CASE OF M.A. v. ICELAND

SECOND SECTION

CASE OF M.A. v. ICELAND

(Application no. 59813/19)

JUDGMENT

Art 8 • Positive obligations • Private life • Failure to conduct an effective investigation into applicant’s complaints of domestic violence • Delays and administrative confusion leading to the expiry of limitation periods

Art 14 (+ Art 3 and Art 8) • Discrimination • Alleged gender-based discrimination in handling of domestic violence cases • Various legislative and policy measures adopted to combat sexual and domestic violence • Insufficient prima facie evidence of structural bias or disproportionate effect capable of shifting the burden of proof to the State

Prepared by the Registry. Does not bind the Court.

STRASBOURG

26 August 2025

FINAL

26/11/2025

This judgment has become final under Article 44 § 2 of the Convention.
It may be subject to editorial revision.

In the case of M.A. v. Iceland,

The European Court of Human Rights (Second Section), sitting as a Chamber composed of:

Arnfinn Bårdsen, President,
Saadet Yüksel,
Tim Eicke,
Gediminas Sagatys,
Stéphane Pisani,
Juha Lavapuro, judges,
Ragnhildur Helgadottir, ad hoc judge,

and Hasan Bakırcı, Section Registrar,

Having regard to:

the application (no. 59813/19) against the Republic of Iceland lodged with the Court under Article 34 of the Convention for the Protection of Human Rights and Fundamental Freedoms (“the Convention”) by an Icelandic national, Ms M.A. (“the applicant”), on 8 November 2019;

the decision to give notice to the Icelandic Government (“the Government”) of the complaints concerning the allegations of an ineffective investigation of domestic violence and gender-based discrimination, and to declare inadmissible the remainder of the application;

the decision not to have the applicant’s name disclosed;

the withdrawal of Oddný Mjöll Arnardóttir, the judge elected in respect of Iceland, from the case (Rule 28 of the Rules of Court), and the appointment of Ragnhildur Helgadóttir to sit as ad hoc judge (Article 26 § 4 of the Convention and Rule 29 § 1);

the parties’ observations;

Having deliberated in private on 8 July 2025,

Delivers the following judgment, which was adopted on that date:

INTRODUCTION

1. The case concerns the alleged failure of the Icelandic authorities to conduct an effective investigation into the applicant’s complaints of domestic violence, as well as allegations of gender-based discrimination in the handling of such cases. The applicant complained under Articles 3 and 8 of the Convention, taken alone and in conjunction with Article 14.

THE FACTS

2. The applicant was born in 1972 and lives in Kopavogur. She was represented before the Court by Sigrún Ingibjörg Gísladóttir, a lawyer practising in Reykjavík.

3. The Government were initially represented by Einar Karl Hallvarðsson, Agent, and subsequently by Fanney Rós Þorsteinsdóttir.

4. The facts of the case may be summarised as follows.

5. Alleged violence

6. In December 2017 the applicant lodged a complaint with the police against her former boyfriend, H.M.S., concerning two distinct incidents of physical assault allegedly committed in February and July 2016, as well as a threat made against her in May 2017.

7. The first alleged incident took place at her boyfriend’s residence on 29 February 2016. According to the applicant, during an altercation, he shoved her by applying pressure to her chest and swung her back and forth by alternately pulling and pushing her. The second alleged incident occurred at their joint residence on 4 July 2016. The applicant stated that H.M.S. shoved her in the chest, forced her out of the bathroom and later attempted to take her bag from her. In the course of this altercation, he allegedly shoved her, threw her back and forth and pulled her forward by the bag.

8. Following both incidents, the applicant sought medical attention, and her injuries were documented. After the February incident, doctors noted a suspected broken toe and requested an X-ray which was not conclusive. Following the July incident, the applicant took photographs of her injuries.

9. The applicant initially contacted the police in January 2017 but at that time did not wish to file a formal complaint. On 4 May 2017, after learning that she had contacted the police, H.M.S. threatened to send intimate photographs of her to her employer.

10. Police investigation

11. The central investigation department of the police, which investigates serious violent and sexual offences, conducted an initial assessment before transferring the case to a police station where domestic violence cases are generally handled. On 14 February 2018 a police officer at the station referred the case back to the central investigation department, reasoning that the investigation should be completed where it had begun. In March 2018 the case was discussed, and the alleged offence analysed, at a department meeting before being returned to the police station for investigation. During the course of the subsequent investigation, the relevant prosecutor sent the case back for further investigation on 22 October 2018, 14 February and 28 March 2019. Each time, the prosecutor noted that the case was old and that its investigation should be conducted without delay.

12. The applicant was questioned at a police station on 3 March 2018, providing details regarding the alleged physical assaults. Additional questioning took place by telephone on 25 May 2018 in relation to the threats and on 3 April 2019 concerning the assaults and her injuries.

13. In early June 2018 the applicant sent emails to the investigator, expressing concern that the statute of limitations in her case might expire. On 7 June the investigator responded, stating, “I understand your point of view and will proceed with the case as soon as I have an opportunity”, and inquiring whether the applicant had a new telephone number for H.M.S. On 12 June the investigator informed the applicant that attempts to contact H.M.S. had been unsuccessful. Police records indicate that H.M.S. was summoned for questioning on 12 July but failed to appear. A second summons was issued, and H.M.S. was ultimately questioned on 27 August. According to the Government, the email exchanges suggested that the investigator’s summer leave played a part in H.M.S. not being questioned until the end of August.

14. During questioning, H.M.S. partially admitted to having physical contact with the applicant during the July 2016 incident. He stated that, following an exchange of words, the applicant had begun striking him, at which point he had grasped her arms and pushed her into the living room. He acknowledged that she had sustained bruises on her arms as a result and expressed regret for this. However, he denied the applicant’s description of the incident as set out in her complaint and did not recognise that any altercation had taken place in February 2016. With regard to the threat concerning intimate photographs, he admitted making the threat but asserted that he had never intended to act upon it.

15. Between 24 October 2018 and 11 March 2019 the police took statements from four witnesses, some of whom were questioned more than once. The witnesses testified to having seen injuries on the applicant following the incidents in February and July 2016, and two of them stated that they had witnessed H.M.S. pushing the applicant against a door in July 2016.

16. Decisions by domestic authorities

17. On 23 April 2019 the police discontinued the investigation. In relation to the alleged assaults, the police determined that, as H.M.S. had not been questioned until August 2018, the two-year statute of limitations for offences under Article 217 of the General Penal Code (the “GPC”) had expired. The available evidence was not considered sufficient to support prosecution under Article 218b on domestic violence.

18. The applicant appealed this decision to the State Prosecutor, who upheld the decision regarding the assaults, confirming that they were time‑barred since H.M.S. was questioned in relation to the case on 27 August 2018. The limitation period had thus expired when his testimony was taken. The State Prosecutor further found that, while the applicant’s allegations were supported by evidence, Article 218b could not be applied retroactively to the alleged assault of February 2016 and that the July 2016 incident alone did not satisfy the requirement of repeated or serious conduct necessary for Article 218b to apply. However, the State Prosecutor invalidated the police’s decision in respect of the alleged threat and ordered that prosecution be initiated under Article 233 of the GPC.

19. On 12 November 2019 the police issued an indictment concerning the threat. H.M.S. was convicted of this offence by the Reykjanes District Court on 7 July 2020. The Court of Appeal upheld his conviction on 17 September 2021, noting that the fact that the threat was directed at his former cohabiting partner aggravated the seriousness of the offence.

20. In August 2021, the National Commissioner of Police, who had served as the chief of the metropolitan police during the investigation of the applicant’s case, publicly apologised to the applicant for the way her case had been handled. In particular, he acknowledged that procedural errors had resulted in the assaults becoming time-barred.

RELEVANT LEGAL FRAMEWORK

1. Domestic legal framework

   1. Constitution of the Republic of Iceland

2. Article 65 provides that everyone shall be equal before the law and enjoy human rights irrespective of, inter alia, sex. The second paragraph states that men and women shall enjoy equal rights in all respects.

3. General Penal Code No. 19/1940 (Almenn hegningarlög)

   1. Provisions on assault

4. Article 217 provides:

“Any person convicted of assault, providing it is not as serious as is described in Article 218, shall be fined or imprisoned for up to six months, and imprisoned for up to one year if the conduct involved is particularly reprehensible”.

20. Article 218 provides:

“If by a deliberate assault someone causes another person physical injury or health damage and these consequences can be regarded as his or her fault in terms of intention or negligence, the person shall be imprisoned for up to three years, or fined if there are extenuating circumstances.

Where serious physical injury or health damage results from an assault or where the offence is particularly dangerous in view of the method, including the implements, used, and also where the assault victim dies as a consequence of the attack, punishment for the offence shall take the form of up to sixteen years’ imprisonment”.

2. Domestic violence provisions

3. Article 218b, introduced into the GPC by Amending Act no. 23/2016 and in force from 5 April 2016, provides:

“Any person who, repeatedly or seriously poses a threat to the life, health or well‑being of his or her present or former spouse or cohabiting partner, to his or her descendant or the descendant of his or her present or former spouse or cohabiting partner, to an older person in his or her direct blood-line, or to other persons who live with him or her in the home or are in his or her care, by means of violence, threats, deprivation of freedom, coercion or in another manner, shall be imprisoned for up to six years.

A gross violation may be punishable by up to sixteen years’ imprisonment. When the seriousness of the violation is assessed, particular consideration shall be given to whether the injured party suffered major physical injury or damage to his or her health or whether the violation was fatal. Furthermore, consideration shall be given to whether the violation was committed in a particularly painful or injurious manner, whether it lasted a long time or whether the perpetrator grossly abused his or her superior position vis-à-vis the injured party”.

22. The Explanatory Report to Act No. 23/2016 clarifies that, although Article 218b highlights that domestic violence should not be regarded as a series of isolated incidents but rather as a continuous situation creating a climate of threat and fear, a single incident, if serious enough, can also engage criminal liability under the provision. It further explains that the means by which domestic violence might be committed are not limited to physical violence, threats, deprivation of freedom or coercion, but can also include forms of social, psychological and financial violence.

23. Article 233 provides that anyone who makes a threat of committing a criminal act designed to cause another person to fear for their life, health or well‑being, or that of other persons, shall be fined or imprisoned for up to two years.

24. Article 233b provides that anyone who insults or denigrates their spouse or ex-spouse, child or other closely related person, where the offence constitutes gross defamation, shall be imprisoned for up to two years.

25. General provisions

26. Article 2 provides that where penal legislation is amended between the commission of an offence and the delivery of judgment, the case shall be adjudicated in accordance with the newer legislation with respect to both criminal liability and punishment. However, no punishment may be imposed unless authorised by law at the time the act was committed, nor may a more severe punishment be imposed than that which would have been applicable under the law in force at that time.

27. Article 70(3) stipulates that where an offence is committed against a person closely related to the perpetrator and the nature of their relationship is deemed to have aggravated the seriousness of the offence, this shall normally be considered an aggravating factor in determining the punishment.

28. Article 81 establishes statutory limitation periods: two years for offences carrying a maximum penalty of one year’s imprisonment, five years for those punishable with up to four years’ imprisonment, ten years for offences punishable with up to ten years’ imprisonment and fifteen years for offences subject to over ten years’ maximum imprisonment.

29. Article 82 provides that the initiation of an investigation into a criminal case against a person as a suspect interrupts the statute of limitations. According to the explanatory report to the 1998 amendment, the limitation period is interrupted when a public investigation begins against a specific person as a defendant, which may occur through various means including arrest, custody, search, coercive measures, or questioning on suspicion of a criminal offence. The burden of proof regarding when the limitation period was interrupted rests with the prosecution.

30. Criminal Procedure Act No. 88/2008

31. The Act contains detailed provisions on criminal investigations and prosecutions. These include the requirements that investigations shall be carried out expeditiously (Article 53), that the police shall investigate when necessary based on knowledge or suspicion of crime regardless of whether a complaint has been received (Article 52), and that the burden of proof regarding guilt lies with the prosecution (Article 108). It also provides that, after receiving the case, the prosecutor may instruct the police to undertake additional investigative measures (Article 57), and that an indictment shall not be issued unless the investigative material is considered sufficient or likely to secure a conviction (Article 145).

32. Article 41 provides that the police shall appoint a legal representative for victims in sexual offence cases upon request, and in all cases where the victim is under eighteen. In cases concerning domestic violence, the police shall also appoint a legal representative when it is considered necessary. Upon the initiation of court proceedings, the legal representative shall be appointed by the court.

33. Act on Restraining Orders and Removal from the Home No. 85/2011

34. The Act permits the imposition of restraining orders where there is reasonable suspicion of the commission, or the risk of commission, of a criminal offence, or where the perpetrator otherwise disturbs the victim’s peace (Article 4). The perpetrator may also be arrested and removed from the home if there is reasonable suspicion of the commission, or the danger of commission, of certain punishable offences, including sexual and physical violence as well as offences under Articles 233 and 233b and, since its entry into force, Article 218b of the GPC (Article 5).

35. Council of Europe

36. The Council of Europe Group of Experts on Action against Violence against Women and Domestic Violence (GREVIO), in its Baseline Evaluation Report on legislative and other measures giving effect to the provisions of the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (Istanbul Convention) in the Republic of Iceland, GREVIO/Inf(2022)26, published on 14 November 2022, noted the persistence of gender stereotypes and the continued need to address violence against women in Iceland, while highlighting “the clear commitment” by the Icelandic Government to “combat gender-based violence, to improve the legal status of victims [and] to prioritise the striving for gender equality” (§§ 2-3). It welcomed Iceland’s ranking in 2021, for the twelfth consecutive year, at the top of the Global Gender Gap Index, and commended the Icelandic authorities for the “official anchoring of a gender equality perspective in all areas of governance and policy making” (§ 16).

37. As regards coordinated and comprehensive measures to prevent and combat violence against women, GREVIO noted that “the first action plan on domestic and sexual violence in Iceland [had been] introduced in 2006” and was followed by further action plans on gender-based violence, which focused on combating domestic violence and sexual violence against women and children and aimed to improve the legal status of and facilities for victims. Emphasis was also placed on “strengthening preventive efforts against violence” (§ 30).

38. As regards the collection of data on violence against women, GREVIO noted that, according to the authorities, statistics on such violence were “not currently compiled in a single location” and that “there [was] no overview available on the extent of the different forms of violence covered by the Istanbul Convention in Iceland”. Nevertheless, the police collected data on domestic violence “disaggregated by sex, age, nationality, type and place of the offence, and relationship of the perpetrator to the victim”. Statistics on sexual and domestic violence, including yearly trends, were published on the police website and updated quarterly. Overall, the police and police prosecutors had “a solid data system in place for recording reported offences of sexual violence and domestic violence”. However, GREVIO regretted that “no data on indictments or convictions [were] being collected or made available publicly” (§§ 53-55).

39. As regards substantive criminal-law provisions and the initiation of proceedings, GREVIO noted with satisfaction that the Icelandic GPC provides for ex officio initiation of legal proceedings in relation to all of the offences listed in Article 55, paragraph 1, of the Istanbul Convention (§ 286). It also noted legislative developments following a 2012 report on the GPC’s conformity with the Convention and remarked that they demonstrated “Iceland’s serious commitment to tackling violence against women and stemming impunity” (§ 192). GREVIO observed that, with one exception, the aggravating circumstances contained in Article 46 of the Convention form part of the Icelandic legislation, either in Article 70 of the GPC or as a part of the provisions of the substantive criminal law (§ 234). It further welcomed the adoption of the new Article 218b and noted the judicial clarification of the threshold for its application in cases of a single violent incident. It observed, however, that acts of psychological violence occurring between partners who do not share a residence were not covered within its scope, and that its efficiency in terms of protecting against psychological violence remained to be seen (§§ 193 and 199).

40. As regards the response of law-enforcement agencies, GREVIO welcomed the issuance of protocols for police officers dealing with cases of domestic violence by the National Commissioner for Police “as early as 2005”, with subsequent updates in 2014 and 2018. It was satisfied that these protocols were “comprehensive and provided a model of good practice consistent with the requirements of the Istanbul Convention”, although it noted that their implementation “remain[ed] inconsistent and further effort [was] required in that respect” (§ 242). It also noted the State Prosecutor’s issuance, in 2017, of instructions prioritising the investigation of rape cases, cases involving violence against children and violence in close relationships and, in 2018, of instructions requiring the implementation of a standardised and timed investigation plan for such cases (§ 251). GREVIO further noted that “following a study conducted in 2013 and 2014, which found that acquittal rates for sexual offences were high in Iceland, an Action Plan on Sexual Offences was passed by the parliament in 2017” with “several measures to improve the situation, in particular aimed at increasing the rate of prosecutions and convictions of perpetrators of sexual offences”. It also provided for “15 new full-time police officer positions ... added to police departments all over the country”, and additional funding to update procedures and investigative equipment (§ 250).

41. Noting the concerns raised by NGOs and women’s organisations over the number of cases concerning sexual violence that did not proceed to prosecution, and how the deployment of more resources to their investigation created “a new bottleneck” in prosecutions (§§ 252-53), GREVIO urged the authorities “to significantly reinforce their investigative and prosecutorial capacity and to take immediate measures to ensure a prompt and appropriate response by law-enforcement agencies in all cases of violence against women” (§ 259). GREVIO further observed that, as regards domestic violence, “the lack of statistics in respect of interventions, prosecutions and convictions [made] it difficult to assess the effectiveness of the system”. It noted that “numerous concerns were raised by NGOs and women’s organisations in respect of the delays in getting to court, leading to reduced or suspended sentences, and unduly negative approaches to women victims of violence at District Court and Court of Appeal level” (§ 263).

42. United Nations

43. The United Nations Committee on the Elimination of Discrimination against Women (CEDAW), in its Concluding Observations on the Combined Seventh and Eighth Periodic Reports of Iceland, adopted at its sixty‑third session on 15 February - 4 March 2016 (CEDAW/C/ISL/CO/7-8), noted with concern the increasing number of cases of violence against women and welcomed “the new rules, procedures and risk assessment tool issued by the Reykjavik Metropolitan Police in March 2015 for cases of domestic violence”. It expressed particular concern about “the high number of discontinuances in criminal proceedings on charges of violence against women, in particular rape and sexual violence, by the State Prosecutor and the low number of convictions” and recommended, among other measures, that Iceland incorporate the Istanbul Convention into national legislation to strengthen legal protection for women, including by criminalising online harassment and psychological violence. It further called upon Iceland to enhance efforts to prosecute and convict perpetrators of rape and sexual violence and analyse and address the causes of the high acquittal rates (§§ 19‑20).

44. In its Concluding Observations on the Ninth Periodic Report of Iceland, adopted at its eighty-fifth session (8-26 May 2023) (CEDAW/C/ISL/CO/9), the CEDAW Committee noted with appreciation Iceland’s top ranking in the 2022 Global Gender Gap Index and welcomed the progress achieved since the last report, including the criminalisation of digital sexual violence and stalking, increased protection for victims of psychological violence and the definition of rape based on absence of consent. However, it expressed concern over “the high number of criminal proceedings concerning gender-based violence against women, in particular rape and sexual violence, discontinued by the State Prosecutor and the low number of convictions, notwithstanding the adoption and implementation of the action plan concerning sexual offences for the period 2018-2022”. It also highlighted “the lack of disaggregated data on all forms of gender-based violence against women and girls, including domestic violence”. Among other measures, it recommended addressing the causes of underreporting and high acquittal rates in cases of gender-based violence and ensuring the collection and analysis of data on all such forms of violence (§§ 4, 25-26).

THE LAW

1. ALLEGED VIOLATION OF ARTICLES 3 AND 8 OF THE CONVENTION

2. The applicant complained that the authorities failed to conduct an effective investigation into her allegations of ill-treatment and to provide adequate protection against gender-based violence, in breach of Articles 3 and 8 of the Convention, which read as follows:

Article 3

“No one shall be subjected to torture or to inhuman or degrading treatment ...”

Article 8

“1. Everyone has the right to respect for his private and family life ...”

1. Admissibility

2. The Government acknowledged that the applicant had exhausted effective remedies in the criminal proceedings by appealing to the State Prosecutor. However, in their view, the applicant had not exhausted domestic civil remedies. As the National Commissioner had publicly admitted mistakes that resulted in the assaults becoming time-barred, she could have lodged a civil claim for damages, which might have led to an award of compensation. The Government also claimed that the applicant’s complaint was manifestly ill-founded.

3. The applicant emphasised that her complaint concerned the ineffective criminal investigation and resulting impunity of her alleged attacker, not compensation. In particular, she maintained that pursuing civil claims would not have addressed the State’s failure to effectively investigate and prosecute gender-based violence.

4. As regards the Government’s argument that the applicant could have instituted civil proceedings for damages, the Court notes that civil proceedings would not concern the same subject matter as her application to the Court, which relates to the authorities’ failure to conduct an effective criminal investigation into her allegations of domestic violence. The Court reiterates that effective deterrence against serious attacks on physical and psychological integrity requires efficient criminal-law mechanisms that would ensure adequate protection in that regard (see Vučković v. Croatia, no. 15798/20, § 41, 12 December 2023, and Volodina v. Russia, no. 41261/17, § 100, 9 July 2019). The Court therefore dismisses the Government’s objection as to the non-exhaustion of domestic remedies.

5. The Court considers that this complaint is neither manifestly ill‑founded nor inadmissible on any other grounds listed in Article 35 of the Convention. It must therefore be declared admissible.

6. Merits

   1. Submissions by the parties

(a) The applicant

45. The applicant submitted that the domestic legal framework had failed to afford practical and effective protection of her rights. She asserted that, prior to the entry into force of Article 218b of the General Penal Code (the “GPC”) in 2016, victims of domestic violence had not been adequately protected under Icelandic law, a deficiency identified in the observations of the CEDAW Committee. In her view, equating domestic violence with random physical assaults, particularly in relation to the two-year statute of limitations for offences under Article 217 of the GPC, was incompatible with Articles 3 and 8 of the Convention, as well as with the Istanbul Convention. She emphasised that Article 218b had extended the limitation period to ten years and allowed for viewing incidents as a continuous pattern rather than isolated events, demonstrating that the previous framework had been inadequate.

46. The applicant disputed the Government’s claim that domestic violence cases were prioritised by the police, pointing to significant delays in her own case. For example, H.M.S. was not questioned until nine months after she filed her complaint and witness statements were not taken until eleven months later. While acknowledging that some time had elapsed between the incidents and her complaint – approximately twenty-two months since the February 2016 incident and seventeen months since the July 2016 incident, she maintained that this made it more crucial to commence an investigation immediately to secure available evidence. She specifically noted her June 2018 email exchange with the investigator expressing concern about the approaching statute of limitations, to which the investigator responded that they would proceed at the earliest opportunity and subsequently took summer leave. The various guidelines and procedures for the expedited handling of such cases, including the State Prosecutor’s Instructions 4/2017 prioritising domestic violence cases, were of little value if they were not effectively implemented in practice. Furthermore, the Government’s decision to increase funding to ensure effective investigations after the investigation in her case further demonstrated that the delays in her proceedings had been unacceptable.

47. The applicant further maintained that her account was credible and corroborated by additional evidence, including medical records documenting a suspected broken toe after the February 2016 incident and photographs of injuries following the July 2016 assault. Two witnesses had directly observed H.M.S. pushing her against a door during the July 2016 incident, and others had seen her injuries following both assaults. She argued that the police systematically discounted or dismissed victims’ evidence in domestic violence cases, while according undue weight to the denials of the accused, despite H.M.S.’s partial admission to physical contact and his acknowledgment that she sustained bruises. The dismissal of witness testimony as merely “retelling” the victims’ accounts inappropriately diminished crucial evidence in domestic violence cases, where direct witnesses were rare by nature.

48. The applicant submitted that, although she filed charges in December 2017, an analysis of the applicable legal provisions was not conducted until March 2018, by which time the first physical attack had already become time‑barred. While the second attack had not yet been time-barred at that stage, the police nevertheless delayed taking H.M.S.’s statement for several more months, ultimately leading to its time-barring as well. No evaluation took place as to whether the July 2016 violence, together with the subsequent threats, constituted a pattern of repeated abuse, reflecting a failure to properly investigate domestic violence as a continuing pattern rather than as isolated incidents. These delays were particularly serious given that the police were aware that she feared for her safety, having been threatened over her contact with them, and that she maintained regular contact with H.M.S. due to shared custody arrangements. Her account’s credibility was demonstrated by H.M.S.’s subsequent conviction for threats, and the time-barring reflected not mere procedural irregularities but a complete failure to investigate, exposing her to continued risk and preventing her credible allegations from ever reaching court.

(b) The Government

49. The Government submitted that investigations into sexual assault and domestic violence had been prioritised by the police and prosecution for decades and were subject to continuous review by the State Prosecutor. They pointed to Instructions 4/2017 expressly prioritising domestic violence cases and Instructions 2/2018 requiring digital investigation plans for such cases. Before the enactment of Article 218b of the GPC in 2016, protection had been provided through Articles 217 and 218 on physical assault and Article 70(3), which prescribed aggravated punishment for offences committed within close relationships. Although psychological violence had not been a punishable offence, the law had still offered protection, which increased significantly after Article 218b came into force with its longer limitation periods and comprehensive approach to domestic violence.

50. The Government emphasised that investigators were required to establish the truth and give equal weight to evidence supporting both acquittal and conviction, in accordance with Article 70 of the Constitution and Article 6 § 2 of the Convention. While accepting that the burden of proof in domestic violence cases was difficult, the Government rejected the applicant’s claim that it was “almost impossible to fulfil”. The legislative framework allowed for consideration of various types of evidence, including medical certificates concerning psychological effects and accounts of events before and after the alleged incidents.

51. While acknowledging an increase in the influx cases during 2017‑18 and a lack of manpower, the Government maintained that this had been met with organisational changes and increased funding, noting in particular the addition of six full-time positions at the Reykjavík Metropolitan Police Department. This did not, therefore, indicate that sexual and domestic violence cases had not been prioritised. Urgent cases reported immediately after an incident were given priority over those reported later, such as the applicant’s case. The delays were not due to the nature of the offences, nor did they suggest a lack of police prioritisation. Instructions issued by the State Prosecutor placed clear emphasis on expediting proceedings and prioritising such cases.

52. Turning to the investigation in the applicant’s case, the Government submitted that the level of severity required to engage Articles 3 and 8 of the Convention had not been met. While acknowledging procedural shortcomings, including the almost nine-month delay between the applicant’s statement and the questioning of H.M.S., they maintained that these, although regrettable, did not in themselves breach Convention obligations since the investigation had otherwise been conducted effectively. The applicant had lodged her complaint almost two years after the first alleged assault in February 2016 and fifteen months after the second in July 2016. Although her report did not concern recent offences, it was undisputed that the delay had led to the expiry of the statute of limitations. Nonetheless, all relevant facts had been established within eighteen months of the complaint. Given the time elapsed since the events, the Government maintained that the case had not required higher priority and that they had no information suggesting H.M.S. was aware of the complaint before his questioning.

53. The first alleged assault had occurred before the enactment of Article 218b of the GPC on 5 April 2016 and, accordingly, Article 217 had applied with its two-year limitation period. Had the applicant obtained an X‑ray confirming the suspected toe fracture, the case might have been classified under Article 218, which carried a longer limitation period. The second alleged assault had taken place after Article 218b came into force. However, the Government argued that it had not met the severity threshold for that provision to apply, as Article 218b required either repeated or serious conduct and the July 2016 incident, viewed in isolation, did not satisfy these criteria given the non-retroactivity principle prevented consideration of the February 2016 incident.

54. The Government emphasised that witness statements had been taken, though belatedly between October 2018 and March 2019, and that they had supported the applicant’s account to the extent that witnesses confirmed seeing injuries and two witnessed H.M.S. pushing the applicant. However, the witnesses could only testify to what the applicant or H.M.S. had told them, and these statements were contradictory. Even if the statements had been collected earlier, the evidence would still have been insufficient to support the applicant’s accusations given H.M.S.’s partial denial: while he admitted physical contact and causing bruises, he disputed the applicant’s characterisation of the events. While acknowledging that H.M.S. should have been questioned sooner, the Government pointed out that the case had not been ignored, as H.M.S. had ultimately been convicted of making threats against the applicant, an offence carrying heavier penalties (up to two years’ imprisonment) than those under Article 217 (up to one year).

55. The Government also addressed the administrative confusion that occurred when the case was transferred between the central investigation department and a police station, with multiple referrals back and forth in February and March 2018. They acknowledged that the investigator’s summer leave had contributed to the delays, as evidenced by the June 2018 email exchanges. However, they maintained that these operational issues, while unfortunate, resulted from resource constraints affecting the police generally rather than any specific reluctance to investigate domestic violence cases. They emphasised that, despite these delays, a comprehensive investigation was ultimately conducted, including questioning of eleven witnesses and consideration of all available evidence.

56. The Court’s assessment

(a) General principles

56. The Court reiterates that the issue of domestic violence, which can take various forms – ranging from physical assault to sexual, economic, emotional or verbal abuse – transcends the circumstances of an individual case. It is a general problem which affects, to a varying degree, all member States and which does not always surface since it often takes place within personal relationships or closed circuits and affects different family members, although women make up an overwhelming majority of victims. The particular vulnerability of the victims of domestic violence and the need for active State involvement in their protection have been emphasised in a number of international instruments and the Court’s case-law (see Kurt v. Austria [GC], no. 62903/15, §§ 161-62, 15 June 2021, and Opuz v. Turkey, no. 33401/02, §§ 72-86, ECHR 2009).

57. Since in cases concerning violence inflicted by private parties the distinction between the requirements of Articles 3 and 8 of the Convention is not clear-cut, the Court may examine the applicant’s complaints simultaneously under both provisions. Indeed, both provisions impose an obligation on the State to safeguard an individual’s physical and psychological integrity and, together with Article 2, form a continuum that triggers the State’s duty to provide protection once it has been established that attacks on an individual’s integrity were sufficiently serious to necessitate a response (see Hanovs v. Latvia, no. 40861/22, § 45, 18 July 2024, with further references, and Vučković v. Croatia, no. 15798/20, § 54, 12 December 2023).

58. The positive obligation of the authorities to protect victims of violence has three key aspects. First, they must establish a legislative and regulatory framework of protection. Second, in certain well-defined circumstances, they are required to respond promptly to reports of domestic violence and take operational measures to protect individuals at risk of ill-treatment. Third, they must conduct an effective investigation into arguable claims concerning each instance of such ill-treatment (see X and Others v. Bulgaria [GC], no. 22457/16, § 178, 2 February 2021; M.C. v. Bulgaria, cited above, § 153; and Kurt, cited above, § 165, with further references).

59. Regarding the existence of a legal framework, the Court’s case-law and relevant international materials reflect a common understanding that comprehensive legal and other measures are necessary to ensure effective protection for victims of domestic violence. These measures must include, in particular, the criminalisation of acts of violence within the family through effective, proportionate and dissuasive sanctions (see Ž.B. v. Croatia, no. 47666/13, §§ 50-51, 11 July 2017, and Galović v. Croatia, no. 45512/11, § 114, 31 August 2021).

60. The procedural obligation to conduct an effective investigation involves the duty of the domestic authorities to apply in practice the criminal-law mechanisms established to prohibit and punish conduct contrary to Articles 3 and 8 of the Convention. For an investigation to be effective, it must be prompt and thorough throughout the proceedings. The authorities are required to take all reasonable steps to secure evidence related to the incident, including forensic evidence. Particular diligence is necessary in cases of domestic violence, and the specific nature of such violence must be taken into account (see Tunikova and Others v. Russia, nos. 55974/16 and 3 others, § 114, 14 December 2021, and Vieru v. the Republic of Moldova, no. 17106/18, § 81, 19 November 2024).

61. Nonetheless, the Court reiterates that the procedural obligation is a requirement of means and not of results. There is no absolute right to the prosecution or conviction of a particular person where no culpable failures have occurred in the authorities’ efforts to hold perpetrators accountable. The mere fact that an investigation has yielded limited or inconclusive results does not, in itself, indicate any failing. While the authorities must take all reasonable steps to gather evidence, clarify the circumstances, and conduct a thorough, objective and impartial analysis of all relevant elements without neglecting any obvious lines of inquiry, the procedural obligation must not be construed as imposing an impossible or disproportionate burden. The Court needs not be concerned with allegations of errors or isolated omissions and cannot replace the domestic authorities in the assessment of the facts of the case or decide on the alleged perpetrators’ criminal responsibility. Instead, it must focus on whether there were significant shortcomings in the proceedings, namely those capable of undermining the investigation’s ability to establish the circumstances or identify those responsible (see X and Others v. Bulgaria, cited above, § 186, and S.M. v. Croatia [GC], no. 60561/14, § 315-20, 25 June 2020, with further references).

(b) Application of the principles to the present case

62. The applicant submitted that her former partner subjected her to physical violence on more than one occasion, which posed a threat to her physical integrity and well-being. She alleged a failure to comply with the first and third aspects of the State’s positive obligations outlined in paragraph 58 above. The Court does not need to determine whether the alleged violence was sufficiently serious to meet the minimum level of severity required to bring it within the scope of Article 3 of the Convention because, in any event, the domestic authorities had an identical positive obligation to protect victims of violence under Article 8 of the Convention. Accordingly, it will first assess whether the domestic legal framework afforded adequate protection against domestic violence at the material time, before turning to the question of whether the investigation into the applicant’s specific allegations met the standards required under the Convention.

(i) Legislative framework

63. As noted in paragraph 59 above, the States’ obligations in cases of domestic violence generally require the establishment of effective criminal-law provisions. It is well established that the Court’s role is not to substitute its own assessment for that of national authorities in selecting among the various measures that could ensure compliance with their positive obligations under the Convention (see Ž.B. v. Croatia, cited above, § 58). Different legislative approaches may fulfil these requirements, provided they offer effective and practical protection against domestic violence. As reflected in the Court’s case-law concerning various Contracting States, domestic violence may be appropriately addressed either through specific criminal offences or by treating it as an aggravating factor in other violent crimes (ibid., § 57; see also E.M. v. Romania, no. 43994/05, § 62, 30 October 2012, and Valiulienė v. Lithuania, no. 33234/07, § 78, 26 March 2013).

64. In assessing the adequacy of the protection framework in Iceland during the relevant period, the Court observes that although Article 218b, which specifically criminalises domestic violence, was not introduced into the GPC until 2016, the legal framework had already provided for enhanced punishment in such cases. Article 70(3) of the GPC stipulated that a close relationship between the perpetrator and the victim should be considered an aggravating circumstance where the relationship increased the gravity of the offence (see paragraph 26 above). This provision thus allowed for heavier penalties in domestic violence cases prosecuted under the general assault provisions of Articles 217 and 218 (see Ž.B. v. Croatia, cited above, § 54). The Court reiterates that while the Istanbul Convention requires the criminalisation of domestic violence, it does not prescribe specific standalone offences. Rather, it allows for the inclusion of domestic violence as either a constituent element of particular offences or an aggravating circumstance in sentencing for other offences established by the Convention (ibid., § 56; see also the GREVIO Baseline Report on Iceland in paragraph 35 above).

65. Insofar as criminal proceedings could be initiated ex officio, without requiring a formal complaint from the victim, the legislative approach adopted by Iceland conforms with the relevant international standards (see Volodina, cited above, §§ 82-84, and the GREVIO Baseline Report on Iceland in paragraph 35 above).

66. As regards physical violence, the Court notes that Articles 217 and 218 of the GPC criminalised assault, with a two-year limitation period for less serious offences under Article 217. While a longer limitation period might arguably be considered desirable in domestic violence cases given its nature and victims’ particular vulnerabilities, the Court does not find that the two-year period was, in itself, contrary to the requirements of the Convention, insofar as States enjoy a wide margin of appreciation in setting limitation periods (see Stubbings and Others v. the United Kingdom, 22 October 1996, §§ 55-56, Reports of Judgments and Decisions 1996-IV). What is at issue in the present case is not the length of the limitation period as such but whether the authorities took all reasonable steps to investigate the applicant’s allegations with the expediency required by the applicable limitation period, in order to ensure practical and effective protection.

67. Having found that the framework in place during the relevant period provided adequate mechanisms for protection against domestic violence through criminal law (see Ž.B. v. Croatia, cited above, § 58), the Court must examine next whether, within this legal framework, the authorities conducted an effective investigation into the applicant’s allegations and what part the alleged delays may have played in its outcome.

(ii) Adequacy of the investigation

68. The Court acknowledges that domestic violence cases present inherent evidentiary challenges when the offences occur in private settings without any witnesses, and these challenges are further compounded when the events are reported only after a significant passage of time. While the authorities must remain sensitive to the particular dynamics of domestic violence that may explain delayed reporting, such delays can affect the available evidence and may legitimately influence the prioritisation of cases by the authorities. However, resource constraints and operational priorities cannot justify excessive tardiness in basic investigative steps once allegations have been brought to the authorities’ attention. In each case, the fundamental question is whether, in the circumstances, including the timing of the complaint, the relevant limitation period and the risk of evidence being lost, the authorities demonstrated the requisite diligence.

69. The Court notes that the applicant’s allegations concerned events that had occurred approximately seventeen to twenty-two months before she lodged her complaint in December 2017. It acknowledges that victims of domestic violence often hesitate to file charges until some time has passed, such as after the end of their relationship with the alleged perpetrator. While such delays should not in themselves prejudice victims’ claims, given the particular dynamics of domestic violence, they may nonetheless affect the evidentiary position.

70. However, the Court emphasises that delay on the part of the victim to file charges does not in itself absolve the authorities of their obligation to conduct an effective investigation once the allegations have been brought to their attention. A context-specific and comprehensive approach to domestic violence requires a high level of diligence and expedition, taking into account the legal and factual circumstances of each case.

71. In the present case the applicant lodged her criminal complaints in December 2017, yet she was first interviewed by the police only in March 2018, three months later (see paragraphs 5 and 10 above). It is also undisputed that further investigation of the reported violence against the applicant was warranted, particularly given the need to interview H.M.S., the suspect. In view of the imminent expiry of the limitation period the authorities were required to proceed with particular expedition, and H.M.S should have been interviewed as a matter of urgency to prevent the statutory limitation from expiring (see paragraph 28 above). Had H.M.S been interviewed as a suspect at any point before the end of February 2018 none of the offences would have become time‑barred. An interview by the end of June 2018 would, at the very least, have prevented the second alleged incident from becoming time‑barred. None of these measures was taken, notwithstanding the prosecutor’s repeated instructions that the case be investigated without delay and despite the authorities’ knowledge that the applicant remained in regular contact with H.M.S owing to shared custody arrangements and that he had previously threatened her over her communications with the police.

72. Initial delays appear to have arisen from administrative confusion, with the case being transferred between different police departments before a determination was made as to which unit should handle the investigation. Even after the applicant expressly raised concerns about the impending expiry of the limitation period in correspondence with the investigator in June 2018, it does not appear that any urgent measures were taken to expedite the investigation. The investigator’s response that they would proceed as soon as practicable and their decision to take summer leave during this critical period without arranging for a replacement appears to have contributed to the delayed questioning of H.M.S. and ultimately resulted in the second assault becoming time-barred as well.

73. The Court has consistently held that where criminal proceedings are discontinued due to the expiry of limitation periods as a result of shortcomings in the actions of the relevant State authorities, the objective of providing effective protection against acts of ill-treatment cannot be achieved (see Mehmet Yaman v. Turkey, no. 36812/07, § 71, 24 February 2015). Even if the facts of the case and the identity of the alleged offenders have been established, an investigation can hardly be regarded as effective and capable of leading, if appropriate, to the punishment of those responsible, where prosecution becomes time-barred due to the slow pace of proceedings (see Valiulienė, cited above, § 85, and P.M. v. Bulgaria, no. 49669/07, § 66, 24 January 2012).

74. As regards the Government’s contention that the delays resulted from staff shortages and an increased caseload during 2017-18, the Court reiterates that it is for the Contracting States to organise their legal systems in such a manner as to enable their authorities to comply with the requirements of the Convention, including those arising from the positive obligations under Article 8. The subsequent increase in funding and staffing levels, while commendable, does not absolve the State of its responsibility for the deficiencies in the applicant’s case.

75. The Court has thus identified a key deficiency in the authorities’ response to the applicant’s allegations. The investigation was marked by delays and administrative confusion, leading to the expiry of the limitation periods despite the authorities’ awareness of the approaching time-limits and the prosecutor’s repeated instructions to expedite the case. That H.M.S. was later convicted of making threats against her over her communication with the police only underscores the gravity of the authorities’ failure to ensure an effective investigation into the alleged physical violence.

76. Having regard to the above considerations, the Court finds that the authorities failed to demonstrate the diligence required in investigating allegations of domestic violence and to provide the level of protection to which the applicant was entitled.

(c) Conclusion

77. There has accordingly been a violation of the procedural limb of Article 8 of the Convention.

78. ALLEGED VIOLATION OF ARTICLE 14 OF THE CONVENTION READ IN CONJUNCTION WITH ARTICLES 3 AND 8

79. The applicant complained under Article 14 of the Convention, read in conjunction with Articles 3 and 8, that she had suffered discrimination as a woman in the enjoyment of her Convention rights. Article 14 provides:

“The enjoyment of the rights and freedoms set forth in [the] Convention shall be secured without discrimination on any ground such as sex ...”

1. Admissibility

2. The Court notes that this complaint is neither manifestly ill‑founded nor inadmissible on any other grounds listed in Article 35 of the Convention. It must therefore be declared admissible.

3. Merits

   1. Submissions by the parties

(a) The applicant

80. The applicant submitted that she had suffered discrimination as a woman due to the authorities’ systematic failure to effectively investigate and prosecute domestic violence. She maintained that, as a rule, violence against women was treated differently within the system than other forms of violence where the victims were male. Despite the reforms introduced by the Icelandic authorities in recent years, the overall unresponsiveness of investigations and the low rate of prosecutions demonstrated an insufficient commitment on their part to take appropriate action to address violence against women. The Government’s references to women in senior positions and Iceland’s generally progressive stance on gender equality did not offset the failures in prosecuting gender-based violence.

81. According to official statistics for the years 2015 to 2021 in the capital area, 84 to 86% of domestic violence perpetrators were men, while 75.5 to 89% of victims were women. Statistics from other regions of the country were similar. However, data from Stígamót indicated consistently low reporting and prosecution rates. Between 2014 and 2019, approximately 85 to 88% of victims seeking assistance were women, yet only 8 to 13% of cases were reported to the police. Of those reported, an average of only 30% resulted in prosecution, and of those, an average of 47% led to conviction. The applicant highlighted specific examples from Stígamót’s annual reports: in 2016, only 23.1% of cases reported to police led to prosecution, with 58.3% of those resulting in conviction; in 2017, the figures were 25% and 29.4% respectively. She contrasted these with prosecution rates for physical assault cases, which according to the State Prosecutor’s Annual Report 2019, stood at 61% in 2019, 57% in 2017, and 74% in 2014. The disparity in prosecution rates between gender-based violence and other violent crimes required an explanation from the Government, which had not been provided.

82. The applicant submitted that gender-based violence presented unique challenges, as such crimes were typically committed without witnesses, over an extended period, and with forensic evidence often difficult to obtain. Victims were thus at a disadvantage in contributing to prosecutions compared to victims of other violent acts. She argued that applying exactly the same procedures to these cases as to other forms of violence was ineffective and that investigations and the regulatory framework must account for the nature of the available evidence to ensure victims’ rights under Articles 3 and 8. Evidence such as psychological effects confirmed by specialists, accounts of events before and after the violence, and testimonies about patterns of controlling behaviour must be seen as important. The evidentiary threshold in domestic violence cases was, in her view, effectively impossible to meet, as authorities systematically diminished or dismissed evidence specific to such cases, including psychological assessments and witness testimonies on patterns of controlling behaviour. Her own case illustrated this discrimination, as credible allegations supported by medical evidence, photographs of injuries, direct witness observations of H.M.S. pushing her, and witness statements were not prosecuted due to delays leading to the expiry of limitation periods.

(b) The Government

83. The Government disputed the existence of any systematic discrimination against women in the Icelandic justice system. They maintained that the evidence submitted by the applicant did not substantiate any bias or discriminatory nature in Icelandic legislation or practices. The legislative framework was entirely gender-neutral in terms of the same burden of proof and procedural requirements applying irrespective of the victim’s gender. Iceland prioritised the protection against sexual and domestic violence and the system for protecting women’s rights was among the most advanced globally. The domestic authorities had systematically implemented effective and targeted measures to combat domestic and sexual violence, as evidenced by the submitted documents on various action plans and targeted programmes. These included information on specialised investigative teams, a priority policy, coordination with social services and specialised risk assessments in cases of domestic violence, as well as targeted police and prosecutorial training. The Government further submitted that two specialised hospital emergency units for victims of sexual violence existed in the country and that three centres had been set up, where victims could seek multidisciplinary help and police assistance in one location.

84. The Government maintained that the predominance of women among victims of sexual and domestic violence was a global phenomenon and not unique to Iceland. While women constituted the majority of such victims, the Chief Commissioner of Police’s analysis of data for 2015-2020 showed that there was no measurable difference in prosecution rates between cases involving male and female victims of domestic and sexual violence. Any disparity in prosecution rates between homicide and assault cases, where direct witnesses and physical injuries were more common, on one hand, and sexual offences, on the other, reflected inherent evidentiary challenges in the latter type of cases rather than any discriminatory practice. The Government further submitted comparative statistics on rape (Article 194 of the GPC) and grave physical assault (Article 218 of the GPC) cases for the capital area in the years 2013-21, demonstrating comparable prosecution rates. Lastly, they asserted that statistics from Stígamót could not replace official data, particularly as they included alleged offences that had never been reported to the authorities.

85. The Government emphasised that the delays in the applicant’s case were exceptional and not representative of domestic violence investigations. They pointed to Instructions 4/2017 placing clear emphasis on expediting proceedings and giving priority to these types of cases. In any event, the delay was unrelated to the fact that the victim was a woman or to any differential treatment by the police compared to cases involving male victims. They maintained that the finding of procedural errors in an individual case did not demonstrate institutional discrimination. The fact that the National Commissioner had apologised for the handling of the case demonstrated the authorities’ commitment to accountability and improvement rather than any systematic bias. They noted that the applicant had failed to produce any reports or statistics demonstrating discrimination against women in Iceland, unlike the extensive documentation in cases such as Opuz, cited above.

86. The Court’s assessment

(a) General principles

86. The principles concerning discrimination in the context of domestic violence are well established in the Court’s case-law (see Opuz, cited above, §§ 184-91; Volodina, cited above, §§ 109-14, and, for a recent summary, Y and Others v. Bulgaria, no. 9077/18, § 122, 22 March 2022, and M.S. v. Italy, no. 32715/19, § 54, 7 July 2022). The Court has held that violence against women, including domestic violence, constitutes a form of discrimination. Such discrimination may arise not only from explicit policies of different treatment but also from de facto situations in which general policies or measures have disproportionately prejudicial effect on women, irrespective of discriminatory intent.

87. Once an applicant has shown that there has been a difference in treatment or disproportionate effect, it is for the respondent Government to show that it was justified (see D.H. and Others v. the Czech Republic [GC], no. 57325/00, §§ 177 and 189, ECHR 2007-IV). As regards the question of what constitutes prima facie evidence capable of shifting the burden of proof on to the respondent State, in proceedings before the Court, there are no procedural barriers to the admissibility of evidence or predetermined formulae for its assessment (see Volodina, cited above, § 112). Prima facie evidence of discrimination may include reports by non-governmental organisations or international bodies, as well as statistical data illustrating that the general attitude of the authorities has fostered a climate conducive to domestic violence. The Court has found violations of Article 14 where systemic failures to address violence against women were demonstrated by patterns of inadequate investigations in a series of cases (see Vieru, cited above, § 129, and A.E. v. Bulgaria, no. 53891/20, § 119, 23 May 2023), where the domestic authorities’ approach amounted to condonation of domestic violence (see Eremia v. the Republic of Moldova, no. 3564/11, § 89, 28 May 2013, and Talpis v. Italy, no. 41237/14, § 145, 2 March 2017) and in cases of documented deficiencies in the efforts to prevent and combat domestic violence (see Opuz, cited above, § 200, and Tkhelidze v. Georgia, no. 33056/17, §§ 56-57, 8 July 2021), or persistent legislative gaps in protection, reflecting the authorities’ reluctance to acknowledge the seriousness of the issue (see Tunikova and Others, cited above, § 129).

88. As regards statistical evidence, the Court has established that statistics which, upon critical examination, appear to be reliable and significant may be sufficient to constitute the prima facie evidence of indirect discrimination (see D.H. and Others, cited above, §§ 180 and 187-88). However, not all statistical disparities will trigger a presumption of discrimination. Fragmented, incomplete or inconclusive statistical data cannot suffice to establish a prima facie case that the effect of a measure or practice was discriminatory (see Hugh Jordan v. the United Kingdom, no. 24746/94, § 154, 4 May 2001, and Oršuš and Others v. Croatia [GC], no. 15766/03, § 152, ECHR 2010). Where statistical evidence was limited or not available, the Court has had regard, as evidence of indirect discrimination, to reports of institutional attitudes or practices documented by national and international authorities (see D.H. and Others, cited above, §§ 191-93; Oršuš and Others, cited above, §§ 153-154, and Opuz, cited above, §§ 193-97).

(b) Application to the present case

89. The applicant complained of structural bias, arguing that violence against women was treated differently in the legal system than violent crimes against male victims. She also complained of the disproportionate effect of applying the same approach to evidence in cases of sexual and domestic violence as in other types of crimes. Since the applicant did not allege individual discriminatory treatment, the Court must examine whether she has presented prima facie evidence of structural bias or disproportionate effect capable of shifting the burden of proof onto the Government.

90. It is undisputed that sexual and domestic violence in Iceland predominantly affects women; it is also the case in other Member States of the Council of Europe. The statistics presented by the applicant indicate that the vast majority of victims were women, while most perpetrators were men. However, the fact that women are disproportionately affected by sexual and domestic violence does not, in itself, establish that this reflects discriminatory policies or conduct by the authorities or the disproportionate effect of general measures rather than a broader societal issue.

91. The Court notes that Iceland ranks highly in international gender equality assessments and has implemented numerous reforms aimed at combating sexual and domestic violence, including the legislative reforms in 2016 introducing Article 218b and the creation of specialised investigation teams in 2018. While such general measures do not preclude the possibility of discrimination in practice, they provide relevant context for determining whether apparent disparities result from discriminatory intent or omission.

92. As regards the legal framework, domestic violence had not been comprehensively criminalised until the introduction of Article 218b in 2016 but this does not in itself demonstrate either discriminatory intent or effect. As noted above, the previous legal framework nonetheless provided various protections through provisions on sexual violence, assault, threats, and gross defamation of closely related persons, as well as enhanced sentencing for assaults committed within a close relationship. The 2016 reforms reflect the authorities’ commitment to strengthening protection against domestic violence rather than any indication of discriminatory complacency.

93. The Court further notes that the present case does not suggest that police officers or prosecutors attempted to dissuade the applicant from pursuing her complaint, implied that she was at fault, or displayed prejudicial attitudes towards female victims (see J.L. v. Italy, no. 5671/16, §§ 129‑30, 27 May 2021, and contrast with Eremia, cited above, §§ 86-89).

94. The applicant sought to demonstrate structural bias and disproportionate effect by referring to comparative data from the State Prosecutor’s annual reports, which showed that prosecution rates for sexual violence cases were lower than those for homicide and physical assault. She also cited statistics from the Stígamót centre for victims of sexual abuse, presenting prosecution and conviction rates in cases handled by the NGO. The Court considers that, while the statistical data referred to by the applicant may indicate differential outcomes, they must be interpreted with considerable caution.

95. The Court notes, first, that the domestic authorities do not make available the statistics on indictment and convictions in domestic and sexual violence cases (see also the limited data referred to by the Government in paragraph 84 above). While international monitoring bodies have identified a need for improvements in this regard (see paragraphs 34, 37 and 39 above), nothing in the present case indicates that the absence of comprehensive official statistics reflects deliberate indifference to a widespread societal problem (contrast with A.E. v. Bulgaria, cited above, § 120 in fine). Admittedly, incomplete data complicates the assessment of the applicant’s complaint of discrimination. Yet it cannot, on its own, establish either discriminatory bias or effect. It further entails that the fragmented statistics relied upon by the applicant may not offer a complete or fully accurate depiction of how such cases are handled by the authorities.

96. Turning, secondly, to the data from the State Prosecutor’s annual reports, the Court notes that the lower prosecution rates in sexual violence cases compared to those concerning homicide and physical assault may be explained by objective factors unrelated to discriminatory attitudes or effect. Homicide and physical assault cases cited by the applicant for comparison more frequently involve incidents in public settings with witnesses and/or forensic evidence. The evidential position in such cases is materially different from that in typical scenarios of sexual or domestic violence, where incidents usually occur in private settings without any witnesses present. Physical evidence may also be lacking, particularly where victims do not immediately report incidents or seek medical attention. The establishment of facts often turns on conflicting accounts between the victim and perpetrator. These inherent evidentiary challenges, common to sexual and domestic violence cases across all jurisdictions, may affect prosecution and conviction rates without necessarily reflecting discrimination. For that reason, comparing raw prosecution rates across different categories of offences, without accounting for the specific characteristics and circumstances of each case type, may be misleading. The statistical evidence comparing different types of crimes therefore lacks the requisite specificity to allow the Court to determine whether the observed disparity in prosecution rates is attributable to the inherent characteristics of the offences or to bias or disparate effect.

97. Thirdly, while the statistics referred to by the applicant show that women constitute the majority of victims of sexual and domestic violence, they do not establish that cases involving male victims of sexual assault or domestic violence are treated differently by the authorities or lead to different prosecution rates. In the absence of more detailed data illustrating how comparable evidentiary situations are addressed based on the gender of the victim, the Court cannot conclude that the statistics reflect a biased practice or stem from a gender-based disparity in the application of facially neutral measures.

98. The Court has previously found violations of Article 14 in cases where statistical evidence was part of a broader pattern demonstrating systematic failures by the authorities. In Opuz, official statistics regarding violence against women were combined with reports from Amnesty International and other organisations documenting community attitudes tolerant of violence against women, attitudes frequently shared by judges and senior officials. In A.E. v. Bulgaria, statistics were considered alongside documented patterns of inadequate investigations in multiple cases before the Court. In contrast, the present case presents limited statistical evidence without corresponding documentation of institutional attitudes or systematic patterns of discrimination.

99. Turning to the applicant’s remaining submissions in support of her prima facie case of bias or disproportionate effect, the Court reiterates that it has concluded above that, but for the delays, the investigation of her case was sufficiently thorough and led to the conviction of the perpetrator on one count. The Court further observes that the authorities have reformed the GPC in line with evolving international standards and implemented numerous other measures to protect victims of sexual and domestic violence, including the adoption of the Act on Restraining Orders and Removal from the Home, the implementation of specific investigation protocols, a prioritisation policy, risk assessment, specialised police and prosecutorial training, and coordination with health and social services. While the effectiveness of these measures may vary, they indicate institutional commitment to addressing sexual and domestic violence rather than systematic disregard for female victims.

100. As regards the applicant’s argument that applying the same standard of proof to cases of sexual or domestic violence as to other violent crimes fails to account for their specific nature, the Court notes that, although special investigative approaches and other measures tailored to the needs of victims of sexual and domestic violence may be desirable and have, in fact, been implemented by the domestic authorities with varying degrees of success (see paragraphs 36 and 83 above), the Convention does not require the application of a different standard of proof in such cases.

101. The applicant also argued that the authorities systematically disregarded evidence specific to sexual and domestic violence cases, such as psychological assessments and witness accounts of events before and after the incidents in question. Once again, however, the Court finds no indication that such evidence has been disregarded or not given adequate weight. The fact that such evidence has not always been considered sufficient or likely to secure a conviction, especially when balanced against defence rights and the presumption of innocence, does not indicate discrimination.

102. There remains the applicant’s contention that domestic violence investigations have been marked by systematic delays. While investigative delays and implementation issues may indicate institutional shortcomings requiring remedy, there is no indication in the present case that they result from gender bias or disproportionate effect rather than resource constraints affecting criminal investigations generally. In this respect, the Court also notes the priority policy adopted by the authorities and the allocation, following an influx of complaints in 2017-18, of additional resources to improve the handling of cases, which demonstrate recognition of, and commitment to, addressing any operational challenges (see paragraphs 36 and 51 above).

103. The Court acknowledges the particular challenges in prosecuting domestic violence cases and the importance of ensuring effective protection for victims. However, these challenges, which are common across many jurisdictions, cannot in themselves establish a prima facie case of discrimination absent evidence of prejudicial treatment or disproportionate effect. While raising concerns about data collection that warrant attention from the authorities, the limited statistical evidence adduced by the applicant is insufficient to establish prima facie evidence of structural bias or disproportionate effect. Furthermore, the observations of international monitoring bodies, while identifying areas for improvement, do not suggest the existence of institutional attitudes or practices indicating discriminatory treatment of female victims of domestic violence (compare and contrast D.H. and Others, cited above, § 192; Oršuš and Others, cited above, § 154, and Opuz, cited above, § 197). Taking also into account the various legislative and policy measures adopted by the authorities to combat sexual and domestic violence, stem impunity and protect victims, the Court finds that the applicant has not sufficiently established a prima facie case of structural bias or disproportionate effect, capable of shifting the burden of proof to the State.

104. There has accordingly been no violation of Article 14 of the Convention read in conjunction with Articles 3 and 8.

105. APPLICATION OF ARTICLE 41 OF THE CONVENTION

106. Article 41 of the Convention provides:

“If the Court finds that there has been a violation of the Convention or the Protocols thereto, and if the internal law of the High Contracting Party concerned allows only partial reparation to be made, the Court shall, if necessary, afford just satisfaction to the injured party.”

1. Damage

2. The applicant claimed 25,000 euros (EUR) in respect of non‑pecuniary damage, and also, in respect of pecuniary damage, ISK 434,450 (EUR 2,926) for therapy costs she had incurred following the time-barring of her case and ISK 448,520 (EUR 3,020) for private legal representation after losing her State-appointed lawyer when the domestic violence charges became time-barred.

3. The Government submitted that finding a violation should in itself constitute sufficient just satisfaction for any non-pecuniary damage sustained. In any event, they considered the amounts claimed excessive and inconsistent with the Court’s case-law in similar cases, maintaining they should be reduced significantly if the Court found that awarding such damages was warranted. Regarding the applicant’s pecuniary damage claims, the Government noted these were supported by invoices and submitted that it was for the Court to assess whether such damages should be awarded if a violation was found.

4. Regarding the pecuniary damage claimed, the Court notes that while the applicant submitted invoices for therapy costs and private legal representation, it is unable to establish a direct causal link between these expenses and the procedural violation found. The therapy costs appear to relate to treatment for the underlying trauma from the alleged assaults rather than specifically arising from the investigation’s procedural failures. Similarly, it cannot be established that additional legal costs were a direct and necessary consequence of the delays during the investigation. The Court therefore makes no award for pecuniary damage.

5. As regards non-pecuniary damage, the Court, ruling on an equitable basis, awards the applicant EUR 7,500 plus any tax that may be chargeable.

6. Costs and expenses

7. For costs and expenses incurred before the Court, the applicant claimed ISK 762,602 (EUR 5,135). That amount had been paid on her behalf by Stígamót, the centre for victims of sexual abuse. In line with her agreement with Stígamót, the applicant had declared that any costs awarded by the Court as just satisfaction due to the costs and expenses incurred by the proceedings before the Court would be reimbursed to Stígamót.

8. The Government, while noting that the applicant’s written agreement with Stígamót was not submitted (only her declaration to reimburse), accepted that the costs were not excessive and submitted that if the Court found a violation, it was for the Court to evaluate whether costs and expenses should be awarded based on the submitted documents.

9. The Court reiterates that, in accordance with Article 41 of the Convention, an applicant is entitled to the reimbursement of costs and expenses only to the extent that such costs have been actually incurred by the applicant and are reasonable as to quantum (see Merabishvili v. Georgia [GC], no. 72508/13, §§ 370-71, 28 November 2017). In the present case, as the expenses were borne by a third party and the applicant has not demonstrated the existence of a legal obligation to reimburse Stígamót unless and until an award in respect of costs is made, the Court finds that the applicant has not actually incurred these expenses. A mere declaration of willingness to reimburse, contingent upon an award of costs by the Court, is insufficient to establish that the applicant has personally borne the costs in question. Accordingly, the Court makes no award under this head.

FOR THESE REASONS, THE COURT, UNANIMOUSLY,

1. Declares the application admissible;
2. Holds that there has been a violation of Article 8 of the Convention under its procedural limb;
3. Holds that there has been no violation of Article 14 of the Convention, taken in conjunction with Articles 3 and 8;
4. Holds

(a) that the respondent State is to pay the applicant, within three months from the date on which the judgment becomes final in accordance with Article 44 § 2 of the Convention, EUR 7,500 (seven thousand five hundred euros), plus any tax that may be chargeable, in respect of non‑pecuniary damage, to be converted into the currency of the respondent State at the rate applicable at the date of settlement:

(b) that from the expiry of the above-mentioned three months until settlement simple interest shall be payable on the above amounts at a rate equal to the marginal lending rate of the European Central Bank during the default period plus three percentage points;

5. Dismisses the remainder of the applicant’s claim for just satisfaction.

Done in English, and notified in writing on 26 August 2025, pursuant to Rule 77 §§ 2 and 3 of the Rules of Court.

Hasan Bakırcı Arnfinn Bårdsen
Registrar President