Law, International Public Policy (So-called) and Arbitral Choice in International Commercial

Arbitration

W. Michael Reisman (*)

Public policy – it is an unruly horse and when once you get astride it, you never know where it will
carry you. It may lead you from the sound law. It is never argued at all but when other points fail.

Burroughs J. in Richardson v. Mellish (1)

I. Limits to Arbitral Choice

Policy is a central and inescapable part of the enterprise of law. Policies are authoritative political
objectives; in some instances, they also include the authoritative legal means for their achievement.
All legislation as well as all the rules and principles of a body politic which have not originated in
legislation may be seen as instruments for achieving that body politic's policies.

Two recurring and quite distinct questions in legal theory are, first, the extent to which those charged
with applying legislation may themselves make policy and, second, the extent, if any, to which they
may depart, in their applications, from the explicit directions of legislation or rules in an effort to
implement the policies which, the applier believes, actually animate the legislation and rules.

The second question – which presents itself frequently and seems to be the more complex – actually
admits of a simple answer: do what the law which governs the matter prescribes. In some
circumstances, legal prescriptions are designed to allow relatively little or even no contextual
adjustment by the person designated to apply them in particular cases. To take an extreme example,
the sergeant sitting in a silo housing ICBMs is given the strictest instructions about the circumstances
under which he is to flip a switch and send the missiles in his charge hurtling toward their target.
Those who have established the rules (and indeed all the rest of us) do not want that soldier to do
any contextual appraisal, even less to do a Derrida-style deconstruction of his commands. The
sergeant is expected to confine himself to a drastically limited normative universe in which one and
only one relevant rule contains a single binary choice which is to be taken on the occurrence of one
and only one specific contingency.

In other circumstances, the legislature frames its prescriptions in a more open-textured fashion. Then
the implication is clear: the persons assigned to apply them are authorized and required to make
certain adjustments in their application so as to ensure that the policies animating the prescriptions
are given effect in that particular situation. In these circumstances, which are the more common, the
applier, whether judge or arbitrator, must view the prescriptions which he or she is called upon to
apply not as black-letter rules but as authoritative communications conveying information about (i)
the policy or policies to be achieved as well as (ii) an indication of the range of circumstances in
which those policies are to be applied. Where a rule formulation is the vehicle chosen for the
achievement of a policy and it proves ill-adapted to do so in a particular case, whether because of the
anachronism of the rule in the face of subsequent social changes or the idiosyncratic character of the
fact situation, the appliers, whoever they may be, must consult and apply the policy. They must, as
our French colleagues put it, engage in actualisation.
A conception of application which tried to model itself only on the sergeant in the silo and to insist
exclusively upon strict application of the rules is likely, in these circumstances, to produce
anachronistic decisions at odds with the contemporary objectives of the legislature. Sensitivity to
policy is also important in complex cases: Because real life controversies rarely engage a single law,
rule or, indeed, a single policy, the judge or arbitrator may have to make extensive adjustments in
the accommodation of a policy in order to shape the package of remedies to be awarded so that they
approximate all the policy objectives of the legal system which are engaged in that case.

In these respects, the international commercial arbitrator, like all other appliers, is obliged to be
sensitive to policy, but it must be emphasized that it is the policy of that legal system – its authorized
political objectives and the means for their achievement – which the parties have selected or to
which the arbitrators are directed by principles of conflict of laws. This use of policy is a highly
disciplined teleological exercise. It bears repeating that it is the policy of the legal system which
governs the dispute and not an artifacted policy created by the arbitrators for that particular case.

It is important to emphasize that none of these tasks involves making new policy, i.e., creating new
political goals and new instruments for achieving those goals. Creating new political goals and new
instruments for achieving those goals is a province assigned to others in the community whose law is
being applied. To be sure, some civil codes explicitly authorize and require a judge confronted by a
lacuna to act as if he or she were the legislator rather than return a judgment of non liquet. An
arbitrator applying that choice of law of the parties would similarly have a derivative competence to
“legislate” pro hac vice but without effects beyond that case. Some legal systems – one thinks of the
United States in certain historical periods – are more generally comfortable with the vocation of
judge-made law or judicial legislation. There, the legislature is the designated law-maker but judges
are deemed to have a contingent law-making role. Whatever one's view of that particular
conception, it seems to me not to be part of the province and vocation of the international
commercial arbitrator, other than in those circumstances in which the applicable law explicitly
authorizes an applier to act as the legislator where there is a real lacuna. The reason for this is hardly
recondite. It lies in the essential role of the international commercial arbitrator.

Arbitrators execute the mandate of the parties, whose common consent is the source of their
jurisdiction. If parties expressly authorize arbitrators to decide ex aequo et bono, they may depart
from law otherwise applicable and the terms of otherwise applicable contracts and decide according
to their individual sense of justice. National courts, which play an indispensable role in international
commercial arbitration, should have no hesitation in lending the powers of the state to the
enforcement of such an ex aequo et bono award. After all, the parties had effectively authorized the
arbitrators to rewrite their contract so the action of doing so is within the boundaries of their
contractual consent. And of course, the state has already decided that if the parties consent to waive
their right to adjudication and elect instead to privatize the resolution of their dispute, the state
apparatus will implement that agreement.

The situation is entirely different when there has been no such authorization by the parties. In that
situation, the arbitrator purporting to decide ex aequo et bono or an arbitral system that purports to
endorse such excursions may still expect the apparatus of the state to ignore both its own
commercial and constitutional policies as well as the fact that its own citizen has not agreed to waive
them (insofar as they are jus dispositivum) yet still to implement the idiosyncratic vision of the
arbitrators. This is, I submit, an unreasonable expectation which is likely to remain unrequited. And
properly so. After all, national jurisdictions exist to protect the legitimate interests of the community
concerned, which is the raison d'être both of the state and of the international law of jurisdiction. It is
entirely reasonable to ask governments to yield judicial jurisdiction (i) when the parties manifest
national diversity; (ii) when they wish, for licit reasons, to privatize the mode of resolution of their
dispute and clearly indicate their wish; and (iii) when those private proceedings, including the
application by the arbitrators of the parties' choice of law, are fair. But it is unreasonable to ask
governments to surrender the application of their law and policy to matters within their jurisdiction
just because arbitrators, without being so authorized by the parties, assign themselves a general
legislative power or, more modestly, a norm-making power on a case by case basis.

My point is not that international commercial arbitrators should behave like the sergeant in the silo.
Far from it. Every lawyer quickly learns that, though the law is beautiful in its complexity, reality is
often more complex, while particular legislative or general law arrangements are insufficiently so. By
its very nature, every application of the policies expressed in an open-textured rule to new situations
supplements existing law. Particularly in modern industries, like biotechnology or information
technology, in which national legislation lags behind technical or scientific developments, the most
scrupulous application of the parties' choice of a national law may still require the identification of
the relevant policies of that state and their extrapolation to new situations. In this sort of situation,
supplementing law infra and praeter legem is sometimes required and proper; fashioning
prescriptions contra legem is not. Wholly aside from the jurisdictional limit, aggressive law-making by
international commercial arbitrators is ill-advised, because of the lack of aptitude of the international
commercial arbitral process for law-making. I turn to that.

Modern law-making incorporates the efforts of many participants or stake-holders, all of whom bring
both information about preferred policy and their respective views of the consequences of different
arrangements on the aggregate goals of a community. This openness of the process of law-making is
a critical part of the legitimacy of the page "851"process. Now obviously, the many stake-holders
have their own preferred outcomes, which will influence the information they gather and they will
present it in a way most favorable to themselves. But the openness of the process and the wide
range of participation ensure that a rich trove of information and a wide diversity of proposals will be
available to the law-maker. Thus, the modern legislature, thanks to the structured openness of the
law-making process, benefits from a wide range of viewpoints and rich and detailed projections of
the probable consequences of the different alternatives available to it. In addition, the modern
legislature, thanks to this open process, develops an acute sense of the relative power positions of
different stakeholders, a factor which will be critical in securing economic implementation. This also
alerts the law-maker to the intensity of the resistance which certain possible prescriptions are likely
to encounter and, hence, the quantum of political capital which will have to be diverted for
enforcement. Contrast this polycentric and information-rich process with three arbitrators, deciding
a bilateral international commercial dispute in a process which is closed to the participation of all
outsiders and could well be unknown to them. This arbitral process benefits from none of the
advantages enjoyed by the modern law-maker. Hence arbitral efforts to play the role of Solon are
likely to produce bad law, unenforceable law or both. (2)
II. A Few Concerns

Contracts presuppose a legal system. Ordinarily, one would expect those engaged in the competitive
collaborative process of designing international commercial agreements between parties of different
jurisdictions to select a system of national law which can recommend itself for its statutory precision,
rich and relevant judicial illumination and page "852"comparative research accessibility. So it can
occasion no surprise, for example, that the law of New York or Delaware will be selected by American
lawyers for contracts about corporate transactions or the law of Texas or Oklahoma will be selected
for “farming-in” options in oil exploration contracts. The point is that national legal systems will be
chosen.

There have been some curious exceptions to this common-sense practice. In an interesting sub-
chapter in the history of international commercial arbitration, petroleum concessions concluded
through the middle of the twentieth century – a period in which the jurisdictional theory of
territoriality still prevailed – were subject to the law of the host state. Petroleum companies and,
indeed, even arbitrators called upon to decide disputes arising under them chafed under this regime.
The governing law was sometimes evaded or sometimes simply ignored by the arbitrators, as in Lord
Asquith's astonishing award in the Abu Dhabi case. Governing law for these types of agreements was
eventually contractually wrested from the host states by applicable law formulae on the order of
“the law of the host state and international law and in case of a conflict between them general
principles of law”. This assured the application of international law or of general principles of law, a
part of international law. But these were investment agreements with a state party and not
commercial agreements between two private parties. In an investment, context, the imprecision of
this sort of formula was apparently deemed preferable to the application of Libyan or Saudi law. The
trend was finally confirmed by Maître Dupuy in CALTEX v. Libya.

On rare occasions, one will still find choices of international law as applicable or governing law in
international commercial contracts or, indeed, governing law provisions that are even more
idiosyncratic, like the inimitable clause in the Aminoil v. Kuwait arbitration:

The law governing the substantive issues between the Parties shall be determined by the Tribunal,
having regard to the quality of the Parties, the transnational character of their relations and the
principles of law and practice prevailing in the modern world.

Clearly, in cases in which the parties have selected international law, the same type of policy analysis
and application I have proposed earlier would be applied, but now with respect to the animating
policies of the pertinent parts of the corpus of international law. In addition, where a system of
national law which has been selected by the parties incorporates international law, either as a
supplement or a corrective to its own law, the same sorts of policy analyses with respect to
international law would apply.

But when international law has not been adopted as governing law by the parties to an international
commercial transaction nor directly incorporated and self-executing in the system of national law
which was selected, would it not be inappropriate to allow a norm of international law to override
the applicable norm of the national law selected by the parties on the ground that the international
norm is “different” and “higher”? The issue does not turn on grand theories of monism or dualism
but on common sense. That the norms are different is obvious. The predicate of the selection of
governing law – the whole idea of bothering to make a selection – is that different legal systems
address particular legal and factual issues differently. And, as for the relative “spatial” positions page
"853"of different systems of law, when parties have the power to select the law which will govern
their transaction, whether the law which they select is “higher” or “lower” is irrelevant. Imagine an
international commercial arbitration tribunal deciding to apply to the construction of the commercial
contract in dispute Art. 31 of the Vienna Convention on the Law of Treaties rather than the
interpretative regime of the national governing law which the parties had selected on the ground
that international law is “higher”. Would not that be bizarre, if not a ground for annulment?

But is there some “transnational (or truly international) public policy”, to borrow the title of Maître
Lalive's famous report to ICCA in 1986, which can trump the national law that would otherwise
govern the transaction and which should be applied by an international commercial arbitration
tribunal? I am considerably more cautious than Maître Lalive about the space for – and propriety of –
application of so-called international or transnational public policy in international commercial
arbitration, though I acknowledge that it has on occasion been done. My own reservation is not
simply Lord Burrough's acerbic observation that “public policy is never argued at all but when other
points fail.” My opposition to the use of international public policy in international commercial
arbitration arises from systemic features of international commercial arbitration and of international
law itself. Let me set out a few of my concerns:

First, international public policy is a term of almost unlimited and protean potential. In a decision this
year by the Swiss Federal Tribunal, albeit on the use of public policy in the context of Art. V of the
New York Convention and Art. 190(2) of the Swiss Law on Private International Law, the Court
observed:

The fleeting character of public policy may be inherent to the concept, due to its excessive generality;
the wide scope of the almost countless opinions proffered in this regard would tend to prove it…. As
a commentator pointed out, all attempts to answer the numerous recurring questions raised by the
interpretation of this concept merely resulted in raising further thorny or polemical question…. (3)

If that is the situation in Swiss law, imagine a fortiori what it is in international law! International law
is rife with precatory statements. Moreover, not all of them were made with the intention of being
effective; indeed, some were made with the intention of not being effective. International law abets
this by maintaining categories of legal statements that are intended not to be binding or are subject
to derogation on conditions to be determined by the derogating state itself. So-called “soft law” has
become an industry in its own right. International public policy is indeed fleeting and the
authorization of its application by international commercial arbitrators would lead to great
uncertainty.

Second, even where international law has established a firm and not soft policy, it will often carefully
exclude national and even specific international agencies from enforcing it. Recall George W. Bush's
not implausible presumption that when important Chapter VII Security Council Resolutions
concerning the maintenance of international security have been repeatedly ignored by the state to
which they were directed, they may then be coercively enforced by self-selected willing states or
coalitions of such states without additional UN authorization. As we all know, his presumption was
resoundingly rejected. And it was no surprise to students of international law. Or consider the issue
of slavery, apparently a favorite of those who urge transnational public policy to trump national law
in arbitrations. Slavery is a practice which still persists and on the prohibition of which the
international legal system has not distinguished itself. One of the reasons why is clearly explained in
the famous nineteenth century Le Louis judgment. There Lord Stowell (then Sir William Scott)
affirmed that British ships could not enforce a slavery prohibition against flag vessels of other
nations, on the principle that no matter how reprehensible the practice, it was more important to
maintain the freedom of the oceans which depended on the principle that no state could purport to
exercise jurisdiction on the vessels of any other state. So the slave ship and its human cargo was
permitted to go its way. In 1982, the United Nations Law of the Sea Convention in Art. 99 prescribed,
in relevant part, that

Every State shall take effective measures to prevent and punish the transport of slaves in ships
authorized to fly its flag and to prevent the unlawful use of its flag for that purpose.

Plus ça change, plus c'est la même chose. Yet using transnational public policy to hoist his bootstraps,
the arbitrator who is so inclined may take the softest international law and simply present it as
transnational public policy and thereby make it directly effective without regard to the way it would
be handled in international law itself or the reasons why it would be handled that way.

Third, and related to the previous observation, is the extraordinary caution of international law about
disturbing agreements, because they constitute one of its most important and most fragile lines of
defense, itself built on the doctrine of pacta sunt servanda. While the Vienna Convention on the Law
of Treaties nonetheless created a category of so-called peremptory norms or jus cogens, which have
the effect of annulling any treaties that are inconsistent with them, the only jus cogens upon which
the drafters of the Vienna Convention could agree was the prohibition of the use of force in Art. 2(4)
of the UN Charter. Though the human rights advocacy bar and its literature are constantly
discovering new jus cogens, to date only one international tribunal has applied the Vienna
Convention's provision and, at that, in a way which seems quite bizarre. If international law is so
careful about using a potentially disruptive device lest it disturb the effectiveness of international
agreements, it would seem curious that international commercial arbitrators would invoke the
“public policy” of the same international law to disturb the stability of expectation and effectiveness
of international commercial agreements.

Fourth, public policy in domestic law is a legal concept with a verifiable judicial history. Not so in
international law. When an alleged international public policy is given page "855"the force of law by
an international commercial arbitral tribunal, it is not subjected to the discipline of customary
international law analysis which requires that the state practice justifying the inference that a
customary rule has been produced must be shown to be wide and extensive and to be accompanied
by opinio juris. Without this discipline, the invocation of “transnational public policy” becomes an
easy way for those claiming to have an insight into the heart and the soul of international law to
effect their own preferences without having to prove that they have become customary international
law.

Fifth, does international commercial arbitration really need such a slippery and malleable concept in
order to protect its virtue? After all, what practice before an international commercial arbitration
tribunal that has been alleged to violate an international or transnational public policy was permitted
by the national governing law? Is there a national legal system that does not prohibit bribery of
public officials? A national legal system that does not prohibit slavery….
If international public policies are treaty-based and their application extends to parties before an
international commercial arbitral tribunal, then they can be applied as part of the governing law. If
the policies have been confirmed by customary international law, then their application may be
appropriate insofar as the governing national law incorporates customary international law. But if
they are simply invoked as international public policy and that is their only basis in authority, I doubt
that they have a proper role in international commercial arbitration.

And, finally, a personal plea. As a public international lawyer who believes that international law is a
real and important system of law, I object to a concept whose notorious imprecision and subjectivity
gives international law a bad name. The parties to an international commercial arbitration and the
system of international law itself deserve better.