1 Lee Kwang Peng (`the appellant`) was committed for trial on nine charges of using criminal force against a person intending or knowing it to be likely that he would thereby outrage the modesty of that person under s 354 of the Penal Code. There were three complainants who each brought three charges against the appellant. Before the commencement of trial, the prosecution stood down three of the nine charges, all in respect of the same complainant. The trial therefore proceeded on six charges of offences committed against two persons, the first complainant or PW1 and the second complainant or PW2 respectively. The third alleged victim (the third complainant or PW3), gave evidence at trial, but not in respect of the alleged offences committed against him. The complainants were aged between 12 and 14 at the time of the offences, which spanned a period of around two years.

2 At the close of the prosecution`s case, it was apparent that the charges framed against the appellant would have to be amended as certain particulars contained in the charges did not correspond with evidence given by the prosecution`s witnesses. The district judge therefore allowed the prosecution to amend the charges against the appellant.

3 The appellant was convicted on all six amended charges and sentenced to four months` imprisonment for each offence, with three of the sentences to run consecutively and the remainder to run concurrently, making a total period of imprisonment of 12 months. No caning was ordered.

4 The appellant appealed against his conviction and sentence. The public prosecutor filed a cross-appeal against sentence.

5 Facts

6 The appellant was a male taekwondo instructor and was 33 years old at the date of the trial. He did not continue his studies beyond primary school but had some vocational training. He took up taekwondo in 1978 and obtained his instructor`s licence in 1984. He has since taught at various institutions, including the Mountbatten Community Centre (Mountbatten CC).

7 All three complainants were taught by the appellant at the Mountbatten CC. All three boys were also Secondary 3 students at Tanjong Katong Secondary School at the time of the trial.

8 The three boys were introduced to the appellant by one Lin Jing Ze (hereafter referred to as Jing Ze or DW8), who had been at one time or another classmates with all three boys. Jing Ze had undergone taekwondo training under the appellant at the Mountbatten CC since 1985. Jing Ze testified that he introduced the second complainant to the appellant in February 1994 and introduced the first and third complainants to the appellant in May 1995 as they were considering taking up taekwondo lessons under the appellant. In respect of the first and third complainant, the introduction took place at the Mountbatten CC. According to Jing Ze, before registering new students for taekwondo class, it was customary for them to attend an informal `trial training session`. It was at this `trial training session`, which both the first complainant and the third complainant attended, that they first met the appellant. Jing Ze stated at trial that this took place on either 21 or 25 May 1995.

9 The first complainant`s allegations against appellant

10 The first complainant alleged that he was molested by the appellant on three separate occasions.

11 The first incident involving the first complainant

12 The original charge in respect of the first incident involving PW1 and the appellant read as follows:

13 The first complainant decided after the `trial training session` to register and sign on for the taekwondo course. He felt awkward about commencing training sessions without the taekwondo outfit or `gi`. The appellant thus arranged for the first complainant to collect the `gi` from the former`s business office at Gay World Complex on a Saturday. The first complainant alleged that, at the office, the appellant required him to try on the `gi` for size. The appellant then wanted to check the first complainant`s stomach muscles. The first complainant allowed him to do so. The appellant then asked him if he knew how to masturbate, pulled down the first complainant`s pants and fondled his penis. After this, the appellant called for food and drinks to be delivered to the office, a seemingly minor detail which turned out to be critical in the trial judge`s assessment. The first complainant said in evidence that he ate the food for fear of being physically harmed by the appellant. After this, he left for home on his own. The first complainant claimed that he reported this incident to the third complainant a short while after the event.

14 The first complainant alleged that this took place shortly after having been introduced to the appellant. In his evidence, he also claimed that the first incident took place in February 1995. However, when shown his application form for taekwondo classes at the Mountbatten CC dated 18 June 1995, he admitted that he could not have met the appellant earlier than May 1995. The charge against the accused was therefore amended to read as follows:

15 The second incident involving the first complainant

16 The first complainant said at trial that, notwithstanding this incident, he continued with taekwondo lessons at Mountbatten CC because of his interest in the sport and also because his friends had taken it up. He claimed, however, only to have attended classes when he was sure the third complainant would be present. It was in this context that the second incident took place. The charge against the appellant in relation to this second incident originally read as follows:

17 The first complainant gave evidence that this took place `a few months after the first incident.` However, in light of his admission that he had only joined the taekwondo class in May or June 1995, the date reflected in the charge was inaccurate and the charge was amended to read:

18 According to the first complainant, the appellant instructed him to collect training equipment from the store room. This was a task normally assigned to one Mai De Kang (Mai) or (DW11) who had been appointed as the class quartermaster by the appellant. It was Mai`s job to obtain the store room key and to fetch the equipment for the class. He was occasionally assisted by other students. The first complainant testified that he protested that fetching the equipment was Mai`s job. The appellant insisted that he do it and threatened him that there would be consequences if he did not. Mai went with the first complainant to take the key as the first complainant did not know where it was. Mai then left the first complainant to retrieve the equipment himself. The first complainant testified that the appellant confronted him in the store room and told him to `Whun wo tong see`, which translated from Mandarin means `Return my things.` At trial the first complainant explained that this was the appellant`s euphemism for expressing his desire to masturbate the first complainant or to fondle his genitals. The appellant then closed the store room door, pulled down the first complainant`s pants and fondled his penis. The first complainant tried to, in his own words, `procrastinate in returning his things` by telling the appellant he would do it next time. The appellant desisted when the first complainant did not get an erection. The first complainant then fetched out the equipment and continued with the training lesson. He did not tell anyone what had transpired until very much later.

19 The third incident involving the first complainant

20 After the second incident, the first complainant continued with taekwondo lessons, but only sporadically. He did this to avoid the appellant. However, in December 1995, he was again molested by the appellant. The original charge against the appellant read as follows:

21 The incident took place during a taekwondo camp held at the People`s Association Holiday Camp in Pasir Ris. The first complainant alleged that on the second day of the camp he was washing dishes outside the male toilet. After doing so, he went into the toilet to turn off the tap. Leaving the toilet, he ran into the appellant, who told him he wanted him to `return his things` and commanded him to go into the toilet. He then told the first complainant to go into the shower room, where the appellant pulled down the first complainant`s bermudas. The charge framed by the investigating officer initially alleged that the appellant had caressed the first complainant`s testicles. The first complainant, however, gave evidence that the appellant had fondled his penis. When the first complainant objected to the appellant`s unwanted attention, the appellant threatened to spread rumours that he had a small penis. The molestation then went on for a few minutes before the appellant ceased, because he was unable to arouse an erection in the first complainant. The first complainant claimed to have reported this incident to the third complainant. In light of the first complainant`s evidence that the appellant had fondled his penis (as opposed to having caressed his testicles), the charge was amended to read:

22 The second complainant`s allegations against appellant

23 The second complainant also alleged three incidents of molest committed by the appellant.

24 The first incident involving the second complainant

25 The first incident took place after a training session at Mountbatten CC. The second complainant was not certain when exactly, but stated it was `after I signed the form but after a few weeks.` By `the form` the appellant was referring to his application form for taekwondo lessons. The charge against the accused initially alleged that the incident took place sometime in 1993:

26 However, the second complainant admitted in cross-examination that he had only submitted his application to take up classes on 19 February 1994. The incident therefore had to have taken place in February or March 1994 and not 1993 as the charge had stated. The charge was therefore amended as follows:

27 The second complainant stated that the incident took place after a training session. During a sparring match with one Xie Chang Yong (Chang Yong or DW6), the second complainant was accidentally kicked in the groin. He fell to his knees in pain. After instructing him on how to help the blood to circulate, the appellant continued with the lesson. After the lesson, the second complainant went to the toilet to change. It was here that the appellant approached him and asked if he was all right. He told the second complainant to take off his pants so he could check. He pulled his pants down. The appellant then checked the groin area and massaged his testicles for three to four minutes. The second complainant said he consented because the appellant was his instructor and he thought that he knew what he was doing. He then put on his pants and his father came to fetch him. He told his father of the injury he sustained but not of the events afterwards.

28 The second incident involving the second complainant

29 According to the second complainant, the second incident took place a few weeks after the first. The charge initially stated that this took place in January 1994. However, in light of the second complainant`s evidence in relation to the first incident, it appeared that the date was probably between February and April 1994. The charge which originally read:

30 was therefore amended to read:

31 On this occasion, the second complainant was putting equipment back in the store room. The appellant entered and closed the door and locked it. He then tried to pull down the second complainant`s pants saying he wanted to check his testicles and his lower abdominal muscles to see if they were hard enough to sustain another kick. The second complainant pulled his pants down himself. The appellant checked his abdominal muscles and then moved to his testicles and massaged them for three to four minutes. For the same reason as the previous incident, the second complainant did not think much of it.

32 The third incident involving the second complainant

33 The third incident took place early in 1996. The appellant was charged as follows:

34 At the close of its` case, the prosecution amended the charge in minor respects to read as follows:

35 At the time of the third incident, the community centre was under renovation. The equipment was therefore kept not in a store room but in a large metal container not unlike temporary mobile offices commonly used at construction sites. The second complainant claims he was again putting training equipment back after a lesson. Although he was helped by others, he was the last one in the container. The appellant approached him and said he wanted to check his stomach muscles. The second complainant consented. The appellant first examined his lower abdomen then asked about his injury. The appellant then checked his testicles, first looking at them and then massaging them for three to four minutes. The second complainant gave evidence that he felt `weird` as this was so long after the training accident in early 1994. However he did not object or ask questions as he felt that the appellant was his instructor and he trusted him. After this he put on his pants and went out of the container.

36 The appellant denied all six charges. In relation to the first incident involving the second complainant, he admitted having examined the second complainant`s genitals, but stated that this was purely out of concern that he might have been injured after the kick he received. He claimed it was a brief examination for a genuine medical purpose. In relation to the remaining charges, the appellant put forth a total denial. He was convicted on all six charges.

37 The appeal

38 The appellant raised numerous arguments in his appeal, which might be classified broadly into procedural and evidential issues which I now summarise for convenience.

39 Procedural issues

40 The thrust of the appellant`s case was that the district judge had not properly directed herself on the issue of corroboration. The appellant contended that the district judge wrongly found the complainants` evidence to be `unusually convincing` and that she should have not convicted the appellant in the absence of corroboration. Counsel contended that it was unsafe to convict the appellant where the only evidence against him was the uncorroborated evidence of child complainants.

41 The second point canvassed by counsel for the appellant was whether the trial judge admitted prejudicial evidence, namely, the evidence of PW1, PW2 and PW3 supporting the charges of offences committed against the third complainant. As the charges against the third complainant had been stood down before trial, such evidence was irrelevant and prejudicial.

42 The appellant`s third contention was that the district judge erred in allowing a single trial for charges based on complaints brought by two separate persons and that there should have been separate trials for each of the alleged victims.

43 The appellant also argued that the district judge wrongly failed to draw an adverse inference from the prosecution`s failure to call one Karine Yeo as a witness.

44 Evidential issues

45 As far as the evidence against him was concerned, the appellant argued most forcefully that too little weight was given to the evidence of the defence witnesses and that, if their evidence had been properly considered, the prosecution would not have established it`s case beyond a reasonable doubt.

46 The appellant also contended that the statements given by himself to the police had not been properly recorded, and that the district judge placed too much weight and attributed too much probative value to them.

47 Finally, the appellant argued that too much credibility was attributed to the evidence given by PW1, PW2 and PW3, especially given that they had motive and opportunity to fabricate evidence against the appellant. Counsel for the appellant highlighted that, where the accused outright denied the charges against him and alleged that the charges had been fabricated by the complainant(s), the burden of proving lack of motive for fabrication was on the prosecution rather the burden of proving improper motive being on the defence, and he claimed that the prosecution had failed to discharge this burden.

48 Prejudicial evidence - similar fact evidence

49 Counsel for the appellant quite rightly pointed out that PW1, PW2 and PW3, in giving evidence, made reference to the incidents of molest involving the third complainant. It is helpful to set out the following passages from the examination-in-chief of PW1, which are illustrative of these references:

50 And

51 It is a cardinal rule of our law that when an accused person is being tried for an offence, the fact that he has committed a similar offence or has at some other time in his life behaved similarly is not a relevant fact. The court is not allowed to infer from a person`s past crimes that he has a propensity or tendency to commit similar crimes, for such an inference colours the court`s ability to assess the evidence before it objectively. If, therefore, the learned district judge decided to convict the appellant because the allegations of the third complainant persuaded her that the appellant was a man likely to commit the offences for which he had been charged, this would have amounted to a clear error of law. For that matter, the district judge was under an obligation to give the same treatment to the evidence of the second complainant viz the first complainant and vice versa.

52 However, the impermissible inference discussed in the antecedent paragraph must be distinguished from cases where a fact which is independently and of itself relevant to the criminal liability of the appellant incidentally reveals that he has in the past committed similar crimes or acts. Such `similar fact evidence` (as it is frequently referred to at common law) is made admissible by virtue of its relevance (for example, under ss 14 and 15 of the Evidence Act (Cap 97)) and in spite of its` prejudicial effect.

53 This principle was most recently enunciated by the Court of Appeal in Tan Meng Jee v PP [1996] 2 SLR 422 , 432H where Yong Pung How CJ, delivering the judgment of the court, said of similar fact evidence that:

54 The court in Tan Meng Jee held that, before a judge may consider a similar fact relevant by virtue of s 14 or s 15, that fact must first satisfy the test for the admissibility of similar fact evidence laid down in Boardman v DPP [1975] AC 421 , 434E, namely, that the probative value of the evidence must exceed its prejudicial effect.

55 At first glance, this may appear to have undercut the supremacy of s 5 of the Evidence Act, which provides that evidence may only be given of a fact in issue or a relevant fact. The logical argument is that if a fact is relevant under s 15 of the Evidence Act it is by definition admissible as evidence by virtue of s 5. The degree of probative force and the potential prejudice to the appellant are non-factors as far as s 5 is concerned.

56 However, I am of the view that the superimposition of the Boardman test on s 14 and s 15 poses no such conceptual difficulty. The reason is that the Boardman test, as the court explained in Tan Meng Jee , has already been encapsulated in the wording of s 14 and s 15 as they now stand: see pp 434I to 435C. In order to qualify as `relevant` under ss 14 and 15, the fact must necessarily be `probative`. And as the court elaborated in Tan Meng Jee , where the prejudicial effect is great, a greater corresponding degree of similarity (and hence, probity) is required to satisfy the test.

57 It is, however, the question left unanswered in Tan Meng Jee which was presented to the court in this appeal, namely, does our law accommodate the admission of similar fact evidence which does not fall within the ambit of ss 14 and 15, and if it does, to what extent would the test in Boardman govern the admissibility of such evidence? This question arose because by and large the similar facts recounted by PW1, PW2 and PW3 did not establish the appellant`s mens rea but only the actus reus of the offences charged. The similar facts thus did not qualify for inclusion under ss 14 and 15, with the possible exception that they could have rebutted the appellant`s defence of innocent association to the first incident involving the second complainant (the medical examination).

58 One solution is to answer the `unanswered question` affirmatively and without reference to the Evidence Act, but by saying that, to draw a distinction between probative similarities of intention, knowledge or states of mind and probative similarities of action is too artificial to be sustainable. Attractive as this approach may appear to be, I considered it to be overly dismissive of the provisions of the Evidence Act, particularly in light of s 2(2) of that Act, which provides that all rules of evidence not contained in any written law, so far as such rules are inconsistent with any of the provisions of the Act, are repealed. As ss 14 and 15 contemplate the inclusion of certain similar facts, other similar facts must also be admitted under a provision of the Act. If similar facts were admitted other than under one of the relevancy provisions, it would make a mockery of the Act and extend the ambit of the similar fact rule beyond the extent intended by the legislature.

59 A second, more preferable, solution would be to declare that such facts would be relevant by virtue of s 11(b) of the Act, which provides that a fact which would not otherwise be relevant is relevant if, by itself or in connection with other facts, it makes the existence or non-existence of any fact in issue or relevant fact highly probable or improbable. This solution has the added advantage that the phrase `highly probable or improbable` also answers the question whether the Boardman test should be applicable: the phrase marries relevance to probity arguably to a greater extent than either s 14 or s 15.

60 The principal difficulty with this approach is that to construe s 11(b) in such a manner is at odds with the draftsman`s commentary: see Stephen, Digest of the Law of Evidence at p 19 at which he commented on similar fact evidence that `the fact that [a man had] formerly committed another crime of the same sort, and had a tendency to commit such crimes, is deemed to be irrelevant`. This comment itself offers nothing objectionable, for, as I have explained, the admission of similar facts under s 11(b) would be restricted to such facts as have independent relevance and probative value, and would not extend to facts that merely show a disposition towards crime.

61 However, Stephen elaborated at pp 188s is one of the `four great exclusive rules of evidence`, to which he acknowledged only two exceptions, namely, those set out in ss 14 and 15. He did not accommodate the possibility that similar facts might be probative of the actus of a crime in addition to it`s mental element. It would therefore be contrary to the draftsman`s intention to construe s 11(b) as admitting facts that Stephen never considered appropriate for admission.

62 Nonetheless, without detracting from the utility of Stephen`s Digest as an interpretative aid, the application of the doctrine in Pepper v Hart [1993] AC 593 does not necessarily sanction the elevation of such commentary to the level of Hansard. The incorporation of that doctrine into our law as s 9A of the Interpretation Act (Cap 1) does not warrant the use of academic texts in construing the intention of Parliament. This approach seems to be endorsed by s 9A(3) of the Interpretation Act, a list of interpretative aids from which academic texts and the private works of draftsmen is conspicuously absent. I am of the view that to affirm this interpretation of s 11(b) would also pave the way for future treatment of the Evidence Act as a facilitative statute as opposed to a mere codification of Stephen`s statement of the law of evidence. Particularly in this area, it must remembered that Stephen drafted the Indian Evidence Act long before the first authoritative expression of judicial opinion on this subject in Makin v A-G of New South Wales [1894] AC 57 . Thus, in the present case, I do not think it appropriate to sustain an artificial distinction between similar facts which are probative of intention (or other states of mind) and similar facts which are probative of acts done by the accused, nor do I consider such a distinction to have been intended by Parliament.

63 Moreover, this use of s 11 has been previously the subject of judicial reference. In PP v Teo Ai Nee [1995] 2 SLR 69 , 94G Yong Pung How CJ, hearing an appeal from the magistrate`s court against the respondents` acquittal from offences under s 136 of the Copyright Act, discussed s 11 in the context of similar fact evidence and the Boardman test although he stopped short of deciding that similar fact evidence pertaining to the actus reus of a crime was relevant under that section, as that question was not before the court. See also Abubakar bin Ismail v R [1954] MLJ 67 .

64 Coming to the present case, however, I am prepared to hold that the district judge would have been entitled to consider the allegations made against the appellant by all three complainants in deciding the appellant`s liability for charges in respect of any one of them. The degree of similarity in this case is by no means `striking` in the legal sense widely attributed to the House of Lords` decision in the Boardman case. However, as Lord Mackay of Clashfern LC pointed out in R v P [1991] 3 All ER 337 (which was applied by the Court of Appeal in Tan Meng Jee ):

65 R v P was an appeal to the House of Lords against the acquittal of the accused father on four counts of rape and four counts of incest in respect of his two daughters. The trial judge had refused the respondent`s application for separate trials and proceeded to admit the evidence of each girl in respect of the offences allegedly committed by the respondent against her. The Court of Appeal reluctantly acquitted the respondent and quashed his conviction. Lord Lane CJ came to the conclusion that as there were no unusual features such as to make the account given by one girl more credible because those features were mirrored by the statement of the other, the evidence, not being `strikingly similar`, was wrongly admitted and the law as it stood demanded an acquittal. The question before the House of Lords was whether, in cases where a series of offences involving multiple complainants was alleged against a pederast or incestuous father, one victim`s evidence can be properly admitted in the trial of another victim in the absence of `striking similarities` in the evidence of both. The House answered the question in the affirmative.

66 Having reviewed the authorities, the Lord Chancellor concluded that where A wishes to give evidence of offences committed against A and B also wishes to give evidence of offences committed against B, the judge must decide whether:

67 He went on to say:

68 I am entirely in agreement with the approach adopted by my learned friend. In cases such as these, where, to adopt the words of Lord Lane CJ in the lower appellate court, a person accused of child molestation:

69 Similarities in the present case include the capacity in which the boys were known to the appellant, the age of the boys, the nature of the acts (in particular the references to masturbation and the fondling of testicles), certain similarities in the manner the boys claim they were approached (for example on the pretext of physical examination and in the context of returning gym equipment), and a commonality as to the location where the acts took place (four of the six incidents took place at the community centre). Although the similarities in this case would not have justified admitting the evidence if the identity of the perpetrator was in question, I consider that the facts of the present case fall neatly into the type of situation envisaged by the House of Lords in R v P . The district judge would therefore (subject to a finding that there was a real risk of collusion between the victims) have been entitled to consider the evidence of both victims together in deciding whether to convict the appellant. The issue of collusion is of significance and it will be addressed separately.

70 I wish to add that, even if I am wrong on this point, I consider that the district judge did in fact consider the evidence against each victim separately and was not minded to make inferences as to the appellant`s guilt on any one charge from the evidence given by the other complainants. Limited evidence was presented of the incidents alleged by the third complainant. There seemed little doubt that once the prosecution stood down its charges, it was prepared to refrain from presenting evidence relevant to those charges. Whether the prosecution was mindful of the dangers in adducing similar fact evidence or it simply was not prepared to show its hand in the event the charges would be revived in subsequent proceedings is not clear. At any rate it is evident that the learned district judge was mindful of the potential pitfalls in that she did not allow PW5 to recount what the third complainant narrated to him concerning the latter`s experiences with the appellant.

71 As to whether the learned district judge considered the evidence of PW1 in deciding liability as regards the charges involving PW2 and vice versa, there is nothing in the learned district judge`s grounds of decision to suggest that she would not have convicted the appellant of the charges involving one boy but for the evidence of the other. Her treatment of the allegations made by PW1 and PW2 and the respective denials made by the appellant and the supporting evidence adduced by the defence witnesses are completely independent. In deciding the appellant`s liability for each charge, no reliance was placed on evidence of the other alleged victims. Although she stated in her grounds of decision that `Looking at the evidence in totality I had no doubt that the appellant had committed the acts alleged in the six charges`, I am not prepared to conclude that the district judge meant that the conviction of the appellant on charges involving PW1 was a sine qua non of PW2 and PW3`s evidence.

72 Joinder of offences

73 It is convenient to deal briefly with the question whether the district judge erred in law in ordering a joinder of the six charges and in refusing defence counsel`s application for separate trials. I again take note that a similar application was at the heart of the issues in R v P .

74 As an important preliminary consideration, I wish to highlight that English law and Singapore law diverge on this point. In our law, the question whether a judge should order a joinder is governed by wholly different considerations from the question whether similar fact evidence should be admitted. It is entirely possible for a judge to order a joinder but with a view to treating that the evidence of the two victims separately. The trier of fact in our system is endowed with the judicial ability to preserve and apply the rule against similar facts. In England, the difficulties faced by a jury in preserving that rule where evidence of two victims is adduced at one trial almost invariably justifies separate trials. It is only where there are striking similarities or similarities sufficient to create a probative nexus between the evidence of the two victims that a single trial is appropriate - this is the result of the decision in R v P .

75 In Singapore law, whether a joinder is appropriate is governed by s 169 of the Criminal Procedure Code. Section 169 provides that where a person is appellant of more offences than one he may be charged with and tried at one trial for any number of those offences if they form or are part of a series of offences of the same or similar character .

76 I am of the view that the six charges proceeded with by the prosecution met the requirements of s 169. The only real question, which was argued before the district judge before the commencement of trial, was whether this was a case in which the court should have ordered a separate trial as it is empowered to do under s 171 of the Criminal Procedure Code where a joinder of charges in a single trial would prejudice the appellant or embarrass him in his defence.

77 As this was a case in which it would have been appropriate for the district judge to consider similar fact evidence, in the absence of some other source of prejudice or embarrassment other than the rule against similar facts, s 171 has no operation and I therefore found that the district judge rightly refused defence counsel`s application for separate trials. I wish to reiterate, however, that, even if I had come to the conclusion that similar facts could not rightly be admitted in the present case, that would not have conclusively necessitated separate trials. For even where similar facts must not be considered in determining liability, because their prejudicial effect exceeds their probative force, the trial judge retains the discretion under s 171 to decide whether the degree of prejudice presented by a single trial justifies an order for separate trials. In such cases, the judge must ask himself whether he would be so influenced by the evidence presented by both victims that he would be unable to preserve the sanctity of the rule against similar facts.

78 Corroboration

79 Counsel for the appellant premised the thrust of his appeal on the argument that the district judge ought not to have convicted the appellant as the evidence of PW1 and PW2 was not corroborated and it was accordingly unsafe to render a conviction. There are two bases on which corroboration may be said to be required - the first is that the present case involves sexual offences to which the complainants are the only witnesses and, secondly, that those complainants are in fact child witnesses.

80 Child witnesses

81 Although s 136 of the Evidence Act provides that `no particular number of witnesses shall in any case be required for the proof of any fact`, it is a well-established rule of practice in our law that where evidence is given by a child witness, that evidence is not to be accepted at face value without some measure of corroboration: see Goh Liong Lam & Ors v R [1958] MLJ 254 , Chao Chong & Ors v PP [1960] MLJ 238 and Ng Kwee Piow v R [1960] MLJ 278 .

82 This rule was of more significance in the days before bench trials were introduced in Singapore, as an improper direction to the jury on the issue of corroboration warranted an acquittal on appeal. The rationale for this rule was conveniently set out in the judgment of Thomson CJ in the Chao Chong decision, namely, that:

83 Naturally, this rationale applies to different children in different degrees. Where, therefore, evidence is given by older children whose intellectual faculties are more developed, the danger in convicting without corroboration is diminished. The rationale of the rule makes it very difficult to lay down a guideline as to the point at which a maturing individual, in his progress towards adulthood, crosses the line past which the judicial process considers his testimony credible without independent evidence in support of it and this must therefore be a matter for the judge`s assessment in each case.

84 The court`s discomfort with requiring a corroboration warning in all cases involving a child witness manifested itself in Tham Kai Yau & Ors v PP [1977] 1 MLJ 174 in which the Federal Court considered that a formal warning on the issue of corroboration need not be issued to the jury if they were advised to pay particular attention to or to scrutinise with special care the evidence of young children, explaining the tendencies of such children to invent and distort.

85 In Tham Kai Yau , the appellants were convicted of murder on the sole evidence of the deceased`s son, who was aged 13 years at the date of the murder and 14 years at the date of the trial. The Federal Court considered that as the child `possessed sufficient intelligence to understand the meaning and significance of an oath` a warning was not required. It did not, however, dispense with the need for corroboration in that case.

86 It is in accordance with the approach of the Federal Court that I consider that there is no special rule requiring a trial judge to direct himself as to the dangers of convicting without corroboration where the only evidence is that of a child witness, although he or she must remain sensitive to the requirement of corroborative evidence or alternatively consider that corroboration is not required because of the maturity and reliability of the witness.

87 Complainants in sexual offences

88 Counsel for the appellant drew my attention to the decision of the Court of the Appeal in PP v Tang Kin Seng [1997] 1 SLR 47 . In Tang Kin Seng , the Court of Appeal held that:

89 In PP v Teo Keng Pong [1996] 3 SLR 329 , it was held that the phrase `unusually convincing` meant that the complainant`s evidence was so convincing that the prosecution`s case was proven beyond reasonable doubt solely on the basis of that evidence.

90 Tang Kin Seng and Teo Keng Pong were cases involving a single complainant who made allegations against an accused person. The principle emerging from these cases was that, unless the complainant`s evidence alone is so strong that it proves the prosecution`s case beyond reasonable doubt, corroboration is required. By extension, where the complainants` evidence is not so strong as to prove the case beyond reasonable doubt, but where the other prosecution witnesses, for example, an eye-witness, can provide direct evidence of the offence, such that the combined effect of the evidence is to satisfy the burden of proof, corroboration is similarly not required.

91 It is next necessary to consider situations in which two complainants make accusations against the same accused. Where each complainant`s evidence standing on its own is `unusually convincing` but the appellant makes an allegation of collusion between the complainants, I am of the view that it is only safe to convict if (i) the allegation of collusion is considered and discounted as a sham defence or (ii) where the allegation of collusion cannot be totally discounted, there must be corroboration, that is, independent direct evidence implicating the accused of the offences with which he is charged. The reason for this is that, because an allegation of collusion impugns the persuasive force of the complainants` evidence, `something else` is required to justify a conviction, the nature of which will be the subject of later discussion. At this juncture, I wish only to make a reference to the High Court`s dicta in PP v Khoo Kwoon Hain [1995] 2 SLR 767 that in cases of sexual offences:

92 I note that there may be some conceptual difficulty in the proposition that where the complainants` evidence satisfies the prosecution`s burden of proof, there will be situations in which an additional element - corroborative evidence - is required to justify a conviction. This objection also will be considered together with my conclusions on the issue of collusion.

93 Yet a third scenario warrants consideration, that in which the complainants` evidence on its own is not unusually convincing and there is an additional allegation of collusion. I am of the view that in such cases two elements are required before a conviction may be justified. The first is that each complainant`s evidence must be supplemented by some other evidence capable, when considered together, of meeting the standard of proof required by the prosecution. As mentioned, this may take the form of eye-witness evidence, or in cases such as the present, similar fact evidence. The second element which is required is that the allegation of collusion be obviated either by a finding that there was in fact no collusion, or if that possibility remains live because the prosecution could not counter it, then there must be some corroboration of the complainant`s evidence. The principles are the same as those discussed above.

94 Based on the district judge`s findings, I was of the opinion that the present case fell into the second category enumerated, that is, where two complainants` evidence satisfies the standard of proof but is challenged on the basis of possible collusion. The district judge found the evidence of both the first complainant and the second complainant `unusually convincing and accepted that they had been molested by the appellant on the occasions alleged`. The High Court in Khoo Kwoon Hain was not prepared to find the complainant`s evidence in that case `unusually convincing`, a matter to which I had to give some consideration because the factors highlighted by the High Court in coming to that conclusion were repeated in the present case, namely, that (i) the victims in both cases did not scream out or draw attention to the acts committed against them and (ii) the victims continued their association with the accused person after the alleged incidents. Additionally, in the present case, the victims did not register their complaints immediately after the incidents but only made the allegations against the accused after the passage of some time.

95 Having considered these factors in the context of the evidence adduced in the case before me, I was not prepared to overturn the district judge`s finding that the complainants` evidence was unusually convincing. Although the alleged victims did not scream out or protest when molested, this was easily explained by the fact that they were either threatened (in the case of the first complainant) or thought the whole matter to be innocent (as in the case of the second complainant). Similarly, although the alleged victims continued taekwondo classes after being molested, the first complainant explained that he made sure the third complainant was always there. The second complainant had no reason to avoid classes as he never suspected the incidents to be criminal until much later. The same fact explains why the second complainant did not make any complaints after the incidents involving him. The first complainant did in fact confide in the third complainant, but also mentioned in evidence that he did not talk about the incidents as he was embarrassed and afraid of what people might think of him if they knew he had been sexually abused.

96 Notwithstanding that I am prepared to uphold the district judge`s conclusion that the complainants` evidence was sufficiently convincing to warrant a conviction, the allegation of collusion activates the second limb of the approach I have outlined, that is, the possibility of collusion must be discounted or if it cannot be, there must be some form of corroboration. I dealt first with the question of whether there was in fact corroboration in the present case.

97 Whether complaints amount to corroboration

98 It was convenient first to consider the High Court`s decision in PP v Khoo Kwoon Hain , in which the question whether a complaint of an offence amounts to corroboration was discussed.

99 The principal corroborating evidence in the present case was the numerous incidents of complaints made by the victims. In evidence, the first complainant said that a few weeks after the first incident he told the third complainant about it. The third complainant confirmed this but said the time period was days rather than weeks. The first complainant said that after the third incident he told the third complainant about it at the camp itself. The third complainant again confirmed this. The first complainant also informed the second complainant of the incidents (although not in detail) in May 1996, around the same time that he informed Ranjeet Prasad. He thereafter made a police report.

100 The second complainant did not complain after any of the three incidents for some time. It was not until the first complainant confided in him in May 1996 that he made it known that the appellant had molested him. He also informed Ranjeet Prasad. He informed his father and made a police report.

101 At common law, complaints cannot amount to corroboration because the rule against self-corroboration prevents this. As the High Court noted in Khoo Kwoon Hain , complaints are capable of amounting to corroboration in our law because of s 159 of the Evidence Act. Be that the case, I did not consider that decision to be of very much use here, for the complaints made by the first and second complainants did not even fall within the ambit of s 159, because they were made so long after the alleged incidents. Even if that difficulty could be circumvented, I would still have to conclude, as did the High Court in Khoo Kwoon Hain that such corroboration, not being independent, could only be of `little additional evidential value`. See also John Benjamin Cadawanaltharayil v PP [1995] 3 SLR 805 and PP v Anuar bin Arshad [1996] 2 SLR 52 .

102 Whether similar facts may amount to corroboration

103 Looking beyond the complaints registered by the respective complainants, I was mindful of the possibility that, in certain circumstances, similar fact evidence, where properly admitted, may also corroborate a victim`s evidence.

104 DPP v Kilbourne [1973] AC 729 was a case in which the accused was charged and convicted of seven counts of buggery, attempted buggery or indecent assault. The victims were young boys aged between 9 and 12 and fell into two groups, one group for offences committed in 1970 and the second group for offences committed in 1971. All the boys gave evidence at the same trial. Counsel for the accused conceded that similar fact evidence was rightfully admitted. The question before the House of Lords was whether evidence given by boys in one group could corroborate the evidence of boys in the other group. The Court of Appeal had felt constrained by earlier authorities to conclude that such evidence could not amount to corroboration of the other boys` testimonies.

105 The House of Lords reversed the decision of the Court of Appeal on this point. Their Lordships were of the opinion that once evidence directly relevant to one group of boys was relevant as similar fact evidence which, although circumstantial, supported the evidence of the other group of boys, the evidence of the first group was corroboration of the evidence of the second group. Said Lord Reid:

106 To their Lordships, if evidence was admitted as similar fact evidence, it necessarily had to implicate the accused by supporting the principal evidence in a material particular. It made no sense to draw an artificial distinction in pronouncing that evidence given by A of the accused`s behaviour towards A which could be used to directly prove the accused`s guilt in respect of an offence against B could not equally be capable of amounting to evidence that merely supported or reinforced B`s evidence. In this case the similar fact evidence and corroborative evidence were one and the same - to have come to any other conclusion would have been illogical.

107 I wish to add that there will be situations in which that distinction is sustainable. For example, even if the evidence given by A is inadmissible as similar fact evidence because it fails the Boardman test, it may nonetheless be admissible as corroborating evidence. In such a case, the evidence adduced by A is not of itself admitted as a relevant fact or proof of a fact in issue: it is not in and of itself probative. The evidence merely lends support to B`s evidence of a relevant fact or fact in issue so as to make B`s evidence more believable. The inclusion of evidence as similar facts necessarily renders it corroborative. However, the exclusion of evidence as similar facts does not necessarily rob it of its corroborative quality.

108 This dichotomy is confirmed by close examination of s 158(1) of the Evidence Act, in which there is a ready classification of types of evidence into (i) `relevant facts` and (ii) `questions as to any other circumstances` which were observed at or near the time or place at which the relevant fact occurred. The `questions as to any other circumstances` are referred to as corroborating evidence as contradistinct from evidence which has a direct bearing on liability (ie `relevant facts` or `facts in issue`, including similar facts which are `relevant` under ss 11, 14 or 15).

109 I did not, however, find myself inclined to hold that the issue of corroboration in the present case should be decided in the same way as in Kilbourne , that is to say, that PW1 and PW2`s evidence may amount to corroboration. For where there is an allegation of collusion, the evidence of another complainant who is alleged to have participated in the conspiracy against the accused can hardly amount to independent corroboration. To look to the co-complainant for corroboration merely begs the question whether there is corroboration in the first place. This was the conclusion of the trial judge in Kilbourne and the principle also appears to have been approved when the matter went before the House of Lords: see for example the italicised words in the passages I have quoted from the judgment of Lord Reid, at para 83, above, and also the following passage from the judgment of Lord Hailsham:

110 I also made reference to the passage I have quoted from the judgment of Lord Lane CJ in R v P at para 52. The principle was stated most clearly and succinctly in the recent House of Lords decision in R v H [1995] 2 All ER 865 , to which I will refer later.

111 I did not consider that my conclusion that two complainants against whom conspiracy is alleged may not corroborate each other is contrary to the wording of s 158(2) which provides that `any rule of law or practice whereby in criminal proceedings the evidence of one witness is incapable of corroborating the evidence of another witness is hereby abrogated.` It should be remembered that, although the common law rule against mutual corroboration has been expressly abrogated by s 158(2) of the Evidence Act, the dating of the Act does not extend the effect of s 158(2) to the rule in Kilbourne which is a doctrine of much more recent origin.

112 Whether there was any oher evidence amounting to corroboration

113 The district judge did, however, make one observation which might be considered critical. She observed that Tommy Kok (hereafter referred to as `Tommy` or `DW3`), who shared office premises with the appellant, was emphatic that he and the appellant ordered lunch from outside to be delivered to their office at Gay World Complex. The appellant insisted that he did not order lunch the day that the first complainant collected the `gi` from his office. However, as the district judge was correct to point out that the appellant must have ordered food by telephone, otherwise the first complainant would not have known of this fact, much less be mindful to include it in his testimony. This suggested the appellant was lying in his testimony and the credibility of his denial that the incident ever took place was damaged as a consequence.

114 I therefore took the view that the first complainant`s observations taken together with the statements made by the appellant and Tommy Kok fell into the category of `questions as to any other circumstances` which were observed at or near the time or place at which the relevant fact occurred within the meaning of s 158(1) and corroborated his story as regards the first incident. I had directed myself to the possibility that the lie told by the appellant (relating to the ordering of food) might constitute corroboration, but considered that such a holding would be straining the four-part test applied in the Khoo Kwoon Hain decision, in particular the requirement that the lie must relate to a material issue.

115 Effect of absence of corroboration

116 In the absence of corroboration in respect of the remaining five charges, it was necessary to examine the allegations of collusion to determine if they were made out. It was only if they were not that a conviction was justified.

117 Collusion

118 It was convenient first to refer to the judgment of Lord Mustill in the House of Lords` decision in R v H where his Lordship distinguished between two forms of `collusion`:

119 For future guidance, I also thought it necessary to clarify that the manner a trial judge should approach evidence which is tainted by allegations of conspiracy or innocent infection differs depending on the allegation. Where the allegation is that the complainants are telling deliberate untruths, the established principle is that the prosecution has the burden of proving beyond reasonable doubt that there was no real risk of collusion, for as the High Court held in Khoo Kwoon Hain ,

120 The standard of proof to be applied is the criminal standard. Unless the prosecution can prove lack of motive beyond reasonable doubt, the judge must disregard the evidence of the complainant or complainants and, if that is the only evidence before him, acquit the accused, or if there are other complainants in respect of whom the allegation of collusion is not made or not made out, consider the accused`s liability in respect of those complainants without regard to the condemned testimony. I did not consider it appropriate that the admissibility of the evidence be considered at a separate hearing akin to a voir dire. This proposition was put forth by the appellant`s counsel in R v H and rightly rejected by the House of Lords.

121 By contrast, where the allegation is one of innocent infection, and there is some evidence suggesting the opportunity for contamination, the court must always be alive to that possibility even if it considers the possibility to be slight. Where an allegation is one of infection (as opposed to conspiracy), the accused is objecting to the probative value of the evidence and not its` validity. He is not challenging the evidence as an outright lie, but saying that the degree of veracity of the evidence and the strength of recollection of the witness must be subject to scrutiny.

122 To require the prosecution to prove beyond reasonable doubt that there was no risk of contamination would be oppressive as - and this has already been pointed out - the possibility of contamination must as a necessity always be in the court`s mind so long as the opportunity for contamination was present. If the court must remain sensitive to the possibility of contamination as a matter of legal principle and fairness to the accused, it would be illogical to adopt a test requiring the prosecution to remove all doubt from the court`s mind.

123 The duty of the trial judge in such a position is to consider the evidence of the complainants and to assess the risk and opportunity for contamination. The trial judge must then ask himself or herself what weight is to be attached to the testimony of these witnesses in light of that assessment. The final question is whether that evidence, together with the other evidence before the court, satisfies the standard of proof required of the prosecution.

124 In the present appeal, counsel for the appellant contended that the district judge erred in `failing to consider that there was sufficient evidence before the court to reasonably infer that [the first, second and third complainants] had a motive to fabricate evidence against the appellant.` By this, I understood the appellant to be alleging that the charges against him were completely false and that the stories had been made up by the three complainants purely out of spite or some other illegitimate reason, in other words, there had been, in the words of Lord Mustill, a `conspiracy`.

125 I was of the view that the district judge in this case was correct in allowing the evidence to be adduced and in leaving the question of collusion or fabrication to be determined subsequently. This approach was preferable to any suggestion that the evidence should be subjected to some inquiry as to its admissibility because it does not result in the exclusion of potentially probative evidence at the very outset of trial. Rather, the trial judge may decide at the close of the case and in the light of the evidence whether the allegation of collusion has been made out. In the present case, the trial judge considered that the appellant`s contention that the complainants had fabricated evidence against him was not made out. The question before this court was whether the trial judge was correct in arriving at that conclusion, to which I will now turn.

126 The appellant alleged that the complaints were all false and instigated by the second complainant, who bore a personal animosity towards him. This animosity, the appellant said, stemmed from the appellant having disciplined the second complainant whilst on a taekwondo tour of Taiwan, Hong Kong and Korea from 29 May 1996 to 7 June 1996.

127 According to the appellant, he learnt that the second complainant had spent the night sleeping in the same bed as one Karine Yeo. The following day he spoke to both of them and reproached them. For this, the second complainant bore resentment against the appellant. Subsequently, the appellant found one of the hotel rooms in a disorderly state. The other roommates indicated that it was the second complainant`s fault. The appellant found the second complainant in Karine`s room. The appellant `gave him a dressing as to why he did not listen to me`. The second complainant did not respond. Seeing this, two other team-members who were with the appellant became infuriated and attempted to strip off the second complainant`s pants. The appellant told them to stop because a female was present in the room. The two team-members were Brandon Thian Yuang Loong (DW9) and Tang Jun Jie (DW12). The appellant said that the second complainant harboured a grudge against him as a result of these incidents and that the present charges were a fabrication. In support of this, his witnesses adduced evidence of a third incident, in which a branded shirt went missing from Karine`s bag. The second complainant told Karine that the appellant took the shirt. Dr Wong Lai Ping (DW2)`s version of the facts was that the second complainant took the shirt himself when Karine was sleeping and put the blame on the appellant. This evidence was intended to show that the second complainant `had it in` for the appellant.

128 The second complainant`s version of the facts was that he, Karine, her room-mate Lin Chang Yi (DW10) and DW10`s two brothers, Lin Jing Ze (DW8) and Lin Jing Tian (not called as a witness) decided to sleep in the same room. He admitted being reproached by the appellant but claimed he felt more guilty than annoyed. The second complainant also claimed that it was the appellant who instigated Brandon and Jun Jie to attempt to strip him and that the appellant also participated in the incident. It stopped when someone tried to take photos and the camera flash went off. However, there was no film in the camera. The second complainant denied that he ever told Karine that the shirt was taken by the appellant.

129 The district judge found that:

130 Having reviewed the evidence, I was of the opinion that the district judge was justified in coming to that conclusion. Even if the appellant`s witnesses` version of the events on the trip to Taiwan, Hong Kong and Korea was closer to the truth than that suggested by the second complainant, I was not minded to disagree with the trial judge`s finding that those events did not influence the present charges against the appellant. Counsel for the appellant both in this appeal and in the lower court also did not elaborate on how the second complainant could have persuaded the first and third complainant to bring false charges, nor did the appellant make any suggestions as to why the latter two boys might have been motivated to undertake this ordeal. This fact alone - the absence of a unifying motive amongst the complainants - eliminated any doubts that may have arisen out of the second complainant`s evidence and the evidence adduced by the appellant`s witnesses in contradiction of it.

131 In coming to this conclusion, I was also persuaded by the fact that the respective complainants admitted at trial that they were not fully acquainted with the details of the incidents involving the other complainants. Had this been a true `conspiracy`, the case would have been one in which the `details chime because that is what they are designed to do`.

132 However, quite separate from the question of whether the complainants had conspired to falsely bring these charges against the appellant was the equally important question of whether the complainants` evidence was contaminated through `innocent infection`. The three complainants knew each other well. They attended the same school and were in the same standard. They took taekwondo lessons together. The second and third complainants were especially close friends and it was to the third complainant that the second complainant turned to after the events that transpired involving the appellant.

133 It is necessary to state for the record that this matter was neither canvassed in the lower court nor raised in this appeal. It was therefore not even necessary for me to decide the issue. However, bearing in mind the approach I have suggested in cases such as this, I considered that even had the matter been raised, I would have been of the opinion that the risk of contamination was not so overwhelming as to justify an acquittal before this court. For this conclusion I relied on the reasons I have set out in para 105.

134 I now return to the conceptual conundrum I referred to earlier - that is, whether the standard of proof required of the prosecution is raised above the conventional standard if we were to require corroboration where there is an allegation of collusion but the complainants` evidence either standing on its` own or in conjunction with other evidence establishes guilt beyond reasonable doubt.

135 Where an allegation of `conspiracy` is concerned, then obviously this requirement is a non sequitur - for if the possibility of conspiracy is not discounted by the prosecution, the complainant`s evidence is to be dismissed and an acquittal is in order. If, however, there is independent evidence that may be capable of supporting or verifying the evidence of the complainants, I do not think it matters whether this be classified as corroboration or as evidence that goes to prove the prosecution`s case that there was no motive for fabrication - as either way, provided the independent evidence is of sufficient probative value, the allegation of conspiracy would be defeated.

136 Insofar as an allegation of `innocent infection` is concerned, so long as there has been proof of some opportunity of contamination, the weight of the respective complainants` evidence is accordingly diminished. In cases where the opportunity of contamination is so slight as to be insignificant, then the allegation of collusion is not made out and it is safe to convict on the evidence of the complainants provided it is `unusually convincing`. Where, however, the opportunity of contamination is a real opportunity, such that the truth of the complainants` evidence is subject to scepticism, something more is required to justify a conviction. It is conceivable that there will be cases in which the very strength of the complainants` `unusually convincing` evidence derives entirely from the innocent infection. As a result of the potential for injustice which springs from such conspiracy, the law demands evidence originating from some independent source which is capable of confirming the credibility of the complainants` evidence.

137 If the appellant`s allegations of collusion and fabrication are not borne out in the light of the evidence, then the rationale for requiring corroboration in sexual offence cases, such as the present one, no longer applies. As the possibility of improper motive on the part of the second complainant was fully explored and correctly dealt with at trial, and for the reasons I have set out above on the question of innocent infection, the court could dispense with the need for corroboration insofar as the rationale concerning complainants of sexual offences is concerned. Insofar as the rule requiring corroboration of children`s evidence was concerned, having discounted the possibility of collusion, I considered there to be no obstacle to the similar fact evidence adduced by the complainants standing as corroboration of the evidence of each individual complainant in accordance with the reasoning in R v Kilbourne .

138 Prosecution`s failure to call Karine Yeo

139 The appellant also contended that the district judge should have been prepared to draw an adverse inference from the prosecution`s failure to call Karine Yeo. This argument was based on illustration (g) of s 116 of the Evidence Act, which states that the court may presume that evidence which could be and is not produced would if produced be unfavourable to the person who withholds it.

140 Reference should be made to the decision in Khoo Kwoon Hain in which the court had to deal with a similar argument. In that case, the alleged victim claimed she had been molested by the appellant and that she had complained of this to a person identified as `aunty`. The prosecution did not produce `aunty` as a witness or reveal her identity to the defence. Accordingly, the court was prepared to draw an adverse inference.

141 To do the same here, however, would be to overstate Karine`s importance as a witness. She could offer no evidence which would have any bearing on the appellant`s guilt. It was never contended that she had any knowledge of any of the alleged incidents. All she could do was to rebut certain facts alleged by the appellant in his effort to establish that the charges brought against him were fabricated. The prosecution was entitled not to call her as a witness, in accordance with the principles enunciated by the court in Lim Young Sien v PP [1994] 2 SLR 257 , at p 266 where it held that:

142 This must a fortiori be the case in light of the conclusion that the appellant`s allegation of collusion was not borne out by the evidence. Karine`s contribution to the prosecution`s case would therefore have been all the less.

143 Appellant`s statements to the police

144 I will now deal with the factual issues, starting with whether the police statements were properly recorded. Some reliance seems to have been placed on incriminating statements made by the accused in his statements to the police. Counsel for the appellant argued at length that the manner in which the statements were recorded rendered them unreliable, and, even if not excluded completely from the record, the statements should have been accorded less weight by the trial judge, who held that the long statement `showed without a doubt that the accused had intended to molest the victims.` She did not believe that the accused had been tricked into signing a statement he did not agree with and concluded that `the accused having been confronted with his highly incriminating long statement ... felt it necessary to claim he was misunderstood and that the statement was not interpreted to him.`

145 Of the six s 122(6) statements made by the appellant, five consisted of denials of the charges. The exception was the charge setting out the first incident involving the second complainant. This was the purported medical examination after a training injury. In response to the charge the accused said `Maybe during that time I was playing with him and touch his testicles (below). Nothing else to say.` In his long statement, the accused admitted that he had taken the second complainant into the toilet and when asked if he had rubbed his testicles replied `I flick his balls`. When asked of his intention in doing so, he replied that he was `playing a fool`. He then admitted having done the same thing two or three times. He then said `I do not know why I do such things. After touching him at the private part, I feel very bad.` In the interests of space I do not propose to set out the remaining parts of his statement except to add that in the last paragraph of his statement he added:

146 The appellant sought at trial to establish that he did not understand the interview and that what was recorded by the investigating officer was not what he had intended to say. For example, he claimed that in his s 122(6) statement he had said `Maybe during that time I was training with him` not `I was playing with him`. In relation to the last paragraph of his long statement, he claimed that what he meant was that he frequently scolded his students if they did not take training seriously and he wished not to offend anyone. Although some of his claims were incredible, the district judge rightly directed her attention to the procedure adopted by the police officers in taking his statement.

147 The appellant`s s 122(6) statements were allegedly recorded through the Hokkien interpretation of one Leow Hua Keong (PW8). Subsequently, the appellant allegedly gave a long statement through the Mandarin interpretation of Corporal Choo Kwang Meng (PW9). The recording officer was one Staff Sergeant Shepherdson Ignatius (PW7), who speaks neither Mandarin nor Hokkien. There were certain inconsistencies in the evidence concerning the taking of the statements.

148 As far as the s 122(6) statements are concerned, it appeared that there was some confusion as to whether the interviews were conducted in Hokkien. According to PW7, the appellant spoke to him in English. PW8, however, remembered that the appellant spoke in Hokkien.

149 At least as far as the first s 122(6) statement was concerned, which was the only one in which the appellant said anything other than to deny the charge, this was translated into Hokkien for the appellant by his own admission, a fact which was consistent with PW7`s evidence. The evidence given by PW8 was that he assisted in the making of the first three statements but not the other three as he was called off to do something else. The appellant alleged the other s 122(6) statements were not translated, but this need not be regarded as material as it appeared that, even given the appellant`s limited knowledge of English, he would have understood the statements, which were very simple.

150 In respect of the long statement, it appeared at trial that no translation was provided at any time during the interview. The questions posed to the appellant when PW7 took his long statement were not in fact translated to him. When the interpreting officer, PW9, was asked whether he heard what had transpired between the appellant and PW7, his reply was `Yes, but not exactly`. However, PW7 stated that the statement was translated to the appellant before he signed it although the appellant alleged that it was not.

151 A further irregularity was that although the appellant signed the statement, his residential and business addresses were wrong, suggesting the statement might not have been translated to him. PW9 attempted to explain this by saying that he did not translate the first paragraph as he expected the appellant to be familiar with his personal particulars.

152 The questions before the court were therefore, where an appellant claims not to understand English very well, (i) what is the effect of the interview for the long statement being conducted in English without interpretation; (ii) what is the effect if the interview was conducted in English without interpretation but the statement translated to him in a language he understands before he signed it; and (iii) what is the effect if the interview was conducted in English without interpretation and the statement was not translated to him at all?

153 I am of the view that, unless the accused person`s understanding of English is so poor that he cannot understand what is being asked of him or cannot communicate with the interviewer, there is nothing wrong with the interview being conducted in English so long as the statement is translated to him and he is given an opportunity to amend or make additions to it. He may then do so in whatever tongue he is comfortable with, and the translator should dictate this to the recording officer. Unless there is bad faith on the part of the translator or recording officer, for example in refusing to incorporate the accused`s amendments or in falsely recording what the accused wished to add, the statement must be good and valid as it has been adopted by the accused. If, however, the statement is not translated to him at all, then it cannot be his statement as the risk of him having adopted as his own something he did not understand is too great to warrant admission of the statement.

154 Therefore, as long as PW7 and the appellant were able to communicate, the language in which the interview itself was conducted was not relevant provided it was translated to him before he signed it.

155 In the present case, two factors suggested that the long statement was correctly admitted. The first was that the district judge found as a fact that PW7 and PW8 were truthful, that is, the statements were properly translated. The second was that it appeared from the notes of evidence that the appellant in fact understood more English than he would have liked the court to believe. There were several times when he answered the prosecutor`s questions before they were translated to him. He was also able to read out from his long statement in English. I therefore found that the district judge was correct to admit the statement and there was no basis for the court to accept the invitation of the appellant`s counsel to disregard the incriminating statements in the light of subsequent evidence, as it did in Neo Ah Soi v PP [1996] 1 SLR 534 . It was not necessary for me to decide the question whether the approach taken by the court in Neo Ah Soi would apply in a case other than one in which the subsequent evidence showed the appellant`s statement to be involuntary.

156 Evidence of defence witnesses

157 The appellant adduced witnesses to contradict five elements of the prosecution`s case, which will be addressed individually. Counsel for the appellant submitted that the district judge did not put sufficient weight on the evidence of these witnesses and that, if she had, she would have ordered an acquittal.

158 Incident at Gay World Complex

159 The appellant produced Philip Choo Beng Hoe (`DW4`), who shared office premises with the appellant and claimed he was in the office every afternoon from 1pm to 7 or 8pm. He testified that, in the whole of 1995, only one person came to collect a `gi` from the appellant`s office premises, and that he was present at the time, and that nothing happened. This was intended to rebut the first complainant`s allegations as regards the first incident. Philip Choo was adamant that this happened two days before the trip to Taiwan, Hong Kong and Korea, that is, on Saturday, 27 May 1995. Because the first complainant was completely mistaken as to the date of the incident (it will be remembered that he initially chose a date long before he even met the appellant), the prosecution`s case on this was never quite clear.

160 The third complainant stated that the first complainant collected his uniform sometime after he `joined`. He appeared to have been referring to a date sometime after the `trial training sessions`, which Lin Jing Ze insisted were on 21 or 25 May 1995. It was possible that the first complainant did not in fact collect his uniform on 27 May 1995, that he waited until the appellant returned from his training trip (the appellant did mention that he was very busy planning for the trip) to collect his uniform on Saturday, 10 June 1995 and registered the following Sunday, 18 June 1995. Philip Choo could not have been around on this date as he was in Europe on holiday. I was inclined to hold that, on balance, this latter theory should be accepted, taking into consideration my earlier discussion of the incriminating effect of the first complainant`s knowledge that the appellant ordered food in.

161 Trip to Taiwan, Hong Kong and Korea

162 The incidents alleged at the trip have already been set out above. Lin Jing Ze and his sister Lin Chang Yi gave evidence that the second complainant had all along wanted to sleep in the same bed as Karine, and that he had threatened to jump into Chang Yi`s bed in the middle of the night. Therefore Jing Ze and his brother Jing Tian slept on either side of Chang Yi`s bed, on the floor, to protect her. They said that the beds were never joined together. This evidence was not of much use to the defence as it did not show anything. The fact that the second complainant was reproached by the appellant had already been admitted. At the most it showed that the context in which he shared the bed with Karine was different. Nonetheless, Chang Yi testified that, when she asked Karine if she had any objections, Karine said she had none. If the district judge gave little weight to this incident, she was entirely correct in doing so. Whatever the true version of those events might have been, I did not think it would have had much bearing on the question of collusion. They would have shed no light on the second complainant`s reaction to having been disciplined by the appellant for sleeping in Karine`s bed, a fact which was admitted on both sides. In fact, if the appellant`s story was to be believed, that would have favoured the second complainant - for if he had genuinely slept in Karine`s room against the protests of the others - he would have had more reason to feel guilt and less reason to feel resentment when disciplined by the appellant. In rejecting the appellant`s version, I was mindful of the fact that the Lin family had long been associated with the appellant, who frequently gave them lifts home in his car.

163 Brandon Thian and Tang Jun Jie gave a different account of the stripping incident. However, I found that their story that they were so infuriated that the second complainant did not show respect to the appellant that they tried to strip off his pants in front of a female team member (Karine) hard to believe. It was not conceivable that their actions were a reasonable response to how they allegedly felt. The incident seemed to me more a youthful prank than a serious exercise in discipline undertaken by exponents of the martial arts.

164 Dr Wong Lai Ping testified that the second complainant in fact took a shirt from Karine`s bag and later said the appellant had taken it. Very little cross-examination was made on this point. The second complainant`s evidence on this was so confused as to be of little assistance. The impact of Dr Wong`s evidence, even if accepted, was limited, for the reasons discussed above regarding the appellant`s allegations of fabrication.

165 Taekwondo camp at People`s Association Resort

166 Dr Wong Lai Ping gave evidence that, on the second day of the camp, the appellant was with her in the kitchen until around 3 or 3.30pm when he went off to attend to his father who was admitted to hospital. This was meant to rebut the first complainant`s allegation of the third incident involving the appellant. However, she admitted that the appellant did go off from time to time for a few minutes and that she did not know where he went. Given the proximity of the boys` toilet to the kitchen (they were adjacent to each other), there would have been time for the appellant to accost the first complainant without DW2 noticing it.

167 Equipment store at Mountbatten CC

168 The appellant adduced several witnesses, all of whom attested to the fact that Mai De Kang was the team quartermaster and in charge of equipment. This tended to show that it was not possible that the first complainant or the second complainant could have been appointed to fetch or return equipment. Many witnesses also testified that the appellant never entered the storeroom except on one occasion when a snake was reported to have been found inside. Mai, however, testified that there was quite a lot of equipment - eight sets of head gear and body protectors and 15 hand target pads. It must have been the case that he would occasionally seek assistance in fetching or returning equipment. This would have been consistent with the second complainant`s testimony that he returned equipment together with others and that, on the occasions he was molested by the appellant, he was simply the last one left in the storeroom. However, there having been only three incidents of molest over two years taking place in the storeroom, the appellant`s occasional absence (especially where the incident took place after training) would not have been sufficiently missed to have been noticed and remembered. Helpful as the evidence of these witnesses might have been, it could not have been conclusive of the appellant`s innocence. There had to be some other material defect in the prosecution`s case to create a reasonable doubt.

169 Date of third incident involving second complainant

170 Quite a lot of evidence was adduced by the appellant that he stopped teaching classes in December 1995 and that the class was taken over by Tommy Kok. This was intended to rebut the second complainant`s allegation that the third incident took place in 1996.

171 It was not disputed that DW3 assisted the appellant and occasionally took over classes. As far as Mountbatten CC was concerned, however, Tamilselvam s/o Kaliannan (PW6), the manager of Mountbatten CC, gave evidence that the appellant was the only instructor at the CC until the end of May 1996. PW6 stated that whilst it might have been possible for DW3 to take over lessons occasionally, it would not have been acceptable for him to take over classes completely from the appellant. The prosecution also highlighted that DW3 was not allowed to do so as he was not the holder of an instructor`s licence, having obtained his black belt only in January 1996.

172 I was also mindful of the fact that PW6 was a completely independent witness compared to DW3, who was a friend and former student of the appellant. DW3 also received remuneration when the appellant gave him the money credited into the appellant`s bank account for the lessons conducted by DW3 on the appellant`s behalf. The district judge therefore had adequate grounds on which to reject this aspect of the appellant`s case.

173 Conclusion

174 For the reasons set out above, I considered that the district judge was justified in convicting the appellant on all six charges and that the appellant`s appeal should accordingly be dismissed.

175 In view of the age of the victims and the fact that the appellant had abused the trust reposed in him by the complainants and their parents, I allowed the public prosecutor`s cross-appeal and ordered the sentence on all six charges to be increased from four to nine months with two of the sentences to run consecutively and the remainder to run concurrently for a total of 18 months` imprisonment. In so doing, I would add that I considered that the gender of the victims was not a factor to be taken into account and that the same standard should apply to victims of both genders. For this reason I adopted the benchmark of nine months` imprisonment which applies in cases of this kind.