Kurt v Austria: A missed chance to tackle intersectional discrimination and gender-based stereotyping in domestic violence cases

By Lisa Maria Weinberger*

On 15 June 2021, the Grand Chamber of the European Court of Human Rights (ECtHR) rendered its judgment on the domestic violence case Kurt v Austria. This case concerned a woman in Austria who experienced domestic violence at the hands of her husband, which resulted in his murdering their son. Based on a highly divisive ten to seven Court decision, the majority of the Grand Chamber endorsed the preceding Chamber’s assessment and rejected the applicant’s claim of an Article 2 violation of the European Convention of Human Rights (ECHR) on account of the authorities’ failure to ensure the protection of her son’s life from his violent father. Nevertheless, important general principles regarding domestic violence were clarified.

In the first part, this contribution will present the majority’s decision and the newly established general principles, which constitute an extension of the Osman test for cases of domestic violence. The second part of the contribution explores different points of criticism regarding the judgment, with a particular focus on the majority’s lack of sensitivity towards the specificities of domestic violence when applying the general principles to the concrete case. It is argued that this resulted in the Court reinforcing gender-based stereotyping and victim-blaming. Furthermore, the missed opportunity to recognise the case as one of intersectional discrimination as it concerned a migrant woman will be discussed.

Facts and judgment

The case concerned the murder of an 8-year-old boy who was killed by his father on school premises in May 2012. While the younger sister was left unharmed, the father shot his son with a gun, which he acquired a few days earlier, before leaving the school and committing suicide (paras 35-39). This incident was preceded by the son’s mother, the applicant, requesting a barring order in 2010 (which was granted), the father’s criminal conviction for bodily harm and dangerous threatening behaviour against the applicant in 2011, the applicant filing for divorce and reporting to the police that her ex-partner had raped her and beaten both her and their children, and that he had voiced threats to kill her and their children in May 2012 (paras 12-34). Following this latest complaint to the police, another restraining order was issued and criminal proceedings against the father were initiated. In subsequent interviews, the children confirmed the beating by their father.

The applicant claimed that the Austrian authorities should have known that the father posed a real and immediate risk to their son’s life outside the restricted areas of the barring order, and therefore should have taken the father into pre-trial detention. Failing to do so constituted a breach of Article 2 ECHR. The applicant pointed to several risk factors which should have triggered preventive measures, such as her husband’s violent history, the serious death threats targeting the applicant as well as the children, the heightened violence in comparison to 2010 (including rape and beatings of the children) and the filed divorce, which the father declared that he would not accept.

While the ECtHR acknowledged the particular context of domestic violence as a form of gender-based violence, and recognised children who are victims of domestic violence as particularly vulnerable individuals, the Court’s majority nevertheless concluded that the authorities could not have known or ought not to have known that the applicant’s husband would kill their son. The majority held, therefore, that the Austrian authorities had ordered appropriate measures to prevent a risk of renewed violence against the applicant and her children.

New general principles based on the Osman test

In the present case, the Court introduced new general principles on the basis of the Osman test. The latter established the positive obligation of Member States to ensure effective criminal law provisions as well as to take appropriate preventive measures to protect an individual whose life is at risk from the criminal acts of another individual (Osman v United Kingdom (1998)). For a positive obligation to arise,

‘it must be established (…) that the authorities knew or ought to have known at the time of the existence of a real and immediate risk to the life of an identified individual or individuals from the criminal acts of a third party and that they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to avoid that risk.’

(Osman, para 116)

The ECtHR held in Kurt that the authorities must assess the reality and immediacy of any risk to life by taking due account of the particular context of domestic violence. If the outcome of the risk assessment is that there is a real and immediate risk to life, the authorities’ obligation to take preventive operational measures is triggered. Such measures must be adequate and proportionate to the level of the risk assessed (para 177). This is a refined reiteration of a principle which was already established in Talpis v Italy in 2017, where the Court added a key criterion to the Osman test, stating that:

‘the risk of real and imminent threat must be assessed taking due account of the particular context of domestic violence. In such a situation it is not only a question of an obligation to afford general protection to society […] but above all to take account of the recurrence of successive episodes of violence within the family unit.

(Talpis, para 122)

This approach was also endorsed in the ill-treatment case Volodina v Russia, a judgment which came only days after the Chamber judgement of Kurt was published in 2019.

In this sense, Kurt can be characterised as a concretisation of this principle. On a general level, the authorities must provide an immediate response to allegations of domestic violence and apply special diligence when dealing with such cases. More specifically, this means that Member States have the duty to conduct a comprehensive risk assessment of the potentially lethal risks faced by any victim of domestic violence. This lethal risk assessment must be autonomous, proactive and comprehensive (para 168). While a standardised risk assessment is not strictly required under the Osman test, the Court highlighted the usefulness of using research-based tools with pre-established questions that the authorities should systematically ask and answer (para 167). It also emphasised that the majority of Member States already uses standardised risk assessment tools (para 171). Additionally, the Court highlighted that the notion of ‘imminence’ means that ‘the common trajectory of escalation in domestic violence cases’ and other factors that adhere to the specific context of domestic violence must be considered (para 176).

Lastly, the Court held that authorities dealing with victims of domestic violence must receive regular training, particularly in respect of risk assessment tools, in order to understand the dynamics of domestic violence, thus enabling them to better assess and evaluate any existing risk, respond appropriately and ensure prompt protection (para 174).

These general principles were endorsed by all judges of the Grand Chamber. The divergence emerged in the application of these principles to the concrete case. The Court’s majority found that the authorities had duly considered the domestic violence context in their risk assessment (para 199). In particular, the majority noted that ‘police officers with significant relevant experience and training were involved in making this assessment, which the Court should be careful not to question in a facile manner with the benefit of hindsight’ (para 204).

Dissenting opinion

The dissenting judges were of a different view. They held that the risk assessment procedures in place were flawed in certain aspects and that the lethality risk assessment was inadequate in substance as ‘the authorities unjustifiably overemphasised certain factors and underemphasised others’ (joint dissenting opinion, para 7).

More specifically, the dissenting opinion disagreed with the majority’s view that there was no discernible risk to the children’s lives. The dissenting judges criticised the application of the danger assessment, which was only done in relation to the issuance of a barring order and thus focusing on the general risk of further violence of the alleged perpetrator, but not necessarily regarding the lethal risks to the family members (para 9). Furthermore, a standardised lethality risk assessment was not in place, which the dissenting opinion acknowledged – in line with all third-party interveners – as a serious problem. This led for the authorities to disregard any risks in relation to the applicant’s children and to the failure to identify significant and internationally recognised risk factors such as the father’s gambling addiction and mental health problems, his economic dependence on the applicant and the applicant’s job loss as well as his recent suicidal ideation in addition to his homicidal threats (para 11).

A case of gender stereotyping and victim-blaming

The majority endorsed the reasoning of the Austrian authorities, which, from a critical gender perspective, based their decision on several problematic factors. For example, the authorities put emphasis on the fact that any prior violence to the son’s murder happened exclusively in the marital home and that the father had previously been compliant with the restraining order. This reflects an overly simplified and unrealistic understanding of domestic violence, which is always ‘person-dependent, not place-dependent’, as the dissenting opinion rightly pointed out (para 25).

Furthermore, the authorities claimed that the reported offences of raping the applicant and beating her and their children, and the death threats in 2012 were of the same nature and gravity as the offences in 2010, thus concluding that a barring order should have been sufficient to protect the applicant’s and children’s lives. This dismisses the nature of escalating violence, a principle which the Court declared as essential in carrying out a risk assessment in cases of domestic violence. A risk assessment would have identified several incidents in Kurt as aggravating factors of domestic violence. For example, one of the commonly used tools by executive bodies worldwide is the Danger Assessment Instrument by Jacqueline Campbell. This tool includes the following risk factors associated with intimate partner homicide, which were all present in Kurt: past death threats against the victim or their children, emotional abuse, the partner’s employment status, separation or filing for divorce, rape, choking or strangling and the partner’s access to a gun.

Another decisive factor was that the applicant did not immediately report the rape and strangulation by her husband. The Chamber had put particular emphasis on this aspect in its judgment. While the authorities already downplayed the sexual violence and noted that the husband ‘did not hold her down and did not use violence during the act, and she did not scream‘ (para 28), the Chamber highlighted that:

‘the applicant herself seemed not to have been aware of the acute danger emanating from her husband in respect of their children following the incident reported […]. She remained in the marital home for three days after that incident before going to the authorities.’

(para 79)

The latter seems to suggest the following logic: if the applicant was not even aware of the real and immediate risk, how could the authorities be? This ignores the particularities of domestic violence, which, unlike other forms of violence, is an abuse of power. As explained by the Domestic Abuse Intervention Project, domestic violence is often linked to specific tactics using acts of intimidation, emotional abuse, coercion and threats followed by isolation, manipulation and limiting personal freedoms and choices. Factors such as economic dependency, the effects of prior isolation and psychological abuse, concerns for the children’s well-being and fear of stigma can lead to hesitation or even prevent victims of domestic violence to report their perpetrators to the police.

The reasoning of the Chamber as well as the authorities constitutes clear victim-blaming. While the Grand Chamber did not explicitly apply the same logic, it did not reject it either and rather resorted to formalistic arguments, which nevertheless endorsed the authorities’ (unjustified) under- and overestimation of certain factors without further explanation. In particular, this concerned underestimating the applicant’s concerns while overestimating the husband’s behaviour towards the authorities, for example his calm and cooperative demeanour during interrogations, despite the many objective risk factors present in the case. Adhering to the authorities’ logic, the majority stated that, regarding the possibility of pre-trial detention, they simply did not find the husband’s threats‘sufficiently serious or credible […] to point to a lethality risk that would have justified pre-trial detention or other more stringent preventive measures than the barring and protection order’ (para 207, emphasis added). Therefore, the majority unfortunately missed the opportunity to stand up against gender-based stereotyping and is, moreover, complicit in reinforcing a distorted view on domestic violence.

Victim-blaming is problematic regardless of its explicit or implicit nature and not only because it makes women and their children responsible for their own safety and for the violence they suffer. Such stereotyping can have far-reaching consequences in legal proceedings including judicial impartiality, misunderstanding the nature of the criminal offense, having a distorted view regarding witness credibility, limiting legal capacity, preventing effectively handling allegations, holding offenders legally accountable and preventing access to legal rights and protection (Cusack, Eliminating Judicial Stereotyping (2014)). Furthermore, this practice violates Art 5a of CEDAW, which obliges State Parties to eliminate wrongful gender stereotyping.

Ultimately, a first important step was made by establishing a gender-sensitive approach to cases of domestic violence in theory. However, the application of these principles in the concrete case shows how deeply imbedded a misogynistic culture is and how difficult it is for society – including the legal system – to move from theory to practice. This is why systemic changes such as standardised risk assessments, obligatory coordination between authorities and schools in cases of domestic violence as well as sensibility trainings for the executive and judiciary are essential for achieving progress in this area. Crucially, the latter was emphasised by the Court as part of the elaborated general principles. This is important as a mechanical usage of risk assessment and risk management tools such as the use of checklists can lead to ‘downgrading’ violence and ignoring the victim’s particular situation (Amesberger/Haller, The protection of IPV victims (2016)).

Missed opportunity to discern intersectional discrimination

While the Court highlighted that the issue of domestic violence has a clear gender dimension, the Article 14 claim based on sex discrimination was declared inadmissible as the issue was only raised by the applicant before the Grand Chamber and thus violated the six-month time limit rule. This constitutes another missed opportunity to frame the issue of domestic violence as the result of a historically established unbalance of power between men and women. Moreover, this specific case entailed an intersectional dimension as the woman was born in Turkey and only migrated to Austria when she was fourteen. As a migrant woman, her experience with the police cannot simply be equated to that of a (White) woman born in Austria, which should have been duly considered. While it is already a significant hurdle to report a case of domestic violence to the police due to fear of stigma, concerns for the children’s well-being or the effects of prior isolation and psychological abuse, it is all the more difficult as a migrant woman who is not fluent in German and may face gendered racism. In fact, the applicant was not provided an interpreter when she was questioned by the police – a fact which only emerged at the oral hearing as Judge Elósegui pointed out in her dissenting opinion. Moreover, a survey conducted by the Fundamental Rights Agency regarding racism in 12 European countries showed that respondents in Austria had the lowest level of trust in the police and reported the highest rates of perceived racial profiling by the police. Persisting racism in the legal system and the lack of trust in law enforcement heightens the under-reporting of crimes of domestic violence by women of colour and migrant women.

Applying Article 14 ECHR could have allowed the burden of proof to shift to the government after the applicant’s establishing a prima facie case of discrimination. This means that the government would have needed to show that a difference in treatment was objective and reasonably justified. This approach would have helped to shift the overall perspective from questioning the applicant’s credibility and aspects such as her hesitance to report the rape and strangulation toward the authorities’ duty to justify an approach that does not take the particular context of domestic violence as well as the specificities of intersectionality into account. At the same time, the Article 14 claim does not come without challenges in practice given the Court’s general reluctance to acknowledge cases of discrimination. This is particularly the case in the context of intersectionality where the issue is usually framed as one of heightened vulnerability (or multi-discrimination) rather than intersectionality (understood as a unique/specific disadvantage experienced due to different intersecting categories such as gender and race).

Conclusion

Kurt v Austria was an opportunity for the Court to clarify its approach in cases of domestic violence. While the Grand Chamber used the chance to establish general principles, the majority decision fell short on applying a sensitive approach regarding gender-based violence in the concrete case. It also missed the opportunity to highlight the issues of gender-based stereotyping and victim-blaming, which would have been best done through the lens of intersectional discrimination.

At the same time, it is important to point out that seven dissenting judges were in agreement that if the established general principles were adequately applied to the concrete case, an Article 2 violation should have been found because the Austrian authorities should have known that there was a real and immediate risk to the life of the applicant’s son. The unity in the dissent’s reasoning certainly gives hope that the current minority may become a majority position in the near future. In the meantime, human rights advocates may further their cause more effectively by putting more emphasis on Article 14 ECHR, intersectional discrimination and the vulnerability of children in cases of domestic violence.

* As a representative of the Women’s Popular Initiative 2.0 (Frauenvolksbegehren 2.0), Lisa Weinberger was involved in the third-party intervention in Kurt v Austria. She lives and works in Vienna in the field of environmental and constitutional law as well as gender equality.

One thought on “Kurt v Austria: A missed chance to tackle intersectional discrimination and gender-based stereotyping in domestic violence cases

  1. Dear Lisa. Thanks for bringing this important case, and the critique, to our attention. I find interesting the factual similarities, and yet apparently very different conclusions, between this case and CEDAW Committee’s González-Carreño v Spain (2014). About the CEDAW case and its domestic recognition https://www.ejiltalk.org/supreme-court-of-spain-un-treaty-body-individual-decisions-are-legally-binding/ (This is from August 2018; note the Spanish Supreme Court changed its position on the binding nature of UN Treaty Body decisions in another ruling of February 2020).

Leave a Reply