This page uses so called "cookies" to improve its service (i.e. "tracking"). Learn more and opt out of tracking
I agree

HSBC Trustee (Singapore) Ltd v Lucky Realty Co Pte Ltd, [2015] SGHC 93

Title
HSBC Trustee (Singapore) Ltd v Lucky Realty Co Pte Ltd, [2015] SGHC 93
Table of Contents
Content

HSBC Trustee (Singapore) Ltd v Lucky Realty Co Pte Ltd [2015]
SGHC 93

Suit No : Originating Summons No 391 of 2014

Decision Date : 13 April 2015

Court : High Court

Coram : Vinodh Coomaraswamy J

Counsel : Edwin Tong, Lee Bik Wei and Lee May Ling (Allen & Gledhill LLP) for the plaintiff; Julian Tay and Mark Cham (Lee & Lee) for the defendant.

Contract - Contractual terms - Rules of construction - Contextual approach to contractual interpretation

Equity - Estoppel by convention

Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251 (refd)

Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101 (refd)

Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 (refd)

Philip Collins Ltd v Rahmlee Davis and Louis Satterfield [2001] ECDR 17 (refd)

Sembcorp Marine Ltd v PPL Holdings Pte Ltd [2013] 4 SLR 193 (refd)

Zurich Insurance (Singapore) Pte Ltd v B-Gold Interior Design & Construction Pte Ltd [2008] 3 SLR 1029 (refd)

Evidence Act (Cap 97, 1997 Rev Ed) s 2(1), s 2(2), s 5, s 62, s 94, s 94(f), s 95, s 96, s 97, s 98, s 99, s 100

Rules of Court (Cap 322, R 5, 2014 Rev Ed) O 14, O 18 r 19, O 27, O 28 r 2, O 38 r 2(2), O 38 r 2(5), O 38 r 3, O 41 r 5(1)

13 April 2015

Vinodh Coomaraswamy J:

1 In 1975, a lessor let a parcel of land on a 60-year lease. In exchange, the lessee agreed to pay a fixed yearly rent. In 1976 and 1977, the lessee erected upon the land a strata development comprising four buildings. The lessee then retained all of the strata units in one of the buildings for itself and sold virtually all of the strata units in the other three buildings to third parties. In 1995, to compromise a dispute, the lessor and lessee varied the lease to permit the lessor to increase the rent every five years to the prevailing market rent or by 10%, whichever was higher. The two first rent increases were uncontentious. The parties have, however, been deadlocked since the time for the third rent increase in 2009. The lessor's position is that the lessee is obliged to pay rent under the lease for the entire parcel of land demised to it under the lease. The lessee's position is that it is obliged to pay rent only for the one building that it retains. In these proceedings, the lessor seeks as its primary relief a declaration that its position is correct and that the lessee's position is wrong.

The facts

The parties

2 The lessor and plaintiff is HSBC Trustee (Singapore) Limited, a trust company and a member of the HSBC group.1 The lessee and defendant is Lucky Realty Company Pte Ltd, a property developer and a member of the Far East Organisation.2 The parcel of land demised under the lease is in Bedok. At the time of the lease, it was registered as Lot 3041 of Mukim 27. The freehold of Lot 3041 was owned by one Mr Koh Sek Lim until he passed away in 1948.3Clause 1 of Mr Koh's will settled Lot 3041 on certain trusts.

3 In 1975, the trustee of Lot 3041 was the Public Trustee. In 1991, OCBC Trustee Limited took over as the trustee.4 In 2006, HSBC Trustee (Singapore) Limited took over as the trustee. The change of trustees over the years is of no legal relevance to the dispute before me. I shall therefore refer to the trustee of the demised land at any given time simply as "the Trustee".

The lease

4 On 25 February 1975, the Trustee demised Lot 3041 to Lucky Realty for a term of 60 years commencing 1 March 1977 and ending on 28 February 2037.5 The lease obliged Lucky Realty to pay a yearly rent of $3,877.15 for the demise. This rent is in fact a ground rent in that it was to be paid solely for the use of the land. The rent was fixed at that annual figure for the entire 60-year term of the lease: it contained no provision for the rent to be increased nor, for that matter, for it to be decreased.

5 The operative part of the lease provides as follows:6

I, THE PUBLIC TRUSTEE ... the proprietor of the land hereinafter described HEREBY LEASE unto LUCKY REALTY ... all such my estate or interest in the said land ...

DESCRIPTION OF LAND

Reference to Land-Register  
Volume Folio   Mukim Lot Description of Land (whether whole or part)
102 138   XXVII 3041 The whole of Lot 3041 of Mukim XXVII together with the buildings in the course of construction thereon.

TO BE HELD by the Lessee as tenant for the term of sixty (60) years commencing from the 1st day of March One thousand nine hundred and seventy-seven (1977) YIELDING AND PAYING therefor on the 19th day of March 1975 for fifty-nine (59) years and on the same date [on every] succeeding year during the said term at the office of The Public Trustee the yearly rent of $3,877-15 (Dollars Three thousand eight hundred and seventy-seven and cents fifteen only) SUBJECT to the covenants conditions and powers implied by law in instruments of lease (or to such of them as are not hereinafter expressly negatived or modified) AND SUBJECT also to the covenants and conditions hereinafter contained.

The land is developed and subdivided

6 Clause 1(iii) of the lease obliged Lucky Realty to develop Lot 3041 by erecting buildings upon it. In 1976 and 1977,7 Lucky Realty duly erected four two-storey buildings on the land. The buildings are known as Blocks A to D. The position of these four blocks in Lot 3041 can be seen in the diagram set out at [7].

Compulsory acquisition

7 Lot 3041 was eventually subdivided in two ways. First, Lot 3041 was subdivided into one large lot and two smaller lots.8 A diagram showing this subdivision is set out below. (Diagram not available, editor's note)

8 The entire shaded area is Lot 3041 as it stood at the time of the lease. Lot 5247L, halfway up the left side of the shaded area, was then the location of a septic tank and sewage works. The State compulsorily acquired this lot for a sewage treatment plant. Lot 5246X at the bottom of the shaded area was a road. The State compulsorily acquired this lot also. The remainder of Lot 3041 was known after the subdivision as Lot 5245N.

9 The subdivision and compulsory acquisition reduced the area of the demised land by 2,042.8 square metres, or about 11%, from 18,947 square metres to 16,904.2 square metres. But the rent under the lease, being a fixed rent, did not change. Lucky Realty continued to be obliged to pay the Trustee $3,877.159 yearly. It duly did so.

Strata development

10 The second subdivision was carried out to enable Lucky Realty to realise a capital sum from its development of the land. It divided Lot 5245N into a strata subdivision comprising a total of 119 strata units: (i) 33 strata shops on the first storey of Blocks A, B and C; (ii) 33 strata flats on the second storey of those three blocks; (iii) a single strata market on the first storey of Block D; and (iv) 52 strata shops on the second storey of that block.10

11 Lucky Realty eventually sold 64 out of the 66 strata units in Blocks A, B and C to third party purchasers. It effected the sale by assigning to the purchasers the benefit of its remaining term under its 60-year lease over those units.

12 Lucky Realty did not sell the 53 units in Block D or the two remaining units in the other blocks.11 Instead, it retained these 55 strata units for itself, preferring to generate a recurring income from them rather than a one-off capital sum.

The 1995 variation

13 In February 1994, a dispute arose between the Trustee and Lucky Realty.12 The Trustee discovered that Lucky Realty was redeveloping Block D from a two-storey market with shop units above into a two-storey shopping centre.13 The Trustee considered this to be a breach of cl 1(vi) of the lease. That clause prohibits Lucky Realty from making any alterations or additions to any of the buildings on the demised land without the prior consent in writing of the Trustee. Solicitors appointed by the Trustee wrote a letter before action to Lucky Realty.14 Lucky Realty took the position that it was not in breach of the lease; alternatively, that any breach was, at the very most, a technical breach.15

14 The Trustee and Lucky Realty then entered into negotiations to resolve their dispute. The negotiations were protracted but ultimately resulted in a binding compromise. By an exchange of letters ending on 23 September 1995,16 the Trustee and Lucky Realty agreed to vary the obligation to pay rent under the lease. This variation is today evidenced by a Deed of Variation signed, sealed and delivered by lessor and lessee on 17 December 1996.17

15 The 1995 variation effected only one textual change to the original lease. It deleted the existing habendum and replaced it with a new and more detailed one comprising three limbs. The last of these three limbs expressly provides that, save as varied by the 1995 variation, all other express and implied provisions of the original lease remain in full force and effect.

16 The result of the new habendum was two-fold. First, Lucky Realty accepted an immediate increase in the rent from $3,877.15 per annum to $120,000 per annum backdated to take effect from 15 June 1994. Second, Lucky Realty granted the lessor the right,18 on 15 June 1999 and on that date every five years thereafter, to increase the yearly rent "to the market rent prevailing at the time" or by 10% of the existing rent, whichever was higher. It is the scope of this right of the Trustee to increase the rent to the "market rent" which lies at the heart of this dispute.

The Trustee reviews and increases the rent

17 The first two rent increases - for the five year periods from 1999 to 2004 and from 2004 to 2009 - were uncontentious. With effect from 1999, the Trustee increased the rent by exactly 10%, from $120,000 per annum to $132,000 per annum. The Trustee relied for this increase on the minimum 10% increase specified in the rent escalation clause.19 With effect from 2004, the Trustee increased the rent by 13.6%, to $150,000 per annum. The Trustee based this increase "on the market rental value of the said property in 2004".20 Lucky Realty has duly paid all rent as claimed by the Trustee right up to 2009.21

18 The third rent increase was to take place in 2009. The Trustee, by now having changed to the plaintiff, took the view that the lease entitled the Trustee to charge rent - and obliged Lucky Realty to pay rent - for the entirety of Lot 5245N. On this basis, about a month before the rent increase in 2009 was due to take effect, the Trustee gave notice that it intended to increase the rent from $150,000 per annum to $1.3m per annum.22 The Trustee supported this figure with a valuation23 showing that $1.3m was indeed the prevailing market rent for the whole of Lot 5245N.24 Astonished,25 Lucky Realty contended in response that it was obliged to pay rent only in respect of Block D. It presented a valuation showing that the prevailing market rent for Block D alone was only $168,000 per annum.26

19 Each party has doggedly maintained its position since 2009. Lucky Realty has in the meantime continued paying $150,000 per annum to the Trustee, being the rent fixed at the last uncontentious rent increase.

The Trustee commences these proceedings

20 On 28 April 2014, the Trustee commenced these proceedings against Lucky Realty. In summary, the Trustee seeks:

(a)

A declaration that the five-yearly increase of the yearly rent is to be calculated by reference to the whole of Lot 5245N and not by reference only to Block D.

(b)

An order that every future five-yearly revision of the yearly rent be carried out on the same basis.

(c)

An order that Lucky Realty pay a yearly rent recalculated on this basis for the five-year periods commencing in 1999 and 2004.


21 At the commencement of the hearing before me, counsel for the Trustee withdrew its claim for a retrospective increase in the yearly rent summarised at [20(c)] above. He concedes - rightly in my view - that any such claim is time-barred. The only claim before me, therefore, relates to the rent that Lucky Realty is obliged to pay from 2009 onwards.

Construction of the rent escalation clause

22 Each side relies for its case only on the true construction of the express terms of the lease. More specifically, Lucky Realty does not rely on any of the following to oppose the Trustee's submissions on the true construction of the lease: an argument that the lease does not set out the parties' complete agreement, an implied term in the lease (whether of law or of fact), a collateral contract, an actionable misrepresentation or an operative mistake. Equally, Lucky Realty does not seek the equitable remedy of rectification. This is an important point. If there were a basis to do so, it was at all material times open to Lucky Realty to raise any or all of these grounds by way of counterclaim against the Trustee pursuant to Order 28 Rule 2 ofthe Rules of Court (Cap 322, R 5, 2014 Rev Ed).

23 Accordingly, the single question before me is whether, on a true construction of the lease, Lucky Realty's obligation to pay the "yearly rent" and the Trustee's right to achieve the prevailing "market rent" extends to the whole of Lot 5245N or whether it is confined to Block D only.

24 The Trustee submits that the ordinary meaning of the plain words of the lease is that the "yearly rent" and "market rent" relates to the whole of Lot 5245N. Lucky Realty submits in response that the rent escalation clause must be construed in the light of the facts and circumstances surrounding the 1995 variation, including the parties' pre-contractual negotiations and their subsequent conduct. Its case is that recourse to this extrinsic evidence shows that the terms "yearly rent" and "market rent" as used in the lease refer to the rent for Block D only.27

The contextual approach

25 Our law on the proper approach to construing a contract is set out in two seminal decisions of the Court of Appeal: Zurich Insurance (Singapore) Pte Ltd v B-Gold Interior Design & Construction Pte Ltd [2008] 3 SLR 1029 ("Zurich") and Sembcorp Marine Ltd v PPL Holdings Pte Ltd [2013] 4 SLR 193 ("Sembcorp"). These cases firmly entrench the contextual approach to construing contracts in our law and clearly explain how that approach is to be applied. Given the breadth of the survey and the depth of the analysis in each of these two cases, I need look only at them in order to derive the principles which I must apply to construe the lease contextually.

Zurich

Zurich: the facts

26 Zurich involved a dispute over the terms of a policy of insurance between an insured and its insurer.28 The insured commenced action in the District Court claiming that the insurer was obliged under the policy to indemnify it against certain loss and damage.29 The District Court30 found in favour of the insurer and against the insured, holding that the insured's claim was excluded by the natural and ordinary meaning of a special condition of the policy.31

27 The High Court allowed the insured's appeal.32 It agreed with the District Court that the insured's claim was excluded by the natural and ordinary meaning of the special condition.33 But having received extrinsic evidence of the genesis of the insurance policy, the High Court held that that special condition was inoperable. The extrinsic evidence showed that the "insured had relied upon the insurer to provide cover for a specific purpose which was made known to the [insurer] prior to the issue of the policy".34 The High Court therefore held that it would be "contrary to all sense of justice and fair play"35 to give effect to the natural and ordinary meaning of this special condition in the light of its findings on the extrinsic evidence.

Four important points extracted from Zurich

28 The Court of Appeal reversed the High Court and restored the District Court's decision.36 In doing so, the Court of Appeal began by setting out the five principles of the contextual approach from Lord Hoffmann's judgment in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 at 912-913 ("Investors Compensation Scheme"). These principles are quoted in Zurich at [56] and need not be repeated here. The Court of Appeal then refined the contextual approach as it is to be applied in Singapore.

29 From the Court of Appeal's judgment in Zurich, I extract four important points which govern my approach to answering the deceptively simple question before me.

30 First, the goal of construing a contract is to determine and give effect to the intention of the parties, objectively ascertained. Thus:

It is true [that] the objective of the construction [of contracts] is to give effect to the intention of the parties. But our law of construction is based on an objective theory. The methodology is not to probe the real intentions of the parties but to ascertain the contextual meaning of the relevant contractual language. Intention is determined by reference to expressed rather than actual intention.37


My focus in construing the lease, therefore, must remain on ascertaining the objective intention of the parties even "in the face of conflicting subjective interpretations advanced by ingenious counsel."38 That is so even on the modern and more liberal contextual approach to construction.39 The subjective intentions of the parties, even if they could be determined with certainty, are ordinarily immaterial to construing their contract. That is obviously the case if one party leaves its subjective intentions wholly unexpressed at the time of contracting. That is also the case even if one party expresses its subjective intention to the other party at the time of contracting, but fails to reflect that subjective intention within the contract in an objectively ascertainable manner. That is even the case even if both parties share that subjective intention. This does not leave entirely without remedy a party to a contract who finds a unilateral or even a shared subjective intention defeated by the objectively ascertainable meaning of the contract. It merely means that it is not the function of the contextual approach to contractual interpretation to supply that remedy.

31 The second important point which I extract from Zurich is that achieving a result which is just and fair in all the circumstances of the case is not the purpose of construing a contract. The purpose of construing a contract is to give the contract a meaning which represents fairly the parties' objectively ascertained intention. That outcome is axiomatically the just and fair result of construing a contract. Achieving justice and fairness which goes beyond that outcome is the objective - to varying degrees and with varying flexibility - of other contractual doctrines which come into play, both at law and in equity, after ascertaining the parties' objective intention. The difficulty in Zurich was that the High Court, asked to do no more than to construe the insurance policy, relied on extrinsic evidence to nullify the parties' objectively ascertained intention in order to achieve what it considered to be a just and fair result. As the Court of Appeal said, "amorphous notions based on justice and fairness did not warrant the Judge straying beyond the confines of the Policy" in that way.40

32 The third important point I extract from Zurich is that an exercise in construction must ascribe to the words and phrases chosen by the parties in their contract some meaning, and that that meaning must be a legitimate meaning. It is therefore wrong to give those words a meaning which is beyond the contours of their penumbral meaning. It is equally wrong to give those words no meaning at all. Thus, the Court of Appeal held in Zurich that treating an exclusion clause as entirely inoperable went beyond the limits of what a court can legitimately do when construing a contract:41

... By holding an entire exclusion clause ... to be inoperable, the Judge - with respect - strayed far into the realm of varying a contract in contravention of s 94 [of the Evidence Act]. This was not a case where the Court read down broad words within the scope of their penumbral meaning ... nor could one argue that this was merely a case of aggressive or intrusive interpretation. Instead, it was a case where the Court read an entire provision out of the contract being construed. There is a conceptual difference between attributing a meaning to words or phrases that might strain the contours of their penumbral meaning and simply ignoring a provision altogether. ...
[emphasis in original]


33 The fourth and final important point I extract from Zurich is that the contextual approach not only permits but obliges the court to have regard to evidence extrinsic to the words used by the parties within the four corners of a written contract in order properly to give effect to the intention of the parties objectively ascertained. But there are limits. The extrinsic evidence must  in every case satisfy the criteria set out in Zurich.

34 The criteria established by Zurich which are relevant to the present case are the following (internal cross-references omitted):

132

To summarise, the approach adopted in Singapore to the admissibility of extrinsic evidence to affect written contracts is a pragmatic and principled one. The main features of this approach are as follows:

(a)

A court should take into account the essence and attributes of the document being examined. The court's treatment of extrinsic evidence at various stages of the analytical process may differ depending on the nature of the document. In general, the court ought to be more reluctant to allow extrinsic evidence to affect standard form contracts and commercial documents ...

(b)

...

(c)

Extrinsic evidence is admissible under proviso (f) to s 94 to aid in the interpretation of the written words. Our courts now adopt, via this proviso, the modern contextual approach to interpretation, in line with the developments in England in this area of the law to date. Crucially, ambiguity is not a prerequisite for the admissibility of extrinsic evidence under proviso (f) to s 94 ...

(d)

The extrinsic evidence in question is admissible so long as it is relevant, reasonably available to all the contracting parties and relates to a clear or obvious context ... However, the principle of objectively ascertaining contractual intention(s) remains paramount. Thus, the extrinsic evidence must always go towards proof of what the parties, from an objective viewpoint, ultimately agreed upon. Further, where extrinsic evidence in the form of prior negotiations and subsequent conduct is concerned, ... there should be no absolute or rigid prohibition against evidence of previous negotiations or subsequent conduct, although, in the normal case, such evidence is likely to be inadmissible for non-compliance with the requirements set out ... above. (We should add that the relevance of subsequent conduct remains a controversial and evolving topic that will require more extensive scrutiny by this court at a more appropriate juncture.) Declarations of subjective intent remain inadmissible except for the purpose of giving meaning to terms which have been determined to be latently ambiguous ...

(e)

In some cases, the extrinsic evidence in question leads to possible alternative interpretations of the written words (ie, the court determines that latent ambiguity exists). A court may give effect to these alternative interpretations, always bearing in mind s 94 of the Evidence Act. In arriving at the ultimate interpretation of the words to be construed, the court may take into account subjective declarations of intent ... Furthermore, the normal canons of interpretation apply in conjunction with the relevant provisions of the Evidence Act, ie, ss 95-100 ...

(f)

A court should always be careful to ensure that extrinsic evidence is used to explain and illuminate the written words, and not to contradict or vary them. Where the court concludes that the parties have used the wrong words, rectification may be a more appropriate remedy ...


The Zurich criteria applied to Zurich's facts

35 The Court of Appeal in Zurich, having reversed the decision of the High Court, had to go on to ascertain objectively the parties' intention. To do so, the Court of Appeal started with the natural and ordinary meaning of the words of the contract.42 It then considered whether the extrinsic evidence proffered as an aid to construction could be taken into account. The particular type of extrinsic evidence proffered by the insured in Zurich was evidence of the parties' pre-contractual negotiations. The Court of Appeal held that our law of contract, unlike English law, has no absolute rule prohibiting the use of this type of extrinsic evidence as an aid to construction. Like all other extrinsic evidence, this type of extrinsic evidence may be used as an aid if, but only if, it satisfies the three criteria set out at [132(d)] of Zurich itself.

36 On the facts of Zurich, however, the Court of Appeal held that the extrinsic evidence did not satisfy these criteria43 and therefore had to be rejected as an aid to construction. The Court of Appeal's construction exercise could therefore go no further. Having started with the plain and ordinary meaning of the parties' words, it ended with that meaning.

Sembcorp

37 Sembcorp, decided five years after Zurich, strongly endorses Zurich (at [46]) while revisiting and supplementing three points from Zurich which are of significance to the present case.

38 First, Sembcorp draws a distinction between "interpretation" and "construction" of a contract. "Interpretation" is "the ascertainment of the meaning which the expressions in a document would convey to a reasonable person having all the background knowledge which would reasonably have been available to the parties at the time of the contract" (at [33]) [emphasis in original]. "Construction", on the other hand, "is the composite process that seeks to ascertain the parties' intentions, both actual and presumed, arising from the contract as a whole without necessarily being confined to the specific words used" (at [31]) [emphasis in original]. I have endeavoured to observe this distinction when using those terms in this judgment.

39 Second, Sembcorp emphasises the importance of drawing the distinction in every case between the substantive law of contract and the adjectival law of evidence. The contextual approach to construction and the supporting principles and guidelines are part of our substantive law of contract (Sembcorp at [43]). Their objective is to determine the rights and liabilities of contractual counterparties. The objective of the law of evidence, on the other hand, is to determine whether the existence or non-existence of a particular fact may be proved (s 5 of the Evidence Act (Cap 97, 1997 Rev Ed)); and, if so, to regulate the means by which it may be proved (Sembcorp at [40], [43]).

40 The third point of importance to emerge from Sembcorp builds on this insight to analyse more closely the inevitable interplay when construing a contract between the law of contract and the law of evidence. This analysis leads the Court of Appeal to articulate two factors, one legal and one pragmatic, which are implicit in its earlier analysis in Zurich and which militate against taking too robust an approach to the use of extrinsic evidence as an aid to construction. The legal constraint arises from our statutory law of evidence. The pragmatic constraint is necessitated by the nature of our system of adversarial litigation. Taken together, these constraints justify not only the Zurich criteria but also justify applying those criteria with particular caution when a litigant proffers certain types of extrinsic evidence as an aid to construction.

41 The legal constraint on taking too robust an approach to using extrinsic evidence as an aid to construction arises from the fact that our law of evidence is primarily statute-based whereas our law of contract is primarily common law. The nature of statutory rules as compared to common law rules means that our law of evidence is less flexible than our law of contract. The result is that the principles of both bodies of law cannot develop or evolve in parallel through judicial decision at the same time and in the same way. The position in English law is, of course, different. In England, the law of contract and the law of evidence are both primarily common law and can therefore evolve together, virtually without any need to draw a distinction between the two, in the same judicial decisions.

42 Our statutory law of evidence therefore has a controlling influence on the development of our common law of contract when it comes to the contextual approach to construction. It would be pointless for our law of contract to permit - or for us even to consider allowing it to evolve so as to permit - particular types of extrinsic evidence to be used as an aid to construing a contract if our law of evidence lays down statutory rules which operate to exclude that evidence, either absolutely or ordinarily. To put it another way, the universe of extrinsic evidence which our common law of contract accepts, or can evolve to accept, as legitimate aids to construction can never be anything other than a subset of the evidence rendered admissible by our statutory law of evidence.

43 The pragmatic constraint which militates against taking too robust an approach to relying on extrinsic evidence as an aid to construction arises from the realities of our system of adversarial litigation. That system gives each party wide latitude to choose, marshal and present to the court all the evidence which it wishes to rely on in order to establish its case, so long as the evidence is at least arguably relevant, is arguably admissible and is an arguably legitimate aid to construing a contract. That latitude is compounded by each party's entitlement to compel production of documents from the opposing party through discovery, a process which is not necessarily limited by relevance or admissibility. Each adversary therefore has the freedom, the power and the incentive to inundate its opponents with requests to compel production of extrinsic evidence and to inundate the court by proffering extrinsic evidence as an aid to that party's favoured construction. The court must then sift through and resolve all of the arguments about all of the proffered evidence to see whether it can be received as an aid to construction, both under the law of evidence and under the law of contract. Even if that painstaking process ultimately determines that some of that extrinsic evidence is indeed a legitimate aid to construction, the evidence may yet offer only marginal assistance in ascertaining objectively the parties' intentions, assistance which is out of all proportion to the effort taken in that process. Too robust an approach to relying on extrinsic evidence has systemic consequences. It decreases the certainty with which legal advisers are able to advise parties in order to avoid litigation and engenders cost and delay when litigation proves unavoidable (Sembcorp at [66] and [71]).

Four types of extrinsic evidence

44 For present purposes, it is useful to consider how the two constraints identified in Sembcorp operate on particular types of extrinsic evidence by dividing that evidence into four broad types:

(a)

Evidence of the circumstances surrounding the contract;

(b)

Evidence of the subjective intentions of the parties;

(c)

Evidence of the parties' pre-contractual negotiations; and

(d)

Evidence of the parties' subsequent conduct, ie their conduct after entering into the contract.



Circumstances surrounding the contract

45 The first type of extrinsic evidence is evidence of the circumstances surrounding the contract. The legal and practical constraints to receiving extrinsic evidence of this type as an aid to construction are significantly lower than for the other three types. Extrinsic evidence of this type is proffered to place the court in the same position as the parties when they entered into the contract. It enables the court, in order to construe objectively the words used by the parties in their contract, to have regard to the "facts and circumstances which were (or ought to have been) in the mind of the [drafter] when he used those words" (Sembcorp at [64]). Zurich establishes that these surrounding circumstances are not limited to matters of fact but can also include the legal and regulatory background (at [131]) and the object or purpose for which the parties entered into the contract (at [53]). Zurich also establishes that there is no need to find any sort of ambiguity in the parties' contract before turning to evidence of this nature as an aid to construction.

46 There is no tension between the law of contract and the law of evidence with regard to this type of extrinsic evidence. The law of evidence, by virtue of s 94(f) of the Evidence Act, admits evidence of this type for this purpose without restriction (Zurich at [121]; Sembcorp at [55] and [63]). The law of contract then accepts all of the evidence of this type which is admitted by the Evidence Act also to be a legitimate aid to construction. Indeed, acknowledging the importance of this type of extrinsic evidence lies at the very heart of the contextual approach. It may even be argued that the effect of the contextual approach is to take evidence of this type entirely outside any exclusionary rule, whether as a matter of evidence or as a matter of contract. Thus, s 94(f) of the Evidence Act takes this evidence outside the exclusionary evidential rule set out in s 94 of the same Act. Further, the second of Lord Hoffmann's principles in Investors Compensation Scheme, as modified in Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251 ("BCCI v Ali") at [39] permits evidence of this type to be used as an aid to construction as a primary rule of construction rather than as an exception to an exclusionary rule.

47 The significance of the pragmatic constraint when using evidence of this type as an aid to construction is much reduced. As Lord Hoffmann said in Chartbrook Ltd v Persimmon Homes Ltd [2009] 1 AC 1101 (cited in Sembcorp at [67]), "surrounding circumstances are, by definition, objective facts which will usually be uncontroversial."

The parties' subjective intentions

48 The second type of extrinsic evidence is evidence of one or both parties' subjective intentions. As in the case of evidence of the first type, but for the opposite reason, there is no tension between the law of contract and the law of evidence with regard to this type of extrinsic evidence. The law of contract permits evidence in this type to be used as an aid to construction, but only if there is an ambiguity in the contract, and even then, only if the ambiguity is latent (Zurich at [51] and [132(d)]; Sembcorp at [59]). This type of extrinsic evidence therefore cannot be used as an aid to construction if there is no ambiguity at all or if there is an ambiguity which is patent rather than latent. There is no tension with the law of evidence because the Evidence Act mirrors these limitations on the admissibility of this type of extrinsic evidence in s 94 to s 100.

49 As in the case of the evidence of the first type, the significance of the pragmatic constraint when using evidence of this type as an aid to construction is much reduced, once again for the opposite reason. Given that evidence of this type is ordinarily inadmissible under the law of evidence, the scope for a litigation adversary and the court to be inundated with this type of evidence is regulated primarily by the rules of evidence. They operate as a prior constraint to exclude the evidence even before the law of contract can operate upon it. Pre-contractual negotiations and subsequent conduct

50 It is convenient to take the third and fourth types of extrinsic evidence together. Extrinsic evidence of these two types are by their nature highly susceptible to being shaped towards self-serving ends with the benefit of hindsight, highly likely to be highly contentious and very often of only marginal assistance in construing a contract. The time, the cost and the effort of resolving the subsidiary legal and factual disputes thrown up in any attempt to admit evidence of these two types are therefore often disproportionate to the assistance that this evidence might supply to ascertaining objectively the parties' intentions at the time of contracting. English law therefore has a blanket rule excluding extrinsic evidence of both types. New Zealand, on the other hand, takes what Sembcorp refers to (at [36]) as the "robust approach". It receives evidence of both of these types freely as an aid to construction. We chart a middle course. We reject a blanket exclusionary rule which operates despite relevance while also rejecting a blanket admissibility rule which operates subject only to relevance. We therefore require extrinsic evidence in these two classes, like all extrinsic evidence, to satisfy the three Zurich criteria.

51 Sembcorp's reiteration (at [72]) of Zurich's warning (at [127] and [129]) against taking too liberal an approach to the use of extrinsic evidence is made especially forcefully with respect to extrinsic evidence of the third and fourth types (at [38]). Indeed, it could be said that the practical significance of Sembcorp, like Zurich, is to permit evidence of these two types to be used as an aid to construction while warning of the serious consequences of being too liberal in doing so.


Evidence Act does not apply to these proceedings

52 I am undoubtedly bound by the contextual approach set out in Zurich and Sembcorp in construing the parties' contract. That is because the body of principles which comprises the contextual approach is part of the law of contract. These principles operate whenever and wherever a court must ascertain the objective meaning of a contract in order to make a final determination of the parties' rights and liabilities in accordance with law. They operate whether that final determination is made in an action commenced by writ or upon an application commenced by originating summons. They operate whether that final determination is made summarily upon affidavit evidence alone or after the court receives evidence viva voce.

53 The provisions of the Evidence Act regulating the admissibility of extrinsic evidence do not, however, apply to the proceedings before me. Those provisions are found in Part II of the Evidence Act. Section 2(1) of the Act expressly provides that Part II of the Act "shall apply to all judicial proceedings in or before any court, but not to affidavits presented to any court or officer nor to proceedings before an arbitrator". Part II of the Act applied to the proceedings at first instance in Zurich and Sembcorp because they were judicial proceedings commenced by writ and resolved at trial. When the parties' rights and liabilities are finally determined at trial, the underlying findings of fact are made on viva voce evidence received in accordance with the Evidence Act. What is before me are judicial proceedings commenced by originating summons. In proceedings of this type, designed for litigation in which it is unlikely that there will be any substantial dispute of fact, the court ordinarily determines the parties' rights and liabilities summarily, upon affidavit evidence alone (see Order 38, Rule 2(2) of the Rules of Court). The effect of s 2(1) of the Evidence Act is therefore to disapply the entirety of Part II of the Evidence Act to all of the evidence before me.

54 The relationship between the law of evidence and the law of contract in the contextual approach to construing a contract is therefore of especial significance in the proceedings before me. The question arises as to whether I am bound to apply the law of evidence at all in determining the Trustee's application. The answer must be yes. It cannot be that the law, taken as a whole, permits me to reach a final determination of the parties' rights and liabilities in these proceedings free of all rules of evidence when I would have been bound by the Evidence Act if the parties had chosen instead to litigate their dispute at a trial in proceedings commenced by writ.

55 When then is the source of the law of evidence which I apply, if it is not the Evidence Act? I am bound by our common law of evidence. The common law continues to be one of the sources of our law of evidence Sembcorp at [38]. This is so even though s 2(2) of the Evidence Act repeals common law rules of evidence. That is because it does not repeal all common law rules of evidence. It repeals only those common law rules of evidence which are inconsistent with the rules of evidence set out in the Evidence Act (see Robert Margolis, "The Concept of Relevance: In The Evidence Act And The Modern View" (1990) 11 Sing LR 24 at 26).

56 Common law rules of evidence which are not inconsistent with the Evidence Act continue to form part of our law of evidence, although their operation is now masked in proceedings to which the Evidence Act applies. To put it another way, to the extent that the rules of evidence at common law at the time the Evidence Act was enacted in Singapore in 1893 found their way into the Act, those common law rules - at the very least - were not repealed when the Act came into force. Those rules remain part of our common law of evidence. They continue to apply to judicial proceedings which fall outside the Evidence Act. Sembcorp's historical analysis of s 94 to s 100 of the Evidence Act (Sembcorp at [39] - [65]) shows that the principles captured in those rules are indeed consistent with the common law of evidence as it stood at the time our Evidence Act was enacted. Those rules therefore continue to form part of our common law of evidence. I am bound, therefore, to apply those common law rules to determine whether the existence or non-existence of a particular fact may be proved in the proceedings before me and, if so, how it may be proved.

57 This is an important point. There is only one body of principles applicable to construing contracts in our substantive law of contract. No doubt that body of principles is independent of the law of evidence. But there should also be only one body of principles which determines what evidence that body of contractual principles can operate upon. If we had no common law of evidence, or if our common law of evidence were not consistent with sections 94 to 100 of the Evidence Act, there would be potential for the final judicial determination of a particular dispute to differ depending on how a plaintiff chooses to commence proceedings and how a plaintiff chooses to end proceedings with a final judgment. That should not be and cannot be the case. A dispute over how a contract is to be construed must yield the same final judicial determination whether the contract is construed at trial in an action (to which Part II of the Evidence Act does apply) or whether it is construed on a summary judgment application, on a striking out application, on an originating summons or even in arbitration (to all of which Part II of the Evidence Act does not apply).

58 For these reasons, I am bound not only to apply the contextual approach to construction set out in Zurich and Sembcorp but also to apply the common law rules of evidence which find their statutory analogue in s 94 to 100 of the Evidence Act and which are preserved by s 2(2). Any other approach risks countenancing arbitrary differences in outcome resulting from procedural happenstance.

Applying the contextual approach

Text before context

Starting with the parties' words

59 Neither Zurich nor Sembcorp is authority for the proposition that the contextual approach starts with extrinsic evidence. Even under the modern contextual approach, and all the more so with the cautionary notes sounded in both Zurich and Sembcorp, it remains the case that "the primary source for understanding what the parties meant is their language interpreted in accordance with the conventional usage." (See Zurich at [57] citing Lord Hoffmann's speech in BCCI v Ali at [39].) Text comes before context. Thus, even the contextual approach must start with a consideration of the words chosen by the parties themselves to express their agreement.44 This is clear from Zurich's quotation, with approval, of the following passage from an academic text likening the "context" of a contract to a series of concentric circles of meaning:45

The context is a series of circles: the phrase, the sentence, the paragraph, the part of the [contract], the whole of the [contract], and then, outside the [contract] itself, the past dealings of the parties [subject to the rule against the admissibility of pre-contractual negotiations ... ], the trade context, and the objects which the [contract] was intended to achieve.


60 As the Court of Appeal said in Zurich at [130], "the way in which the task of interpretation is to be carried out"46 is that:

"[T]he court will first take into account the plain language of the contract together with relevant extrinsic material which is evidence of its context. Then, if, in the light of this context, the plain language of the contract becomes ambiguous (ie, it takes on another plausible meaning) or absurd, the court will be entitled to put on the contractual term in question an interpretation which is different from that demanded by its plain language."
[emphasis in original]


61 Neither Zurich nor Sembcorp, therefore, permits a party to throw all available extrinsic evidence into a heap together with the words of the contract and then invite the court to rummage with it through the heap to see what turns up. The position is quite the contrary. The contextual approach starts with the innermost of the concentric circles of meaning: the natural and ordinary meaning of the words, phrases and sentences chosen by the parties to express their contractual intention in the document. What Zurich and Sembcorp do is threefold. First, they make clear that although the construction exercise starts with the natural and ordinary meaning of the words chosen by the parties, it does not necessarily end there. Second, they remove a legal rule which sets an artificial a priori barrier in moving from a consideration of the parties' language in the contract itself to any consideration of the external context. That barrier is replaced by the Zurich criteria, which regulate how the contextual approach can legitimately transition from the internal to the external. The end point of every construction exercise now lies, at least potentially, in the last of the series of circles of meaning adverted to in the passage quoted at [59] above.47 Third, both Zurich and Sembcorp reconciled the modern contextual approach, developed in England alongside the English common law of evidence, with the statutory strictures of our Victorian Evidence Act.

Starting with the parties' words is not unduly narrow

62 It could be argued that it is unduly technical and narrow to apply Zurich by starting with the parties' words and then progressing in sequence through concentric circles of meaning. There are indeed passages in Zurich which could be read as suggesting a more liberal approach. But read in context and read with Sembcorp, these passages do not point against the manner in which I intend to apply Zurich.

63 At [110] of Zurich, the Court of Appeal says as follows:

110 In their zealous resistance to extrinsic material, the proponents of the traditional approach to contractual interpretation (as outlined at [47]-[49] above) have neglected the commercial reality and logic that reference to such evidence is necessary even before the actual interpretive exercise takes place (see [52] above).
[emphasis added]


64 These italicised words might be read to suggest that the court must have reference to the extrinsic circumstances even before it interprets the words of the contract. But read in context, and bearing in mind the Court of Appeal's reference in the paragraph quoted to the earlier passage in its judgment at [47]-[49], it is clear that all that the Court of Appeal is saying here is that the traditional approach to construction was wrong to require a finding of ambiguity as a prerequisite to have regard to any type of extrinsic evidence at all, even evidence of the first type (see [44] above). This must be so: it is often impossible to make a finding of ambiguity without reference to material extrinsic to the document. Therefore, the traditional rule is potentially circular. This passage from Zurich does not, however, support a broader proposition that the contextual approach permits or requires a court to begin with extrinsic evidence rather than with the language of the agreement being construed.

65 So too, the passage at [130] of Zurich, cited more fully at [60] above, might be read to suggest that the words of the contract and the extrinsic material must be considered together at the first circle of meaning. But the opening words of [130] refer to "the second issue outlined at [124] above." That second issue is "the way in which the task of interpretation is to be carried out." Those words - "the court will first take into account the plain language of the contract together with relevant extrinsic material which is evidence of its context" - are therefore not intended to suggest that the plain language of the contract and the relevant extrinsic material are interchangeable and equivalent starting points in the construction exercise. The two are put together in that sentence only in order to explain how the entire process of construction - encapsulated in those words - can under the modern contextual approach legitimately lead a court to "put on the contractual term in question an interpretation which is different from that demanded by its plain language." This passage, too, does not support the proposition that the construction exercise under the modern contextual approach permits or requires a court to start with extrinsic evidence rather than with the language of the agreement being construed.

66 Further, as a matter of principle and logic, it must be the case that the starting point in the construction exercise must be the language chosen by the parties. One cannot call extrinsic evidence in aid in a vacuum. The purpose of the construction exercise is not for a court to conduct a roving inquiry into every aspect of the parties' bargain extrinsic to their contract. The purpose of the construction exercise is to ascertain objectively the parties' contractual intention. It is only in light of the words which the parties actually use that it becomes possible to identify those crevices of meaning which extrinsic evidence might illuminate.

67 In short, there is no longer in our law a bright dividing line between the plain language of the contract and extrinsic evidence of its context. But that is not to say that there is no dividing line at all. Instead, there is now a principled dividing line which permits a court to have regard to extrinsic evidence if and only if the evidence satisfies the Zurich criteria and, of course, is admissible in evidence. If the result is that no extrinsic aids to construction are available, a court is left with nothing but the words the parties themselves chose. Even under Zurich, therefore, the starting point remains - and indeed must be - the natural and ordinary meaning of the plain language which the parties have used to express their rights and obligations.

68 In accordance with contextual approach, therefore, I start with the parties' language and progress through the circles of meaning in the principled way signposted by Zurich, and subject to the Sembcorp constraints. In the course of that, I will consider whether I can legitimately take into account the extrinsic evidence proffered by Lucky Realty, and if so, its effect on the construction of the lease.

The parties' intention objectively ascertained

The first circle of meaning: the operative part of the lease.

69 As I noted at [22] above, both parties have framed the dispute before me purely as a matter of construction. Further, the construction which is disputed is of the lease subject to the 1995 variation. The operative part of the lease as varied in 1995 provides as follows:48 


I, the Public Trustee ... HEREBY LEASE unto LUCKY REALTY ... [t]he whole of Lot 3041 of Mukim XXVII together with the buildings in the course of
construction thereon,

(a) TO BE HELD by the Lessee ... for ... sixty (60) years ... YIELDING AND PAYING therefor ... the ... yearly rent ...

(b) The yearly rent shall be revised every five (5) years from the 15th June 1994 to the market rent prevailing at the time of such revision or an increase of 10% of the immediately preceding yearly rent, whichever is the higher. ... The market rent prevailing at the time of such revision means that the valuation of the rental would be based on the existing development and not on an imaginary highest and best use consideration.

(c) ...
[emphasis added] 

The crucial issue I have to determine is the meaning to be ascribed to the terms "yearly rent" and "market rent" in cl (b) of the habendum. The Trustee claims that the plain meaning of both terms is the yearly rent and the market rent for the whole of Lot 5245N. Lucky Realty maintains that both terms understood in context mean the market rent for Block D only.

70 I start with the first circle of meaning: the phrases, sentences and paragraphs used by the parties in this part of the lease (see the passage quoted at [59] above). These words comprise a description of the demised land followed by the habendum. It describes the demised land as "[t]he whole of Lot 3041 of Mukim XXVII together with the buildings in the course of construction thereon". I make four points. These points do no more than state the obvious, but these obvious points set the foundation for what follows. First, the lessor demises to Lucky Realty Lot 3041. Second, it demises to Lucky Realty not just a part of Lot 3041 but the whole of Lot 3041. Third, it demises to Lucky Realty not just the whole of the land comprised in Lot 3041 but also the buildings in the course of construction upon it. Finally, it is in exchange for the demise that Lucky Realty agrees to pay the "yearly rent". That is the significance of the word "therefor" in paragraph (a) of the habendum which I have italicised at [69] above.

71 All of this suggests that the objective meaning of the terms "yearly rent" and "market rent" in cl (b) of the habendum is the yearly rent and market rent for the whole of Lot 3041.

The second circle of meaning: the rest of the lease

72 The words of the operative part of the lease, however, leave unanswered what is meant by "the buildings in the course of construction" on Lot 3041. I therefore move then to the next circle of meaning: the lease as a whole. A consideration of the whole of the lease illuminates what is meant by those words.

73 The word "buildings" appears in many provisions outside the operative part of the lease. All of these provisions speak of Lucky Realty's obligations as they stood at 1975. Thus, cl 1(iii) of the lease specifically obliged Lucky Realty "to erect ... the buildings on the demised land ... in conformity with plans ... submitted by [Lucky Realty] to the Local Government Building Authority." Clause 1(v) obliged Lucky Realty to keep the buildings, once erected, in tenantable repair throughout the term of the lease. Clause 1(vi), as I have said, obliged Lucky Realty not to make any alterations or additions to the buildings without the lessor's consent in writing. Clause 1(vii) obliged Lucky Realty to permit the lessor to enter onto the demised land to view the state of repair of the "buildings". And cl 1(viii) of the lease obliged Lucky Realty to insure the "demised buildings" in the joint names of lessor and lessee.

74 The parties' intention objectively ascertained as at the time of the 1995 variation, therefore, is that the word "buildings" in the description of the demised land is to comprise the buildings that Lucky Realty bound itself contractually in 1975 to construct upon Lot 3041. Further, the parties in their lease consistently refer only to "buildings", ie in the plural and collectively. There is no attempt to distinguish any one of the buildings from the others. The ordinary meaning of "buildings" in this context is therefore "all buildings".

75 All of this again suggests that "yearly rent" and "market rent" in cl (b) of the habendum is to be a rent for not just the whole of the demised land, but is also to be a rent referable to all of the buildings on that land.

Third circle of meaning: extrinsic evidence

76 I move next to the third circle of meaning: evidence of matters which are extrinsic to the contract. I have before me extrinsic evidence that, by the time of the 1995 variation, Lot 3041 had been subdivided, with two of the subdivisions compulsorily acquired by the State. I have before me also extrinsic evidence that the result of that subdivision and compulsory acquisition was to extinguish the parcel of land known as Lot 3041 and to create a new and smaller parcel of land known as Lot 5245N which the Trustee continues to own.

77 It is common ground that the term "Lot 3041" in the lease should be construed as "Lot 5245N". Indeed, that is selfevident. But I prefer to arrive at that conclusion through explicit analysis rather than by concession or by treating it as axiomatic. This is because Lucky Realty submits that if I can legitimately have recourse to extrinsic evidence to conclude that "Lot 3041" ought to be construed as "Lot 5245N", it is equally legitimate for me to have recourse to other extrinsic evidence to conclude that "yearly rent" and "market rent" ought to be construed as "yearly rent for Block D only" and "market rent for Block D only". I do not accept that submission. I therefore now analyse why the law permits me to have recourse to the extrinsic evidence, summarised at [76] above, to construe "Lot 3041" as "Lot 5245N".

"Lot 3041" ought to be construed as "Lot 5245N"

78 In order for me to use the extrinsic evidence of the extinction of Lot 3041 and the creation of Lot 5245N as an aid to construing the lease, I must first consider the admissibility of this evidence and, if it is admissible, I must then apply the Zurich criteria bearing in mind the Sembcorp constraints. Although, as I have explained above, the Evidence Act does not apply to these proceedings, I shall, purely for ease of exposition, refer to each common law rule of evidence which I must apply by the section number of the equivalent statutory rule.

79 As a matter of the law of evidence, this extrinsic evidence is prima facie excluded by s 94 of the Evidence Act because the evidence is adduced for the purpose of varying or contradicting a written agreement which both parties accept is complete. Indeed, not only is the written agreement complete, there is no ambiguity at all on the face of the document as to the identity of the demised land. It is "Lot 3041".

80 Zurich, however, is authority that extrinsic evidence of this type is admissible under the proviso to s 94 found in s 94(f) of the Act. This evidence shows how the language of the lease is related to facts existing at the time of the 1995 variation. It is also admissible under s 97 of the Act. It shows that this language in the lease, although plain, was used in a peculiar sense because it is meaningless in reference to the facts existing at the time of the 1995 variation. Lot 3041 did not exist at the time of the 1995 variation. A reference to it is meaningless. This evidence is also not excluded by s 96 of the Evidence Act. That rule of evidence excludes extrinsic evidence meant to show that the plain language of a document which does apply accurately to existing facts was not meant to apply to those facts. The 1995 variation did not apply accurately to existing facts because no parcel of land known as Lot 3041 then existed. This evidence is therefore outside the scope of any exclusionary rule of evidence.

81 As a matter of the law of contract, this evidence is a permissible aid to construction because it satisfies the Zurich criteria. First, it is clearly relevant, adduced to elucidate the very subject-matter of the lease. Second, it is evidence that was reasonably available to both parties at the time of contracting. The compulsory acquisition and renaming of the remainder of Lot 3041 as Lot 5245N was (and is) an undisputed fact and a fact which was known to both parties at the time of the 1995 variation. Finally, it relates to a clear and obvious context. Given that Lot 3041 did not exist at the time of the 1995 variation, what else, objectively speaking, could the parties have intended the subject-matter of that variation to have been other than that part of Lot 3041 which remained in the lessee's ownership after the compulsory acquisition, ie Lot 5245N?

82 The clear and obvious context supplied by this extrinsic evidence is also supported by returning to the second circle of meaning. The subdivision of Lot 3041 and the compulsory acquisition of parts of it is an event which the parties contemplated and accommodated in their lease. Thus, cl 1(iv) obliged Lucky Realty to maintain the roads, drains, sewers and septic tank on Lot 3041 but only "until the roads, drains, sewers, septic tank made or laid on or in the demised land shall be taken over by the Government or other competent authority". The parties took the compulsory acquisition of the road and septic tank into account prospectively, when they struck their bargain in 1975. Looked at retrospectively in 1995, those events provide a clear and obvious objective context to the words the parties used when they adjusted their bargain then.

83 Finally, neither of the Sembcorp constraints which militate against taking too robust a view of extrinsic evidence apply to this evidence. Evidence of surrounding circumstances is the least problematic of the four types of extrinsic evidence. As Sembcorp makes clear, relevant evidence in this class is admissible virtually without restriction. Further, this extrinsic evidence, as is typical of evidence of this type, is evidence of objective fact and is, in this case, undisputed. Receiving it does not lead to a proliferation of subsidiary issues of fact or law and to the associated increase in cost and delay.

84 A consideration of the whole of the lease in light of the surrounding circumstances therefore demonstrates that the parties' intention objectively ascertained in 1995 was for Lot 5245N to constitute the demised land, being the only part of the extinguished Lot 3041 still owned by the Trustee in 1995. That in turn informs the meaning to be ascribed to the "yearly rent" that Lucky Realty bound itself to pay under the lease and to the "market rent" which Lucky Realty agreed to pay, subject to conditions, with effect from 1999.

85 It is important to note that the extrinsic evidence I am now considering permits me only to construe "Lot 3041" as "Lot 5245N". There is nothing whatsoever in this extrinsic evidence which permits me to construe what was demised to Lucky Realty under the lease, and what Lucky Realty was obliged to pay the yearly and market rent for, as being anything other than the whole of Lot 5245N or as being referable to anything other than all of the buildings erected upon it.

The extrinsic evidence relied upon by Lucky Realty

86 It is at this point in the construction exercise that Lucky Realty relies on additional extrinsic evidence to argue that "yearly rent" and "market rent" ought to be construed as "yearly rent for Block D only" and "market rent for Block D only".49 Lucky Realty arranges this evidence under four heads:

(a)

First, Lucky Realty relies on extrinsic evidence of the parties' pre-contractual negotiations to argue that the common basis of the parties' negotiations leading up to the 1995 variation was that its obligation to pay rent was confined to Block D.

(b)

Second, Lucky Realty relies on the fact that the increased rent under the 1995 variation was backdated to the date on which it received the temporary occupation permit ("TOP") for Block D.

(c)

Third, Lucky Realty relies on the term "existing development" in cl (b) of the habendum as referring to Block D only.

(d)

Fourth, Lucky Realty relies on the parties' conduct subsequent to the 1995 variation in deciding not to procure all the assignees of strata units in Blocks A, B and C to enter into a variation of all of their leases to reflect the 1995 variation to their head lease.50 


87 I hold that I cannot take any of this extrinsic evidence into account in construing the lease. All of this evidence is either inadmissible or fails to satisfy the Zurich criteria.

Two points on admissibility

88 Sembcorp is a reminder that the first step in considering whether any extrinsic evidence may be used as an aid to construction is to apply the evidential rules as to admissibility. I therefore consider admissibility of the evidence proffered by Lucky Realty as a preliminary issue, before examining the content of the extrinsic evidence.

89 A large part of the extrinsic evidence relied upon by Lucky Realty appears in the affidavit of its principal deponent, Ms Ong Sook Mor. I hold that evidence to be inadmissible. A large part of the extrinsic evidence relied upon by Lucky Realty is adduced for the purpose of establishing the parties' subjective intent. I hold that evidence also to be inadmissible.

Extrinsic evidence from Ong Sook Mor is inadmissible

90 Ms Ong Sook Mor is a Director in the Sales Admin and Leasing Division of the Far East Organisation.51 In the third paragraph of her affidavit, Ms Ong asserts the following:

"The facts deposed to herein are based on my personal knowledge and/or the documents from the Defendant. Where the facts are based on my personal knowledge, they are true. Otherwise, the facts are true to the best of my knowledge, information and belief.


A paragraph to this effect is routinely included in all affidavits. When it is included in an affidavit sworn and filed for the purposes of interlocutory applications, it is innocuous enough. It is merely a convenient way of alerting the reader to the deponent's intention to rely on Order 41 Rule 5(2) of the Rules of Court. That rule provides that "[a]n affidavit sworn for the purpose of being used in interlocutory proceedings may contain statements of information or belief with the sources and grounds thereof". Of course, that rule does not require the deponent to include a paragraph to this effect in order to rely on that rule. And including a paragraph to this effect does not, in itself, comply with the requirements of that rule. It is still necessary for the deponent, in the body of her affidavit, to identify clearly which evidence is within her personal knowledge and which evidence is not and, where evidence falls into the latter category, to identify equally clearly her sources for the information she deposes to and her grounds for the beliefs she deposes to. But if an affidavit filed for the purposes of interlocutory proceedings does all of that, no serious complaint can be made if it includes a paragraph to the same effect as Ms Ong Sook Mor's paragraph 3.

91 A paragraph to this effect, however, should not form any part of an affidavit sworn and filed for the purposes of a final determination of the parties' rights and liabilities in any proceedings. Evidence on information or on belief is prima facie inadmissible for that purpose. That is so whether the proceedings are commenced by originating summons or by writ and whether the final determination is to take place in open court or in chambers.

92 The rights and liabilities of the parties in proceedings commenced by writ are typically determined with finality at trial. The contents of affidavits filed as the deponents' evidence in chief for trial are governed by s 62 of the Evidence Act read with Order 38 Rule 2(5) of the Rules of Court. The general rule is that evidence of fact at trial must be given from personal knowledge in order to be admissible. Any exceptions to that general rule must be found in the Evidence Act or in the common law of evidence which was not repealed by s 2(2) of the Act. Those exceptions can be raised and applied pre-trial pursuant to Order 38 Rule 3 of the Rules of Court or in the course of the trial itself, as the evidence is adduced and objection is taken.

93 The rights and liabilities of the parties in proceedings commenced by originating summons are determined with finality in chambers. The rights and liabilities of the parties in proceedings commenced in writ may also be determined with finality in chambers - under Order 14, under Order 18 Rule 19 or under Order 27. Where that determination takes place in chambers, the contents of the affidavits are governed by Order 41 Rule 5(1). That rule is to the same effect as s 62 of the Evidence Act. It provides that "an affidavit may contain only such facts as the deponent is able of his own knowledge to prove". The exception to this rule in Order 41, Rule 5(2) is unavailable because proceedings aimed at securing judgment are not interlocutory proceedings. The exceptions in Parts I to III of the Evidence Act are also unavailable, as they have no application to affidavits. Any exceptions to the general rule found in Order 41, Rule 5(1) must therefore be found only in the common law of evidence.

94 If a deponent of any affidavit sworn for the purpose of being used at a hearing which finally determines the parties' rights and liabilities intends to rely on any exception to the requirement for direct evidence, whatever the source of that requirement, the deponent should identify that exception in her affidavit and set out the evidence necessary to establish that the evidence proffered comes within the exception. Thus, if Lucky Realty wishes me to receive and consider the extrinsic evidence set out in Ms Ong Sook Mor's affidavit in determining finally the parties' rights and liabilities, it must establish that her evidence of fact is direct evidence as required by Order 41 Rule 5(1) or comes within a recognised exception to that requirement.

95 It is clear from the contemporaneous documents that Ms Ong Sook Mor was not personally involved in the historical events on which Lucky Realty relies as extrinsic evidence. Having acknowledged in paragraph 3 of her affidavit that some parts of her evidence are not within her personal knowledge, she makes no attempt to distinguish in the body of her affidavit between the evidence which is within her personal knowledge and the evidence which is not. More specifically, at no point in her evidence of these historical events does she assert that that evidence is within her personal knowledge. I therefore have no basis on which to find that Ms Ong Sook Mor has any personal knowledge of the historical extrinsic evidence at all. That evidence, to the extent that it is contentious, is therefore wholly inadmissible.

96 In closing, I should note that the Trustee's affidavits also contain a paragraph to the same effect as paragraph 3 of Ms Ong Sook Mor's affidavit. The Trustee's deponent too fails to make clear which parts of her evidence are within her personal knowledge and which parts are not. But the Trustee is not relying on contentious extrinsic evidence to support its construction of the lease. It relies simply on the plain words of the lease. The significance to the Trustee's case of this evidential defect is much less significant than it is to Lucky Realty's case.

Evidence of subjective intent is inadmissible 

97 The second point of admissibility I consider is in relation to the extrinsic evidence proffered by Lucky Realty insofar as its purpose is to prove the parties' subjective intent. Both Zurich and Sembcorp make clear that even on the modern contextual approach, extrinsic evidence of subjective intent is excluded as an aid to construction unless there is an ambiguity and unless also that ambiguity is latent. This is the effect of both the law of evidence and of the law of contract. 

98 The only material ambiguity in the lease is a latent ambiguity as to the meaning of "Lot 3041" in light of the facts and circumstances surrounding the 1995 variation. That ambiguity arises because no lot with the number 3041 existed in 1995. I have resolved that ambiguity definitively by receiving, without objection but also in accordance with the law, extrinsic evidence of the extinction of Lot 3041 and the creation of Lot 5245N.

99 Once that has been determined, there is no remaining ambiguity as to the extent of Lucky Realty's rights and obligations in respect of the demised land. That is so whether one construes the lease as it stood when the parties entered into it in 1975 or as it stood after they varied it in 1995. When the parties entered into the lease in 1975, it was clear on the face of the lease that Lucky Realty took on lease the whole of Lot 3041 and all of the buildings then in the course of construction on it. It was also clear on the face of the lease that in exchange for that, Lucky Realty obliged itself to pay rent for the whole of that land and referable to all of those buildings. That obligation is unambiguous. The 1995 variation left this bargain unchanged save only that the parties' reference to Lot 3041 is to be construed as a reference to Lot 5245N. Lucky Realty's obligation to pay rent was unambiguous before the 1995 variation and remained unambiguous after the 1995 variation is properly construed. In those circumstances, the fact that Lucky Realty had, by the time of the 1995 variation, subdivided Blocks A, B, C and D into strata units and sold virtually all of the units in three of those blocks on long leases cannot, as a matter of construction, introduce an ambiguity into the words defining Lucky Realty's obligation to pay rent where none previously existed.

100 In the absence of any ambiguity, extrinsic evidence is wholly inadmissible for the purpose of establishing the parties' subjective intent in entering into the 1995 variation. That is so whether the subjective intent alleged is a subjective intent of Lucky Realty which was never expressed to the Trustee, a subjective intent of Lucky Realty which was expressed to the Trustee or even a subjective intent which was expressed and shared by both Lucky Realty and the Trustee. That rule appears harsh but is necessary in order to achieve certainty in contract, particularly where (as here) the parties' contract creates proprietary rights and where those rights are alienable to third parties. The nature of the document being construed and its commercial character are legitimate considerations in construing it (see Zurich at [132(a)], cited at [34] above). Every contract has only one true construction. A third party who purchases the freehold of Lot 5245N from the Trustee has no way of extracting from the lease any of these subjective intentions insofar as they remain unexpressed in the parties' objective intention manifested by the words which they chose to define their rights and obligations read in context. If subjective intention - even a shared subjective intention - determines the one true construction of the lease, the purchaser will find itself bound by an unexpressed subjective intention to confine to Block D alone the obligation to pay rent, even though the purchaser was never a party to that intention and would ordinarily be in no position to discover it. That cannot be correct.

101 If the heart of Lucky Realty's complaint is that the 1995 variation failed to capture the subjective intent of one or both of the parties, its remedy lies in the contractual principles relating to mistake or in the equitable doctrine of rectification. Its remedy does not lie in stretching the contextual approach to construction beyond the principled limits set in Zurich and Sembcorp. I therefore hold that none of the extrinsic evidence proffered by Lucky Realty may be used for the purpose of establishing the parties' subjective intent in entering into the 1995 variation.

102 I now turn to consider one by one the four heads of extrinsic evidence upon which Lucky Realty relies.

First head: basis of fixing $120,000 in 1995

Available evidence of pre-contractual negotiations 

103 Under the first head, Lucky Realty seeks to adduce extrinsic evidence of the parties' pre-contractual negotiations to show that its obligation to pay the "yearly rent" and the lessor's right to revise the yearly rent to a "market rent" should be construed as being referable to Block D alone. Lucky Realty's case is that this was the common basis of the parties' negotiations preceding the 1995 variation. In view of my holding at [100] above, I consider this evidence only insofar as it casts a different light on the meaning of the words used by the parties objectively ascertained and not insofar as it is evidence of the parties' subjective intent.

104 In order for me to use extrinsic evidence of the parties' pre-contractual negotiations as an aid to construing "yearly rent" and "market rent", the evidence must be admissible in accordance with the law of evidence and must satisfy the Zurich criteria, bearing mind the Sembcorp constraints.

105 The only deponent from either party who is able to give direct evidence of the pre-contractual negotiations is Ms Lydia Sng, a property valuer with Knight Frank Cheong Hock Chye & Bailieu. She advised the Trustee and the beneficiaries in these negotiations. She, together with a senior colleague, also represented the Trustee in conducting the negotiations with Lucky Realty. She has filed two affidavits in these proceedings. Both of them support Lucky Realty's case. Her evidence is that both the Trustee and Lucky Realty proceeded in the negotiations leading up to the 1995 variation on the basis that the increased "yearly rent" of $120,000 and the "market rent" in the rent escalation clause were referable only to Block D and not to the whole of Lot 5245N.

106 Apart from Ms Sng's direct evidence, the only other evidence of the pre-contractual negotiations before me are the minutes of a meeting of the committee of the beneficiaries of the trust convened by the Trustee on 26 May 1994. This document was adduced through the affidavit of the Trustee's only witness. That witness has no personal knowledge of that meeting. But Ms Sng does: she attended that meeting. She addresses the contents of these minutes directly in her second affidavit. In it, she does not suggest that the minutes are inaccurate. Her only contention is in respect of the inferences to be drawn from the minutes. Ms Sng's evidence of the pre-contractual negotiations and the minutes of the meeting of 26 May 1994 are therefore prima facie admissible as being direct evidence. The next question is whether it is admissible under the law of evidence as an aid to construction.

Evidence is of subjective intention and therefore inadmissible 

107 I do not consider that this extrinsic evidence is admissible. What Lucky Realty seeks to adduce under this branch of its argument is evidence that the parties took a common position, known to each other, during their negotiations. That common position was that the yearly rent and the market rent was not referable to the Trustee's demise of the whole of Lot 5245N to Lucky Realty but was referable only to Block D. Assuming that the parties indeed negotiated from that common position, evidence of it is inadmissible. There is no chain of reasoning that connects this evidence and Lucky Realty's desired conclusion - as to the meaning that ought to be ascribed to "yearly rent" and "market rent" - that does not require a journey into the parties' subjective intentions. That is the only relevance of this extrinsic evidence. But, as I have pointed out at [97]- [101] above, there is in this case no latent ambiguity in the lease which renders extrinsic evidence of subjective intention admissible.

108 I also do not consider that this evidence is admissible under the proviso to s 94 of the Evidence Act set out in s 94(f), to show in what manner the language of the lease is related to facts existing at the time of the 1995 variation. The parties' subjective intentions cannot be "existing facts" within the meaning of s 94(f). That interpretation of s 94(f) would deprive of all meaning the rule requiring latent ambiguity before evidence of subjective intention can be received.

109 By way of example, the result would be different if the dispute in this case were over whether the objectivelyascertained meaning of the words "[t]he whole of Lot 3041" in the lease required Lucky Realty to pay rent to the Trustee referable even to the parts of Lot 3041 which had, by the time of the 1995 variation, been compulsorily acquired. The evidence of the compulsory acquisition and of the creation of Lot 5245N would be admissible in those circumstances to show how the language of the varied lease is related to facts existing at the time the variation was agreed, ie to the reduced parcel of land which remained in the Trustee's ownership in 1995. The connection between "Lot 3041" in the lease and Lot 5245N on the ground can be made objectively from the extrinsic evidence of the surrounding circumstances, without a chain of reasoning which requires a journey into the parties' subjective intentions. That extrinsic evidence is therefore within the scope of s 94(f) and outside the scope of the rule barring evidence of subjective intention.

110 So too, extrinsic evidence would be admissible if the dispute were over whether the objectively-ascertained meaning of the words "yearly rent" and "market rent" required Lucky Realty to pay rent to the lessor not only for the use of the whole of the land comprised in Lot 5245N but also for the use of all of the buildings erected upon the land. If the Trustee took that position, extrinsic evidence would then be admissible to establish the fact that the buildings in question were erected at Lucky Realty's own cost and that "rent" in the lease should therefore be construed as "ground rent", ie rent only for the use of the land occupied by the buildings and not for the buildings themselves. That construction would prevail despite any apparent contradiction with the lease's description of the demised land, which refers not only to Lot 3041 but also to all of the buildings which were in the course of construction on the land in 1975. The connection between ground rent as a legal concept and the rent payable under the lease can be made objectively from the surrounding circumstances and the nature of the lease, without a chain of reasoning which requires a journey into the parties' subjective intentions. This extrinsic evidence too would be within the scope of s 94(f) and outside the scope of the rule barring evidence of subjective intention.

111 The extrinsic evidence proffered by Lucky Realty under this first head is wholly inadmissible. It serves no purpose other than to establish subjective intention. There is no latent ambiguity to render evidence of subjective intention admissible.

No clear or obvious context 

112 In case I am wrong on this, I consider also whether the extrinsic evidence proffered under the first head satisfies the Zurich criteria. I first set out a summary of the parties' pre-contractual negotiations taken from the affidavits of Ms Sng and from the minutes of the 26 May 1994 meeting she attended before applying the criteria.

113 On 26 May 1994, the Trustee convened a meeting. In attendance were representatives of the Trustee, representatives of the committee of beneficiaries of the trust and Ms Sng. The purpose of the meeting was for the committee to seek Ms Sng's advice on how to proceed in its negotiations on rent with Lucky Realty. The opportunity to renegotiate the rent had arisen in the context of trying to resolve without litigation the Trustee's dispute with Lucky Realty over its decision to redevelop Block D without seeking the Trustee's consent.

114 At the meeting, Ms Sng presented her valuation of the ground rent payable for Block D as $269,310 per annum. She derived this figure by estimating the area of the land occupied by Block D, assuming a plot ratio of 1.2, valuing the land at $5,386,200 and taking 5% of that figure as the annual ground rent. Ms Sng's valuation was a preliminary valuation for three reasons: (i) she did not know the exact area of the land occupied by Block D; (ii) she did not know Lucky Realty's cost in redeveloping Block D; and (iii) she did not know the exact plot ratio of the land.

115 The committee then asked Ms Sng to prepare a valuation of the ground rent for the whole site, ie Lot 5245N. The minutes record that the reason for this request was because "the breach is on the whole site". In context, this reflects the committee's view that Lucky Realty's breach of the lease had triggered the lessor's right to re-enter upon the whole site, and not just in respect of Block D. The breach thereby put Lucky Realty's lease at risk of forfeiture over the whole site, and not just in respect of Block D. That, in turn, created an avenue for the Trustee to let the whole site to a new lessee at the market rent in 1994 and thereby reopened Lucky Realty's obligation to pay rent for the whole site in order to avoid that consequence. The minutes further record that the committee intended to use the ground rent for the whole site as a bargaining chip to negotiate not just for an increase in the fixed rent, but also for a new clause permitting the lessor to increase the rent during the currency of the lease.

116 Ms Sng explained to the committee that the numerous limitations on the information available to her would make it difficult if not impossible to do a valuation of the ground rent payable for the whole site. She further said that a valuation subject to those limitations would not be acceptable to the court. For the purposes of discussion, however, she estimated that the ground rent for the whole site was $900,000 per annum. Her view, however, was that Lucky Realty would not accept a figure of $900,000 per annum as rent because she had already had discussions with Lucky Realty in which they were prepared to offer a new rent of only $60,000 per annum.

117 The committee then had a long discussion. It concluded by instructing Ms Sng to do three things: (i) to provide an indicative value for the ground rent for the whole site; (ii) to seek the necessary information from Lucky Realty to refine her valuation of the ground rent for the area of the breach, ie for Block D; and (iii) to recommend to the committee a minimum rent that the Trustee should accept. The committee's intention was to give Ms Sng and OCBC Trustee a mandate to negotiate with Lucky Realty to arrive at an agreed revised rent and a rent escalation clause. The committee also resolved to take their dispute with Lucky Realty to court if these negotiations failed.52

118 Ms Sng cannot recall preparing a valuation of the rent for the whole site. Instead, she prepared a final valuation of the rent for Block D alone. Taking into account the actual land area of Block D and its actual plot ratio, she valued the rent for the land it occupied at $300,000 per annum.53 There is a dispute of fact as to whether Ms Sng sent this final valuation to the Trustee in 1994, when she initially prepared it, or later in 1995, when she sent that valuation to Lucky Realty, apparently for a second time, at its request. That, to my mind, is a red herring. The extrinsic evidence that Lucky Realty relies upon under this head goes beyond a particular valuation and arises from Ms Sng's evidence as to the parties' common position in the negotiations. It is therefore not necessary for me to resolve this difficult issue of fact.

119 Ms Sng and OCBC Trustee thereafter engaged Lucky Realty in negotiations. Her evidence is that the Trustee instructed her to negotiate with Lucky Realty on the basis of the rent for Block D only and that that was the basis on which both parties negotiated. After various offers and counter-offers, the committee of beneficiaries instructed her to accept Lucky Realty's proposal of a ground rent of $120,000 per annum from 1994 to 1999 with the rent increasing according to the formula set out in cl (b) of the habendum every five years from 1999 onwards (see [69] above).

120 The Zurich criteria require me to consider whether particular extrinsic evidence proffered by one party is relevant, was reasonably available to all the contracting parties at the time they contracted and relates to a clear or obvious context. I assume, without deciding, that this evidence is relevant to the issue at hand, ie to determining the meaning objectively ascertained of "yearly rent" and "market rent" in cl (b) of the habendum.

121 The second of the three Zurich criteria is somewhat difficult to apply to the extrinsic evidence Lucky Realty proffers under its first head. It is difficult to see in what way a common basis for negotiation can be said to have been reasonably available to both parties in the same way, for example, as the information about the subdivision and compulsory acquisition of parts of Lot 3041 was reasonably available to both parties. Therein lies the root of the difficulty in applying the second criterion to this extrinsic evidence. The only purpose this evidence can serve is as evidence of the parties' subjective intention. Nevertheless, I will assume in Lucky Realty's favour that this criterion is made out.

122 I hold, however, that the third Zurich criterion is not satisfied. The extrinsic evidence of the parties' pre-contractual negotiation does not supply a clear or obvious context to justify ascribing to the terms "yearly rent" and "market rent" in cl (b) of the habendum a meaning which limits it to the rent referable to Block D only, in order to give effect to the parties' intention, bearing in mind always that that intention is to be ascertained objectively rather than subjectively.

123 Assume an objective observer who is aware of the circumstances surrounding the 1995 variation, who witnesses the pre-contractual negotiations summarised at [113]-[119] above and who is also aware of the language chosen by the parties to embody the agreement that results from those negotiations. Assume further that this objective observer is prohibited from looking inside either party's mind. I have no hesitation in concluding that an observer in that position would find it neither clear nor obvious that Lucky Realty's preferred construction of the lease is the true construction of the lease. 

124 That observer would be aware that OCBC Trustee instructed Ms Sng to value the rent for Block D only. The observer would note, however, that the lessor was well aware that its right to receive rent under the lease - and its right to renegotiate rent under the lease - extended to the whole of Lot 5245N. That is why, at the meeting on 26 May 1994, the committee took the position that Lucky Realty's breach, although in relation only to Block D, permitted it to reopen the issue of rent payable for the whole of Lot 5245N (see [115] above). That observer would also note that, by the end of that meeting, Ms Sng had placed before the Trustee three widely differing figures for a new rent to be charged under the lease. The highest figure was $900,000 per annum, being Ms Sng's indication of the rent for the whole of Lot 5245N. The intermediate figure was $269,310 per annum, being Ms Sng's valuation of the rent for Block D only. Ms Sng later revised this intermediate figure, with the benefit of more accurate data, to $300,000 per annum. The lowest figure was $60,000 per annum, being the last offer from Lucky Realty which Ms Sng reported to the meeting.

125 That observer would also note that the meeting ended with the Trustee instructing Ms Sng to refine her figure for the rent payable for the whole of Lot 5245N and to value more accurately the rent for Block D. The latter valuation was prepared but the former was not. Ms Sng does not explain in her affidavit why the former valuation was not prepared. That omission does not supply the clear or obvious context that Lucky Realty advocates. The omission could indeed have been because OCBC Trustee and Lucky Realty decided to negotiate on the basis that the "yearly rent" and the "market rent" would be the rent referable to Block D only. But it could also have been because Ms Sng was unable to get the information necessary to arrive at a valuation of the rent for the whole of Lot 5245N, or to arrive at a valuation which was more precise than her indication at the meeting of $900,000 per annum. Or it could have been because Lucky Realty's offer of $60,000 was so far away from even the intermediate figure of $300,000 that it was not then worthwhile incurring the time and cost of setting an upper limit for the negotiations which was even further away from $60,000.

126 That observer would then note the parties agreed on a figure of $120,000 per annum and that that figure was well below both the highest and the intermediate figure and closest to Lucky Realty's opening figure. From that, he might infer that that figure was indeed derived and agreed at on the basis that it was the rent referable to Block D only. But he would note that in reducing their agreement on a figure of $120,000 into writing in the 1995 variation, the parties included no indication of how that figure had been derived or why it had been agreed. Instead, they chose to leave the description of the land in the lease unchanged, referring to the whole of Lot 5245N, and to insert the new figure of $120,000 into the lease as a gross figure. The new gross figure replaced the existing gross figure of $3,877.15, which was the rent for the whole of Lot 5245N. The observer would be aware also that Lucky Realty retained not only Block D under the lease but also two strata units in Blocks A to C and wonder why in that context, it should be clear or obvious that Lucky Realty's obligation to pay rent should be referable to Block D only.

127 Finally, that observer might be concerned that the parties would be unlikely to have agreed on a rent to be paid from 1994 to 1999 at the rate of $120,000 per annum and agreed thereafter to pay a market rent which, based on Ms Sng's opinion at the 26 May 1994 meeting, would be a multiple of that figure by seven or eight times. That too could be because Lucky Realty's obligation to pay rent was referable to Block D only. But it could also be because the fixed sum of $120,000 was agreed on a commercial basis as a gross figure, representing no more than what Lucky Realty was prepared to pay from 1994 to 1999 to avoid litigation and no less than the lessor was prepared to accept from 1994 to 1999 to avoid litigation, with the ground rent thereafter to be calculated and paid for the whole of Lot 5245N. There is no reason why one meaning should, in this context, be clear or obvious and the other not.

128 The search for indications of objective intent in pre-contractual negotiations is an endless one, layering hypothesis upon hypothesis. This search is no exception. I stop here. The exercise undertaken thus far demonstrates that it is quite impossible to say that extrinsic evidence of the parties' pre-contractual negotiations supplies an objective context which is either clear or obvious as an aid to construing the words which they eventually chose to embody their agreement in the way that Lucky Realty puts forward.

Second head: rent was backdated to the TOP date for Block D

129 I move now to consider the second head of extrinsic evidence. Lucky Realty relies on the fact that the parties agreed in the 1995 variation to impose the fixed rent with retrospective effect by reference to the TOP date for Block D. This is, in point of fact, true. In a letter dated 25 July 1995 from Ms Sng's superior to Lucky Realty, it was proposed on behalf of the Trustee that the increase in rent which was then being negotiated should take effect from the TOP date for Block D.54

130 As a matter of the law of evidence, this extrinsic evidence is admissible. It shows how the language of the 1995 variation is related to existing facts. It is extrinsic evidence of surrounding circumstances. It is within the first of the four types of extrinsic evidence. It is ordinarily admissible without restriction.

131 As a matter of the law of contract, however, I hold that it may not be used as an aid to construction. It certainly satisfies the first two Zurich criteria. It is relevant to the issue at hand, which is construing the words "yearly rent" and "market rent". It was also available to both parties. Indeed, it was the Trustee, and not Lucky Realty, who proposed to backdate the new rent to the TOP date for Block D. But once again, I find that it does not supply a clear or obvious context.

132 It is certainly true that an objective observer could view the context supplied by this fact as context suggesting that the increased rent was to be referable to Block D only. But that is not the clear or obvious context. The increased rent could have been backdated to 15 June 1994 because that was the date on which Lucky Realty, in the Trustee's view, breached the lease and because it was that breach which was the trigger for the renegotiation of rent for the whole of Lot 5245N. Or it could have been backdated to 15 June 1994 because at least part of the contractual objective of any lessee's obligation to seek its lessor's consent before redeveloping any part of the demised land is to give the lessor an opportunity to renegotiate the rent in order to capture for itself some of the enhanced economic benefit that will accrue to the lessee from that redevelopment and which was not contemplated when the parties set the existing rent. That economic benefit accrued to Lucky Realty from the date on which it could begin to earn an enhanced income from the redeveloped Block D, ie upon the issue of the TOP on 15 June 1994. And so its obligation to pay the Trustee the increased rent also accrued from that date.

133 Again, I find it impossible to say that this extrinsic evidence supplies an objective context which is either clear or obvious as an aid to construing in the way that Lucky Realty puts forward the words which the parties eventually chose to embody their agreement.

Third head: "existing development"

134 Under its third head, Lucky Realty switches its focus from the introductory words to cl (b) of the habendum to the concluding words of that clause. Those concluding words define "market rent" and provide as follows: "... market rent ... means that ... the rental would be based on the existing development and not on an imaginary highest and best use consideration". Lucky Realty's argument is that the objective meaning to be ascribed to the words "existing development" is Block D, when those words are construed in light of two particular items of extrinsic evidence. The result is that the market rent construed in context means the market rent referable to Block D only.

135 The two items of extrinsic evidence that Lucky Realty relies on55 are, first, that Block D was the subject of the alleged breach which led to the 1995 variation and, second, that OCBC Trustee was well aware that Lucky Realty had assigned to third parties its interest under the lease insofar as it related to all the strata units in Blocks A to C except for two.56

136 This extrinsic evidence is admissible under the law of evidence. It shows how the language of the 1995 variation is related to existing facts. It is extrinsic evidence of surrounding circumstances. It is within the first of the four types of extrinsic evidence. It is ordinarily admissible without restriction. Applying the Zurich criteria, I accept also that this evidence is relevant and was available to both contracting parties. Again, however, I hold that it does not supply the clear or obvious context necessary to support Lucky Realty's argument.

137 If one looks at these concluding words of cl (b) alone, there are of course no words limiting its scope to Block D. Read in context but within the four corners of the lease, the concluding words sit comfortably together with the description of the land and the obligation to pay the "yearly rent". These words provide that the "market rent" is to be the ground rent payable for the whole of Lot 5245N based on the existing intensity of Lucky Realty's development of the whole of Lot 5245N at the time the rent is to be revised and not on a hypothetical maximum intensity that Lot 5245N could be developed to.

138 Neither of the two facts that Lucky Realty proffers under this head provide a clear or obvious context to advance the construction of these concluding words beyond this point. When the lease comes into effect in 1975, the Trustee is entitled to yearly rent over the whole of Lot 3041. After the subdivision and compulsory acquisition of parts of Lot 3041, the Trustee is entitled to yearly rent for the whole of Lot 5245N. Lucky Realty assigns its leasehold interest in almost all - but crucially not all - of the units in the Blocks A to C. Almost twenty years later, Lucky Realty allegedly breaches the lease. This causes the lessor to reopen the issue of how much rent Lucky Realty should pay going forward.

139 There is nothing in the objective context supplied by either of the two facts which makes it clear or obvious that the Trustee, in reaching the agreement with Lucky Realty embodied in the 1995 variation, intended "existing development" to mean Block D, and Block D alone, thereby giving up its right to be paid ground rent for those parts of Lot 5245N which were not occupied by Block D. Indeed, an objective observer would wonder why the Trustee, which was well aware that it had a right to ground rent for the whole of Lot 5245N (see [115] above) and who intended to use that right as a bargaining chip in the ensuing negotiations, would give up that right even if it knew that Lucky Realty had assigned virtually the entirety of its interest in the other blocks. It is of course possible that the Trustee gave up this right in exchange for a new right to charge a higher rent and to increase the rent. But there is no suggestion in the evidence that this is how the Trustee chose to play this bargaining chip.

140 Further, it is significant that cl (c) of the new habendum expresses clearly the parties' intention to preserve the contractual force of the original lease save only to the extent that it was varied in the 1995 variation. Nothing in the 1995 variation cuts down the lessor's right to ground rent for the whole of Lot 5245N. There is nothing in the original lease or in the varied lease to suggest that the Trustee's right to collect ground rent for the whole of Lot 5245N was in any way contingent on Lucky Realty's continued ownership of the strata units in Blocks A to C, or upon Lucky Realty's right to pass that ground rent on to those subsidiary proprietors.

141 Once again, it is impossible to say that these two facts supply a clear or obvious context which justifies ascribing the limited meaning which Lucky Realty contends for to the words "existing development".

Fourth head: subsidiary proprietors' leases not varied

142 Under the fourth head, Lucky Realty relies on extrinsic evidence in the form of the parties' conduct subsequent to the 1995 variation to show that its obligation to pay rent and the lessor's right to revise the rent to a market rent should be construed as being referable to Block D only. Some further background information is necessary to understand Lucky Realty's argument under this head.

143 Following the exchange of letters by which the 1995 variation was agreed, a formal instrument to record the variation to the lease was drawn up and executed by the parties.57 That instrument was registered with the Registry of Land Titles and Deeds on 18 December 1995 as instrument number I/026520K.58 On 23 February 1996, however, the Registry rescinded the registration and rejected the instrument as being "not in order." The Registry's reasoning was that Lucky Realty's leasehold interest under the lease had been strata subdivided with subsidiary certificates of title being issued to all of the subsidiary proprietors to whom Lucky Realty had sold the strata units. Lucky Realty had thus divested all of its interest in the strata units under its lease with the Trustee. The Registry's view, therefore, was that the variation had to be lodged not against Lucky Realty's lease with the Trustee but against each of the subsidiary proprietors' leases.59 The instrument of variation was withdrawn, but the next step suggested by the Registry was never taken. The leasehold interests of the subsidiary proprietors in Blocks A, B and C were never varied to reflect the 1995 variation. Lucky Realty submits that this shows that the 1995 variation has nothing to do with the land occupied by Blocks A, B and C. This subsequent conduct is therefore said to supply the necessary context to read "yearly rent" and "monthly rent" in the habendum as being rent referable to Block D only.

144 Like the extrinsic evidence under Lucky Realty's three other heads, this extrinsic evidence too fails to satisfy the third Zurich criterion. It fails to supply a clear or obvious context. It is of course objectively true that the parties decided not to register the 1995 variation against the leasehold interest of each of the subsidiary proprietors who owned strata units in Blocks A to C. That is indeed capable of indicating that the parties' intent objectively ascertained at the time of the 1995 variation was for the variation to deal with the rent referable to Block D only, and not at all with the rent referable to Blocks A to C. But it is not clearly and obviously so. It is equally, if not more, possible that the parties saw no need to incur the time, cost and inconvenience of varying the leasehold interests of all of the subsidiary proprietors when the variation affected only Lucky Realty's contractual obligation to pay rent to the Trustee for the whole of Lot 5245N and affected none of the subsidiary proprietors' contractual obligations or proprietary rights. I bear in mind too that extrinsic evidence of this type - subsequent conduct - is one of the types of extrinsic evidence to which Sembcorp attached special caution.

145 Lucky Realty then submits that the "Property Address" given in the formal description of the land in the instrument of variation refers to Block D alone, because it gave its name as "Bedok Shopping Centre".60 Lucky Realty asserts that "Bedok Shopping Centre" is the name of Block D, whereas all four of Blocks A to D were always known collectively as "Bedok Shopping Complex".61 I reject this argument for a number of reasons. First, and foremost, how the parties described the land in question in a non-contractual document prepared more than a year after their contract is of little or no relevance to ascertaining the parties' objective intention at the time of contracting. Second, to the extent that it is relevant, the instrument describes the land as "Vol 227, Fol 2, Mk 27, Lot 5245". Opposite this, in the "Property Address" column, appears the word "Whole" above "Bedok Shopping Centre". The instrument supplies no clear or obvious context. It can equally plausibly be read as referring to the whole of Lot 5245N. Third and finally, there is no evidence of invariably consistent usage to support Lucky Realty's submission on the meaning of "Bedok Shopping Centre" as opposed to "Bedok Shopping Complex".

Beyond penumbral meaning 

146 Finally, even if I am wrong in all of this, and this welter of extrinsic evidence does supply a clear and obvious context to assist in ascertaining the objective intention of the parties, I hold that to construe "yearly rent" and "market rent" in cl (b) of the new habendum as meaning "yearly rent for Block D" and "market rent for Block D" goes so far beyond even the penumbral meaning of these words that it amounts to rewriting the parties' contract under the guise of applying the contextual approach to construction. In a legal sense, "ground rent" is within the penumbral meaning of "rent" (see [110] above). In a conceptual sense, "Lot 5245N" is within the penumbral meaning of "Lot 3041" (see [109] above). In no sense is "yearly rent for Block D only" or "market rent for Block D only" within even the penumbral meaning of "yearly rent" or "market rent". The true effect of Lucky Realty's extrinsic evidence is to contradict, vary, add to or subtract from the terms of the parties' lease, a contract which the parties reduced into writing in the form of a document. Zurich makes clear at [134] that that is impermissible.

Result is not absurd

147 Lucky Realty says that finding it liable to pay ground rent on the whole of Lot 5245N produces an absurd result. Lucky Realty's evidence is that its income from Block D falls below $1m and is therefore insufficient to cover the market rent for the whole of Lot 5245N of $1.3m. This is not an absurd result. It may be an uncommercial result. But if Lucky Realty is faced with an uncommercial result, it is because it failed to reserve for itself a right to charge its assignees a ground rent, back to back with its obligation to pay a ground rent under its lease from the Trustee.

148 A lease creates both contractual rights and a proprietary interest. When Lucky Realty created a strata subdivision over Lot 5245N and assigned to purchasers its interest in all but two of the strata units outside Block D, it conveyed to each assignee both a proprietary interest in the strata unit in question and the contractual benefits of its lease with the Trustee. But Lucky Realty remained subject to the contractual burdens of its lease with the Trustee. One of these burdens is the obligation to pay ground rent to the Trustee for the whole of Lot 5245N for the entire 60-year term of the lease.

149 Every developer who carves subsidiary leasehold interests in land out of its own superior leasehold interest with a view to assigning those subsidiary interests to purchasers has to make a commercial decision as to how it wishes to extract value from them. It can assign the remainder of its lease to them in exchange for a one-time capital sum paid upon the assignment. Or it can do so in exchange for a capital sum for the building and a ground rent for the land underneath it. If a developer chooses not to reserve a contractual right to collect a ground rent from its assignees even though the developer is contractually liable to pay a ground rent to its lessor, it takes a commercial risk. If a developer who has not reserved a contractual right to collect a ground rent from its assignees later agrees to vary its contract to allow the ground rent which it has to pay to its lessor to rise with the market, it is taking an even bigger commercial risk. In either case, when that risk eventuates, it may give rise to a result which the developer finds to be uncommercial. But it does not give rise to a result which is objectively absurd. It is simply the result of a failure to foresee and to cater for a foreseeable commercial risk.

Estoppel by convention

150 As an alternative to its construction argument, Lucky Realty argues that the doctrine of estoppel by convention applies to preclude the Trustee from calculating the agreed yearly rent by reference to the whole of Lot 5245N. Even if, as I have found, the lease on its true construction obliges Lucky Realty to pay rent for the whole of Lot 5245N, Lucky Realty submits that the Trustee has always charged a rent which is referable to Block D only. It is therefore said to be unconscionable for the Trustee now to contend otherwise.62

151 I reject this argument. Assuming that the parties indeed proceeded on a common assumption as to the basis of rent, the effect of that would be to preclude the Trustee, quite apart from the issue of limitation, from revisiting the amount of rent it claimed and collected from Lucky Realty from 1975 to 2009. But it cannot extinguish the Trustee's contractual right to collect rent in accordance with the lease, properly construed, from 2009 onwards.

152 Support for this proposition of law can be found in the decision of Parker J in Philip Collins Ltd v Rahmlee Davis and Louis Satterfield [2001] ECDR 17 at [66]:

In my judgment the defendants' contention that estoppel by convention (assuming it applies in the instant case) can have prospective effect by, in effect, changing the meaning of the Agreement for the future, is wrong in law. As Lord Donaldson M.R. said in Hiscox v. Outhwaite [1992] 1 A.C. 562 at 575, referring to the judgment of Bingham L.J. in Norwegian Cruises A/S v. Paul Mundy Ltd [1988] 3 Lloyd's Rep. 343 (a passage cited with approval by the Court of Appeal in Republic of India v. India Steamship Co Ltd [1998] A.C. 878 at 891F): "... once a common assumption is revealed to be erroneous, the estoppel will not apply to future dealings."


153 Given that the Trustee has withdrawn its attempt to revisit the rent claimed and collected from 1975 to 2009, there is no longer any aspect of the Trustee's claim against which the doctrine of estoppel by convention can operate as a defence.

Conclusion

154 The dispute between the parties arose primarily because of their competing constructions of the rent escalation clause which was inserted into the lease by the 1995 variation. For the reasons set out above, I am wholly unable to conclude that the parties' objective intention was to restrict the area of the demised land upon which Lucky Realty was obliged to pay rent to the land occupied by Block D only. The language used in the lease is clear. Lucky Realty agreed to pay rent over the whole of the demised land. There is no extrinsic evidence which permits me to arrive at a different conclusion as to the parties' objective intention. That objective intention, now that it has been ascertained by a final judicial determination, will govern the parties' unresolved dealings.

155 For the reasons which I have set out in these grounds, I therefore declare that on a proper construction of the Deed of Variation between OCBC Trustee Limited and Lucky Realty Co Pte Ltd dated 17 December 1996, read with the terms of the Lease between the Public Trustee and Lucky Realty Co Pte Ltd dated 25 February 1975, the revision to the yearly rent every five years is to be calculated by reference to the whole area of Lot 5245N Mukim 27 (formerly Lot 3041 Mukim 27) ("the Land") and not a part of the Land only.

156 Lucky Realty also prays for an order to the following effect:

An Order that ... for the remainder of the term of the Lease, every revision to the yearly rent be based on the proper construction of the Deed of Variation as set out ... above. 


This relief is, in the absence of any counterclaim by the defendant in these proceedings, the direct legal consequence of the declaration which I have already made. I therefore decline to make an additional order to this effect; it is unnecessary.

157 The Trustee having succeeded in these proceedings, it is entitled to its costs. I therefore order that Lucky Realty shall pay the costs of and incidental to these proceedings to the Trustee, such costs to be taxed if not agreed.

1Agatha Chee Bee Hong's affidavit dated 28 April 2014 ("AC's 1st affidavit") at para 1.
2Ong Sook Mor's affidavit dated 19 May 2014 ("OSM's affidavit") at para 2.
3Defendant's submissions, para 5. 
4OSM's affidavit at paras 13 to 14.
5AC's 1st affidavit at para 7.
6AC's 1st affidavit, Tab AC-2.
7OSM's affidavit at para 8.
8OSM's affidavit at para 9.
9OSM's affidavit at paras 9 to 11.
10OSM's affidavit at para 8.
11OSM's affidavit at para 12.
12OSM's affidavit at para 15.
13OSM's affidavit at p 57.
14OSM's affidavit at p 52.
15OSM's affidavit at para 16 and p 61.
16OSM's affidavit at p 72.
17AC's 1st affidavit, Tab AC-3.
18OSM's affidavit at p 72.
19OSM's affidavit on p 105.
20OSM's affidavit at p 107.
21OSM's affidavit at paras 31, 33, 35.
22OSM's affidavit at p 109.
23OSM's affidavit at pp 111, 113.
24OSM's affidavit at p 115.
25OSM's affidavit at p 110.
26Wong Loo Kuan Lydia's affidavit dated 15 May 2014 ("LW's 1st affidavit") at para 22 and p 29.
27Defendant's skeletal submissions dated 18 July 2014 ("DSS") at para 28.
28Zurich at [5].
29Zurich at [26].
30Zurich at [29].
31Zurich at [19] to [21].
32Zurich at [30].
33Zurich at [30(d)].
34Zurich at [30(e)].
35Zurich at [150].
36Zurich at [172].
37Zurich at [126] quoting Lord Steyn in Deutsche Genossenschaftsbank v Burnhope [1996] 1 Lloyd's Rep 113 at 122.
38Zurich at [1].
39Zurich at [125] to [126].
40Zurich at [150].
41Zurich at [134].
42Zurich at [130], [166] to [171].
43Zurich at [135] and [139].
44Zurich at [57], [125] and [130].
45Zurich at [53].
46Zurich at [124] outlining the "second issue" which is the subject of [130].
47Zurich at [53].
48AC's 1st affidavit, Tabs AC-2 and AC-3.
49DSS at para 28.
50DSS at para 18(j).
51OSM's affidavit at para 2.
52Agatha Chee Bee Hong's affidavit dated 24 June 2014 ("AC's 2nd affidavit") at pp 41to 43.
53Wong Loo Kuan Lydia's affidavit dated 15 July 2014 ("LW's 2nd affidavit") at para 17, OSM's affidavit at pp 68 to 69.
54OSM's affidavit at page 65.
55DSS at para 60.
56OSM's affidavit at para 12.
57OSM's affidavit at pp 87 to 90.
58OSM's affidavit at p 79.
59OSM's affidavit at p 95.
60OSM's affidavit at p 87.
61Defendant's reply submissions dated 25 July 2014 ("DRS") at para 7.
62DSS at para 75.

Referring Principles
Trans-Lex Principle: IV.5.2 - Context-oriented interpretation
A project of CENTRAL, University of Cologne.