Arbitral tribunals, like State courts, apply legal rules. How the applicable law is identified is a classical topic, as well as the question of whether the applicable law is necessarily that of a State, or may consist in an anational, or transnational law, such as the lex mercatoria. Most often the applicable law is chosen by the parties in the framework of the contract that binds them, and almost always the law chosen by the parties is that of a State. The arbitrators are expected to abide by the choice made by the parties. In the absence of such choice, there are various methods by which the arbitrators are able to identify the applicable law. All this is well known, and is outside the scope of the present book.

However the difficulties in finding the applicable legal rule do not end with the identification of the applicable law. Firstly, since the applicable law is very often not the national law of all three arbitrators, its contents must be proven. By what means it is proven was the subject dealt with by Phillip Capper (see p.31). It also raises the issue, discussed by Antonias Dimolitsa (see p. 22), whether and to what extent arbitrators may introduce new issues ex officio. Secondly, it is often difficult to construe the provisions of the applicable law, which raises two issues. Are arbitrators forced to follow the case law established by the courts of the country the law of which is applicable? Klaus-Peter Berger illustrated the debate by describing the German experience regarding the Law on Standard Terms (AGB-Law) (see p. 80). On the other hand, should arbitrators take into account arbitral jurisprudence? That issue was addressed by William W. Park (see p. 60). Thirdly, Horacio Grigera Naόn discussed the difficult question of whether arbitrators can deal with hierarchical conflicts of law (for instance, between a law and the constitution): see p. 97.

However the whole conference, the proceedings of which are compiled in the present book, in fact focused on a subject that is of great importance, but is rarely discussed: the degree of liberty enjoyed by arbitrators to set aside a particular rule of the applicable law (even if that law had been chosen by the parties themselves), for the reason that according to them said rule does not correspond to the expectations of the parties, or to the needs of international commerce. That was the object of François Perret’s contribution (see p. 109), and of a table ronde which I chaired, and in which Matthieu de Boisséson, Antonio Crivellaro, Tarek Riad and Mohanadass Kanagasabai participated (see p. 116, p. 128, p. 143). In fact the bases for the discussion having been introduced by Bernard Audit at the outset of the conference (see p. 10), most speakers dealt with it more or less abundantly, and there were heated discussions involving other participants. The general view of the speakers was in favour of the arbitrators’ freedom to make the expectations of the parties, or the needs of international commerce, prevail, but from the audience, several voices were heard, saying that if the parties have agreed that a certain law should be applied, that covers all its relevant rules, whether or not the parties were aware of the contents of each of them — which is rarely the case — , and that if one party, for instance, argues that a given clause in the contract is null and void, the relevant rule is to be found within the applicable law and nowhere else.

The problem is extremely complex. I seize this opportunity to mention some elements of the discussion, taking as an example one provision of the Swiss Code of Obligations (CO) which François Perret used in his contribution: Article 404, CO.

Pursuant to said provision, an agency relationship may be revoked or repudiated at any time, this possibility being considered by the case law of the Federal Court as not subject to derogation. What should one decide if the contract, which the parties have agreed is subject to Swiss law, states that it is made for a fixed term? At least two awards have set aside the application of Article 404 CO in such circumstances, with varying arguments.

The obligation to apply the chosen law — and thus, normally, every relevant rule of the law invoked by either party — is a constraint, but it is a weak one since it is not obvious that an award will be less convincing and less effective if the arbitrator sets aside Article 404 CO. It will not be less effective if there is no control by a state court, and it is difficult to say what is the more convincing in this situation, as the contractual clause reflects the will of the parties but the chosen law annuls it. It is therefore necessary to set two objectives: to meet the legitimate expectations of the parties on the one hand, and on the other hand, to not seem to judge arbitrarily. But what results from their collation is not so clear.

On the one hand, and prompting an arbitrator to take liberty, it is noted that the choice of Swiss law in a fixed-term agency contract can only result from the parties’ ignorance of this very special mandatory rule, rarely existent in comparative law, and criticized by some Swiss authors. Between, on the one hand, a will relating to a very general subject, that of choosing a set of rules that one cannot imagine each of which was known, and on the other hand, a will expressed in a specific clause, there would be no cause for hesitation. The legitimate expectations of the parties would be those based on the clause.

On the other hand, one would first say that the normal, banal and predictable reconciliation between the law and the contract is that the validity of the second - and that of its clauses - is dependent on the first. From there it is dangerous to posit that an arbitrator may set aside a rule of the chosen law because he finds it inappropriate to the specific situation: there is a risk of arbitrariness here. Finally, the expectations of the parties should not be considered only at the time of conclusion of the contract, but also on the eve of the proceedings, when all elements are known: the parties need to know what risk they face of losing any eventual proceedings, which would possibly allow for the avoidance of such proceedings. But in this regard the hope that the rule would be thrown out by the arbitrator is much more uncertain than that which would be based on the straightforward application of the law, which is a simpler and therefore more predictable guideline.

It may be expected that the majority of arbitrators would prove more inclined to consider respecting the legitimate expectations of the parties, and would easily set aside a rule cancelling a clause freely agreed between them. In doing so, they would introduce a seed of insecurity in arbitrations to come, but that is not their primary concern. An arbitrator is the judge of the situation before him, his goal is not to participate in the development of a system favouring general predictability.

The editors hope that the arbitration community will find interest in reading the present Dossier XI. Hope for that can be found in the fact that the conference itself had met with considerable success.