Commercial Arbitration


Critically evaluate the seat and delocalisation theories and the extent to which these theories have impacted on the principle of party autonomy in international commercial arbitration.




Arbitration is a private justice that is born by the will of parties to settle their disputes, usually but not, exclusively of commercial nature and out of court. Theoretically parties are free to agree to arbitrate, choose arbitrators, design procedures and select substantial laws governing the merits of their case. Finally, they could be rendered an award to decide their disputes. Arbitration as an alternative form of dispute resolution existed in domestic or national arbitration for many years. In domestic arbitration all disputes took place in the jurisdiction of a country and municipal laws naturally either governed or over saw the process of arbitration. But with the emergence of global trade more and more disputes became complex with international dimensions and conflict of state laws became apparent. Both national and international Commercial arbitration have several defining characteristics. Firstly arbitration is mostly consensual and thus the type and form of the arbitration agreement controls the process of arbitration. Secondly arbitrators do not act as judges or government agents and infect are agreed by the parties. Finally, in its nature arbitration is completely flexible, confidential and autonomous[1].

Thus the harmonisation of domestic and international arbitration laws required theoretical foundations and two very distinct approaches were formed. First it was argued that the law of the country where arbitration is to take place, lex arbitri  or the country where the arbitral award is enforced should govern commercial arbitration. This is called the ‘seat theory’. Second approach is to detach an international commercial arbitration from the control of national law of the place where arbitration proceedings are held. This is called delocalisation theory. The rationale for such an autonomous approach is that there should only be one point of control and that is the place where arbitral award is to be enforced instead of a dual control system. It is therefore argued that this way, the whole world or most of it, would be available for international commercial arbitrations and international commercial arbitration itself would be supra-national, a-national, transnational, delocalised, or even expatriate. More poetically, such arbitration would be a floating arbitration, resulting in a floating award[2].


Seat Theory and Party Autonomy

This is perhaps the most obvious and important theoretical approach in international commercial arbitration. This section is limited to the explanation of different components of seat theory as well as of its critical analysis from party autonomy point of view. As mentioned above the concept of arbitration is based upon party autonomy and that means parties can decide, by way of an agreement how they would like to resolve their dispute. However it is equally important that the commercial aspects of a dispute should never be over looked. Before a critical analysis is presented it is imperative to ascertain that seat theory is not only endorsed by UK courts but also has a support of various international jurisdictions.

English courts have implemented the concept of seat theory and there are various cases that could be looked into it.  Starting point will be comments of Steyn J. in the English case of Paul Smith Ltd v H&S International Holding Inc[3] where he explained the concept of lex arbitri as a necessary supportive role of the courts to promote arbitration. Secondly F.A. Mann[4] who is a fundamentalist supporter of seat theory noted the court of appeals decision in Naviera Amazonica Peruana SA v Compania Internacional de Seguros de Peru[5]. Mann not only supports courts decisions but also professes that in fact international arbitration is national arbitration. It could be argued that it is an orthodox yet simple view that ignores party autonomy as a foundation of international commercial arbitration. It is further argued that parties also should be free to choose the supervisory, supportive and conclusive role of national courts as parties to an arbitration agreements know in advance that by choosing the judicial seat of arbitration[6] they are agreeing to principles of lex arbitri. Therefore freedom to choose should not be limited to the exclusion of lex arbitri but should be an inclusive option of choice. Nevertheless it seems that as far as English courts are concerned, national courts will have inherent jurisdiction over arbitration agreements if the agreement stipulates England as its judicial seat of arbitration.

Various international courts have also sided with seat theory. For example in the case of Rupali Polyester Ltd[7]Malik Muhammed Qayyum J. of the Supreme Court of Pakistan supported the view that the national court should have jurisdiction over arbitral process. Furthermore in National Thermal Power Corp[8] the Indian supreme court not only endorsed to lex arbiti but also went on to say that it is an inherent jurisdiction of the Indian court to govern the arbitration agreement. Finally Chao Hick Tin J.A.'s judgment in PT Garuda Indonesia[9] adheres to seat theory by putting forward territorial arguments. However, in the context of international commercial arbitration, it is somewhat ambiguous whether courts of other jurisdictions are following the decision of English courts or have genuinely relying upon philosophical arguments of seat theory that in fact compliments their domestic legal jurisprudence.

In addition to that the seat theory is also entrenched in the UNICITRAL Model Law on International Commercial Arbitration. According to Article 1 of the Model Law and by virtue of its sovereign power, it is to supervise over the commercial arbitration[10]. Furthermore Article 8 and 9 states very clearly that court’s role is to assist and to provide support as oppose to intervene. Therefore to evaluate the seat theory and its impact on party autonomy the following four points are made.

Firstly the seat theory, although is not only enforced by national and international courts but also has been recognised by various treaties and conventions, approach has been criticised because it has limited the scope of party autonomy which is the central pillar of commercial arbitration. It has been expressed by Mance L.J[11] that because of a limited scope of seat theory the concept of party autonomy may only become a ‘myth’ if domestic courts do not stop interference in the arbitral process. However the author would argue that domestic courts do not interfere but are there to support and assist should the parties to the arbitral contract require. Most importantly it is just not enough to merely say that lex arbitri should not prevail due to the autonomous nature of the Commercial Arbitration and not present a valid and credible alternative process to deal with all arbitral anomalies and eventualities. This has also been well stated by Lord Mustills judgement in Channel Tunnel Group Ltd v Balfour Beatty Construction Ltd[12]. His lordship elaborated that the parties are perceived to have the freedom to choose the formation of the arbitral agreement any way they please but by choosing the seat of arbitration means expressly or impliedly that the lex arbitri principles apply and therefore national courts will have inherent jurisdiction to assist and to support the arbitral process. This seems a reasonable approach especially if the parties to the arbitration agreement have a dead lock over issues akin to the contract than there must be a sovereign authority that should resolve the issues by way of making legally binding decisions. Furthermore Massood Ahmed[13] commenting on Lord Mustills judgement notes that this supervisory role of the courts is not necessary for the smooth running of the actual arbitral proceeding but also parties could potentially avoid non enforcement of arbitral award.

Secondly there is a commercial argument upon which, in authors view the concept of party autonomy rests. It is argued that it is easier to forget why those parties to an arbitration agreement are in arbitration proceedings at all. In commercial terms claimant usually is claiming for damages by way of an award. Therefore if the parties know that after going though the hurdles of arbitration, considerable expense and economic loss award may not be rendered or set aside or non enforceable than they might consider commercial litigation instead. Therefore to promote the idea of International Commercial Arbitration not only the process needs to be efficient, confidential, coherent and cheaper but also have the power to enforce award.

Thirdly and most importantly the concept of party autonomy also requires justifiable limitations for the process to be credible, in expensive, quick and confidential. Redfern and Hunter[14] commented on Article 19(1) of the UNICITRAL Model Law that provides that “subject to the provision of this law, the parties are free to agree on the procedure to be followed by the arbitral tribunal in conducting the proceedings”. Furthermore where the applicable rules are agreed before the constitution of the tribunal it is unlikely that those rules could be amended without the consent of the arbitral tribunal. Therefore it seems that in practice and its application the party autonomy does have limitations that are imperative for fair and justice process[15].

Finally in the course of excising party autonomy parties must also be allowed to choose, if they may wish, to have the assistance of the national court as a safety net. Therefore it could be argued that this protection cannot be achieved if there is no ‘fall back’ position available for the parties. William Park has maintained that some control of international commercial arbitration by municipal courts is necessary in order to deal with deficiencies in the arbitral award and the arbitral process[16].


Delocalised Theory and Party Autonomy

In essence the delocalisation theory maintains that Commercial Arbitration should be free from the constrains of the national laws and therefore should not be influenced or governed in any shape or form by those laws but only for the purpose of enforcement of arbitral awards. There are number of cases and treaties and conventions that have illustrated the delocalised view but for the purpose of this essay three main cases shall be critically analysed. The section shall also comment on the view point of other notable authors and there learned views.

Jan Paulsson[17], a renowned supporter of delocalised theory explains the concept that “The sometimes-used expression ‘floating arbitration’ is not entirely satisfactory, because all arbitral awards may, and frequently do, float”. Delocalised theory found its strength in the flowing three cases. In Gotaverken Arendal AB v Libyan General National Maritime Transport Co[18] claimants enforced the arbitration agreement and sought the defendant to take delivery of three oil tankers. Defendants refused to take the delivery and contended that claimant has breached the arbitration agreement. The arbitral tribunal, under the rules of International Chamber of Commerce (ICC) decided against the defendant. Defendant appealed to the Paris court to have the award set aside on the grounds of tribunal not providing sufficient reasons for the award. The Paris court rejected defendant’s arguments and maintained that due to the ‘international’ nature of the award it does not have jurisdiction. Claimant eventually enforced its tribunal award in Sweden. Massod Ahemed explains that the heart of the court of appeal’s decision was “The place of the arbitral proceedings, chosen only in order to assure their neutrality, is not significant; it may not be considered an implicit expression of the parties' intent to subject themselves, even subsidiary, to the ‘loi procedural francaise”. This case was particularly encouraging for the supporters of delocalised theory but they failed to appreciate the uncertainty that it spread in the arbitration world. If this case did not create enough sensation, confusion and uncertainty than in Hilmarton Ltd v Omnium de Traitement et de Valorisation[19] the Cour de Cassation recognised an arbitral award that was set aside by the Swiss court on the grounds of public policy. The court held “the award is not integrated in the legal system of that State, so that it remains in existence even if set aside and its recognition in France is not contrary to international public policy.” This case was very significant as it brought out the real impractical nature of the delocalised theory. Most importantly international arbitration perhaps could deal with the floating awards but drifting awards are degrading the very integrity of the arbitration process. Further in PT Putrabali Adyamulia v Rena Holding[20] where the parties to an arbitration agreement started proceeding in London as their chosen seat of arbitration. The Tribunal made an award in favour of the claimant and defendant appealed to the high court under the Arbitration Act 1996 s (69)[21] and high court referred the matter back to the tribunal which issued a second award in favour of the defendant. Claimant made an application to the French court for the recognition and enforcement of their award and it, following its earlier decisions, recognised the award. The Cour de Cassation held “an international arbitral award, which is not linked to any national legal order, is a decision of international justice, whose validity is reviewed in accordance with the rules applicable in the country where its recognition and enforcement is sought.”

It appears that the delocalisation argument is not only circular but also presents conflicts within. To elaborate this notion Masood Ahmed states in his well balance and meticulous article thatwhy should states not be given the right to govern the arbitral process before the award is rendered in another state? In the interest of justice, enforcement should not be allowed where an award has already been set aside or annulled at the seat of arbitration this would clearly lead to unfairness to the parties and will have the undesired effect of undermining international commercial arbitration”. The reality is that due to those French decisions the concept of party autonomy perhaps needs to be revised in the context of delocalisation. Furthermore it is argued that theorists have to also focus on the objectives of arbitration and those are that commercial entities require, cheap, private, efficient, reliable, consistent, protective and institutional justice that could never shy away to assist or to enforce awards rendered instead of making them disputed[22].

Further the delocalists always point toward the New York Convention 1958 as the convention that supports their views. It is considered one of the successful conventions if not the only convention in recognition and enforcement of arbitral awards. The delocalists argue that the states those are party to this convention have not only the duty to recognise the arbitral award but also have to recognise and enforce the award. Therefore this notion points out the fact that states must recognise the award regardless of where it was rendered. However writer would pose this question as to how could a sovereign court’s power be restricted to only recognising and enforcing the arbitral awards without accepting court’s jurisdiction to supervising the arbitral award?

The delocalised arbitration is further criticised as being “wholly unrealistic” because it attempted to put international commercial arbitration in a legal vacuum. For example a contract, this is derived from the parties’ autonomy but is a necessarily subject to local contract law and other relevant laws. The autonomy of the contractual parties is not absolute and is controlled by mandatory rules and public policy of a state. The same rationale also applies to national as well as international arbitration[23].

Finally it seems that as Lainé has argued that it is a fact that no arbitration can legally exist or be performed without the parties' agreement. However, he does not accept that the parties' arbitration agreement constitutes or stipulates the ‘jurisdiction of the arbitration’. It is also a misconception that arbitrators are judges, in reality the judges assert their power directly from the sovereign whereas the arbitrators derive their authority from the sovereign but their nomination is a matter for the parties[24]. Similar point has also been made by Pillet who stated in relation to the arguments that arbitration agreement is irrelevant once the arbitrator has been appointed “The arbitration agreement is necessary to give the arbitrators their authority, but once that authority has been conferred on them, provided they keep within the limits of the task given to them, their freedom is absolute and the arbitration agreement has no influence on their award which is based on quite different matters.”[25] . Therefore from the party autonomy’s point of view delocalised approach or not, once arbitration is structured on the jurisdictional framework, parties are allowed to rely on party autonomy to decide how arbitration procedures shall be followed. However, the scope of autonomy is only recognised within the jurisdictional framework[26].



Party autonomy is the central pillar of international commercial arbitration but has obvious limitations and cannot be absolute. Because if the concept of party autonomy is stretched to the point of an absolute private law and justice mechanism than it will have two immediate effects. Firstly expressly or impliedly the concept of sovereignty from which all state and court powers are derived, shall by default, be limiting itself within its own jurisdiction. This will represent a serious conflict of legal jurisprudence theories. Secondly who will be observing the implied conditions necessary to exercises ‘freedom of contract’ and if there is a un resolvable issue between the parties how could it be settled justly without the assistance of the courts.

However it is obvious from the case law that it is French and English courts that are divided in their decisions. It seems that French courts prefer the delocalised approach whereby English courts have sided with the seat theory. In the mist of this confusion the argument of commerciality is being ignored and it is the international business entities that are the real corner stone of Commercial arbitration and party autonomy, because without the arbitration agreement, international commercial arbitration will simply not exist. The most important objective of any arbitration proceeding is that after business relationship between the parties has been broken down and claimant want a just and equitable award for its grievance. Therefore the above arguments presented by the seat theory and delocalisation theory could be summarised that in practice seat theory is more likely to produce a coherent and enforceable award than the delocalised theory. Nevertheless confusion and inconsistency in arbitral awards and proceedings shall remain until the match between those theories is not decided and the principles are accepted by the courts of major countries, such as France and England.

Perhaps a point of contention could be the hybrid theory that allows the autonomous approach of the parties but limiting only to the assistance of the courts for the enforcement of an arbitral award. Briefly, it has the right components of delocalised and seat theories.




[1]   International Commercial Arbitration, Commentary and Materials, Gary B. Born, 2001 Transnational Publisher INC

[2]  See in particular Lew, Achieving the Dream: Autonomous Arbitration (2006) 22 Arb Intl 178, at 202; Fouchard, L'Arbitrage Commercial International (Litec, 1965), 2227; Paulsson, Arbitration Unbound: Award Detached from the Law of its Country of Origin (1981) 30 ICLQ 358; and Paulsson, Delocalisation of International Commercial Arbitration: When and Why it Matters (1983) 32 ICLQ 53. For a continuation of the debate see Nakamura, The Place of ArbitrationIts Fictitious Nature and Lex arbitri 15(10) Mealey's Intl Arb Rep 2329 (October 2000); Rubins, The Arbitral Seat is No Fiction: A Brief Reply to Tatsuya Nakamura's Commentary, The Place of ArbitrationIts Fictitious Nature and Lex arbitri 16(1) Mealey's Intl Arb Rep 2328 (January 2001); Pinsolle, Parties to An International Arbitration With the Seat in France are at Full Liberty to Organise the Procedure as They See Fit: A Reply to the Article By Noah Rubins 16(3) Mealey's Intl Arb Rep 30 (2001); and Nakamura, The Fictitious Nature of the Place of Arbitrationa May not Be Denied 16(5) Mealey's Intl Arb Rep (2001).

[3]Paul Smith Ltd v H&S International Holding Inc [1991] 2 Lloyd's Rep. 127 QBD (Comm).

[4] F.A. Mann, “Lex Facit Arbitrum” (1967) International Arbitration 160.

[5] Naviera Amazonica Peruana SA v Compania Internacional de Seguros de Peru [1988] 1 Lloyd's Rep. 116 CA.

[6] The Arbitration Act 1996 s.3 provides a statutory definition of the seat of the arbitration: “In this Part ‘the seat of the arbitration’ means the juridical seat of the arbitration designated--(a) by the parties to the arbitration agreement, or (b) by any arbitral or other institution or person vested by the parties with powers in that regard, or (c) by the arbitral tribunal if so authorised by the parties, or determined, in the absence of any such designation, having regard to the parties' agreement and all the relevant circumstances.” Professor Goode has explained the seat of the arbitration as: “the jurisdictional seat designated by the parties to the arbitration agreement or by any arbitral or other institution or person vested by the parties with powers in that regard or by the arbitral tribunal if so authorised by the parties. The judicial seat is the place whose law is intended by the parties to govern the arbitral proceedings, not the place where the proceedings are actually held or the award given” (R. Goode, Commercial Law, 3rd edn (London: Penguin, 2004), at p.1170). Also, in Union of India v McDonnell Douglas Corp [1993] 2 Lloyd's Rep. 48 the court held: “Although the choice of a ‘seat’ also indicates the geographical place for the arbitration, this does not mean that the parties have limited themselves to that place. As is pointed out in a passage approved by the Court of Appeal in Naviera Amazonica Peruana SA v Compania Internacional de Seguros de Peru … it may often be convenient to hold meetings or even hearings in other countries. This does not mean that the ‘seat’ of the arbitration changes with each change of country. The legal place of the arbitration remains the same even if the physical place changes from time to time, unless of course the parties agree to change it.” This understanding of the seat of arbitration can also be taken from the comments of the Supreme Court of Singapore in Swift-Fortune Ltd v Magnifica Marine SA [2007] 4 L.R.C. 819; [2006] SGCA 42 in which Chan Sek Keong C.J. stated: “The choice of venue is dictated by diverse factors and the stipulated seat of arbitration may not ultimately be the chosen venue. PT Garuda Indonesia v Birgen Air [2002] 5 LRC 560 is an example of a case where the seat of arbitration was Jakarta, but the arbitration itself was conducted in Singapore.” (See [2002] 1 S.L.R. 393 (CA); [2001] SGHC 262.)

[7] Rupali Polyester Ltd v Bunni [1995] 3 L.R.C. 617.

[8] National Thermal Power Corp v Singer Co [1992] 2 Com. L.J. 256; [1992] INSC 146.

[9] PT Garuda Indonesia v Birgen Air [2002] 1 S.L.R. 393 (CA); [2001] SGHC 262.

[10] “(1) A court before which an action is brought in a matter which is the subject of an arbitration agreement shall, if a party so requests not later than when submitting his first statement on the substance of the dispute, refer the parties to arbitration unless it finds that the agreement is null and void, inoperative or incapable of being performed.

(2) Where an action referred to in paragraph (1) of this article has been brought, arbitral proceedings may nevertheless be commenced or continued, and an award may be made, while the issue is pending before the court.

[11] Department of Economic Policy and Development of the City of Moscow v Bankers Trust Co [2004] EWCA Civ 314; [2005] Q.B. 207 at 314. Further, the English Arbitration Act 1996 Pt I s.1 (General principles) reflects the principle of party autonomy albeit with some limitations. Section 1(b) provides: “the parties should be free to agree how their disputes are resolved, subject only to such safeguards as are necessary in the public interest”

[12] Channel Tunnel Group Ltd v Balfour Beatty Construction Ltd [1993] 2 W.L.R. 262.

[13] A. Massood, “The influence of the delocalisation and seat theories upon judicial attitudes towards international commercial arbitration”  2011Arbitration

[14] Redfern and Hunter, with Blackby and Parasides, Law and Practice of International Commercial Arbitration, 4th Edition, 2004 at p315

[15] P Michael ‘Limits to Party Autonomy in Arbitral Procedure’ President, Australian Centre for International Commercial Arbitration;

[16] W. Park, “The Lex Loci Arbitri and International Commercial Arbitration” (1983) 32 I.C.L.Q. 21.

[17] J. Paulsson, “Arbitration Unbound: Award Detached From the Law of Its Country of Origin” (1981) 30 I.C.L.Q. 358

[18] Gotaverken Arendal AB v Libyan General National Maritime Transport Co Cour d'Appel de Paris (Feb. 21, 1980), reprinted in (1980) J. Dr. Int. 660 and [1980] Revue de l'Arbitrage 524.

[19] (1997) XXII Ybk Comm. Arbn 696 (Cour de cassation, June 10, 1997).

[20] CA Paris, Société PT Putrabali Adyamulia v SA Rena Holding, C. cass, 1ère civ, June 29, 2007. See also M. Haravon, “Enforcement of Annulled Foreign Arbitral Awards: The French Supreme Court Confirms The Hilmarton Trend” (2007) 22 Mealey's International Arbitration Report ; and G. Cuniberti, “The French Like It Delocalised: Lex Non Facit Arbitrum”, available at Conflicts of [Accessed September 9, 2011].

[21] Section 69(1) provides: “Unless otherwise agreed by the parties, a party to arbitral proceedings may (upon notice to the other parties and to the tribunal) appeal to the court on a question of law arising out of an award made in the proceedings.”

[22] Goode, “The Role of the Lex Loci Arbitri in International Commercial Arbitration” (2001) 17 Arbitration Int. 19.

[23] A. Tweeddale and K. Tweeddale, Arbitration of Commercial Disputes, International and English Law and Practice (Oxford: OUP, 2007), para.7.78.

[24] J-P. Niboyet, Traité de Droit International Privé Français, para.1985, 137; cited from J. Lew, Applicable Laws in International Commercial Arbitration (The Hague: Kluwer Law International, 1978), p.53.

[25] A. Pillet, Traité Pratique de Droit International Privé (Grenoble: Imp. Allier, 1923), vol.2 at p.537; translation from A. Samuel, Jurisdictional Problems in International Commercial Arbitration. A Study of Belgian, Dutch, English, French, Swedish, USA and West German Law (Zurich: 1989), 52.

[26] Hong-  Lin Yu “How far can party autonomy be stretched in setting the grounds for the refusal of arbitral awards” International Arbitration Law Review 2011.