How Should One Define the Principles of ‘Kompetenz-Kompetenz’

1. Introduction

The concepts of separability and “Kompetenz-Kompetenz” are among the most significant in the field of international arbitration. The reason for their importance is that those are the concepts on which basis the jurisdiction of the arbitrators is defined and can work on a practical level. In fact, without them, the practical possibility of the arbitral proceeding itself, would be at risk and arbitration would cease to be the favourite way of solving disputes among the international business community.

It could even be said, on the one hand, that without the principle of separability, or the autonomy of the arbitration agreement from the main contract in which it is contained, there might not be any arbitration proceedings at all. In the absence of the concept of “Kompetenz-Kompetenz”, on the other hand, the arbitrators would not be able to rule, at least in the first instance, on their own jurisdiction – every time that an issue arises in this respect – and thus it would be very easy for the party which would like to avoid the arbitration proceeding to go to the national courts.

These concepts are often analysed in conjunction with each other, because they are both related to the scope and the form of the arbitrators’ jurisdiction. The aim is to give to the arbitration proceeding the maximum possible effect [1] .

While the separability principle allows the very existence of the arbitral proceeding, preserving the arbitration clause and thus making it possible to resolve the dispute even when the main contract is invalid, the aim of the “Kompetenz-Kompetenz” principle is to avoid paralyzing the proceeding when the competence of the arbitral tribunal is questioned by one of the parties.

It is essential, however, to remember that they do not have the same meaning nor do they have equal recognition all over the world.

The following example could help in order to understand the differences between the concepts:

The principle of separability of the arbitral agreement from the main contract allows the arbitrators to decide the dispute even when the main contract is null and void, assuming that the invalidity does not affect the arbitral agreement itself. However, when the relevant point is, for instance, related to the validity of the signature on the arbitration agreement and consequently with the competence of the arbitral tribunal itself, it is the “Kompetenz-Kompetenz” concept – and not the separability principle – that allows the arbitrator to first rule on their own jurisdiction.

The two concepts differ in their theoretical basis; the separability principle is based on a fictitious construct which is nevertheless essential for the arbitral proceeding to work and exist.

It would not make sense to consider the arbitration agreement as really separated from the main agreement (when the former is not formally a distinct agreement) of which it forms part of and, in fact, broadly speaking, no party signs a commercial contract thinking that he is signing two different agreements.

As has been said [2] , it is not inconceivable that two different agreements are contained in the same contract and there could well exist a single instrumentum that contains two negotia, as usually happens in the case, for example, of a sale agreement and a loan agreement. In the case of the arbitral agreement, however, the circumstances are different, because one of the two negotia is related to the resolution of the disputes that could arise out of the contract and hence can be seen as a procedural – or jurisdictional – clause to be treated differently from one that defines rights and obligations in the context, for example, of a sale agreement.

As for the “Kompetenz-Kompetenz” concept, this can be seen as a concession of the power to rule on jurisdiction, made by national legal systems to the arbitral tribunals. The aim is to prevent proceedings being stayed or delayed – often even when the issue related to the jurisdiction is before a national court – wherever the jurisdiction is allegedly lacking. In general terms, however, this power to rule granted to the arbitral tribunals is limited to the first review of the issue, and could later be challenged before a Court. As stated above, the “Kompetenz-Kompetenz” rule, unlike the separability principle, is less widely recognized by different legal systems. France, the U.S., England and other countries give this concept slightly different meanings and there are countries, i.e. China, that do not even recognize the existence of this rule and have enacted statutes which make its future acceptance unlikely.

2. The Separability Principle

The autonomy of the arbitration agreement has been defined as a “fundamental legal principle” [3] in the field of international commercial arbitration.

The separability concept allows the arbitration clause to be independent to the greatest possible extent of the fate of the main contract of which it forms part [4] and from the law applicable to the main contract itself.

Theoretically, this is extremely meaningful because, as has been authoritatively stated, even if “the purposes of the contract have failed […] the arbitration clause is not one of the purposes of the contract” [5] . However, the most important outcome of the separability rule is to be found in its practical applications: it is its presence that often makes possible the continuation of the arbitral proceedings.

2.1. Separability, Institutional Rules, International Treaties and Domestic Statutes

The paramount importance of the separability principle in the field of international commercial arbitration, although it is not expressly stated in major treaties on international arbitration, (e.g. the 1958 New York Convention) is recognized by almost all the institutional arbitration rules [6] .

In fact, it can be found in the UNCITRAL Rules art. 23 (as revised in 2010), in art. 16 of the Model Law, in art. 23 of the LCIA art. 23.1, in Rules 6(4) ICC in art. 36 of the WIPO Arbitration Rules and in art. 5 (4) of the CIETAC Rules.

Thus, when the parties to an arbitration agreement make reference to one of these sets of rules in their contract, the separability principle is included without the need to make a specific reference to it.

As for domestic legislation, the principle is recognized in the arbitration statutes of many countries, which have either adopted the UNCITRAL Model Law or included the separability principle in their legislation.

For instance, Sec. 7 of the 1996 English Arbitration Act clearly states that “Unless otherwise agreed by the parties, an arbitration agreement which forms or was intended to form part of another agreement (whether or not in writing) shall not be regarded as invalid, non-existent or ineffective, because that other agreement is invalid, or did not come into existence or has become ineffective, and it shall for that purpose be treated as a distinct agreement.”

In other systems of law, the separability concept is also known as the “autonomy of the arbitration clause”; this is what it is called in France and Italy (where the principle is recognized in art. 808 of the Code of Civil Procedure – as reformed in 2006).

In France, although the principle was not recognized in the 1804 Napoleonic Code, which outlawed the enforcement of arbitral clauses, it is now widely accepted. Many authors suggest that France is in fact the country in which the autonomy of the arbitration agreement is strongest (as will be discussed later) [7] .

In Switzerland, after the decision of the Supreme Court in 1931, which declared that the invalidity of the main contract could not affect the arbitration clause [8] , the principle was formally accepted by the 1987 Swiss Private International Law Act (PILA). Art. 178 paragraph 3 PILA reads as follows: “the validity of an arbitration agreement cannot be contested on the ground that the main contract may not be valid”.

In the U.S., the 1925 Federal Arbitration Act makes no reference to the separability principle; nevertheless it has been increasingly recognized in American case law, as will be discussed in more details later.

To sum up, the principle has reached a high degree of recognition in almost all countries of the principle in recent decades. Even Latin American countries, that historically have been suspicious of it, have now changed their statutes in order to adapt them to the separability principle, and only some Arab countries (i.e. Saudi Arabia and Kuwait) seem not to recognize it. Nevertheless, the trend is towards a full recognition of the autonomy of the arbitration agreement, and it is to be be expected that this will be reached in the next decade. [9]

Moreover, it is worth mentioning that the importance of the principle has been recognized in the jurisprudence of the European Court of Human Rights. In the decision Stran Greek Refineries and Stratis Andreadis v Greece, it was held that “the unilateral termination of a contract does not take effect in relation to certain essential clauses of the contract, such as the arbitration clause”. [10]

2.2. Origin and evolution of the separability principle

Although the 1930 Lena Goldfield v USSR case is often cited as one of the most important decision related to the applicability of the separability principle, it has been observed that actually this decision does not offer any historical support for the autonomy doctrine. In fact, in that dispute “The arbitration clause under Soviet law was a special law and no mere civil procedural contract capable of unilateral dissolution; the Soviet Union was not actually impugning the validity of the arbitration clause” [11]

2.2.1. United Kingdom

Most likely, the first statute which recognized the separability principle was the 1698 English Arbitration Act. This statute granted the arbitration agreement the status of a rule of court, whenever the parties so agreed. However, the subsequent case law reversed this initial development; in the 1746 Kill v. Hollister case, it was held that arbitration agreements which were not covered by the 1698 Arbitration Act, were not to be recognized, because they oust the jurisdiction of the court [12] .

During the 1st half of the 20th century the idea of the independence of the arbitration agreement from the main contract began to be recognized. It was held that the arbitration clause could be governed by a different law from that applicable to the main agreement, but it was not fully autonomous, because, should the latter fall, the arbitration agreement would have fallen as well.

The evolution of the principle in the English system was much less problematic later on: since the 1942 Heyman v Darwins case, the discharge of the main contract because of breach of performance obligations was deemed to have no effect on the arbitral clause [13] .

However, the full acceptance of the principle of separability was reached only in 1992 in the case Harbour Assurance Co. (UK) v. Kansa Gen. Int’l Ins. Co. [14]

The rationale underlying the decision is “the imperative of giving effect to the wishes of the parties unless there are compelling reasons of principle why it is not possible to do so”. Moreover, the threat posed by the lack of recognition of the principle was clearly identified by the House of Lords, which stated that “if the arbitration clause is not held to survive the invalidity of the contract, a party is afforded the opportunity to evade this obligation to arbitrate by the simple expedient of alleging that the contract is void.”

The importance of this decision, however, lies in the fact that – for the first time – the arbitration agreement was deemed to be valid and effective even when the main contract was void ab initio, as long as the invalidity was not related to the arbitration clause itself [15] .

2.2.2. The U.S. approach

The evolution of international commercial arbitration in the U.S., has been sped up by the landmark decision Prima Paint Corp. v. Flood & Conklin Mfg. Co. [16] in which the doctrine of separability was recognized for the first time [17] . In that decision was held that“arbitration clauses are“separable” form the contract in which they are embedded, and that were no claim is made that fraud was directed to the arbitration clause itself, a broad arbitration clause will be held to encompass arbitration of the claim that the contract itself was induced by fraud”

However, the Prima Paint doctrine was later interpreted in a restricted way [18] . In the decision Three Valleys Municipal Water District v. E.F. Hutton & Co. it was held that the applicability of the separability doctrine was “limited to challenges seeking to avoid or rescind a contract – not to challenges going to the very existence of a contract that a party claims never to have agreed to” […] “a party who contests the making of a contract containing an arbitration provision cannot be compelled to arbitrate the threshold issue of the existence of an agreement to arbitrate. Only a court can make that decision”.

The distinction between void and voidable contracts in relation to the autonomy of the arbitration agreement was later dismissed by the decision Buckeye Check Cashing Inc. v John Cardegna, et. Al. [19] In this case, involving the alleged illegality of the main contract, it was held that “unless the challenge is to the arbitration clause itself, the issue of the contract’s validity is considered by the arbitrator in the first instance”.

2.2.3 The French approach

The development of the French approach to “l’autonomie de la clause compromissoire” begins in 1963 with the Gosset v Carapelli [20] decision. In that case, it was stated that “The arbitration agreement, whether concluded separately or included in the contract to which it relates, shall […] have full legal autonomy and shall not be affected by the fact that the aforementioned contract may be invalid” [21] .

Later, the concept of autonomy of the arbitration clause, acquired its status as a fundamental legal principle in the case Comité populaire de la municipalité de Khoms El Mergheb v Dalico Contractor (the Dalico case) [22] . In this decision it was held that “pursuant to a substantive rule of international arbitration law, the arbitration agreement is legally independent of the main contract which incorporates it either directly or by reference and […] its existence and validity are to be appreciated subject to the mandatory rules of French law and international public policy, based on the mutual intent of the parties, without a need for a reference to any national law.”

In the Dalico decision, the French Supreme Court laid down a new approach to the issue of separability; the so-called “principle of validity”. By this principle, in order to judge the validity of the arbitration clause, a substantive rules method is applied: no reference is made to the law governing an arbitration agreement, which is valid and effective as far as it is not contrary to French law and international public policy.

2.3 Limits and Difficulties Related to the Separability Principle

The problems and difficulties that could arise with reference to the separability principle are twofold. The first is related to the term used, and the second to some critical flaws in the main contract that could affect the arbitration clause, e.g. the illegality of the main contract.

2.3.1. Separability, autonomy or severability?

Although many authors make reference to the separability or autonomy of the arbitration agreement, it has been noticed that this terminology could lead to ambiguities and mistakes that would be solved using a slightly different word; severability [23] .

The reasons to reject the use of the term “autonomy” derive mainly from the fact that the arbitration agreement cannot actually be separated from the main contract because “one cannot provide for arbitration in a vacuum” [24] . This means that the arbitration clause cannot exist without the rest of the agreement, since it represents only one of its terms, even if its procedural nature makes it clearly different from the other terms in the agreement, which are commercial in nature.

One of the most problematic aspects of the “autonomy” concept is that it could entail the idea that an arbitration clause is a separate agreement and, hence, that there are two separate agreements, one related to the commercial terms and the other to the agreement to arbitrate. If this were the case, however, that could lead to the conclusion that unless the arbitration clause is accepted separately, it is not binding.

Another problematic aspect of the terminology, which will be examined in more detail later, is that should the arbitration agreement be considered fully autonomous from the contract of which it forms part of, it could not be transferred along with it. Hence, every time the main agreement is assigned or there is subrogation, the arbitration clause might not form part of the assignment or subrogation. Clearly, this is not a satisfactory result, and this is the reason why it would be better to speak of severability rather than separability or autonomy [25] .

However, since “autonomy” and “separability” are expressions much more widely used in doctrine and jurisprudence, they will be utilized in this analysis, with the caveat that the meaning of these words is to be understood in a way more similar to that of severability, (Where something is ‘severable’ if it is capable of being divided into legally independent rights or obligations) rather than something that really is completely autonomous or separate.

2.3.2 Illegality of the main contract: consequences on the arbitration agreement

In some countries, difficulties with regard to the autonomy principle have arisen in relation to some serious flaws of the main contract; namely, its illegality. For a long time it was held that the arbitral tribunal could never judge the validity of the contract. Wherever the contract was deemed invalid, so was the arbitration clause. In Overseas Union Insurance Ltd v AA Mutual International Insurance Co Ltd [26] , it was said that this rule “owes as much to logic as it does to authority”.

This was the case which the House of Lords had to face in the 2007 case Fiona Trust & Holding Corp. v Privalov [27] . The case was related to a charter party contract, containing an arbitration agreement, which was later rescinded for alleged bribery.

Their Lordships, following the trend inaugurated in the Harbour v Kansa case, held that “An arbitration clause that makes reference to “any dispute arising under this charter” is wide enough to cover any disputes relating to the existence or legality of the contract” i.e. even disputes related to the illegality of the main agreement. However, the House of Lords went on and clarified when an arbitration agreement could be held invalid: “Of course there may be cases in which the ground upon which the main agreement is invalid is identical with the ground upon which the arbitration agreement is invalid. For example, if the main agreement and the arbitration agreement are contained in the same document and one of the parties claims that he never agreed to anything in the document and that his signature was forged, that will be an attack on the validity of the arbitration agreement. But the ground of attack is not that the main agreement was invalid. It is that the signature to the arbitration agreement, as a “distinct agreement”, was forged. Similarly, if a party alleges that someone who purported to sign as agent on his behalf had no authority whatever to conclude any agreement on his behalf, that is an attack on both the main agreement and the arbitration agreement”

2.3.3. Assignment of the main contract: which consequences for the arbitration agreement?

One of the difficulties that could arise with regard to the application of the separability principle, and one that could threaten its very applicability, at least on a theoretical level, is related to the assignment of the main contract. Should the arbitration clause be deemed to be a completely separate agreement, one could infer that it does not follow from the fate of the main contract whether the arbitration agreement is assigned.

Institutional rules and international conventions say nothing with regard to this possibility, and the different legal systems do not have a homogeneous view of the problem.

In the U.S. the trend is to consider that the assignment of the arbitration clause with the main contract requires the consent of the parties, and the same is the prevailing view in Italy (in which, unlike other civil law countries, the express consent of the assignee is deemed to be necessary in order to validly assign the arbitration agreement along with the main contract). [28]

The French approach is probably the most problematic on this point, since on the one hand it holds that the arbitration clause is completely and fully autonomous and, on the other hand, its circulation (and therefore its assignment) is closely related to that of the main contract.

In order to justify and reconcile these conflicting views, the French Cour de Cassation, stated that “an international arbitration agreement, legally independent of the main contract, circulates with it regardless of the validity of the assignment of rights under the main contract”. [29] After this decision, the degree of autonomy of the arbitration clause was at its greatest. The French Supreme Court reaffirmed not only the autonomy of the existence of the arbitration clause from the main contract, but also the independence of its circulation.

3. The “Kompetenz-Kompetenz” principle

When it comes to the “Kompetenz-Kompetenz” principle, confusion and misunderstanding may arise by the very choice of that term. The German meaning of this expression is to some extent different from that which has become widely accepted in the field of international commercial arbitration.

In its literal German sense, the term “Kompetenz-Kompetenz” entails that arbitrators, and they alone, are authorized to rule on their own jurisdiction [30] .

However, the traditional meaning of this expression is not the same as that which is acknowledged in arbitral proceedings [31] , and this is why some authors prefer the term “competence-competence”, which does not have ambiguous implications [32] .

Thus, it is appropriate to consider the “Kompetenz-Kompetenz” principle as a “rule of chronological priority [33] ” that allows the arbitrators to rule first on their own jurisdiction.

There are two aspects of the “Kompetenz-Kompetenz” principle [34] : the first one is related to the ability of the arbitrators to rule on their own jurisdiction, and the second, and more problematic one, is related to the fact that national courts can rule on the arbitral tribunal’s jurisdiction only after the arbitrators has done so. This second characteristic of the principle, however, is less accepted in national systems, and, as will be illustrated in details later, has given rise to several problems, especially in the U.S.

3.1. The “Kompetenz-Kompetenz” principle: international, institutional rules and domestic statutes

The recognition of the “Kompetenz-Kompetenz” principle is quite widespread, even if, unlike the separability principle, it is not universally accepted and the extent of its implementation at the national level is still diverse [35] .

With regard to international conventions, the principle is recognized in art. VI (3) of the 1961 European Convention on International Commercial Arbitration (the “Geneva Convention”) and in the 1958 Convention on the Recognition and Enforcement of Foreign Arbitral Award (the “New York Convention”). Article II (3) of the New York Convention reads as follows: “The Courts of the Contracting State, when seized of an action in a matter in respect of which the parties have made an agreement within the meaning of this article, shall, at the request of one of the parties, refer the parties to arbitration, unless it finds that the said agreement is null and void, inoperative or incapable of being performed” [36] .

The principle is also recognised in the rules of the major arbitral institutions, such as Art. 21 (1) of the UNCITRAL Arbitration Rules, Art. 6 (2) of the ICC Arbitration Rules, Art. 23 of the LCIA Rules. It is also recognised in Art. 16, of the Model Law which, significantly, is called “Competence of arbitral tribunal to rule on its jurisdiction”, and clearly states that:

“(1) The arbitral tribunal may rule on its own jurisdiction, including any objections with respect to the existence or validity of the arbitration agreement. For that purpose, an arbitration clause which forms part of a contract shall be treated as an agreement independent of the other terms of the contract. A decision by the arbitral tribunal that the contract is null and void shall not entail ipso jure the invalidity of the arbitration clause.

(2) A plea that the arbitral tribunal does not have jurisdiction shall be raised not later than the submission of the statement of defence. A party is not precluded from raising such a plea by the fact that he has appointed, or participated in the appointment of, an arbitrator. A plea that the arbitral tribunal is exceeding the scope of its authority shall be raised as soon as the matter alleged to be beyond the scope of its authority is raised during the arbitral proceedings. The arbitral tribunal may, in either case, admit a later plea if it considers the delay justified. The arbitral tribunal may rule on a plea referred to in paragraph (2) of this article either as a preliminary question or in an award on the merits. If the arbitral tribunal rules as a preliminary question that it has jurisdiction, any party may request, within thirty days after having received notice of that ruling, the court specified in article 6 to decide the matter, which decision shall be subject to no appeal; while such a request is pending, the arbitral tribunal may continue the arbitral proceedings and make an award”

Until 2005, the only major exception to the recognition of the “Kompetenz-Kompetenz” principle was China. It was not enshrined in the 1994 Arbitration Law of the People’s Republic of China and only the People’s Court or the CIETAC itself were allowed to rule on the jurisdiction of the arbitral tribunal [37] .

However, with the introduction in 2005 of the new CIETAC Rules, a restricted form of the “Kompetenz-Kompetenz” rule was introduced. While CIETAC retains the power to rule on issues related to the arbitral tribunal’s jurisdiction, an option to delegate such power to the arbitral tribunal itself is now provided for. However, the People’s Court preserved its power to rule on jurisdictional issues, and is not bound by the decision made by the arbitration commission [38] .

Although the new CIETA