- New Zealand
- SC 65/2008  NZSC 5
- Supreme Court of New Zealand
- Vector Gas Ltd. v. Bay of Plenty Energy Ltd.
LONG-TERM CONTRACTS - GAS SUPPLY CONTRACT - BETWEEN TWO NEW ZEALAND COMPANIES - REFERENCE TO UNIDROIT PRINCIPLES AS A MEANS OF INTERPRETING APPLICABLE DOMESTIC LAW (NEW ZEALAND LAW)
CONTRACT INTERPRETATION - ADMISSIBILITY OF EXTRINSIC EVIDENCE AND IN PARTICULAR OF PRE-CONTRACTUAL NEGOTIATIONS FOR INTERPRETATION OF WRITTEN CONTRACTS - REFERENCE TO UNIDROIT PRINCIPLES (ART. 4.3)
Two New Zealand companies entered into a gas supply contract, according to which the buyer had to pay the supplier, “for each GJ supplied, the difference between the price set out in the Agreement and $6.50 per GJ”, plus interest.
A dispute arose between the parties as to whether the expression “$6.50 per gigajoule” meant that the buyer must pay the supplier for each gigajoule of gas supplied $6.50 plus transmission charges of approximately $1.86 per gigajoule, or simply $6.50 per gigajoule on the basis that supplier was responsible for supplying the gas at its own expense to buyer’s premises within that price.
The High Court held that transmission costs were not included in the price, while the Court of Appeal ruled in favor of the buyer, stating that transmission costs were included in the price as indicated in the contract. The supplier appealed.
The Supreme Court allowed the supplier's appeal and restored the decision of the High Court. In so doing, the majority of the Court relied on the correspondence between the parties' lawyers before the final agreement had been reached. The Court was however divided as to the basis on which that evidence was referred to. Two of the judges (Blanchard J and Gault J) would have admitted evidence of pre-contractual negotiations only as part of the overall context and commercial purpose of the agreement while another member of the Court (McGrath J) thought pre-contractual negotiations should generally be excluded from consideration except where they gave rise to an estoppel by convention or a claim in rectification. On the contrary, the remaining two judges (Tipping J and Wilson J) concluded that pre-contractual negotiations should be admissible in contract interpretation as a general principle. In particular, Tipping J, while recalling that English law has always required interpretation issues to be addressed on an objective basis and rejected the subjective approach as provided for instance in Article 4.3 of the UNIDROIT Principles, pointed out that even under English law extrinsic evidence, such as prior negotiations, is excluded only if intended to demonstrate the subjective intention or understanding of only one of the parties and not when it tends to establish a fact or circumstance capable of demonstrating objectively what meaning both or all parties intended their words to bear. On the basis of this argument the judge, supported by Wilson J, decided in favour of admitting the evidence of the parties’ intentions as expressed in the course of their negotiations thereby, though without expressly admitting it, in substance reaching the same conclusion as would follow from Article 4.3 of the UNIDROIT Principles.
JUDGMENT OF THE COURT
A The appeal is allowed.
The respondent is to pay to the appellant costs of $15,000 together with reasonable disbursements as fixed by the Registrar.
Costs in the lower Courts are to be fixed by those Courts in the light of this Court’s judgment.
Blanchard J 
Tipping J 
McGrath J 
Wilson J 
Gault J 
 This case concerns the interpretation of a contract for the supply of gas. The question is whether the expression “$6.50 per gigajoule” means that the respondent purchaser (BoPE) must pay the appellant supplier (NGC) $6.50 plus transmission charges of approximately $1.86 per gigajoule, for each gigajoule of gas supplied, or simply $6.50 per gigajoule on the basis that NGC is responsible for supplying the gas at its expense to BoPE’s premises within that price. The High Court held that it was the former, that is $6.50 plus transmission costs. The Court of Appeal allowed an appeal by BoPE, accepting its submission that it was the latter. NGC appeals to this Court seeking the reinstatement of the High Court’s determination.
 The case raises issues concerning the proper approach to interpretation questions and the admissibility of extrinsic evidence, particularly evidence concerning prior negotiations when a written contract is being interpreted. Before addressing the facts in more detail, I will examine the legal principles against which the interpretation question must be determined.
 The ultimate objective in a contract interpretation dispute is to establish the meaning the parties intended their words to bear. In order to be admissible, extrinsic evidence must be relevant to that question. The language used by the parties, appropriately interpreted, is the only source of their intended meaning. As a matter of policy, our law has always required interpretation issues to be addressed on an objective basis. The necessary inquiry therefore concerns what a reasonable and properly informed third party would consider the parties intended the words of their contract to mean. The court embodies that person. To be properly informed the court must be aware of the commercial or other context in which the contract was made and of all the facts and circumstances known to and likely to be operating on the parties’ minds. Evidence is not relevant if it does no more than tend to prove what individual parties subjectively intended or understood their words to mean, or what their negotiating stance was at any particular time.
 Although subjective evidence would be relevant if a subjective approach were taken to interpretation issues, the common law has consistently eschewed that approach. The common law focuses strongly on the agreement in its final form as representing the ultimate consensus of the parties. Hence it is regarded as irrelevant how the parties reached that consensus. To inquire into that process would not be consistent with an objective inquiry into the meaning of a document which is generally designed to be the sole record of the final agreement. A party cannot be heard to say – never mind what I signed, this is what I really meant.
 The objective approach is regarded as having two principal advantages. These are greater certainty and the saving of time and cost: greater certainty, because the subjective approach is apt to undermine the security of the written words by means of which the parties recorded their consensus; and saving time and cost, because a subjective approach is generally thought to require a fuller search for and examination of extrinsic evidence. A lesser, but still significant, perceived advantage is avoiding the effect a subjective approach might have on third parties who may have relied on what the words of the document appeared objectively to mean. But, despite its eschewing a subjective approach, the common law does not require the court, through the objective method, to ascribe to the parties an intention that a properly informed and reasonable person would not ascribe to them when aware of all the circumstances in which the contract was made.
 Nor does the objective approach require there to be an embargo on going outside the terms of the written instrument when the words in issue appear to have a plain and unambiguous meaning. This is because a meaning that may appear to the court to be plain and unambiguous, devoid of external context, may not ultimately, in context, be what a reasonable person aware of all the relevant circumstances would consider the parties intended their words to mean. An example of that situation is when plain words, read contextually, lead to a result which does not make sense, whether commercially or otherwise: a meaning that flouts business commonsense must yield to one that accords with business commonsense. The appropriate contextual meaning, if disputed, will, almost invariably, involve consideration of facts and circumstances not apparent solely from the written contract. While displacement of an apparently plain and unambiguous meaning may well be difficult as a matter of proof, an absolute rule precluding any attempt would not be consistent either with principle or with modern authority.
 The proposition that a party may not refer to extrinsic evidence “to create an ambiguity” is at least potentially misleading. It does not mean context is irrelevant unless there is a patent ambiguity. Context is always a necessary ingredient in ascertaining meaning. You cannot claim to have identified the intended meaning without reference to context. Hence it is always permissible to go outside the written words for the purpose of identifying the context in which the contract was made and its objective purpose. While there are no necessary preconditions which must be satisfied before going outside the words of the contract, the exercise is and remains one of interpretation. Subject to the private dictionary and estoppel exceptions to be mentioned below, it is fundamental that words can never be construed as having a meaning they cannot reasonably bear. This is an important control on the raising of implausible interpretation arguments. Furthermore, the plainer the words, the more improbable it is that the parties intended them to be understood in any sense other than what they plainly say. In some recent cases it has been suggested that contractual context should be referred to as a “cross-check”. In practical terms this is likely to be what happens in most cases. Anyone reading a contractual document will naturally form at least a provisional view of what its words mean, simply by reading them. That view is, in a sense, then checked against the contractual context. This description of the process is valid, provided the initial view is provisional only and the reader is prepared to accept that the provisional meaning may be altered once context has been brought to account. The concept of cross-check is helpful in affirming the point made earlier that a meaning which appears plain and unambiguous on its face is always susceptible to being altered by context, albeit that outcome will usually be difficult of achievement. Those attempting the exercise unsuccessfully may well have to pay for the additional costs caused by their attempt.
 A private dictionary meaning is a meaning the words linguistically cannot reasonably bear. It is, nevertheless, open to a party to show that, despite that fact, the parties intended their words to have that special meaning for the purposes of their contract. This represents a consensual parallel with cases in which words have a special meaning by trade custom. It can also be regarded as a linguistic example of estoppel by convention. The estoppel prevents the accepted special meaning from later being disavowed. Estoppels can also arise in interpretation cases not involving a special meaning. They are then normally based on a common assumption or representation as to meaning.
 If parties wish, they may contract on the basis that black means white; albeit the unlikelihood of their doing so, without expressly saying so, will no doubt be a powerful factor when it comes to questions of proof. Whether the parties have adopted a private dictionary meaning must be determined objectively in the same way as other disputes as to meaning are determined.
 Against that background I come to the subject of the admissibility of prior negotiations. Some of the difficulties in this area may derive from the concept of “prior negotiations” being employed in a more or less expansive way. Sometimes the concept seems to be used as if it encompassed all conduct and circumstances associated with negotiations towards the formation of a contract. It is necessary, however, to distinguish between the subjective content of negotiations; that is, how the parties were thinking, their individual intentions and the stance they were taking at different stages of the negotiating process on the one hand, and, on the other, evidence derived from the negotiations which shows objectively the meaning the parties intended their words to convey. Such evidence includes the circumstances in which the contract was entered into, and any objectively apparent consensus as to meaning operating between the parties.
 The vice in admitting subjective evidence of negotiations, is that doing so would be inconsistent with the objective basis on which interpretation issues are resolved. As already seen, evidence of a party’s subjective intention is not relevant to an objective resolution of interpretation issues. Although the common law takes the view that it is only the final written contract which records the ultimate consensus of the parties, the way that consensus is expressed may be based on an agreement as to meaning reached during negotiations.
 There is no problem with objective evidence directed to the context, factual or linguistic, in which the negotiations were taking place. That kind of evidence can properly inform an objective approach to meaning. Whereas evidence of the subjective content of negotiations is inadmissible on account of its irrelevance, evidence of facts, circumstances and conduct attending the negotiations is admissible if it is capable of shedding objective light on meaning. It is often said in contract interpretation cases that evidence of surrounding circumstances is admissible. Circumstances which surround the making of the contract can operate both before and after its formation. In either case irrelevance should be the touchstone for the exclusion of evidence. I do not consider there are any sufficiently persuasive pragmatic grounds on which to exclude evidence that is relevant. Indeed to do so would require reconciliation with s 7 of the Evidence Act 2006.
 In Gibbons Holdings Ltd v Wholesale Distributors Ltd I expressed the view that evidence of subsequent conduct should be admissible, if capable of providing objective guidance as to intended meaning. I suggested that, in order to be admissible, post-contract conduct should be shared or mutual. I saw that as a way of emphasising the need to exclude evidence which demonstrated only a party’s subjective intention or understanding as to meaning. I now consider that the approach I am taking in these present reasons is a simpler and clearer articulation of the appropriate principle, but one which still preserves the essential line between subjectivity and objectivity of approach.
 There is no logical reason why the same approach should not be taken to both post-contract and pre-contract evidence. The key point is that extrinsic evidence is admissible if it tends to establish a fact or circumstance capable of demonstrating objectively what meaning both or all parties intended their words to bear. Extrinsic evidence is also admissible if it tends to establish an estoppel or an agreement as to meaning. Such an agreement can demonstrate a special (private dictionary) meaning or an accepted meaning of words which would otherwise be ambiguous. I should expand a little on the latter proposition.
 If the parties have reached agreement on what meaning an otherwise ambiguous word or phrase should have for their purposes, that definitional agreement is itself an objectively determinable fact. When the issue is which of two possible meanings is objectively the more probable, the existence of a definitional agreement is obviously relevant, indeed it should be decisive. There is no logic in ascribing a meaning to the parties if it is objectively apparent they have agreed what that meaning should be.
 The foregoing analysis recognises that, generally speaking, issues of contractual interpretation arise in three circumstances: mistake; ambiguity; and special meaning. A mistake can represent either a drafting error or a linguistic error. Errors of this kind are primarily the subject of rectification. But a clear drafting or linguistic error, combined with equal clarity as to what was intended, can be remedied by way of interpretation, and in that respect context can and should be taken into account. An ambiguity arises when the language used is capable of more than one meaning, either on its face or in context, and the court must decide which of the possible meanings the parties intended their words to bear. A special meaning exists when the words used, even after the contractual context is brought to account, are linguistically still capable of only one meaning or are wholly obscure; but it is nevertheless evident from the objective context that the parties, by custom, usage or agreement, meant their words to bear a meaning which is linguistically impossible (for example, black means white), or represents a specialised and generally unfamiliar usage.
Although an estoppel will usually arise from the adoption of a special meaning, it is in cases where words are capable of bearing more than one meaning that estoppel is likely to have its primary application. A party may be estopped from denying that one of two possible meanings was the meaning the parties intended their words to bear. This, or an agreement as to meaning, is the best analysis of the controversial decision of Kerr J in the Karen Oltmann.
 That case concerned a time charter of a vessel for two years. The issue was whether the words in a break clause “after 12 months trading” meant “on the expiry of 12 months” or “at any time after the expiry of 12 months”. Evidence of negotiations (in the form of telexed exchanges) was admitted by Kerr J on the following basis:
If a contract contains words which, in their context, are fairly capable of bearing more than one meaning, and if it is alleged that the parties have in effect negotiated on an agreed basis that the words bore only one of the two possible meanings, then it is permissible for the Court to examine the extrinsic evidence relied upon to see whether the parties have in fact used the words in question in one sense only, so that they have in effect given their own dictionary meaning to the words as the result of their common intention.
 The Karen Oltmann is sometimes referred to as a special (private dictionary) meaning case because of Kerr J’s reference to the parties’ “own dictionary meaning”. But I agree with the House of Lords in Chartbrook that this is not its true basis. The case was one where the word “after” was, on its face, capable of two meanings. If the parties agreed or represented to each other in the telexes that the word “after” meant “on the expiry of” and the agreement or representation was relied on when they entered into the time charter, the parties were each estopped by that agreement or representation from contending that the word “after” bore the alternative meaning. Indeed, on the basis discussed earlier, they were bound by any such definitional agreement.
 Of course, the court must be satisfied that an agreement or representation as to meaning, reached or made during negotiations, was still operating at the time the contract was formed and represented a linguistic premise on which it had been formed. The Karen Oltmann was correctly decided; but on the basis of agreement or estoppel as to meaning, not on the basis of special meaning. There was nothing special about the meaning of the word “after”. It was, however, capable of two meanings. The parties had consensually resolved which meaning was to apply, or an estoppel had been created, and evidence to that effect was admissible.
 Against that legal background, I return to the facts. Wilson J has described the general circumstances in which the interpretation issue has arisen. I will not repeat that exercise beyond what is necessary to explain my views. BoPE disputed NGC’s termination of the principal agreement. It was necessary for the parties to resolve issues of supply pending the resolution of that dispute. Had BoPE sought an interim injunction requiring supply to be maintained in the meantime, it would have been obliged to give an undertaking to pay damages should its challenge to NGC’s termination fail. Those damages would have been calculated on the basis of the difference between the price payable under the principal contract and the market price for the period in question. Whereas the price under the principal contract was a dollar figure inclusive of supply costs, NGC’s pricing at the time the contract was terminated had become a dollar figure plus supply costs.
 It is inherently most unlikely that NGC would have been willing to enter into any interim arrangement on a basis which was less favourable to it than an undertaking as to damages would have been. This means that commercially the inherent probabilities were that the interim agreement would be on the basis of a dollar figure plus supply costs. This is because the interim agreement was designed to be a commercial substitute for the undertaking BoPE would have been obliged to give if it obtained an interim injunction requiring supply in the meantime. That is the context in which the exchange of correspondence between the parties which led to the agreement should be interpreted.
 Before I address that correspondence, I should make a point about the crucial expression “$6.50 per gigajoule”. Although it does not ultimately matter for interpretation purposes, I do not consider it can be said, as BoPE suggested, that the expression had a plain and unambiguous meaning. In itself the expression could have meant either $6.50 per gigajoule of gas supplied to BoPE’s premises or $6.50 for a gigajoule of gas with transmission/delivery costs payable in addition. When one views the expression in the context of the key letter of 5 October read as a whole, the ambiguity remains. The reference to the interim supply agreement being “based on the terms of” the principal agreement suggests that the sum of $6.50 included delivery costs. On the other hand, the earlier reference to $6.50, on the first page of the letter, as I shall explain, clearly suggests that the sum of $6.50 was exclusive of delivery costs. Hence, both in itself and in the context of the key letter, the expression $6.50 per gigajoule was ambiguous. It is only when reference is made to the wider commercial context and the previous correspondence that it becomes clear in what sense the parties should be taken to have been using the disputed expression.
 On 28 September 2004 NGC wrote to BoPE offering to settle the dispute overall and, if that were not possible, to supply gas in the interim at $6.50 per gigajoule. In its letter, under the separate heading “transportation and metering”, NGC said:
If the above offer is acceptable to you, we should meet [to] discuss what the transmission and metering arrangements will be. These are likely to take into account historic usage and transmission posted pricing.
 BoPE replied to NGC’s letter through its solicitors, Russell McVeagh. Their letter of 5 October 2004 contained the following paragraphs:
The issue therefore arises as to a supply of gas for BoPE pending determination of the litigation. In that regard, we record the parties’ positions as follows:
(i) Has, by your letter noted above, confirmed, on the record:
(aa) that it has sufficient gas available to supply to BoPE for at least the interim period (and, indeed, for the remainder of the term of the Agreement, and beyond); and
(bb) the sale price of the gas in question (ie $6.50 per GJ), thereby quantifying the loss that NGC perceives either party will suffer, depending on the outcome of the litigation.
(ii) Is probably protected from any claim BoPE brings against it for actual loss suffered by way of increased price, by the limitation of liability clause in the Agreement.
(b) BoPE, on the other hand:
(i) cannot currently source an equivalent alternative supply of gas for the remainder of the term of the Agreement; and
(ii) would in any event probably be barred from recovery of its loss on such an alternative supply by the limitation clause, such that damages for wrongful breach by NGC would not in this case be an adequate remedy.
In those circumstances, it would seem that the best course for the parties, in lieu of Todd having to apply for interim injunctive relief, would be for:
(a) NGC to undertake, without prejudice to its position, to simply continue to supply gas on the basis of the Agreement pending determination of BoPE’s proceeding as above.
(b) BoPE to undertake to:
(i) file that proceeding on or before 31 October 2004; and
(ii) in the event that BoPE is unsuccessful on its proceeding or withdraws it:
(aa) pay NGC on demand, for each GJ supplied, the difference between the Agreement price and $6.50 (or the current market price, whichever is the lower), plus interest, or
(bb) at BoPE’s election, provide NGC with an equivalent amount of gas as taken by BoPE from 1 November 2004, (if: (i) such gas is available from other gas fields; and (ii) Maui open access allows that gas to be transmitted to market.
 Paragraph 4(a)(i)(bb) refers to the sale price of the gas in question as being $6.50 per GJ (gigajoule). That, as a recital of NGC’s position, can only have meant $6.50 for the gas itself with transmission costs to be paid in addition. If it did not mean that, it would have misrepresented the position NGC took in its letter of 28 September 2004. The point is reinforced by the subsequent words “thereby quantifying the loss”. Unless $6.50 per GJ meant that amount for the gas plus transmission costs, it would not truly have quantified NGC’s loss. BoPE did not ultimately dispute that the reference to $6.50 per GJ in paragraph 4(a)(i)(bb) was a reference to $6.50 for the gas itself with transmission costs being payable in addition.
 In paragraph 5, which is on the second page of Russell McVeagh’s letter of 5 October 2004, there is a further reference in (b)(ii)(aa) to $6.50 for each GJ supplied. BoPE’s argument is that, whereas the reference on the first page of the letter in paragraph 4 to $6.50 meant $6.50 plus transmission costs, the same reference to $6.50 on the second page in paragraph 5 meant $6.50 inclusive of transmission costs. This was signalled, so the argument ran, by the words in paragraph 5(a) “on the basis of the Agreement”, meaning on the basis of the principal agreement where the dollar figure for gas supplied was expressed on a basis which included transmission costs.
 I am of the view that no reasonable reader of Russell McVeagh’s letter of 5 October, aware of the commercial context in which the correspondence was taking place, and reading that letter as a whole, would have made the astute distinction which BoPE’s argument involves. The first reference to $6.50 per GJ in that letter, reciting NGC’s position, is beyond doubt a reference to $6.50 plus transmission costs, because that is what NGC was proposing in its letter of 28 September, and Russell McVeagh’s letter of 5 October was, in paragraph 4, purporting to record that position. It is altogether too subtle, indeed some might say too clever, a point to suggest that the second reference to $6.50 for each GJ supplied should have been understood to mean $6.50 inclusive of transmission costs. The words “on the basis of the Agreement” cannot possibly, in context, be regarded as flagging such a fundamental change in meaning. In context they can only be understood to have meant otherwise on the basis of the principal agreement, that is, in all respects other than price. I consider a reasonable and properly informed reader would undoubtedly have understood BoPE to be agreeing to pay $6.50 plus transmission costs for the gas to be supplied in terms of any acceptance of the offer made in Russell McVeagh’s letter of 5 October. At this point there was objectively a consensus as to what the expression $6.50 per GJ meant.
 That construction is entirely consistent with and supported by the commercial context, the agreement being a substitute for an injunction and undertaking as to damages. The construction proposed by BoPE would be wholly inconsistent with the commercial context. Indeed it would involve a concession by NGC for which there was no commercial rationale. Like Wilson J, I regard the proposition that there was a valid reputational reason for NGC to make the concession inherent in BoPE’s suggested meaning as far fetched and unpersuasive. When agreement was ultimately reached in subsequent correspondence on the terms of the interim agreement with the supply price being expressed as $6.50 per GJ, that expression of the price was clearly intended to carry the same meaning as it did in the crucial letter of 5 October.
 The letters to which I have referred were part of a sequence containing negotiations leading up to the final agreement reached by means of a letter from Chapman Tripp, now representing NGC, to Russell McVeagh written on 15 October 2004. It is clearly apparent, on an objective basis, that the parties had reached an agreement during the negotiations as to what meaning the expression $6.50 per GJ should bear for the purposes of the contract they ultimately entered into. That agreement was still operating when the final agreement was reached. At the very least, BoPE, by means of Russell McVeagh’s letter of 5 October, read as a whole and in context, represented to NGC that $6.50 per GJ was to be understood as meaning $6.50 exclusive of transmission costs which were to be paid in addition. The parties proceeded to conclude their arrangement on that premise.
 If the case is analysed as one of estoppel, I do not consider this estoppel had to be pleaded. There can be no question of surprise. It was inherent in NGC’s case that the parties were either agreed on the meaning NGC asserted or that BoPE could not deny NGC’s asserted meaning. In any event, BoPE cannot be prejudiced by NGC now being allowed to raise an estoppel. The point is not one which could have been countered by the calling of further evidence. It arises from the terms of the correspondence which passed between the parties.
 The conclusion I have reached differs from that reached by the Court of Appeal. The key point of principle on which I respectfully differ from that Court concerns their decision to limit reference to the correspondence which led up to the agreement. The Court excluded reference to the crucial letter of 5 October. The Court of Appeal took as its starting point the letter of 8 October, which replied to that letter. The Court’s reason for this approach was that it saw the letter of 8 October as effectively resuming negotiations after Russell McVeagh, for BoPE, had rejected the proposal made by NGC in its letter of 28 September. That analysis overlooks the fact that NGC’s settlement proposal was two-pronged. First, it contained a proposal for an overall settlement which BoPE rejected. But it also contained a proposal for an interim settlement which was the subject of reply in Russell McVeagh’s letter of 5 October. The letter of 8 October written by Chapman Tripp, on behalf of NGC, was in response to the letter of 5 October. The effect of the Court of Appeal’s approach was to draw a line between what was in effect offer and acceptance. That, with respect, represents a most unusual way of approaching the ambit of correspondence leading up to the formation of a contract.
 I also consider that the Court of Appeal erred in the weight it placed on so-called “reputational matters” as a reason why NGC might be willing to accept, by way of agreement, an outcome substantially less favourable than what it was likely to have achieved pursuant to an undertaking as to damages. For this reason the Court of Appeal failed to take sufficient account of the commercial context in which the agreement was made.
 Before closing these reasons, I wish to associate myself with the remarks made by Wilson J at the end of his reasons. Those who practice as litigators in firms and who have been involved in correspondence between the parties, should not appear as counsel if the correspondence with which they are associated is in issue in the litigation. In this case Mr McIntosh is shown as the lead author of the key letter of 5 October which, on BoPE’s stance, could be viewed as setting a trap. That is the letter in which the expression “$6.50 per GJ” was said to have fundamentally changed its meaning from page 1 to page 2. The text of this letter was central to the litigation.
 For the reasons I have given, I consider that the Court of Appeal erred in reversing Harrison J’s interpretation of the agreement between the parties. I would allow the appeal with costs, and reinstate the determination of the High Court.
Note 17: Such as is taken in French law which treats what the parties intended as a question of subjective fact unaffected by any constructional rules. The French approach was heavily influential in the drafting of the Unidroit Principles of International Commercial Contracts (1994 and 2004 revision) which allow reference to prior negotiations (art 4.3). But that is not the common law way.}}