1 Thangarajan Elanchezhian (“the Appellant”) was charged with one count of the offence of outrage of modesty under s 354(1) of the Penal Code (Cap 224, 2008 Rev Ed) (“Penal Code”) and claimed trial. He was convicted and sentenced to six months’ imprisonment. Dissatisfied with the decision of the learned District Judge (“the DJ”), the Appellant appealed against his conviction and sentence, contending that the DJ had erred in his assessment of the evidence and in his application of the relevant sentencing framework.

2 At the end of the hearing on 18 September 2024, I dismissed the appeal in its entirety as I found no basis to interfere with the decision of the DJ, both in respect of the conviction and the sentence. I now furnish the detailed grounds for my decision.

3 On 13 September 2021, between 1.30pm and 2pm, the complainant (who I will refer to as “PW1”) and the Appellant boarded Bus 242 from Boon Lay Bus Interchange (“the Interchange”). Bus 242 runs a loop service, which starts and ends at the Interchange. The entire route takes about 30 minutes.

4 At the time, PW1 was 16 years old. She was on her way home from school after sitting for an examination. The Appellant, a 42-year-old software engineer, was also on his way home after getting his second Covid-19 vaccination at around 12pm at a clinic called AcuMed Medical Group (“AcuMed”) in Jurong Point Mall (“Jurong Point”).

5 Upon boarding the bus, PW1 sat on the window seat at the second-last row on the driver’s side. There were several seats around her which were unoccupied; these included the two seats immediately in front of her, the two seats across the aisle from her, and four of the five seats behind her at the last row of the bus. After PW1 had taken her seat, the Appellant sat beside her on the seat to her left.

6 About ten minutes into the journey, the Appellant moved his right elbow outward and in an up-and-down motion. In doing so, his right elbow pressed against the side of PW1’s body, from her waist to the area just below her armpit. The Appellant does not deny that this occurred several times. According to PW1, the Appellant did so with some “pressure”, which she felt on the left side of her body (the “Elbow Contact”). While PW1 initially thought the Elbow Contact was accidental, she suspected it was deliberate when it persisted for several minutes. PW1 did not change seats or alight from the bus because she was afraid this would enable the Appellant to initiate more physical contact if she crossed in front of him to the aisle. As a result, she missed her intended stop. The Elbow Contact allegedly lasted for about ten minutes in all.

7 According to the Appellant, he had moved his elbow to relieve the pain in his arm caused by the vaccination. In his long statement to the police, the Appellant stated that he had moved his right elbow whilst seated on the bus to relieve the discomfort he felt in his right arm as a result of the vaccination. This was purportedly in keeping with the advice of the doctor who administered the vaccination (“Dr Vikram”) that the Appellant should “shake [his] right hand to [sic] every now and then to ease the pain”. He further claimed that as he bent his right elbow, it “started to shake uncontrollably”, causing it to rub against the left side of PW1’s body. At trial, the Appellant’s narrative changed slightly. While he persisted in his claim that any such contact was purely accidental, he asserted at trial that he was not even aware at the time that his right elbow had come into contact with PW1 at all.

8 PW1 on the other hand stated that, following the Elbow Contact, the Appellant then placed his right hand between his right thigh and her left thigh and started to use his finger to stroke her left lower thigh (the “Finger Contact”). He then allegedly proceeded to touch her left knee with his right hand (the “Knee Contact”). PW1 testified that the Knee Contact lasted for “less than 1 minute” before she said to the Appellant: “Can you please stop touching me?” The Appellant then stopped touching PW1 and moved his body away from her but remained in the seat beside her until the bus returned to the Interchange, where he alighted. PW1 subsequently started crying and then also alighted.

9 The Appellant denied that the Finger Contact had taken place. As for the Knee Contact, he testified at trial that his right hand had only made contact with PW1’s knee once. This happened by accident when he stretched his right arm by straightening his elbow. The Appellant explained that he had continued his journey back to the Interchange because he wanted to get some pain-relief tablets from the nearby Guardian outlet in order to alleviate the pain in his right arm. However, he eventually decided that the tablets (which cost $12) were too expensive and that he would instead obtain them from a clinic on the next day, using his Integrated Healthcare Solutions card. The Appellant then boarded Bus 242 again and returned home.

10 As for PW1, she called her close friend (“PW3”) when she alighted from the bus and told her what had transpired on the bus. PW3 then met PW1 at the linkway from Boon Lay MRT station to Jurong Point and accompanied her home on Bus 242.

11 After arriving home, PW1 took a nap but woke up after a nightmare. She decided to send her teacher (“PW2”) a message on WhatsApp to tell her what had happened. PW2 had been PW1’s secondary school teacher for about three years at that time and PW1 trusted her. PW2 comforted PW1 and encouraged her to file a police report and to tell her mother about the incident.

12 On the next day, PW2 accompanied PW1 to the police station to lodge a police report.

13 The Appellant was subsequently charged as follows:

14 For ease of reference, s 354(1) of the Penal Code provides:

15 The Appellant denied the charge. His principal claims were that the Finger Contact did not take place, and that the Elbow Contact and the Knee Contact were accidental. He also pointed to errors in Dr Vikram’s clinical notes and his own long statement which, in his view, rendered them inaccurate and unreliable.

16 These arguments were rejected by the DJ, who convicted the Appellant and sentenced him to six months’ imprisonment.

17 The DJ observed that two elements had to be proved by the Prosecution beyond a reasonable doubt for the charge to be made out:

In assessing the evidence, the DJ noted that he had to consider whether PW1’s evidence was “unusually convincing”, because the alleged acts of outrage of modesty were only witnessed by PW1, and so the evidence came down to the competing testimonies of the two witnesses.

18 On the facts, the DJ found that PW1 provided a cogent and consistent account of the alleged acts of outrage of modesty committed by the Appellant; and also accepted her explanation as to why she did not confront the Appellant earlier or change seats. Her distress following the incident was also evident from the testimonies of PW2 and PW3.

19 In contrast, the Appellant’s defence – that all contact was accidental and resulted from the stretching of his arm to ease the pain from the vaccination – was completely contradicted by Dr Vikram’s evidence. Dr Vikram testified that the vaccination had been administered on the Appellant’s left arm and this was supported by his clinical notes. There was no reason to think that Dr Vikram had not updated his notes accurately. The Appellant was also inconsistent in his assertion that he had been advised to “shake his hand” to ease the pain. He first attributed this to Dr Vikram but then subsequently to the staff at the clinic. Dr Vikram rejected this assertion and pointed out that this did not fall within the Ministry of Health’s guidelines on advice to be given to patients following their vaccinations. The DJ also agreed with the Prosecution that the Appellant could not adequately explain why he chose to sit beside PW1 when other seats were available and why he chose to return to the Interchange.

20 The DJ rejected the Appellant’s version of events and found that the Prosecution had proved its case beyond a reasonable doubt.

21 On sentence, the DJ applied the framework set out in Kunasekaran s/o Kalimuthu Somasundara v Public Prosecutor [2018] 4 SLR 580 (“Kunasekaran”). Under this framework, the court first considers the following offence-specific factors: (a) the degree of sexual exploitation, (b) the circumstances of the offence and (c) the harm caused to the victim. The court will then place the offence within one of three sentencing bands to determine the indicative starting sentence for the accused. As a final step, the court will consider any aggravating and mitigating factors relating to the offender and calibrate the sentence to be imposed.

22 The DJ placed the offence just above the lowest end of Band 2, which attracted a sentence of six months’ imprisonment. While the contact did not intrude upon any private parts of PW1’s body, the DJ thought that there were four aggravating circumstances:

The DJ thought there were no significant mitigating factors in this case to warrant a downward calibration of the sentence. The sentence of six months’ imprisonment, which was consistent with that meted out in Kunasekaran, was therefore appropriate.

23 The Appellant submitted that the DJ erred in law and in fact in convicting him of the charge. His arguments broadly fell within two categories.

24 First, he submitted that the DJ had erred in accepting Dr Vikram’s evidence, in particular that the vaccination was administered on the Appellant’s left arm. This was because Dr Vikram had vaccinated about 20 persons in two hours and it was accordingly open to doubt whether his clinical notes had been accurately updated.

25 Second, the DJ failed to take the serious errors in the Appellant’s long statement into account. Various particulars pertaining to the Appellant were wrongly recorded and the interpreter, Senior Investigation Officer Mohammad Hussein (“SIO Hussein”), did not countersign the statement as he was required to. The Appellant also pointed to the fact that although the Appellant was recorded as having said that his right elbow “started to shake uncontrollably”, it emerged at trial that SIO Hussein had in fact interpreted the Appellant’s words as “started to shake many times” but this was wrongly recorded by the investigation officer, Senior Staff Sergeant Nurulain Binte Mohamed Rafie (“IO Nurulain”). These errors, taken together, raised doubts as to the accuracy of the long statement.

26 If the Appellant’s arguments on conviction were not accepted, the Appellant submitted that the circumstances surrounding the charge warranted the imposition of a four-month imprisonment term instead of the six months imposed by the DJ. Applying the Kunasekaran framework, it was submitted that the offence fell within Band 1, which warrants a starting sentence of less than five months’ imprisonment, instead of Band 2. This is because the degree of sexual exploitation was very low and there was no harm caused by the Appellant, no premeditation, no use of force or violence, no abuse of a position of trust and no deception. Given the lack of any egregious aggravating factors, a four-month imprisonment term would be appropriate.

27 The Prosecution, on the other hand, submitted that the appeal should be dismissed.

28 The Prosecution maintained that the conviction was sound. First, the DJ did not err in accepting Dr Vikram’s evidence and rejecting the Appellant’s version of events. Dr Vikram was a non-partisan medical professional who had no reason to lie. There was no reason to think that Dr Vikram had failed to update his clinical notes accurately because that was a part of his ordinary practice. The accuracy of Dr Vikram’s clinical notes – which stated that the vaccination was administered on the Appellant’s left arm – was buttressed by his undisputed evidence that he had administered vaccinations on the Appellant’s left arm on two other occasions. As against this, the Appellant was unable to explain why he had purportedly asked for the vaccination to be administered on his right arm on just this occasion.

29 Second, while the Prosecution accepted that there were errors in the recording of the Appellant’s long statement, this did not detract from the fact that the substantive content of the statement had been accurately recorded. This was confirmed not only by IO Nurulain and SIO Hussein, but also by the Appellant himself. The DJ was therefore correct to rely on the long statement and to consider that the material inconsistencies between it and the Appellant’s testimony at trial undermined the Appellant’s credibility.

30 As for the sentence, the Prosecution submitted that six months’ imprisonment was just and appropriate. The DJ had correctly applied the sentencing framework in Kunasekaran. Cases which fall within Band 1 include cases that do not present any, or at most one, of the offence-specific factors. Here, the Prosecution pointed to three: (a) the vulnerability of PW1, as a young girl who was not in a position to resist or protest effectively, (b) the duration of the offending period and (c) the fact that the offence took place on public transport. The DJ was therefore correct in finding that the present case fell slightly above the lowest end of Band 2. The sentence imposed by the DJ was also consistent with the precedents.

31 Two issues arose for my determination:

32 It was clear to me, having carefully considered the parties’ submissions and the evidence, that the DJ had correctly found that the Appellant had intentionally initiated the Elbow Contact, the Finger Contact and the Knee Contact.

33 Before me, the Appellant did not contest the DJ’s finding that the Elbow Contact, the Finger Contact and the Knee Contact had taken place. Instead, the crux of his defence was that all contact was purely accidental, and a result of his attempt to alleviate the pain in his right arm caused by the vaccination. The immediate and considerable obstacle that the Appellant faced in trying to advance this contention was the evidence of Dr Vikram, which was that the vaccination had been administered on the Appellant’s left arm. Indeed, counsel for the Appellant, Mr R Kalamohan, candidly and very fairly accepted at the hearing that if I saw no reason to reject Dr Vikram’s evidence in this regard, the Appellant’s entire defence would crumble.

34 It was not disputed that Dr Vikram’s evidence was based on his contemporaneous clinical notes and not on his memory. I could not find any reason for thinking that Dr Vikram might have failed to update his clinical notes accurately. According to Dr Vikram, it is his practice to update his clinical notes immediately after a patient leaves his consultation room and before the next patient enters. He also testified that he administered vaccinations on his patients’ left arm by default, because their left arm would be closer to him given how they would be seated in the consultation room. He would only administer a vaccination on a patient’s right arm if they specifically asked him to do so.

35 The Appellant also accepted that Dr Vikram had, on two other occasions, administered Covid-19 vaccinations on his left arm. The Appellant was not able to offer a sensible explanation as to why he had allegedly asked for the vaccination on 13 September 2021 to be administered on his right arm. This was also incongruous in that the Appellant was right-handed, and if he was concerned about pain, it would not make sense for him to specifically ask that the vaccination be administered on his dominant arm.

36 I was satisfied in the circumstances that there was nothing to suggest that Dr Vikram’s clinical notes had been incorrectly recorded. It followed that the DJ’s finding that the Appellant had in fact been vaccinated on his left arm could not be said to be against the weight of the evidence and that effectively was the end of the Appellant’s defence.

37 Even beyond this, the surrounding facts clearly support this conclusion. First, it was undisputed that PW1 had to ask the Appellant to stop touching her. It is not normal for one to consciously come into physical contact with a stranger when this is entirely avoidable. There were several other seats available where the Appellant could have sat and stretched his allegedly hurting arm as much as he wished without disturbing anyone else. It struck me as incredible that he would choose to sit next to a stranger and “accidentally” be in repeated physical contact with her over a prolonged period. It is also clear that by the time PW1 plucked up the courage to tell him to “stop touching” her, she was convinced he was doing this intentionally. That is not the way one puts it if one thinks there has only been accidental and harmless contact. That, to my mind, was a clear indication that PW1 was convinced that the Appellant had deliberately made contact with her body given the duration and escalation of the contact. And significantly, the Appellant did not challenge PW1 or deny that he had been touching her, when she confronted him.

38 Further, even after PW1 asked him to stop touching her, the Appellant did not change his seat. If the contact had indeed been purely accidental and a consequence of the pain he was in, it seemed incredible to me that he would remain in the seat and continue his journey all the way back to the Interchange, supposedly to get medication to alleviate the pain, while enduring the discomfort of not moving his arm. The Appellant’s decision not to move to another seat was also particularly telling given that he would have known at this point that he had caused (and indeed was continuing to cause) PW1 distress.

39 Finally, the Appellant’s ostensible concerns over the inconsistencies in his long statement were overstated. While it is true that certain particulars (such as the Appellant’s date of birth, nationality and handphone number) were incorrectly recorded and SIO Hussein had failed to countersign the statement, the Appellant did not deny that he understood the questions which had been posed to him by IO Nurulain, that the statement was read back to him in Tamil after it was recorded and, pertinently, that the substantive contents of the statement were accurately recorded. Both IO Nurulain and SIO Hussein were also called as witnesses at trial and cross-examined by counsel for the Appellant. In these circumstances, the DJ had not erred in placing weight on the contents of the statement despite the minor errors and procedural missteps. Simply put, the Appellant was unable to point to any material aspects of the long statement which were inaccurate and which therefore rendered the statement unreliable.

40 For these reasons, I dismissed the appeal against the conviction.

41 I was equally of the view that the DJ had correctly applied the sentencing framework in Kunasekaran, which is applicable to offences under s 354(1) of the Penal Code: see Public Prosecutor v Tan Chee Beng and another appeal [2023] SGHC 93 (“Tan Chee Beng”) at [144].

42 Before addressing the Appellant’s arguments on sentence, I observe that the maximum imprisonment term for an offence under s 354(1) was increased with effect from 1 March 2022 from two years’ imprisonment to three years, pursuant to the Criminal Law (Miscellaneous Amendments) Act 2021 (Commencement) Notification 2022. This was not material to the present appeal because the offence was committed before the amendment came into effect. For offences committed on or after 1 March 2022, the relevant sentencing ranges for each band may need to be readjusted to account for the increase in the maximum sentence: see for example, Public Prosecutor v Wong Teck Guan [2023] SGMC 64 at [48]–[55].

43 As outlined at [21] above, under the Kunasekaran framework, the court will first place the offence within one of three sentencing bands based on its consideration of the relevant offence-specific factors to arrive at an indicative starting sentence or range for the case at hand:

(a) Band 1 includes cases that do not feature any, or at most one, of the offence-specific factors, and typically involves cases that involve a fleeting touch or no skin-to-skin contact, and no intrusion into the victim’s private parts. This would attract a sentence of less than five months’ imprisonment.

(b) Band 2 includes cases where two or more offence-specific factors are present. The lower end of the band involves cases where the private parts of the victim are intruded upon, but there is no skin-to-skin contact. The higher end of the band involves cases where there is skin-to-skin contact with the victim’s private parts. It would also involve cases where there was some deception on the part of the offender. This would attract a sentence of five to 15 months’ imprisonment.

(c) Band 3 includes cases where there are numerous offence-specific factors, especially those such as the exploitation of a particularly vulnerable victim, a serious abuse of a position of trust, and/or the use of violence or force on the victim. This would attract a sentence of 15 to 24 months’ imprisonment.

The court will then consider any aggravating and mitigating factors relating to the offender and calibrate the sentence accordingly.

44  In my judgment, the DJ was correct to assess the offence in this case as falling at the lower end of Band 2. This was because the offence here clearly involved three offence-specific factors.

45 First, the offence involved a young and vulnerable victim. Not only had the Appellant essentially cornered PW1 within a tight space from which she could not easily extricate herself, but he also escalated his offending conduct when he realised that she was afraid or unable to remove herself from the situation or seek immediate help.

46 Second, the offence took place on public transport. This is well-recognised as an aggravating offence-specific factor: see Kunasekaran at [58].

47 Third, the offence caused PW1 to suffer a degree of emotional harm. This was evident from PW1’s nightmare, the panic attack she experienced while the offence took place and the fact that she cried several times after the incident. The absence of a victim impact statement was not an impediment to the DJ’s finding that PW1 had suffered emotional harm.

48 It followed that the DJ was correct to place the offence at the lower end of Band 2. I therefore dismissed the appeal against sentence.

49 Before leaving this point, I should clarify the types of cases which may fall within Band 2 of the framework. In Kunasekaran (at [45(b)(ii)]), the court elaborated on the scope of Band 2 in these terms:

50 This might suggest that a case might fall within Band 2 only if it involves an intrusion of the victim’s private parts. I do not consider that was what the court in Kunasekaran intended. In my judgment, there are a number of disjunctive classes of cases encapsulated in that formulation of Band 2. The first is where there are two or more offence-specific factors. The second is where the intrusion concerns the victim’s private parts, but without skin-to-skin contact. The third is where there is skin-to-skin contact with the victim’s private parts. Finally, there are cases involving deception possibly with one or the other of the factors already noted. It is therefore altogether possible for a case to fall within Band 2 even though it does not involve an intrusion upon the victim’s private parts.

51 This position is also borne out by the court’s application of the Kunasekaran framework in other cases. For example, in Tan Chee Beng, Vincent Hoong J placed the offence at the lower end of Band 2 even though it was not apparent that the victim’s private parts had been intruded upon.

52 This disposed of the present appeal. But as I reviewed the evidence led at the trial in the course of preparing for this appeal, it struck me that some aspects of the cross-examination of PW1 invited close attention. These included questions that concerned various areas. I focus on one, which was why PW1 did not seek immediate help from the bus driver when the incident occurred. Given the heightened sensitivities at play when a complainant of a sexual offence gives evidence in court, it is imperative that the process of cross-examination be approached with greater care. I therefore take this opportunity to provide some further guidance on the need for the court to manage the examination and cross-examination of the complainant in trials of sexual offences. For the avoidance of doubt, I adopt the definition of sexual offences that is set out in s 2 of the Criminal Procedure Code 2010 (2020 Rev Ed) (“CPC”).

53 The broad principles which govern this area have been helpfully set out by Vincent Hoong J in his recent decision in GHI v Public Prosecutor [2024] 5 SLR 607 (“GHI”):

54 The permissible scope of questioning, as pointed out by Hoong J (at [74]–[76]), is demarcated by the relevant provisions in the Evidence Act 1893 (2020 Rev Ed) (“Evidence Act”), which, for example, prohibits questions that appear to be needlessly offensive in form or intended to insult or annoy, as well as the Evidence (Restrictions on Questions and Evidence in Criminal Proceedings) Rules 2018 (“Evidence Rules”), which prohibit counsel from asking a complainant about his or her sexual behaviour or physical appearance without the permission of the court.

55 Hoong J’s observations in GHI (particularly at [77]) correctly recognise the tensions that uniquely underlie the cross-examination of complainants of sexual offences. Victims of such offences – which are committed against not only their person but also their dignity – often describe cross-examination as the most distressing part of their experience within the criminal justice system. Indeed, it is not uncommon for these victims to report that the suspicion and disbelief they encounter in the course of cross-examination can feel like a repeat of the trauma pertaining to the sexual offence itself – a phenomenon sometimes referred to as “secondary victimisation”: see Sarah Zydervelt et al, “Lawyers’ Strategies for Cross-Examining Rape Complainants: Have We Moved Beyond the 1950s?” (2017) 57(3) The British Journal of Criminology 551 (“Zydervelt”) at 553. Complainants of sexual offences should be protected from undue and insensitive harassment in cross-examination as much as possible; and they should not be penalised for speaking out against their alleged attackers, much less be forced to relive their trauma.

56 Such protection must, however, be balanced against the accused person’s right to a fair trial, which necessarily includes the right to challenge the complainant’s testimony. This is of especial importance in the context of cases involving sexual offences, where the complainant’s testimony will often be the central (and perhaps only) evidence against the accused person. It is for this reason that courts have generally recognised a need for close scrutiny of allegations of sexual abuse, given the ostensible ease with which such allegations may be fabricated and the concomitant difficulty of rebutting such allegations (see for example, AOF v Public Prosecutor [2012] 3 SLR 34 at [112]). Accordingly, the court must be careful not to place unwarranted constraints on cross-examination, so as to ensure that the accused person’s due process rights are preserved.

57 Striking the right balance between an accused person’s right to a fair trial and the interests of a complainant of a sexual offence in cross-examination is not an easy task. While the parameters of cross-examination are defined by the existing statutory framework, the relevance or permissibility of a particular question or line of questioning will necessarily require fact-specific consideration. It will depend on matters such as the complainant’s narrative, the specific facts of the case, the defence that is being run at trial and even the manner in which the questions are being put to the complainant. What this means is that the attempt to set out granular rules which dictate how complainants of sexual offences ought to be questioned in cross-examination will generally be unhelpful and unworkable.

58 Instead, a more sensible approach would be for the court to take on a more active supervisory role in managing the giving of evidence by complainants of sexual offences. This, in my view, is a fundamental component of the court’s duty to engage in active case management. The prevailing judicial philosophy of our courts places a heavy emphasis on judicial case management (see U Myo Nyunt (alias Michael Nyunt) v First Property Holdings Pte Ltd [2021] 2 SLR 816 at [57]). This is evident in the civil context with the implementation of the Rules of Court 2021, which sought to accord the courts greater control over the litigation process (see Civil Justice Commission Report (29 December 2017) (Chairperson: Tay Yong Kwang J) at para 1), and in the criminal context by the increasing number of tools at the court’s disposal to better manage proceedings. For example, the amendments introduced by the Criminal Justice Reform Act 2018 (Act 19 of 2018) allow the court, in appropriate situations, to implement shielding measures for vulnerable victims and to order the evidence of such individuals to be given in camera. It is clear that while case management processes may be procedural in form, they advance objectives which extend beyond concerns of process. At its core, case management is an exercise undertaken by the court to administer expeditious and efficient justice by ensuring that matters which come before it are handled in a manner which is procedurally fair: see Cockerill v Collins [1999] 2 Qd R 26 at 28. Given the attendant implications on procedural justice, which in turn impinges upon the parties’ right to a fair trial (see V Nithia (co-administratrix of the estate of Ponnusamy Sivapakiam, deceased) v Buthmanaban s/o Vaithilingam and another [2015] 5 SLR 1422 at [39]), it follows that every judge has not only the prerogative, but indeed the duty to properly manage the cases before him or her.

59 The approach a judge takes to effective case management will necessarily be influenced by the particular facts and issues in the case, and how the judge sees these issues and their importance. It is not my intention to curtail the judicial discretion that is an integral aspect of this process. But I nonetheless think it may be useful to set out some broad guidelines on how this might be approached.

60 It bears emphasising that this duty (which in this context is ultimately concerned with the questioning of complainants of sexual offences in cross-examination) is engaged even before the trial commences. At this pre-trial stage, the judge conducting the trial should consider convening a pre-trial conference with the Prosecution and the Defence to identify the main issues in contention and, where possible, to elicit the general nature of the defence. I am mindful of the observations of the Court of Appeal in Muhammad Nabill bin Mohd Fuad v PP [2020] 1 SLR 984 (“Nabill”), in particular at [168] and [173]. It is indeed the case that before the accused person is called on to present his defence, the Prosecution must first discharge its burden by establishing a prima facie case. While the Defence may not be compelled to depart from this position, as a matter of sensible case management, it may well be the case that the Defence does not contest certain factual positions. This can be seen in the fact that there may at times be a statement of agreed facts or an agreement for certain pieces of evidence to be admitted. At times the accused person’s own statements may be admitted as part of the Prosecution’s case. The short point is that without derogating from the rights of the Defence, as set out in Nabill, there may be considerable scope for the court to clarify and establish the areas that are in dispute.

61 To the extent this can be done, it will help focus the parties’ attention on the points of potential relevance and significance at the trial and, on this basis, inform their preparations for the cross-examination of the complainant. Such a discussion would provide an opportunity for the judge to remind the parties of the unique tensions at play where sexual offences are concerned and of the duties of counsel in cross-examination (see [75] below). The judge, having read the relevant papers, should in any case be in a position to flag areas where care will be especially needed to avoid unfair or inappropriate questioning.

62 Assuming the criminal case disclosure procedures apply (see ss 159 and 211A of the CPC), the pre-trial conference should be convened after the deadlines for the service of the relevant documents (namely, the Case for the Prosecution, the Case for the Defence and the copies of statements, exhibits and/or criminal records). At this point, the parties will have attended the criminal case disclosure conferences, at which the issues of fact or law which are to be tried by the judge conducting the trial would have been identified with the benefit of the disclosure of the relevant documents (see ss 160(1)(b) and 163(2) of the CPC). At the pre-trial conference, therefore, parties should be able to apprise the court of the issues which will need to be resolved at trial with a reasonable degree of specificity.

63 Where the criminal case disclosure procedures do not apply, the pre-trial conference should be convened after trial dates have been fixed. While the issues in these cases will likely be crystallised to a less defined extent than in cases governed by the criminal case disclosure procedures, the parties may nevertheless consider agreeing on the broad issues in contention as far as possible. At the very least, this will provide the judge and the parties with some indication as to the potential areas of relevance in cross-examination. I emphasise that such pre-trial conferences are distinct from the case conferences provided for in ss 171 and 220A of the CPC, which are primarily aimed at settling administrative matters pertaining to the trial and need not be presided over by the judge conducting the trial.

64 To the extent there has been any agreement on these points at the pre-trial stage, the judge’s duty of active case management will extend to framing the contentious points which are likely to be explored in cross-examination before the commencement of the trial, in consultation with counsel. This would enable the judge subsequently to shut out lines of questioning which are either completely irrelevant to the issues identified at the pre-trial stage (unless the position has, for some legitimate reason, changed) or which are clearly intended to insult or annoy the complainant. In this way, the complainant may be shielded from the effects of questions which are unwarranted or inappropriate. The significance of this should not be understated, given that some mischief may be done the moment inappropriate questions are posed to the complainant, even if the judge ultimately disallows it (see Sudipto Sarkar & V Kesava Rao (eds), Sarkar on the Law of Evidence: In India, Pakistan, Bangladesh, Burma, Ceylon, Malaysia & Singapore (Vol 2) (LexisNexis, 20th Ed, 2021) at p 3136).

65 At the trial itself, the judge will have to assess the permissibility of each question or line of questioning as it is being posed to the complainant in cross-examination. In this regard, a two-step framework, which is premised on the applicable rules in the Evidence Act and the Evidence Rules, may be instructive.

66 First, the judge should determine whether the question or line of questioning relates to facts in issue or matters that have to be dealt with, in order to determine the facts in issue. The focus at this stage is on relevance, which is the threshold requirement for questions which may be posed to a witness in cross-examination. It is for the party asking the questions to demonstrate their relevance. Evidence can only be given of relevant facts, and it must therefore be incumbent on the examining party to justify how the question is relevant, and to that end it may have to explain, at least broadly, what its case is.

67 To this end, s 140(2) of the Evidence Act states that the cross-examination of a witness must relate to relevant facts, even though the cross-examination need not be confined to the facts to which the witness has testified in his or her examination-in-chief. In assessing relevance, the judge should test the question or line of questioning against the broad issues and the areas of potential questioning that might have been identified by the parties at the pre-trial stage and before the commencement of the hearing. Relevance may also pertain to matters touching on the witness’s veracity or credibility, the accuracy of his or her evidence, his or her position in life and his or her character (see s 148 of the Evidence Act). Where necessary, the judge should seek clarification from counsel as to the direction and/or aim of a specific question or line of questioning and its relevance to the identified issues.

68 Second, if the particular question or line of questioning is found to be relevant, the judge should go on to consider whether it nevertheless falls within a specific prohibition in the Evidence Act or the Evidence Rules. For example, ss 3 and 4 of the Evidence Rules, read with s 154A of the Evidence Act, prohibit questions pertaining to the complainant’s sexual behaviour or physical appearance unless otherwise permitted by the court. Such questions must therefore be entirely disallowed by the judge unless an application for permission has been made in the absence of the complainant of the offence, and been allowed.

69 Also of particular significance in this context is s 154 of the Evidence Act, which prohibits any question which appears to the court to be intended to insult or annoy, or which though proper in itself, appears to the court to be needlessly offensive in form. In deciding whether the prohibition in s 154 applies, the judge will need to determine the underlying aim of the particular question or line of questioning which is being posed to the complainant. This prohibition is easy to apply in extreme cases. For example, in Law Society of Singapore v Wong Sin Yee [2018] 5 SLR 1261 (“Wong Sin Yee”), counsel for the accused person had cross-examined the complainant on her physical appearance and sought to establish a correlation between her attractiveness and the accused person’s “motive” to molest. As pointed out by the Court of Three Judges, these questions were clearly intended to insult or annoy the victim and were needlessly offensive in form (at [43]–[45]).

70 However, the scope of permissible cross-examination in many other cases involving the alleged commission of sexual offences may well fall in an area where it is not immediately apparent whether specific questions or lines of questioning are indeed intended to insult or annoy. This case (as I elaborate later at [80]) presents one such example. In this specific context – where a complainant of a sexual offence is cross-examined – I consider that questions or lines of questioning which rely on a harmful stereotype or are otherwise based upon or intended to further a harmful stereotype will generally fall foul of s 154 of the Evidence Act. These harmful stereotypes, also known as “rape myths”, refer to unjustified beliefs about the context, causes and consequences of sexual violence which serve to deny, downplay or justify such behaviour (see Zydervelt at 553).

71 The need to avoid the perpetuation of these harmful stereotypes has been recognised not only by our courts (see for example, GHI at [73]), but also by Parliament. Indeed, this was the impetus for the enactment of the Evidence Rules, which sought to tackle the misogynistic view that a complainant’s sexual behaviour, sexual history or physical appearance may have induced the commission of a sexual offence against him or her (Singapore Parliamentary Debates, Official Report (19 March 2018) vol 94 (Indranee Rajah, Senior Minister of State for Law)). Reliance on these stereotypes in the course of cross-examination re-traumatises victims of sexual offences by unfairly insinuating responsibility on their part for the offences and may, in doing so, negatively affect the recovery and treatment for such victims (Jessica Woodhams et al, “Behavior Displayed by Female Victims during Rapes Committed by Lone and Multiple Perpetrators” (2012) 18(3) Psychology, Public Policy, and Law 415 at p 445). This may also discourage other victims from reporting offences committed against them, thereby creating a culture of blaming and consequently of silencing victims. If, therefore, a question relies on a harmful stereotype or is otherwise based upon or intended to further such a stereotype, this may be a strong indication that the question is intended to insult or annoy.

72 Examples of these harmful stereotypes include the following:

(a) That victims of sexual offences will inform someone or report the attack immediately. Research has shown that the timing at which sexual abuse is disclosed or reported is, by itself, inconclusive as to the credibility of the complainant (see Elaine Craig, “The Ethical Obligations of Defence Counsel in Sexual Assault Cases” (2014) 51(2) Osgoode Hall Law Journal 427 at p 433). A myriad of factors affect disclosure; these include the victim’s age, gender, relationship to the perpetrator and cultural and religious influences (see Mitru Ciarlante, “Disclosing Sexual Victimization” (2007) 14(2) The Prevention Researcher 11 (“Ciarlante”) at pp 11–12). Further delays may be occasioned by fears of retaliation or harm as well as the shock, shame and stigma attached to being a victim of such offences (Ciarlante at p 13).

(b) That victims of sexual offences will react in a uniform, specific or predictable manner and that differences in victims’ responses to the offences are probative of their credibility or lack thereof. It is now widely accepted that there is no one “typical” emotional or behavioural reaction that is expected to be exhibited by victims of sexual offences (see GBR v Public Prosecutor and another appeal [2018] 3 SLR 1048 at [20]). For example, rape victims display a diverse range of reactions in the immediate aftermath of the offence: some display great distress while others are quiet and controlled. It is therefore important to avoid making generalisations about how a victim should have responded to an offence (see Dean Kilpatrick, Patricia Resick and Lois Veronen, “Effects of a Rape Experience: A Longitudinal Study” (1981) 37(4) Journal of Social Issues 105 at p 119).

(c) That the sexual history of a complainant of a sexual offence bears on the likelihood of his or her having consented to sexual activity and/or his or her credibility. This myth has been recognised and rejected by the courts in Australia (see for example, Bull v R [2000] HCA 24 at [47]) as well as in Canada (see for example, R v Seaboyer; R v Gayme [1991] 2 SCR 577 at [164]–[167]).

(d) That victims of sexual offences who dress in a certain way invite the commission of the offences (see GHI at [69]).

(e) That victims of sexual offences who were voluntarily intoxicated at the time of the offence are at least partly responsible for the commission of the sexual offence. The mere fact that a victim’s actions has put him or her at a greater risk of victimisation does not make him or her responsible for the commission of the offence (see “Rape Myths and ‘Rational’ Ideals in Sex Offence Trials” in Olivia Smith, Rape Trials in England and Wales (Palgrave Macmillan Cham, 2018) at p 58).

(f) That victims of sexual offences are able to resist attackers if they really want to (see Singapore Parliamentary Debates, Official Report (19 March 2018) vol 94 (Patrick Tay Teck Guan, Member of Parliament)). On the contrary, many victims enter an involuntary temporary state of inhibition when sexual offences are committed against them, causing them to “freeze” in intense fear (see Public Prosecutor v GCK and another matter [2020] 1 SLR 486 at [111]).

73 The heuristic I have outlined (namely, that questions which rely on or perpetuate harmful stereotypes may be seen as generally intended to insult or annoy) is not intended to function as a legal test to trigger the prohibition in s 154 of the Evidence Act. Its aim is instead to serve as a tool to alert the court to questions which may, in essence, be relying on a premise which is not only false but also (as a result) disparaging to the complainant. In other words, it is but one factor (albeit one of significant weight) to be considered by the judge in his or her assessment as to whether a question or line of questioning falls foul of s 154 of the Evidence Act based on a holistic assessment of the circumstances. In this regard, the judge may also consider other aspects pertaining to the questioning of the complainant, such as the specific phraseology and tone of the questions, the demeanour of counsel and the overall duration of the cross-examination process. The judge should be especially mindful of the need to avoid prolix, unfocused and repetitive cross-examination. Any relevant condition, characteristic or disability of the complainant may also be taken into account. As a rule, the judge should not permit the complainant to be subjected to the strain brought about by extended hostile or pointless questioning and poorly managed court proceedings.

74 At trial, therefore, the judge overseeing the cross-examination of a complainant of a sexual offence should satisfy himself or herself that the questions or lines of questioning posed to the complainant are relevant, do not fall within the statutory prohibitions set out in the Evidence Act and the Evidence Rules, and are generally within the bounds of reason in terms of their manner, duration and focus.

75 Of course, the onus to ensure that the cross-examination of complainants of sexual offences is conducted in an appropriate manner does not fall on the court alone; counsel is obliged to be acutely cognisant of their duties to the witness and the court in this regard. As observed in Wong Sin Yee at [26], a lawyer’s ethical duties in their conduct of cross-examination are informed by both the Evidence Act (at ss 140 to 154A) and the Legal Profession (Professional Conduct) Rules 2015 (“PCR”). Rule 12(5) of the PCR further provides that an advocate and solicitor shall not make any statement, or ask any question, which is scandalous, is intended or calculated to vilify, insult or annoy a witness or any other person, or is otherwise an abuse of the function of the legal practitioner. These duties, whilst owed to the witnesses who are the subject of cross-examination, are integral to the lawyers’ overriding duty to the court and the administration of justice (see Wong Sin Yee at [35]).

76 Counsel must therefore seek to challenge the evidence of the complainant in a manner which is respectful not only in form but also in substance. They should seek to craft their questions in cross-examination in line with the framework set out above to ensure that these questions are not only relevant but also appropriate in such contexts. Counsel would also be well-advised to consult the “Best Practices Toolkit” for the cross-examination of vulnerable witnesses (the “Toolkit”) prepared by the Law Society of Singapore on 21 August 2019. This Toolkit reminds counsel not only of their specific duties under the Evidence Act and the PCR, but also of techniques such as maintaining appropriate body language, tone and eye contact to put the complainant at ease during cross-examination.

77 Ultimately, it is the responsibility of counsel to exercise his or her own judgment to perform cross-examination in a respectful and honourable manner – this duty is not lessened in any way by the existence of the court’s powers to forbid irrelevant or improper questions (see Wong Sin Yee at [34]).

78 In that light, I illustrate how this might (or ultimately, depending on counsel’s explanation, might not) have impacted the cross-examination below.

79 At the pre-trial stage, the DJ should first have identified what the main issue in dispute between the parties was. This was whether the Appellant possessed the necessary mens rea to make out the offence under s 354(1) of the Penal Code. The thrust of the Appellant’s defence, which was maintained on appeal, was not that the contact did not take place or that it was consensual, but that it was purely accidental in nature. This should have been the basis on which parties should have prepared for cross-examination.

80 At the trial, the application of the two-step framework (as set out above at [66]–[68]) would have alerted the DJ to lines of questioning which were, likely, impermissible. An example of such a line of questioning pertained to the reason why PW1 did not seek help on the bus immediately after the incident took place:

81 At the first stage of the framework, the foregoing line of questioning might well have failed to cross the threshold of relevance. As would have been established at the pre-trial stage, the main plank of the Appellant’s defence was that the touching was accidental. However, this line of questioning did not go towards establishing whether the touching was accidental or intentional. Counsel for the Appellant did not suggest that this failure to seek help immediately was an indication that PW1 viewed the touching as accidental – this was not and could not have been the Appellant’s case as he did not dispute that PW1 had explicitly told him to stop touching her (see [37] above). Nor was it ever suggested, and indeed, it could not have been, that the touching was consensual. As such, this line of questioning seems to me to have been irrelevant to the issues in dispute at trial.

82 Further, at the second stage, this line of questioning might have been caught under the prohibition in s 154 of the Evidence Act because it sprang from the assumption that victims of sexual offences, such as PW1, would report the offence immediately and/or react in a uniform manner, that being what they were taught in school. It also appeared to cast blame on PW1 for the incident by insinuating that she had failed to learn from other outrage of modesty cases which had been reported in the newspapers or featured on the news. This amounted to the perpetuation of harmful stereotypes (see [72(b)] above). Even if counsel for the Appellant wanted to make the point that PW1’s evidence was inconsistent with what she had been taught at school – which in any case would not have been directly relevant to the issues at hand – there was no need to press PW1 on the reasons why she had failed to act in a manner which she ostensibly must have known was “correct” or “reasonable”. To do so would be to ignore the prevailing understanding that victims react to the occurrence of sexual offences in different ways and to insinuate that it was PW1’s fault that she did not react in the “right” manner. Accordingly, the DJ could have disallowed that line of questioning to proceed beyond establishing the inconsistency between what PW1 was taught and how she in fact reacted. The DJ should have pointed out to counsel that his focus on the reasons why PW1 reacted the way she did had no real impact on the Appellant’s defence and sought to clarify the relevance of the questioning there and then. This could have saved PW1 from unnecessary distress while affording counsel with the opportunity to explain his purpose for pursuing the point.

83 In every criminal case the court must consider a triangulation of interests – those of the accused person, the victim and the public (see Attorney General’s Reference (No 3 of 1999) [2001] 2 WLR 56 at 63). Striking the right balance between these interests is crucial to the dispensation of justice, because that is when the court upholds the accused person’s fundamental right to a fair trial, the victim’s right to give his or her evidence without undue harassment and the public’s faith in the criminal justice system. Where sexual offences are concerned, the interests of the complainant should be accorded greater attention, specifically in the course of cross-examination. This is because the nature of such offences magnifies the harm which irrelevant and inappropriate questions may inflict. Such questions will not only annoy or irritate the victim but may also cause her to blame herself for the occurrence of the offence or question her recollection of the incident. This may in turn result in further psychological harm being inflicted on the victim, which may last long after the trial has concluded.

84 The court must, as part of its duty of active case management, exercise its oversight over the entire process through which a complainant gives evidence, bearing in mind the competing interests identified above.

85 At the same time, counsel too must do the needful to ensure that cross-examination is carried out in a respectful and focused manner, thereby fulfilling their ethical duties and upholding the integrity of the judicial process.

86 Finally, I do not think these remarks can readily apply in the admittedly rare category of such cases where the accused person is unrepresented. That was not the nature of the case before me, and it is therefore not appropriate for me to make observations in that regard. But it does seem to me to be potentially problematic if a self-represented accused person were to cross-examine the victim. In my view, this may warrant special provision being made either as a matter of legislative amendment or by way of rules of criminal procedure.

87 For the reasons set out in these grounds, I dismissed the Appellant’s appeal against conviction and sentence. I am grateful to Mr R Kalamohan, counsel for the Appellant, and Mr Niranjan Ranjakunalan, who appeared for the Prosecution, for their helpful submissions not only on the merits of the appeal but also on the scope of questions which may be appropriately directed to complainants of sexual offences, which were instructive in the formulation of the framework above.