Speech by Mr Justice Foxton: Arbitration and the Rule of Law – the Role of the Court

Commercial CourtSpeeches

Skip to related content

A Speech to the ICC UK Annual ADR and Arbitration Conference

5 October 2023


1. It is a great privilege to have been asked to address the ICC UK’s Annual ADR and Arbitration Conference, at which it celebrates its 100th anniversary. The conference’s theme this year is, of course, “Promoting the Rule of Law”. I regard myself as a rather unlikely choice of speaker for an occasion such as this, for I must confess that in my early years of practice, I was somewhat of a sceptic, indeed some might even have said a cynic, about the world of international arbitration.

2 In the days when the preceding speaker, Toby Landau KC, and I were members of the same chambers, he would regularly circulate emails complaining about unauthorised borrowings from his collection of exotically titled arbitration monographs. I would respond with emails complaining about the removal of spoof volumes from my own collection – works such as ‘Mañana, mañana’ – The Problem of Delay in Spanish Arbitration, and French, Indian or Chinese?: A Gastronomic Guide to the Major Arbitral Seats of the World. I have a distinct recollection that, in the intervening years, I have seen real books with those titles reviewed in the pages of Arbitration International.

3. However, I am pleased to say that I have long since seen the error of my ways, and, of course, there are none more devout than the converted. So it is that I now find myself following one of the well-established entry routes to a career in international arbitration – full-time on the job training, of indeterminate duration, as a Judge of the Commercial Court.

4. Arbitration and the rule of law is a topic which has attracted contributions from some very distinguished speakers. Notable among the contributions of common lawyers are those of Lord Hoffmann, Lord Neuberger, and Chief Justice Sundaresh Menon. Those speeches offer valuable insights into how different conceptions of the rule of law apply to the resolution of disputes by arbitration, and the speeches debate how far the application of the rule of law, at least in its Bingham iteration, is germane to, and realised by, the international arbitration process.

5. You will pleased to hear that you will not be subjected to a bantamweight contribution to that heavyweight debate, although I regret missing the opportunity to pose Tony Hancock’s question about what many regard as the historical origin of the rule of law in this jurisdiction: “Does Magna Carta mean nothing to you? Did she die in vain?”

6. Fortunately, the organisers of this conference have allowed for the fact that I am only 4 years into my period of indeterminate training, and have described this slot in the programme as addressing “the role of the court”. And so prompted, what I intend to do is consider three of the areas in which the Commercial Court engages with the international arbitration process, and how that engagement reflects the rule of law.

Patrolling the Arbitral Perimeter

7. The first topic I wanted to consider is the court’s role in patrolling the arbitral perimeter. Sometimes we build walls to keep people in, sometimes to keep them out. The arbitral perimeter serves both those functions.

8. The court’s containment function is intended to ensure that the promise to arbitrate is kept, when one of the parties who has made that promise seeks to pursue an arbitrable claim in court. The court most obviously guards that side of the arbitral perimeter when it stays proceedings before it to the extent to which they raise a matter which the parties have agreed to refer to arbitration – one of the few occasions when the specific performance of a non-monetary obligation is mandatory, rather than a matter for the court’s discretion, and which applies regardless of the applicable law of the arbitration agreement.

9. For reasons of comity, that is not the case for the other principal means by which the Commercial Court is asked to secure this side of the arbitral perimeter – the anti-suit injunction. However, where the first element of the applicable test is satisfied – a “high degree of probability” that there is an arbitration agreement which governs the dispute in question – it will be a rare case in which the injunction is refused because there are strong reasons for refusing the relief, or it would not be just and convenient to grant the injunction.

10. While this aspect of the court’s engagement with international arbitration undoubtedly has an ethical underpinning, it does not obviously give effect to the rule of law. The ability of parties to agree the process by which their disputes will be determined reflects the values of personal liberty or autonomy, the freedom to make choices and take actions subject only to such constraints as are necessary for the proper functioning of society. The enforcement of such agreements reflects the moral imperative of holding people to their promises: pacta sunt servanda. In giving effect to those values, it might be said that the court is supporting the fifth aspect of Lord Bingham’s formulation of the rule of law: affording adequate protection for fundamental human rights – something I shall refer to as “Bingham 5”. However, it is in the assertion that the rule of law has a substantive as well as a procedural content that the concept has proved most contentious. Even if the enforcement of the agreement to arbitrate can be seen as an aspect of the rule of law in the Bingham 5 sense, that is equally true of the law of contract generally, rather than something particular to the law of arbitration.

11. What of the court’s exclusionary role – ensuring that those who have not agreed to submit a particular dispute to arbitration do not find it determined there nonetheless? The court patrols this side of the arbitral perimeter when it refuses to stay court proceedings because it determines that the parties have not entered into a binding agreement to arbitrate the dispute in question, when it sets aside an award for lack of jurisdiction, and when it refuses to enforce an arbitration award on jurisdictional grounds. The exclusionary role may be said to support the rule of law in two ways. First, it can be seen as a more convincing application of the Bingham 5 aspect of the rule of law: reflecting the fundamental human right of access to a fair and public hearing by a tribunal “established by law” enshrined in most charters of fundamental human rights, including Article 6 of the European Convention of Human Rights. Second, it reflects the fact that, within the arbitral perimeter, there is an attenuated application of other aspects of the rule of law. This is the case for Lord Bingham’s seventh aspect of the rule of law (“Bingham 7”): that the adjudicative process will be fair. The threshold for interference on grounds of procedural irregularity is tightly confined by s.68 of the Arbitration Act 1996, and can be lost altogether by failing to object in time, or by delay in bring the s.68 challenge. It also arguably the case for Lord Bingham’s sixth aspect (“Bingham 6”) – resolution of disputes without prohibitive cost or inordinate delay – to the extent that an arbitrating party has related claims against third parties to the arbitration agreement, and faces the risk of trying the dispute twice, with potentially different outcomes as a result. It is the consent of the parties to the arbitral process which justifies these differences between the application of the rule of law within and outside the arbitral perimeter. The court’s “exclusionary” patrolling is intended to ensure that the attenuated form of the rule of law applicable within the perimeter only applies to those who have agreed to enter it.

12. There have been persistent calls in some quarters for the Commercial Court to reduce the intensity of its exclusionary perimeter patrols. In particular, it has been suggested that English law should adopt the “prima facie” standard of review when determining whether to stay court proceedings said to have been brought in breach of an arbitration agreement, and which applies in  many UNCITRAL Model law jurisdictions, including Singapore and Hong Kong. It has also been suggested that the court should adopt a “review” rather than re-hearing standard when faced with a challenge to jurisdiction which has already been considered by the arbitral tribunal in the exercise of its kompetenz kompetenz jurisdiction under s.30 of the Arbitration Act 1996. That would involve reviewing the tribunal’s decision by reference to the factual findings it has made, and or at least only by reference to the evidence before the tribunal.

13. On both issues, I favour maintaining English law as currently formulated, albeit there may be scope for making more extensive use of procedural controls to prevent the misuse of the court’s process. A starting point in which it is necessary for someone who seeks to abrogate a claimant’s Article 6 rights in their fullest sense to establish the latter’s consent to that abrogation more effectively promotes the rule of law. I acknowledge that the application of a “prima facie” standard to stay applications would preserve the opportunity for the Article 6 right to be asserted by way of a post-award jurisdiction challenge. However, requiring a party to undertake the expense of fighting an arbitration under a jurisdictional reservation before it can pursue its claim in court imposes a significant restriction on the exercise of that right. That will frequently be the practical effect of staying court proceedings on the basis of a “prima facie” standard, given the significant risks involved in not engaging in the arbitration at all. The right of challenge under s.67 is itself subject to a significant time constraint, with time running from a point when the award is made, even though the award may only become available to the parties after they have paid the arbitral tribunal’s fees. Further, if the s.67 hearing is turned from a re-hearing into a review, the Article 6 right would become more attenuated still – and it is surprising how many people advocating reform of the s.9 standard of review support reform of the s.67 standard of review as well!

14. The principal argument in favour of s.9 reform is that there will be many cases in which determining the s.9 issue wholly or partly in favour of the party seeking the stay will effectively have determined the substance of the dispute which the parties had agreed to arbitrate. Where the court is “virtually certain” that there is an arbitration agreement, or the dispute is only as to its scope, the court has used its case management powers to grant a discretionary stay of court proceedings until the arbitration is concluded. To date, the courts have set their face against the use of the case management stay in this context, save in exceptional cases of this type. However, if there is a case for re-adjusting the balance between the s.9 stay and the tribunal’s s.30 kompetenz kompetenz jurisdiction, revisiting the circumstances in which a case management stay might be appropriate seems a more promising avenue for development than the wholesale adoption of the “prima facie” standard.

15. Turning to the nature of a s.67 hearing, the same issue of principle arises. However, the powers of the court to strike out hopeless s.67 challenges on paper, and to control the evidence before the court on a s.67 challenge, provide considerable scope for curbing attempts to abuse the s.67 process. I see merit in the Law Commission’s proposals to provide an express procedural basis for these powers in the CPR. That will remove any lingering doubt as to their basis, and act as a “behavioural nudge” to judges to exercise them. Where I part company with the Law Commission’s recommendations is in the suggestion that the Arbitration Act 1996 should itself foreshadow the content of those rules the Civil Procedure Rules Committee might make, and the terms in which the draft bill does this. That limits the freedom of action on the part of those charged with formulating the new rules, and attempts to amend them in the light of the experience of applying them. In particular, the provision that “evidence that was heard by the tribunal must not be re-heard by the court, unless the court considers it necessary in the interests of justice”, seems to apt to generate prolonged interlocutory battles. There are significant practical difficulties in attempting to hold the court to the evidential record in the arbitration. Even where transcripts are available – and this is not the case for all arbitrations – the disadvantage of determining contested factual issues on transcripts alone conditions the Court of Appeal’s relationship with the puisne judge when it undertakes a review rather than a rehearing. It is not obvious why the relationship of the judge and the arbitral tribunal in the context of a s.67 review should be approached on the basis that the court must defer to the tribunal’s assessment of the oral evidence, when determining whether the parties had actually agreed to trust this issue to the tribunal’s assessment. Further, the arbitration may have been conducted with no, or limited, disclosure, or with limited time for cross-examination. In my experience, it is the greater prospect of obtaining disclosure, rather than a desire to call additional evidence which was always within its own control, which most animates the claimant who seeks to challenge a jurisdictional decision reached on factual grounds.

“Guarding the guards”

16. The second topic I wanted to consider is the court’s role in the formation of the arbitral tribunal. The Commercial Court can become involved in the constitution of the arbitral tribunal in two ways. First, where the parties’ appointment procedure has failed, the court may be called onto make the appointment. Second, the court might be asked to remove an arbitrator, including where “circumstances exist that give rise to justifiable doubts as to his impartiality.” In both contexts, however, the court’s role is  to advance the rule of law in its Bingham 7 articulation – to ensure a fair hearing for the parties.

17. The two contexts are obviously different – the first exercise is essentially prospective in nature, whereas the second has implications both for the arbitral process to date, and for any arbitrator removed. For that reason, associations which might render someone an undesirable court appointee for s.18 purposes would not justify their removal under s.24. In both contexts, the arbitrator is under a duty to disclose facts and circumstances known to them which would or might reasonably lead the fair-minded and informed observer, having considered the facts, to conclude that there was a real possibility that the arbitrator was biased. That duty currently arises as an implied term of the contract of appointment, but the Law Commission has recommended placing it on a statutory footing, so that it will apply regardless of the applicable law of the arbitration agreement.

18. The identification of the kinds of connection which might be capable of giving rise to justifiable doubts as to an arbitrator’s impartiality has been immeasurably assisted by the International Bar Association Guidelines on Conflicts of Interest in Arbitration, adopted in 2004 and updated in 2014. The IBA Working Party who produced the 2004 version said that the Guidelines were intended to be “a beginning rather than an end of the process”, and so it has proved. The 2014 edition identified further matters which prospective arbitrators were recommended to disclose, reflecting changes in practice and expectations over the intervening decade. It would be idle to suppose that best practice has not further developed over the last 9 years, or that we have seen the last iteration of the Guidelines.

19. Against that background, I want briefly to take stock of one continuing controversy: the issue of repeat appointments of the same arbitrator by one the parties or their lawyers. The IBA Guidelines’ “Orange List” includes the fact that “the arbitrator has, within the past three years, been appointed as arbitrator on two or more occasions by one of the parties, or an affiliate of one of the parties” and “the arbitrator has, within the past three years been appointed on more than three occasions by the same counsel, or the same law firm.“

20. Repeat appointments by a party will frequently arise in cases in which litigation is a regular feature of an arbitrating parties’ business – insurance or reinsurance being an obvious example. However, a party who very rarely arbitrates may well be represented by a law firm which regularly advises its clients on which arbitrator to appoint. Multiple appointments can raise concerns for a number of reasons, particularly where the appointments are in arbitrations with a related subject-matter, with only one party or set of lawyers privy to all arbitrations. The arbitrator’s duty of disclosure in those circumstances was the subject of detailed consideration in Halliburton Co v Chubb Bermuda Insurance Ltd. That decision has generated considerable debate, and criticism from some quarters. I am conscious that the Supreme Court did not adopt the submissions made by the ICC, who intervened in the appeal. I hope you will not think me too timorous in those circumstances if I say no more about it.

21. What, however of repeated sequential, appointments? In most areas of commercial life, those who provide a good professional service expect to generate repeat customers, just as those who fail to do so will find that the customer does not return. Arbitrators who are attentive to detail, conduct hearings fairly and efficiently, and promptly produce awards which are clear, cogent and not susceptible to challenge are providing a good service. It can be asked why they should be denied the usual benefits of doing so, and why those who have experienced a high quality service should be inhibited from coming back for more. Equally, where there is a right of party-appointment, it is difficult to see why there should be a legitimate objection to a party appointing an arbitrator who, as a matter of genuine conviction, approaches particular legal issues in a manner which happens to best suit its case – an arbitrator who favours a literalist approach to contractual interpretation, or a more purpose one, for example. I am told litigating parties often make the same choice when deciding whether to commence proceedings in the Chancery Division of the Commercial Court!

22. That is not, unfortunately, the whole story. I expect most of us in this room have at some point encountered a pattern of repeated appointments of an arbitrator where the arbitrator’s appeal to the appointing party was not their commitment to the duty under s.33 of the Arbitration Act 1996, nor their general outlook on a particular topic of legal enquiry. Rather it was their expertise in sandwich making. At least to the extent of knowing which side their bread was buttered.

23. How can this issue be tackled? Disclosure will no doubt assist, and in some cases, the disclosed facts may be sufficiently extreme to provide the basis for a successful s.24 application, at least where the arbitrator is sufficiently guileless to lend support to the application, for example through reluctance to disclose the facts or by sending hostile communications. But successful applications of this kind are vanishingly rare. Removal of an arbitrator is a blunt tool, with significant implications for the cost and duration of arbitration, and the significant consequences of removal understandably engender great caution before an order is made. The issue is much better addressed at the appointment stage.

24. It would be possible for arbitration institutions such as the ICC to adopt a rule restricting the number of appointments any arbitrator appointed to an arbitration conducted under its rules (or at all) could accept from the same party within a particular period – such as the “3 and 8” rule of the International Cotton Association which limits the number of appointments an arbitrator can accept from the same party within a calendar year to 3. A rule of that kind might also help broaden the pool of regularly appointed arbitrators, and in the process assist in the diversification of international arbitration which the ICC is rightly so keen to achieve.

25. But of course, the problems to which the party-appointment of arbitrators can give rise are not limited to repeat appointments. Alan Redfern’s survey of dissenting awards famously found that 95% were in favour of the appointing party. Respondents to a 2021 survey found that 55% of respondents who had sat as arbitrators  and 70% of arbitrating counsel had experience of party-appointed arbitrators trying to favour the appointing party by some means. Perhaps more revealing than the findings themselves is the fact that they were greeted with the same level of surprise as statements about bears’ toilet habits and the Pontiff’s religion.

26. A more radical response would be for the major arbitration institutions to remove the right of party-appointment altogether. That innovation has a number of notable proponents, including Chief Justice Menon and Professor Jan Paulsson. It has always struck me as strange that parties who have agreed to arbitrate their disputes in part to avoid them being determined in the national courts of one of them should agree to a tribunal where the opposing party can chose one member, although only, of course, in return for a similar right of appointment for themselves. As we ponder the means by which international arbitration can move ever closer to the ideal of the rule of law, the phenomenon of party appointment will merit particular attention.

Section 68 challenges

27. The final topic I wanted to consider is also concerned with the Bingham 6 aspect of the rule of law: “due process” challenges to arbitration awards brought under s.68 of the Arbitration Act 1996. I doubt that there is any other type of court application with such a high failure rate. The Commercial Court Report for 2021-2022 recorded that 26 applications had been brought the previous year, and only 4% – 1 application – succeeded. Nine of the 26 applications – just under 35% – were dismissed on paper.

28. One complaint which tends to get a little more traction in the s.68 context is that the arbitral tribunal decided the dispute on a basis which had not been raised in the arbitration, and it is a particular type of that kind of complaint which I wanted to briefly mention today. We are all familiar with cases in which a claimant, who has sustained a legal wrong, comes to the arbitration claiming the moon. In response, the respondent does not advance a rival, more realistic, calculation of loss, but limits itself to undermining the claimant’s figures, and argues that no damages should be awarded because the claimant has failed to put a sufficiently credible assessment before the tribunal. Now it would of course be possible for an arbitral tribunal faced with such a scenario to award everything or nothing. However, outside the rarefied context of baseball salary arbitration, most tribunals faced with two extremes prefer to seek something closer to the actual loss in the middle. In my experience, that is as true of judicial as it is of arbitral tribunals.

29. It is of course always open to the tribunal to go back to the parties to ask for further submissions on alternative bases. However, a tribunal’s reluctance to do so is easy to understand. There is not only a statutory duty, under s.33 of the Arbitration Act 1996, but a professional impetus, to resolve an arbitration as expeditiously as possible. The closing of the record and filing of post-hearing memorials are generally regarded as a watershed moment in the arbitral process. Further, for the tribunal to “show” its hand in this way, for the purpose of inviting further submissions, risks the parties adjusting their behaviour in anticipation of the terms of the award. Lewison LJ’s description of the trial is equally applicable to the merits hearing in an arbitration: it is not “a dress rehearsal. It is the first and last night of the show.”

30. The courts accord the  arbitral tribunal some leeway when faced with difficulties of this kind, a room for manoeuvre accommodated by references to the question of whether the point was “in play” or the “building blocks” of the tribunal’s approach appeared on the record, if not the precise structure the tribunal conjured with them. However, there have been occasions when the tribunal’s approach has been held to involve a sufficient departure from the submissions and evidence put before them to give rise to a serious irregularity.

31. There is a difficult line for the arbitral tribunal to walk here. In my view, it is important when the court is asked to determine whether they have stepped over it to pay appropriate regard to the reasons why the arbitral tribunal found themselves in such a position, and the pressures of finality which they face. I recently heard a s.68 application raising an issue of this kind in a case called LMH v EGK. In dismissing the application, I noted that:

“It must have been obvious to the parties, and wholly within their contemplation, that if the case reached the question of damages, then unless the Tribunal either awarded the entire amount claimed or nothing (and I cannot believe any lawyer with experience of international arbitration would have regarded either outcome as a likely scenario), then the Tribunal would have to make its own adjustments or allowances to reflect LMH’s attacks on EGK’s quantum case to the extent that they were regarded as having merit.”

I also noted that “a respondent who chooses to run a ‘scorched earth’ response to an overstated quantum case necessarily accepts the possibility of the tribunal fashioning a reduced award from the ‘building blocks’ provided.” Perhaps that decision will show how far I have come since sending my spoof emails to Toby Landau a quarter of a century ago.


32. At an event earlier this year, a member of the ICC UK board suggested to me that the Commercial Court had become, if anything, a little too “pro-arbitration”. As I hope will be clear, my own perspective is that the Commercial Court is not pro-arbitration for its own sake. To the extent that it can be said to be “pro-arbitration”, it is because it is the policy of English private law to respect the autonomy of individuals and legal persons to organise their affairs as they see fit subject to minimally invasive and well-defined limits; to enforce promises which parties have voluntarily made to each other; and to favour the resolution of commercial disputes by a fair and efficient process. Supporting international arbitration serves all of those important values, and I do not think any change in the court’s approach is likely, or desirable.

33. It is high time I drew to a close, and returned to my on-the-job training. Until the day comes when I can finally ascend to your ranks, I shall have to content myself with administering the “faint praise” with which the judiciary traditionally rejects challenges to arbitral awards. Informing arbitral tribunals that their decision on a point of English law could not be said to be “obviously wrong”, and as one Commercial Court judge observed of an award from a former member of the Court of Appeal, might possibly be right. And however novel the procedure which the tribunal has chosen to follow, it cannot be said to be one of those “extreme cases where the tribunal has gone so wrong in its conduct of the arbitration that justice calls out for it to be corrected.”

Thank you