Poland's constitutional court vs EU law on same-sex marriage
- Authority
- Polish Constitutional Tribunal
- Rule type
- constitutional court ruling
- Jurisdiction scope
- EU
- Effective date
- Jul 28, 2026
- Source text
- Read primary rule text ↗
EU law requires recognition of foreign same-sex marriages for free movement purposes despite TK ruling.
The practical problem is not a single ruling. It is a collision between a July 28, 2026 judgment of Poland’s Constitutional Tribunal in case U 2/26 and two European Court of Justice judgments from late 2025 that were already reshaping the legal position before Warsaw’s latest constitutional move.
In U 2/26, the Constitutional Tribunal unanimously held that the government’s May 22 regulation on transcribing foreign same-sex marriages was inconsistent with Article 92(1) in conjunction with Article 18 of the Polish Constitution, the provision defining marriage as “a union of a man and a woman.” The panel was composed of Bogdan Święczkowski as presiding judge, Stanisław Piotrowicz as rapporteur, and Jarosław Sych.[1] For a Polish official who treats the Tribunal as a binding constitutional court, that is not background noise. It is the direct answer to whether a ministerial regulation can be used to move foreign same-sex marriages into the Polish civil-status registry.
For an EU-law adviser, that is only half the answer. On December 18, 2025, the Court of Justice ruled in Commission v. Poland, C-448/23, that the Polish Constitutional Tribunal “does not meet the requirements of an independent and impartial tribunal established by law,” because of the unlawful appointment of judges during the PiS period.[2] The Tusk government had already refused to publish or implement Tribunal rulings it regards as illegitimate. That refusal has not yet been formally tested against this specific July 28 ruling, which is why the safe practitioner answer is neither “the Tribunal settled it” nor “the ruling can be ignored.”

The collision is real because the authorities answer different questions
U 2/26 addresses the domestic constitutional status of a ministerial regulation. Its strongest point is formal hierarchy: a regulation issued at ministerial level cannot rewrite constitutional or statutory definitions of marriage. That objection had already appeared in public criticism before the judgment.[3] The Tribunal’s July ruling gives that objection the form of a constitutional judgment, at least for institutions that accept the Tribunal’s authority.
The EU-law line answers a different question: what legal effects must a member state give to a same-sex marriage validly concluded elsewhere in the Union when EU free movement rights are engaged. The key 2025 development was Trojan, C-713/23, decided on November 25, 2025. Commentary on the judgment reads it as going beyond the earlier Coman residence-rights logic by requiring full legal effects of marriage for free movement purposes, including matters such as health insurance, name rights, and third-party status, under Article 21(1) TFEU and Articles 7 and 21(1) of the EU Charter.[4]
That distinction is not academic. A registry clerk asked to transcribe a certificate, a social-security officer asked to attach spousal coverage, and an employer deciding whether to process a benefit enrollment are not necessarily applying the same legal instrument. Transcription is a civil-status act. Recognition for a derived EU-law right may be required even where the domestic marriage-law category remains contested.
A workable compliance map starts with forum, not slogans
The present risk map has two tracks. They overlap in the same office inbox, but they do not have the same source of authority.
| Question | Polish constitutional-law track | EU-law track |
|---|---|---|
| Can the May 22 ministerial regulation be treated as constitutionally valid? | U 2/26 says no for actors who accept the Constitutional Tribunal as binding. | EU law does not depend on the validity of that specific Polish regulation. |
| Must a foreign same-sex marriage have legal effects in a cross-border EU situation? | Article 18 remains the domestic constitutional obstacle relied on by the Tribunal. | Trojan supports recognition of full legal effects where free movement rights are engaged. |
| Can Polish agencies safely refuse all effects after U 2/26? | Some officials may view refusal as constitutionally required. | Blanket refusal creates EU-law exposure, especially for derived social and personal-status effects. |
| Is the government’s nonrecognition of the Tribunal a complete shield? | Not for institutions or officials that still treat Tribunal rulings as authoritative. | The CJEU’s independence ruling strengthens the government’s position but does not automatically settle every domestic implementation step. |
The unpleasant part for compliance teams is that both columns can matter on the same file. A multinational employer may want one answer for payroll and benefits; the employee may need another answer from ZUS, NFZ, the tax office, and the civil registry. The legal consequence of a marriage certificate is therefore being fragmented by institution.

What Trojan changes, and what it does not
Trojan matters because it moves the discussion away from the narrowest version of recognition. After Coman, many administrations could treat the issue as primarily about residence rights for a spouse. Trojan is more difficult to cabin. Its significance lies in the idea that free movement is impaired if a couple’s marital status loses practical legal effects when they cross an internal EU border. The effects identified in commentary include health insurance, name rights, and third-party status.[4]
That does not mean every Polish domestic route is now clear. EU law can require recognition of effects without requiring Poland to introduce same-sex marriage in its domestic family law. It can also require an agency to disregard a national obstacle in a particular EU-law setting while leaving another office fighting over the registry consequences. The distinction is irritating because it is operationally messy, but it is the distinction on which many files will turn.
The Polish administrative courts had already shown how narrow this hinge can be. On March 20, 2026, the Supreme Administrative Court, in II OSK 216/21, held through Judge Kirnaszek that Article 18 “does not prohibit recognizing marriages concluded in other EU countries.”[5] That is not the same as a blanket domestic marriage-equality ruling. It is a recognition route tied to foreign marriages and, in practice, to cross-border legal effects.
Where the ruling lands first: registries, ZUS, NFZ, tax, and employers
The civil registry is the most exposed point because U 2/26 is aimed at transcription. A registrar who accepts the Constitutional Tribunal’s authority can say the ministerial regulation has been struck down. A registrar whose institution follows the government’s nonrecognition line may be instructed to continue processing. That is not a comfortable place for a municipal official: the legal exposure depends less on the certificate in front of them than on which institutional hierarchy their office treats as real.
ZUS and NFZ sit in a different posture. Before the July 28 ruling, expert commentary reported that the social-security and healthcare authorities had announced they would recognize transcribed same-sex marriages, while some legal scholars warned that those announcements could be unlawful if the Constitutional Tribunal invalidated the underlying regulation.[6] After U 2/26, the same tension becomes sharper. A benefits officer may see an EU-law obligation to recognize spousal effects for free movement purposes, while a constitutional-law actor may say the transcription mechanism has lost its domestic legal basis.
Health insurance is the clearest practical category because Trojan-style reasoning maps directly onto the loss of spousal coverage when a couple moves. Name rights and third-party status also have identifiable legal effects: hospital access, administrative representation, and private-sector forms that ask whether a person is a spouse or a stranger. These are not symbolic incidents of status; they are the points at which a clerk either opens the file or rejects the document.
Tax treatment is less settled on the available materials. The research record supports saying that it remains unresolved; it does not support saying that joint tax treatment, deductions, or succession-adjacent consequences must already be processed in a uniform way. For tax advisers, the honest memo is likely to separate EU-derived recognition arguments from any representation that the Polish tax administration has adopted a stable post-U 2/26 position.
Employers should not confuse their own benefits administration with civil-status adjudication. If a Polish payroll team receives a foreign marriage certificate from an employee who has exercised EU free movement rights, refusing every spousal benefit because of U 2/26 may create a different risk profile from declining to opine on registry transcription. The harder cases will be those where an employer benefit depends on a state agency record that may now be contested.
The resistance track is organized too
Officials who refuse transcription will not necessarily be acting alone. Ordo Iuris has pledged free legal assistance to officials who resist transcription on the basis of the Constitutional Tribunal’s ruling.[3] That matters for risk assessment because it turns individual refusal into a supported litigation posture. It also makes uneven implementation more likely: two offices may receive similar documents and make opposite decisions, both with legal arguments ready to hand.
The government’s nonrecognition strategy cuts the other way. The CJEU’s December 2025 finding on the Tribunal’s lack of independence gives Warsaw and EU-law actors a serious basis for treating the Tribunal’s judgments as defective under EU rule-of-law standards.[2] But that is not the same as an administrative instruction that has already been absorbed by every local registry, benefits office, tax desk, and court that might later review an official’s decision.
Context matters, but it should not be allowed to do the legal work
Poland remains an outlier in the EU family-law landscape. In November 2025 reporting, DW described Poland as one of only four EU member states without same-sex marriage or civil unions.[7] The number is relevant because it explains why cross-border recognition does so much work in Poland: for many couples, the foreign certificate is the only formal marital document they have.
The domestic political route has also narrowed. On July 17, 2026, Poland’s president vetoed a civil-partnership bill, leaving the recognition fight to be carried through courts, agencies, and ministerial instruments rather than a settled legislative compromise.[8] That does not prove how U 2/26 will be implemented, but it explains why offices are being asked to solve a problem the legislature has not resolved.
The scale is significant but not precisely established. Courthouse News Service, citing AFP and rights organizations, reported an estimate of 30,000 to 40,000 same-sex marriages by Polish citizens abroad.[9] That figure should be treated as an advocacy-sector estimate reported through media, not as official registry data. It is enough to show that the issue is not hypothetical; it is not enough to calculate exposure with actuarial confidence.
The current practitioner position
A cautious legal-risk position now has to separate four questions before giving an answer.
- Is the decision about registry transcription, or about a derived effect such as healthcare, social security, name use, or third-party status?
- Has the person or couple exercised EU free movement rights, making Article 21 TFEU and Charter-based recognition arguments central?
- Does the institution handling the file follow the government’s nonrecognition of the Constitutional Tribunal, or does it still treat TK rulings as binding?
- Is there a published agency position after July 28, 2026, or only pre-ruling statements and expert commentary?
The answer will be strongest where the request concerns a concrete EU-law effect of a foreign marriage in a cross-border setting. It will be weaker where the request asks a Polish registry to transcribe the marriage solely through the May 22 ministerial regulation, because U 2/26 directly targets that instrument. It will be most uncertain where the file depends on tax treatment or on a local office whose post-ruling instructions are not yet public.
The ruling therefore cannot be safely ignored by anyone operating inside Polish institutions that treat the Constitutional Tribunal as binding. It also does not erase EU-law obligations for cross-border recognition and derived rights. The immediate legal position is dual-track, institution-sensitive, and still moving.
Last verified: July 31, 2026, UTC. Risk status: Regulation & Ethics. The position may shift as Polish agencies, Polish courts, the Tusk government, or EU institutions respond specifically to U 2/26.
References
- Wyrok w sprawie U 2/26, Trybunał Konstytucyjny, July 28, 2026.
- Polish Constitutional Tribunal violated European law, rules EU court, Notes from Poland, December 18, 2025.
- Poland’s government recognition of foreign same-sex marriages at ministerial level, Brussels Signal, June 5, 2026.
- Trojan: Towards Full Recognition of Same-Sex Marriages in Europe, Verfassungsblog, December 22, 2025.
- The Curious Life of Article 18, I·CONnect, April 10, 2026.
- What will Poland’s recognition of foreign same-sex marriages mean in practice? Five expert views, Notes from Poland, June 19, 2026.
- Poland: Same-sex couples hope for equal rights, DW, November 2025.
- Polish president vetoes civil partnership bill, Notes from Poland, July 17, 2026.
- Poland court strikes down recognition of same-sex marriages performed abroad, Courthouse News Service.
Operationalizing workflow
No workflow has been explicitly linked to this obligation yet. See Workflows generally.
Illustrative cases
No illustrative case is currently tracked for this obligation. See Risk Digest for documented incidents generally.
← Back to RegulationReport a correction or tip
Spotted an outdated figure, a misstated fact, or a ruling this regulation entry should reflect? Public comments are disabled for this content given the professional cost of a misreported case outcome, penalty amount, or rule text — use the structured correction channel instead.
Report a correction or tip for this record →