[Akash Kumar Surya is a IV year student at the National Law School of India University, Bengaluru]
When parties agree to resolve disputes through arbitration, three distinct laws typically come into play: the lex contractus (governing the rights and obligations under the contract), the law governing the validity and interpretation of the arbitration agreement (LGAA), and the lex arbitri (governing the arbitration procedure at the seat). Where parties expressly specify all three, interpreting their intentions is straightforward. However, complications arise when only some are stated. This issue was central to the Supreme Court of India’s (SCI) decision in Disortho v. Meril, where the lex contractus and lex arbitri were clearly chosen, but the LGAA was not. This post critiques the Court’s reasoning in Disortho, arguing that it is flawed and undermines India’s ambition to be seen as an arbitration-friendly jurisdiction.
The SCI’s Reasoning in Disortho: An Uncritical Adoption of Enka
In Disortho, the parties had entered into a Distributor Agreement for the supply of medical products in Colombia. Two clauses are particularly relevant: Clause 16.5 stated that the contract would be governed by Indian law (lex contractus), and Clause 18 provided that disputes would be resolved through arbitration “in accordance with the Rules of Arbitration and Conciliation of the Chamber of Commerce of Bogota DC.” The petitioner approached the Indian courts under section 11(6) of the Arbitration and Conciliation Act, 1996 for appointment of arbitrators. The respondent objected, arguing that Indian courts lacked jurisdiction. Since the contract did not specify the LGAA, the court had to determine it.
The SCI began by referring to the judgment in Enka Insaat Ve Sanayi A.S. v. OOO Insurance Company Chubb where the same issue was dealt with by the Supreme Court of the United Kingdom (UKSC). The SCI understood Enka to be laying down the following principle in this regard: when parties do not choose the LGAA but choose the lex contractusas well as the lex arbitri, the lex contractus shall be presumed to be the LGAA. This presumption, however, can be rebutted in limited circumstances, such as when applying the lex contractus would invalidate the arbitration agreement.
The SCI approved of these conclusions on the basis that consistency and uniformity in applying legal principles were “crucial for ensuring fairness and comity in international commerce and dispute resolution mechanisms.” However, it offered no further reasoning to justify adopting Enka in the Indian context. It also cited a host of its own precedents like Arif Azim, Mankatsu Impex, and Roger Shashoua II. However, none of these cases involved the determination of the LGAA in the absence of an explicit choice of law in that regard. These cases, as argued previously, dealt with the issue of determination of seat in the absence of an explicit choice by the parties. Therefore, these judgments were of no help to the SCI in this case and have been incorrectly relied on.
Notably, the Court also failed to engage with alternative approaches to determining the LGAA. While this post does not assess the merits of those alternatives, it highlights the limitations in the Court’s reasoning in uncritically relying on Enka.
The Pitfalls of Uncritically Applying Enka in the Indian Arbitration Landscape
There are several issues with uncritically adopting the Enka approach. In Enka, the principle laid down by the UKSC can be understood as follows: the LGAA should first be determined on the basis of the choice of the parties. This choice can be made either specifically or by choosing the lex contratus (para 118 of Enka). Therefore, the first step is to look for an explicit choice of law made by the parties or, if the parties have chosen a lex contractus, treat it as an implied choice of law for LGAA as well. If neither is specified, the so-called default rule kicks in which states that the law of the seat, which is generally the closest system of law to the arbitration agreement, will be treated as the LGAA. However, blindly applying this framework in India is problematic for several reasons.
Firstly, the lex arbitri does not govern only the procedural aspects of arbitration – it often also governs several important aspects of the arbitration agreement itself. In many cases, these provisions are mandatory and cannot be overridden by party agreement. Now, if the Enka principle is applied blindly, it means that whenever parties choose a lex contractus but remain silent on the LGAA, the courts will assume the lex contractus also governs the arbitration agreement. This can create serious conflicts. To illustrate, suppose a contract specifies Brazilian law as the lex contractus, Indian law (the Arbitration and Conciliation Act, 1996) as the lex arbitri, and makes no mention of the LGAA. According to Enka, Brazilian law would be presumed as the LGAA. However, section 2(2) of the lex arbitri, which in this case would be the Arbitration and Conciliation Act, 1996, clarifies that any arbitration seated in India shall be governed by Part I of the Act. This would mean that aspects like the validity of the arbitration agreement would necessarily have to be governed by section 7 of the Act. Similarly, if parties have not agreed upon a procedure for appointment of arbitrators, the appointment has to necessarily comply with section 11(6). Therefore, applying the Enkaprinciple would lead to problematic outcomes.
Secondly, the rationale behind the Enka principle – treating the lex contractus as the LGAA – was clarified by the UKSC in UniCredit v. Ruschemalliance. The Court explained that when parties choose a governing law for the contract, it is reasonable to infer that this law also governs the arbitration clause because the lex contractus typically applies to all clauses in the contract, even if they are not individually mentioned. The arbitration clause, being one of them, is presumed to be subject to the same law. In essence, the general (lex contractus) includes the particular (arbitration clause) (para 27).
However, this reasoning is flawed as it overlooks the fundamental differences between an arbitration clause and other contractual provisions. Unlike other clauses, an arbitration clause does not define rights or obligations but sets out a dispute resolution mechanism. Importantly, it can be governed by a different law from the rest of the contract – an option rarely available for other clauses. Therefore, treating the arbitration clause like any other clause governed by the lex contractus is logically inconsistent. It assumes a functional similarity that does not exist. Other aspects in the UNCITRAL Model Law itself also highlight such differences. For example, the emphasis on consent underscores the unique nature of an arbitration clause. Unlike other contractual clauses, an arbitration clause must meet stricter formal requirements – such as being in writing and typically signed by both parties – to establish valid consent. In contrast, other clauses can often be unwritten. Moreover, arbitration clauses can, in certain cases, bind non-signatories, whereas ordinary contractual clauses generally do not bind third parties unless through assignment or if the third party is a beneficiary under the contract.
Additionally, the core flaw in treating an arbitration clause like any other contractual clause is that it risks rendering the concept of a separate LGAA meaningless. The very purpose of recognising an LGAA is to allow the arbitration agreement to be governed by a law different from the lex contractus – a principle supported by the doctrine of dépeçagein common law. Therefore, even where the lex contractus is expressly chosen, the proper approach is first to assess whether the parties have impliedly chosen a different LGAA, rather than automatically presuming the lex contractus to be the LGAA as well. Failing to do so blurs the critical distinction between the lex contractus and the LGAA.
Thirdly, this approach conflicts with key international arbitration instruments. Article V(1)(a) of the New York Convention provides that, in the absence of a party-chosen law, the validity of the arbitration agreement is governed by the law of the seat. Similarly, Article 34(2) of the UNCITRAL Model Law states that where no law is chosen, the validity of the arbitration agreement must be assessed under the law of the seat (referred to as “this State”). Admittedly these provisions respectively pertain to the stage of enforcement of the award or its setting aside. However, in the absence of a chosen law, if the arbitration agreement is subject to the law of the seat at the stage of enforcement or setting-aside, it is unclear as to why it should be subject to the lex contractus at the arbitration-commencement stage.
A possible counter-argument is that the choice of lex contractus implies an intention to apply it as the LGAA as well. However, this reasoning is flawed. As already discussed, the choice of lex contractus should not automatically extend to the arbitration agreement. This distinction is also evident in the UNCITRAL Model Law, which recognises two separate laws: one governing the substance of the dispute (Article 28(1)) and another governing the arbitration agreement (Article 34(2)). Article 28 refers to the “rules of law applicable to the substance of the dispute,” while Article 34(2) speaks only of the law governing the arbitration agreement. This clear textual separation reinforces that the two are distinct, and equating them is therefore untenable.
Fourthly, Enka has been subsequently questioned by the UK Law Commission itself. In its recent Review of the Arbitration Act 1996, the Commission noted that “the law in Enka v Chubb was complex and unpredictable” (para 12.20). It went on to propose that unless the parties “expressly agree otherwise”, the LGAA is “the law of the seat” (para 12.15). In Disortho, the SCI overlooked this significant criticism of Enka. The Enka principles, therefore, have been met with criticism even in the very jurisdiction they had been laid down in. Hence, it is submitted that the SCI committed an error in adopting these principles in the Indian context.
Lastly, the approach followed in Enka stands in stark contrast to the judgment of the House of Lords in Compagnie d’Armement Maritime SA v. Compagnie Tunisienne de Navigation which is a leading authority on determination of the lex contractus when parties fail to choose one. In such a scenario, according to Lords Morris and Wilberforce, it is a sound general rule that the law of the seat be treated as the proper law of the contract. If that be so, it is not clear why the LGAA should also not be the same as the law of the seat. After all, when parties choose a lex contractus, it is only for the purposes of governing the substantive rights and liabilities under the contract. The purpose of the LGAA, on the other hand, is significantly different. An automatic extension of the lex contractus as the LGAA, as argued above, is, therefore, untenable.
Conclusion
The SCI’s reliance on Enka in Disortho to determine the LGAA is both legally unsound and practically problematic. By presuming the lex contractus to be the LGAA in the absence of an express choice, the Court ignored the distinct nature and function of an arbitration agreement, the potential conflict with mandatory provisions of the lex arbitri, and key international instruments such as the New York Convention and the UNCITRAL Model Law. Moreover, the approach in Enka has itself been questioned in its home jurisdiction, with the UK Law Commission recommending a return to seat-based determination. If India is to position itself as a truly arbitration-friendly jurisdiction, it must adopt a more nuanced and internationally consistent approach that respects the autonomy of the arbitration agreement and the importance of the law of the seat.
– Akash Kumar Surya
