A Vote Against Making Things Worse

On 28 January 2026, the Parliamentary Assembly of the Council of Europe (PACE) unanimously rejected a draft Additional Protocol to the Convention on Human Rights and Biomedicine (Oviedo Convention). The draft is supposed to create legally binding rules governing involuntary psychiatric placement and treatment, including seclusion and restraint. While many Member States of the Council of Europe (CoE) already have legal safeguards protecting persons subject to involuntary measures in place, others do not, and the Protocol seeks to fill those gaps. After more than a decade of drafting and four prior recommendations against it (2016, 2019, 2022, 2024), the Protocol was ultimately rejected by PACE.

Civil society organisations welcomed the vote enthusiastically, with some even calling it a landmark decision. This blogpost argues that, while understandable, that characterisation is premature. The vote does not resolve the existing fundamental tension between the CRPD’s abolitionist approach and the European human rights system’s regulatory approach to involuntary psychiatric measures. To understand why the vote might not (yet) be a turning point, it is helpful to look at the gap between the two human rights regimes.

Abolishing Coercion: The CRPD’s Approach

The UN Convention on the Rights of Persons with Disabilities (CRPD), as interpreted by its Committee, prohibits involuntary admission and treatment outright, regardless of safeguards, and regardless of whether it is a measure of last resort. Article 14 (1) (b) of the CRPD provides that the existence of a disability shall in no case justify a deprivation of liberty. The CRPD Committee’s Guidelines on Article 14 clarified this includes psychosocial disabilities such as mental illnesses, and that legislation authorising admission based on actual or perceived impairment remains incompatible with the CRPD even when combined with additional criteria such as dangerousness or therapeutic necessity (paras. 6, 13). Notably, during the CRPD’s negotiations, a proposal to insert a qualifier like “solely” or “exclusively” into Article 14 (1) (b) CRPD was deliberately rejected to prevent impairment from being reintroduced as one factor among several (Guidelines on Article 14, para. 7). Involuntary commitment on the basis of disability is classified as inherently discriminatory and thus constitutes an arbitrary deprivation of liberty (Guidelines on Article 14, para. 6). According to the CRPD Committee, it strips the person of the legal capacity to decide about their own admission, implicating Article 12 CRPD (equal recognition before the law) and the requirement of free and informed consent under Article 25 (d) (Guidelines on Article 14, para. 10). Involuntary treatment is likewise considered inadmissible under Articles 12 and 25 (d) CRPD, and may also violate the right to personal integrity (Article 17 CRPD), freedom from torture (Article 15 CRPD) and freedom from violence, exploitation and abuse (Article 16 CRPD) (General Comment No. 1, para. 42).

Regulating Coercion: The European Model

The European Court of Human Rights (ECtHR) and the Oviedo Convention take the opposite starting point: involuntary admission and treatment are, in principle, legitimate, provided certain safeguards are met.

Article 5 (1) (e) of the European Convention on Human Rights (ECHR) permits the detention of persons of “unsound mind”, and the Court has confirmed this authorises compulsory confinement as a preventive security measure rather than a punitive one (Denis and Irvine v. Belgium, paras. 141, 168). The Court operationalises this through the Winterwerp criteria: a competent national authority must establish the existence of a true mental disorder based on objective medical expertise, the mental disorder must be severe enough to warrant compulsory confinement, and continued confinement must be justified by the disorder’s persistence (Winterwerp v. the Netherlands, para. 39). In contrast to the CRPD, the ECtHR thus refers to the severity of a mental disorder as a justification for the deprivation of liberty (cf. Alexandrov/Schuck, p. 3). More recent case law added the requirement that confinement must serve a genuine therapeutic function, with treatment going beyond basic care (Rooman v. Belgium, paras. 208, 209).

The requirement of a therapeutic purpose highlights how intertwined involuntary admission and treatment are for the ECtHR. The Court has considered treatment administered within lawful hospitalisation as generally justified by extension (Fiala-Butora, p. 7), mainly subjecting it to scrutiny under Articles 3 and 8 ECHR. Under Article 3 ECHR, however, forced treatment is not seen as inherently inhuman or degrading as long as there is convincing proof of genuine medical necessity and adequate procedural safeguards (ECtHR, para. 41; Jalloh v. Germany, para. 69). Under Article 8 ECHR, compulsory psychiatric treatment interferes with private life and the right to refuse psychiatric medication (Guide on Article 8, para. 157), but this interference can be justified if it is in accordance with the law, pursues a legitimate aim, and is necessary in a democratic society, as required by Article 8 (2). Physical restraint and seclusion are tolerated only as measures of last resort to prevent imminent harm (ECtHR, para. 32).

The Oviedo Convention, ratified by 31 states, draws on the principles of the ECHR and specifies them as regards the field of biology and medicine. Article 6 (3) Oviedo Convention allows interventions on adults lacking capacity with authorisation from a representative, with the person participating “as much as possible”. Article 7 permits treatment without consent for persons with a serious mental disorder where necessary to prevent serious harm to their health.

A Missed Case for Convergence

Given the profound structural differences between the CRPD and the ECHR, it is worth asking whether the European human rights system has even engaged with the CRPD and to what extent the CRPD should be considered in case of collision.

It can persuasively be argued that the CRPD, ratified by every ECHR Member State, should take interpretive priority where the two regimes contradict each other. The CRPD is a “relevant rule of international law” to be taken into account under Article 31 (3) VCLT and, as the later treaty on the same subject matter, should prevail according to Article 30 (3) VCLT (cf. Fiala-Butora, p. 9). Article 53 ECHR reinforced this priority by prohibiting interpretations that limit rights secured under other treaties (cf. Fiala-Butora, p. 9).

The ECtHR, however, has not embraced this approach but rather engaged with the CRPD inconsistently (in depth, Fiala-Butora), initially treating it as a relevant disability-rights instrument without always clarifying what legal effect that reference carries. Yet in its Article 5 jurisprudence, the Court relies solely on its own Winterwerp criteria and accepts confinement as long as these are met, without really engaging with the standards of the CRPD (Fiala-Butora, p. 5). Most tellingly, in Rooman v. Belgium, the Court stated expressively that it does not currently wish to interpret Article 5 ECHR in line with the CRPD Committee’s interpretation of Article 14 (para. 205).

This reluctance is not confined to the ECtHR. The Council of Europe’s Steering Committee on Bioethics concluded that the CRPD should be interpreted as meaning that the existence of a disability may not in itself justify a deprivation of liberty or an involuntary treatment, but that such measures can be justified where the absence of intervention would likely cause serious harm to the person or a third party. While this formula might sound like a concession to the CRPD, “in itself” ultimately carries much the same meaning as the qualifiers “solely” or “exclusively” deliberately excluded from the CRPD, as noted above. Reading it back into the provision does not bridge the gap between the two regimes but rather reproduces an interpretation the CRPD Committee has already rejected under a different label.

The Draft Additional Protocol

The draft Additional Protocol must be read against this backdrop. It was framed as a harmonisation project rather than an expansion of state power. Its alleged aim was to bring Member States that still lack adequate safeguards up to the level already achieved by ECtHR case law, not to authorise anything new. While the ECHR and the Oviedo Convention already permit involuntary admission and treatment under mental health laws, the draft Additional Protocol seeks to develop and concretise these provisions in a codified form. To that end, draft Article 11 sets out criteria for involuntary placement and treatment. It explicitly permits such measures where a person’s current mental health condition poses a significant risk of serious harm either to their own health, and their ability to decide on the measure is severely impaired, or a significant risk of harm to others, provided the measure serves a therapeutic purpose and no voluntary alternative is sufficient. Draft Article 12 sets out a standard procedure for deciding on such measures, and draft Article 17 permits seclusion and restraint subject to safeguards.

Although this largely reflects European practice, the Additional Protocol would consolidate coercive interventions in a more elaborate regulatory framework, thereby further widening the gap between this approach and the CRPD’s position. Case law can evolve incrementally, be distinguished, or be reversed by a differently composed chamber, so the door to convergence with the CRPD, however narrow, remains open. A ratified treaty is far more entrenched and would lock in the current divergence, giving Member States, including the very states the Protocol targeted for not yet having safeguards, a new, authoritative anchor point around which to align domestic legislation.

The CRPD Committee, in an open letter, also classified the draft Additional Protocol as inconsistent with the CRPD and warned State parties to the CRPD against becoming party to the Additional Protocol since this would breach their general obligation under Article 4 (1) (f) CRPD to refrain from acts incompatible with the Convention. The Committee was explicit that ratifying states may not use the Additional Protocol as a means of derogation from CRPD obligations.

Adopting the Protocol would not just preserve the existing European discrepancy with the CRPD but deepen it and make it harder to reverse.

Concluding Thoughts

Seen in this light, PACE’s vote is best understood as an important signal, but not a resolution to the underlying tension between the CRPD and the ECHR. Its rejection is a recommendation, not a decision, as the Committee of Ministers is the body that ultimately decides whether to adopt the Additional Protocol and may proceed regardless of PACE’s opinion. Even a final rejection by the Committee of Ministers would only preserve the status quo, not improve it. It would not undo the underlying case law and domestic legislation which both already implement coercive measures, nor would it close the legal gaps the Protocol was meant to address. Judged against that bar, PACE’s decision on 28 January 2026 was a vote against making things worse, not necessarily a sign that things are getting better.

The Committee of Ministers has not stood entirely still since, though its steps point in different directions, leaving it unclear how it intends to respond to PACE’s rejection. On 17 June 2026, it adopted a Recommendation on respect for autonomy in mental healthcare, reaffirming free and informed consent as the rule in mental healthcare, a step toward the CRPD even though it still permits capacity-based exceptions inconsistent with the CRPD.  Yet a classified study commissioned by its Secretariat for the Ministers’ Deputies’ discussions on the draft Additional Protocol struck a different tone again, reportedly concluding that the Protocol does not run counter to the CRPD. Which of these directions will ultimately prevail remains to be seen.

The “Bofaxe” series appears as part of a collaboration between the IFHV and Völkerrechtsblog.