Arbitrator and Counsel, Schellenberg Wittmer, Zurich
Iura novit curia, or here, iura novit arbiter, means, put literally, that the arbitrator knows the law. But with a non-participating party, the question can quickly become: how much of that knowledge may, or even must, the arbitrator actually use?
Non-participating parties in arbitration are a regular occurrence. The reasons for this are manifold. A party may choose not to participate for tactical reasons; it may do so to save costs in a losing case; or there may be practical or legal impediments preventing it from participating.
Traditionally, arbitrations involving a defaulting party were more often associated with smaller disputes (the proverbial first case for a young arbitrator). However, that has arguably changed with the recent increase in anti-arbitration injunctions and sanctions-related cases, where parties may deliberately abstain from participating or even face serious legal obstacles to doing so. Such disputes can be financially significant and legally highly complex.
The Tension Between Iura Novit Arbiter, the Right to Be Heard and Equal Treatment of the Parties
Particularly in such cases, the non-participation of one party may give rise to difficult procedural and case-management questions. One of the more challenging questions is how to reconcile iura novit arbiter with the tribunal’s duty to ensure the participating party’s right to equal treatment and its right to be heard.
Many arbitrators will have faced some version of the following scenario. The claimant is represented by sophisticated counsel and pleads its case on the facts and the law. The respondent does not participate. On reviewing the claim, the tribunal concludes that the claimant’s legal analysis on one or more claims is untenable and cannot support the requested relief.
If the respondent were represented, this would rarely pose a serious difficulty. Its counsel would likely have identified the weakness and put it before the tribunal, providing helpful authorities on which the tribunal could rely in dismissing the claim.
But if the respondent is in default, that task may partially fall on the tribunal.
The question is easier to answer if the claimant's interpretation of the law is plainly wrong, as in the example above, and/or is not supported by the authorities the claimant itself has provided. If, even assuming the facts as pleaded to be true, the applicable law manifestly does not support the relief sought, the tribunal must dismiss the claim.
The issue is harder to tackle where the law is not crystal clear. There may be conflicting appellate decisions. The relevant provision of the law may be controversial or little tested. Or there may be no domestic jurisprudence on the precise issue at all.
Yet the arbitrator's obligation to apply the law as it is, rather than simply as it has been pleaded by one party, remains. The Swiss Supreme Court held, for example, that, like state courts, tribunals are “obliged” to assess ex officio the legal consequences of the properly pleaded or established facts, “without being bound by the legal views of the parties” [Swiss Federal Supreme Court, 2 March 2001, 4P.260/2000, consid. 5(b)–(c)]. Most civil-law jurisdictions, including Germany, Belgium, Sweden and France, likewise take the view that the principle of iura novit curia extends to tribunals, though, as we will see, its scope and limitations differ.
For the tribunal, this presents a particular conundrum where one of the parties is not participating: how much independent legal research and legal reasoning is too much?
The Limits of Iura Novit Arbiter
The tension is between three principles. The tribunal must apply the law correctly. At the same time, it must treat the parties equally, avoid taking over the role that counsel for the non-participating respondent would otherwise have performed and respect the participating party’s right to be heard.
Publicly available jurisprudence from tribunals or courts reviewing awards on that issue specifically involving non-participating parties is scarce. But certain conclusions can be drawn from the existing general jurisprudence on iura novit arbiter and the right to be heard in setting-aside proceedings.
Here, different jurisdictions draw the line in somewhat different ways.
As noted above, Swiss jurisprudence takes a strong view that iura novit curia is an obligation of the tribunal. In the case referred to above, the tribunal had recharacterized the legal basis of the claimant’s case from contractual performance to damages without first hearing the parties on that issue. The Swiss Supreme Court held that this did not violate either the right to be heard or the prohibition of ultra petita. On the other hand, the Swiss Supreme Court has recognized limits to iura novit curia and set aside a CAS award where the tribunal relied on Swiss employment-agency legislation that had not been discussed by the parties and whose relevance was not reasonably foreseeable, such that its application came as a surprise to the affected party [Swiss Federal Supreme Court, 9 February 2009, 4A_400/2008, consid. 3.1–3.2].
German courts have taken a similar view. In a 2020 decision, the Frankfurt Higher Regional Court accepted that a tribunal could rely on a provision of law that one of the parties had never invoked. The Court, however, took the opportunity to stipulate the limits of iura novit arbiter: while, according to the Court, there is no obligation on the tribunal to share its views of the law with the parties before rendering a decision, a diligent and knowledgeable party should reasonably have been able to anticipate the legal view adopted by the tribunal [OLG Frankfurt am Main, 17 December 2020, 26 Sch 15/19].
By contrast, France and Belgium place greater emphasis on the right to be heard. The French Cour de Cassation held that tribunals need not disclose their legal reasoning in advance, but must hear the parties before relying on a genuinely new legal basis [Cass. 1re civ., 14 March 2006, No. 03-19.764]. Belgian law similarly requires respect for the principe du contradictoire in arbitral proceedings, meaning that each party must be given a genuine opportunity to address the factual and legal grounds on which the tribunal intends to base its decision [Art. 1699 Belgian Judicial Code]. However, there appear to be no publicly available Belgian court decisions illustrating how that principle should be applied in arbitral proceedings.
The common thread is therefore clear: a tribunal has the power, and depending on the jurisdiction at least to some extent also an ex officio obligation, to conduct its own research on the applicable law, but the parties must have a fair opportunity to address a materially new decisional route that the tribunal intends to take.
The legislative history of the Swedish Arbitration Act illustrates this point particularly well [Proposition 1998/99:35]. The travaux préparatoires, later endorsed by the Svea Court of Appeal, confirm that iura novit curia applies to tribunals and that tribunals may independently seek guidance on the applicable law. At the same time, the travaux emphasize that tribunals should provide clear procedural guidance to the parties on legal issues so as to avoid taking them by surprise, particularly in international arbitrations seated in Sweden.
Practical Considerations for Arbitrators
From the above examples, several practical considerations emerge. Tribunals should take iura novit arbiter seriously. The legitimacy of arbitration as an alternative to domestic courts also depends on tribunals applying the law correctly and ensuring that claims with genuine legal merit succeed. At the same time, legitimacy depends equally on predictability and the parties’ confidence in the arbitral process. The tribunal’s decision on the law therefore must not take the parties by surprise.
Similar conclusions are reflected in the 2008 International Law Association (ILA) Report on Ascertaining the Contents of the Applicable Law in International Commercial Arbitration. The Report suggests that arbitrators should primarily obtain information on the content of the applicable law from the parties (Recommendation 5). As a consequence, they should, in principle, refrain from introducing new legal issues, namely propositions of law capable of affecting the outcome of the dispute, that the parties have not raised (Recommendation 6).
The ILA Recommendations draw a useful distinction as to when tribunals may rely on new authorities. A tribunal may consult authorities that the parties have not cited. If those authorities merely confirm or reinforce a proposition that has already been fully argued, it may not be necessary to bring to the parties’ attention every further judgment or commentary. However, where those authorities go “meaningfully beyond” the sources already invoked and may significantly affect the outcome, the parties should be granted an opportunity to comment [Recommendation 10].
Finally, the Recommendations contemplate a more active role for the tribunal where it is dealing with a defaulting party or expedited interim relief, including by scrutinizing the parties’ submissions ex officio more closely (Recommendation 14).
While not intended as a procedural manual, this appears to provide a sensible framework for tribunals applying iura novit curia, bearing in mind that Recommendation 14 should inform their overall approach where one party does not participate in the proceedings.
In practice, particularly where one party does not participate, tribunals should scrutinize the participating party's submissions and conduct their own legal research at an early stage. This allows the tribunal to use the hearing to test the participating party’s submissions on the content and interpretation of the law and to raise any legal theory the tribunal considers relevant but that has not been pleaded, thereby avoiding surprise.
An issue identified early can be put to the parties through a question, a procedural order or a focused request for submissions. By contrast, an issue identified only at a late stage of deliberations, particularly after the proceedings have closed, may delay the arbitration and even require the proceedings to be reopened.
From the above jurisprudence, it transpires that a tribunal is unlikely to be criticized merely for conducting its own legal research, even where that research ultimately supports the position of a non-participating respondent. The real procedural risk arises where the tribunal adopts a legal route that the participating party had no opportunity to address.
Ensuring that the claimant is heard on such issues is therefore not only a matter of procedural fairness but arguably also serves the claimant’s own interest in a robust award. A tribunal that rigorously tests the legal basis of the claim and gives the claimant an opportunity to engage with the decisive legal issues is better placed to render a well-reasoned award that can withstand challenge and enforcement scrutiny.