Tartsinis v Navona Management Company

In this case the English Commercial Court is asked to interpret a contract for the sale of shares in a company which owned a fleet of ships. The shares were sold by two Greek businessmen, Mr Mihail Tartsinis and Mr Antonis Nikolaou, to Navona Management Company (“Navona”), a company incorporated in the Marshall Islands which is represented and controlled by a third Greek businessman, Mr Kriton Lentoudis. The contract was made in Athens where all three of the protagonists live. The contractual document was drafted by a Greek lawyer. The document, entitled “Share Transfer Agreement”, is however written in the English language and provides that it shall be interpreted in accordance with English law and that “all disputes hereunder shall be submitted to the exclusive jurisdiction of the Courts of England and Wales”.

The dispute which has arisen concerns the price payable for the shares. The shares in question represented 40% of the issued share capital of Newport Holdings Limited (“Newport”), another company incorporated in the Marshall Islands. Before the sale, Mr Tartsinis and Mr Nikolaou each owned 20% of the shares of Newport; the other 60% of its share capital was owned by Navona, which after the sale therefore owned all the shares of Newport. Newport in turn owned all the shares of Global Oceanic Carriers Limited (“GO Carriers”), a Jersey company which had five wholly owned special purpose subsidiaries incorporated in the Marshall Islands which each owned one ship. Those five ships formed the GO Carriers fleet.

At the same time as selling their shares in Newport to Navona, Mr Tartsinis and Mr Nikolaou purchased (through a special purpose company) one of the ships in the GO Carriers fleet, called the mv “GO Patoro”. The Share Transfer Agreement was executed on the same date (23 February 2011) as the closing of the sale and purchase of the “GO Patoro” – a date referred to in the Share Transfer Agreement as the “Delivery Date”.

Under the Share Transfer Agreement Mr Tartsinis and Mr Nikolaou sold their shares in Newport (the “Shares”) for a price “equal to 40% of the NAV [i.e. “net asset value”] of GO Carriers”. Payment for the Shares was to be made in two stages. A “Provisional Purchase Price” based on a provisional calculation of 40% of the NAV of GO Carriers was payable on the Delivery Date (and was set off against the price of the “GO Patoro”). The second stage of payment involved adjustment of the Provisional Purchase Price of the Shares “upwards or downwards following the final determination and settlement of the NAV of GO Carriers following the issuance of the audited accounts of GO Carriers” for the period ending on the Delivery Date. (I will refer to these accounts as the “Delivery Date Accounts”.) The dispute between the parties is about exactly how this mechanism for adjusting the price of the Shares was to operate and, in particular, whether it was intended to apply to the valuation of the fleet which was the main component of the NAV of GO Carriers or whether the fleet valuation used to determine the Provisional Purchase Price was intended to be a final figure not subject to adjustment in the same way as the other elements of the NAV of GO Carriers.

Mr Tartsinis, who is the claimant in this action, contends that on the proper interpretation of the Share Transfer Agreement the final determination of the value of the fleet, along with all the other assets and liabilities of GO Carriers, was agreed to be based on the value stated in the Delivery Date Accounts. That value is substantially higher than the value of the fleet mentioned in the Share Transfer Agreement, which on the claimant’s case was only provisional. If this case is correct, a further sum of over US$6m is due to Mr Tartsinis from Navona.

Navona’s primary case is that, on the proper interpretation of the Agreement, the fleet value mentioned in it, unlike the other elements of the NAV of GO Carriers, was a final figure not subject to adjustment. Alternatively, if the fleet value was subject to adjustment, Navona contends that the figure stated in the Delivery Date Accounts is incorrect and that there is in fact a substantial adjustment of some US$2.3m due in its favour. In the further alternative Navona asks the court to rectify the Agreement to reflect what it says was the common intention of the parties that the fleet value would not be subject to adjustment.

A complicating feature of the case is that, when the action was begun, Mr Nikolaou whose position as a seller is identical to that of Mr Tartsinis was also a claimant. However, Mr Nikolaou later abandoned his claim and at the trial was called as a witness by Navona.

I have mentioned that, as well as arguments about the interpretation of the Share Transfer Agreement, there is a claim for rectification. The potential need for such a claim comes about because of two distinctive features of the rules of English law governing the interpretation of contracts.

The first is that, in deciding what a contract means, English law does not attempt to identify what the parties actually understood or intended the language used in the contract to mean. Instead, the law adopts an “objective” approach to interpretation. As Lord Hoffmann might have said, I do not think that the extent to which this is so is always sufficiently appreciated. It is not simply that a court, in interpreting a contract, has no window into the minds of the parties and must therefore necessarily draw inferences about what the parties were using the language of the contract to mean, adopting the standpoint of a reasonable observer. What the parties to the contract actually meant, or whether they had any pertinent subjective intention at all, is irrelevant to the task of interpretation. Rather, the court identifies the meaning of the language used by assuming that the parties were reasonable people using the language of the contract to express a common intention. As Lord Wilberforce said in Reardon Smith Line Ltd v Hansen-Tangen (The “Diana Prosperity”)[1976] 1 WLR 989, 996:

“When one speaks of the intention of the parties to the contract, one is speaking objectively ... and what must be ascertained is what is to be taken as the intention which reasonable people would have had if placed in the situation of the parties.”

A second important feature of the applicable rules of English law is that evidence of what was said during the negotiation of the contract is not admissible for the purpose of interpretation. One reason for this is that such evidence is generally of no help in ascertaining the objective meaning of the document. Even where such evidence could potentially bear on that meaning, however, it is not admissible: see Chartbrook v Persimmon Homes [2009] 1 AC 1101, 1120-1, para 41. Evidence of the subsequent conduct of the parties is also inadmissible to interpret a contract: see e.g. James Miller & Partners Ltd v Whitworth Street Estates (Manchester) Ltd [1970] AC 583.

These rules have many advantages. Such advantages include: (i) enabling a party to predict with a reasonable degree of certainty when entering into a contract how its provisions will be interpreted, without having to probe or be concerned about whether the other party shares this understanding or with trying to lay a favourable paper trail in pre-contractual correspondence; (ii) enabling a lawyer advising a party, or a judge or arbitrator required to interpret the contract, to do so on the basis of relatively little information and without the need for an extensive and expensive factual inquiry; and (iii) respecting the autonomy of the contracting parties by treating them as rational agents who have chosen the words of their document to give appropriate expression to their bargain. There are some cases, however, in which the application of these rules would lead to injustice. In particular, that is potentially so in circumstances where the parties to the contract had a common intention or understanding when they made the contract as to what it meant which was different from the objective meaning of the contractual document as ascertained in accordance with the rules of interpretation.

The solution which English law adopts is to allow the wording of the document to be rectified in circumstances where it is necessary and just to do so in order to give effect to a common intention of the parties (or in certain cases one party’s unilateral intention). I will mention later the specific requirements of this doctrine and the extent to which rectification, as English law now stands, is itself an “objective” exercise. Importantly, however, on a claim for rectification evidence is admissible which is not admissible for the purpose of interpreting a contractual document. In particular, evidence of what was said in pre-contractual negotiations is admissible and indeed is generally essential to prove the existence of a common intention which the document mistakenly failed to express. Evidence of the parties’ subjective aims and intentions is also admissible, as is evidence of their subsequent conduct insofar as it sheds light on their relevant intentions.

Because interpretation and rectification are two such very different exercises, it is important, though not always easy, to keep them separate. For this reason the rest of this judgment is in two parts. In the first part I will consider the issues of interpretation. For that purpose I will mention only such background facts as are in my view properly admissible in interpreting the Share Transfer Agreement. In the second part of this judgment I will address the issue of rectification. At that stage I will consider the further evidence which is admissible for that purpose including, most importantly, evidence about what was said in negotiating the Share Transfer Agreement and what the individuals who entered into the Agreement understood that they had agreed.

Although evidence of pre-contractual negotiations is not admissible to draw inferences about what a contract means, it is well established that the court can and should consider other aspects of the factual background known to both parties at the time when the contract was made. This includes evidence of the “genesis” of the transaction: see Prenn v Simmonds [1971] 1 WLR 1381, 1385; The “Diana Prosperity”)[1976] 1 WLR 989, 995.

The claimant, Mr Tartsinis, has been involved in the shipping industry for more than 20 years. He started his career as a chartered accountant specialising in shipping. In 2003 he bought his first vessel and in 2004 he set up Antares Shipmanagement SA (“Antares”) with Mr Nikolaou, who has also been involved in the shipping industry for many years.

GO Carriers was founded in December 2004 and listed on the London Stock Exchange Alternative Investment Market in May 2005. On 7 June 2006 Mr Tartsinis and Mr Nikolaou became directors of GO Carriers, with Mr Tartsinis being appointed as Chief Executive Officer. At that time GO Carriers owned three ships through special purpose companies. When Mr Tartsinis and Mr Nikolaou became directors of GO Carriers, the management of these ships was transferred to Antares.

Mr Nikolaou had for many years known Mr Lentoudis, a successful ship owner who is well known in the Greek shipping community. In November 2007, following an approach from Mr Nikolaou, Mr Lentoudis (through his company, Navona) agreed to invest in GO Carriers and also to lend money to Mr Nikolaou and Mr Tartsinis to enable them to increase their stakes in GO Carriers. By this time, GO Carriers owned (through special purpose companies) a fleet of seven vessels. Shortly after this investment was made, GO Carriers was de-listed and its entire share capital was acquired by Newport. As mentioned, 20% of the shares of Newport were owned by Mr Tartsinis, 20% by Mr Nikolaou and 60% by Navona. Navona had lent a sum of over £14m to Mr Tartsinis and Mr Nikolaou to enable them to acquire shares in Newport, and their shares were pledged to Navona as security for this loan. Mr Tartsinis and Mr Nikolaou were the only directors of GO Carriers but the company was controlled by Mr Lentoudis, through Navona’s ownership of 60% of the shares of Newport and its control of the voting rights, via the pledges, of the remaining 40% of the shares.

At the end of 2008 Mr Lentoudis decided to transfer the management of the ships in the GO Carriers fleet from Antares to Evalend Shipping Company SA (“Evalend”), a company under his control. Mr Tartsinis and Mr Nikolaou were unhappy about this but they had no choice in the matter.

The investment in GO Carriers did not prove to be the success that the three participants had hoped. This was because in August 2008 the shipping market crashed. In 2009 two of the ships in the GO Carriers fleet were sold at a substantial loss. By the time of the Share Transfer Agreement in February 2011, ship values had still not recovered.

During 2010 Mr Tartsinis and Mr Nikolaou approached Mr Lentoudis about selling their shares in Newport to Navona. The transaction which ultimately took place had the following basic structure:

Two contracts were made to give effect to the transaction: the Share Transfer Agreement and a separate contract for the sale and purchase of the mv “GO Patoro”.

The parties to the Share Transfer Agreement were Mr Tartsinis and Mr Nikolaou, referred to as the “Sellers”, and Navona, referred to as the “Buyer”. The Agreement contains a number of recitals including the following recital (E):

“The Parties have agreed that the Net Asset Value (“NAV”) of GO Carriers and of Newport equals provisionally the sum of 57.000.000 as of today, and accordingly Seller’s A share (20%) to the NAV of Go Carriers amounts to the provisional amount of US$11,406,263.50 and Seller’s B share (20%) to the NAV of Go Carriers amounts to the provisional amount of US$11,406,263.50.”

The Agreement also contains the following definition:

“NAV means the Net Asset Value of Go Carriers which as of today has been provisionally calculated to amount to U.S.D. 57m approximately as follows: Fleet Value $96,5m plus cash $28m less loan obligations $67,5m” [emphasis in original]

Article 1 of the Agreement provided for the sale and transfer of the “Shares” (defined to mean the shares of Newport owned by Mr Tartsinis and Mr Nikolaou) to Navona. Clause 1.2 stated:

“Subject to Clause 1.4, the Shares are transferred to the Buyer for a total consideration equal to 40% of the NAV of GO Carriers, therefore provisionally at the total price of 22.812.527 USD (the “Provisional Purchase Price”), i.e. 114.062,635 USD per share. The Provisional Purchase Price is subject to adjustment in accordance with clause 1.4 hereof, in order to ensure that the consideration that will be actually due for the sale of the Shares hereunder will be 40% of the NAV of GO Carriers as finally determined.”

The clause at the centre of the dispute is clause 1.4, which stated as follows:

“It is hereby agreed between the Parties that the Provisional Purchase Price may be adjusted upwards or downwards following the final determination and settlement of the NAV of Go Carriers following the issuance of the audited accounts of Go Carriers (pursuant to the I.F.R.S. standards) for the period ending on the Deliver[y] Date of the M/V GO PATORO pursuant to the provisions of Clause 2.2 hereunder, which accounts should be delivered to the Parties the latest by 10.4.2011. Such determination will take place latest by 30.4.2011. Upon final determination of the NAV as of the Delivery Date, the Provisional Purchase Price will become “Final Purchase Price” (equal to 40% of the NAV of Go Carriers as finally determined) and the party responsible for payment (Buyer or Sellers), must pay the difference between the Final Purchase Price and the Provisional Purchase Price within ten (10) working days following the date of such final determination. Solely for the purpose of determining the NAV of Go Carriers as of Delivery Date, the Sellers are granted access to the books and records of Go Carriers. ... Following such final determination, a further payment between the Parties to reflect the adjustment to the NAV of Go Carriers will take place. The Parties agree that the Final Purchase Price as above calculated (on the basis of the Provisional Purchase Price and the adjustments set out in this clause) is fair, reasonable and at market value”

Article 2 of the Agreement specified how payment of the purchase price of the Shares was to be effected. Under Clause 2.1 the parties agreed to set off “their respective claims against each other, namely Sellers’ claim for the provisional amount of $22.812.527 against the Buyer for the Purchase Price of the Shares” and the amount due to the Buyer to repay the loan, converted into US dollars. The clause recorded that, following this set-off, the provisional amount of approximately US$2.5m remained to be paid by the Buyer to the Sellers (referred to as the “Balance of the Provisional Purchase Price”).

Clause 2.2 recorded that the Sellers had agreed (as buyers) with Navona, Newport, GO Carriers and Dion Maritime Company SA (as seller) to purchase, through a special purpose vehicle, the mv “GO Patoro” for a total consideration of US$17.5m. It was agreed that, at the date of closing of the sale and purchase of this vessel (the “Delivery Date”), the Balance of the Provisional Purchase Price would be set off against the amount payable for the purchase of the “GO Patoro”.

The final article of the agreement (Article 3) contained some miscellaneous provisions, including the choice of English law and jurisdiction clauses mentioned earlier.

As referenced in the Share Transfer Agreement, the terms of the sale of the mv “GO Patoro” were contained in a Memorandum of Agreement dated 11 February 2011 (the “MoA”). Addendum No 2 to the MoA provided for payment of the purchase price of the vessel on the Delivery Date in the following manner:

The sale of the Shares and the purchase of the “GO Patoro” were duly completed on the Delivery Date of 23 February 2011. It remained necessary to determine the Final Purchase Price of the Shares under clause 1.4 of the Share Transfer Agreement following the production of the Delivery Date Accounts.

Although clause 1.4 of the Agreement provided for these accounts to be delivered to the parties by 10 April 2011, that did not happen. The Delivery Date Accounts were, however, eventually issued and were received by Mr Tartsinis on 6 July 2011. These accounts showed a Net Asset Value for GO Carriers of US$89,301,535. This included an amount of US$110,632,485 for the value of the vessels. That figure represented the net book value of the vessels (excluding the “GO Patoro”), i.e. their acquisition cost less accumulated depreciation.

On 15 November 2011 solicitors instructed by Mr Tartsinis and Mr Nikolaou wrote a letter to Navona claiming that a sum of over US$13.6m was payable to their clients under the Share Transfer Agreement. Most of this sum represented the difference between the value of the vessels shown in the Delivery Date Accounts and the value of the vessels used to calculate the Provisional Purchase Price under the Agreement (US$96.5m, including the “GO Patoro”).

Navona rejected the claim, asserting that the fleet value of US$96.5m mentioned in the Agreement was an absolute figure which was not subject to adjustment by reference to the Delivery Date Accounts. These proceedings followed.

In interpreting the Share Transfer Agreement, there are two main issues:

I will consider the issues in this order, albeit that there is some inter-relationship in the points of interpretation which bear on them.

The case of the claimant, Mr Tartsinis, is that the fleet value of US$96.5m referred to in the definition of “NAV” in the Share Transfer Agreement was clearly and unambiguously agreed to be provisional and to be subject to adjustment under clause 1.4 on the basis of the Delivery Date Accounts. Navona disputes this and contends that the fleet value of US$96.5m was intended to be a final figure, not subject to adjustment.

On behalf of Mr Tartsinis, Mr Peters submitted that:

As mentioned, Navona’s primary case is that, on the proper interpretation of the Agreement in its factual context, the fleet value of US$96.5m referred to in the definition of “NAV” was agreed to be a final figure. In support of this case, counsel for Navona pointed out that it is only the total NAV which is said in the definition to have been “provisionally calculated” and to be an “approximate” amount, and not the individual elements of the NAV. While this is true, if the intention was to differentiate between the elements of the NAV and to treat the most important element (the fleet value) as final while the other elements were provisional, it is reasonable to expect that there would be some wording in the Agreement to indicate this. Yet there is not. There are no words anywhere in the Agreement which say in terms or can be taken to imply that the figure of US$96.5m was intended to be a final figure or to have a different status from the other elements of the NAV.

Counsel for Navona also pointed out that clause 1.4 provides for the final determination and settlement of the NAV of GO Carriers to take place “following” the issuance of the Delivery Date Accounts. They submitted that this means only that the one event was to take place after the other, and does not stipulate how the final determination and settlement of the NAV of GO Carriers is to be carried out. In particular, the Agreement does not say that all the figures used in the final determination of the NAV are to be taken from the Delivery Date Accounts. Again this is true, and I will need to consider later exactly how the mechanism provided in clause 1.4 was meant to work. Again, however, there is nothing in the language of clause 1.4 (or anywhere else in the Agreement) which states or provides any basis for inferring that one element of the NAV – the fleet value – may be dealt with in a different way or stands in a different relationship to the Delivery Date Accounts from any other element. Again, it is reasonable to expect that, if the intention had been that the fleet value should be treated differently, the Agreement would have contained some wording to that effect.

In the absence of any such wording, Navona is driven to contend that it was unnecessary to state that the fleet value of US$96.5m was a final figure, as this was so obvious that it went without saying. Navona argues that treating the fleet value as subject to adjustment on the basis of the Delivery Date Accounts makes no commercial sense. To make this argument, Navona relies on features of the context in which the Agreement was made.

There was ultimately little dispute between the parties about the relevant background which can be taken into account. Navona relies on the following facts, all of which were within the parties’ knowledge when they entered into the Agreement and which I accept are admissible for the purpose interpreting the Agreement:

In the light of these background facts, Navona argued that the obvious purpose of the adjustment mechanism in clause 1.4 of the Share Transfer Agreement was to ensure that exact figures were used in the final determination of the purchase price of the Shares for those elements of the net asset value of GO Carriers which were not quantifiable when the Agreement was executed or which could sensibly be left to accountants to verify. That did not apply to the fleet value, which was a matter which the parties could negotiate and agree themselves. Furthermore, it would make no sense to adjust the figure of US$96.5m referred to in the Agreement, which was obviously an assessment made by the parties of the vessels’ market value, by reference to a figure in the accounts which the parties knew would represent the net book value of the vessels – a figure of no commercial relevance which bore no relationship to the market value of the vessels.

I accept, and I did not understand Mr Peters in his closing submissions on behalf of Mr Tartsinis to dispute, that it could not sensibly have been intended that the fleet value used to calculate the final NAV would be the net book value of the vessels. This is so for at least the following reasons:

Recognising the force of these considerations, Mr Tartsinis did not seek to argue that the fleet was properly valued in the Delivery Date Accounts at net book value. Rather, he accepted that an estimate of the recoverable amount of the ships should have been made given the requirement in clause 1.4 for the Delivery Date Accounts to be prepared “pursuant to the IFRS standards”. To follow this point through, it is necessary to identify the relevant International Financial Reporting Standards (“IFRS”) and how they applied to the valuation of the vessels. This was the subject of evidence from expert accountants.

There was no or no significant disagreement between the experts either as to the relevant accounting standards or as to how those standards applied to the valuation of the vessels. Based on their evidence, I find the position to be as follows:

Applying these standards to the valuation of the GO Carriers fleet, both experts agreed that, if the cost model was used in preparing the Delivery Date Accounts, there were indications that the vessels were impaired. In particular, the accounts of GO Carriers for the year ended 31 December 2009 showed that during that year two vessels (“GO Faith” and “GO Pride”) were sold. The accounts recorded a loss on disposal of those vessels of US$10.5m – which was more than half of their net book value. Furthermore, the Delivery Date Accounts had to take account of the sale of the “GO Patoro” which resulted in a loss on disposal of US$10,655,452. These were clear indications of impairment. It was therefore necessary to estimate the recoverable amount of the vessels by ascertaining (a) their fair value less cost to sell and (b) their value in use. If, instead of the cost model, the revaluation model was used, it would also be necessary to estimate the fair value of the vessels. (For present purposes, the concept of “fair value” can be treated as synonymous with “market value”.)

It follows that, on the information available at the time when the Share Transfer Agreement was concluded, it was to be expected that Delivery Date Accounts prepared in compliance with IFRS would include a figure for the value of the fleet which represented an estimate either of the market value of the vessels or of their value in use. It would not be consistent with IFRS to value the vessels in the accounts at their net book value (i.e. simply the original acquisition cost less accumulated depreciation).

This finding undermines Navona’s argument that the fleet value of US$96.5m included in the definition of “NAV” in the Agreement must have been intended as a final and not a provisional figure.

In their closing submissions, counsel for Navona relied on the fact that in the 2009 accounts the fleet was carried at net book value, even though those accounts recorded that the “GO Faith” and the “GO Pride” had been sold during the year for an amount which was less than half of their net book value. Furthermore, the 2009 accounts had included a statement that “there is no indication of impairment” – which on the face of the accounts themselves was palpably false. Counsel for Navona argued that in these circumstances it was to be expected that the Delivery Date Accounts would be prepared in the same way as previous financial statements, and would again show the vessels at their net book value.

Navona’s argument shows that the 2009 accounts did not comply with IFRS. I do not consider, however, that the Share Transfer Agreement can properly be interpreted on an assumption that the Delivery Date Accounts would not be prepared in accordance with IFRS, when the Agreement made compliance with IFRS a specific contractual requirement. Given that requirement and the relevant accounting standards, which the parties either knew or could readily have found out, it seems to me that they must be taken to have anticipated that the Delivery Date Accounts would need to value the GO Carriers fleet on the basis of market value or value in use, and not net book value.

In their closing submissions, counsel for Navona also argued that nevertheless no reasonable people in the position of Mr Tartsinis and Mr Nikolaou would have agreed to an arrangement which made the final determination of the value of the vessels dependent on the value shown in the Delivery Date Accounts. To do so would be to put themselves in the hands of Mr Lentoudis, who would have control over the preparation of the Delivery Date Accounts and would be able to ensure without either contravening IFRS or meeting objection from the auditors that the accounts put a low value on the vessels.

I think there is force in this argument. There is no doubt, and the parties clearly understood, that when the Delivery Date Accounts were prepared GO Carriers would be under the sole control of Navona, which was controlled by Mr Lentoudis. That was already the de facto position before the Share Transfer Agreement was concluded. After Mr Tartsinis and Mr Nikolaou had transferred their shares, the entire share capital of Newport, which was the sole shareholder of GO Carriers, would be owned by Navona. Mr Tartsinis and Mr Nikolaou would no longer have any interest in those companies. If they did not immediately resign as directors of GO Carriers, they could be removed.

Responsibility for preparing the financial statements of a company of course lies with its management. Where figures included in the accounts involve an element of judgment, there is often more than one judgment which the management of the company may make which is consistent with the accounts showing a true and fair view. In preparing the Delivery Date Accounts that would apply in deciding what figures to include for the value of the vessels. Provided that the management of GO Carriers adopted a valuation of the fleet which was reasonable and supported by the opinion of an independent professional valuer or other reasonable evidence of market value, the auditors would be bound to certify that the accounts had been prepared in accordance with IFRS and showed a true and fair view of the company’s financial position.

These considerations show that agreeing to have the value of the fleet finally determined by reference to the Delivery Date Accounts was unlikely to be advantageous to the Sellers. However, as an argument directed at the meaning of the Agreement, it in my view falls a long way short of showing that reasonable people in the position of the parties could not have agreed that the final determination of the fleet value would be based on the Delivery Date Accounts. It is possible to envisage scenarios in which this could realistically have been agreed, particularly if Mr Tartsinis and Mr Nikolaou were very keen to sell their shares in Newport and were in a weak bargaining position. They might, for example, have believed that the figure of US$96.5m used in calculating the Provisional Purchase Price was at the bottom of or below the range of values which could reasonably be attributed to the fleet in preparing the Delivery Date Accounts. (It was in fact the evidence of Mr Tartsinis that he did believe this.) It is thus conceivable that they would prefer the fleet value to be determined on the basis of accounts which were audited and contractually required to comply with IFRS rather than accept as final a figure which they regarded as insupportable. Alternatively, they might have been desperate to sell and expecting the market to fall still further and willing in these circumstances to accept the best price that they could get, even on terms that the fleet value would be based on accounts prepared by GO Carriers, with at least the protection afforded by the requirement that the accounts must be audited and compliant with IFRS. To exclude these scenarios and all others that might possibly explain agreeing to treat the figure of US$96.5m as provisional it would be necessary to investigate facts about the motivations of the parties and the course of their negotiations which are not admissible for the purpose of interpreting their agreement.

There is also a need for caution in relying on arguments of “commercial common sense”, particularly when they conflict with the intention naturally to be inferred from the language which the parties have chosen to express their bargain. As Neuberger LJ said in Skanska Rashleigh Weatherfoil Ltd v Somerfield Stores Ltd [2006] EWCA Civ1732 at para 22:

“the court must be careful before departing from the natural meaning of the provision in the contract merely because it may conflict with its notions of commercial common sense of what the parties may must or should have thought or intended. Judges are not always the most commercially-minded, let alone the most commercially experienced, of people, and should, I think, avoid arrogating to themselves overconfidently the role of arbiter of commercial reasonableness or likelihood.”

More recently, in BMA Special Opportunity Hub Fund Ltd v African Minerals Finance Ltd [2013] EWCH Civ 416 Aikens LJ said at para 24:

“The starting point is the wording of the document itself and the principle that the commercial parties who agreed the wording intended the words used to mean what they say in setting out the parties' respective rights and obligations. If there are two possible constructions of the document a court is entitled to prefer the construction which is more consistent with ‘business common sense’, if that can be ascertained. However, I would agree with the statements of Briggs J in Jackson v Dear (Footnote: ¹),first, that ‘commercial common sense’ is not to be elevated to an overriding criterion of construction and, secondly, that the parties should not be subjected to ‘…the individual judge's own notions of what might have been the sensible solution to the parties' conundrum’. I would add, still less should the issue of construction be determined by what seems like ‘commercial common sense’ from the point of view of one of the parties to the contract.”

With these cautionary statements in mind, I return to the fact that parties who intended the fleet value to be fixed by the Share Transfer Agreement at US$96.5m and not subject to adjustment in the same way as the other elements of the NAV could reasonably be expected to have said so, and yet all the indications in the language of the Agreement are to the opposite effect. In these circumstances I find it impossible to say that an agreement to base the final valuation of the fleet on the Delivery Date Accounts is so plainly nonsensical that rational parties who contracted on the terms of the Share Transfer Agreement could not have made it. I conclude that on the proper interpretation of the Agreement the fleet value of US$96.5m was agreed to be a provisional and not a final figure.

The second question of interpretation which then arises is how the value of the fleet is to be finally determined and, in particular, whether Navona is bound by the value shown in the Delivery Date Accounts

Clause 1.4 states that the Delivery Date Accounts “should be delivered to the Parties” by 10 April 2011. It does not say which of the parties is responsible for ensuring that the accounts are produced. It would make no sense to suppose that neither the Sellers of the Shares nor Navona had such an obligation, so that they would both be left without any recourse and unable to determine the Final Purchase Price of the Shares if no accounts were produced. That would make the Agreement unworkable, and neither party contended that it should be interpreted in that way.

Navona argued that, in circumstances where the accounts in question are those of an entity (GO Carriers) which is not a party to the Agreement and clause 1.4 does not in terms impose the obligation to arrange for the accounts to be issued on one contracting party or the other, the appropriate inference is that the obligation rested on them both. Mr Tartsinis argued that, in circumstances where GO Carriers was under the exclusive control of Navona so that only Navona had the power to arrange for the accounts to be produced, the necessary implication is that the obligation was solely that of Navona.

In my view, the latter interpretation is to be preferred. Not only does it reflect the legal and practical reality of Navona’s control over GO Carriers, but it also finds support in the wording of clause 1.4 itself. The clause contains the sentence:

“Solely for the purpose of determining the NAV of GO Carriers as of [the] Delivery Date, the Sellers are granted access to the books and records of GO Carriers.”

The grant of this right of access (impliedly by the other party to the Agreement, Navona) shows that it was understood (a) that the Sellers would have no involvement in the production of the Delivery Date Accounts and no ability to verify the figures in the accounts after they had been produced unless granted access to the books and records of GO Carriers, and conversely (b) that Navona not only did not need to be granted a right of access to the books and records of GO Carriers but was in a position to grant such a right to the Sellers. This confirms that responsibility for the production of the accounts was understood to fall on Navona.

The next question is how the Delivery Date Accounts are to be used once they have been produced. On behalf of Mr Tartsinis, Mr Peters submitted that the provision in clause 1.4 that the final determination of the NAV of GO Carriers is to take place by, at the latest, 30 April 2011 indicates that there was a window of time after the accounts were delivered within which the Sellers had to decide whether to dispute any of the figures in the accounts – for which purpose they were granted access to the books and records of GO Carriers. If no such dispute was raised, the final determination of the NAV of GO Carriers was required to be made using the figures in the Delivery Date Accounts. If, on the other hand, any of the figures in the Delivery Date Accounts was disputed by the Sellers, it was for the parties to try to resolve the dispute and arrive at a final determination of the NAV by agreement, failing which the dispute would have to be resolved by the court. On this analysis the right to challenge the figures stated in the Delivery Date Accounts is a right of the Sellers only. There is no scope for Navona to take issue with the figures or to refuse to agree a final determination of the NAV of GO Carriers and Final Purchase Price of the Shares which reflect those figures.

Navona’s case is that nothing in clause 1.4 of the Agreement justifies differentiating between the parties in this way. Mr Bright QC on behalf of Navona did not dispute that, if the Delivery Date Accounts were properly prepared in accordance with IFRS, clause 1.4 of the Agreement required the figures in those accounts to be used for the final determination of the NAV of GO Carriers. If that were not so, the arrangements for having the accounts prepared and audited would indeed be pointless. He argued, however, that it is equally open to either party to assert that the accounts produced did not satisfy the requirements of clause 1.4, including the requirement of compliance with IFRS, and for that reason were not conclusive of the NAV of GO Carriers. It is therefore open to Navona to rely on the point that, as I have found, the fleet was not valued in the Delivery Date Accounts in accordance with IFRS. It follows that Navona is not bound to accept the fleet value stated in the Delivery Date Accounts and, as the parties have been unable to agree the appropriate figure, the court must decide it.

Counsel for Navona sought to draw support for their interpretation from the decision of the Court of Appeal in Universities Superannuation Scheme Ltd v Marks & Spencer Plc [1999] L & TR 237. That case concerned the interpretation of provisions in a lease under which the landlord was required to calculate the amount of the service charge payable by the tenant and to issue a certificate setting out the calculation. I will not take a detour to discuss the reasoning in that case because it is seldom, if ever, helpful in deciding how to interpret particular contractual provisions to refer to a case in which a court has interpreted different provisions of a differently worded contract made in a different factual context. This case is no exception. I am unable to derive any assistance in construing clause 1.4 of the Share Transfer Agreement from analysis of the service charge provisions at issue in the Universities Superannuation Scheme case.

I cannot accept that on the proper interpretation of clause 1.4 the position of the parties is symmetrical in the way contended for by Navona. I have already held that, interpreted against the background that the production of the Delivery Date Accounts was under Navona’s exclusive control, clause 1.4 is to be construed as imposing on Navona the obligation to ensure that accounts were produced in accordance with the clause. As part of that obligation, Navona had to ensure that the accounts were prepared in accordance with IFRS. In these circumstances, I do not think it could reasonably have been intended that, when making the final determination of the NAV of GO Carriers, Navona should be allowed to dispute figures contained in the Delivery Date Accounts by arguing that it was in breach of its own contractual obligation. Navona has no need for any such protection, and it would frustrate the clear intention that the Delivery Date Accounts should provide an effective means of finally determining the NAV of GO Carriers if Navona, after issuing the accounts, could then disown them in this way.

I also consider that the provision in clause 1.4 granting the Sellers access to the books and records of GO Carriers for the purpose of determining the NAV of GO Carriers as of the Delivery Date is an indication that the right to scrutinise and challenge figures in the Delivery Date Accounts after they had been delivered was intended to be a right of the Sellers only.

In my view, therefore, on the proper interpretation of the Agreement it is only the Sellers who are entitled to challenge the accuracy of the Delivery Date Accounts and to argue that figures contained in the accounts should not be used in the final determination of the NAV because the accounts contain an error or have not been prepared in accordance with IFRS.

I would add that, although Mr Peters was inclined to concede in his closing oral submissions that, in order to be valid, any such challenge to the figures had to be made by 30 April 2011, which is specified in the Agreement as the latest date by which the final determination of the NAV was to take place, I see no reason to interpret the Agreement as having that effect. Clause 1.4 does not say that the figures in the Delivery Date Accounts are to be treated as final and binding except in so far as the Sellers dispute them by 30 April 2011. In the absence of any wording which expressly requires that result, I consider the straightforward meaning of the words used to be that there is an obligation (on both parties, since it takes both parties to agree the NAV) to determine the final NAV by the date specified. In practice, a breach of this obligation – as with a breach of the obligation to produce the Delivery Date Accounts by 10 April 2011 – is unlikely to give rise to any separate remedy.

I would not exclude the possibility that if, after the Delivery Date Accounts have been issued and before the NAV has been finally determined and agreed, Navona notices an error in the accounts, it would be open to Navona to arrange for corrected accounts to be produced. In its Defence, Navona has in fact stated that it will provide audited accounts prepared in accordance with IFRS “in due course”. However, this has not been done. GO Carriers was dissolved on 1 October 2011, and Navona has chosen not to have the company restored and new accounts prepared and audited. On the facts, therefore, the question of whether or in what circumstances Navona may arrange for corrected accounts to be issued does not arise.

I do not accept that Navona can ignore or bypass the need to arrange for revised accounts to be produced by seeking to show that, if such accounts were to be produced in accordance with IFRS, they would be bound or likely to contain different figures from the Delivery Date Accounts actually issued. That would be equivalent to allowing Navona to dispute the accounts which it had the responsibility for issuing – an interpretation which I have rejected. It would also deprive the Sellers of the safeguard that the accounts should be audited, which is an essential part of the contractual mechanism.

Even if (contrary to my view) Navona is entitled in principle to object to the fleet value stated in the Delivery Date Accounts on the ground that the accounts were not prepared in accordance with IFRS, Navona faces an evidential problem. It has not adduced the evidence necessary to show that the failure to comply with IFRS resulted in an understatement of the fleet value.

Navona’s case is that the value of the fleet shown in the Delivery Date Accounts should have been its market value on the Delivery Date. Navona has obtained three retrospective valuations from reputable professional valuers which value the fleet (excluding the “GO Patoro”) on 23 February 2011 at US$65.6m, US$66m and US$66.75m respectively and produce an average value of US$66,116,666. On this basis the NAV of GO Carriers on the Delivery Date was US$45,041,442 and Navona is entitled to recover US$2,397,975 from Mr Tartsinis.

The response of Mr Tartsinis to this counterclaim is that IFRS allows an entity to use the cost model to value fixed assets, as GO Carriers did. Where there are indications of impairment, the cost model, as mentioned earlier, requires the asset to be valued at the recoverable amount – which is the higher of (a) its fair value less the cost of sale and (b) its value in use. Although the valuations obtained by Navona are evidence of the fair value of the vessels, Navona has not adduced any evidence at all of their value in use. In their joint statement the accounting experts agreed that no value in use calculation had been disclosed in the proceedings, and this remained the position at the end of the trial. Accordingly, Navona has failed to establish the recoverable amount of the fleet and, specifically, to prove that it was less than the net book value. It has therefore failed to prove that Delivery Date Accounts misstated the fleet value.

Navona sought overcome this gap in its evidence in three ways. The first was to argue that it was not in fact permissible to prepare the Delivery Date Accounts using the cost model. Navona contended that, although IAS 16 allows a company to choose as its accounting policy either the cost model or the revaluation model, in preparing the Delivery Date Accounts that choice was constrained by clause 1.4 of the Share Transfer Agreement, which required the use of the revaluation model. That contention was based principally on the final sentence of clause 1.4, which states:

“The Parties agree that the Final Purchase Price as above calculated (on the basis of the Provisional Purchase Price and the adjustments as set out in this clause) is fair, reasonable and at market value.”

Mr Bright QC submitted that the parties could not sensibly have agreed this if they did not believe it to be true, and that the parties could not have believed that the Final Purchase Price was “fair, reasonable and at market value” if it was going to be calculated on the basis of a valuation of the fleet using the cost model.

This argument might have force if using the cost model required the vessels to be valued at their net book value. However, as I have indicated, it did not. If the cost model had been properly applied, an estimate should have been made of the recoverable amount of the vessels and that amount should have been shown in the accounts if it was less than the net book value. In these circumstances I see no reason why the parties could not sensibly have agreed that a valuation of the Shares which used the cost model to value the fleet was “fair, reasonable and at market value”.

Navona’s second response was to invite the court to infer that, if a calculation had been made of the value in use of the vessels, it would have yielded a figure approximately equivalent to their market value, and that it is legitimate in these circumstances to treat the evidence of market value as good evidence of the recoverable amount.

Mr Bright QC advanced the proposition that the value in use of an income-generating asset such as a ship should in theory correspond fairly closely with its market value. The operative words here are “in theory”. Whatever the economic theory of the matter under idealised assumptions, I see no reason to assume that the two valuation methods will in practice produce a similar result. I certainly do not feel able to make that assumption without evidence to that effect, of which there was none. Mr Bright QC sought in re-examination of Navona’s accounting expert, Mr Wilkinson, to elicit some evidence to support this contention, by asking Mr Wilkinson whether he was able to “give us any approximate indication as to the discrepancy or likely discrepancy or range of possible discrepancies between value in use and recoverable market value”. Mr Wilkinson’s answer was: “No, I am not”. There is no calculation or evidence of the value in use of the vessels, and no basis on which it is possible or permissible to speculate about what the result of such a calculation would be.

Navona’s third, fall-back argument, first put forward in closing submissions, was that, even if Navona cannot prove that an adjustment to the Provisional Purchase Price should be made in its favour following the issuance of the Delivery Date Accounts, nor can Mr Tartsinis. In circumstances where the accounts (and in particular the vessel valuations) were not prepared in accordance with IFRS as required by clause 1.4, neither party can require the figures in the accounts to be used to effect a final determination of the vessel values. The claim for an adjustment made by Mr Tartsinis must therefore fail as he is unable to rely on the Delivery Date Accounts to establish that an adjustment is due to him.

This argument rests on the premise that if or at any rate insofar as either party shows that the Delivery Date Accounts were not prepared in accordance with IFRS, the accounts cannot be used for the final determination of the NAV of GO Carriers, even if it has not been shown that the failure to comply with IFRS made any material difference to the figures stated in the accounts by causing any of those figures to be misstated. I see no basis for that interpretation either in the language of clause 1.4 or commercial common sense. There is nothing in the wording of the clause which says or implies that, if the Delivery Date Accounts do not comply with IFRS in some respect, this makes them a nullity. Furthermore, it would be absurd if Navona could prevent any upwards adjustment of the Provisional Purchase Price from taking place simply by procuring the issuance of Delivery Date Accounts which do not comply with IFRS and then refusing to correct them. I have already held that the requirement to ensure compliance with IFRS is an obligation resting on Navona (alone). Thus, to allow Navona to defeat a claim by relying on the failure of the accounts produced to comply with IFRS would be to enable Navona to take advantage of its own breach of contract.

Even if (contrary to my view) it is in principle open to Navona to defeat a claim for an upwards adjustment of the Provisional Purchase Price based on the Delivery Date Accounts by showing that the accounts which were produced did not comply with IFRS, it must in my view be incumbent on Navona to show that such non-compliance was material – that is, that it made a demonstrable difference to the NAV of GO Carriers. An interpretation of the Agreement which would allow Navona to take advantage of a failure to comply with IFRS to defeat a claim without even having to show what figures properly prepared accounts would have stated, and whether the figures in such accounts would in fact have been any different, is in my opinion untenable.

I conclude that, although Navona has shown that the valuation of the fleet for the purpose of the Delivery Date Accounts was not carried out in accordance with IFRS, it is not open to Navona to dispute the valuation on that basis in circumstances where Navona was responsible for the production the accounts and has failed either to produce corrected accounts or to prove that, if the accounts had been prepared in accordance with IFRS, the fleet value would have been different (and what it would have been).

It cannot be said that there is any merit in this result. I have found that reasonable people in the position of the contracting parties could not sensibly have intended that the NAV of GO Carriers and final purchase price of the Shares should be calculated using the net book value of the vessels. I have concluded, however, that this is precisely what, on the evidence, the Agreement on its proper interpretation requires. The reason for that conclusion is that, although the Delivery Date Accounts were not prepared in the way required by the Agreement, Navona has failed either to arrange for the accounts to be corrected or to prove what value would have been stated for the fleet if the accounts had been properly drawn.

In these circumstances, unless the Agreement can be rectified, I would hold that Mr Tartsinis is entitled to have the final determination of the NAV of GO Carriers and Final Purchase Price of the Shares calculated using the fleet value shown in the Delivery Date Accounts.

Navona contends that by 31 January 2011, at the latest, the parties had agreed that, for the purposes of determining the NAV of GO Carriers which would in turn determine the price of the Shares, the fleet would be valued at US$96.5m, and that this remained the common intention of the parties when the Share Transfer Agreement was executed. In these circumstances if – as I have held – the Agreement on its proper interpretation provided for the fleet value as well as the other elements of the NAV of GO Carriers to be adjusted by reference to the Delivery Date Accounts, Navona claims that this was a mistake and the Agreement should be rectified to reflect the parties’ common intention.

It is established law that, to succeed in a claim for rectification, a party must show that: (1) the parties had a common continuing intention, whether or not amounting to an agreement, in respect of a particular matter in the instrument to be rectified; (2) there was an outward expression of accord; (3) the intention continued at the time of the execution of the instrument; and (4) by mistake, the instrument did not reflect that common intention: see Chartbrook v Persimmon Homes [2009] 1 AC 1101, 1123, para 48, approving the summary of the law by Peter Gibson LJ in Swainland Builders Ltd v Freehold Properties Ltd [2002] 2 EGLR 71, 74, para 33.

It is commonly said, and was common ground between counsel in this case, that a party seeking rectification must prove its case to a high standard. In Joscelyne v Nissen [1970] 2 QB 86 the Court of Appeal endorsed the phrase “convincing proof”. In The Olympic Pride [1980] 2 Lloyd’s Rep 67 Mustill J referred to the need to prove the mistake “with a high degree of conviction” and to the agreement of counsel in that case that the court must be “sure”. The latter formulation reflects the criminal standard of proof but the case law does not support the proposition that, anomalously, the criminal standard of proof applies in rectification cases. In Joscelyne v Nissen [1970] 2 QB 86, 98, the Court of Appeal declined to import from the criminal law the test of proof beyond reasonable doubt; and in Agip SpA v Navigazione Alta Italiana SpA (The “Nai Genova”) [1984] 1 Lloyd’s Rep 353, 359, the Court of Appeal confirmed that the standard of proof required to establish the common intention of the parties is the civil standard of balance of probability. Furthermore, recent cases of high authority have made it clear that there is only one civil standard of proof and no higher standard which applies in particular categories of civil cases: see McMeel, ‘The Construction of Contracts’ (2^nd Edn) paras 17-81–17-88. Thus, in the case of In re B (Children)(Care Proceedings) (CAFCASS intervening) [2009] 1 AC 11, a decision of the House of Lords concerning whether allegations of sexual abuse of children made in care proceedings must be proved to a higher than usual standard, Lord Hoffmann (with whose speech Lords Rodger and Walker agreed) said at para 13:

“I think that the time has come to say, once and for all, that there is only one civil standard of proof and that is proof that the fact in issue more probably occurred than not.”

The explanation for the statements that “convincing proof” is needed where rectification is claimed lies in the very nature of the allegation that the written instrument does not record the parties' common intention. It is not, in truth, the standard of proof which is high, thereby differing from the normal civil standard, but that sufficiently strong proof is needed to counteract the evidence of the parties' intention displayed by the instrument itself: see Thomas Bates & Son Ltd v Wyndham’s (Lingerie) Ltd [1981] 1 WLR 505, 521, per Brightman LJ; The “Nai Genova” [1984] 1 Lloyd’s Rep 353, 359. As Lord Hoffmann has observed in relation to the interpretation of contracts, in words which are however equally applicable to claims for rectification, “we do not easily accept that people have made linguistic mistakes, particularly in formal documents”: Investors Compensation Scheme Ltd v West Bromwich Building Society[1998] 1 WLR 896, 913. The fact that the parties to a contract have approved particular language as the appropriate expression of their bargain is thus often itself cogent evidence that the document correctly records their common intention, so that convincing proof will be needed to displace that inference.

The evidential weight of the document itself clearly varies, however, according to the circumstances. This was recognised by Simonds J in observations in Crane v Hegeman-Harris Co Inc [1939] 4 All ER 662 at 664, which appear to be the origin of the expression “convincing proof”:

“[The jurisdiction to rectify an instrument] is a jurisdiction which is to be exercised only upon convincing proof that the concluded instrument does not represent the common intention of the parties. That is particularly the case where one finds prolonged negotiations between the parties eventually assuming the shape of a formal instrument in which they have been advised by their respective skilled legal advisers. The assumption is very strong in such a case that the instrument does represent their real intention ...” [emphasis added]

The point made here is simply the factual one that it is less likely that a document prepared after long negotiations with the help of skilled legal advisers will fail to record the common intention of the parties. Other things being equal, this is surely true. Much may also depend, however, on the clarity with which the document is worded. Where the meaning of the document is itself a matter of controversy and the subject of competing arguments of interpretation, it may be easier to displace the assumption that the meaning which the court identifies as the proper interpretation correctly records the parties’ common intention than where the meaning of the document is clear and unambiguous. A further relevant factor in the present case is that, although the Share Transfer Agreement was drafted in English by a lawyer, he was not a lawyer practising in an English-speaking jurisdiction or whose first language was English. Likewise, the three principals who signed the Agreement, although they all speak and read English well, do so as their second language.

As I described earlier, the essential function of the doctrine of rectification, as the law has long been understood, is to provide equitable relief in circumstances where the objective approach of the common law results in a document being interpreted in a way that does not reflect the actual intentions of its makers. Lord Wright explained the position very clearly in Inland Revenue Commissioners v Raphael [1935] AC 96, 143:

“...the principle of the common law has been to adopt an objective standard of construction and to exclude general evidence of actual intention of the parties; the reason for this has been that otherwise all certainty would be taken from the words in which the parties have recorded their agreement or their dispositions of property. If in some cases hardship or injustice may be effected by this rule of law, such hardship or injustice can generally be obviated by the power in equity to reform the contract, in proper cases and on proper evidence that there has been a real intention and a real mistake in expressing that intention: these matters may be established, as they generally are, by extrinsic evidence. The Court will thus reform or re-write the clauses in order to give effect to the real intention. But that is not construction, but rectification.”

The decision of the Court of Appeal in Joscelyne v Nissen [1970] 2 QB 86 established the need for “an outward expression of accord” between the parties. This requirement signifies the need for the parties to have manifested their intention. The importance of the requirement is clear. It would be capricious if a contractual document could be reformed to reflect the private intention of a party just because, although unknown to that party at the time, it turns out that the other party had a similar intention. What in principle is needed to justify rectification is a shared intention, that is to say a common intention which, as a result of communication between them, the parties understood each other to share. Such a shared intention is most obviously created by means of an express agreement. An express agreement is not, however, essential. As stated in Chitty on Contracts (31^st Edn) at 5-117: “the accord may include understandings that the parties thought so obvious as to go without saying, or that were reached without being spelled out in so many words”. To borrow an analogy from the philosopher David Hume: “Two men, who pull the oars of a boat, do it by an agreement or convention, tho’ they have never given promises to each other”. (Footnote: ²)

The requirement of an outward expression of accord may be said to introduce an element of objectivity into the test for rectification. But it only makes the test objective in the limited sense that objectively manifested intentions are required, and a mere coincidence of uncommunicated subjective intentions is not enough. It does not detract from the principle that rectification for mutual mistake depends upon proving that both parties actually were mistaken about the effect of the instrument. However, in Chartbrook v Persimmon Homes[2009] 1 AC 1101, the House of Lords, albeit obiter, endorsed a much more strongly objective test. Lord Hoffmann, with whose speech the rest of the House agreed, expressed the opinion (at paras 57-66) that, to establish a case for rectification, it is both necessary and sufficient to show a prior consensus which differs from the objective meaning of the contract and that for this purpose the meaning of the prior consensus is what a reasonable observer would have understood it to be and not what one or even both of the parties understood it to be. On this basis, had he not accepted Persimmon’s interpretation of the final contract, Lord Hoffmann would have held that Persimmon was entitled to rectification of the contract so that it reflected the way in which a reasonable observer would have understood earlier letters from Persimmon setting out the terms offered. That was so even though Chartbrook’s understanding of what both the letters and the final contract meant was different and, on the assumption made, the contract meant what Chartbrook thought it meant.

I am bound to say that, in company with Morgan J in Crossco No 4 Unltd v Jolan Ltd [2011] EWHC 803 (Ch) at para 253, I have difficulty with this statement of the test of rectification for common mistake. It is one thing to say that a contract should not be rectified just because both parties privately intend it to bear a meaning different from its meaning objectively ascertained. It is quite another thing, however, to say that a contract should be rectified to conform to what a reasonable observer would have understood the parties previously to have agreed, irrespective of the parties’ own understanding.

In the first place, I find it hard to see any equity in a doctrine which allows a party to obtain rectification of a document to reflect a view of what had been agreed that the party himself did not actually have, just because a reasonable observer would have taken this to be his view. Equally, I find it difficult to see the equity of imposing the view that a hypothetical reasonable observer would have formed of what had been agreed on a party who did not have that understanding of what had been agreed and whose understanding is reflected in the proper interpretation of the final document.

In these situations rectification is not serving to avoid the injustice that would otherwise be caused when the objective principle of interpretation leads to a result which fails to reflect the parties’ real intention. Instead, the effect is to treat the objective meaning of communications which were not intended to be legally binding as superior to the objective meaning of the document intended to record the parties’ final agreement and to allow the former to displace the latter. The looking-glass logic of this approach was exposed by Hobhouse LJ in Britoil plc v Hunt Overseas Oil Inc [1994] CLC 561, 573, when he said:

“It is contended that this sentence in this informal document which is not intended to have legal effect is to be treated as a superior statement of the parties’ agreement and is to displace the clear language of the considered and carefully drafted definitive agreement.

It can immediately be seen that this proposition needs to be carefully examined. As a matter of logic it can lead to the result that where there is a succession of documents of increasing formality but without legal effect leading up to a final considered legal document, the ascertainment of the actual agreement between the parties can be thrown back to the successively less formal, less considered and less carefully drafted earlier documents. This cannot be right.”

In the Britoil case the Court of Appeal – by a majority, with Hoffmann LJ dissenting – held that rectification for common mistake is available only where it is proved that both parties were in fact mistaken about the effect of the final document. Although the decision in the Britoil case was not disapproved, that requirement was not satisfied on the facts assumed in the Chartbrook case.

A further objection to the approach adopted in Chartbrook is that it allows a party who is mistaken about the effect of a contractual document although the other party is not to obtain rectification of the document without satisfying the requirements which must generally be satisfied before rectification will be granted on the basis of a unilateral mistake. Those requirements are the subject of a distinct body of case law. Broadly speaking, rectification will only be granted in cases of unilateral mistake where the other party knows of the mistake and it is inequitable for him to take advantage of it: see Chitty on Contracts (31^st Edn) at 5-122–5-129; Thomas Bates & Son Ltd v Wyndham’s (Lingerie) Ltd [1981] 1 WLR 505. The Chartbrook approach bypasses these established principles.

The view of the law endorsed by the House of Lords in Chartbrook v Persimmon Homes [2009] 1 AC 1101 has since been considered and applied by the Court of Appeal in Daventry District Council v Daventry & District Housing Ltd[2012] 1 WLR 1333. That was another case where, as was assumed to be the position in Chartbrook when discussing rectification, the parties were each mistaken in believing that their final agreement reflected their prior consensus but for different reasons: see paras 146 and 154. One party was wrong about the meaning of the pre-contractual documents (objectively understood) but right about the meaning of the legally binding contract (objectively construed). The other party was right about the meaning of the pre-contractual documents but wrong about the meaning of the legal contract. Applying the approach endorsed in Chartbrook, the latter party’s view prevailed. Toulson LJ said (at para 117) that the case “highlighted some real difficulties in the present state of the law governing rectification”. He pointed out some of those difficulties in cogent terms at paras 176-177. Nevertheless, he ultimately concluded that, despite his concern about the correctness of the principle in the Chartbrook case, the Court of Appeal ought to follow it. That was because (i) it would be a “bold course” not to follow the opinion of the House of Lords, (ii) the appeal had been argued on the basis that the approach in the Chartbrook case was correct, and (iii) that approach did not on the particular facts lead to an unjust result, although it would have been more satisfactory to reach that result on the basis of unilateral mistake (see paras 179-182 and 185).

Lord Neuberger MR reached the same conclusion as Toulson LJ and also thought, for similar reasons, that it was right to proceed on the basis of the analysis of rectification in Chartbrook, even though he agreed that “the analysis is not without difficulties” and “may have to be reconsidered or at least refined”: see paras 195-196.

Etherton LJ dissented in the result but in his judgment gave a summary of the law with which Lord Neuberger MR agreed. Etherton LJ at paras 85-88 distinguished four different factual situations and concluded (para 89) that this analysis showed “why it is good policy to favour objective accord or objective change of accord over subjective belief and intention in cases of mutual mistake”. I hope I may be forgiven for saying that the analysis seems to me much more compelling if an “objective accord” is taken to refer to a common understanding based on communication between the parties (i.e. what I have called a shared intention) rather than to an intention which an objective observer would have attributed to the parties, irrespective of what the parties themselves understood. Of potential relevance in the present case is the third scenario, described at para 87, namely:

“where there was objectively a prior accord, but one of the parties then subjectively changed their mind, but objectively did not bring that change of mind to the attention of the other party. It is right that, if the documentation gives effect to the objective prior accord, the formal documentation should not be rectified to reflect the changed but uncommunicated subjective intention; and if the documentation as executed reflects the changed but uncommunicated subjective intention, it should be rectified to give effect to the objective prior accord. To do otherwise would be to force on one of the parties a contract which they never intended to make on the basis of an uncommunicated intention and belief.”

The postulate that “one of the parties then subjectively changed their mind” presupposes that the “prior accord” reflected both parties’ subjective intentions; otherwise there would be no scope for one party to change their subjective intention. I can well see that where the parties have through communications “crossing the line” established a shared intention which the final document does not reflect, a claim for rectification should not be defeated just because, at some point before the document was executed, the defendant privately changed their mind without telling the claimant. On one view of the facts this was in effect the situation in the Daventry case. The disagreement between Etherton LJ and the majority of the Court of Appeal was essentially over whether the defendant had sufficiently communicated a change of intention by requesting a change in the wording of the draft contract in correspondence between solicitors. On the facts found, the defendant’s negotiator caused the change of wording to be made by misleading the defendant’s board and solicitors as to what had been agreed, and could not reasonably have believed when the contract was concluded that the change had been brought to the attention of the claimant’s negotiator. I cannot help wishing that the majority had felt able to analyse the situation as one of unilateral mistake.

Since the Daventry case was decided, Lord Toulson (as he now is) has stated extra-judicially that “in Daventrythe majority made it clear ... that we were not deciding that Chartbrookshould be followed in other cases” and that, when a similar problem arises:

“it will be a matter for argument whether a court should follow the reasoning in Britoilor in Chartbrook. In principle a court should follow a binding decision of the Court of Appeal rather than a later opinion expressed obiter by the House of Lords.” (Footnote: ³)

Lord Toulson nevertheless realistically acknowledged that a lower court may “feel hesitant” about deciding that the opinion expressed in the Chartbrook case was wrong.

Despite my very real misgivings, if I had concluded in this case that there is a disjunct between how the parties actually understood their pre-contractual discussions and what an objective observer would have thought their intention to be, I would have considered myself bound to follow the approach endorsed in Chartbrook and applied in Daventry. Happily, however, in the light of my factual findings reached below, this dilemma has not arisen.

As mentioned earlier, the exclusionary rules which apply when interpreting contractual documents do not apply on a claim for rectification. Thus, evidence of what was said in the negotiations is admissible, and generally essential. Evidence of subjective intentions is also admissible. In Chartbrook, at para 65, Lord Hoffmann confirmed that this is so even where the claim for rectification is based on a prior consensus said to have been expressed entirely in writing – although it is hard to see the justification for this if, as he considered, the only relevant perspective is that of an objective observer. Where, as in the present case, the relevant communications were largely oral, it is clear that the subjective understandings of the parties may provide evidence of what was in fact said and agreed.

The three people who signed the Share Transfer Agreement each gave evidence of how the Agreement came to be made and what was discussed during the pre-contractual negotiations. The only other person involved in some of the relevant discussions was Mr Vekris, the lawyer who drafted the Agreement. He has unfortunately died since the transaction took place. There are, however, various drafts of the Agreement which were found on his file containing his handwritten annotations. The negotiations were almost entirely oral and the only relevant emails are certain valuations of the ships and emails from Mr Vekris to the parties attaching successive drafts of the Agreement. The other main relevant document is a diary kept by Mr Tartsinis in which he wrote reminders to himself of things he had to do.

In expressing some opinions on the quality of the testimony of the three main witnesses, I should make it clear that my views are based not only on how they gave their evidence and its inherent probability but also on its consistency with the evidence as a whole.

It was clear to me that Mr Tartsinis has made such an emotional (as well no doubt as financial) investment in his claim and has so heavily reconstructed his beliefs about what happened as to deprive his testimony of any value. He has convinced himself that he never agreed to the fleet value of US$96.5m which is mentioned in the Agreement; that he never attended any meetings with Mr Lentoudis or Mr Vekris to discuss the preparation of the Agreement; that he never spoke to Mr Lentoudis during the relevant period, though he tried to do so on numerous occasions, as Mr Lentoudis refused to speak to him; and that he had minimal input into the drafting of the Agreement. If this evidence were true, it is difficult to imagine how any deal could have been done at all, let alone how the contract could have assumed the form that it did.

Having committed himself to these claims, Mr Tartsinis stuck to them in face of all evidence to the contrary. To illustrate the obduracy of his approach, it is sufficient to give two examples.

In the diary that Mr Tartsinis disclosed there is a column for each day with times set out down the side of the page. In the column for 3 February 2011 opposite 7pm Mr Tartsinis wrote “7pm KL/JV for GO Carriers sale stake?” It is not in dispute that “KL” refers to Mr Lentoudis and “JV” to Mr Vekris. A natural interpretation of this entry is that it records a meeting arranged – or which at least, given the question-mark, was potentially going to take place – between himself, Mr Lentoudis and Mr Vekris to discuss the sale of his stake in GO Carriers. If this interpretation is correct, it is difficult to square this entry with the evidence of Mr Tartsinis that he did not attend any meeting with Mr Lentoudis and Mr Vekris and that no such meeting was possible because Mr Lentoudis was not willing to meet or even speak to him. Rather than acknowledge that he might have been mistaken in making those claims, Mr Tartsinis sought to interpret this entry (and other entries) in his diary in any way that he could think of that might fit his case. In relation to this particular entry he claimed that he had written “7pm” (in the diary row for 7pm), not because a meeting was arranged for 7pm on that day, but to record that 7pm was the time when he was writing the note. I regard this explanation as absurd.

A second example of Mr Tartsinis’ approach concerns evidence that he gave about the chartering position of the vessels. A major problem for him was to explain why it made sense to agree that the value of the ships should be determined by accounts prepared by GO Carriers instead of by the parties agreeing their value. At one stage in his evidence Mr Tartsinis suggested that he could not agree a final figure for the fleet value until the accounts were prepared as he did not know until then if any of the vessels were on long term charters, which could dramatically affect their value.

It seems improbable to say the least that Mr Tartsinis would have entered into a contract to sell his interest in the GO Carriers fleet (held via his shares in Newport) without having found out whether any of the vessels were on long term charters – given his evidence that this could make a dramatic difference to the value of his shares. Mr Tartsinis attempted to suggest that he was unable to find out the chartering arrangements of the vessels. However, he had already accepted that during the period of the negotiations, as shown by entries in his diary, he was obtaining from Evalend information about the vessels such as the amount of bunkers on board and their class status. Mr Tartsinis sought to explain this by saying that he was in contact with the operations department of Evalend, whereas chartering was a matter handled by Mr Lentoudis who refused to speak him. However, it would be astonishing if the operations department of the company managing the vessels did not know their chartering position. Confronted with that point, Mr Tartsinis maintained that he did not in fact know whether the operations department knew about the vessels’ charters, as he did not ask them. Thus, to try to prop up his case Mr Tartsinis went from one improbable assertion to another. I regard the evidence of Mr Tartsinis that he did not know the chartering position of the vessels when he signed the Share Transfer Agreement as unbelievable.

A conspicuous feature of his testimony was the resentment that Mr Tartsinis clearly felt about Mr Lentoudis’ decision to depart from what he considered to be a ‘gentlemen’s agreement’ reached in November 2010 about the value of the vessels. That resentment was obvious throughout and was evidently shared to some extent at least by Mr Nikolaou. I think it likely that Mr Tartsinis has used his conviction, which itself has probably strengthened over time, that Mr Lentoudis dishonourably went back on their ‘gentlemen’s agreement’ to justify to himself the notion that the figure later proposed by Mr Lentoudis was never one that he actually agreed to or was bound to stick to. Overall, I consider the evidence of Mr Tartsinis to be self-serving and entirely unreliable.

Mr Lentoudis gave his evidence in a straightforward manner. While he did not recall or claim to recall details of the negotiations, he was clear and in my view entirely credible about the key points that mattered to him in the transaction. In particular, he was clear that no deal could have been done unless the value of the ships had been agreed.

Although I have approached the evidence of Mr Nikolaou with caution, I have in fact found his account of the negotiations to be the most informative of all the witnesses. He had a better memory of the relevant events than the others. He recalled some particular details of discussions which I consider entirely plausible. He also gave an insight, which made sense psychologically, into the thinking of himself and Mr Tartsinis when they entered into the transaction.

An attack was made on Mr Nikolaou’s credibility on the basis that he started the action as a claimant, with Mr Tartsinis, and in that capacity approved statements of case which contained assertions inconsistent with the evidence that he gave at the trial. Mr Peters on behalf of Mr Tartsinis submitted that on his own evidence Mr Nikolaou had allowed proceedings to be commenced and pursued in his name which were founded upon propositions of fact which he knew to be false, and that in these circumstances his evidence should be afforded little or no weight.

I do not see Mr Nikolaou’s conduct in such a dark light as Mr Peters sought to suggest. The thrust of his explanation of why he brought the proceedings with Mr Tartsinis was that Mr Tartsinis told him that there was a good technical argument based on the wording of the Share Transfer Agreement that they were entitled to have the final determination of the fleet value used to calculate the final price of their shares based on the Delivery Date Accounts, and that he went along with this hoping that the proceedings would put pressure on Mr Lentoudis to agree to a settlement in line with what Mr Nikolaou and Mr Tartsinis had both considered to be their original gentleman’s agreement with Mr Lentoudis. I am sceptical of Mr Nikolaou’s evidence that he did not realise until it was explained to him at the mediation held in February 2014 that the basis of the claim was that the vessels had been valued in the Delivery Date Accounts at their net book value, which was substantially higher than their market value: I think it likely that he understood this perfectly well all along. However, I do not consider it wholly unreasonable for Mr Nikolaou to have supported the case advanced by Mr Tartsinis as to the meaning of the Agreement in circumstances where I have found that case to be not merely arguable but legally sound. Whatever view is taken of the moral position, I do not think it was unreasonable from a legal point of view for Mr Nikolaou to make the claim that has been made in these proceedings, any more than it was for Mr Tartsinis to do so. The difference between them is that Mr Nikolaou, unlike Mr Tartsinis, later decided that it was wrong to pursue a claim based on contractual wording which did not reflect what had actually been agreed and understood between the three contracting parties.

I reject as unfounded the suggestion, somewhat faintly made, that Mr Lentoudis persuaded or pressed Mr Nikolaou to drop his claim. It was further suggested that in doing so Mr Nikolaou was motivated by hopes of doing business with Mr Lentoudis. Whether (which he denied) Mr Nikolaou has such hopes or not, I perceive his main motive as being loyalty to someone whom he had known personally for many years and whom he was not ultimately prepared to double-cross. Mr Tartsinis has never known Mr Lentoudis well and has evidently felt no similar inhibition.

I may also mention that Mr Nikolaou struggled to provide an answer as to what he had meant in a sentence of his witness statement which he had asked to delete. After stating that the price of the Shares was only intended to be provisional as regards the cash and loan figures, Mr Nikolaou had said that the parties were, however, still talking about the transaction in terms of valuation and it was his understanding that there could be further valuations of the fleet in order to finalise the figures. I take Mr Nikolaou to have meant by this that there was still the possibility after the fleet value of US$96.5m had been agreed that the parties could in the final determination of the purchase price agree a different figure. This is obviously correct and it may be that Mr Nikolaou and Mr Tartsinis held out a hope that Mr Lentoudis might be persuaded to revisit the fleet valuation and agree to an adjustment to it in their favour when the price of the Shares was finally determined. That is not to say, however, that the value of the fleet was ever intended to be determined on the basis of the figure included in the Delivery Date Accounts.

There is no material disagreement about what happened in the initial stage of the negotiations which took place in the autumn of 2010. It is clear that the impetus for the transaction came from Mr Tartsinis and Mr Nikolaou who wanted to sell their shares in Newport and hoped that Mr Lentoudis would agree to Navona buying them out. I accept the evidence of Mr Lentoudis that initially he was not interested in this proposal but that he subsequently agreed to buy the Shares, provided the transaction was structured in a way which did not require Navona to make any substantial payment. To avoid the need for such a payment, it was agreed that Mr Tartsinis and Mr Nikolaou would buy one of the vessels in the GO Carriers’ fleet and that the price payable for this vessel would be set off against the purchase price of the Shares.

In September 2010 two valuations of the GO Carriers’ fleet were obtained which are documented in emails sent to Mr Nikolaou. The first valued the fleet as at 7 September 2010 at US$108m (including a figure of US$20m for the “GO Patoro”). The email containing this valuation was forwarded by Mr Nikolaou to Mr Lentoudis on 9 September 2010. The second valuation was sent to Mr Nikolaou by email on 29 September 2010. This gave approximate values for the five ships as at 24 September 2010 totalling US$104.5m (including US$20.5m for the “GO Patoro”). It is clear that this valuation was also provided to Mr Lentoudis, as a hard copy of the email was found by Navona when conducting a search for the purpose of disclosure.

At some point after these valuations were obtained, agreement was reached that in principle Navona was willing to buy the shares of Mr Tartsinis and Mr Nikolaou as part of a deal which would also involve Mr Tartsinis and Mr Nikolaou buying the mv “GO Patoro”. According to Mr Tartsinis and Mr Nikolaou, there was a “gentleman’s agreement” as to the value of the fleet which would be used as the basis for the transaction. They had different recollections of what this value was. According to Mr Tartsinis, it was US$108m. According to Mr Nikolaou, it was US$106m. Mr Lentoudis agreed that figures of this kind were discussed but denied that there was any “gentleman’s agreement”.

I make no finding as to whether the term “gentleman’s agreement” is an appropriate description of the point which the deal reached. I do find, however, that there was discussion of a particular figure as one which at that time was agreed to represent a fair value for the vessels. I think it probable that the figure was US$106m. That is first because I consider the evidence of Mr Nikolaou generally more reliable than that of Mr Tartsinis, and second because it makes more sense to envisage that the parties selected a figure roughly midway between the two valuations which had been obtained rather than the higher of the two figures.

According to Mr Tartsinis and Mr Nikolaou, the value of the “GO Patoro” was set at US$18m. I am somewhat sceptical of this, since the two valuations used in the discussions valued that vessel at US$20m and US$20.5m, respectively. However, nothing turns on this point.

At some time in November Navona obtained verbal valuations from a broker for the four vessels apart from the “GO Patoro”. A record of the valuation was kept by the broker on its computer system and was provided in response to a request made in the course of these proceedings. The total amount of the valuations was US$80m. As no valuation was obtained for the “GO Patoro”, I infer that by the time these valuations were obtained the structure of the deal had been agreed and probably also the figure which was being used, at least provisionally, for the purpose of the overall fleet valuation. It is apparent that this November valuation placed lower values on the vessels than the two valuations obtained in September.

To buy the “GO Patoro”, and therefore to be able to conclude the deal, Mr Tartsinis and Mr Nikolaou needed a bank loan. They obtained agreement for such a loan from Marfin Egnatia Bank in an amount of US$13.5m, to be secured by a mortgage on the ship and personal guarantees from themselves. The terms of the loan were set out in a facility letter dated 15 December 2010. It was the evidence of Mr Nikolaou, which I accept, that in negotiating this loan from the bank he and Mr Tartsinis were proceeding on the basis of the fleet valuation as to which they thought they had a ‘gentleman’s agreement’ with Mr Lentoudis. I also accept his evidence that they were hoping to get a written agreement in place before Christmas but that by the time they had managed to arrange the loan Mr Lentoudis said that they should wait until after Christmas and New Year to have a contract prepared.

When the transaction was discussed again in early January 2011, however, Mr Lentoudis told Mr Nikolaou that, because ship values had fallen, he was not willing to proceed on the basis of the fleet value previously discussed and instead required a lower figure. Mr Nikolaou reported this to Mr Tartsinis. They were both extremely unhappy that Mr Lentoudis was, as they saw it, reneging on their “gentleman’s agreement” and seeking to renegotiate the deal.

It is at this point that the accounts given by the three main witnesses diverge. Whereas Mr Lentoudis and Mr Nikolaou say that Mr Tartsinis agreed to accept a value of US$96.5m for the fleet, Mr Tartsinis maintains that he did not.

Mr Tartsinis accepted that he was told by Mr Nikolaou some time after Christmas that Mr Lentoudis would only agree to a value for the fleet of US$96.5m. According to Mr Tartsinis, this figure was completely unacceptable to him and he told Mr Nikolaou that he would not to agree to it. He did, however, agree to a lower value of US$17.5m for the “GO Patoro”. His evidence was that he tried to speak to Mr Lentoudis to discuss the fleet value, but Mr Lentoudis would not answer his telephone calls. Mr Tartsinis then received a draft agreement, which Mr Vekris had prepared on instructions from Mr Lentoudis. According to Mr Tartsinis, he did not attend any meeting with Mr Lentoudis or with Mr Vekris to discuss the draft agreement. In his witness statements Mr Tartsinis said that his input into the drafting of the agreement was minimal, apart from insisting that the Delivery Date Accounts should comply with IFRS. In his oral evidence Mr Tartsinis said that he may have spoken to Mr Vekris on the telephone but he made very few comments on the draft agreement.

Mr Lentoudis could not remember exact dates or in exactly what conversations agreement on the fleet value had been reached; but he was clear that at some time during January 2011 the parties had agreed that the fleet value was to be US$96.5m. He thought that probably the agreement was initially concluded in discussion with Mr Nikolaou and was then confirmed at a meeting with Mr Tartsinis at Evalend’s offices.

According to Mr Nikolaou, he had two conversations with Mr Lentoudis after the New Year holiday. In the first conversation Mr Lentoudis said that, as the market had dropped since the fleet value had originally been discussed, he was only prepared to go ahead on the basis of a reduced figure of US$96m. Mr Nikolaou said that he discussed this change of stance with Mr Tartsinis. They were both unhappy with it. They also agreed that it was unfair to reduce the value placed on the whole fleet without also reducing the price of the “GO Patoro”. Mr Nikolaou then had a second conversation with Mr Lentoudis in which Mr Lentoudis agreed to amend the figures slightly, to US$96.5m for the fleet as a whole and US$17.5m for the “GO Patoro”. Mr Nikolaou reported this to Mr Tartsinis, and they agreed that if they wanted to proceed with the transaction, which they did, they had no choice but to agree to these figures. Their agreement was communicated by Mr Nikolaou to Mr Lentoudis by telephone. The Share Transfer Agreement was then drafted. In order to give instructions to the lawyer, Mr Vekris, who drafted the agreement, at least two meetings took place at the offices of Evalend which Mr Lentoudis, Mr Tartsinis and Mr Vekris all attended. Mr Nikolaou said that he was also present at the first meeting but might not have attended the later meeting or meetings.

I have no hesitation in rejecting the account of the negotiations given by Mr Tartsinis. On the whole of the evidence I am sure: (1) that he agreed to accept the valuation of the fleet at US$96.5m on which Mr Lentoudis insisted; (2) that he had no difficulty in speaking to Mr Lentoudis while the contract was being drafted (or at any other time) and attended at least one and in all probability two meetings with Mr Lentoudis and Mr Vekris to discuss the draft agreement; and (3) that at all times up to, and indeed after, the signing of the Share Transfer Agreement Mr Tartsinis – in common with Mr Nikolaou and Mr Lentoudis – understood that the fleet value of US$96.5m would be the value used to determine the final price of the shares and would not be subject to adjustment on the basis of the Delivery Date Accounts.

I have reached these conclusions for the following eight principal reasons.

First, I accept the evidence of Mr Lentoudis that without an agreement on the value of the vessels there could not have been a deal. As he said:

“That was the most important thing. The agreed value for the fleet. Otherwise what kind of deal we could have?”

And later in his evidence:

“If there would not be a final agreement on a value we could not have a deal. It is very, very simple. How would we construct otherwise a deal?”

I accept that in the way that the parties did business the matter was indeed as simple as this and that commercially, in another phrase used by Mr Lentoudis, it is “beyond any logic” that an agreement for the sale of the Shares could have been concluded without an agreement on the value of the vessels.

Mr Tartsinis accepted in cross-examination, albeit after some equivocation, that agreeing a value for the vessels would have been the normal way to proceed. It was, moreover, how the parties had proceeded in the autumn when they had agreed on what at that time was an appropriate value for the fleet. It is common ground that in the New Year Mr Lentoudis insisted on reducing the figure to US$96.5m on the footing that ship values had fallen. Mr Tartsinis claimed that he refused to agree to this figure but nevertheless waited for a draft agreement to be prepared with a view to seeing what it said. I do not regard this claim as coherent. I am satisfied that if Mr Tartsinis and Mr Nikolaou had not agreed the value of US$96.5m for the fleet on which Mr Lentoudis was insisting, Navona would not have agreed to buy their Shares and there would not have been a deal to embody in a draft agreement.

Second, it is not in dispute that there was agreement reached on the value of the “GO Patoro” at US$17.5m – which was lower than the value which had previously been discussed. It is unlikely that such agreement could have been reached in relation to one vessel unless the parties were able to agree a figure for the whole fleet. As Mr Lentoudis said when the proposition was put to him that only the value of the “GO Patoro” was fixed:

“How is that possible? How I agree only one ship and the other ships do not agree a value? There is no logic in that.”

One obvious reason why there would be no logic in such an agreement is that the value of the whole fleet and the value of each individual vessel in the fleet were related: indeed, the one figure included the other. Furthermore, the price of the “GO Patoro” was to be set off against a price for the Shares which depended on the value of the whole fleet, and that arrangement could only operate fairly if the values harmonised and were consistent with each other. It is difficult in these circumstances to see how any prudent businessman would have agreed to fix the value to be placed on the “GO Patoro” independently of the valuation of the other vessels.

Third, I have no doubt that Mr Lentoudis understood there to be a final agreement on the fleet value which was not capable of being affected by the Delivery Date Accounts. In particular:

It is in theory possible that there could have been a misunderstanding such that Mr Lentoudis believed that the parties had reached a final agreement on fleet value, whereas Mr Tartsinis believed that they had not. In practice, however, in a personal negotiation of this kind in which the fleet value was the key element in the price of the Shares, the realistic likelihood of such a misunderstanding seems to me vanishingly small.

Fourth, perhaps because he recognised the improbability that the transaction could have proceeded without an agreement on fleet value or at least without the parties knowing that they had not reached such an agreement if they were in communication with each other, Mr Tartsinis was insistent that during the negotiation of the Agreement he had no direct communication with Mr Lentoudis, who refused to meet him or answer his calls. I reject this evidence as untrue.

I accept that the negotiation of the fleet value, on which the deal depended, was conducted with Mr Lentoudis by Mr Nikolaou (who went back to Mr Tartsinis before representing their joint position to Mr Lentoudis). I also accept that the reason for this was that Mr Nikolaou knew Mr Lentoudis much better than Mr Tartsinis. I do not accept, however, that Mr Lentoudis refused to speak to Mr Tartsinis. In the first place, there was no reason for Mr Lentoudis to take that attitude. Second, it is difficult to conceive how they could have negotiated the contract if he had refused to speak to Mr Tartsinis. Third, the claim is contradicted by Mr Tartsinis’ own diary which contains a number of notes to himself to call Mr Lentoudis about a variety of matters. In particular there are such entries on 1 February (“Call KL for GO Patoro charts with bunkers”), 2 February (“Call KL for update”), 9 February (“Call KL for class status etc with MoA”), 14 February (“KL for details m/v GO Patoro”) and 17 February (a note of KL’s office telephone number). Mr Tartsinis sought to explain these entries either on the basis that “KL” did not mean Mr Lentoudis himself but someone in his office or on the basis that he kept trying to call Mr Lentoudis about various things but never managed to speak to him. I regard both these suggestions as incredible. My rejection of Mr Tartsinis’ evidence on this point undermines both the realistic likelihood that no consensus was reached on fleet value and the general credibility of his evidence.

Fifth, the same applies to Mr Tartsinis’ evidence that he did not attend any meetings with Mr Lentoudis or Mr Vekris to discuss the drafting of the Agreement. Again, I reject that evidence as untrue, in particular because:

Sixth, I regard the evidence of Mr Tartsinis that he was content to enter into the Share Transfer Agreement without an agreed fleet value as incapable of belief. As is shown by the figures in Addendum No 2 to the MoA, on the basis of the provisional NAV there was a shortfall of US$1.5m between the balance of the purchase price of the Shares (after repayment of the loan to Navona and drawdown of the new bank loan) and the price payable for the “GO Patoro”. That shortfall had to be funded by Mr Tartsinis in cash, as Mr Nikolaou did not have the money. There is evidence that they explored the possibility of a further loan secured by a second mortgage on the vessel but without success. The numbers were therefore tight. In these circumstances I think it plain that Mr Tartsinis and Mr Nikolaou could not afford to take a risk that the final purchase price of the Shares might be the subject of a substantial downwards adjustment because a value lower, perhaps significantly lower, than US$96.5m might subsequently be placed on the fleet. That could potentially have required them to pay a further sum in cash which would render the transaction unviable. In these circumstances it would not have made sense to conclude the deal without certainty as to the value placed on the vessels.

Nor would it have made sense for Mr Tartsinis and Mr Nikolaou to put themselves in the hands of Mr Lentoudis by agreeing that the final price of the Shares should depend on the valuation of the fleet in accounts prepared under his control. The explanation given by Mr Tartsinis for why he was prepared to do this was to the effect that: (i) Mr Tartsinis knew that the accounts would have to satisfy the auditors and comply with IFRS and that he could object to them if they did not; (ii) to comply with IFRS it would be necessary to get a proper independent professional valuation of the ships (and, if the cost model was used, to calculate their value in use); and (iii) Mr Tartsinis was confident that the fleet was worth more than the US$96.5m which Mr Lentoudis ascribed to it and it was therefore to his advantage not to agree this figure and instead to have the value determined by reference to the Delivery Date Accounts.

The key assertion here is the last. If Mr Tartsinis was going to stake his fortunes on a belief that the Delivery Date Accounts, if compliant with IFRS, could be counted on to include a fleet value greater than US$79m (when the “GO Patoro” was excluded), it is natural to expect that he would have contacted one or more valuers and obtained estimates of the value of the ships; otherwise he could have no confidence that a professional valuation was going to produce a favourable result. However, there is no credible evidence that he did. It is also reasonable to expect that Mr Tartsinis would have made or obtained an estimate of the vessels’ value in use. Again, however, there is no suggestion that he did.

What is more, the valuations obtained by Navona in the course of this litigation from three different professional valuers have, as mentioned earlier, produced total figures (excluding the “GO Patoro”) as at 23 February 2011 ranging from US$65.6m to US$66.75m – all that is, very substantially less than the figure of US$79m (again excluding the “GO Patoro”) which Mr Lentoudis wanted. On the basis of this evidence I find that, if Mr Tartsinis had taken soundings from valuers, he would have known that a valuation of the fleet obtained by Evalend for the purpose of the Delivery Date Accounts would be almost certain to produce a result to his considerable financial detriment. I think it inconceivable that Mr Tartsinis would have agreed to be bound by whatever IFRS-compliant value was used in the Delivery Date Accounts either (a) without any real idea what the number was likely to be or (b) knowing that it was likely to require him to make a substantial payment. I am sure that the reality is that Mr Tartsinis never in fact gave any thought at all to how the fleet could or should be valued in the Delivery Date Accounts, the reason being that it never occurred to him that the figure shown in the accounts would be used to determine the final price of the Shares.

Seventh, I am confirmed in that view by the reaction of Mr Tartsinis after he received the Delivery Date Accounts on 6 July 2011. On 14 July 2011 a letter was sent to Evalend by the Greek branch of Watson, Farley & Williams, solicitors instructed by Mr Tartsinis, which said that Mr Tartsinis was currently in the process of reviewing the accounts and that, from his initial review, he had identified one or two items which required further clarification. There no suggestion in this letter that on the basis of the accounts a large sum of money was owing to Mr Tartsinis. Mr Tartsinis acknowledged in his evidence that it was only after he reviewed the position several weeks later with the assistance of Price Waterhouse that this became apparent to him. I regard it as inconceivable that this realisation would not have struck him immediately on receiving the accounts if he had believed that the final fleet value would be based on the value shown in the accounts.

Eighth, I accept the evidence given by Mr Nikolaou as a substantially accurate account of the negotiations. Amongst other points, I find convincing his description of the thought process which led him and Mr Tartsinis to agree to the fleet value on which Mr Lentoudis insisted:

“what was happening was that the market was dropping, the values were dropping, the market values were dropping and the loan was standing still. So we knew that after a while, if the market drops more, we would owe to Mr Lentoudis, we would come to a point, it’s purely a simple mathematics, that we would owe money to him, and instead of coming to that point I said, ‘Mihail, let’s close the thing now, at least we have a deal,’ and he agreed.”

By far the strongest objection to Navona’s case that the fleet value had been finally agreed – to which I have given full weight – is the wording of Share Transfer Agreement itself. There is real force in the point that, if the fleet value of US$96.5m was intended to be a final figure, not subject to adjustment, it is natural to expect that this would have been recorded in the Agreement and that the Agreement would not have been drafted in a way which draws no distinction between different elements of the NAV of GO Carriers and appears to make them all subject to adjustment following the issuance of the Delivery Date Accounts. Forceful as this point is, however, it cannot stand in the face of the convincing proof that the fleet value figure of US$96.5m was agreed and understood by all three parties to be a final figure.

I am led to conclude that this understanding was so basic that no one actually thought that it needed to be spelt out, and that when discussing the need for an adjustment mechanism based on audited accounts it was simply taken for granted that the relevance of this mechanism was only to items which were seen as matters of accountancy rather than the ship values – which were seen as a commercial rather than an accounting matter. This is of course the explanation that I rejected as a valid interpretation of the Agreement. Interpretation, however, is a different exercise and, as I have indicated, the evidence of what was actually discussed in negotiating the Agreement and what was in the minds of the parties could not be considered in that context. When this evidence is taken into account, I am persuaded that it did not in fact occur to the parties that it was necessary to articulate that the fleet value was agreed and was not an element of the NAV of GO Carriers to which the Delivery Date Accounts would be relevant.

As well as putting understandable emphasis on the wording of the Agreement, Mr Peters on behalf of Mr Tartsinis submitted that there was no clear or satisfactory evidence as to when or how an agreement on the fleet value was allegedly reached: he argued that the testimony of Mr Lentoudis on these matters was extremely vague and that of Mr Nikolaou was untrustworthy. Mr Peters submitted that in these circumstances the evidence comes nowhere near to the level of clarity and cogency which is needed to prove a claim for rectification.

For the reasons given earlier, I do not accept the criticisms of the evidence of Mr Lentoudis and Mr Nikolaou. Nor, more importantly, do I accept the premise that the dispute comes down essentially to a swearing match and a question of one person’s (or two people’s) word against another’s. When the commercial realities of the negotiations are understood and the evidence is viewed as a whole, it is in my view clear that there was an agreement reached on fleet value. It is not necessary in order to be satisfied of this to establish exactly when and how the agreement was reached. In fact, however, with the assistance of Mr Tartsinis’ diary and the various drafts of the Agreement I think it possible to reconstruct with reasonable accuracy the chronology of what occurred.

On 19 January 2011 Mr Tartsinis made an entry in his diary: “AN GO Carriers at $96.5 and $17.5m for Patoro”. I think it probable that the negotiation of the value of the ship after Mr Lentoudis insisted on a lower valuation took place before this note was made and that Mr Tartsinis was here recording the final outcome of the negotiation, being the figures agreed by Mr Nikolaou on behalf of them both with Mr Lentoudis for (i) the value of the whole fleet and (ii) the value of the “GO Patoro”. The next step was therefore to arrange for an contractual document to be prepared to implement the deal.

The first draft of the Share Transfer Agreement is dated 31 January 2011. Mr Vekris sent it by email to Mr Tartsinis, copied to Mr Lentoudis, on 2 February 2011. I accept the evidence of Mr Lentoudis and Mr Nikolaou that the draft agreement was prepared following a meeting attended by Mr Lentoudis, Mr Nikolaou, Mr Tartsinis and Mr Vekris, at which the three principals explained the basic structure of their deal to Mr Vekris and discussed with him the shape of the agreement needed to implement it.

The diary kept by Mr Tartsinis has an entry for Tuesday, 25 January 2011 at 5pm which states: “Evalend-Vekris GO”. On Wednesday, 26 January Mr Tartsinis has written notes to himself saying “which lawyer to see Vekris Agreement??” and “arrange AM for cancelled meeting what/when??” The second entry in the diary on Monday, 31 January 2011 states: “Vekris Agreement to read/prepare”. A reasonable interpretation of these entries is that a meeting was initially arranged for 5pm on 25 January to take place at Evalend’s offices for Mr Tartsinis, Mr Nikolaou and Mr Lentoudis to give instructions to Mr Vekris to enable him to prepare the draft agreement. Mr Tartsinis was contemplating instructing another lawyer of his own to review the draft agreement, though in the event he chose not to do so. For some reason, the proposed meeting was cancelled and had to be rearranged. It may have been rearranged over the weekend of 29/30 January to take place first thing on the Monday morning, which could explain why the time of the rescheduled meeting was not entered by Mr Tartsinis in his diary. The note of 31 January “Vekris Agreement to read/prepare” may be a reminder which Mr Tartsinis wrote to himself after the meeting had taken place that Mr Vekris would be preparing a draft agreement that Mr Tartsinis would need to read.

The first draft of the Share Transfer Agreement contains a recital which states:

“The Parties and AN have agreed that the Net Asset Value (“NAV”) of Go Carriers and of Newport equals the sum of 57.000.000 [PLEASE CONFIRM] as of today, and accordingly Seller’s share (20%) to the NAV of Go Carriers amounts to U.S$11.400.000 [PLEASE CONFIRM].”

The “Parties” were defined as Mr Tartsinis and Navona; it seems that it was not intended at this stage that Mr Nikolaou (“AN”) would be selling his shares. Clause 1.2 of the draft provided for the shares of Mr Tartsinis to be transferred to Navona for a total price of US$11.4m, “which is equal to 20% of the NAV of GO Carriers”. Clause 1.4 provided that the price of the shares “may be adjusted upwards or downwards following the final determination and settlement of the NAV of GO Carriers” following the issuance of management accounts of GO Carriers for the first three months of 2011. There were square brackets around this clause. I infer that it was uncertain at this stage whether the figure of US$11.4m would be the final price of the shares or whether it would be subject to adjustment after up-to-date accounts of GO Carriers had been prepared. I also infer that there must have been discussion not only of the fleet value, but also of the other elements of the Net Asset Value of GO Carriers in order to arrive at the sum of US$57m mentioned in the draft.

There is a copy of the first draft of the agreement found on the file of Mr Vekris which contains his handwritten notes. On the first page above the recital he has written:

“NAV = value πλοίων [i.e. of the ships] – liquid + cash”.

Under the reference in this formula to the fleet value Mr Vekris has written “96.5”. Mr Vekris also made notes on the draft agreement to indicate that the figures of US$57m for the NAV of GO Carriers and US$11.4m for the price of the shares were to be provisional and that there was indeed to be an adjustment mechanism (based on audited, not management accounts).

Mr Vekris produced a second draft of the Agreement dated 4 February 2011 in which changes from the first draft were marked up. There was then another iteration of this second draft in which further changes were included. A clean copy of the second draft was sent by email to Mr Tartsinis and Mr Lentoudis on 10 February 2011. The email also attached a draft of Addendum No 2 to the MoA for the sale and purchase of the “GO Patoro”.

The first entry made by Mr Tartsinis in his diary for Thursday, 3 February 2011 states: “review Vekris agreement + prepare MOA for GO Patoro with Evalend”. Opposite the time of 7pm on that day is the entry to which I referred earlier: “7pm KL/JV for GO Carriers sale stake?” On Friday, 4 February there is an entry which says: “MOA GO Patoro to receive from Evalend”. On Monday, 7 February there is a reminder to “call Vekris for agreement”. I infer that after the first draft of the Share Transfer Agreement was circulated by Mr Vekris on the afternoon of 2 February 2011 a further meeting was arranged between Mr Tartsinis, Mr Lentoudis and Mr Vekris to discuss the draft agreement. In preparation for this meeting Mr Tartsinis reviewed the first draft (see his first note on 3 February). The meeting either took place at 7pm on 3 February or possibly was deferred until first thing the next morning. It is probable that Mr Nikolaou was not present at this meeting. After this meeting Mr Tartsinis made a note to himself that he was to receive a draft of the MoA for the sale and purchase of the “GO Patoro” (as well as the second draft of the Share Transfer Agreement). On 7 February he had not yet received the second draft of the Share Transfer Agreement and made a note to chase up Mr Vekris. He probably saw the first iteration of the second draft on that day or the next day, as on 8 February there is an entry in his diary “comments on MOA + Vekris agreement”.

The main body of the MoA was signed on 11 February 2011. Some time before 14 February the decision was made that Mr Nikolaou would also be selling his shares in Newport. Mr Vekris prepared a third draft of the Share Transfer Agreement dated 14 February 2011 which reflected this change. He sent the draft by email to Mr Tartsinis on that day. It is not clear whether any further meeting took place to discuss the agreement either before or after this draft was produced, and I make no finding that there was a further meeting. The third draft was substantially similar in form to the final version of the agreement, save that there was not yet a requirement for the Delivery Date Accounts to be prepared in accordance with IFRS. As I have mentioned, it was the evidence of Mr Tartsinis, which I accept, that the reference to IFRS was inserted at his request. I think it likely that this and any other comments made by Mr Tartsinis on the third draft of the agreement were conveyed in a telephone call to Mr Vekris.

As mentioned, the Share Transfer Agreement was signed in its final form on 23 February 2011, shortly after the closing of the sale and purchase of the “Go Patoro”. Both meetings took place at the offices of Marfin Bank.

I have found that, before the process of drafting the Share Transfer Agreement began, the parties had agreed the value to be placed on the fleet for the purpose of the transaction. It has not been suggested by Mr Tartsinis (or anyone else) that during the process of drafting the Agreement there was any further negotiation of the fleet value or of how it should be determined. It was, as I have indicated, the evidence of Mr Tartsinis that he had no communication with Mr Lentoudis at all during this period and that his input into the drafting of the Agreement was minimal. I have rejected that evidence, but there is no indication that Mr Tartsinis raised the subject of the fleet value at any point either with Mr Lentoudis or with Mr Vekris. I have no doubt that this was because Mr Tartsinis recognised that there was no possibility of persuading Mr Lentoudis to revisit the fleet value and that, if he tried to do so, he would be putting the deal at risk.

I conclude that at no point before the Share Transfer Agreement was signed did Mr Tartsinis signal any intention to depart from the agreement made orally on or about 19 January 2011 that the GO Carriers’ fleet was to be valued at US$96.5m for the purpose of calculating the price of the Shares.

For the reasons given, I have reached the following conclusions:

In the light of these conclusions, I am satisfied that it is just and equitable to make an order for rectification of the Agreement. I consider that the appropriate way of doing so is to add a further sentence in the definition of “NAV” to state: “Unlike the other elements of the Net Asset Value, the Fleet Value of US$96.5m is a final figure, not subject to adjustment in accordance with Clause 1.4”. I understand it to be agreed that the consequence of rectifying the Agreement will be that a further sum of US$178,681.70 is payable by Navona to Mr Tartsinis.

I would finally add that, if I had accepted Navona’s alternative case on the interpretation of the Agreement and found that under the terms of the Agreement Navona is entitled to a payment from Mr Tartsinis representing the amount by which the market value of the fleet on the Delivery Date was less than US$96.5m, I would not have given judgment for that sum. Although Mr Tartsinis has not made a claim for rectification, Navona has; and I have concluded on the evidence given at the trial that the Agreement ought to be rectified. It would not be just to decline to rectify the Agreement because of the order in which Navona has chosen to advance its various alternative arguments.