Arbitration agreements, governing law and contractual interpretation

by Dr Mark Campbell, University of Bristol Law School


This is currently a busy time for the UK Supreme Court when it comes to international commercial arbitration. The Court has recently handed down its judgment in Enka Insaat Ve Sanayi AS v OOO ‘Insurance Company Chubb’ (‘Enka v Chubb’).[1] That judgment followed an expedited appeal from a decision of the Court of Appeal in April 2020.[2] Added to that, there a judgment in Halliburton Co v Chubb Bermuda Insurance Ltd, an important case involving an attempt to remove an arbitrator in the context of non-disclosure.[3] There are also appeals pending in Kabab-Ji SAL (Lebanon) v Kout Food Group (Kuwait)[4]  and General Dynamics United Kingdom Ltd v State of Libya.[5]

It is with Enka v Chubb, and the law governing an international arbitration agreement, that this blog post is concerned. In many, if not most, cases an arbitration agreement will be contained in a clause within the parties’ main agreement. In most situations the parties will have expressly chosen the law to govern their transaction; they might e.g. state that ‘This contract is governed by Swiss law.’ And, although it might come as a surprise to those unfamiliar with international commercial arbitration or private international law, a choice of law for the main contract will not necessarily reach over into the clause containing the arbitration agreement. One reason might be the doctrine of separability whereby an arbitration clause can, at least for certain purposes, be treated as an agreement distinct from the main contract: see the Arbitration Act 1996, s 7 and the UNCITRAL Model Law on International Commercial Arbitration, art 16(1). Another reason—this time specific to European jurisdictions—is the conflict of laws regime under the Rome I Regulation which expressly excludes arbitration agreements from its remit.[6] So, within English law the law applicable to the main contract will be determined in accordance with the provisions of the Rome I Regulation, while the law applicable to the arbitration agreement (even if a clause within that contract) is determined by a traditional common law test. That common law test directs the adjudicator to apply (1) an express choice of law if one has been made, or (2) an implied choice of law if one can be discerned from the agreement, or (3) the law of closest connection. For an arbitration agreement the law of closest connection will almost always be the seat of arbitration.

Due in large part to the influence of the New York Convention (and also the UNCITRAL Model Law), there is a significant degree of consensus around the world on many aspects of international commercial arbitration. But there are, of course, legal issues on which things differ from jurisdiction to jurisdiction; and even within a particular jurisdiction, on certain matters legal opinion may be divided. Until the Supreme Court decision in Enka v Chubb, English law was divided on a crucial point concerning the law governing arbitration agreements. Where the law to govern an arbitration agreement has not been clearly stated, should the law governing the main contract be the starting presumption for the arbitration agreement? Or, should it be the law of the seat? There was case law from the English courts supporting those opposing views and it was, in effect, a difference of opinion over how the law should be applied to the facts at the second stage of the common law test. The Supreme Court in Enka v Chubb has opted for the law of the main contract as the starting position, observing that such a presumption fits with ‘principle and authority’ and also finds international support beyond the English law context: see [54]–[58]. For a summary of the English legal principles to be applied when determining the governing law for an arbitration agreement, see the majority judgment at [170].

A comparison of the Court of Appeal and Supreme Court decisions in Enka v Chubb will reveal three key things. One is the starting presumption when looking for an implied choice of law in relation to the arbitration agreement. As noted above, the Supreme Court has said it should be the law of the main contract; that contrasts with the Court of Appeal in Enka v Chubb which unanimously opted for the law of the seat. Another point is the final outcome on the facts. The Court of Appeal and (the majority of) the Supreme Court agreed the arbitration agreement was governed by English (rather than Russian) law. But, finally, that same conclusion (English law) was reached by different means. In the Court of Appeal English law applied as a result of an implied choice of the law of the seat (London). The majority in the Supreme Court held that English law applied as the law of closest connection; they used the law of closest connection as, in their view, there had been no choice of law by the contracting parties in relation to the main contract.[7]

I have had the opportunity to discuss the Court of Appeal decision in Enka v Chubb in a case comment published in International Arbitration Law Review: Mark Campbell, ‘The Law Applicable to International Arbitration Agreements: the English Court of Appeal Departs from Sulamérica (2020) 23 Int ALR 193. In that piece, I expressed the view that an interpretative presumption (or, perhaps, better still a rule of law) favouring of the law of the seat may be the most straightforward way of dealing with the matter. Given the debate over the previous two decades in both case law and commentary, I queried whether the law governing arbitration agreements might at some point in the future be the subject of legislative intervention. Since the Supreme Court’s decision in October 2020, I have been working on another piece for publication (submitted and currently awaiting peer-review) which focusses on the way the Supreme Court approached the law governing arbitration agreements largely as a matter of contractual interpretation and what that decision (especially the reasoning of the majority) might mean for the interpretation of arbitration agreements where the validity or scope is called into question.

Having given much thought to Enka v Chubb during 2020, I have been taken back twenty years to the summer of 2000 when I was about to begin studying law for the first time. At that point I received a book in the post from the institution where I would be studying. The book was Learning Legal Rules by James Holland and Julian Webb and we used it as part of the introduction to law course that occupied the first two or three weeks. On opening the book, my attention was drawn to a sentence in a foreword written by Lord Templeman, a judge in the House of Lords in the 1980s and 1990s. Lord Templeman said: ‘The lawyer is a manipulator of words; this is an assertion and not a criticism.’[8] I was initially baffled by the idea of the lawyer as ‘a manipulator of words’ but it has fascinated me ever since. If ever a court judgment exemplified Lord Templeman’s observation, it is, in my opinion, the decision of the UK Supreme Court in Enka v Chubb where words were interpreted—and pro-arbitration principles and presumptions fashioned and applied—to ensure that an international arbitration agreement was applicable to the dispute in question.


[1] [2020] UKSC 38, [2020] 1 WLR 4117.

[2] [2020] EWCA Civ 574, [2020] 3 All ER 577.

[3] [2020] UKSC 48.

[4] On appeal from [2020] EWCA Civ 6, [2020] 1 Lloyd’s Rep 269.

[5] On appeal from [2019] EWCA Civ 1110, [2019] 1 WLR 6137.

[6] Art 1(2)(e). Although the UK has now left the EU, the Rome I Regulation still applies given that the European Union (Withdrawal) Act 2018 retains EU law as UK domestic law.

[7] For the minority judges in the Supreme Court, the contracting parties had selected Russian law for the main contract and, applying the starting presumption of the law of the main contract, the arbitration agreement was also governed by Russian law.

[8] Foreword to the first edition of James Holland and Julian Webb, Learning Legal Rules, now in its 10th edition (OUP 2019).


Leave a Reply

Your email address will not be published. Required fields are marked *