Domestic Violence in International Protection Claims: Death Threats Treated Within the Domestic Violence Spectrum and “Women in Georgia” Not Automatically a Particular Social Group

Case: N.K. v The International Appeals Tribunal and Anor (Approved)
Citation: [2026] IEHC 116 (High Court of Ireland, Ferriter J.)
Date: 11 March 2026

1) Introduction

This judicial review challenged a decision of the International Protection Appeals Tribunal (“the Tribunal”) dated 13 March 2024 affirming an International Protection Office (“IPO”) recommendation that N.K., a Georgian national, be refused both a refugee declaration and subsidiary protection.

The applicant’s protection claim arose from serious domestic violence by her father, including an alleged escalation to threats to kill her and a severe assault in March 2023. The core legal issues before the High Court were:

• whether the Tribunal unlawfully narrowed/mischaracterised the claim by treating it as fear of “assault” rather than fear of being killed (including “femicide” framing); and
• whether the Tribunal irrationally rejected a Refugee Convention nexus by finding no “membership of a particular social group” under ss. 2 and 8 of the International Protection Act 2015.

2) Summary of the Judgment

The High Court refused relief and upheld the Tribunal’s decision.

Ferriter J. held that:

• The Tribunal did not commit a material error by not explicitly repeating “death threats” in its formulation of the risk; its analysis treated domestic violence as a broad spectrum capable of encompassing threats to kill and potentially fatal assaults, and it considered COI that included gender-based killings of women.
• The Tribunal’s finding that the applicant was not a member of a “particular social group” was not irrational: on the Tribunal’s factual findings, the risk arose from her father’s general violence (also directed at male persons), not because she was a woman in Georgia; and alternative formulations (“victims of domestic violence”, “daughters of violent men”) were considered not to exist independently of the feared persecution or were not supported as socially distinct by the COI.
• Two additional challenges (on State protection and COI contradictions) were not pursued orally and, in any event, failed on the reasoning of the Tribunal decision.

3) Analysis

A) Precedents Cited

i) Case C‑756/2, X v IPAT

Ferriter J. treated this CJEU authority as establishing the decision-maker’s duty to assess protection claims “with vigilance and care”, because such claims engage “integrity of the person”, “individual liberties”, and fundamental EU values. The judgment uses Case C‑756/2, X v IPAT to:

• confirm a high standard of careful, individualised assessment; and
• frame the applicant’s argument that the Tribunal allegedly failed to engage with the “real” claim (risk of being killed).

The High Court accepted the legal standard, but found that (on the facts) the Tribunal met it: the Tribunal addressed the feared serious violence within its domestic-violence analysis and consulted COI expressly referencing lethal outcomes.

ii) BC v IPAT [2019] IEHC 763

This decision was relied upon in the Tribunal decision (and noted by Ferriter J.) for summarising the test for effective State protection under s. 31 of the 2015 Act. While the applicant did not press the State-protection grounds at the hearing, BC v IPAT [2019] IEHC 763 still matters in the architecture of the Tribunal’s reasoning:

• it anchors the legal test the Tribunal purported to apply; and
• it reinforces that State protection must be examined with focus on the applicant’s “particular circumstances”.

Ferriter J. did not disturb the Tribunal’s approach, noting the Tribunal ultimately found “effective” (not merely “functioning”) protection in its conclusion.

iii) BA v IPAT [2020] IEHC 589

This authority was pivotal to rejecting the applicant’s “death threats versus domestic violence” distinction. Ferriter J. approved Burns J.’s analysis in BA v IPAT [2020] IEHC 589 that domestic violence in a relationship setting spans a continuum “from minor assaults to vicious fatal attacks”, so the proper inquiry is whether State protection is available against “all serious physical attacks” rather than artificially isolating “threats to kill”.

The High Court extended that logic to the present facts: parsing “credible death threats” away from serious assault within a domestic violence pattern was characterised as an inappropriate dissection of the essence of serious domestic violence, which may escalate to potentially fatal violence.

B) Legal Reasoning

i) Characterisation of the claim: “death threats” within the domestic violence risk

The applicant’s primary submission was that the Tribunal focused on “assault” and failed to engage with her expressed fear of being killed if returned to Georgia. Ferriter J. rejected that challenge for three connected reasons:

• Substance over phrasing: although the Tribunal used language such as “further violence” and “assault”, it accepted serious past violence and assessed a real risk of serious harm in the form of violent abuse by the father. The court treated that as capturing the essential risk.
• COI addressed lethal outcomes: the Tribunal’s COI review included references to “gender-based murder of women” and discussion of systemic shortcomings and improvements. This supported the conclusion that the Tribunal’s State-protection analysis was aimed at the correct “target”.
• Doctrinal coherence (via BA v IPAT): the court endorsed the idea that threats to kill sit within the domestic-violence spectrum and should not be separated in a way that would distort the assessment of protection.

In effect, the judgment establishes a pragmatic approach: a Tribunal decision will not be quashed merely because it does not explicitly recite “death threats”, provided the decision engages with the applicant’s feared harm in substance and uses relevant COI encompassing the severity alleged.

ii) Convention nexus and “membership of a particular social group”

The applicant argued (a) that “women in Georgia” should qualify as a particular social group given disproportionate domestic violence; and (b) that the family unit should be treated as a particular social group.

Ferriter J. upheld the Tribunal’s approach on irrationality review:

• “Women in Georgia” too broad on these facts: the Tribunal reasoned that the applicant’s fear did not arise because she was a woman but because her father was violent generally (including to a son and a male friend). On that factual footing, it was open to the Tribunal to find no causal nexus between feared persecution and gender, and thus no s. 8 reason.
• Non-circularity and social distinction: the Tribunal’s rejection of groups like “victims of domestic violence” or “children/daughters of violent men” rested on orthodox concerns reflected in s. 8 analysis: a group should not be defined solely by the persecution feared, and COI must show the group has a distinct identity perceived by surrounding society. The High Court found that conclusion not irrational on the record.
• New “family as PSG” argument not properly raised: the court refused to entertain a complaint that the Tribunal failed to identify “family” as a social group when the applicant had not advanced that case before the IPO or in the Tribunal appeal grounds. The court was sceptical that a cooperative duty could require the State to construct an unpleaded Convention nexus.

The judgment therefore reinforces two practical propositions in Irish judicial review of protection decisions: (1) the Convention nexus inquiry is fact-sensitive and may fail where the persecutor’s violence is found to be indiscriminate rather than gender-targeted; and (2) applicants must articulate their proposed “particular social group” theory in the protection process, not for the first time on judicial review.

C) Impact

The likely impacts of N.K. are these:

• Framing of risk in domestic violence claims: tribunals may describe the feared harm as “assault” or “violence” without separately itemising “death threats”, where the overall analysis (including COI) is broad enough to cover escalation to lethal harm. Challenges based on linguistic omission alone may be harder unless the omission changes the legal test applied or the evidential analysis undertaken.
• PSG litigation strategy: applicants will likely need to plead and evidence (i) why the harm arises “by reason of” membership of the proposed group and (ii) social distinction (where relevant), particularly if relying on formulations other than “women” (e.g., “victims of domestic violence”).
• Judicial review discipline: the decision exemplifies the High Court’s reluctance to substitute its own evaluative judgment for the Tribunal’s where the challenge is framed as “irrationality”, especially on mixed questions of fact, COI assessment, and nexus characterisation.

4) Complex Concepts Simplified

5) Conclusion

N.K. v The International Appeals Tribunal and Anor confirms that, in domestic-violence-based protection claims, a Tribunal’s failure to explicitly restate “death threats” will not necessarily invalidate its decision where the Tribunal substantively assesses the feared serious violence and relies on COI encompassing lethal domestic violence. It also underscores that “women in Georgia” will not automatically provide a “particular social group” nexus where the Tribunal, on the evidence, finds the risk stems from a perpetrator’s general violence rather than gender-targeted persecution, and that unpleaded PSG theories (such as “family” as a social group) are unlikely to succeed if raised only on judicial review.