1 The appellant, the Russian Federation (“Russia”), invokes state immunity in response to proceedings brought to enforce three Final Arbitration Awards (“Final Awards”) by three claimants, who are the respondents to this appeal (“Claimants”). The Claimants’ case is that Russia enjoys no such immunity, because Russia had agreed to submit the disputes, the subject of the three Final Awards, to arbitration; and that whether it had done so has already been examined and decided in proceedings in the Netherlands, the seat of the arbitration, in which Russia sought, unsuccessfully, to have the Final Awards set aside.

2 The central issue now is whether or how far the Singapore courts must re-determine the Claimants’ case that Russia enjoys no such immunity for themselves de novo – a phrase deployed here to mean that the Singapore courts must make that re-determination by themselves as if there had been no Netherlands proceedings. The Claimants accept the importance of the principle of state immunity, but submit that the doctrine of transnational issue estoppel enables the Singapore courts to conclude, without more in the circumstances of this case, that Russia enjoys no immunity in respect of their claims. In a joint judgment dated 25 July 2025 (by Andre Maniam J and Anthony Meagher IJ with which James Allsop IJ also agreed, while adding supplementary observations of his own) (“Judgment”) now under appeal, the Singapore International Commercial Court (“SICC”) agreed with the Claimants and held that Russia enjoys no state immunity in the present proceedings.

3 The Final Awards were issued in consolidated arbitrations by a tribunal (“Tribunal”) which assumed jurisdiction over Russia under the investor-state dispute resolution procedure contained in Art 26 of the Energy Charter Treaty (17 December 1994), 2080 UNTS 95 (entered into force 16 April 1998) (“ECT”). Article 26 entitles an investor of one Contracting State who has a dispute with another Contracting State which cannot be settled within three months to invoke one of the forms of arbitration specified in the ECT. The three Claimants were, directly or indirectly, majority shareholders in OAO Yukos Oil Company (“Yukos Oil”). The disputes which they referred to arbitration under the ECT arose from their claims that Russia had failed to protect and had expropriated their investment in Yukos Oil in breach of the ECT. By the three Final Awards, each dated 18 July 2014, the three Claimants were respectively awarded sums in the region of US$50 billion, US$1.8 billion and US$8.2 billion, together with compound interest and costs.

4 In addition to failing to have the Final Awards set aside in the Netherlands, Russia has also failed to establish state immunity as an answer to the recognition of the Final Awards in England: Hulley Enterprises v The Russian Federation [2025] KB 1 (“Hulley (EWCA)”) (from which an application for permission to appeal to the UK Supreme Court was refused by an appeal panel on 25 June 2025 on the ground that the issue of whether an exception to a foreign state’s right to adjudicative immunity under s 1 of the State Immunity Act 1978 (c 33) (UK) (“State Immunity Act 1978 (UK)”) can be established by an issue estoppel arising from a decision of a foreign court “does not raise an arguable point of law”). The Claimants do not rely upon Hulley (EWCA) as giving rise to a further issue estoppel. But they do rely on the reasoning by which the English Court of Appeal concluded that the judgments of the District Court of The Hague, The Hague Court of Appeal and the Supreme Court of the Netherlands (“Dutch Courts”) gave rise to an issue estoppel in England. Russia’s response is again that state immunity must be considered de novo (in the sense already mentioned) in each and any jurisdiction in which the recognition or enforceability of the Final Awards arises for decision.

5 Mindful of waiver, Russia’s response to the present Singapore proceedings is carefully limited to an application for: (a) a declaration that it is immune from the jurisdiction of the Singapore courts, pursuant to s 3(1) of the State Immunity Act 1979 (2020 Rev Ed) (“SIA”); (b) a consequential order setting aside the SICC’s order dated 20 May 2024 giving leave to serve the proceedings on Russia; and (c) in the event that it is determined that Russia is not immune, directions for it to file a challenge to the leave order on the merits. As the SICC noted at [23] of the Judgment, if Russia had submitted to the Singapore courts’ jurisdiction, it could still, under s 31(2) of the International Arbitration Act 1994 (2020 Rev Ed), have raised jurisdictional objections to the Final Awards, raising factual issues similar to those arising in relation to its claim to state immunity, but would then have borne the onus of proof, which under the SIA rests on the Claimants to displace state immunity.

6 The SIA provides, inter alia:

7 Unless the dispute to which the Final Awards relate was one which Russia had agreed in writing to submit to arbitration, the Claimants will have failed to bring the case within s 11 of the SIA and Russia will enjoy state immunity. Russia identifies the following four respects in which, it submits, the Claimants have fallen short in showing that s 11 applies:

(a) Article 45 Argument: The Claimants have failed to establish any agreement by Russia to arbitrate any claims by the Claimants (or anyone) under the ECT. Russia signed, but never ratified the ECT. Article 45(1) of the ECT provides that:

Russia’s case is that provisional application of the dispute resolution provisions of the ECT prior to its ratification is not permitted under Russian law, and so is ineffective under Art 45.

(b)  Investor/Investment Argument: The Claimants were not “investors” within the meaning of the ECT, being no more than offshore companies owned and controlled by Russian interests; and they made no “investment” within the meaning the ECT, since, if and in so far as they put up any money, it came from Russia.

(c) Article 21 Purported Jurisdiction Argument: The Claimants’ claims in the arbitrations dealt with “Taxation Measures”, and so fall outside the protection of the ECT by virtue of Art 21(1) which provides that:

(d) Article 21 Mandate Argument: If the claims fell otherwise within the ECT, they related to taxes, in respect of which Art 21(5)(b)(i) provides that:

Referral to a Competent Tax Authority was relied on as a precondition to the Tribunal’s continuing authority to decide the dispute.

8  In the Judgment, the SICC set out the way in which the Dutch Courts disposed of these objections as follows (at [38]–[45]):

9 In summary, as the SICC further explained, the Claimants, having lost on jurisdiction in the District Court of The Hague, succeeded before The Hague Court of Appeal in establishing that the Tribunal had jurisdiction, and succeeded before the Supreme Court of the Netherlands in resisting all but one ground of cassation. The one ground on which cassation was granted (ground 1) related to The Hague Court of Appeal’s finding that it was not open to Russia to challenge the Final Awards on the additional ground that the Claimants acted fraudulently in the arbitration proceedings. The Supreme Court of the Netherlands referred that question to the Amsterdam Court of Appeal for consideration. Meanwhile, however, the judgments of The Hague Court of Appeal stand and Russia has not challenged them before either the SICC or this court. Ground 1 itself was on 20 February 2024 rejected by the Amsterdam Court of Appeal on the basis that it involved an “untimely reliance on fraud” and, further, on the basis that if the fraud had been relied upon in time, the Final Awards would not in any event have been set aside. Russia filed a further cassation appeal against that decision, and that appeal appears still to be pending. On 11 April 2025, the Procurator General at the Supreme Court of the Netherlands submitted an advisory opinion recommending that this appeal be dismissed, on which the parties could make submissions. The present appeal must in the present circumstances be considered on the basis that the Final Awards were made by the Tribunal within its jurisdiction, although no doubt the position would have to be reviewed if the pending cassation appeal were to succeed.

10 On 23 April 2025, the SICC identified the following issues as being “broadly agreed” by the parties as arising before it, and directed that it would hear argument first on issues described in paragraph (a) below. The issues, as set out at [11] of the Judgment, are:

(a) The “Preliminary Issues”: Whether Russia is precluded by the operation of transnational issue estoppel and/or the Primacy Principle (as defined in The Republic of India v Deutsche Telekom AG [2024] 1 SLR 56 (“Deutsche Telekom”)) from litigating matters concerning the Tribunal’s jurisdiction which it had unsuccessfully raised before The Hague Court of Appeal and Supreme Court of the Netherlands (“Dutch Appellate Courts”).

(b) The “Immunity Issues”: Whether Russia is immune from the Singapore court’s jurisdiction by reason that it had not “agreed in writing to submit a dispute which has arisen, or may arise, to arbitration” under s 11 of the SIA, on the grounds that:

(i) Russia has only provisionally applied (and never ratified) the ECT, which provisional application never extended to the dispute resolution procedure pursuant to which the arbitrations were ostensibly commenced;

(ii) Russia never agreed to arbitrate any disputes with the Claimants who were not protected “Investors” and did not make any protected investments in Russia within the meaning of the ECT; and/or

(iii) the arbitrations dealt with “Taxation Matters” which are excluded from protection under the ECT and, if it be found that they dealt with “Taxes”, the Tribunal “side-stepped” the preconditional referral mechanism under the ECT.

(c) The “Postliminary Issue”: In the event that Russia does not succeed in the Preliminary Issues or the Immunity Issues, whether it is entitled to directions for the filing of challenges against the enforcement of the Final Awards on the merits.

11 If the doctrine of issue estoppel applies, that effectively answers question (b) as well as question (a). Question (c) is outside the scope of both questions (a) and (b).

12 In the Judgment, the SICC examined the applicability of the doctrine of transnational issue estoppel under a series of headings. It started by considering:

(a) whether this court’s prior decision in Deutsche Telekom was “binding authority for the proposition that the doctrine of transnational issue estoppel can be applied to a question of state immunity arising under the [SIA]”. Having held that it was, it went on to consider:

(b) whether the requirements for the application of the doctrine were satisfied;

(c) whether transnational issue estoppel contradicted a de novo review of questions of state immunity;

(d) whether it could not be used to enlarge the statutory jurisdiction of Singapore courts;

(e) whether there were any policy considerations which require the Singapore courts not to apply the doctrine of transnational issue estoppel to any question relating to state immunity; and

(f) finally, whether there were any other special circumstances which rendered transnational issue estoppel inapplicable.

13 The SICC stated its conclusions on these points at [146]:

14 The SICC added that, “[i]n view of our decision above, these grounds cannot include state immunity, or any other ground precluded by the application of the transnational issue estoppel doctrine to the decisions of the Dutch Appellate Courts” (at [147]).

15 The existence and effect of the doctrine of issue estoppel is not in dispute. Nor is it in dispute that it can extend transnationally to bind one party by reference to findings made between the same parties in a foreign jurisdiction. What is in dispute in this appeal is whether there is a place for transnational issue estoppel when it is relied upon under Singapore domestic law (here, s 3 of the SIA) to bind a state by jurisdictional findings made in a foreign jurisdiction (here, the Netherlands). Russia submits that there is not, on the following grounds:

(a) First, s 3 of the SIA requires a Singapore court to inquire and determine for itself (and if necessary, proprio motu) whether state immunity exists.

(b) Second, jurisdiction cannot be conferred by an estoppel.

(c) Third, and in any event, the balance of policy considerations takes issues of state immunity outside the proper scope of transnational issue estoppel.

(d) Fourth, and not raised before the SICC or in this appeal until the hearing, there are policy considerations special to this particular case, taking it outside the proper scope or application of any transnational issue estoppel.

16 In so far as the SICC held itself bound to a contrary conclusion on all or any of these grounds, by this court’s previous decision in Deutsche Telekom, Russia further submits that the SICC was wrong to hold itself bound, since the above four points were not argued, but, if the SICC was right to hold itself bound, this court should come to a different conclusion about the above four points, and that decision should not be followed now that these have been raised and argued.

17 The Claimants in response submit that there is no incongruity about applying the doctrine of transnational issue estoppel in relation to a plea of state immunity and that its application does not mean that the Singapore court is not deciding the issue of state immunity for itself. Rather, transnational issue estoppel is one of the tools which the court has at its disposal for deciding issues, including jurisdictional issues like those raised before the Dutch Courts and now the Singapore courts. The SICC was correct to treat itself as bound by this court’s decision in Deutsche Telekom, and this court should, in any event, treat transnational issue estoppel as applicable in relation to the current jurisdictional issues.

18 The present case is not concerned with “cause of action” estoppel, which arises where the outcome of a prior judgment is sought to be relitigated in later proceedings. Cause of action estoppel has long been recognised as the basis for refusing to allow relitigation of a prior judgment, whether domestic or foreign. The present case is, in contrast, concerned with issue estoppel, relied upon as arising where an issue has been decided in prior proceedings, which issue a party seeks to relitigate in later proceedings. In a domestic context, the requirements for issue estoppel are: (a) a final and conclusive judgment on the merits; (b) by a court of competent jurisdiction; (c) identity of the parties in the two proceedings; and (d) identity of subject matter in the two proceedings. For there to be any issue estoppel, “the decision on the issue must have been a ‘necessary step’ to the decision or ‘a matter which it was necessary to decide, and which was actually decided, as the groundwork of the decision’”: see Lee Tat Development Pte Ltd v Management Corporation of Strata Title Plan No 301 [2005] 3 SLR(R) 157 at [14]–[15]. ¹ The fact that the decision was reached by default or even in absentia does not mean that it cannot give rise to an issue estoppel. Otherwise, the clearer the outcome, or the more a party was able to evade the court process, the less force the decision would have.

19 Transnational issue estoppel is now a relatively well-established principle. It denotes a form of issue estoppel derived from the judicial decision of an issue in another jurisdiction: see, eg, Carl Zeiss Stiftung v Rayner & Keeler Ltd (No 2) [1967] 1 AC 853 (“Carl Zeiss Stiftung”); The Sennar (No 2) [1985] 1 WLR 490 and, recently in India, Nagaraj V Mylandla v PI Opportunities Fund-I 2026 INSC 298. ² In that context its application may require a more nuanced approach. The position in Singapore was examined in Merck Sharp & Dohme Corp v Merck KGaA [2021] 1 SLR 1102 (“Merck Sharp”), where this court (echoing the caution advised by the House of Lords in Carl Zeiss Stiftung), said at [1]:

20 Later at [25], the court said:

21 The application of transnational issue estoppel was also examined by this court in Deutsche Telekom where this court said at [64]–[69]:

22 This court in Deutsche Telekom also noted potential exceptions to the applicability of transnational issue estoppel, saying at [177]:

23 Finally, the court in Merck Sharp considered at [62]–[65] the “Arnold exception”, referred to at [68] of Deutsche Telekom (above). The court said that this is a potential exception to issue estoppel that could be invoked “‘in the special circumstance that there has become available to a party further material relevant to the correct determination of a point involved in the earlier proceedings’, provided that the further material in question ‘could not by reasonable diligence have been adduced in those [earlier] proceedings’ (Arnold at 109 per Lord Keith of Kinkel; …)”. In the event, the court in Merck Sharp found it unnecessary to decide whether that exception applied, and whether, if it did, it was to be viewed broadly or narrowly, a subject of some differing views in the cases of Lee Tat Development Pte Ltd v Management Corporation Strata Title Plan No 301 [2009] 1 SLR(R) 875, Management Corporation Strata Title Plan No 301 v Lee Tat Development Pte Ltd [2011] 1 SLR 998 and The Royal Bank of Scotland NV v TT International Ltd [2015] 5 SLR 1104 (“The Royal Bank of Scotland”). The Arnold exception only requires further reference in this appeal in relation to Russia’s fourth ground of appeal: see [75] and [78] below.

24 The present case and Deutsche Telekom are for present purposes closely aligned. In both cases the claimants invoked the dispute resolution procedure in an investment treaty. In Deutsche Telekom the treaty was the bilateral investment treaty between India and Germany. In the present case, the treaty was the multilateral ECT. For present purposes, that difference is unimportant. In Deutsche Telekom the claimant had obtained in its favour a Swiss arbitral award. India had challenged the award unsuccessfully in the Swiss courts on jurisdictional grounds. Before the Singapore courts, India resisted enforcement on the same grounds, namely that the bilateral investment treaty only applied and its dispute resolution procedure could only be invoked, where there had been a (lawful) “investment” made “directly” by a German entity, and that Deutsche Telekom had not made any or any lawful investment within the meaning of the BIT or had at most only made an indirect investment through a non-German subsidiary. These grounds were described as the Illegality Argument, the Pre-Investment Argument, the Indirect Investment Argument and the Essential Security Interests Argument. India failed on all these jurisdictional grounds.

25 In proceedings to enforce the award in Singapore, India claimed state immunity under s 3 of the SIA, advancing in Singapore the same jurisdictional issues on which it had failed in the Swiss court of the seat. This court upheld the SICC’s decision dismissing the claim to state immunity. It held that the principle of transnational issue estoppel meant that India was “precluded from relitigating the Illegality Argument, the Pre-Investment Argument, the Indirect Investment Argument and the Essential Security Interests Argument” (at [176]). This court went on at [177] to refer “[f]or completeness” to the potential limits to the doctrine of transnational issue estoppel suggested in Merck Sharp at [54] –[58] and said at [178] that:

26 The difference between Deutsche Telekom and this case is that it was not submitted in Deutsche Telekom that the principle of state immunity took the case outside the proper scope of the doctrine of transnational issue estoppel, and this court did not raise or examine that as a possibility. The present appeal therefore raises a point which was not argued or considered in Deutsche Telekom. Whether the SICC was bound by Deutsche Telekom (as it concluded, after careful examination, that it was) is not relevant in this court. Now that the point has been raised, the point must be addressed on its own merits, without direct guidance from Deutsche Telekom, but now with the benefit of the full submissions, oral and written, and much other material.

27 Taking the first ground which Russia advances, as summarised in [15] above, its force consists in the straightforward imperative, imposed on the Claimants by the combination of ss 3 and 11(1) of the SIA, to show to the Singapore courts that Russia agreed in writing to submit the dispute arising from or relating to the alleged (mis)treatment of Yukos Oil to arbitration. In Russia’s submission, the SICC’s analysis ignores the Singapore court’s duty to satisfy itself that an exception to state immunity applies; it amounts to an illegitimate delegation or “outsourcing” of issues regarding state immunity to the Dutch Appellate Courts, and to a failure to approach such issues independently or de novo in each jurisdiction in which they arise. In terms of the possible qualifications of transnational issue estoppel mentioned in Merck Sharp and summarised in Deutsche Telecom, as cited in [22] above, this submission can be seen as invoking a principle that “transnational issue estoppel should not arise in relation to any issue that the court of the forum ought to determine for itself under its own law”.

28 We can start with the concept of de novo review.

(a)  First, we accept at the outset that each jurisdiction must give effect to the law governing state immunity in its own jurisdiction, in whatever terms this may be formulated, either in a statute like the SIA or, in the absence of any statute, in common law. If there are differences in the formulation or interpretation of the relevant principles governing state immunity between different jurisdictions, then this may lead to a difference in the outcomes in such jurisdictions of any plea or point raised on state immunity. Counsel for the Claimants, Ms Wendy Lin, was right to note this point in the course of her submissions. In the present case, it is not in fact suggested that there are any relevant differences in the law of state immunity as applied in the Netherlands and as applied under the SIA in Singapore. ³

(b) Second, it follows that we are concerned with the application of accepted principles of state immunity on the facts. Russia’s case is that the Dutch Courts erred in their findings of fact (including Russian law) and that Russia should be able to relitigate the same issues of fact in Singapore. Whether that is so is the critical question before us.

(c) Third, we can therefore put aside a number of cases cited by Russia where recognition or enforcement has been sought in state B of a judgment given in state A in plain conflict with the principles of state immunity recognised in state B. The leading International Court of Justice (“ICJ”) case of Jurisdictional Immunities of the State (Germany v Italy: Greece intervening) [2012] ICJ 99 (“Jurisdictional Immunities”) and a number of national court judgments stand for the proposition that a national court must in such circumstances give effect to the principles of state immunity as established or interpreted in its own jurisdiction. The issue in such cases about the applicable principles of state immunity can be seen as an issue which the national court of the forum “ought to determine for itself under its own law”: see [22] above, citing [177(a)] in Deutsche Telekom. Even if transnational issue estoppel had had any potential role in such cases, we see them as cases where the foreign judgment could have fallen to be ignored in the interests of justice and/or on the ground of perversity: see [21]–[22] above, citing from Deutsche Telekom at [69(d)] and [177(d)].

29 The question identified in [28(b)] must be answered having regard both to the terms and effect of the SIA and, more widely, to the nature and effect of the customary international law principle of state immunity, which is the international legal background to the SIA. It is less helpful in our view to seek to gain assistance from different legislation and caselaw, addressing entirely different contexts and situations. That applies to the case of TSH v TSE [2017] SGHCF 21, which Russia relied on, and which concerned the statutory duty to have regard to the welfare of the child as a “paramount consideration” under s 3 of the Guardianship of Infants Act 1934 (Cap 122, 1985 Rev Ed). It also applies to other cases relied upon as showing that public policy may exclude reliance on any doctrine of estoppel per rem judicatam: see, eg, K v P [1995] 1 FLR 248 (considering the effect of prior magistrates’ court findings in later Children Act 1989 (c 41) (UK) proceedings in the Family Division), Rowe v Rowe [1979] 3 WLR 101 (custody proceedings, effect of prior admissions by a husband in the context of an undefended divorce that the children were his) and Kok Hoong v Leong Cheong Kweng Mines Ltd [1964] AC 993 (considering the effect of a prior default judgment on a later plea that the transactions involved illegal moneylending).

30 The term de novo is commonplace in an arbitral context, when considering the courts’ approach to an arbitral tribunal’s conclusions regarding the scope of its own jurisdiction. The kompetenz-kompetenz which an arbitral tribunal (like most other tribunals) possesses to consider the scope of its own jurisdiction has no legal force when a court has to decide on the scope of such a tribunal’s jurisdiction or on whether the tribunal acted within it. The SICC referred in this context to a number of well-known authorities, including the UK Supreme Court decision in Dallah Real Estate & Tourism Holding Company v The Ministry of Religious Affairs, Government of Pakistan [2011] 1 AC 763: see Judgment at [22] and [125]. The SICC went on, correctly, to distinguish such cases from the present. An arbitral tribunal cannot give itself a jurisdiction which it does not have. But it is an important part of the role of the courts of an arbitral seat to determine, when necessary, the scope of an arbitral tribunal’s jurisdiction and the extent to which the tribunal has acted within it when issuing its award. Such a determination may of course be subject to appeal (as were the decisions of the lower instance Dutch Courts in the present case). But that is an ordinary incident of a legal system. Once domestic proceedings have taken their course, whatever determination is reached in them will bind the parties in any later proceedings in that jurisdiction.

31 That creates no problem for Russia’s case, because, as Russia points out, any prior determination will have been by the courts of the same jurisdiction as any later determination. When a court of a different jurisdiction is asked to revisit an issue of state immunity which has already been decided by a prior court of the seat, the different question arises as to how far the later court can simply apply the principles of issue estoppel which it would apply if no possibility of state immunity existed, eg, as if it were a commercial case between ordinary persons or concerns. Russia’s complaint in the present case is that the English Court of Appeal in Hulley (EWCA) and the SICC at first instance approached the factual issues arising under s 11 of the SIA as if this were a simple domestic commercial dispute, without focusing on the special context and fundamental jurisdictional nature of the plea of state immunity under s 3, to which s 11 is no more than an exception.

32 The Claimants and the English Court of Appeal in Hulley (EWCA) took as their starting point that a claim to state immunity operates subject to the domestic procedural and evidential rules of the state in which immunity is in issue. Both cite the English Court of Appeal in Zhongshan Fucheng Investment Co Ltd v Federal Republic of Nigeria [2023] EWCA Civ 867 (“Zhongshan”), where Nigeria had failed to raise state immunity in time. The court said at [33]–[34]:

33 Russia points out, however, that Zhongshan cannot be taken as endorsing an across-the-board application of ordinary procedural rules in the face of either a plea of state immunity or a situation where, without such a plea, the possibility of state immunity is apparent to a court. First, the court must raise and consider the issue of state immunity proprio motu, of its own motion: see s 3(2) of the SIA, set out at [6] above. It must moreover do this, even if the possibility of state immunity only becomes apparent on appeal: Royal Embassy of Saudi Arabia (Cultural Bureau) v Costantine [2025] 1 WLR 1207 (“Royal Embassy”) at [44]. In Royal Embassy, s 1(1) of the State Immunity Act 1978 (UK) was the UK equivalent of s 3(1) of the SIA (see [6] above). Saudi Arabia did not appear, but the UK Supreme Court held that the domestic court had to inform itself of the circumstances and to determine whether there was state immunity of its own motion. Second, it has been held that the ordinary procedural rules (under Ladd v Marshall [1954] 1 WLR 1489) relating to the filing of an appeal or the admission of fresh evidence on appeal may be departed from in order to examine a plea of state immunity: see United Arab Emirates v Abdelghafar [1995] ICR 65 and Arab Republic of Egypt v Gamal-Eldin [1996] 2 All ER 237 followed in Republic of Yemen v Aziz [2005] ICR 1391 (“Aziz”). ⁴ The present appeal cannot therefore be approached on the simple basis that a claim to state immunity falls to be addressed with precisely the same tools as a court would have to address other, purely domestic, issues.

34 Russia submits that ss 3 and 11 of the SIA, read in terms and in the light of customary international law, impose on the Singapore courts a similar duty to forgo or displace any principle of transnational issue estoppel that would otherwise apply in a case where state immunity was not an issue. Here, the application of a principle of preclusion would prevent issues raised by state immunity being raised by Russia or addressed by the court. The courts, if they gave effect to such a principle, would fall short in their positive duty to apply any mandatory law of the forum. Section 3(2) of the SIA is such a law and requires them to “give effect” to state immunity. In contrast, it is of the nature of an issue estoppel that it may preclude the court from reaching the right conclusion. That, Russia submits, would be inconsistent with the court’s duty to investigate any potential state immunity, of its own motion if necessary.

35 These submissions by Russia in our view overlook the fundamental distinction which we have already drawn, and to which we will return, between the identification and application of the principles governing state immunity in Singapore, which are matters for this court, and the effect in this court of the Dutch Courts’ determination of the factual issues, including issues regarding Russian law, relevant to the application of such principles. They also identify only one aspect of transnational issue estoppel. Transnational issue estoppel does operate negatively to preclude or make irrelevant certain evidence and submissions. But it also has a positive aspect, in that it may, where its preconditions are satisfied, establish, in binding fashion as between the parties, that a factual and/or legal position is as it has been determined in prior proceedings. Transnational issue estoppel has in that respect been described as a “rule of substantive law … as to the legal effect” of the prior determination, whereby this determination is, because of “the important public policy of finality in litigation”, treated as creating “an enforceable legal right”: see [34]–[35] and [65] of Hulley (EWCA), citing Associated Electric & Gas Insurance Services Ltd v European Reinsurance Co of Zurich [2003] 1 WLR 1041 at [15], Foxton J in PJSC National Bank Trust v Mints [2022] 1 WLR 3099 and Mills v Cooper [1967] 2 QB 459. ⁵

36 Turning to the second ground of Russia’s case, Russia submits that parties cannot by conduct inter se create an estoppel conferring jurisdiction on the part of a court or tribunal which does not otherwise have this. Russia cites among other authorities Koh Zhan Quan Tony v Public Prosecutor [2006] 2 SLR(R) 830 (“Koh Zhan Quan Tony”), Griffiths v Davies [1943] KB 618 (“Griffiths v Davies”) and J & F Stone Lighting and Radio Ltd v Levitt [1947] AC 209 (“J & F Stone”). But these are cases involving very different contexts and legislation from the present. The first point to note is that, in so far as they address the creation of an estoppel by parties’ conduct, there is a fundamental difference between parties’ agreement or conduct, on the one hand, and a considered decision by a court on an issue of jurisdiction, on the other. The point was made very clearly by Lord Hoffmann in Watt (formerly Carter) v Ahsan [2008] 1 AC 696, at [30]–[31]:

37 Without the distinction drawn by Lord Hoffmann, even a decision of the Singapore courts on an issue of jurisdiction, in the face of a claim to state immunity, would appear vulnerable to repeated relitigation. That this is not the law appears to have been accepted by Russia’s counsel before the English Court of Appeal in Hulley (EWCA) at [51]. If that was so, then the question would be not whether jurisdiction can be based on a finding of an issue estoppel arising from a prior court judgment, but whether there is any reason to distinguish between the position where the prior court is a court of the same jurisdiction and where it is a foreign court.

38 The second point to note is that Lord Hoffmann’s statement also requires some further refinement in a context such as the present. The “extended” forms of issue estoppel, including the principle in Henderson v Henderson (1843) 3 Hare 100; (1844) 6 QB 288, may be less aptly applicable where all that there has been is a common though mistaken assumption by both the parties and the court as to the existence of jurisdiction, as opposed to a considered decision by the court that jurisdiction exists. This point was made by this court in The Royal Bank of Scotland at [126], when considering the reasoning in Koh Zhan Quan Tony.

39 The third point to note is that the cases and statements on which Russia relies, to submit that any court must always be prepared to revisit the correctness of a prior assumption of jurisdiction, must be seen in their statutory context. This may supply, or point to, an answer for reasons which have no parallel in the present case. This further point was made by this court in The Royal Bank of Scotland at [127] with reference to Koh Zhan Quan Tony, where the death penalty was at stake. It is also illustrated by Griffiths v Davies and J & F Stone – both Rent Restrictions Act cases where it is unsurprising that the courts treated the intended statutory protection as continuing for the future, unaffected by erroneous prior judicial orders made on mistaken but unreasoned assumptions about the applicability of such protection.

40 Thus in Griffiths v Davies the tenant failed in initial proceedings to request the court to assess the standard rent, and the court made accordingly an order for suspended possession subject to payment of arrears, which were duly paid. The tenant subsequently applied to the court for the standard rent to be assessed. A plea of res judicata or issue estoppel failed, on the ground that the legislative policy and effect was to allow a tenant to apply at any time during his or her tenancy for the standard rent to be assessed. The English Court of Appeal drove this point home by making clear expressly that it was not addressing what the position might have been if the tenant had been trying to reverse the prior order or recover the arrears ordered to be paid and paid under it, rather than to establish the continuing legal position. There is in the present case no parallel to the continuing statutory protection throughout a tenancy in issue in Griffiths v Davies. A reading of the SIA entitling the state to relitigate the very same factual issues relevant to immunity has, on the contrary, no compelling basis in our view in either the text or the policy of the SIA.

41 In J & F Stone an erroneous order made by the county court judge in respect of a period from April 1943 to 5 December 1944 was again argued to give rise to an estoppel in subsequent proceedings to the effect that the rent payable was 10s. a week in respect of a later period running from at earliest 18 January 1945 to 28 March 1945. Consistently with the approach taken in Griffiths v Davies, the House of Lords had no difficulty in deciding on the applicability or not of the relevant Rent Restriction Act during the later period “untrammelled by the previous decision”; ⁶ and, in doing so, it concluded on the evidence that the parties had in fact agreed on rent in the later period at a level which took the case outside the continuing protection of the relevant Rent Restriction Act. Again, the relevant legislation had continuing protective aims without parallel before us.

42 The case of Koh Zhan Quan Tony is also from a very different – very complex criminal law – context. The Prosecution had appealed successfully to this court against a trial court decision to convict the applicants of robbery with hurt, rather than murder. The applicants sought, belatedly, to object that the Prosecution’s appeal lacked any proper basis, since s 44(3) of the Supreme Court of Judicature Act (Cap 322, 1999 Rev Ed), on which the Prosecution relied only provided in context for an appeal “against the acquittal of an accused person”. This court (a) dismissed the Prosecution’s submission that the applicants were too late to raise their objection and were now estopped from arguing that the Prosecution’s prior appeal fell outside the scope of s 44(3), but (b) went on to hold that the Prosecution’s application was within s 44(3). However, it (c) emphasised that the position would have been different had the applicants been challenging a prior decision on the substantive merits of the case, rather than raising a jurisdictional point. We have already noted the latter comments on this case by this court in The Royal Bank of Scotland at [126]–[127]. It is impossible to derive from it any general principle that Russia must be able to relitigate de novo in any jurisdiction, Dutch or Singaporean, the factual issues determined in the Dutch courts when deciding whether Russia enjoyed state immunity. We agree with the English Court of Appeal in Hulley (EWCA) at [65], that the doctrine of issue estoppel is not limited so as to require such a result.

43 The third ground of Russia’s case is more general, and capable of being relied on together with or in support of both the prior limbs. It is that the balance of policy considerations takes issues of state immunity outside the proper scope of issue estoppel. Russia in this context relies on international law and authority. It submits that, even if it be accepted that a court relying on a foreign decision to conclude or prevent any further examination of an issue relevant to state immunity, is making a substantive decision, it is not doing so “for itself”, as it is suggested that international law does or may require. ⁷ Russia further submits, with reference to statements in Holland v Lampen-Wolfe [2000] 1 WLR 1573 at 1588, that state immunity is a creature of customary international law and derives from the equality of sovereign states and is not a self-imposed restriction on its own jurisdiction. In relation to that, we note only that customary international law regarding state immunity has proved at times both open to doubt and capable of development, changing over quite recent decades to meet current conditions, particularly in the area of commercial activity undertaken by States. National jurisprudence has, we think, also played a role in such shifts of perception: see in this connection the Judgment at [17]–[19]. We recognise however that national courts should in their development and application of domestic law consider and look, at least generally, to act consistently with customary international law as it currently appears, wherever possible.

44 In considering international law and what it might expect in the present situation, Russia relied upon the weighty authority of the ICJ’s judgment in Jurisdictional Immunities. We note at the outset that that was a case relating to a very different immunity from that involved in the present case. What was in issue was immunity from pursuit of a civil claim for war crimes, rather than, as here, immunity from a civil claim for damages for mistreatment of an alleged investment.

45 In the Jurisdictional Immunities case, the First Instance Court of Livadia in Greece rendered a judgment by default on 25 September 1997 against Germany awarding damages to the successors in title of the victims of a massacre taking place in the Greek village of Distomo on 10 June 1944. The judgment was upheld by the Areios Pagos (the Greek Supreme Court) on 4 May 2000. The report does not disclose the issues argued in the Areios Pagos, but it may be presumed that they included state immunity. The justification later advanced for the Greek courts’ action in maintaining the claim against Germany appears to have been that the crimes committed were of such gravity as to remove state immunity.

46 In default of satisfaction of the judgment, the claimants sought and obtained from the Italian courts an order making it enforceable in Italy. Germany claimed against Italy that the order involved a breach by Italy of its jurisdictional immunity. The claim was put on a basis that suggested that the question of whether Italy was in breach in accepting the Greek judgment as enforceable depended on whether Germany’s immunity had been respected by the Greek courts (see [126]). The ICJ recognised the difficulty arising from the fact that it was not concerned with a judgment given by the Italian courts against Germany, but with a judgment by the Italian courts enforcing a judgment against Germany given by the Greek courts. It said (at [125]):

47 The ICJ went on to make clear that, in such circumstances, Italy’s liability did not depend on the ICJ ruling on the question (at [127]):

48 On the contrary, the relevant question was simply “whether the Italian courts did themselves respect Germany’s immunity from jurisdiction in allowing the application for exequatur” (see [127]). The ICJ continued at [128]:

49 The reasoning of the ICJ can be seen as emphasising the duty of every State to consider for itself whether the principle of state immunity is being properly respected; and, in the context of the case before it, it can also be read as implying that Italy might have been liable for breach of that duty, even if it could say that it was enforcing “a decision already rendered by a court of another State, which is deemed to have itself examined and applied the rules governing the jurisdictional immunity of the respondent State”. In other words, Italy could not rely on Greece and the Greek courts having done what they could be deemed or assumed to have done, namely satisfy themselves that there was no state immunity. It had to consider that for itself. The ICJ went on to make this clear from the viewpoint of international law at [130]:

50 To read these passages as decisive of the present case is, however, going too far in our view. First, and above all, they were uttered in a context which involved, on its face, a highly contestable claim that there was no domestic state immunity in respect of war crimes. Horrific though the alleged war crimes were, that was self-evidently an attempt to reframe traditional views of state immunity, which would be drastically affected as a principle, if the seriousness of any alleged breach were the hallmark of the existence or not of immunity. Not surprisingly, the ICJ took the view that Germany’s immunity from civil claims for war crimes allegedly committed some seventy or so years before by Nazi Germany should have been recognised by the Italian courts, whatever the basis on which the claimants had persuaded the Greek courts to ignore it and to give the Greek judgment which was sought to be enforced in Italy. The case was, in summary, one where the Italian courts were concerned with a foreign judgment, issued contrary to basic principles of state immunity, and which should have been refused enforcement accordingly. The ICJ was, self-evidently, bound to apply international law regarding state immunity as it held it to be, whatever view might have been taken about its scope by or in any national court(s).

51 The present case is very different. The issues arising are not about the relevant principles governing state immunity. The Claimants rely on a well-recognised exception to state immunity, represented in the law of Singapore by s 11(1) of the SIA where a “State has agreed in writing to submit a dispute which has arisen, or may arise, to arbitration”. The dispute is about whether, on the facts, including the content and effect of Russian law, the case is shown to fall within that section. The Dutch Courts have determined the relevant issues of fact, and, on the basis on which they determined them, there was and is no breach of international law, which it was or is incumbent on the Dutch Courts or the Singapore courts to recognise.

52 Second, there was no estoppel in play in Jurisdictional Immunities. The question in that case concerned the enforcement of a foreign judgment. There was no suggestion that Italy or its courts were or could be bound in any relevant sense by anything done or said by Greece or its courts.

53 Third, it is not right to read the ICJ as pre-judging or ruling out, in tangential dicta relating to quite different issues, the possibility of transnational issue estoppel in domestic legal proceedings such as the present where the issues are issues of fact relating to the application of unchallenged principles of state immunity. The Westphalian system requires States to respect each other’s autonomy, but not we think entirely to ignore what other States have done or decided. In a world composed of multiple centres and jurisdictions, international law cannot be insensitive to the imperatives of finality: see also the Convention on Choice of Court Agreements (30 June 2005), 3110 UNTS 9 (entered into force 1 October 2015) and the Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters (2 July 2019) (entered into force 1 September 2023), coupled with steps towards a possible further convention on parallel proceedings and related actions, all under the aegis of the Hague Conference, not to mention regional systems such as the EU’s Regulation on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial matters (recast), Regulation (EU) No 1215/2012, [2012] OJ L 351/1.

54 That said, it is of interest to look at the two domestic authorities that the ICJ cited at [130], as supporting its view that each State must consider for itself whether it would have been obliged to accord immunity under international law, if it had been seized of the merits of a dispute identical to that which was the subject of the foreign judgment. The first of these, Kuwait Airways Corp v Iraq [2010] 2 SCR 571, proves to be particularly instructive. Kuwait Airways sought to enforce in Canada an English judgment which it had obtained against Iraq in England, relating to costs of English litigation against Iraqi Airways Company (“IAC”) in circumstances where the English Commercial Court held that Iraq had controlled, funded and supervised a defence by IAC which was marked by perjury and tactics intended to deceive the English courts. Having made findings of fact to that effect, the English Commercial Court held that the commercial exception in the State Immunity Act 1978 (UK) applied. ⁸ In the Canadian enforcement proceedings, Kuwait Airways failed in its primary submission that there was, in the light of the English decision, no issue of Canadian state immunity law to be tried. The Supreme Court of Canada held (at [11]–[12], [19]–[20]) that the State Immunity Act, RSC 1985, c S-18 (Can) (“State Immunity Act 1985 (Canada)”) applied, and that the court had to determine whether its provisions precluded recognition of the English judgment. ⁹ This had to be decided under Canadian law. However, the Supreme Court accepted Kuwait Airways’ secondary argument that the Canadian courts could not review the findings of fact of the English Commercial Court. More fully, it said:

55 The case is therefore not authority for a proposition that where there is a plea of state immunity, there can be no issue estoppel. On the contrary, Iraq was held to be bound in the Canadian proceedings by the findings of fact of the English Commercial Court relating to the conduct of the English proceedings. The only issue which it was for the Canadian courts to decide was whether, in the light of these (binding) findings, the commercial exception in the State Immunity Act 1985 (Canada) applied. The State Immunity Act 1978 (UK) and State Immunity Act 1985 (Canada) being differently worded, there was something for the Canadian courts to consider and determine which had not already been decided by the English court: see also [28(c)] above. The ICJ, when citing this case with approval, cannot have overlooked that it accords only a limited role to the enforcement state, and expressly affirms the legitimacy of an enforcement state treating as binding, for the purpose of establishing whether or not state immunity existed, the prior determination by the English Commercial Court of the relevant issues of fact in the English proceedings.

56 The other domestic authority cited by the ICJ is NML Capital Ltd v Republic of Argentina [2011] 2 AC 495. The claimant in that case sought leave to serve proceedings out of the jurisdiction on the Republic of Argentina (“Argentina”) with a view of enforcing in England a New York judgment obtained on certain bonds. Argentina invoked state immunity. The UK Supreme Court held by a majority that, whilst the commercial exception applied in the New York proceedings, the English proceedings were held not to be premised on the bonds, but on the New York judgment, so that the commercial exception in s 3 of the State Immunity Act 1978 (UK) did not apply. The Supreme Court held however that another provision, s 31(2) of the Civil Jurisdiction and Judgments Act 1982 (c 27) (UK), constituted a further exception to state immunity, and afforded the English courts with jurisdiction, and that the independent waiver of immunity in the bonds was also wide enough to amount to a submission to the English courts for the purposes of an action on the New York judgment. The case illustrates the discharge by UK courts of their duty to give careful effect to the principles and exceptions governing state immunity in each state in which an issue of immunity arises. It does not in our view really assist on the issues now before this court.

57 It is convenient at this point to examine some other cases which Russia cites as showing that other domestic courts “have refused to defer to [a] foreign court’s determination” of an issue of state immunity:

(a)  The first is a decision of the District Court of The Hague in Tereschenko v The Russian Federation No C/09/659657 (2 April 2025) (“Tereschenko”). In that case, the claimant had in an action before the Kyiv-Svyatoshynskyi District Court (“Kyiv Court”) obtained a judgment in absentia against Russia for damage to his apartment by a Russian missile attack in Spring 2022. The Kyiv Court justified its assumption of jurisdiction over Russia on the basis that Russia had infringed both Ukrainian and international law and Ukrainian sovereignty, and could not therefore invoke state immunity. As we have already noted, a decision reached by default or in absentia may give rise to an issue estoppel, but whether it does or does not must involve careful analysis, particularly in the context of an alleged transnational issue estoppel. In the event, in Tereschenko the District Court of The Hague held, citing the Dutch law governing immunity from exequatur as well as Jurisdictional Immunities, that it could not simply adopt the Kyiv Court’s ruling on immunity, but had to form its own opinion on the matter. That was and is, in the present court’s view, in no way surprising. The Kyiv Court’s analysis was plainly unsustainable. The Russian acts complained of were acta jure imperii, not acta jure gestionis. There was and is no principle limiting immunity in cases of serious violations of international humanitarian law or international military law. Even if Art 12 of the draft UN Convention on Immunity removed Russia’s immunity in Ukraine to actions for personal injury or damage to or loss of property, that had no application in the Netherlands, and, in the light of Jurisdictional Immunities, could not in any event apply to armed military actions. The case is in these circumstances one where the decision of the Kyiv Court was plainly inappropriate to give rise to a transnational issue estoppel: see [28(c)] above.

(b) The second case is a decision of the French Cour de Cassation No ECLI:FR:CCASS:2023:C100443 (28 June 2023). In that case, the French courts were being asked to enforce an American judgment against Iran for damages resulting from the death of the claimant’s daughter following an attack committed by a vehicle loaded with explosives, for which a Palestinian faction of Islamic jihad claimed responsibility. The American judge had adjudicated upon and “had himself dismissed such immunity from jurisdiction, under his own law” [emphasis added in italics], but, as the Cour of Cassation went on at [6] to say: “[that] did not exempt the French judge from exercising his jurisdictional power in order to assess whether the Islamic Republic of Iran was admissible and well-founded in invoking this immunity before him”. The Cour de Cassation went on to hold that Iran was immune, notwithstanding any breach of jus cogens and also having regard to what was only indirect involvement on Iran’s part. The decision is further explained at [13], where the Cour de Cassation recorded that: “the Iranian state’s immunity from jurisdiction had been dismissed by the American judge in application of the law on immunity from jurisdiction of foreign states, which provides a specific exception for legal actions relating to personal injury or death resulting from acts of terrorism sponsored by a foreign state and thus allows for the investigation of the civil liability of that state”. The italicised words “under his own law” are significant. The case is one where the principles of state immunity in the prior jurisdiction and in the later enforcement jurisdiction were radically different, so that the French courts were bound to re-evaluate for themselves the claim to state immunity, in the light of French law, in a context where the factual circumstances do not appear to have been in issue.

(c) In the third case, Magnat Agrarindustrielles Unternehmen LLC v The Russian Federation No 14 W 13/24 (27 June 2024, Higher Regional Court of Berlin), the Court upheld a plea of state immunity in German proceedings which were evidently brought with a view of enforcing a foreign judgment. The brief report says little, although it does state that “[t]he view … that there should be no immunity from claims for human rights violations, even where the state acted in a sovereign capacity, does not reflect the general principles of international law and has not been accepted by the courts”, citing, among others, the Jurisdictional Immunities case. This accords with our analysis of the Jurisdictional Immunities case at [50] above; the case falls within the category identified in [28(c)] above.

(d) The fourth case, Mironenko v The Russian Federation No 2T-5-881/2025 (13 February 2025, Lithuanian Court of Appeal), is a parallel to the first above case. The Lithuanian courts were asked to enforce a Ukrainian judgment given against Russia in absentia for, in this case, non-pecuniary damages. The Lithuanian courts gave effect to Russia’s plea before them of state immunity. For similar reasons to those given above (at [57(a)]) in relation to the case of Tereschenko, the case falls within the category identified in [28(c)] above.

(e) The fifth case, Agrotek-Invest v The Russian Federation, No e2T-32-912/2025 (18 June 2025, Lithuanian Court of Appeal), is a direct parallel to the fourth, involving a claim for loss of profits and compensation due to the actions of Russian armed forces. The reasoning in the judgment reflects and to some degree mirrors that in the fourth case. The Ukrainian judgment was again in absentia, but Russia raised state immunity before the Lithuanian courts, which gave effect to the plea. The case again falls within the category identified in [28(c)] above.

(f) We are told that there are parallel proceedings to the present proceeding before the courts of the District of Columbia, ¹⁰ but that they have not, as yet, given rise to any guidance on the issues before us.

58  None of these cases therefore, in our view, presents a parallel to the present, or suggests that it would be wrong to hold Russia bound by the determination by the Netherlands courts of the factual issues before them arising in the course of applying basic principles of state immunity, about which there is no difference in or between the Netherlands and Singapore.

59 Standing back in the present case, there is a judgment to be made as to whether the principle of state immunity is of such fundamental jurisdictional importance that every State must determine its applicability on the facts in every case for itself only – de novo in the sense set out at the outset of this judgment. The alternative is that there is place for application of the principle of transnational issue estoppel, where there has been a prior determination of the relevant factual issues by the courts of the arbitral seat. Russia’s case offers the prospect of repeated litigation of the same issues in a series of countries, with no means of achieving the finality which is normally, or at least domestically, seen as so important an outcome of litigation. It conflicts in that respect with the spirit of trust which in other respects underpins legal relations between like-minded jurisdictions in the modern world. Counsel for Russia, Mr Abraham Vergis SC, sought to achieve a more harmonious and, we would add, palatable position, by saying that he saw no harm if the Singapore courts were to “take account” of the decisions of the Dutch Appellate Courts. Of course, courts value the opportunity to see the submissions before and reasoning of other courts. But that does not, on Russia’s case, give such submissions and reasoning any legal force, or avoid the requirement for a full de novo relitigation of issues already determined in the Netherlands.

60 In his separate concurring judgment below, Allsop IJ raised the possibility that there might be developed some “nuanced and flexible approach in being persuaded of the correctness or at least lack of error in the conclusion of the seat court”. That would suggest that there could be some form of half-way house review between full de novo re-determination and the outright application of transnational issue estoppel. We are not ourselves, however, persuaded that this would prove satisfactory or workable, or that transnational issue estoppel, with its potential qualifications, is so rigid as to be unjust or insufficiently respectful of state immunity. Allsop IJ says, critically, of transnational issue estoppel that it has as its premise that the relevant decision may be wrong. That appears to us to overstate the position. Such an estoppel makes it irrelevant and inappropriate to relitigate the correctness or otherwise of the original decision, save to the extent that the outer limits of transnational issue estoppel under Singapore law, as set out in [19]–[23] above, are exceeded. That was not suggested in the present case. There was no suggestion that it would be contrary to the public policy of Singapore to find that the decisions of the Dutch Appellate Courts gave rise to an estoppel. Nor was there any suggestion that the decision could be considered to be perverse or reflect a sufficiently serious and material error to justify the disapplication of transnational issue estoppel. Russia does not agree with those decisions and seeks to persuade us to come to a different view; but it is no higher than that, and, if transnational issue estoppel means anything, it must exclude such an attempt to relitigate a point already decided elsewhere.

61 We also note the thoughtful differentiation in Allsop IJ’s judgment, between the issue whether Russia ever subscribed to the ECT’s arbitral dispute resolution mechanism at all and the other issues as to whether, if it did, the Claimants were investors making protected investments and able under the terms of the ECT to invoke the dispute resolution mechanism in relation to any such investments. All such issues can, as Allsop IJ acknowledges, be seen as jurisdictional. But Allsop IJ suggests that there is a difference in kind or character between a situation where a state can be shown to have consented to a particular treaty dispute resolution mechanism for at least some disputes and a situation where it has not so consented for any disputes at all. In the former situation, he suggests, the investor will receive a ruling from the tribunal, or the seat court of a tribunal, which it has chosen; in the latter the tribunal and its seat court will have no claim to decide anything. The difficulty about this is that the fact that the former tribunal has jurisdiction by agreement over some “investment” disputes gives it no greater legitimacy to decide in binding fashion that it has jurisdiction over “non-investment” disputes than a tribunal which has not been agreed at all. It is true that the former tribunal and seat court may wrongly decide that something is an “investment” dispute, when it is not. But that is a legal risk which exists just as much in the latter situation; a purported tribunal and its purported seat court may on occasion also reach a wrong answer to the question whether there was any valid agreement to submit any dispute at all to an arbitral dispute resolution mechanism in a treaty. The law must however assume, in both situations, that the law will reach the right answer.

62 We do not therefore consider that it is or would be appropriate to draw a distinction, between the two situations. All the factual issues (including the issues of Russian law), whether viewed as more or less fundamental, go to the jurisdiction of the arbitral tribunal, and were considered in depth, on the basis of evidence and submissions, by the Dutch Courts. We do not see a basis for a logical distinction between the force of those courts’ decisions in respect of different aspects of the jurisdictional issues. Whether a state has committed itself to any dispute resolution procedure does not appear axiomatically to merit different adjudicatory treatment, in comparison with an issue whether a dispute falls within the ambit of a dispute resolution procedure to which the state has committed itself.

63 Standing back, we ask ourselves what there is about the principle of state immunity, which necessitates or argues for so great a departure, as Russia suggests, from the principles underlying transnational issue estoppel in contexts not involving a State. Our answer is that we do not see anything that leads or should lead to such a conclusion in the terms of the SIA read by themselves or read against the background of general international law. Transnational issue estoppel is a recognised doctrine of Singapore law. There is no reason to read the language of ss 3 and 11 of the SIA as excluding or precluding its application. It also contains in its currently accepted form important and appropriate safeguards or qualifications, directly addressing the possibility that the foreign judgment may not be worthy of recognition. As stated by the English Court of Appeal in Good Challenger Navegante SA v Metalexportimport SA [2004] 1 Lloyd’s Rep 67, there is also the overriding consideration that the application of the principle of transnational issue estoppel must work justice and not injustice, and this (as the SICC noted) means that its application may have to yield to special circumstances: see [21], [22], and [28(c)] above. These qualifications depend however on a full examination of all relevant material, not upon examination of the existence or not of “facial” or “manifest” error. As we have already noted, such qualifications were not invoked by Russia in relation to the Dutch Courts’ determination of the factual issues in respect of which transnational issue estoppel is now relied upon. What is said is that issues regarding state immunity should be treated as falling generally outside the ambit of transnational issue estoppel.

64 There is in our view also no clear policy reason why a State alone should be relieved of the consequences of losing on fully litigated factual issues in the courts of the arbitral seat, by permitting it to relitigate precisely the same issues in another – indeed every other – jurisdiction in which such issues may arise again. Once an issue has been decided in the courts of the arbitral seat, the rights or wrongs of that issue have been determined; the winning party has acquired a substantive right accordingly; and, unless one of the recognised qualifications on the application of transnational issue estoppel applies, the State has no special prerogative to enable it to ignore or relitigate that substantive right. Russia submits that the number of cases in which issues of state immunity will be relitigated, on their approach to state immunity, will be small, but that the risk of a State being wrongly deprived of state immunity, and in the present case being held wrongly to be subject to the ECT dispute resolution mechanism, is such that we should recognise that transnational issue estoppel has here no role to play. We disagree. What Russia appears to be suggesting is that the court must undertake an exhaustive search for the truth. But while accuracy and rectitude are undoubtedly imperatives in the legal process (especially in the present context), these must be balanced against other considerations: Sundaresh Menon, “Introduction: Justice in a Globalised Age”, Transnational Commercial Disputes in an Age of Anti-Globalism and Pandemic (Sundaresh Menon and Anselmo Reyes ed) (Hart Publishing, 2024) at pp 21–22. We consider that transnational issue estoppel is, carefully applied to its terms, an appropriate and applicable mechanism, for the promotion of a fair and final outcome in those cases where it applies.

65 We have limited what is said in the previous paragraph to situations where the prior determination has been in the courts of the arbitral seat. That is all that is necessary to decide in this case. The decisions of a court of the arbitral seat have, in our view, a particularly strong claim to respect, bearing in mind especially the role assigned to that court by Art 5.1(e) of the Convention on the Recognition and Enforcement of Foreign Arbitral Awards (10 June 1958), 330 UNTS 38 (entered into force 7 June 1959, accession by Singapore 21 August 1986). Article 5.1(e) provides that recognition and enforcement of an award may be refused upon proof by the party against whom it is invoked that:

66 We would not however wish it to be thought that transnational issue estoppel may not also apply where an issue of state immunity has been determined in litigation between the same parties in one jurisdiction, eg, in enforcement proceedings, and is then sought to be relitigated in another jurisdiction, eg, in further enforcement proceedings, or indeed in the jurisdiction of the seat of the arbitration. We decide in this case no more than is necessary to determine the issues between these parties, but we note the same powerful considerations of finality appear to us potentially applicable in other areas where the doctrine of transnational issue estoppel may foreseeably be deployed. For the same reason, because it is unnecessary, we also refrain from any examination of or comments on a possible Primacy Principle: cf the discussion in Deutsche Telekom.

67 We turn to Russia’s fourth submission, not explored below or developed until its oral submissions at the hearing. It raises the question whether there are policy considerations special to this particular case and militating against recognition of any transnational issue estoppel. Russia submits that the decisions of the Dutch Courts cannot or should not be recognised, because they did not have before them relevant material, particularly a ruling no 2867 of the Constitutional Court of Russia issued 24 December 2020.

68 The order of events is as follows. The Final Awards dated 18 July 2014 were followed by Russia’s application in November 2014 to set them aside. The District Court of The Hague acceded to this application in April 2016, but the Hague Court of Appeal quashed the District Court’s decision on 18 February 2020.

69 One of the issues, indeed the most fundamental issue, determined by the Hague Court of Appeal was whether Russia had by signing the ECT without any ratification brought into effect provisionally the dispute resolution provisions in the ECT. Article 45.1 of the ECT provides that:

70 Russia’s case before the Dutch Courts was that the provisional application of the dispute resolution provisions in the ECT was inconsistent with its constitution or laws. That case was fully argued before the Dutch Courts, on the basis of evidence of the Russian legal position adduced on both sides. Russia’s case was in the event rejected by the Hague Court of Appeal.

71  In response (significantly, even if not exclusively) to the Hague Court of Appeal’s decision, Art 125(5.1)(b) of the Russian Constitution was amended to provide that:

72 At a working group on the proposals for constitutional amendments on 26 February 2020, one of the drafters of the amendment of Art 125, Senator Konstantin Kosachev said:

President Vladimir Putin, present at the meeting, responded:

73 The matter duly went to the Constitutional Court of Russia, but not before the federal constitutional law governing the procedure of that Constitutional Court had been amended. Professor Ekaterina Mishina (whose evidence was adduced to explain the Russian legal position, should the Singapore courts reach the point of considering that) states that, with effect from 9 November 2020:

74 Four days after these amendments took effect, ie, on 13 November 2020, the request for clarification contemplated at the 26 February 2020 meeting was submitted to the Constitutional Court, and on 24 December 2020 the Constitutional Court issued its ruling on Art 23(1) of the Russian Federal Statute “On International Treaties of the Russian Federation”. Article 23(1) provides that an international treaty or part of a treaty may be applied provisionally by the Russian Federation before its entry into force if this is provided for in the treaty or agreed upon with the parties that signed the treaty. It effectively reproduces Art 25(1) of the Vienna Convention on the Law of Treaties (23 May 1969) 1155 UNTS 331 (entered into force 27 January 1980). The Constitutional Court ruled that Article 23(1) does not allow the provisional application of treaty provisions providing for the referral of disputes with foreign investors without the adoption of a federal statute on the treaty’s ratification, even if the treaty has been officially published, and does not imply Russian consent to the referral of any such disputes to international arbitration.

75 Russia now seeks to rely on the Constitutional Court’s ruling as the best, indeed conclusive, evidence of the true Russian constitutional or legal position for present purposes and as showing that the constitution of the Tribunal occurred without any jurisdictional basis in Russian consent. They submit that they should be allowed to argue out the resulting issue of jurisdiction. They would do this, as we understand, either on the basis that the ruling falls within the Arnold exception (see [23] above) or because it suggests a sufficiently serious and material error or a sufficiently grave risk of an unjust result, to bring the case within one of the qualifications on transnational issue estoppel identified in Deutsche Telekom at [69(d)] and [177(d)]: see [21]–[22] above.

76 The Claimants resist Russia’s case on the ruling. They point out that, although the ruling has been available since December 2020, and was the subject of much evidence put before the SICC, it was not deployed in the context of the preliminary issues ordered and considered by the SICC, and was not therefore considered in the SICC’s judgment currently under appeal. The ruling was also not deployed before the English courts. It was sought to be relied upon before the Supreme Court of the Netherlands, but that court sits as a cassation court, adjudicating upon legal issues arising under the law of the Netherlands, not upon issues arising as a matter of fact and evidence under Russian law. We were told that no attempt was made to re-open the matter before the Hague Court of Appeal, as we would have expected could have been possible. In the absence of any evidence on this point, we cannot proceed on the basis of counsel’s “instructions” that no such possibility exists under Dutch law. Had Russia sought to place any reliance on the ruling before the SICC or before this court in written submissions, such points might have been elucidated. They have not been, and the consequences are to undermine any attempt to rely on them at the last stage, as Russia now seeks to do.

77 Still more pointedly, the Claimants refer to the basis on which and manner in which the ruling came into existence, summarised in [68]–[74] above. They point out that the only “parties” whose representations are recorded as having been heard were “the State Duma, the Federation Council, the President of the Russian Federation, the opinions of the General Prosecutor’s Office of the Russian Federation, the Ministry of Justice of the Russian Federation and the Ministry of Foreign Affairs of the Russian Federation”. There was no inter partes argument, and no hearing, or opportunity for submissions or a hearing afforded to the parties actually interested in the proceedings which gave rise to the question which the Constitutional Court was asked to address. Under amendments passed just before the request for a ruling was submitted to that Court, there was a deliberate exclusion of any such outside intervention or influence, and of any transparency at all. The court accepts that some legal systems recognise the possibility of ex parte announcements or definitions of the legal position on particular points. But the court doubts whether they are ever quite so obviously delivered in circumstances where they bear, and are obtained in order to bear, directly on ongoing contentious legal issues, without inviting submissions from or giving any opportunity for hearing both sides of such issues. The Russian Constitutional Court’s reasoning in ruling no 2867 stresses the importance of the right to judicial protection, the principle of unimpeded access to justice, the right to a fair and public hearing by an independent, impartial and competent court, as well as the separation of powers. All these principles appear to us to be challenged by the circumstances giving rise to the Russian Constitutional Court’s ruling, after hearing only the Russian Federation’s side of the argument. We are unable to assimilate the Russian Constitutional Court’s ruling with ordinary jurisprudence, or to think that it could, would or should lead (or, if it had been available, have led) the Netherlands or Singapore courts to arrive at any different conclusions about Russian law than those which the Netherlands courts have actually reached.

78 Having regard to the course of events and circumstances, which we have set out above, we do not consider it either appropriate or just to permit Russia to rely, for the first time before this court, upon ruling no 2867 in order to try to bring the case outside the scope of transnational issue estoppel, by undermining the Netherlands Court of Appeal’s conclusion that the dispute resolution mechanism was operative and had been validly agreed to by Russia by agreeing provisionally to apply the ECT. But, even if we had permitted Russia to rely on the ruling, we are unable to regard it as a new development that should persuade us to treat the doctrine of transnational issue estoppel as inapplicable, whether under the Arnold exception or on any other basis.

79 It follows from the above that we hold that:

(a)  The decisions of the Dutch Appellate Courts establish that the Tribunal had jurisdiction to issue the Final Awards dated 18 July 2014 – by virtue of the parties having agreed to submit the relevant dispute to arbitration within s 11 of the SIA – and that Russia is precluded by the principle of transnational issue estoppel from relitigating the issue whether Russia did so agree before the Singapore court.

(b) Russia’s appeal against the Judgment is therefore dismissed, with costs to be assessed by this court if not agreed.

(c) Russia will have liberty within 14 days of the delivery of this judgment to address the Postliminary Issue, by setting out and explaining the directions which it seeks in order to file any challenge that it may be permitted to make on the merits to the grant of any enforcement orders.