For dispute resolution practitioners familiar with the concept of the seat of arbitration, it may come as a surprise that the new UN Convention on International Settlement Agreement Resulting from Mediation does not include provisions in relation to the ‘seat’ of mediation. Why, you may ask? The Convention includes no provisions on ‘seat’ simply because there has never been the need for a ‘seat’ of mediation when cross-border disputes are brought before a mediator, and this remains the case. In this blog post, we will develop a hypothetical scenario to explore the issues.

What is the concept of ‘seat’ in international arbitration law and practice?
To illustrate the role that the seat in international arbitration plays, let’s start with Mia’s story. Mia runs a graphic design business (sole proprietorship) in Germany. For the past five years she has been working with Maha Pty Ltd (‘Maha’), a company registered in New South Wales, Australia. In accordance with an arbitration clause in her contract with Maha, Mia takes Maha to the Singapore International Arbitration Centre (SIAC) alleging breach of contract involving inter alia infringement of IP rights. The seat of arbitration is agreed to be in Singapore, with the choice of law agreed to be Singapore law. Throughout the hearing of the case, one of the appointed arbitrators on the tribunal, George, is constantly distracted by his cell phone, anxiously awaiting news of the birth of his first grandchild. At one point the representative of Maha notices that George is texting on his phone whilst its lawyers were making oral submissions. When the tribunal issues an award in favour of Mia, the representative notes that George’s signature was absent on the award: he had not participated in the deliberations among the other tribunal members as the award was drafted.

Mia is likely to want to enforce the award against Maha’s assets in Germany, and in the alternative, against their assets in New South Wales. To challenge the award, however, Maha would take the award to the court of the seat of arbitration (i.e., the Singapore High Court) to seek to have it set aside for breach of natural justice according to Singapore law on the basis that there is real evidence to show that not all members on the arbitral tribunal (i.e., George) provided their reasons for adjudicating against it. If Maha’s challenge at the seat of arbitration were successful and the award was set aside by the Singapore High Court, Mia may not be able to enforce the award in Germany or New South Wales, as Article V(1)(e) of the New York Convention provides for refusal of enforcement of arbitral awards if it has been set aside by a competent court at the seat of arbitration.

The roots of the ‘seat’ of arbitration can be found in what is conventionally known as ‘localisation theory’ – an approach generally embraced by most jurisdictions.

The ‘localisation theory’ gives rise to an alignment of the law of the ‘seat’ of arbitration where the arbitration physically takes place (lex loci arbitri) with the governing law of the arbitration (lex arbitri). In other words it is the law of the place where arbitration takes place that ultimately governs the arbitration, through the courts of that place (seat).

The ‘localisation theory’ reflects the reality of international arbitration practice, as it is routine for courts at the seat of arbitration to scrutinise international arbitral awards, and sometimes set them aside if they do not, for instance, comply with mandatory rules of the lex loci arbitri, or for breach of natural justice considerations. The parties at arbitration may also require the assistance of the courts at the seat to issue interim orders such as freezing orders (or, as it is known in common law jurisdictions, Mareva orders): this underscores the need for convergence of the lex loci arbitri with the lex arbitri.

What about international mediation? Why is there no ‘seat’ of mediation when cross-border disputes are brought before a mediator?
As counterpoint to the ‘localisation theory’, there is the ‘delocalisation theory’. The latter theory coheres with the procedurally and geographically flexible and fluid process of international mediation (although delocalization has certainly has had its followers in international arbitration circles – but we digress).

The ‘delocalisation theory’ envisions a dissociation of mediation from the laws of the geographical location where mediation takes place. This approach puts a premium on party autonomy and the ability of parties to self-regulate by the laws and rules they have selected. The theory rejects the idea of any supervision by a putative ‘seat’ of mediation.

So how does this pan out in practice? Let’s re-imagine that Mia and Maha agree to go to mediation instead of arbitration. They select an Australian-based German mediator fluent in the languages and culture of both parties. The mediator conducts the first mediation meeting via video conference and the second via Skype to accommodate the diverse travel schedules of parties and lawyers. Encouraged by the early indications of progress, the Australian and German parties decide to come together and meet in Bali for two days of mediation.

Increasingly, international mediation processes are not limited to one geographical location but take place in different locations; alternatively mediation processes may make use of online dispute resolution platforms, video-conferences or generally available technology such as Skype, Zoom and even email. Under these types of circumstances it becomes impossible to identify the “seat” of mediation.

During the Bali mediation meeting, Mia and Maha find themselves reflecting on what led to the breakdown in their business relationship and the resulting legal claim. They both agree that up until the current dispute their business relationship had been going well; they remain open to the possibility of continuing it, provided they can settle the current claim amicably and to their mutual satisfaction. As the mediation progresses they move between negotiating a solution to the current problem and re-negotiating the terms of their business relationship to better address each party’s business needs. This type of dynamic problem-solving approach takes Mia and Maha beyond the legal parameters of the IP claim, which was the initial catalyst for dispute resolution. Mediation effectively moves the parties into a negotiation space where they can choose to redefine their legal dispute in a way that reflect commercial realities and other priorities; they can also renegotiate the terms of their current contract – in terms of substantive provisions and also potentially including choice of law and forum. It is a space where they will assume ultimate responsibility for the substantive outcome of their dispute. In short, mediation is a fundamentally different process to international arbitration, and one where the notion of a “seat” would appear misplaced.

The careful language of the Singapore Convention on Mediation reflects the fact that there is no seat in international mediation — it provides for enforcement of international settlement agreements arising from mediation. This language differs from the New York Convention on Arbitration which provides for enforcement of foreign arbitral awards. To define an award as foreign, it must be foreign in respect to something or somewhere – and this is where the notion of a seat of arbitration is useful.

The internationality of a mediated settlement agreement is based on the relative geographies of the disputing parties, in respect to each other as well as in respect to the loci of their cause of action. Article 1 of the Singapore Convention on Mediation has established generally that it would cover international settlement agreements arising from mediation where:
(a) At least two parties to the settlement agreement have their places of business in different states; or
(b) The State in which the parties to the settlement agreement have their places of business is different from either:
(i) The State in which a substantial part of the obligations under the settlement agreement is performed; or
(ii) The State with which the subject matter of the settlement agreement is most closely connected.

The absence of the concept of a ‘seat’ in cross-border mediation underscores the fact that mediation and arbitration are fundamentally different processes. As cross-border mediation practice increases, lawyers must be mindful not to assume that principles applicable to international arbitration will automatically apply to mediation.


To make sure you do not miss out on regular updates from the Kluwer Mediation Blog, please subscribe here.

Kluwer Arbitration
This page as PDF


  1. The mere difference the author highlighted between this two conventions is that the manner of international mediation is more flexible than international arbitration, which soly leads to the outcome that there is no word “seat” in Singapore Convention. While we can see that it is also common that a international arbitratral proceeding takes place in different spots of various contries. Therefore, the changeability of places where an arbitration proceeding shouldn’t be the main reason for this action. Fom my perspective, Singapore Concention abandoned the use of “seat” exactly shows the great progress made in Singapore Convention compared to its sister convention, namely NYC. In terms of traditional localized theory, the seat of arbitration is relevant because it concerns which procedure law should be applied and which court has the right of to set aside the award. While to not adding “seat” in mediation, the convention tries to avoid the situation that a specific court holds the exclusive power to annull the settlement agreement and also the arbitration law of the seat automatically apply to the arbitration by default. This is indeed a great step in delocalization of international ADR.

  2. What is the scope for challenge to the settlement agreement in light of the Singapore Convention? Is it the position that there will be no challenge and the only space for questioning the settlement agreement will be when it is enforced or brought up in defence?
    If not, will this be determined by domestic laws? If so, will parties to the settlement agreement not have to identify the governing law of the settlement agreement (provided for in the SIngapore Convention), and the jurisidiction (since one would not want multiple actions in different states)?
    Do share views on this.

Leave a Reply

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