New Developments at the Intersection of Domestic Violence and the Hague Convention on International Child Abductions: a Historic Forum in South Africa

From 18 June 18 to 21 June 2024, the Permanent Bureau of the Hague Conference on Private International Law (Hague Conference – Conférence de La Haye, or HCCH) held the first-ever Forum on Domestic Violence and the Related Operation of Article 13(1)(b) of the Hague Convention on Civil Aspects of International Child Abduction (the “Forum on Domestic Violence”, or “the Forum”). Co-hosted by the South African government and the University of Pretoria’s Centre for Child Law in Sandton, South Africa, the Forum brought together more than 100 participants, including legal practitioners, judges, psychological experts, policymakers and advocates, along with ‒ perhaps most importantly ‒ survivors of domestic violence who shared their experiences living through a Hague Convention proceeding. This historic event focused on the intersection of domestic violence and Article 13(1)(b) of the Convention of 25 October 1980 on the Civil Aspects of International Child Abduction (the “1980 Hague Convention”), which allows a court to refuse the return of a child if such a return would expose the child to a “grave risk of physical or psychological harm”.

The significance of the Forum cannot be overstated. Domestic violence in the context of the 1980 Hague Convention presents complex challenges for the legal system. The original salutary goal of the 1980 Hague Convention – namely, to return children wrongfully removed from their home country – seemed completely unobjectionable. Yet, in practice, when children are moved cross-border to protect them from abuse, the wisdom of their return becomes questionable.

Indeed, Article 13(1)(b) cases now represent the majority of the 1980 Hague Convention cases that go to trial, and harm from domestic violence is invoked in the great majority of those. Although the drafters of the 1980 Hague Convention anticipated that the typical abductor would be a non-custodial parent disappointed by or fearing an adverse custody decision, practice has shown that more than 75% of the taking parents are mothers – the vast majority of whom (94%) are the primary carers of their children.

Domestic violence is inherently complex as compared to other forms of violence. Not only are its victims often reluctant to report their abusers, but cultural norms and societal attitudes play a crucial role in what counts as impermissible violence between family members in the first place, as well as how a country responds to it. While any modern society considers certain acts – such as forced sex or the corporal punishment of women or children –criminal in other contexts, it may deem them acceptable if they happen at home, between members of a family.

Thus, for a country to provide victims of domestic violence with effective protection, there must be a proper recognition of the problem not only at the national level – leading to the adoption of laws specifically tailored to address domestic violence – but also an implementation and proper enforcement of such laws at the local level, through the court system, state enforcement and social services agencies. Officers of these systems must all be trained to recognise and understand the problems and complexities of domestic violence, and be willing and equipped to help victims of this abuse.

However, the system for addressing domestic violence varies from country to country, often dramatically. This variation poses an especially difficult challenge for courts in international child abduction cases tasked with ordering measures designed to protect a child upon being returned to the child’s home country. The issuing court often lacks the necessary understanding of the legal system in the child’s country of habitual residence and may be powerless to enforce these measures once the parties leave its jurisdiction.

This article provides a brief overview of the 1980 Hague Convention and the Article 13(1)(b) defence in the context of domestic violence. It also highlights the main points of discussion and the takeaways from the Forum.

1980 Hague Convention and Article 13(1)(b) defence

The 1980 Hague Convention is a multilateral treaty to which the USA and more than 100 other countries are signatories. It is designed to protect children internationally from the harmful effects of their wrongful removal by establishing an expedited process for the courts or administrative agencies of the country to which the child is removed to return the child back to the child’s home country (“state of habitual residence”). The 1980 Hague Convention is not a mechanism for resolving custody disputes and, in that expedited proceeding, custody issues are not to be addressed. Indeed, the fundamental purpose of the 1980 Hague Convention is to ensure that ‒ by promptly returning the child ‒ custodial issues are decided by the child’s state of habitual residence, rather than by the country to which the child was abducted by a parent.

Each of the signatory member states of the 1980 Hague Convention has a Central Authority, which helps to locate abducted children, encourages resolutions of parental abduction cases, and processes requests for the return of children in what are known as both “incoming” and “outgoing” cases. In the USA, the Office of Children’s Issues within the Department of State serves as the Central Authority for the US government.

The 1980 Hague Convention requires that a child determined to have been “wrongfully removed or retained” be promptly returned. The taking or retention is considered “wrongful” if the petitioner proves by a preponderance of the evidence that:

• the child was removed from or retained outside the child’s country of habitual residence;
• the removal or retention was in breach of the petitioner’s custody rights; and
• those custody rights were actually exercised at the time of removal or retention or would have been exercised but for the removal or retention.

If this burden is met, the 1980 Hague Convention requires that the child be returned.

Despite the emphasis on the return, the 1980 Hague Convention also recognises that “the interests of children are of paramount importance”. As the Explanatory Report to the 1980 Hague Convention explained, “the interest of the child in not being removed from its habitual residence… gives way before the primary interest of any person in not being exposed to physical or psychological danger…”. Guided by this consideration, the 1980 Hague Convention allows a court of a country to which a child has been wrongfully removed to refuse the mandatory return under certain circumstances, including ‒ in the Convention’s Article 13(1)(b) ‒ where such return would expose the child to a “grave risk of physical or psychological harm or otherwise place the child in an intolerable situation”. This is known as the “Article 13(1)(b) defence” or the “grave risk of harm” defence.

Most cases invoking the “grave risk of harm” defence arise in the context of domestic abuse. In the USA, the International Child Abduction Remedies Act (ICARA) ‒ through which the 1980 Hague Convention is implemented ‒ adds an extra burden on the parent opposing the return, requiring that the grave risk of harm be proven by “clear and convincing evidence”. It is a high bar, which until recently could be met only in cases of severe physical abuse directed at the child. The USA is the only signatory to the 1980 Hague Convention to impose this heightened evidentiary standard.

Domestic violence in the USA: a complex legal landscape

Domestic violence is a complex, persistent and pervasive problem. The response to it across the country is complicated, in part, by the lack of a uniform definition and understanding of what constitutes domestic violence. Mental health professionals define domestic violence as a long-standing pattern of control and intimidation in the context of an intimate relationship. It is recognised that this dynamic in a relationship is created and maintained through multiple vehicles of control across many areas of the victim’s personal life, including physical, sexual, emotional and psychological abuse, medical neglect, financial and legal manipulation, social isolation, threats to a child of the relationship, and threats to deploy others in service of the abuser’s goals.

In contrast, there is no one accepted legal definition in the USA. Each state uses its own standards and procedures for addressing domestic violence. In many US states, the legal definition of domestic violence is narrowly focused on physical abuse and thus neglects other forms of control and intimidation, such as emotional, psychological or financial abuse. These forms of non-physical abuse are often harder to prove in court, yet they are just as damaging to victims, particularly when children are involved. The variability in state definitions and the often-narrow focus on physical violence contribute to the legal system’s inadequate response to the complexities of domestic violence.

California, for example, has one of the broadest definitions of domestic violence, which includes physical harm, the threat of harm, and behaviours such as harassment, stalking, and the destruction of personal property. The state’s definition also encompasses both current and former spouses, cohabitants, and individuals in dating relationships, as well as a child of a party. Similarly, in Florida, domestic violence is defined as any assault, battery, sexual assault, or other forms of physical harm between family or household members. Florida law also allows for protective orders based on the reasonable belief of imminent danger of becoming the victim of any act of domestic violence. This takes into account numerous factors, including prior attempts to harm and patterns of abusive, threatening, intimidating or controlling behavior.

Even though South Carolina’s definition of domestic violence is more limited, focusing ‒ like Florida ‒ primarily on physical harm or the threat of harm between household members, it does also consider a reasonable fear of imminent peril (although it does not explicitly include non-physical forms of abuse such as emotional or psychological coercion). Conversely, Alabama’s legal definition is more restrictive still, breaking domestic violence down into three statutory degrees. Alabama’s first two degrees of domestic violence are the most restrictive and extreme, primarily emphasising physical violence and specific criminal acts such as assault or sexual abuse, with less emphasis or recognition to emotional or physical violence. Its third degree of domestic violence recognises not only physical injury, but also menacing (or the threat of physical injury), and the victim can be connected to a defendant merely by dating. And New York does not have any explicit statutory definition of domestic violence at all ‒ although it is recognised through various family offences and real property matters.

The US legal system’s historical tendency to minimise or overlook allegations of domestic violence, especially when it involves non-physical forms of abuse, has further compounded these issues. Courts often require evidence of severe physical harm to the child, ignoring the broader spectrum of abuse that can have equally devastating effects on the victim and their child.

The lack of uniformity leads to a patchwork of protections that can vary dramatically depending on the jurisdiction. The inconsistency in the definition of domestic violence is even greater at an international level, where it is further complicated by the divergent legal, social and cultural norms and standards. This lack of common understanding poses significant challenges domestically and internationally in the context of the 1980 Hague Convention, with the outcome of a particular case often depending on which US state the case is litigated and which country constitutes the child’s “state of habitual residence”. In essence, the same victims subjected to the same abuse may or may not be protected depending in large part on the forum and the victims’ country of origin.

The legal community has also been slow to recognise the impact of domestic violence on children, especially in cases where the children are not direct targets of the violence but are exposed to it. This lack of recognition has contributed to the misguided reliance on protective measures in 1980 Hague Convention cases, even when domestic violence is a factor.

Moreover, until recently, even after finding a grave risk of harm, the US courts required that the parent opposing the return prove that there were no “protective measures” that would mitigate such risk. Protective measures, also frequently referred to as “ameliorative measures” in court decisions and scholarly articles, are usually in the form of voluntary undertakings by the left-behind parent. They typically consist of promises to provide financial assistance, stay away from the other parent and the child, and not co-operate in the criminal prosecution of the “abductor” parent. Such promises may be reduced to an order issued by the court deciding the return petition, with mirror orders in the country of the child’s habitual residence. The insistence on protective measures after finding a grave risk of harm underscores the emphasis US courts placed on granting the return petition. The Permanent Bureau’s Guide to Good Practice on Article 13(1)(b) also strongly recommends protective measures when considering returning children to their habitual residence.

Research shows, however, that abusers are prone to recidivism and are likely to ignore or defy interventions (such as court orders) intended to mitigate the recurrence of abuse. Many domestic violence offenders revert to their abusive behavior within months following law enforcement or social service interventions. Thus, reliance on protective measures to facilitate return in cases involving domestic violence despite a finding of grave risk reflects a fundamental lack of understanding of the complexity of domestic violence and the needs of its victims (the children and their caretaker parents), putting children in real danger.

Three studies of the effectiveness of protective measures provide ample reason to be hesitant to ever rely on such measures. By way of example, the Reunite International Child Abduction Centre’s study of cases in the UK revealed that two-thirds of the undertakings issued ‒ including all of those focused on a child’s safety upon return ‒ were not implemented in the country of habitual residence (see The Outcomes for Children Returned Following an Abduction (Reunite International Child Abduction Centre, UK, 2003)). Even when judges issued mirror orders in the child’s home country, only one in five of those mirror orders was implemented as planned.

Research into US incoming cases has also revealed that judges and attorneys were skeptical of the enforcement of these orders by another country’s courts and that mothers who returned with their children to the country of habitual residence would frequently face violations of previously agreed undertakings by their abusive ex-husbands or find that mirror orders were seldom enforced. Finally, in the recent online survey conducted by two UK charities, mothers from a number of countries reported that protective measures ‒ even those in which mirror orders were obtained ‒ were not enforced or were very difficult to enforce.

Fortunately, in the USA, there has been a significant shift away from reliance on protective measures. In 2022, the US Supreme Court recognised the complexity of domestic violence and the limitation of US courts to issue orders that would protect victims of domestic violence overseas. In the seminal case of Golan v Saada, 142 S Ct 1880 (2022) (“Golan”), the US Supreme Court rejected the appropriateness of protective measures in cases involving domestic violence: “A court may [] decline to consider imposing ameliorative measures where it is clear that they would not work because the risk is so grave. Sexual abuse of a child is one example of an intolerable situation. Other physical or psychological abuse, serious neglect, and domestic violence in the home may also constitute an obvious grave risk to the child’s safety that could not readily be ameliorated. A court may also decline to consider imposing ameliorative measures where it reasonably expects that they will not be followed.”

Main takeaways from the forum on domestic violence

Several main themes emerged in discussion at the Forum.

First, while there was no immediate agreement on whether victims of domestic violence and their children are already sufficiently protected under the 1980 Hague Convention, there was a uniform recognition that domestic violence harms and endangers children. Decades of research have clearly established that exposure to domestic violence, including adult-to-adult violence, has significant negative physical and psychological outcomes for children. These outcomes include depression, anxiety, sleep disturbances, lower social and emotional competence, poorer academic performance, and a higher tolerance for aggression and violence. The effects are particularly severe when the violence is directed at the child’s primary caregiver.

Second, the fragmented legal framework results in a child’s exposure to domestic violence being viewed differently depending on the location in which the case is heard. Some member states may prioritise the child’s well-being and recognise the harmful impact of witnessing domestic violence, whereas others may focus solely on direct physical harm to the child. This inconsistency can lead to vastly different outcomes, even when the facts of the case are similar. By way of example, if a protective measure is issued in one state with a broad definition of domestic violence, but the child is returned to a state or country with a narrower definition or weaker enforcement mechanisms, the protection may be rendered ineffective. This disconnect between different legal jurisdictions can leave victims and their children vulnerable to ongoing abuse.

This is a highly unsatisfactory outcome. There is no doubt that a uniform definition of domestic violence in the 1980 Hague Convention context (both nationally and internationally) ‒ a definition which is expansive enough to encompass all of the vehicles of abuse, including physical, psychological, emotional, financial and legal ‒ would be of great benefit to the implementation of the Article 13(1)(b) defence.

Third, the appropriateness of reliance on protective measures is another issue that was discussed in detail at the Forum. Although no consensus was reached, given the prevalence of the protective measures, it was an important conversation to begin. The international community should take note of the US Supreme Court’s decision in Golan and prioritise the safety of the children over their return. As Secretary General Christoph Bernasconi specifically acknowledged in his closing remarks, “it is not about the sheer number of returns but really about the correct application of the [1980 Hague] Convention”. Given the volume of evidence that protective measures are ineffective and do not adequately protect the victims, reliance on these measures is misplaced and should be discontinued.

The journey toward adequately protecting victims of domestic violence in international child abduction cases is far from complete. Importantly, the conversation will continue in 2025, when the Forum will convene again ‒ this time hosted by the Brazilian government. Ongoing education, legal reform, and societal changes are essential to ensuring that the legal system can effectively address these issues and protect those most vulnerable, while balancing important competing interests.