of Australia, most notably Kirby J, on the basis that ‘[a]n overbroad interpretation of the exceptions would tend to undermine the achievement of the Convention’s core purposes and defeat its underlying policy.’^[7]

In the small number of Convention cases that have been heard by the High Court of Australia, the majority judges have consistently criticised the Family Court’s approach when interpreting the exceptions as being overly narrow or restrictive.^[8] In DP v Commonwealth Central Authority (‘DP’),^[9] the majority of the High Court said that establishing the grave risk of harm exception does not warrant the conclusion that the wording of the exception should be construed narrowly rather than broadly.^[10] If an abducting parent raises the grave risk of harm exception but is unsuccessful, there is no guarantee that post-return judicial proceedings will take place to determine the parenting dispute or examine any risk of harm concerns. Consequently, the High Court has emphasised that the fact that there ‘may’ be judicial proceedings post-return in the child’s habitual residence should not limit the applicability of the exception to the full extent that its language permits.^[11] This has been demonstrated by the High Court’s preparedness to examine the potential outcomes awaiting a child post-return, when determining the exception’s applicability during Convention return proceedings.

First, this article explains the balancing act that must be performed by the Australian judiciary when the grave risk of harm exception is raised during Convention return proceedings. Second, this article critiques several Family Court and all High Court judgments that concern the interpretation of the grave risk of harm exception. Finally, conclusions will be drawn about whether or not an appropriate balance between upholding comity among Convention countries, and protecting the individual child’s best interests and welfare, is achieved.

II BALANCING COMITY AND THE BEST INTERESTS OF THE INDIVIDUAL CHILD DURING RETURN PROCEEDINGS

The Convention’s core objective is to deter international parental child abduction and protect children from its harmful effects,^[12] after a parent has used ‘force to establish artificial jurisdictional links on an international level, with a view to obtaining custody.’^[13] Fostering comity^[14] between Convention countries by promptly returning children to their habitual residence is said to enliven the most appropriate forum by restoring the geographical status quo.^[15] The child’s habitual residence is considered to be the most appropriate forum in which to determine the substantive parenting dispute if the parties choose to seek to resolve it post-return.^[16] Comity has been aptly described as

a rule of choice of law, courtesy, politeness, convenience or good will between sovereigns, a moral necessity, expediency, reciprocity, or considerations of high international politics concerned with maintaining amicable and workable relationships between nations.^[17]

Restoring the status quo through prompt return is perceived as the ideal approach to thwart an abductor’s attempt to gain an unfair advantage, by forum shopping to obtain a more favourable custody arrangement in a different jurisdiction.^[18]

Restoration of the geographical status quo seeks to ensure that any decision regarding a child’s best interests is informed by moral and cultural assumptions appropriate to the child.^[19] Respect for a Convention country’s domestic laws requires that the judicial system of the Convention country to which a child is taken does not engage in determining parental responsibilities and rights. This is because in doing so a court would risk

expressing particular cultural, social etc. attitudes ... thus basically imposing their own subjective value judgements upon the national community from which the child has recently been snatched.^[20]

The Family Court, and dissenting judges in the High Court, have chosen to interpret the grave risk of harm exception narrowly as a way of promoting this rationale.^[21]

The promptness of the return process is, however, explicitly qualified in the sense that the abducting parent can raise several limited exceptions to a child’s return.^[22] These exceptions focus on ensuring that the individual child’s best interests inform the decision about whether or not to return the child in appropriate circumstances. The High Court of Australia has challenged the assumption that the return of a child under the Convention is return for the purpose of determining the parenting dispute (otherwise known as custody proceedings).^[23] The orthodox view of the High Court has been to recognise the reality of many returns by explaining that:

the content of the exceptions must be understood against the other provisions of the Regulations which ... make plain that there may be an order for return with no expectation that there will be any judicial process in the country to which the child will be returned in which any question about what is in the best interest of the child will be raised or addressed. ... the construction of the Regulations cannot proceed from a premise that they are designed to achieve return of children to the place of their habitual residence for the purpose of the courts of that jurisdiction conducting some hearing into what will be in that child’s best interests.^[24]

To what extent should an individual child’s best interests be examined during Convention return proceedings? Return proceedings have traditionally been viewed as summary in nature which in itself creates a practical dilemma. Findings of fact are often made ‘on the papers’ without the benefit of oral evidence and the cross-examination of witnesses on disputed facts. It can be problematic when an exception to the child’s return is raised by an abducting parent due to the limited ability to explore conflicting evidence.^[25] In MW v Director-General, Department of Community Services,^[26] the High Court recognised that, although these applications are typically dealt with via affidavit evidence without the benefit of cross-examination,^[27] the Convention’s prompt return policy does not preclude issues of disputed fact (including risk of harm concerns) from being examined through the expeditious giving of oral evidence, which is subject to cross-examination.^[28] Despite this, the extent to which a child’s best interests should be considered when interpreting the grave risk of harm exception during return proceedings remains contentious.

III INTERPRETING THE GRAVE RISK OF HARM EXCEPTION

For some time academics and commentators have explored the effects of the feminisation of international parental child abduction on the Convention’s operation.^[29] Since the Convention’s inception there has been a trend away from abducting non-custodial fathers to abducting primary-caregiving mothers.^[30] Abducting mothers are often principally motivated by: a need to flee domestic violence and/or child abuse; a desire to return to their homeland; a longing to return to family and social support networks; and a desire to improve their economic situation.^[31] This change in the gender dynamics underpinning abductions has had an impact on the operation and effectiveness of the Convention’s exceptions to return. Given that the grave risk of harm exception is most often raised in circumstances of alleged domestic and family violence, how we interpret this exception moving forward is particularly important.

The grave risk of harm exception requires the existence of a grave risk that the child’s return to their habitual residence would expose them to physical or psychological harm, or otherwise place them in an intolerable situation.^[32] The assessment relates specifically to the return of the child to the Convention country that was their habitual residence immediately prior to the abduction rather than the left-behind parent. The Convention does not define the gravity of risk required to successfully establish the exception. Courts have interpreted the degree of physical or psychological harm required as being restricted by the words ‘or otherwise place the child in an intolerable situation.’^[33] This means that the exception’s interpretation is restricted in the sense that the courts have required that the situation of grave risk must be intolerable. In Friedrich v Friedrich,^[34] the United States Court of Appeals provided two examples of circumstances which will qualify as grave risks of harm frequently cited by Australian courts.^[35] The first is where the return will place the child in imminent danger, such as returning the child to famine, disease or a war zone. The second is where the child will be subjected to serious neglect, abuse or extraordinary emotional dependence, and authorities in the child’s habitual residence are either incapable or unwilling to adequately protect the child.^[36]

The Australian judiciary’s consistent position prior to the High Court’s decision in DP in 2001 was that a grave risk could be appropriately dealt with in a child’s habitual residence post-return, once the Convention process is complete.^[37] This approach is most often applied on the basis that a child can be afforded protection by the relevant authorities post-return. Alternatively, an undertaking can be given by the left-behind parent promising, for example, not to perpetrate acts of domestic violence, to provide financial support, or to allow the abducting parent and child exclusive use of a residence.^[38] This restrictive approach protects the Convention’s objective of prompt return, and promotes comity between Convention countries. It means that during return proceedings the Court can avoid making determinations on the individual child’s best interests when interpreting the gravity of harm required.^[39]

In Bassi, DK and Director General of Community Services (‘Bassi’),^[40] the Family Court interpreted the exception narrowly to find that the degree of risk of physical or psychological harm deriving from family violence was not sufficiently grave to warrant non-return.^[41] This case concerned the abduction of two girls, aged 13 and 6, from the United Kingdom to Australia by their mother. There was a history of family violence and the father had been convicted of assault on the mother only one and a half months before the abduction. The severity of the violence was such that there was

sufficient material for the [C]ourt to reach the view that the husband engaged in violent, drunken, obsessional behaviour in the presence of the children and that he made threats to the life of their mother, the children and himself in their presence.^[42]

The wife alleged that on two occasions the husband had threatened her with a kitchen knife, and one of the children had intervened in an attempt to protect her. This resulted in the child’s hand being cut by the knife.^[43] This incident was indicative of direct harm to the children. The wife regarded

the husband’s threats to kill her as being not just a personal vendetta against her but also his cultural reaction to a situation where he would consider it necessary to kill her to protect his own dignity and family name.^[44]

The Court accepted affidavit evidence from witnesses in support of the mother’s claims:

[T]hat the husband drinks alcohol to excess, that he is of violent disposition and has frequently been violent, that he has actually physically injured the children in the course of physical attacks on the wife, that he has threatened to kill the children and that by reason of his cultural and social background his threats should be taken seriously.^[45]

However, the Court noted that the decision of the Full Court of the Family Court in Director-General of Family and Community Services v Davis^[46] is authority for the proposition that the degree of harm must be substantial and to a level comparable to an intolerable situation.^[47] This requisite degree of risk had not been established.

In assessing the gravity of the risk at hand the Court considered that, despite the wife’s allegations, she ‘continued to allow the children to visit the husband and his parents on weekends prior to her removal of the children.’^[48] Furthermore, one of the children had stated to the Family Report writer that she ‘did not believe her father would hurt her or N [her younger sibling] however, she believe[d] he would hurt her mother and this cause[d] her anxiety.’^[49] Importantly the Court also made reference to Murray v Director of Family Services ACT,^[50] in which the Full Court of the Family Court concluded that ‘it would be presumptuous and offensive in the extreme’^[51] to assume that a country such as New Zealand [the child’s habitual residence in that case] was unable to protect a child from a grave risk of harm upon return. In that case Nicholson CJ and Fogarty J expressed the view that welfare considerations are not relevant to return proceedings because the Court hearing the application is solely concerned with determining forum.^[52]

Ultimately, in Bassi, the Family Court exercised its discretion not to return the eldest child. However, the Court did this on the basis of a different exception: that the child objected to being returned and her maturity was such that her wishes should be considered.^[53] Discretion not to return was also exercised in relation to the youngest child. However, again not on the basis of there being a grave risk of harm, but rather that she would be placed in an intolerable situation if returned to England without her sibling.

Similarly, in Gsponer v Johnstone,^[54] the Family Court interpreted the grave risk of harm exception restrictively in the context of family violence. This restrictive interpretation was that the child’s welfare could be considered post-return in the child’s habitual residence once the Convention process was concluded. The mother had abducted her child from Switzerland to Australia. She submitted evidence that during the marriage she had been subjected to what was described as significant episodes of violence at the hands of her husband.^[55] She also claimed that the child had been the direct target of assault and mistreatment on several occasions.^[56] She argued that these circumstances constituted a grave risk of harm to the child. The Full Court of the Family Court stated that, once the child had been returned:

[N]o doubt the appropriate court in that country [Switzerland] will make whatever orders are then thought to be suitable for the future custody and general welfare of that child, including any interim orders.^[57]

The Family Court adopted the view that:

[C]ourts should not assume that once a child is returned, the courts in that country are not appropriately equipped to make suitable arrangements for the child’s welfare.^[58]

These cases reveal an inherent tension between the promotion of comity between Convention countries, and assessing a child’s best interests during return proceedings. Here the Family Court is resolving this incompatibility by facilitating comity and the best interests of children generally. This is being done on the basis that the individual child’s best interests can be reserved for consideration in the child’s habitual residence post-return.

The judiciousness of this approach is best explained by Kirby J, one of the dissenting judges in DP.^[59] His Honour agreed with the Family Court’s narrow interpretation of the exception whilst warning of the dangers of interpreting the exceptions to a child’s return too broadly. His Honour explained why comity should be promoted as the principal objective to safeguard the best interests of children generally by stating:

Unless 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. ... To the extent that Australian Courts, including this Court, do not fulfil the expectations expressed in the rigorous language of the Convention and the Regulations, but effectively reserve custody decisions to themselves, we should not be surprised if other countries, noting what we do, decline to extend to our Courts the kind of reciprocity and mutual respect which the Convention scheme puts in place. And that, most definitely, would not, in aggregate, be in the best interests of children generally and of Australian children in particular.^[60]

However, the majority of the High Court in DP said that greater weight should be placed upon whether a grave risk exists in fact.^[61] In this case, two actions (the appellants being DP and JLM) were heard concurrently because they both concerned the interpretation of the grave risk of harm exception. The High Court said that courts should assess the consequences of return when the grave risk of harm exception is raised, and that this requires ‘the kind of inquiry and prediction that will inevitably involve some consideration of the interests of the child.’^[62] In DP’s case,^[63] the abducting mother claimed that the child would be at grave risk if returned to Greece, because Greece lacked appropriate medical facilities to treat her son’s autism. In JLM’s case,^[64] the abducting mother claimed that the child would be at grave risk if returned to Mexico because she was suffering from a major depressive disorder. The child’s return could put her at serious risk of committing suicide.

In DP, the High Court majority overturned the Full Court of the Family Court’s preference for a restrictive interpretation of the grave risk of harm exception.^[65] In each case the Family Court had decided that the fact that judicial proceedings ‘may’ take place post-return, along with the father’s willingness to provide undertakings, addressed the contention of grave risk.^[66] On appeal the High Court explicitly said that the fact that there may be judicial proceedings in a child’s habitual residence does not in itself address the assertion of a grave risk of harm.^[67] Gaudron, Gummow and Hayne JJ stated:

What must be established is clearly identified: that there is a grave risk that the return of the child would expose the child to certain types of harm or otherwise place the child in ‘an intolerable situation’. That requires some prediction, based on the evidence, of what may happen if the child is returned. In a case where the person opposing return raises the exception, a court cannot avoid making that prediction by repeating that it is not for the courts of the country to which or in which a child has been removed or retained to inquire into the best interests of the child. The exception requires courts to make the kind of inquiry and prediction that will inevitably involve some consideration of the interests of the [individual] child.^[68]

In DP’s case, the High Court said that the issues that needed to be explored during return proceedings included: whether or not the child was at grave risk due to unavailability of appropriate and accessible facilities for the treatment of his autism; and what facilities were available in Greece, in particular the region where the child would be returned.^[69] Gleeson CJ asked:

[A]s a practical matter, what would be the circumstances in which the child and the mother would live upon return to Greece? How accessible would any facilities for treatment be?^[70]

The High Court emphasised that these questions should not be reserved for consideration post-return once the Convention process is complete, but rather answered during Convention return proceedings. In JLM’s case, the High Court said that it was not open to the Full Court of the Family Court to find that there was no evidence before the primary judge to support his Honour’s finding that a grave risk existed.^[71] The primary judge reached the conclusion that a grave risk existed in fact based on the evidence of a psychiatrist treating the mother, who had said that the mother had no will to live beyond the time when she had to hand the child to the father.^[72] In addition, the primary judge had relied on evidence given by another mental health professional, who was a friend of the mother, that there was a real risk that the mother would commit suicide if the child was returned to Mexico.^[73] There had been no challenge to this evidence. Neither had any additional evidence been presented before the Full Court of the Family Court. Consequently, the High Court said that the fact that the Full Court of the Family Court had not reviewed the evidence, and accepted or rejected it, meant that it was not open to conclude that the primary judge had erred in his findings.^[74] The High Court also expressed that the Full Court of the Family Court’s view that the mother was the source of the risk of harm did not reflect an understanding of her major depressive illness.^[75]

The High Court also considered whether undertakings given by the fathers in DP and JLM were adequate to address the claim that there was a grave risk of harm to the child if returned.^[76] In DP’s case the father undertook that he would not remove the child from the mother’s care until a court in Greece heard the custody matter.^[77] He also agreed that he would not enforce a pre-existing custody order that he had obtained from a Greek Court.^[78] Importantly the High Court questioned the adequacy and enforceability of undertakings given by left-behind parents explaining that:

For our part we gravely doubt the efficacy of an undertaking in this form. If the undertakings to be given by the father about his future conduct in Greece were to be enforceable, it would seem to have been necessary to suspend the order for return until production of evidence to the Family Court of the giving of undertakings by the father which would be enforceable in Greece at the suit of the mother.^[79]

In JLM’s case the father undertook to cooperate with the mother to ensure that custody proceedings took place in Mexico post-return.^[80] The High Court noted that the mother had submitted uncontested evidence that bribery may be a necessary prerequisite for success in such proceedings. Consequently, the High Court said that accepting such an undertaking as a condition for the child’s return was wrong.^[81] In both DP’s case and JLM’s case the majority of the High Court overturned the Full Court of the Family Court’s decision to order the child’s return based on a narrow interpretation of the grave risk of harm exception.^[82]

Despite the High Court’s decisions concerning DP and JLM in DP, the extent to which an individual child’s best interests should be examined during return proceedings still remains contentious. This may be because a degree of incompatibility exists between the promotion of comity between Convention countries, and assessing the individual child’s best interests during return proceedings. The divergent judicial opinions of the Family Court of Australia and the High Court of Australia, about the extent to which the grave risk of harm exception should permit consideration of a child’s best interests, may be a manifestation of the intricacies of the balancing act that must be performed by courts during Convention return proceedings. However, arguably, the High Court’s broad approach to interpreting the exception provides an appropriate response that goes some way to remedy the problematic social dynamics arising from the changing face of international parental child abduction

It is not surprising that the broad approach to interpreting the grave risk of harm exception advocated by the High Court in DP has not been applied consistently.^[83] Post DP, the Family Court has applied both a narrow and broad approach to interpreting the exception. For example, in State Central Authority, Secretary to the Department of Human Services v Mander,^[84] the parents’ relationship was characterised by a history of violence perpetrated in their children’s presence. The mother abducted the children from the United Kingdom to Australia. The Family Court at first instance held that the grave risk of harm exception was satisfied.^[85] The Court applied a broad interpretation of the exception and exercised judicial discretion to refuse to order the children’s return. This was despite Kay J noting that ‘the English legal system provides ample legal protection, and the English police and social services provide excellent care for battered women.’^[86] The Family Court applied DP whilst acknowledging its potential to produce outcomes that conflict with the Convention’s principle objectives. Kay J stated:

Although I found it a difficult case to come to grips with in light of the very strong underlying message within the Convention, ultimately I am satisfied of the existence of a grave risk of harm in this case. ... There have been years of sporadic violence in the presence of the children. It has necessitated constant court proceedings, and regular invocation of criminal sanctions. The problem persisted until the mother left England. I am confident a return to England would most likely lead to a continuation of the problems that have dogged these children for all of their lives in England. It is beyond argument the exposure of children to violence between their parents cannot be seen to be in the children's interests. I feel ... discomfort ... in light of the strong underlying currents of the Convention and the need to overcome the scourge of wrongful removal. But the High Court has reminded us on several occasions that the Convention is to be read as a whole. It is a Convention with exceptions.^[87]

Another post-DP example is State Central Authority v Papastavrou.^[88] In this case the Full Court of the Family Court was satisfied that the primary-carer mother had suffered very serious physical abuse at the hands of the father over a prolonged period of time.^[89] The Court also accepted that this past behaviour constituted a serious and weighty risk for the children in the future.^[90] Despite this the Court construed the exception narrowly, and held that the mother had not established the grave risk of harm exception.^[91] The Court explained that its decision was based on the conclusion that where and with whom the children should live, and whether or not the mother had good cause to leave the relationship, were all matters for the courts in the children’s habitual residence.^[92] The mother raised as a concern the ability of Greek authorities to respond appropriately to protect her and the children if there was another incident of family violence post-return.^[93] The Court accepted that she presented convincing evidence which established a prima facie case on this issue,^[94] and the Central Authority had not offered evidence in response.^[95] Despite this, the Full Court of the Family Court was satisfied by the father’s willingness to offer what were unenforceable undertakings. He agreed to permit the mother and children to have exclusive occupation of a flat, and to not enter the premises without her permission. He also undertook to pay expenses and not initiate criminal proceedings against the mother in Greece.^[96] Upon this basis the Court held that the mother had not established the grave risk of harm exception. The undertakings were relied upon because the Court said that the only period with which it was concerned was the time up until when the mother was able to put the father’s promises before a Greek court. This was even though the Court acknowledged that Greek courts do not have a similar system providing for the enforcement of these undertakings.^[97]

Again post DP, in HZ v State Central Authority,^[98] the Full Court of the Family Court held that the grave risk of harm exception was not established in the context of family violence. Throughout the parties’ marriage they had lived with their children in the paternal grandparents’ home in Greece. At the end of a 10-week holiday in Australia the mother informed the father that she would not be returning with the children. It was accepted that the mother and children had been subjected to constant violent and inappropriate behaviour. However, at first instance and on appeal, the Family Court held that the grave risk of harm exception was not established. The Full Court of the Family Court explained that:

Greece was clearly the appropriate forum for issues relating to the welfare of the children to be determined. In the circumstances, it was appropriate for the trial judge to place significant weight on the first of the objects of the Convention, namely the prompt return of the children who had been wrongfully retained in Australia.^[99]

The Court explored the international jurisprudence on cases with similar facts of family violence. The Court noted that non-return orders were only made when the facts were very compelling, and determined that there was no clear statement of principle.^[100] Despite the authority of DP, ultimately the fact that the children did not have to return to the grandparents’ home, and could in principle seek protection under Greek law, was determinative. Whether or not a grave risk of harm existed in fact was not explored on the basis that post-return a Greek court may not necessarily find that the children had to reside permanently in Greece, or require the children to live in circumstances that put them at physical or emotional risk.^[101]

Conversely in Department of Communities (Child Safety Services) v Garning (‘Garning’),^[102] the grave risk of harm exception was not established. It is submitted that Garning turns on its facts however, as the Family Court reached this finding after a detailed examination of whether or not a grave risk of harm existed in fact. The Court did not eschew addressing the contention of risk. In this case four girls were abducted from Italy to Australia by their Australian born mother. This case was appealed to the High Court after a return order was made by Forrest J, however, on grounds other than the grave risk of harm exception.^[103] The exception was only considered and rejected by the Family Court at first instance. The grave risk of harm was said to include the father’s state of mental health. He had suffered from depressive episodes that appeared to have been brought on by the death of one of the couple’s children.^[104] The Family Court examined a history of physical and verbal violence perpetrated by the father against the mother and children, prior to the couple’s separation in 2007 (some four years before the return proceedings).^[105] In addition, the Court accepted the evidence of a child psychologist who had interviewed the children, who concluded that the children had experienced a degree of authoritative and inappropriate physical disciplining by the father. Yet the Court also noted that the children had expressed positive interactions and a level of attachment with their father post-separation. This was evidenced by the children ‘warmly describing positive memories and activities that they participated in together.’^[106]

Following the broad approach advocated by the High Court in DP, the Family Court in Garning^[107] placed weight upon whether a grave risk existed in fact.^[108] Forrest J assessed the consequences of return for the four children by engaging in a detailed inquiry of affidavit evidence provided by the parents and witnesses, a letter from the father’s psychiatrist, and reporting by the child psychologist who assessed the children. The Court accepted the mother’s evidence that she was subjected to emotional, verbal and physical violence prior to and leading up to separation.^[109] However, the abuse was mostly historical.^[110] The Court also accepted that the father had been hospitalised for his depression on three occasions before March 2007. Since that time the evidence showed that he had received outpatient treatment, and had progressed well despite the stress of a difficult separation with his wife.^[111] Forrest J found the mother’s evidence to be ‘internally contradictory’^[112] because she had expressed that she was prepared to allow the children to spend holiday time with the father if they were permitted to remain in Australia. In addition, the Court also considered it significant that the mother had not taken steps prior to the abduction to amend the existing Italian agreement by consent that the father have contact with the children.^[113]

In DP, the High Court said that Australian courts should not avoid interpreting the gravity of harm exception; that is, they should not make a determination about the individual child’s best interests by assuming that the child will be afforded protection by the authorities in their habitual residence post-return. In Garning’s case,^[114] the Family Court’s rejection of the grave risk of harm exception was not based upon an unwillingness to examine the factual circumstances awaiting the children upon return. The Court did not avert addressing the contention of risk by simply saying that the Italian authorities and courts could deal with the risk. Forrest J examined the circumstances awaiting the children upon return in detail. His Honour then decided that despite concerns about the father’s overly authoritative parenting style, he could not find evidence before him

that returning the girls to Italy, where their ongoing parenting arrangements [could] clearly be subject to further consideration in the courts of Italy, place[d] them at a risk of physical or psychological harm that can be described as reaching the level of ‘grave’.^[115]

IV CONCLUSION

The Convention’s drafters acknowledged a teleological connection between the action of prompt return, and the promotion of comity between Convention countries and the best interests of children generally.^[116] However, they also recognised that the Convention must still strike a delicate balance between protecting children generally from the harmful effects of international parental child abduction, and accommodating situations where a child’s unilateral removal is justified and in fact in their best interests.^[117] Perhaps the differing judicial opinions of the Family Court and High Court, about the extent to which the grave risk of harm exception should permit consideration of an individual child’s best interests, are a manifestation of the intricacies of the balancing act that must be performed by courts hearing return proceedings.

The Family Court has exhibited a propensity to resolve the incompatibility between facilitating comity between Convention countries and examining a child’s best interests in favour of the former. This choice has been rationalised with the assumption that the individual child’s best interests are most appropriately reserved for consideration post-return in the child’s habitual residence.^[118] From a practical perspective accepting this assumption as accurate has been the easy solution. This is because Convention return proceedings are suited to being summary in nature and ‘they are not ideally designed to determine contradicted issues of fact.’^[119] Contradicted issues of fact can include each party’s submissions concerning their child’s best interests when the grave risk of harm exception is raised by the abducting parent.

The feminisation of international parental child abduction, and the social implications that this has had, arguably justifies the High Court’s broader approach to the exception’s interpretation. A true state of balance is the attainment of the best interests of all children. Realistically this balance may be difficult to achieve. Perhaps the restrictive interpretation of the exception advocated by the Family Court of Australia and dissenting High Court judges is a display of the tightrope becoming unsteady. Comity between Convention countries is swayed towards, whilst the child’s welfare is sacrificed. The High Court has arguably managed to master the art of maintaining balance whilst traversing the tightrope, by expeditiously examining the child’s welfare and the potential consequences awaiting the child if they are returned to their habitual residence.

The Vexed Question of Agency and Torrens Fraud: The High Court in Cassegrain

DILAN THAMPAPILLAI^[∗]

Abstract

The question of whether a principal may be held liable for the fraud of his or her agent is not a simple matter. There are difficult questions of law and legal policy at play. These issues arose in Cassegrain v Gerard Cassegrain & Co Pty Ltd. However, as the High Court in both the majority and minority opinions found that no relevant agency relationship existed, these issues passed largely unexplored. In setting out their reasons, both the majority and minority opinions of the High Court endorsed the approach that Street J took in Schultz v Corwill Properties Pty Ltd in determining the scope of an agent’s authority. This article sets out two criticisms of the High Court’s decision in Cassegrain. First, the article contends that there were in fact sound reasons to find that the husband acted as agent for the wife in effecting both the initial sales transaction and first transfer to himself and his wife as joint tenants. Those who set up an agency relationship, however slightly or informally, should bear the consequences of that relationship where the agent causes harm to third parties in the furtherance of his and the principal’s interests. It is imperative that agency law should look closely at the question of whom should bear the risk of the agent’s fraud. Second, the article sets out three criticisms of the approach of Street J in Schultz v Corwill Properties Pty Ltd.

I INTRODUCTION

In its recent decision in Cassegrain v Gerard Cassegrain & Co Pty Ltd,^[120] the High Court considered whether an agency relationship had arisen between a husband and wife in relation to the registration of a title under the Torrens system. Had such an agency been found to exist and had the actions of the husband been within the scope of the agency, then the title of the wife might have been rendered defeasible due to the fraud perpetrated by her husband.^[121] However, the High Court’s findings on agency are rather equivocal with the majority judgment appearing to suggest that some limited agency might have existed.^[122]

There is precious little literature or jurisprudence available on the intersection of agency law and Torrens fraud. In particular, there is little clarity as to the precise circumstances under which a fraud committed by an agent in the course of obtaining a registered title for his or her principal should be attributed to the latter under Torrens law. It is generally accepted that the fraud may be attributed to the principal where the impugned act falls within the scope of the authority that the former has conferred upon the agent.^[123] However, there is some controversy concerning how the question of the scope of the agent’s authority should be approached. This particular issue arose in Schultz v Corwill Properties Pty Ltd.^[124] It was considered again by both the majority and minority opinions of the High Court in Cassegrain and the reasoning of Street J in Schultz was cited with approval in both judgments.^[125]

Regrettably, the endorsement of Schultz leaves unexplored some rather complex issues involving agency and Torrens fraud.^[126] In particular, it is unclear whether the sensible approaches of the New Zealand Supreme Court in Dollars & Sense Finance and the eminent commentator Atiyah,^[127] can be usefully applied in Australian law to a situation where an agent commits Torrens fraud in furtherance of the interests of his or her principal.^[128] In Cassegrain, the High Court has quite arguably missed an opportunity to revisit and clarify these particular matters. Given the criticisms that the New Zealand Supreme Court made of Schultz in its decision in Dollars & Sense Finance, it would have been helpful if the High Court could have indicated whether Dollars & Sense Finance would be applicable within Australian jurisdictions.

This article advances two criticisms of the High Court’s decision in Cassegrain. The first is that even though there may have only been a threadbare agency in place between Felicity and Claude Cassegrain, it should have been sufficient to warrant holding the principal liable for the fraud committed by the agent even though the principal might not have been aware of the fraud at the time. The second criticism is that the Court’s support for the approach in Schultz is problematic. This article first provides a brief overview of some of the fundamental principles of agency law. The article then examines the treatment of agency law in Cassegrain by the NSW Court of Appeal and the High Court. The article lastly advances three criticisms of the approach in Schultz.

II THE LAW OF AGENCY

For the most part agency law is a creation of commercial convenience. Where a person finds a particular task difficult or inconvenient to do for themselves, they may assign it to another to act on their behalf in the matter. Though it is primarily a creature of commercial law, agency is now so well established that it may arise in a wide variety of contexts. All that is required is the existence of a principal who consents to an agent acting on his or her behalf in their dealings with third parties.^[129] In International Harvester Co, the High Court stated that agency is, ‘a word used in law to connote an authority or capacity in one person to create legal relations between a person occupying the position of principal and third parties.’^[130]

It is important to note that agency law is both a relationship between the parties, where one delegates responsibilities to the other, and a consequence dictated by law.^[131] In Branwhite, Lord Wilberforce noted that the relevant parties may act in such a manner as to create, ‘a state of fact upon which the law imposes the consequences which result from agency.’^[132] That the parties themselves have not explicitly recognised the existence of an agency relationship is of no great importance.

In South Sydney District Rugby League Football Club, Finn J stated:

The consents so given need not necessarily be to a relationship that the parties understand, or even accept, to be that of principal and agent. It is sufficient if ‘they have agreed to what amounts in law to such a relationship’; notwithstanding that they may have ‘artfully disguised’ it by express disclaimers.^[133]

Fundamentally, substance triumphs over form in agency law.^[134] The scope of an agency relationship may be limited by the parties. That an agent may have authority to act in one capacity does not mean that they have the authority to act in other capacities.^[135] More pertinently, an agency relationship is not without its risks or complications. By creating authority in the agent the principal has set up a situation in which the agent might potentially harm the principal’s interests. This can happen through fraud,^[136] negligence^[137] or simply by poor judgment. Moreover, an agent might knowingly act beyond the limits of his authority in good faith so as to allow his principal to gain the benefit of an advantageous transaction.^[138]

It is well settled that a principal may bear liability for the actions of his or her agent where the latter has exceeded the scope of his or her authority.^[139] The doctrine of ostensible authority essentially provides that where a principal represents to a third party that an agent has authority to act on behalf of the principal and the third party, in reliance of that representation, alters his position through dealings with the agent, the principal will be bound to the third party.^[140] Claims of ostensible authority have been entertained by the courts even where the agent has forged documents.^[141] For example, in Klement v Pencoal Ltd^[142] a claim of ostensible authority was upheld even though the agent had forged documents.

It follows then that holding a principal liable for the consequences of her agent’s fraud is not an altogether controversial proposition provided that the agent is acting within the scope of his actual or apparent authority.^[143] That is not the same thing as saying that the principal has directed the agent to commit fraud. Instead, where the fraudulent actions of an agent have created a loss, which one of two innocent parties must bear, there are sound reasons for ascribing the loss to the principal on the basis that he or she demarcated the scope of the agent’s authority or failed to properly oversee the agent’s actions.

III FINDING AN AGENCY RELATIONSHIP IN CASSEGRAIN

The facts of Cassegrain are not uncomplicated. Gerard Cassegrain & Co Pty Ltd (GCC) was the registered proprietor of a dairy farm in New South Wales. Claude Cassegrain was a director of the company. After GCC had received a settlement amount of $9.5 million from the Commonwealth Scientific and Industrial Research Organization (CSIRO), Claude contrived to make a debt of $4.25 million appear in the books of the company. The debt appeared to arise due to a loan made by Claude to GCC, though it was recognised during the various court proceedings that he was never entitled to any such amount from GCC. In 1996, Claude and his sister Anne-Marie Cameron, who was a co-director of GCC, decided that the loan would be settled by the sale of the dairy farm (the sale transaction) and other assets to Claude and his wife Felicity as joint tenants.^[144]

In February 1997, Claude gave instructions to his solicitor Chris McCarron to register the transfer of the dairy farm. There was no allegation that McCarron had any part in the fraud that Claude was perpetrating and he acted in good faith. The transfer was registered in March 1997. No evidence was adduced to demonstrate that Felicity Cassegrain had given any instructions to McCarron. Neither Claude nor Felicity gave evidence at trial. Consequently, there was no direct evidence as to the communications between Claude and Felicity. GCC did not call McCarron during the trial before Barrett J in the NSW Supreme Court.

Later, the other siblings of the Cassegrain family sued Claude and Anne-Marie in 1996 claiming harsh and oppressive dealings in relation to the $4.25 million loan. In July 1998, Davies J in the Federal Court ruled in their favour.^[145] This fact colours the next major dealing in the whole saga. On the 24^th of March 2000, Claude Cassegrain transferred his interest in the dairy farm to Felicity for the princely sum of just $1.

There might have been a particular cunning in Claude’s decision to initially transfer the land to both himself and Felicity as joint tenants, rather than just to himself as sole proprietor. The effect of s 118(1)(d)(ii) of the Real Property Act 1900 (NSW) is to deprive a volunteer of their registered title where such title derives from a prior owner who has committed Torrens fraud. In the High Court, Keane J noted:

If Claude had procured the registration of the transfer from the respondent to himself as sole registered proprietor of the land, and then transferred the land to the appellant for one dollar, there could be no doubt that the respondent would be entitled to recover the land from the appellant under s 118(1)(d)(ii) of the Real Property Act.^[146]

There was no evidence that Felicity was aware of Claude’s fraud at the time of the first transfer. However, it was generally accepted by the various courts in Cassegrain that Claude had committed Torrens fraud within the meaning of s 42 of the Real Property Act 1900 (NSW). Section 42 provides:

Notwithstanding the existence in any other person of any estate or interest which but for this Act might be held to be paramount or to have priority, the registered proprietor for the time being of any estate or interest in land recorded in a folio of the Register shall, except in case of fraud, hold the same, subject to such other estates and interests and such entries, if any, as are recorded in that folio, but absolutely free from all other estates and interests that are not so recorded.

The combined operation of s 42 and s 118(1)(d)(i) effectively provides that the title of a registered proprietor is defeasible where fraud can be brought home to the registered proprietor’s title.^[147] The system of title by registration under the Torrens system confers immediate indefeasibility upon registration.^[148] Subject to some qualifications,^[149] the fraud that would render a title defeasible must occur prior to registration.^[150] In Stuart v Kingston,^[151] it was noted that Torrens fraud requires, ‘something in the nature of personal dishonesty or moral turpitude.’^[152] Similarly, in Farah Constructions Pty Ltd v Say-Dee Pty Ltd,^[153] the High Court noted the need for ‘actual fraud, moral turpitude.’^[154] This obviously suggests conscious wrongdoing and the knowing perpetration of some scheme of deception.^[155]

Given the nature of Torrens fraud, it is submitted that in most cases a person who lacks either knowledge or active participation in a fraud cannot have their title set aside under s 118(1)(d)(i). However, where the fraudulent party was the agent of the person who became the registered proprietor, there are two ways in which a Torrens fraud can be brought home to such a registered proprietor. The first arises where the fraud falls within the scope of the agency.^[156] The second occurs where the agent knew of the fraudulent conduct and that knowledge can be imputed to the principal. Accordingly, GCC sought to argue that Felicity was the principal and that Claude acted on her behalf as agent. The basis for the agency argument emerges from a famous quote by Lord Lindley in Assets Co Ltd on the subject of Torrens fraud. Lord Lindley stated:

[T]he fraud which must be proved in order to invalidate the title of a registered purchaser for value ... must be brought home to the person whose registered title is impeached or to his agents. Fraud by persons from whom he claims does not affect him unless knowledge of it is brought home to him or his agents.^[157]

Lord Lindley’s statement appears applicable to two scenarios. In the first situation, the person’s agent is directly involved in the fraud. In the second, Lord Lindley appears to suggest that if a person’s agent has knowledge of a fraud then that too necessarily binds the principal so as to place him or her within the parameters of Torrens fraud.^[158] In that position the principal’s registered title would then be defeasible.

Where this leaves a passive principal, such as that which Felicity Cassegrain was alleged to be, is altogether unclear.

On its bare terms the first limb of Lord Lindley’s statement suggests that where the agent is the architect and perpetrator of the fraud in whole or part for the benefit of the principal,^[159] the existence of an agency is no shield against fraud being brought home to the title of the registered proprietor. Indeed, where the first scenario is concerned, the active participation of the agent in the fraud should be enough to deprive the principal of an indefeasible title. Crucially, there is no suggestion in Lord Lindley’s statement that the limited nature of the agency is any defence against the defeasibility of the principal’s title. This is why the existence of an agency relationship was one of the most crucial matters in Cassegrain.

In the High Court in Cassegrain, the Justices in the majority, French CJ, Hayne, Bell and Gageler JJ questioned whether Lord Lindley had knowingly used the term ‘agents’. Their Honours stated:

What Lord Lindley meant by his reference to "agents" was not explored then or in later decisions of this Court. It may be thought that the reference to "agents" was intended to do no more than refer to those natural persons through whom the corporation, Assets Company Limited, had acted in acquiring the registered title that it did.^[160]

However, their Honours acknowledged that later courts had not interpreted Lord Lindley’s comments in this manner.^[161] Nonetheless, it is well known that an agency relationship can arise in a number of ways. An agent might be appointed by the principal, or the agency relationship might arise from the manner in which the parties have conducted themselves. An agency relationship can arise by implication from the actions of the relevant parties.^[162] In Equiticorp, Clarke and Cripps JJA noted that the parties, ‘may conduct themselves in such a way that it is proper to infer that the relevant authority has been conferred on the agent.’^[163] That said, the existence of a marital relationship alone does not give rise to an agency relationship between the husband and wife.^[164] What must occur is that the principal must consent to the agent representing them in some way.^[165]

At some point prior to the first registration, Felicity Cassegrain must have known that Claude Cassegrain was going to purchase the farm from the respondent and that he would register it in both their names as joint tenants. In the absence of evidence to the contrary, and with Felicity having chosen not to give evidence, it is hard to resist the conclusion that Claude acted on Felicity’s behalf. Further, with nothing to suggest that Felicity ever gave any instructions to the solicitor, it must be assumed that Claude gave the instructions for registration. On this point the appellant’s arguments were exceedingly coy. The appellant submitted in her written submissions that the respondent could easily have called the solicitor as a witness to definitively prove that she had given him no instructions. Leaving matters of confidentiality aside, where the presumption of agency had arisen the burden should surely have shifted to the appellant to disprove such a relationship.

On the facts before the NSW Court of Appeal and the High Court, it could reasonably be inferred that Claude was Felicity’s agent for the purposes of the initial sale transaction. Felicity was certainly a passive recipient of her property interest in the farm. However, as the registration could not be brought about without the existence of the sale transaction, it is not unreasonable to suggest that at some point she had consented to Claude procuring both the sale transaction and subsequent registration on her behalf. Having drawn that inference, it can then be said that as well as acting for himself, Claude acted as Felicity’s agent in achieving the sale transaction and registration.

Indeed, in the NSW Court of Appeal, Beazley P and Macfarlan JA found that Claude was Felicity’s agent. In doing so Beazley P and Macfarlan JA relied in part upon the principle in Blatch v Archer,^[166] that an inference could be drawn where certain facts are wholly within the knowledge of one party.^[167] However, notwithstanding the principle in Blatch, Beazley P found sufficient basis upon which to draw the inference of an agency relationship.^[168] Macfarlan JA concurred with her Honour’s reasoning on this point.^[169]

In the High Court, the majority took issue with the findings of both Beazley P and Macfarlan JA on the issue of agency. In their judgment, the majority of the High Court noted that the term ‘agent’ is much abused. Further, the majority stated:

At least for the most part, the word "agent" appears to have been used in the Court of Appeal as a term explaining how events happened rather than as a term attributing legal responsibility for those events. The relevant question was treated as one of fact.^[170]

In the view of the majority, a finding that an agency relationship existed should have been the beginning rather than the end of the inquiry. On this point a valid criticism of the judgments of the Court of Appeal may lie in the lack of attention given to the question of the scope of Claude’s authority. Nonetheless, a close reading of the judgments of Beazley P and Macfarlan JA makes it clear that their Honours understood the full import of a finding of agency between Claude and Felicity.

It is possible to read the majority opinion in the High Court of French CJ, Hayne, Bell and Gageler JJ as accepting that there was an agency relationship between Felicity and Claude with respect of registration. The language employed by the majority is decidedly equivocal and it is also possible to read the judgment as suggesting that there was no agency.

The majority opinion did acknowledge that Felicity acquiesced to Claude registering the title in both their names.^[171] The majority stated that ‘Felicity was no more than the passive recipient of an interest in land which her husband had agreed to buy, but which he wanted (with her acquiescence) put into their joint names.’^[172] The majority did not expressly state that Claude was Felicity’s agent for the act of registration and this omission must be reconciled with the majority’s finding that there was no agency in ‘any relevant sense’.^[173] With respect of Claude’s actions in procuring registration, the majority also stated that ‘without more’ these actions, ‘showed no more than that Claude had performed tasks that were of advantage to Felicity.’^[174] As such, apart from their rejection of the reasoning of the NSW Court of Appeal, the majority’s actual findings on agency are decidedly ambiguous.

If Felicity consented to Claude doing an act on her behalf, which the majority did appear to concede when they noted her acquiescence to the registration in both their names, then at least for the mere purposes of that particular act, he was her agent. It might be a very limited type of agency, but nonetheless it should be regarded as an agency relationship. The majority stated:

Without more, the conclusion that Claude had taken the steps necessary to procure registration of the transfer from the company to Felicity and him as joint tenants did not show that his fraud was within the scope of any authority she had, or appeared to have, given to him. Without more, it did not show that knowledge of his fraud was to be imputed (in the sense of ‘brought home’) to her.^[175]

However, this statement appears almost to conflate the first and second limbs of Lord Lindley’s famous statement on Torrens fraud and agency. Moreover, if the majority did indeed hold the view that Claude was Felicity’s agent for the act of registration, but not the sale, then two problems arise. The first is that this suggests some sort of sharp distinction between the sale transaction and the registration. Yet, in a Torrens context registration is the end-point of any sale transaction. As such, it may not be tenable to attempt to draw a distinction between the sales transaction and the registration as one cannot exist without the other.

Second, to the extent that it does appear to acknowledge a slight agency, the majority judgment seems to suggest that the limited nature of this agency tells against imputing any knowledge of the fraud to Felicity. However, this runs counter to Lord Lindley’s statement which appears on its first limb to require no more than the conscious participation of the agent in the fraud. Moreover, it is difficult to suggest that Claude’s fraud ceased prior to registration as the fraudulent scheme that he was perpetrating in its entirety consisted of a fake loan, a sale and subsequent registration. The achievement of a registered title was the end point of the fraud.

The majority opinion in Cassegrain should be criticised for failing to make plain what capacity it believed that Claude was acting in when he procured registration over the property for himself and Felicity. It could be possible to construe Claude’s actions in registering the title to the property in both his and Felicity’s names as a gift to Felicity of a joint interest.^[176] Neither the majority nor minority opinions of the High Court in Cassegrain expressly contemplate this possibility, though this does appear to be the assumption upon which the members of the Court based their judgments. Nonetheless, even if they had done so, there would be no reason at all to suppose that the making of a gift precludes a finding of agency in circumstances where the purported principal is aware of the nature of the gift that is being procured.

It is important to note that in procuring the sale transaction, Claude was acting both for himself and Felicity. Though it is submitted that a neat distinction between the sales transaction and the registration is not entirely viable in a Torrens context, it does follow that an argument could have been advanced to the effect that Felicity’s actions in retaining her title after the first registration amounted to ratification by conduct of the actions done on her behalf in the sales transaction itself.^[177] In effect, this would have been a ratification of Claude’s ability to act as agent on Felicity’s behalf and all of his actions in obtaining registration.^[178] Moreover, the absence of any explicit authorisation by Felicity is not problematic as an agency relationship can be retrospectively ratified by conduct.^[179] It is quite likely that retaining the benefit of a registered title, in circumstances where the principal has taken no positive steps to procure the registration, is in effect ratification by conduct.^[180] It would not have been necessary to suggest that Felicity was ratifying Claude’s fraud. However, this would have raised an issue concerning the requirement that the principal must know all the material circumstances in order to ratify an act.^[181] In turn, this would have brought the doctrine of ratification into conflict with the Mair principle,^[182] in that Felicity would be declaiming the existence of the agency and the authorisation of the transaction whilst retaining the benefit of the fraud. It is notable that those cases where an agent’s fraud or misconduct have precluded ratification have been those concerning the acts of an agent which have seriously disadvantaged or defrauded the principal.^[183]

It was actually Felicity who raised the issue of ratification before the NSW Court of Appeal when she submitted that asserting the indefeasibility of her title did not amount to a ratification of Claude’s conduct. Her Honour, Beazley P, briefly noted this point,^[184] but as she found that an agency existed by other means it was not developed further. Curiously, GCC did not advance a ratification argument before the NSW Court of Appeal or the High Court. Whilst Felicity is undoubtedly correct in stating that claiming indefeasibility does not amount to ratification, this does not preclude the suggestion that the act of retaining a registered title after that registration has been procured for the ‘principal’ by the acts of another amounts to a ratification of both the agency relationship and the transaction in question.

IV THE VEXED QUESTION OF THE SCOPE OF AN AGENT’S AUTHORITY

In the course of discussing the agency relationship between Claude and Felicity, the High Court relied upon the reasoning of Street J in Schultz.^[185] The remarks of Street J in Schultz are relevant to the question of the scope of an agent’s authority. The Court’s approval of the position of Street J in Schultz is not unproblematic. In particular, the continued reliance upon Schultz might put Australian courts at odds with the approach of the New Zealand Supreme Court if matters arise in the future in which the scope of an agent’s authority is directly relevant. This is an area of Torrens law that has not been fully explored.

In Schultz, Street J based his reasoning on Lord Lindley’s consideration of fraud and agency in Assets Co Ltd. Justice Street noted that the discussion of agency and Torrens fraud in Assets Co Ltd contemplated two different situations:

The first is one in which the fraud is actually committed by (‘brought home to’) the person whose title is impeached or his agent. And the second is one in which he or his agents have knowledge that a fraud has been committed whereby the previous registered proprietor is being deprived of some or all of his interests.

Each of these two concepts is capable of being applied in accordance with settled principles of law. The first, namely, fraud on the part of the person whose title is impeached or his agents, involves the application of the ordinary principles governing the responsibility of a principal for the fraud of his agent. If the fraud in question is the immediate act of the person whose title is impeached, then the position is not open to doubt. If, however, the fraud is that of an agent for the person whose title is impeached, the principle of respondeat superior, with all its limitations and qualifications, is applicable. The matter is to be tested by investigating whether or not the principal is, in the particular circumstances under consideration, liable to the person who has been defrauded for the acts of the agent.^[186]

In Schultz, the defendant company was the registered proprietor of a parcel of land in NSW. Mrs Schultz sought to invest £3000 by securing a mortgage over the land. She gave the money to her lawyer, Clive Galea, to further the grant of the mortgage and to register it on her behalf. The defendant company had two shareholders and directors, one of whom was Clive Galea’s mother. Galea himself was the secretary of the company.^[187] Though he did register a mortgage, Galea misappropriated Mrs Schultz’s money for his own purposes. Given the fraud of Galea and his role as her agent, the issue that came to be in dispute was whether Mrs Schultz’s mortgage was indefeasible.

On the issue of knowledge of a fraud, Street J further stated:

It is not enough simply to have a principal, a man who is acting as his agent, and knowledge in that man of the presence of a fraud. There must be the additional circumstance that the agent's knowledge of the fraud is to be imputed to his principal. This approach is necessary in order to give full recognition to (a) the requirement that there must be a real, as distinct from a hypothetical or constructive, involvement by the person whose title is impeached, in the fraud, and (b) the extension allowed by the Privy Council that the exception of fraud under s 42 can be made out if ‘knowledge of it is brought home to him or his agents’.^[188]

Justice Street did not elaborate upon the circumstances under which the knowledge of the agent might be imputed to the principal. There are two important principles of agency law that might govern this matter. The first is the long-standing exception, which is now rather controversial, that the knowledge of an agent is not to be attributed to the principal where the former is defrauding the latter.^[189] The second is the equitable rule that a principal cannot benefit from his or her agent’s fraud.^[190] It should ordinarily be the position that knowledge of an agent’s fraud is to be imputed to a principal where he or she stands to benefit from the fraud. This would be consistent with other rules of agency law.^[191] Though he might easily have done so, Street J did not expressly refer to the fraud exception to the principal’s liability for the conduct of their agents or any of the cases that have dealt with that exception.^[192] Nor did Street J discuss any of the general agency law cases that have dealt with the issue of whether an agent is acting within the scope of his authority.^[193] Street J did cite Bowstead on Agency, which states the general principle that a principal may be liable for the acts of an agent acting within the scope of his authority, but no other cases were discussed.^[194]

A Demarcating the agent’s authority too narrowly

There are three criticisms that can be made of Justice Street’s opinion in Schultz. The first is that in Schultz, Street J defined the scope of the agent’s authority in a manner that was too specific.^[195] This places Street J’s approach at odds with later decisions. In Schultz, Street J stated that with regard to the principal’s liability to the third party:

On this topic one need not delve more deeply than the general statement in Bowstead on Agency, 13th ed., p.242:-

‘An act of an agent within the scope of his actual or apparent authority does not cease to bind his principal merely because the agent was acting fraudulently and in furtherance of his own interest.

This principle is general, applicable to cases of actual and apparent authority; in tort; in the disposition of property; a similar result even appears in criminal cases. But the mere fact that the principal, by appointing an agent, gives that agent the opportunity to steal or otherwise to behave fraudulently does not without more make him liable: the agent must normally be acting within the scope of his actual or apparent authority for the principal to be responsible’.^[196]

Much confusion in this area of the law might well have been averted had Street J delved deeper on this topic than the general statement in Bowstead. In Schultz, Street J dealt with the issue of the scope of the agent’s authority by simply stating that Schultz had given authority to Galea to register a valid mortgage.^[197] The approach that Street J took in Schultz almost amounts to treating the fraud as if it automatically severs any connection between the action of the agent and their actual or apparent authority.

Other courts have not treated the existence of fraud as being decisive as to whether the agent is acting outside or within the scope of his or her authority. In Perpetual Limited v Barghachoun,^[198] the defaulting mortgagor, Mr Abdul Barghachoun, challenged the indefeasibility of Perpetual's mortgage on the basis that the latter’s agent had acted ‘fraudulently’ by procuring the former’s signature when he was unwell and not in a fit state to sign a mortgage. Notwithstanding the question of whether these acts fell within the scope of Torrens fraud, Rothman J was prepared to accept:

for present purposes that relate to whether Mr Barghachoun’s proposition is arguable, that the act of the agent, within the scope of its actual or apparent authority, does not cease to bind Perpetual, merely because the agent may have been acting fraudulently and in furtherance of its own interests.^[199]

At the very least, the existence of a fraud should not immediately place the agent’s act outside the scope of his authority.

In Dollars & Sense Finance, the New Zealand Supreme Court took issue with the narrow treatment of the scope of authority in Schultz. In Dollars & Sense Finance, a finance company, Dollars & Sense Finance Ltd (DSF), agreed to lend $245,000 to Rodney Nathan to assist him in purchasing shares in a business. However, DSF asked Rodney to get a security for the loan from his parents in the form of a mortgage over their home. DSF’s lawyer gave Rodney the relevant mortgage papers including the papers his parents were to sign. Rodney did not tell DSF that his parents were in fact separated. Rodney procured his father’s signature, but he forged his mother’s signature on the papers. Rodney returned the documents to DSF and the mortgage was duly registered.

At the time that the case was litigated Rodney’s father had passed away, so his mother challenged the validity of DSF’s registered mortgage on the grounds that it was defeasible for Torrens fraud. During the proceedings it was argued that Rodney had acted as DSF’s agent in procuring the mortgage over his parents’ home. On the facts in Dollars & Sense Finance there was stronger evidence of an agency than there was in Cassegrain. However, this is attributable to the fact that the key parties actually gave direct evidence as to what had transpired. Crucially, because DSF, through its lawyer, gave Rodney the relevant documents and asked him to procure his parents’ signatures, an agency relationship was created between the parties for this purpose. It was acknowledged that DSF was not instigating a fraud and was unaware of what actually transpired at the relevant time.^[200]

Having established that an agency relationship existed, the New Zealand Supreme Court turned its attention to the scope of the agency. The Court found that it would be misguided to rely on Schultz, in part due to the overly specific construction of the scope of the authority.^[201] The Court stated:

No one suggests of course that D & S actually authorised the particular forgery or any forgery or fraudulent act at all. But it does not follow from that fact alone that the forgery was beyond the scope of the agency.^[202]

In essence, the suggestion that a fraud immediately places the actions of the agent outside the scope of their authority, except where they have been authorised to commit the fraud, takes ‘too narrow a view of an agent’s task.’^[203]

The Court relied on Atiyah in constructing the scope of the authority.^[204] Atiyah has suggested that determining the scope of an agent’s authority depends on: (i) what acts the principal has authorised; and (ii) whether what the agent has actually done is sufficiently connected to those acts as to constitute a mode of performing them.^[205] In Dollars & Sense Finance the Court found that what was authorised was the obtaining of the signatures and related acts.^[206] The Court formed the view that an agent’s acts could have a close connection to the authorised acts even though it is criminal and fraudulent.^[207] The Court stated:

A fraudulent act impacting on a third party may vis-à-vis the third party be seen as done within the scope of an agency even if done exclusively for the benefit of the agent, and a fortiori may be seen as an act within the agency if it is done for the benefit of the principal as well as for the benefit of the agent.^[208]

In developing its position the Court drew on vicarious liability cases in tort, notably Lister v Hesley Hall Ltd,^[209] where a boarding school was found vicariously liable for the sexual abuse of children by a school warden. The Court also placed weight on the obiter remarks of Lord Millett in Dubai Aluminium Co Ltd v Salaam,^[210] where his Lordship had suggested that the question of connection had to be dealt with by looking at the authorised acts and the actual acts of the agent while keeping the rationale underpinning vicarious liability in mind.^[211] In Dollars & Sense Finance, the Court suggested that this rationale was

a loss distribution device based on grounds of social and economic policy by which liability is imposed for all those torts which can fairly be regarded as reasonably incidental risks to the type of business being carried on.^[212]

The Court in Dollars & Sense Finance appears to have been influenced greatly by the concept of risk, in the sense that the party that created the risk should bear the consequences if it materialised.^[213] Notably, in Dollars & Sense Finance, Blanchard J delivering the opinion of the Court stated:

The tenor [of the relevant authorities] is that someone who creates an agency in which there is a risk of improper behaviour by an agent (or, as in this case, by someone entrusted with a sub-agency) should expect to bear responsibility where that risk eventuates and loss is thereby caused by the agent to a third party. The nature of that risk and the extent of the liability will depend upon the nature and scope of the agency.^[214]

Subsequently, in Hickman v Turn and Wave Ltd,^[215] Young J expressed support for this view. If a criticism could be made of the way in which the New Zealand Supreme Court addressed the question of the scope of authority it would be that it draws a little too heavily on the nexus between agency law and vicarious liability in tort, with too little consideration for the internal doctrinal concerns of Torrens fraud.

In Dollars & Sense Finance, the New Zealand Supreme Court saw no difficulty in relying upon principles of vicarious liability to develop the law of agency. This broadly reflects the position under Australian law.^[216] However, in the UK Supreme Court in Jetivia SA and another (Appellants) v Bilta (UK) Limited (in liquidation),^[217] Lord Sumption stated in obiter:

Vicarious liability does not involve any attribution of wrongdoing to the principal. It is merely a rule of law under which a principal may be held strictly liable for the wrongdoing of someone else. This is one reason why the law has been able to impose it as broadly as it has. It extends far more widely than responsibility under the law of agency: to all acts done within the course of the agent’s employment, however humble and remote he may be from the decision-making process, and even if his acts are unknown to the principal, unauthorised by him and adverse to his interest or contrary to his express instructions ... indeed even if they are criminal ... personal or direct liability, on the other hand, has always been fundamental to the application of rules of law which are founded on culpability as opposed to mere liability.^[218]

The statement by Lord Sumption does not necessarily mean that vicarious liability cannot influence the development of agency law. Rather, it means that some care must be taken to ensure that the two do not completely overlap. This safeguarding might well be achieved by the two-step test that was employed in Dollars & Sense Finance to construct the scope of the agent’s authority. In particular, the sufficient connection requirement in the second step means that not every act of the agent can be attributed to the principal for the purposes of liability. Moreover, in Dollars & Sense Finance the New Zealand Supreme Court made it abundantly clear that a finding that the fraud was within the scope of the agency did not amount to a suggestion that the principal had authorised the fraudulent act.^[219]

B The Schultz approach ignores valid reasons for finding the principal liable

The decision of the New Zealand Supreme Court in Dollars & Sense sets the basis for the second criticism of the approach in Schultz. That is, the narrow approach taken to constructing the scope of authority in Schultz more or less obscures the valid reasons for holding a principal liable for the wrongful acts of an agent. Notwithstanding the objection raised by Lord Sumpton in Jetivia, the parallels with vicarious liability in tort are hard to miss in this area of law.^[220] The rationale for holding a principal liable for their agent’s wrongful act was set out in Bayley v Manchester, Sheffield and Lincolnshire Railway Co,^[221] where Willes J stated:

A person who puts another in his place to do a class of acts in his absence, necessarily leaves him to determine, according to the circumstances that arise, when an act of that class is to be done, and trusts him for the manner in which it is done; and consequently he is held answerable ... provided that what was done was done, not from any caprice of the servant, but in the course of employment.^[222]

Dal Pont sensibly notes that it is immensely difficult to demonstrate that there was an express authority to commit fraud.^[223] Nevertheless, a principal might still be liable for a fraud committed in the scope of an agent’s authority.^[224] For example, in Lloyd a fraud committed by a managing clerk in the course of dealings with a widow, was attributed to his firm because it fell within the broad parameters of the conveyancing business which the latter had entrusted to him. The fact that the firm was innocent did not prevent the House of Lords from attributing responsibility to it as principal to a rogue agent.^[225] Quite pertinently, in a passage cited with approval by McHugh J in Scott^[226] Bramwell LJ in Weir v Bell,^[227] stated:

[E]very person who authorizes another to act for him in the making of any contract, undertakes for the absence of fraud in that person in the execution of the authority given, as much as he undertakes for its absence in himself when he makes the contract.^[228]

It is quite notable that where situations similar to those faced in agency law arise in the field of tort law, the fact that an ‘agent’ has completed an authorised task in an unauthorised manner is not problematic. Moreover, where agency and vicarious liability in tort have overlapped in the past this issue has not created great difficulty. In Hollis McHugh J, commenting on an agency argument that was raised in the dispute, stated, ‘it is not necessary for the principal “specifically” to “instigate, authorise or ratify” the agent’s wrongful act.’^[229] In other words the principal does not need to tell the agent specifically to commit fraud on his or her behalf. He or she may simply give the agent permission to procure the title to a property.

There is an issue that arises here that requires some thought. Given the remarks of various judges of the High Court in Hollis and Scott, and the endorsement of Schultz in Cassegrain, there is now a difference between the treatment of a person’s responsibility for the wrongs of another on the one hand in the fields of agency law and vicarious liability in tort, which on the one hand seem to be quite similar, and Torrens fraud on the other hand, which appears almost radically different. In fact, it appears as if tort law takes agency in one direction and Torrens takes it in another, with little thought given by Australian courts to the resulting confusion. The question is whether this differential treatment is necessary or desirable. I submit that where Torrens principles and agency law combine, it is important for the principles of agency law to be more carefully applied. Given that the fundamental question of the liability of a principal for the wrongful actions of an agent, is very much the same across Torrens, tort law and the general law of agency (that the principal is responsible for the wrongful actions of an agent), the law should try to move in the same direction. If there are genuine reasons for applying agency law differently within the Torrens system then this needs to be clearly articulated.

Indeed, it is worth asking the question as to whether there is something so special about Torrens fraud that it warrants the differential treatment of agency law in a Torrens context. The issue could be that safeguarding the robust nature of immediate indefeasibility under Torrens requires a restrictive approach to the fraud exception and that in turn has conditioned the operation of agency law.^[230] If this is indeed the case then the High Court did not explore it in Cassegrain.

Where an agent has done some wrong which will result in a loss to one of two innocent parties, either the third party or the principal, agency law invariably asks which of the two is more deserving of bearing the loss.^[231] The reasoning in both Bayley v Manchester, Sheffield and Lincolnshire Railway Co and Lloyd bear out this point. In a sense, agency asks who has contributed more to the loss. In Skandinaviska, the Singapore Court of Appeal found that the third party should bear the loss for a loan it made to Asia Pacific Breweries (APBS). The third party, Skandinaviska Enskilda Banken (SEB), a commercial bank, made the loan to APBS on the basis of representations made by an agent of APBS. SEB knew that APBS had not conferred on the agent any authority to accept the loan. However, the agent was a rogue and he forged documents that made it appear as if he had the authority to accept the loan. He then misappropriated the monies to support his gambling habit. The Singapore Court of Appeal held that SEB could easily have checked with APBS to determine the true extent of the agent’s authority. In this instance SEB’s failure to take precautions contributed the most to the loss.

Such an argument could quite easily have been raised in Cassegrain. It could have been argued that GCC could have been more diligent in checking the bona fides of the claim that monies were owed to Claude.^[232] There should have been better oversight of Claude and his sister in their role as directors. Yet, it could easily be said that Felicity’s suspicions ought to have been aroused at least by the time of the second transfer.^[233] In this sense, Felicity could hardly have been unaware of the litigation in Cassegrain v Cassegrain.^[234] Had such matters been raised the arguments would have been more finely balanced.

In the context of fraud, the Torrens system approaches the question of allocating loss between two innocent parties in a manner that is somewhat different to the law of agency. For example, in that crucial passage in Assets Co Ltd, Lord Lindley stated:

Fraud by persons from whom he claims does not affect him unless knowledge of it is brought home to him or his agents. The mere fact that he might have found out fraud if he had been more vigilant, and had made further inquiries which he omitted to make, does not of itself prove fraud on his part. But if it be shewn that his suspicions were aroused, and that he abstained from making inquiries for fear of learning the truth, the case is very different, and fraud may be properly ascribed to him.^[235]

In other words, a person who is innocent of actively seeking to commit a fraud does not become liable for Torrens fraud through mere want of care,^[236] but rather through wilful blindness.^[237] In this sense, Torrens law does look at contribution to the loss, but it has to balance it against the concept of immediate indefeasibility. Courts which have dealt with Torrens matters have in the past had to decide between an innocent buyer for value who becomes registered and an innocent former registered proprietor. This was the case in Frazer where an innocent owner had been defrauded by his wife and daughter and had lost his registered title. In that case the Privy Council chose to support the innocent buyer in order to implement the principle of immediate indefeasibility. The policy choice in favour of immediate indefeasibility was made in order to make the Torrens system a workable system of title by registration. This also involved an explicit rejection of the deferred indefeasibility concept that was raised in Gibbs v Messer.^[238]

It might well be that where agency law mixes with Torrens fraud, its application must be carefully managed to avoid the spectre of deferred indefeasibility. That is, if it were to be otherwise, the principal’s title would be defeasible until it were abundantly clear that his or her agent had neither knowledge nor involvement in any fraud. In Schultz, Street J did suggest that Torrens fraud could be attributed to a principal where the agent acts within the scope of his actual or apparent authority. However, he resolved the matter by more or less suggesting that the scope of the authority would have had to have contemplated fraud. This resolution does not sit well with agency law, though it might help Torrens law evade a type of deferred indefeasibility.^[239] It does appear that in situations such as those in Schultz, the courts are looking not just at the nature of the parties’ contribution to the loss, whether it might be a lack of prudence or wilful blindness, but also at the need to maintain coherency within the Torrens system itself. If there are concerns about the impact that agency law might have upon immediate indefeasibility then they were not raised and addressed in Schultz or Cassegrain.^[240] The best that can be said here is that the paucity of the reasoning in Schultz on the nexus between agency law and Torrens fraud has left too much unanswered.

It might also be that Torrens fraud, as a statutory concept, necessarily conditions the application of the common law of agency. Yet, here the looseness of the language used by Lord Lindley in Assets Co Ltd, which was adopted again in Schultz and Cassegrain, causes difficulties. There is no doubt that Torrens fraud requires conscious wrongdoing. However, if it is admitted that the conscious wrongdoing of an agent can be imputed to a principal,^[241] then what explains the reluctance of the courts to do so when the opportunity has arisen?

C A principal should not benefit from the fraud of their agent

One of the more troubling aspects of Cassegrain is that Claude’s fraud was done for the benefit of himself and Felicity. If an agency relationship had been found to exist, then the equitable principle, referred to in Williams as the Mair principle,^[242] which provides that a principal cannot retain the benefit of a fraud committed by their agent might have been applicable.^[243] As noted, the equitable principle prevents a principal from relying upon their own innocence in order to take advantage of the fraud that has been done by their agent.^[244] How this principle should interact with a Torrens statute is somewhat less clear in light of immediate indefeasibility. In Williams, Hodgson JA stated:

For the Mair principle to apply, in my opinion the principal must have a real choice whether or not to take or retain the benefit, and the benefit must be more than trivial. I refer to an analogy from the law of contract. If a person does work altering a house without being requested to do so by the house owner and without the house owner's knowledge ... the house owner will not be taken to have undertaken to pay for that work merely because the house owner has the benefit of the work through ownership and occupation of the altered house. This is because the house owner has no real choice in the matter. In my opinion, much the same approach should be taken in connection with the Mair principle.^[245]

It is arguable that but for Claude’s fraud, Felicity would never have acquired the dairy farm. Moreover, when she eventually discovered fraud she could have ceded the title to GCC. As the facts of Cassegrain did not suggest that the dairy farm was Felicity’s only possible residence, a real choice did exist. Her position was quite different to the facts of Williams where the benefit in question was insubstantial.^[246] While this matter was not pursued before the High Court or the NSW Court of Appeal, Felicity did suggest that her assertion of indefeasibility did not amount to an act of ratification.^[247] Whether the same argument would work to preclude the application of the Mair principle seems less likely, at least to the extent that the NSW Court of Appeal did not dismiss the issue out of hand when it arose in the context of indefeasibility in Williams. The concept of ‘real choice’ that Hodgson JA outlines above appears a workable model for balancing the Mair principle with the imperatives around immediate indefeasibility.

It is submitted that in any Torrens transaction there is an implied representation made by each of the parties to the other that the transaction itself is free from fraud. Nobody would rationally choose to deal with a party who refused to give a guarantee that they were not acting fraudulently. It is then almost self-evident that parties who choose to deal with each other are assuming the absence of any fraud and would also know that the other is acting on the same basis. In the context of Cassegrain, if Felicity was the principal and Claude the agent at the commencement of the sales transaction then she implicitly made a representation to GCC that there was no fraud at play. When this representation later turned out to be untrue and Felicity chose to retain the benefit of the fraud the Mair principle should have applied. The Torrens system has contemplated in Bahr that post-registration conduct that repudiates a pre-registration representation should give rise to an in personam exception. In Bahr, the Court was equally split as to whether this conduct should be considered Torrens fraud.

In his considered analysis of agency law, Dal Pont has suggested that the Third Restatement might be a sensible compromise between the differing approaches in Schultz and Dollars & Sense Finance.^[248] In the United States, the Third Restatement of the Law of Agency provides:

For purposes of determining a principal's legal relations with a third party, notice of a fact that an agent knows or has reason to know is not imputed to the principal if the agent acts adversely to the principal in a transaction or matter, intending to act solely for the agent's own purposes or those of another person. Nevertheless, notice is imputed: (a) when necessary to protect the rights of a third party who dealt with the principal in good faith; or (b) when the principal has ratified or knowingly retained a benefit from the agent's action.^[249]

The Restatement attempts to balance the competing goals of protecting the principal against the fraud of the agent and protecting third parties who unwittingly deal with a rogue agent. There is something about Felicity knowingly retaining the benefit of Claude’s fraud, in circumstances where she must have had some inkling by the time of the second transfer that something was amiss, that is contrary to both the principles of the Torrens system and the law of agency. There have been a number of cases in which parties have relied upon indefeasibility despite the fact that some fraud, of which they themselves had no part, was present in the transaction that provided them with their indefeasible title. However, in these cases, such as Mercantile Mutual v Gosper^[250] and Frazer, the indefeasible title was not procured by their agent.

V CONCLUSION

The High Court’s decision in Cassegrain leaves a number of key issues concerning the relationship between Torrens and agency law unanswered. While this is understandable given the majority’s view of the agency question, it will be incumbent upon future courts to untangle these issues. At the very least, Schultz should no longer be regarded as the best approach to determining the scope of an agent’s authority in light of Torrens fraud. The two step process outlined by Atiyah and adopted by the New Zealand Supreme Court in Dollars & Sense Finance is an altogether more workable approach.

The McMullan Principle: Ministerial Advisors & Parliamentary Committees

LORRAINE FINLAY^[∗]

The information can be provided without calling a member of ministerial staff before the committee. In my view, ministerial staff are accountable to the minister and the minister is accountable to the parliament and, ultimately, the electors.^[251] – Senator McMullan

Abstract

Parliamentary committees play an increasingly important role in ensuring government accountability. A question that has proven controversial in recent times is whether ministerial advisors can be compelled to appear before parliamentary committees. There have been a number of examples in which Government Ministers have invoked the so-called McMullan Principle, claiming that a constitutional convention exists that precludes ministerial advisors from being compelled to appear and give evidence. This paper will consider whether such a convention in fact exists, and will consider recent examples in which the principle has been invoked, such as the Children Overboard incident, Hotel Windsor inquiry and Orange Grove inquiry. It concludes that even if such a convention did exist at one time, the role of a ministerial advisor has now evolved to such an extent that it calls into question the underlying rationale for the McMullan Principle.

I INTRODUCTION

Parliamentary committees are increasingly an ‘important vehicle of ministerial accountability’.^[252] One of the key powers that parliamentary committees have at both the Commonwealth and State levels is the power to summon witnesses, with the failure to obey a summons potentially leading to a person being held in contempt of Parliament. A question that has proven controversial in recent times is whether ministerial advisors can be compelled to appear before parliamentary committees. There have been a number of examples in which Ministers have invoked the so-called McMullan Principle, claiming that a constitutional convention exists that precludes ministerial advisors from being compelled to appear.

This paper will consider whether such a constitutional convention in fact exists. It will begin by addressing the preliminary question of what a constitutional convention actually is, and how to determine whether a particular practice has moved into the category of being a constitutional convention. The McMullan Principle will then be specifically considered, with a particular focus on examples in which the principle has been invoked in order to reach a conclusion about whether or not it has moved into the convention category. This paper concludes that the McMullan Principle has not attained the status of an entrenched constitutional convention, but will then go on to consider whether it should be entrenched in the future as an accepted practice.

This final section of the paper involves a consideration of the policy reasons underpinning the McMullan Principle and, in particular, discussion of the evolving role of ministerial advisors and any lessons that can be drawn from the practices that have developed in relation to the appearance of public servants before parliamentary committees. It will be submitted that the role of the ministerial advisor has evolved in a manner that is inconsistent with the policy rationale underpinning the McMullan Principle and that it is not a principle that should therefore be routinely applied. It is important, however, to balance the importance of parliamentary accountability with the importance of confidentiality and trust in the relationship between individual Ministers and their advisors. Given this, it will be suggested that where a ministerial advisor does appear before a parliamentary committee, there are sound policy reasons for applying similar protocols to those which currently apply to public servants in similar circumstances.

II ESTABLISHING A CONSTITUTIONAL CONVENTION

The first preliminary point to consider is what we mean by a constitutional convention and how do we determine whether a particular principle should be accorded this status? A constitutional convention is an unwritten rule or constitutional practice that is consistently acknowledged and routinely followed, despite not being expressly provided for in the text of the Constitution. These unwritten customs ‘provide the flesh which clothes the dry bones of the law’.^[253] Constitutional conventions are not legally enforceable, but there is a broad expectation that they will be followed and that political consequences will attach to their breach.

Indeed, Ian Killey observes that:

in Westminster-based constitutions, conventions are so important to the system of government that actions inconsistent with these conventions are considered to be just as unconstitutional as they would be if inconsistent with the words of the constitution.^[254]

These unwritten principles and practices often govern important aspects of the Australian political system, and have enormous practical significance in the operation of Australia’s political institutions. The need for constitutional conventions becomes obvious when we realise that many key offices and institutions are not actually mentioned in the Australian Constitution. For example, the offices of the Prime Minister and Leader of the Opposition, the Cabinet and the Council of Australian Governments all play an important role in Australian politics, and yet none are actually mentioned in the Australian Constitution. Examples that are frequently cited as constitutional conventions in Australia include the reserve powers of the Crown, ministerial responsibility (both collective and individual) and the caretaker conventions that apply during an election campaign.^[255]

How then do you identify a constitutional convention? While there is not a single test that has been uniformly recognised as determinative, there are two key elements that are routinely emphasised as being central to identifying constitutional conventions. These are, firstly, an acknowledgement or acceptance by participants of the existence of a convention and, secondly, acceptance of the practice as binding.^[256] The first element of acknowledgment refers to a general understanding by the group of political participants as a whole that the convention exists. That is, ‘the terms of conventions are what the actors believe them to be’.^[257] In effect, this means that a principle will only be accorded the elevated status of a convention if the relevant actors consent to this understanding. The second element distinguishes conventions from political practices which are followed for some other reasons (such as, for example, political expediency or popularity). Constitutional conventions are consistently followed and applied because participants feel bound to do so, despite conventions not being legally enforceable.

Therefore, when considering whether the McMullan Principle should be accorded the status of an established constitutional convention, the key factors to consider are whether the principle is consistently acknowledged and accepted by political participants, and whether they consider the practice to be binding. An examination of recent examples in which the McMullan Principle has been invoked will be insightful in relation to both of these factors. However, before examining these recent examples, it will be useful to consider what the McMullan Principle actually is, its history and its underlying rationale in order to provide the necessary surrounding context within which to consider the debates that have recently occurred about its application in practice.

III THE MCMULLAN PRINCIPLE

The origins of the name, the McMullan Principle, stem from the debate surrounding a motion moved in the Australian Senate in 1995. Amongst other things, the motion highlighted the decision of the Minister for Finance to refuse to allow the Director of the National Media Liaison Service – a Ministerial staff member – to appear before the Finance and Public Administration Legislation Committee during its estimates hearings. During the course of debate, then-Senator McMullan (who, it should be pointed out, was not the individual Minister responsible for originally refusing permission to allow the Ministerial staff member to appear before the Committee) sought to justify the refusal by claiming that requiring ministerial staff to appear would set a dangerous precedent and was not the correct method by which the Senate should obtain the information it was seeking. He outlined what has subsequently been referred to as the McMullan Principle when he stated:

[t]he information can be provided without calling a member of ministerial staff before the committee. In my view, ministerial staff are accountable to the minister and the minister is accountable to the parliament and, ultimately, the electors.^[258]

Ironically, given that this is the case from which the McMullan Principle takes its name, on this particular occasion the Senate decided to take matters further and compel the appearance of the ministerial staffer before a parliamentary committee. The Senate passed a resolution instructing the Finance and Public Administration Legislation Committee to hold a supplementary estimates hearing and to direct the Director of the National Media Liaison Service (then Mr David Epstein) to appear before the committee. This resolution was opposed by the Government, however it passed with the support of the Australian Democrats who noted that whilst they had ‘grave reservations about what we are doing’,^[259] they supported the motion on the basis of repeated assurances by the Opposition that no precedent was being set.

The idea that ministerial staff members should not be compelled to appear before parliamentary committees – which is known as the McMullan Principle – has its basis therefore in the basic structures of Westminster parliamentary democracy and the doctrines of responsible government and ministerial responsibility. The foundation of these fundamental principles is the idea that the executive government is subject to parliamentary scrutiny and is accountable to Parliament (and, by extension, to the people). Accountability is maintained through the Ministers, who are presumed to be responsible for all actions and decisions taken by their staff and department, and who may be required by the Parliament to account directly for these actions and decisions. In short, under the doctrine of ministerial responsibility, it is Ministers themselves who are required to be accountable to Parliament, and not their ministerial staff.^[260]

The idea of ministerial responsibility is one of the key arguments that is generally relied upon in support of the McMullan Principle as a principle that should be consistently applied. The second key argument that is regularly invoked is that it is important to guarantee a level of confidentiality in discussions between Ministers and their ministerial advisors, which would be undermined if those advisors could be compelled to appear before parliamentary committees.^[261]

On the other hand, it would be an obvious concern in terms of accountability if the McMullan Principle operated in such a way as to prevent parliamentary committees from being able to obtain relevant information and allow governments to avoid legitimate parliamentary scrutiny. This would not be a problem if the doctrine of ministerial responsibility was rigorously applied such that Ministers were answerable for not only their own actions, but also those of their staff. The difficulty arises when Ministers are able to claim ignorance of the actions of their advisors. In such a case, the McMullan Principle undermines accountability by allowing Ministers to assert plausible deniability while the advisors themselves are not subject to questioning by parliamentary committees.^[262]

The potential tension here was identified by Lucy Hare when considering the position of ministerial advisors in New Zealand (being a role identified within that jurisdiction as personally appointed advisors to Cabinet Ministers or ministerial assistants). Hare acknowledged that the effectiveness of a ministerial advisor depends largely on the strength of their relationship with their individual Minister, and that confidentiality is a significant aspect of this. On the other hand, it is obviously important that executive power – whether exercised by the Minister or their staff – is subject to accountability checks and parliamentary scrutiny. As Hare observed:

full accountability where personal appointees are concerned may be impeded by the secrecy of their role – particularly where a personal appointee undertakes functions without explicit instruction from the Minister. There is thus a tension between political reality and the constitutional imperative of accountability.^[263]

There are significant parallels that can be drawn between the position of ministerial staffers and public servants in terms of accountability. The arguments made in favour of the McMullan Principle are also applicable in the case of public servants. For example, ministerial responsibility extends to actions and decisions taken by their departments. It is obviously also important for there to be a relationship of trust between ministers and public servants in order for public servants to be able to meet the mandated Australian Public Service values of providing the government with frank and honest advice.^[264] While public servants are compellable witnesses, there are protocols that govern the evidence they can give (which are discussed in further detail below), with some key limitations being established in an effort to accommodate the concerns related above regarding ministerial responsibility and retaining relationships of trust and confidence.^[265] The fact that public servants routinely give evidence before parliamentary committees, and that this is no longer seen as particularly controversial, provides an interesting comparison when considering the debate surrounding the McMullan Principle.

It may be, however, that these parallels between ministerial advisors and public servants only extend so far. In this context, it is interesting to note the differences between ministerial staff and public servants that was highlighted in the Prime Minister’s Guide on Key Elements of Ministerial Responsibility:

Ministers’ direct responsibility for actions of their personal staff is, of necessity, greater than it is for their departments’. Ministers have closer day-to-day contact with, and direction of the work of, members of their staff. Furthermore, ministerial staff do not give evidence to parliamentary committees, their actions are not reported in departmental annual reports, and they are not normally subject to other forms of external scrutiny, such as administrative tribunals.^[266]

In relation to ministerial advisors, Ian Killey, in Constitutional Conventions in Australia: An Introduction to the Unwritten Rules of Australia’s Constitutions, has concluded that the McMullan Principle has gained the status of a constitutional convention. He argues that a constitutional convention does exist that ministerial advisors do not appear before parliamentary committees.^[267] This is not, however, a conclusion without controversy.

The existence of a constitutional convention precluding ministerial advisors from appearing before parliamentary committees has increasingly been called into question in recent times.^[268] For example, Dr Yee-Fui Ng conducted interviews with a number of current and former Ministers and Members of Parliament about their beliefs as to whether such a constitutional convention exists. Dr Ng notes that the existence of such a belief amongst parliamentarians is an essential factor to consider when assessing whether such a constitutional convention has formed.^[269] This accords with the long-standing observation that ‘the terms of conventions are what the actors believe them to be’.^[270] The interviews conducted by Dr Ng showed that there was no positive consensus that such a constitutional convention existed.^[271]

Interestingly, interviewees appeared willing to change their position on the issue depending on whether they were in government or opposition!^[272] This in itself does tend to suggest that the McMullan Principle may be more of a political principle invoked when convenient rather than an entrenched constitutional convention applied consistently. Indeed, it may reflect a broader trend that can unfortunately be discerned when it comes to the power (and willingness) of the Parliament, and particularly the Upper House, to hold the Executive Government to account. That is, powers and processes designed to ensure accountability ‘tend rigorously to be asserted by the opposition of the day, and resisted by whomever is in government’.^[273]

Dr Ng concluded that ‘the conventional requirement that the rule be considered binding by political participants is not satisfied at the Commonwealth and Victorian level’.^[274] A similar conclusion was also reached in Odgers’ Australian Senate Practice, which provides that ministerial staff ‘have no immunity against being summoned to attend and give evidence, either under the rules of the Senate or as a matter of law’.^[275] To illustrate this point, it was noted that there were past examples of ministerial staff giving evidence before Senate committees, with one of the earliest examples being back in 1975 when the Private Secretaries to the Prime Minister and the Minister for Labour and Immigration appeared before a Senate Standing Committee on Foreign Affairs and Defence.^[276]

Therefore, there appears to be some disagreement about whether the McMullan Principle should be considered a constitutional convention amongst academics and commentators. As noted above, however, the key questions are whether the political participants as a group acknowledge the existence of the principle, and whether it is accepted as being binding in practice. To this end, an examination of recent examples where the McMullan Principle was invoked will provide an important insight into whether or not the principle can be considered to be an established constitutional convention or not.

IV INVOKING THE MCMULLAN PRINCIPLE IN PRACTICE

The question of whether ministerial advisors can be compelled to appear before parliamentary committees has arisen periodically in recent decades, and consideration of these past examples of practice is central to determining whether the McMullan Principle has become entrenched as a constitutional convention. An examination of these past examples, however, demonstrates that there is not the consistent and uniform practice that would be expected if the McMullan Principle had been accepted as a binding constitutional convention. Further, the routinely diametrically opposed views taken by those in government and those in opposition (and the readiness of individuals to alter their stated view depending on whether they are in government or opposition at the particular time in question) strongly suggests that the McMullan Principle could be better described as a principle of political expediency, rather than an entrenched constitutional convention.

To begin with, there are examples of ministerial advisors voluntarily appearing before parliamentary committees. One early example was in 1989 when the head of the National Media Liaison Service (then Mr Colin Parks)^[277] agreed to appear before Senate Estimates Committee D and answered questions when his presence was requested by members of the Committee.^[278] There have also been examples where Ministers have initially refused to allow ministerial advisors to appear, but have then declined to test the resolve of Parliament when it has insisted upon such an appearance. One example of this is the situation that gave the McMullan Principle its name, which was briefly outlined above.^[279] While in 1995 the Minister for Finance initially refused to allow the head of the National Media Liaison Service (then Mr David Epstein) to appear before the Finance and Public Administration Legislation Committee, the Senate decided to take the matter further and passed a resolution directing him to appear, which he subsequently did.^[280]

There have also been a number of high profile examples in which Ministers have refused to allow ministerial staff to appear before parliamentary committees, citing the existence of a constitutional convention that precludes them from being called as witnesses. One interesting aspect of each of these examples is the interplay between the Executive and the Parliament, with the former tending to rely on the McMullan Principle and claim that it is a well-established constitutional convention, while the latter tend to dispute this claim. Recent examples include: the Children Overboard inquiry in 2002, the Hotel Windsor inquiry in 2010, and the Orange Grove inquiry in 2004. Each of these will be discussed below in turn.

A The ‘Children Overboard’ Inquiry

The ‘Children Overboard’ inquiry arose following claims that asylum seekers who had been intercepted in Australian territorial waters by the Australian Navy had thrown children from their boat into the water. These claims were originally made by the then Minister for Immigration, the Hon. Phillip Ruddock MP, during a media briefing on 7 October 2001, and were later repeated by other senior government members, including the Prime Minister. Photographs were also presented that initially appeared to support these claims. The ‘Children Overboard’ claims became a significant issue during the election campaign, being a campaign in which border protection issues were considered by many to have been a central factor.

It subsequently became clear that no children had been thrown overboard and that the evidence originally thought to support these claims did not actually do so. A central question related to precisely when the Government had become aware that the ‘Children Overboard’ claims were inaccurate, and whether they knew of this during the election campaign. The Senate Select Committee on a Certain Maritime Incident (‘Senate Select Committee’) was established to inquire into the ‘Children Overboard’ incident^[281] with a key aspect of this being ‘the flow of information about the incident to the Federal Government, both at the time of the incident and subsequently’.^[282]

In relation to the ‘Children Overboard’ incident the Senate Select Committee concluded that ministerial advisors:

played a significant part in the failure of ministers to correct the public record. Their interactions with public servants and Defence officials, and the way in which they managed information flows in and out of ministers’ offices, raise numerous questions about the appropriateness of their performance, let alone matters of courtesy and fair dealing.^[283]

The Senate Select Committee was, however, unable to hear evidence from, or to question, any ministerial advisors. Throughout the ‘Children Overboard’ inquiry, a whole-of-government decision was taken to refuse to allow ministerial staff to appear before the Senate Select Committee.

For example, Prime Minister John Howard stated that:

In my view, ministerial staff are accountable to the minister and the minister is accountable to the parliament and, ultimately, the electors ... What we are doing in relation to this issue is following the convention, and the convention is that Ministerial staff do not appear.^[284]

The Senate Select Committee (unsurprisingly) adopted a different view, concluding in its Main Report that this type of blanket refusal was ‘anathema to accountability’^[285] and that ‘[t]he time has come for a serious, formal re-evaluation of how ministerial staff might properly render accountability to the parliament and thereby to the public’.^[286] This conclusion was supported by advice received from the Clerk of the Senate who concluded that the Senate (and comparable houses of legislatures) had not recognised any immunity attaching to ministerial advisors, and furthermore:

there is a strong case for subjecting ministerial personal staff to compulsion in legislative inquiries, on the basis that their role is manifestly now not confined to advice and personal assistance ... they act as de facto assistant ministers and participate in government activities as such ... Moreover, ministers no longer accept full responsibility for the actions of their staff.^[287]

The advice received by the Senate Select Committee was not however uniform. The advice provided by the Clerk of the Senate was itself based on legal advice provided by Bret Walker SC who concluded that ‘former Ministers and Ministerial staff have no immunity from compulsory attendance to give evidence and produce documents to a Senate committee’.^[288] This contrasted with advice given by the Clerk of the House of Representatives who concluded that there was a ‘reasonable case’^[289] for ministerial immunity also being extended to ministerial advisors.

B The ‘Hotel Windsor’ Inquiry

The ‘Hotel Windsor’ inquiry was established by the Standing Committee on Finance and Public Administration, which is a committee of the Victorian Legislative Council. The inquiry was designed to investigate the Windsor Hotel redevelopment planning process and was based on the inadvertent release to the media through email of a media plan by the Office of the Minister for Planning that appeared to outline a strategy of using a manufactured public consultation process to provide grounds for the rejection of the development despite the formal planning process being expected to recommend its approval.

In March 2010 the Standing Committee resolved to conduct public hearings and to take evidence from a number of witnesses, including four ministerial media advisors.^[290] The Victorian Attorney-General formally refused to allow the ministerial advisors to appear before the Standing Committee, claiming that:

[t]he conventions concerning Ministerial advisors are well established and acknowledged in all Australian Parliaments. The convention, to put it simply, is that advisors are not summonsed by Committees.^[291]

The Attorney-General argued that the attempts to compel the ministerial advisors to give evidence were a ‘total breach of practice and a total breach of procedure’.^[292] The key reason put forward by the Attorney-General in support of this claimed convention was ‘the public interest in the security of communications between Ministers and their advisors’.^[293]

The Standing Committee issued a number of summonses compelling the ministerial advisors to appear and give evidence, but on each occasion correspondence was received indicating that the Attorney-General had directed the witnesses not to attend the hearing.^[294] The Standing Committee reported to the Legislative Council that the intervention of the Attorney-General in directing witnesses not to give evidence ‘represents a significant interference in the Committee’s functions and ability to fully investigate the issues within its terms of reference’.^[295]

It is interesting, but perhaps not at all surprising, to note that in both this and the ‘Children Overboard’ example, while the relevant Ministers strongly asserted the existence of a long-standing constitutional convention, serious doubt as to the existence of such a convention was expressed by the relevant Parliamentary Committees. In this case the Standing Committee, in its Second Interim Report (August 2010), cited earlier advice that had been given to the Legislative Council by both the Clerk of the Legislative Council and Bret Walker SC in 2007. Both confirmed that ministerial staff had no immunity at law against being summoned to give evidence before a parliamentary committee.^[296] The Standing Committee itself concluded that:

in the event that a Minister denies knowledge of a state of affairs or has distanced him/herself from a public servant or advisor’s actions and that public servant or advisor has a direct involvement in the state of affairs in question, there may be grounds for the public servant or advisor to be answerable to the Committee.^[297]

The Standing Committee in this case sought further advice from the Clerk of the Legislative Council. They were advised that, in general, any action that obstructed the Legislative Council in the performance of its functions could be treated as contempt and, more specifically, that directing somebody not to attend a hearing in response to a summons would be contempt. However, the Clerk further advised in relation to the direction issued by the Attorney-General that ‘no further action is possible in the Legislative Council as the Attorney-General is a member of the Legislative Assembly and responsible only to that House and not to the Legislative Council’.^[298]

The final recommendation of the Standing Committee in its Second Interim Report was that the Legislative Council should resolve to order the four witnesses to appear before the Standing Committee.^[299] This set the stage for a potentially serious conflict between the Parliament and Executive, which was only averted by the fact that the 56^th Victorian Parliament expired on 2 November 2010, with a State election held shortly thereafter and resulting in a change in government.

The change in government did not entirely end the matter. The Standing Committee had previously (in June 2010) also referred the matter to the Victorian Ombudsman for investigation.^[300] During the course of this investigation 38 witnesses were interviewed, including the relevant Minister, several Members of Parliament and ministerial staff. This included a number of ministerial staff who had previously refused to give evidence before the Standing Committee.^[301] The Ombudsman noted that all witnesses cooperated with the investigation,^[302] although it was also remarked that under the Ombudsman Act 1973 (Vic) the Ombudsman had the statutory power to summon witnesses and interview any person who has information relevant to the investigation.^[303] The Ombudsman ultimately concluded that there were concerns regarding the probity of the planning and heritage approval processes for the Windsor Hotel redevelopment. Amongst a number of recommendations made by the Ombudsman included recommendations relating to the accountability of ministerial advisors for checking the accuracy of information, and that the Victorian Government conduct a comprehensive review of the Ministerial Staff Code of Conduct ‘to ensure that ministerial staff are held accountable to appropriate standards of ethical and professional conduct’^[304] and to ‘recognise the executive decisions are the preserve of Ministers and public officers and not ministerial staff acting in their own right’.^[305]

C The ‘Orange Grove’ Inquiry

Another example of the constitutional convention not being accepted in practice stems from the 2004 ‘Orange Grove’ inquiry in NSW. The inquiry was conducted by the General Purpose Standing Committee No. 4, which is a committee of the NSW Legislative Council. It involved an examination of the approval process relating to the Designer Outlets Centre on Orange Grove Road in Liverpool, and in particular involved an examination of the circumstances surrounding the approval and the role of relevant Ministers in dealing with the development. In its Report the Standing Committee noted that it had invited a number of ministerial staff to appear before the Committee, and observed:

[a]lthough there are no restrictions on the power of a committee of the Legislative Council to invite ministerial staff as witnesses before a committee, there has been a general political convention, which has resulted in ministerial staff not being called as witnesses.^[306]

During the inquiry, a number of ministerial staff did appear voluntarily before the Standing Committee after being invited to do so. However, the Chief of Staff to the Assistant Planning Minister (Michael Meagher) indicated that he would decline his invitation to appear on the basis that the Assistant Planning Minister was invoking the constitutional convention and had not authorised his appearance. In fact, Meagher declined an invitation to appear on two occasions,^[307] and the Standing Committee then issued a summons for his appearance under the Parliamentary Evidence Act 1901 (NSW).^[308] This was the first time that any Standing Committee of the NSW Parliament had ever taken the step of issuing a summons to compel the appearance of a ministerial advisor.^[309] The summons was ultimately answered, and Meagher appeared before the Standing Committee on 30 August 2004.^[310]

V A CURRENT CONSTITUTIONAL CONVENTION?

The above examples highlight the diametrically opposed views that the Executive and the Parliament appear to take regarding the existence of the McMullan Principle as an established constitutional convention. While these cases are not a comprehensive examination of every example of the McMullan Principle being invoked in recent years, they are illustrative of the lack of consistent practice and the lack of general acceptance by political participants as a whole. These examples are therefore themselves sufficient to strongly suggest that the McMullan Principle is more of a political principle than an entrenched constitutional convention.

It was identified above that the key factors to consider when determining whether a particular practice has become entrenched as a constitutional convention are whether the principle is consistently acknowledged and accepted by political participants, and whether they consider it to be binding in nature. On both of these grounds the McMullan Principle fails to meet the threshold. In each of the above examples the invocation of the principle by the Executive has been met by an equally forceful rejection of the principle by the relevant Parliamentary Committee. There appears to be habitual disagreement between the Executive and Parliamentary arms of government regarding the existence of the McMullan Principle as an entrenched constitutional convention. The contestable nature of the principle is itself evidence that it has not attained the status of an established constitutional convention.

There is no doubt that, to the extent it is asserted by the Executive, the McMullan Principle has the practical impact of placing ministerial advisors in an invidious position when an individual ministerial advisor is invited to appear before a parliamentary committee but is then instructed by their relevant Minister to either not appear or not to answer particular questions. The untenable position that ministerial advisors find themselves in when confronted with this situation was recognised by the Senate Select Committee for an Inquiry into a Certain Maritime Incident. The Senate Select Committee ultimately found that whilst it did have the power to compel attendance by ministerial staff, it had ‘decided not to exercise its power to compel their attendance, and thereby expose the advisors to the risk of being in contempt of the Senate should they not respond to the summons’.^[311]

Part of the decision by the Senate Select Committee not to issue summons for ministerial advisors was based on the previously expressed view that it would be ‘unjust for the Senate to impose a penalty on a public servant who declines to provide evidence on the direction of a Minister’.^[312] Indeed in 1994 the Senate Committee on Privileges had called the exercise of parliamentary powers to impose penalties on public servants in such circumstances ‘untenable’.^[313] Instead, as Peter Hanks QC observed:

any confrontation between the witness’s duty to her or his Minister and the witness’s obligation to the House or the Committee is likely to be diverted into a confrontation between the House or the Committee and the Minister. That is what happened in 2007, when the Attorney-General openly instructed witnesses summoned before the Select Committee on Gaming Licensing not to answer questions or produce documents; and the Legislative Council then took formal (albeit symbolic) action, not against the witnesses, but against the Government in the Council.^[314]

This discretionary decision to look for alternative solutions is itself important to consider when examining whether a constitutional convention exists. Ian Killey has pointed to a number of examples where parliamentary committees have asserted the power to call ministerial staffers to appear before the committee, but have then failed to take any action to enforce that appearance. It is suggested that this reflects ‘the silent recognition of a convention by non-government parties’.^[315] It may be, however, that it instead reflects a realistic assessment of the practical difficulties of enforcement in these circumstances and the surrounding political sensitivities that inform such decisions. As was noted by the Finance and Public Administration References Committee when considering the question of enforcement in relation to a related topic:

[t]here are no effective deterrents for non-compliance with the order. The Senate has no remedies to enforce its powers against Ministers who are members of the House of Representatives; its penalties in the Senate, such as suspending Ministers from the chamber, are ineffective; and it would be unfair for the Senate to punish public servants for following Ministers’ directions.^[316]

Indeed, in Western Australia the Policy for Public Sector Witnesses Appearing Before Parliamentary Committees produced by the Public Sector Commissioner recognises that where public servants appearing before a Parliamentary Committee are given a lawful directive by a Minister not to release information or documents or answer a question they must comply with that directive. In such circumstances ‘it will be for the Committee to take the matter up with Minister or the Minister representing the Minister in the relevant House’.^[317]

The failure of Parliament to take enforcement action does, however, suggest an implicit acceptance by the Parliament that an Executive direction should take priority over an express order from Parliament. The unwillingness of the Parliament to assert its authority and to test the resolve of the Executive in these cases is perhaps further evidence of the growing dominance of the Executive within the Australian political landscape. This point has previously been made by Dr Ian Holland, who observed:

[i]t is worth considering exactly why it is sometimes claimed that public servants should not be confronted by the powers of the chambers of parliament. Implicit in the statement that one should not penalize a public servant who is acting on the directions of a Minister is a concession that the Minister has the legal authority to issue directions to someone to defy the Senate or House of Representatives. In making this concession, those who claim to be seeking to assert the power of the Senate are in fact deferring to the power of the executive and are encouraging the public servants (and probably ministerial staff) to do the same. This seems to rest uneasily with the Parliament’s declarations on powers and immunities, and the limited case law that exists in this area.^[318]

Phil Larkin sees this inability of Parliament ‘to put into force its power to compel individuals without immunity from appearing before committees’^[319] as a practical limitation that renders the question about whether there is a constitutional convention ‘somewhat academic’.^[320]

VI THE FUTURE OF THE MCMULLAN PRINCIPLE

It has been concluded above that the McMullan Principle has not – despite the claims periodically made by members of Executive Governments – attained the status of an entrenched constitutional convention. It may not currently be an accepted constitutional convention, but should it be? In this concluding section the paper will go on to consider whether there are any underpinning policy reasons that support the future entrenchment of the principle.

The arguments in favour of the McMullan Principle, which have been outlined above, emphasise the unique nature of the role of the ministerial advisor. The importance of preserving ministerial accountability and the confidentiality and trust between an advisor and their individual Minister have both been put forward as key arguments in favour of the McMullan Principle. To consider the continued persuasiveness of these claims it is necessary to briefly examine the role of the ministerial advisor and, in particular, the way that this role has evolved in recent times.

A The Evolving Role of the Ministerial Advisor

Ministerial advisors were first introduced by the Whitlam Government in the early 1970s. They are distinct from both electoral and departmental staff, and are employed to provide direct support to Ministers in their ministerial roles. Ian Killey describes ministerial advisors as being different from other public servants in that ‘they are neither apolitical, impartial, and certainly not independent’.^[321] The role of a ministerial advisor has expanded over time, with the job of the modern ministerial advisor being described as follows:

[t]hey play an important part in the management of the flow of information into and within the Ministerial office – ensuring that relevant information is obtained in a timely manner, that accurate records are kept, and that where necessary that information reaches the Minister in a suitable form. They may also interface with the bureaucracy, other ministers’ offices and other stakeholders, including giving directions to departments and agencies. If they have been employed because of their particular expertise they may help shape the policy agenda, give the Minister policy advice, and write speeches on that area. Increasingly, they manage media perceptions and reporting. And they are often instrumental in ‘delivering’ policy initiatives – keeping major players focused and cooperating until the initiative has been implemented.^[322]

An increasingly important aspect of this role is liaising between Ministers and the public service.^[323] Whereas traditionally the ministerial advisor had no delegated authority to direct public servants, this has gradually changed over time. Examples such as the ‘Children Overboard’ incident^[324] illustrate the potentially independent role that may be played by modern ministerial advisors, particularly in terms of their possible ‘gatekeeper’ role in determining exactly what specific information is passed from the public service through to the Minister. The modern ministerial advisor will now frequently be seen to ‘act as a conduit between the [Departmental] Secretary and the Minister, often injecting policy advice along the way’.^[325]

Andrew Alexandra and Clare McArdle describe the role of the advisor in Australia as being inherently dualistic.^[326] That is, while ministerial advisors are partisan appointments who are employed by the Minister, they work within the broader parliamentary sphere and have a role within both the executive and legislative arms of government.

These potentially competing interests have been expressly noted in the United Kingdom in the Code of Conduct for Special Advisors which describes special advisors (as they are known in that jurisdiction) in the following way:

They add a political dimension to the advice and assistance available to Ministers while reinforcing the political impartiality of the permanent Civil Service by distinguishing the source of political advice and support.

Special advisors should be fully integrated into the functioning of government. They are part of the team working closely alongside civil servants to deliver Ministers’ priorities. They can also help Ministers on matters where the work of government and the work of the government party overlap and where it would be inappropriate for permanent civil servants to become involved. They are appointed to serve the Prime Minister and the Government as a whole, not just their appointing Minister.^[327]

Alexandra and McArdle note that there are a ‘general lack of clear conventions or guidelines regarding the nature of the role of ministerial advisor and the responsibilities consequent on that role’^[328] and that this is reflective in a lack of consistency at a practical level across Ministers’ offices. This was also noted by the Senate Select Committee examining the ‘Children Overboard’ incident, which remarked that while the Prime Minister’s Guide on Key Elements of Ministerial Responsibility provided some guidance at the Commonwealth level, it dealt almost entirely with ‘possible conflicts between their individual self interest and the interests of their minister’ and did not consider ‘problems that might arise through the ministerial advisor’s pursuit of what they perceive as the interests of their minister or their party’.^[329] Alexandra and McArdle have concluded that:

[s]uch lack of clarity as to the role of the ministerial advisor within the parliamentary system of government must make it very difficult for advisors to know how to act when faced with conflicts between their immediate political loyalties and the broader commitment to open and accountable government.^[330]

There has been some forward movement in this area in recent years with, for example, the introduction of the Code of Conduct for Ministerial Staff at the Commonwealth level in 2008. This relevantly provides that ministerial staff employed under the Members of Parliament (Staff) Act 1984 (Cth) must:^[331]

• Acknowledge that ministerial staff do not have the power to direct [public servants] in their own right and that [public servants] are not subject to their direction;

• Recognise that executive decisions are the preserve of Ministers and public servants, and not ministerial staff acting in their own right; and

• Facilitate direct and effective communication between their Minister’s department and their Minister.

While the Code of Conduct for Ministerial Staff clearly recognises the evolution of the role played by ministerial staff, it may not be a particularly effective mechanism in terms of providing for enhanced accountability. The first issue is that:

it makes no provisions for accountability to Parliament, instead limiting advisors’ executive role on the basis that if they have no executive role, they need have no direct accountability to Parliament.^[332]

The second issue is that there are no clear sanctions provided for any breaches of the Code, with breaches being dealt with by the Prime Minister’s Chief of Staff and the relevant Minister. Perhaps more fundamentally, the Code does not mention any obligations that ministerial staff may have towards the Parliament other than a general obligation that they make themselves aware of both the APS Values and Code of Conduct which bind Parliamentary Service employees, and that they do not knowingly or intentionally encourage or induce a public official to breach the law or parliamentary obligations. This is despite ministerial staffers being formally employed under the Members of Parliament (Staff) Act 1984 (Cth), and highlights the possible tension that exists given the duality of the role played by ministerial staffers as highlighted by Alexandra and McArdle above.

Those advocating the McMullan Principle as an entrenched constitutional convention and preferred practice appear, however, to base this assertion on a traditional understanding of the advisor’s role, namely that ministerial advisors are working solely under the direct instruction of their responsible Minister. When this is the case ‘the chain of accountability running through the Minister to Parliament may hold’^[333] as the Minister will be able to answer questions relating to the activities of the advisor and can be held accountable for those actions.

Increasingly, however, it appears that the role of ministerial advisors has evolved

to a point where they enjoy a level of autonomous executive authority separable from that to which they have customarily been entitled as the immediate agents of the Minister.^[334]

The Senate Select Committee in the Children Overboard example found this evolution of the role of ministerial advisors to be a key factor that has led to what they identified as ‘a serious accountability vacuum at the level of ministers’ offices’.^[335] Similarly, the Victorian Ombudsman recommended following his investigation into the Hotel Windsor redevelopment that the Ministerial Staff Code of Conduct in Victoria should be reviewed to recognise that ministerial advisors acting in their own right do not have the power to direct public servants or to make executive decisions.^[336]

Modern ministerial advisors ‘actually have a good deal of independence and discretion in much of their work, and this seems to be integral to the role’.^[337] This creates a potential accountability vacuum, as Ministers are able to claim ignorance of the actions of their advisors while the advisors themselves are not subject to parliamentary scrutiny by being required to answer questions before parliamentary committees. Phil Larkin has argued that examples such as the ‘Children Overboard’ inquiry suggest that this ‘accountability gap is not simply potential or hypothetical but very real, and that it has indeed been used by ministers to dodge accountability’.^[338]

Indeed, during the original Senate debate from which the McMullan Principle takes its name, the Opposition noted that the Government had been careful to distinguish between the role of the National Media Liaison Service^[339] and ‘the more political role that ministerial staff often perform’.^[340] Over time, the role of the ministerial advisor has evolved into one that is highly political and increasingly autonomous. As was noted by Dr Stanley Bach:

[i]n principle, these advisors are not supposed to have managerial authority. So when an advisor tells public servants that their Minister wants something done, the advisor is assumed to be speaking for the Minister and acting at the Minister’s direction. In practice, however, some ministerial advisors have been given managerial responsibilities, and it can be very tempting for other advisors to give directions to public servants on the basis of what the advisor thinks the minister would want if he had been consulted or on the basis of what the advisor believes is in the minister’s best interests.^[341]

When this is combined with the significantly increased number of ministerial staffers,^[342] it becomes clear that if the McMullan Principle is recognised as an established practice that should be routinely applied, this will, in light of the way that ministerial advisor roles have evolved, have significant implications for executive accountability and the effectiveness of the parliamentary committee system.

Indeed, Bob McMullan himself later recognised the potential accountability gap when he reconsidered these issues from the perspective of Opposition and argued that whilst the McMullan Principle was ‘in the normal course’^[343] correct:

that means you have to accept responsibility for what your staff do. You cannot say: ‘They’re responsible to me but I do not care what they do; I am not going to tell you what they do. If they make a mistake, it is nobody’s business.’ Then there is a black hole of accountability because they deal with the departments. They give instructions; they receive directions ... There is a big black hole in Australian accountability, and either ministers have to accept responsibility for what their staff do or staff have to be accountable. It cannot be that nobody is accountable.^[344]

The importance of ensuring that ministerial advisors are now subject to some formal level of accountability was expressly recognised in 2010 by the Victorian Public Sector Standards Commissioner in the Review of Victoria’s integrity and anti-corruption system. The Commissioner noted that while accountability for ministerial advisors had traditionally been addressed through the convention of individual ministerial responsibility, the evolution of the role played by ministerial advisors has meant that this may no longer be the best approach:

This makes the traditional approach of ministerial accountability for the actions of their staff increasingly challenging and increases the likelihood that ministers could distance themselves from the actions (or inactions) of their staff. It has resulted in suggestions that ministerial officers should be more directly subject to scrutiny.

As ministerial officers, parliamentary advisors and electorate officers are publicly funded, the Review considers that they should be subject to external scrutiny to ensure they act with integrity.^[345]

The evolving role of the ministerial advisor, and in particular the evidence suggesting that advisors are increasingly able to act with a significant degree of independence and discretion, provides strong support for the view that the McMullan Principle should be revisited and that its primary rationale (in terms of the doctrine of ministerial responsibility) is no longer always applicable.

B Improving the Accountability of Ministerial Advisors

The changing nature of the ministerial advisor role is one factor that weighs strongly against the continued application of the McMullan Principle into the future, with the increased independence being exercised by some ministerial advisors undermining arguments that the principle is a necessary application of the doctrine of ministerial responsibility. The above statement by Bob McMullan identifies two potential options when it comes to improving accountability in relation to ministerial advisors – ‘either ministers have to accept responsibility for what their staff do or staff have to be accountable’.^[346]

That is, if the McMullan Principle is going to be invoked, it should only be in circumstances where the Minister themselves is prepared to appear before the Committee and accept responsibility for an issue. There is, however, no guarantee that this will occur in every situation. For example, while the Senate may order a Senator to attend and give evidence before a Senate Committee,^[347] there is an accepted limitation (whether based on comity or law) on the power of either House of Parliament to summons witnesses from either the other House or from a State or Territory Parliament, although their attendance may be requested in a voluntary capacity.^[348] Indeed, in the Children Overboard incident, the Senate Select Committee noted that the relevant Minister had failed to submit information or appear before the Inquiry (indeed, he declined three successive invitations to appear), and argued that this ‘further eroded public confidence in the government’.^[349]

Absent this direct ministerial accountability, an insistence upon the McMullan Principle does appear to expose a concerning accountability gap when it comes to ministerial advisors. However, it is also important to recognise that even if parliamentary committees are able to compel ministerial advisors to appear and answer questions, this does not mean that there are no limits surrounding that questioning. To this end, as initially noted above, parallels may be drawn between ministerial advisors and public servants. Indeed, the accepted limitations that are applied to questioning of public servants by parliamentary committees would seem to be equally appropriate in the case of ministerial advisors.

For example, Bret Walker SC has drawn a distinction between the ability of a parliamentary committee to summons a public servant to appear before the committee and their ability to compel that public servant to answer particular questions when they do appear. It was suggested that there is a convention that public servants cannot be compelled to answer questions about policy ‘in such a way as to endanger the necessary confidence between Ministers and public servants’.^[350] Similarly, the Clerk of the Legislative Council observed that, in this respect, ministerial advisors ‘should generally not be held accountable for matters of opinion on policy, which is the domain of Ministers’.^[351] This reflects the Parliamentary Privilege Resolutions agreed to by the Australian Senate which provides that:

[a]n officer of a department of the Commonwealth or of a state shall not be asked to give opinions on matters of policy, and shall be given reasonable opportunity to refer questions asked of the officer to superior officers or to a minister.^[352]

Similarly, Alexandra and McArdle have recommended that whilst the McMullan Principle should be weakened in order to ensure appropriate accountability and parliamentary scrutiny, protocols should be established similar to those applying to public servants. For example:

Public servants cannot give information on matters of policy; certain material is protected by public interest immunity, and in camera evidence may be given where desirable ... Advisors should also be shielded from questioning about the personal life of the minister, opinions expressed in the office, policy or media advice offered or discussed, and party matters. They should however be required to respond honestly to questions as to whether advice from the bureaucracy was passed onto the minister, whether information was received in the minister’s office, and whether instructions from the Minister were passed to the bureaucracy.^[353]

While it appears strongly arguable that ministerial advisors are able to be summonsed before a parliamentary committee – with this paper establishing above both that the McMullan Principle is not an entrenched constitutional convention and that it has diminishing applicability given the evolution of the ministerial advisor role – this does not mean that a parliamentary committee should consider itself to have carte blanche when questioning such a witness. Given the significant parallels that can be drawn between the position of ministerial advisors and public servants there would appear to be sound policy reasons for applying similar protocols to those which currently apply to public servants when giving evidence before parliamentary committees.

VIII CONCLUSION

The examples discussed above suggest that the McMullan Principle has not attained the status of an entrenched constitutional convention. The result of this conclusion is that a parliamentary committee may well have the legal authority in a particular case to compel ministerial advisors to appear and give evidence, although whether a committee has the political will to enforce such an appearance is another question. However, where a ministerial advisor does appear before a parliamentary committee, there are sound policy reasons for applying similar protocols to those which currently apply to public servants when they give evidence before parliamentary committees. This reflects the importance of striking a balance between parliamentary accountability and the need to also preserve a level of confidence in the relationship between individual Ministers and their advisors.

---

* Dr Danielle Bozin, Lecturer, Queensland University of Technology, Australia.

¹ Hague Convention of 25 October 1980 on the Civil Aspects of International Child Abduction, opened for signature 25 October 1980, 1343 UNTS 89 (entered into force 1 December 1983); The Convention is given effect to by the Family Law (Child Abduction Convention) Regulations 1986 (Cth).

^[2] Convention art 13(b). See also Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(b). See generally Carol S Bruch, ‘The Unmet Needs of Domestic Violence Victims and Their Children in Hague Child Abduction Convention Cases’ (2004) 38(3) Family Law Quarterly 529; Merle H Weiner, ‘International Child Abduction and the Escape from Domestic Violence’ (2000) 69(2) Fordham Law Review 593; Miranda Kaye, ‘Hague Convention and the Flight from Domestic Violence: How Women and Children are Being Returned by Coach and Four’ (1999) 13(2) International Journal of Law, Policy and the Family 191; Regan Fordice Grilli, ‘Domestic Violence: Is it Being Sanctioned by the Hague Convention?’ (1997) 4(1) Southwestern Journal of Law and Trade in the Americas 71; Roxanne Hoegger, ‘What if She Leaves – Domestic Violence Cases Under the Hague Convention and the Insufficiency of the Undertakings Remedy’ (2003) 18(1) Berkeley Women’s Law Journal 181; Taryn Lindhorst and Jeffery L Edleson, Battered Women, Their Children, and International Law: The Unintended Consequences of the Hague Child Abduction Convention (Northeastern University Press, 2012); Suzanne Christie ‘The ‘Grave Risk’ or ‘Intolerable Situation’ Defence in Cases of Domestic Violence Under the Hague Child Abduction Convention’ (2013) 3(4) Family Law Review 191; Galit Moskowitz ‘The Hague Convention on International Child Abduction and the Grave Risk of Harm Exception’ (2003) 41(4) Family Court Review 580.

^[3] See, eg, Bassi, DK and Director General of Community Services [1994] FLC 92–465; Murray v Director of Family Services ACT [1993] FamCA 103; [1993] FLC 92-416; Gsponer v Johnstone [1988] FamCA 21; (1988) 12 Fam LR 755; Laing v Central Authority [1996] FLC 92-709; Director-General Department of Families, Youth and Community Care v Hobbs [1999] FamCA 2059 (24 September 1999); P v Commonwealth Central Authority [2000] FamCA 461 (19 May 2000); Director-General, NSW Department of Community Services v JLM [2001] FamCA 1338; (2001) 28 Fam LR 243; State Central Authority, Secretary to the Department of Human Services v Mander [2003] FamCA 1128 (17 September 2003); State Central Authority v Papastavrou [2008] FamCA 1120 (22 December 2008); State Central Authority v Sigouras [2007] FamCA 250 (23 March 2007); HZ v State Central Authority [2006] FamCA 466 (6 July 2006).

^[4] [2000] FamCA 461 (19 May 2000).

^[5] Elisa Peréz-Vera, Explanatory Report of the Convention on the Civil Aspects of International Child Abduction, Actes et Documents of the XIVth Session, Vol III, 1980, 426, 430–2.

^[6] Michael Freeman, ‘The Best Interests of the Child? Is the Best Interests of the Child in the Best Interests of Children?’ (1997) 11(3) International Journal of Law Policy and the Family 360.

^[7] Michael Kirby, ‘Children Caught in Conflict – The Child Abduction Convention and Australia’ (Speech delivered at the Inaugural Peter Nygh Memorial Lecture, Halifax, Nova Scotia, Canada, 23 August 2009) <http://www.michaelkirby.com.au/images/stories/speeches/2000s/2009+/2424B.Peter_Nygh_Lecture-Halifax_2009.pdf> . See, eg, the minority opinions in De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640; DP v Commonwealth Central Authority (2001) 206 CLR 401; MW v Director-General of the Department of Community Services [2008] HCA 12 (28 March 2008); LK v Director-General, Department of Community Services [2009] HCA 9; (2009) 237 CLR 582.

^[8] De L v Director-General, NSW Department of Community Services (1996) 187 CLR 640, which was an appeal from De Lewinski v Director-General, New South Wales Department of Community Services [1996] FLC 92-674 (Full Court of the Family Court of Australia); DP v Commonwealth Central Authority (2001) 206 CLR 401, which was an appeal from Director-General, NSW Department of Community Services v JLM [2001] FamCA 1338; (2001) 28 Fam LR 243 (Full Court of the Family Court of Australia); MW v Director-General of the Department of Community Services [2008] HCA 12 (28 March 2008), which was an appeal from Wencelslas v Director-General, Department of Community Services [2007] FamCA 398; (2007) 37 Fam LR 271 (Full Court of the Family Court of Australia); LK v Director-General, Department of Community Services [2009] HCA 9; (2009) 237 CLR 582, which was an appeal from Kilah v Director-General, Department of Community Services [2008] FamCAFC 81; (2008) 39 Fam LR 431 (Full Court of the Family Court of Australia); RCB as litigation guardian of EKV, CEV, CIV and LRV v The Honourable Justice Colin James Forrest [2012] HCA 47; (2012) 247 CLR 304; Garning v Director-General, Department of Communities (Child Safety Services) [2012] FamCAFC 35 (9 March 2012).

^[9] (2001) 206 CLR 401. This decision involved hearing two appeals together, with the appellants being DP and JLM respectively.

^[10] Ibid 418.

^[11] Ibid 423.

^[12] Convention preamble.

^[13] Pérez-Vera, above n 5, 426, 428.

^[14] See generally Joel R Paul, ‘Comity in International Law’ (1991) 32(1) Harvard International Law Journal 1; Lord Collins of Mapesbury et al (eds), Dicey, Morris and Collins: The Conflict of Laws (Sweet & Maxwell, 14th ed, 2010); Ronald Harry Graveson, Conflict of Laws (Sweet & Maxwell, 7^th ed, 1974).

^[15] Danielle Bozin-Odhiambo, ‘Reexamining Habitual Residence as the Sole Connecting Factor in Hague Convention Child Abduction Cases’ (2012) 3 Family Law Review 4.

^[16] Pérez-Vera, above n 5, 426, 431.

^[17] Paul, above n 14, 3.

^[18] Pérez-Vera, above n 5, 426; See also Guido Rennert, ‘Is Elimination of Forum Shopping by Means of International Uniform Law an ‘Impossible Mission?’’ [2005] MqJlBLaw 5; (2005) 2 Macquarie Journal of Business Law 119.

^[19] Pérez-Vera, above n 5, 426, 429.

^[20] Ibid 431.

^[21] Ibid 430.

^[22] The exceptions are as follows:

i. the grave risk of harm exception, see Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(b);

ii. the child objects to being returned to their habitual residence (if it is appropriate to take into account the child’s views given the child’s age and degree of maturity) exception, see Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(c)(i)−(iii);

iii. the left-behind parent was not exercising rights of custody or consented to, or acquiesced in, the removal or retention exception, see Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(a)(i)−(ii);

iv. the child’s return would offend basic principles of human rights and fundamental freedoms exception, see Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(d); and

v. if more than a year has passed and the child has become settled in their new environment exception, see Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(2)(c). Note that even if an exception to return is established by the abducting parent, judicial discretion to dismiss the return order application must still be exercised.

^[23] This is demonstrated by the fact that the High Court of Australia has overturned the Full Court of the Family Court of Australia’s narrow reading of the Convention’s exceptions to return in all Convention cases that have come before it.

^[24] DP v Commonwealth Central Authority (2001) 206 CLR 401, 414 (Gaudron, Gummow and Hayne JJ).

^[25] The problems associated with this were articulated in Laing v Central Authority (1996) 21 Fam LR 24, 33.

^[26] [2008] HCA 12 (28 March 2008).

^[27] Ibid [38].

^[28] Ibid [46]–[56].

^[29] Lindhorst and Edleson, above n 2; Jeffrey Edleson and Taryn Lindhorst, ‘Research for the Real World: Mothers and children seeking safety in the U.S.: A Study of International Child Abduction Cases Involving Domestic Violence’ (Speech delivered at the NIJ Research for the Real World Seminar, The National Institute of Justice, 12 October 2011) <http://www.nij.gov/multimedia/presenter/presenter-edleson/pages/presenter-edleson-transcript.aspx> Linda Silberman, ‘The Hague Child Abduction Convention Turns Twenty: Gender Politics and Other Issues’ (2000) 33 New York University Journal of International Law and Politics 221; Grilli, above n 2; Kaye, above n 2; Weiner, above n 2.

^[30] Nigel Lowe, A Statistical Analysis of Applications made in 2003 under the Hague Convention of 25 October 1980 on the Civil Aspects of International Child Abduction, Part 1 – Overall Report, Preliminary Document No 3, for the attention of the Fifth Meeting of the Special Commission to review the operation of the Hague Convention of 25 October 1980 on the Civil Aspects of International Child Abduction, October 2006, 22. See also Nigel Lowe, A Statistical Analysis of Applications made in 2003 under the Hague Convention of 25 October 1980 on the Civil Aspects of International Child Abduction, Part 1, Preliminary Document No 3, of September 2008 (2007 update).

^[31] See Danielle Bozin, ‘Equal Shared Parental Responsibility and Shared Care Post-Return to Australia under the Hague Child Abduction Convention’ (2014) 37(2) University of New South Wales Law Journal 603, 617.

^[32] Convention art 13(b). See also Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(b).

^[33] See, eg, Harris v Harris [2010] FamCAFC 221; (2010) 245 FLR 172; Director-General, Department of Families, Youth and Community Care v Bennett [2000] FamCA 253; (2000) 26 Fam LR 71; Director-General, Department of Families, Youth and Community Care v Hobbs [1999] FamCA 2059; [2000] FLC 93-007.

^[34] [1996] USCA6 311; 78 F 3d 1060 (6th Cir 1996).

^[35] See Director-General, Department of Families, Youth and Community Care v Bennett [2000] FamCA 253; (2000) 26 Fam LR 71; Director-General, Department of Families, Youth and Community Care v Hobbs [2000] FLC 93–007.

^[36] [1996] USCA6 311; 78 F 3d 1060, 1069 (6^th Cir 1996).

^[37] (2001) 206 CLR 401.

^[38] Undertakings will be discussed within this article in the context of an examination of specific cases. See Hoegger, above n 2.

^[39] The case law will be discussed within this article. See Jeanine Lewis, ‘Hague Convention on the Civil Aspects of International Child Abduction: When Domestic Violence and Child Abuse Impact the Goal of Comity’ (2000) 13 Transnational Lawyer 391.

^[40] [1994] FLC 92–465.

^[41] Ibid [35].

^[42] Ibid [60] (Johnston J).

^[43] Ibid [31].

^[44] Ibid [7] (Johnston J).

^[45] Ibid [33] (Johnston J).

^[46] [1990] FamCA 119; (1990) 14 Fam LR 381.

^[47] [1994] FamCA 163; [1994] FLC 92-465 [34].

^[48] Ibid (Johnston J).

^[49] Ibid.

^[50] [1993] FLC 92–416.

^[51] Ibid [176] (Nicholson CJ and Fogarty J, with Finn J substantially agreeing).

^[52] Ibid [161].

^[53] See Family Law (Child Abduction Convention) Regulations 1986 (Cth) reg 16(3)(c).

^[54] [1988] FamCA 21; (1988) 12 Fam LR 755. See also Laing v Central Authority [1996] FLC 92-709; Director-General Department of Families, Youth and Community Care v Hobbs [1999] FamCA 2059 (24 September 1999).

^[55] [1988] FamCA 21; (1988) 12 Fam LR 755, 767.

^[56] Ibid.

^[57] Ibid 768 (Fogarty, Frederico and Joske JJ).

^[58] Ibid.

^[59] (2001) 206 CLR 401.

^[60] Ibid 449.

^[61] (2001) 206 CLR 401.

^[62] Ibid 417.

^[63] P v Commonwealth Central Authority [2000] FamCA 461 (19 May 2000). The initial ‘P’ was the pseudonym given to the applicant in this case as reported by the Full Court of the Family Court. This case reached the High Court under the name DP v Commonwealth Central Authority (2001) 206 CLR 401.

^[64] Director-General, NSW Department of Community Services v JLM [2001] FamCA 1338; (2001) 28 Fam LR 243.

^[65] Ibid 418 (Gaudron, Gummow and Hayne JJ).

^[66] Director-General, NSW Department of Community Services v JLM [2001] FamCA 1338; (2001) 28 Fam LR 243; P v Commonwealth Central Authority [2000] FamCA 461 (19 May 2000).

^[67] DP v Commonwealth Central Authority (2001) 206 CLR 401, 423.

^[68] Ibid 417.

^[69] Ibid 408 (Gleeson CJ).

^[70] Ibid.

^[71] Ibid 424–6.

^[72] Ibid 426.

^[73] Ibid.

^[74] Ibid.

^[75] Ibid 427.

^[76] Ibid 420–1 .

^[77] Ibid 420.

^[78] Ibid.

^[79] Ibid 421 (Gaudron, Gummow and Hayne JJ).

^[80] Ibid 425.

^[81] Ibid.

^[82] Ibid 417–18, 426–7, 453, 456.

^[83] Ibid.

^[84] [2003] FamCA 1128 (17 September 2003).

^[85] The left-behind applicant father did not appeal this decision to the Full Court of the Family Court.

^[86] [2003] FamCA 1128 [114] (17 September 2003).

^[87] Ibid [111]–[112].

^[88] [2008] FamCA 1120 (22 December 2008). See also State Central Authority v Sigouras [2007] FamCA 250 (23 March 2007); Sally Nicholes, ‘Family Court Spreads Cloak of Protection – Case Note; State Central Authority and Papastavrou’ (2009) 83(8) Law Institute Journal 32.

^[89] [2008] FamCA 1120 (22 December 2008) [59], [125].

^[90] Ibid [125].

^[91] Ibid [130].

^[92] Ibid.

^[93] Ibid [77].

^[94] Ibid [106]. The mother presented expert evidence explaining the inadequacies of Greece domestic violence protections: at [87]–[89].

^[95] Ibid [106].

^[96] In Greece the act of international parental child abduction is a criminal offence.

^[97] [2008] FamCA 1120 (22 December 2008) [106].

^[98] [2006] FamCA 466 (6 July 2006).

^[99] Ibid [45] (Kay, Coleman and Warnick JJ).

^[100] Ibid [57].

^[101] Ibid 418.

^[102] [2011] FamCA 485 (23 June 2011).

^[103] See RCB as Litigation Guardian of EKV, CEV, CIV and LRV v Hon Justice Colin James Forrest [2012] HCA 47; (2012) 247 CLR 304, where proceedings were brought on behalf of the children by their maternal aunt as litigation guardian on grounds of due process, but importantly for the purposes of this article the mother did not appeal the Family Court’s interpretation of the exceptions to return.

^[104] Garning [2011] FamCA 485 (23 June 2011) [8].

^[105] Ibid [83].

^[106] Ibid [96] (Forrest J).

^[107] [2011] FamCA 485 (23 June 2011).

^[108] For post-Garning cases where the Family Court continued to assess the degree for risk see, eg, Department of Family and Community Services and Viduka [2015] FamCA 640 (4 August 2015); State Central Authority v Castillo [2015] FamCA 792 (10 September 2015).

^[109] Garning [2011] FamCA 485 (23 June 2011) [89].

^[110] Ibid [87].

^[111] Ibid [84].

^[112] Ibid [95] (Forrest J).

^[113] Ibid [100].

^[114] [2011] FamCA 485 (23 June 2011).

^[115] Ibid [101] (Forrest J) (emphasis added).

^[116] Pérez-Vera, above n 5, 426, 429–30.

^[117] Ibid 432.

^[118] Ibid 431.

^[119] Rhonda Schuz, ‘Policy considerations in determining the habitual residence of a child and the relevance of context’ (2001) 11(1) Journal of Transnational Law and Policy 101, 104.

^[∗] Senior Lecturer, ANU College of Law, Australian National University. The author would like to thank Heather Roberts for her comments and advice in the preparation of this paper.

^[120] (2015) 254 CLR 425 (‘Cassegrain’). The case has already drawn some useful commentary. See Penny Carruthers and Natalie Skead, ‘Confirming Torrens orthodoxy: The High Court decision in Cassegrain v Gerard Cassegrain & Co Pty Ltd’ (2015) 24 Australian Property Law Journal 211.

^[121] The agency issue primarily related to the first transfer in which both the husband and wife received the property as joint tenants. Thereafter, by way of a second transfer the husband transferred his share to his wife for nominal consideration. The High Court held that the interest that the wife received from her husband could be recovered on the basis that she was not a bona fide purchaser for value. See Cassegrain (2015) 254 CLR 425, 445 (French CJ, Hayne, Bell and Gageler JJ), 458 (Keane J).

^[122] See Cassegrain (2015) 254 CLR 425, 439, 445 (French CJ, Hayne, Bell and Gageler JJ). Notably, at 445 the majority states, ‘Claude was not her "agent" in any relevant sense.’ By implication this suggests that some other less relevant form of agency might have existed. However, see the discussion below nn 56–63.

^[123] FMB Reynolds, Bowstead and Reynolds on Agency (20^th ed, 2014) 8-062. See also Ruben v Great Fingall Consolidated [1906] UKLawRpAC 29; [1906] AC 439; Morris v CW Martin & Sons Ltd [1966] 1 QB 716.

^[124] [1969] 2 NSWLR 576 (‘Schultz’).

^[125] Cassegrain (2015) 254 CLR 425, 439 (French CJ, Hayne, Bell and Gageler JJ), 452 (Keane J).

^[126] See Carruthers and Skeed, above n 1. The authors briefly acknowledge some of the differences between the approaches of Street J in Schultz and the Supreme Court of New Zealand in Dollars & Sense Finance Ltd v Rerekohu Nathan [2008] NZSC 20; [2008] 2 NZLR 557 (‘Dollars & Sense Finance’). This article takes the argument much further and sets out three detailed criticisms of the way Schultz engages with the task of constructing the scope of an agency. Carruthers and Skeed adopt the terminology of ‘substantive agency’ and ‘procedural agency’ to deal respectively with the questions of whether an agency exists, and, if so, what the scope of that agency may be. This terminology has no basis in the law of agency and will not be adopted in this paper.

^[127] PS Atiyah, Vicarious Liability in the Law of Torts (Butterworths, 1967).

^[128] Some doubt has already been raised about Schultz in light of the equitable principle that a principal should not benefit from the fraud of an agent. See further, Davis v Williams [2003] NSWCA 371 (16 December 2003) [38] (Hodgson JA) (‘Williams’). See below nn 133–4.

^[129] International Harvester Co of Australia Pty Ltd v Carrigan’s Hazeldene Pastoral Company [1958] HCA 16; (1958) 100 CLR 644 (‘International Harvester Co’).

^[130] Ibid 652.

^[131] Branwhite v Worcester Works Finance Ltd [1969] 1 AC 552 (‘Branwhite’).

^[132] Ibid 587. See also South Sydney District Rugby League Football Club v News Limited [2000] FCA 1541; (2000) 177 ALR 611 (‘South Sydney District Rugby League Football Club’).

^[133] South Sydney District Rugby League Football Club [2000] FCA 1541; (2000) 177 ALR 611, [133]–[134].

^[134] Technology Leasing Ltd v Lennmar Pty Ltd [2012] FCA 709 (6 July 2012).

^[135] Petersen v Moloney [1951] HCA 57; (1951) 84 CLR 91.

^[136] Skandinaviska Enskilda Banken AB (Publ), Singapore Branch v Asia Pacific Breweries (Singapore) Pte Ltd and another and another appeal [2011] 3 SLR 540 (‘Skandinaviska’).

^[137] Scott v Davis (2000) 204 CLR 333 (‘Scott’); Colonial Mutual Life Assurance Society Ltd v Producers and Citizens Co-operative Assurance Co of Australia Ltd [1931] HCA 53; (1931) 46 CLR 41 (‘Colonial Mutual Life Assurance Society’).

^[138] Cheng-Han Tan, ‘Unauthorised agency in English law’ in Danny Busch and Laura J. Macgregor (eds), The Unauthorised Agent: Perspectives from European and Comparative Law, (Cambridge University Press, 2009) 187.

^[139] Freeman and Lockyer v Buckhurst Park Properties (Mangal) Ltd [1964] 2 QB 480; Egyptian International Foreign Trade Co v Soplex Wholesale Supplies Ltd, and P S Refson & Co Ltd [1985] 2 Lloyd’s Rep 36 (‘The Raffaella’); First Energy (UK) Ltd v Hungarian International Bank [1993] 2 Lloyd’s Rep 194; Pacific Carriers v BNP Paribas [2004] HCA 35; (2004) 218 CLR 451.

^[140] Ostensible authority is sometimes referred to as apparent authority or agency by estoppel. See further Hely-Hutchison v Brayhead Ltd [1968] 1 QB 549; Crabtree-Vickers Pty Ltd v Australian Direct Mail Advertising & Addressing Co Pty Ltd [1975] HCA 49; (1975) 133 CLR 72.

^[141] See Skandinaviska [2011] 3 SLR 540. See also Cavenagh Investment Pte Ltd v Kaushik Rajiv [2013] SGHC 45.

^[142] [2000] QCA 152 (5 May 2000).

^[143] Lloyd v Grace, Smith & Co [1912] AC 716 (‘Lloyd’); Clancy v Prince [2001] NSWSC 85 (26 February 2001) [61].

^[144] There is an argument that Claude could be seen as an agent for both GCC and his wife. However, he does appear to have acted on his own behalf and Felicity’s in the transaction, and not as an agent of GCC. Even were he to be regarded as an agent for GCC in the transaction, there is an exception within agency law to the rule that the knowledge of an agent is imputed to the principal to the effect that the rule does not apply where the agent is seeking to defraud the principal. See Re Hampshire Land Co [1896] UKLawRpCh 122; [1896] 2 Ch 743. See also Reynolds above n 4, 8–213. This particular exception has been applied for over 150 years. See Kennedy v Green [1834] EngR 1072; (1834) 3 My & K 699; Re European Bank (1870) 5 [1870] UKLawRpCh 27; LR 5 Ch App 358; Re Hampshire Land Co Ltd [1896] UKLawRpCh 122; [1896] 2 Ch 743; JC Houghton & Co v Nothard, Lowe and Wills Ltd [1928] AC 1, 15 (Viscount Dunedin); Belmont Finance Corp Ltd v Williams Furniture Ltd (No 1) [1979] 1 Ch 250, 260–1 (Buckley LJ); Cricklewood Holdings Ltd v CV Quigley & Sons Nominees Ltd [1992] 1 NZLR 463, 482 (Holland J); Duncan v McDonald [1997] NZCA 318; [1997] 3 NZLR 669, 679 (Blanchard J); Beach Petroleum NL v Kennedy [1999] NSWCA 408; (1999) 48 NSWLR 1, 99 (Court of Appeal). However, in recent times the application of the exception has been doubted. See Nathan v Dollars & Sense Finance Ltd [2007] NZCA 177; [2007] 2 NZLR 747, [99] (Glazebrook and Robertson JJ); Tonto Home Loans Australia Pty Ltd v Tavares [2011] NSWCA 389 (21 December 2011) [208]–[209] (Allsop P). See further Peter Watts, ‘Imputed Knowledge in Agency Law: Excising the Fraud Exception’ (2001) 117 Law Quarterly Review 300.

^[145] Cassegrain v Cassegrain (Unreported, Federal Court of Australia, Davies J, 15 July 1998).

^[146] Cassegrain (2015) 254 CLR 425, 453. Keane J did note that if the dairy farm had initially been transferred solely to Felicity then she would have received an indefeasible title. However, this would have deprived Claude from ever having enjoyed a legal estate in the property.

^[147] In the High Court in Cassegrain, Keane J dissented from the majority decision on the interpretation of s 118(1)(d)(i). In particular, the phrase “registered through fraud” which appears in s 118(1)(d)(i) has been interpreted to refer to fraud by the registered proprietor. See for example, Register of Titles (WA) v Franzon [1975] HCA 41; (1975) 132 CLR 611, 618 (Mason J). Such an interpretation of s 118(1)(d)(i) is consistent with the protection offered under s 42(1) to registered proprietors who achieve registration without fraud. However, in Cassegrain, Keane J at [106]–[116] pointed out that the statutory language of s 118(1)(d)(i) could regard the statutory title of a joint tenant as defeasible even though he or she was innocent of the fraud perpetrated by the other joint tenant. The statutory interpretation issue that arose in Cassegrain is an important matter. However, it is beyond the scope of the article.

^[148] Frazer v Walker [1967] AC 569, 580–1 (Lord Wilberforce) (‘Frazer’); Breskvar v Wall (1971) 126 CLR 376, 385 (Barwick CJ) (‘Breskvar’).

^[149] See Bahr v Nicolay (No 2) [1988] HCA 16; (1988) 164 CLR 604 (‘Bahr’) where assurances that were given prior to registration were later repudiated. Mason CJ and Dawson J were willing to regard these actions, which in their view amounted to equitable fraud, as fraud within the meaning of s 42.

^[150] Mayer v Coe [1968] 2 NSWR 747, 754 (Street J).

^[151] [1923] HCA 17; (1923) 32 CLR 309.

^[152] Ibid 329 (Knox CJ).

^[153] (2007) 230 CLR 89.

^[154] Ibid [192] (Gleeson CJ, Gummow, Callinan, Heydon and Crennan JJ).

^[155] Frazer [1967] AC 569; Breskvar (1971) 126 CLR 376; Assets Co Ltd v Mere Roihi [1905] UKLawRpAC 11; [1905] AC 176 (‘Assets Co Ltd’).

^[156] Schultz [1969] 2 NSWLR 576, 582–3. See also Reynolds, above n 4.

^[157] [1905] UKLawRpAC 11; (1905) AC 176, 210 (emphasis added).

^[158] Even then Street J in Schultz suggested that something more was required. See below n 73.

^[159] Frazer [1967] AC 569, 580–1 (Lord Wilberforce) (‘Frazer’); Breskvar (1971) 126 CLR 376, 385 (Barwick CJ).

^[160] Cassegrain (2015) 254 CLR 425, 438. Nonetheless, the law of agency would likely apply here as well. There is a degree of overlap between the general rules on attribution of acts to a corporation and the law of agency. See Colonial Mutual Life Assurance Society [1931] HCA 53; (1931) 46 CLR 41.

^[161] Frazer [1967] 1 AC 569, 580 (Lord Wilberforce). See also Schultz. In each of these cases Lord Lindley’s statement was interpreted as applying directly to the law of agency.

^[162] See Branwhite [1969] 1 AC 552, 587 (Lord Wilberforce); Cadd v Cadd [1909] HCA 59; (1909) 9 CLR 171. See also Equiticorp Finance Ltd (in liq) v Bank of New Zealand (1993) 32 NSWLR 50 (‘Equiticorp’).

^[163] (1993) 32 NSWLR 50, 132.

^[164] Pollard v Wilson [2010] NSWCA 68 (8 April 2010) [113] (McClellan CJ).

^[165] International Harvester Co [1958] HCA 16; (1958) 100 CLR 644. Of course, the agent must also consent to the agency relationship.

^[166] [1774] EngR 2; (1774) 98 ER 969 (‘Blatch’).

^[167] Gerard Cassegrain & Co Pty Ltd v Cassegrain [2013] NSWCA 453 (18 December 2013) [26]–[31] (Beazley P) (‘Gerard Cassegrain & Co Pty Ltd’). However, it should be noted that her Honour did not wholly rely on Blatch v Archer, but rather drew reasonable inferences from the known facts. See also Gerard Cassegrain & Co Pty Ltd [2013] NSWCA 453 (18 December 2013) [155] (Macfarlan JA).

^[168] Gerard Cassegrain & Co Pty Ltd v Cassegrain [2013] NSWCA 453 (18 December 2013) [31].

^[169] Gerard Cassegrain & Co Pty Ltd [2013] NSWCA 453 (18 December 2013) [155].

^[170] Cassegrain (2015) 254 CLR 425, 438 (French CJ, Hayne, Bell and Gageler JJ).

^[171] Ibid 439.

^[172] Ibid.

^[173] Ibid 445.

^[174] Ibid 439.

^[175] Ibid.

^[176] However, the making of a gift is not necessarily inconsistent with the existence of an agency. It is also possible to regard Claude as the donor and Felicity as the donee. Other than acknowledging Felicity as a ‘passive recipient’ the majority fails to ascribe any given legal category to Claude and Felicity’s roles in the transfer.

^[177] Keighley, Maxsted & Co v Durant [1901] UKLawRpAC 21; [1901] AC 240. The absence of dissent to the act of the first registration is also a relevant consideration in assessing ratification. See Lamshed v Lamshed [1963] HCA 60; (1963) 109 CLR 440, 448 (Kitto J).

^[178] Ibid.

^[179] Davison v Vickery’s Motors Ltd (in liq) [1925] HCA 47; (1925) 37 CLR 1, 19 (Isaacs J).

^[180] Cox v Isles, Love & Co [1910] St R Qd 80; Australian Blue Metal Ltd v Hughes [1962] NSWR 904, 925; McLaughlin v City Bank of Sydney [1912] HCA 16; (1912) 14 CLR 684.

^[181] Victorian Professional Group Management Pty Ltd v Proprietors ‘Surfers Aquarius’ Building Units Plan No 3881 [1991] 1 Qd R 487; Sinclair v Hudson (1995) 9 BPR 16, 259. Though there is an argument that knowledge of the registration is sufficient to dispense with the knowledge requirement even though Felicity may have been unaware of the fraud. This interpretation of the rules on ratification would be consistent with the Mair principle: see below n 63.

^[182] Mair v Rio Grande Rubber Estates Ltd [1913] UKLawRpAC 32; [1913] AC 853, 870 (Lord Shaw) (‘Mair’); Refuge Assurance Co. Ltd v Kettlewell [1909] UKLawRpAC 9; [1909] AC 243; Williams [2003] NSWCA 371 (16 December 2003) [128]–[129] (Young CJ). See below nn 132–5. In Williams, Young CJ stated, ‘[w]here a person receives a benefit from the fraud of another, that person is not permitted to deny the agency.’

^[183] Taylor v Smith [1926] HCA 16; (1926) 38 CLR 48, 54–5 (Knox CJ); Fried v National Australia Bank Ltd [2001] FCA 907; (2001) 111 FCR 322, 361; Porteous v Donnelly [2002] FCA 862 (8 July 2002).

^[184] Gerard Cassegrain & Co Pty Ltd [2013] NSWCA 453 (18 December 2013) [29].

^[185] Cassegrain (2015) 254 CLR 425, 439 (French CJ, Hayne, Bell and Gageler JJ).

^[186] [1969] 2 NSWR 576, 582–3 (emphasis added).

^[187] It seems also to have passed without comment in Schultz, that Galea was agent for both the plaintiff and the defendant.

^[188] Schultz [1969] 2 NSWLR 576, 583 (emphasis added).

^[189] Frazer [1967] AC 569; Breskvar (1971) 126 CLR 376; Assets Co Ltd [1905] UKLawRpAC 11; [1905] AC 176.

^[190] See below nn 132–3.

^[191] For example, the rule that where by a wrongful act the money of a third party is acquired by the agent for the benefit of the principal it is inequitable for the principal to retain such moneys. See Reynolds, above n 4, 8–201. See Campden Hill Ltd v Chakrani [2005] EWHC 911.

^[192] See above n 68.

^[193] See for example Lloyd [1912] AC 716.

^[194] Schultz [1969] 2 NSWLR 576, 582–3

^[195] Pamela O’Connor, Immediate Indefeasibility for Mortgagees: A Moral Hazard?’ (2009) 21(2) Bond Law Review 133, 142–3.

^[196] Schultz [1969] 2 NSWR 576, 583 (emphasis added).

^[197] Ibid.

^[198] [2010] NSWSC 108 (26 February 2010).

^[199] Ibid [30] (Rothman J).

^[200] Dollars & Sense Finance [2008] NZSC 20; [2008] 2 NZLR 557, [31].

^[201] Ibid.

^[202] Ibid. See also Carruthers and Skeed, above n 1, 218–219. The authors suggest that the New Zealand Supreme Court relied upon the principle of respondeat superior to determine the liability of the principal. However, the judgment of the Court makes no mention of this principle.

^[203] Dollars & Sense Finance [2008] NZSC 20; [2008] 2 NZLR 557, [46].

^[204] Ibid.

^[205] Atiyah, above n 8, 178; ibid [32].

^[206] Dollars & Sense Finance [2008] NZSC 20; [2008] 2 NZLR 557, [34].

^[207] Ibid.

^[208] Ibid [35].

^[209] [2002] 1 AC 215.

^[210] [2003] 2 AC 366.

^[211] Ibid [124].

^[212] Dollars & Sense Finance [2008] NZSC 20; [2008] 2 NZLR 557, [37].

^[213] Ibid [42].

^[214] Ibid [48].

^[215] [2012] NZSC 72; [2013] 1 NZLR 741.

^[216] Hollis v Vabu Pty Ltd [2001] HCA 44; (2001) 207 CLR 21, [99] (McHugh J) (‘Hollis’). See the discussion below nn 114–18.

^[217] [2015] UKSC 23.

^[218] Ibid [70]. His Honour was not joined by the other members of the Court on this point.

^[219] Dollars & Sense Finance [2008] NZSC 20; [2008] 2 NZLR 557, [31].

^[220] Gino Dal Pont, Law of Agency, (2^nd ed), (LexisNexis Butterworths, 2008) 595.

^[221] [1872] UKLawRpCP 52; (1872) LR 7 CP 415.

^[222] Ibid 420.

^[223] Dal Pont, above n 101, 599.

^[224] Lloyd [1912] AC 716.

^[225] Ibid 738 (Lord Macnaughten), 725 (Earl Loreburn).

^[226] (2000) 204 CLR 333, [49]. Scott concerned a claim for vicarious liability in tort on the basis of an agency argument. In Scott, the alleged principal owned several light planes which he used for his own enjoyment. At a party that he organised, he allowed another pilot to use one of his planes for a joyride involving some passengers. The plane crashed killing the pilot and severely injuring a young boy. The plaintiffs alleged that the pilot was an agent of the plane’s owner. In the absence of any employment relationship or contract, all of the judges of the High Court, apart from McHugh J, declined to find any agency relationship. See also Sweeney v Boylan Nominees Pty Ltd [2006] HCA 19; (2006) 226 CLR 161. See Dal Pont, above n 101. Dal Pont’s criticisms of the way in which McHugh J in Scott and Kirby J in Sweeney handle agency law are worth noting. Dal Pont at 597 suggests that their Honours deal with agency law in a ‘cursory’ manner, ‘ostensibly driven more by securing a particular outcome than necessarily grounded in principle.’ In other words, agency law is used as a means to an end, but receives too little attention to its own internal rules. This problem appears to persist in Cassegrain.

^[227] [1878] UKLawRpExch 25; (1878) 3 Ex D 238.

^[228] Ibid 245.

^[229] Hollis [2001] HCA 44; (2001) 207 CLR 21, [99]. See also Dal Pont, above n 101, 597.

^[230] However, this idea is not directly addressed in the jurisprudence.

^[231] This is a question of contribution to the loss rather than connection to authorised tasks.

^[232] It is worth observing that at least in the initial stages of the sales transaction Claude was an agent for both GCC and Felicity.

^[233] This in of itself would not have rendered Felicity’s title defeasible with respect of the first title. With regard to the second title, it might have raised a question of ratification. This was raised before the NSW Court of Appeal by Felicity herself, but curiously GCC did not press this point. Felicity argued that the concept of ratification was not available with regard to indefeasibility. There is nothing to suggest that the operation of the Torrens system precludes the operation of the doctrine of ratification.

^[234] [1999] NSWSC 1165 (1 December 1999).

^[235] Assets Co Ltd [1905] UKLawRpAC 11; [1905] AC 176, 210.

^[236] Grgic v Australian and New Zealand Banking Group Ltd (1994) 33 NSWLR 202.

^[237] Young v Hoger [2000] QSC 455 (6 December 2000).

^[238] [1891] UKLawRpAC 2; [1891] AC 248. Also Clements v Ellis [1934] HCA 18; (1934) 51 CLR 217.

^[239] Under deferred indefeasibility the title of a registered proprietor who obtains their title as a consequence of a fraud is not entitled to indefeasibility even though they themselves are completely innocent of any involvement in the fraud. In this situation the indefeasibility is deferred until they pass on good title to another buyer. See Gibbs v Messer [1891] UKLawRpAC 2; [1891] AC 248.

^[240] While this is understandable in light of the High Court’s findings on agency it is undesirable.

^[241] Assets Co Ltd [1905] UKLawRpAC 11; (1905) AC 176, 210.

^[242] Though, the principle is older than the Mair case discussed above n 63. See Dixon v Olmius (1787) 1 Cox 414; Huguenin v Baseley (1807) 14 Ves 273, 290 (Lord Eldon LC).

^[243] See Homeward Bound Gold Mining Co NL v McPherson (1895) LR (NSW) Eq 281, 319 (Owen CJ).

^[244] Mair [1913] UKLawRpAC 32; [1913] AC 853, 872–3 (Lord Moulton). See also Welch v Handcock (1907) SR (NSW) 404.

^[245] Williams [2003] NSWCA 371 (16 December 2003) [39].

^[246] Ibid [40].

^[247] However, in my view the relevant act of ratification is not the pleading of indefeasibility at trial, but rather the retention of title after the first registration.

^[248] Dal Pont, above n 101, 633. In this passage Dal Pont refers to a well-established exception to the presumption that the agent’s knowledge is to be imputed to the principal. This principle operates on the basis that it is unfair to impute knowledge of a fraud to a party when they are one of the victims of that fraud.

^[249] American Law Institute, Restatement (Third) of Agency (2006) § 5.04.

^[250] Mercantile Mutual Life Assurance Co v Gosper (1991) 25 NSWLR 32.

^[∗] BA (Hons) (UWA), LLB (Hons) (UWA), LLM (NYU), LLM (NUS). Law Lecturer,

Murdoch University. The author would particularly like to acknowledge and thank the anonymous referees for their constructive comments and feedback concerning an earlier draft of this article.

^[251] Commonwealth, Parliamentary Debates, Senate, 7 February 1995, 610 (Bob McMullan).

^[252] Phil Larkin, ‘Ministerial Accountability to Parliament’ in Keith Dowding and Chris Lewis (eds), Ministerial Careers and Accountability in the Australian Commonwealth Government (ANU E Press, 2012) 102.

^[253] Sir W. Ivor Jennings, The Law and the Constitution (University of London Press, 1959).

^[254] Ian Killey, Constitutional Conventions in Australia: An Introduction to the Unwritten Rules of Australia’s Constitutions (Australian Scholarly Publishing, 2009) 4.

^[255] For a full discussion on each of these see Killey, above n 4.

^[256] Ibid; Brian Galligan and Scott Brenton (eds), Constitutional Conventions in Westminster Systems: Controversies, Changes and Challenges (Cambridge University Press, 2015). It is important to note that these are not the only elements that have been identified as relevant, although they are uniformly identified as key considerations when identifying constitutional conventions. For example, some commentators have suggested that the existence of a sound constitutional rationale underpinning the constitutional convention is an essential element of its existence, although there is debate as to exactly how this consideration should be approached in practice. Compare, for example, Richard McGarvie, Democracy, Choosing Australia’s Republic (Melbourne University Press, 1999), 54–5 and Michael Coper, Australian Constitutional Convention Standing Committee D, The Senate and Supply: Special Report to Working Committee (Australian Government Printer, 1977) 103, 110.

^[257] Andrew Heard, Canadian Constitutional Conventions, The Marriage of Law and Politics (Oxford University Press, 2^nd ed, 2014) 12.

^[258] Commonwealth, Parliamentary Debates, Senate, 7 February 1995, 610 (Bob McMullan).

^[259] See Commonwealth, Parliamentary Debates, Senate, 7 February 1995, 612 (Cheryl Kernot).

^[260] See, eg, Australia, Parliamentary Debates, Senate, 7 February 1995, 610 (Bob McMullan).

^[261] See, eg, Andrew Alexandra & Clare McArdle, ‘Accountability and Ministerial Advisors’ (2003) 5(2) Australian Journal of Professional and Applied Ethics 71, 76.

^[262] This potential problem was raised and discussed in Anne Twomey, ‘Executive Accountability to the Australian Senate and the New South Wales Legislative Council’ (Legal Studies Research Paper No 07/70, Sydney Law School, November 2007) 30.

^[263] Lucy Hare, ‘Ministers’ Personal Appointees: Part Politician, Part Bureaucrat’ (2004) 2 New Zealand Journal of Public and International Law 315, 316.

^[264] Public Service Act 1999 (Cth) s 10.

^[265] See, eg, Department of the Prime Minister and Cabinet, Government Guidelines for Official Witnesses Before Parliamentary Committees and Related Matters (2015).

^[266] Department of the Prime Minister and Cabinet, A Guide on Key Elements of Ministerial Responsibility (1998) ch 6.

^[267] Killey, above n 4, 127.

^[268] See, eg, Yee-Fui Ng, ‘The Nexus Between Law and Politics’ (Paper presented at Legalwise CPD Seminar, 17 June 2015); Greg Tayor, ‘Book Review – Constitutional Conventions in Australia: An Introduction to the Unwritten Rules of Australia’s Constitution’ [2010] AdelLawRw 14; (2010) 31 Adelaide Law Review 271.

^[269] Ng, above n 18.

^[270] Andrew Heard, Canadian Constitutional Conventions: The Marriage of Law and Politics (Oxford University Press, 1st ed, 1991) 12, quoted in Killey, above n 4, 8.

^[271] Ng, above n 18.

^[272] Ibid.

^[273] Ian Holland, ‘Accountability of Ministerial Staff?’ (Research Paper No. 19, Department of the Parliamentary Library, 2001–02) 14.

^[274] Ibid.

^[275] Harry Evans and Rosemary Laing (eds), Odgers’ Australian Senate Practice (Parliament of Australia, 13^th ed, 2012) ch 17.

^[276] Ibid.

^[277] The head of the National Media Liaison Service being a position that was employed under the Members of Parliament (Staff) Act 1984 (Cth) and was effectively therefore a ministerial advisor position.

^[278] Holland, above n 23, 14.

^[279] See above at Part III.

^[280] Although it is worth noting in relation to this specific example that the Senate itself expressly attempted to distinguish this case from that of other ministerial advisors by noting that the stated role of the National Media Liaison Service was to provide ‘accurate and timely information on government policies’. As such, it was not meant to be party political and, given its unique role, could be distinguished from the usual circumstances in which the McMullan Principle might otherwise be invoked. See Holland, above n 23, 16.

^[281] The Terms of Reference required the Senate Select Committee to look into the ‘Children Overboard’ incident and a range of issues directly associated with that incident; operational procedures to ensure the safety of asylum seekers on vessels attempting to enter Australian waters; and issues relating to agreements between Australia, Nauru and Papua New Guinea relating to the Pacific Solution.

^[282] Senate Select Committee for an Inquiry into a Certain Maritime Incident, Parliament of Australia, Main Report (23 October 2002) [Terms of Reference (b)(ii)].

^[283] Ibid [7.124].

^[284] See Ng, above n 18.

^[285] Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 32, [7.126].

^[286] Ibid [7.149].

^[287] Ibid [7.138].

^[288] Ibid [7.140].

^[289] Ibid [7.139].

^[290] See Victorian Legislative Council Standing Committee on Finance and Public Administration, Parliament of Victoria, Inquiry into Victorian Government Decision Making, Consultation and Approval Processes (Second Interim Report – August 2010) [10]–[33].

^[291] Ibid.

^[292] Victoria, Parliamentary Debates, Legislative Assembly, 20 March 2002, 376 (Rodd Hulls). See also Holland, above n 23, 1.

^[293] Quoted in Killey, above n 4, 126.

^[294] Victorian Legislative Council Standing Committee on Finance and Public Administration, above n 40.

^[295] Ibid [36].

^[296] Ibid.

^[297] Ibid [62].

^[298] Ibid [57].

^[299] Ibid [46].

^[300] One interesting aside is that shortly after commencing his investigation the Ombudsman received legal advice from the then Solicitor-General arguing that he did not have the authority to investigate the actions of Ministers and that he had only limited jurisdiction to investigate ministerial advisors. The Ombudsman concluded that he did have jurisdiction. See Victorian Ombudsman, Ombudsman Investigation into the Probity of the Hotel Windsor Redevelopment, Ombudsman Investigation Report (2011) 8.

^[301] Ibid 9.

^[302] Ibid [61].

^[303] Ibid 19.

^[304] Ibid 14, 84.

^[305] Ibid 14, 84.

^[306] General Purpose Standing Committee No 4, NSW Legislative Council, The Designer Outlets Centre, Liverpool Report 11 (2004) [1.19].

^[307] Although it should be noted that he did offer to assist by answering questions on notice: see ibid [1.21].

^[308] Section 4 of the Parliamentary Evidence Act 1901 (NSW) provides that any person, other than a Member of Parliament, may be summoned to attend and give evidence by the Parliament or a Parliamentary Committee. Sections 7 and 8 go on to provide that if they fail to attend without just cause or reasonable excuse an arrest warrant may be issued.

^[309] General Purpose Standing Committee No 4, above n 56, [1.22]. See also Beverly Duffy and David Blunt, ‘Information is power: recent challenges for committees in the NSW Legislative Council’ (Paper presented at the 45^th Presiding Officers’ & Clerks’ Conference; Apia, Samoa, 30 June – 4 July 2014).

^[310] See General Purpose Standing Committee No 4, above n 56.

^[311] Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 32 [7.146].

^[312] Ibid [7.149].

^[313] Senate Committee of Privileges, Parliament of Australia, Parliamentary Privileges Amendment (Enforcement of Lawful Orders) Bill 1994 (49^th Report) (1994) 5.

^[314] Peter Hanks, ‘Parliamentary Privileges in Victoria: The power of a House of Parliament to demand production of documents and the giving of evidence’ (Paper presented at Legalwise CPD Seminar, 17 June 2015).

^[315] Tayor, above n 18. See also Killey, above n 4, 126.

^[316] Finance and Public Administration References Committee, Parliament of Australia, Independent Arbitration of Public Interest Immunity Claims (2010).

^[317] Office of the Public Sector Commissioner (Western Australia), ‘Policy for Public Sector Witnesses Appearing Before Parliamentary Committees’ (Public Sector Commissioner’s Circular 2010-03, 29 March 2010) <https://publicsector.wa.gov.au/sites/default/files/documents/2010-03_policy_for_public_sector_witnesses_appearing_before_parliamentary_committees_3.pdf>.

^[318] Holland, above n 23, 23.

^[319] Larkin, above n 2, 109.

^[320] Ibid.

^[321] Killey, above n 4, 124.

^[322] Alexandra and McArdle, above n 11, 74–5. For further discussion regarding the nature of this role see Malcolm Abbott and Bruce Cohen, ‘The Accountability of Ministerial Staff in Australia’ (2014) 49 Australian Journal of Political Science 316.

^[323] Noting, of course, the role that Departmental Liaison Officers also play in this regard, being public servants who are seconded to a ministerial office from their Department.

^[324] For a discussion of the ‘Children Overboard’ incident see part IV(A).

^[325] John Halligan et al, The Australian Public Service: The view from the top (Coopers & Lybrand, University of Canberra, 1996) 71.

^[326] See Alexandra and McArdle, above n 11.

^[327] United Kingdom Cabinet Office, Code of Conduct for Special Advisors (October 2015) Gov.uk [1]–[2] <https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/468340/CODE_OF_CONDUCT_FOR_SPECIAL_ADVISORS_-_15_OCTOBER_2015_FINAL.pdf >.

^[328] Alexandra and McArdle, above n 11,77.

^[329] Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 36, [7.117]. See also Larkin, above n 2, 109.

^[330] Alexandra and McArdle, above n 11, 77.

^[331] Australian Government, Code of Conduct for Ministerial Staff (2008).

^[332] Larkin, above n 2, 109–10.

^[333] Ibid 109.

^[334] Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 32, [7.107].

^[335] Ibid.

^[336] Victorian Ombudsman, above n 50, 84.

^[337] Alexandra and McArdle, above n 11, 76–7.

^[338] Larkin, above n 2, 109.

^[339] With the roles and function of the National Media Liaison Service stating that ‘representatives will not work with local or state party branches ... and will not be involved in party programs, nor electoral campaigning’. See Commonwealth, Parliamentary Debates, Senate, 7 February 1995, 607 (Rod Kemp).

^[340] Ibid.

^[341] Dr Stanley Bach 'Strengthening Australia’s Senate: Some Modest Proposals for Change' (Papers on Parliament No 50, Parliamentary Library, 2010).

^[342] As noted, for example, by the Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 32, [7.113]. The Select Committee noted that ‘[t]he number of staff working in ministers’ offices has at least doubled in thirty years’, although it was noted that this growth appeared ‘primarily to be the consequence of decisions about the machinery of government, and attempts to make government more ‘professional’, rather than the desire of governments to secure partisan advantage or gain further dominance over parliament’.

^[343] Commonwealth, Parliamentary Debates, House of Representatives, 28 November 2006, 178 (Bob McMullan).

^[344] Ibid.

^[345] Victorian Public Sector Standards Commissioner, Review of Victoria’s integrity and anti-corruption system, Review (2010).

^[346] Ibid.

^[347] Commonwealth of Australia, Standing Orders and other orders of the Senate (August 2015) Standing Order 177.

^[348] Evans and Laing (eds), above n 25, chs 2, 17. See also Abbott and Cohen, above n 84, 328; Anne Twomey, above n 21, 26; Ian Holland, ‘Reforming the conventions regarding parliamentary scrutiny of ministerial actions’ (2004) 63 Australian Journal of Public Administration 3.

^[349] Senate Select Committee for an Inquiry into a Certain Maritime Incident, above n 36, [7.175].

^[350] Legal opinion obtained by Mr Bret Walker SC, Select Committee on Gaming Licensing, Legislative Council, First Interim Report (2007) 47 quoted in Victorian Legislative Council Standing Committee on Finance and Public Administration, above n 40, [46].

^[351] Quoted in Victorian Legislative Council Standing Committee on Finance and Public Administration, above n 40, [44].

^[352] Australian Senate, Parliamentary Privilege Resolutions (1988) 1(16) <http://www.aph.gov.au/Parliamentary_Business/Chamber_documents/Senate_chamber_documents/standingorders> .

^[353] Alexandra and McArdle, above n 11, 80.