Multiparty arbitration is a well-trodden topic that, over the course of the last several decades, has been the subject of multiple international conferences, abundant doctrinal writings and the scrutiny of arbitral institutions, courts and users. It is a subject that has long commanded the attention of the ICC, in particular. In 1991, the Institute of International Business Law and Practice published a collection of papers from its 1989 conference in Stockholm, under the title Multi-Party Arbitration: Dossiers of the Institute of International Business Law and Practice. 1 This publication was followed in 1994 by the Final Report on Multi-Party Arbitrations of the Working Party established by the ICC Commission on Arbitration under the chairmanship of Jean-Louis Delvolvé. 2This report was the culmination of more than 15 years of study by the ICC Commission. In 2003, the ICC revisited the topic once more in a publication entitled Complex Arbitrations: Perspectives on their Procedural Implications. 3

So why another ICC conference on this subject? Is it the case, as Stephen Bond has quipped, that this is a subject about which everything has already been said, but simply not yet by everyone? The large number of attendees (more than 150), the presentations of the speakers and the discussions that followed demonstrated that, irrespective of the attention that the topic of multiparty arbitration has already received, it continues to fascinate and to challenge practitioners, courts and arbitral institutions.

Part of the reason, of course, is that multiparty arbitration is not a single subject but a multiplicity of different and continually evolving topics. The conference programme was a testament to this. Some of the topics traversed familiar territory (e.g. groups of contracts, groups of companies, the separation and consolidation of disputes and the extension of arbitration agreements to non-signatories), but others covered relatively new ground,

[Page235:]

such as the development of 'class action' arbitration in the united States and the possible reform of the ICC Rules to cater expressly for some of the issues raised by multiparty arbitration.

During the 20 years since the Institute last explored the different facets of multiparty arbitration, there have been important shifts in both the legal landscape and attitudes towards multiparty arbitration.

Perhaps the most important shift has been the emergence, in the wake of the well-known decision of the French Court of Cassation in the Dutco case in 1992, 4 of a widely-accepted solution to the potential problems posed by the constitution of the arbitral tribunal in a multiparty context. Twenty years ago, the difficulties associated with the constitution of the arbitral tribunal in such cases were a source of considerable concern. The subsequent widespread amendment of international arbitral rules (including those of the ICC) to deal with this particular issue appears to have given multiparty arbitration a new lease on life, opening up avenues that previously remained largely unexplored in the predominantly bipolar world of international arbitration. It is notable how the mood appears to have swung, emboldening the ICC, as part of its current review of the ICC Rules of Arbitration (described by Simon Greenberg at the conference and in this publication), to consider incorporating into its Rules new provisions intended to regulate issues arising in a multi-polar arbitration context, such as the joinder of parties and cross-claims.

When the ICC last amended its arbitration rules in 1998, these changes were widely resisted. At that time, multiparty arbitration was generally viewed as a matter that was best left to the parties themselves and one that should not be the subject of regulation. Moreover, the international arbitration community was generally wedded to the view, consistent with the then architecture of nearly all arbitration rules, that arbitration was and should be a bipolar process, with a claimant (or claimant group) on one side and a respondent (or respondent group) on the other. Departing from this model was widely perceived to be fraught with difficulties and risks, beginning with the constitution of the arbitral tribunal, and likely to lead to inefficiency and delay. At the time of the 1989 Stockholm conference, one prominent practitioner noted: 'There is some law […] and […] there is some practice. In my view, however, both deter the sensible use of multi-party arbitration. The Holy Grail remains as elusive as ever.' 5

Twenty years have passed, and while this view still has its adherents - multiparty arbitration (of the multi-polar variety, at least) was described as a 'snake pit' by one of the speakers at this year's conference - the practical

[Page236:]

issues and potential difficulties arising from the organization and conduct of multiparty cases have generated far less comment and fewer expressions of concern than was the case in Stockholm in 1989. Whether or not, and if so to what extent, the ICC is close to finding the Holy Grail, given its resolve to deal with some of the related issues in its next set of arbitration rules, remains to be seen.

While there appears to be a new openness towards the possible accommodation of multiparty procedures in the ICC Rules of Arbitration, one aspect of the multiparty arbitration discussion that has not changed over the years is the central place occupied by the issue of the parties' consent. Speaker after speaker posited that the consent of the parties is fundamental. However, it was also repeatedly acknowledged that, notwithstanding decades of consideration of multiparty arbitration issues, parties often still do not cater for multiparty (or multicontract) arbitration in their arbitration agreements. This has inevitably led to tension between the recognized need for consent and the widespread desire for arbitration to be able to accommodate multiparty disputes in the interests of efficiency and the good administration of justice. This, in turn, has led to a debate about how faithful courts and arbitrators have actually been to the requirement of consent when finding or imputing consent that has not been explicitly expressed. Although there no longer appears to be much support, as was once the case, for mandating the consolidation of arbitrations by legislative action, the contributions to the conference and this publication demonstrate the continuing vitality of this debate and the differing ways in which consent may or may not be established by courts and arbitrators.

Professor karim youssef, the author of a new book on arbitral consent in a multiparty and multi-contract context, 6 set the stage for the discussion by raising the provocative question, based on his review of arbitral case law, of whether consent as a jurisdictional criterion has declined in importance. Do arbitrators and commentators merely pay lip service to consent as a 'fundamental pillar' of international arbitration, while, in reality, stretching the concept in order to do justice in specific cases? youssef argues that the case law with respect to consent is in a state of flux, claiming that an analysis of the case law reveals the existence of not one, but a multitude of approaches to jurisdiction in multiparty and multicontract cases. Among the doctrines that he considers is the so-called 'group of companies' doctrine, as formulated in the Dow Chemical case nearly three decades ago, with its controversial reference to the group as 'a single economic reality'. 7

[Page237:]

The group of companies theory, often perceived as an anomalous French doctrine, is in turn the subject of the contribution of yves Derains, who, intent on emphasizing the importance of consent in French jurisprudence, stresses that the existence of a 'single economic reality' in the Dow case was merely an indication of (and an ambivalent one at that) rather than a substitute for consent. The group of companies theory, he says, has today outlived its usefulness ('a fait son temps').

Whether or not that is the case, there is nevertheless ample room for discussion on who is best suited, and in what circumstances, to seek to divine the parties' intent when it has not been made explicit: the arbitrators, the courts or the arbitral institution? What presumptions, if any, should inform the decision-makers' attempt to determine the parties' intent? Do different or special considerations come into play when one (or more) of the parties is a state or state entity? What laws or rules of law are to be applied in making the related decision?

These difficulties are illustrated by two recent and very different cases, on different continents, that were the subject of considerable discussion during the conference.

One was the decision of the english Court of Appeal in the case of Dallah Estate and Tourism Holding Company v. The Ministry of Religious Affairs, Government of Pakistan. 8 In this case, as discussed in greater detail by Georgios Petrochilos, the english Court of Appeal affirmed the decision of a lower court refusing to enforce an ICC arbitration award made in Paris against the government of Pakistan. The lower court's refusal to enforce the award was founded on its determination that the arbitral tribunal had not properly applied principles of French law in deciding that the government of Pakistan was a party to the arbitration agreement. Given that the arbitral tribunal was chaired by an eminent english jurist, Lord Mustill, the decision of the lower english court, as affirmed by the Court of Appeal (but now on appeal to the Supreme Court), demonstrates in a particularly spectacular fashion how a general appreciation of the importance of establishing the common intent of the parties in determining jurisdiction in multiparty cases is nevertheless susceptible to widely different possible applications in specific cases.

Similarly, in a different context, the united States Supreme Court determined, in a much anticipated decision issued on 27 April 2010 in the case of Stolt- Nielsen S.A. et al. v. AnimalFeeds International Corp., 9 that an arbitral tribunal sitting in new york (and chaired by Gerald Aksen) had wrongly found that an arbitration clause that was silent on the subject permitted a class action

[Page238:]

arbitration. 10 The issue before the uS Supreme Court was whether the arbitration tribunal had properly found, in accordance with the Class Rules developed by the American Arbitration Association, that the arbitration clause in question permitted AnimalFeeds to bring a 'class action' arbitration in a case where the parties had stipulated that their arbitration clause was silent on that issue. While the Supreme Court has now found, by a majority decision, that the arbitral tribunal erred in the circumstances of that case, its decision does not exclude the possibility of class action arbitration. More litigation aimed at determining the consent of parties to such an extreme form of multiparty arbitration is thus likely to arise in the united States.

Twenty years after the Institute's last conference on the subject of multiparty arbitration, the community of international arbitration practitioners appears to have grown more comfortable with the concept. nevertheless, it appears that determining whether all the parties in a specific case have consented to multiparty arbitration will continue to present challenges that are likely to divide practitioners, courts and arbitrators for years to come.

[Page239:]

1
ICC Pub. no. 480/1.

2
ICC International Court of Arbitration Bulletin 6(1) (May 1995).

3
ICC International Court of Arbitration Bulletin, Special Supplement (2003).

4
Cour de Cassation, Sociétés BKMI et Siemens c/ société Dutco, 7 January 1992, Rev. arb.(1992) p. 470.

5
H. Lloyd, 'A national experience', in Multi-party Arbitration: Dossiers of the Institute of International Business Law and Practice, ICC Pub. no. 480/1, p. 62.

6
Consent in Context: Fulfilling the Promise of International Arbitration (West, 2009).

7
Cour d'appel de Paris, Société Isover-Saint Gobain c/ société Dow Chemical France et autre, 22 october 1983, Rev. arb. (1984) p. 98.

8
20 July 2009, EWCA Civ. 755.

9
559 U.S. ___, 2010 WL 1655826, at *4 (2010).

10
By coincidence, the oral argument in the Stolt-Nielsen case was to be conducted the day after the conference, on 9 December 2009.