by the Arbitration Association

Dissenting opinions and why they should be tolerated

March 12, 2019

Many years ago I heard a remarkable story from a seasoned arbitrator; this was from a time when he acted as counsel for a claimant party in a major arbitration at a seat which was not his home seat. At the end of the arbitration – when sitting in his hotel room in the evening – there was a knock at the door and it appeared that one of the arbitrators wanted to raise an issue with him. He said: We have progressed somewhat in our deliberations and we have come to a realization which prompts the following question: You may obtain an award for the entire claim presented with one dissenting arbitrator and you may receive a unanimous award for 50% of the amount. Which one do you want?

One may immediately be struck by the utter inappropriateness of such an initiative from one of the arbitrators (this was long ago, arbitrators today would not stray so totally off course). But still, on some level there was some justification in the approach: This party was going to win the case – and in such circumstances it would not seem so unfair to give it the advantage of choosing between a 100% success with a dissent and a 50% success with a unanimous tribunal. Because, what does the tribunal know, perhaps a unanimous award for half is worth more that an award for the full amount tainted by a dissent? A winning party is not involved in some lofty quest for justice but will seek to enforce the award at the best possible terms. Often the enforcement has to take place in the opposing party’s home jurisdiction and a dissent from a “local” arbitrator may well impact negatively on efforts to enforce.

So, this demonstrates one thing – dissents are not desirable. They detract from the authority of the award and they may jeopardize the enforcement. They may also give grounds for – or at least encourage – losing parties to seek annulment of the award.

Looking at it this way one may well ask whether dissents should be allowed at all?

The answer is that dissents necessarily must be allowed, while, at the same time, they certainly should not be encouraged. And they should be allowed, however grudgingly, despite their generally negative influence on the authority of the award proper.

Dissenting opinions are a relatively rare occurrence

One may note that national arbitration laws generally do not mention dissents. Neither do the UNCITRAL ad hoc arbitration rules, or institutional rules[1]. In the investment arbitration area one may note that the ICSID Convention in its Article 48(4) provides that “ [a]ny member of the Tribunal may attach his individual opinion to the award, whether he dissents from the majority or not, or a statement of his dissent”. 

The general silence of arbitration laws and rules on the topic of dissenting opinions does not operate as a bar against such manifestations of arbitrator opinion, but testifies to a general feeling of discomfiture at the phenomenon as such. And arbitral institutions and tribunals do deal with such occasional discord. A dissenting arbitrator is entitled to have his or her opinion made known, normally in the context of the award being handed out to the parties.

Dissenting opinions are relatively rare occurrences. According to statistics by the ICC in the period 2009-2011, the number of unanimous awards represented 83% of the total number of awards, while dissenting opinions represented a number of 16 %. Cases where the party appointed arbitrators differed from the presiding arbitrator where exceedingly rare, actually less than 1%.[2]So, it is practically always a party-appointed arbitrator that dissents, but occasionally it happens that the party-appointed arbitrators take issue with their presiding arbitrator.[3]

A dissent is not part of the award, and, consequently, will not be scrutinized by the ICC. It will be reviewed, however, a s part of the scrutiny process relating to the award itself.

Dissenting opinions are not awards and do not form part of an award. This means that they will not be the subject of scrutiny in an ICC arbitration. The award should, in all circumstances, be signed also by the dissenter, and a dissenting arbitrator should not refuse to sign the award on accountof the dissent. In fact, some jurisdictions require that the reason for an arbitrator’s refusal to sign is noted in the award.[4]

Different categories of opinions

Dissenting opinions find expression in a wide-ranging variety of ways, but most frequently in a separate document which is “limited to giving a different view on discrete issues that have caused disagreement between arbitrators”.[5]

Dissenting awards may express differences of opinion not only in respect of the conclusion of the tribunal but also as regards the reasoning where the conclusion will be unaffected. There are also separate opinions which differ in neither respect, but where the arbitrator has felt a compelling need to expand on a certain line of reasoning or with respect of a particular legal issue raised by the case. These categories of opinions are named in different ways, e g “Separate and Dissenting Opinion” (Hochtief), “Concurring and Dissenting Opinion” (Impregilo), “Declaration” (SGS v Philippines), depending on their general thrust and direction.

Ways to express disagreement

As is well known, dissents are usually expressed in the form of a separate declaration. But disagreement may not necessarily be expressed in this way. Oftentimes a dissent (or the elaboration of a particular issue) is included in the running text of the award proper. This is particularly the case where the disagreement is limited to a discrete issue that may play a less dominant role overall and where a separate declaration would be unnecessarily cumbersome. Usually one would see language such a as “…the arbitrator [name] disagrees in respect of the relevance of…”. The anonymous alternative of writing “…one arbitrator begs to differ as regards the computation….”is also seen occasionally. There is of course nothing that proscribes the practice of such anonymous dissents, but based on the view expressed in this paper misses the point. The rare instance of dissent must reasonably be driven by an arbitrator’s wish not to be associated with a particular point of view to which he or she cannot subscribe. This sole, legitimate reason to dissent is of course not achieved by way of anonymity. And, in any event, it does not normally require much acumen to figure out the identity of the anonymous dissenter.

Obviously, dissents ought not to engage in criticism of the majority opinion. This will only depreciate the award with no redeeming benefit. Fortunately, the prevailing practice is rather the opposite: dissenters normally show restraint, expressing reverences, even solicitously, to the majority arbitrators in its introductory language. Usually the majority does not comment on the dissent at all[6]. Occasionally, however, the rifts between the minority and the majority are too great to prevent a certain degree of acidity between the dissenter and the majority. Such, to the extent it occurs, is usually expressed in exquisite form[7].

There is also a rare class of endemic dissenters, a species which luckily seems to be on its way to extinction, if not perdition, who systematically attach their unreasoned “Dissenting” scribbled on each award. This is of course a singularly unhelpful act of an arbitrator normally driven by a misconceived – but improper – urge to show allegiance to the appointing party.

There are fundamental differences between commercial arbitration and investment arbitration

In commercial arbitration the need for precedent is small. Commercial cases tend to be fact-driven and distinguishable from one another by the particular circumstances of each case. The precedential interest is small or non-existent (also looking away from the fact that there is, theoretically no such thing as “precedent” in arbitration other than as a source of persuasive argument).

So, the combination of non-publicity and fact-variety makes for a very small role for dissenting opinions as a source of alternative legal reasoning in commercial arbitration, lessening or reinforcing the persuasive value of legal precedent. The need to dissent can only be dictated by the concerns of the individual arbitrator to announce – to the limited audience of his or her co-arbitrators and party representatives – a difference of opinion on a matter of principle.

The situation in investment arbitration is radically different. Investment disputes proceed on the strength of international treaties, which use essentially the same techniques in terms of dispute resolution and contain very much the same substantive standards of protection. From this follows that identical procedural and substantive issues tend to re-surface from case to case. Typical examples are: Does an “umbrella clause” make every contractual dispute susceptible to investment jurisdiction? Does an MFN-clause allow “import” of procedural features of another investment treaty? Is there a distinction in valuation between lawful and unlawful expropriations? This puts the need to dissent into a very different light. The much greater publicity of investment awards also makes dissents take on another degree of importance[8].

Although there is a movement- supported by arbitral institutions, practitioners and scholars – to increase transparency and openness also in the area of commercial arbitration, the fact remains that the very great majority of awards, not to mention the arbitral proceedings themselves, are private and confidential. Generally speaking, the pull toward increased transparency in commercial arbitration is weak. This is easily explainable by such awards having scant persuasive value and offer dull reading.

In investment arbitration the situation is different. It is a fact that the majority of ICSID awards are made public (by agreement of the parties) and this also applies, albeit to a lesser extent, to investment awards emanating from the second most relied-on forum – the SCC of Stockholm – in the investment dispute area. 

This generates a quandary in its own right, and of an intractable nature. A few fundamental concepts and their application (the effect of waiting periods, the reach of MFN clauses) are of a binary nature. They will open access to international arbitration or they will not, depending on the professed affinity of the particular arbitrator. By reviewing prior awards and also dissenting opinions a putative litigant may select an arbitrator who, in prior practice, has adopted a particular application of a concept which will be favourable to the desired outcome on the threshold question of jurisdiction.

What is the use of dissenting opinions?

One may well ask what use dissenting opinions serve in arbitration. In the adjudication taking place in national judiciaries it is sometimes pointed to the fact that dissents may be helpful in the appeals process in that they contribute to give an additional dimension to the legal issues involved. By so doing, the argument goes, they give a more all-sided and thorough assessment of the case. However, in arbitration there is no appeals procedure and, moreover, there is generally no publicity with brings dissents to the knowledge of the wider public. There is no rule of precedent, so each case is dealt with on a strictly-stand-alone basis. Or so at least it is postulated as a principle.

In investment arbitration it has occasionally been said that arbitrators should at all cost refrain from dissenting. This may be so but the fact remains that dissents are much more frequent in the realm of investment arbitration than in commercial arbitration.

There are certainly arguments canvassed that underscore benign effects of dissenting opinions. They will usually posit that dissents may incentivise the majority of the tribunal to a more careful analysis in its reasoning, or that it will promote party confidence in the arbitral process in that it will “enhance the legitimacy of the process by showing losing party that alternative arguments were considered, even if ultimately rejected”.[9]

However, on a balance, it is fair to say that neither of these considerations carry any persuasive weight. The aspiration of arbitrators to produce carefully reasoned awards should certainly not (and does not normally) require the confrontation of a dissenting opinion against a majority view to ensure the required quality of the ensuing award. 

Neither does the argument, sometimes touted, that dissents may contribute to the development of law deserve much traction. It is certainly not the business of arbitrators to dedicate their intellectual efforts to the development of law (but to solve the particular dispute at hand). 

A particular feature of dissents that may cast their even-handedness into doubt is the fact they hardly ever disfavour the party that appointed the dissenting arbitrator. This raises the important concern that arbitrators may not always, after all, adopt an entirely independent position in the arbitration. This is absolutely a valid concern, even if it is easy to understand the issue. An arbitrator which is not prepared to agree with the co-arbitrators’ positive approach to the appointing party finds herself between a rock and a hard place. The arbitrator will certainly experience discomfort at being associated with a particular legal proposition on which the case turns but will likewise be uncomfortable with taking an express stand in opposition to the party that showed the confidence of making the appointment. 

So, why should dissenting opinions be allowed?

One may well ask why dissenting opinions are allowed at all, considering that they do not enhance the authority of the arbitral award and have no business to promote then administration of justice on a general plane.

The reason why dissenting opinions must be allowed is to ensure arbitrators of the fundamental right to freely express their opinion untrammelled by any dictate from the co-arbitrators where there is no unanimous view. In a wider sense it engages the principle of freedom of expression. One may expect that arbitrators may even feel unease at being bandied together with other arbitrators – whom they may well not have met before the particular appointment – whose ways of approaching legal issues they may have no possibility to foresee. 

The freedom of announcing dissent stems from the fundamental – and inalienable – right of an arbitrator to freely vent his orher opinion and not to risk being forcibly shoehorned into any particular view of which he or she does not partake. This, indeed, compels allowing dissent to be articulated also in appreciation of the negative effects such expression of discord has for the integrity of the arbitral award.

This having been said, it does not mean that efforts should not be made by a tribunal to seek common ground in the course of deliberations. With sufficient portions of patience, perseverance and resilience dedicated to good faith efforts serious attempts should be made to consensus. A draft of the dissenting opinion should be provided to the majority so that conflicting views can be examined and possibly reconciled. This may even operate as an invitation for the arbitrators to resume deliberations. No efforts should be spared to explore all avenues to seek common ground between the arbitrators and to render the dissent superfluous. 

Christer Söderlund,
Senior Counsel, Morssing & Nycander, Stockholm

[1]On a topical note it may be mentioned that the 1999 SCC Arbitration Rules explicitly stated that ”[a]n arbitrator may attach a dissenting opinion to the Award”. This provision was innocuously left out in the next revision of the Rules in 2007 with no change in substance intended.

[2]J Fry, S Greenberg, F Mazza, The Secretariat’s Guide to ICC Arbitration, ICC Publication 729 (Paris,2012), Table 35 at section 3-1131

[3]This happened in the well-known Tokios Tokele-case, where the Chairman, Prof Prosper Weil, wished to assert jurisdiction but could not bring his co-arbitrators aboard. Prof Weil, as a consequence, resigned from the Tribunal.

[4]This is the case, for instance, of Sweden, Article 31 of the Arbitration Act

[5]Ibid, at section 3-1143(b)

[6]Although it certainly happens, e.g. Iskale v Turkmenistan

[7]A highly readably dissent thanks to its literary, scholarly and, not least, pedagogical approach (in addition to its incisive analysis) is the dissent by Prof Georges Abi-Saab to the Abaclat investment arbitration. 

[8]The recurrent feature of certain issues in investment arbitration have given rise to the concept of ”issue conflict” – should an arbitrator excuse him- or herself for having taken a certain position on such a recurrent feature because it will re-emerge in the present arbitration? (The answer is no, but the subject is outside the scope of this paper.) 

[9]Mosk & Ginsberg, Dissenting Opinions in International Arbitration, 15, Mealy’s Int’l Arb. Rep 6 (2000)]