The First Claimant (“GDE”) is a US company incorporated on 21 April 2009 in Georgia, originally under the name of Anglia Autoflow North America LLC. The Second Claimant (“Mr Goffe”) is and was the only member and shareholder of the company. The Defendant (“AAL”) is an English company which manufactures machinery and associated products for the poultry processing industry.

The underlying dispute between the parties arises out of an Agency Agreement concluded on about 9 April 2009 whereby the Defendant appointed a company to be incorporated by Mr Goffe (in the event, GDE) as its commercial agent in various territories including Canada, the United States, Mexico and the Caribbean. It is unnecessary to go into the details of the dispute. It is sufficient for present purposes to note only (1) that the Claimants’ claim is for unpaid commission and damages for alleged repudiatory breach of the Agency Agreement by AAL; and (2) that AAL admits non-payment of the invoices but denies that the non-payment was repudiatory. AAL asserts that in so far as any amounts were due and owing to Mr Goffe/GDE, it is entitled to set off its own claim for damages for breach of the agreement.

The parties are also in dispute as to the governing law of the Agency Agreement by which these claims and cross-claims should be determined. AAL alleges that the governing law is that of Ontario while the Claimants allege that the Agency Agreement is governed by English law. The point is of critical importance because the Claimants concede that, if AAL is correct, their claim is time-barred under Ontario law.

In the light of that concession, HHJ Rawlings ordered at a CMC held on 24 May 2019 that the following two questions be tried as preliminary issues:

This is the trial of those issues.

Anglia Autoflow North America LLC to be the exclusive agent for Anglia Autoflow Ltd for the territories of Canada, Caribbean Islands and United States of America.

The effective commencement date of the agency to be 10 April 2009 and will run for an initial period of 3 years and thereafter may be continued.

Anglia Autoflow Ltd normally achieves terms with customers of:

A sales commission, of 10% of the equipment nett ex works price, England, charged to the customer, will be paid by Anglia Autoflow Ltd to Anglia Autoflow North America LLC. Anglia Autoflow Ltd will forward to Anglia Autoflow North America LLC copies of all invoices raised on customers – be they for spares, deposits with order, progress payments etc.

Anglia Autoflow Ltd. currently manufactures modules frames and plastic drawers in Canada. In the event that Anglia Autoflow Ltd, and Anglia Autoflow North America LLC, make a joint decision to source items of Anglia Autoflow Ltd equipment (such as module frames) in any of the other territories in order to reduce freight costs etc., a separate agreement, and understanding, will need to be mutually agreed. This being on the principle that Anglia Autoflow North America LLC would still receive a minimum sales commission of 10% based on the nett ex factory price, charged to the customer, of the equipment sold.

Anglia Autoflow Ltd will also pay a commission of 10% to Anglia Autoflow North America LLC on all spare parts sold ex England to customers in any of the territories during the course of this agency – this calculated on the ex works England price. This commission payable to Anglia Autoflow North America LLC upon receipt of settlement, in full, from the customer…

Both parties have discussed and have agreed that close co-operation will be necessary, particularly in the early stages of this agreement. Anglia Autoflow Ltd will supply full documentation and information to Anglia Autoflow North America LLC and will also provide extensive sales support by joint visits to potential customers in any of the territories…

In the event of serious disputes, both parties agree to subject themselves to the jurisdiction of the English Courts.

It is common ground that the governing law of the Agency Agreement is to be determined in accordance with the provisions of the Rome Convention which was incorporated into English law by section 2 of the Contracts (Applicable Law) Act 1990 and continues to apply to all contracts concluded prior to 17 December 2009.

The Rome Convention provides in material part as follows:

To the extent that the law applicable to the contract has not been chosen in accordance with Article 3, the contract shall be governed by the law of the country with which it is most closely connected. Nevertheless, a severable part of the contract which has a closer connection with another country may by way of exception be governed by the law of that other country.

Subject to the provisions of paragraph 5 of this Article, it shall be presumed that the contract is most closely connected with the country where the party who is to effect the performance which is characteristic of the contract has, at the time of conclusion of the contract, his habitual residence, or, in the case of a body corporate or unincorporate, its central administration. However, if the contract is entered into in the course of that party's trade or profession, that country shall be the country in which the principal place of business is situated or, where under the terms of the contract the performance is to be effected through a place of business other than the principal place of business, the country in which that other place of business is situated.

It is not in dispute (i) that the performance which was characteristic of the Agency Agreement was that of the agent; (ii) that the agent contemplated by the Agency Agreement was GDE; but (iii) that the contract was concluded prior to GDE’s incorporation by Mr Goffe. This gives rise to interesting and difficult questions as to whether, if there is no choice of law under Article 3(1), the presumption in Article 4(2) applies and, if so, how.

The Claimants’ primary case was that:

The parties had made a tacit or implied choice of English law for the purposes of Article 3(1) of the Rome Convention by virtue of their express agreement to English jurisdiction in clause 11 of the Agency Agreement.

In the alternative, if there was no implied choice of law, Article 4(2) expressly looked to the situation as it existed at the date of conclusion of the contract and, since GDE did not exist at that date, this necessarily meant looking at the situation of Mr Goffe. Mr Goffe’s principal place of business and habitual residence at the date of the contract were both in Georgia and the prima facie presumption under Article 4(2) was therefore that the contract was most closely connected with Georgia, whether or not it was concluded in the course of his trade or profession.

Nonetheless, the presumption in Article 4(2) should be displaced under Article 4(5) because, taking the circumstances as a whole, the Agency Agreement was more closely connected with England with the result that English law was the governing law.

As a final fall-back position (although this was controversial: see paragraphs 17ff. below), the Claimants contended that if the presumption could not be displaced, then Georgia law would apply.

AAL’s case, by contrast, was that:

The jurisdiction clause in the Agency Agreement did not give rise to any implied choice of law. Article 3(1) therefore had no application.

The principal place of business of both GDE (and, if relevant, Mr Goffe) at the date of the agreement was Ontario.

Alternatively, if it was Georgia, the circumstances as a whole demonstrated that the Agency Agreement had its closest connection with Ontario.

Either way, the governing law of the Agency Agreement was Ontario law.

Whether the parties made a tacit choice of law pursuant to Article 3(1) of the Rome Convention. This includes considering the nature and effect of the jurisdiction clause in the Agency Agreement;

The application of Article 4 of the Rome Convention if there was no tacit choice of law under Article 3, including:

Whether Article 4(2) can sensibly be applied to the circumstances of this case;

If it can, whether the relevant connecting factor for the purposes of Article 4(2) is:

What, if any presumption, arises by virtue of Article 4(2);

Whether any presumption arising under Article 4(2) is to be displaced under Article 4(5) and, if so, in favour of which country.

Both sides took pleading points against the other. Some were minor and are dealt with in the discussion of the issues. The more substantial points were as follows.

The first arose out of the skeleton argument of Professor Jonathan Harris on behalf of the Defendant. Having set out in paragraph 36 the Defendant’s primary case that the relevant connecting factor under Article 4(2) pointed to Ontario, paragraph 37 continued “Further or alternatively, Mr Goffe should be bound by his representations to the Defendant to the effect that he was living and working in Ontario.” Paragraph 38 put forward a further alternative case that by virtue of these representations “the

The Claimants initially objected to both these paragraphs on the grounds that they introduced for the very first time an estoppel argument which had been neither pleaded

Mr Craig Ulyatt, who appeared for the Claimants, quite properly did not object to this manner of presenting the argument provided that it was so confined. However, he made it clear that he would vigorously oppose any case which strayed into estoppel territory as in his submission it was far too late to be raising a new point of this nature. I agree. It is accordingly not open to the Defendant to go beyond the ambit of the case on construction/interpretation as clarified by Professor Harris.

The second pleading point was whether it was open to the Claimants to rely on anything other than the jurisdiction clause in the Agency Agreement in support of their case that there had been a tacit choice of law under Article 3(1). Professor Harris pointed to paragraph 14(2) of the Claimants’ Reply which read:

He submitted that since the only factor relied on by the Claimants in relation to Article 3(1) was the jurisdiction clause, it was simply not open to them on the pleadings to rely on other circumstances, notwithstanding the wording of Article 3(1) which expressly refers to “the terms of the contract or the circumstances of the case”. As a matter of strict pleading, Professor Harris may well be right about this. However, it is the most technical and unmeritorious of arguments given that the circumstances on which the Claimants sought to rely in addition to the jurisdiction clause were, without exception, all expressly pleaded in relation to Article 4(5) and explored in cross-examination in that context.

In his reply, Mr Ulyatt made a precautionary application to amend paragraph 14(2) by inserting the prefatory words “In the circumstances of the present case (including, inter alia, the matters referred to in paragraph 14.3 below)”. Professor Harris was hardpushed to suggest that the Defendant would suffer any real prejudice were the amendment to be allowed and I have no hesitation in granting permission on the usual terms that any costs of and occasioned by the amendment should be paid by the Claimants to the Defendant.

The final pleading point was the most difficult and potentially important. As appears from paragraphs 10 and 11 above, neither side is positively inviting me to find that the Agency Agreement is governed by Georgia law. The Claimants say that there was an implied choice of English law. Alternatively, if that is not correct, the presumption in Article 4(2) of the Rome Convention, which would otherwise point to Georgia law, falls to be disapplied in favour of English law. The Defendant says that there was no implied choice and that application of Article 4(2) leads to Ontario law. Alternatively,

In these circumstances, Professor Harris invokes the well-known procedural rule that a party wishing to rely on the provisions of a foreign system of law must plead and prove that law. He argues that, far from pleading the application of Georgia law, the Claimants have expressly pleaded that Georgia law does not apply. It is therefore simply not open to the court to find that the Agency Agreement is governed by Georgia law, even as a fallback position. He says that the Defendant assessed the merits of its position by reference to the pleaded case and prepared its defence on the reasonable basis that the only choice was between English law and Ontario law. It would therefore be severely prejudiced if it were now to be faced with the possible application of Georgia law – a result for which neither party is contending.

Whether or not the Defendant – or, indeed, the Claimants – would be prejudiced by the application of Georgia law I am unable to say, as neither side adduced any evidence as to its substantive provisions. My instinctive reaction to the Defendant’s point was sceptical; both parties had invited the court to apply the Rome Convention (which applies as a matter of law under the 1990 Act) and it seemed to me that once the wheels of the Convention had been put in motion, they could not be stopped short of their ultimate destination. The idea that the process dictated by the Convention should be hijacked halfway, as it were, on the basis of a pleading point was, to my mind, deeply unattractive.

Nonetheless, Professor Harris insisted that my instinctive reaction was wrong and set out so to convince me with the assistance of copious reference to both authority and learned academic commentary. There were a number of steps in his argument:

Article 1(2)(h) of the Rome Convention expressly provides that (with one immaterial exception) the rules of the Convention do not apply to “evidence and procedure”. These are governed by the lex fori.

Under English law as the lex fori, a foreign law must be pleaded and proved as a fact, failing which the case will be decided by the application of English domestic law: Dicey, Morris & Collins, The Conflict of Laws (15^th ed., 2012, OUP) paragraphs 9-002, 9-003; Briggs, Private International Law in English Courts (2014, OUP) paragraphs 2.59, 3.24-3.25.

The Claimants are not arguing for an application of Georgia law. Although it is accepted that the Claimants’ Reply contains an implicit plea that Article 4(2) raises a presumption of Georgia law, this is only put forward as a stepping stone and the Claimants’ affirmative case is that Georgia law should be positively disapplied.

There being no positive plea by either side that Georgia law applies, the court cannot find that it does, irrespective of what the Rome Convention might otherwise say. The court is not bound, and indeed is obliged not, to apply the Convention of its own motion: Briggs (op. cit.) §§3.25, 3.26; McParland, The Rome I Regulation on the Law Applicable to Contractual Obligations (OUP) paragraphs 8.73-8.77; The Alexandros T, [2013] UKSC 70; [2014] Bus L. Rep. 873.

In short, the argument was that once the Claimants had advanced from Article 4(2) to Article 4(5), the drawbridge was pulled up behind them and they could not retreat back again if Article 4(5) ultimately proved to be hostile territory.

I accept the first step in Professor Harris’ argument. I also accept the second, noting, as the extract from Dicey, Morris & Collins makes clear, that two conceptually distinct issues arise in relation to the application of foreign law. First, identification of the foreign law in question, which must be pleaded. Second, the substantive content of the foreign law so identified, which must be proved to the satisfaction of the court as a matter of fact in the usual way.

I further accept that, even though the Rome Convention applies as a matter of law, on the orthodox common law approach it is not the function of a judge to apply the Convention ex officio. While Professor Briggs expresses some doubt as to how long this approach can be maintained ((op. cit.) paragraph 2.59 and fn 10 to paragraph 3.25), it has nonetheless recently been endorsed by the decision of the Supreme Court in The Alexandros T.

This concerned a dispute arising out of a settlement agreement between the owners of the Alexandros T and their insurers. The owners commenced proceedings in Greece in breach of an exclusive jurisdiction clause in the settlement agreement and the insurers commenced proceedings in England seeking damages for the breach. The owners applied to stay the English proceedings on the grounds that the Greek courts were first seised for the purposes of Article 28 of Council Regulation No. 44/2001. They expressly disclaimed reliance on the lis alibi provisions of Article 27. The stay was refused and the owners appealed, seeking to rely, inter alia, on Article 27 for the first time. Article 27 expressly provides that where parallel proceedings involving the same cause of action between the same parties are on foot in different member states, any court other than that first seised shall “of its own motion” stay its proceedings. The Court of Appeal held that it was bound to apply Article 27 of its own motion but this was doubted in the Supreme Court, where Lord Clarke accepted the submissions of the insurers that the object of the Brussels Convention was not to unify the rules of substantive law and procedure of the different contracting states and that national rules should continue to govern matters of procedure provided that the effectiveness of the Convention was not thereby impaired. Reluctantly, he came to the conclusion that the words “of its own motion” were not acte claire and he would have been prepared to make a reference to the European Court of Justice had it been necessary to resolve the appeal: see especially at paragraphs 98-123.

The case is thus not precisely analogous to the present situation. In particular, the words “of its own motion” which particularly concerned Lord Clarke do not appear in the Rome Convention. However, it is quite clear that had it not been for these words he would have held that the court had a discretion as a matter of English procedural law whether or not to allow the owners to rely on Article 27 and that this discretion was not in principle overridden by the provisions of the Regulation. The case therefore provides substantial support for Professor Harris’ argument; and if neither party had pleaded that a foreign law applied, or relied on the Rome Convention at all, I accept that the court would have been bound to apply English law: see Dicey, Morris & Collins (op. cit.) paragraph 9-011; Briggs (op. cit.) paragraph 2.59.

However, that is not this case. On the contrary, both parties have relied on the Rome Convention as the appropriate mechanism for determining the governing law. So far as I am aware, there is no authority on the question of whether a mere plea that the Rome Convention applies is a sufficient pleading of whichever foreign law emerges at the end of the Convention process so as to satisfy the English rule of procedure set out above.

In principle, it seems to me that it ought to be sufficient. I asked Professor Harris on several occasions what the position would be on his case if I were to find that the presumption in Article 4(2) pointed to Georgia law, but if I then rejected the arguments of both parties that the contract was more closely connected with England or Ontario respectively for the purposes of Article 4(5). I received no very satisfactory answer beyond a submission that I nonetheless had to choose between English law and Ontario law.

This seemed to me to involve parting company with all common sense. The idea that the court should be obliged to hold that the contract was governed by one of two systems of law which it had just rejected as being applicable was startling, to say the least. If the submissions of both sides on Article 4(5) were rejected and I was not entitled to apply Georgia law under Article 4(2), by what system of law would the claim be governed? It cannot simply disappear. Presumably, I would have to apply English domestic law in default of any other pleaded applicable system, yet that was precisely the result which the Defendant was trying to avoid.

If it had been necessary to do so, I would therefore have held that the invocation of the Rome Convention by the Claimants was in and of itself a sufficient plea of foreign law to enable the court to follow the Convention process to its logical conclusion without it being necessary to plead expressly each and every possible system of law that might apply thereunder. The court would accordingly have been entitled to apply Georgia law under Article 4(2), notwithstanding that it had only been pleaded as a stepping stone to the Claimants’ ultimate goal. Pace Professor Briggs (op. cit.) paragraphs 3.28, 3.40, this is not a question of the court thinking that it knows better than the parties; rather it is a question of the court dutifully proceeding down the path prescribed for it by the parties. I reject the suggestion that there is anything unfairly prejudicial to the Defendant in this. The Defendant was the party which first invoked the Convention and it is inherent in the Convention that the Article 4(2) presumption might not be displaced. That possibility should at least have been recognised.

In the event, however, I do not need to rest my decision on this basis since I agree with Mr Ulyatt that the Claimants’ Reply contains within it a sufficient plea (albeit implicit) that Georgia law is the presumptive applicable law under Article 4(2) and that this is sufficient to support a fallback case in the event that his primary case under Article 4(5) is rejected.

One way or the other, therefore, I hold that it is open to me to find that the Agency Agreement is governed by Georgia law although, if that is indeed the conclusion to which I come, the substantive provisions of Georgia law will need to be proved as a fact, failing which they will be taken to be the same as English law: Iranian Offshore Engineering and Construction Co. v Dean Investment Holdings SA, [2018] EWHC 2759 (Comm); [2019] 1 W.L.R. 82. Neither side has yet pleaded a positive case on

For completeness, I reject Professor Harris’ secondary argument that the pleadings themselves constitute an agreement not to apply Georgia law and that such agreement fell within Article 3(2) of the Convention which permits the parties to agree at any time to subject the contract to a law other than that which previously governed it.

I heard factual evidence from Mr Goffe and from AAL’s Sales Director, Mr Barry Landymore. On behalf of the Claimants, Mr Ulyatt launched a wholesale attack on the credibility of Mr Landymore, suggesting that he had set out deliberately to mislead the court by suppressing disclosable documents and giving an unfairly one-sided version of events in his witness statements. It is true that AAL’s initial disclosure given on 12 July 2019 consisted of only 14 documents and that further voluntary disclosure was given on 2 October 2019. The explanation for this, set out in correspondence at the time and confirmed by Mr Landymore in the witness box, was that he carried out the disclosure search on his computer. This had been replaced earlier in 2019 and he had been informed by AAL’s IT team that all archived material had been transferred from the old computer to the new one. It was not until he came to review Mr Goffe’s disclosure that he realised there should have been more documents than he had found. Upon making enquiries, he discovered that not all the archived material had in fact been transferred and the Defendant’s further disclosure was generated by a search of this archived material, which included not only his own historic documents but those of the entire business.

There were apparently some problems with arrangements for inspection of the Claimants’ documents but on any view they were available for inspection from at least the end of July 2019. I can therefore well understand the Claimants’ surprise and frustration that AAL does not appear to have reviewed their disclosure until late September. Even allowing for the fact that Mr Landymore was apparently abroad for more than one period, the documents should have been inspected before then and the gaps in AAL’s own disclosure accordingly revealed much sooner.

Nonetheless, I am not satisfied that Mr Landymore was personally at fault in this respect. Still less am I satisfied that he deliberately set out to mislead the court. I was informed by Mr Ulyatt [Day 5 page 7/19-23] that disclosure in relation to the relevant issues was agreed to be given by both sides up to the end of April 2009. It is therefore unsurprising that neither side initially disclosed any significant documentation from after that date. I do not know the basis on which the Defendant’s supplemental list of documents was prepared, but I have no evidence to suggest that it was compiled selectively or in defiance of the Defendant’s duty of disclosure. Accordingly, I reject the suggestion that the Defendant attempted to suppress disclosable documents.

As regards Mr Landymore’s testimony, the criticisms of his witness statement and oral evidence were in my judgment wholly overstated and largely unfair. By way of example, Mr Ulyatt suggested that Mr Landymore had deliberately downplayed the extent of AAL’s connection with the United States by failing to make any mention in his first statement of AAL’s warehouse facility in North Carolina. However: (i) the

I am equally unpersuaded that Mr Landymore’s failure to take steps to verify the accuracy of his recollection about having met Mr Goffe at a trade show in January 2007 warranted a finding that he was careless with the truth and therefore not to be believed unless supported by contemporaneous documentation.

As to Mr Ulyatt’s complaint that Mr Landymore’s written evidence unfairly concentrated on Mr Goffe’s work in the Canadian market to the exclusion of his work in the United States, as stated above, Mr Landymore’s first statement was directed at matters as they stood at the date of the contract, i.e., before Mr Goffe had done any work for AAL at all. With the exception of paragraphs 21-23, it did not stray beyond the end of April. As to these:

Paragraph 21 was demonstrably incorrect in so far as the email to which it refers was sent by Mr Landymore, not by Mr Goffe.

Paragraph 22 merely stated that AAL had been unable to find any evidence of brochures and marketing material being sent to Mr Goffe in Georgia. There is nothing to suggest that this bare statement was anything but true. However, in Mr Goffe’s second statement it had metamorphosed into a positive averment by Mr Landymore that AAL had never sent marketing material to Georgia, and was relied on by Mr Ulyatt as an “egregious” example of selective reliance on cherrypicked documents. As will be clear, however, it did not reflect what Mr Landymore had actually said.

Paragraph 23 stated only that Mr Goffe was working with AAL’s existing customers in Ontario by September 2009. This was undoubtedly true. It did not say or even imply (as Mr Ulyatt sought to suggest it did) that he was exclusively so involved.

Mr Landymore’s second witness statement was simply responsive to points made by Mr Goffe in his second statement, some of which seemed to me to be based on a misreading of Mr Landymore’s first statement. (Footnote: ¹) It was therefore Mr Goffe, rather than Mr Landymore who expanded the scope of the factual evidence into matters post-dating the contract. In so far as Mr Landymore was seeking to address the Defendant’s supplementary disclosure, I have found above that there is no evidence that this was deliberately cherrypicked to support the Defendant’s case.

In any event, it is entirely commonplace for witnesses to emphasise evidence which contradicts the evidence of the opposing side and supports rather than undermines their own case. There is nothing necessarily objectionable in this, although confession and avoidance may often be a more effective way of dealing with unhelpful evidence. However, that is ultimately a question of tactics and, having observed Mr Landymore give evidence, I am satisfied that he was a witness of truth such that I can safely place reliance on his testimony. He answered the questions put to him (some of which, it has to be said, unfairly sought to put words into his mouth or mischaracterised his evidence) carefully, frankly and fully, even when his answers were not necessarily helpful to AAL’s case. He was not argumentative or evasive; he readily made concessions where appropriate and Mr Ulyatt’s suggestion that such concessions had to be wrung out of him does not reflect the reality of his evidence. On other occasions, he stood his ground, but I cannot find any basis for criticism in that. On the contrary, I find that his testimony represented his genuine understanding and belief, subject to understandable lapses of recollection in relation to events which occurred between 10 and 12 years ago. For the most part, his evidence was supported by the contemporaneous documentation and it did not in any event differ significantly from Mr Goffe’s account save in matters of emphasis.

I similarly find that Mr Goffe was a generally truthful and honest witness. Nonetheless, I was unconvinced by his attempts in cross-examination to distance himself from some of his contemporaneous statements as to where he was living and working at the relevant time and to downplay his obvious interest in finding a job which would give him a reason to work at least some of the time in Ontario. In particular, his insistence that he only ever led Mr Landymore to believe that Ginette alone was living in Canada was wholly at odds with the impression created by his emails at the time. Mr Ulyatt invited me to give Mr Goffe credit for candidly admitting that he had deliberately exaggerated his connections with Ontario in order to make himself more attractive to AAL. But that presupposes that Mr Goffe had in fact exaggerated the situation. If in fact the documents accurately reflected the reality of his position, then no credit is due; quite the reverse.

On the basis of the oral evidence and the contemporary documentation, I find the relevant facts to be as follows.

AAL was founded by Mr David Wills and is based in Norfolk in East Anglia. It was one of four main suppliers of poultry processing equipment to North America. The other three were Stork, Linco and Meyn, all of which were much bigger companies. Mr Landymore joined the company in 1979 and became a director in 1989.

AAL is the manufacturer of, in particular, the Easyflow handling system. This was described to me by Mr Landymore as a “shed to shackle” system for the catching, handling and transport of live chickens from the time they are caught until they reach the processing line. Part of the system includes a mild steel galvanised frame (known

That said, some US producers who used water chilling nonetheless purchased Easyflow systems. These tended to be producers who were interested in a high quality product or who handled very large birds where careless handling might cause, not just bruising, but also dislocated limbs. Dislocations are very expensive to deal with.

Easyflow systems sold in the UK and Europe were manufactured entirely in the UK. In the case of systems sold in the United States and Canada, the modules and drawers were manufactured in Ontario. This was principally in order to reduce freight costs but customers also liked the idea that a major element of the system was manufactured locally.

AAL had a set of standard terms and conditions which provided for English law and non-exclusive English jurisdiction. Mr Landymore did not know who originally drafted them, although they have all the hallmarks of having been drafted by lawyers. All he knew was that they had been in use for many years. AAL would always try to use these terms and conditions but many customers wanted to use their own terms in which case there would be a negotiation. Beyond putting their own terms forward, however, AAL did not consciously seek to contract on English law terms.

From 1996, AAL had operated in the Canadian market through Meyn Food Equipment who were based in Ontario. Subsequently it recruited an agent, Ian Taylor, who was also based in Ontario but who covered the entirety of the North American market. In March 1998, AAL incorporated a company in Delaware called Anglia Autoflow Inc. Mr Landymore was not directly involved in the decision to set up the Delaware company as Mr Wills himself handled North America. He understood that it was set up primarily as a spares facility in response to pressure from some major US customers who wanted rapid supply of spares without having to hold stocks themselves or wait for shipment from England. Anglia Autoflow Inc. had a large warehouse in North Carolina and Mr Landymore agreed that the location could well have been chosen because it was ideally situated for the US poultry industry which was concentrated in the so-called “broiler belt” of Arkansas, Mississippi, Alabama, Georgia and North Carolina.

Anglia Autoflow Inc was never a large operation and was never used for manufacturing. Initially it had only one employee dealing with spares but, once it was incorporated, it was thought that it might also be used as a sales base for North America (sales to the

At that time Canada was the main focus of AAL’s sales efforts. Mr Landymore readily accepted that the US poultry market was “massive” and about 18 times bigger than the Canadian market in terms of volume. However, whilst it was theoretically a more valuable market for AAL, this was not the case in practice because the majority of US producers used water chilling and until the market changed predominantly to air chilling there was limited potential for making any further sales of the Easyflow system.

By contrast, the majority of Canadian producers used air chilling and AAL already sold to nearly all the producers in Western Canada. It had not yet acquired any customers in Ontario but this was because the producers there, although they used air chilling, were still using very old and inefficient handling equipment. AAL’s hope was that if Maple Lodge (one of the biggest poultry companies) could be persuaded to change to Easyflow from their old transport system, this would provide the catalyst for other producers also to change. Mr Taylor had not had any great success in selling into Ontario, despite being based there, but in Mr Landymore’s view this was because the market was not then ready to make the move. Eastern Canada was accordingly perceived as a very valuable untapped market (“Eastern Canada was still there to be had”) and was very much the focus of AAL’s efforts in terms of North American sales. In 2007, AAL’s equipment sales in Canada amounted to about $2.3 million, compared to around $500,000 in the United States.

Although Mr Ulyatt sought to ridicule the suggestion that Canada was a more important market for AAL than the United States, this was also a view shared by Mr Goffe himself at the time: see Mr Goffe’s email to AAL dated 30 January 2009 in which he commented that “I believe that the market, at least in Canada, looks to AA as a major equipment supplier, no less than Stork or Meyn. In the US Market that may be different, but in time it ought to be possible to change that perception.” (Emphasis added.) In my view, that comment entirely supports Mr Landymore’s evidence on this point.

Mr Taylor had been very popular and successful with customers in Canada and following his untimely death AAL was actively looking to replace him. Meanwhile, liaison with local manufacturers and customers was being handled by AAL from England, although with some difficulty as it depended on Mr Landymore and another employee travelling out as and when required. AAL therefore wanted someone with the requisite experience to replace Mr Taylor and this in practice meant someone who was already working in the industry. I accept that AAL ideally wanted someone who was based in Canada. Not only would this make liaison with the local manufacturers easier, but AAL’s Canadian customers preferred not to deal with a US-based agent and had an expectation that any replacement for Mr Taylor would likewise be based in Canada.

Mr Goffe is a British national who was born in Somerset in 1962. Virtually the entirety of his working life has been spent in the poultry processing industry. After graduating, he worked in England for a company which exported equipment to the United States and Canada. In 1986 he moved to the United States, somewhat reluctantly, on secondment to the company’s US agent in Kansas City. In the event, however, he found that he liked America and decided to stay. He married an American woman in 1990 and acquired a Green Card in 1991. In 1991 he moved to a job with Jamesway Incubator Company in North Carolina handling worldwide sales. From 1995 to 2005 he was employed by Dapec Inc in Georgia where his role involved both management and sales, the latter almost exclusively to clients in the United States.

In December 1995, Mr Goffe and his wife acquired a vacant plot of land in Woodstock, Georgia where he set about building what he described as his “dream house”. He and his wife divorced in 2005 at which point he acquired sole ownership of the Georgia house. In early 2005, he left Dapec to become Chief Operating Officer of Linco Food Systems Inc., which was also based in Georgia. Again he was involved in both management and sales, principally in the United States although he also travelled to Western Canada.

In late 2004 Mr Goffe met a Canadian citizen, Ginette Denomme (“Ginette”), who was later to become his second wife. She had been working in Ontario as Vice-Principal of a High School in Richmond Hill. In June 2005, she came to stay with Mr Goffe at the Georgia house on a temporary basis during a period of medical leave. They decided that they would like to live there together permanently. Accordingly, she enrolled on a Master’s programme in Community Psychology at Argosy University and in September 2006, with the benefit of a student visa, she moved to Georgia from Ontario with all her belongings. She and Mr Goffe were married in Toronto on 18 August 2007. Ginette’s children continued to live in Ontario, although her daughter also spent some time in Europe.

In August 2007, the Linco group was acquired by a German company called Baader, following which Mr Goffe started looking for a new job.

In about mid-September 2007, Mr Goffe approached Mr Landymore whom he had met on previous occasions. He knew that AAL was interested in the Canadian market and had previously been represented by Mr Taylor whom it was looking to replace. For his part, Mr Landymore knew that Mr Goffe was British and that he had lived in Georgia for some years, although he did not know his precise legal status there. He knew that Mr Goffe had previously worked for Linco and Dapec.

It seems that a telephone conversation took place between the two men in which Mr Goffe made Mr Landymore aware that he was also considering an offer from another company. No detailed discussions took place on this occasion but Mr Landymore recollected speaking about the potential to sell more equipment into Western Canada and the even greater potential for sales in Eastern Canada. They agreed that Canada had a lot of potential compared to the United States and Mr Goffe expressed interest in handling the Canadian market. For the avoidance of doubt I find that neither Mr Goffe

On 14 September 2007, Mr Goffe sent Mr Landymore an email recording the terms of a job offer which, subject to clarification of certain outstanding matters, had been agreed in principle. These included a gross annual salary of US$100,000 (albeit Mr Goffe would be declared as self-employed for US tax reasons), plus 3% commission on sales in North America. Mr Goffe apologised for putting pressure on AAL to answer the outstanding points immediately but explained that he had promised the other company a decision that day.

Mr Landymore responded saying that he appreciated that Mr Goffe had a deadline to accept the other offer but that this was an important decision and that AAL would prefer to have the weekend to consider it. In his oral evidence he explained that he had not been convinced that Mr Goffe was prepared to move to Canada but that he did not push the point because he did not want to scare him off altogether. Nonetheless, it would have been a big step for AAL to employ a representative in Georgia given its previous loss-making operation in the United States. In Mr Landymore’s view, AAL’s best tactic was therefore to delay responding in order to see whether Mr Goffe accepted the other offer. Mr Landymore realised that this was likely to be from one of AAL’s larger competitors whom AAL would be unable to outbid, but if Mr Goffe nonetheless turned it down and came back to AAL, this would be a good indication that he was really interested in working for AAL, in which case Mr Landymore hoped to persuade him to make the move to Canada. Ultimately, however, his geographical location was less important than his willingness to devote time to the Canadian market. I accept this evidence which I find entirely plausible.

In the event, Mr Goffe did accept the other offer which was in fact from Stork Gamco Inc in Georgia, for whom he became Regional Sales Manager covering the Eastern US and Western Canada. However, he was unhappy at Stork from the outset and in January 2008 he contacted Mr Landymore again, although apparently without any tangible outcome.

By April/May 2008, it had transpired that Ginette’s student visa did not permit her to undertake an internship in the United States. Although she was eligible to apply for a Green Card, she would have been unable to work in the United States or travel to and from Canada during the pendency of the application without running the risk of having her application rejected and having to start all over again. She and Mr Goffe therefore decided that the only solution was for her to move back to Ontario for one year while her application was processed, whereafter she would return to Georgia permanently. Meanwhile, Mr Goffe would try to spend as much time as possible with her in Ontario, principally at weekends and during holidays. In May 2008, Ginette therefore accepted a job as Vice-Principal of Aurora High School in Ontario. Also in May 2008, Mr Goffe acquired US citizenship thereby becoming a dual citizen. Thereafter he registered to vote in Georgia.

In an email to one of his friends dated 13 May 2008, Mr Goffe stated that “Ginette has to go back for a year for immigration reasons… Essentially we plan to be based in Canada (we are going to move the dogs) but I will stay in Georgia.” In an email sent to another friend on 30 May 2008, he expressed frustration that his current job did not include Ontario and that he therefore had no good reason to travel to Toronto on business.

In his witness statement, Mr Goffe said that by early 2008 he was unhappy at Stork and had decided to look for a job which would give him greater flexibility and allow him to spend more time in Ontario. I do not accept, as he sought to suggest in his oral evidence, that he was only looking for a job which would allow him to spend more weekends in Ontario. On the contrary, I find that he would have been overjoyed to find a job that gave him a reason to work in Ontario, at least in the short term.

In these circumstances, it is unsurprising that Mr Goffe emailed Mr Landymore yet again on 26 May 2008 to find out what the position was regarding AAL’s North

Although Mr Goffe initially considered buying a property in Ontario, this turned out not to be financially viable and in July 2008 he and Ginette found a property in Richmond Hill which Ginette leased in her own name. She alone paid the rental and although the telephone and internet connections were set up in Mr Goffe’s name and he may have paid the installation costs, Ginette paid for all other running expenses of the property. She moved to Richmond Hill in about July 2008 with the couple’s dogs and some of the furniture that she had previously taken to Georgia in 2006. There is no evidence to suggest that Mr Goffe ever made AAL aware that he had not in fact bought a house in Canada. Subsequent references to “my house in Canada” (see below) would therefore only have served to reinforce any impression that he owned property there.

I accept that in 2008, this was intended only to be a temporary arrangement and that it was still Mr Goffe’s intention eventually to live permanently in Georgia with Ginette. Meanwhile, however, it is clear to me that he moved his domestic base to Ontario in the latter part of 2008 and spent as much time there as he could. In practice this meant not only weekends but also other periods when he was either not working or was working in Eastern Canada. When it was put to him in cross-examination that he was unhappy about living apart from Ginette, he denied this saying, “We did live together pretty much.” This is consistent with his further evidence that even though he regarded the Georgia house as also being a family home, “We just weren’t there.”

By September 2008, Mr Goffe was even more dissatisfied with his job at Stork. An attempt to meet Mr Landymore during a visit to England in July 2008 had been unsuccessful and on 23 September 2008 (shortly after the financial crash), Mr Goffe emailed Mr Landymore again, commenting that the US market continued to be very poor and that 2009 looked bleak. By contrast, “the Canadian market has been pretty good, the exception is likely to be Maple Leaf who will be struggling to recover from a costly recall of cooked meat after listeria has killed over a dozen… If you would like to talk further I am still very interested, I feel that I have good experience and contacts that would be very useful for AA.” In his witness statement, Mr Goffe said that he was very keen to work for AAL and needed Eastern Canada to be within any area given to him. It was for this reason that he included market commentary on Eastern Canada specifically.

On 14 December 2008, Mr Goffe emailed Mr Landymore suggesting a meeting at the Atlanta show in January. He referred to the dire state of the market in the US, noting that “Canada is still OK”, but that in his view there was a good chance of a return to profitability in the early part of 2009 and that “In the long term though I think that there is still great potential in North America for superior live bird handling and stunning and also catching systems.” Mr Landymore confirmed that he would be attending the show and on 18 December 2008, Mr Goffe emailed to say that he would like to speak further. He said “Tonight I am flying to Toronto and Friday I will be working at my house in Canada” and gave his phone numbers there. Mr Landymore’s evidence was that he was reluctant to talk in advance of the show given that Mr Goffe was working for one of AAL’s major competitors. However, he was encouraged by the email as he understood from it that Mr Goffe now had a base in Canada.

The two men did indeed meet on 29 January 2009 at the Atlanta show where they had a general conversation during which Mr Goffe expressed great enthusiasm for becoming AAL’s sales representative in North America and handling the Canadian market. Mr Goffe did not speak in any detail about his personal circumstances although he explained that he was splitting his time between Georgia and Ontario. On 9 February 2009, he sent a follow-up email which emphasised his interest in talking further and stated that “I think that there is a lot of immediate potential in the Canadian market”. In his response on 11 February 2009 Mr Landymore said:

I find that this very specific query was designed to obtain explicit confirmation of what Mr Landymore had inferred from the 18 December 2008 email about Mr Goffe now living in Canada.

Mr Goffe responded the following day. As this is an important email, I set it out at some length. He noted that:

In his first witness statement, Mr Goffe categorised this email as a sales pitch. He said that he was giving Mr Landymore the answers he thought AAL wanted to hear, and this included leading Mr Landymore to believe that he would be living in Canada. On his own admission, therefore, he was fully aware that AAL’s primary interest was in Canada and that it ideally wanted someone based there. In a further attempt to make himself more appealing to AAL, he also emphasised his English connections. I find that the Defendant in the person of Mr Landymore reasonably understood both from this and from Mr Goffe’s previous emails that: (i) he had moved to Canada; (ii) he only required the Georgia house for the purposes of his work with Stork; (iii) but for current market conditions, he would have sold it; and (iv) Mr Goffe shared AAL’s view that Canada was currently a more important market for it than the United States.

On 25 March 2009, Mr Goffe visited AAL’s premises in Norfolk. It is common ground that agreement in principle was reached by the end of the day that Mr Goffe would be employed by AAL as an independent agent remunerated on a commission basis and that he would operate through a company yet to be incorporated. Mr Goffe was shown a draft agreement before he left and possible changes were discussed. I accept Mr Landymore’s evidence that the draft was based on a form of agreement which AAL had used for many years with other agents although there is no evidence that Mr Goffe was aware of this at the time. Mr Landymore did not know who had drafted this agreement or why clause 11 had been included. While changes to some of the commercial terms were discussed, Mr Goffe and Mr Landymore agreed that neither of them mentioned or focused on clause 11 or the governing law of the agreement.

Names for the new company were also canvassed and on 29 March 2009, Mr Goffe suggested that “Anglia Autoflow North America” would be preferable to “American Autoflow” which was unlikely to appeal to Canadians. On 31 March 2009, Mr Landymore sent Mr Goffe a draft of the agreement incorporating the changes they had discussed together with some further points that had occurred to him. The agent was identified simply as “PG Co.” in the draft. As far as clause 11 was concerned, Mr Goffe’s written evidence was that he could not recall when he first read it, but that he understood it to mean that disputes would be dealt with in the English courts. “I thought this meant the English Courts applying English Law (I had no comprehension that it was even possible for a Court of one country to apply the laws of a different country).” He had no objection to this, not least because he did not expect to have any disputes, so he did not think it would ever be used.

The only further discussions which took place related to the name of Mr Goffe’s company which was agreed on about 3 April 2009. On 8 April 2009, Mr Landymore emailed a final draft of the agreement incorporating the name “Anglia Autoflow North

On 9 April 2009, Mr Goffe signed the agreement while physically present at the Richmond Hill house over the Easter weekend. It has not been possible to locate any copy of the Agency Agreement signed by AAL, but it is not disputed that there was a concluded agreement from that date.

It is apparent from Mr Goffe’s reconstructed diary that from the beginning of March until 9 April 2009, he had spent 3 nights in Georgia, 13 nights in Ontario and 24 nights travelling. This was the subject of much heated debate on both sides and I shall have to return to it later. However, it is not controversial (or, if it is, I find) that:

Mr Goffe was formally employed by Stork until 23 April 2009 although he ceased to carry out any actual work for them after handing in his notice on 11 April 2009.

His work for Stork in March/early April involved only 3 nights in Georgia, the last of which was on 13 March. He only returned to Georgia thereafter for his exit interview.

From 14-20 March he was on holiday in Mexico with his wife.

From 21-27 March he was in England for the purpose of visiting AAL as described above. I assume that he was on leave during this time.

He undertook work trips to British Columbia (2-3 March), Ontario (6-7 March), Philadelphia (9 March), Washington DC (12 March) and British Columbia (2-7 April).

Otherwise, he was in Ontario with his wife.

Professor Harris was therefore correct to submit that, as a matter of fact, Mr Goffe carried out no work for Stork in Georgia after 13 March 2009.

The extent to which matters post-dating the contract can or should be taken into account in determining the issues identified in paragraph 12 above was a matter of some debate before me and is addressed further below. It is therefore necessary for me to address the subsequent history of events, albeit relatively briefly.

On 11 April 2009, Mr Goffe handed in his notice to Stork and informed Mr Landymore that he could deal with any visits or calls that needed to be made in Ontario the following week. On the same day he purchased a new Canadian cell phone so that it would be easier for Canadian customers to call him. On 14 April 2009, he gave Mr Landymore his new Canadian cell phone number and said that he was planning to use it as his primary number although he would be keeping his US number as well. He also gave Mr Landymore instructions for both Canadian and US business cards and was in due course sent two sets of cards which he used for Canadian and US customers respectively.

On 15 April 2009, Mr Goffe emailed AAL’s Business Development Manager giving his contact information. This comprised his Richmond Hill postal address and land line, and both his Canadian and US mobile phone numbers. He also enquired about leasing a Mercedes car from a car dealership in Ontario to which he gave his Canadian address.

On 21 April 2009, GDE was incorporated as Anglia Autoflow North America LLC. Mr Goffe carried this out himself via a website called Legal Zoom without taking any legal advice. His understanding was that an “LLC” type corporation was the most suitable for small businesses and he chose the cheapest option he could find. He did not consider incorporating the company in Canada as he was a US citizen and intended to continue being permanently resident in Georgia. He did not have a residency permit for Ontario; he had no idea whether or how he could incorporate a company there and he did not want to spend any money investigating the position. Accordingly, he used Legal Zoom’s templates for all the company’s documents and specified the Georgia house as its principal mailing address as this was where he intended the company be to be legally based. However, GDE had no trading premises. To all intents and purposes, it was synonymous with Mr Goffe who was a travelling salesman and ran the business on a laptop from wherever he happened to be.

On 23 April 2009 Mr Goffe attended his exit interview at Stork, having returned to the United States on 21 April 2009. He flew back to Ontario on 25 April 2009 and emailed Mr Landymore the following day to say that he was “back in Toronto, ready to start letting customers know where I am…” and enquiring about his AAL email address. His personal diary for 27 April to 1 May 2009 is annotated“home office” and he did not

Mr Goffe’s first invoice to AAL was rendered on 12 September 2009. It specified the Georgia house as GDE’s address and nominated a US bank account in Virginia for payment. It does not appear that Mr Goffe in fact had any bank accounts in Georgia itself; his rather careful written evidence said only that “All of my bank accounts were in the US” and the clear inference, given the nature of the present dispute, is that he did not. On any view, however, Mr Goffe did not open a Canadian bank account until 2010. Thereafter, invoices in Canadian dollars used Mr Goffe’s Canadian address as GDE’s address.

Under Georgia law, GDE was not required to file a separate tax return or pay corporation tax. All its revenues were attributed to Mr Goffe as its sole member and recorded on his personal tax returns. It is not disputed that at all material times Mr Goffe paid all his taxes in Georgia, both state and federal.

Although, as I accept, Mr Goffe and Ginette had originally intended that she only move back to Canada for one year, their plans subsequently changed. For one thing, it transpired that she needed to complete 4 years work in order to qualify for a full state pension. In addition, Ginette unfortunately suffered a seizure later in 2009 which required specialist medical attention. They therefore decided that the arrangement would continue for a further three years. For a short period of about 6 months from July 2009 to December 2009/January 2010, part of the Georgia house was rented out to a friend of a friend. In July 2010, Ginette purchased a house in Sharon, Ontario. Mr Goffe contributed to the purchase price although the house and mortgage were in Ginette’s sole name. The Sharon house was bigger than the Richmond Hill house and allowed Mr Goffe to have a proper home office for the first time.

I accept that following the conclusion of the Agency Agreement, Mr Goffe pursued leads throughout North America and not exclusively in Canada. I also accept that (1) when dealing with US customers he would give his US contact details, and when dealing with Canadian customers he would give his Canadian contact details; and (2) that he ran the business from his laptop and that if he was required to do any work when he was not travelling for work, he would do it wherever he happened to find himself, be that Georgia or Ontario or elsewhere. His unchallenged evidence was that he procured the sale of five Easyflow systems during the course of the agency: two in the United States in the second half of 2010; two in Canada in mid-2011 (including Maple Lodge) and a further one in California in May 2012.

Nonetheless, Mr Goffe himself accepts that, although he continued to travel a lot for his work and stayed at the Georgia house from time to time, he spent more time in Ontario than in Georgia from April 2009. Although he attempted in his oral evidence to confine the time he spent in Ontario to “most weekends”, I find the reality was, as he accepted, that whenever work commitments did not require otherwise, he was – for entirely understandable reasons – in Ontario with his wife. This is confirmed by an email he sent to Sandra Smith at AAL on 10 December 2010 referring to his “Home office” in Sharon and his “US Office” in Georgia and asking her to note that “usually I am in Canada”. Although Mr Goffe had no residence or work permit for Canada, his US passport entitled him to visit for up to six months at a time for personal reasons or for business, such as making sales. As set out above, I accept that Mr Landymore and

In his oral evidence, Mr Goffe said that he had originally intended to use his Canadian phone as his primary number for calls and his US Blackberry as his primary number for emails and internet access. In the event, however, this turned out to be too expensive, and he ended up using the US phone as his primary number for both emails and calls in both the United States and Canada.

The evidence before me shows that supplies of leaflets and other marketing material were sent by AAL to Mr Goffe in Ontario save for one occasion in January 2012 when he asked for leaflets to be sent directly to Georgia for use at the Atlanta show. This was because he was flying down from Ontario and could not bring any with him as he would otherwise normally have done.

As the agency progressed there is also evidence of Mr Goffe performing a liaison function between customers and the local manufacturers in Ontario, such as Stratus. Mr Landymore’s evidence was that AAL had always intended Mr Goffe to perform this role. It had been explained to Mr Goffe that this was part of what Mr Taylor had done and AAL understood that he would be based in eastern Canada and thus well-placed to do the same. Mr Goffe said it was more a question of him gradually assuming the role as he became more knowledgeable and experienced. On any view, it was not expressly stated to be part of his job in the Agency Agreement.

There is a particular sequence of correspondence in May 2010 which was relied on by the Claimants. This concerned a problem with the colour of certain drawers manufactured by Stratus and was complicated by a degree of confusion between two different orders. On 13 May 2010, Mr Landymore emailed Mr Goffe saying that “To avoid any confusion it would be better if Cheryl only takes instructions from one of us, as we have always placed the orders in the past I think it is best that we carry on doing so.” So far as relevant, I find that this was not an instruction to Mr Goffe not to involve himself with the local manufacturers at all, simply a direction that instructions should come from either Mr Goffe or AAL and that, in relation specifically to the initial placing of any orders, AAL had always done this historically and it would probably be best if it carried on doing so.

It is common ground that the agency agreement came to an end at the latest by 6 April 2013, although the circumstances in which this occurred form part of the underlying dispute. In September 2013, Mr Goffe purchased a condominium in Florida, which thereafter became his and Ginette’s permanent residence. The Georgia house was sold in June 2014 following which GDE’s principal mailing address was also changed to the new address in Florida.

The relevant terms of the Rome Convention have been set out above. As is plain from the Preamble and Article 18, one of the main objectives of the Convention was to introduce certainty and predictability into the identification of the governing law of a contract. It has been confirmed in numerous authorities and was common ground before me that the court should therefore adopt an international, purposive approach to

He expressed his agreement and, in para 26, approved my own view expressed in Egon Oldendorff v Liberia Corp [1996] CLC 482 at 505, where I said:

Although those views were expressed in the context of Article 3, they seem to me to apply equally to Article 4: see also to the same effect Plender & Wilderspin on the European Contracts Convention (2001) at para 2-01. I should perhaps stress that in applying Article 4 an English court should not be influenced by the old common law approach to the proper law of the contract because the nature of the enquiry under Article 4 is fundamentally different: see e.g. Credit Lyonnais v New Hampshire Insurance Co [1997] CLC 909.”

Section 3(3) of the Contracts (Applicable Law) Act 1990, which enacts the Rome Convention into domestic English law, provides that the report on the Convention by Professor Mario Guiliano and Professor Paul Lagarde is an authoritative aid to interpretation, which the court is entitled to take into account in ascertaining the meaning and effect of any provision of the Convention.

So far as the general scheme of the Convention is concerned:

Priority is accorded to the intention of the parties, who have complete freedom of choice under Article 3 as to the law to be applied;

In default of any such choice, the court must determine the governing law on the basis of the connecting criteria set out in Article 4;

The general principle underlying Article 4 is set out in Article 4(1), namely that the contract should be governed by the law of the country “with which it is most closely connected”;

The application of that general principle is achieved with the assistance of Articles 4(2)-4(4), containing a series of mandatory presumptions as to the country with which a contract is deemed to be most closely connected;

Article 4(5) is an exceptions clause which nonetheless permits the presumptions to be disregarded if the circumstances as a whole show that the contract is more closely connected with another country.

There are a number of points to note about this general scheme.

First, as explicitly recognised by Clarke LJ in Iran Continental Shelf Oil Co. v IRI International Corp (supra), there is an important distinction between Articles 3 and 4. Article 3 is looking for a mutual choice of law by the parties, whether or not expressly stated. Article 4 on the other hand is looking for a connection between the contract and a particular country. Thus, specific factors may carry more or less weight depending upon the context in which they are being considered.

Secondly, the overriding objective of Article 4 is to identify the country with which the contract is most closely connected. It is for that reason that Article 4(5) positively requires the presumption in Article 4(2) to be disregarded if it leads to a result which is not consonant with the general principle of closest connection set out in Article 4(1): Intercontainer Interfrigo SC v Balkenende Oosthuizen BV, (supra) at [58]-[60]. I therefore agree with Mr Ulyatt that the presumptions in Article 4 are relatively blunt instruments which may readily be displaced. In the words of the Guiliano-Lagarde report at page 22:

Thirdly, despite some superficial similarities, the scheme established by the Convention for the identification of proper law is not the same as the approach previously adopted at common law. This was made clear by the Court of Appeal in Lawlor v Sandvik Mining and Construction Mobile Crushers and Screens Ltd, [2013] EWCA Civ. 365;

The court then continued:

In view of the potential difficulty in drawing a line between inferring an unexpressed intention and imputing an intention, the requirement of article 3 that the choice must be demonstrated with reasonable certainty is significant. The party asserting that there has been a choice of law has the burden of establishing it with reasonable certainty.

It would be a mistake to attempt to apply article 3 through the prism of the preceding common law. Article 18 of the Convention requires the court to have regard to its international character and to the desirability of achieving uniformity in its interpretation and application.”

Finally, there is the extent to which the court may take account of supervening events and factors in applying the Rome Convention. As to this, it was common ground that Article 3(1) looks to the date of the contract but that it is permissible to take account of subsequent events to the extent that they demonstrate what, if anything, the parties impliedly agreed at that date: Egon Oldendorff v Liberia Corp, [1996] 1 Lloyd’s Rep. 380, 384 per Clarke LJ; Lawlor v Sandvik Mining and Construction Mobile Crushers and Screens Ltd, [2012] 2 Lloyd’s Rep.25 at [12], [47] per HHJ Mackie QC (noted without adverse comment by the Court of Appeal). See also the obiter comments of Simon J in Lürssen Werft GmbH & Co. KG v Halle, [2009] EWHC 2607 (Comm); [2010] 2 Lloyd’s Rep. 20 at [38]. (The point was left open in the Court of Appeal.)

Since Article 4(2) likewise expressly looks to the time of conclusion of the contract, I accept, by parity of reasoning, that the same approach applies.

Less certain is the correct approach to Articles 4(1) and 4(5). Unlike Article 4(2), these articles do not expressly refer to any particular point in time for determination of the closest connection and according to the Giuliano-Lagarde report (page 20): “In order to determine the country with which the contract is most closely connected, it is also possible to take account of factors which supervened after the conclusion of the contract.” While this was said in the context of Article 4(1), it cannot sensibly be suggested that the approach to Article 4(5) should be any different.

A possible justification for the distinction is that under Article 4(5) the court is not looking for indications of a positive choice of law – a situation where it would make sense to confine the enquiry to the date of the contract (given, in particular, the express contemplation of subsequent consensual change) – but is expressly required to take account of the circumstances as a whole as part of a much broader investigation into factors which connect the contract with a particular country. Certainly, HHJ Mackie in Lawlor v Sandvik Mining and Construction Mobile Crushers and Screens Ltd (supra) was prepared to take account of supervening factors in arriving at his decision under Article 4(5) and that aspect of his decision was not appealed to the Court of Appeal.

The idea that supervening events are relevant to the identification of the proper law of a contract may look odd to English eyes where the common law requires a governing law to be established at the date of the contract and eschews the idea of a floating proper law. Mr Ulyatt accordingly submitted that the court should adopt a similar approach as applies to Articles 3(1) and 4(2), namely to distinguish between “backward looking” events which cast light on the expectations of the parties at the date of the contract, and completely unanticipated events. He referred to the 14^th edition of Dicey, Morris & Collins at paragraph 32-059 (maintained in the 15^th edition at paragraph 32-037), which recognised that it would not be in keeping with the spirit of the Convention to allow the English common law approach to defeat the intentions of the parties and hesitantly suggested that the court should be entitled to take account of subsequent conduct “at least to the extent that it sheds light on … the country with which the contract is most closely connected at the time the contract was concluded.”

The point is also discussed in Plender & Wilderspin The European Private International Law of Obligations (4^th ed., 2015) at paragraphs 7-073 to 7-074 where the learned authors discern no good reason for taking supervening events into account at all. They suggest that if the Giuliano-Lagarde report is correct, subsequent events can only be considered to the extent that this would be permitted by all possible putative governing laws. If those laws differ on the point, then the law of the forum should apply.

It is clear that I must adopt an international and purposive approach to the interpretation of the Convention, but this particular point is one which appears to be open on the authorities. I am not convinced that the answer lies in the solution suggested by Plender & Wilderspin which would entail a potentially costly and burdensome investigation into at least one and possibly several foreign laws regarding the admissibility of postcontractual conduct. With all due respect to the learned authors, this seems to me to be straying rather too far from the Convention’s objective of achieving uniformity in the application of the rules for the ascertainment of governing law.

Nonetheless, the distinction suggested by Mr Ulyatt clearly has some validity, although it must be balanced against the fact that the Convention itself mandatorily requires the court to consider “the circumstances as a whole” without imposing any temporal limitation on that consideration. In my view, therefore, while there is no conceptual limit to the circumstances which may be taken into account for the purposes of applying Articles 4(1) and 4(5), less weight is to be attached to unforeseen and purely adventitious post-contractual events than to those which may simply be said to represent the normal working out of the contract.

Against that background, I turn to the specific articles of the Convention.

There was no express choice of law in the Agency Agreement. The question is therefore whether the parties nonetheless made an implied choice of law. As made clear by the Court of Appeal in Lawlor, it would be a mistake to apply Article 3 through the prism of the preceding common law, notwithstanding that the respective tests have a passing resemblance and raise many of the same considerations. Accordingly, I do not accept Mr Ulyatt’s submission that the similarity in wording between passages in the speeches in Cie d’Armement Maritime SA v Cie Tunisienne de Navigation SA, [1971] A.C. 572 and passages in the Giuliano-Lagarde Report either authorises or justifies the court in adopting the common law approach to the ascertainment of an implied choice. The very real change in approach which the Rome Convention introduced is made clear in the Giuliano-Lagarde Report at page 17 where is it explicitly stated that:

Moreover, Article 3(1) itself stipulates that a choice which is not express must be demonstrated with “reasonable certainty” by the terms of the contract or the circumstances of the case. There was some debate before me as to whether the requirement for “reasonable certainty” was the same as the requirement in the later Rome I Regulation for any choice of law to be “clearly demonstrated”. The view taken by Dicey, Morris & Collins (op.cit.) (15^th ed.) paragraph 32-059 fn 217 and approved by Popplewell J in Aquavita Interational SA v Ashapura Minecham Ltd, [2014] EWHC 2806 (Comm) at [20] is that the change was effected simply in order to bring the English and German texts into line with the equally authoritative French text which reads “de façon certaine”. Since Article 33 of the Convention provides that all the approved language texts are equally authentic, it seems to me that this is overwhelmingly the likely reason.

I accordingly proceed on the basis that no significance is to be attached to the change in wording and that “reasonably certainty” requires any implied choice to be clearly demonstrated. In the words of the Giuliano-Lagarde Report, there must have been a real choice which the parties had the clear intention to make.

The application of the test has been further clarified by the Court of Appeal in Lawlor v Sandvik Mining and Construction Mobile Crushers and Screens Ltd (supra) at [31][33] as follows:

Logically there may be a certain artificiality in attributing to the parties a tacit choice in circumstances which do not suggest that they gave actual thought to the matter, as Redfern and Hunter comment in their book on International Arbitration, 5th Edition, 2009, at para 3.206. However, one can see the justice of inferring a choice of law in circumstances where it would not reasonably have occurred to the parties to suppose that a different law might apply. It would lack practical sense to require that they should have contemplated that which would not reasonably have occurred to them.

The objective nature of the test means that the party asserting an implied choice of law has to satisfy the court to the required standard that, on an objective view, the parties must have taken it without saying that their contract should be governed by that law – or, in Lord Diplock's formulation, that the contract taken as a whole points ineluctably to the conclusion that the parties intended it to be governed by that law. He does not have to prove that there was in fact a subjective conscious choice (for, as I have said, evidence of subjective intention would be inadmissible), but he does have to satisfy the court that the only reasonable conclusion to be drawn from the circumstances is that the parties should be taken to have intended the putative law to apply.”

This is a high hurdle. The court is not looking for the choice that the parties probably would have made if they had turned their minds to the question: Giuliano-Lagarde Report page 17; Egon Oldendorff v Liberia Corp (supra) at 387-388. It is equally insufficient to show that a particular choice would have been reasonable: Amico v Cellstar Corp, EWCA Civ. 206 at [44] per Mance LJ. Moreover, given the difference between the nature of the investigation under Article 3 and that which is required by Article 4, the court can and should not derive an implied choice merely from the fact that there are considerations which link the contract with a particular country: see paragraph 103 above.

I also agree with Simon J in Lürssen Werft GmbH & Co. KG v Halle at [33(3)] that the court should not strain to find a choice of law given the safety net provided by Article 4.

In the present case the evidence established that there was no reference by the parties to the question of governing law at all. I accept that no adverse inference can be drawn against the Claimants from that fact alone; it is merely one of the objective circumstances to be taken into account. However, as indicated above, the Claimants’ primary case was that a real choice of English law could be implied from the agreement of the parties to the jurisdiction clause in clause 11 which, for convenience, I set out again here:

A large part of the argument before me on both sides was accordingly devoted to submissions that this clause either was (Mr Ulyatt) or was not (Professor Harris) sufficient to constitute or evidence the necessary “real choice”. (Footnote: ³) Two basic points were not in dispute:

A jurisdiction clause may permit an inference that the parties have chosen the law of the selected jurisdiction: Dicey, Morris & Collins (op.cit.) (14^th ed.) at paragraph 32-095. The position is the same under the Rome I Regulation where a proposal for a positive presumption to this effect was rejected: Dicey, Morris & Collins (op.cit.) (15^th ed.) at paragraph 32-063. See also Marubeni Hong

The inference is stronger where the jurisdiction clause in question is exclusive: Dicey, Morris & Collins (op.cit.) (15^th ed.) paragraph 32-063 fn 239. Clearly a clause which imposes an obligation to litigate all disputes in England is a more significant pointer towards a choice of English law than a jurisdiction clause which merely permits proceedings in England without precluding suit being brought elsewhere.

In these circumstances, I was referred to a substantial body of authority and learned commentary on whether clause 11 was an exclusive or non-exclusive jurisdiction agreement. I note that a potential threshold question might have arisen as to the appropriate system of law that I should apply in construing clause 11. The orthodox approach would be to apply the putative proper law, i.e., English law on the Claimants’ case and Ontario law on the Defendant’s case. However, neither party suggested that I should determine the point otherwise than by reference to English law and that is the basis on which I accordingly proceed.

In support of his argument, Mr Ulyatt referred to a number of authorities on Article 25 of the Brussels Recast Regulation and its predecessor, Article 23(1) of Regulation 44/2001 (which was accepted to be materially the same for present purposes). He also took me to case law on the inferences to be derived from arbitration and jurisdiction clauses at common law although I found these to be of limited utility in circumstances where I am required to adopt an autonomous approach under the Rome Convention.

Moreover, I doubt in any event whether any real assistance can be derived from shipping cases concerning arbitration clauses as these seem to me to raise different considerations. As Dicey, Morris & Collins (op.cit.) (14^th ed.) points out at paragraph 32-096, the inferences to be drawn from an arbitration clause depend on the circumstances, and a significant factor in many of these cases was that London is a preeminent centre for maritime arbitration which had been chosen as a “neutral” forum by parties of different nationalities in circumstances where there was a specific requirement for arbitrators with specialist shipping expertise. In such a case, it is not difficult to infer that a choice of London arbitration is also a real choice of English law: see Egon Oldendorff v Liberia Corp (supra) at page 386-387, which was just such a case. But even at common law, a London arbitration clause, albeit a weighty pointer towards English law, was not conclusive and might have to yield to other indications: Cie d’Armement Maritime SA v Cie Tunisienne de Navigation SA (supra).

Turning to the Brussels Recast Regulation, Article 25 provides as follows:

Mr Ulyatt argued that this created a rebuttable presumption of exclusivity, while Professor Harris submitted that there was no such presumption and that it was simply a question of construing the clause. Ultimately it seemed to me that the parties were dancing on the head of a pin in relation into this point. Article 25 is undoubtedly inelegantly phrased and it raises a number of difficulties. In particular, it is unclear to me whether “any disputes” means “any” in the sense of “some”, or “any” in the sense of “any and all”. That may not matter, however, as I accept Professor Harris’ submission based on the comments of Professor Briggs in Private International Law in English Courts (2014, OUP) at paragraph 4.186 that the words “unless the parties have agreed otherwise” were likely to have been inserted to remedy the unsatisfactory situation under the predecessor provision in the Brussels Convention whereby even an expressly non-exclusive jurisdiction clause was arguably rendered exclusive.

Nonetheless, I find it difficult to see how the current wording in Article 25 does not create a presumption, and this may or may not affect the burden of proof depending on which party is arguing the point. That said, I agree with Professor Harris that whether or not the parties have “agreed otherwise” is primarily a question of construing their agreement: Briggs (op.cit.) paragraphs 4.185-4.186; Briggs on Civil Jurisdiction and Judgments (6^th ed., 2015, Informa) paragraph 2.139; (Footnote: ⁴) Dicey, Morris & Collins (op.cit.) (15^th ed.) paragraph 12-105; Joseph on Jurisdiction and Arbitration Agreements and their Enforcement (3^rd ed., 2015, Sweet & Maxwell) paragraph 4.13.

Whilst theoretically possible, it seems very unlikely that the court would need to resort to a presumption in order to determine a question of construction but, in so far as it may be necessary, I hold that the burden is on the Defendant to satisfy me that clause 11 is non-exclusive.

I start by rejecting Mr Ulyatt’s argument that the presumption can only be rebutted by an express provision for non-exclusivity. That would obviously be the clearest and simplest way of making the position clear, but in my view it is not the only way. Moreover, the argument that the parties could not impliedly stipulate for non-exclusive jurisdiction sat very uneasily with his case that they could nonetheless impliedly choose a governing law. Nor was it supported by any of the authorities or commentators on which he relied. These, on the contrary, suggested that it was simply an ordinary question of construction: Perella Weinberg Partners UK LLP v Codere SA, [2016] EWHC 1182 (Comm) at [23]; Dickinson & Lein: The Brussels I Regulation Recast (2015, OUP) paragraph 9.83 fn 177.

In any event, I agreed with Professor Harris that this debate, interesting though it was, risked losing sight of the wood for the trees. For the purposes of Article 3(1), the only relevant question is whether, construing the contract as a whole, the parties demonstrated with reasonable certainty a genuine positive choice of English law. Whether clause 11 is to be categorised as exclusive or non-exclusive is therefore only a means to that end and not an end in itself: Perella Weinberg Partners UK LLP v Codere SA (supra) at [20].

The same thought finds resonance in the very recent comments of Baker J in Enka Insaat ve Sanayi AS v OOO Insurance Company Chubb, [2019] EWHC 3568 (Comm) where he counselled that “there is limited utility in seeking to formulate or proceed to a decision from generally stated rules or presumptions as to the strength with which a choice of seat does nor may convey or imply a choice of governing law for the arbitration agreement.” While the point at issue before him was slightly different, the sentiments expressed are equally apposite.

With those cautionary remarks well in mind, I turn to clause 11 itself. Professor Harris argued that it contained two fundamental conceptual uncertainties as a result of which it could not conceivably demonstrate a choice of English law with any degree of certainty, reasonable or otherwise:

First, on its face the clause only applied to “serious disputes” without providing any yardstick for assessing whether a dispute was serious or not. Was it referring to the amount in dispute? The nature of the dispute? The seriousness of the disagreement? Or something else and, if so, what?

Secondly, an agreement “to subject themselves” is inapt to create any obligation to refer disputes to the English courts. The clause therefore amounted only to an agreement by the parties not to contest English jurisdiction without imposing any corresponding obligation not to sue elsewhere.

I should record Mr Ulyatt’s submission that the first point was not open to the Defendant on the pleadings, although he did not formally object to it. In my view, he was right not to do so, as both parties had expressly reserved the right in their pleadings to rely on all documents for their full terms, meaning and effect. In any event, I was not persuaded by Professor Harris’ argument on the point. My initial view, which subsequent reflection has done nothing to change, is that on an objective construction, particularly in the light of clause 10 which imposed an obligation of mutual cooperation, the clause was referring to any dispute which the parties were unable to resolve amicably.

Support for this view can be found in the approach of the House of Lords in Fiona Trust & Holding Corp v Privalov, [2007] UKHL 40; [2007] Bus. L.R. 1719, where the question was whether claims in tort for bribery, conspiracy and breach of fiduciary duty fell within the scope of an arbitration clause referring to “any dispute arising under this charter.” Against a background of fine distinctions in the case law between arbitration clauses referring to disputes “under this charter”, “in connection with this charter”, “arising out of this charter” and the like, Lord Hoffman held that although it was ultimately a question of construction, the court should start from the presumption that rational businessmen were likely to have intended any dispute arising out of the relationship into which they had entered to be decided by the same tribunal. Likewise in this case, I can see no rational reason why the parties should have agreed that some but not all unresolvable disputes should be determined by the English courts.

I saw more force in Professor Harris’ second point. Obviously, the parties could have provided expressly for exclusive jurisdiction if they had wished. Equally, they could have provided expressly for non-exclusive jurisdiction. That point is therefore neutral. One is left with the rather curious wording whereby the parties agree to submit themselves, rather than any disputes, to the jurisdiction of the English courts.

In this context a question arose as to significance, if any, to be attached to the use of intransitive rather than transitive language. I was referred to several authorities on this point but was not really assisted by any of them. Ultimately, they seemed to me to establish no more than that it is a question of construction in each case and not one of mechanistic grammatical pedantry. The language used by the parties, whether transitive or intransitive, may cast light on the intentions of the parties when read in context, or it may not. The only really relevant question is whether the parties undertook a positive obligation to submit disputes to the chosen court or whether they were merely agreeing to accept its jurisdiction: BNP Paribas SA v Anchorage Capital Europe LLP, [2013] EWHC 3073 (Comm) at [85]-[88]; Global Maritime Investments

Looking specifically at clause 11, it might be said that there was no point in specifying the English courts at all unless the parties had intended a positive obligation to submit all disputes to them. However, AAL was always amenable to English jurisdiction and there was therefore no compelling necessity to specify English jurisdiction for the purposes of any claim that Mr Goffe might wish to bring against AAL. On the other hand, the natural forum in which to sue Mr Goffe was either Georgia or Ontario. GDE had not been incorporated at the date of the contract and it was therefore not possible to identify its seat. The likelihood was that it would be incorporated somewhere in North America but it could have been elsewhere. A possible purpose of the clause might therefore have been to ensure that AAL had the option of suing the Claimants in England whatever their respective home courts: see the discussion in Briggs: Civil Jurisdiction and Judgments (op.cit.) paragraph 4.50. I reject Mr Ulyatt’s submission that the only claims contemplated at the date of the contract would have been claims brought by the Claimants against AAL. In my view it would have been impossible at the outset of the Agency Agreement to predict in terms of likelihood where any claims might fall. For example, it was quite foreseeable that AAL might wish to sue the Claimants for breach of fiduciary duty, such as selling trade secrets. The Fiona Trust case, although relied on heavily by Mr Ulyatt, was not concerned with the exclusive/non-exclusive distinction and seemed to me to be of only marginal assistance in this context.

This is a difficult question but, on balance, I have concluded that clause 11 did impose a mandatory obligation on both parties to litigate their disputes before the English courts. As I have concluded that “serious disputes” means all disputes which are incapable of amicable resolution, it follows that there is no practical difference between the parties undertaking to submit themselves to English jurisdiction and undertaking to submit all such disputes to English jurisdiction: BNP Paribas SA v Anchorage Capital Europe LLP (supra) at [86]-[87]; Austrian Lloyd Steamship Co. v Gresham Life Assurance Co., [1903] 1 KB 249; Briggs: Civil Jurisdiction and Judgments (op.cit.) paragraph 4.50.

That, of course, does not conclude the enquiry, since I must now go on to ask whether the parties thereby clearly demonstrated a positive choice of English law.

In this context Mr Ulyatt submitted that AAL was an English company which was choosing its home court and must therefore be taken to have intended to choose its home law. In my view this is a point which goes nowhere. The evidence was that in

Moreover, even on an objective approach, a choice of English jurisdiction may simply have been to ensure that AAL could sue the Claimants in England in circumstances where the place of GDE’s incorporation was as yet unknown. What can be said is that a choice of AAL’s home court effectively rules out such inference as might have been drawn from the choice of a neutral jurisdiction.

Mr Ulyatt further submitted that Mr Goffe was an English national who was happy to agree that disputes could be dealt with under the English legal system. No doubt that was true, but again that is evidence of Mr Goffe’s subjective state of mind which is inadmissible on any objective assessment of the circumstances.

The present case is not quite like Lawlor where there was no written agency agreement at all. In that case HHJ Mackie considered it unlikely that choice of law was even considered, let alone discussed, and he held that in those circumstances there had been no clear intention of making a choice of law. The Court of Appeal did not expressly disagree with this conclusion, although it emphasised that it was not necessary to prove a conscious subjective choice.

Here, by contrast, there was a written agreement, a draft of which was considered by the parties. Although it is common ground that there was no express reference to or discussion of clause 11, I agree that the natural inference is not that the clause was overlooked but that it was read and accepted by both parties: see Egon Oldendorff v Liberia Corp (supra) at page 383. I also accept that, artificial as the exercise may seem, it is possible to find a positive tacit choice of governing law even where (as here) there is no evidence to suggest that either party gave the question of governing law a moment’s thought. The hurdle to be surmounted is, however, high: it must be the only reasonable conclusion, in other words one which “goes without saying”.

I have little doubt that, having agreed to English jurisdiction, the parties may objectively be taken to have assumed that the English courts would probably apply English law. However, I have grave doubts as to whether this is sufficient to constitute a clear positive choice of English law. In Marubeni Hong Kong & South China Ltd v Mongolian Government (supra), the question before the court was whether the claimant could show a good arguable case under the Rome Convention that the governing law of a guarantee was English law in circumstances where the guarantee contained an English jurisdiction clause but an express English law clause proposed in an earlier draft had been deleted. The claimant asserted that the deletion was irrelevant, while the defendant alleged that it was highly material.

Of particular interest are the learned judge’s comments about the approach to Article 3(1) of the Rome Convention and the inferences to be drawn from an agreement to English jurisdiction:

Aikens J then went on to consider the significance of the deletion of the English law clause and held that it did not necessarily exclude the possibility that the parties intended English law to apply. In the event, he held that there was a good arguable case that the guarantee was governed by English law notwithstanding the deletion, mainly because the evidence suggested that England had been chosen as a neutral forum and it would be odd if the parties nonetheless intended Mongolian law to apply. However, he declined to decide the point finally as it depended on the circumstances of deletion which could not be determined at an interlocutory stage.

In my view, the comments quoted above draw a very real distinction between a mere expectation or assumption that English law will be applied and the positive choice required under Article 3(1) of the Rome Convention. The same diffidence as to whether a choice of court necessarily betokens a choice of law appears in an article written by Professor Lagarde shortly after the entry into force of the Convention: Le nouveau droit international privé des contrats après l’entrée en vigueur de la Convention de Rome du 19 juin 1980 (Rev. crit. dr. internat. Privé, avr-juin 1991). In a section entitled (in translation) “Problems concerning the choice of law”, he comments as follows:

Mr Ulyatt sought to disparage this text on the basis that it was not an official report and merely represented Professor Lagarde’s personal views which, in his submission, were idiosyncratic and incorrect. It is true that Professor Lagarde’s comments have been doubted by Plender & Wilderspin (op.cit.) paragraph 6-033, but provided they are not read as saying that a choice of jurisdiction can never indicate a choice of law, I see nothing objectionable in them. As recognised in Plender & Wilderspin itself (see fn 96) European practice regarding the inferences to be drawn from a jurisdiction clause in a choice of law context appears to have been divergent. There is therefore no overarching European approach to be applied. It is, moreover, difficult to dismiss Professor Lagarde’s doubts when they are echoed by such eminent lawyers as Aikens J (as he then was) and Professor Briggs: see Civil Jurisdiction and Judgments (op.cit.) paragraphs 7.104-7.108. (Footnote: ⁶)

On the other side of the coin, the Court of Appeal in Lawlor suggests that a choice of law may fairly be implied from a jurisdiction clause where it would not reasonably have occurred to the parties to suppose that a different law might apply, pointing out that “It would lack practical sense to require that they should have contemplated that which would not reasonably have occurred to them.” As a bare proposition, that is no doubt true. The parties here were relatively unsophisticated. They were laymen in legal terms and almost certainly in the happy position of knowing nothing about the finer points of English private international law. I therefore readily accept that it may not have occurred to them that the English courts might apply something other than English law.

With the very greatest of respect, however, that does not entirely meet the point made by Aikens J in Marubeni that this is nonetheless a matter of assumption rather than positive choice. Moreover, in the same way that Mance LJ in Amico (supra) was not prepared to assume that either party in that case would have had any inkling that the warranties relied upon would have been invalid under one (but not the other) of the putative proper laws, so here I am not at all sure that I can accede to Mr Ulyatt’s invitation to infer that a businessman can reasonably be expected to understand that a choice of court is likely to entail a choice of law.

As all the authorities and commentators agree, everything ultimately depends on the circumstances. I remind myself that the burden of proving a real choice with the requisite degree of clarity and certainty rests with the Claimants. I also heed the fact that it is unnecessary to strain to find an implied choice under Article 3 given the default provisions of Article 4. Finally, I remind myself that I am above all not concerned with the intentions of the individual parties, deemed or otherwise: I am looking for an objective mutual choice.

Ultimately, I have concluded that the jurisdiction clause in this case falls short of a positive choice of English law demonstrated with reasonable certainty. In my view, the most that can be inferred from clause 11 is an agreement that the English courts would determine any disputes in such appropriate way as they saw fit. No doubt this involved private (inadmissible) assumptions on both sides that English law would be applied,⁷ but even if it could be said that there was an objective shared assumption to this effect, an assumption is not the same thing as a positive choice of English law to the exclusion of all other laws. Adopting the words of the Giuliano-Lagarde Report, I find that in agreeing to clause 11 the parties have not shown “in no uncertain manner” that they intended the contract to be governed by English law.

Clause 11 was, of course, the lynchpin of Mr Ulyatt’s case under Article 3(1). He did nonetheless also seek to rely on a number of other factors, although in truth, most of them were more relevant to the question of closest connection under Article 4 than to

The Agency Agreement was largely negotiated and agreed at AAL’s premises in England, was drafted in England in the English language using English terminology and spelling. However, I can attribute no weight to these factors for the purposes of Article 3. In the context of an agency agreement between an English company and an English salesman resident in English-speaking North America it would have been surprising to find anything the parties communicating in anything other than English. Moreover, the agreement was based on AAL’s standard agency contract, and it is therefore equally unsurprising that it adopted English spelling conventions and terminology. As for the place of negotiation, this seems to me to have been entirely coincidental. It just so happened that this was the first opportunity Mr Goffe had had to visit AAL’s premises. Had he been able to visit in July 2008 as he originally proposed, it might well not have been necessary to negotiate and conclude any subsequent agreement in England; it could just as well have been dealt with in correspondence.

The Agency Agreement contemplated the sale of equipment at least half of which was to be manufactured in England. This likewise I find to be irrelevant to the question of mutual choice. The equipment had to be manufactured somewhere; precisely where was a matter of insignificance so far as the Agency Agreement was concerned. In any case, significant elements of the Easyflow system were manufactured in Ontario.

Commission was to be paid on the basis of the nett ex works price in England.

The fact that AAL endeavoured to contract wherever possible on the basis of its standard terms (which contained an English law clause and a non-exclusive English jurisdiction clause) is neither here nor there. Any sensible manufacturer would want its relationships with its customers governed by the same law if possible. But the same considerations do not necessarily apply to its relationship with its agents where local factors, such as mandatory employment legislation may be more relevant. I note that in Lawlor (supra), the employer likewise contracted with its customers on English law terms whenever it could. HHJ Mackie held that this was an indication that the parties probably would have chosen English law if asked, but that it did not point to any actual choice. His decision on this point was approved in the Court of Appeal. AAL’s relationship with its customers does not therefore say anything about any choice of law that

No-one has suggested that the parties made a positive choice of any law other than English law. It is therefore unnecessary to discuss the factors on which Professor Harris relied to negative any choice of English jurisdiction. It is sufficient for me to conclude that, while the above factors might have confirmed a choice of English law had I been prepared to infer such a choice from the jurisdiction clause, none of them was sufficient in my view to demonstrate a positive choice of law, whether taken individually or cumulatively. In reality, without the jurisdiction clause, there is nothing which necessarily and ineluctably suggests a choice of English law to govern this agreement.

For the sake of completeness, I should add that the post-contractual conduct of the parties does not take the matter any further so far as implied choice is concerned.

I therefore hold that there was no implied choice of English law under Article 3(1).

In the absence of any choice of law under Article 3, the overarching principle in Article 4(1) is that the contract shall be governed by the law of the country with which it is most closely connected. However, the court is required to give effect to that principle by applying the presumption in Article 4(2) (Footnote: ⁷) unless the case falls within one of the two exceptions set out in Article 4(5):

The first exception is where the characteristic performance of the contract cannot be identified. In that situation, the court must apply Article 4(1): see the Giuliano-Lagarde Report at page 22.

The second is where the circumstances as a whole demonstrate that the contract is more closely connected with another country than that dictated by the presumption.

One way or another, therefore, the ultimate objective of Article 4 is to find the country with which the contract is most closely connected.

The starting point is of course Article 4(2). This sets out a number of possible connecting factors, all of which require the court first to identify the person who is to effect the performance which is characteristic of the contract.

Article 4(2) then caters for two different categories of contract: those that are entered into in the course of the characteristic performer’s trade or profession and those that are not. In the case of contracts which are not concluded in the course of the characteristic performer’s trade or profession, the first sentence of Article 4(2) presumes that the most closely connected country is the country where the characteristic performer has, at the time of conclusion of the contract, his habitual residence or, in the case of a company, its central administration. By contrast, where the contract is concluded in the course of the characteristic performer’s trade or profession, the second sentence of Article 4(2) instead presumes the country of closest connection to be that in which its principal place

While the first sentence of Article 4(2) expressly looks to the date of the contract, Professor Harris argued that the second sentence was not similarly so constrained. I disagree. It seems to me that the most coherent interpretation (and certainly the interpretation most consonant with the objective of laying down clear, predictable rules) is that the entirety of Article 4(2) is concerned with the position at the date of the contract, except where the contract itself stipulates a place of business through which performance is to be effected. This is the view also taken in the Giuliano-Lagarde Report at page 21 and by Dicey, Morris & Collins (op.cit.) (14^th ed.) paragraph 32-119.

In the present case, it was common ground that the characteristic performer for the purposes of Article 4(2) was the agent, i.e., GDE. The difficulty, of course, is that GDE did not exist at the date of the contract and therefore had neither a central administration nor a principal place of business at the relevant time.

In these circumstances, Mr Ulyatt urged me to adopt a broad autonomous approach to the interpretation of Article 4(2) as follows:

EU law adopts a tolerant approach to pre-incorporation contracts whereby the promoter (in this case Mr Goffe) is personally liable if the company fails subsequently to adopt the contract: see Article 7 of the First Council Directive of 9.3.68.

Mr Goffe was accordingly under a contingent personal liability at the date of the contract, notwithstanding that GDE adopted the contract when it was incorporated with the result that there was a subsequent novation on 21 April 2009.

Mr Goffe as the sole proprietor of GDE should therefore be treated as the relevant actor for the purposes of applying Article 4(2), although the fact that the intended counterparty was GDE can be taken into account where relevant and necessary.

If Mr Goffe concluded the contract in the course of his general trade/profession as a salesman in the poultry processing equipment trade, then his principal place of business at the date of the contract was Georgia because that was where he was employed by Stork. (Mr Ulyatt accepted that this was an unsatisfactory and somewhat illogical result.)

Alternatively, if his relevant trade/profession at the date of the contract was specifically that of a Stork employee, then he plainly did not conclude the contract in the course of that trade/profession since it was no part of his duties to do so. In that case, the relevant connecting factor was Mr Goffe’s habitual residence which was likewise Georgia.

On behalf of the Defendant, Professor Harris argued that:

The Agency Agreement was concluded in the course of GDE’s trade/profession.

The relevant principal place of business was therefore that of GDE, which was in Ontario.

Alternatively, under the terms of the contract, performance was to be effected through GDE and this was equivalent to a provision that performance was to be effected through GDE’s place of business for the purposes of the exception in the last four lines of Article 4(2).

If it was necessary to look at the principal place of business of Mr Goffe at all then:

The court should look at his principal place of business for the business contemplated by the Agency Agreement. It would be absurd to apply Article 4(2) by reference to an employment in which Mr Goffe was deeply unhappy and which he was desperate to leave. Mr Goffe’s principal place of business for the purposes of the Agency Agreement was Ontario.

Even if that analysis was wrong, Mr Goffe as a travelling salesman worked in variety of locations. At least some of his work for Stork was performed from his Ontario home. The evidence showed that he was primarily based in Ontario when the Agency Agreement was signed and he had hardly ever been present in Georgia during March and April 2009. Moreover, after he resigned from Stork, his principal place of business can only have been Ontario.

I was unconvinced by either analysis in its entirety.

It is uncontroversial that, having identified the characteristic performer, Article 4(2) then asks whether the contract was entered into in the course of that person’s trade or profession. Despite some initial doubts, I am inclined to agree with Mr Ulyatt that, once it has been determined that the relevant person entered into the contract in the course of a trade or profession, there is no requirement that his principal place of business for the purposes of Article 4(2) be limited to his principal place of business for the performance of the contract in question. In the vast majority of cases, there will be no difference between the two, since the contract will be entered into in the course of a single pre-existing and continuing business. The problem is therefore likely to arise only in unusual cases such as the present where the contract concerns an employee who is in the course of quitting his employment to set up a new business which has not yet commenced.

Professor Harris sought to argue to the contrary by a process of “reverse engineering” from the exception in the last four lines of Article 4(2). This, it will be recalled, provides that if under the terms of the contract, performance is to be effected through a place of business other than the principal place of business, the country in which the specified place of business is situated shall be presumed to be most closely connected. On the basis that this exception specifically focuses on performance under the contract in question, Professor Harris submitted that references to “principal place of business” in the earlier part of the Article were similarly to be so qualified. I do not agree. Where

At this point, I should also dispose of the suggestion that the exception in the last four lines of Article 4(2) in fact has anything to do with this case. I agree with Mr Ulyatt that it does not, since it is only triggered where there is an actual contractual requirement, whether express or implied, that performance take place through a particular place of business such that performance anywhere else would be a breach of contract. If there is no such contractual requirement, it is irrelevant what the parties might have anticipated would be the case: Ennstone Building Products Ltd v Stanger Ltd, [2002] EWCA Civ 916; [2002] 1 W.L.R. 3059 at [29]-[31]. Despite Professor

In my view, there can be no serious doubt that the contract in this case was concluded by Mr Goffe on behalf of GDE in the course of “that party”, i.e., GDE’s trade/profession, albeit only a contemplated trade at that point in time. It is therefore GDE’s principal place of business at the date of the contract, not Mr Goffe’s principal place of business, which is notionally relevant.

What, then, is the position where there is no such principal place of business because the agent does not exist at the date of the contract? In my view the answer is much simpler than either party suggested. The reason why Article 4(2) is expressly disapplied where the characteristic performance cannot be identified must be because it cannot sensibly be applied in those circumstances. The court is then thrown back on Article 4(1): see paragraph 160.i) above. The same reasoning must equally apply if Article 4(2) cannot sensibly be applied because, even though the characteristic performance can be identified, the relevant actor does not exist at the date of the contract. In such circumstances, it seems to me that the court should not strain to apply a presumption which is incapable of sensible application and should likewise instead simply apply the default principle of Article 4(1).

Support for this approach can be found in Intercontainer Interfrigo SC v Balkenende Oosthuizen BV (supra) at [56] where the European Court of Justice held that Article 4(5) authorised the disapplication of Article 4(2), not only where the characteristic performance could not be identified, but also where the place of residence of the relevant performer could not be identified.

Article 4(1) provides a perfectly sensible and straightforward way of resolving the dilemma in this case. On the one hand, it avoids the need to resort to the law on preincorporation contracts, which in my view would be far too complicated an approach: on the other, it does not require any tortured construction of Article 4(2). In my view, it is the correct approach to adopt in the circumstances of this case where the characteristic performer had no principal place of business at the date of the contract.

I take the nature of the enquiry under Article 4(1) to be the same as that required under Article 4(5) where the European Court has stated that the court must carry out “an overall assessment of all the objective factors characterising the contractual relationship and determine which of those factors are, in its view, most significant.”: Haeger & Schmidt GMbH v Mutuelles de Mans assurances LARD, [2015] Q.B. 319 at [49]. In other words, I must determine the centre of gravity of the contract on an objective basis, for which purpose I regard myself as entitled to take account of postcontractual events to the extent discussed in paragraphs 109-114 above.

As noted, Article 4(1) seeks to find the country, not the system of law, with which the contract is most closely connected, although that does not mean that the court is concerned only with purely geographical or physical features: Dicey, Morris & Collins (op.cit.) (14^th ed.) paragraph 32-110. It is also important to note that it is the contract which must be most closely connected with the country, not the parties. The principal place of business of the characteristic performer is therefore not determinative although it is not wholly irrelevant where (as here) neither Claimant had any other business apart from the Agency Agreement. Nor am I looking for the closest connection with AAL’s business, even though the nature and location of that business may be a relevant consideration.

The following factors were all canvassed by one side or the other as bearing on the question of closest connection:

The purpose of the Agency Agreement: This was to establish a commercial agency for the North American market, selling to North American customers. Nonetheless, although it was a multi-territory agreement, it is clear from my findings above that Mr Goffe knew and accepted that AAL required a principal focus on the Canadian market and, specifically, on eastern Canada which it regarded as ripe for plucking and where it hoped to make a breakthrough. On any view, the focus of the Agency Agreement was not England.

The nationality of the parties: It is true that Mr Goffe was an English national and that AAL was an English company, but so far as the contract was concerned, they might just as well have been Rwandan or Uzbek, Finnish or Samoan. It was Mr Goffe’s residence in North America and his knowledge of the North America poultry market which qualified him to act as a sales agent for AAL, not his nationality. Likewise with GDE. As an SPV for Mr Goffe, its place of incorporation, whilst always likely to be somewhere in North America, was irrelevant. It is difficult to see that any more weight is to be attached to AAL’s nationality. It happened to be an English company, but it would have made no difference to the performance of the contract had it been incorporated in Portugal. I regard the nature and purpose of the contract as being far more important than the nationality of the parties.

Negotiation and signature of the agreement: For substantially the same reasons as given in paragraph 156.i) above when discussing Article 3(1), the mere fact that the Agency Agreement was negotiated during Mr Goffe’s visit to England in March 2009 and was drafted in the English language using English terminology, does not point to a particularly close connection with England – certainly not in the context of an agency agreement covering English-speaking

Place of agent’s performance: It follows from what I have said above that the characteristic performance under the contract was to take place in North America with a primary focus on Canada and eastern Canada in particular. It may well be, as Mr Ulyatt submitted, that Mr Goffe/GDE did on occasion have to perform duties in England, for example by bringing clients to England during the course of a sale. However, that can only have been a very minor part of their duties and on any view was wholly incidental to the primary purpose of the agreement which was to effect sales in North America.

Mr Goffe/GDE principal place of business: GDE was a SPV for Mr Goffe. It could only work through Mr Goffe and the evidence showed that Mr Goffe conducted his business from his laptop wherever he happened to be at the time. I have found above that by the date of the contract, his domestic base had moved to Ontario and for the reasons given in more detail in paragraphs 182-202 below, I find that that his principal place of business was also in Ontario at that date. Moreover, for the purposes of Article 4(1) and 4(5), I am entitled to take account of supervening events, and not only was Mr Goffe’s immediate post-contract conduct (see paragraphs 84-88 above) entirely consistent with his principal place of business being in Ontario at and from the date of the contract, it is quite clear – and indeed was admitted by Mr Goffe – that he spent more time in Ontario during the course of the Agency Agreement than in Georgia (and, by legitimate inference, anywhere else). I should also add that, as appears from the discussion below, Mr Ulyatt conceded that the only basis on which he could contend that Mr Goffe’s principal place of business at the date of the contract was Georgia was because he was still employed by Stork at that date. Selfevidently, that is irrelevant when considering the country with which the Agency Agreement is most closely connected.

Temporary nature of Mr Goffe’s residence in Ontario: The fact that Mr Goffe only ever intended his residence in Ontario to be temporary relates to his subjective state of mind and is therefore inadmissible and irrelevant in the absence of any evidence to show that this intention was appreciated and shared by AAL. In any event, as this case demonstrates, intentions and circumstances can change and while Mr Goffe and Ginette had originally intended to return to Georgia after a year, they ended up staying in Ontario until 2013. Even if the prolongation of their stay was unforeseen at the date of the contract and so a matter of less significance in itself (see paragraphs 109-114 above), I am nonetheless entitled to place some weight on the fact that Mr Goffe was primarily based in Ontario for the entire duration of the agreement.

AAL’s obligations: It was argued that the Agency Agreement required AAL to do many things from England. However, it is difficult to see that these included anything of substance. The shipment of equipment and spares to which Mr

Payment of commission: The contract expressly provided for payment of commission into GDE’s bank account. The natural inference (given that payment by AAL’s customers was to be in local currency) was that the commission would likewise be paid in local currency. There was no contractual requirement to pay in England or, indeed, anywhere else, and the objective likelihood at the date of the contract was that commission would be paid to a bank account in North America, as in the event it was. I note that GDE’s bank accounts were initially located only in Virginia, although a Canadian bank account was also opened in 2010, following which payments in Canadian dollars were made to the Canadian account and payments in US dollars to the Virginia account. As I have said, I regard the fact that commission was calculated on the basis of an ex works price in England as irrelevant. All that meant was that GDE was not entitled to commission on the freight element of the price paid by the customer. The price itself (on which the commission was based) was still calculated in local currency.

Place of manufacture of the equipment: For substantially the same reasons as stated in paragraph 156.ii) above, I find the place of manufacture to be largely irrelevant to the centre of gravity of the Agency Agreement. The Agency Agreement was not primarily concerned with manufacture but was focused on the effecting of sale contracts in North America. In any event, drawers and modules were manufactured in Ontario and while no doubt each system incorporated thousands of individual nuts, bolts, screws, brackets and fixings which were manufactured in England, the drawers and modules were critical components of the Easyflow system. To my mind, it does not matter that they were manufactured using AAL’s moulds and drawings. The fact that they were manufactured in Ontario demonstrates the importance of the Canadian market to AAL and underlines the fact that a necessary part of GDE’s role would necessarily have involved liaison with the Ontario manufacturers even without this being expressly stated in the contract. It should also be noted that clause 8 of the Agency Agreement required specific mutual agreement in order to source equipment from any other territory.

AAL’s sales contracts were governed by English law: For substantially the same reasons as I have given in relation to Article 3(1), this fact does not in my view support a connection between the Agency Agreement and any particular jurisdiction. The only relevance of AAL’s sales contract to the Agency Agreement was that they contained the base price on which the Claimants’ commission would be calculated. The law by which they were governed as between AAL and its customers was neither here nor there.

The jurisdiction clause: This to my mind was the strongest factor in favour of a connection between the contract and England. However, as a clause dealing

Mr Goffe’s personal connections to Georgia: It is true that Mr Goffe had a house in Georgia, was registered to vote there, paid all his taxes there and that his doctor, dentist and tennis club membership were all there. However, his 12 February 2009 email linked his retention of the Georgia house specifically to his employment with Stork, which on any view is irrelevant to the Agency Agreement. Moreover, these are all essentially personal factors which have nothing to do with the contract itself.

On an objective assessment of all these factors, I have concluded that the centre of gravity of this contract was Ontario. The two factors which to my mind carried the most weight were that (i) both parties contemplated that the agency would focus primarily on Canada and on eastern Canada in particular; and (ii) Mr Goffe’s domestic base was in Ontario and, given the nature of his work as a travelling salesman and his entirely understandable desire to spend as much time as possible with his wife, that was the base from which he worked when he was not on the road, both at the date of the contract and thereafter. The most powerful factor in favour of England was the jurisdiction clause, but as I have said, this was an ancillary dispute resolution provision which had nothing to do with the substance of the contract and in my view is outweighed by the considerations above. Apart from the fact that Georgia was part of the sales territory covered by the Agency Agreement, there was no particular connection between the agreement and Georgia. Such connections as existed, were essentially personal to Mr Goffe and unrelated to his relationship with AAL.

I therefore find that the country with which the Agency Agreement was most closely connected for the purposes of Article 4(1) was Ontario and that the governing law is therefore the law of Ontario.

If my analysis above is wrong and it is after all appropriate to apply Article 4(2) and for that purpose to look at the position of Mr Goffe at the date of the contract, I would have found that the contract was entered into in the course of Mr Goffe’s trade/profession. To my mind, the purpose of drawing a distinction between contracts which are concluded in the course of a trade/profession and those which are not is (broadly) to distinguish commercial contracts from non-commercial contracts. On any view this was not a private arrangement or a consumer contract, and I therefore have little doubt that the relevant connecting factor under Article 4(2) on this hypothesis is Mr Goffe’s principal place of business at the date of the contract.

So far as concerns the correct interpretation of “principal place of business”, I was referred by Mr Ulyatt to The Rewia, [1991] 2 Lloyd’s Rep.325 at 334. This was a decision under the Brussels Convention for a different purpose and is therefore of marginal relevance, although I do not dissent from the proposition that a principal place of business is not necessarily the same as the place where the bulk of the work is performed, or (in the case of a company) the place where it is incorporated. Often the relevant principal place of business will be the same as the place of performance, but this is not necessarily so: Dicey, Morris & Collins (op.cit.) (14^th ed.) paragraph 32-118.

Identification of the relevant place of business is therefore a question of fact which depends on the circumstances of each case: Young v Anglo American South Africa Ltd, [2014] EWCA Civ. 1130 at [39]. The parties were agreed that the test was objective.

However, they could not agree what the relevant business was. Was it Mr Goffe’s business as an employee of Stork? or the business contemplated by the Agency Agreement? In support of his case that Mr Goffe’s principal place of business at the date of the contract was Georgia, Mr Ulyatt drew attention to the email of 12 February 2009, in which Mr Goffe said that he maintained a home in Georgia because he needed to do so for his current position. He frankly accepted in his closing submissions [Day 3 page 112] that his argument in support of Georgia as Mr Goffe’s principal place of business was determined by Mr Goffe’s employment with Stork (which did not terminate until 23 April 2009) and conceded that this was an unsatisfactory result. Indeed, his primary case was that this was an unfortunate consequence of Article 4(2) being a “blunt tool” and a powerful reason why the resulting presumption should be displaced under Article 4(5) in favour of England.

In response, Professor Harris objected that it was absurd to determine Mr Goffe’s principal place of business by reference to an employment which he was desperate to quit. I agree that there is an element of absurdity involved, particularly when clause 1 of the Agency Agreement expressly provides that Mr Goffe is to commence working full time for GDE. Nonetheless, I do not find absurdity in itself to be a particularly weighty argument, given that Article 4(5) exists precisely in order to ensure that any such absurdity resulting from the presumptions can be corrected.

Moreover, I found this to be a somewhat arid debate. The objective of the Rome Convention is to establish clear, simple, uniform rules and I do not see that Article 4(2) requires the court to conduct a detailed investigation in order to distinguish between Mr Goffe’s duties as an employee and his wider business interests as a salesman. For the reasons given in paragraph 170 above, it is necessary only to identify Mr Goffe’s principal place of business at the date of the contract, whatever business he may have been carrying on at the time. If, for the sake of argument, that was a completely different type of business from the business contemplated by the contract, then the presumption can be displaced under Article 4(5). If he had no business at the date of the contract because it was yet to be established, it is Article 4(1) which will then apply.

That said, the contortions and evident absurdities involved in trying to apply Article 4(2) to the present facts only serve to confirm me in my view that this is not at all a sensible approach and that Article 4(1) is instead the correct provision to apply.

There was no evidence that Mr Goffe’s employment with Stork contractually required him to be based in Georgia, although this was no doubt the most convenient base for him to adopt. Identifying the principal place of business of a travelling salesman is somewhat elusive since travelling salesmen necessarily work while they are on the road, which could be anywhere, and as I have said above it is clear that a principal place of business is not necessarily the same as the place of performance. In this case, as I have found above, Mr Goffe had moved his domestic base to Ontario by the date of the Agency Agreement and his family circumstances were such that he needed to and did spend as much time there as possible. Although he had no Canadian work permit, it was not disputed that he was entitled to carry out his duties as a travelling salesman in

I am also entitled for this purpose to take into account subsequent events and conduct in so far as they cast light on the position at the date of the contract. As noted above, Mr Goffe’s immediate post-contract conduct is entirely consistent with his principal place of business being Ontario at that date: see, in particular, the reference in his email of 26 April 2009 stating that he was now “back” in Toronto.

The fact that Mr Goffe did not have a Canadian bank account at the date of the contract does not undermine this conclusion. As already noted, his bank account could have been anywhere. In fact it was in Virginia, which was not a jurisdiction that either party suggested had any relevance to this case. Mr Ulyatt also pointed to Mr Goffe’s personal connections with Georgia. Although these had no necessary relationship with the contract itself, I accept that in the case of an effectively self-employed agent, they might be of more significance in determining that agent’s principal place of business. Nonetheless, in circumstances where Mr Goffe had clearly moved his domestic base to Ontario by the date of the contract, they can bear little, if any, weight.

On the basis of the facts as I have found them, I therefore find that Mr Goffe’s principal place of business at the date of the contract was Ontario. However, Professor Harris also relied on another consideration, namely the representations that Mr Goffe made in pre-contractual correspondence to the effect that he was now living and working in Ontario, in particular in his email of 12 February 2009. It was not seriously disputed that these representations formed part of the objective evidence that I could take into account in determining Mr Goffe’s principal place of business. I have already stated that I was not convinced by Mr Goffe’s attempts to downplay their significance and to that extent they confirm my conclusion.

But Professor Harris’ argument went slightly further than this. He submitted that European authority required European rules of private international law to be construed so as to uphold the expectations of the counterparty and that Mr Goffe’s principal place of business should be determined accordingly. He referred me in particular to two cases.

In Berghoefer GmbH & Co. KG v ASA SA, [1986] 1 C.M.L.R. 13, the issue was whether an amendment to a jurisdiction clause was sufficiently evidenced in writing for the purposes of Article 17(1) of the Brussels Convention. The amendment in question had been agreed orally and written confirmation had been sent by the beneficiary of the change to the counterparty who had not responded but had nonetheless taken advantage of more favourable terms which had been negotiated in return. The European Court rejected the argument that the amendment had to be confirmed by the party against whom the clause was invoked and accordingly found that there had been sufficient compliance with the formality requirements. However, it also said that it would be bad faith for the counterparty to dispute the application of the oral agreement when it had raised no objection to the confirmation.

In my view, Berghoefer was not really on point. The question there was whether one party could rely on a lack of formality to avoid the effect of an agreement which was

The more pertinent case was Gruber v BayWa AG, [2006] Q.B. 204. Here it was necessary to decide whether a contract which the claimant had entered for both business and private purposes was a “consumer contract” so as to permit derogation from the general rule of jurisdiction under the Brussels Convention. The European Court of Justice noted that (i) in accordance with established practice any such derogations had to be strictly construed; and (ii) the purpose of derogation in consumer cases was to ensure adequate protection for consumers who were deemed to be economically weaker parties. In accordance with these principles, it held that a dual-purpose contract would only be considered a consumer contract where any business use was merely negligible, since it was only in those circumstances that special protection was justified.

The judgment is of interest in the present case in two respects. First, the court held that in determining whether or not the contract was a consumer contract, the court should look at the content, nature and purpose of the contract and the objective circumstances in which it was concluded. It should not take account of the subjective situation of the person involved: see paragraphs [36], [47].

Secondly, it held that even if the objective circumstances showed no non-negligible business purpose, the claimant could not take advantage of any protection for consumers where the counterparty was reasonably unaware of the private purpose of the contract because the claimant had given the legitimate impression that he was acting for business purposes:

In such a case, the special rules of jurisdiction for matters relating to consumer contracts enshrined in articles 13-15 are not applicable even if the contract does not as such serve a non-negligible business purpose , and the individual must be regarded, in view of the impression he has given to the other party acting in good faith, as having renounced the protection afforded by those provisions.

In the light of all the foregoing considerations, the answer to the first three question must be that the rules of jurisdiction laid down by the Brussels Convention are to be interpreted as follows:… (ii) it is for the court seised to decide whether the contract at issue was concluded in order to satisfy, to a non-negligible extent, needs of the business of the person concerned or whether, on the contrary, the trade or professional purpose was negligible; (iii) to that end, that court must take account of all the relevant factual evidence objectively contained in the file; on the other hand, it must not take account of facts or circumstances of which the other party to the contract may have been aware when the contract was concluded, unless the person who claims the capacity of consumer behaved in such a way as to give the other party to the contract the legitimate impression that he was acting for the purposes of his business.” (Emphasis added.)

Mr Ulyatt sought to distinguish Gruber on the basis that the knowledge of the counterparty was always relevant to a determination of whether a contract was a consumer contract or not. I disagree. The definition of “consumer” makes no reference whatsoever to the understanding of any counterparty.

More pertinently, he submitted that Gruber was effectively a case of estoppel which I have held is not open to the Defendant in this trial. I agree with him that relying on a party’s conduct and representations to disentitle him from asserting what might otherwise be the true position comes perilously close to an estoppel but, as I understood it, Professor Harris’s argument was altogether more subtle. He submitted that Gruber was not a case of estoppel, but rather about the European meaning of European legal concepts. Thus, a contract was in fact a non-consumer contract if one party led the other reasonably to believe that it was a non-consumer contract. So too here, I should interpret “principal place of business” as meaning the principal place of business represented as such by Mr Goffe, unless AAL could reasonably have been aware that it was not.

I have to say that I have real doubt as to whether this was indeed what the European Court was saying in Gruber. I am also far from convinced that it was intending to a state a principle of general application. There was some force in Mr Ulyatt’s submission that the Court’s approach was conditioned by the fact that the protective scheme in question was a derogation (see paragraph 51). In any event, if there is any distinction between what the European Court said in Gruber and a classic estoppel, it is wafer-thin.

Fortunately, in view of my findings on the facts, this is not a nettle which it is necessary for me to grasp and I decline to do so. The answer to Professor Harris’ objection that it would be unattractive to permit Mr Goffe to disavow his deliberate representations to the effect that he was living and working in Ontario is that the Defendant should have pleaded a case in estoppel from the outset.

As it is, I have reached the conclusion, independently of the European authorities on which Professor Harris relied, that Mr Goffe had indeed established a principal place of business in Ontario by the date of the contract. The fact that he subjectively only intended his residence in Ontario to be temporary, is inadmissible for the reasons already given.

I was not addressed at any length on the question of habitual residence. Habitual residence as a connecting factor formed no part of the Defendant’s case and it was not at the forefront of the Claimants’ submissions, although Mr Ulyatt did not abandon it as a fallback position if I were to find that Mr Goffe did not conclude the contract in the course of his trade/profession. In his skeleton argument he referred to the test summarised by the European Court of Justice in Fernandez v European Commission, [1994] ECR -4301 (ECJ) as follows:

Since the question of habitual residence does not arise on my findings, I do not propose to lengthen this judgment further by addressing it. I will say only that in circumstances where Mr Goffe had moved his domestic base to Ontario by the date of the contract, it would have been necessary to consider what was meant by “lasting character” in the context of an intention that his residence there should be as short as possible but as long as necessary, and the fact that even though he initially only planned to stay in Ontario for one year, he ended up living there for nearly five years.

It will be appreciated that Article 4(5) is only relevant if:

I am wrong in my analysis that the court should apply Article 4(1) in circumstances where GDE had no principal place of business at the date of the contract;

the Article 4(2) presumption should be applied by reference to Mr Goffe’s principal place of business or habitual residence.

In my view the correct approach to Article 4(5) does not differ from what I have set out at paragraphs 176-177 above when discussing Article 4(1). However, although the enquiry is clearly similar under each article, it is not in all respects identical because the outcome under Article 4(5) depends on both the starting point under Article 4(2) and the putative rival country.

Thus if, as I have held, the Article 4(2) presumption leads to Ontario, the relevant comparison is between Ontario and England. If am wrong about that, however, and

I mention at this point the Defendant’s submission that the Claimants had only pleaded that a presumption in favour of Ontario was to be displaced in one solitary situation, namely where GDE’s central administration was situated there. This was a highly technical and unmeritorious point. It was (rightly) not pressed and I proceed on the basis that the Claimants are entitled to argue for displacement in favour of England whatever the basis on which Article 4(2) may have led to a presumption in favour of Ontario.

I therefore turn to consider each of the possible permutations.

This requires no separate consideration, since my decision under Article 4(1) necessarily involves a finding that the contract was more closely connected with Ontario than with England.

This permutation requires me to revisit the factors considered in paragraph 178 but this time with a view to making a specific comparison between Georgia and Ontario. Clearly factors which I have found to be irrelevant (for example, because they are personal to Mr Goffe and nothing to do with the Agency Agreement) or which I have rejected as demonstrating any connection with any particular country can be ignored in this context also.

As for the rest, it seems to me that they point to a closer connection with Ontario than with Georgia. Once again, I attribute most weight to the particular focus on Canada, and specifically eastern Canada, required by AAL and accepted by Mr Goffe. For the reasons previously given, Mr Goffe’s principal place of business at the date of the contract which, on this hypothesis, was Georgia is incapable of demonstrating any connection between the Agency Agreement and Georgia. Moreover, subsequent events which I am entitled to take into account show that Mr Goffe’s performance of his duties under the agreement was more closely connected with Ontario than with Georgia. Other factors pointing to Ontario rather than Georgia are:

the manufacture of significant elements of the Easyflow system in Ontario requiring at least some liaison by Mr Goffe with the manufacturers;

the fact (albeit of comparatively minor significance) that commission was never paid to the Claimants in Georgia, whereas it was paid to them in Ontario.

I have found this comparison the most difficult of all.

The only relevant factor pointing to a connection between the Agency Agreement and Georgia derives from the fact that this was a sales agreement covering North America. It was thus a multi-territory agreement, and Georgia was only one of the territories to be covered by Mr Goffe. It cannot be said that Georgia was a particular focus of the agreement in the same way that eastern Canada clearly was. Nonetheless, the Agency Agreement clearly contemplated performance in Georgia in a way that it did not in England and to that extent the centre of gravity of the agreement was closer to Georgia than to England. Mr Goffe’s principal place of business in Georgia for the purposes of his employment with Stork is irrelevant here as well.

As against that, there is the English jurisdiction clause which, for lack of any other factors favouring Georgia, carries proportionately more weight in this context than in any comparison with Ontario.

In these circumstances, I find the scales to be evenly balanced. The wording of Article

This would be truly bizarre and to my mind merely underlines the absurdity of trying to force this camel of a case through the eye of the needle that is Article 4(2) when in truth it will simply not fit.

Nonetheless, to the extent relevant, I hold that:

if the presumption under Article 4(2) is in favour of Ontario, the facts do not justify displacement in favour of England;

if the presumption under Article 4(2) is in favour of Georgia, it is to be displaced in favour of Ontario, but not in favour of England.

Given my primary conclusion, however, the preliminary issues are to be answered as follows:

The law governing the Agency Agreement is the law of Ontario.

Does not arise.

Despite its initial one and a half day estimate being more than trebled, this proved to be a most interesting case and I cannot conclude without paying tribute to the high quality of the submissions of counsel on both sides who I think can fairly be said to have left no stone unturned in their efforts to assist me.