DHL Project & Chartering Limited v Gemini Ocean Shipping Co Limited

It is convenient to begin by considering the effect of the “subject” in this case. This topic was considered in some detail by Mr Justice Foxton in The Leonidas [2020] EWHC 1986 (Comm), [2021] 2 Lloyd’s Rep 165 in a judgment which Mr Holroyd described as “a tour de force”, a description which I would endorse. In that case the parties reached an agreement for a voyage charter which was stated to be “on subjects”. There were four such subjects, three of which were subsequently lifted by the Charterer in exchange for a reduction in the demurrage rate which had been negotiated. The one remaining subject was “Suppliers’ Approval”. Mr Justice Foxton drew a distinction between what he called a “pre-condition”, which prevents a binding contract coming into existence, and a “performance condition”, which has the effect that performance does not have to be rendered if the subject is not satisfied for reasons other than a breach of contract by one of the parties. He said that:

“52.

While each case will depend on its own individual facts and commercial context, it is clear that a ‘subject’ is more likely to be classified as a pre-condition rather than a performance condition if the fulfilment of the subject involves the exercise of a personal or commercial judgment by one of the putative contracting parties (e.g. as to whether that party is satisfied with the outcome of a survey or as to the terms on which it wishes to contract with any third party).

53.

While these general principles apply to contracts whether they pertain to the domain of land rats or water rats, there is a particular feature of negotiations for the conclusion of contracts for the employment of ships which should be noted. When the main terms for a charterparty have been agreed but the parties have yet to enter into contractual relations, this is generally referred to by shipowners, charterers and chartering brokers as an agreement on ‘subjects’ or ‘subs’, an expression which signals that there are pre-conditions to contract which remain outstanding. The conclusion of a binding contract in respect of such an agreement is seen as dependent on the agreement of the relevant party or parties to ‘lift’ (i.e. remove) the subjects.”

This is borne out by the leading textbooks. Thus Carver on Charterparties, 2^nd Ed (2021), states at para 2-031 that:

“The parties may agree the terms of a charterparty and one such term may be a condition precedent that unless and until the condition precedent is satisfied, no binding contract comes into being. In charterparty negotiations, such conditions precedent are often referred to as ‘subjects’ and the satisfaction of those conditions precedents is referred to as the ‘lifting of the subjects’.”

To the same effect, Time Charters, 7^th Ed (2014), states at para 1.11:

“In practice, parties very often indicate that they do not yet intend to make a binding contract by saying that their agreement is ‘subject to’ conditions. To say an agreement is ‘on subjects’ means that it is not binding until the ‘subjects’ in question have been ‘lifted’. Generally, only when all subjects are lifted does an agreement become a binding contract. At that point the ship is ‘fully fixed’.”

Examples of terms which have been held to negative contractual intent include “subject to details” (The Junior K [1988] 2 Lloyd’s Rep 583), “subject to board approval” (The Palladium [2018] EWHC 1056 (Comm)) and “subject to stem” (i.e. the availability of a cargo) (Kokusai Kisen Kabushiki Kaisha v Johnson (1921) 8 Ll LR 434). The use of a “subject” in the context of charterparty negotiations is therefore well known as a device to ensure that a binding contract is not yet concluded, just as is the case with the term “subject to contract” in other contexts.

Although an agreement “on subjects” leaves either party free to withdraw until the subject is “lifted”, such an agreement nevertheless serves a useful commercial purpose. It provides a convenient summary of the state of the parties’ negotiations; it identifies what remains to be done in order for the agreement to become binding; and it provides commercial (albeit not legal) pressure on the parties not to withdraw for reasons unrelated to the subject in question, such as a change in the market.

That being the usual effect of an agreement on subjects, Mr Justice Foxton in The Leonidas considered next the effect of the particular subject in that case, namely “Suppliers’ Approval”. His conclusion was that it too was a pre-condition which negatived contractual intent. While one of his reasons for reaching this conclusion was that at least two of the other subjects in that case were clearly pre-conditions (“stem” and “management approval”), a consideration which does not arise in the present case, this was not a critical factor. Of particular relevance was that “Suppliers’ Approval” was a matter which bore on the commercial desirability for the Charterer of entering into the charterparty which it was for the Charterer to determine, and that the Charterer might have a number of potential suppliers in mind to supply a cargo:

“61.

I prefer [the Charterer’s] contention that the phrase encompasses all those approvals which the charterer commercially wishes to obtain on the supply side (with the Receivers' Approval Subject having an equivalent meaning so far as the delivery side is concerned), and it can only be said to have been satisfied when the charterer lifts or waives the term. …

62.

The conclusion I have reached as to the true scope of the Suppliers' Approval Subject provides further strong support for the classification of this phrase as a pre-condition and not a performance condition. It is for the charterer to determine who its contractual supplier will be. It may be in discussions with more than one potential supplier at the same time or in quick succession, or have a choice between loading a cargo it already owns or buying cargo in from a third party. In these decisions, a wide range of commercial considerations will be in play. It would be wholly unreal against that background to suggest that the charterer was under an obligation to the owner to obtain the suppliers’ approval from ‘whoever the defendant intended to be the suppliers’ or ‘the approval of the supplier who they said they were waiting for the approval of’ when the subject was imposed (which was Mr Pearce's submission if I rejected his construction of the Suppliers' Approval Subject). This would constrain the charterers’ choice of supplier, hinder its ability to ‘change horses’ during a negotiation and commit it to obtaining the approval of a particular supplier even after negotiations with that supplier had broken down and the ‘supplier’ had no reason to engage with requests that it approve a vessel to lift a cargo which it was not going to supply.”

That reasoning applies with equal force in the present case to the subject of “shipper/receivers approval”. In particular, the subject was one which it was for the Charterer to “lift”, it being a commercial decision for the Charterer whether to do so, as distinct from being dependent on the actions of a third party. The same applies here. What matters is not whether the shipper has in fact given its approval (although the judge’s express finding and the arbitrator’s implicit finding was that it had not), but whether the Charterer has communicated to the Owner that the subject is lifted.

I would conclude, therefore, that the subject in this case was a pre-condition the purpose of which was to prevent a binding contract coming into existence.

As I understood his submissions, Mr Young did not seriously dispute that this would be so, subject to two points. The first was that because of the separability principle, the “subject” did not negative contractual intent so far as the arbitration clause in the recap was concerned. The second was that, in accordance with the arbitrator’s reasoning, the “subject” should be read together with clause 20 of the attached proforma so as to provide that approval could not be unreasonably withheld. I therefore consider next each of these points.

The separability principle, holding that an arbitration agreement is, or must be treated as, a contract which is separate from the main contract of which it forms part, is widely accepted internationally. It is an important concept for arbitration lawyers, although it may be questioned how many business people who include an arbitration clause in their contracts are aware that it exists.

The application of the separability principle to the initial validity of a main contract in English law was confirmed by the important case of Harbour v Kansa. In that case the parties entered into a reinsurance contract containing an arbitration clause. However, the reinsurer contended that, because the reinsured was not registered or approved to carry on reinsurance business in Great Britain under the Insurance Companies Act 1974 and 1981, the contract was void for illegality. The question was whether an arbitrator appointed pursuant to the arbitration clause had jurisdiction to determine this issue.

At first instance Mr Justice Steyn held that the arbitrator did not have jurisdiction, but his reasoning is important for an understanding of this case and the later case of Fiona Trust. He said this ([1992] 1 Lloyd’s Rep 81, 86 col 2):

“The foundation of an arbitrator’s authority is the arbitration agreement. If the arbitration agreement does not in truth exist, the arbitrator has no authority to decide anything. Similarly, if there is an issue as to whether the arbitration agreement exists, that issue can only be resolved by the Court. For example, if the issue is whether a party ever assented to a contract containing an arbitration clause, the issue of lack of consensus impeaches the arbitration agreement itself. Similarly, the arbitration agreement itself can be directly impeached on the ground that the arbitration agreement itself is void for vagueness, void for mistake, avoided on the ground of misrepresentation, duress, and so forth. All such disputes fall outside the scope of the arbitration agreement, no matter how widely drawn, and are obviously outside the arbitrator’s jurisdiction. The scope of the principle of the separability of the arbitration agreement only arises for consideration where the challenge is directed at the contract, which contains an arbitration clause. This fundamental distinction requires the Court to pay close attention to the precise nature of each dispute.”

This passage appears to be the origin in English law of the concept of an issue which “impeaches” an arbitration agreement. It is an important passage, which needs to be unpacked. Mr Justice Steyn distinguished between two situations. The first is where the dispute is whether a party ever assented to a contract containing an arbitration clause, that is to say where the argument is that “I never agreed to that” or “our negotiations never got as far as a binding contract”. That is an issue of contract formation, concerned with issues such as offer and acceptance and intention to create legal relations. The second situation is where the parties did assent to the terms of the contract containing an arbitration clause, but their agreement is invalidated on some legal ground which renders the contract void or voidable. That is an issue of contract validity. The parties did agree, but one of them is contending that the agreement is invalidated.

The passage makes clear that, where the issue is one of contract formation, it will generally impeach the arbitration clause: the argument “I never agreed to that” applies to the arbitration clause as much as it does to any other part of the contract. But where the issue is one of contract validity, that is not necessarily so. It is necessary “to pay close attention to the precise nature of each dispute” in order to see whether the ground on which the main contract is attacked is one which also impeaches the arbitration clause.

Harbour v Kansa was a case where the issue was one of contract validity, not contract formation. There was no doubt that the parties had in fact agreed the reinsurance contract, which included an arbitration clause. In those circumstances Mr Justice Steyn would have wished to hold that the arbitrator had jurisdiction to determine the parties’ dispute. The particular issue of illegality (the fact that the reinsured was not registered) did not impeach, and in fact had nothing to do with, the validity of the separate arbitration agreement contained in the reinsurance contract. But he held that he was constrained to hold otherwise by the decision of this court in David Taylor & Son Ltd v Barnett Trading Co [1953] 1 WLR 562.

An appeal to this court was allowed, on the ground that the earlier decision in the David Taylor case did not require the conclusion that an arbitrator had no jurisdiction to determine an issue of contractual illegality. Two points should be noted. First, all three members of the court approved the reasoning of Mr Justice Steyn in all respects, save for his view that he was bound by the David Taylor case. Second, all three members of the court emphasised that Harbour v Kansa was a case of what I have called contract validity, not contract formation.

Lord Justice Ralph Gibson approved the judgment of Mr Justice Steyn in these terms (708G-709B, 709F):

“In brief summary, the judge held as follows. (i) The principle of the separability of the arbitration clause or agreement from the contract in which it is contained exists in English law; and, provided that the arbitration clause itself is not directly impeached, the arbitration agreement is, as a matter of principled legal theory, capable of surviving the invalidity of the contract so that the arbitrators could have jurisdiction under the clause to determine the initial validity of the contract. Further, it would be consistent to hold that an issue as to the initial illegality of the contract is also capable of being referred to arbitration, provided that any initial illegality does not directly impeach the arbitration clause. (ii) The illegality alleged in this case does not impeach the arbitration clause. (iii) The arbitration clause in its proper construction is wide enough to cover a dispute as to the initial illegality of the contract. (iv) To his evident regret, however, Steyn J was driven to hold that the principle of separability could not apply when the alleged ground of invalidity of the contract the decision of this court in David Taylor & Son Ltd v Barnett Trading Co [1953] 1 WLR 562 compelled him to hold that the separability principle does not extend to initial illegality. …

For my part, for the reasons which follow, I would uphold the reasoning and conclusions of Steyn J on all aspects of his judgment, save for his final conclusion that he was bound by the decision in David Taylor & Son Ltd v Barnett Trading Co [1953] 1 WLR 562 to hold as he did. I would hold that this court can properly distinguish the decision in the Taylor case, and I would therefore allow this appeal.”

Lord Justice Leggatt also adopted the concept of impeachment of the arbitration clause (715G):

“I also agree that it would be consistent with [the judge’s] general approach to say that the initial illegality of the contract is capable of being referred to arbitration, provided that it does not impeach the arbitration clause itself …”

Lord Justice Hoffmann did not use the term “impeachment”, but did use language having the same effect (723E-G, 724A-B):

“Mr Longmore [counsel for the reinsurer] therefore accepts, as he must, that for some purposes the arbitration clause is treated as severable and may survive the termination or even the avoidance with retrospective effect of all the other obligations under the contract: see Mackender v Feldia AG [1967] 2 QB 590. He submits, however, that the separability doctrine cannot apply to any rule which prevents the contract from coming into existence or makes it void ab initio. In particular, it does not apply to a statute or other rule of law which makes the contract void for illegality.

It seems to me impossible to accept so sweeping a proposition. There will obviously be cases in which a claim that no contract came into existence necessarily entails a denial that there was any agreement to arbitrate. Cases of non est factum or denial that there was a concluded agreement, or mistake as to the identity of the other contracting party suggest themselves as examples. But there is no reason why every case of initial invalidity should have this consequence. …

In every case it seems to me that the logical question is not whether the issue goes to the validity of the contract but whether it goes to the validity of the arbitration clause. The one may entail the other but, as we have seen, it may not. …”

He later repeated that “the most common examples of cases in which the ground of invalidity of the substantive obligations of the contract also necessarily entails the invalidity of the arbitration clause are cases of initial invalidity, such as the absence of consensus ad idem, non est factum, mistake as to the person and so forth” (725D-E).

It follows, in my judgment, that Harbour v Kansa provides no support for any argument that an arbitrator has jurisdiction to determine an issue of contract formation. On the contrary, the clear tenor of the judgments, expressly stated by Mr Justice Steyn (whose reasoning was approved by Lord Justice Ralph Gibson) and by Lord Justice Hoffmann, is to the opposite effect.

Following Harbour v Kansa Parliament enacted the Arbitration Act 1996, which included section 7. This provides:

“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.”

This section gives statutory recognition to the separability principle. The DAC Report on the Arbitration Bill at para 43 noted that the section “sets out the principle of separability which is already part of English law (see Harbour Assurance v Kansa [1993] QB 701)”, with the explanation that “the doctrine of separability is confined to the effect of invalidity etc of the main contract on the arbitration agreement”. It is clear, therefore, that Parliament did not intend to go further than what Harbour v Kansa had already decided. The section is concerned with issues of contract validity, not contract formation. It applies where the main contract is invalid, non-existent or ineffective, but there must still be “an arbitration agreement”. This is apparent also from the definition of “arbitration agreement” in section 6 of the Act:

“(1)

In this Part an ‘arbitration agreement’ means an agreement to submit to arbitration present or future disputes (whether they are contractual or not).”

As Mr Young accepted, this refers to an agreement which is legally binding. It follows that the same term in section 7 must also refer to an arbitration agreement which is legally binding, and thus to one which satisfies the principles of contract formation. As Mr Holroyd put it, if there is no binding arbitration agreement in the first place, there is nothing to which the separability principle can apply.

Fiona Trust is the case on which Mr Young principally relied. That too was a case of contract validity, not contract formation, the question being whether arbitrators had jurisdiction to determine whether the contract was voidable as a result of being induced by bribery. There were two principal arguments. The first was concerned with the construction of the arbitration clause. The argument was that a clause referring to “any dispute arising under this charter” did not extend to a dispute as to whether the contract was voidable. The House of Lords dealt with this argument by sweeping away the distinctions between different forms of arbitration clause based on linguistic nuances (described by Lord Hope at [27] as “fussy distinctions”) which in its view had disfigured the previous case law. Instead, the right approach to the construction of an arbitration clause was to “start from the assumption that the parties, as rational businessmen, are likely to have intended any dispute arising out of the relationship into which they have entered or purported to enter to be decided by the same tribunal”, a presumption which was only to be rebutted by clear language (Lord Hoffmann at [13]). We are not concerned with this argument of construction of an arbitration clause in the present appeal.

The second argument was that if the owner was entitled to rescind the charterparty because of bribery, the whole contract including the arbitration clause would be rescinded, so that an arbitrator would have no jurisdiction. It was here that the principle of separability came into play.

Lord Hoffmann dealt with this argument in these terms:

“16.

The next question is whether, in view of the allegation of bribery, the clause is binding upon the owners. They say that if they are right about the bribery, they were entitled to rescind the whole contract, including the arbitration clause. The arbitrator therefore has no jurisdiction and the dispute should be decided by the court.

17.

The principle of separability enacted in section 7 means that the invalidity or rescission of the main contract does not necessarily entail the invalidity or rescission of the arbitration agreement. The arbitration agreement must be treated as a ‘distinct agreement’ and can be void or voidable only on grounds which relate directly to the arbitration agreement. 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.

18.

On the other hand, if (as in this case) the allegation is that the agent exceeded his authority by entering into a main agreement in terms which were not authorised or for improper reasons, that is not necessarily an attack on the arbitration agreement. It would have to be shown that whatever the terms of the main agreement or the reasons for which the agent concluded it, he would have had no authority to enter into an arbitration agreement. Even if the allegation is that there was no concluded agreement (for example, that terms of the main agreement remained to be agreed) that is not necessarily an attack on the arbitration agreement. If the arbitration clause has been agreed, the parties will be presumed to have intended the question of whether there was a concluded main agreement to be decided by arbitration.

19.

In the present case, it is alleged that the main agreement was in uncommercial terms which, together with other surrounding circumstances, give rise to the inference that an agent acting for the owners was bribed to consent to it. But that does not show that he was bribed to enter into the arbitration agreement. It would have been remarkable for him to enter into any charter without an arbitration agreement, whatever its other terms had been. Mr Butcher QC, who appeared for the owners, said that but for the bribery, the owners would not have entered into any charter with the charterers and therefore would not have entered into an arbitration agreement. But that is in my opinion exactly the kind of argument which section 7 was intended to prevent. It amounts to saying that because the main agreement and the arbitration agreement were bound up with each other, the invalidity of the main agreement should result in the invalidity of the arbitration agreement. The one should fall with the other because they would never have been separately concluded. But section 7 in my opinion means that they must be treated as having been separately concluded and the arbitration agreement can be invalidated only on a ground which relates to the arbitration agreement and is not merely a consequence of the invalidity of the main agreement.”

Lord Hope’s speech is to similar effect:

“33.

The appellants' case is that, as there was no real consent to the charterparties because they were induced by bribery, there was no real consent to the arbitration clauses. They submit that a line does not have to be drawn between matters which might impeach the arbitration clause and those which affect the main contract. What is needed is an analysis of whether the matters that affect the main contract are also matters which affect the validity of the arbitration clause. As the respondents point out, this is a causation argument. The appellants say that no substantive distinction can be drawn between various situations where the complaint is made that there was no real consent to the transaction. It would be contrary to the policy of the law, which is to deter bribery, that acts of the person who is alleged to have been bribed should deprive the innocent party of access to a court for determination of the issue whether the contract was induced by bribery.

34.

But, as Longmore LJ said at para 21 of the Court of Appeal's judgment, this case is different from a dispute as to whether there was ever a contract at all. As everyone knows, an arbitral award possesses no binding force except that which is derived from the joint mandate of the contracting parties. Everything depends on their contract, and if there was no contract to go to arbitration at all an arbitrator's award can have no validity. So, where the arbitration agreement is set out in the same document as the main contract, the issue whether there was an agreement at all may indeed affect all parts of it. Issues as to whether the entire agreement was procured by impersonation or by forgery, for example, are unlikely to be severable from the arbitration clause.

34.

That is not this case, however. The appellants' argument was not that there was no contract at all, but that they were entitled to rescind the contract including the arbitration agreement because the contract was induced by bribery. Allegations of that kind, if sound, may affect the validity of the main agreement. But they do not undermine the validity of the arbitration agreement as a distinct agreement. The doctrine of separability requires direct impeachment of the arbitration agreement before it can be set aside. This is an exacting test. The argument must be based on facts which are specific to the arbitration agreement. Allegations that are parasitical to a challenge to the validity to the main agreement will not do. That being the situation in this case, the agreement to go to arbitration must be given effect.”

Both speeches distinguish, therefore, between cases where the argument is that no contract was concluded at all (in Lord Hoffmann’s example, “one of the parties claims that he never agreed to anything in the document and that his signature was forged”; or as Lord Hope put it, approving what Lord Justice Longmore had said in the Court of Appeal, where there is “a dispute as to whether there was ever a contract at all”) and cases where the argument is that the contract apparently agreed is void or voidable. This is the distinction between what I have described as contract formation and contract validity. In the former case, the argument that no contract was ever agreed necessarily affects the arbitration clause because it means that the arbitration clause was not agreed either. In the latter case, however, it is necessary to consider whether the ground of invalidity in question is one which amounts to “an attack on” or “impeaches” the arbitration clause. It will not necessarily do so and, indeed, will be presumed not to do so unless the point relates directly to the arbitration agreement.

This is entirely consistent with the approval by the Court of Appeal in Harbour v Kansa of the reasoning of Mr Justice Steyn at first instance in that case, including his recognition that “if the issue is whether a party ever assented to a contract containing an arbitration clause, the issue of lack of consensus impeaches the arbitration agreement itself”. It is inconceivable that the House of Lords intended to depart from the reasoning in Harbour v Kansa, which included recognition of the distinction between contract formation and contract validity, without expressly saying so, not least as Lord Hoffmann had been a party to the decision in Harbour v Kansa, to which he referred with approval at [9].

The distinction between contract formation and contract validity has been applied in two first instance cases in this jurisdiction, albeit not in these terms. In Pacific Inter-Link Sdn Bhd v EFKO Food Ingredients Ltd [2011] EWHC 923 (Comm) the parties were negotiating contracts for the sale of palm oil. They had reached agreement on the terms to be included, which included a FOSFA arbitration clause, but the proposed seller had made clear that the contracts would only become effective once formal documents were returned to it signed and stamped by an authorised representative of the proposed buyer, which did not happen. It was, therefore, in effect a “subject to contract” case. Nevertheless the proposed buyer maintained that a binding arbitration agreement had been concluded. Mr Justice David Steel rejected this submission as “not arguable”:

“47.

It is not arguable that the arbitration clause contained within the original offer was nevertheless accepted by the counter-offer. The submission advanced by EFKO was that, in sorting out the details of the contracts, the parties agreed FOSFA arbitration even if the terms of the main contract were never concluded. But this is a contradiction in terms. There was no contract the details of which needed to be sorted out.”

The decision of Mr Justice Eder in The Pacific Champ [2013] EWHC 470 (Comm), [2013] 2 Lloyd’s Rep 320, on facts very similar to those of the present case, is to the same effect. The parties were negotiating a time trip charter and had reached agreement on the terms to be included, which included an arbitration clause. A recap was sent which set out those terms, “SUB CHTRS RECONFIRMATION 0900 HRS NYT Feb 13, 2008”. The issue was whether a binding arbitration agreement had been concluded, giving an arbitral tribunal jurisdiction to determine whether a binding charter fixture had been concluded. Mr Justice Eder held that it had not, rejecting a submission that the principle of separability meant that the arbitration agreement could only be challenged on grounds relating directly to the arbitration agreement and not merely to the main agreement. Rather, he held that the provision for arbitration was, like all the other terms set out in the recap, conditional on the subjects being satisfied. Thus the fixture and the arbitration agreement stood or fell together.

A similar approach has been adopted in Singapore. In BCY v BCZ [2016] SGHC 249, [2016] 2 Lloyd’s Rep 583 the parties were negotiating an agreement for the sale of shares. Their negotiations were expressed to be subject to the execution of a mutually acceptable agreement. The drafts exchanged between them contained an arbitration clause, but they continued to negotiate on other points. The plaintiff decided not to proceed, but the defendant commenced arbitration proceedings, contending that an arbitration agreement had been concluded, relying on the principle of separability. The plaintiff brought an action for a declaration that there was no arbitration agreement in existence. Mr Justice Steven Chong agreed. He introduced his judgment as follows:

“1.

When the jurisdiction of an arbitral tribunal is challenged on the basis that there is no binding arbitration agreement, the usual ground for such a challenge is that the contract which incorporates the arbitration clause was itself never concluded. In this familiar situation, it has been held that the validity of the arbitration agreement and the existence of a binding contract would ‘stand or fall together’ and the court would usually determine both issues collectively (see Hyundai Merchant Marine Co Ltd v Americas Bulk Transport Ltd (The Pacific Champ) [2013] 2 Lloyd’s Rep 320 at paras 35 and 36, cited in Jiangsu Overseas Group Co Ltd v Concord Energy Pte Ltd [2016] SGHC 153, [2016] 4 SLR 1336 at para 48).”

Mr Justice Steven Chong dealt first with principle of separability as it affects the governing law of the arbitration agreement. A feature of the case was that the latest draft of the proposed agreement provided for New York law to govern the main agreement, but with an arbitration clause providing for arbitration in Singapore. As it was common ground that the question whether an arbitration agreement had been concluded was to be decided in accordance with the governing law of that putative arbitration agreement, identification of that governing law was a necessary first step. The judge said:

“60.

The suggestion that the arbitration agreement is a distinct agreement with a governing law distinct from that of the main contract is often justified by the doctrine of separability. However, the doctrine of separability serves to give effect to the parties’ expectation that their arbitration clause -- embodying their chosen method of dispute resolution -- remains effective even if the main contract is alleged or found to be invalid. It does not mean that the arbitration agreement forms a distinct agreement from the time the main contract is formed. Resort need only be had to the doctrine of separability when the validity of the arbitration agreement itself is challenged. This is clear from article 16 of the UNCITRAL Model Law on International Commercial Arbitration set out in the first schedule of the [International Arbitration Act] (“Model Law”) …

61.

Separability serves the narrow though vital purpose of ensuring that any challenge that the main contract is valid does not, in itself, affect the validity of the arbitration agreement. This is necessary because the challenge to the validity of the arbitration agreement often takes the form of a challenge to the validity of the main contract. However, as Moore-Bick LJ noted in Sul América [Sulamérica Cia Nacional de Seguros SA v Enesa Engenharia SA [2012] EWCA Civ 638, [2013] 1 WLR 102], separability does not ‘insulate the arbitration agreement from the substantive contract for all purposes’ (at para 26). It is one thing to say that under the doctrine of separability, a party cannot avoid the obligation to submit a dispute to arbitration by merely denying the existence of the underlying contract; it is quite different to say that because of this doctrine, it is intended to enter into an arbitration agreement independent of the underlying contract. This does not reflect commercial reality.”

Having concluded that the governing law of the putative arbitration agreement was the law of New York, Mr Justice Steven Chong turned to the question whether there was a valid and binding arbitration agreement at all. He held that the separability principle was not relevant to this question, which fell to be decided applying the usual requirements for the formation of a contract:

“79.

Mr Jeremiah also submitted that the doctrine of separability supported his argument that the arbitration agreement could have been concluded prior to the SPA. However, as I have explained above, the doctrine of separability is only relevant where an arbitration agreement forms part of a main contract -- the doctrine prevents a party from impugning the arbitration agreement simply by alleging that the main agreement was invalid. In this case, Mr Jeremiah’s case is that the arbitration agreement was concluded before the conclusion of the SPA. There is no need to invoke the doctrine of separability. The court’s task, in deciding whether the arbitrator had the jurisdiction to hear the dispute, is to consider the usual requirements for the formation of a contract under the applicable law. This was the way both parties agreed to address the jurisdictional issue.”

Because the governing law of the putative arbitration agreement was the law of New York, the applicable “usual requirements for the formation of a contract” were those provided for by New York law, namely offer, acceptance, consideration, mutual assent and intent to be bound (see at [82]). These are no different, of course, from those which apply under English law. Applying those principles, the judge held at [82] that there was “no objective manifestation of any mutual intention by the parties to be bound by the arbitration agreement”. In particular, the fact that the parties had agreed on the wording of the arbitration clause and made no further changes to it thereafter did not mean that they intended to be bound by it as an independent contract (see at [86]), while the fact that the negotiations were “subject to contract” prevented any of the terms of the proposed agreement, including the arbitration clause, from becoming binding (see at [93]).

The judge found support for this view in Born, International Commercial Arbitration, a leading work on international arbitration (2^nd Ed (2014), pages 795-6):

“In many instances, it will be difficult to show that the parties did not agree to be bound by an underlying commercial contract, but nonetheless intended to conclude an arbitration agreement associated with that contract. For example, parties not infrequently exchange drafts of proposed contracts, including comments on both draft arbitration provisions and draft commercial terms; sometimes, parties reach agreement on the terms of an arbitration clause before doing so on commercial terms. If no agreement is ever reached on the commercial terms of the underlying contract, it is sometimes argued that the exchange of identical drafts of an arbitration clause, whose terms both parties accept, evidences an agreement on the arbitration provision (notwithstanding the lack of agreement on the underlying contract).

Although dependent on the fact of individual cases, arguments of this sort are generally difficult to sustain. The parties’ agreement on the terms of an arbitration clause does not typically amount to a mutual intention to be legally bound by that provision, absent conclusion of the underlying contract. Rather, such exchanges typically indicate agreement on the text of an arbitration clause, but an intention to be legally bound by that arbitration provision when, but only when, the underlying contract is also concluded. That conclusion is often reinforced by inclusion of caveats on negotiating materials indicating that the drafts are ‘subject to contract’, ‘without prejudice’, or otherwise conditional upon final agreement and formal execution of the contracts in question.”

Finally, Mr Justice Steven Chong rejected a submission that a different approach to the applicability of the doctrine of separability was required as a result of the decision of the House of Lords in Fiona Trust. After citing what Lord Hoffmann had said at [18] which I have already set out, he said:

“89.

The challenge to the arbitration clause in that case took the form of a challenge to the validity of the main contract on the ground that it had been procured by bribery. Given that the arbitration clause was not impugned on the basis that the main contract was not concluded, this observation takes us no further than the question of what it means for an arbitration clause to be ‘agreed’ -- whether that means agreement to the wording of the clause, or agreement to be bound by it.

90.

The underlying principle, as I have found, is that agreeing to the wording of the arbitration clause does not per se equate to an intention to be contractually bound to arbitrate absent the conclusion of the contract under which the arbitration clause was negotiated.”

I have cited from the judgment in BCY v BCZ at some length because it represents the fullest (and if I may say so, clearest) treatment cited to us of the application (or rather non-application) of the separability principle to issues of contract formation; it demonstrates that the distinction which I have sought to explain between issues of contract formation and contract validity is recognised in at least one other leading jurisdiction which is a centre of international arbitration; it demonstrates that the distinction is recognised in at least one widely used textbook on international arbitration (incidentally described by the majority in the Supreme Court in Enka v Chubb at [107] as a “monumental work”); it recognises in terms that Fiona Trust was a case of contract validity, not contract formation; and it was referred to with approval by the Supreme Court in Enka v Chubb by both the majority (at [57] and [58]) and the minority (at [224] and [225]), albeit on the issue of the governing law of the arbitration clause. It demonstrates also the value of dialogue between courts exercising jurisdiction in commercial cases, in which we can each learn much from the other.

As Mr Justice Gross observed in UR Power GmbH v Kuok Oils & Grains Pte Ltd [2009] EWHC 1940 (Comm), [2009] 2 Lloyd’s Rep 495 at [35], there is a “vast literature on the topic of separability and its true scope”. The current third edition (2021) of Mr Born’s book contains in Chapter 3 an extensive discussion of the way in which this topic has been treated in numerous jurisdictions, including the United States where, it appears, the Supreme Court has also acknowledged the distinction between contract validity (for the arbitrator to decide) and contract formation (for the court): see in particular the discussion at pages 442-3 of the case of Buckeye Check Cashing, Inc v Cardegna 546 US 440 (2006). However, the materials cited to us in this appeal did not even scratch the surface of this literature.

The discussion in Mr Born’s book includes at pages 493-5 the valuable insight that the non-existence of the main contract does not necessarily mean that an arbitration agreement is also non-existent. Rather, the separability principle means that the question of contract formation must be asked twice, once in relation to the main contract and again in relation to the arbitration agreement. In most cases the same answer will be given to both questions, although it is theoretically possible for parties to conclude a binding agreement to arbitrate even if they have not (or not yet) agreed on the main contract. But in both cases the issue is one of contract formation, in particular whether, applying usual principles, the parties have evinced an intention to be bound. As Mr Born puts it:

“It is true that the non-existence of an underlying contract may be accompanied by the nonexistence of the arbitration agreement. Thus, where two parties never met or negotiated in any way, there will be no arbitration agreement and no underlying contract. This is not, however, in any way inconsistent with the separability presumption; on the contrary, properly analyzed, this type of case is a useful illustration of the separability presumption’s application.

As discussed above, the separability presumption does not provide that, where the underlying contract is non-existent or invalid, the arbitration agreement is nonetheless necessarily existent and valid. Rather, that the arbitration agreement may be existent and valid even if the underlying contract is not; that is because the arbitration agreement is presumptively a separate agreement, distinct from the underlying contract, whose terms and status differ from those of the underlying contract. The relevant question, therefore, is whether the parties did or did not negotiate and conclude a valid agreement to arbitrate their disputes even if they did not also conclude the underlying contract.

In general, given the close relationship between the underlying contract and the arbitration agreement, defects in the formation of the former are likely to affect the latter: parties do not ordinarily agree to arbitration provisions in the abstract (‘floating in the legal ether’), without an underlying contract. Nevertheless, there will be instances where the parties are held to have concluded their negotiations, and reached a valid binding agreement, on an arbitration clause, but not on the underlying contract. …

The most difficult issues arise when a particular alleged defect in formation affects both the arbitration clause and the underlying contract (e.g. the contract, including the arbitration clause, was never executed, or the contract was affected by forgery, or a party lacks mental capacity). These are cases of ‘doubly relevant facts’ or ‘identities of defects’, is simultaneously relevant to the validity or existence of both the underlying contract and the associated arbitration agreement.

In these cases, absent special or additional circumstances, the reasons for the defect in the underlying contract almost always also affects the substantive validity of the arbitration agreement. There is seldom a credible basis for arguing that forgery of a signature on a contract, affecting the underlying contract, does not also impeach the arbitration clause: unless the arbitration clause was separately signed, or agreed in some other manner, then a forged signature on the underlying contract evidences the absence of agreement on anything in that document. Similarly, the failure to execute the underlying contract will generally evidence a failure to agree upon the associated arbitration clause; there may be cases where separate expressions of assent exist with regard to the arbitration agreement, but the circumstances will be unusual, and must be established through allegations directed specifically at the existence of an arbitration agreement. …”

I do not accept that the approach which I have set out is (as Mr Young submitted) “antithetical to the modern ‘one-stop’ dispute resolution presumption in contractual interpretation”. That presumption is concerned with the interpretation of dispute resolution clauses, as made clear in Fiona Trust. But there is no issue about the interpretation of the arbitration clause in this case. The presumption has nothing to do with the question whether the parties have concluded a contract (including a contract to arbitrate) in the first place. On the contrary, to hold that the question whether a binding arbitration agreement has been concluded is subject to ordinary principles of contract formation is a principled approach. It recognises that an arbitration agreement is a contract like any other, so that there is no justification for treating the question whether such an agreement has been concluded as subject to special presumptions uniquely applicable in arbitration cases. One-stop shopping is all very well, but if the parties have not entered into an arbitration agreement, the shop is not open for business in the first place.

The only case cited to us which comes anywhere near calling into question the approach which I have set out above is UR Power v Kuok Oils & Grains. In this case Mr Justice Gross held that there was a binding sale contract between the parties, which was subject to a FOSFA arbitration clause. The contract was subject to a requirement that the buyer should open a letter of credit, but this was (in the terms later used by Mr Justice Foxton in The Leonidas) a performance condition and not a pre-condition to the formation of a binding contract. Accordingly the arbitrators had jurisdiction to determine the parties’ dispute. As Mr Justice Gross noted, it was therefore unnecessary to determine whether the separability principle would have given the arbitrators jurisdiction even if the letter of credit term had been a pre-condition. He observed at [35] that “this issue raises points of no little difficulty and of some general interest” and that as the issue did not arise, “this does not seem to be the occasion to explore such an issue in depth or to express a concluded view”. Nevertheless, he went on to express at [40] the “provisional inclination” that it was “strongly arguable” that the arbitrators would have had jurisdiction:

“In principle, therefore, an arbitration agreement may be binding even though the underlying contract has not come into existence. With respect to Mr Collett’s argument to the contrary, it does not follow that in every case where pre-contractual negotiations have not resulted in a binding (underlying or matrix) contract, an arbitration clause discussed in the course of those negotiations would [sc. would not] be binding. Whether it is or not will necessarily be a question of fact and degree, depending on the circumstances of the individual case.”

I do not regard this avowedly provisional and tentative view as detracting from the analysis which I have set out.

The applicable principles of contract formation are those set out in the leading case of RTS Flexible Systems Ltd v Molkerei Alois Müller GmbH & Co KG [2010] UKSC 14, [2010] 1 WLR 753, applying the earlier case of Pagnan SpA v Feed Products Ltd [1987] 2 Lloyd’s Rep 601. In brief, and so far as relevant for present purposes, Lord Clarke approved the principle stated in Pagnan that:

“Even if the parties have reached agreement on all the terms of the proposed contract, nevertheless they may intend that the contract shall not become binding until some further condition has been fulfilled. That is the ordinary ‘subject to contract’ case.”

Drawing these points together, and leaving to one side for the moment Mr Young’s submission that the “subject” in this case needs to be read together with clause 20 of the proforma, in my judgment the position in the present case is clear.

It can be summarised as follows:

It remains to consider the submission that the “subject” in this case needs to be read together with clause 20 of the proforma charterparty which was attached to the recap message. It will be recalled that this was the view which the arbitrator took. He concluded that, reading the “subject” of “shippers/receivers approval” together with clause 20 of the proforma, the effect of the “subject” could be restated as follows:

“subject shippers/receivers’ approval within one working day after fixing main terms and receipt of all required/corrected certificates/documents such approval not to be unreasonably withheld.”

I would agree with Mr Young that if this were the correct reading, it would cast the “subject” in a very different light. It would mean that the “subject” was properly to be regarded as a performance condition, rather than a pre-condition which had the effect of negativing contractual intent. In that case, the arbitration agreement would have been concluded as part of a binding charterparty contract, and the arbitrator would therefore have had jurisdiction over the parties’ dispute.

However, I agree with Mr Justice Jacobs that this is an untenable reading. The judge gave a number of reasons for that view. In my judgment it is sufficient to say that clause 20 of the proforma, which is concerned with the nomination of a vessel under a charterparty for a vessel to be nominated, has no application to a charterparty for a named vessel. There is, therefore, no question of seeking to read the two provisions together.

For the reasons which I have given I would dismiss the appeal. The judge was right to conclude that the arbitrator had no substantive jurisdiction and that his award should be set aside pursuant to section 67 of the Arbitration Act 1996.

There are two other matters which should be mentioned. The first is that, in view of my conclusion on section 67, the appeal under section 69 does not arise. However, if I had concluded that the separability principle meant that the arbitrator did have jurisdiction in this case, it would have been necessary to remit the matter to Mr Justice Jacobs so that he could decide the section 69 appeal. Mr Young submitted that, in that event, the remission should be to a different judge. He referred to the practice in the Commercial Court that the judge who grants permission to appeal does not usually hear the substantive appeal (see CVLC Three Carrier Corp v Arab Maritime Petroleum Transport Co [2021] EWHC 551 (Comm) at [23] and [24]). But that sensible practice has no application to a case such as the present, where there was a rolled up hearing, with the application for permission to appeal and the appeal heard together. The position is that Mr Justice Jacobs has heard the appeal, albeit that in view of his conclusion on section 67, there was no need for him to give judgment on it. If we were to hold that he was wrong on the issue of jurisdiction, it would become necessary for him to give that judgment. In that event, he made clear that he would have allowed the appeal. For what it is worth, I agree that he would have been right to do so, although the matter could only have come before this court if the court below had given permission for a second appeal under section 69(8) of the 1996 Act. It follows that a successful appeal on the issue of separability would not ultimately have availed the Owner.

The second matter arises out of a concern expressed at the hearing that, in practice, a party who wishes to contend that a binding contract including an arbitration agreement has been concluded has no real choice other than to commence arbitration, and therefore faces the expense of proceedings before arbitrators, followed by a section 67 challenge in court. However, that is not necessarily so. As I said in The Kalisti at [4], one obvious solution in circumstances such as the present case would be for the parties to make an ad hoc agreement to submit the issue whether a binding contract has been concluded to arbitration, without prejudice to any issue whether such an ad hoc agreement is necessary. There could be no serious objection to that course. One party wants to arbitrate its claim, while the other party was prepared to arbitrate any dispute if a contract had been concluded and can therefore have no sensible objection to arbitration as a process to resolve a dispute of this nature. If that course is taken, the arbitrators will produce an award which, subject only to the possibility of a section 69 appeal on a question of law, will be final and binding. That ought to satisfy both parties. In the present case that course was not taken, and was not even explored because the Owner’s notice of arbitration never reached the Charterer’s management as a result of the misconduct of one of the Charterer’s employees. But that is an unusual circumstance. It should not be thought that a claimant in circumstances such as the present case is necessarily faced with multiple proceedings.