“The Gang of Four” (and I’m not making this up) is the name taken by four of Europe’s leading arbitration institutions to describe their loose affiliation for discussing common issues and sharing best practices.

Arbitration's Gang of Four
Arbitration’s Gang of Four

The “Gang” consists of the German Institution of Arbitration (DIS), the Milan Chamber of Arbitration, the Arbitration Institution of the Stockholm Chamber of Commerce, and the Vienna International Arbitration Centre, as represented by their secretary generals Francesca Mazza (designate), Stefano Azzali, Ulfe Franke, and Manfred Heider.

They rotate their meetings among their seats, with the host organizing a public seminar on an important arbitration theme chosen for the cycle of all four meetings. This captures in perceptive light how the institutions manage similar issues in sometimes different ways. And the atmosphere of open dialogue gives the broader arbitration community an opportunity to contribute to the way they operate.

I had the honor of moderating the seminar at the Gang’s recent meeting in Milan, a lively discussion about the problems created by pathological clauses and how each institution addresses them. This was the second session on this theme.

Pathology Through the Eyes of the Four Institutions

As most practitioners know, pathological clauses are arbitration agreements whose unclear or inconsistent wording runs counter to the setting in motion of a proceeding that will lead to an enforceable award.

According to the Gang’s leaders, these problems clauses are more frequent than most practitioners realize. “At least 30 percent of cases have a threshold dispute over arbitrability due to poor drafting of the arbitration clause,” stated one secretary general. Another responded that this just refers to those cases where the language is disputed, and that “some 70-80 percent of all arbitration clauses are pathological in some way.”

The panelists all concurred that their institutions see a lot of badly drafted arbitration agreements.

As moderator, I had the good fortune that heavy lifting on this important topic had been done by my predecessor, Rolf Trittman, who kicked off discussion at the Gang’s seminar in Cologne this past April. Rolf astutely summarized how pathological clauses are managed by each of the four institutions, by tribunals appointed under their rules, and by the national courts in which they are located.

Notably, only two of the institutions (Stockholm and Milan) have rules that expressly authorize them to engage in a prima facie analysis of whether an arbitration clause is valid. In contrast, Rolf had pointed out, Vienna has a rule that only touches upon jurisdiction, Article 9(6), and the DIS rules are silent on the topic.

And yet, his research indicated, the absence of an express rule may be of little concern in Austria and Germany, as courts in those countries have an established record of finding an agreement to arbitrate even where the parties’ intentions were ambiguous. Rolf reported the same for Sweden. By contrast, the Milan rule could be seen as fitting a local need. Italy has only recently modified its civil procedure rules to encourages courts to recognize the validity of poorly drafted arbitration agreements, a change that is yet to be extensively tested.

The Epidemiology of the Pathology

Against the study by the previous session of the effects of pathological clauses, our Milan seminar focused on the different ways a clause can be problematic, the causes, and the role that institutions can play in preventing or at least in reducing the impact of drafting mistakes.

The panel considered various causes of defects in arbitration agreements: lack of expertise among the contract negotiators, imperfect compromises over which rules and institutions to apply, poor translations of otherwise good clauses, a natural tension between the parties’ need for simplicity versus their desire to customize the arbitral process to suit the parties’ needs, and the occasional tendency to over-lawyer the language of a dispute clause.

The panel had no shortage of real-life examples to draw upon, ranging from parties agreeing to have their disputes resolved by an “abritrary institute” to one of my own favorites: a page-long arbitration clause specifying every conceivable detail of what should happen in the stages leading up to and during an arbitration except (rather importantly) a requirement to actually resolve disputes by arbitration.

The Gang’s secretary generals also considered the range of defects, from those that can be easily overcome through an institution’s involvement to others than can be fatal to the arbitral process. Below are some examples:

Misnamed institutions and rules. Both the Vienna and Milan institutes noted that, within reason, this is not typically an obstacle. They each gave the humorous example of parties that had specified the “arbitrary court” of their cities, yet the courts had concluded this was intended to refer to their institutions.

Clauses that appoint one arbitral institution to administer proceedings under the rules of a different institution. Among the four institutions present, only the SCC indicated that they would conduct arbitrations under the rules of a different institution.

Over-specificity of arbitrator requirements. For example, how would the institutions handle an arbitration clause that specified that a chair or sole arbitrator be fluent in Estonian, Chinese, and French, but not a national of any of those countries? Not easily. The leaders of all four institutions concurred that they would never ignore such a requirement without the agreement of the parties. As moderator, I pushed back on this, as clearly intended to refer their dispute to arbitration; but all four insisted that, when it comes to choice of arbitrator, they would feel bound by any party specifications.

Finally, the panelists all questioned whether there is more that they, as institutions, can do to reduce the apparent plague of pathological clauses, such as stressing the utility of model clauses (vs bespoke drafting) and developing more models to fit specific party needs.

For more on pathological clauses and how institutions are responding to the problem, catch the seminars scheduled for January 17, 2014 in Stockholm and March 14 in Vienna, when the Gang of Four rides again.
White cowboy hat


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration
This page as PDF


  1. Isn’t it better to allow jurisdictional questions, which include pathological clauses, to be dealt with by the arbitral panels when constituted?

    There are two reasons for that: one is that the issue of the clause has to be dealt with by the panel in any event; each arbitrator must be satisfied that his or her actions are founded by the agreement of the Parties. They cannot rely on the administrative opinion of the institution for that.

    The second reason is that the Court (certainly at the seat, debatably elsewhere) always has the last word on these matters. Arguably the institution is redundant for this purpose.

    1. A good point, as always, and this was amply discussed at the seminar. The consensus would not only agree with you, but would say this what in fact usually occurs.

      During the panel’s exchange with the public, however, a well-known Italian arbitrator made a plea for the opposite, for institutions to play a more active role in deciding threshold issues before a case is sent to the tribunal. He provided an example from a recent case before an international institution in which he noted that the arbitrators were rather irritated by the passive role played by the institution, leaving an issue to the tribunal that the institution could easily have decided. Speaking as a party, I think it is fair to expect institutions to do more in this regard, as part of the value they add to the arbitral process.

      1. Do they add value to the arbitral process?

        The ICC system can work well in introducing new arbitrators (I recall my first appointment in 1979/80 when ICC Counsel was most helpful). The Engineering Institutions, however, had appointing powers but (apart from Civils) no legal knowledge. Others seem in between.

        It’s difficult to see how an Institution can have the authority to make a legal decision (as opposed to an administrative one) absent a specific term in the Rules invoked by the arbitration agreement.

        Most Arbitration Laws seem to expect panels to look after these issues, don’t they? Isn’t that part of the task?

        1. I think the question, from a party perspective, is that they could add value by doing more than current practice. If defective clauses are so prevalent, then wouldn’t arbitrability be more predictable for parties if institutions told us how they would handle the problems?
          Yes, specific rules giving the institution the authority to make prima face assessments (as discussed in the above post) would be the best approach.

          1. I wonder about the legal basis for saying in advance how an institution (or its panels) will deal with pathological clauses.

            Would or could the Rule preclude the panel from looking into an Institution’s decision and deciding otherwise?

            Won’t the Court of competent jurisdiction have the last word (which will resolve the matter for the future) in any event?

            While there is competition for Institutions to expend their scope, won’t there be an incentive to decide that a clause gives the institution the right to take a case?

  2. Dear Michael

    With reference to this extract from your article:
    “Rolf Trittman, who kicked off discussion at the Gang’s seminar in Cologne this past April. Rolf astutely summarized how pathological clauses are managed by each of the four institutions, by tribunals appointed under their rules, and by the national courts in which they are located.”
    Where could I find the report compiled by Rolf?

    1. Chantel, apologies for not seeing your question until now. I do not know if Rolf or the institutions have published his paper from the seminar (I hope they will), and I will put you touch off-line to see if he consents to sharing it.

      If it turns out the institutions will publish his paper, then I will also update my blog post to reflect this. I’m glad you raised this.


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.