I. Introduction

Although, as far as I have been able to ascertain, none of the arbitration rules of the leading arbitration institutions, such as the ICC International Court of Arbitration, the American Arbitration Association, the London Court of International Arbitration, and the Arbitration Institute of the Stockholm Chamber of Commerce, provide for the submission of dissenting opinions, all of them accept them and distribute them to the parties. In this regard, the Anglo-American practice of permitting dissenting, as distinct from concurring,1 opinions has prevailed in ICC arbitration over the civil law tradition of collegial decision-making. 2 A dissenting opinion does not, however, form part of an ICC award.

I am a fervent follower of the school that believes arbitrators should make every effort to arrive at an award that commands the concurrence of all involved in the enterprise. Nevertheless, upon occasion, I have felt compelled to write dissenting opinions. 3 This has prompted me to consider in greater detail whether dissenting opinions have a place in arbitration and, if so, what that place is. That in turn has made me weigh the advantages and disadvantages of dissenting opinions in arbitration, as contrasted with court adjudication. The final judgment, I submit, cannot be made until these are carefully considered. The following attempts to provide this consideration. It will pay particular attention to the differences between arbitration and court adjudication. 4

II. Advantages of dissenting opinions

A. Intellectual honesty

The dissenting opinion permits an arbitrator or judge to provide his or her personal judgment, arrived at with due consideration of the facts and the law, and does not force him or her to concur in a decision to which he or she cannot fully subscribe. In [Page38:] a real sense, the dissenting opinion represents the prevalence of individual integrity over collegial solidarity. 5 Indeed, the concurring opinion affords the same advantage. This advantage applies to both arbitration and judicial adjudication. 6

B. Contributing to development of the law

In the United States, dissenting opinions have frequently marked the evolution of new approaches, different solutions, and analyses more in tune with new developments. They have thus contributed to the continuing adjustment of the law to new conditions. The writers of these opinions have been more ready to take the next step and to propose novel solutions, perhaps in part because, as dissenters, they did not bear responsibility for the final result. As a consequence, one of them, Justice Oliver Wendell Holmes, even became known as the 'great dissenter', and, in many cases, his opinions eventually became the law. This spark of innovation cannot as readily be found in collegial decisions, like those of the European Court of Justice, that must provide cover for all of the judges involved in the decision. However, this benefit is of reduced importance in arbitration, because dissenting opinions are generally not published, even in the few instances in which the plurality opinions are. 7

I have long argued that arbitral decisions should be published, in their entirety, as a matter of course. 8 This will promote transparency, social acceptability of arbitration and orderly development of the law. For the same reason, dissenting and concurring opinions should be published. 9 International arbitration is increasingly developing and applying procedural and substantive rules specially adapted to international situations. 10 These special rules, sometimes called lex mercatoria, need the impetus that dissenting and concurring opinions can give them. They potentially play the same role that the great dissenters played in the development of law in American courts.

C. Promoting judicial and arbitral responsibility

The dissenting opinion, and also the prospect of it, is likely to force the plurality to be more responsible about discharging its adjudicatory functions. The dissenter is likely to stress the weaknesses in the plurality's decision and force the plurality to address them in the factual and legal analyses in its decision.

This is especially important in arbitration in which the award is normally not subject to review, except on very limited grounds, and the prospect of review is therefore not as likely as in ordinary litigation to motivate the decision-makers to provide a cogently constructed award. 11

The value of the dissenting opinion in this regard is not to be underestimated. Pluralities in ordinary courts often change when confronted with a draft dissent. When they do, a unanimous opinion is likely to be the result. This is a significant advantage, especially in arbitration, for a unanimous opinion, in which the arbitrator appointed by the losing party concurs, is likely to be more readily accepted and not to cause an attack in court, and thus promote the aim of arbitration to settle disputes efficiently and in one instance. [Page39:]

D. Creation of a more complete record

In American court proceedings, a verbatim record of all proceedings, including the examination of witnesses, is normally prepared. A dissenting opinion is therefore not needed for the purpose of creating a more complete record.

In arbitration, however, verbatim transcripts are frequently not prepared and reserved only for disputes involving relatively large interests or amounts. The plurality opinion in arbitration is likely to pay less attention to facts and arguments that are not helpful in supporting the ultimate conclusion reached but that the dissent is likely to highlight. The dissent may thus provide the premises for a successful attack in court. It might be argued that this is not a benefit and that the dissent, especially a dissent prepared by the arbitrator appointed by the losing party, may be more intent upon creating a record on which an attack in court can be based than upon presenting a dispassionate view of the case. Indeed, as discussed below, this may be a disadvantage of a dissent peculiar to arbitration. However, the circumstance that the dissent may be abused for this purpose is to be taken into account, but does not justify disallowing it. 12 After all, the plurality may similarly be motivated by its desire to protect its award from attack by divulging a less than complete, or even a misleading, record. A watchful eye cast by the institution when reviewing draft awards would appropriately redress arbitral abuses in this regard. 13

E. Relevance to the review of awards

Under the ICC Rules of Arbitration, the ICC Court may require revisions of a proposed award on matters of form and draw attention to matters of substance. 14 However, the staff counsel and the members of the Court who fulfill this function are, in the nature of things, not as familiar with the case as the arbitrators. An arbitral dissent may alert them to problems that may be redressed before the award is approved, but that might otherwise have remained undetected. Indeed, the Court, in the discharge of its review functions, has had occasion to invite the drafters of awards prepared by the majority to reconsider its draft in the light of the simultaneously submitted dissent.

III. Disadvantages of dissenting opinions

A. The dissent may be the platform for the expression of partisan views

It is obviously preferable that arbitral tribunals render unanimous decisions. Unanimity adds to what may be called the moral force of the award. This is particularly true when members of the tribunal have been appointed by the parties. The circumstance that the arbitrator appointed by the losing party concurred in the decision will necessarily promote its acceptance and temper the desire to attack it in court.

When none of the arbitrators has been appointed by a party, the dissent is more likely to reflect the dissenter's dispassionate view of the case. 15 But when the dissenting arbitrator was appointed by the losing party, the dissent becomes naturally suspect. 16[Page40:] It may then serve the purpose of assuring the party that appointed the arbitrator that its arguments have found resonance with at least one of the arbitrators. By itself, that may not be particularly objectionable, but the desire to please the party that appointed him or her may move the dissenting arbitrator to forego a real effort to reconcile his or her views with those of the majority and to use the dissent in order to provide the basis for an attack upon the award in court.

B. Abuse of the dissent to create a basis for attack

The party-appointed arbitrator knows full well that the party that appointed him or her is moved by the desire for victory in the arbitration. A party-appointed arbitrator must face this undeniable fact and deal with it. 17 One way of doing this is to disclose a basis for attack on the award in the dissent. 18 This is most regrettable if, in the effort to do so, the dissenter discloses confidential matter, and particularly the deliberations among the arbitrators, in the dissent. The confidentiality of the internal deliberations of the tribunal is essential to full and frank arbitral discussions. The notion that arbitration is confidential should have the greatest strength when applied to the tribunal's internal deliberations. Nevertheless, disclosure of deliberations could serve a purpose if the arbitrators misbehaved and their misconduct would remain secret unless disclosed. For example, arbitrators who disclose that they were denied an opportunity to present their views may properly disclose this in their dissents. But they should be particularly careful to be accurate in their disclosures. In some recent cases, dissenting arbitrators have accused their colleagues of failing to allow them to express their position in the arbitral tribunal's deliberations when, in fact, the dissenting arbitrators had been given an adequate opportunity to present their views. 19

C. In arbitration, the dissent may not reflect the true views of the dissenter

As indicated, one of the significant benefits of a dissent is that it may point the way to novel solutions and approaches. In arbitration, it is less likely to play that role unless the novel approach has been advocated by the party that appointed the dissenter. 20 The dissenter may be reluctant to advance the novel approach on his or her own, because it may imply a shortcoming on the part of the counsel who appointed him or her, but did not advance it. 21 The appointing counsel would not appreciate this implied criticism and, as a consequence, be less likely to choose the arbitrator in the future or recommend the arbitrator to others. On the other hand, dissents that mirror merely the views and arguments advanced by the party that appointed the dissenter may give rise to a reasonable doubt as to the dissenter's having made appropriate efforts to reach a collegial decision. This problem would be avoided if arbitrators were not appointed by the parties. 22

D. The prospect of a dissent may harden views

While the prospect of a dissent may have the wholesome effect of promoting a more careful adjudication, it may also harden positions and prevent better collegial decision-making. The arbitrator who meets resistance to his or her views may too easily give up trying to persuade the other members of the arbitral tribunal, who in turn may be [Page41:] less willing to give adequate consideration to the dissenter's views because they expect he or she will dissent anyway. In fact, the disadvantage of the dissenting opinion in this respect may well outweigh its advantage.

IV. Conclusion

All in all, the advantages of dissents appear to outweigh, both in number and in substance, their disadvantages. However, for me this remains a close call. As long as the party-appointed arbitrator remains part of the scene, potent arguments can be made in favor of dissents. 23 Abolition of dissents would eliminate the possibility that the dissent is written only to appease the party that appointed the dissenter or that the dissenter seeks to lay a basis, whether or not appropriately, for an attack in court. However, abolition of dissenting opinions might incline the majority not to give adequate consideration to the views of the dissenting arbitrator. The prospect of a dissenting opinion will ordinarily move the majority carefully to consider the dissenter's views. To that extent, the potential dissent serves a most useful purpose. However, when the prospect of a dissent does not move the majority to adjust its views, the potential dissenter must weigh carefully whether the issuance of a dissent will serve a proper purpose.

In my opinion, a potential dissenter should issue a dissent only after having made all possible efforts to reach an accommodation with his or her co-arbitrators. After all, things are normally not clearly black or white. Both on the facts and the law, reasonable arbitrators may differ. Proper respect for alternative views is the hallmark of an effective and conscientious arbitrator. But if the majority is uncompromising on essential points, the arbitrator who is in the minority has little choice. Surely, the minority arbitrator can seek consolation in the fact that he or she gave it his best effort, but not writing a dissent would appear to undermine the sincerity of the arbitrator's differing views. If the minority arbitrator writes a dissent, it is his or her professional obligation to do as thorough and competent a job as possible. A perfunctory dissent will serve neither the institution of arbitration nor any of the parties. A responsibly formulated dissent is also required to encourage the majority to proper reconsideration of its award in the light of the dissent. The circumstance that the ICC Court, upon occasion, has invited such reconsideration illustrates the validity of this point.

If there are no party-appointed arbitrators on the panel, these considerations lose a large measure of their pertinence. Arbitral adjudication will then parallel to a far greater extent judicial adjudication. No arbitrator can then reasonably be suspected of espousing only a partisan view. Getting a collegial decision becomes less important, because the concurrence of the non party-appointed arbitrator is less meaningful. The civil law model of collegial decision-making may also comport better with the aim of arbitration to settle disputes in a single instance with an unqualified decision. And as long as dissenting opinions are not published, their contribution to development of the law is less likely to be significant.

I therefore conclude that, under current practice, dissents should remain permitted and should be published, but that they should not be used to undermine the aims and principles of arbitration, especially its neutrality and confidentiality, or to lay an improper basis for judicial attack on the award. Only responsible use of dissenting opinions can be to the benefit of arbitration.

1
Concurring opinions may be concurring in part or in whole. A separate opinion may also be concurring in part and dissenting in part. I have never been involved in an arbitration in which an arbitrator wrote a concurring opinion. However, although arbitration is structurally inclined to collegial decision-making, there would appear to be no formal objection to submitting concurring opinions, which, to the extent they do not accept the majority's opinion, become dissenting.

2
The usual civil law rule is that all judges sign the court's decision, even though they do not agree with it. They, in this sense, accept the majority rule. This is also the rule in the Court of Justice of the European Communities. In my commentary on the EC Treaty, I criticized the drafters of the treaty for not providing for dissenting opinions which I had found an attractive novelty in Anglo-American jurisprudence. Since then, I have changed my mind. Since the European Court's justices serve for only limited terms and cannot be reappointed without the concurrence of their native country, it would appear preferable not to permit dissents; for justices facing re-election might otherwise feel compelled to dissent in order to curry favor with their home states which would have to propose their re-election.

3
Dissenting opinions are normally written by the arbitrator appointed by the party that lost on the issues addressed in the dissent; for the arbitrator appointed by the winning party will usually join the majority. Some years ago, I asked Judge Feinberg, then the Chief Judge of the prominent Federal Court of Appeals for the Second Circuit, whether any of his dissents had ever swayed the majority to change its mind. To my surprise, he recalled a recent case in which Judge Henry Friendly had submitted a draft for the majority and he for the dissent and in which each judge informed the other that the other's draft had caused him to reconsider his position. This is unlikely to occur in the world of party-appointed arbitrators. On the question of what should move an arbitrator to write a dissenting opinion, see my conclusion below.

4
Interestingly enough, I have not found specific authorization for the rendering of dissenting opinions in Anglo-American practice either. It most probably finds its basis in the original English practice of the judges' giving their views orally right after the conclusion of the trial, which did not offer an opportunity for prior collegial deliberation and formulation of a collegial decision.

5
One might have thought that the absence of dissenting opinions in civil law judicial adjudication contributed to the very succinct and brief decisions that are the hallmark of the case law of the French, Belgium, and Dutch courts of cassation. However, the highest German and Swiss courts do write more elaborate decisions.

6
The relative absence of concurring opinions in arbitral practice may be due to the greater pressure in arbitration than in court adjudication to reach a unanimous award.

7
I strongly favor the publication of arbitral dissenting opinions, especially when they indicate novel approaches. After all, dissenting opinions often augur the development of new law that otherwise might not be easily discernible.

8
Arbitral awards issued under the auspices of the ICC International Court of Arbitration are published on a selected basis, primarily in the Collection of ICC Arbitral Awards, co-published by ICC Publishing and Kluwer Law International, the ICC International Court of Arbitration Bulletin, the ICCA's Yearbook Commercial Arbitration and the French Journal du droit international, commonly known as Clunet.

9
Dissenting opinions are probably not published, because they are not formally part of the awards, even though they may be included in the text of the award.

10
I have described these trends in H. Smit, 'Substance and Procedure in International Arbitration' (1991) 65 Tulane Law Review 1309.

11
In fact, in the USA arbitrators in domestic cases often do not issue reasoned awards in order to reduce attacks in court.

12
Dissenting opinions may also be used for the purpose of improperly disclosing internal deliberations of the tribunal that might then be seized upon by the award loser in an attack upon the award. See section III.B, below.

13
The ICC Court has an especially important position in this regard, because the arbitrators in the majority may not discern this abuse when it is not embodied in the draft dissent sent to them.

14
ICC Rules of Arbitration, Art. 27.

15
I have strong reservations about the desirability of permitting party-appointed arbitrators. Even when a party-appointed arbitrator properly discharges his or her obligation impartially to decide the case, an appearance of partiality on the part of the party-appointed arbitrator is almost inevitable. The presence of a party-appointed arbitrator may also impede full and frank deliberations, especially when the partiality of the party-appointed arbitrator manifests itself.

16
When the party-appointed arbitrator is part of the majority, he or she is unlikely to write a separate concurring opinion.

17
Of course, the arbitrator who is asked by a party to accept an appointment should make clear that he or she will properly discharge his or her obligations, but the party making the appointment may not take this disclosure at face value. In my opinion, a party's appointment of an arbitrator should be moved principally by the perceived competence of the arbitrator and the arbitrator's ability to function effectively as a member of the tribunal.

18
This is appropriate only if a proper basis for attack actually exists.

19
If improper disclosure is made in the final award, the appropriate sanction is for state courts to disregard it. If this is to become the accepted rule, it would be unnecessary to have the other arbitrators come forward to contest the allegations made. Unfortunately, at the present stage of international arbitration practice, it is not clear whether state courts are ready to disregard improper disclosures and to issue an interlocutory order to that effect.

20
However, in that case the dissenter is merely endorsing an approach that the plurality has rejected. The arbitrator's adding his or her voice to that of the party that appointed him or her would serve only the purpose of signifying his or her loyalty.

21
It would appear preferable for the arbitrator to put that approach on the table during the arbitration for the parties' consideration. In my experience, co-arbitrators usually concur in an arbitrator's doing so.

22
An analytical and empirical study of the advantages of party-appointed arbitrators would appear long overdue.

23
It should be noted that in ICC arbitration arbitrators are nominated by the parties and confirmed by the ICC Court.