1 This case concerned the continued liability of a guarantor where the main contract had been varied. The outcome turned on whether a term of the guarantee covered the variations in question, or alternatively, whether the guarantor had given her consent to the variations by way of a separate letter. I found against the guarantor as I was satisfied that the guarantee contained a clause permitting the variations. In any event, the guarantor had indeed signed the letter giving her consent.

2 The plaintiff, an insurance company, provided a credit facility to the first defendant, a construction company called Capstone Construction Pte Ltd (“Capstone”), under which the plaintiff agreed to issue bonds or guarantees in respect of construction work done by Capstone. Personal guarantees were given by the remaining defendants to secure this facility. The second defendant, Suardi @ Chew Seng Nan (“the 2nd Defendant”) and the third defendant, Yew San Ho (“the 3rd Defendant”), were directors of and shareholders of Capstone. The fourth defendant, Priscilla Kua Bee Guat (“the 4th Defendant”) is married to the 3rd Defendant. She is a homemaker and did not appear to be involved in the affairs of Capstone.

3 The present trial only involved the plaintiff and the 4th Defendant. The other defendants did not participate.

4 In 2011, pursuant to the credit facility, a performance bond was issued by the plaintiff in favour of another construction company, Expand Construction Pte Ltd (“Expand”), who was the main contractor for a building project at Punggol West. Capstone was a sub-contractor for this project. This performance bond included, as required by the credit facility, a proviso to the effect that no demand was to be made under that bond unless a similar demand was made by the Housing and Development Board (“the HDB”) against the main contractor. Expand was not, however, agreeable to this proviso. Capstone then requested the plaintiff to issue the performance bond without that proviso. This was done through a performance bond numbered 4073-12-201101360, dated 9 December 2011.

5 In May 2012, a different project was awarded to Capstone. This project concerned the demolition of an existing hawker centre and the erection of a new hawker centre and town plaza at Bedok. For this, Capstone needed guarantee facilities. Subsequently in June 2012, the plaintiff, through an additional agreement (“the Supplemental Agreement”), increased the facility to $4.6m. This increased facility was to be guaranteed personally by the 2nd to 4th Defendants. While letters of acceptance of the variations and guarantee of all sums owing by the Capstone were signed by the 2nd and 3rd Defendants, a dispute arose as to whether the 4th Defendant had similarly signed the letter from the plaintiff dated 15 June 2011 thus consenting to the variations (“the Consent Letter”).

6 Under this increased guarantee facility, the plaintiff issued a performance bond, numbered 4073-12-201201285 and dated 22 June 2012, to the HDB for $1,242,506 in respect of work done at Bedok.

7 Capstone came to grief and was wound up on 16 August 2013.

8 At trial, issues arose about evidence not brought into court. The 4th Defendant sent in the signature on the Consent Letter for the opinion of a handwriting analyst at the Health Sciences Authority (“the HSA”). The report prepared by the analyst from HSA, Yap Bei Sing, stated that the signature on the Consent Letter was probably that of the 4th Defendant. The 4th Defendant did not, however, call the HSA expert to give testimony. Neither did the plaintiff. However, that report was included in the Agreed Bundle. Additionally, the person who was supposed to have witnessed the 4th Defendant’s signature on the Consent Letter and signed the Consent Letter as a witness, one Ms Soh Chow Ping (“Ms Soh”), was also not called. Her answer to interrogatories, denying that the 4th Defendant signed, was however included in the Agreed Bundle. In this respect, another handwriting expert who was called by the plaintiff did testify that Ms Soh had probably signed on the Consent Letter.

9 Bearing in mind these evidential issues, I invited further submissions after receiving the closing submissions from the parties on the following matters:

(a) What is the effect of the inclusion of the HSA report by Yap Bei Sing in the Agreed Bundle?

(b) How should such inclusion be construed in light of the decision of the Court of Appeal in Jet Holding Ltd and others v Cooper Cameron (Singapore) Pte Ltd and another and other appeals [2006] 3 SLR(R) 769 (“Jet Holding”) that the truth of the contents would still need to be proven?

(c) How would the position in Jet Holding apply to an Expert Report or other opinion evidence?

(d) Can the inclusion of a document in the agreed bundle be the basis of the drawing of an adverse inference under s 116 of the Evidence Act (Cap 97, 1997 Rev Ed) against the Party that should have called the maker? If so, in what circumstances can such an adverse inference be drawn?

10 The plaintiff argued that the evidence showed that the 4th Defendant had signed the Consent Letter. The 4th Defendant herself had in fact submitted the Consent Letter for analysis by an analyst from the HSA who concluded that she had signed the Consent Letter. However, the 4th Defendant did not call that expert and an adverse inference should be drawn against her for this failure. The 4th Defendant also did not call Ms Soh, who was the witness to her signature on the Consent Letter, to testify in her defence. There was also the testimony of Kelvin Toh, who was an employee of Times Insurance Brokers Pte Ltd (“Times”), the insurance broker acting in these transactions, that after the Supplemental Agreement and the Consent Letters had been prepared, he collected them from the plaintiff and delivered them to Capstone for them to arrange execution. He also testified that he had chased for the return of the documents over the telephone and was subsequently told that these documents had been signed. The documents were subsequently collected by the brokers.

11 Alternatively, the plaintiff submitted that even if the 4th defendant did not sign the Consent Letter, the variation of the facility through the Supplemental Agreement and the removal of the proviso did not discharge her from liability under the personal guarantee executed by her on 5 December 2011 (“the Personal Guarantee”). Clause 7 of the Personal Guarantee permitted such variations. The contra proferentem rule did not operate in this instance since there was no ambiguity.

12 As to the questions raised by the court, the plaintiff submitted that the HSA expert report was part of the evidence, referring primarily to Goh Ya Tian v Tan Song Gou and others [1981-1982] SLR(R) 193 (“Goh Ya Tian”) at first instance; Tan Song Gou v Goh Ya Tian [1982-1983] SLR(R) 584, as well as Press Automation Technology Pte Ltd v Trans-Link Exhibition Forwarding Pte Ltd [2003] 1 SLR(R) 712 (“Press Automation”). Jet Holding was only authority for the proposition that the court was not bound to accept the contents as true. In any event, even if Jet Holding held that the truth of the contents had to be proved, in the present case, as the HSA report was the 4th Defendant’s own document, an adverse inference should be drawn from the absence of the expert, citing Cheong Ghim Fah and another v Murgian s/o Rangasamy [2004] 1 SLR(R) 628 (“Cheong Gim Fah”).

13 The 4th Defendant contended that she had been asked by the 3rd Defendant to give the Personal Guarantee because the 3rd Defendant, her husband, was not a Singaporean at the time of the facility agreement. Subsequently, when the 3rd Defendant acquired citizenship, there was less of a reason for her to be involved in the transactions.

14 No consent was given by the 4th Defendant for the removal of the proviso, and the increase in the amount of the facility which she guaranteed. Accordingly, the 4th Defendant could not be liable for more than the original facility of $3,511,261.50. The contractual clauses could not make the 4th Defendant liable to any increase in the amount of the facility that may have been agreed between the plaintiff and Capstone. The plaintiff could only succeed if it showed that she signed the Consent Letter. The burden lay on the plaintiff, and they did not call anyone to show that the signature on the Consent Letter was the 4th Defendant’s. They also failed to call the witness to the signature. Calling an expert to show that the signatory witness had signed was not sufficient. The other evidence relied upon by the plaintiff, such as Kelvin Toh’s evidence, was inadmissible as it was hearsay; Kelvin Toh had not seen the signing or collected the documents himself. It was further argued by the 4th defendant that adverse inferences should not be drawn in the circumstances, particularly as evidence was not withheld from the court, citing Yeo Choon Huat v Public Prosecutor [1997] 3 SLR(R) 450 (“Yeo Choon Huat”).

15 In the circumstances, the 4th Defendant was discharged from her guarantee as there had been a material variation, following the rule in Holme v Brunskill (1873) 3 QBD 495 (“Holme v Brunskill”). She was prejudiced by the variations. Clause 7 of the Personal Guarantee did not operate to permit the variations. The contra proferentem rule also operated against the plaintiff. Clause 13 of the Appendix to the Guarantee Facility Agreement (ie for the performance bond itself) also required the plaintiff to give notice of the variation.

16 As to the questions posed by the Court, Jet Holding applied, which meant that documents in the Agreed Bundle were only agreed as to authenticity. There was no distinction between expert evidence and other forms of evidence. Press Automation only stood for the proposition that the inclusion of the documents did not excuse proof of such documents. The Court of Appeal disapproved of Goh Ya Tian in Jet Holding.

17 The Personal Guarantee between the plaintiff and 4th Defendant contemplated variations to the agreement between the plaintiff and Capstone. That being so, the 4th Defendant remained bound even after the increase in the amount covered by the performance guarantee and the removal of the proviso. Neither was the plaintiff required to serve written notice on Capstone of the variation before it became effective. Additionally, the evidence showed that, on the balance of probabilities, the 4th Defendant did sign the Consent Letter agreeing to these variations.

18 The Personal Guarantee, to which the 4th Defendant was a party, contemplated through cl 7, that there could be variations in the agreement between the plaintiff and Capstone. The 4th Defendant had thus contracted out of the general rule that a variation of the main agreement would discharge her from liability under the guarantee.

19 Clause 7 was apparently drafted by a commercial drafter imbued with the usual traditional abhorrence of commas and of a reader-friendly layout. It read:

The material portion of cl 7 read:

The effect of cl 7, on a plain reading was to ensure that the obligations of the guarantor remained even if there were any changes to the contract between the insurance company and Capstone. It must be stressed that cl 7 was broad – it covered any term and any change. Contrary to what was argued by the 4th Defendant, ie, that cl 7 was limited to issues of enforcement only, on a plain reading, the clause was not on the face of it limited to rights of enforcement, or specific types of clauses. The clause was broad, but there was nothing wrong with breadth if that was indeed the bargain struck between the guarantors, including the 4th Defendant, and the plaintiff.

20 Clause 7 was clearly there to take the contract out of the rule in Holme v Brunskill, which held that in a guarantee a material variation of the principal contract not consented to by the guarantor would discharge the latter. Holme v Brunskill has been followed in many local decisions, including American Home Assurance Co v Hong Lam Marine Pte Ltd [1999] 2 SLR(R) 992. Certainly, the principle is clear and will be applied where it is relevant. However, parties can contract out of the rule: British Motor Trust Co Ltd v Hyams (1934) 50 TLR 230. This has in fact been done in very many, if not all, contemporary commercial insurance contracts. As noted by the plaintiff, clauses similar to cl 7 have been upheld locally.

21 In Standard Chartered Bank v Neocorp International Ltd [2005] 2 SLR(R) 345, the clause in question provided that the obligations and liabilities of the guarantor would not be “abrogated, prejudiced, affected or discharged … by any increase, amendment, or variation to any of the credit, banking or other accommodation extended to the Customer…” (at [8]).

22 In that case, V K Rajah J noted that similar clauses were generally included in guarantees to address the rule that any material variation without the guarantor’s consent would otherwise release the guarantor. Rajah J found that through that clause the guarantor had expressly agreed that the lender and the borrower could vary their agreements without obtaining the guarantor’s consent. Rajah J stated (at [40]):

23 Rajah J’s opinion encapsulates the general position at law that clauses of this nature will be given full force. If the language is clear, there is nothing to obstruct the continued liability of the guarantor even after a variation is effected. The approach of the courts is simply a reflection of freedom of contract accorded to the parties – if the variation was objectionable, the guarantor should not have agreed to such a clause.

24 Another case considering the effect of a variation clause was SAL Industrial Leasing Pte Ltd v Lin Hwee Guan [1998] 3 SLR(R) 31, which involved a guarantee of a factoring agreement. The variation clause read in part:

The Court of Appeal found no difficulty in finding that this variation clause covered an increase in liability.

25 Similar propositions are found in other cases cited by the plaintiff, showing that the courts in Singapore do give effect to variation clauses: Overseas-Chinese Bank Corporation v Tan Geok Ser and Another [2000] SGHC 263 and Development Bank of Singapore v Yeap Teik Leong and others [1988] 2 SLR(R) 201 (“DBS v Yeap Teik Leong”).

26 The 4th Defendant argued against giving cl 7 a wide reading, submitting that it should be construed contra proferentem, or against the plaintiff; and that on a proper construction, cl 7 was limited to matters of enforcement only. The 4th Defendant’s arguments, invoking the maxim contra proferentem, went too far. As noted in The Interpretation of Contracts (Sweet & Maxwell, 5th Ed, 2011) (“The Interpretation of Contracts”), the operation of this maxim of construction captures three different situations, in which construction would lean against:

(a) the person who prepared the document in question;

(b) the person who prepared the particular clause; or

(c) the person for whose benefit the clause in question is to operate.

But the common thread is ambiguity in the words used, as is captured by the paragraph in The Interpretation of Contracts paraphrasing the maxim:

Thus the maxim may apply to a contract of guarantee, but only when there is ambiguity. When the provision is clear, there is no occasion to require that something be read against one party or the other. The various cases cited by the 4th Defendant as showing the adoption of contra proferentem, such as Cohlan v SH Lock (Australia) Ltd (1987) 8 NSWLR 88, and Bank of Montreal v Korico Enterprises et al (2000) 50 OR (3d) 520, must be understood in this light. They are authoritative only in so far as they deal with ambiguities in the contract before them.

27 The 4th Defendant tried to argue that cl 7 was not all that broad. The focus of the 4th Defendant was on an earlier part which read:

It was argued that variation was thus limited to the specific types of security mentioned in that clause. I could not accept this argument as it ignored the breadth of the next portion, which was clearly not limited to any form of security, and instead referred to “terms or conditions in respect of any transaction with the [the plaintiff] in any manner whatsoever whether under or in connection with the Agreement or otherwise…”. There was nothing to limit its operation only to the specific type of security mentioned in cl 7.

28 There may be a limit to the extent to which variation clauses may operate; one instance would be exemplified by Triodos Bank NV v Dobbs [2005] EWCA Civ 630 (“Triodos”), in which entirely new agreements were entered into between the bank and the debtor. These were found by the English Court of Appeal not to be covered by the relevant variation clause. However, the present case did not come within that factual scenario. The variations in question here, namely the increase in amount guaranteed because of the Supplemental Agreement and the removal of the proviso in the guarantee facility agreement, did not change the essential basis of the guarantee, which was to cover the exposure incurred by Capstone through the guarantee facility agreement. Neither were any of the variations concerned with guarantee facility agreements that were distinct from the original one, unlike the situation in Triodos.

29 The Supplemental Agreement increased the overall exposure of the plaintiff. The 4th Defendant’s Personal Guarantee created an obligation on her part, as a joint and several guarantor of all sums incurred by the plaintiff for payments made in favour of Capstone (see paragraph 1 of the Personal Guarantee). Clause 1 of the Personal Guarantee was sufficiently broad in its proper interpretation to render the 4th Defendant liable for the increase in exposure that Capstone incurred in this case. The liability of the 4th Defendant arose to pay under the Personal Guarantee regardless whether any action was taken against Capstone (see paragraph 5(i) of the Personal Guarantee). The obligation of the guarantors, including the 4th Defendant, was for all sums owed by Capstone to the plaintiff, under or in connection to the Agreement between them; further that obligation was on a continuing basis (see cl 1 and 2 of the Personal Guarantee). The increased obligation of the 4th Defendant was not, in these circumstances, of such a nature as to take the new obligation out of the initial contemplation of the original agreement.

30 There was nothing to prevent the application of the variation clause (cl 7) in this case to the removal of the proviso. The removal of the proviso did not render the guarantee obligation significantly different from what had been originally agreed. The proviso was stipulated in the guarantee facility letter, ie, in the contract between the plaintiff and Capstone. The proviso stated that no demand would be made, and hence no payment due, unless a similar demand was made by the customer of the construction contract against the main contractor. Effectively, this would ensure that the demand would only be made against Capstone when a similar demand was made against the main contractor.

31 The removal of the proviso removed linkage between the liabilities of Capstone and the main contractor. While this may enlarge the situations where Capstone’s performance bond is triggered, it did not follow that this removal rendered the obligations of the personal guarantors wholly different from what was contemplated at the start. It was a question of degree only, and was adequately covered by the broad language of the variation clause. In any event, the variation clause was sufficiently wide to cover the removal of the proviso. The plain words of the clause were not limited in any way. Certainly, there was no distinction drawn between contractual and performance obligations, as the 4th Defendant seemed to suggest. Such a distinction was not supported by authority either. The removal of the proviso was also not a change of such a degree that it should be interpreted as falling outside the ambit of the original contract. The proviso was just one part of the guarantee facility covering obligations owed by Capstone under its building contracts. Those contracts continued in existence, and continued to be covered by the guarantee facility, even after the removal of the proviso. There was thus nothing to lead to the conclusion that there was such a significant change that the main contract supported by the guarantee given by the 4th Defendant was no longer in existence.

32 As noted by the plaintiff, in DBS v Yeap Teik Leong, a variation allowed the use of guaranteed funds for a purpose that was originally prohibited. The variation clause in that case was found by the court to cover such a broad variation. A fortiori, the variation clause here should at least cover the removal of the proviso.

33 In any event, in practical terms, it was highly likely that a call on the main contractor’s own performance bond would have been a question of a matter of time only once circumstances were such that a call on Capstone’s performance bond could be made. In view of this, the removal of the proviso could not be regarded as something that substantially or wholly altered the scope of liability of the personal guarantors, including the 4th Defendant.

34 I also accepted that cl 13 of the Appendix to the Guarantee Facility Agreement (ie for the performance bond itself), was a clause for the benefit of the plaintiff, thus the fact no written notice of variation was given does not hinder the plaintiff’s claim. The 4th Defendant argued that this clause required that variations be notified in writing.

35 Clause 13 read:

The clause as a whole was intended to be for the benefit of the plaintiff – it was concerned with ensuring that the plaintiff was only taken to have waived an obligation in its favour if there was a record in writing. It was possible for the plaintiff to waive this very obligation, without doing so in writing – whether or not it can be easily proved by an obligor in an appropriate case was another matter. Although the clause referred to variation, it was clear in the context of that clause as a whole that the subject that it addressed was waiver: this was what the rest of the clause addressed. Variation in this context therefore was nothing more than an expansion of waiver, probably inserted ex abundante cautela. In addition, clause 13 was a term of the agreement between the plaintiff and Capstone, not the Defendant. Furthermore, even if it should be taken into account in this case, interpreting this clause as imposing a requirement that all variations must be covered by notice by writing went against the broad aspect of cl 7, which operated automatically whenever there was a variation.

36 My conclusion that cl 7 covered the variations in this case such that the 4th Defendant remained liable was sufficient to dispose of the case. The need for the Consent Letter to be signed was thus a requirement ex abundante cautela. However, if I was wrong in finding that cl 7 applied to the situation, I also concluded that the Consent Letter had indeed been signed by the 4th Defendant.

37 While the legal burden of showing that the Consent Letter was signed, fell on the plaintiff, the 4th Defendant could not in the end just rest on a denial that this was her signature, especially when the plaintiff adduced a document purportedly signed by the 4th Defendant as well as other evidence which pointed to her having signed it: the collection of the document by the broker; her inconsistent behaviour in these proceedings and adverse inferences that could be drawn against her. In this context, it behoved her to bring into play evidence in support of her contention: she had to discharge her evidential burden in the face of what was against her. She did little to rebut the evidence arrayed against her, and in the end, weak though each strand may have been on its own, cumulatively, these various pieces of evidence worked together to sufficiently establish the plaintiff’s position against her on the balance of probabilities.

38 Specifically, I found that the evidence as a whole pointed to the 4th Defendant having signed the Consent Letter. There was the evidence of the insurance broker, Kelvin Toh, who testified that the letter was picked up signed. Evidence pointing to signing also came from the failure of the 4th Defendant herself to take an unequivocal stand until late in the day. Additionally, adverse inferences were drawn from the failure of the 4th Defendant to call:

(a) The expert who had prepared a report indicating that the 4th Defendant had signed the Consent Letter. This report was in fact prepared at the 4th Defendant’s behest; and

(b) The person who supposedly witnessed the 4th Defendant’s signing. While this person gave an answer in an interrogatory included as part of the Agreed Bundle that the 4th Defendant had not signed the Consent Letter, this went up against the expert evidence that was received in court that that person had indeed appended her signature as a witness. Against this backdrop, her absence from court led to an inference being drawn against the 4th Defendant. In addition, I found that little weight should be attached to her interrogatory answer.

Thus my conclusion that the 4th Defendant had signed the letter was based on the cumulative effect of the above.

39 The insurance broker who had arranged the facility from the plaintiff gave evidence through PW2, Kelvin Toh. PW2 was a director of the insurance brokers, Times, appointed by the company. As mentioned above, his testimony was that when the Supplemental Agreement and Consent Letters were to be delivered to the plaintiff, these documents, including the Consent Letter from the 4th Defendant, had been signed.

40 The 4th Defendant argued against the acceptance of Kelvin Toh’s evidence, primarily on the basis that it was hearsay, as he had no personal knowledge the collection of documents. He did not collect the documents himself; it was someone else from his firm. It was correct that as regards the truth of the facts stated, namely that the documents were signed at the point of collection, this was not something he had perceived himself. He testified as to what was told to him by others about what they perceived, and thus it was hearsay. However, the fact that this was said to him could be relevant just as much as whether what they said was true. Thus the fact that he was told of certain things would support his testimony that nothing out of the ordinary arose in the collection of documents. Such evidence, that a person was told of things and that nothing went wrong, may be of little weight usually, but in some circumstances, where there are other circumstantial evidence, it may form part of the matrix of circumstantial evidence which, taken as a whole, may be sufficient to discharge the burden of proof.

41 Mere inconsistency in explanations given or positions taken by a witness about a material fact would not necessarily be fatal to the credibility of that witness, and would generally not be a reason for reaching a finding that that fact was proved or not. However, where the explanations in question go against what would normally be expected in a given situation, and no reasons were given justifying the different explanations, this could be a basis to infer that the fact was proved or not, as the case may be, when the inference is made against a backdrop of other grounds.

42 In the present case, as submitted by the plaintiff, the 4th Defendant had taken various positions as to whether or not she had signed the Consent Letter. The following were outlined by the plaintiff:

(a) in her affidavit of evidence in chief (“AEIC”), the 4th Defendant did not deny categorically that the signature on the Consent Letter was hers, or that someone must have forged it. What she stated was that she believed she did not sign as she did not recall doing so;

(b) when she sought to set aside a statutory demand issued against her, the 4th Defendant maintained that she did not recall signing the document, and did not think the signature was hers; and

(c) it was in the summary judgment proceedings in this case that the 4th Defendant first raised the possibility that the signature was a forgery. The plaintiff argued that this was because the Assistant Registrar had indicated that only conditional leave to defend would otherwise be granted if there was no allegation of forgery.

43 The 4th Defendant argued that she had adequately denied that the signature was hers. It was contended that the 4th Defendant had sufficiently explained that she had said what she did because this was her manner of speaking – that when she said that she did not recall, it really meant that she did not do it.

44 I did accept that persons might express themselves in indirect ways, and not categorically deny something outright. However, the 4th Defendant’s position and statements in these and related proceedings would be at odds with what would be expected had she truly not signed the document. It would not be sufficient on its own to merit a finding against her, but it would legitimately constitute part of a series of facts which would have to be weighed against her.

45 Adverse inferences could be drawn from a number of facts in this case. They were:

(a) The inferences from the failure of the 4th Defendant to call the expert who produced a report indicating that the 4th Defendant probably signed the consent letter; and

(b) The evidence that the witness to the signing of the consent letter had herself signed that document.

46 The inferences that were drawn were those contemplated by s 116 of the Evidence Act (Cap 97, 1997 Rev Ed).

47 Section 116 of the Evidence Act read:

Section 116 and illustration (g) is part of a section dealing with inferences of fact.

48 At common law, an adverse inference may be drawn against a party failing to call a witness to testify on an issue: Phipson on Evidence (Sweet & Maxwell, 18th Ed, 2013), at para 11–15, referring to cases, including Wisniewski v Central Manchester Health Authority [1998] 5 PIQR P324 (“Wisniewski”). V K Rajah JC in Cheong Ghim Fah at [42] cited the following propositions from Wisniewski:

Rajah JC was of the view that these should apply to absentee witnesses under illustration (g) of s 116 of the Evidence Act. In addition, Rajah JC added that the reasons should be put during cross-examination, to allow an explanation to be given for the absence.

49 I would respectfully agree with Rajah JC’s approach, save though I do not think it is necessary for the reasons or lack of reasons to be put in all cases. Much depends on the context; whereas here it is evident that the absence was the result of a conscious decision not to call a witness, then putting that very issue was not essential. The party in question would be aware of the issue and would be expected to have chosen to explain or not to explain.

50 The 4th Defendant tried to argue against the application of adverse inference, citing Yeo Choon Huat for the proposition that the failure to call a witness did not trigger the application of illustration (g) of s 116 if the other party could have called that witness itself. It is important to bear in mind the context of that decision. That was a criminal case in which the prosecution (among other allegations) did not call as a witness, one Koh, who was initially a co-accused alongside the appellant there. The appellant was charged with trafficking drugs to Koh. The trial judge declined to have a joint trial, and the charges against Koh were stood down. In that case, the prosecution’s case was not at all dependent on Koh, or any inference dependent on Koh’s absence or presence. Thus, in that situation, the question of whether evidence has been withheld or suppressed would be material as an additional consideration whether illustration (g) of s 116 was triggered. Where however the evidence is an essential part of the case for one side, its absence can trigger illustration (g) of s 116; that was the underlying approach of Cheong Ghim Fah. In any event, it is clear that Yeo Choon Huat actually concerned an allegation of withholding evidence by the prosecution, a matter which went beyond s 116 and touched on the scope of the ethical duty of the prosecution as ministers or guardians of law to adduce all relevant evidence, though with respect, the Court of Appeal in that case couched it otherwise. The holding in Yeo Choon Huat did not readily translate to a civil setting, where neither side is in the same position as the prosecution.

51 In the present case, the signing of the document was an essential part of the plaintiff’s case. The 4th Defendant’s denial that she had signed, and expert evidence as to whether she did append the signature that was on the document, were evidence to be raised by the 4th Defendant and thus essential parts of her defence. Similar inferences may be drawn in respect of the absence of the witness to the signing of the document.

52 I found that an inference should be drawn against the 4th Defendant in respect of whether she signed the Consent Letter. No expert evidence was adduced by either side at trial on whether the signature on the consent was indeed the 4th Defendant’s, though there was an expert’s report prepared for the 4th Defendant. The parties each chose to leave it to the other side to call the expert who had produced the report on the 4th Defendant’s signature.

53 That report was part of the Agreed Bundle. I invited and received further arguments on that point. As noted in the Court of Appeal decision of Jet Holding, the inclusion of a document in the Agreed Bundle only dispenses with formal proof; at the least, the truth of the contents needs to be proven (at [44]). In the present case, the parties had expressly agreed only to the authenticity of the documents. In the context of opinion evidence, as in a report, I am of the view that this principle applies such that, in the absence of anything else, the inclusion of such a report is only acceptance by the other side of the authenticity, and that it does not lead to the automatic inclusion of such a report as opinion evidence to which weight ought to be given by the court. In the circumstances, the presence of the report could not be used to determine the signing of the Consent Letter.

54 The 4th Defendant did not call the expert who had produced a report on the 4th Defendant’s signature. Instead, the expert it called, PW1, Ms Yang Chiew Yung, from a forensics agency, testified as to her opinion on the signatures of the witness, Ms Soh, to the 4th Defendant’s own signature on the Consent Letter. However, even for this expert witness, the focus of questioning ultimately was less about Ms Soh’s signature, but more about the 4th Defendant’s. She was asked for evidence about the terms used to indicate confidence on similarity of signature. Based on this evidence, the plaintiff argued that the report about the 4th Defendant’s signature showed that it was likely that the 4th Defendant had indeed signed the Consent Letter.

55 As I have noted above, that report about the 4th Defendant’s signature could not be used in that way. Be that as it may, the failure to call the 4th Defendant’s expert gave rise to an adverse inference. As stipulated by Cheong Ghim Fah, the question was put to her in the course of cross-examination, in the context of the HSA report:

The relevance of the expert report was apparent and clear. It pointed to the 4th Defendant being the one who signed the Consent Letter. Furthermore, it was commissioned by her. While the plaintiff could have approached that expert directly in view of the posture of the 4th Defendant, the failure of the 4th Defendant to call that expert witness to court led to the inference that that evidence would have been unfavourable to her. The possible sub-inferences included the following:

(a) calling the expert would allow the expert report to come into evidence, implicating the 4th Defendant;

(b) calling the expert would show up the 4th Defendant’s denial that she signed the consent form; and

(c) calling the expert may allow the expert’s testimony in court to reinforce the contents of the report and the credibility of the expert.

The position of counsel faced with an adverse expert report is not to be envied. But counsel cannot avoid the repercussions of such a report by declining to call the expert in question to court and leaving it to the other side to consider bringing in that evidence. Unless such evidence is part of the positive case for the other side, there is no obligation on the other side to call that witness, and it would be open, as was the case here, for the other side to submit that adverse inferences should be drawn against the party that failed to bring in the adverse witness and evidence.

56 The witness to the signature on the Consent Letter was not called as a witness at trial. The 4th Defendant relied on answers to interrogatories that Ms Soh did not witness the signing of the Consent Letter. It was contended that the answer was evidence as it was in the Agreed Bundle. However, as in the case of the expert report on the 4th Defendant’s signature, though the answer to interrogatories was part of the Agreed Bundle, in this case there was only agreement as to authenticity. It was thus not in evidence. In any event, even if it were, the weight to be accorded to it is a matter for the court to determine; in such a case in the absence of testimony in court and testing by cross-examination, I was of the view that negligible weight should be given to her denial. Additionally, all of this had to be weighed against the evidence of the expert, PW1, who gave evidence on the stand and in her report that Ms Soh had indeed signed on the Consent Letter.

57 In any event, in the present circumstances, given the expert evidence adduced by the plaintiff, an explanation or a contrary testimony from the witness would have been expected. And it would have been for the 4th Defendant to call her, since the 4th Defendant’s case was a denial of both her signature as well as the witness’.

58 The inference that would be drawn from the failure to call Ms Soh in the face of the strong evidence that Ms Soh had signed on the Consent Letter was that Ms Soh’s evidence would be detrimental to the 4th Defendant, and would contradict her interrogatory answer.

59 The overall impact of the various matters considered above gave strong credence to the contention of the plaintiff that the Consent Letter was indeed signed. Her denial and what explanation there was for her behaviour were insufficient to act against the conclusion, on inherent probabilities, that she had indeed signed that letter. The differences could not to my mind be the product of mere lapses in memory, vagueness, or a desire to hold back pending the crystallisation of issues in a trial.

60 Though the HSA report was not in evidence, its existence, and the clear fact that it was contrary to the 4th Defendant’s case have to be taken into account. The hostile report was not used, in itself, to find that the 4th Defendant signed the form, but that the failure, tantamount to a refusal to bring it into evidence, led to an inference against the 4th Defendant that she had something to hide and therefore that she was not telling the truth. The inference to be drawn becomes stronger in its effect when the adverse nature of the omitted evidence is apparent. Another way of putting this is that it becomes much more readily inferable that the reason why a witness is not called, namely the expert on the signature the 4th Defendant, is that the testimony will be clearly hostile or adverse. Similarly, the fact that Ms Soh was not called in the circumstances of the case, and in the face of the adverse report against her, raises the ready inference that her testimony would not be favourable. The impact of that inference may be lessened by other circumstances; but here, the fact that Ms Soh had given an answer to an interrogatory could not assist very much. Her evidence here could, for the reasons I have given, only be accorded minimal weight.

61 Counsel for the 4th Defendant was correct that the legal burden was on the plaintiff to show that the 4th Defendant had indeed given her consent to the Supplemental Agreement. However, the standard of proof in civil cases is that of the balance of probabilities, and I found, in light of the evidence that was before me, that it was more probable that not that the 4th Defendant had indeed signed that document. The evidence was largely circumstantial but it was sufficiently strong for a finding against her.

62 The cumulative and combined effect of the evidence from the insurance broker that there was nothing untoward in the collection of the forms, the muted position of the 4th Defendant herself as to whether she had signed, and the inferences drawn from her failure to bring in the expert’s report, and obtain Ms Soh’s testimony in light of the other expert’s findings, was that she had indeed signed that Consent Letter. Certainly, any one of these on their own may not have been sufficient to lead to that conclusion. And even together, they may not be enough to establish in a criminal case against all reasonable doubt that she indeed signed that document. However, what was needed in this case was the preponderance of evidence such that it was more likely than not that she had indeed signed it.

63 The 4th Defendant tried to argue that there was a rational reason that made it unlikely that she signed the Consent Letter. She gave evidence that it was unnecessary for her to give the guarantee, so this pointed against her signing the Consent Letter. She alleged that she was only brought in to give a guarantee because her husband could not do so as he was not a Singaporean citizen at the time. However, by the time the Consent Letter was to be signed, he had become a citizen; the implication being her involvement was not needed. I could not accept this as an adequate explanation that would lead to the conclusion against all the inferences and other evidence. If this was indeed a reason, the probable course of action would have been for her to discharge herself completely from the guarantee she had given.

64 All in all, I had to reject her evidence, and I found, bearing in mind the evidence from the plaintiff’s side, as well as the fact that the Consent Letter had a signature in the name of the 4th Defendant, and the inferences and circumstantial evidence, that the probabilities showed that she did sign the Consent Letter.

65 In the course of submissions a number of points were put forward by the plaintiff which I could not accept. First, the fact that other defendants had not taken similar points could not be taken against her. That their position may be factually different also could not support any inferences against her in this case. Such inferences would be too remote and there were too many possible counter explanations. Second, the fact that the 4th Defendant’s husband was not called to support her case could not be lead to an inference against her. There again could have been many reasons, including innocent ones, why he was not called.

66 The 4th Defendant at various points raised the requirement that the guarantee had to be in writing. In the event, given my findings, the 4th Defendant was bound by an agreement in writing, including the variations, as these would have been covered by the original guarantee given by her, or by the Consent Letter signed by her.

67 I noted that it was not part of the 4th Defendant’s case that there was any undue influence or unconscionability, or any other vitiating factor, in her signing of the Personal Guarantee.

68 While the 4th Defendant found herself in a difficult position because of the Personal Guarantee given by her, the court needed to give effect to the contract freely entered into. Too loose an approach would only increase the transaction costs and hinder commerce, which relies on guarantees to give the necessary assurance to counterparties, especially in complex or long drawn out dealings.

69 I accordingly found that judgment should be given in favour of the plaintiff against the 4th Defendant, jointly and severally with the other Defendants, in respect of the plaintiff’s claims and interest. Costs on an indemnity basis were also awarded to the plaintiff, following the terms of the contract and the tendered schedule.

70 The 4th Defendant appealed against my decision. In the circumstances of the case, I granted stay of execution pending appeal.