ARBITRATION INTERNATIONAL, Vol. 26, No. 4
© LCIA, 2010
441
And/Or: The Problem of Qualiﬁcation in
International Arbitration
by VEIJO HEISKANEN*
Arbitration International
Kluwer Law InternationalLCIA0957-04112005264
ABSTRACT
The problem of qualiﬁcation is one of the classic problems of conﬂict of laws and indeed has been
characterised by leading scholars as the ‘fundamental’ problem. In short, it is about how a
particular legal relation should be conceptualized or, more speciﬁcally, whether it should be
qualiﬁed as a relationship of law and fact or rather as a conﬂict of laws. While the problem has
attracted only limited attention among international arbitration scholars and practitioners, it
provides an interesting perspective to conﬂict of laws in international arbitration.
In classic conﬂict of laws, the problem of qualiﬁcation was associated with the question of the
relationship between legal categories (contract, tort, procedure, etc.) and a connecting factor (place
of performance, place of tort, forum, etc.) In international arbitration, the relationship between
the arbitration proceedings and the seat of arbitration, and other relevant connecting factors, raises
similar issues of qualiﬁcation. Measured by the relevant connecting factors such as the seat of
arbitration, the nationality or domicile of the parties, the place of performance of the contract, and
the subject matter of the dispute, an international arbitration may be more or less ‘international’
– or perhaps more accurately, ‘transnational.’ The less transnational the arbitration is in terms
of the relevant connecting factors, the more appropriate it arguably is to resolve any conﬂict of
laws issues that may arise on the basis of the standards of the seat. The more transnational the
arbitration is in terms of such connecting factors, the more appropriate it arguably is to resolve
any conﬂict issues by reference to transnational standards. In other words, the transnationality of
international arbitration is a sliding scale, or a difference in degree, and as such a matter of
policy.
While qualiﬁcation may or may not be the ‘fundamental’ problem of international arbitration,
depending on one’s intellectual viewpoint, it does provide an instructive framework for developing
an understanding of conﬂict of laws issues in international arbitration. Although true conﬂicts of
laws remain rare, they may arise, and when they do arise, they tend to raise sensitive issues of
public policy, precisely because an arbitral tribunal is not necessarily bound by the public policies
of the seat.
I
* Partner, Lalive, Geneva. I would like to thank David Bonifacio of Lalive for diligent research assistance.
442 Arbitration International, Volume 26 Issue 4
I. INTRODUCTION
THE PROBLEM of qualiﬁcation (or ‘characterisation’ or ‘classiﬁcation’ or
‘categorisation’) is considered to be one of the classic problems of conﬂict of laws,
or private international law. It is also considered to be one of the most difﬁcult
problems in the ﬁeld and indeed has been qualiﬁed by leading scholars as a
‘fundamental problem’.1 Given the traditional links between conﬂict of laws and
international arbitration, how does the problem of qualiﬁcation arise in the
context of international arbitration? Or does it – or should it be allowed to – arise
at all?
The problem of qualiﬁcation is said to have been ‘discovered’ independently
and almost simultaneously by Franz Kahn, a German jurist, and Etienne Bartin,
a French scholar, at the end of the nineteenth century.2 Since then, the issue has
been recognised in other jurisdictions, including in the common law world, and
has given rise to voluminous legal literature, ‘much of it highly theoretical’.3
Indeed, one is tempted say that, in the course of this debate, the problem of
qualiﬁcation itself has become an instance of the problem of qualiﬁcation, as
reﬂected in the variety of the terms used – ‘qualiﬁcation’, ‘characterisation’,
‘classiﬁcation’, ‘categorisation’, even ‘interpretation’ – to conceptualise the
problem.
However, since the 1960s, the issue has lost some of its appeal, in particular as
a result of the conﬂict-of-laws ‘revolution’ in the United States, and would today
be considered by few as one of the cutting edge issues in the ﬁeld.4 Moreover,
arbitration scholars and practitioners, who tend to be more pragmatic than their
private international law colleagues, appear to have always approached the
problem of qualiﬁcation with caution. While there are authors who have ventured
1 See e.g., Dicey, Morris and Collins on The Conﬂict of Laws (Sir Laurence Collins (ed.), 14th edn, 2006), p. 37.
2 Franz Kahn, ‘Gesetzkollisionen’ in 39 Jherings Jahrbücher für die Dogmatik des heutigen römischen Privatrechts 1
(1891); Etienne Bartin, De l’impossibilité d’arriver à la suppression déﬁnitive des conﬂits de loi in (1897) Journal de droit
international privé 225. For subsequent discussion, see e.g., Ernst Rabel, Das Problem der Qualiﬁcation (1931);
A.H. Robertson, Characterization in the Conﬂict of Laws (1940); William Beckett, ‘The Question of Classiﬁcation
(“Qualiﬁcation”) in Private International Law’ in (1934) 15 Brit. YB Int’l L 46; Walter Wheeler Cook,
‘ “Characterization” in the Conﬂict of Laws’ in (1941) 51 Yale LJ 191; Albert A. Ehrenzweig, ‘Characterization
in the Conﬂict of Laws: an Unwelcome Addition to American Doctrine’ in Kurt H. Nadelman et al. (ed.),
XXth Century Comparative and Conﬂicts Law: Legal Essays in Honor of Hessel E. Yntema (1961), p. 395; John D.
Falconbridge, ‘Characterization in the Conﬂict of Laws’ in (1937) 53 LQ Rev. 235; Ernest G. Lorenzen, ‘The
Qualiﬁcation, Classiﬁcation, or Characterization Problem in the Conﬂict of Laws’ in (1941) 50 Yale LJ 743;
Ernest G. Lorenzen, ‘The Theory of Qualiﬁcations and the Conﬂict of Laws’ in (1920) 20 Colum. L Rev. 247;
Joseph Morse, ‘Characterization: Shadow or Substance’ in (1949) 49 Colum. L Rev. 1027; Joseph Unger, ‘The
Place of Classiﬁcation in Private International Law’ in (1937) 19 The Bell Yard 3.
3 Dicey, Morris and Collins on The Conﬂict of Laws, supra n. 1 at p. 38.
4 While the United States initially had adopted what was effectively a continental European approach to the
conﬂict of laws, this was abandoned as a result of the choice of law ‘revolution’ that took place in the United
States beginning in the 1960s. For a fundamental contribution, see e.g., Brainerd Currie, Selected Essays on the
Conﬂict of Laws (1963). The interest-based approach (to the exclusion of the ‘rules-based’ approach) was
largely incorporated into the Second Restatement, in particular in relation to conﬂict issues. See Restatement
(Second) of the Conﬂict of Laws (1971). For further discussion, see e.g., Symeon C. Symeonides, The American
Choice-of-Law Revolution in the Courts: Today and Tomorrow (2005); Symeon C. Symeonides, ‘The American
Revolution and the European Evolution in Choice of Laws: Reciprocal Lessons’ in (2008) 82 Tul. L Rev. 1;
Louise Weinberg, ‘Theory Wars in the Conﬂict of Laws’ in (2005) 103 Mich. L Rev. 1631.
And/Or: The Problem of Qualiﬁcation in International Arbitration 443
to tackle the issue, they appear to have been less than successful in managing to
provoke the intellectual interest of their colleagues.5 One could say that, from the
point of view of international arbitration, the problem of qualiﬁcation remains to
be discovered. The question that arises is whether it should indeed be
rediscovered or rather left dormant like the proverbial sleeping dog.
This article does not propose to raise the problem of qualiﬁcation beyond
the relatively narrow conﬁnes of the ﬁeld of international arbitration, except to
the extent that this appears to be necessary for the purpose of introducing the
problem. Limited in this way, what is the problem of qualiﬁcation all about?
What does ‘qualiﬁcation’ mean, or refer to, in the ﬁrst place? What is so
fundamentally problematic about it? What is its potential relevance – if any – to
the law and practice of international arbitration? And ﬁnally, can it still be
considered a ‘fundamental’ problem of conﬂict of laws, or indeed of international
arbitration, in circumstances where the traditional rule-based approach to
conﬂict of laws no longer dominates and has effectively been abandoned in
favour of more policy-oriented (interest-based) approaches in certain jurisdictions,
in particular in the United States?
II
II. THE PROBLEM OF QUALIFICATION
IN CONFLICT OF LAWS
A standard textbook approach is to deﬁne the problem of qualiﬁcation as
consisting of ‘determining which juridical concept or category is appropriate in
any given case’.6 In this account, the problem of qualiﬁcation is a fundamental
problem in all traditional systems of conﬂict of laws which rely on conﬂict rules
that are composed of two elements: juridical concepts or categories and localising
elements or connecting factors.7 This duality is reﬂected in conﬂict rules such as
‘succession to immovables is governed by the law of the situs’, ‘contract is
governed by the law of the place where it was made’, ‘tort is governed by the law
of the place where it occurred’, ‘procedure is governed by the lex fori’ or, closer to
the home of international arbitration, ‘arbitration proceedings are governed by
law of the seat (lex arbitri)’. Here, succession to immovables, contract, tort,
procedure and arbitration proceedings are the categories, whereas the law of the
situs, the place where the contract was made, the place where the tort occurred,
lex fori, and the law of seat, respectively, are the connecting factors. Nussbaum
summarises the doctrine in these terms:
5 For contributions, see e.g., Mathieu Boisséson, ‘L’arbitre international et le problème de la qualiﬁcation’ in
Liber Amicorum Claude Reymond: Autour de l’Arbitrage (2004), p. 29; Fabrizio Marrella, ‘Choice of Law in ThirdMillennium
Arbitrations: the Relevance of the UNIDROIT Principles of International Commercial
Contracts’ in (2003) 36 Vanderbilt J Transnat’l L 1138 at pp. 1148–1152; Claude Reymond, ‘Conﬂits de lois en
matière de responsabilité délictuelle devant l’arbitre international’ in Travaux du Comité français de droit
international privé (1991), p. 97.
6 Dicey, Morris and Collins on The Conﬂict of Laws, supra n. 1 at p. 38.
7 The classic system is essentially derived from Friedrich Karl von Savigny, System des heutigen römischen Rechts
(1849).
444 Arbitration International, Volume 26 Issue 4
The crucial point … is the role of qualification in the choice of laws. Choice-of-law rules consist
of two elements. They link together a legal relationship (property right, contract claim, tort
claim, etc.) and a connecting factor which one might call the localizator, such as the situs of the
res, the place of performance of a contract, the place of wrongdoing … In respect of each
element, a conflict of qualifications may arise.8
Accordingly, the problem of qualiﬁcation is related to the relationship between
the relevant legal category and the connecting factor, i.e. the localising element
that links the legal category to the relevant jurisdiction. The problem of
qualiﬁcation is therefore bound to arise in any conﬂict of laws system which
employs rules that operate so as to allocate a particular legal category to a proper
jurisdiction, or to a particular ‘place’. The problem may arise in respect of the
legal category, or in respect of the factual category (the ‘place’), or indeed in
respect of both.
When deﬁned in these terms, it is evident that the problem of qualiﬁcation is
not limited to the conﬂict of laws, but is a reﬂection of a more general issue.
Indeed, the problem of qualiﬁcation is bound to arise in any ﬁeld of law which
seeks to establish rules governing the relationship of a legal category (or more
broadly, a legal issue) and a factual category (or more broadly, an issue of fact). In
this broad sense, the problem of qualiﬁcation deals with the question of whether
a particular legal category covers a particular fact, or conversely, whether a
particular fact falls under a particular legal category – and indeed, whether a
particular category, in itself, should be characterised (qualiﬁed) as a legal or as
a factual category.9
From this more general (some might say ‘philosophical’) perspective it is
therefore arguable that the issue of qualiﬁcation is ultimately not a matter of
conﬂict of laws, nor indeed purely a matter of any law. It is about making the
distinction between law and fact and therefore is not limited to the ﬁeld of conﬂict
of laws. It tends to arise more frequently in the context of private international
law because this is a ﬁeld of law where not only the differences between different
laws – domestic and foreign law, on the one hand, and international and
municipal law, on the other – but also the distinction between law and fact are
easily blurred. Thus, for instance, private international law doctrine tells us that
under certain laws, such as English law, foreign law is said to be a matter of fact
8 Arthur Nussbaum, ‘Book Review: Characterization in the Conﬂict of Laws. Harvard Studies in the Conﬂict
of Laws, Vol. IV. By A.H. Robertson’ in (1940) 40 Colum. L Rev. 1461.
9 Ibid. p. 1462 (‘Qualiﬁcation may be understood in a much broader sense … Any subsumption of facts under
legal concepts, or of legal concepts under broader categories can be called qualiﬁcation or characterization’).
See also, Restatement (Second) of the Conﬂict of Laws (1971), s. 7 (‘Two aspects of characterization.
Characterization is an integral part of legal thinking. In essence, it involves two things: (1) classiﬁcation of a
given factual situation under the appropriate legal categories and speciﬁc rules of law; and (2) deﬁnition or
interpretation of the terms employed in the legal categories and rules of law. The factual situation must be
classiﬁed to determine under what legal categories and rules of law it belongs. Likewise, the terms employed
in the legal categories and rules of law must be interpreted in order that the factual situation may be placed
under the appropriate categories and that the rules of law may be properly applied.’).
And/Or: The Problem of Qualiﬁcation in International Arbitration 445
and as such a matter of evidence,10 whereas under certain others, foreign law is
said to be a matter of law.11 In other words, whether a particular issue is to be
considered an issue of law or an issue of fact is, precisely, a matter of qualiﬁcation
in the strict conceptual sense of this term: a qualiﬁcation (that is, a restriction) of
the meaning of the concept or category at issue.12 If a particular issue, such as the
status of foreign law before the forum, is qualiﬁed as a matter of fact rather than
law, or vice versa, such a qualiﬁcation necessarily delimits the meaning of the
concept of foreign law. After qualiﬁcation, it no longer means law and/or fact, but
law or fact. By implication, given the way in which the process of qualiﬁcation
operates, there is a conceptual state – a state of ‘and/or’ – before qualiﬁcation
takes place where the distinction between law and fact does not yet exist; rather it
is suspended in a sort of ‘superposition’, to borrow a term of quantum mechanics,
where law and/or fact coexist in one, unqualiﬁed or suspended state. In such a
state, law and fact are effectively the same, a sort of Urbegriff of law and/or fact.13
It is only the process of qualiﬁcation that produces the distinction between the
two concepts by qualifying, i.e. by limiting, the scope and meaning of the Urbegriff
and by distinguishing law from fact, and vice versa.
This preliminary analysis suggests that qualiﬁcation is fundamentally about the
issue of how to distinguish between law and fact in a particular context, including
in the context of conﬂict of laws. The issue is this: when two laws are in conﬂict
and cannot be applied at the same time, which law should be given effect as law,
and which should be treated rather as non-law, effectively as a mere fact? The
distinction between law and fact in the context of conﬂict of laws does not exist
before this determination has been made. It emerges as a result of the process of
qualiﬁcation, i.e. as a result of a conceptual delimitation of the particular conﬂict
of laws at issue. In this sense qualiﬁcation operates a bit like measurement in
10 See e.g., Dicey, Morris and Collins on The Conﬂict of Laws, supra n. 1 at p. 255, rule 18-(1) (‘In any case to which
foreign law applies, that law must be pleaded and proved as a fact to the satisfaction of the judge by expert
evidence or sometimes by certain other means’). Similarly, international law contains a conﬂict rule that
qualiﬁes municipal law as a ‘mere fact’ before an international court or tribunal; see e.g., Certain German
Interests in Polish Upper Silesia (Germany v. Poland) (1926) PCIJ (ser. A.) No. 7, 19, Judgment on Merits, 25 May
1926 (‘From the standpoint of International Law and of the Court which is its organ, municipal laws are
merely facts which express the will and constitute the activity of States, in the same manner as do legal
decisions or administrative measures. The Court is certainly not called upon to interpret the Polish law
as such, but there is nothing to prevent the Court’s giving judgment on the question whether or not, in
applying that law, Poland is acting in conformity with its obligations towards Germany under the Geneva
Convention.’). Of course, this does not prevent an international tribunal from interpreting and applying
domestic law as a preliminary matter. For further discussion of this and other transnational conﬂict rules, see
Veijo Heiskanen, ‘Forbidding Dépeçage: Law Governing Investment Treaty Arbitration’ in (2009) 32 Suffolk
Transnat’l L Rev. 367 at pp. 395–406.
11 See e.g., Swiss Private International Law Act, 1987, art. 16(1) (‘The content of the applicable foreign law shall
be established ex ofﬁcio. The assistance of the parties may be requested. In the case of pecuniary claims, the
burden of proof on the content of the foreign law may be imposed on the parties.’).
12 See e.g., Webster’s Encyclopedic Unabridged Dictionary of the English Language (1989): ‘Qualiﬁcation: … modiﬁcation,
limitation or restriction: to endorse a plan without qualiﬁcation … an instance of this: He presented his
argument with several qualiﬁcations’).
13 The concept of ‘Urbegriff ’ (or ‘primordial’ or ‘original’ concept) was aptly deﬁned by Immanuel Kant as ‘the
sum-total of all possibility’. See Immanuel Kant, Critique of Pure Reason (Norman Kemp Smith (trans.), 1965),
p. 489.
446 Arbitration International, Volume 26 Issue 4
quantum physics. Just as measurement destroys the quantum state (superposition)
of an elementary particle, qualiﬁcation destroys the ‘quale’ state of the Urbegriff
law and/or fact by qualifying one of its meanings and disqualifying the other.14 In
other words, pushing the analogy a bit further, just as physical phenomena such
as electromagnetic radiation, once measured, emerge as quanta, i.e. in discrete
quantities of energy rather than as continuous waves, legal phenomena, once
qualiﬁed, always appear as qualia, i.e. in terms of singular concepts rather than as
a stream of disembodied information.15
The problématique of the classic conﬂict of laws has been muddled somewhat by
the emergence of interest-based approaches, in particular in the United States.
These approaches, which began to gain momentum in the 1960s, have
challenged the role of the classic conﬂict rules and the importance of the problem
of qualiﬁcation. However, while a concerted effort has been made by proponents
of these approaches to eliminate, or at least to reduce the importance of, the
problem of qualiﬁcation, these attempts have been less than successful. Indeed, it
is arguable that the interest-based approaches to conﬂict of laws tend to beg – or
rather, suspend – the question of qualiﬁcation rather than resolve it. This is
because even when a choice of law is made on the basis of weighing of competing
interests rather than a pre-existing rule that links a particular legal category to a
particular jurisdiction, such a weighing will by deﬁnition result in a decision to
apply one law or another. In other words, while the choice of law in such a
situation is made on the basis of a weighing of a variety of criteria and is therefore
essentially ad hoc and as such should not raise qualiﬁcation issues, this does not
change the fact that, as a result of the choice, one of the competing laws will be
categorised as law and the other as ‘non-law’ – effectively as a fact.
Consequently, such a choice-of-law decision necessarily involves a process of
qualiﬁcation, even if not one based on rules, which suggests that the interestbased
approach is not fundamentally different from the rules-based approach.
The principal difference between the two approaches is, in substance, the timing
of qualiﬁcation. While the rules-based approach seeks to make a conceptual
distinction between law and fact in advance, on the level of the rule, the interestbased
approach suspends the making of this distinction until the concrete context
14 The concept of ‘qualia’ (in singular, ‘quale’) was introduced into modern philosophy by C.I. Lewis; see C.I.
Lewis, Mind and the World Order (1929), p. 121 (deﬁning qualia as ‘recognizable qualitative characters of the
given’). The term is used here in a broader sense to refer to the interface between information and
conceptualisation, or the ‘collapse’ or decoherence of the ﬂow of information into discrete and concrete
concepts. For further discussion of the relationship of information and physical phenomena, see e.g., Hans
Christian von Baeyer, Information: The New Language of Science (2003). The quale (unqualiﬁed) state of
information is reﬂected in certain concepts that still retain their opposite meanings; among legal concepts,
perhaps the most qualiﬁed example is ‘sanction’.
15 Conceptually speaking, neither outcome is surprising and indeed both follow virtually by deﬁnition: it should
be no more surprising if quantiﬁcation (i.e. measurement) produces quanta, or measurable units, than if
qualiﬁcation (i.e. conceptualisation) produces qualia, or intelligible concepts. From this (conceptual)
perspective, there is no ‘quantum weirdness’. Indeed, any other outcome of quantiﬁcation and qualiﬁcation
(i.e. other than quanta or qualia), respectively, would be weird. Just as concepts do not exist before they have
been formulated out of information, quanta do not exist until they have been meted out of the wave of
energy. Information and energy, no less than bread, do not naturally exist in a sliced condition.
And/Or: The Problem of Qualiﬁcation in International Arbitration 447
of judicial decision-making. Accordingly, since both approaches are based on the
assumption that there is a fundamental conceptual distinction between law and
fact, and that this distinction must eventually be made, they cannot be said to be
fundamentally different from each other.16
It should not come as a surprise, therefore, that the interest-based approaches
have been unable to dispose of legal categories altogether. Thus, the Restatement
(Second) of the Conﬂict of Laws, which is much inﬂuenced by the interest-based
approaches and has largely disposed of formal conﬂict rules for the purpose of
determining the applicable law,17 continues to apply traditional legal categories
such as jurisdiction and procedure for the purpose of categorising the relevant
legal issues.18 Indeed, it is arguable that even an extreme interest-based approach
would have difﬁculty in disposing of these legal categories since a choice of
law can arise as an issue of substance only after preliminary issues involving
jurisdiction and admissibility, on the one hand, and procedure and substance, on
the other, have been resolved. No substantive choice of law issues can arise, as a
practical matter, before it has been determined that the particular dispute in
question falls under the court’s jurisdiction, and/or that the claim is admissible.
Similarly, no substantive choice of law issues can arise, as a matter of law, before
it has been determined that there is indeed a conﬂict of laws and not merely an
apparent conﬂict. If it turns out that one of the apparently conﬂicting laws is a rule
of procedure whereas the other qualiﬁes as a norm of substance, there is, legally
speaking, no conﬂict of laws. Both of these preliminary determinations require
qualiﬁcation, or proper conceptualisation, of the issue at hand. Obviously such
preliminary determinations, whether they concern jurisdiction or admissibility, or
the categorisation of a particular law as a rule of procedure or as a norm of
substance, are made in international arbitration as a matter of daily routine,
although probably often without any deeper reﬂection on the question of whether
one is engaged in a process of qualiﬁcation. This does not make the issue less
important – rather the contrary is arguably the case.
The policy-oriented (interest-based) approaches have certainly made a lasting
contribution to conﬂict of laws in the sense that they have shifted the focus of
conﬂict of laws from the application of formal conﬂict rules to the substance of the
issue. Even accepting that there cannot be, as a matter of law, any conﬂict of laws
unless both of the applicable laws qualify as rules of procedure or as norms of
substance, respectively, there cannot be any conﬂict of laws as a matter of fact unless
16 It is hard to say, therefore, which one of the two approaches deserves more credit, in abstracto. While the
interest-based approach increases the likelihood that the ultimate choice of law better reﬂects the facts of the
case, the increased accuracy in judicial decision-making comes at a price, as it reduces predictability.
Conversely, the rules-based approach puts a premium on predictability, leaving the courts to struggle with
concrete cases in circumstances where the abstract rule that they have been instructed to apply tends to
capture only a sub-set of the relevant facts and thus may produce arbitrary results.
17 See Restatement (Second) of the Conﬂict of Laws (1971), ss. 145 (torts) and 188 (contracts) (adopting the test
of the ‘most signiﬁcant contact’).
18 Ibid. ch. 3 (judicial jurisdiction). See also, ibid. ch. 6 (procedure), s. 122 (discussing the ‘substance–procedure
dichotomy’).
448 Arbitration International, Volume 26 Issue 4
the two laws conﬂict also in substance and not merely in form. In other words, it
cannot be said that there is any conﬂict of laws until it is established, as a matter
of fact (as opposed to formal law), that the conﬂict at hand is not a ‘false’ conﬂict
but rather a ‘true’ conﬂict of laws in the sense that both of the applicable laws
incorporate compelling public policies that cannot be superseded or displaced by
formal conﬂict rules or by agreement between the parties. In the words of
Brainerd Currie, for a true conﬂict of laws to exist, there must be a conﬂict as to
the ‘rule of decision’:
Conflict of laws, as we practice it, is concerned with references to foreign law for quite different
purposes. Our failure to distinguish between them is to a considerable degree responsible for our
troubles. The distinction needs to be clarified and better stated, so that it can be more easily
applied; and we need to know more about the class of cases in which foreign law is referred to
for some purpose other than that of finding the rule of decision. For the present, however, I
divided all conflict cases into (1) those in which the purpose of the reference to foreign law is to
find the rule of decision, and (2) those in which the reference has some other purpose … The
central problem of conflict of laws may be defined then, as that of determining the appropriate
rule of decision when the interests of two or more states are in conflict – in other words, in
determining which interest shall yield.19
Consequently, if one of the laws in question is a non-mandatory (‘dispositive’)
law from which the parties can contract out, and the other is a mandatory law
which applies regardless of the terms of the contract, there is, as a matter of
substance (or public policy), no conﬂict between the applicable laws. Similarly,
if the applicable law by its terms supersedes the law that would otherwise
be applicable under the relevant conﬂict rule, there cannot be, as a matter of
substance, any conﬂict between such law and the law that would otherwise be
applicable, but for the displacement of the applicable conﬂict rule.
The determination of whether the applicable law, in any particular case,
incorporates public policies that cannot be displaced by conﬂict rules or set aside
by agreement between the parties is, in itself, a matter of qualiﬁcation or
characterisation. It involves a determination of whether the law in question
qualiﬁes, in substance, as a matter of public policy applicable in the context, and
whether, in the circumstances, the otherwise applicable law should be
‘disqualiﬁed’. It may be said, therefore, that whether one prefers the traditional
rule-based approach or the more modern interest-based approach to conﬂict of
laws, the problem of qualiﬁcation remains, to an extent, unavoidable. It tends to
arise not only in the context of preliminary determinations, such as whether a
particular preliminary issue should be characterised as an issue of jurisdiction or
admissibility, or whether a particular law qualiﬁes as a rule of procedure or as a
norm of substance; it also tends to arise when determining whether there is a true
conﬂict between rules of decision. In other words, given that legal categories such
19 Brainerd Currie, ‘Notes on Methods and Objectives in Conﬂict of Laws’ in Currie, supra n. 4 at pp. 178–
179.
And/Or: The Problem of Qualiﬁcation in International Arbitration 449
as jurisdiction and admissibility, or procedure and substance, are not deﬁned in
identical terms in all jurisdictions, and given that conceptions of whether a
particular law qualiﬁes, in terms of its substance, as a matter of public policy are
likely to diverge between different jurisdictions, the problem of qualiﬁcation
remains part and parcel of conﬂict of laws, whatever one’s preferred intellectual
approach.
III
III. THE PROBLEM OF QUALIFICATION AND
CONFLICT OF LAWS IN THE CONTEXT OF
INTERNATIONAL ARBITRATION
a
(a) The Context of International Arbitration
The fact that the problem of qualiﬁcation has never been considered a fundamental
problem of international arbitration suggests that the traditional
problématique of conﬂict of laws tends to play only a limited role in modern
international arbitration.20 This is so largely for two reasons. First, an international
arbitral tribunal has a much more tenuous link with the seat (or the ‘place’) of
arbitration than a local court has with its own forum; indeed, it is often said that
an international arbitral tribunal does not have a forum.21 Secondly, the
distinction between domestic and foreign law is not relevant to an international
arbitral tribunal in the same way that it is relevant to the local court. The legal
system of the seat of arbitration is not the ‘domestic’ jurisdiction of an
international arbitral tribunal.
These two important differences between an international arbitral tribunal
and a local court are interlinked. To say that an international arbitral tribunal
does not have a forum is another way of saying that the relationship between the
seat and the tribunal is ad hoc rather than inherent or systemic. An international
arbitral tribunal has not been established to resolve disputes in any particular
territorially deﬁned jurisdiction, or in any particular ‘place’, and this applies to
the seat of arbitration. It may be said that an international arbitral tribunal is
substantially ‘delocalised’ in the sense that, even if it has a seat in a particular
jurisdiction, the fact that the seat is located in one jurisdiction rather than another
is often largely fortuitous, and at best a matter of convenience or practicality,
20 For reasons set out infra nn. 40–41 and the accompanying text, although the focus of this article is on
international commercial arbitration, this does not imply that there is a strict conceptual distinction between
international commercial arbitration and investment arbitration. This is not to say that the conﬂict rules or
approaches applicable in the two regimes are identical. See Heiskanen, supra n. 10.
21 See e.g., Pierre Lalive, Les règles de conﬂit de lois appliquées au fond du litige par l’arbitre international siégeant en Suisse in
(1976) Rev. Arb. 155 at p. 159; Reymond, supra n. 5 at p. 99; Boisséson, supra n. 5 at p. 30. See also, Gabrielle
Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’ in (2003) 36 Vanderbilt J Transnat’l L 1313 at p. 1315
(‘Two contrary ﬁndings arise from a review of the evolution of arbitration law over the last decades. On one
hand, it is now commonly accepted that an arbitration is governed by national arbitration law of the place
or seat of the arbitration, though not by local rules of civil procedure. On the other hand, such national law
has less and less actual bearing on the arbitration proceedings.’).
450 Arbitration International, Volume 26 Issue 4
rather than an indication of the territorial scope of the arbitral tribunal’s
jurisdiction.22
Thus, an international arbitral tribunal sitting in Geneva and composed of,
say, an English chairman and Egyptian and Canadian co-arbitrators, and dealing
with a contractual dispute arising between a Turkish and a Delaware company,
can hardly approach its task by distinguishing between domestic and foreign
law. Indeed, it may be said that the relevant laws, including the law of the seat
(lex arbitri), are all equally ‘foreign’, or equally ‘domestic’, from the perspective of
the arbitral tribunal. While the arbitration law of the seat governs the arbitral
proceedings in the sense that any action for setting aside of the arbitral award
is to be brought before the courts of the seat, this does not mean that the law of
the seat plays a substantial role in the conduct of the arbitral proceedings, in
particular in jurisdictions that have adopted the UNCITRAL Model Law on
Commercial Arbitration or another, similar modern arbitration law.23 The
private arbitration rules designated by the parties invariably provide much more
detailed guidance for the arbitral tribunal than the few minimum standards
established in the applicable international arbitration law. Nor does the law of the
seat necessarily have anything to do with the substance of the dispute, which in
practice is often governed by a law other than the law of the seat. Nor does the
law of the seat necessarily govern the validity of the arbitration agreement, or the
capacity or authority of the parties to enter into an arbitration agreement – and
this is because the seat is often chosen precisely because it has no connection with
the dispute. In other words, while a local court is generally considered competent
because the dispute has arisen within its jurisdiction, the seat of an arbitral
tribunal is often intended to be as ‘foreign’ to the parties and the subject matter of
the dispute as reasonably possible.
The principal consequence of these important differences between local court
litigation and international commercial arbitration is that an international
arbitral tribunal cannot be considered to be bound by the conﬂict rules of the seat
of the arbitration – a conclusion generally accepted and reﬂected in most modern
international arbitration rules.24 While an international arbitral tribunal may seek
22 The controversy about whether parties to arbitration agreements would or should be able to ‘delocalise’ the
arbitration and have the arbitration conducted without regard to the mandatory rules of law of the seat or,
in any event, have the arbitral award recognised and enforced even if it had been set aside by the court of the
seat of arbitration was a hot topic among arbitration scholars and practitioners in the 1980s. For discussion,
see e.g., Jan Paulsson, ‘Arbitration Unbound: Award Detached from the Law of Its Country of Origin’ in
(1981) 30 Int’l and Comp. LQ 358; Hans Smit, ‘A-National Arbitration’ in (1981) 63 Tul. L Rev. 629; Jan
Paulsson, ‘Delocalization of International Commercial Arbitration: When and Why It Matters’ in (1983) 32
Int’l and Comp. LQ 53; William W. Park, ‘The Lex Loci Arbitri and International Commercial Arbitration’ in
(1983) 32 Int’l and Comp. LQ 21; Jan Paulsson, ‘The Extent of Independence of International Arbitration from
the Law of the Situs’ in J. Lew (ed.), Contemporary Problems in International Arbitration (1986), p. 297; William W.
Park, ‘Judicial Controls in the Arbitral Process’ in (1989) 5 Arb. Int’l 230.
23 See e.g., Swiss Private International Law Act, arts. 182–184, 190.
24 See e.g., UNCITRAL Arbitration Rules, art. 33(1) (‘The arbitral tribunal shall apply the law designated by the
parties as applicable to the substance of the dispute. Failing such designation, the arbitral tribunal shall apply
the law determined by the conﬂict of laws rules which it considers to be applicable’); Swiss Rules of
International Arbitration, art. 33(1) (‘The arbitral tribunal shall decide the case in accordance with the rules
xxxXSPxxx
And/Or: The Problem of Qualiﬁcation in International Arbitration 451
guidance from the local conﬂict rules, it is not required to do so. It may also
choose to apply a particular rule, norm or principle that it considers appropriate,
without having to engage in any conﬂict-of-laws analysis (voie directe).25 It is this
latter development, in particular, that accounts for the relative decline in
importance of the classic conﬂict of laws in the law and practice of international
arbitration; indeed, it effectively amounts to a displacement of the rule-based
approach in favour of the interest-based (or perhaps more appropriately, purely
contextual)26 approach to conﬂict of laws issues in international arbitration.27 If
the arbitrators may choose any conﬂict rule they consider appropriate and even
disregard conﬂict of laws rules and choose the applicable law directly, without
having to engage in any conﬂict of laws analysis, is there any real role or function
left for the traditional conﬂict of laws methods in modern international
arbitration?
It is indeed the case that conﬂict of laws, in its classical sense as a set of rules
designed to facilitate the choice of the applicable law before a local court, is only
of law agreed upon by the parties or, in the absence of a choice of law, by applying the rules of law with
which the dispute has the closest connection’). See also, UNCITRAL Model Law, Art. 28(2) (‘Failing any
designation [of the applicable law] by the parties, the arbitral tribunal shall apply the law determined by the
conﬂict of laws rules which it considers applicable’). The traditional approach, which was reﬂected for
instance in the resolution adopted by the Institut de droit international at its 1957 and 1959 sessions, was to
apply the conﬂict rules of the seat of arbitration. See art. 11 (‘Law Applicable to the Substance of the
Difference’) of the resolution on ‘Arbitration in Private International Law’ in 47 Annuaire de l’Institut de Droit
International (1957), vol. II, pp. 491, 496 (‘The rules of choice of law in force in the state of the seat of the
arbitral tribunal must be followed to settle the law applicable to the substance of the difference’). While the
traditional approach is no longer considered as the sole available or indeed permissible approach, it
continues to have its supporters; see e.g., G.C. Moss, ‘International Arbitration and the Quest for the
Applicable Law’ in (2008) 8(3) Global Jurist 1. See also, J.-F. Poudret and S. Besson, Comparative Law of
International Arbitration (2007), p. 573 (‘Although there is no evident link between the choice of the seat and the
choice of the rules of conﬂict for determining the law applicable to the merits, making a presumption in this
sense hazardous, this method does at least have the merit of assuring a certain degree of foreseeability with
regard to the determination of the lex causae in the absence of a choice of law’).
25 See e.g., Arbitration Rules of the International Chamber of Commerce, art. 17(1) (‘The parties shall be free to
agree upon the rules of law to be applied by the Arbitral Tribunal to the merits of the dispute. In the absence
of any such agreement, the Arbitral Tribunal shall apply the rules of law which it determines to be appropriate’)
(emphasis added); Arbitration Rules of the Arbitration Institute of the Stockholm Chamber of Commerce,
art. 22(1) (‘The Arbitral Tribunal shall decide the merits of the dispute on the basis of the law or rules of law
agreed upon by the parties. In the absence of such agreement, the Arbitral Tribunal shall apply the law or
rules of law which it considers to be most appropriate’) (emphasis added); Arbitration Rules of the American
Arbitration Association, art. 28 (‘The tribunal shall apply the substantive law(s) or rules of law designated by
the parties as applicable to the dispute. Failing such designation by the parties, the tribunal shall apply such
law(s) or rules of law as it determines to be applicable’) (emphasis added).
26 This is because the approach does not necessarily require any weighing of ‘interests’ between the relevant
jurisdictions; it simply requires the determination of the ‘appropriate’ rule of law, taking into account all the
circumstances. See W. Laurence Craig, William W. Park and Jan Paulsson, International Chamber of Commerce
Arbitration (3rd edn, 2000) (‘Th[e] empowerment to use the “voie directe” in choice of law also coincides with
the tendencies of arbitral practice. The freedom of the arbitral tribunal, like that of the parties, to apply rules
of law other than those of a single state provides a ﬂexibility to meet the intentions of the parties and to
respond to all the circumstances of a case.’).
27 It should be noted that the voie directe approach remains a conﬂict-of-laws approach in the sense that it results
in a choice of law and accordingly the arbitrators must provide reasons for their (contextual) choice of law.
See Horacio Grigera Naón, ‘Choice-of-Law Problems in International Commercial Arbitration’ in (2001)
289 Recueil des Cours 377.
452 Arbitration International, Volume 26 Issue 4
of limited relevance in international arbitration. This is not to say that conﬂict of
laws approaches have no relevance at all in international arbitration, or that there
has been a wholesale displacement of the rule-based approach by a purely
contextual approach. It is to say that, although conﬂicts of laws continue to arise
in the context of international arbitration, they do not arise out of the distinction
between domestic and foreign law, which as such has no relevance in
international arbitration. Instead of having to choose between domestic and
foreign law, an international arbitral tribunal is often faced with a much broader
range of options – it must choose between the law(s) that the parties have either
directly or indirectly designated as applicable and any other laws that may be
considered to be applicable.
Unlike a local court, an international arbitral tribunal is faced with what may
be termed a ‘transnational’ conﬂict of laws. Such a conﬂict is ‘transnational’ in
the sense that it arises in terms of the law or the laws that the parties have either
directly (by including a governing law clause in the contract) or indirectly (by
designating a seat of arbitration and/or a set of private arbitration rules that
authorise the arbitral institution or the arbitral tribunal to choose the seat of
arbitration) chosen as being applicable, and any other laws, including laws of
third jurisdictions, that may be applicable. Such other laws include rules of law
that may be considered applicable under the appropriate conﬂict rule or voie
directe, or mandatory rules of law that may be applicable to the contract regardless
of the parties’ choice of law, or public laws (lois de police) that may apply to the
contract or its performance, or they may consist of national, international or
transnational public policies or, in certain instances, of public international law.
The potential for conﬂict between such laws in any particular case depends on
how ‘international’ the arbitration process is in terms of the various relevant
criteria, including the place of the arbitration, the nationality of the parties, the
principal places of business of the parties, the law governing the contract, the
place of performance of the contract, and the place or places where the effects of
the performance of the contract are felt. In other words, there is no one standard
case of international arbitration; some international arbitrations are more
international (or more accurately, more ‘transnational’)28 than others, depending
on the diversity of the relevant criteria. There are international arbitrations
where the only cross-border element may be the seat of the arbitration or, more
28 The term ‘international’ implies that the arbitral tribunal is established under and derives its jurisdiction
from international law, which is not the case. F.A. Mann famously argued that ‘[t]here is very strong reason
to fear that [the word “transnational”] means nothing and is incapable of deﬁnition’. F.A. Mann, ‘Book
Review: Resolving Transnational Disputes through International Arbitration, Sixth Sokol Colloquium,
Edited by Thomas E. Carbonneau’ in (1986) 2 Arb. Int’l 378. Dr Mann was certainly right in that, unlike the
distinctions between international law and municipal law, on the one hand, and between domestic and
foreign law, on the other, the term ‘transnational’ does not have a clear-cut conceptual opposite but rather
occupies a conceptual continuum where the particular phenomenon under consideration (in this case,
arbitration) may be more or less transnational, depending on the context. Since it tends to muddle the
conceptual distinctions between international and municipal, on the one hand, and domestic and foreign, on
the other, it may indeed seem, for some, too unwieldy as the conceptual basis of international arbitration.
And/Or: The Problem of Qualiﬁcation in International Arbitration 453
frequently, the nationality of one of the parties; but there are also others where
the diversity of connecting factors is such that the arbitration process may be
characterised as truly transnational. In such a truly transnational setting, there is
a variety of laws, rules of law and/or public policies that are potentially relevant:
• local arbitration law, to the extent that it establishes mandatory rules of law
for arbitrations taking place in the jurisdiction;29
• other mandatory rules of law of the seat of arbitration;
• laws of the places of organisation and/or of the principal places of business
of the parties;
• law governing the arbitration agreement;
• law of the place where the contract was made;
• law governing the contract;
• law of the place of performance of the contract, or of a jurisdiction where
the effects of the performance of the contract are felt, and/or of the place
of recognition and enforcement of the award;
• national, international and/or transnational public policy;
• public international law.
Any problems of qualiﬁcation or conﬂicts of laws that may arise in such a
transnational context cannot be approached or resolved exclusively in terms of
the traditional conﬂict rules or approaches, which focus on the distinction
between domestic and foreign law. While the conﬂict of laws in each of these
instances may be characterised in terms of the law of the seat and the law of a
jurisdiction other than the seat, such other law is not necessarily more ‘foreign’ in
the context of the arbitration than the law of the seat, in particular where the sole
connecting factor between the arbitration and the seat is the seat itself. Thus, for
instance, while it might be tempting to argue that in case of a conﬂict between the
mandatory law of the seat and the mandatory law of any other jurisdiction,
priority should invariably be given to the law of the seat, this solution is not
necessarily justiﬁed in circumstances where, again, the sole link between the seat
and the arbitration is the seat itself. In such circumstances the arbitral tribunal
cannot be said to be part of the ‘domestic’ system of the seat of arbitration and
29 Such rules of law may be procedural or substantive, e.g., to the extent that they regulate arbitrability or
reserve the seat’s public policy. See e.g., UNCITRAL Model Law, Art. 1(5) (‘This Law shall not affect any
other law of this State by virtue of which certain disputes may not be submitted to arbitration or may be
submitted to arbitration only according to provisions other than those of this Law’), and Chapter V
(‘Conduct of Arbitral Proceedings’), esp. Art. 19(1) (‘Subject to the provisions of this Law, the parties are free
to agree on the procedure to be followed by the arbitral tribunal in conducting the proceedings’). See also,
ibid. Art. 34(b)(i) (‘An arbitral tribunal may be set aside by the [competent court] if: … the court ﬁnds that …
the subject matter of the dispute is not capable of settlement by arbitration under the law of this State’) and
Art. 34(b)(ii) (‘An arbitral tribunal may be set aside by the [competent court] if: … the court ﬁnds that … the
award is in conﬂict with the public policy of the State’).
454 Arbitration International, Volume 26 Issue 4
therefore cannot be expected to automatically resolve the conﬂict in favour of the
law of the seat.30
This is not to suggest that no distinction can or should be made between
international and domestic arbitration. It is to suggest that the ‘internationality’
(or rather ‘transnationality’) of an arbitration is a sliding scale. The more tenuous
the link between the seat and the arbitration in terms of the various connecting
factors (the place of negotiation and/or conclusion of the contract, the places of
organisation and/or the principal places of business of the parties, the place of
performance of the contract, the law governing the arbitration agreement, the
law governing the contract, and the place where the effects of the performance of
the contract are felt) the more ‘international’ is the arbitration. Conversely, the
stronger the link between the seat and the arbitration in terms of such connecting
factors, the more ‘domestic’ is the arbitration. In other words, the transnationality
of an arbitration is a matter of difference in degree, depending on the presence or
absence of the relevant connecting factors, rather than a strict conceptual
distinction between ‘international’ and ‘domestic’ arbitration. In these
circumstances, distinguishing between the two forms of arbitration is essentially a
matter of policy rather than strict law.31 While one jurisdiction may draw the
distinction between domestic and international arbitration based on one set of
criteria, and another jurisdiction may draw it on the basis of another set of
30 This suggestion is not particularly novel or radical as it has been endorsed even in international civil
litigation. See e.g., the Rome Convention on the Law Applicable to Contractual Obligations, Art. 1(2)(a), and
the EU Regulation that replaced the Convention, Regulation 593/2008/EC of 17 June 2008 on the Law
Applicable to Contractual Obligations (Rome I), art. 9 (‘Effect may be given to the overriding mandatory
provisions of the law of the country where the obligations arising out of the contract have to be or have been
performed, insofar as those overriding mandatory provisions render the performance of the contract
unlawful. In considering whether to give effect to those provisions, regard shall be had to their nature and
purpose and to the consequences of their application or non-application.’). See also, Swiss Private
International Law Act, art. 19 (‘Taking into account of mandatory provisions of foreign law’) (‘1. If, pursuant
to Swiss legal concepts, the legitimate and manifestly preponderant interests of a party so require, a
mandatory provision of a law other than that designated by this Code may be taken into account if the
circumstances of the case are closely connected with that law. 2. In deciding whether such a provision must
be taken into account, its purpose is to be considered as well as whether its application would result in an
adequate decision under Swiss concepts of law.’).
31 In this purely formal or conceptual sense F.A. Mann certainly had a point when arguing that the term
‘international arbitration’ is a ‘misnomer’ since ‘[i]n the legal sense no international commercial arbitration
exists’. This is because ‘[j]ust as, notwithstanding its notoriously misleading name, every system of private
international law is a system of national law, every arbitration is a national arbitration, that is to say, subject
to a speciﬁc system of national law’. F.A. Mann, ‘Lex Facit Arbitrum’ in Pieter Sanders (ed.), International
Arbitration Liber Amicorum for Martin Domke (1967), pp. 157, 159. Mann’s controversy in the course of the 1980s
with those, like Jan Paulsson, who argued in favour of delocalisation of international arbitration (or rather,
the delocalisation of arbitral awards) is analogous to the controversy between conceptualists and modernists
in conﬂict of laws. Just as the conceptualists, Mann focuses on the link between a ‘category’ (international
arbitration proceedings) and a particular jurisdiction (the seat of the tribunal), whereas Paulsson, like the
modernists, focuses on a ‘weighing’ of the substantive public policies of the seat and the place where
recognition and enforcement of the arbitral award is sought. See Paulsson, ‘Delocalization of International
Commercial Arbitration’, supra n. 22 (‘So the award is set aside in country A. Why does this necessarily have
to mean that the award is annulled erga omnes? Could not country B recognize the award simply by holding
that under its law the result of the arbitration is perfectly valid and the reason for annulment in country A
is so peculiar to country A that the annulment is properly deemed to be limited to that country?’).
And/Or: The Problem of Qualiﬁcation in International Arbitration 455
criteria, one cannot say that one jurisdiction has drawn a proper distinction and
the other has acted improperly, so long as there is at least an arguable basis for
making the distinction in the way it has been made.32
However, while it is essentially a matter of legal policy as to how precisely to
distinguish between international and domestic arbitration, and indeed while
each jurisdiction remains free to decide whether international and domestic
arbitration should be governed by separate legal regimes, it is arguable that the
more transnational an arbitration is in terms of the relevant criteria, the weaker
the claim that mandatory laws and public policies of the seat should
automatically be given priority in the arbitration over any other applicable
mandatory laws and policies.33 Obviously, the public policy issues arising in the
context of a purely domestic arbitration are different from those arising in the
context of a truly transnational arbitration. The forum state arguably has a
legitimate interest in ensuring that parties to purely domestic arbitrations do not
abuse their right to choose the applicable law and the seat of the arbitration, for
the sole purpose of seeking to avoid mandatory rules of law of their domestic
jurisdiction. On the other hand, the state of the seat has only a very limited
interest in regulating and interfering with a truly transnational arbitration which
has no connection whatsoever with the seat of arbitration, apart from the seat
itself.
In any event, the fact that a particular jurisdiction has made such a policy
choice, and has drawn a distinction between domestic and international
arbitration as set out in the local arbitration law, does not fully settle the matter
for the arbitral tribunal. The degree of trasnationality of arbitrations varies not
only in terms of the distinction between ‘international’ and ‘domestic’ arbitration;
it also varies between arbitrations that qualify as ‘international’ arbitrations
within the meaning of the local arbitration law. What are the characterisation
issues and conﬂicts of laws that may arise in the context of such ex lege
‘international’ arbitrations?
b
32 Thus, while important venues of international arbitration such as France and Switzerland have both
distinguished in their domestic legislation between domestic and international arbitration, the relevant
distinctions are not identical. See French Civil Code of Procedure, art. 1492 (‘An arbitration is international
when it involves the interests of international trade’) and Swiss Private International Law Act, art. 176(1)
(‘The provisions of this chapter shall apply to arbitrations if the seat of the arbitral tribunal is in Switzerland
and if at least one of the parties at the time the arbitration agreement was concluded was neither domiciled
nor habitually resident in Switzerland’). Cf. UNCITRAL Model Law, Art. 1(3) (‘An arbitration is
“international” if: (a) the parties to an arbitration agreement have, at the time of the conclusion of the
agreement, their places of business in different States; or (b) one of the following places is situated outside the
State in which the parties have their place of business: (i) the place of arbitration if determined in, or
pursuant to, the arbitration agreement; (ii) any place where a substantial part of the obligations of the
commercial relationship is to be performed or the place with which the subject-matter of the dispute is most
closely connected; or (c) the parties have expressly agreed that the subject matter of the arbitration
agreement relates to more than one country.’).
33 As noted infra, this is recognised in Regulation 593/2008/EC and Swiss Private International Law Act,
art. 19, supra n. 30.
456 Arbitration International, Volume 26 Issue 4
(b) The Problem of Qualiﬁcation in International Arbitration
The shift in the context of international arbitration from the domestic/foreign
law framework to the transnational context has resulted in a corresponding shift
in the context of the problem of qualiﬁcation. The principal qualiﬁcation issue in
such a transnational context is not whether the relevant issues, categories or laws
should be characterised on the basis of domestic or foreign law; the principal
issue is how such qualiﬁcations should be made in a situation where the law of the
seat is only one among many applicable laws, the applicability of such other laws
depending on the context.
In such a transnational context, the problem of qualiﬁcation may arise on two
different levels. First, it may arise as a preliminary legal issue in terms of whether
a challenge to the arbitral tribunal’s authority over the claim should be
characterised as an issue of jurisdiction or admissibility. Secondly, it may also arise
in terms of whether there is any conﬂict of laws, both as a matter of law and as a
matter of fact. As noted above, for a conﬂict of laws to exist, as a matter of law,
both of the applicable laws must qualify as procedural rules or substantive norms,
respectively, and for a conﬂict of laws to exist as a matter of fact, both of the
applicable laws must qualify as mandatory rules of law or public policies in the
broad sense that they cannot be superseded or displaced by conﬂict rules or by
agreement between the parties.34
In the traditional conﬂict of laws, the question of whether such qualiﬁcation
issues should be resolved by reference to the law of the forum or the applicable
foreign law remains controversial, and there appears to be no generally accepted
position. Indeed, there appear to be compelling arguments against both
approaches. It has been argued that, if foreign law is to be qualiﬁed based on the
law of the forum, by reference to the corresponding domestic rule or institution,
the court may end up failing to apply a rule of foreign law in all cases where there
is a conﬂict of characterisations, or applying a rule of foreign law in cases where
such law (according to its own terms) should not be applied, thus effectively
distorting the foreign law and ending up applying neither the foreign law nor the
34 See e.g., George A. Berman, ‘Introduction: Mandatory Rules of Law in International Arbitration’ in (2007) 18
Am. Rev. Int’l Arb. 1 (‘In the private international law literature, a norm or rule of law is most often described
as “mandatory” when a court must apply it, even if the court, under the operation of its conﬂicts of laws
rules, would ordinarily apply some other body of law (often referred to casually as “the otherwise applicable
law”). As indicated …, other terms – often in other languages (lois de police in France, for example) – capture
much the same idea. On other occasions, authors deﬁne as “mandatory” those rules of law that cannot be
derogated from by private parties in the exercise of their autonomy’) (footnote omitted). See also, Pierre
Mayer, ‘Mandatory Rules of Law in International Arbitration’ in (1986) 2 Arb. Int’l 274 at p. 275 (noting that
the issue ‘has scarcely been analyzed in a serious manner’ and listing ‘competition laws; currency controls;
environmental protection laws; measures of embargo, blockade, or boycott; or laws falling in the rather
different category of legislation designed to protect parties presumed to be in an inferior bargaining position,
such as wage earners or commercial agents’ as examples of mandatory laws); Swiss Private International
Law Act, art. 17 (‘The application of provisions of foreign law shall be precluded if it would produce a result
which is incompatible with Swiss public policy (ordre public)’) and art. 18 (‘This Code does not prevent the
application of those mandatory provisions of Swiss law which, by reason of their particular purpose, are
applicable regardless of the law designated by this Code’).
And/Or: The Problem of Qualiﬁcation in International Arbitration 457
law of the forum.35 Similarly, it has been pointed out that if qualiﬁcation is
performed in accordance with the foreign law, this would mean that the forum
would lose control over the application of its own conﬂict of laws and would no
longer be the master of its own home.36 The problem gets even more complicated
in situations where the relevant categories in the foreign conﬂict of laws are not
the same as those of the forum.37
Preliminary qualiﬁcation issues involving jurisdiction and admissibility, on the
one hand, and substance and procedure, on the other, are also bound to arise in
the context of international arbitration, and when they do arise, they may be
complex.38 In the absence of an obvious ‘domestic’ law to fall back on, the sole
viable solution that would appear to be available to an international arbitral
tribunal is to approach such issues by reference to the degree of transnationality
of the arbitration. The less transnational (in other words, the more ‘local’) the
arbitration is in terms of the relevant connecting factors, the more appropriate it
would be for the arbitral tribunal to approach such issues on the basis of the
applicable local standards, i.e. on the basis of characterisations adopted in the law
of the seat. Conversely, the more transnational the arbitration is in terms of the
relevant connecting factors, the more appropriate it would appear to be for the
arbitral tribunal to resolve any such issues on the basis of transnational rather
than local standards.
This raises two distinct questions. First, how should an international arbitral
tribunal distinguish between contexts where characterisations should be made on
the basis of local standards, as opposed to those where such issues should be
resolved on the basis of transnational standards? And secondly, once the tribunal
has determined that it should indeed refer to transnational standards, where will
it ﬁnd such standards?
There is no easy or obvious solution to the ﬁrst question. Given that the degree
of transnationality of an international arbitration is a sliding scale, there is, again,
no standard case of international arbitration. Each arbitration tends to be unique
in terms of the relevant connecting factors: the seat of arbitration, the nationality
and/or principal place of business of the parties, the place of negotiation and
conclusion of the contract, the place of performance of the contract, the place or
places where the effects of the arbitration are felt, and the law governing the
contract. An arbitral tribunal must exercise its best judgment when determining
whether any characterisation issues that may arise should be resolved by
reference to local standards (the law of the seat) or transnational standards, and
take into account all the relevant circumstances. In other words, the matter is one
35 Dicey, Morris and Collins on The Conﬂict of Laws, supra n. 1 at pp. 39–40.
36 Ibid.
37 Ibid.
38 See e.g., V.V. Veeder, Towards a Possible Solution: Limitation, Interest and Assignment in London and Paris, ICCA
Congress Series No. 7 (1996), p. 268 (noting that the distinction between procedural law and substantive law
‘in England and elsewhere’ ‘is a twilight zone where even legal angels fear to tread’).
458 Arbitration International, Volume 26 Issue 4
of weighing the relevant factors and determining whether the links with the seat,
taken together, outweigh those with all other relevant jurisdictions.39
As to the second question – where can one ﬁnd the relevant transnational
standards, once the arbitral tribunal has determined that the transnational links
indeed outweigh the local links? – an international arbitral tribunal need not
necessarily go too far. It is suggested (as has been suggested by others)40 that this
is a largely a matter of comparative law and therefore a ﬁeld where there is no
need to make any strict conceptual distinctions between domestic and foreign
law or indeed between private and public international law. Qualiﬁcation of a
particular issue as a jurisdictional issue or as an issue of admissibility, on the one
hand, or as an issue of procedure or as an issue of substance, on the other, is
arguably a matter of legal methodology (or ‘lawyers’ law’) rather than a matter of
governmental regulation, or public policy, and as such need not be, and indeed
has no compelling policy reason to be, different in different jurisdictions. While
there may be instances where a local legislator has expressly qualiﬁed a particular
issue in particular methodological terms (for instance, as a matter of procedure
rather than as a matter of substance), there should generally be no compelling
public policy issues raised by such qualiﬁcations that would prevent an
international arbitral tribunal from recharacterising such issues, if required for
compliance with transnational legal standards.41
The problem of qualiﬁcation may also arise when determining whether there
is, as a matter of law and fact, any conﬂict of laws. There is a variety of laws that
may be applicable in the context of international arbitration, and there is
39 In other words, in respect of this issue a contextual approach seems more appropriate than a traditional
rules-based conﬂict-of-laws methodology.
40 See e.g., Jan Paulsson, ‘Jurisdiction and Admissibility’ in Gerald Aksen, Karl-Heinz Böckstiegel, Michael J.
Mustill, Paolo Michele Patocchi and Anne-Marie Whitesell (eds., 2005), Global Reﬂections on International Law,
Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, p. 616 (noting that ‘[t]his essay does
not treat private and public international law as separate worlds. It deals with a generic problem without
troubling itself with abstractions that are rapidly becoming obsolete, particularly in the ﬁeld of investment
arbitration under international treaties’).
41 If such an approach were to be adopted, there would be no basis to make any strict conceptual distinctions
between international commercial arbitration and international investment arbitration in terms of legal
methodology, i.e. in terms of the conceptual distinctions between jurisdiction and admissibility, on the one
hand, and procedure and substance on the other. Indeed, while the conﬂicts of laws that arise in the context
of investment treaty arbitration tend to concern the relationship between international law (i.e. the
investment treaty and applicable customary international law) and municipal law (i.e. the law of the host
state), this does not necessarily have any bearing on how the relevant legal issues should be conceptualised.
Moreover, in cases where an investment dispute arises out of an investment contract rather than an
investment treaty, the distinction between investment arbitration and international commercial arbitration
tends to collapse because in such instances the principal substantive issues tend to arise out of the
interpretation and application of the contract rather than the applicable law.
This is not to say that there are no differences; for instance, an ICSID tribunal dealing with contractual
disputes must take into account the jurisdictional limitations in the ICSID Convention. Another source of
differences may be the applicable law; indeed, the question of whether such ‘state contracts’ between states
and foreign investors are governed by national or international law was one of the most debated issues in the
ﬁeld until it was settled, at least in formal terms, in Art. 42 of the ICSID Convention. For further discussion,
see Heiskanen, supra n. 10.
And/Or: The Problem of Qualiﬁcation in International Arbitration 459
therefore also a variety of conﬂicts of laws that may arise in such a context,
including the following:
• conﬂicts between the law of the seat of arbitration and the law under which
one or both of the parties are organised and/or the law of their principal
place of business, concerning the capacity and/or authority to enter into
an arbitration agreement;42
• conﬂicts between the law of the seat and the law of the place where the
contract was concluded, concerning the validity of contracts;43
• conﬂicts between the law of the seat and the law governing the contract
concerning the validity of contracts;44
• conﬂicts between the law of the seat and the law governing the contract
concerning the enforceability of certain terms of the contract;45
• conﬂicts between the law of the seat and the law of the place of
performance of the contract concerning the validity of the contract and/or
the enforceability of certain terms of the contract;46
• conﬂicts between the law of the seat and the law of the jurisdiction where
the effects of the performance of the contract are felt concerning the
validity of the contract and/or the enforceability of certain terms of the
contract.47
However, such apparent conﬂicts between the applicable laws may not
necessarily amount to a true conﬂict of laws. A conﬂict of laws may be false in
two different ways. First, it may be false in formal terms if the two conﬂicting laws,
42 The possibility of conﬂicts between the law of the forum and the otherwise applicable laws is recognised in
Regulation 593/2008/EC, art. 9(1), (2) (‘1. Overriding mandatory provisions are provisions the respect of
which is regarded as crucial by a country for safeguarding its public interests, such as its political, social or
economic organization, to such an extent that they are applicable to any situation falling within their scope,
irrespective of the law otherwise applicable to the contract under this Regulation. 2. Nothing in this
regulation shall restrict the application of the overriding mandatory provisions of the law of the forum’), and
art. 21 (‘The application of a provision of the law of any country speciﬁed by this Regulation may be refused
only if such application is manifestly incompatible with public policy (ordre public) of the forum’). See also, Swiss
Private International Law Act, art. 177(2), which seeks to regulate such conﬂicts, to the extent that one of the
parties is a state (‘If one of the parties is a State or an enterprise dominated by or an organization controlled
by a State, it may not invoke its own law to contest the arbitrability of a dispute or its capacity to be subject
to an arbitration’).
43 Cf. Regulation 593/2008/EC, arts. 9(1), (2), 21, supra n. 42.
44 Ibid.
45 Ibid.
46 Ibid. See also, ibid. art. 3(3) (‘Where all other elements relevant to the situation at the time of the choice are
located in a country other than the country whose law has been chosen, the choice of the parties shall not
prejudice the application of provisions of the law of that other country which cannot be derogated from by
agreement’).
47 See Regulation 593/2008/EC, arts. 9(1), (2), 21, supra n. 42. See also, ibid. art. 3(4) (‘Where all other elements
relevant at the time of the choice are located in one or more Member States, the parties’ choice of applicable
law other than that of a Member State shall not prejudice the application of provisions of Community law,
where appropriate as implemented in the Member State of the forum, which cannot be derogated from by
agreement’).
460 Arbitration International, Volume 26 Issue 4
once applied, in fact produce the same result, and secondly, it may be false in
substantive terms because only one of the two laws is mandatory in the broad sense
of this term.48 In other words, a true conﬂict of laws may arise only when there is
a conﬂict between laws that are both mandatory in the sense that each of them
incorporates public policies that cannot be displaced by the otherwise applicable
law (i.e. the law that would be applicable under the relevant conﬂict rule, but for
the mandatory rule) or by private contract between the parties.49
The principal qualiﬁcation issue in such a context is, therefore, whether both
of the laws in question can be characterised as mandatory in the broad sense of
this term. A law may be considered mandatory for two different reasons: because
of its form, i.e. the law itself qualiﬁes itself as being mandatory, or because of its
substance in the sense that it lays down public policies that must be considered
applicable to the dispute regardless of the otherwise applicable law, or the terms
of the contract.50 It is the latter characterisation that raises the more complex
issues in the context of international arbitration. While the determination of
whether a particular law is formally mandatory can be made on the basis of the
face of the law itself, the question whether a particular law sets out mandatory
public policies applicable to the dispute cannot be resolved in the abstract, or
outside the concrete context. Again, since there is no one standard case of
international arbitration, the degree of transnationality of each arbitration depending
on the context, the more ‘local’ the arbitration is in terms of the relevant
connecting factors, the more justiﬁable it would be for the arbitral tribunal to
make the determination of whether the law in question is mandatory based on its
substance, or sets out public policies applicable in the context, by reference to the
international and national public policies applicable at the seat of arbitration.51
Conversely, the more transnational the arbitration is in terms of the relevant
connecting factors, the more justiﬁable it would be for the arbitral tribunal to
48 See e.g., Weinberg, supra n. 4 at p. 6 (noting the distinction between the two types of false conﬂicts and
crediting Currie for its discovery).
49 There has been some debate among international arbitration scholars and practitioners as to whether
mandatory laws are applicable in international commercial arbitration. There is broadly speaking a
consensus that they do apply, although it is less than clear how such laws are to be chosen. See e.g., Mayer,
supra n. 34 at p. 280 (‘An afﬁrmative answer to this question [i.e. to the question of whether the arbitrator
must apply mandatory rules of law] can hardly be doubted whenever the following three statements are true:
the mandatory rule belongs to lex contractus, the parties have not expressly excluded its application (which is
doubtless exceedingly rare in practice), and ﬁnally one of the parties has invoked it before the arbitrators’);
Marc Blessing, ‘Mandatory Rules of Law versus Party Autonomy in International Arbitration’ in (1997) 14(4)
J Int’l Arb. 23 (‘[A] substantial and growing percentage of cases are affected by the interference of mandatory
rules of law which claim or demand to be respected or to be applied directly, irrespective of any law or rules
of law chosen by the parties or determined by arbitral tribunal’) (footnotes omitted). For recent contributions,
see e.g., the articles published in the 2007 issue of the American Review of International Arbitration.
50 See supra n. 19 and accompanying text.
51 The law of the seat may or may not make a distinction between ‘national’ and ‘international’ public policy.
National public policy refers to the same public policy that would apply to a domestic arbitral award; it is the
‘internal’ public policy of the seat. International public policy, on the other hand, refers to a public policy
adopted in certain jurisdictions, which distinguish between public policies applicable to purely domestic
arbitral awards and ‘international’ arbitral awards. France is a well-known example of a jurisdiction that
applies a narrower concept of public policy to international arbitral awards. See French Code of Civil
Procedure, arts. 1498, 1502 and 1504.
And/Or: The Problem of Qualiﬁcation in International Arbitration 461
make such determinations by reference to transnational standards, i.e. by
reference to transnational public policy.52
The determination of whether a particular public policy qualiﬁes as transnational
public policy raises the question of how to distinguish between transnational
public policy, on the one hand, and national or international public policy, on the
other. Again, because there is no standard case of international arbitration, what
qualiﬁes as transnational public policy in one context does not necessarily amount
to transnational public policy in another context – even if the relevant issues on
their face appear to be the same.53 By way of an example, the transnational
public policy governing the regulation of competition between a German and an
Italian company in the context of an arbitration that takes place in The Hague is
not necessarily the same as the competition policies applicable in an arbitration
between a Chilean and a Brazilian company that takes place in Buenos Aires.54
This is because the substance of the relevant mandatory rules and public policies
applicable in the context, in light of all the relevant connecting factors, is not
necessarily identical.
In sum, because the determination of whether a particular public policy
qualiﬁes as transnational public policy depends on the substance of the laws in
question, it will always depend on the context. Consequently, the qualiﬁcation of
a particular public policy as transnational public policy in a particular context
does not amount to the creation or endorsement of a formal ‘category’ of
transnational public policy within the meaning of the classic conﬂict of laws. The
public policy of international arbitration is ultimately contextual.
c
(c) Conﬂict of Laws in International Arbitration
The shift in the context of international arbitration from the distinction between
domestic and foreign law to the transnational context has also produced a shift in
the substance of the relevant conﬂict of laws. The conﬂict issues faced by
international arbitral tribunals have become more complex – and also potentially
more sensitive – than they ever were in international civil litigation. This is not to
suggest that conﬂicts between mandatory laws and/or public policies are
common, or more common than, say, conﬂicts between the terms of the contract
52 For a seminal article on the concept of transnational public policy, see Pierre Lalive, ‘Transnational (or Truly
International) Public Policy and International Arbitration’ in ICCA New York Arbitration Congress (1986), p. 259.
53 Ibid. pp. 295–296 (‘The question also arises – although this may be rare in practice – whether the
transnational public policy of the arbitrator should lead him, in a given case, to ignore the international
public policy of a certain State, be it that of the place of arbitration or one of the States concerned in the
dispute … A general and abstract answer does not appear to have much sense, especially since the concept
of public policy … has a dynamic and evolutive character and must be considered in concreto, in the light of
all the circumstances of the case’) (footnote omitted).
54 In practice, differences regarding the characterisation of European competition law as a public policy have
arisen even within Europe. See e.g., Eco Swiss China Time Ltd v. Benetton International NV (Case C-126/97) (1999),
[1999] All ER (D) 574 (where the European Court of Justice held that Art. 81 of the EC Treaty constituted
a matter of public policy within the meaning of the New York Convention) and the decision of the Swiss
Federal Tribunal in Tensacciai v. Terra Armata, 8 March 2006 (ﬁnding that EC competition law did not
constitute public policy).
462 Arbitration International, Volume 26 Issue 4
and applicable mandatory laws.55 In order for a true conﬂict of laws to exist,
there must be a conﬂict of laws both as a matter of law and as a matter of fact.
Deﬁned in this way, true conﬂicts of laws are a rarity in international arbitration.
However, the possibility of conﬂicts between mandatory rules of law, in the
broad sense of this term, cannot be excluded. Given the sensitivity of the issue,
it would therefore seem important that there is an understanding among
international arbitration professionals as to how to approach such conﬂicts. In
circumstances where an international arbitral tribunal has no obvious ‘domestic’
law to rely upon, it must approach such issues keeping in mind the degree of
transnationality of the arbitration. In other words, the stronger the links between
the seat and the arbitration, based on relevant criteria such as the nationality of the
parties and the place of performance of the contract, and/or the place where the
effects of the performance of the contract are felt, the more justiﬁed it will be for
the arbitral tribunal to resolve conﬂicts of mandatory laws in favour of the law or
the public policies of the seat. Conversely, the more tenuous the link between the
seat and the arbitration, the more justiﬁed it will be for the arbitral tribunal to
resolve such conﬂicts by reference to a transnational standard.56 In other words,
instead of automatically giving priority to the mandatory law of the seat, the
tribunal should determine which of the two conﬂicting mandatory rules better
reﬂects transnational public policy on the subject and designate that law as
applicable in the matter.
Not every government regulation deserves to be recognised and given effect
by an international arbitral tribunal, in particular if it appears to be designed to
deal mainly with local policy concerns and interests. Such a regulation must
incorporate or reﬂect a transnational standard in order to deserve recognition
and application by an international arbitral tribunal, in case of a true conﬂict. It
does not necessarily or automatically deserve recognition and application merely
because the parties chose to locate the seat of the arbitration in that particular
jurisdiction, or merely because it forms part of the law of the jurisdiction in which
the contract was negotiated, or merely because it forms part of the law governing
the contract, or merely because it forms part of the law of the jurisdiction
in which the effects of the performance of the contract are felt. While these
are relevant criteria and the arbitral tribunal must consider the laws of such
jurisdictions, as applicable, it should not apply them blindly, but should consider
the substance of all such applicable laws. The law that best reﬂects a standard
that may be considered transnational in the context, taking into account all the
relevant connecting factors, should be the one that is given effect and applied to
resolve the conﬂict.
55 This is a subject that is much more debated than the conﬂict of mandatory laws. See supra n. 49.
56 This approach is also endorsed in international civil litigation, see Regulation 593/2008/EC, art. 9 (‘Effect
may be given to the overriding mandatory provisions of the law of the country where the obligations arising
out of the contract have to be or have been performed, insofar as those overriding mandatory provisions
render the performance of the contract unlawful. In considering whether to give effect to those provisions,
regard shall be had to their nature and purpose and to the consequences of their application or non-
application.’).
And/Or: The Problem of Qualiﬁcation in International Arbitration 463
When determining, in case of conﬂict between mandatory rules of law (in the
broad sense of this term), which of the competing mandatory rules better reﬂects
transnational public policy, international arbitral tribunals are not seeking to
create or establish any global transnational public policy. They are only
concerned with the transnational public policy that is applicable in the particular
context they are dealing with. To the extent that international arbitral tribunals
cannot avoid making policy decisions – and in case of a conﬂict between two
mandatory laws such a policy decision is virtually unavoidable – this does not
mean that they are engaged in the elaboration of transnational public policy
beyond the concrete context in which the particular policy conﬂict arises. In
order to exercise their function, and in order to resolve the disputes before them,
it is sufﬁcient if international arbitral tribunals choose between the applicable
policies, without attempting to formulate or establish any novel policies. Indeed,
they could not establish such policies even if they wanted to: given the contractual
basis of their jurisdiction, the legal effects of their decisions will in any event be
limited to the relationship between the parties and cannot affect the rights and
interests of third parties.57
Whether international arbitral tribunals will eventually adopt a rule-based
conﬂicts methodology or a more contextual approach, or indeed both, in any
particular context to resolve the conﬂict does not really matter.58 Such a choice
will not compromise the system of international arbitration since there is no
one standard case of international arbitration and therefore no ‘system’ of
international arbitration in the ﬁrst place. The degree of transnationality of each
arbitration depends on the context, and it would be misconceived to say that an
international arbitration taking place in Singapore between a Korean and a
Chinese company involving a large construction project governed by English
law is part of the same ‘system’ of international arbitration as an arbitration
taking place in Chile between a Chilean and a Peruvian company involving a
distribution agreement governed by the law of Chile. The two arbitrations each
form their own ‘system’. What may be legitimately expected and indeed required
by the parties to such arbitrations is that any conﬂicts of mandatory laws that may
arise are settled in a manner that takes properly into account their transnational
context. However, the ways in which these conﬂicts are resolved need not
necessarily be the same in each context because the applicable mandatory laws,
and consequently the applicable public policies, are not necessarily the same.59
Therefore, such transnational standards or approaches need not be, and indeed
57 Indeed, this is in substance what arbitrability means in practical terms: the subject matter (ratione materiae)
jurisdiction of an international arbitral tribunal only governs the relationship between the parties and the
legal effect of its decisions is limited accordingly. This is the case even when the decision is based on
mandatory law or resolves a substantive conﬂict of laws.
58 In other words, the proposed answer to the principal question of conﬂict of laws as a discipline – ‘whether
we should have rules or an approach’ – is that we should have rules and/or an approach. For further
discussion, see Willis Reese, ‘Choice of Law: Rules or Approach’ in (1972) 57 Colum. L Rev. 459.
59 It has been rightly said that the concept of transnational public policy is vague. See Lalive, supra n. 52 at
p. 259. Indeed, it is arguable that the concept of transnational public policy is by deﬁnition vague because its
content depends on the context.
464 Arbitration International, Volume 26 Issue 4
cannot be, identiﬁed in abstracto. They can only be identiﬁed in the concrete
context.
However, even though it is not possible to categorise such transnational
standards in abstracto, it is important that international arbitration scholars and
practitioners apply their minds to the issue and are prepared to deal with it, as it
is likely to present an increasingly important challenge to the function that
international arbitral tribunals play in the settlement of international commercial
disputes – a function for which there is currently no effective alternative.60
International arbitration has substantially displaced local courts as fora for the
settlement of international commercial disputes in many jurisdictions, but it
may not be advisable to assume that this trend will continue ad inﬁnitum. While
international arbitral tribunals, precisely because they operate in a transnational
context, are currently in a better position to deal with international commercial
disputes than local courts, this is at least in part because local courts are being
seen as more parochial than arbitral tribunals, and as such more likely to apply
local rather than transnational standards, even when this is arguably not
necessary, or indeed justiﬁed.
However, a transnational approach to conﬂict of laws need not be limited to
international arbitration. Nothing prevents local courts from following the lead of
international arbitral tribunals in shifting the context of their decision-making
from the domestic/foreign law framework to the transnational context, taking
into account the degree of transnationality of the dispute. For instance, nothing
would prevent local courts from adopting a transnational approach, within the
limits of the applicable local law, when dealing with challenges to arbitral awards
or requests for recognition and enforcement of arbitral awards. The existing legal
framework is certainly sufﬁciently ﬂexible to accommodate such a shift. Thus, the
UNCITRAL Model Law and many modern international arbitration laws set out
public policy as a basis for setting aside an arbitral award, and it is also one of the
bases for refusing recognition and enforcement of an arbitral award in the New
York Convention. It is notable that both the UNCITRAL Model Law and the
New York Convention use permissive and not mandatory language; public policy
may be a basis for setting aside an arbitral award, or refusing recognition and
enforcement, but it does not have to be.61
60 See Jan Paulsson, ‘International Arbitration is Not Arbitration’ in (2008) Stockholm Int’l Arb. Rev. 3 (stressing
that international arbitration is ‘the only game in town’ as there are no compulsory global commercial
courts, and that accordingly, ‘[i]f we do not deliver justice, if we do not close the door to abuse, we should
understand that sharp reactions are likely – sharp reactions which may harm a very valuable tool’).
See also, Mayer, supra n. 34 at pp. 285–286 (‘Although arbitrators are neither guardians of the public order
nor invested by the State with a mission of applying its mandatory rules, they ought nevertheless to have an
incentive to do so out of a sense of duty to the survival of international arbitration as an institution.’).
61 See UNCITRAL Model Law, Art. 34(2)(b)(ii) (‘An arbitral award may be set aside by the court speciﬁed in
Article 6 [the competent court] only if: … the court ﬁnds that: … the award is in conﬂict with the public
policy of this State’) and New York Convention, Art. V(2)(b) (‘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: … [t]he recognition or enforcement of the award would be contrary to the public policy
of that country.’).
And/Or: The Problem of Qualiﬁcation in International Arbitration 465
Consequently, even if a local court were to ﬁnd that a particular arbitral award
is incompatible with the forum’s public policy, this does not necessarily mean that
the court must set aside the award, or refuse recognition and enforcement, as
the case may be. There is ample room for discretion in the language of the
UNCITRAL Model Law and the New York Convention to adopt an alternative
approach, and it is arguable that this regulatory ﬂexibility should be applied by
taking into account the degree of transnationality of the award. The more
transnational the award is in terms of the relevant criteria, the narrower should
be the concept of public policy applied by the court. In case of a truly
transnational award, where the sole connecting factor between the award and the
seat is the seat itself, it is arguable that the local court should interpret the concept
of public policy accordingly and set aside the arbitral award, or refuse recognition
and enforcement, as the case may be, only if the award is incompatible with
transnational public policy.62
IV
IV. CONCLUSION
The problem of qualiﬁcation may or may not be the fundamental problem of
international arbitration, depending on one’s intellectual approach to conﬂict of
laws, but it does appear to provide an instructive framework for developing an
understanding of the role and function of conﬂict of laws in international
arbitration. While true conﬂicts of laws are and are likely to remain rare in
international arbitration, they may arise, and when they do arise, they tend to
raise sensitive issues of public policy, precisely because such true conﬂicts involve
conﬂicts between mandatory laws, in the broad sense of this term. Since the
jurisdiction of an international arbitral tribunal is based on a private contract,
international arbitrators are well advised to carefully consider the context of the
dispute when dealing with such conﬂicts.
The fundamental policy issue underlying conﬂict of laws in international
arbitration is in substance this: how transnational is the arbitration, in terms of
the relevant connecting factors, vis-à-vis the seat of arbitration? The more
transnational the arbitration is in light of these factors, the more justiﬁed it
arguably is to resolve the relevant qualiﬁcation issues and conﬂicts of laws by
reference to a transnational rather than a local standard. Just as the difference
between international and domestic arbitration is a difference in degree and as
such a matter of policy, the transnationality of an international arbitration
remains a sliding scale, or a difference in degree, and as such a matter of policy.
62 Such an enlightened (and as such somewhat unrealistic) approach would involve taking a step beyond the
recommendations made by the International Arbitration Committee of the International Law Association,
which took the view that recognition and enforcement may be refused if the award would be against the
international public policy of the forum. See Final Report on Public Policy as a Bar to Enforcement of International
Arbitral Awards, recommendation 1(b). On the other hand, the distinction between international public policy
and transnational public policy is ﬂuid rather than categorical, and not much would be lost if the concept of
international public policy were abandoned, or equated with transnational public policy, at least in the
context of truly international arbitrations. See Lalive, supra n. 52 at p. 311.
466 Arbitration International, Volume 26 Issue 4
Consequently, it is the degree of transnationality of the arbitration that
determines the dependency of an international arbitral tribunal on the public
policies of the seat.
In this context, the principal function of the process of qualiﬁcation is to
narrow down the issues to a point where a true conﬂict of laws becomes
inevitable. In other words, its principal function is to postpone the inevitable.