Justice Judith Prakash: Speech delivered at the Delhi Arbitration Weekend 2024

Delhi Arbitration Weekend (DAW) 2024
6 March 2024

Understanding the Unsaid: Biases in Arbitration and the Role of Tribunals and Courts

The Honourable Justice Judith Prakash 
Senior Judge
Supreme Court of Singapore

I. Introduction

1.         The Honourable Dr Justice D.Y. Chandrachud, Chief Justice of India, The Honourable Mr Justice Manmohan, Acting Chief Justice, High Court of Delhi, the Honourable Mr Justice Vibhu Bakru, Chairperson, Organising Committee, Delhi Arbitration Weekend, Honourable Judges of the Supreme Court of India and the High Court of Delhi, Distinguished Guests, Ladies and Gentlemen, Good Evening.

2.         First, I would like to thank the organisers of DAW for inviting me to participate in this most prestigious occasion. Since receiving the invitation, I have been reading up on arbitration in India and have learned that the Delhi International Arbitration Centre is at the forefront of a most exciting expansion of arbitration here. The establishment of this annual conference, DAW, is just one example of how the DIAC is a premier thought leader and educator in this process. I am deeply honoured to have been asked to address this distinguished gathering.

3.         My topic today is “Understanding the Unsaid: Biases in Arbitration and the role of Tribunals and Courts”. It is a topic that needs some unpacking because the issue of bias is one that the courts, tribunals and arbitral institutions have had to deal with for some time. Considering it more, it seemed to me that the topic implied two things: first that parties who are being advised to choose arbitration as a dispute resolution mechanism may be reluctant because they fear bias on the part of any arbitrator appointed by the other side. The second implication is that arbitral tribunals acting in good faith may not be aware that their actions may reflect bias.

4.         The thesis of this speech is that as arbitration has moved away from its roots of trade disputes settled quickly and informally by other merchants, towards the heavily judicialized arbitration proceedings we see today, so too has it been recognised that great care must be exercised to remove any taint of bias. All practitioners of arbitration are acutely aware that one of the pillars of a respectable and popular system of dispute resolution is a reputation for probity. If disputants do not have confidence that the dispute resolution system adopted will be fair, the system will be rejected and will collapse from disuse. Thus, as this address will illustrate, legislatures and arbitration practitioners have gone to great lengths to identify and deal with bias.

Types of bias

5.         A fair process of dispute resolution requires that the adjudicator be independent and impartial and that he or she not only be fair but be seen to be fair. This requirement applies across the board to courts and to tribunals. However, bias is an issue that arises more often in the context of arbitration than in judicial proceedings. The view has been expressed that, because arbitral awards are subject to extremely limited curial review, the process itself is fundamental and its fairness needs to be jealously guarded (see Chan, Tan and Poon, The Law and Theory of International Commercial Arbitration at [4.87]). Further, in many countries, arbitrators and counsel involved in arbitrations are drawn from a smallish pool of practitioners. While this situation may be changing due to the increasing popularity of arbitration, the relatively small number of experienced arbitrators, coupled with the expectation of many nominating parties that the arbitrator nominated by them should take their side, often gives rise to challenges to the independence of a nominated arbitrator.

6.         Challenges to the suitability of an adjudicator are not, in the modern context, restricted to allegations of actual bias. Of course, persons who are biased in favour of one of the parties because they have a personal interest in the matter, or have been promised a reward for procuring a particular outcome, will be ejected from arbitral tribunals when the bias is proved. But, in addition, an arbitral appointment can be challenged on the ground of what is known as “apparent” bias. And when such challenges succeed, it is not because the challenger has established that the appointee is in fact biased, but because the challenger has been able to show a ground for fearing that the appointee may not be independent or impartial. Such challenges are, in the ultimate analysis, decided by the court of the seat of the arbitration, the supervisory court. Therefore, it is that court that decides on, first, the standard of proof required to establish apparent bias and, secondly, whose perspective should be adopted to find such bias.

7.         I would like to illustrate my point that the problem of bias is a central concern of modern legislatures and practitioners. In earlier arbitration legislation, the problem of bias was dealt with very briefly. For example, the Arbitration Act 1950 of the United Kingdom by s 24(1) allowed a party to an arbitration to apply to court to remove a person named or designated as arbitrator in the arbitration agreement, on the ground that he was not or might not be impartial. That was a very limited right of challenge for apparent bias. Today arbitration legislation usually deals with the issue of bias in considerably wider terms.

8.         I think it is safe to say that almost all modern legislation on international commercial arbitration is based on or greatly influenced by the Model Law on International Commercial Arbitration adopted by the United Nations Commission on International Trade Law in 1985. My country Singapore, for one, incorporated the Model Law as part of its domestic law on international arbitration in 1994 through the enactment of the International Arbitration Act. In regard to bias, Article 12 of the Model Law provides an articulation which has been widely adopted internationally. Article 12 (1) imposes on all persons who are approached in connection with a possible appointment as an arbitrator the duty to “disclose any circumstances likely to give rise to justifiable doubts as to his impartiality or independence”. I repeat “justifiable doubts as to his impartiality or independence”.

9.         This is a duty that is also imposed by legislation with which you are all familiar. I am referring to section 12 (1) of the Arbitration and Conciliation Act, 1996 of India (the Indian Arbitration Act”). Indeed, that section has built on the Model Law and goes into more detail about the nature of the circumstances that must be disclosed. It states explicitly that such circumstances include the existence, either direct or indirect, of any past or present relationship with, or interest in, any of the parties or in relation to the subject matter in dispute, which is likely to give rise to justifiable doubts as to the candidate’s independence or impartiality.

10.         Further guidance is provided in the Fifth Schedule of the Indian Arbitration Act. This specifies a whole series of relationships and circumstances involving an arbitrator that would give rise to “justifiable doubts” as to partiality. Among the grounds stated in the Fifth Schedule are not only connections with the parties or the dispute but also less common situations. One example is that the arbitrator has been appointed to a tribunal during the previous three years more than 3 times by the same counsel or the same law firm or that the arbitrator holds a position in an arbitral institution with appointing authority over the dispute. If I may say so, it is clear from the Fifth Schedule that Indian legislators have no fear of expressing the unsaid fears of lay people about bias.

11.         And it is not only the legislators, but also the arbitrators themselves who recognise the importance of impartiality and independence to the health of arbitration. Thus, bodies of arbitration practitioners like the International Chamber of Commerce, the Singapore International Arbitration Centre (SIAC) and the International Bar Association have set out rules and practices in this area that go beyond the legislation to guide their members in the proper practices. For example, the SIAC has produced a Code of Ethics for Arbitrators which is sent out to all prospective arbitrators in an SIAC administered arbitration. The very first paragraph of the Code states peremptorily that the appointment should be accepted only if the appointee is himself fully satisfied that he can discharge his duties without bias (para1.1). Then para 2.1 reminds the prospective arbitrator that he has an obligation to disclose all facts or circumstances that “may give rise to justifiable doubts as to his impartiality or independence”.

The Duty of Disclosure

12.         The Code of Ethics thus repeats and emphasises the prime tool that the SIAC, the DIAC and other arbitral bodies have adopted to ensure there is no bias whether apparent or otherwise in the arbitral procedure. I am referring of course to the duty of disclosure. In Singapore, the duty of disclosure is set out in the Model Law and repeated by Rule 13.4 of the SIAC Rules 2016. That Rule emphasises that the duty applies both before appointment and after the arbitration commences. If, during the course of the arbitration, the arbitrator discovers any circumstances that might give rise to such justifiable doubts he must disclose them at that point of time.

13.         The SIAC Rules do not, however, specify who has to entertain the “justifiable doubts”. They are also not as clear as the Indian Arbitration Act as to what needs to be disclosed. This more ambiguous situation is also the position elsewhere. As the cases show, there are multiple circumstances where an arbitrator may appear to or in fact lack independence and whose appointment may be susceptible to challenge. The most comprehensive guidelines as what circumstances require disclosure are contained in the International Bar Association (“IBA”) Guidelines on Conflicts of Interest in International Arbitration 2014 (“IBA Guidelines”). These provide a non-exhaustive list of circumstances where disclosure may be required. The circumstances are categorised into four lists of varying gravity identified by colour from a non-waivable red list of situations where is clear bias to a green list of circumstances that need not be disclosed. The lists are rather fluid and the cases show that each challenge must be judged on its own facts.

Test For Bias

14.         Whether a particular situation gives rise to justifiable doubts over an arbitrator’s impartiality depends, to some extent, on the test that is applied to the analysis of the situation. Of course, some situations are extremely obvious and admit of no doubt whatsoever whatever test is applied. Others are more nuanced and depend, amongst other things, on whose perspective is being considered. For example, the disclosure requirements under the IBA Guidelines come into play if circumstances exist that may “in the eyes of the parties” give rise to doubts about impartiality or independence. The parties to an arbitration are likely to be the most suspicious or fearful of bias on the part of a tribunal and to use their perspective may lead to findings of apparent bias that would not be made by a disinterested observer.

15.         In Singapore, the test which has been adopted deals both with the standard which is applied to assess apparent bias and with the person whose perspective has to be considered when the standard is being applied. Since the case of Re Shankar Alan s/o Anant Kulkarni [2007] 1 SLR(R) 85, decided in 2007, the Singapore courts inquire whether a decision maker is beyond reproach from the perspective of a reasonable member of the public even though the court itself may think that on the facts there was no real danger of bias (at [75]). The inquiry is directed to determining whether the circumstances complained of provide a reasonable basis for a fair-minded observer to suspect that a tribunal might be biased. The court considered two legal tests for bias. The first test was the English common law test of whether there was “a real danger (meaning possibility) of bias” (deriving from the case of R v Gough [1993] AC 646). The second test, “the reasonable suspicion or apprehension of bias” test had been used previously in Singapore and also by the Australian courts (see, the High Court in Webb v The Queen (1993-1994) 181 CLR 41). It explained that Singapore had adopted the “reasonable suspicion” of bias test, for very good reasons. Among these, the first is that the test of reasonable suspicion on the part of the fair-minded observer makes it clear that the court is making no adverse finding on the question whether it is possible or likely that the tribunal in question was infected by bias (at [81]). Secondly, it subjects the work of decision makers to the most exacting scrutiny to ensure that their work is or will be beyond reproach by disinterested fair-minded members of the public.

16.         Shankar Alan’s case arose from the decision of a disciplinary committee, but its principles have been widely applied. In the arbitration context, apparent bias was alleged in the 2014 case of PT Central Investindo v Franciscus Wongso and 2 Others [2014] SGHC 190. The applicant there applied to remove the sole arbitrator on the basis that there were justifiable doubts as to his impartiality. The complaint was that the circumstances leading to certain directions that the arbitrator had made in April 2013, and the directions themselves, showed the Arbitrator to be guilty of partiality.

17.         It was accepted that the test to be applied for determining apparent bias was the reasonable suspicion test set out in Shankar Alan. In applying this test, the court pointed out (at [19]) that a two-stage inquiry had to be undertaken. First, the applicant would have to establish the factual circumstances that would have a bearing on the suggestion that the tribunal was, or might be seen to be, partial. The second stage would be to ask if the hypothetical fair-minded and informed observer would view those circumstances as bearing on the tribunal’s impartiality in the resolution of the dispute before it. Having undertaken this exercise in relation to the facts of the case, the court dismissed the application, finding that the complaint was basically in respect of a “stand-alone” procedural matter and the application to remove the arbitrator was extreme. Arbitrators will not be removed simply because a party disagrees with their procedural rulings.

18.         I understand, from my admittedly brief research, that India adopts a slightly different position to that of Singapore. In 1987, and again in 2005, Indian courts have stated that the relevant test for arbitrator impartiality is whether there is a reasonable apprehension of bias from the viewpoint of the concerned party rather than that of the arbitrator (Murlidhar Roongta v S Jaganath Tribewala [2005] 57 SCL 128 Bom and Ranjit Thakur v Union of India (2018) 12 SCC 471). The test is an objective one, but it employs the perspective of the parties rather than that of a disinterested fair-minded person.

Examples of challenges on grounds of bias

19.         I move on to some examples of cases where it seems to me that the outcome of a challenge on the ground of bias would have been different had the challenge been made in a different jurisdiction.

20.         First, in the 2011 English case of A & others v B & another [2011] 2 LLR 501, the solicitors for the parties to the arbitration, A and B, had in March 2009 jointly appointed Mr X as sole arbitrator under the rules of the LCIA. Mr X had previously been appointed in different matters by A’s solicitors in 2005 and by B’s solicitors in 1999 and 2004. The 2004 matter was in abeyance at the time of the 2009 appointment, but it revived later at a time when work on the arbitration award was mostly complete. In December 2010, Mr X wrote to both parties disclosing that he was acting for B’s solicitors in the 2004 matter. A sought to remove Mr X on the ground that there were justifiable doubts as to his impartiality. This application was rejected. The judge applied the English test for bias and held that there was no real possibility that a fair-minded and informed observer would conclude that Mr X was biased. If he had applied the Singapore test of what a reasonable member of the public would have reasonably suspected, the outcome may well have been different. If the case had arisen in India, I think the outcome would definitely have been different since the arbitrator’s relationship with the solicitors would likely have been impugned under various grounds specified in the Fifth Schedule of the Indian Arbitration Act.

21.         Another English case offers an example of how the right of a party to appoint an arbitrator can sometimes be exploited. In Cofely Ltd v Bingham & Another [2016] EWHC 240 (Comm), there was a successful challenge of a party-appointed arbitrator on the ground of apparent bias. The arbitrator had been appointed 25 times in three years by the second defendant and these appointments comprised 18% of his appointments and 25% of his income during this period. This situation would never have arisen in India, however. Ground 22 of the Fifth Schedule specifies that justifiable doubts as to impartiality would arise if an arbitrator has within the previous three years been appointed as arbitrator on two or more occasions by one of the parties or by an affiliate of one of the parties.

22.         This question of whether an arbitrator accepting a multiplicity of appointments gives rise to apparent bias has most recently been considered by the UK Supreme Court in the case of Halliburton Co v Chubb Bermuda Insurance Ltd [2020] 3 WLR 1474. There, the claimant had commenced arbitration against Chubb pursuant to an arbitration clause in an insurance policy. The claimant and Chubb each appointed an arbitrator and the third arbitrator, one Mr Kenneth Rokison QC, was appointed by the High Court. Subsequent to that appointment, Mr Rokison accepted appointments as an arbitrator in two related arbitrations, in one of which he was Chubb’s appointee. Those appointments were not disclosed to the claimant. When the claimant found out about them it sought to remove Mr Rokison as an arbitrator on the ground that circumstances existed that gave rise to “justifiable doubts as to his impartiality”.

23.         The application went before the Supreme Court after it had been dismissed in two rounds before the lower courts. The Supreme Court applied the English common law test for apparent bias which I mentioned earlier. It held that there might be circumstances in which an arbitrator’s acceptance of appointments in multiple references concerning the same or overlapping subject matter with only one common party, might reasonably cause the objective observer to conclude that there was a real possibility of bias, although whether the objective observer would reach that conclusion would depend on the facts of the particular case and especially on the custom and practice in the relevant field of arbitration. Further, an arbitrator’s failure to make disclosure when required to do so was a relevant factor for the fair-minded observer to take into account in assessing whether there was a real possibility of bias. It should be noted that the Supreme Court held that the arbitrator was under a legal duty to disclose his subsequent appointments and accepted that repeated appointments could give rise to the appearance of bias. But it also held that, on the facts of this case, the non-disclosure itself was accidental and was insufficient to support a finding of bias.

24.         The UK Supreme Court specifically recognised (at [87-91]) that there are practices in maritime, sports and commodities arbitration in which engagement in multiple overlapping arbitrations does not need to be disclosed because it is not generally perceived as calling into question an arbitrator’s impartiality or giving rise to unfairness. The Court accepted evidence from the Grain and Feed Trade Association (GAFTA) and The London Maritime Arbitrators’ Association (LMAA) that it is a feature of their arbitrations that arbitrators will act in multiple arbitrations, often arising out of the same events. Parties which refer their disputes to these institutions are taken to accede to this practice and to accept that such involvement by their arbitrators does not call into question their fairness or impartiality. A similar sort of situation has been catered for in the Indian Fifth Schedule too. Explanation 3 under the rubric “Other Circumstances” reads:

For the removal of doubt it is clarified that it may be the practice in certain specific kinds of arbitration, such as maritime or commodities arbitration, to draw arbitrators from a small, specialised pool. If in such fields it is the custom and practice for parties frequently to appoint the same arbitrator in different cases, this is a relevant fact to be taken into account while applying the rules set out above.

25.         The Halliburton case is worth reading in detail. There are many useful observations on the question of apparent bias as it relates to international commercial arbitration. I would like to point out, however, that if the English Courts had used the “reasonable suspicion” of bias test that is applied in Singapore and India, the outcome could have been very different. In India, in particular, the arbitrator would likely have been disqualified since the claimant’s perspective would have been considered.

The Role of the Court

26.         I move on to the role of the court, albeit somewhat briefly. The court plays its role in ensuring that arbitration is free from bias at two points in the process. The first comes at the beginning of arbitral proceedings when the court adjudicates on challenges to the appointment of arbitrators. In Singapore this is pursuant to Articles 12 and 13 of the Model Law which are very similar in substance to the equivalent provisions in India, namely sections 12 and 13 of the Indian Arbitration Act.

27.         The second point at which bias in an arbitral proceeding may make an appearance is after an award is issued and one of the parties to the arbitration applies to the supervisory court to set aside the award. In this regard, one of the major differences between the arbitration law of the past and modern arbitration law makes its appearance. In the past, awards could be appealed against on a question of law. This allowed judges to look into the merits of the dispute and set aside the award if the tribunal had gone wrong in law. Now that licence no longer exists. Awards can only be set aside if one of the limited grounds set out in Article 34 of the Model Law is satisfied. Additionally, in Singapore, an award can be set aside if there has been a breach of natural justice or if enforcing it would be against the public policy of Singapore.

28.         The judiciary in Singapore has taken a strong pro-arbitration stand by adopting a policy of minimal judicial interference in arbitration proceedings. When faced with a setting-aside application, the preferred approach of the court is to read the award supportively and in a manner that is likely to uphold it rather than to destroy it (CNQ v CNR [2022] 4 SLR 1150 at [49]). The court eschews interference with the merits of an award; however, it is zealous in ensuring that a fair process has been afforded to the parties. I understand that this is the position of the Delhi High Court as well.

29.         Bias is not expressly stated to be a ground on which an award can be set aside. In my view, however, many challenges based on an assertion of breach of natural justice, at their heart, involve allegations of bias which have resulted in the parties having been denied the right to be heard. In Singapore, there is a high threshold for setting aside an arbitral award for a breach of natural justice: it is only in “exceptional cases that a court will find the threshold crossed” (China Machine New Energy Corp v Jaguar Energy Guatemala LLC and another [2020] 1 SLR 695 at [87]).

30.         One such exceptional case in Singapore was Sai Wan Shipping Ltd v Landmark Line Co Ltd [2022] 4 SLR 1032 (“Sai Wan Shipping”). It concerned a dispute between an owner of a vessel and the charterer of that vessel. The court set aside the final award issued in favour of the Owner on the basis that breaches of natural justice in the making of a peremptory order had prejudiced the Charterer’s rights.

31.         The court found that the breach of natural justice began when the arbitrator fixed the time period for the Charterer’s service of defence submissions without any input from the Charterer (at [57]); it continued when he did not give the Charterer an opportunity to address him on the latter’s alleged failure to serve the defence submissions within the stipulated timeframe. That alleged failure was the premise of the arbitrator’s decision to issue a peremptory order (at [64]). The arbitrator then failed to give the Charterer the opportunity to address him in relation to the Charterer’s breach of the peremptory order (at [66]). The arbitrator was in breach, once again, when he failed to give the Charterer the chance to be heard on whether the peremptory order ought to be enforced (at [69]). On enforcement of the peremptory order, the arbitrator disallowed the Charterer from making submissions on the merits of the Owner’s claim, which led to the making of the final award in favour of the Owner (at [74]).

32.         The litany of breaches that I have recited may justify an inference of actual bias. It can also be looked at as demonstrating a closed mind on the part of the arbitrator regarding the charterer’s right to answer the claim against it. Such a closed mind is a classic example of bias.

33.         Another case where an award was set aside for breach of a party’s right to be given a full opportunity to present its case was CBS v CBP [2021] 1 SLR 935. There, the respondent in arbitral proceedings requested a hearing for witnesses to give oral evidence on an alleged agreement between the parties. The respondent’s defence to the claim was that no contract existed. The oral evidence was seen by it to be integral to its case as there was no documentary evidence of the agreement. The arbitrator directed the respondent to submit its proposed witness statements so that he could decide if the evidence had substantive value. Since the respondent refused to do so (insisting on its right to call witnesses without such condition), the arbitrator eventually convened a hearing for oral submissions only. Following the respondent’s withdrawal from any further participation in the arbitration, the arbitrator proceeded to find in favour of the claimant.

34.         The respondent applied to set aside the award. The Singapore Court of Appeal held that while the arbitrator had broad case management powers under the applicable rules, such powers did not go so far as to empower the arbitrator to unilaterally dispense with an evidential hearing altogether, especially where one party had unequivocally asked for such hearing. Consequently, the respondent was denied the opportunity to present its evidence in the manner it had requested, and it was materially prejudiced. This case, I would submit, is another example of a closed mind and it shows that while it is generally accepted in arbitration that the arbitrator is the master of procedure, that mastery must be exercised without bias.

35.         An interesting case for students of bias is CAJ v CAI [2022] 1 SLR 505, where the dispute arose out of a construction contract. The claimant in the arbitration sought liquidated damages for 99 days delay against the respondent contractor claiming that the contractor had caused delay in completion. In its defence the contractor pleaded that first, the construction had been completed on time and, secondly, that the claimant was estopped from claiming liquidated damages (the “estoppel defence”). This was a complete rejection of liability. The contractor’s pleadings did not contain any assertion that it was entitled to an extension of time so as to reduce the liquidated damages payable, if any. The issue of extension of time (the “EOT Defence”) was raised for the first time in the contractor’s written closing submissions in the arbitration. The Tribunal found that the contractor had failed to achieve completion on time and it also rejected the estoppel defence. However, despite the claimant’s objections, the Tribunal accepted the EOT Defence. The Tribunal considered the substance of the EOT Defence and decided to extend the time for completion by a period of 25 days, thus reducing the number of days for which liquidated damages were payable from the 99 days claimed to 74 days.

36.         claimant successfully applied to the Singapore court for part of the award to be set aside. On appeal by the respondent, the Court of Appeal agreed with the Judge that there had been a breach of natural justice. Among other things, in granting a 25-day extension of time, the tribunal had not relied on any of the evidence adduced by the contractor, but had relied on its own experience. This was a breach of natural justice because in any construction project a claim for an extension of time is fact sensitive and cannot be assessed by reference to a tribunal’s experience with other such projects. A failure of the tribunal to inform the parties as to how its “experience” would bear on the extension of time issue was a classic case of breach of natural justice (at [55]). In my view, this was also a case of unconscious bias. A tribunal must decide a case based on the evidence before it and cannot fill in any evidential gaps by recourse to its own knowledge no matter how extensive the same may be.

37.         Like Judges, arbitral tribunals may sometimes find it disconcerting to have to send away a worthy claimant without the full recovery to which it may have been entitled had it been more diligent in the adduction of evidence. But, if the evidence is lacking, no matter how worthy the claimant and recalcitrant the respondent, it is likely biased for the tribunal to give an unsubstantiated remedy. And such bias will often translate into a breach of natural justice.

38.         Finally, how the tribunal conducts the hearing itself may also result in allegations of bias leading to a breach of natural justice. One clear illustration of this is the Shankar Alan case itself. That case involved a solicitor who had been found by a disciplinary committee of the Law Society of Singapore ( the DC) to be guilty of grossly improper conduct. The solicitor applied to court for the finding to be set aside on the basis that the DC had conducted itself in such a manner as to give rise to apparent bias. The court found that the treatment meted out by the DC to the complainants on the one hand, and to the solicitor’s witnesses on the other, was unequal. Further, the questions went beyond clarifying aspects of the evidence and seemed to be directed at extracting concessions from the witnesses. The court therefore concluded that the DC’s conduct of the proceedings would have led a fair-minded observer to have a reasonable suspicion that the DC was infected by disqualifying bias.

39.         Thus far, I have been speaking about the role of the supervisory court. I would like to add a few words about whether there is a role for the enforcement court in relation to bias. In the Model Law world, an award can only be set aside by the court of the seat. In international commercial arbitration,however, it is common for the parties to the dispute not to have any assets in the seat jurisdiction since that jurisdiction would have been chosen for its perceived neutrality vis-à-vis each party. Thus, after a challenge to the award has failed in the seat court, the winning party will most likely have to go to another jurisdiction to enforce the award. When that happens, can the other party resist enforcement on the basis of breach of natural justice or bias arguments that have been rejected by the seat court? In Singapore, the answer is no because the doctrine of transnational issue estoppel will be invoked to bar the repetition of the rejected arguments.

40.         This principle was developed by the Singapore Court of Appeal in the recent decision of The Republic of India v Deutsche Telekom AG [2023] SGCA (I) 10. The case involved an arbitration arising out of a bilateral investment treaty between the Republic of India and the Federal Republic of Germany. The arbitration was seated in Switzerland. The Tribunal issued both Interim and Final Awards in favour of Deutsche Telekom. India applied to the seat court, the Swiss Federal Supreme Court, for the awards to be set aside primarily on the basis that the Tribunal lacked jurisdiction over the dispute. These applications were dismissed.

41.         In September 2021, Deutsche Telekom obtained leave from the Singapore High Court to enforce the Final Award in Singapore. India then applied for the leave order to be set aside. When the application failed, India renewed its arguments in the Court of Appeal, substantially repeating arguments against the validity of the Awards that it had made before the Swiss Courts.

42.         In its judgment, the Court of Appeal emphasised (at [58]) that the threshold question was whether India was precluded from making arguments before the Singapore court, as the enforcement court, which arguments had already been argued before and determined by the Swiss Federal Supreme Court. To answer this question, the court first considered the doctrine of transnational issue estoppel which is a part of Singapore law when foreign judgments are in issue. It then considered whether that doctrine also applied in the context of international commercial arbitration. After a discussion of academic views and English case law, the court held that the doctrine of transnational issue estoppel can, and should, be applied by a Singapore enforcement court when determining whether preclusive effect should be accorded to a seat court’s decision going towards the validity of an arbitral award (at [97]).

43.         Among the reasons the Court gave for coming to this conclusion are the following: First, the International Arbitration Act is silent on this matter and what is not governed by that legislation must be governed by the rules of domestic law which, in Singapore, include the principles of transnational estoppel (at [97]). Secondly, to do so respects the parties’ choice of the arbitral seat since, by making that choice, parties have chosen the system of law that will have primacy in relation to many matters concerning the arbitration. In line with that choice, an enforcement court should ordinarily give effect to the prior decision of the seat court (at [98]). Thirdly, it coheres with the notion that courts co-exist as part of an international legal order and should respect each other’s decisions in the fullest sense and avoid duplication, repetition and inconsistency in decision making (at [99]). Fourthly, the sensible use of the doctrine can help alleviate the problem of inconsistent judicial outcomes and limit re-litigation, thus reducing wastage of time and resources (at [100]).

44.         What I have said just gives you the essence of the judgement. I am sorry that I do not have the time to discuss it in more detail. But I must disclose that I was a member of the coram and I was in full agreement with the adoption of this doctrine in the arbitration sphere. In my opinion, application of the doctrine of transnational issue estoppel can only enhance the efficiency and repute of arbitration by bringing consistency and finality to the court’s decisions on Awards and reducing protracted litigation by dissatisfied litigants.

Conclusion

45.         The topic of bias is an important one for the arbitration industry as I noted at the beginning. That said I am heartened by the amount of attention that has been paid to it in recent years by the legislatures, the academics and the arbitration practitioners and institutions. The more that is said, the more we will be able to understand and deal with the issue. And the better the experience of parties in arbitration will be. Thank you for your patience and attention.