Imagine you are an arbitrator being asked to decide on the validity of a flawed international arbitration agreement.

The parties have spent great time and effort negotiating and finally agreeing on the arbitration agreement in question. The parties’ representatives engaged in marathon negotiation sessions. The representatives exchanged correspondence throughout the negotiation. They also preserved each draft of the arbitration agreement leading up to the final agreed version, and the negotiators’ personal notes.

In the end they agreed that any dispute, controversy, or claim arising out of or relating to their main contract, or its breach, termination, or invalidity, will be settled by arbitration in accordance with the UNCITRAL Arbitration Rules in effect on the date of their contract. The arbitration will be administered by the UNCITRAL Arbitration Institute at The Hague. It will be seated in San Francisco. The parties further agree that a U.S. federal court with competent jurisdiction may exercise de novo review of the arbitral tribunal’s legal conclusions.

The parties entered into their contract in November 2003. Now, ten years later, a dispute arises. The claimant initiates ad hoc proceedings, using the Permanent Court of Arbitration as the appointing authority.

The respondent resists arbitration on the grounds that the arbitration agreement is invalid. First, the parties agreed on institutional arbitration that would be administered by a non-existent entity. Second, the U.S. Supreme Court prohibited so-called heightened judicial review in 2008. It will be impossible for the parties to benefit from their agreed upon appeals mechanism.

The respondent argues that giving effect to the arbitration agreement would require finding consent to arbitrate where none exists. The respondent consented to Process X, but is being forced to accept Process Y.

You, the arbitrator, are well aware of the fundamental principle that parties cannot be compelled to arbitrate when they have not consented to arbitrate. A true statement indeed, but one that Alan Scott Rau correctly described as, “so banal, so commonplace, so formulaic, that readers justifiably wince when they see it repeated.”1)Alan Scott Rau, Arbitral Jurisdiction and the Dimensions of “Consent”, 24 ARB. INT’L. 199 (2008) It raises, but does not solve, a very important issue: how do you, as the arbitrator, determine whether consent exists in this case?

How you answer this question will depend on your approach to interpreting the arbitration agreement. You may want to adopt a pro-arbitration approach, and interpret the agreement expansively under the principle of in favorem validitatis.

Or you may want take special care to protect the respondent from being forced into a procedure to which it did not consent. In that case, you may choose to interpret the agreement more restrictively than you would a normal contract.

Or you may simply want to give effect to the parties’ true intentions.

All of these approaches,2)Fouchard, Gaillard, Goldman, ON INTERNATIONAL COMMERCIAL ARBITRATION 253–261(Emmanuel Gaillard & John Savage eds., Kluwer Law International 1999) among others, are discussed in Fouchard, Gaillard, and Goldman’s 1999 treatise, ON INTERNATIONAL COMMERCIAL ARBITRATION. The authors reject the principles of expansive and restrictive interpretation. Instead, they write, “All that matters is the parties’ common intention,”3)FOUCHARD, GAILLARD, GOLDMAN at 260–261 which is established using accepted principles of interpretation.

But even if you are persuaded by this assertion, you are still in the same position. You need a method for establishing what the parties commonly intended. Or, put another way, to what they consented.

The question of establishing the parties’ consent is necessarily an evidentiary question. Gary Born addresses this issue in his 2009 treatise, INTERNATIONAL COMMERCIAL ARBITRATION. He lays out three possible standards of proof that the claimant bears in establishing the existence of a valid arbitration agreement: the heightened standard of proof, the reduced standard of proof, and the neutral standard of proof.4)Gary B. Born, INTERNATIONAL COMMERCIAL ARBITRATION 643–648 (Kluwer Law International 2009

Each standard is relative to the standard of proof needed to demonstrate the existence of the main contract. Therefore, the question is whether proving the existence of the arbitration agreement should require the same amount of proof as proving the existence of the main contract (the neutral standard), a greater amount of proof (the heightened standard), or a lesser amount of proof (the reduced standard).

Born advocates the reduced standard.5)Born at 652–654 He reasons that, “there is no reason to assume generally that parties would be inclined to enter into particular sales or other contracts. In contrast, there are very serious reasons to presume . . . that commercial parties are predisposed to enter into international arbitration agreements, in order to obtain the benefits that such agreements provide.”

This argument is compelling. But it runs the risk of confusing the standard of proof needed to demonstrate the parties’ true intent with the interpretation principle of in favorem validitatis, which is not necessarily concerned with giving effect to the parties’ true intentions.

A better argument for adopting a reduced standard of proof is demonstrated by considering the absurdity of the scenario presented at the beginning of this post. How often are arbitration agreements negotiated at great length? And how often do negotiations over an arbitration agreement leave an extensive paper trail? Requiring a reduced standard of proof may be preferable simply because the evidence needed to demonstrate the existence of the arbitration agreement will often be severely limited.

In light of this, it is helpful to consider how the claimant may demonstrate the validity of the hypothetical arbitration clause without contemporaneous evidence.

Regarding the first flaw, the claimant could use the wording of the arbitration agreement to demonstrate its validity despite the reference to the UNCITRAL Arbitration Institute. It is clear that the parties intended to arbitrate, agreed upon a set of rules, defined the types of disputes to be arbitrated, and designated an arbitral seat. This would likely meet even the neutral or heightened standards of proof, let alone the reduced standard.

Your task as arbitrator would then be to interpret the clause in a way that would give effect to the parties’ proven intent to arbitrate. You may find that, by writing “administered by the UNCITRAL Arbitration Institute,” the parties imperfectly expressed their true intent to use the Permanent Court of Arbitration as the appointing authority under an ad hoc proceeding pursuant to the UNCITRAL Rules.

The second flaw—the now-invalid heightened judicial review provision—poses a greater challenge. The respondent could argue that this is a question concerning the very existence of the arbitration agreement. The reasoning is that the parties consented to Process X (arbitration followed by an appeals process), but never consented to Process Y (arbitration without an appeals process).

The issue could also be seen as one of interpreting an otherwise existing arbitration agreement. As stated above, the parties indicated a clear intent to arbitrate. The question would then be whether the parties intended to arbitrate even if part of their arbitration agreement needed to be severed.

In that case, your unenviable task will be to find some way to determine whether the parties would have consented to arbitrate had they known heightened judicial review would not be available.

The claimant is unlikely to be able to present much evidence in support of its position. So you may choose to decide the issue purely on evidentiary grounds. Without evidence, the claimant fails to bear its burden of proof and the agreement must be viewed as invalid.

Or you may choose to adopt a pro-arbitration presumption. In that case, you could presume that the parties would not have intended to lose all of the commercial benefits that come from arbitration because one piece of the agreed procedure was unavailable. Making this presumption, however, removes your inquiry from the realm of discerning the parties’ common intent and instead places it in the realm of policy making.

But if you are inclined to make presumptions in the absence of evidence, why presume validity?

A heightened judicial review clause is a rare and specific addition to an arbitration agreement. Furthermore, it preserves the right of appeal that the parties would have had if they had never agreed to arbitrate. It demonstrates that the parties intended to preserve that right while at the same time taking advantage of the benefits of arbitration.

A pro-arbitration presumption may be appropriate when part of the arbitration agreement needs to be severed under other circumstances. But a heightened judicial review provision may be so unique and so connected to the parties’ procedural rights that it requires making the opposite presumption: That, absent evidence to the contrary, the parties would not have consented to arbitrate without access to an appeals process.

This result may be unsatisfactory in light of the prevailing pro-arbitration environment. But it just may be the fairest result.

Jeremy L. Zell is an associate in the Dispute Resolution group at Vinge in Stockholm. Jeremy examined similar issues from a U.S. law perspective in Discerning the Validity of Arbitration Agreements Containing Heightened Judicial Review Clauses After Hall Street Associates, L.L.C. v. Mattel, Inc., 40 LOY. U. CHI. L.J. 959 (2009).


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Profile Navigator and Relationship Indicator
Access 17,000+ data-driven profiles of arbitrators, expert witnesses, and counsels, derived from Kluwer Arbitration's comprehensive collection of international cases and awards and appointment data of leading arbitral institutions, to uncover potential conflicts of interest.

Learn how Kluwer Arbitration can support you.

Kluwer Arbitration


1 Alan Scott Rau, Arbitral Jurisdiction and the Dimensions of “Consent”, 24 ARB. INT’L. 199 (2008)
2 Fouchard, Gaillard, Goldman, ON INTERNATIONAL COMMERCIAL ARBITRATION 253–261(Emmanuel Gaillard & John Savage eds., Kluwer Law International 1999)
4 Gary B. Born, INTERNATIONAL COMMERCIAL ARBITRATION 643–648 (Kluwer Law International 2009
5 Born at 652–654
This page as PDF


  1. Jeremy, a fascinating and, regrettably, not improbable scenario. But I think you need to provide some additional information before concluding what outcome is most appropriate: where are the parties from, what is the object of their agreement and place of performance, and what will be the consequence of a determination that there is no valid agreement to arbitrate?

    The parties did agree their dispute would be decided in the first instance by arbitration in San Francisco. This brings a degree of predictability, especially as to what procedures would NOT be followed (determination of facts by available courts in different countries that the parties could not agree as suitable to assess them). It also brings international enforceability under the NY convention.

    If the arbitrator determines the arbitration agreement is not valid, she may be returning them to all of the concerns that they sought to avoid, however imperfectly. At a minimum, her decision may inject uncertainty as to where (and how) their dispute will be decided. Or, worse, she may be rendering the underlying contractual obligations essentially unenforceable for a claimant.

    How could that be a fair result?

  2. I feel that Jeremy has provided an intriguing puzzle (and, as Michael suggest, one that is alarmingly likely). I would like to suggest that the agreement to arbitrate and the arrangement for appeal are separable.

    The Supreme Court frustrated the appeal facility. It is lost. But that’s all.

    Would it be permissible – or wise – for an arbitrator to invite the parties to devise a new means – an appellate arbitration by recent holders of judicial office? If they agree,fine; if not, there is no facility.

    Again, if they agreed, for example, to take the seat to London, they could have the Court at their service for serious errors of law (s.69, AA 1996), jurisdiction (s.67), or serious irregularity (s.68). That seems to cover everything except the facts and as any good lawyer will say: “what have the facts to do with it”. Again if they fail to agree a new scheme, tant pis. They’re left with a vanilla arbitration. The chocolate sauce has been taken by the US Supreme Court!

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.