4

The Legal Framework of Arbitration: International

and National Sources
f a b i e n gél i n a s *

4.1 Introduction
The ambitions of this chapter are considerably more modest than its title suggests.
Although the history of arbitration as a legal institution is one that has important sites
and moments in both domestic and international settings, this chapter will focus on
international arbitration. This focus will exclude local legal norms intended to govern the
arbitration of domestic or intra-national disputes. However, this focus still leaves much
within the scope of this chapter because domestic sources remain relevant for inter-
national arbitration and because international arbitration is pervasive and multi-faceted.
It is practised between states or state entities, between private parties – be they natural
persons or corporations – and between states or state entities and private parties.
International arbitration is also conducted between sub-federal units and international
governmental as well as non-governmental organizations, and it can adapt to situations
involving associations or groupings which would not readily fit within any of those
descriptions.1 Arbitration may be used in and across most areas of international
activity, legal domains, and subject matters. Importantly, it is used to apply norms
that arise in any source, public or private, and any conceivable combination of these
norms.
The literature on international arbitration has been fragmented because the
practice of international arbitration has, for some time, also been fragmented. In
the space between the well-established (but relatively isolated) communities of legal
practice that operate in the ‘public-law’ arbitration of interstate disputes and the
‘private-law’ arbitration of international business disputes, there has been the rela-
tively recent development of ‘mixed’ investment arbitration that has caused those
two legal communities to move closer to one another, and to overlap. This has
enabled the consideration of international arbitration and its legal framework as

*
The preparation of this paper was facilitated by SSHRC and FQRSC grants. The author wishes to
thank Dr G. Marchisio, Dr L. Vanhonnaeker and Dr E. Bateman for their assistance, and his
colleague and friend Prof. A. Bjorklund for her comments on a previous version. All errors or
omissions are the author’s.
1
See, for example, Permanent Court of Arbitration, The Government of Sudan v. The Sudan People’s
Liberation Movement/Army (Abyei Arbitration), Award, 22 July 2009, Case no. 2008–07.
81

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 82 f ab ie n gé linas

a unified subject of inquiry, albeit not an easy one.2 In this chapter, I will endeavour
to focus on the commonalities that this perspective highlights.
In the first part of the chapter, I briefly consider the development of the different
worlds of international arbitration since the late nineteenth century by dwelling on
elements often overlooked when the recent story of international arbitration is told.
In the second part of the chapter, I provide an overview of the potential features and
the hurdles that stand in the way of a unified representation of international
arbitration.

4.2 Arbitration as an International Legal Construct

4.2.1 Accounts of International Arbitration: Public, Private, and Mixed
If one steps back and recalls that the state as we know it is a relatively recent
construct, it becomes clear that something like arbitration must have been the
default means of adjudication throughout much of human history. Others have
looked, and will continue to look, at ancient and medieval civilizations to discover
and better understand the countless instantiations and forms of arbitration that have
existed.3 However, the recent re-emergence of arbitration as an ‘international’ legal
institution is a more interesting story for our purposes. This story often places
emphasis on private arbitration by focusing on the creation of an international
commercial arbitration system in the first half of the twentieth century that is
supported by a succession of treaties ensuring its cross-border recognition and
enforceability. International commercial arbitration, having met with considerable
success as an effective cross-border system of dispute resolution, naturally served as
a model for the kind of ‘mixed’ arbitration of investment disputes that was multilat-
erally formalized in the 1960s by the adoption of the International Centre for the
Settlement of Investment Disputes (ICSID) Convention.4 Such an account is useful
to understand many of the now common features of the practice of ‘international
arbitration’. However, what this account often leaves out is a discussion of what
international arbitration owes to prior efforts in the international realm of interstate
arbitration.
A popular representation of contemporary international arbitration is that it
emerged after the First World War as part of an effort led by the International
Chamber of Commerce (ICC), a private organization dedicated to the facilitation

2
For an earlier call for a unified study of international arbitration, see M. Chapman, ‘The Interlaced
History of Public and Commercial Arbitration, 1794 to 1999’, in International Bureau of the PCA
(ed.), International Alternative Dispute Resolution: Past Present and Future. The Permanent Court of
Arbitration Centennial Papers (Kluwer, 2000), 75, 80 (‘The interlocking events of public and private
dispute resolution, to me at least, emphasise the tragedy that the two are not taught side by side. They
grew up together. As adults they may not have maintained contact, but to understand them you must
understand their development. To do that, the influences of one upon the other must be studied.’).
3
See, for example, H. S. Fraser, ‘A Sketch of the History of International Arbitration’, Cornell Law
Quarterly, 11 (1926), 179, 188.
4
Convention on the Settlement of Investment Disputes Between States and Nationals of Other
States, 18 March 1965, entered into force 14 October 1966, 575 UNTS 159.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 the l ega l framework of arbi trati on 83

of international commerce in the name of peace.5 At the beginning of the 1920s,

the ICC set up a service for conciliation through good offices and the arbitration
of international business disputes.6 The arbitration system was made effective by
first relying not only on the limited means of enforcement then provided by
national laws,7 but also on disciplinary measures taken by trade associations
(where available),8 and on the economic pressure resulting from the publication
of information about instances of non-compliance with arbitral awards.9 While
taking advantage of these pre-state normative mechanisms associated with the lex
mercatoria, or law merchant, the ICC also successfully campaigned in favour of
an international legal regime, which came to be known as the Geneva regime, for
the cross-border recognition by national laws of arbitration agreements,10 and for
the enforcement of the resulting awards through national legal processes.11 In the
1950s, the ICC proposed a new treaty framework for arbitration that would
consolidate and build on the Geneva regime, ensuring party autonomy in matters
of procedure and making arbitral awards more easily enforceable across
borders.12 A 1953 ICC proposal was taken up by the United Nations, through
its Economic and Social Council (ECOSOC), and this led to the adoption of the
New York Convention for the Recognition and Enforcement of Foreign Arbitral
Awards (New York Convention) in 1958.13 The story goes that, with the intro-
duction of the New York Convention, arbitration grew exponentially and a model
of international adjudication was established that was eventually taken up in
other contexts, most notably that of investment law.14 In this account, investment
arbitration broadly adopted the commercial arbitration model, including its key
practices and procedural legal framework. Not only were investment arbitration
instruments inspired (in large measure) by the rules and procedures initially

5
See, generally, G. L. Ridgeway, The Merchants of Peace: Twenty Years of Business Diplomacy
through the International Chamber of Commerce (Columbia University Press, 1938).
6
International Chamber of Commerce, Rules of Procedure for Conciliation and Arbitration, ICC
Publication No. 21, Paris, 1923 (1923 ICC Rules).
7
1923 ICC Rules, 21 (‘in a number of countries, it is not as yet possible to obtain legal sanction for
the enforcing of arbitration awards’).
8
1923 ICC Rules, s. C, Art. XLI b).
9
1923 ICC Rules, s. C, Art. XLI c):
The Court of Arbitration of the International Chamber of Commerce shall also have
the right to request that the name of the defaulting party be published in the official
publications of the International Chamber of Commerce, in those of the National
Committees, together with the text of the award so remaining unexecuted.
10
Geneva Protocol on Arbitration Clauses, 24 September 1923, entered into force 28 July 1924, 27
LNTS 157.
11
Geneva Convention on the Execution of Foreign Awards, 26 September 1927, entered into force
25 July 1929, 92 LNTS 301.
12
ICC, Report and Preliminary Draft Convention, Publication no. 174, adopted on 13 March 1953
(ICC, 1953), 9–10 (reprinted in ICC International Court of Arbitration Bulletin, 9(1) (1998), 32).
13
Convention on the Recognition and Enforcement of Foreign Arbitral Awards, 10 June 1958,
entered into force 7 June 1959, 330 UNTS 38.
14
G. B. Born, ‘A New Generation of International Adjudication’, Duke Law Journal, 61(4) (2012),
775, 826–7.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 84 f ab ien g é linas

designed for commercial arbitration,15 but the investment treaties themselves

generally provide arbitration as an option under sets of rules originally drafted
with international commercial arbitration in mind.16
Absent from this story is the significant contribution of public international law to
today’s mainstream practice of international arbitration. In fact, by the end of the
nineteenth century, there was already a significant public law heritage to draw upon
when creating and steering the norms of arbitration practices that would persist into the
twentieth century.17 That heritage is still of relevance today, even if the establishment of
the United Nations has given the International Court of Justice pride of place in the
public consciousness of public international law adjudication. The keen observer will
note that the establishment of the International Court of Justice, and before it, the
Permanent Court of International Justice, cannot obscure the fact that arbitration has
been – and remains to this day – the most favoured mechanism for the adjudication of
disputes under public international law.18 What is more, even the most institutionalized
international courts, including the International Court of Justice, are still premised, for
the most part, on the consent model that defines arbitration: a model where there is no
jurisdiction unless both parties consent to it, in relation to a specific dispute or more
generally. The rich public law arbitration heritage upon which international commercial
arbitration was built consisted of a shared and evolving interstate arbitration practice,
accompanied by normative formulations of principles and procedures, for steps such as
the constitution of tribunals, the conduct of proceedings, and the rendering of decisions.
It is difficult to assess with any degree of precision the actual extent of this practice
prior to the twentieth century without engaging in historical research that would
take me beyond the scope of this chapter. However, by any measure, it is far from
insignificant. Leaving aside the large number of panels established under the Jay
Treaty of 1794, at least 158 interstate arbitral tribunals were constituted between
1814 and 1899.19 Towards the end of that period, a Code of International Arbitration
was adopted by the Peace Congress at Antwerp in 1894.20 The Peace Congress,
which looked to arbitration as a means of promoting peace among nations, was the

15
See C. McLachlan et al., International Investment Arbitration: Substantive Principles (Oxford
University Press, 2017), 48, para. 3.01 (observing that regarding the internal procedure of invest-
ment treaty arbitrations, ‘such arbitrations closely resemble international commercial arbitrations’).
16
The UNCITRAL Rules of Arbitration, which feature as an option in a large number of BITs, were
initially designed for commercial disputes and are a prime example. Other examples include the
Stockholm Rules (see, for example, Energy Charter Treaty, 17 December 1994, entered into force
16 April 1998, 2080 UNTS 100, Art. 26(4)(c)) and the ICC Rules (see, for example, Agreement
between the Government of the Hashemite Kingdom of Jordan and the Republic of Poland on the
Reciprocal Promotion and Protection of Investments, signed 4 October 1997, entered into force
14 August 1999, Art. 7(2)).
17
Born, ‘A New Generation of International Adjudication’ (2012), 795–800.
18
Born, ‘A New Generation of International Adjudication’ (2012), 859 et seq.
19
Note that many of those tribunals dealt with the claims of private parties through public international
law espousal of claims. See A. M. Stuyt (ed.), Survey of International Arbitrations: 1794–1989 (Kluwer,
1990). On the historical significance of the Jay Treaty for international arbitration, see, for example,
Chapman, ‘The Interlaced History of Public and Commercial Arbitration’ (2000), 76–7.
20
W. Evans Darby, International Arbitration: Treaties and Procedure; A Supplement to
‘International Tribunals’ (Peace Society, 1897).

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 the l ega l framework of arbi trati on 85

initiative that eventually led to the creation of the Permanent Court of Arbitration
(PCA) at The Hague in 1901 and the adoption of the Convention on the Pacific
Settlement of International Disputes in 1907.
However, during the 1870s, the practice of interstate arbitration had already been
codified to a remarkable extent, prompted in particular by the peaceful arbitral
resolution of the Alabama claims21 and, on the heels of this success, the work of the
Institut de droit international.22 Some of the most important concepts of international
arbitration were shaped and formulated in this public international law setting before
they were taken up by commercial arbitration and adopted in the resulting practice of
investment arbitration. So, if we take a bird’s eye view of international arbitration over
a few centuries, it appears as a subject worth studying as one whole phenomenon
comprised of shared and evolving normative practices, even if those practices have at
times been experienced, viewed, and presented as distinct.

4.2.2 Principles from a Public Law Past

To help dispel the all-too-common notion that international arbitration is essentially
a private instrument created and mobilized by private actors to further private interests,
it is useful not only to recall that the resolution of disputes – even of purely private
disputes – is widely considered a public good but also to reflect on the contributions that
public law has made to some of arbitration’s most important principles. This historical
perspective sheds light on arbitration as an international legal institution, whose
framework of rules and principles crosses the public-private boundary to provide
a modest – but crucial – measure of the rule of law to transnational relations.23 Three
examples demonstrate the contributions made by public law in this regard.24
21
Arbitral Tribunal established under Art. I of the Treaty of Washington of 8 May 1871, Alabama
claims of the United States of America v. Great Britain, Award, 14 September 1872, United
Nations, Reports of International Arbitral Awards, vol. XXIX, 125–34, available at legal.un.org
/riaa/cases/vol_XXIX/125-134.pdf (last accessed 30 September 2019).
22
See Draft Regulation for International Arbitral Procedure, Annuaire de l’Institut de droit inter-
national, vol. 1 (1875), 31. Available in English in Resolutions of the Institute of International Law
Dealing with the Laws of Nations (collected and translated by J. Brown Scott, Oxford University
Press, 1916) (Institut de droit international, 1875 Draft Regulation for International Arbitral
Procedure).
23
On the theme of public law’s contribution to the principles of arbitration, see J. Crawford,
‘Continuity and Discontinuity in International Dispute Settlement: An Inaugural Lecture’,
Journal of International Dispute Settlement, 1(1) (2010), 3.
24
There are many examples, including the arbitral competence-competence principle, discussed by
Crawford, ‘Continuity and Discontinuity in International Dispute Settlement’ (2010). For early
examples of the competence-competence principle in arbitration rules, see, for example, Arts.
53(1) and 73 of the 1907 Convention for the Pacific Settlement of International Disputes,
18 October 1907, entered into force 26 January 1910, available at www.pca-cpa.org/wp-content/
uploads/sites/6/2016/01/1899-Convention-for-the-Pacific-Settlement-of-International-Disputes.pdf
(last accessed 30 September 2019) (Hague Convention I), and Art. 14 of the Institut de droit
international’s 1875 Draft Regulation for International Arbitral Procedure). Other examples include
the binding effect of arbitral agreements for future disputes (see, for example, Art. 39 of the Hague
Convention I); the mechanism of party-appointed arbitrators (see, for example, Art. 45 of the Hague
Convention I and Art. 10 of the Code of International Arbitration approved by the Peace Congress,

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 86 f ab ien g é linas

A first example is the notion that an agreement to arbitrate future disputes is

binding and irrevocable. This may seem obvious today, but it was not so in the late
nineteenth and early twentieth centuries, because in several legal systems, theoretical
objections traditionally prevented the recognition or enforcement of agreements to
arbitrate future disputes.25 In order for an arbitration procedure to be set in motion,
it was often considered necessary for the parties to sign what is commonly referred to
as a compromis, i.e., an agreement submitting an existing dispute to arbitration. This
requirement made arbitration largely ineffective because the defending party could
disregard a prior undertaking to arbitrate by refusing to sign a compromis and thus
prevent the arbitral procedure from being set in motion. This situation prevailed in
many municipal legal systems (in respect of all arbitrations, domestic or inter-
national) well into the twentieth century.26 This was seen as a key impediment,
one that was addressed for non-treaty arbitration clauses by the Geneva Protocol of
1923.27 However, at the international level, the Institut de droit international had
already recognized this principle, for the purpose of interstate arbitration, as early as
1875.28 Article 1 of the Draft Regulations makes it clear that an arbitration

held at Antwerp, at its sitting of 30 August 1894 (1894 Code of International Arbitration)); the
freedom of the parties to agree on the rules applicable to the arbitration procedure (see, for example,
Art. 51 of the Hague Convention I and Art. 19 of the 1894 Code of International Arbitration); the
freedom of the arbitration tribunal to require the production of evidence (see, for example, Art. 68 of
the Hague Convention I); decisions by a majority of the members of an arbitration tribunal (see, for
example, Art. 78 of the Hague Convention I and Art. 21 of the Institut de droit international’s 1875
Draft Regulation for International Arbitral Procedure); the obligation to give the reasons on which an
award is based (see, for example, Art. 79 of the Hague Convention I, Art. 23 of the Institut de droit
international’s 1875 Draft Regulation for International Arbitral Procedure, and Art. 42 of the 1894
Code of International Arbitration); and the finality of awards, except for issues of interpretation (see,
for example, Art. 81 of the Hague Convention I).
25
This was notably the case in the civil and Islamic law traditions and was related to the uncertainty
of the agreement’s object before a dispute arose. See generally V. Powell-Smith, ‘Settlement of
Disputes by Arbitration under Sharī‘ah and at Common Law’, Islamic Studies, 34(1) (1995), 5, 11–
2 and M. Rubino-Sammartano, International Arbitration: Law and Practice, 3rd ed. (Juris, 2014),
154 et seq.
26
See, for example, the survey conducted by the ICC in 1927: Supplement to International Chamber
of Commerce Journal no. 12, Arbitration Report no. 7, January 1927.
27
Geneva Protocol on Arbitration Clauses, 27 LNTS 157, Art. 1:
Each of the Contracting States recognizes the validity of an agreement whether
relating to existing or future differences . . . by which the parties to a contract agree
to submit to arbitration all or any differences that may arise in connection with such
contract . . .
28
Institut de droit international, 1875 Draft Regulation for International Arbitral Procedure.
Opposition to this principle continued to be felt until the late 1950s, when International Law
Commission Special Rapporteur Georges Scelle was able to confirm that ‘the idea cannot . . . be
entertained that if there is no compromis, or if the compromis has not yet been drawn up, a “bare
undertaking” to arbitrate is not binding because it is only an abstract promise relating to
hypothetical or future disputes’, while lamenting that, ‘[u]nfortunately, Governments accustomed
to the diplomatic technique of arbitration frequently incline to the view that, until a compromise
relating specifically to a particular dispute has been reached or made final, no legal obligation
exists’. In some cases, this understanding was encouraged by the absence of an agreed mechanism
whereby a tribunal could be constituted through a neutral third party in case of disagreement

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t he l e ga l f ra m ew o r k of a rb i tr at i o n 87

agreement made by means of a valid international treaty may be concluded ‘in

advance, either for all differences or for differences of a certain kind to be deter-
mined, that may arise between the contracting States’, or, ‘[f]or one difference or
several differences already arisen between the contracting States’.29 This was
confirmed in the 1907 Convention for the Pacific Settlement of International
Disputes, where Article 39 recognizes that an arbitration agreement ‘is concluded
for questions already existing or for questions which may arise eventually’. Thus,
the principle of the legal validity of an undertaking to arbitrate future disputes, i.e.,
the principle that such an undertaking is legally binding, was established as
a matter of public international law for the purpose of interstate arbitration long
before the Geneva Protocol on Arbitration Clauses came about in 1923.30
A second example is the principle of party autonomy in the determination of the
procedural rules that will govern the arbitral procedure, such as the rules for the
constitution of the arbitral tribunal, the rules concerning the proceedings before
the tribunal, including rules for the taking of evidence, and the rules governing the
making of the award. Party autonomy in this context implies a form of procedural
contract, the recognition of which may be constrained in some ways by the law
governing that contract. The Geneva Convention on the Execution of Foreign
Awards clearly recognized party autonomy in the constitution of the tribunal. In
accordance with Article 1 al. 2 c) of that instrument, enforcement of an arbitral
award can be granted, provided ‘the tribunal was constituted in the manner agreed
upon by the parties’.31 However, the Geneva regime did not establish that the
agreement of the parties on procedural matters (other than the composition of the
tribunal) could depart from or replace the procedural law of the seat. This was
achieved, albeit indirectly, by Article V(1)(d) of the New York Convention, which,

between the state parties. The Rapporteur clarified the situation as follows: ‘The truth is that, even
then, Governments are bound by an implicit obligation, namely, the obligation to conclude the
compromis and thereafter to comply with the decision delivered under it’ (Report of the
International Law Commission to the General Assembly Concerning a Draft on Arbitral
Procedure, 6 March 1958, Doc. A/CN.4/113, in Yearbook of the International Law Commission,
vol. II (UN, 1958), 1, 3). The Draft was eventually adopted by the Commission at its 473rd
meeting, in the form of a set of model draft articles entitled ‘Model Rules on Arbitral Procedure’
(Yearbook of the International Law Commission, vol. II (UN, 1985), 83). Not mentioned by the
Rapporteur here is the view reported by James Crawford, which he calls the ‘contingency of
consent’, that a party was in some contexts free to withdraw even after entering a compromis, at
any time before the rendering of an award (withdrawal would represent a breach of the agreement
but would have the effect of revoking the tribunal’s authority to decide). Crawford cites only
domestic sources of English law, however: Crawford, ‘Continuity and Discontinuity in
International Dispute Settlement’ (2010), 10–2.
29
Institut de droit international, 1875 Draft Regulation for International Arbitral Procedure.
30
Note that the practical impact of the principle was dependent on the establishment of mechanisms
which, by resorting to third party appointments where needed, would ensure the constitution of
a tribunal where the respondent failed to make the required appointment(s) within the appropri-
ate time. This was resolved in the ICC Rules of Arbitration in 1931: F. Eisenmann, ‘The Court of
Arbitration – Outline of its Changes from Inception to the Present Day’, in ICC (ed.),
International Arbitration: 60 Years of ICC Arbitration – A Look at the Future (ICC Publication
no. 412, 1984), 410.
31
Geneva Convention on the Execution of Foreign Awards, Art. 1 al. 2 (c).

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 88 f ab ien gé linas

within the bounds of public policy, makes the law of the arbitral seat subsidiary to
party autonomy on all procedural matters.32
In interstate arbitration, party autonomy in matters of arbitral procedure – and
indeed in all matters of adjudicative jurisdiction – has always been the governing
principle. This principle was nevertheless explicitly recognized for interstate arbitra-
tion in the Draft Regulations of 1875. Article 3 deals with the constitution of the
tribunal, which it subjects directly to the agreement of the parties, or to an agreement
‘that the arbitral tribunal should be formed by a third person designated by them’. In
terms of the procedure to be followed before the tribunal, Article 12 clearly lays out the
role of party autonomy in this arbitral context (note that the word ‘compromis’ in this
document encompasses agreements to arbitrate either existing or future disputes). ‘If
the compromis or a subsequent convention between the parties prescribes for the
arbitral tribunal the procedure to be followed, or the observance of a determined and
positive law of procedure, the arbitral tribunal must conform to that provision.’
Although the provision could be read as referring to a municipal procedural law as
a possible choice of the parties, the fallback position is not the municipal law of the seat
(which could, problematically in the context of interstate arbitration, subject the state
parties to the municipal law of another state) but rather the procedure chosen by the
tribunal. In the absence of provisions made by the parties, ‘the procedure to be
followed shall be freely chosen by the arbitral tribunal’. This position is precisely
that which commercial arbitral institutions such as the ICC strove to attain in the
period between the Geneva regime and the New York Convention regime. A clear
formulation of the now established position can be found in the hierarchy of sources
for procedural rules, laid out in Article 19 of the ICC Rules of Arbitration:
The proceedings before the arbitral tribunal shall be governed by the [ICC] Rules
and, where the Rules are silent, by any rules which the parties or, failing them, the
arbitral tribunal may settle on, whether or not reference is thereby made to the
rules of procedure of a national law to be applied to the arbitration.
Therefore, we see that the possibility of an ‘internationalized’ procedural frame-
work based on party autonomy which this provision represents, and which
commercial arbitration formally established only in the late 1950s, had a long
and well-documented model in interstate arbitration.33

32
This feature was a key point in the 1953 draft prepared by the International Chamber of
Commerce, which was instrumental in establishing the now current notion of an international-
ized arbitration procedure governed primarily by a set of international arbitration rules agreed by
the parties and intended to displace the default provisions of the law of the seat. See ECOSOC,
‘Statement Submitted by the International Chamber of Commerce, a Non-governmental
Organization Having Consultative Status in Category A’, 10 September 1953, E/C.2/373, 8.
33
Article 52 of the Hague Convention I affords another example of party autonomy being treated as
the starting point for the settlement of all procedural matters in interstate arbitration. The second
paragraph, in particular, states that the compromis between the parties defines ‘the manner of
appointing Arbitrators, any special powers which may eventually belong to the Tribunal, where it
shall meet, the language it shall use, and the languages the employment of which shall be
authorized before it, and, generally speaking, all the conditions on which the parties are agreed’.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t h e l eg a l f r a m e wo r k of a r b i tr a t i o n 89

Our third and last example also focuses on party autonomy. It is the principle
whereby the parties to an arbitration enjoy the freedom to designate the substantive
rules of law to be applied by the tribunal. In interstate arbitration proceedings, the
governing substantive law is usually public international law. General public inter-
national law may, of course, be departed from by treaty, within the boundaries set by
jus cogens, so party autonomy in the designation of substantive law appears unsur-
prising. The provisions of a treaty applicable between the parties are law as between
them, such that the parties can replace general law with the particular law, or lex
specialis, of a treaty. However, in the arbitral context, the principle of party autonomy
goes even further, for it allows the parties to designate, after the fact, the substantive
rules that the tribunal will apply to an existing dispute. This principle was recognized
on the occasion of the Alabama claims arbitration between Great Britain and the
USA.34 In that case, the British government had refused to acknowledge the pre-
existence of international legal obligations, including a duty to exercise due diligence
in preventing its ports or waters from being used to facilitate naval operations against
a power at peace, but was at the time of the arbitration willing to abide by such an
obligation and to have it applied to the facts surrounding the Alabama claims. With
a view to implementing this diplomatic position, the detailed rules to be applied by the
arbitrators were set out in writing in a compromis which, inter alia, stated:35
Her Majesty’s Government can not assent to the foregoing rules as a statement of
principles of International Law which were in force at the time when the claims
mentioned in Article I arose; but Her Majesty’s Government, in order to evince
its desire of strengthening the friendly relations between the two countries and of
making satisfactory provision for the future, agrees that, in deciding the ques-
tions between the two countries arising out of those claims, the Arbitrators
should assume that Her Majesty’s Government had undertaken to act upon
the principles set forth in these rules.
A few years later, the principle of party autonomy in this interstate arbitral context
was formulated as follows: ‘[t]he arbitral tribunal gives judgment according to the
principles of international law, unless the compromis imposes upon it different rules
or leaves the decision to the free discretion of the arbitrators’.36

34
Arbitral Tribunal established under Art. I of the Treaty of Washington of 8 May 1871, Alabama
claims, United Nations, Reports of International Arbitral Awards, vol. XXIX, 125. Alabama was
a confederate raider used for attacks on Union ships during the US-American Civil War.
35
Treaty of Washington of 1871 between the United States and Great Britain, Art. VI, available at
www.trans-lex.org/502000/highlight_treaty_of_washington_1871/treaty-between-the-united-
states-and-great-britain-1871-/#head_0 (last accessed 30 September 2019). The background law
to be applied remained public international law, as stated in Art. VI:
In deciding the matters submitted to the Arbitrators, they shall be governed by the
following three rules, which are agreed upon by the High Contracting Parties as
rules to be taken as applicable to the case, and by such principles of International
Law not inconsistent therewith as the Arbitrators shall determine to have been
applicable to the case.
36
Note, again, that in this document, the word compromis refers to an arbitration agreement
concerning either existing or future disputes.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 90 f abie n gé linas

Again, in crucial respects, this is the position which commercial arbitration strove
to achieve over time by gradually allowing parties not only to designate a governing
law once the dispute has arisen – (a position now recognized in national judicial
contexts37) – but also to designate as governing law ‘rules of law’ that are not the
substantive rules of a recognized legal system:38 (a position still resisted in national
judicial settings).39 In both respects, we see that interstate arbitration provided a model
with which commercial arbitration, and arguably investment arbitration,40 came to be
aligned.
The three examples given here not only illustrate public law’s contribution to
international arbitration but also show that the contemporary forms of international
arbitration may usefully be viewed as instantiations of one legal construct built upon
a common framework. This view is understandably the subject of resistance because it
sets up arbitration as a challenge to many entrenched distinctions in the global
representation of law. The challenge arises from the fact that arbitration effectively
bridges the public/private divide and, though this is largely beyond the scope of this
chapter, the international/municipal divide.41 Thinking about arbitration’s legal
framework forces us to break out of some of the perceived ‘watertight compartments’
of legal orders and legal systems and offer a better representation of international legal
practice that accounts for significant crossover between those compartments.

4.3 International Arbitration’s Legal Framework: A Unified

Representation?
The time has certainly come for a less fragmented representation of international
legal practice.42 The phenomenal development of investment arbitration challenges
37
See Regulation (EC) No. 593/2008 of the European Parliament and of the Council of 17 June 2008
on the law applicable to contractual obligations (Rome I), OJ L 177, 4 July 2008, 6–16, Art. 3.
38
On ‘rules of law’, or ‘non-state law’ in the national judicial context, see generally, for example,
L. Gama Jr and G. Saumier, ‘Non-State Law in the (Proposed) Hague Principles on Choice of Law
in International Contracts’, in D. P. Fernandez Arroyo and J. J. Obando Peralta (eds.), El Derecho
Internacional Privado en los Procesos de Integracion Regional (Editorial Juridica Continental,
2011).
39
Gama Jr and Saumier, ‘Non-State Law in the (Proposed) Hague Principles’ (2011). See also, for
example, K. Boele-Woelki and V. Lazić, ‘Where Do We Stand on the Rome I Regulation?’, in
K. Boele-Woelki and W. Grosheide (eds.), The Future of European Contract Law (Kluwer, 2007),
27–30.
40
Although I am not aware of an investment case where the state party agreed during the arbitration
to designate governing rules departing from the relevant treaty, the wording of Art. 42 of the
ICSID Convention suggests that this is possible.
41
Nowhere is this more apparent than in the governing law provision of Art. 42 of the ICSID
Convention, which brings together the whole range of possible legal sources: private and public,
domestic and international. On the historical tensions and developments that led to Art. 42, see
notably D. Di Pietro, ‘Applicable Law under Article 42 of the ICSID’, in T. Weiler (ed.),
International Investment Law and Arbitration: Leading Cases from the ICSID, NAFTA, Bilateral
Treaties and Customary International Law (Cameron May, 2005), 223.
42
See, for example, Chapman, ‘The Interlaced History of Public and Commercial Arbitration’
(2000), 84 (‘I spoke earlier of the dimorphic classification imposed on arbitration.
Developments over the last couple of decades are very much against such a cosy view.’).

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 the l egal framework of arbitra tion 91

existing conceptions of international law in crucial ways, and commentators, as well

as actors, are now adapting mainstream representations to new realities. Obviously,
this adaptation cannot happen overnight, for law’s inclination in the face of change is
to carve a way forward by slowly drawing strength and legitimacy from existing
artefacts (concepts, principles, decisions, text) found in legally recognized practices.
Law succeeds in stabilizing human relations over time by requiring that change be
justified at least in part by reference to its own constrained resources. This means
that legal change is often more easily resisted than effected.
However, the potential for a unified representation of international legal practice
is real. The three examples above illustrate the contributions of public law to the
principles of international arbitration and suggest that a unified representation of
international arbitration’s legal framework may have to recognize the primary
importance of the principle of party autonomy. After all, the three examples are
linked to that principle known as pacta sunt servanda, and arbitration relies first and
foremost on the agreement of the parties who subject their disputes to it. I will,
therefore, take it as my starting point for the second part of this chapter.

4.3.1 Consent and Law as a Basis of Obligation

It is trite to say that international arbitration is based on consent at every stage and
every level. When referring to consent to describe the basis of arbitration, one is
usually taken to mean that the jurisdiction of an arbitral tribunal comes from an
agreement between the parties rather than from a legal rule imposed on them. This is
the most basic sense in which arbitration rests on consent, and it cuts across all
conceivable sites of arbitration. As we saw in our first example, consent is now
recognized as a valid basis for a legal obligation to arbitrate future and existing
disputes. As we saw in our second and third examples, the principle of consent
allows the parties to choose the arbitrators or the method of selecting arbitrators, the
procedure to be followed, and the substantive rules governing the merits.
The power of consent is particularly visible in the ability of the parties to
determine the rules of law governing the merits. Unlike most courts, arbitrators
will recognize an agreement to have general principles govern the merits or an
agreement to have a combination of sources as applicable law that bring together
rules both national and international, and public and private. The parties may, of
course, also give arbitrators the power to decide in amiable composition or ex aequo
et bono. In other words, the rules of law that will in the main govern the parties’
relationship may be determined by agreement. However, the pervasiveness of
consent as a basis of legal obligation can be misleading at least in two respects.
The first respect in which consent may mislead stems from its universal recogni-
tion as a basis for legal obligation. In most settings, this universal recognition relieves
us of the need to query and identify the legal basis for enforcing an agreement, be it
public or private, international or domestic. In arbitral practice, there are countless
instances of procedural and substantive decisions relying on decision rules based
directly on party agreement, where arbitrators do not feel compelled to identify the

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 92 fab ien g é linas

legal basis for enforcing the agreement. This is misleading because it conceals the
basis upon which the agreement, other than by virtue of its own terms, may be given
or refused legal force. It is universally recognized that not all agreements are valid
and enforceable, and that the legal bases for enforcing agreements also place condi-
tions and limits on pacta sunt servanda. These limits notably concern the conditions
relating to the formation of an agreement and the limits placed on party autonomy
by peremptory norms. Certainly, there is a significant extent to which an agreement
may be viewed as an instrument governed by its own rules. As seen earlier, parties
enjoy broad autonomy in the designation of the rules of law that govern their
agreement or dispute. However, there are rules that place conditions and limits on
the recognition and validity of agreements and those rules are not fully harmonized:
different rules may apply depending on factors that may not relate to party
agreement.
The second respect in which consent is misleading is in the choice of procedure.
In spite of a broadly recognized rule of party autonomy in respect of procedure, in
practice, parties rarely design their own arbitral procedure. To begin with, parties
frequently rely on standard-form arbitration agreements rather than negotiating
their own.43 Then, arbitration agreements most often refer to pre-established sets of
rules.44 The rules that end up applying in most cases tend to be formulated by
a variety of public and private institutional actors such as the ICC, the ICSID, the
United Nations for International Trade Law (UNCITRAL), and the PCA. Although
the sets of rules proffered by such institutions are usually applied by agreement
between the parties (they are often viewed as incorporated in the parties’ agreement
to arbitrate), they clearly reflect, and to an extent steer, broad normative under-
standings drawn from international practice. On a trajectory of convergence, these
understandings then form a kind of background law whose broad principles might
be invoked by default in the absence of agreement. It may be said that this back-
ground law is at least as important as the countless agreements embedded within it.
Therefore, if party autonomy does cut across the whole range of legal experience,
it does not follow that it can singlehandedly provide a basis for the unifying account
of international legal practice that international arbitration seems to demand. The
consent base for most of the rules followed and applied in international arbitration is
itself supported by background law that helps to identify enforceable agreements and
default substantive and procedural positions, failing a valid agreement.
43
This is demonstrated by empirical research, at least in the commercial context: P. Friedland and
L. Mistelis, 2010 International Arbitration Survey: Choices in International Arbitration (White & Case
and Queen Mary University, 2010), 10, available at http://www.arbitration.qmul.ac.uk/media/arbitra
tion/docs/2010_InternationalArbitrationSurveyReport.pdf (last accessed 30 September 2019);
T. H. Webster, ‘Party Control in International Arbitration’, Arbitration International, 19 (2003),
119, 120.
44
G. B. Born, International Commercial Arbitration, 2nd ed. (Kluwer, 2014), 2143–4. See also
E. A. O’Hara O’Connor and C. R. Drahozal, ‘Carve-Outs and Contractual Procedure’, Public
Law & Legal Theory Research Paper 13–29, Law & Economics Research Paper 13–16, Vanderbilt
University Law School, 14 June 2013, available at https://papers.ssrn.com/sol3/papers.cfm?abstrac
t_id=2279520 (last accessed 30 September 2019) (though parties have the ability to design their
own procedure, they rarely do so in practice).

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t he l e ga l f ra m ew o r k of a rb i tr at i o n 93

The background law that gives meaning and legal force to agreements on the
international plane is not as easily identifiable as one might wish. The contemporary
doctrinal focus on Article 38 of the Statute of the International Court of Justice (ICJ
Statute) in defining the sources of (public) international law certainly contributes to
this difficulty. I suggest that the focus on Article 38 is misleading in at least two ways.
First, there is no basis for saying that Article 38 was intended to isolate the whole of
international law as a legal system that only governs public law relations in the
narrow sense of interstate relations. The Statute concerns the International Court of
Justice, and the reality of legal practice is that interstate disputes, like other inter-
national disputes, are mostly resolved in venues other than the court.45 Those
venues, most significantly arbitral venues, hear parties irrespective of whether they
are public or private entities, and they apply rules defined by reference to ‘relevant’
practices and usages that can go beyond the confines of state practice, and by
reference to all relevant agreements, not only treaties. Second, the focus on Article
38 masks the unwritten norms of recognition that the text of the Statute was
intended to merely reflect (and only in the context of legal relations between states).
In other words, the focus encourages confusion between the rules of recognition
themselves – which lie outside treaties, in the form of custom or general principles –
and their narrow formulation at a particular point in time, for the particular purpose
of interstate relations. Obviously, rules of recognition, be they customs or general
principles, are subject to change depending on the evolution of practice and norma-
tive understandings.
While it is possible to accommodate Article 38 within a broader conception of
sources of law, the doctrinal focus on the Statute has entrenched a conception of
international law as a closed legal system whose only subjects are states.46 If we step
back from this narrow view and contemplate international legal practice as we find
it, we should discern that international law is better viewed as encompassing the
governance of the full range of relations that Phillip Jessup famously placed under
the umbrella of ‘transnational law’, namely:
all law which regulates actions or events that transcend national frontiers. Both
public and private international law are included, as are other rules which do not
wholly fit into such standard categories.47
This precisely captures the scope of the unified legal practice associated with
international arbitration in the first part of this chapter. The unified legal basis for
international arbitration, i.e., the background law introduced in the first part, should

45
Born, ‘A New Generation of International Adjudication’ (2012), 860.
46
For a scathingly critical view of the concept of ‘subject’, see R. Higgins, Problems and Process:
International Law and How We Use It (Clarendon Press, 1995), 45 (‘the whole notion of “subjects”
and “objects” has no credible reality, and, in my view, no functional purpose. We have erected an
intellectual prison of our own choosing and then declared it to be an unalterable constraint.’). For
her high-level account of the increasing role of non-state actors in the international system, see
also R. Higgins ‘International Law in a Changing International System’, Cambridge Law Journal,
68(1) (1999), 78, 84–6.
47
P. Jessup, Transnational Law (Yale University Press, 1956), 1 (footnote omitted).

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 94 f abi en g é linas

be identified as international law writ large, including, but not limited to, so-called
‘public’ international law.

4.3.2 Challenges and Paths to a Unified Representation

Commentators and actors are clearly working towards a more capacious representa-
tion of international legal practice, notably by uncovering existing legal roots to
explain and justify what may be termed the ‘new normal’. This ‘new normal’ is
a reality where, for example, private entities routinely sue states before international
tribunals for treaty violations and are able to enforce the resulting decisions through
national courts. James Crawford, for example, points out that ‘the “BIT revolution”,
reinforced in the United Kingdom and beyond by the Occidental case (which recog-
nised a further exception to the non-justiciability doctrine), has deeper roots than may
be thought’. He refers in particular to the Permanent Court of International Justice’s
1928 decision in the Jurisdiction of the Courts of Danzig case, and the Alabama claims
arbitration.48 In the Occidental case, the UK Court of Appeal decided that an arbitral
award rendered under a BIT that is not part of English law is reviewable by English
courts, with the implication that an award based on a foreign BIT is enforceable
through those courts.49 The UK Court of Appeal referred, in its reasons, to the 1928
decision where the Permanent Court of International Justice wrote:50
It may be readily admitted that, according to a well established principle of
international law, the [treaty], being an international agreement, cannot, as such,
create direct rights and obligations for private individuals. But it cannot be
disputed that the very object of an international agreement, according to the
intention of the contracting Parties, may be the adoption by the Parties of some
definite rules creating individual rights and obligations and enforceable by the
national courts.
The second sentence here may be said to rely on the principle of sovereignty in the
creation of treaty obligations.51 In this case, the principle was combined with
the principle expounded in the Alabama claims arbitration that a state cannot invoke
the deficiencies of its own law to excuse its breach of international rules, ultimately to
reach a conclusion of justiciability in the national courts.52 This provides an example

48
Crawford, ‘Continuity and Discontinuity in International Dispute Settlement’ (2010), 8–9;
Permanent Court of International Justice, Pecuniary Claims of Danzig Railway Officials who
have passed into the Polish Service v. Polish Railways Administration, Advisory Opinion,
3 March 1928, Publications of the Permanent Court of International Justice Series B – No. 15,
3; Arbitral Tribunal established under Art. I of the Treaty of Washington of 8 May 1871, Alabama
claims, United Nations, Reports of International Arbitral Awards, vol. XXIX, 125.
49
UK Court of Appeal (Civil Division), Ecuador v. Occidental Exploration & Production Co,
9 September 2005, [2005] EWCA Civ 1116.
50
UK Court of Appeal (Civil Division), Ecuador v. Occidental, [2005] EWCA Civ 1116, para. 19.
51
Permanent Court of International Justice, Case of the SS Wimbledon (UK, France, Italy, Japan and
Poland (intervening) v. Germany), judgment, 17 August 1923, Publications of the Permanent
Court of International Justice, Series A – No. 1, 15.
52
See Crawford, ‘Continuity and Discontinuity in International Dispute Settlement’ (2010), 8–9.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t he l e ga l f ra m ew o r k of a rb i tr at i o n 95

of an element in the existing legal corpus being uncovered to give current practice
a different appearance and direction, one in which it is recognized that ‘[s]tates
can . . . confer upon individuals . . . international rights strictu sensu, i.e., rights
which they can acquire without the intervention of municipal legislation and
which they can enforce in their own name before international tribunals . . .’53
However, there is still active resistance to the broad view outlined above. One
excellent example of such resistance in the arbitral context is afforded by the
reasoned decision on the challenge in the Chagos Archipelago arbitration between
Mauritius and the United Kingdom.54 In that case, where an arbitral tribunal of five
members was constituted under Annex VII of the Convention on the Law of the
Sea,55 an extensive debate took place concerning the standard applicable to the
challenge of an arbitrator. Mauritius argued that an arbitral standard of indepen-
dence and impartiality focused on a reasonable apprehension or appearance of bias
was generally applicable to international arbitration, whereas the United Kingdom
insisted that an interstate standard drawn from the practice of international bodies
focused on interstate disputes governed the challenge.
The tribunal, constituted as a four-member tribunal to hear the challenge, took
a hard and traditional line against the broad view advocated by Mauritius by settling
on Annex VII of the Convention as the applicable law, as ‘supplemented by the law
and practice of international courts and tribunals in interstate cases’.56 The tribunal
went on to rely on the narrow view of international law built up from the ICJ
Statute:57
The Tribunal recalls that the system of inter-State dispute settlement is based
upon the consent of the Parties, and more specifically upon the rules of public
international law, the sources of which are set out in Article 38(1) of the Statute
of the ICJ. In the Tribunal’s view, Mauritius has not demonstrated that the rules
adopted by non-governmental institutions such as the IBA have been expressly
adopted by States, nor do they form part of a general practice accepted as law, nor
fall within any other of the sources of international law enumerated in Article
38(1) of the Statute of the ICJ.
This refusal to engage with arbitral practice considered more broadly and at least
contemplate the idea that international arbitral tribunals may respond to, or con-
verge towards, a common standard of practice seems surprising. The tribunal flatly

53
UK Court of Appeal (Civil Division), Ecuador v. Occidental, [2005] EWCA Civ 1116, para. 19
(quoting R. Jennings and A. Watts, Oppenheim’s International Law, 9th ed. (Longmans, 1992)
para. 375).
54
Arbitral Tribunal established under Annex VII of the 1982 United Nations Convention on the
Law of the Sea, The Republic of Mauritius v. The United Kingdom of Great Britain and Northern
Ireland, Reasoned Decision on Challenge, 30 November 2011.
55
United Nations Convention on the Law of the Sea, 10 December 1982, entered into force
16 November 1994, 1833 UNTS 3.
56
Arbitral Tribunal established under Annex VII of the 1982 United Nations Convention on the
Law of the Sea, Mauritius v. United Kingdom, 30 November 2011, para. 165 (emphasis added).
57
Arbitral Tribunal established under Annex VII of the 1982 United Nations Convention on the
Law of the Sea, Mauritius v. United Kingdom, 30 November 2011, para. 167.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 96 f abie n g é linas

states that ‘[t]here is no reason . . . for considering challenges to arbitrators

appointed under Annex VII of the Convention . . . on grounds other than those
contained in the law and practice of international courts and tribunals concerned
only with inter-state cases.’ But is there a reason for applying a different standard to
international ‘tribunals concerned only with inter-state cases’?58 If there is such
a reason – a possibility that should not lightly be excluded – it is not one with which
the tribunal deigns to engage.59
Of course, a unified representation of the international arbitration framework can
accommodate some distinctions, such as the one drawn in the Chagos Archipelago
arbitration, thus allowing the application of different standards depending upon the
context, institutional or otherwise, in which the arbitrators are constituted as an
international tribunal. These distinctions do not represent significant challenges to
a unified conception as long as they are not tied to rigid conceptual positions
enforced without reference to the reasons why such positions should be held.
The issues that pose a more significant challenge, to which I now turn, concern the
role of the seat of arbitration (where there is a seat) and the choice of the law governing
the arbitration agreement. In the Occidental case, the UK Court of Appeal wrestled
with the question of the law that governs an arbitration agreement between a state and
an investor pursuant to an investment treaty and incorporating the UNCITRAL
Arbitration Rules. Looking at the English cases on the law governing arbitration
agreements and considering the connection between the investor-state arbitration
agreement at issue and the treaty supporting the scheme, the court was inclined to
settle on international law as the appropriate law to govern the arbitration agreement.
It was, in the court’s view, ‘hard to see why it should not be possible for a state and an
investor to enter into an agreement to arbitrate of the type contemplated by the
present Bilateral Investment Treaty subject to international law’.60 Although the court
was uncertain as to whether English law was quite ready for this move, and accord-
ingly offered the alternative basis of municipal laws (Ecuadorian or US law in this
case) to reach the same result,61 the court’s recognition of international law as
a possible legal basis for investor-state agreements to arbitrate signals a key point of
acceptance in the convergence of the public and private arbitration frameworks.
For a long time, commentary surrounding the New York Convention has debated
the question of which law should by default govern an arbitration agreement that
incorporates an international set of arbitration rules. One of the candidates that has

58
Arbitral Tribunal established under Annex VII of the 1982 United Nations Convention on the
Law of the Sea, Mauritius v. United Kingdom, 30 November 2011, para. 165.
59
See generally F. Gélinas, ‘The Independence of International Arbitrators and Judges: Tampered
With or Well-Tempered?’, New York International Law Review, 24 (2011), 1 (exploring the
sources of independence and impartiality standards for a range of international adjudicators, in
private and public settings).
60
UK Court of Appeal (Civil Division), Ecuador v. Occidental, [2005] EWCA Civ 1116, para. 34.
61
UK Court of Appeal (Civil Division), Ecuador v. Occidental, [2005] EWCA Civ 1116, para. 35:
However, if this is not possible and any such agreement must, under English private
international law, be subject to a municipal law, then, since the present agreement was
clearly intended to be binding, it must be subject to Ecuadorian or United States law.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t h e l e ga l f ra m ew o r k of a rb i tr at i o n 97

systematically been put forward is ‘international’ or ‘transnational’ law.62 This was in

response to a need to subject international arbitration agreements to international
rules, a need that, unsurprisingly, was felt most strongly in the context of arbitral
matters involving states. In that context, the shift in the international position is
strikingly illustrated by reference to the evolution that took place in the period
between two resolutions of the Institute of International Law, both dealing with
international arbitration: the Amsterdam Resolution of 195763 and the Santiago de
Compostela Resolution of 1989.64 The 1957 Resolution was part of an attempt to
localize all arbitrations by subjecting them to a national law determined by the seat of
the tribunal.65 The combined effect of the resolution’s provisions would have been to
exclude the possibility of a tribunal without a seat and subject the arbitration
agreement and the arbitral proceedings to the law of the seat. This attempt never
fully succeeded. The New York Convention, adopted the following year, is widely
seen, for example, as allowing (at least in some cases) the enforcement of an arbitral
award declared void under the law of the seat.66 The position of the 1957 Resolution
is difficult to reconcile with a unified representation of international arbitration’s
legal framework, and it has had an undeniable influence on the representation of
international arbitration over time. However, the Santiago de Compostela
Resolution of 1989, which deals with arbitrations involving a state or a state entity,
took a fundamentally different view, in that it essentially substitutes the law of the
seat as the legal basis of the arbitration for the provisions of the agreement to
arbitrate. It then subjects that agreement to arbitrate to the law most likely to
make it effective, including ‘public or private international law’.67 Thus, approaching
the issue from the perspective of mixed arbitration, there seems to be no

62
G. B. Born, International Arbitration: Law and Practice, 2nd ed. (Kluwer, 2015), para. 35. See also,
for example, Cour de Cassation Civ. 1e, France, Municipalité de Khoms El Mergeb v. Société
Dalico, 20 December 1993, Revue de l’Arbitrage (1994), 116, 117; US Court of Appeals, Third
Circuit, Rhone Mediterranee Compagnia Francese di Assicurazioni e Riassicurazioni v. Achille
Lauro, 6 July 1983, 712 F.2d 50; US District Court for the Southern District of California, Chloe
Z Fishing Co. v. Odyssey Re (London) Ltd, 26 April 2000, 109 F.Supp.2d 1236.
63
Institut de droit international, ‘Arbitration in Private International Law’, Session of Amsterdam –
1957, available at www.idi-iil.org/app/uploads/2017/06/1957_amst_03_en.pdf (last accessed
30 September 2019).
64
Institut de droit international, ‘Arbitration Between States, State Enterprises, or State Entities, and
Foreign Enterprises’, Session of Santiago de Compostela – 1989 (1989 Santiago de Compostela
Resolution), available at www.idi-iil.org/app/uploads/2017/06/1989_comp_01_en.pdf (last accessed
30 September 2019).
65
This goal can be inferred from the combined effect of the Resolution’s provisions.
66
See, for example, M. Thadikkaran, ‘Enforcement of Annulled Awards: What Is and What Ought to
Be?’, Journal of International Arbitration, 31(5) (2014), 575.
67
1989 Santiago de Compostela Resolution, Art. 4. Art. 4 of the Resolution states that in selecting the
law governing the validity of the agreement to arbitrate, ‘the tribunal shall be guided in every case
by the principle in favorem validitatis’. This formulation combines the principles of effective
interpretation and party autonomy in the choice of the governing law: a law that gives effect to the
agreement is to be applied in preference to one which would nullify the agreement. See also
K. P. Berger, ‘Re-Examining the Arbitration Agreement, Applicable Law Consensus or
Confusion?’, in A. J. van den Berg (ed.), International Arbitration 2006: Back to Basics?
(Kluwer, 2007), 312–4.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 98 f ab ien g é linas

insurmountable impediment to a unified framework with international law writ

large as its unifying basis.
Returning to Occidental, we see that the court was prepared to view the investor-
state arbitration agreement as one that can be governed by international law. This
could mean that any international arbitration agreement is liable to be governed by
international law writ large. This would normally place interstate and investor-state
arbitration agreements on the same footing, but the court in Occidental appears to
treat such agreements differently, at least for the purpose of its jurisdiction. Speaking
of the investor-state arbitration agreement, the court states:68
[We see] no good reason why any arbitration held pursuant to such an agree-
ment, or any supervisory role which the court of the place of arbitration may
have in relation to any such arbitration, should be categorized as being con-
cerned with ‘transactions between states’ so as to invoke the principle of non-
justiciability.
This is the main challenge to a unified framework. Under the New York Convention
scheme, reflected in the UNCITRAL arbitration instruments, there is an assumption,
though not strictly a requirement, that arbitral tribunals will have a seat, providing
the arbitral proceedings with a national lex arbitri. In most respects, the law of the
seat is subsidiary and can be displaced by an agreement of the parties, implicit or
explicit. In other respects, however, including the supervisory jurisdiction of local
courts and the imposition of some public policy requirements, the law of the seat is
broadly considered to be mandatory.
That interstate arbitration should be subject to the supervisory jurisdiction of the
local courts of the seat is not easily reconciled with the expectations of state parties.69
And yet, even the instruments intended for interstate70 and investor-state71 proceed-
ings tend to call for a seat or a place to be fixed. Understandably, there is resistance to
this in interstate cases. The Iran-US Claims Tribunal brought this difficulty to light,
with the jurisdiction of the domestic courts at the Hague and the relevance of Dutch

68
UK Court of Appeal (Civil Division), Ecuador v. Occidental, [2005] EWCA Civ 1116, para. 41.
This issue of the internationalization of the arbitration agreement raises for procedural law the
questions widely discussed in respect of substantive law in and around the award rendered by
René-Jean Dupuy in the Texaco-Calasiatic case. The award leaves room for the concession
agreement to be governed by ‘international law’ without necessarily being treated exactly as an
interstate agreement would be (Ad Hoc Tribunal, Texaco Overseas Petroleum Company v. The
Government of the Libyan Arab Republic, Award, 19 January 1977, International Legal Materials
(1978), 1).
69
For a thorough analysis of ‘delocalization’ in this context, see S. J. Toope, Mixed International
Arbitration (Grotius, 1990), 17–44.
70
See, for example, the Permanent Court of Arbitration’s Optional Rules for Arbitrating Disputes
Between Two States of 1992, Art. 16 (‘Place of Arbitration’). The ‘Notes to the Text’ of the Rules
state that they are based on the UNCITRAL Arbitration Rules, with modifications intended ‘to
reflect the public international law character of disputes between states, and diplomatic practice
appropriate to such disputes’. Another older example is the General Act for the Pacific Settlement
of International Disputes, adopted at Geneva on 26 September 1928 (Pact of Bogota), 93 LNTS
344, Art. XLIII (‘Seat of the Tribunal’).
71
ICSID Additional Facility Rules, Arbitration (Additional Facility) Rules, Art. 20.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 the l eg al framework of arbitration 99

law being questioned.72 The award in Republic of Ecuador v. United States of America
provides another example of this resistance: the award does not mention a place of
arbitration,73 in spite of the apparent requirement of the UNCITRAL Arbitration Rules
under which it was made.74 The Chagos Archipelago case (discussed above) provides
another example. In that case, an arbitral tribunal constituted under the Convention on
the Law of the Sea, hearing an interstate case at the Hague, viewed the proceedings as
having no legal seat. Not only does the reasoned challenge decision, in that case, make no
mention of a seat; the tribunal also expressly states that it ‘does not consider that there is
any basis under the Convention for the application of the Netherlands Arbitration Act or
the jurisdiction of the Dutch courts in these proceedings’.75
I suggest this resistance to the jurisdiction of the courts and application of the
arbitration law at the place of arbitration can and should be accommodated under
a unified representation. However, this alone should not prevent the enforceability
of the resulting award under the New York Convention where the required condi-
tions are otherwise met.76
First, the distinction between geographical place and legal seat should be
respected, maintained and cultivated, so as to prevent mere references to

72
See J. J. van Haersolte-van Hof and E. V. Koppe, ‘International Arbitration and the Lex Arbitri’,
Arbitration International, 34 (2015) 27. See also D. D. Caron et al., The UNCITRAL Arbitration
Rules – A Commentary (Oxford University Press, 2006), 38–44. Note that the tribunal handles
both mixed and interstate cases.
73
Arbitral Tribunal established under Art. VII of the Treaty between the United States of America
and the Republic of Ecuador Concerning the Encouragement and Reciprocal Protection of
Investment, 27 August 1993, and the UNCITRAL Arbitration Rules 1976, The Republic of
Ecuador v. The United States of America, PCA Case no. 2012–5, Award, 29 September 2012.
The case information provided by the PCA leaves a blank under the rubric ‘Seat of Arbitration (by
Country)’. See www.pca-cpa.org/en/cases/83/ (last accessed 30 September 2019).
74
UNCITRAL Arbitration Rules, 1976, adopted by the General Assembly on 15 December 1976,
Art. 18 (emphasis added):
1. Unless the parties have agreed upon the place where the arbitration is to be held, such place shall be
determined by the arbitral tribunal, having regard to the circumstances of the arbitration.
2. The arbitral tribunal may determine the locale of the arbitration within the country agreed by the
parties. It may hear witnesses and hold meetings for consultation among its members at any place it
deems appropriate, having regard to the circumstances of the arbitration.
...
4. The award shall be made at the place of arbitration.
75
Arbitral Tribunal established under Annex VII of the 1982 United Nations Convention on the
Law of the Sea, Mauritius v. United Kingdom, 30 November 2011, para. 170. In making this
statement, the tribunal makes no reference to the Rules of Procedure prepared for the specific
needs of the case dated 29 March 2012 and whose Art. 9 reads as follows:
Place of and Language of Arbitration
Article 9
1. The place where the arbitration is to be held shall be The Hague, the Netherlands.
2. The place of hearings shall be Dubai, United Arab Emirates.
3. The Award shall be made at the place of arbitration.
4. The language of the arbitration is English.
76
In all cases involving states, effective enforceability may depend on the extent of immunity
protection from jurisdiction in and execution.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 100 fab ie n gé l i n a s

a geographical place (for a registry, institutional seat, hearings or deliberations) from

having unintended legal consequences.77 Second, the first paragraph of Article 1 of
the New York Convention should be recalled, to highlight the fact that a seat is not
required for the Convention to apply:
This Convention shall apply to the recognition and enforcement of arbitral
awards made in the territory of a State other than the State where the recognition
and enforcement of such awards are sought, and arising out of differences
between persons, whether physical or legal. It shall also apply to arbitral awards
not considered as domestic awards in the State where their recognition and
enforcement are sought. (emphasis added)
Subject to the reciprocity and commerciality reservations under paragraph 3 of
Article 1, awards made pursuant to arbitration agreements governed by inter-
national law or international rules may appropriately fall in the category of awards
not considered domestic in the state where their recognition and enforcement are
sought.78 The courts of many jurisdictions have already accepted that international
awards need not be made under national procedural law and that such non-national
awards can be enforced under the New York Convention.79
Once it is accepted that an agreement to arbitrate found in a treaty or any other
international instrument can lead to an award, without a legal seat, within the
meaning of the New York Convention, I suggest that the main impediment to
a unified framework can be said to have been overcome. The outcome for each
case and category of cases then becomes a matter of establishing what the parties
77
A very good example of careful drafting is found in the procedural rules adopted as Procedural
Order no. 1 (6 December 2013) in the Timor-Leste v. Australia Arbitration (held under Para. (b) of
Annex B to Art. 23 of the Timor Sea Treaty between the Government of East Timor and the
Government of Australia of 20 May 2002). No mention is made of a ‘seat’ and, in Art. 11, under
the heading ‘Place of Hearings and Meetings’, one finds the following wording:
1. Unless the parties agree otherwise, the place where hearings will be held shall be The Hague, The
Netherlands.
2. The arbitral tribunal may hold meetings for consultation among its members at any place it deems
appropriate, having regard to the circumstances of the arbitration.
3. After inviting the views of the parties, the arbitral tribunal may meet at any place it deems
appropriate for the inspection of property or documents. The parties shall be given sufficient notice
to enable them to be present at such inspection.
The PCA acted as registry in that case. The case information leaves a blank under the ‘seat’ rubric,
and the Termination Order of 20 March 2017 makes no mention of a seat. The case information
and documents are available at www.pca-cpa.org/en/cases/37/ (last accessed 30 September 2019).
78
See UN, UNCITRAL Secretariat Guide on the Convention on the Recognition and Enforcement of
Foreign Arbitral Awards (New York, 1958) (United Nations, 2016), paras. 57–63.
79
See, for example, Netherlands Supreme Court, Société Européenne d’Etudes et d’Entreprises (S.E.E.
E.) v. Federal Republic of Yugoslavia, 7 November 1975, P. Sanders (ed.), Yearbook Commercial
Arbitration, vol. I (ICCA, 1976), 195; US Court of Appeals, Ninth Circuit, Ministry of Defense of
the Islamic Republic of Iran v. Gould Inc., Gould Marketing, Inc., Hoffman Export Corporation, and
Gould International, Inc., 23 October 1989, 887 F.2d 1357; Rouen Court of Appeal, France, Société
Européenne d’Etudes et d’Entreprises (S.E.E.E.) v. République Socialiste Fédérale de Yougoslavie,
13 November 1984, Case no. 982/82; Paris Court of Appeal, France, Société Aksa v. Société
Norsolor, 9 December 1980, Revue de l’Arbitrage (1981), 306.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 t h e l eg al fra mew or k of a rb itr at ion 101

intended when they agreed to arbitrate, notably by refining the default positions
through which legal practice relies on consent.80
From a cursory observation of practice, one might say, for example, that in
commercial matters, a designation of the place of arbitration is, by default, taken
to set the legal seat, bearing in mind that the parties enjoy opportunities in some
cases to exclude the main avenues of recourse against arbitral decisions even at the
legal seat.81
In the case of investor-state arbitration, although the same assumption has been
widely made, further thought seems called for as to whether the designation of a place
of arbitration should be taken as having the legal effects which the English Court of
Appeal found it to have in terms of both the applicability of the local arbitration law
and supervisory function of the courts. This would certainly depend on the specific
terms of the agreement between the parties, including the wording of the relevant
arbitration rules. However, as a matter of policy, considering the broadly agreed
position under the ICSID Convention where ‘[t]he laws of the physical place of
arbitration have no bearing whatsoever on the arbitration procedure’,82 it may well
be more appropriate to assume a preference, at least of state parties, for a completely
delocalized tribunal. The court in Occidental likely wanted to ensure that its jurisdic-
tion would not be denied for the purpose of enforcement under the New York
Convention and may not have realized that enforceability under the Convention
could be achieved without courts necessarily assuming supervisory jurisdiction
under local arbitration law. The better view, suggested tentatively here, may be that,
unless there are grounds to indicate otherwise, investment arbitration should nor-
mally be taken to be delocalized so as to deny any supervisory role to the law or courts
of the place of arbitration. The opposite assumption is widely made with respect to
proceedings conducted not only under the UNCITRAL Rules but also under the
ICSID Additional Facility Rules.83 However, the textual bases advanced for the

80
In the case of interstate arbitration, the consent of the host state (in whose territory the place of
arbitration is found) may come into play, as suggested by Art. 60 of the 1907 Hague Convention
(stating that a tribunal constituted under the convention may sit ‘in the territory of a third Power’
only ‘with the latter’s consent’). See on this point and more generally J. J. van Haersolte-van Hof
and E. V. Koppe, ‘International Arbitration and the Lex Arbitri’, Arbitration International, 31
(2015), 27, 44. Note that the Kingdom of the Netherlands may be taken to have consented to host
all cases administered by the Permanent Court of Arbitration: see the 30 March 1999 Host
Agreement concerning the Headquarters of the Permanent Court of Arbitration (Netherlands
Treaty Series 1999, 68).
81
In particular, in some countries, the parties can opt out of the action for annulment and/or setting
aside the award. This is possible, for example, in France (Code of Civil Procedure (consolidated
version of 29 January 2019), Art. 1522), Switzerland (Draft Bill for the Swiss Private International
Law Act 1982, Art. 192), Belgium (Judicial Code 1998 (as amended on 2 July 2018), Art. 1718), and
Sweden (Arbitration Act 1999, s. 51). See also A. J. van den Berg, ‘Should the Setting Aside of the
Arbitral Award Be Abolished?’, ICSID Review, 29(2) (2014), 263, 276–7.
82
McLachlan, International Investment Arbitration (2017), 65, para. 3.71. Note that the ICSID
Convention refers in ch. VII to the ‘place of proceedings’.
83
See notably F. Nitschke and K. Aït-El-Hadj, ‘Determining the Place of Arbitration in ICSID
Additional Facility Proceedings’, ICSID Review, 30(1) (2015), 243, 244–5.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 102 f a bie n gé linas

assumption, namely, that Additional Facility awards are considered to be ‘made at the
place of arbitration’84 or that the Additional Facility Rules apply ‘save that if any of
these rules is in conflict with a provision of the law applicable to the arbitration from
which the parties cannot derogate, that provision shall prevail’,85 are not entirely
determinative. As seen above, a ‘place’ of arbitration does not necessarily refer to
a legal seat giving local courts supervisory powers under the local law, and ‘the law
applicable to the arbitration’ need not be the law of the seat.
This leaves us with interstate cases. In the current state of affairs, one might say
that the parties should at least be presumed not to intend an interstate tribunal to
have a legal seat that would unnecessarily subject the procedure and the award to
national laws and courts.86 This seems relatively straightforward. The outstanding
question for interstate arbitration, as viewed from the perspective of a unified
framework, is whether an award made by an interstate arbitral tribunal could ever
be enforced under the New York Convention. Under a unified framework, we may
reach the point where the answer would be ‘yes, provided this was the intention of
the parties to the treaty or other agreement’.
It should be recalled that the difficulty of the questions relating to the arbitral seat
is compounded by the fact that each seat has its own public policy and mandatory
rules. Where a tribunal has a legal seat, one of the consequences is that the procedure
and the award are considered to be subject to the mandatory rules of that seat, no
matter how weak the seat’s connection to the arbitration may be. This is certainly
a reason to view delocalized arbitration favourably where a state is one of the parties.
However, in the jurisdiction where an award – even a delocalized award – is sought
to be enforced, the enforcing state has every reason to wish to ensure, through its
courts, that the award is consistent with its public policy. Thus, a procedurally
delocalized arbitration does not fully eliminate the possibility of judicial review; it
shifts the control function from the state in which the award is ‘made’ to the state or
states in which it is sought to be enforced.87 As Stephen Toope puts it (in the context
of mixed arbitration), ‘[s]uch a shift is both practical and sensible’.88 This is the
compromise reached where international decisions are intended to be enforceable
through a national judicial process.89

84
ICSID Additional Facility Rules, Arbitration (Additional Facility) Rules, Art. 20(3).
85
ICSID Additional Facility Rules, Arbitration (Additional Facility) Rules, Art. 1.
86
For a careful argument to the effect that consent must be explicit for an interstate case to be
subjected to the courts and arbitration laws of the place of arbitration, see J.J. van Haersolte-van
Hof and E.V. Koppe, ‘International Arbitration and the Lex Arbitri’ (2015), 34–42. The authors
also usefully review the materials concerning the use, actual and potential, of the ICJ and its
predecessor in the exercise of the award review function, see 35–42.
87
See J. Paulsson, ‘Arbitration Unbound: Award Detached from the Law of Its Country of Origin’,
International and Comparative Law Quarterly, 30 (1981), 358, 375.
88
Toope, Mixed International Arbitration (1990), 39.
89
The sufficiency of judicial review of awards at the place of enforcement was famously argued by
Phillippe Fouchard at the ICCA Congress of 1998: ‘Suggestions to Improve the International
Efficacy of Arbitral Awards’, in A. J. van den Berg (ed.), Improving the Efficiency of Arbitration
Agreements and Awards: 40 Years of Application of the New York Convention (Kluwer, 1999),
601–16.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press

 the l egal framewo rk o f arbitration 103

4.4 Conclusion
Although a unified legal framework for international arbitration has not yet been
achieved, it appears to be within reach. With the development of investor-state
arbitration over the last fifty years, the continued separation of arbitral practice into
private and public realms seems increasingly implausible. We have seen that
a unified framework will have been achieved once it is fully and clearly recognized
that an arbitration agreement may be subject to international law or international
rules and that an international arbitral tribunal does not require a domestic legal
seat.90
One may speculate whether states will accept the position that international
arbitral awards can be enforced under the New York Convention even in inter-
state settings. After all, states have already consented to treat ICSID awards made
against them as enforceable judgments.91 States have also consented to treat the
decisions of the specialized Seabed Disputes Chamber of the International Tribunal
for the Law of the Sea (an interstate forum) in the same way as enforceable national
court judgments,92 thereby giving momentum to what Gary Born has called the
‘second generation of international tribunals’.93 Of course in both cases, as the law
stands in most jurisdictions, states can invoke immunity from execution, at least in
some respects, where they decide that this is expedient in the exercise of their
sovereign powers.94 In addition, under the New York Convention regime, enforce-
ment would remain liable to resistance in accordance with broader public policy
considerations recognized by the enforcement court.95
Whatever the future may hold, I hope to have shown that the unified representa-
tion of international legal practice surrounding arbitration tentatively outlined in
this chapter, while certainly not the path of least resistance, is a path worth pursuing.

90
For the arguments against the recognition of delocalized awards under the New York Convention,
see A. J. van den Berg, ‘When Is an Arbitral Award Nondomestic Under the New York
Convention of 1958?’, Pace Law Review, 6 (1985), 25, 62–5 (‘It is true that the text of the
Convention, as far as its field of application is concerned, does not require that the award be
governed by a national arbitration law. However, if the Convention’s scope is read in conjunction
with the Convention’s other provisions, it becomes evident that this requirement is implied.’).
91
ICSID Convention, Art. 54.
92
Statute of the International Tribunal for the Law of the Sea (Annex VI of the United Nations
Convention on the Law of the Sea), 10 December 1982, entered into force on 16 November 1994,
1833 UNTS 561, Art. 39.
93
Born, ‘A New Generation of International Adjudication’ (2012).
94
ICSID Convention, Art. 55. See A. K. Bjorklund, ‘State Immunity and the Enforcement of
Investor-State Arbitral Awards’, in C. Binder et al. (eds.), International Investment Law for the
21st Century: Essays in Honour of Christoph Schreuer (Oxford University Press, 2009); see also M.
Brenninkmeijer and F. Gélinas, “The Problem of Execution Immunity and the ICSID Convention”
Journal of World Investment and Trade, 22 (2021) 429. .
95
In some jurisdictions, immunity from ‘enforcement jurisdiction’ (as distinguished from ‘execu-
tion’) may also be recognized where the award is not already deemed a judgment.

https://doi.org/10.1017/9781108304467.005 Published online by Cambridge University Press