Singapore Management University

Institutional Knowledge at Singapore Management University

Research Collection Yong Pung How School Of Yong Pung How School of Law
Law

6-2024

Re-calibration of curial intervention in public policy challenges

against arbitral awards
Darius CHAN
Singapore Management University, [email protected]

Elias Ngai Hum KHONG

Singapore Management University, [email protected]

Follow this and additional works at: https://ink.library.smu.edu.sg/sol_research

Part of the Dispute Resolution and Arbitration Commons, and the Public Affairs, Public Policy and
Public Administration Commons

Citation
CHAN, Darius and KHONG, Elias Ngai Hum. Re-calibration of curial intervention in public policy challenges
against arbitral awards. (2024). Journal of International Arbitration. 41, (3), 271-316.
Available at: https://ink.library.smu.edu.sg/sol_research/4448

This Journal Article is brought to you for free and open access by the Yong Pung How School of Law at Institutional
Knowledge at Singapore Management University. It has been accepted for inclusion in Research Collection Yong
Pung How School Of Law by an authorized administrator of Institutional Knowledge at Singapore Management
University. For more information, please email [email protected].
 KluwerArbitration

Document information Re-calibration of Curial Intervention in Public Policy

Challenges Against Arbitral Awards
Publication Darius Chan ; Elias Khong
Journal of International (*) (**)
Arbitration
When an award debtor challenges an award on public policy grounds, usually the principle
of finality prevails, and courts will consider the award debtor bound by the decision of the
tribunal. However, because public policy has implications beyond the disputing parties
Key words themselves, some courts consider themselves justified in reviewing the award. There is
public policy therefore a tension between finality versus the court’s duty to stand as the guardian of
curial intervention public policy. Whether a review of an award should be allowed under this ground, and if so,
Betamax the extent of permissible review, differs across various jurisdictions. For instance, common
AJU law authorities have generally preferred a very strict approach where a court may review an
maximal review award on public policy grounds only in extremely limited situations. This paper considers
contextual review the prevailing approaches taken across different jurisdictions and ultimately proposes an
minimal review alternative approach for the common law to strike a better balance between all competing
enforcement of award interests.

1 INTRODUCTION
Bibliographic 1 In arbitration, parties must lie in the bed that they have made for themselves. (1) Those
reference who opt for arbitration ‘must live with the decision of the arbitrator, good or bad’. (2) This
Darius Chan and Elias underscores the principle of finality in arbitration. (3) Under this principle, the
Khong, 'Re-calibration of substantive merits of an arbitral award generally cannot be reviewed by a national
Curial Intervention in P 272 court – even if the tribunal may have made an error of fact and/or law. (4) Meanwhile,
Public Policy Challenges national courts also have a duty to safeguard their state’s public policy. (5) Indeed, when
Against Arbitral Awards', in confronted with an attack on its legal system’s most fundamental principles, national
Maxi Scherer (ed), Journal courts ought to be in a position to act.
of International Arbitration, 2 The tension between these two competing interests – the principle of finality on one
(© Kluwer Law hand and the duty to safeguard a state’s public policy on the other – is most stark when a
International; Kluwer Law national court is faced with an award where the tribunal has already pronounced on a
International 2024, Volume public policy argument. Betamax (6) is a case in point. In that case, the defendant
41 Issue 3) pp. 271 - 316 attempted to argue that a contract made with a Mauritian-owned entity was illegal for
infringing the state’s procurement process. The tribunal dealt with this issue, and found
that the contract was not illegal as it fell within an exemption provided under the
Mauritian public procurement legislation. This same point was raised at the setting-aside
stage where the defendant argued that the award was contrary to public policy as it
sought to uphold an illegal contract. In such a case, should the national court be allowed
to review the tribunal’s findings, to prevent the enforcement of an award which is
potentially injurious to the public good? If so, under what circumstances should the
national court be justified in reviewing such findings? On one hand, excessive
intervention in an award may undermine international arbitration as a viable and
effective dispute resolution mechanism that can reduce the caseload of courts. On the
other hand, indifferent rubber-stamping of awards can lead to the enforcement of illegal
agreements and promote illicit activities against the spirit of upholding the rule of law.
3 How then, should this difficult balance be struck? Different jurisdictions have calibrated
this balance differently. (7) Some jurisdictions favour a minimalist review, strictly
upholding award finality except in extremely narrow circumstances, which risks
undermining the state’s public policy. On the other end, a maximalist review approach
intervenes extensively whenever a public policy objection arises, which jeopardizes the
P 273 finality of arbitration awards. In addressing these shortcomings, this paper critically
examines existing approaches and proposes a nuanced solution. The proposed approach
first deals with the severity of any public policy allegation upfront. At this first stage, the
reviewing court must determine if the allegation(s) – even if proven – would trigger the
fundamental public policy of the forum. Only allegations which would trigger the
fundamental public policy of the forum will proceed to the second stage, where the
reviewing court will engage in a full review of the tribunal’s analysis on the public policy
issue(s). Unlike a de novo review, a full review is limited to the evidence that was put
forth before the arbitral tribunal. New evidence will not ordinarily be accepted unless
the evidence was not reasonably available at the time of the arbitration.
4 Section 2 begins by analysing the ground of public policy from a conceptual and
practical perspective. Section 3 deals with the central question of when and how
extensively a national court should be allowed to review a tribunal’s findings related to a
public policy challenge. It critically evaluates three broad approaches that national
courts have taken in this regard, and identifies their main deficiencies. Section 4
explains this paper’s proposed approach. This proposed approach seeks to balance the
individual deficiencies of the three prevailing approaches. Section 5 provides the

1
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 justifications for the proposed approach with reference to the three prevailing
approaches. It also recognizes the risks associated with the proposed approach and
provides suggestions on how to deal with them. Section 6 concludes.

2 THE PUBLIC POLICY GROUND OF CHALLENGE

5 Although the UNCITRAL Model Law on International Commercial Arbitration (Model Law)
(8) and the New York Convention (NYC) (9) do not permit appeals against arbitration
awards, (10) they provide an exhaustive list of grounds to challenge an award. (11) One of
those grounds is premised on the award being contrary to the ‘public policy’ of the forum.
This public policy ground is reflected in Articles 34(2)(b)(ii) and 36(1)(b)(ii) of the Model
Law, which are in pari materia to Article V(2)(b) of the NYC. (12) These provisions provide
P 275 as follows:

Model Law, Article 34. Application for setting aside as exclusive recourse
against arbitral award
…
(2) An arbitral award may be set aside by the court specified in Article 6 only
if:
… (b) the court finds that:
… (ii) the award is in conflict with the public policy of this State.
Model Law, Article 36. Grounds for refusing recognition or enforcement
(1) Recognition or enforcement of an arbitral award, irrespective of the country
in which it was made, may be refused only:
…
(b) if the court finds that:
… (ii) the recognition or enforcement of the award would be contrary to the
public policy of this State.
ARTICLE V of the New York Convention
… 2. Recognition and enforcement of an arbitral award may also be refused if
the competent authority in the country where recognition and enforcement is
sought finds that –
… (b) the recognition or enforcement of the award would be contrary to the
public policy of that country.
The public policy challenge is not only found in these international instruments. Indeed,
even jurisdictions which have not adopted the Model Law provide for this ‘public policy’
exception in their respective national arbitration legislation. (13)
6 This section of the paper will first engage in a conceptual and theoretical analysis of
the public policy ground ( A ). Next, we will look at how the public policy ground typically
manifests itself in practice. In this regard, this paper identifies at least three
manifestations of public policy breaches ( B ).

2.1 Conceptual analysis of the public policy ground

2.1[a] Defining Public Policy
7 Public policy is notoriously amorphous, famously described as ‘a very unruly horse, and
when once you get astride it you never know where it will carry you’. (14) It is therefore no
surprise that there has been no precise definition of the ‘public policy’ ground to date.
(15) In fact, the term ‘public policy’ was intentionally not defined in the Model Law and
the NYC. (16) In the absence of a universal statutory definition, (17) the concept and scope
of the ‘public policy’ ground were left to individual states to develop. (18) This gave rise
to the different expressions of public policy we see today. (19) Rather than aim to provide
a comprehensive definition, this paper sketches out how public policy has been defined
in various jurisdictions in order to sieve out certain common characteristics.
8 Perhaps the best-known definition in the common law world is that of Judge Joseph
Smith in Parsons & Whittemore. (20) According to Judge Smith, public policy will only be
invoked if ‘enforcement would violate the forum State’s most basic notions of morality
and justice’. (21) Common law jurisdictions have largely adopted this concept. The leading
authority in Hong Kong states that awards should be given effect to ‘unless to do so would
violate the most basic notions of morality and justice [which] would take a very strong
case before such conclusion can be properly reached’. (22) Similarly, the Singapore Court
of Appeal (SGCA) held that the public policy exception can only be invoked where ‘the
upholding of an arbitral award would “shock the conscience” … or is “clearly injurious to
the public good or … wholly offensive to the ordinary reasonable and fully informed
member of the public” … or where it violates the forum’s most basic notion of morality
and justice’. (23) In the UK, although the phrase ‘most basic notions of morality and
P 276 justice’ has been referred to occasionally, (24) there is a general reluctance to provide

2
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 a working definition of public policy. (25) As Sir John Donaldson of the English Court of
Appeal puts it:
[c]onsiderations of public policy can never be exhaustively defined, but they
should be approached with extreme caution … It has to be shown that there is
some element of illegality or that the enforcement of the award would be
clearly injurious to the public good or, possibly, that enforcement would be
wholly offensive to the ordinary reasonable and fully informed member of the
public on whose behalf the powers of the State are exercised. (26)
9 Civil law jurisdictions, on the other hand, tend to refer to public policy as a set of
fundamental principles or values which undergird their respective societies. The German
Federal Court of Justice describes a breach of public policy as:
only occur[ing] if the arbitral award violates a norm that regulates the
foundations of state or economic life, or if it is in intolerable contradiction to
German ideas of justice. A mere violation of the substantive law or the
procedural law according to which the arbitral tribunal should decide is not
sufficient for such a violation. (27)
Austria defines its public policy as ‘fundamental values of the Austrian legal system’. (28)
The Belgium Court of Cassation observed that public policy is ‘what touches upon the
essential interests of the State or of the community or sets, in private law, the legal basis
on which rests the society’s economic or moral order’. (29)
10 Despite the diversity in expressions, we can distil at least two key characteristics
which are common to the definitions across the various jurisdictions: (1) it is narrowly
defined; and (2) it is subjectively ascertained through the lens of each jurisdiction. As we
shall see, these two characteristics will be central to the approach proposed in this
paper. (30)
2.2[b] Two Important Characteristics of the Public Policy Ground
2.2[b][i] Narrow Definition
11 Preliminarily, some jurisdictions apply different standards to the public policy
ground – depending on whether it is dealing with an enforcement challenge or an
annulment challenge. (31) For present purposes, this paper proceeds on the assumption
P 277 that the ‘public policy’ ground under both the setting aside and enforcement regimes
are treated in the same way.
12 Putting aside semantic differences, most jurisdictions appear to agree that the public
policy ground should be successfully invoked only in exceptional circumstances. (32)
13 This position is supported by commentators, (33) and it is aligned with the intention of
the Model Law drafters. (34) Indeed, the preparatory works of the Model Law confirm that
the term ‘public policy’ was ‘not equivalent to the political stance or international
policies of a State but comprised the fundamental notions and principles of justice’. (35)
This includes ‘instances such as corruption, bribery or fraud and similar serious cases’.
(36) Article V(2)(b) of the NYC has been interpreted in a similar manner. In October 2015,
the International Bar Association published a ‘Report on the Public Policy Exception in
the New York Convention’ which affirmed that the majority of jurisdictions adopt a narrow
interpretation of what public policy is and that it ‘require[s] a certain level of intensity
for a given circumstance to be held contrary to public policy’. (37)
14 Specifically, numerous jurisdictions have pointed out that not every allegation of
illegality would trigger the public policy of the state. For instance, the Supreme Court of
India stated: ‘In order to attract the bar of public policy, the enforcement of the award
must involve something more than the violation of the law of India’. (38) Likewise, the
P 278 Federal Court of Australia observed that: (39)

[t]he scope of the public policy ground of refusal is that the public policy to be
applied is that of the jurisdiction in which enforcement is sought, but it is only
those aspects of public policy that go to the fundamental, core questions of
morality and justice in that jurisdiction which enliven this particular statutory
exception to enforcement … [The public policy ground] should not be used to
give effect to parochial and idiosyncratic tendencies of the courts of the
enforcement state.
Commentators echo the same views: ‘the narrow construction of public policy implies
that not all failures to apply a nation’s mandatory law will fall within the public policy
exception’. (40)
15 On this point, the SGCA in AJU described illegality and public policy as ‘two strands of
the same principle’. (41) Such a description may give rise to the impression that the two
are equivalent concepts. However, as the authorities above point out, treating illegality
and public policy as equivalent or mirror concepts fail to distinguish between two
related but distinct issues. The first issue relates to the presence or absence of illegality.
The second issue relates to whether that alleged illegality is egregious enough to render

3
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 an award contrary to the fundamental public policy of the forum. Put simply, the two
concepts are not synonymous.
16 Indeed, that the SGCA most likely did not intend to imply that the two concepts are
synonymous is apparent from Westacre Investments Inc v. Jugoimport-SPDR Holding Co Ltd
[1999] 3 All ER 864 (“Westacre (CA)”) – which was cited and endorsed by AJU itself. In
Westacre (CA), the English Court of Appeal agreed that: (42)
… the [public policy] issue raises two separate questions; is it open to the
defendants in the enforcement proceedings to challenge the arbitrators’
findings of fact on the bribery issue, and secondly, if so and if successful in
proving [their] assertions … should the English court enforce the award?
The Court clearly recognized that not every form of illegality would justify a denial of
enforcement; they give rise to ‘two separate questions’. In fact, a crucial part of the
holding in that case was the distinction drawn between breaches of domestic public
policy and breaches of fundamental public policy. (43) Only breaches of the latter would
P 279 justify a denial of enforcement by the English courts. (44)

2.2[b][ii] Subjective Nature of the Ground

17 The second key characteristic is that the application of the public policy ground is
subjective in nature. In other words, whether there is a breach of public policy or not is
assessed through the lens of the relevant state. This is supported by a plain reading of
the Model Law and NYC. Both refer to the award being contrary to the public policy of
‘that country’ or ‘this state’ – which suggests that the notion of public policy is a
subjective one which should be assessed from the perspective of the individual states.
Public policy is ‘a nebulous concept that changes from State to State’. (45) One caveat is
that, instead of being completely subjective, it has been argued that a state must invoke
public policy in good faith. (46)
18 This subjective understanding of public policy is also clear on the face of the
definitions expressed by the various courts above: common law courts refer to public
policy as the forum state’s most basic notions of morality and justice, (47) whilst civil law
courts refer to fundamental principles which undergird their respective societies. (48)
Legislation and commentary also confirm the subjective nature of public policy. For
instance, Portuguese legislation refers to the ‘international public policy of the
Portuguese State’. (49) Likewise, when commenting on French legislation, Fouchard,
Gaillard and Goldman note: (50)
The international public policy to which Article 1502.5 refers can only mean
the French conception of international public policy or, in other words, the set
of values a breach of which could not be tolerated by the French legal order,
even in international cases.
Similarly, Lew describes public policy in the following terms: (51)
[I]t is clear that public policy reflects the fundamental economic, legal, moral,
political, religious and social standards of every State or extra-national
community. Naturally, public policy differs according to the character and
structure of the State or community to which it appertains.
All of this confirms that the content of the public policy ground remains subjective to
each state. Whilst there may be overlap between the public policies of different states,
each state is not necessarily seeking to identify and apply a common or harmonized
P 280 standard.
19 This appears, at first glance, to contradict what is referred to by many jurisdictions as
‘international public policy’ when describing the public policy ground of challenge. (52)
These jurisdictions restrict the scope of the public policy ground to what they describe as
‘breaches of international public policy’, as opposed to ‘breaches of domestic public
policy’. (53) It is suggested here that the word ‘international’ in this context is somewhat
of a misnomer because it gives rise to the misleading impression that the infringement in
question must be condemned by the international community at large. Such a class of
public policy is what some jurists describe as ‘truly international public policy’ or
‘transnational public policy’, (54) and is not to be confused with ‘international public
policy’.
20 Transnational public policy does not refer to the public policy of any one state, but
rather involves public policy that transcends state boundaries. Such public policy
‘aris[es] out of an international consensus regarding universal standards as to norms of
conduct that are generally recognized and agreed upon as unacceptable in most civilized
countries, such as slavery, bribery, piracy, murder, terrorism, and corruption’. (55)
However, as explained above, transnational public policy was not what was
contemplated by the plain wording of the Model Law and NYC. (56)
21 Unlike transnational public policy, international public policy is not meant to connote
a fixed or uniform standard common to all civilised nations. (57) Instead, it is a policy
applied by a state when dealing with foreign arbitral awards (this foreign dimension is

4
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 perhaps why it is described as ‘international public policy’). (58) It contains a
jurisdiction’s most important domestic values – the breach of which cannot be tolerated
by its legal order, even in international cases that otherwise have little connection to that
P 281
jurisdiction. (59) In Sanders’ words: ‘international public policy, according to a
generally accepted doctrine is confined to violation[s] of really fundamental conceptions
of legal order in the country concerned’. (60) This is consistent with a subjective
understanding of the public policy ground.
22 For completeness, it may be part of a state’s practice or policy to recognize or give
effect to certain decisions or policies of foreign states in certain situations (61) – which
indeed finds expression in some forms of the doctrine of comity – but that does not
detract from the proposition that, at least for the purposes of recognition and
enforcement of an arbitral award, each state applies its own public policy.
23 Having dealt with the principles behind the public policy ground, we now turn to
consider the practical manifestations of public policy breaches.

2.3 Three primary manifestations of public policy breaches

24 There are at least three different ways public policy challenges can be raised: (1) the
remedies ordered by the award are contrary to public policy; (2) the award gives effect to
an underlying contract that is illegal or tainted with illegality; and (3) how the award was
procured is contrary to public policy. This section provides an overview of each category
and explains how they may (or may not) give rise to the issue of having to re-open a
tribunal’s findings.
2.3[a] Remedies Ordered by an Award are Contrary to Public Policy
25 The first (and least common) manifestation of a public policy breach is found on the
face of the award itself. Such challenges are premised on the remedies granted in the
award being contrary to public policy. For example, in ICC Case No 5946, a claim for
punitive damages was refused in an arbitration taking place in Geneva, on the basis that
damages beyond compensatory damages constitute a punishment of the wrongdoer
contrary to the public policy of the seat (viz. Swiss public policy). (62) Although this case
dealt with public policy at the pre-award stage, one could certainly imagine such a claim
P 282
arising in a post-award context. More recently, in Hardy Exploration, (63) an award was
refused enforcement in the United States on public policy grounds because the award
issued an order which would violate the national sovereignty of another state. The
tribunal there issued an award which ordered India to allow a private oil and gas
exploration company into its territorial waters for three years to continue its exploration.
The US District Court for the District of Columbia refused enforcement, holding that the
award’s ‘forced interference with India’s complete control over its territory violates
public policy’. (64)
26 In such cases, there will rarely be an issue of re-opening the tribunal’s findings
because tribunals are unlikely to pronounce on the compatibility of its remedies with a
state’s public policy. Generally speaking, the reviewing court would be considering the
public policy challenge in the first instance and thus can pronounce on the issue without
even having to look at the merits of the case.
27 However, that is not to say the issue of re-opening will never arise in this category. The
issue can still arise if the tribunal explicitly finds that its award is not contrary to the
public policy of the reviewing court. By way of illustration, consider ICC Case No 5946
referred to above. (65) Assume the tribunal in that case found that a claim for punitive
damages was not contrary to Swiss public policy, and thus orders punitive damages
against the respondent. If the respondent seeks to set aside the award in the Swiss courts
on the same public policy grounds, that would require the Swiss courts to re-examine an
issue which has already been decided by the tribunal. As shall be seen below, this
situation illustrates why some of the prevailing approaches are not flexible enough to
provide the Swiss courts with an appropriate remedy.
2.3[b] Award Gives Effect to an Underlying Contract That Is Illegal or Tainted with
Illegality
28 The public policy objection in this second category stems from the underlying contract
in the award. In practice, this is the most common manifestation of a public policy
P 283 challenge. (66) The argument is that an award which is underpinned by an illegal
contract should be set aside (or denied enforcement) because recognizing such an award
would amount to the recognition of an illegal contract – and that would be contrary to the
public policy of the forum. (67) This is conceptually one step removed from the first
category because it is the underlying contract that is objectionable, not the award itself.
29 The issue of re-opening an award is most common in such cases because the legality of
the underlying contract would often have been an issue that the tribunal has dealt with.
Therefore, to deal with a public policy challenge in this category, reviewing courts would
generally have to first decide if it should re-open the issue of illegality. Given the
prevalence of such cases, most prevailing approaches were designed specifically with
this category in mind. It is suggested here that a common defect across all the prevailing
approaches (68) arises because they fail to provide a universal solution that can be

5
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 applied across all the three different categories of public policy objections identified
above. (69)
30 As with the first category, there will be exceptional situations where the issue of re-
opening the tribunal’s findings does not arise under the second category. For example,
consider a situation where a tribunal finds that the underlying contract was illegal but
nonetheless permits the claim. The reviewing court in such a case would not have to re-
open the tribunal’s decision on illegality in order to determine that the award is contrary
to its public policy. The court can simply use the tribunal’s own finding of illegality as a
basis to invoke the public policy exception to refuse enforcement. This was precisely
what happened in the English Court of Appeal case of Soleimany. (70) The underlying
contract in that case was for carpets to be illegally exported from Iran and sold in other
countries. The Beth Din (a Jewish tribunal) found that the underlying contract was indeed
illegal under the law of the place of performance (Iranian law), but still enforced the
contract as the tribunal found that the procedural law of the arbitration (Jewish law)
attached no significance to the illegality. The Court refused enforcement because it was
contrary to the public policy of England to enforce an award which upheld an illegal
contract. (71) This would be true regardless of the procedural law of the arbitration. (72)
Since it was the tribunal itself which found that the underlying contract was illegal, the
P 284 Court could refuse enforcement without having to re-open the finding of the tribunal.

2.3[c] The Process of Procuring the Award is Contrary to Public Policy

31 The third and last category involves allegations that the process in which the award
was procured is contrary to public policy. This is sometimes referred to as a breach of
‘procedural public policy’. (73) Examples of such breaches include corruption or fraud
perpetrated on (or by) the tribunal. (74)
32 Unlike the first two categories, re-opening a tribunal’s findings under this third
category is generally unobjectionable. (75) This is because tribunals are not in a position
to adjudicate on their own procedural deficiencies – they are either complicit in the
deficiency or did not even have an opportunity to take notice of it. Hence, when a
reviewing court is confronted with a procedural public policy challenge, it can determine
the issue afresh even in the unlikely event that the tribunal had already pronounced on
the issue. (76) In such cases, concerns of finality usually take a back seat. (77)
33 Given these justifications, there is very little controversy surrounding the re-opening of
an award for breaches of procedural public policy. All of the prevailing approaches are in
agreement that, under this category, the reviewing court should be allowed to reopen the
P 285 findings of the award. (78)

3 WHEN CAN A NATIONAL COURT RE-OPEN A TRIBUNAL’S FINDINGS UNDER

THE ‘PUBLIC POLICY’ GROUND?
34 In the previous section, we have set out the relevant context by discussing the
conceptual basis of the public policy ground, as well as its specific manifestations.
Against this background, we now turn to address the central issue in this paper: under
what circumstances can a national court re-open a tribunal’s findings under the public
policy ground?
35 There are three broad approaches which have emerged from the case law: (1) the
Maximal Review Approach; (2) the Contextual Review Approach; and (3) the Minimal
Review Approach. In this section, we explain these approaches and identify their
deficiencies wherever relevant.

3.1 Maximal Review Approach

3.1[a] Content of the Maximal Review Approach
36 Under this first approach, a national court will always embark on a de novo
examination of the award afresh, whenever a public policy objection is raised (Maximal
Review Approach). As long as there is an allegation to this effect, the court will engage in
a full rehearing of the tribunal’s findings which are related to the public policy objection.
This effectively translates into two defining features. First, the scope for permitting a
review is very wide. The reviewing court does not have to agree with the severity of the
allegation, and the challenging party does not have to substantiate its allegation. (79)
Secondly, the extent of review is relatively intrusive. A de novo examination involves a
complete rehearing of the relevant issues afresh, without being bound by the tribunal’s
findings in any way. (80) The reviewing court can evaluate the evidence as if for the first
time, and make its own factual findings. (81) This may entail accepting new evidence,
even if such evidence was reasonably available (yet not produced) at the time of the
P 286 arbitration. (82) In this regard, new evidence may be contrasted against fresh
evidence. Unlike new evidence, fresh evidence could not have been obtained at the time
of the arbitration despite reasonable diligence. (83) There is therefore less justification
for the introduction of new evidence in the post-award context. As we shall see, this
distinction between new and fresh evidence is important when we consider the other
approaches below. (84)

6
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 37 No court has explicitly adopted such an extreme stance – although there are some
cases which ostensibly apply it. (85) One such case is the Singapore High Court decision in
Rockeby. (86) That case involved an agreement where the claimant was to provide
consultancy services to the respondent. The respondent argued that this agreement was
illegal under the Securities and Futures Act of Singapore, which provides for licensing
requirements for persons who provide capital market services in Singapore. (87) The
claimant, in turn, argued that it was an exempt financial advisor and therefore did not
require a license. The tribunal found in favour of the claimant – that the agreement was
not illegal. The respondent then sought to set aside the award on the same grounds that
it was illegal and thus contrary to Singapore’s public policy. The court then re-opened the
tribunal’s finding on the issue of illegality, without considering whether this was an
appropriate case to do so. (88) In the learned Judge’s view, ‘the court has the inherent
jurisdiction to reconsider the issue of illegality although the arbitral tribunal may have
determined that the contract in question was not illegal’. (89) However, Rockeby did not
involve a de novo examination and was, in any event, quickly overruled by the SGCA in
AJU just a few months later. (90) Since then, the Maximal Review approach has not yet
resurfaced in common law jurisdictions. In civil law jurisdictions, the Maximal Review
approach was celebrated most notably in France during the 1990s. (91) This is presumably
because of the scepticism surrounding arbitration during that time. (92) At that time, it
was believed that commercial arbitrations should not be allowed to usurp the exclusive
P 287
domain of the courts in adjudicating matters and applying the law. (93) However, since
the early 2000s, (94) French judicial attitudes have gradually shifted with the times,
toward a more Minimal Review approach. (95)
38 Interestingly, a recent French Cour de Cassation decision has been described as
endorsing France’s return to the ‘maximalist review’ approach. (96) In Belokon v.
Kyrgyzstan, the highest court in France upheld the Paris Court of Appeal decision to set
aside a USD15m award on public policy grounds. The claimant in that case was a Latvian
citizen who invested in a local bank (‘Manas Bank’) based in the Kyrgyzstan (i.e., the
respondent state). This investment was made in 2007, against the backdrop of a bilateral
investment treaty (the ‘BIT’) between Latvia and Kyrgyzstan – which sought, amongst
other things, to encourage investment across the two countries and safeguard investors’
rights. However, in 2010, the Kyrgyz National Bank issued a decree which essentially
suspended the activities of Manas Bank. This was justified on the basis that the bank was
engaged in money laundering and other criminal activities. The claimant thus brought a
claim under the BIT for the respondent to compensate him for losses. The tribunal issued
an award of roughly USD fifteen m in favour of the claimant, and the respondent sought to
set aside the award in France. After being satisfied that there was a ‘serious, precise and
concordant indication’ (97) of money laundering, the court proceeded to engage in a new
investigation into the merits of the case, ignoring the tribunal’s finding on the same issue.
Although the French courts here may have been very thorough and intrusive in their re-
examination of the issue, (98) this is arguably not an adoption of the Maximal Review
approach in its purest form. The courts did not proceed to re-examine the issue on the
basis of a mere allegation alone. Instead, it only engaged in a re-examination of the
tribunal’s findings after being satisfied that there was a ‘serious indication of illegality’.
(99) Indeed, the Maximal Review Approach not only describes how extensively the courts
review the tribunal’s decision, but also how easily the courts will engage in a review of the
tribunal’s decision.
3.1[b] Deficiencies of the Maximal Review Approach
39 The biggest issue with this approach is that it sets too wide a scope for parties seeking
a backdoor to judicial review on the substantive merits of an arbitral award. This
P 288 contradicts the principle of respecting the finality of arbitral awards and minimal
curial intervention. (100) Reyes IJ aptly describes the issue in the following terms: (101)
Public policy is often invoked by a losing party in an attempt to manipulate an
enforcing Court into re-opening matters which have been (or ought to have
been) determined in an arbitration. The public policy ground is thereby raised
to frustrate or delay the winning party from enjoying the fruits of a victory. The
Court must be vigilant that the public policy objection is not abused in order
to obtain for the losing party a second chance at arguing a case. To allow that
to happen would be to undermine the efficacy of the parties’ agreement to
pursue arbitration. That by itself would not be conducive to the public good.
If all it takes is a simple allegation, that would further perpetuate the issue highlighted
by Reyes IJ. To prevent such abuse, there must be gateways to cross before one can re-
open a tribunal’s findings.
40 The second issue relates to the extent of review permitted under this approach. A full
de novo rehearing of the public policy issue is arguably too intrusive. In particular,
accepting new (and not just fresh) evidence too freely at the post-award stage would
make a mockery of the arbitral proceedings. Uncooperative award debtors would be able
to prolong enforcement unnecessarily by introducing new evidence at the post-award
stage without being able to justify the delay. This erodes the viability of arbitration as an
efficient mode of dispute resolution.

7
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 41 Recent developments in the UK vis-à-vis jurisdictional challenges lend further support
to this critique. After a thorough review of the English Arbitration Act and engagements
with stakeholders, (102) the UK Law Commission has recommended various amendments.
(103) One key proposal relates to the narrowing of jurisdictional challenges – such that
they should operate by way of an appeal and not a rehearing. (104) Unlike a rehearing, an
appeal does not allow parties to adduce new evidence which was not before the tribunal.
(105) In support of this reform, the Commission provides two main reasons. (106) First, it
P 289 reduces delay and costs from repetition. (107) Secondly, it prevents the jurisdictional
hearing before the arbitral tribunal from becoming a ‘dress rehearsal’ for the challenging
party to plug the gaps identified. (108) It is suggested here that these two reasons apply
with equal force when considering the deficiencies of the Maximal Review approach.

3.2 Contextual review approach

3.2[a] Content of the Contextual Review Approach
42 The Contextual Review Approach was first formulated by Waller LJ in Soleimany (109)
which his Lordship further developed in his dissenting judgment in Westacre (CA). (110) In
Westacre, the contract was performed through the exercise of improper influence and
bribery and was therefore void on public policy grounds. The tribunal held that these
allegations were not made out and thus issued an award in favour of the claimant. In
resisting the enforcement of the award in England, the respondent made three separate
public policy arguments:
(1) First, the underlying agreement was a contract for the purchase of personal
influence. (111)
(2) Second, Westacre gave perjured evidence at the arbitration. (112)
(3) Third, the agreement was intended to be used as a vehicle for bribery from its very
inception. (113)
43 In deciding whether (and the extent to which) it should re-open the illegality findings
of the tribunal, Waller LJ formulated the following contextual approach which requires
the court to be satisfied with certain preliminary conditions before it can embark on a
more elaborate enquiry into the tribunal’s findings on illegality. (114)
44 In essence, this approach proceeds in three main stages. (115) First, the court must be
satisfied that there is at least prima facie evidence of illegality. (116) Second, the court
must then conduct a preliminary enquiry and determine whether to give full faith and
P 290 credit to the award, having regard to the following factors:
(1) whether there is evidence on the other side to the contrary; (117)
(2) whether there was anything to suggest that the arbitrator was incompetent to
conduct such an inquiry; (118)
(3) whether the arbitrator expressly found that the underlying contract was not illegal;
(119)
(4) whether it was a fair inference to suggest that he did reach that conclusion; (120)
(5) whether there may have been collusion or bad faith, so as to procure an award
despite illegality; (121) and
(6) whether the level of opprobrium justifies re-opening the case. (122)
Third, it is only if the court is satisfied that the award is unsafe that it should then embark
on a more elaborate inquiry into the issue of illegality. (123)
45 Unfortunately, Waller LJ did not elaborate further on what ‘a more elaborate inquiry’
under the third stage entails. (124) In particular, his Lordship did not specify whether new
evidence (which was reasonably available at the time of the arbitration) could be
considered under this ‘more elaborate inquiry’. (125) Hwang, however, suggests otherwise.
(126) In his description of the Contextual Review approach, Hwang says ‘Waller LJ made
clear that in assessing whether there was illegality (under [the second or third stages]),
the court was not limited to considering fresh or new evidence; it may even consider
evidence that had been put before the tribunal’. (127) He explains in his footnote (128)
that this can be discerned from Soleimany’s departure from Colman J’s sixth proposition
P 291 in Westacre (HC). (129)
46 It is suggested here that Soleimany’s departure only indicates that a reviewing court
may consider facts which were placed before the arbitrators when conducting the ‘more
elaborate inquiry’. (130) It does not suggest that a reviewing court may also consider facts
which are new but not fresh. (131) In fact, Waller LJ most likely did not intend for a
reviewing court to consider new but not fresh evidence. Just a few paragraphs before his
Lordship restated the Contextual Review approach in Westacre (CA), Waller LJ rejected
Jugoimport’s attempt to adduce additional evidence because Jugoimport failed to show
that the evidence was not available at the time of the arbitration. (132) Although this
holding (133) was made in the context of an application for leave to amend arguments,
(134) it demonstrates Waller LJ’s inclination to reject new (as distinct from fresh) evidence
when a court embarks on the ‘more elaborate enquiry’.
47 Putting aside the ambiguity of the third stage, Waller LJ’s approach essentially seeks to
balance the competing interests by only re-opening a limited scope of cases. For

8
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 example, a court applying the Contextual Review approach may reopen a case if there
are strong reasons to doubt the credibility of the award. This appears to be what factors
(1) and (2) address. Another example of when re-opening may be permitted under
Contextual Review is where the tribunal has not even dealt with or pronounced clearly on
the issue of illegality. In such a case, concerns of finality may be overblown since the
reviewing court will not contradict the findings of a tribunal. This situation appears to be
what factors (3) and (4) above seek to address.
48 Re-opening may also be permitted when the level of opprobrium (or seriousness) of
the alleged illegality is sufficiently high. This is of course a direct manifestation of factor
(6). Presumably, Waller LJ included this factor because a national court should be more
inclined to review and ensure that the tribunal’s decision is right when the stakes are
higher. (135) However, the other judges did not agree. Their Lordships viewed this factor
as irrelevant at the second stage, holding that it is something to be taken into account
P 292 only when deciding whether the award should be enforced. (136) Unfortunately, the
majority did not provide any further reasons why this would be an irrelevant factor at the
second stage.
3.2[b] Deficiencies of the Contextual Review Approach
49 Conceptually, this approach may seem like an attractive one to take. It attempts to
strike a balance between respecting the finality of awards, (137) whilst also
acknowledging the role played by national courts in upholding the public policy of the
forum. (138) However, as with all flexible approaches, it is arguably difficult to apply in
practice. (139)
50 First, it is not clear what weight should be placed on the different factors, as well as
the threshold that needs to be crossed before a court can decide not to ‘give full faith
and credit to the award’. In practice, the situations are usually not as neatly delineated
as the examples presented above. (140) The reality (and difficulty) of Contextual Review
is that it is a multi-factorial approach, and the different factors may push and pull in
different directions. This problem is exaggerated by the fact that, unlike other multi-
factorial tests, (141) it is extremely difficult to come to a binary conclusion for most of the
factors under the Contextual Review approach. (142) To illustrate, consider factor
(2) – which deals with the competence of the tribunal. In order to conclude whether this
factor weighs in favour or against re-opening the case, the court would have to draw an
arbitrary threshold for the level of incompetence which would justify re-opening an
award. It is also conceptually difficult to ascribe a value to the competence of a tribunal
solely by reference to other tribunals. Each tribunal may comprise different arbitrators
who bring different strengths and weaknesses to the table.
51 Second, there is also the danger of slipping into a full-scale review at the preliminary
inquiry stage. In Steel J’s words: ‘[t]he difficulty with the concept of some form of
preliminary inquiry is of course assessing how far that inquiry has to go’. (143) Factors (1),
(2), (3) and (4) involve value judgments relating to the robustness of the tribunal’s
decision. This undoubtedly involves some degree of analysis into the merits of the
P 293 dispute, especially if counsel are invited to make submissions on those factors.
52 Hwang and Lim, however, find this second concern overstated. (144) In their view, a
clear enough distinction can be drawn between the preliminary inquiry in the second
stage and the full-scale review in the third stage. (145) In essence, they argue that the
second stage should only be triggered if the award’s lack of credibility can be ‘reasonably
easily perceived’. (146) If ‘lengthy submissions and complex argumentation’ are required
to raise suspicions, then the court should give full faith and credit to the award and not
proceed to a full-scale enquiry at the third stage. (147)
53 However, these guidelines may be too broad to guard against the risk of slipping into a
full-scale review. Lengthy submissions and complex argumentation are not atypical in
commercial disputes where the stakes are high. What do reasonable suspicions entail?
How long and complex do the arguments have to be before a court should reject them
outright? These are questions that will involve difficult value judgments. It is fully
recognized here that a judge’s role is precisely to make difficult value judgments, but in
this context any challenging party will undoubtedly seek to cast as much doubt as
possible on the tribunal’s reasoning. In practice, dealing with such argumentation will
very likely shade into a de facto full-scale review.
54 Third, the preliminary inquiry is somewhat circular in logic. This is particularly true of
factor (1) – which invites counsel to explain why the tribunal was wrong. Considering such
a factor at the preliminary inquiry stage is logically flawed because the whole point of
the preliminary inquiry is to determine if such explanations should even be submitted
for consideration by the reviewing court in the first place.
55 Fourth, as Grierson points out in relation to factor (5), there is much practical difficulty
in showing ‘incompetence’, ‘collusion’, or ‘bad faith’ – especially so when such
deficiencies have to be shown from a cursory audit of the tribunal’s reasoning at the
preliminary enquiry stage. (148)
56 Last, the Contextual Review approach is not flexible enough to deal with all categories
of public policy breaches. Put another way, it does not prescribe a universal solution to

9
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 P 294 deal with all forms of public policy challenges. In fact, it was formulated to deal only with
a specific form of public policy breach (viz. where there is an underlying contract that
is illegal). (149) This is evident from the phrasing of the Contextual Review approach
itself. (150) However, as explained earlier, (151) public policy breaches can take different
forms. It is not necessarily limited to a case which involves an illegal underlying contract.
57 For example, in a case where the public policy objection is directed at the remedies
ordered by an award, (152) the Contextual Review approach is not flexible enough to
provide a satisfactory remedy. This is because the preliminary inquiry stage focuses
primarily on factors relating to the credibility of the award and the seriousness of the
allegation. (153) It pays no attention to the type of pronouncement made by the tribunal.
In other words, the fact that a tribunal pronounced wrongly on the content of the forum’s
public policy is not a relevant factor in itself under the Contextual Review approach. As
long as there is nothing to question the credibility of the award, a reviewing court
applying the Contextual Review approach is expected to yield to a tribunal’s
determination of what its own public policy entails. (154) This will obviously not be
satisfactory to any reviewing court.
58 Another lacuna in the Contextual Review approach emerges when a court deals with an
award where the tribunal erroneously finds that a contract is illegal. Such a situation
would not be captured by the Contextual Review approach because it only permits a
review when the tribunal erroneously finds that a contract is not illegal. (155) This stems
from the first stage, which can only succeed where ‘there is prima facie evidence of
illegality’. (156) It does not contemplate the situation where there is prima facie evidence
P 295 of legality. At first blush, this may seem to be a purely theoretical issue. The
assumption is that a claim would be dismissed in a case where the tribunal finds that a
contract is illegal. Accordingly, one would assume that there would be no award to
enforce in such cases. However, case law has suggested that a negative declaratory award
may still be set aside. (157) This is therefore not a theoretical issue. It is entirely possible
for courts to be faced with such cases, (158) and the Contextual Review approach would
leave those cases without a satisfactory remedy.

3.3 Minimal Review Approach

59 Finally, we consider the most widely endorsed Minimal Review approach, (159) which
gives the greatest degree of deference to the findings of arbitral tribunals. Under this
approach, awards will be upheld (or enforced) as a starting point, and it is only in
exceptional cases that courts will relook the issue of illegality. (160) However, even within
this approach, different courts have adopted different definitions of what constitutes an
‘exceptional case’. (161)
3.3[a] Betamax’s Conception of Minimal Review
60 The most straightforward (and narrow) conception of the Minimal Review Approach is
found in the recent UK Privy Council case of Betamax. (162) That case involved a dispute
between the claimant shipping company (‘Betamax’) and the trading arm of the
Mauritian state (‘STC’). In 2009, the parties entered into a Contract of Affreightment (‘CoA’)
under which Betamax took over the transport of Mauritius’ oil imports from India. A new
government took over STC in 2010 and decided to terminate the CoA on the basis of
serious infringements in the procurement process. In May 2015, Betamax commenced
arbitral proceedings for STC’s breach of contract. STC’s principal defence was that the
CoA breached public procurement rules and was thus illegal and unenforceable.
Therefore, STC argued that Betamax could not rely on the CoA to claim
P 296 damages – because this would be tantamount to enforcing an illegal contract.
Betamax responded by asserting that the CoA was not illegal because it fell within the
exemption prescribed under the Mauritian public procurement legislation. The arbitrator
found that STC was indeed an exempted organization under Mauritian legislation and
that there was accordingly no illegality. The CoA was therefore legal and enforceable, and
Betamax was awarded damages of USD 115 million. STC then applied to set aside the
award for being in conflict with Mauritius’ public policy.
61 The Privy Council held that a national court cannot review the findings of a tribunal
unless the tribunal’s decision-making process has been tainted with ‘fraud, a breach of
natural justice, or any other vitiating factor’. (163) In other words, re-opening a tribunal’s
decision is only justified in cases of procedural public policy breaches. (164) Although the
Board left open the possibility of other ‘exceptional cases’ where a court may be entitled
to review a tribunal’s decision, (165) the line was largely drawn at the point where a
breach of procedural public policy can be established.
62 This approach suffers from several deficiencies. For one, confining the exceptional
cases only to cases of procedural public policy may lead to unpalatable outcomes. For
instance, consider a case that deals with a contract for the sale of human body parts.
Assume the tribunal finds that the contract is not illegal under its proper law and renders
an award on that basis. It would be extremely discomforting to know that reviewing
courts at the award enforcement stage are barred from reviewing that decision unless
there is a breach of procedural public policy.
63 A less extreme illustration can be found in the case of Hardy Exploration. (166) To

10
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 recapitulate, that case dealt with an award where India was ordered to allow a private oil
and gas exploration company into its territorial waters for three years to continue its
exploration. Assume that the tribunal there dealt with a public policy argument, and
erroneously held that it was not against US public policy to issue an award infringing
upon the territorial sovereignty of India. If the Betamax conception of minimal review is
P 297 applied, (167) the US courts would be forced to enforce an award which they believe is
contrary to their state’s public policy. (168) This would be an unsatisfactory outcome.
The US courts – as the judicial arm of the government – should not be expected to yield
to a commercial arbitrator’s determination on what US public policy entails.
64 Another deficiency of Betamax’s conception of the Minimal Review Approach was
noted by Poon. (169) He argues that such a strict approach may result in abuse. In his
view, if a court can only review errors made by the tribunal where there is a recognized
vitiating factor, then ‘parties can endeavour to commit acts which would be a breach of
public policy yet reap the rewards of such a breach by slipping in a favourable
arbitration clause’. (170) This would leave arbitration open to abuse by commercial
parties seeking to hide illegal conduct by laundering it through an arbitral award. (171)
Hwang and Lim echo similar sentiments. (172) They argue that the ‘laissez-faire’ nature of
Minimal Review ‘leaves courts open to shameless exploitation by wily and unscrupulous
claimants seeking judicial assistance to enforce their corrupt schemes and other
nefarious wrongdoing’. (173)
65 Lastly, much like the Contextual Review approach, (174) Betamax’s conception of
Minimal Review is not flexible enough to deal with all forms of public policy challenges.
Lord Thomas’s formulation suggests that the approach applies only where the tribunal
has already pronounced on the (il)legality or public policy issue(s) in question. (175) It is
not clear to what extent the Betamax approach can or should be applied in a case where
the tribunal has not dealt with the specific allegations made before the reviewing court.
This is problematic because crafty award debtors can avoid the restrictive Minimal
Review approach by taking a different factual ‘spin’ on essentially the same public policy
objection at the post-award stage. In fact, this was precisely what happened in Westacre
(CA). (176) There, the award debtor initially argued before the arbitral tribunal that
Westacre had performed the underlying contract through bribery. (177) Before the English
P 298 enforcement court, it instead argued that the entire purpose of the underlying contract
from its inception, was to provide funds for the bribery of governmental officials. (178)
Betamax’s conception of Minimal Review arguably could not be applied on the facts of
Westacre because the tribunal had not yet pronounced on the specific (il)legality or
public policy issue before the court. (179)
3.3[b] Majority in Westacre (CA)’s Conception of Minimal Review
66 In Westacre (CA), Mantell LJ (delivering the judgment of the majority) suggested that
the presence of fresh evidence may also be an exceptional case allowing a court to re-
open an award. (180) However, his Lordship did not specify whether there had to be fresh
evidence of fraud or bribery in particular, or fresh evidence of any breach of public policy.
It is suggested here that Mantell LJ was likely envisioning the former. This is because his
Lordship appears to have restricted the exception of fresh evidence to the
‘circumstances’ of the case he was dealing with. In his words: (181)
From the award itself it is clear that bribery was a central issue. The allegation
was made, entertained and rejected. … [I]n those circumstances and without
fresh evidence, I would have thought that there could be no justification for
refusing to enforce the award.
67 If we construe Westacre (CA) as requiring fresh evidence of bribery in particular, then
this exception is arguably too restrictive. Indeed, Hwang and Lim went as far as to say
that this is the ‘fatal weakness’ of the Minimal Review approach. (182) In their view, the
requirement of fresh evidence of fraud ‘will usually be impossible for victims of
corruption to procure’. (183) As a result, ‘under almost no circumstances will an award be
capable of review … no matter how corrupt the services performed under the contract
were in reality’. (184)
68 However, if Mantell LJ was instead proposing a general exception of fresh evidence,
one can agree with the proposition in principle. Generally speaking, fresh (and not merely
P 299 new) (185) evidence of any public policy breach should be a basis for re-opening an
award. (186) As shall be seen below, (187) this is why the notion of fresh evidence has been
incorporated into the approach proposed by this paper.
3.3[c] AJU’s Conception of Minimal Review
69 AJU’s conception of the Minimal Review Approach is less straightforward. (188) That
case was concerned with the legality of a settlement agreement which required the
respondent in the arbitration to ‘take such action as [was] necessary to withdraw and/or
discontinue certain criminal proceedings’ filed by the respondent against officers of the
claimant with the Thai prosecution authority. This settlement agreement was governed by
Singapore law. At the arbitration, the respondent failed to convince the tribunal that the
settlement agreement was illegal as it sought to stifle the prosecution of forgery in
Thailand.

11
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 70 The respondent then sought to set aside the award on essentially the same
ground – that the award was contrary to public policy as it sought to uphold a settlement
agreement which stifled Thai prosecution. Although the Singapore High Court
acknowledged the principle of finality, it nonetheless thought that this was an
‘appropriate case’ for the court to ‘safeguard the countervailing public interest’ and thus
re-opened the issue on illegality. On appeal, the SGCA had to consider whether the judge
below was right ‘in going behind the … award and reopening the tribunal’s finding that
the [settlement agreement] was valid and enforceable’. Although Chan CJ (as he then was)
ostensibly endorsed the Minimal Review approach, (189) he added one qualification.
Specifically, Chan CJ opined that the public policy ground for setting aside awards only
applied to ‘findings of law made by an arbitral tribunal – to the exclusion of findings of
fact’. (190) In this regard, when dealing with the legality of an agreement governed by
Singapore law, the court held that its:
supervisory power extends to correcting the [t]ribunal’s decision on [the] issue
of illegality … as the court cannot abrogate its judicial power to the [t]ribunal
to decide what the public policy of Singapore is and, in turn, whether or not
the [agreement] is illegal (illegality and public policy being … mirror
concepts) (191)
This appears to suggest that the Singapore courts can freely review all findings of law
P 300 relevant to a public policy challenge, even if no recognized vitiating factor can be
established. Apart from the fact that a bright line distinction between questions of fact
and questions of law can be very hard to draw in practice, (192) the UK Privy Council in
Betamax noted that such a suggestion ‘went further than was necessary for the decision in
the case and is inconsistent with the judgement read as a whole’. (193) As Ng explains,
such an expansive reading of AJU would arguably be inconsistent with the decision of AJU
itself, ‘where the Singapore Court of Appeal held that the interpretation of a Singapore
law agreement should not be reviewed by the Singapore court’. (194)
71 It is suggested here that these criticisms of AJU likely stem from a failure to delineate
clearly between the two distinct issues described above. (195) In particular, issues of
legality and issues of public policy are distinct, and not ‘mirror concepts’. (196) Properly
understood, Chan CJ’s description of an ‘error of law’ should be narrowly construed as an
error relating only to the content of the forum’s public policy. It should not be understood
as any error of law with a public policy dimension. Under this interpretation, an error
relating to the legality of an underlying contract is not capable of review because it does
not deal with the content of the forum’s public policy. A similar view is taken by Ng, who
perceptively points out Chan CJ’s precise description of what a question of law is. (197) In
Chan CJ’s words: ‘[i]t is a question of law what the public policy of Singapore is. An arbitral
award can be set aside if the arbitral tribunal makes an error of law in this regard’. (198)
Put another way, one should understand AJU to stand for the proposition that the
Singapore court can freely review findings made by an arbitral tribunal on the content of
Singapore’s public policy (and not all findings of law with a public policy dimension). This
would reconcile AJU with Betamax’s interpretation of the same. (199)
72 In principle, this paper agrees that this narrow exception, namely the review of a
tribunal’s pronouncement on the content of the reviewing court’s public policy, should
also be a basis for re-opening the tribunal’s award. However, as with the other
P 301 conceptions of Minimal Review, it should not be the only basis for review.

3.3[d] CBX’s Conception of Minimal Review

73 In a more recent decision before the Singapore International Commercial Court, Reyes
IJ in CBX v. CBZ attempted to interpret AJU’s minimal review approach. (200) CBX involved
certain share purchase agreements which were governed by Thai law. The tribunal issued
an award which ordered compound interest on the award sum. The respondent sought to
set aside the award on the basis that the compound interest order was contrary to Thai
mandatory law, and thus upholding that order would be contrary to Singapore public
policy. (201) In support of this argument, the respondent cited AJU for the proposition that
the courts can intervene where the tribunal made an error of law. In this regard, Reyes IJ
interpreted AJU as giving rise to the following four scenarios. (202) These four scenarios
are best represented in the following matrix (see Table 1.1). (203)
Table 1.1 Table summarising different scenarios laid out in CBX v. CBZ
Scenario Law Governing the Tribunal’s Finding on Whether Re-opening
Underlying Contract the Legality of the by Reviewing Court is
Underlying Contract Permissible
#1 Singapore law Not illegal Yes
#2 Singapore law Illegal No
#3 Foreign law Illegal No
#4 Foreign law Not illegal No unless there is
‘palpable and
indisputable
illegality’

12
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 74 Several observations can be made on this matrix. First, Scenario #1 may have
overstretched the holding in AJU. Here, Reyes IJ argues that a Singapore court can
intervene in a tribunal’s finding of legality because ‘as the supervisory court, it “cannot
abrogate its judicial power to the Tribunal to decide what the public policy of Singapore
P 302 is” (AJU v. AJT … at [62])’. (204) However, as argued above, (205) when AJU is understood
in context, the court’s judicial power to scrutinize an award should specifically be limited
to ‘what the public policy of Singapore is’. (206) This is conceptually distinct from
intervening in questions relating to the illegality of the underlying contract. Simply put,
the two are not ‘mirror concepts’. (207) It is therefore arguable that the court intervention
permitted by Scenario #1 may be too broad.
75 The second observation relates to Scenarios #2 and #3. Here, the Court observed that
where a tribunal finds that the contract is illegal, then parties must live with that
decision and have no avenue of appeal. In this regard, the Court essentially distinguished
between ‘false positive’ awards and ‘false negative’ awards. In ‘false positive’ awards, the
tribunal has decided that an underlying contract is not illegal – when in actuality, it may
have been illegal. (208) On the other hand, in ‘false negative’ awards, the tribunal has
decided that an underlying contract is illegal – when in actuality it may have been
perfectly legal. Through Scenarios #2 and #3, the Court categorically held that ‘false
negative’ awards should not be subject to review, but did not provide further justification
for drawing this distinction. (209) Nevertheless, the assumption is that in ‘false negative’
cases there would be no avenue for any national court to review the tribunal’s finding
that the contract was illegal because the claim would have already failed. This makes
sense if the failed claimant does not pursue the matter any further and there is nothing
to enforce. However, that is not always the case. Where a claim is dismissed for illegality,
the tribunal may issue an award denying the claim (i.e., a negative declaratory award).
This is still an award which can be set aside if the failed claimant can show that the
tribunal came to an erroneous conclusion. (210) To the extent the Court believes that a
reviewing court can intervene in ‘false positive’ awards, there is arguably no principled
reason why the Court cannot also intervene in ‘false negative’ awards. After all, both of
P 303 them have equal potential to raise a public policy objection. (211)
76 The last observation relates to the distinction the Court sought to draw between
Scenarios #1 and #4. Under Scenario #4, where a tribunal has decided that the underlying
contract is not illegal under a foreign law, the Singapore court should not intervene in
that decision. This is presumably because the Singapore courts are not in a position to
correct an error of foreign law; that should be left to the foreign court to correct. However,
this view appears to miss the point behind the public policy ground. As seen from the
various possible manifestations described above, (212) breaches of Singapore public
policy are not necessarily confined to cases where the underlying contract is governed by
Singapore law. It can also occur when the underlying contract is governed by a foreign
law. (213) It may be too blunt or artificial to deny intervention in cases simply because
the underlying contract is governed by foreign law.

4 PROPOSED APPROACH
77 Having considered all the deficiencies identified above, this section suggests that
courts can adopt an alternative approach (the Proposed Approach). It is submitted that
this approach strikes the best balance between the concerns highlighted above, whilst
also furthering the courts’ legitimate interest in safeguarding public policy. The Proposed
Approach proceeds in the following two stages.

4.1 Stage One: Would the public policy of the reviewing forum be triggered?
78 Under Stage One, the central question for each court is: do the allegations
made – even if proven – trigger the fundamental public policy of the forum? At this stage,
each court will proceed on the assumption that the allegations made by the challenging
party are true. There is no need to conduct any form of review of the tribunal’s analysis at
this first stage.
79 However, national courts should be careful in applying this stage, always bearing in
mind the two key characteristics of the public policy ground identified above. First, in
order to safeguard the finality of awards, the public policy ground should only be
successfully invoked in exceptional circumstances. (214) Mere illegality does not suffice.
(215) The challenging party will have to demonstrate how the alleged illegality is so
egregious that the forum should not tolerate the award. This will aid in sieving out a
P 304 majority of the applications. Secondly, courts should also remember that the inquiry
here is subjective in nature. (216) In other words, an alleged illegality which involves an
egregious breach of the laws of a foreign jurisdiction can but does not necessarily trigger
the public policy exception in the reviewing forum. Litigants will therefore have to
articulate precisely how the public policy of the reviewing forum has been breached by
the foreign illegality.

4.2 Stage Two: Full review of the tribunal’s analysis

80 Only if the court finds that the public policy of its own jurisdiction has been triggered
should it then proceed to Stage Two. Here, it will perform a full review of the tribunal’s

13
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 analysis on the specific issues giving rise to the public policy objection.
81 However, this should operate as a review and not a rehearing. As explained earlier, a
full rehearing (typically referred to as a ‘de novo’ hearing) involves a re-examination of
the public policy issue afresh. This may entail accepting new evidence. (217) A review,
however, is limited to an evaluation of the tribunal’s existing analysis based on
arguments made by parties during the arbitration. Hence, a court should not ordinarily
accept new evidence, unless the evidence was not reasonably available at the time of
the arbitration. (218) A corollary of this is that a challenging party will have to sustain the
allegations it made at Stage One based on the factual findings of the arbitrator. It cannot
introduce new evidence unless it can show that such evidence could not have been
obtained with reasonable diligence at the time of the arbitration. This will limit a
challenging party’s ability to re-cast or re-characterize its allegations at Stage One. After
all, it will have to substantiate whatever allegations it makes from the limited pool of
evidence.
82 In practical terms, a ‘rehearing’ under Stage Two would only be permitted in cases
where the evidence for the alleged impropriety genuinely becomes known only after the
arbitral proceedings. In such cases, a ‘rehearing’ is justified because the claim is made
entirely on fresh evidence which was not reasonably available at the time of the
arbitration. There can be no rehearing under the Proposed Approach if evidence was
already tendered during the arbitration, but the tribunal decides not to discuss them,
sees them as irrelevant, or deals with them in a cursory manner. Such cases are better
dealt with under a claim for a breach of natural justice, (219) and should not unduly
P 305 broaden the scope of the public policy ground.
83 Critically, Stage Two does not entail a full review of the entire award. Courts should
only re-open an award to the extent needed to decide on the specific allegation(s) that
passed muster under Stage One. Otherwise, the public policy ground would be subject to
abuse by recalcitrant award debtors seeking a back door to re-litigate their entire case.
If, in the course of the review, a court notices another error that is unrelated to the public
policy objection, it cannot correct it. If such errors do not impinge on public policy, there
is no normative reason to go against the principle of finality of awards. After all, those
who opt for arbitration ‘must live with the decision of the arbitrator, good or bad’. (220)

4.3 Application of the Proposed Approach

84 In order to illustrate how the Proposed Approach operates, this next section will apply
the Proposed Approach to the facts of actual cases below.
4.3[a] Application of Proposed Approach to Betamax Ltd v. State Trading Corporation
[2021] UKPC 14
85 Betamax (221) concerned a dispute between a shipping company (‘Betamax’) and the
trading arm of the Mauritian state (‘STC’). In 2009, the parties entered into a ‘CoA’ under
which Betamax took over the transport of Mauritius’ oil imports from India. A new
government took over STC in 2010 and decided to terminate the CoA on the basis of
serious infringements in the procurement process. In May 2015, Betamax commenced
arbitral proceedings for STC’s breach of contract. STC’s principal defence was that the
CoA breached public procurement rules and was thus illegal and unenforceable.
Therefore, STC argued that Betamax could not rely on the CoA to claim
damages – because this would be tantamount to enforcing an illegal contract. Betamax
responded by asserting that the CoA was not illegal because it fell within the exemption
prescribed under the Mauritian public procurement legislation. The arbitrator found that
STC was indeed an exempted organization under Mauritian legislation and that there was
accordingly no illegality. The CoA was therefore legal and enforceable, and Betamax was
awarded damages of USD 115 million. STC then applied to set aside the award for being in
conflict with Mauritius’ public policy.
86 Under Stage One of the Proposed Approach, the public policy of Mauritius would likely
P 306 be triggered by STC’s illegality allegation. Although the Privy Council deliberately
avoided this question, (222) it did suggest that the Mauritius Supreme Court would be
best positioned to decide whether an illegal CoA would be contrary to its forum’s public
policy, (223) and the Mauritius Supreme Court decided that it was. (224) In other words, if
one assumes that the CoA was illegal for failing to comply with public procurement
legislation, then the Mauritius courts will likely find that its public policy has been
infringed. Stage One of the Proposed Approach will therefore be passed in Betamax.
87 Now that one proceeds to Stage Two, the courts will engage in a full review of the
relevant issues. Here, STC would have to substantiate its allegation that the CoA was
indeed illegal. This is where STC would likely fail in its challenge. Indeed, after a very
comprehensive review of Mauritius’ public procurement legislation spanning 40 full
paragraphs, (225) the Privy Council decided that the CoA was not illegal because it was
exempted from the procurement regime. (226) Applying the Proposed Approach, STC will
fail to substantiate its allegation of illegality at Stage Two, and the award will be upheld.
4.3[b] Application of Proposed Approach to Westacre Investments Inc v. Jugoimport-SPDR
Holding Co Ltd [1999] 3 All ER 864

14
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 88 In Westacre, the dispute centred on an arms-trading agreement. Under this
agreement, a Panamanian consultancy firm (‘Westacre’) agreed to help procure for a
Yugoslavian government agency (‘Jugoimport’) contracts to sell military equipment to the
government of Kuwait. A dispute arose as to the fees payable, and Westacre initiated
arbitral proceedings against Jugoimport. In response, Jugoimport alleged that the
agreement was performed through the exercise of improper influence and bribery in
Kuwait, and the agreement is thus void on public policy grounds. The arbitral tribunal
held that Jugoimport’s allegations were not made out and thus made an award in favour
of Westacre.
89 In resisting the enforcement of the award, Jugoimport made three separate public
policy arguments. First, England’s public policy was breached because the underlying
P 307 agreement was a contract for the purchase of personal influence. (227) Second,
Westacre gave perjured evidence at the arbitration, and this fraud is contrary to public
policy. (228) Third, the agreement was intended to be used as a vehicle for bribery from
its very inception. (229)
90 These arguments appeared to have been advanced disjunctively. (230) In other words,
it was argued that each argument alone was sufficient to invoke the public policy of
England. (231) As shall be seen below, (232) the disjunctive structure of the arguments will
influence how Stage One of the Proposed Approach is applied. Interestingly, Jugoimport
also sought to adduce new evidence to support its second and third arguments. This new
evidence took the form of a sworn affidavit by one Miodrag Milosavljevic (‘MM’s Affidavit’).
(233)
91 Under Stage One, the reviewing court will consider these three allegations separately
because they were advanced disjunctively. (234) In other words, the reviewing court will
first consider if each of the allegations are individually capable of invoking the public
policy of the forum under Stage One. Only the allegations which are so capable will
require substantiation under Stage Two.
92 The first argument is likely to fail here. Although the Court of Appeal recognized that a
contract for the purchase of personal influence might be contrary to the domestic public
policy of the place of performance (viz. Kuwait), that in itself was not sufficient to invoke
the fundamental public policy of England. (235) For the fundamental public policy of
England to be invoked, the contract had to be contrary to both the domestic public
policy of the place of performance and the domestic public policy of the country of its
proper law or curial law. (236) This was not satisfied in relation to the first argument. A
contract for the purchase of personal influence was contrary to public policy in Kuwait
(the place of performance). However, it was not contrary to the public policy of
Switzerland (country of the curial law). (237) Consequently, English fundamental public
P 308 policy was not infringed and the first argument would fail under Stage One. There is no
need for the court to enquire into the veracity of this allegation under Stage Two.
93 On the contrary, the second and third arguments are likely to pass muster under Stage
One. This is because the preparatory works of the Model Law explicitly identify ‘bribery or
fraud’ as examples of fundamental public policy breaches. (238) However, they will fail at
Stage Two because they can only be sustained by new evidence. (239) As explained
earlier, (240) Stage Two operates by way of a review of the tribunal’s existing analysis. It
does not permit the admission of new evidence unless the evidence was not reasonably
available at the time of the arbitration. In this case, Jugoimport failed to show how the
new evidence (viz. MM’s Affidavit) could not have been reasonably obtained during the
arbitration. (241) Therefore, it cannot rely on MM’s Affidavit at Stage Two, and without
MM’s affidavit its second and third arguments cannot be sustained. (242) As a result,
applying the Proposed Approach, Jugoimport’s second and third arguments will fail at
Stage Two, and the award will be upheld.
4.3[c] Application of Proposed Approach to AJU v. AJT [2011] 4 SLR 739
94 The facts of AJU are more complicated. It concerned the legality of a subsequent
settlement agreement and not the underlying contract itself. The underlying contract in
this case was between a Thai production company (‘AJU’) and a British Virgin Islands
company (‘AJT’). Under this contract, AJU had the right to stage annual tennis tournaments
in Bangkok for a term of five years. Disputes arose out of this contract, and AJT
commenced arbitration against AJU. In the course of the arbitral proceedings, AJU made a
complaint of fraud and forgery to Thailand’s prosecution authorities against AJT. Under
Thai law, fraud is a compoundable offence whereas forgery was non-compoundable.
Generally, non-compoundable offences are more serious and thus the withdrawal of a
complaint may not guarantee termination of criminal proceedings or investigations. (243)
95 As the investigations were ongoing, parties concluded a settlement agreement. Under
this agreement, AJT agreed to terminate the arbitration and pay AJU a sum of USD 470,000
P 309 upon AJU’s withdrawal and/or discontinuation and/or termination of the Thai criminal
proceedings. However, after AJU withdrew its complaint, AJT did not keep to its side of the
bargain. AJU thus applied to terminate the arbitral proceedings on the grounds that the
parties had reached a full and final settlement. AJT responded by challenging the
settlement agreement on the basis of duress, undue influence, and illegality. In an
interim award, the tribunal rejected AJT’s allegations and upheld the settlement
agreement. AJT was thus compelled to terminate the arbitral proceedings and pay AJU

15
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 the settlement sum of USD 470,000.
96 AJT then applied to set aside the award, arguing that it was in conflict with Singapore
public policy. (244) In support of this ground, AJT made the following two allegations (245)
: (1) that the settlement agreement was an agreement to stifle the prosecution of non-
compoundable offences; and (2) that AJU performed the settlement agreement through
bribery and/or corruption of the Thai public authorities. (246) However, AJT did not
specify whether these two allegations were meant to be considered separately or
cumulatively. (247) For the purposes of this mock application, we will proceed on the
assumption that these allegations were mounted separately. (248)
97 Under Stage One, the reviewing court will thus consider the two allegations in turn.
Here, both allegations are likely to independently invoke the public policy of Singapore.
In fact, it was ‘common ground between the parties’ (249) that the settlement agreement
would be in conflict with the public policy of Singapore if any one of the two allegations
P 310 hold true. (250)
98 Proceeding to Stage Two, we now consider if each of these allegations can be
sustained without the introduction of new evidence. Regarding the first allegation, AJT is
unlikely to succeed in substantiating it. The SGCA shares this view. (251) In particular,
Chan CJ found that the settlement agreement did not require stifling the prosecution of
non-compoundable offences, and provides four reasons in support of his finding. (252) AJT
is also unlikely to succeed in substantiating its second allegation under Stage Two. As
alluded to earlier, (253) this allegation was rejected by the tribunal and the Singapore
High Court. (254) There was simply insufficient evidence to prove that AJU had bribed the
Thai public authorities. (255) Consequently, both allegations will fail at Stage Two, and
the award will be upheld under the Proposed Approach.
4.3[d] Application of Proposed Approach to Gokul Patnaik v. Nine Rivers [2020] SGHC (I) 23
99 Gokul concerned a protracted shareholder dispute between several parties. In
essence, Nine Rivers invested INR 300 million into an Indian company, alongside other
shareholders such as Gokul. Nine Rivers made this investment on the understanding that
it would be able to sell off its shares for a profit of at least INR 1.329 billion by a
stipulated deadline. Unfortunately, this did not happen. After several attempts at re-
negotiation, Nine Rivers still could not secure its exit from the Indian company. As such,
P 311 Nine Rivers sent a put option notice calling upon Gokul and other existing
shareholders to purchase its shares for the originally agreed sum of INR 1.329 billion.
Gokul and the other shareholders did not comply with this put option notice, and Nine
Rivers commenced arbitration. One of the arguments Gokul raised in defence was that
performance of the put option contravened Indian foreign exchange regulations and thus
could not be enforced. The tribunal rejected this argument, (256) finding that Indian
foreign exchange regulations were not contravened. (257) Therefore, Gokul and the other
shareholders were jointly and severally required to purchase Nine Rivers’ shares at the
put option price. Gokul then applied to set aside the award on public policy grounds. In
support of this, he continued to insist that the performance of the put option
contravened Indian foreign exchange regulations.
100 Gokul’s public policy challenge will likely fail at Stage One of the Proposed Approach.
This is because the breach of foreign exchange regulations is insufficient to trigger the
fundamental public policy of Singapore. As Ramsey IJ pointed out:
there is no reason why a breach of the [foreign exchange] [r]egulations or the
laws of India, without more, would ‘shock the conscience’ or violate the ‘most
basic notions of morality and justice’. If Mr Patnaik’s [Gokul’s] submissions are
taken to their logical conclusion, then any minor illegality or regulatory
infringement by a contract in its place of performance would ipso facto lead to
the conclusion that international comity, and thus Singapore public policy,
would be breached so that the arbitral award would have to be set aside. The
public policy ground under Art 34(2)(b)(ii) of the Model Law is a narrow ground
and does not lead to that conclusion. (258)
101 Since Gokul’s public policy challenge fails at Stage One, there is no need to proceed
to Stage Two. Applying the Proposed Approach, the reviewing court can dismiss the
public policy challenge.

5 JUSTIFICATIONS FOR THE PROPOSED APPROACH

102 The Proposed Approach is conceptually and practically preferable to the three
prevailing approaches because it resolves each of their respective deficiencies.

5.1 Justifications over the Maximal Review approach

103 First, in relation to Maximal Review, the two deficiencies identified above were: (1)
P 312 the scope for permitting a review is too wide and open to abuse (259) ; and (2) the
extent of review is too intrusive. (260) The first deficiency is resolved because Stage One
functions as a gateway to prevent the review of issues which do not even fall within the
narrow scope of the public policy ground. (261) This will sieve out abusive challengers

16
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 who seek to re-open a tribunal’s findings under the guise of public policy. It is recognized
here that there might still be a residual risk of abuse. For example, challenging parties
may exaggerate the alleged illegality at Stage One in order to bring the challenge within
the narrow confines of the public policy ground. This may enable them to re-open the
case under Stage Two. However, these potential abusers would have to substantiate their
exaggerated allegations at Stage Two without the introduction of any new evidence. If
they fail to do so, courts can consider imposing cost consequences to deter such abusive
attempts at delaying enforcement.
104 The second deficiency is also resolved because Stage Two restricts the scope of
courts’ review to the findings made by the tribunal itself and/or fresh evidence which was
not reasonably available at the time of the arbitration. (262) Such a level of intrusion is
arguably justified as it strikes the best balance between all interests at stake. On one
hand, the restrictive evidentiary scope balances the interests of award debtors in
repeating legitimate public policy objections, while also respecting the interests of
award creditors in not being unduly burdened with new arguments which should have
been raised at the arbitration. On the other hand, allowing a review of the public policy
challenge allows reviewing courts to safeguard their forum’s most fundamental principles
and values. This makes the Proposed Approach preferable to the Maximal Review
approach.

5.2 Justifications over the Contextual Review approach

105 The Proposed Approach also resolves the deficiencies inherent in the Contextual
Review approach. To recapitulate, the deficiencies of Contextual Review are: (1) it is
difficult to apply in practice; (263) (2) it runs the risk of slipping into a full-scale review;
(264) (3) it is circular in logic; (265) (4) there is a practical difficulty in showing ‘collusion
or bad faith’ under factor (5); (266) and (5) it is not flexible enough to deal with all forms
P 313 of public policy breaches. (267)
106 In relation to the first deficiency, the Proposed Approach is preferable because it is
much easier to apply in practice. Unlike the multi-factorial second stage of the
Contextual Review approach, Stage One of the Proposed Approach only focuses on one
factor (viz. whether the public policy of the forum is invoked). More importantly, in
contrast to the Contextual Review approach, the Proposed Approach does not involve any
arbitrary line drawing or abstract weighing. It simply asks whether the public policy of
the reviewing forum would be invoked. Although this question might involve some value
judgment, it is at least capable of producing binary conclusions – either the public policy
of the reviewing forum is invoked or not. The same cannot be said of the factors under the
Contextual Review approach. (268) In any event, the value judgment inherent in
answering the central question under Stage One must necessarily be answered in any
public policy challenge. This is therefore not a flaw of the Proposed Approach itself, but
rather a difficulty inherent in the very nature of the public policy ground. (269)
107 Likewise, the second deficiency does not feature in the Proposed Approach. There is
no risk of slipping into a full-scale review because Stage Two of the Proposed Approach
precisely entails a full-scale review. In this regard, there is no added benefit to adopting
the Contextual Review approach because it is typically a full review in disguise. As
explained above, (270) the preliminary inquiry from the Contextual Review approach
requires parties to make submissions on issues which are akin to the merits of the
dispute. In practical terms, this means parties do not save time and costs with the
Contextual Review approach because arguments on the tribunal’s findings will emerge
anyway. In contrast, the Proposed Approach can lead to more efficient applications by
expunging weak claims at the outset under Stage One.
108 In relation to the third deficiency, the Proposed Approach presents a more elegant
solution. By assuming the factual premises of the challenging party at Stage One, the
Proposed Approach does not run into the issue of circularity. At this stage, making factual
assumptions is warranted because these are facts which will eventually have to be
established under Stage Two. Indeed, the assumption is merely a tool for courts to
determine if a review of the tribunal’s findings is even worth undertaking in the first
place. This assumption breaks the circularity inherent in the Contextual Review
P 314 approach and makes the Proposed Approach preferable as a matter of conceptual
soundness.
109 In relation to the fourth deficiency, there is no practical difficulty in demonstrating
‘collusion or bad faith’ because its existence is simply not a relevant factor under Stage
One of the Proposed Approach. Under the Proposed Approach, a challenging party is not
faced with the uphill task of showing ‘collusion or bad faith’ solely from a cursory audit of
the tribunal’s findings. He can first make the allegation under Stage One, (271) and then
proceed to substantiate it with the benefit of a full review under Stage Two. This
overcomes the practical difficulty of having to establish ‘collusion or bad faith’ from a
mere preliminary inquiry under the second stage of the Contextual Review approach.
110 Lastly, in relation to the fifth deficiency, the Proposed Approach is capable of being
applied to all forms of public policy challenges. By way of illustration, consider the
applicability of the Proposed Approach to the two hypothetical cases which could not be
resolved under the Contextual Review approach. (272) First, the Proposed Approach can

17
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 apply to a case where the public policy objection is directed at the remedies ordered by
the award. Stage One does not discriminate between different types of public policy
objections. The only limitation is whether the objection would invoke the public policy of
the forum. For the same reasons, the Proposed Approach can also apply to a case where
the tribunal erroneously finds that a contract is illegal. As long as a challenging party can
demonstrate how such an erroneous finding invokes the public policy of the forum, (273) it
may be possible to successfully challenge such an award under the Proposed Approach.

5.3 Justifications over the minimal review approach

111 Finally, the Proposed Approach is arguably preferable to the Minimal Review
approach. As explained earlier, (274) the biggest deficiency of Minimal Review is that it is
too narrow. This can lead to unpalatable outcomes, (275) or perhaps even abuse by
P 315 parties seeking to launder their illegal conduct through an award. (276) The Proposed
Approach overcomes these issues by permitting a full review under Stage Two. (277) Here,
as long as its forum’s public policy is at stake, courts have the power to scrutinize an
award and ensure its compliance with the forum’s fundamental values. It will not have to
yield to a commercial arbitrator’s determination on what public policy entails.
112 In a paper in support of Minimal Review, (278) Ng has argued that Minimal Review
strikes the best balance between the interests of finality and public policy. (279) Most
notably, he points out the scenario where a tribunal reaches a reasonable conclusion that
is actually wrong. (280) In such scenarios, he argues that precluding a review under the
Minimal Review approach is justifiable because ‘all that can be said is that the challenge
court took a different view of the evidence’. (281)
113 However, that is perhaps the precise reason why a review should be permitted in the
first place. As the judicial arm of a government, courts should arguably have the final say
on matters relating to its forum’s public policy – even if it ends up affirming the
erroneous decision of the tribunal. Further, as a matter of principle, a review in such
scenarios is justifiable on the basis that public policy does not hinge on party autonomy;
it can be investigated by the courts ex officio. (282) In other words, where a state’s public
policy is genuinely engaged and at stake (which means the matter passes Stage One of
the Proposed Approach), parties’ preferences for finality should take a back seat. As
Waller LJ observes: ‘[w]here public policy is involved, the interposition of an arbitration
award does not isolate the successful party’s claim from the illegality which gave rise to
it’. (283) The concern of ‘creating an inroad into the interest in finality’ (284) is also
mitigated by the narrow scope inherent in Stage One of the Proposed Approach. Not
every allegation of illegality would trigger the public policy of the state and permit a
review under Stage Two. (285)

6 CONCLUSION
114 The issue of when and how extensively a national court can review a tribunal’s
decision under the public policy ground has always been a difficult issue. This paper has
P 316 attempted to unpack that issue by first analysing the conceptual basis of the public
policy ground of challenge. (286) Two key characteristics were distilled from a thorough
review of the relevant provisions in the NYC, Model Law, and existing case law: (1) that it
is narrowly defined; and (2) that it is subjective in nature. The paper then explored how
public policy challenges can manifest in a number of different ways. (287) Whichever
approach is ultimately chosen by the courts, it is argued that the approach should be
flexible enough to deal with the myriad ways in which public policy challenges may be
advanced.
115 Next, we considered the three prevailing approaches of Maximal Review, Contextual
Review, and Minimal Review. It was then argued that none of them provided a
satisfactory solution. In brief, the Maximal Review approach sets an unduly wide scope
for permitting a review and allows for an overly intrusive extent of review which violates
the finality of awards. The Contextual Review is too difficult to apply in practice, giving
rise to potentially arbitrary outcomes. It also runs the risk of slipping into a full-scale
review, is circular in logic, and is not flexible enough to deal with all forms of public
policy breaches. The Minimal Review approach is arguably too narrow and does not
empower courts with the necessary tools to address potentially serious incursions into
their most fundamental values or notions of morality and justice. Indeed, in adopting any
one approach, courts in the past were forced to ‘pick their poison’ and live with the
respective deficiencies of the chosen approach.
116 This paper suggests an alternative solution which minimizes the deficiencies that
previous approaches had to live with. It proceeds in the following two stages: (1) first,
courts will assume the factual veracity of the public policy allegations and, on that basis,
determine if its fundamental public policy is even triggered in the first place; (2) only
allegations which would trigger the fundamental public policy of the forum will proceed
to the second stage where the court will engage in a review of the tribunal’s analysis on
the relevant public policy issues(s). Unlike a de novo rehearing, a review is limited to the
arguments and evidence that was put forth before the arbitral tribunal. New evidence
will not ordinarily be accepted unless the evidence was not reasonably available at the
time of the arbitration.

18
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 117 It is suggested that this Proposed Approach mitigates the deficiencies of all the
prevailing approaches and offers a universal solution which all reviewing courts can
P 316 apply to all forms of public policy allegations.

References
*) Associate Prof. of Law (Practice), Singapore Management University; Director,
Breakpoint LLC; Door Tenant, Fountain Court Chambers. The author acknowledges
the support of the Singapore International Dispute Resolution Academy Email:
[email protected].
**) LLB (Summa Cum Laude), Singapore Management University; Justices’ Law Clerk,
Supreme Court of Singapore. The author is most grateful for the helpful comments
provided by Mr Devathas Satianathan and Mr Jeremy Chai on an earlier draft. The
views expressed in this article are solely the authors’ alone, and should not be
attributed in any way to their respective organizations Email:
[email protected].
1) I. Ng et al., Five Recurring Problems in International Arbitration: The Relationship
Between Courts and Arbitral Tribunals, 8(2) Indian J. Arb. L. 19, 20 (2020).
2) TMM Division Maritima SA de CV v. Pacific Richfield Marine Pte Ltd [2013] 4 SLR 972 [65].
3) A. Redfern & M. Hunter, Law and Practice of International Commercial Arbitration 422
(4th ed., Sweet & Maxwell 2004) (Redfern & Hunter); G. Born, International
Commercial Arbitration (3d ed., Kluwer Law International 2021) (Born).
4) Born, supra n. 3, at 3760; K. P Berger, The Modern Trend Towards Exclusion of Recourse
Against Transnational Arbitral Awards: A European Perspective, 12 Fordham Int’l L.J.
605 (1988); S. Shackleton, Challenging Arbitral Awards: Part III – Appeals on Questions
of Law (New Law Journal 1834 2002); Redfern & Hunter, supra n. 3, at 412; J. Lew et al.,
Comparative International Commercial Arbitration 731 (Kluwer Law International
2003); BLC v. BLB [2014] SGCA 40 [53]; Burchell v. Marsh [1854] 58 US 344 at 349;
Judgment of 6 October 2010, Albert Abela Family Foundation v. Joseph Abela Family
Foundation, 2010 Rev. Arb. 813, 814 (French Cour de Cassation); Soh Beng Tee & Co
Pte Ltd v. Fairmount Development Pte Ltd [2007] 3 SLR(R) 86 [65]; PT Asuransi Jasa
Indonesia (Persero) v. Dexia Bank SA [2007] 1 SLR(R) 597 (PT Asuransi) [57].
5) This duty is enshrined under Art. 34(2)(b)(ii) and Art. 36(1)(b)(ii) of the Model Law,
which are in pari materia to Art V(2)(b) of the New York Convention.
6) Betamax Ltd v. State Trading Corporation [2021] UKPC 14 (Betamax).
7) See s. 3 below.
8) United Nations Commission on International Trade Law, UNCITRAL Model Law on
International Commercial Arbitration 1985, UN Doc. A/40/17 (Model Law).
9) Convention on the Recognition and Enforcement of Foreign Arbitral Awards, New
York, 10 Jun. 1958, United Nations Treaty Series, vol. 330, no 4739 (NYC).
10) Some domestic legislation, however, permits appeals. For example, the Singapore
Arbitration Act 2001 (2020 Rev ed.) (Singapore AA) s. 49 provides a party to a purely
domestic arbitral proceeding an avenue of appeal to the Singapore courts on a
question of law arising out of an award. However, this right of appeal can be
expressly derogated from by agreement. See also English Arbitration Act 1996
(English AA), s. 69, which provides parties to an arbitral proceeding the right of an
appeal to the English courts on a question of law arising out of the award, unless
parties expressly provide otherwise.
11) Model Law, supra n. 7, Arts 34 and 36; NYC, supra n. 8, Art. V.
12) Born, supra n. 3, at 3603.
13) English AA, supra n. 9, s. 68(2)(g); French Code of Civil Procedure, Art. 1520(5);
Arbitration Law of the People’s Republic of China, 1994, Art. 58(6); Netherlands Code
of Civil Procedure, Art. 1065(1)(e); Swiss Law on Private International Law, Art. 190(2)
(e).
14) Richardson v. Mellish (1824) 130 ER 294, 303 per Burrough J, in the context of an action
to recover damages for the breach of an agreement. Although this phrase did not
originate from an arbitration – related case, many commentators have
subsequently used it to describe the unpredictable nature of the public policy
ground in arbitration. See e.g., S. A. F. Haridi & Mohamed S. Abdel Wahab, Public
Policy: Can the Unruly Horse Be Tamed?, 83(1) Arb. Int’l J. Arb. Mediation & Disp.
Mgmt. 35, 35–47 (2017); H. Arfazadeh, In the Shadow of the Unruly Horse: International
Arbitration and the Public Policy Exception, 13(1–4) Am. Rev. Int’l Arb. (2003); M.
Gearing, The Public Policy Exception – Is the Unruly Horse Being Tamed in the Most
Unlikely of Places? (Kluwer Arbitration Blog 17 Mar. 2011).
15) International Bar Association Subcommittee on Recognition and Enforcement of
Arbitral Awards, Report on the Public Policy Exception in the New York Convention 18
(Oct. 2015) (IBA Report).
16) Ibid., at 1.

19
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 17) Statutory definitions have been provided in the national arbitration legislation of
two notable jurisdictions: Australia and UAE. In Australia, s. 8(7A) of the 1974
International Arbitration Act provides that ‘the enforcement of a foreign award
would be contrary to public policy if: (a) the making of the award was induced or
affected by fraud or corruption; or (b) a breach of the rules of natural justice
occurred in connection with the making of the award’. In the UAE, Art. 3 of the Civil
Transactions Law states in general (not limited to the context of arbitration) that
public order ‘include[s] matters relating to personal status such as marriage,
inheritance and lineage, and matters relating to systems of government, freedom of
trade, the circulation of wealth, rules of individual ownership and the other rules
and foundations upon which society is based, in such a manner as not to conflict
with the definitive provisions and fundamental principles of the Islamic Sharia’.
18) IBA Report, supra n. 15, at 2; New York Convention on the Recognition and
Enforcement of Foreign Arbitral Awards of 19 June 1958 – Commentary 494 (R. Wolff
ed., Beck, Hart & Nomos 2012).
19) See IBA Report, supra n. 15, at 6–10 for a list of definitions provided by various civil
and common law jurisdictions.
20) Parsons & Whittemore Overseas Co. v. Societe Generale de L’Industrie du Papier (1974
US Court of Appeals) 508 F.2d 969 (Parsons & Whittemore).
21) Ibid., at 974 (emphasis added in italics by the authors).
22) Hebei Import and Export Corporation v. Polytek Engineering [1999] 2 HKC 205, 211C–D
(emphasis added in italics by the authors).
23) PT Asuransi Jasa Indonesia (Persero) v. Dexia Bank SA [2007] 1 SLR(R) 597 (PT Asuransi)
[59].
24) See e.g., Westacre Investments Inc v. Jugoimport-SPDR Holding Co Ltd [1999] 1 QB 740
(Westacre (HC)) 752.
25) M. Scherer, Draft IBA Country Report England, The Public Policy Exception under
Article V(2)(b) Methodological Approaches Country Report England, International Bar
Association 6 (2014).
26) Deutsche Schachbau v. Shell International Petroleum Co Ltd [1987] 2 Lloyds’ Rep 246,
254.
27) BGH (German Federal Court of Justice), 12 Jul. 1990 – III ZR 174/89, NJW 1990 3210.
28) OGH (Austrian Supreme Court of Justice), 24 Aug. 2011, 3 Ob 65/11x; OGH (Austrian
Supreme Court of Justice), 12 Oct. 2011, 3 Ob 186/11s; OGH (Austrian Supreme Court of
Justice), 28 Aug. 2013, 6 Ob 138/13g; OGH (Austrian Supreme Court of Justice), 26 Jan.
2005, 3 Ob 221/04b.
29) Belgium Court of Cassation, 18 Jun. 2007.
30) See ss 4 and 5 below.
31) Portugal, e.g., requires a higher level of affront in enforcement challenges compared
to annulment challenges: see IBA Report, supra n. 15, at 3–4. The Singapore Court of
Appeal, on the other hand, has held that ‘there is no difference between [the] two
regimes as far as the concept of public policy is concerned’: AJU v. AJT [2011] 4 SLR
739 (‘AJU’) [37].
32) Parsons & Whittemore, supra n. 20, at 974. RBRG Trading (UK) v. Sinocore International
[2018] EWCA Civ 838 [25]. Hebei Import & Export Corporation v. Polytek Engineering Co
Ltd (1999) 2 HKCFAR 111 per Litton PJ 118; Judgement of 30 October 2008, III ZB 17/08
(German Bundesgerichtshof); Judgment of 20 August 2008, 9 Ob 53/08x (Austrian
Oberster Gerichtshof); Judgment of 9 December 2016, Case No T 2675-14, (Svea Court
of Appeal) ¶40; Judgment of 29 April 2009, CG Impianti v. Bmaab & Son Int’l
Contracting Co., XXXV Y.B. Comm. Arb. 415 (Milan Corte di Appello) 416–7.
33) Born, supra n. 3, at 3611–3618 and 4000–4027; A. G. Maurer, The Public Policy
Exception under the New York Convention: History, Interpretation and Application 54
(2013); M. Campbell, Betamax v. STC: Alleged Illegality, Public Policy and the Model
Law, 17(2) Asian Int’l Arb. J. 183, 184 (2021), doi: 10.54648/AIAJ2021009 (accessed 19
Apr. 2024) (Campbell); A. V. Dicey, J. Morris & L. A. Collins, The Conflict of Laws 16, 150
(15th ed., Sweet & Maxwell 2018) (Dicey, Morris, & Collins).
34) Report of the United Nations Commission on International Trade Law on the Work of
its Eighteenth Session, Official Records of the General Assembly, A/40/17 (21 Aug.
1985).
35) Ibid., para. 296.
36) Ibid., para. 297.
37) International Bar Association Subcommittee on Recognition and Enforcement of
Arbitral Awards: Report on the Public Policy Exception in the New York Convention
(Oct. 2015) 18.
38) Renusagar Power Co. Ltd v. General Electric AIR 1994 SC 860, 880 (emphasis added in
italics by the authors).
39) Traxys Europe S.A. v. Balaji Coke Industry Pvt Ltd (No 2) [2012] 291 ALR 99 (Federal
Court of Australia) 105 (emphasis added in italics by the authors).
40) L. Villiers, Breaking in the ‘Unruly Horse’: The Status of Mandatory Rules of Law as a
Public Policy Basis for the Non-Enforcement of Arbitral Awards, 18 Austl. Int’l L.J. 156,
165 (2011) (Villiers).
41) AJU, supra n. 31, at [19].
42) Westacre Investments Inc v. Jugoimport-SPDR Holding Co Ltd [1999] 3 All ER 864
(Westacre (CA)) 887 (emphasis added in italics by the authors).
43) Ibid., at 876–877.

20
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 44) Ibid.
45) A. I. Okekeifere, Public Policy and Arbitrability under the UNCITRAL Model Law, 2(2)
Int’l Arb. L. Rev. 70 (1999).
46) See Stephan Schill & Robyn Briese, If the State Considers: Self-Judging Clauses in
International Dispute Settlement, 13 Max Planck Y.B. United Nations L. 61 at 97 (2009),
doi: 10.1163/18757413-90000037. (accessed 19 Apr. 2024).
47) See above at [8].
48) See above at [9].
49) Portuguese Voluntary Arbitration Law 2011, Art. 56(1)(b)(ii).
50) E. Gaillard & J. Savage, Fouchard, Gaillard and Goldman on International Commercial
Arbitration (Kluwer Law International 1999) para. 1648 (emphasis added in italics by
the authors).
51) J. Lew, Applicable Law in International Commercial Arbitration 3210 (Oceana 1978)
(emphasis added in italics by the authors).
52) French Code of Civil Procedure, Art. 1520(5°); Lebanese Code of Civil Procedure, Art.
814; Portuguese Voluntary Arbitration Law 2011, Art. 56(1)(b)(ii); Peruvian Arbitration
Law 2008, Art. 75(3)(b); Paraguayan Arbitration Law, Art. 46°(b). Judgement of 4
December 1992 (1997) XXII Yearbook 72 (Court of Appeal of Milan) where the court
held that the public policy referred to in Art V(2)(b) of the NYC is international public
policy.
53) See A. Sheppard, Interim ILA Report on Public Policy as a Bar to Enforcement of
International Arbitral Awards, 19(2) Arb. Int’l 217, 219–220 (2003), doi:
10.1093/arbitration/19.2.217 (accessed 19 Apr. 2024) (Interim ILA Report) (emphasis
added in italics by the authors).
54) P. Lalive, Transnational (or Truly International) Public Policy and International
Arbitration, in Comparative Arbitration Practice and Public Policy in Arbitration 257,
273 (P. Sanders ed., ICCA Congress Series No 3 1987). See also P. Mayer & A. Sheppard,
Final ILA Report on Public Policy as a Bar to Enforcement of International Arbitral
Awards, 19(2) Arb. Int’l 249, 251 (2003), doi: 10.1093/arbitration/19.2.249 (accessed 19
Apr. 2024) (Final ILA Report); H. Fazilatfar, Overriding Mandatory Rules in
International Commercial Arbitration 17–19 (Edward Elgar Publishing 2019) ELECD
2857; Villiers, supra n. 40, at 162; M. Moses, Public Policy: National, International and
Transnational (Kluwer Arbitration Blog 2018) (Moses).
55) Moses, supra n. 54, at 2.
56) See above at [17].
57) Hebei Import and Export Corporation v. Polytek Engineering [1999] 2 HKC 205, 216 C–D.
58) Moses, supra n. 54, at 2.
59) Villiers, supra n. 40, at 162.
60) P. Sanders, ‘Commentary’ in 60 Years of ICC Arbitration: A Look at the Future (ICC
Publishing 1984).
61) For instance, the Singapore Court of Appeal in Anupam Mittal v. Westbridge Ventures
II Investment Holdings [2023] SGCA 1 has taken the view that the term ‘public policy’
in s. 11 of the Singapore International Arbitration Act (Cap 143A, 2992 Rev ed.)
(Singapore IAA), relating to subject matter arbitrability was broad enough for the
Singapore courts to consider the public policy of other states specifically ‘where
this arises in connection with essential elements of an arbitration agreement’ (at
[48]). However, the Court noted that, in contrast, the provisions concerning setting
aside and enforcement of an award uses the term ‘public policy of Singapore’ (at
[48]).
62) ICC Award No 5946 of 1990 (YCA 1991).
63) Hardy Exploration & Production (India), Inc. v. Government of India, Ministry of
Petroleum & Natural Gas Civ. Action No 16–140 (D.D.C. 7 Jun. 2018) (Hardy
Exploration).
64) Ibid., at 113.
65) See above at [25].
66) See e.g., CHY v. CIA [2020] SGHC(I) 3 [18]; CBX v. CBZ [2020] 5 SLR 184 [45]; Gokul
Patnaik v. Nine Rivers Capital Ltd [2021] 3 SLR 22 (Gokul) [135]; Rockeby Biomed Ltd v.
Alpha Advisory Pte Ltd [2011] SGHC 155 (Rockeby) [17]; BBA v. BAZ [2020] 2 SLR 453 [27];
Bloomberry Resorts and Hotels Inc v. Global Gaming Philippines LLC [2021] 3 SLR 725
[100]; Rakna Arakshaka Lanka Ltd v. Avant Garde Maritime Services (Pte) Ltd [2019] 2
SLR 131 [99].
67) Ibid.
68) See below at [55], [63] and [107].
69) See above at [24].
70) Soleimany v. Soleimany [1999] QB 785 (Soleimany).
71) Ibid., at 800.
72) Ibid.
73) IBA Report, supra n. 15, at 12–13; Interim ILA Report, supra n. 53, at 238 et seq.; M. S.
Kurkela & S. Turunen, Due Process in International Commercial Arbitration 22 (2d ed.,
Oxford University Press 2010) (Kurkela & Turunen): ‘It is clear that due process or
procedural public policy also forms a part of public policy or ordre public’.; Report
of the United Nations Commission on International Trade Law on the Work of its
Eighteenth Session, Official Records of the General Assembly, A/40/17 (21 Aug. 1985),
at 296–297.

21
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 74) Singapore IAA, supra n. 61, s. 24(a); Australian International Arbitration Act, s. 19;
Indian Arbitration and Conciliation Act, s. 34.
75) Redfern & Hunter, supra n. 3, ss 10.36, 10.41, 10.75; Jugoimport v. Westacre, Swiss
Federal Tribunal’s Judgment of 30 Dec. 1994, ASA Bulletin (Swiss Decision of
Westacre) 226; Betamax Ltd v. State Trading Corporation [2021] UKPC 14 (Betamax)
[52]; National Oil Corporation v. Libyan Sun Oil Co., 733 F.Sup 800, 813–814 (D. Del.
1990); Biotronik Mess-und Therapiegereate GmbH & Co. v. Medford Med. Instrument
Co., 415 F.Sup 133, 137 (D.N.J. 1976); Westacre Inv. Inc. v. Jugoimport SPDR Holding Co.
Ltd [1998] 4 All E.R. 570 (Q.B.); Judgment of 26 Jan. 2005, XXX Y.B. Comm. Arb. 421
(Austrian Oberster Gerichtshof) (2005) Judgment of 2 Nov. 2000, 2001 NJW 373
(German Bundesgerichtshof); Judgment of 25 May 1992, 1993 Rev arb 91 (French Cour
de Cassation). Compare, Thomson CSF v. Frontier AG, Swiss Federal Tribunal’s
Judgment of 28 Jan. 1997, 16(1) ASA Bulletin 118 (1998) (Thomson v. Frontier Judgment
of 28 Jan. 1997) 130, where the Swiss Federal Tribunal refused to review the award,
even though the challenging party alleged that it was based on non-existent
evidence.
76) Ibid.
77) However, arguments of procedural public policy breaches may still be subject to
other evidentiary restrictions.
78) See s. 3 below for a more thorough analysis into the prevailing approaches.
79) See European Gas Turbines SA v. Westman International Ltd, Paris Court of Appeal
Judgment of 30 Sep. 1993, XX Y.B. Comm. Arb. 198 (1995) (Westman), where the court
embarked on a full analysis of allegations of fraud and perjury upon a simple
allegation.
80) B. A. Garner, Black’s Law Dictionary 837 (10th ed., Thomson Reuters 2014). See also
Sanum Investments Ltd v. Government of the Lao People’s Democratic Republic [2016]
5 SLR 536 [41].
81) A. Sayed, Corruption in International Trade and Commercial Arbitration 407–409 (1st
ed., Kluwer Law International 2004) (Sayed); M. Hwang & K. Lim, Corruption in
Arbitration – Law and Reality, 8(1) Asian Int’l Arb. J. 1 (2012) (Hwang & Lim) [139].
82) N. Enonchong, Enforcement in England of Foreign Arbitral Awards Based on Illegal
Contracts, Lloyd’s Mar. & Com. L.Q. 495, 514 (2000) (Enonchong) The Enonchong
citation should read: N. Enonchong, Enforcement in England of Foreign Arbitral
Awards Based on Illegal Contracts, Lloyd’s Mar. & Com. L.Q. 495, 514 (2000)
(Enonchong)
83) Westacre (CA), supra n. 42, at 878 per Waller LJ; Ladd v. Marshall [1954] 3 All ER 745,
748.
84) See below at [44]–[45] and [64]–[66].
85) See e.g., Westman, supra n. 79; Thomson CSF v. Frontier AG, Paris Court of Appeal’s
Judgment of 10 Sep. 1998, Revue de l’arbitrage (2001) (Thomson v. Frontier French
Judgment of 10 Sep. 1998) 583.
86) Rockeby Biomed Ltd v. Alpha Advisory Pte Ltd [2011] SGHC 155 (Rockeby).
87) Securities and Futures Act (Cap 289, 2006 Rev Ed).
88) Rockeby, at [18]–[19].
89) Ibid., at [18].
90) AJU, supra n. 31, at[60].
91) See e.g., Westman, supra n. 79; Tho mson v. Frontier, French Judgment of 10 Sep. 1998,
supra n. 85.
92) See generally, A. Redfern Chapter 2: The Changing World of International Arbitration,
in Practicing Virtue: Inside International Arbitration 44–51 (D. Caron ed., Oxford
University Press 2015).
93) Ibid.
94) SA Thalès Air Defense v. GIE Euromissile, Paris Court of Appeal’s Judgment of 18 Nov.
2004; SNF SAS v. Cytec Industries BV, French Cour de Cassation’s Judgment of 4 Jun.
2008.
95) The specific contours of the Minimal Review approach are still subject to debate
today and will be discussed further below at [58]–[73].
96) Belokon v. Kyrgyzstan, Appeal No 17–17.981, French Cour de Cassation’s Judgment of
22 Mar. 2022.
97) Ibid., at 14.
98) Ibid.
99) Ibid.
100) Born, supra n. 3, at 3760; Berger, supra n. 4, at 605; Shackleton, supra n. 4; Redfern &
Hunter, supra n. 3, at 412; J. Lew, L. Mistelis & S. Kroll, Comparative International
Commercial Arbitration 731 (2003); BLC v. BLB [2014] SGCA 40 [53]; Burchell v. Marsh
[1854] 58 US 344 349; Abela Family Foundation v. Joseph Abela Family Foundation,
French Cour de Cassation Judgment of 6 Oct. 2010, Rev. Arb. 813, 814; Soh Beng Tee &
Co Pte Ltd v. Fairmount Development Pte Ltd [2007] 3 SLR(R) 86 [65]; PT Asuransi Jasa
Indonesia (Persero) v. Dexia Bank SA [2007] 1 SLR(R) 597 (PT Asuransi) [57].
101) A v. R [2010] 3 HKC 67 [24]–[25].
102) See UK Law Commission’s press release on the consultation,
https://www.lawcom.gov.uk/project/review-of-the-arbitration-act-1996/#related
(accessed 19 Apr. 2024).
103) UK Law Commission, Review of the Arbitration Act 1996 Consultation Paper, Law
Commission Consultation Paper 257 (2022) (UK Consultation Paper).

22
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 104) Ibid., [8.23].
105) Ibid.
106) UK Consultation Paper, supra n. 103, at [8.29]–[8.31].
107) Ibid., at [8.29].
108) Ibid., at [8.31].
109) Soleimany v. Soleimany [1999] QB 785 (Soleimany).
110) Westacre (CA), supra n. 42.
111) Ibid., at 874. Waller LJ refers to this argument as ‘the Lemenda point’.
112) Westacre (CA), supra n. 42, at 871 and 877.
113) Ibid.
114) Soleimany, supra n. 70, 800F–H.
115) Ibid.
116) Ibid.
117) Ibid.
118) Ibid.
119) Ibid.
120) Ibid.
121) Ibid.
122) Waller LJ’s dissent in Westacre (CA), supra n. 42, at 886; Westacre Investments Inc v.
Jugoimport-SPDR Holding Co Ltd [1999] QB 740 (Westacre (HC)) 773A–E. compare,
majority in Westacre (CA), supra n. 42, at 887 who opined that this factor should not
be relevant at this second stage and should instead be considered at the third
stage.
123) Soleimany, supra n. 70, 800F–H.; Waller LJ’s dissent in Westacre (CA), supra n. 42), at
886; Westacre (HC), supra n. 122, 773A–773E.
124) Soleimany, supra n. 70, at 800; Westacre (CA), supra n. 44, at 881–882.
125) Ibid.
126) Hwang & Lim, supra n. 81, at [155].
127) Ibid. (emphasis added in italics by the authors).
128) See footnote 375 of Hwang & Lim, supra n. 81.
129) Ibid. Colman J’s sixth proposition was: ‘[i]f the party against whom the award was
made then sought to challenge enforcement of the award on the grounds that, on
the basis of facts not placed before the arbitrators, the contract was indeed illegal,
the enforcement court would have to consider whether the public policy against the
enforcement of illegal contracts outweighed the countervailing public policy in
support of the finality of awards in general and of awards in respect of the same
issue in particular’ (emphasis added in italics by the authors). Waller LJ in
Soleimany disapproved of Colman J’s sixth proposition in the following terms: ‘But,
in an appropriate case [the court] may inquire, as we hold, into an issue of illegality
even if an arbitrator had jurisdiction and has found that there was no illegality. We
thus differ from Colman J, who limited his sixth proposition to cases where there were
relevant facts not put before the arbitrator’ (emphasis added in italics by the
authors).
130) See n. 125 above, which shows that Waller LJ’s disapproval of Colman J’s sixth
proposition related only to the consideration of facts which were already put before
the arbitrator. It did not relate to the consideration of new but not fresh evidence.
131) The difference between new and fresh evidence has been explained above at [36].
132) Westacre (CA), supra n. 42, at 881.
133) The majority agreed with this portion of Waller LJ’s judgment. See Westacre (CA),
supra n. 42, at 887 per Mantell LJ.
134) Westacre (CA), supra n. 42, at 877–881.
135) See E D & F Man (Sugar) Ltd v. Yani Haryanto (No 2) [1991] 1 Lloyd’s Rep 429, 436,
which Waller LJ cited with approval in Westacre (CA), supra n. 42, at 885.
136) Westacre (CA), supra n. 44, at 887–888.
137) See above at [2].
138) Campbell, supra n. 33, at 183.
139) See Westacre (CA), supra n. 42, at 316 where Mantell LJ expresses his ‘difficulty with
the concept and even greater concerns about its application in practice’.
140) See above at [46]–[47].
141) For instance, consider the multi-factorial approach under the forum conveniens test
in private international law. Under that multi-factorial test, courts can reasonably
come to a binary conclusion for each factor (viz. personal connections of parties;
location of events or transactions; governing law; etc.).
142) Compare, multi-factorial approach under the forum conveniens test in private
international law, where the court can come to a binary conclusion for each factor.
143) R v. V [2008] EWHC 1531.
144) Michael Hwang SC & Kevin Lim, Corruption in Arbitration – Law and Reality, 8(1) Asian
Int’l Arb. J. 1 (2012), doi: 10.54648/AIAJ2012001 (accessed 19 Apr. 2024) (Hwang & Lim)
[182].
145) Ibid., at [184].
146) Ibid.
147) Ibid.
148) J. Grierson, Court Review Of Awards On Public Policy Grounds: A Recent Decision Of The
English Commercial Court Throws Light On The Position Under The English Arbitration
Act 1996, 24 Mealey’s Int'l Arb. Report 2 (2009).

23
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 149) See Soleimany, supra n. 70 800F–H which describes the entire approach from the
perspective of an allegedly illegal contract underlying an award.
150) The seminal formulation of the Contextual Review approach by Waller LJ in
Soleimany, supra n. 70, at 800 states: ‘In our view, an enforcement judge, if there is
prima facie evidence from one side that the award is based on an illegal contract,
should inquire further to some extent. Is there evidence on the other side to the
contrary? Has the arbitrator expressly found that the underlying contract was not
illegal? Or is it a fair inference that he did reach that conclusion? Is there anything
to suggest that the arbitrator was incompetent to conduct such an inquiry? May
there have been collusion or bad faith, so as to procure an award despite illegality?
Arbitrations are, after all, conducted in a wide variety of situations; not just before
high-powered tribunals in international trade but in many other circumstances. We
do not for one moment suggest that the judge should conduct a full-scale trial of
those matters in the first instance. That would create the mischief which the
arbitration was designed to avoid. The judge has to decide whether it is proper to
give full faith and credit to the arbitrator’s award. Only if he decided at the
preliminary stage that he should not take that course does he need to embark on a
more elaborate inquiry into the issue of illegality’ (emphasis added in italics by the
authors).
151) See above at [24], where we describe three primary manifestations of public policy
breaches.
152) See e.g., Hardy Exploration, supra n. 63, where an award was refused enforcement
because it issued an order which would violate the national sovereignty of another
state
153) See above at [43] for a list of the relevant factors under the Contextual Review
approach.
154) By way of example, consider the hypothetical ICC Case No 5946 discussed above at
[25].
155) See Soleimany, supra n. 70, at 800; Westacre (CA), supra n. 42, at 882.
156) Ibid., (emphasis added in italics by the authors).
157) See West Tankers Inc v. Allianz SpA [2012] EWCA Civ 27 where the English Court of
Appeal upheld the High Court’s decision to enforce a negative declaratory award
under s. 66 of the English Arbitration Act 1996.
158) See below at [72], for a deeper analysis into the difference between the two types of
findings.
159) Besides the jurisdictions discussed below, this also appears to be the approach of
courts in India (see Associate Builders v. Delhi Development Authority [2014] 4 ARBLR
307 (SC)); Thailand (see Supreme Court of Thailand Decision No 6411/2560 [2017]);
and Hong Kong (see X v. Jemmy Chien [2020] HKCFI 286). compare, Z v. Y [2018] HKCFI
2342.
160) Hwang & Lim, supra n. 81, at [121].
161) Ibid.
162) Betamax, supra n. 75.
163) Ibid., at [52].
164) See above at [31].
165) Betamax, supra n. 75, at [52].
166) See above at [25].
167) In Betamax, Lord Thomas interpreted AJU as allowing a court to intervene in
‘determination[s] of the nature and extent of public policy’: see Betamax, supra n.
75, at [39]. However, he stopped short of endorsing this narrow exception.
168) The assumption here is that an award infringing upon the territorial sovereignty of a
foreign friendly state falls within the narrow scope of fundamental public policy.
169) N. Poon, Striking a Balance Between Public Policy and Arbitration Policy in
International Commercial Arbit ration, Sing. J. Legal Stud. 186 (2012) (Poon).
170) Ibid., at 190.
171) See Soleimany, supra n. 70, at 800.
172) Hwang & Lim, supra n. 81, at [179].
173) Ibid.
174) See above at [55].
175) In Lord Thomas’ formulation of the Minimal Review approach in Betamax, supra n.
75, at [52], his lordship states: ‘More likely, as appears from the decided cases and
observations made in them, are cases where the arbitral tribunal has expressly
considered issues which have required the arbitral tribunal to inquire into
circumstances suggesting illegality and set out their reasons for holding as a matter
of fact and of law that there was no illegality. In cases of that kind, the arbitral
tribunal’s decision on fact and on law is a decision for the arbitral tribunal, if within
its jurisdiction; if it holds that the contract is not illegal, then that decision will be
final, in the absence of fraud, a breach of natural justice or any other vitiating
factor’ (emphasis added in italics by the authors).
176) Westacre (CA), supra n. 42.
177) Ibid., at 868.
178) Ibid., where Waller LJ states that the arbitrators themselves acknowledge that ‘the
[award debtor] never put their case that way during the arbitration’.

24
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 179) More recently, the Hong Kong Court of First Instance, citing Betamax with approval,
remitted awards back to the tribunal for the tribunal to consider how changes to the
law of illegality in Hong Kong announced just a few days before the release of the
first award would affect the awards: G v. N [2023] HKCFI 3366.
180) Westacre (CA), supra n. 42, at 887.
181) Ibid., (emphasis added in italics and bold italics by the authors).
182) Hwang & Lim, supra n. 81, at [179].
183) Ibid.
184) Ibid.
185) For the differences between fresh and new evidence, see above at [36].
186) However, it should not be the only basis for reviewing an award.
187) See below at [77].
188) The facts of AJU v. AJT are canvassed in fuller detail below at [X].
189) In particular, that an award will only be re-opened if there is 'fraud, a breach of
natural justice or any other vitiating factor’. See Betamax, supra n. 75 at [52].
190) AJU, supra n. 31, at [69] (emphasis added in italics by the authors).
191) Ibid., at [62].
192) See M. J. Beazley, The Distinction Between Questions of Fact and Law: A Question
Without Answer?, Land and Environment Court Conference Kiama (2013), for a flavour
of the blurred distinction between questions of fact and questions of law in the
context of environmental law.
193) Betamax, supra n. 75, at [39].
194) M. Ng, Reviewing the Standard of Curial Review for Findings in Arbitration Involving
Public Policy, Sing. J. Legal Stud. 75, 83 (2022) (Ng).
195) See above at [15].
196) Ibid.
197) Ng, supra n. 194, at 84. See also Rakna, supra n. 66, at [99]–[100].
198) AJU, supra n. 31, at [67] (emphasis added in italics by the authors).
199) See Betamax, supra n. 75, at [39], where Lord Thomas understood AJU as standing for
the following proposition: ‘the determination of the nature and extent of public
policy was a question of Singapore law for determination by the courts of
Singapore’.
200) CBX v. CBZ [2020] 5 SLR 184 (‘CBX v. CBZ’).
201) CBX v. CBZ [52].
202) CBX v. CBZ [67].
203) Ibid.
204) Ibid.
205) See above at [68].
206) As explained above at [15], illegality and public policy are not mirror concepts.
207) Ibid.
208) This was what happened in AJU, supra n. 31; Betamax, supra n. 75; and Gokul, supra n.
66.
209) All Judge Reyes said was: ‘It would not usually be appropriate for the Singapore
court to intervene in such case. The parties should be held to their agreement to
abide by the tribunal’s award, even if that award is wrong as a matter of law’.
However, it is not clear why this exact same reasoning should not also apply to ‘false
positive’ awards.
210) See e.g., West Tankers Inc v. Allianz SpA [2012] EWCA Civ 27, where the English Court of
Appeal upheld the High Court’s decision to enforce a negative declaratory award
under s. 66 of the English Arbitration Act 1996.
211) By way of example, consider a Dutch-seated arbitral tribunal which erroneously
dismissed a claim because it held that gay marriage is illegal in the Netherlands.
The Dutch courts may possibly intervene in this ‘false negative’ award for being
contrary to its pro-LGBT policy.
212) See above at [24].
213) See e.g., Soleimany, supra n. 70.
214) See above at [1]–[13].
215) See above at [14].
216) See above at [17].
217) See above at [36].
218) Westacre (CA), supra n. 42, at 881; Ladd v. Marshall [1954] 3 All ER 745, 748;
Bloomberry Resorts and Hotels Inc v. Global Gaming Philippines LLC [2021] 3 SLR 725
[105]; BVU v. BVX [2019] SGHC 69 [106].
219) Model Law, supra n. 8, Arts 34(2)(a)(ii) or 36(1)(a)(ii).
220) TMM Division Maritima SA de CV v. Pacific Richfield Marine Pte Ltd [2013] 4 SLR 972 [65].
221) Betamax, supra n. 75.
222) Ibid., at [95].
223) Ibid., In particular, the Board opined: ‘Moreover, a determination of the public policy
of the Republic of Mauritius in relation to any such illegality is an issue on which it
would be necessary, particularly in relation to public procurement, to have close
regard to the determination of the Supreme Court when such an issue actually arises’
[emphasis added].
224) Betamax, supra n. 75, at [26]–[27].
225) Ibid., at [54]–[94].
226) Ibid.

25
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 227) Westacre (CA), supra n. 42, at 874. Waller LJ refers to this argument as ‘the Lemenda
point’.
228) Westacre (CA), supra n. 42, at 871 and 877.
229) Ibid.
230) See Westacre (CA), supra n. 42, at 873, where Waller LJ implicitly describes the three
arguments as disjunctive arguments.
231) Ibid.
232) See below at [86] and [91].
233) Westacre (CA), supra n. 42, at 869–871.
234) In theory, the public policy arguments may also be advanced cumulatively. In other
words, a challenging party may argue that although the individual arguments are
not capable of triggering the high threshold of the public policy ground alone, they
would if they are considered cumulatively. In such cases, all of the arguments will
have to be substantiated at Stage Two for the public policy ground to be effectively
invoked.
235) Westacre (CA), supra n. 42, at 877.
236) Ibid.
237) Westacre (CA), supra n. 42, at 874.
238) Report of the United Nations Commission on International Trade Law on the Work of
its Eighteenth Session, Official Records of the General Assembly, A/40/17 (21 Aug.
1985), para. 297.
239) Westacre (CA), supra n. 42, at 881.
240) See above at [77].
241) Westacre (CA), supra n. 42, at 881.
242) Ibid., at 871.
243) AJU, supra n. 31, at [15].
244) AJU also argued that the award was made in breach of natural justice: see AJU, supra
n. 31, at [16] and [18].
245) AJT v. AJU [2010] 4 SLR 649 (‘AJT (HC)’) at [12]; AJU, supra n. 31, at [17].
246) The second allegation was dismissed at first instance, and not followed up on
appeal: see AJT (HC), supra n. 245, at [54]; AJU, supra n. 31, at [26]. However, for the
purposes of this mock application, we will proceed as the first instance court and
deal with both allegations.
247) Compare, Westacre (CA), supra n. 42 where the three allegations appeared to be
mounted separately: see above at [85].
248) Counsel will of course be free to mount the allegations cumulatively. In other words,
they are free to argue that, even if one of the allegations alone is not sufficient to
invoke the public policy of Singapore, the court ought to consider if both
allegations, cumulatively, would be sufficient. However, the danger of mounting
arguments cumulatively is that all the allegations made would then have to be
substantiated under Stage Two. If at least one of the allegations cannot be
sustained, then the entire public policy challenge fails.
249) AJU, supra n. 31, at [19].
250) Ibid., where Chan CJ noted: ‘it was common ground between the parties … that … the
[settlement agreement] would be … in conflict with the public policy of Singapore … if:
(a) the Concluding Agreement required the Appellant to stifle the prosecution of the
Forgery Charges, which were not compoundable under Thai law; and/or (b) the Thai
prosecution authority had been bribed to issue the Non-Prosecution Order with respect
to the Forgery Charges against [O]’ [emphasis added in bold italics].
251) AJU, supra n. 31, at [73].

26
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
 252) Ibid., where Chan CJ held: ‘In our view, this interpretation is, with respect, not justified
for the following reasons. First, the Judge interpreted the Concluding Agreement (which
was not illegal on its face) on the basis that “agreements of an illegal nature [were]
unlikely to be expressly stated[;] [i]nferences ha[d] to be made from the surrounding
circumstances” (see [46] of the HC Judgment). This is tantamount to assuming a fact
which has yet to be proved. Second, none of the provisions of the Concluding
Agreement required the Appellant to take any unlawful action to stop the Thai
criminal proceedings. In particular, cl 1 merely referred to the Appellant receiving
evidence from the Thai prosecution authority (or other relevant authority) of the
withdrawal and/or discontinuation and/or termination of those proceedings. Third,
the Judge relied on the draft cl 5 (as defined at [24] above) as evidence of the parties’
agreement to achieve an illegal purpose when the very fact that that provision was
subsequently discarded would suggest the contrary. In any case, although the draft cl
5 expressly stated that as a condition precedent to the performance of each party’s
obligations, the Appellant was to “take such action as [was] necessary to withdraw
and/or discontinue certain criminal proceedings … filed … by [the Appellant] against
[Q], [P] and [O] with [the Thai prosecution authority]”, it did not necessarily follow that
the action which the Appellant had to take would invariably involve some kind of
unlawful action. As mentioned earlier, the Respondent’s allegation that the Appellant
had procured the issue of the Non-Prosecution Order by bribery was rejected by the
Tribunal, and this finding was affirmed by the Judge (see [25] above). Fourth, the Judge
failed to consider that there was no reason for the Respondent to have entered into an
illegal agreement as it would surely have wanted to be paid the Agreed Final
Settlement Amount of US$470,000: why should the Respondent have taken the risk of
entering into an agreement which the Appellant could have resiled from at any time
on the ground of illegality?’.
253) See footnote 235 above.
254) AJT (HC), supra n. 245, at [54]; AJU, supra n. 31, at [26].
255) Ibid.
256) Gokul, supra n. 66, at [182]-[186].
257) Ibid.
258) Gokul, supra n. 66, at [206].
259) See above at [39].
260) See above at [40].
261) See above at [76].
262) See above at [77].
263) See above at [49].
264) See above at [50].
265) See above at [53].
266) See above at [54].
267) See above at [55].
268) See above at [43]. In particular, factor (2) (viz. whether the tribunal was competent
enough) might not be capable of producing a binary answer.
269) This is perhaps why public policy has been described as ‘a very unruly horse’ by
Burrough J in Richardson, supra n. 14, at 303.
270) See above at [50]–[52].
271) Such an allegation would likely be capable of invoking Stage One of the Proposed
Approach. See Report of the United Nations Commission on International Trade Law
on the Work of its Eighteenth Session, Official Records of the General Assembly,
A/40/17 (21 Aug. 1985) which explicitly identifies 'instances such as corruption,
bribery or fraud' as manifestations of a fundamental public policy breach.
272) See above at [56]–[57].
273) As alluded to at n. 203 above, one possible example is where a Dutch-seated
arbitral tribunal erroneously dismisses a claim because it held that gay marriages
are illegal in the Netherlands. In such a case, the Dutch courts would likely
intervene in this ‘false negative’ award for being contrary to its pro- LGBT public
policy.
274) See above at [60]-[62].
275) Ibid.
276) See above at [62].
277) As long as the objection made at Stage One invokes the public policy of the forum
and is thus worth entertaining.
278) Ng, supra n. 194.
279) Ibid., at 89–94.
280) Ibid., at 90, which he terms as ‘Scenarios 3 and 5’ in his matrix.
281) Ibid.
282) Kurkela & Turunen, supra n. 73, at 18.
283) Soleimany, supra n. 70, at 800.
284) Ng, supra n. 194, at 91.
285) See above at [76].
286) See above at s. 2.A.
287) See above at s. 2.B.

27
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.

Kluwer Arbitration is made available for personal use only. All content is protected by copyright and other intellectual property
laws. No part of this service or the information contained herein may be reproduced or transmitted in any form or by any means, or
used for advertising or promotional purposes, general distribution, creating new collective works, or for resale, without prior
written permission of the publisher.

If you would like to know more about this service, visit www.kluwerarbitration.com or contact our Sales staff at lrs-
[email protected] or call +31 (0)172 64 1562.

KluwerArbitration

28
© 2024 Kluwer Law International, a Wolters Kluwer Company. All rights reserved.