In 2003, the ICC Institute of World Business Law held its annual meeting on oral evidence in arbitration. 1 The present conference, which focuses on the taking of written evidence, naturally complements our work of five years ago.

The importance of the issues discussed today cannot be overstated for the future of international arbitration. As Julian Lew shows in his paper, in contrast to oral evidence, written evidence is generally held to be reliable and admissible. Awards will be based, almost entirely if not exclusively, on documentary evidence. The parties need to prove their case, and they need documents to do so. If they cannot obtain the documents they need, they have no case at all: idem est non esse aut non probari. It is therefore understandable that access to written evidence is one of the main concerns of the parties to a dispute. Is there a right of the parties to obtain disclosure of documents in possession of other parties? What is the proper balance between fairness and the need to resolve disputes expeditiously? Is the current practice of document disclosure satisfactory? Does it need to be changed? Should the parties, for example, be under a duty to disclose all documents in their possession, even if some documents are adverse to their case? And what is the proper balance between the arbitral tribunal's jurisdiction and the role of state courts in obtaining documents? Should court assistance in aid of the arbitration be subject to the consent of the arbitrators? Should parties be allowed to obtain in court discovery orders that are not permitted before the arbitral tribunal? Finally, is the current practice suited for the taking of electronic evidence? All these questions are hardly new, yet they need to be revisited over and over as they are at the heart of the parties' expectation of a fair, neutral and efficient way of solving their international disputes. [Page448:]

Most document disclosure in international arbitration occurs under the control of the arbitral tribunal and between the parties. It is widely accepted that the arbitral tribunal has discretion in assessing the admissibility and evidentiary value of the documents produced in the arbitration. Questions of admissibility may arise, in particular, in the presence of an allegation of fraud. As explained by Markus Wirth, forged documents may be produced. The arbitral tribunal may also have to decide issues of admissibility of documents fraudulently obtained. Other issues may also have to be resolved. For example, in the Libananco case, an ICSID tribunal held that admitting documents obtained by the state through the use of its police and intelligence services would be in breach of the principle of party equality.

Most of the difficulties arise when the parties do not spontaneously disclose documents and the arbitral tribunal is requested to order their production. Bernard Hanotiau makes the point that the parties should be directed to resolve these issues between themselves, without involving the arbitral tribunal. Nonetheless, parties happen to sometimes disagree, and the tribunal's intervention is then required.

The process of document disclosure is governed by the procedural law of the arbitration and the parties' agreement. However, the papers in this collection show that arbitral practice in the field has to a large extent been transnationalized, even though document production is at times still perceived differently from the civil law and common law perspectives. The emergence of a truly international arbitral practice is a great achievement, which is to a large extent attributable to the IBA Rules on the Taking of Evidence. Whether and to what extent the IBA Rules should be amended to take into account developments since 1999 is a question that is currently being debated in the IBA Rules of Evidence Subcommittee, whose ongoing work is discussed by Amy Cohen. The subcommittee will present an interim report at the upcoming IBA Congress in Madrid. V.V. Veeder highlights some of the areas where the rules may be improved. Should they ensure more transparency, for example by requiring witness statements to disclose the level of involvement of counsel in their preparation? This is certainly a controversial proposition. Should the parties be under a general duty do disclose all documents in their possession, whether favourable or not to their case? Even more difficult is the issue of privilege. As Guido Tawil argues in his outstanding paper, it would certainly be desirable to establish harmonized transnational standards of privilege. This has been done in many areas of arbitral practice, so why not in the field of privilege? There are still important differences, however, between national traditions on privilege. Rules on privilege are mandatory. They relate in many respects to the fundamental rights of the parties. Harmonizing them will require patience and caution. [Page449:]

In any case, the rules should preserve the fundamental characteristics that have contributed to the success of arbitration: the flexibility of the process and the wide discretion granted to arbitrators in addressing evidentiary issues.

Of course, the most hotly debated question is always that of the level of specificity that is required from a party seeking the production of documents. We all know the generally accepted solution. The requesting party needs to identify the documents sought. If it seeks the production of a class of documents, such class of documents needs to be specifically identified. The IBA Rules are even more restrictive, in requiring that the requested class of documents should not only be specific but also "narrow". The requesting party also needs to show that the requested documents are relevant to the issues in dispute and material to the outcome of the case. These standards are not civil law or common law standards. It is fair to say that they have been almost universally endorsed. Do they reflect a proper balance between the need of efficiency and that of fairness? Surely, as Lucy Reed explains in her fascinating paper, full-fledged US-style discovery is unsuitable for the particular features of international arbitration. Nonetheless, tribunal-controlled and robust document production is beneficial to arbitration. The parties should be granted a fair opportunity to obtain the documents they need to prove their case, and in case of doubt the default rule should be in favour of document production. As Lucy Reed says: "an alternative rule that would allow document production only in exceptional circumstances would stifle efforts to get the truth".

An overly restrictive approach of arbitral tribunals to document production may lead parties to resort to the assistance of US courts when available. Section 1782 U.S.C. has opened the floodgates of discovery in international arbitration. Laurence Shore nonetheless considers that the sky is not falling on our heads post-Intel. However, much will depend on the level of restraint shown by district courts. The signals coming from the courts in this respect so far are rather contradictory.

It is often said that arbitrators have no imperium. As debatable as that proposition may be, it is certainly true that arbitral orders for the production of documents are rarely enforced. The sanction of a refusal to comply with the order will therefore boil down to the adverse inferences that the arbitral tribunal may draw from the situation. However, drawing adverse inferences is a difficult task. It is, as Vera Van Houtte shows, an exercise of balance and caution in which the arbitrators' discretion needs to be combined with the [Page450:] greatest regard to due process. Drawing adverse inferences from a party's refusal to produce documents amounts in many respects to creating a presumption. Should the tribunal allow the parties to present their views and inform them in advance of the inferences that it is considering to draw? But is it really possible to know beforehand what consequences, in any, the tribunal will draw in its award from the procedural behaviour of a party?

Finally, the disclosure of electronically stored information is currently attracting much attention from the arbitral practice. The Chartered Institute of Arbitrators has recently elaborated a Protocol for E-Disclosure. Loretta Malintoppi informs us of the work of the ICC Task Force on the Production of Electronic Documents, which she co-chairs. As Carole Malinvaud notes, there is no such thing as e-discovery, but the production of electronic documents presents specificities that need to be fully taken into account and should encourage arbitrators to adopt a pro-active approach to managing the case by consulting the parties at the earliest possible stage in order to discuss their expectations.

Ultimately, this annual conference of the Institute of World Business Law will have confirmed that early knowledge of the case by the arbitrators and pro-active management of evidentiary issues is a key element to striking a proper balance between the needs of fairness and efficiency in international arbitration. As Bernard Hanotiau perfectly put it in a recent publication:

"si des problèmes surgissent au cours de l'arbitrage quant à la communication de documents, c'est en général parce que le tribunal arbitral n'a pas adéquatement réglé la question au début de l'arbitrage. En d'autres termes, les problèmes qui peuvent se poser ne sont pas tant dus aux attentes des parties ou à l'étendue de la production qu'à une gestion inadéquate de la procédure par le tribunal arbitral." 2

1
See ICC Publication No. 689 (2005).

2
B. Hanotiau, 'La production de documents dans l'arbitrage international: essai de définition des "meilleures pratiques"', La production de documents dans l'arbitrage international, ICC International Court Bulletin, Special Issue (2006) p. 128.