After Semenya

The Grand Chamber’s decision in Semenya v. Switzerland marks a pivotal development in the European Court of Human Rights’ (ECtHR) evolving engagement with international arbitration. The case concerned Caster Semenya, a two-time Olympic champion in the women’s 800 metres, who challenged World Athletics’ Regulations on Athletes with Differences of Sex Development before the Court of Arbitration for Sport (CAS). These rules require athletes with certain endogenous testosterone levels to undergo medical interventions in order to compete in female categories at international events. CAS upheld the regulations, and the Swiss Federal Supreme Court (‘FSC’), applying the grounds for annulment under Article 190(2)(e) of the Swiss Private International Law Act (‘PILA’), declined to overturn the award. Caster Semenya then applied to the ECtHR, arguing that the restricted scope of Swiss judicial review violated her rights under Article 6(1) and 14 of the European Convention on Human Rights (‘ECHR’).

While the decision ostensibly addressed a discrete procedural defect – the inadequate judicial oversight under Swiss law –, it importantly criticizes the assumptions that underpin private dispute resolution where it is deployed in rights-sensitive domains. The central questions were firstly, whether Swiss law allows a reviewing court to engage meaningfully with the substance of arbitral decisions that determine an individual’s ability to work, and secondly, to give real effect to the procedural guarantees of  Article 6(1) ECHR. The claim prevailed not because the arbitral award itself was held invalid, but because the FSC, constrained by the narrow grounds of review permitted under Article 190(2)(e) PILA, declined to examine whether the arbitral proceeding had given effect to the ECHR’s procedural guarantees (para 226-230).

The Court’s insistence on a “particularly rigorous examination” (para 216) where arbitration is effectively mandatory, as in the field of professional sport, reflects an understanding that institutional power asymmetries cannot be neutralised by formal access to an arbitral forum. This article argues that the decision underscores the need for a reform of Swiss arbitration law. It contends that the current approach under PILA, which treats all forms of arbitration equally, fails to account for the asymmetries and human rights implications of compulsory arbitral regimes. These concerns are not confined to sports law; they arise in other rights-heavy domains such as employment disputes involving statutory protections, consumer contracts with embedded arbitration clauses, and other professional disciplinary proceedings, where the arbitral process functions as a compulsory surrogate for public adjudication. The judgment does not call for the abandonment of Switzerland’s arbitration-friendly posture, but it does require that national law takes into account situations where private adjudication functions as a substitute for public law without its safeguards. This article proceeds first by examining Semenya in the context of other jurisprudence. Second, it interrogates the limitations of Article 190(2)(e) PILA. Finally, it outlines legislative paths towards a more differentiated and rights-sensitive arbitration regime.

Semenya and the Jurisprudence on Compulsory Arbitration

The Grand Chamber’s reasoning in Semenya continues the Court’s recent efforts to delineate the boundaries in cases where recourse to arbitration is effectively compulsory. In Mutu and Pechstein v. Switzerland, the Court held that proceedings before CAS engaged Article 6(1) ECHR and required guarantees of a fair hearing, notwithstanding CAS’s private character and its foundation in contractual consent (paras 96-101). There, the Court explicitly acknowledged that “the requirement of accepting the CAS jurisdiction was imposed on the applicants by their respective sports federations as a condition for their participation in competitions” (para. 98), thereby placing the issue of consent at the centre of its legal analysis. In Ali Riza v Turkey, which involved the Turkish Football Federation’s internal arbitration system, the Court held that where an arbitral tribunal is effectively imposed on one party of the contract by the other, its procedures must conform to the standards of independence and impartiality required under Article 6(1) (paras. 180-81). It also stressed that the lack of such guarantees could not be cured by the mere possibility of subsequent judicial review. In doing so, Ali Riza broadened the principle from the Mutu context of international sports arbitration to domestic arbitral systems, thereby effectively arguing that Article 6(1) safeguards apply whenever arbitration is structurally non-voluntary. This jurisprudence recognises that when arbitration ceases to be voluntary in substance, it assumes a quasi-public character, and the procedural guarantees of the Convention must follow. In Semenya, the ECtHR extended this logic to the review mechanisms that accompany mandatory arbitration.

Before the CAS, Caster Semenya argued that the above-mentioned regulations discriminated against female athletes, particularly those with differences of sex development, on the basis of innate physical, genetic, biological, sex, gender, and appearance-related characteristics, and that they were unnecessary, unreasonable, and disproportionate to the aim of ensuring fair competition in women’s events (para 23). In its award, the CAS panel acknowledged that the regulations were discriminatory, but held that they were a proportionate and necessary means of achieving what it considered the legitimate aim of protecting the integrity of the female category in elite sport (CAS Award, paras 456-463, 583-584). Semenya appealed to the FSC, the sole court competent to review CAS awards. Her arguments focused on Articles 6(1) and 14 ECHR, contending that the CAS award violated her rights to a fair hearing and non-discrimination.

Unlike earlier cases, the claim was not directed against the arbitral tribunal itself but the domestic court that reviewed its award. The FSC had refused to assess the award’s compatibility with the ECHR on the basis that its jurisdiction was limited by Article 190(2)(e) PILA. That provision restricts annulment to cases of manifest incompatibility with public policy, a standard that Swiss courts have construed narrowly. It is defined, in the FSC’s own terms, as conduct that is manifestly incompatible with the fundamental legal order (see para 4.1), a threshold that excludes mere error, even where that error relates to the disregard of international human rights law. As a result, despite the fact that CAS had adjudicated a matter of profound professional and medical consequence for the applicant, the FSC confined itself to a formalistic audit of the process rather than engaging with underlying human rights issues.

The Grand Chamber’s insistence that the review must be a “particularly rigorous examination” (para 216) in such contexts reflects the deepening concern about the procedural legitimacy of arbitration that functions as a surrogate for public adjudication. It recognises that compulsory arbitration in professional sport, far from exemplifying mutual consent, operates within hierarchies of obligation and dependency. When access to a livelihood is conditioned on acceptance of a single forum whose governing structures are dominated by sports bodies, and whose awards receive only minimal state supervision, the standard of review cannot be grounded in assumptions of contractual parity.

Semenya thus raises the question to what degree national courts must scrutinise arbitral awards when fundamental rights are at stake. In doing so, it opens up space for a more substantive model of judicial supervision, calibrated to the real-world effects of arbitration in structurally imbalanced regimes.

Revisiting Article 190 PILA: Toward Rights-Sensitive Judicial Review

The notion of “public policy” under Article 190(2)(e) PILA has been construed with considerable restraint by the FSC. In Semenya, this constraint became outcome-determinative. The Swiss court accepted the factual and legal reasoning of the CAS panel and declined to interrogate whether that reasoning had properly accounted for the ECHR. It abdicated the very form of judicial oversight that the ECtHR found essential (paras 142-145).

The difficulty lies not only in the fact that the PILA restricts Swiss courts from correcting arbitral error. It also assumes that commercial arbitration, where minimal review may be defensible, is directly comparable to other forms of arbitration. In commercial contexts, the parties usually consent freely and the dispute remains private. However, where arbitration is effectively compulsory – such as in professional sport – one party has no genuine choice. The law makes no distinction between a dispute over a transnational supply contract and a case involving coerced medical treatment as a precondition for professional participation. Semenya makes it clear that in the latter case, the reviewing court cannot act as merely a procedural auditor (para 201-205). It must be capable of examining whether the process, and the result, comport with the state’s human rights obligations.

Legislative Guidance for Arbitration in Rights-Heavy Domains

Two forms of reform suggest themselves. First, Article 190(2)(e) could be revised to include a reference to Switzerland’s obligations under the ECHR and other binding human rights instruments. Such an amendment would not entail dismantling arbitration’s autonomy. Rather, it would recognise that the scope of that autonomy is not uniform and must be sensitive to the nature of the dispute and the structural conditions under which consent is given. Second, PILA could be modified to distinguish between commercial arbitration, where parties typically act with relative equality and autonomy, and compulsory arbitration regimes, such as in sport or labour disputes, where judicial oversight is necessary to protect weaker parties. This distinction would allow Swiss courts to review not only the legality and proportionality of such awards, but also whether the arbitral tribunal’s factual findings are supported by a sufficient evidentiary basis, especially where those findings determine access to fundamental rights.

The prevailing architecture of Swiss arbitration law rests on the legal fiction that all types of arbitration may be governed by the same procedural standards. This assumption has sustained the uniform application of PILA, but it has also obscured the important distinctions between arbitration in commercial markets and arbitration in domains that directly affect fundamental rights, livelihoods, or access to essential opportunities.

This requires more than judicial diligence; it demands legislative clarity. The present one-size-fits-all approach of PILA carries the risk of inadequate protection precisely in cases where individuals are most vulnerable to institutional power. While power imbalances and rights-implications can arise even in commercial arbitration and are particularly acute in investment arbitration, these settings at least preserve the possibility of meaningful bargaining over forum and procedure. By contrast, in contexts such as professional sport or consumer disputes, the arbitral forum is imposed as a precondition for participation, leaving no scope for negotiation. To correct this, Switzerland should adopt a differentiated arbitration law, one that preserves the benefits of party-led adjudication in commercial matters while introducing safeguards where arbitration substitutes for public law protections. One approach would be, as explained above, to introduce a category of “rights-sensitive arbitration”. Alternatively, Switzerland could explicitly regulate the conditions under which arbitration may validly replace public adjudication in rights-sensitive domains. Such legislation might require that consent to arbitration in employment or sport be explicit, informed, and freely given; that the arbitral tribunal meets standards of independence analogous to public courts; and that courts retain the power to examine awards for their conformity with enumerated rights instruments, including the ECHR and relevant UN treaties.

Comparable frameworks exist in other areas of Swiss law. In administrative justice, special rules apply under Articles 5 and 49 of the Bundesgesetz über das Verwaltungsverfahren (Federal Act on Administrative Procedure), where state decisions directly affect individual rights. Courts retain review powers to ensure compliance with constitutional guarantees. Of course, judicial oversight of administrative action rests on different foundations than the review of private arbitral awards: it is a core element of the rule of law in public law disputes. The analogy, however, lies not in equating arbitration with administration but in recognising that Swiss law already accommodates differentiated standards of review where structural imbalances or fundamental rights are at stake. Extending this logic to compulsory arbitration would therefore not be alien to the Swiss legal tradition, but a context-sensitive adaptation of it.

Conclusion

Semenya has exposed the structural fragilities in a legal regime that has long assumed the neutrality of process, even where that process functions as a gatekeeper to the exercise of human rights. The finding that Switzerland failed to comply with Article 6(1) ECHR was, in formal terms, a procedural verdict. But in substance, it is a judgment on a legal model that presumes arbitration can remain insulated from accountability under ECHR obligations.

Switzerland’s reputation as a global hub for arbitration is anchored in predictability, limited judicial intervention, and institutional neutrality. Those features remain important, but their preservation cannot come at the expense of meaningful rights protection. Semenya shows that the current application of PILA risks not sufficiently protecting individuals in settings where arbitration is not a product of genuine consent.

Whether reform should come through legislative change or a recalibrated judicial interpretation of PILA is an open question. In principle, Swiss courts could, without statutory amendment, construe Article 190(2)(e) to provide full review in contexts where arbitration is structurally mandatory and rights-sensitive. However, reliance on interpretive evolution alone leaves the scope and consistency of protection uncertain. In the past, Swiss courts interpreted “public policy” restrictively. Codifying a differentiated model would give clarity to litigants, align Switzerland’s framework with its human rights obligations, and strengthen its credibility before international courts. If Semenya is to be more than an isolated correction, it must be a catalyst for such systemic adjustment. The administration of justice, even when delivered through private tribunals, is an international responsibility that the Swiss legal order can no longer discharge through formal neutrality alone.