1 The accused was a private hire vehicle driver at the material time. In the early hours of 19 May 2018, he drove the complainant home after she went drinking at a bar with friends. Certain sexual acts occurred between the accused and the complainant in the car after they arrived at the complainant’s condominium (the “Condominium”). These acts formed the basis of the three charges against the accused, as follows:

(a) one charge of sexual assault by digital-vaginal penetration under s 376(2)(a) of the Penal Code (Cap 224, 2008 Rev Ed) (“PC”), punishable under s 376(3) PC (the “first charge”);

(b) one charge of attempted rape under s 375(1)(a) PC, punishable under s 375(2) read with s 511 PC (the “second charge”); and

(c) one charge of outrage of modesty under s 354(1) PC (the “third charge”).

2 The accused claimed trial to all three charges. He admitted to committing the sexual acts which formed the physical elements of these offences (the “Sexual Acts”), but claimed that the Sexual Acts had been consensual.

3 I acquitted the accused of all charges and provided brief oral grounds. The Prosecution has since appealed against my decision. These are my full grounds.

4 On 18 May 2018, the complainant had been drinking at a bar (the “Bar”) with two friends, PW18 Yong Sern (“LYS”) and PW19 Vanessa Tam (“VT”) from about 10:30pm. ¹ LYS was the complainant’s friend, and VT was LYS’s girlfriend at the material time. ² The complainant had taken dinner earlier that evening and, whilst at the Bar, she consumed five pints of beer over a period of approximately three hours.

5 The complainant left the Bar at around 2.30am on 19 May 2018. By then, she was visibly intoxicated. Her steps were unsteady. ³ She also made a number of trips to the toilet to attempt to vomit. At trial, she explained to the court that this was not because she was nauseous or felt like vomiting, but because she was deliberately trying to make herself vomit so that she would feel more comfortable. ⁴ She also started crying at about the time of her final visit to the toilet. At trial, the complainant told the court that she remembered crying but could not remember why she was crying. ⁵

6 LYS offered a few times to send the complainant home, but she persistently declined and repeatedly reassured LYS that she was okay. ⁶ At trial, the complainant explained to the court that she responded to LYS’s offers in this way because she did not want LYS to worry and also did not want to trouble LYS and VT. ⁷

7 LYS booked a private-hire car for the complainant. The accused’s car (the “Car”) responded to the booking. As the complainant and LYS exited the Bar to walk towards the Car, LYS held the complainant’s arm to steady her. ⁸ They let go of each other when they were about ten metres away from the Car. ⁹ The complainant continued walking to the Car on her own, unassisted. ¹⁰ She also had the awareness to collect her belongings from VT without prompting as she walked past VT on the way to the Car. ¹¹

8 LYS informed the accused that the complainant was a “a little drunk” and asked the accused if he was okay with sending the complainant home. ¹² Upon hearing this, the accused initially declined. ¹³ He eventually relented after LYS repeatedly reassured the accused that the complainant was okay, she would not vomit and that LYS would give her a plastic bag for the car journey just in case. ¹⁴ At trial, the accused also explained that he would normally not ferry people who were drunk, but had agreed to ferry the complainant because he had assessed from her demeanour and her conversation with LYS that she was not “really drunk”. ¹⁵

9 Upon entering the Car, the complainant confirmed her address with the accused. ¹⁶ As the Car left the Bar’s carpark, the accused reminded the complainant to let him know if she wanted to vomit, to which the complainant was alert enough to answer “it’s okay”.

10 After arriving at her destination, the complainant exited the Car and attempted to enter the Condominium through its side gate. After a few minutes, seeing that the complainant still had not entered the Condominium, the accused approached the complainant and suggested that if she did not wish to go home yet, she could return to the Car and rest. ¹⁷ They then walked back to the Car together.

11 When they returned to the Car, the complainant got into the back seat on her own while the accused went to the driver’s seat. ¹⁸ After a while, the complainant began behaving in an agitated manner. This behaviour included sobbing, thumping her chest, and knocking her head against the car window. ¹⁹

12 According to the accused, he initially tried to calm the complainant down by speaking to her from the driver’s seat. ²⁰ After this met with limited success, he went to the back seat to calm her down. When she stopped, he would return to the driver’s seat again. ²¹ But she would then start behaving in an agitated manner again and he would return to the back seat again to calm her down. This cycle was repeated three or four times. ²² Getting impatient, the accused decided to look through the complainant’s bag to search for her handphone and/or her identity card in the hope that he could then get hold of someone to come get her. ²³ As he was searching through her bag, the complainant was resting her head on his left shoulder. ²⁴ He was also holding her hand, which he explained was to calm her down as well as to stop her from thumping her chest. ²⁵ After locating the complainant’s identity card, the accused turned around. By his account, it was at this time that the complainant began kissing him, and they started touching each other. ²⁶

13 The complainant’s moaning associated with sexual activity was captured by the Car’s in-car camera from 3.34am onwards. (This meant that the sexual activity must have commenced sometime between 3.20am and 3.34am, as there was a 14-minute gap in the in-car camera’s recording from 3.20am to 3.34am.) At one point, the accused could be heard saying “I can’t put it in” to the complainant and this was followed by the sound of some movement. ²⁷ The accused explained that this was the sound of the complainant shifting her position in response to what he had said. ²⁸ The accused was also heard saying “quiet” a couple of times to the complainant as her moaning grew louder. ²⁹ At one point, the complainant’s moan grew so loud that it sounded like she was reaching orgasm, ³⁰ although the complainant did not agree with this characterisation at trial. ³¹ During this period, the accused inserted his right finger into the complainant’s vagina and fingered her (which formed the subject of the first charge), and attempted to penetrate her vagina with his penis (which formed the subject of the second charge).

14 There was then a pause in the sexual activity, during which the accused went back to the driver’s seat, drove the Car down the road, made a three-point turn and drove back towards the Condominium. During this drive, the accused could be heard saying “I just take a round, okay?” to the complainant. ³² After stopping the Car on the road opposite the Condominium, the accused joined the complainant in the back seat to resume the Sexual Acts. During this time, he inserted his right hand beneath the complainant’s bra, fondled her breasts and sucked on her left nipple, and also attempted to insert his finger into her vagina (these formed the subject of the third charge). According to the accused, after a few minutes, he felt the complainant disengaging herself from the sexual activity. He then stopped and moved away from her towards the other end of the backseat. ³³

15 Independently of the accused’s testimony, the complainant’s moans associated with this second episode of sexual activity was also captured by the in-car camera. After a few minutes, the complainant was heard saying “no”, after which no further moans from the complainant were heard. ³⁴ This was at about 3.53am. ³⁵

16 The accused and the complainant then engaged in a brief conversation. ³⁶ Thereafter, the accused moved back to the driver’s seat. The complainant was heard saying “go” to the accused. When the accused sought clarification, the complainant told him to “just drive”. ³⁷ The accused then drove around the neighbourhood without incident and arrived back at the Condominium at about 4.05am. ³⁸ The complainant exited the Car unassisted, ³⁹ taking her handbag with her but leaving a paper bag behind in the Car. The accused noticed this and went after her to return the paper bag to her. ⁴⁰ By the time he caught up with her, she had already arrived unassisted at the Condominium gate. ⁴¹ She told the accused that she could not find her access card. The accused found the access card in her bag and used it to tap open the gate for her. ⁴² The complainant walked in unassisted. ⁴³ The accused drove off in the Car after seeing that the complainant had entered the Condominium. ⁴⁴

17 About five minutes later, the complainant was seen on the Condominium’s CCTV leaving the Condominium and walking along the road. ⁴⁵ At about 4.44am a passer-by, PW21 Muhammad Yazid Bin Umar found the complainant lying on the road a few hundred metres from the Condominium. He lodged the first information report via a ‘999’ call. ⁴⁶

18 The accused was arrested on 19 May 2018 at around 12.30pm for attempted rape. The following statements under s 22 of the Criminal Procedure Code (Cap 68, 2012 Rev Ed) were recorded and the accused has not challenged the admissibility of these statements before the court:

(a) one statement recorded by ASP Dan Dunstan Cheang (“IO Cheang”) at about 7.15pm on 19 May 2018 (“First Long Statement”); and

(b) one further statement recorded by IO Cheang at about 4.15pm on 23 May 2018 (“Second Long Statement”).

19 At around 10.05am to 11.40am on the same day, the complainant was examined by PW27 Dr Foo Anqi Sharon (“Dr Foo”) at KK Women’s and Children’s Hospital SingHealth (“KK Hospital”). Her blood and urine samples were taken sometime during this period and sent to the Health Sciences Authority (“HSA") for analysis. According to HSA’s analysis, 26mg/100ml of ethanol was found in the complainant’s blood sample and 102mg/100ml of ethanol was found in her urine sample. ⁴⁷

20 Evidence was also seized from the recordings of the accused’s in-car camera, from which the court could hear certain interactions between the accused and the complainant before, during and after the Sexual Acts. There was also CCTV footage from the Bar and the Condominium, from which the court could observe the complainant’s demeanour and assess her state of intoxication.

21 The only disputed element across the three offences was that of consent. As alluded to above (at [2]), the accused admitted to the physical acts constituting the actus reus of the offences.

22 The Prosecution’s case was that the complainant did not consent to the Sexual Acts and that, in any event, she did not have capacity to consent to the same as she was significantly intoxicated and impaired in her judgement. ⁴⁸

23 The Defence’s case was chiefly that the Prosecution had failed to meet its legal burden of proving lack of consent or lack of capacity to consent beyond a reasonable doubt. Alternatively, the Defence relied on the defence of mistake of fact under s 79 PC, namely, that the accused in good faith believed that the complainant had the capacity to consent and did in fact consent to the Sexual Acts.

24 In light of the statutory requirements and the parties’ cases, the issues that arose for my consideration were:

(a) whether the complainant did not consent to the Sexual Acts, in particular:

(i) whether the complainant lacked the capacity to consent, by virtue of s 90(b) PC;

(ii) if the complainant had the capacity to consent, whether no consent was in fact given; and

(b) if the complainant lacked the capacity to consent and/or did not in fact consent, whether the accused could nevertheless rely on the defence of mistake of fact under s 79(1) PC.

25 Part of the evidence relied on by the Prosecution were video clips from the accused’s in-car camera (or, more accurately, the audio tracks of those video clips, since the in-car camera was outward facing and did not capture video images of what occurred inside the Car). For instance, the Prosecution relied on the absence of audio recording of any “verbalised … express consent” to infer that the complainant did not in fact consent. ⁴⁹ For brevity, I shall refer to these audio tracks as the “in-car audio recordings”.

26 These video clips were tendered in an SD card, ⁵⁰ but the Prosecution also tendered a transcription and translation alongside the video clips. ⁵¹ PW17 Tan See Hua (“Ms Tan”), an interpreter with the Criminal Investigation Department (“CID”) of the Singapore Police Force, had been asked to transcribe the 17 clips. ⁵² She produced a 21-page report (the “Transcription Report”) that: (a) transcribed the English words that she heard, and (b) translated the Mandarin and Hokkien words that she heard from the audio recordings. ⁵³

27 Owing to justifiable concerns raised by the Defence about the accuracy of the Transcription Report as well as its inability to capture crucial nuances, ⁵⁴ I found it unsafe to place reliance on the Transcription Report. ⁵⁵ I indicated to the parties that I would be open to the Prosecution or Defence providing a better set of transcripts. Failing this, the trial would proceed without transcripts being used as evidence; this meant that the trial would proceed on the basis of what the witnesses and I could make out of the audio recordings. ⁵⁶ Both the Prosecution and Defence responded that they would endeavour to have the audio quality enhanced and obtain a better transcription. ⁵⁷ However, neither side sought to admit any further transcript into evidence. In the end, the lack of a replacement transcription did not hinder the trial, as parties were able to present all relevant evidence to the court by having the relevant portions of the in-car audio recording played in court and commented on by the relevant witnesses.

28 A related difficulty was that the in-car camera recordings were an incomplete record of what transpired in the Car that day. In the first place, the in-car camera was, by design, programmed to pause recording at certain points in time (eg, after the car had been idling for more than five minutes or once the car engine had been turned off). ⁵⁸ Furthermore, speech and sounds that were captured were not always clearly audible, despite the best efforts of the court and counsel to discern what had been said. The court’s reliance on such evidence thus had to be tempered by an appropriate appreciation of its limitations.

29 A key evidential feature of this case was that during the period when the Sexual Acts took place, the complainant had been experiencing an alcohol-induced blackout and was substantially unable to recall what she did, said or heard during this period. ⁵⁹ Consequently, she was also unable to explain to the court why she acted the way she did at the material time. Much of her evidence at trial was therefore limited to confirming what she heard in the audio recordings that were played in court and providing, at the time of the trial, her best explanation and reconstruction of what had occurred.

30 As a result of the complainant having no independent recollection of what occurred during the period that the Sexual Acts took place, the only evidence available to the court concerning what occurred during this period were the accused’s testimony and the objective evidence derived from the in-car audio recording.

31 The complainant’s behaviour at the material time was evidenced in three ways: (a) contemporaneous CCTV footages from the Bar and the Condominium, (b) contemporaneous in-car camera recordings from the Car, and (c) eyewitness evidence from various factual witnesses (including the complainant’s own recollections, where available, and the accused’s police statements). There were no challenges to their admissibility or authenticity.

32  The recordings (ie, (a) and (b) in [31] above) constituted real evidence that the court was competent and well-placed to assess. Nevertheless, a court must always be alive to the dangers of assuming that these presented a complete and unambiguous picture. Such dangers stemmed from potential incompleteness (eg, visual blindspots, audio without visual footage, stretches of time that went unrecorded) and lack of context (eg, situational cues, the manner in which a particular witness would normally speak).

33 This was why the evidence referred to at [31(c)] above was important, because it enabled those who witnessed or participated in these events to supply the relevant context. This explained why the key factual witnesses (ie, the complainant, the accused, LYS and VT) were asked both about:

(a) their own recollection of events independent of the recordings – eg, what they saw and heard, how they acted or reacted, and why they chose to act or react in the way they did; and

(b) how their evidence of the events (based on their recollection) matched up against what was shown on the recordings – eg, when and where a particular event took place and how this was situated in the overall chronology of events. This was achieved through the witnesses being shown relevant portions of the recordings in court and being asked questions about what they could see or hear from the footage.

34 At the same time, the dual function performed by these witnesses, and the manner in which this was performed, created its own challenges.

35 Chief among these was that the line between the two functions was not always scrupulously maintained by counsel. The cleanest way to separate both functions would have been to elicit a witness’ entire testimony from recollection before exposing them to any of the footage. However, based on how parties ran their case, witnesses were sometimes exposed to footage before they were asked for their recollection. A related concern was that questions were not always framed to clearly and exclusively relate to one function or the other. For instance, witnesses were sometimes asked to explain why they acted or reacted in a certain way. Yet the answers to these questions did not always specify whether these reflected the contemporaneous reasons that operated on their minds at the material time (ie, statements as to past belief, on a past act), or their present rationalisations or afterthoughts for why they might have acted that way (ie, statements as to present belief, on a past act). In most cases only the former would be relevant.

36 It was understandable that footage could sometimes be used to jog a witness’ memory as to a particular event that took place in the entire sequence. Such an approach, however, posed an obvious risk—that evidence purportedly originating from a witness’ recollection was in fact the product of his or her memory influenced by or reconstructed from the footage itself. Memories lost may now be “recalled”, when in fact the source of this “memory” was really the footage. If so, a witness might appear less forgetful and more reliable than he otherwise would, inflating his credibility and the weight given to his evidence. Additionally, memories inconsistent with the footage may be downplayed or repackaged to present a more coherent and consistent narrative. These potentially undermined two key objectives of witness testimony: one, forensically testing a witness’ credibility (through the completeness, credibility and accuracy of his evidence), and two, elucidating the evidence that a witness is capable of providing to the fullest extent possible.

37 An example that illustrated both these concerns was that on the complainant saying “Yah, yah, yah” in response to LYS asking her to text him when she reached home, LYS initially testified that at that point in time, the complainant’s reply “didn’t stood out so much to me”. It “just sounded like a yes to me. Like an acknowledgement”. ⁶⁰ But after hearing the footage, he added that “now when I hear it, it sounded … like she was pretty damn drunk” [emphasis added]. ⁶¹ Fortunately, this was an instance where the witness was transparent as to how his memory and interpretation of the footage diverged. This might not always be the case.

38 None of this was meant to suggest that any of the witnesses were deliberately lying. Rather, it raised a warning that such choices might have been subconsciously made by the witnesses.

39 As to the second function specifically (see [33(b)] above) (ie, explaining how their recollection matched up with the footage), there was also a concern that some of the questions asked of the witnesses exceeded the scope of what they were called to testify on.

40 These were witnesses of fact called for their observations and, where relevant, their impressions of the events at the material time. These observations and impressions were relevant because of the witness’ presence at and participation in those incidents. Thus, to illustrate, a witness asked to describe how the complainant’s speech was at the time might permissibly answer “she sounded like she was pretty damn drunk”. ⁶² This was because it was his description of what he heard at the time.

41 By contrast, these factual witnesses were not being called to explain what they thought they could see or hear from the audio recordings and footages, except in so far as to relate these observations to their own recollection. The court did not need to be told what a witness heard or thought a footage being played in court showed. That constituted real evidence, and the court viewing or hearing it in real time was competent to discern what it showed. If a CCTV clip showed a red car, and a witness claimed to see a blue car in the clip, it could not be the case that the court was expected to give weight to that. Thus, for instance, the evidence that LYS gave (ie, that upon hearing the audio recording in court, he thought that “she sounded like she was pretty damn drunk”) was but an inference derived from the footage.

42 As noted at [20] above, the evidence in this case included video recordings from the Car’s in-car camera, from the Bar’s CCTV and from the Condominium’s CCTV. These video footages bore timestamps based on the time kept by the internal clocks of the respective recording devices. Unfortunately, the internal clocks of these devices did not all keep the same time. The time recorded by the Bar’s CCTV was about 15 minutes ahead of the time recorded by the Car’s in-car camera, while the time recorded by the Condominium’s CCTV was about ten minutes ahead of the time recorded by the in-car camera. ⁶³ Further, by comparing the time recorded by the in-car camera with the time recorded by the car park gantry at the Bar, it appears that the time recorded by the in-car camera was about five minutes behind the official or correct time. ⁶⁴ This also implies that the time recorded by the Bar’s CCTV was about ten minutes ahead of the correct time while the time recorded by the Condominium’s CCTV was about five minutes ahead of the correct time.

43 In its closing submission, the Prosecution referred to the events captured in the in-car camera recording using the correct time, by adding five minutes to the time shown on the video footage’s timestamp. In contrast, the Defence’s closing submission referred to those events by the time shown on the video footage’s timestamp. In these grounds, where the time of an event is specified without qualification, it refers to the correct time. Conversely, where the time of an event captured in a video recording is given in these grounds by reference to the timestamp on the video recording, this will be indicated.

44 The law places the legal burden on the Prosecution to prove the complainant’s lack of consent beyond a reasonable doubt. In all three offences (see [1] above), the lack of consent would be a requisite element. (The lack of consent is also a requisite element of the outrage of modesty charge as s 354(1) PC refers to the use of “criminal force” which is defined in s 350 PC as the intentional use of force without consent.)

45 The lack of consent may be established in two ways. The first is by proving that the complainant did not in fact consent; in other words, that consent to the acts in question was never manifested. The second way is statutorily codified in s 90(b) PC and involves proving that the complainant was incapable of giving consent at the material time. Section 90(b) PC reads:

46 The effect of s 90(b) PC is that where the absence of consent is an element of the offence, and it is shown that the complainant was incapable of giving consent, then it would not matter whether she ostensibly consented since such a consent would not be valid (Pram Nair v Public Prosecutor [2017] 2 SLR 1015 (“Pram Nair”) at [62]).

47 Regardless of which of the two ways the Prosecution relies on to establish lack of consent, the legal burden and standard of proof remain on the Prosecution to prove beyond reasonable doubt that there was no valid consent to the respective Sexual Acts (Pram Nair at [45]).

48 It is trite, but bears repeating, that not all doubts about the Prosecution’s case are reasonable doubts. What is required is a qualitative appreciation of whether a reasonable doubt has arisen (Jagatheesan s/o Krishnasamy v PP [2006] 4 SLR(R) 45 (“Jagatheesan”)). A “reasonable” doubt, as opposed to a “merely fanciful” doubt, is “a doubt for which one can give a reason, so long as the reason given is logically connected to the evidence” (Jagatheesan at [53], affirming the dictum of Wood JA in R v Brydon (1995) 2 BCLR (3d) 243). Although the observations in Jagatheesan were made in the context of the reasons that a judge must supply to support a conviction, these observations are equally apposite if a judge should instead decide to acquit as a reasonable doubt is a necessary condition for an acquittal (Public Prosecutor v GCK and another matter [2020] 1 SLR 486 (“GCK”) (at [131])). The Court of Appeal in GCK further gave guidance on how the principle of proof beyond a reasonable doubt could be conceptualised in two ways. One, a reasonable doubt could arise from within the Prosecution’s case. Two, a reasonable doubt could arise on an assessment of the totality of the evidence, which includes a holistic assessment of both the Prosecution’s and the Defence’s cases and the interactions between the two.

49 It would be vital to appreciate that the principle of proof beyond a reasonable doubt is grounded in the presumption of innocence. In a passage affirmed by the Court of Appeal in AOF v Public Prosecutor [2012] 3 SLR 34 (at [315]), V K Rajah J (as he then was) observed in Jagatheesan (at [59]):

50 He further observed (at [61]):

51 Accordingly, the court’s task of assessing whether the Prosecution has proven its case beyond reasonable doubt naturally involved assessing whether sufficient doubt had been cast on the Prosecution’s case by the Defence. Should a reasonable doubt arise from within the Prosecution’s case or on the totality of the evidence as to capacity or consent in fact, the accused would be entitled to an acquittal.

52 The Prosecution’s primary case was that the complainant did not consent to each of the Sexual Acts referred to in the charges and, in the alternative, that she did not have the capacity to consent to the same. ⁶⁵ I would, however, first discuss the issue of whether the complainant had capacity to consent, before moving on to discuss whether she did in fact consent. This was the more logical sequence, as recognised by the Court of Appeal in Pram Nair (at [62]), since a negative finding on the first issue would necessarily negate any ostensible consent that the complainant had given. If, however, the complainant was not intoxicated to such degree, the Prosecution could still make out the offences by proving that she did not in fact consent (Pram Nair at [62]).

53 It was undisputed that the complainant had consumed five pints of beer while at the Bar ⁶⁶ and that she had stopped drinking at around 1.30am, or shortly before the staff switched off the lights in the Bar. ⁶⁷ The Prosecution submitted that the complainant did not have capacity to consent to the Sexual Acts as she was significantly intoxicated and impaired in her judgement at the material time. ⁶⁸ She did not understand or could not have understood the nature and consequence of what she was consenting to.

54 The Defence’s response was to point to instances across the night where the complainant demonstrated an awareness of context and ability to judge and decide, despite her intoxication. The Defence thus submitted that a reasonable doubt was raised as to the complainant’s capacity to consent at the material time.

55 Both the Prosecution and the Defence cited Pram Nair on the question of how intoxication affects the ability to consent to sexual activity. The Court of Appeal in Pram Nair noted (at [93]) that the court’s attention had been drawn to Public Prosecutor v Iryan bin Abdul Karim and others [2010] 2 SLR 15 (“Iryan”) where (at [123]) Tay Yong Kwang J (as he then was) cited the following passage from Ratanlal & Dhirajlal’s Law of Crimes: A Commentary on the Indian Penal Code 1860 vol 2 (C K Thakker & M C Thakker eds) (Bharat Law House, 26th Ed, 2007) at p 2061:

56 After noting that Iryan was a case where consent was vitiated because of the complainant’s fear of injury, the court went on to cite various observations from Rook & Ward on Sexual Offences Law and Practice (Sweet & Maxwell, 5th Ed, 2016) at para 1.252 (at [95]), which it synthesised as five general principles (at [96]) as follows:

57 Of particular relevance to the facts of the present case are the second and the fourth principles, namely:

(a) the fact that a complainant has drunk a substantial amount of alcohol, appears disinhibited, or behaves differently than usual, does not indicate lack of capacity to consent. Consent to sexual activity, even when made while intoxicated, is still consent as long as there is a voluntary and conscious acceptance of what is being done; and

(b) capacity to consent requires the capacity to make decisions or choices. A person, though having limited awareness of what is happening, may have such impaired understanding or knowledge as to lack the ability to make any decisions, much less the particular decision whether to have sexual intercourse or engage in any sexual act.

58 Both sides called expert witnesses to assist the court.

59 The Prosecution’s expert was PW35 Dr Christopher Cheok Cheng Soon (“Dr Cheok”), a Senior Consultant at the Department of General and Forensic Psychiatry of the Institute of Mental Health (“IMH”). He obtained his medical degree in 1995 and a Masters of Medicine (Psychiatry) degree in 2000. Prior to joining IMH in 2014, Dr Cheok headed the Department of Psychological Medicine in Khoo Teck Puat Hospital. ⁶⁹ He has appeared as an expert witness in court on about five previous occasions to provide evidence on a person’s blood alcohol content. ⁷⁰

60 The Defence’s expert was DW2 Dr Lim Yun Chin (“Dr Lim”), a consultant psychiatrist at Raffles Hospital. He obtained his medical degree in 1976 and a Masters of Medicine (Psychiatry) degree in 1986. He has served as a consultant psychiatrist at Woodbridge Hospital for 15 years, and has also served as a forensic psychiatrist at Changi Prison Hospital and as a psychiatrist at the Singapore Prisons and Drug Rehabilitation Centre. ⁷¹ He has previously appeared as an expert witness in court both for the Prosecution and for the defence.

61 Dr Cheok prepared a report dated 25 January 2019 at IO Cheang’s request, which was tendered by the Prosecution in evidence. ⁷² Dr Lim prepared two reports, one dated 20 September 2020 and one dated 16 August 2021. ⁷³

62 At trial, Dr Cheok tendered and referred to the following academic articles:

(a) Paul J Perry, Shadi Doroudgar & Priscilla Van Dyke, “Ethanol Forensic Toxicology” (2017) 45 Journal of the American Academy of Psychiatry and the Law 429 (“Exhibit P189”);

(b) Mary Connell, “Expert Testimony in Sexual Assault Cases: Alcohol Intoxication and Memory” (2015) 42-43 International Journal of Law and Psychiatry 98 (“Exhibit P190”);

(c) Alan W Jones, “Alcohol, Its Absoroption, Distribution, Metabolism, and Excretion in the Body and Pharmacokinetic Calculations” (2019) 1 Wiley Interdisciplinary Reviews Forensic Science e1340 (“Exhibit P191”) and

(d) Alan W Jones, “Evidence-Based Survey of the Elimination Rates of Ethanol from Blood with Applications in Forensic Casework” (2010) 200 Forensic Science International 1 (“Exhibit P192”)

63 In addition, during Dr Cheok’s cross-examination, Defence counsel referred Dr Cheok to the following articles:

(a) Aaron M White, “What Happened? Alcohol, Memory Blackouts, and the Brain” (2003) 27 Alcohol Research & Health 186 (“Exhibit D1”); and

(b) Alicia N Justus, Peter R Finn & Joseph E Steinmetz, “The Influence of Traits of Disinhibition on the Association Between Alcohol Use and Risky Sexual Behaviour” (2000) 24 Alcoholism: Clinical and Experimental Research 1028 (“Exhibit D2”).

64 The experts gave evidence of the significance of the complainant’s alcohol-induced blackout on the question of her capacity to consent. Both experts agreed that, while a person in an alcohol-induced blackout would be unable to form memories due to alcohol intoxication, this did not necessarily mean that the person would not be able to perform cognitive functions and understand the nature and consequences of his or her actions. ⁷⁴ Alcohol-induced blackout was not the same as passing out – a person experiencing an alcohol-induced blackout remained conscious and might even continue to talk and socialise. ⁷⁵ Dr Cheok pointed the court to the following passage from Exhibit P189: ⁷⁶

65 The experts also agreed that a person experiencing an alcohol-induced blackout might behave in a way that does not appear intoxicated to those around them or does not cause to those around them to suspect that he or she was so intoxicated to the point of having a blackout. ⁷⁷ The significance of this second feature will be explored under third issue (ie, the defence of mistake) (see [170] below).

66 As noted by the Court of Appeal in Pram Nair (at [96(e)]), expert evidence, such as that showing the complainant’s blood alcohol level, may assist the court in determining whether the complainant had the capacity to consent. I first deal with the experts’ evidence as to the calculation of the complainant’s blood alcohol level, before turning to their evidence as to the implications of the same.

67 Dr Cheok’s report estimated the complainant’s blood alcohol concentration (“BAC”) at 3.00am to be between 132.2 and 155.9mg/100ml. Dr Cheok explained that he gave the BAC as a range because he was not provided with the exact time that the Complainant’s blood sample had been taken. He was only informed by IO Cheang that the blood sample was taken sometime between 10.05am and 11.40am on 19 May 2018. Applying an average rate of 15mg/100ml/hour at which the human body metabolises alcohol to the BAC of 26mg/100ml in the blood sample and then extrapolating backwards in time, Dr Cheok calculated that: ⁷⁸

(a) if the blood sample had been taken at 10.05am, the Complainant’s estimated BAC at 3.00am would be 132.2mg/100ml; and

(b) if the blood sample had been taken at 11.40am, the Complainant’s estimated BAC at 3.00am would be 155.9mg/100ml.

68 During examination-in-chief, Dr Cheok was informed that it had been ascertained that the blood sample was taken at 11.40am. ⁷⁹ The examination-in-chief then proceeded on the basis that 155.9mg/100ml would be the applicable estimate at 3.00am based on an alcohol elimination rate of 15mg/100ml/hour. Dr Cheok also calculated that, using the same alcohol elimination rate, the estimated BAC at 3.30am and 4.00am would be 148.4 and 140.9mg/100ml respectively. ⁸⁰

69  As for why he adopted the alcohol elimination rate of 15mg/100ml/hour, Dr Cheok explained that, for 95% of the population, the alcohol elimination rate would range from 10mg/100ml/hour to 22mg/100ml/hour. ⁸¹ In previous court cases where he appeared as expert witness, both the defence and prosecution expert witnesses were in agreement that 15mg/100ml/hour was a reasonable elimination rate to use for someone who is healthy and not alcohol dependent. ⁸² By way of background, I note that this was also the alcohol elimination rate adopted by the experts in Ong Mingwee (alias Wang Mingwei) v Public Prosecutor [2013] 1 SLR 1217 (“Ong Mingwee”) and in Pram Nair for the purpose of estimating BAC by backward extrapolation.

70 Dr Cheok’s report was prepared on the basis of the complainant’s self-reporting that she did not drink alcohol regularly. ⁸³ At trial, Dr Cheok was informed that the complainant had given evidence in court that she had spent three months in Korea shortly before the incident, and that she would go out about two to three times a week during which she would drink two bottles of soju on each occasion. ⁸⁴ Dr Cheok responded that, based on this new information, he would treat the complainant as a fairly regular alcohol drinker and would use an alcohol elimination rate of 18mg/100ml/hour instead. ⁸⁵ He then recalculated the estimated BAC at 3.00am, 3.30am and 4.00am to be 182 mg/100ml, 173 mg/100ml and 164 mg/100ml respectively. ⁸⁶ (To recapitulate, the significance of these three timings are: 3.00am corresponded to roughly when the Car arrived at the Condominium, 3.30am corresponded to roughly when the first episode of sexual activity commenced, and 4.00am corresponded to roughly when the complainant entered the Condominium.)

71 Dr Lim did not challenge either Dr Cheok’s original estimate or his revised estimate. ⁸⁷

72 Dr Cheok was also asked during examination-in-chief whether the complainant’s BAC at 2.30am would have been higher or lower than at 3.00am. He answered that it would have been higher at 2.30am. ⁸⁸ Prior to arriving at this conclusion, Dr Cheok was informed that the complainant took her last drink at around 1.30am. ⁸⁹ Dr Cheok also explained that alcohol is absorbed into the blood stream mainly through the small intestines, ⁹⁰ and eliminated from the body mainly through the liver. ⁹¹ He also stated that maximum absorption typically occurs between 60 to 120 minutes after consumption. ⁹² Dr Lim’s evidence was that BAC would reach its peak about 30 to 120 minutes after drinking. ⁹³

73 For completeness, I should mention that during a later part of Dr Cheok’s examination-in-chief, while addressing a series of questions concerning alcohol-induced blackout, Dr Cheok suggested that two hours after the last drink (ie, around 3.30am) would roughly correspond to the time when the complainant’s BAC was peaking. However, during cross-examination, Dr Cheok’s attention was brought to a passage from Exhibit P191 which reads: ⁹⁴

After considering this passage, Dr Cheok opined that it would be fair and reasonable to assume that BAC levels would peak at around 60 minutes after drinking. ⁹⁵ Dr Cheok also agreed that (a) the reference in the passage to “pyloric spasm” refers to a situation where the muscle connecting the stomach to the small intestines has cramped up such that the stomach is not releasing its contents into the small intestines, and (b) this meant that the situation of BAC peaking at two hours would be an outlying situation. ⁹⁶

74 To wrap up this point, I would also note that Dr Cheok’s opinion that the complainant’s BAC would have been higher at 2.30am compared to at 3.00am was corroborated by Dr Lim’s observation that the complainant appeared less intoxicated in the CCTV footages from the Condominium than in the CCTV footages from the Bar. ⁹⁷

75 Given the lack of dispute between the experts on the following two points, I accepted that:

(a) the complainant’s estimated BAC at 3.00am to 4.00am would have been between 164 and 182 mg/100ml; and

(b) the complainant’s BAC at 2.30am would have been higher than at 3.00am.

76 The first question to be answered was whether the complainant’s estimated BAC of 164 to 182 mg/100ml at the material time was consistent with the complainant having experienced alcohol-induced blackout. Dr Cheok gave evidence that memory impairment would begin at about 130 to 170 mg/100ml. ⁹⁸ According to Dr Lim, it was believed that alcohol-induced blackout would occur when the BAC exceeded 140 mg/100ml. ⁹⁹ The article tendered by Dr Cheok as Exhibit P190 also stated that “fragmentary blackouts” occur at BACs of 150 mg/100ml or higher. ¹⁰⁰ It may therefore be safely concluded that the complainant’s estimated BAC was consistent with her having experienced alcohol-induced blackout at the material time. Consequently, I accepted the expert witnesses’ assessment that it was due to alcohol-induced blackout (and not any other reasons) that the complainant had no recollection of what had occurred during the time that the Sexual Acts took place.

77 As for the impact of the complainant’s BAC on her behaviour and capacity, Dr Cheok referred to a number of tables in Exhibit P190 setting forth typical behavioural and cognitive impairments at different BAC levels. Dr Cheok then informed the court that, at the complainant’s estimated BAC, there would have been impairment of judgement as well as incoordination in her movements. ¹⁰¹ However, during cross-examination, Dr Cheok’s attention was drawn to a passage from Exhibit P190 which stated: “It should be noted, however, that individuals may differ considerably in what symptoms of intoxication are exhibited at various levels. It is speculative what symptoms of intoxication any specific person would exhibit at BAC levels” [emphasis added]. ¹⁰² I set out the exchange as follows: ¹⁰³

78 Dr Cheok also accepted that there was a need to correlate the complainant’s BAC to her observed behaviour at that time: ¹⁰⁴

79 I agreed with Dr Cheok that the complainant’s estimated BAC must be considered together with her observed behaviour in order that a proper assessment of her level of intoxication may be made. This would be consistent with the approach adopted by the courts in Pram Nair and Ong Mingwee. In Pram Nair, the complainant’s estimated BAC was 219 to 257 mg/100ml while in Ong Mingwee, the complainant’s estimated BAC was 225 mg/100ml. Even though the estimated BAC in Ong Mingwee fell within the ranges of estimated BAC values in Pram Nair, the courts in these two cases came to opposite conclusions on the respective complainant’s capacity to consent. In Pram Nair, the court concluded that the complainant in that case lacked the capacity to consent as it was observed that she could not talk in full sentences, could not stand up by herself and was lacking control of basic motor functions. She was seen pushing the accused away very weakly and lying down with little movement. She was also not able to answer her phone. In contrast, the court in Ong Mingwee concluded that the complainant in that case had the capacity to consent after observing that, while the complainant was tipsy, not very stable and slightly slurred in her speech, she was able to make coherent conversation, walk unassisted and get into the taxi unsupported and unaided.

80 Before turning to consider the complainant’s observed behaviour, there was one other aspect of the experts’ evidence to be considered. When IO Cheang wrote to IMH on 7 December 2018 to request an expert opinion, he stated that he needed an expert “to comment on the victim’s state of intoxication and whether she would have been able to give her consent given the state she was in”. ¹⁰⁵ Dr Cheok’s report of 25 January 2019 responded directly to this request by stating the following in its concluding paragraph: ¹⁰⁶

81 My first observation would be that this conclusion was premised on a number of factual errors. First, although the complainant told Dr Cheok that she did not drink alcohol regularly, the evidence adduced at trial showed otherwise (see [70] above). As a result, Dr Cheok decided during his examination-in-chief to reclassify the complainant as a “fairly regular drinker” (as opposed to a person who did not drink regularly). ¹⁰⁷ Second, the statement “[a]fter sometime [sic], seeing that the victim was not responding, the accused sat at the back seat with her to wake her” was taken from the part of IO Cheang’s summary of facts relating to the period after the complainant had returned to the Car after failing to open the Condominium’s side gate. ¹⁰⁸ The evidence adduced at trial showed that the complainant was not asleep during this period and there was no evidence that the accused woke her up. Third, it was clear from the Condominium’s CCTV footage that, contrary to what the complainant told Dr Cheok about her recollection, she did not fall down at the side gate. ¹⁰⁹

82 Second, it appeared that, when preparing the report, Dr Cheok was operating under a misapprehension of what constituted capacity to consent at law, as demonstrated by the following exchange during Dr Cheok’s cross-examination: ¹¹⁰

83 Even though Dr Cheok subsequently clarified that he was not equating “valid consent” to the concept of informed consent to a medical procedure, it would appear that his understanding was influenced by the latter. This was evident from his explanation that “I don’t think … she would understand … what she was getting herself into fully” [emphasis added]. ¹¹¹ His understanding of “valid consent” thus required the complainant to appreciate the full risks or consequences of her actions. This was also evident from the distinction he drew, between consenting to something simple like crossing the road with someone or getting into a car, versus consenting to something “more sophisticated” such as “buying a car, for example, and to have to sign a loan agreement, a sales and purchase agreement”. ¹¹²

84 Dr Cheok also appeared to conflate the concept of disinhibition (which he referred to as a person “acting in ways that they wouldn’t be when they are sober”) with the inability to comprehend the nature and consequences of one’s actions. He was unable to accept that there was a meaningful distinction between the two: ¹¹³

This failure to distinguish the two concepts does not accord with the principles enunciated in Pram Nair, which drew a clear distinction between mere disinhibition and lack of capacity to consent (at [96(b)]).

85 I was thus not satisfied that Dr Cheok’s understanding of capacity to consent was consistent with the applicable legal principles (see [56] above).

86 Dr Lim was similarly asked by the Defence to comment on the complainant’s capacity to consent in his second report, but he did not do so. When asked about this during cross-examination, Dr Lim candidly acknowledged that he was not able to give a conclusive opinion on the complainant’s capacity to consent. ¹¹⁴

87 At this point, it would be useful to recall the following dictum of G P Selvam J from Gunapathy Muniandy v James Khoo and others [2001] SGHC 165 (at [12.3]):

This dictum was cited with approval by Chan Sek Keong CJ in George Abraham Vadakathu v Jacob George [2009] 3 SLR(R) 631, a case concerning testamentary capacity, when he commented (at [66]) that:

88 With these principles in mind, I would pay less attention to the experts’ conclusions on the ultimate issue of whether the complainant had capacity to consent, and focus my analysis on evaluating the complainant’s observed behaviour with the assistance of the experts’ comments on those behaviour. This was also the approach adopted in the Prosecution’s Closing Submission, where no reliance was placed by the Prosecution on Dr Cheok’s conclusion on the ultimate issue.

89 The Prosecution contended that the complainant had already been intoxicated for some time before she boarded the accused’s Car. ¹¹⁵ The complainant was feeling physically unwell after the Bar closed for the night, had tried (and failed) to vomit three to four times in the Bar’s toilet and had been stumbling as she walked out of the toilet. ¹¹⁶ Reliance was placed by the Prosecution on the CCTV footage and the testimonies of LYS and VT, to corroborate the victim’s account of her physical and emotional state. ¹¹⁷

90 I broadly set out the evidence on the complainant’s observed behaviour at the Bar as follows.

91 Before the Bar’s lights were switched off for the night, LYS described the complainant’s behaviour at about 1.18am (according to the timestamp on the Bar’s CCTV footage) as “perfectly fine” and recalled that she was able to converse with a waitress in the Korean language. ¹¹⁸ VT similarly described the complainant as “sober” even after drinking a few pints of beer, and “still very chatty” and “laughing and smiling with [LYS]” and “did not seem like out of the ordinary at all”. ¹¹⁹

92 After the Bar closed and the lights of the Bar were switched off at around 1.45am or 2.00am, some deterioration in the complainant’s motor control began to emerge. The complainant described herself as feeling “very lightheaded and giddy” ¹²⁰ and “a bit unwell” during this period. ¹²¹ . She was, however, able to key in her address on her phone to book a private hire car by herself (the “first private hire car”). ¹²²

93 When shown a CCTV video clip of her gait from this period, the complainant described herself as taking “very heavy steps” in the first video, and then stumbling to the left and right in the second video. ¹²³ On the other hand, LYS described that her gait “didn’t seem wobbly yet”. ¹²⁴

94 When the first private hire car arrived, the complainant was in the toilet trying to vomit and LYS went to check on her. She then handed her phone over to LYS and told him to cancel the booking on her behalf (which she could not do as she was busy trying to vomit). ¹²⁵ This suggested to me that she was, at the very least, sufficiently aware of her surroundings and able to make decisions despite her intoxication and the impairment in her motor skills.

95 After cancelling the first private hire car booking, LYS returned to the toilet to check on the complainant. She told him that she “needed some time to—to puke”. ¹²⁶ What followed was a brief exchange between LYS and the complainant, which showed some lucidity on the latter’s part: LYS testified that he waited outside the female toilet for about five to ten minutes, periodically asking her “Are you okay?”. ¹²⁷ She occasionally replied with either “I’m okay” or “I need more time” or “I’m trying to puke, so give me a while”. ¹²⁸

96 It was also relevant that the complainant had decided to reject LYS’s offers to send her home and repeatedly emphasised that she was okay. ¹²⁹ The complainant testified that she did this to avoid worrying him, whom she knew to be a very anxious person. She had the presence of mind to recognise that “[s]o if like I tell him like a little bit I’m not okay, he will be very worried and paranoid”. ¹³⁰ She also did not want to trouble both LYS and VT to send her home as she was concerned that they would have to “go a long way before they went back … home” if they did so. ¹³¹ Dr Cheok agreed that this decision demonstrated her ability to look beyond her immediate means and to consider the impact of her words on her friend. ¹³²

97 During this time, the complainant’s motor control was observed to be impaired and unsteady. When she came out of the toilet after the first unsuccessful attempt at vomiting, LYS described her as appearing “a bit lightheaded”, by which he meant that she was “a bit dizzy” and “a bit wobbly” and “like, she needed some help when she was walking”. ¹³³ After yet another unsuccessful attempt at vomiting, LYS testified that the complainant seemed “very unstable, she needed help” and that she “clearly couldn’t walk on her own”. ¹³⁴ CCTV footage showed LYS placing his hands on the complainant’s shoulders to help her walk straight. ¹³⁵ Afterwards, she returned again to the toilet to try to vomit. The complainant testified that she was feeling “lightheaded and giddy” throughout this time. ¹³⁶

98 At about 2.30am, LYS helped the complainant book a second private hire car. The accused responded to this booking, and the Car arrived at the Bar within a few minutes. The evidence showed that the complainant was still unsteady in gait and had started crying and sobbing. When shown a CCTV footage of herself returning to and exiting the toilet for the last time, the complainant described that she had been crying and sobbing, as evidenced by her rubbing her face with both hands, ¹³⁷ but testified that she was unsure why. ¹³⁸ She also described, based on the CCTV footage, that she “couldn’t stand properly” as her knees “went a bit wobbly”. She “almost … fell down” and “almost tripped”. ¹³⁹ LYS helped her by “[h]olding [her] hand and bringing [her]---walk straight.” ¹⁴⁰ Similarly, VT recalled that she had gone to check on the complainant after the Car arrived, and found her crying ¹⁴¹ and squatting down beside a pillar with LYS for about ten minutes. ¹⁴² She was non-responsive to questions such as VT asking her “Are you okay? What happened?” and LYS’s query as to her home address. ¹⁴³ After getting up, she could not walk straight and was swaying unsteadily, and needed to hold onto LYS when walking to the Car. ¹⁴⁴

99 LYS testified that the complainant’s demeanour at this time was “very light-headed, like very lost” and “like very blur”. ¹⁴⁵ VT testified that the complainant did not look “sober” because “firstly of her walking, and then … it’s her facial expression … [i]t was a very sleepy look” [emphasis added] at the time she reached out to grab her bags from VT while walking past.

100 It was clear that the alcohol was having some effect on the complainant. The most obvious effect was on the complainant’s motor control (see [92], [93], [97] and [98] above). The complainant remembered stumbling after her visits to the toilet, before squatting by a pillar and drain. ¹⁴⁶ She was feeling lightheaded at the point in time. ¹⁴⁷ It also impacted her emotional state. She was crying. ¹⁴⁸ According to her, her light-headedness and giddiness had gotten “worse a little bit” by the time she was about to board the Car. ¹⁴⁹

101 However, the complainant was also not completely oblivious of her surroundings or incapable of reacting to others in contextually-relevant ways. LYS accepted that she would have been able to “react to normal conversation or to make decisions as a normal person would … [except] maybe [with a] slightly slower reaction” [emphasis added]. ¹⁵⁰ He also accepted that she had been responsive to his questions, except that he sometimes had to repeat the questions. ¹⁵¹ For instance, once LYS told her that the Car had arrived, she “just stood up by [herself] and walked down back to the car” (albeit with some stumbling). ¹⁵² Both LYS and VT also accepted that the complainant had initiated the collection of her bags and that she had the presence of mind to turn to VT and collect her bags. ¹⁵³ It was also undisputed that the complainant walked to VT unassisted, to collect her bags, and boarded the Car on her own. ¹⁵⁴

102 The Prosecution suggested that the complainant was so intoxicated that she felt like vomiting. ¹⁵⁵ This was contradicted by the complainant’s own testimony. She explained that she had a practice of forcing herself to vomit in order to expel the alcohol from her body. ¹⁵⁶ She testified: “I don’t feel like vomiting, I force myself to vomit”. ¹⁵⁷ She would do this when she had four to five pints to drink and “fe[lt] more lightheaded”, ¹⁵⁸ as it would help her to “feel better” and “more sober and conscious”. ¹⁵⁹

103 Overall, the evidence supported a finding that the complainant was intoxicated. But the relevant inquiry was not whether the complainant was intoxicated per se but whether she was so intoxicated that she was unable to understand the nature and consequence of what she was consenting to. There was sufficient evidence indicating her situational awareness and judgement (see [96] and [101] above), which suggested that, at this stage, she could not be described fairly as being unable to understand the nature and consequences of her choices.

104 The Prosecution relied on VT’s evidence as to the complainant’s demeanour when she boarded the accused’s car. According to VT, the complainant was conscious and not lying down on the seat. But VT also described the complainant’s state as “not responsive”, meaning that “she just [sat] down there and just stare[d] into space” and like “she’s in her own daydream kind of thing”. ¹⁶⁰

105 Non-responsive was, however, not an accurate description of the complainant’s state at the time. The evidence showed that the complainant did respond to some of the exchanges that were taking place around her. For instance, to LYS’s request that she text him after reaching home, the complainant replied “yah yah yah”. ¹⁶¹ The complainant also completed the accused’s sentence “193, ah?” by stating her address. Further, as the Car was exiting the Bar’s carpark, the accused said to the complainant “If you want to puke, just let me know”, to which the complainant respond with “It’s okay, it’s okay”. ¹⁶²

106 The Prosecution submitted that the complainant’s ability to state her address hardly had any bearing on her state of intoxication, relying on Dr Cheok’s evidence. ¹⁶³ Dr Cheok postulated that the complainant was able to understand and respond because this was a “very basic” and “nothing … sophisticated” type of question. ¹⁶⁴ In his view, a person with similar blood alcohol level to the complainant would be able to understand and respond to such basic questions, such as being asked their name, NRIC or address. Dr Lim disagreed and opined that, at the very least, this exchange showed that the complainant’s retrieval process was working to be able to retrieve information such as her address. ¹⁶⁵ On this issue, I preferred Dr Lim’s evidence. However, I would note that, even if the Prosecution was right that not much reliance could be placed by the Defence on this exchange, it did not change the fact that, about 20 seconds after this exchange, the complainant was able to respond with “It’s ok, it’s okay” when the accused said “If you want to puke, just let me know”. When this latter exchange was played to the complainant in court, she agreed that she was responding appropriately to the accused and that she sounded alert when responding. ¹⁶⁶

107 For the reasons above, I was not persuaded that the evidence relating to this period demonstrated lack of capacity to consent.

108 After boarding the Car, the complainant testified that she continued to feel lightheaded and thus closed her eyes, while continuing to cry. ¹⁶⁷ Her evidence was that she remembered just closing her eyes and sleeping after the Car left the Bar, ¹⁶⁸ and that she fell asleep the moment he started driving. ¹⁶⁹ In her conditioned statement and at trial, she maintained that she had only been awoken by the accused when the Car reached her destination. ¹⁷⁰ At trial, she further testified that she was not aware of her surroundings throughout the car journey, including the Car reversing towards the fire engine gate of her condominium. ¹⁷¹ When asked “When were you crying?”, her answer was that she could not remember. ¹⁷² She also could not remember anything else being said by her or the accused during the journey. ¹⁷³ For instance, she did not remember hearing the accused ask “Are you okay?” several times, some of which were captured by the in-car audio recording. ¹⁷⁴

109 The in-car audio recordings captured crying and sniffling sounds from the complainant at various junctures, but did not capture any replies from the complainant to the accused’s questions of “Are you okay?”. ¹⁷⁵ At trial, the complainant described the “sniffling” sounds as “like my nose got blocked by all the crying earlier on”. ¹⁷⁶ The fact that she was sniffling or crying at various times meant that she would have been awake at those times, and therefore she could not have been asleep throughout the entire car journey.

110 Overall, the available evidence of the complainant’s behaviour during this period did not support a finding of lack of capacity to consent, especially when this evidence was considered in the context of the evidence concerning what had occurred immediately before she boarded the Car and immediately after the Car arrived at the Condominium.

111 Upon arrival at the Condominium, the complainant was heard rummaging through her things. ¹⁷⁷ The in-car audio recording also captured the accused asking “Do you need my help?” a number of times, but the complainant did not respond. She testified at trial that she had not heard these questions. The accused then said “I can park my car one side first. Is it okay?” and the complainant replied, “sorry”. ¹⁷⁸ At trial, the complainant agreed that this showed that she knew what he was saying and that she knew that she was inconveniencing him. ¹⁷⁹ She further explained that she had “kept saying sorry” because she heard herself crying and “wanted to faster get down the car”. ¹⁸⁰

112 After the accused moved the Car to park just past the entrance of the Condominium, the complainant was heard telling the accused that she was “not ready”, to which the accused replied “yah, no problem, don’t worry”. ¹⁸¹ At trial, the complainant agreed that this likely meant that she was not ready to leave the Car. ¹⁸² She also suggested an alternative explanation that she had been “not ready” because she was in the process of collecting all her belongings before alighting. ¹⁸³ According to Dr Cheok, it would be reasonable to infer that the complainant had been aware of her circumstances and of whether she was ready to leave the Car. ¹⁸⁴

113 In fact, she was at that point searching for her wallet, intending to pay for the ride. The Prosecution contended that the fact that the complainant attempted to pay for the ride indicated that she was intoxicated at that time, as she had previously seen LYS passing cash to the accused at the Bar’s carpark, but did not recall that fact until the accused returned the change to her and told her that LYS had paid for the ride. ¹⁸⁵ The Prosecution also relied on the fact that, despite her difficulty finding her wallet in the dark, the complainant did not ask the accused to switch on the lights inside the Car because she “wasn’t thinking at that point in time”. ¹⁸⁶ However, as set out above at [57], what was relevant was not whether the complainant was intoxicated per se, but the extent to which this impacted her ability to appreciate her surroundings and understand the nature and consequences of her actions.

114 While the complainant not remembering that LYS had paid for the ride could be a sign of intoxication, I did not regard it as an indication that she was not able to appreciate her surroundings. The complainant testified that, when the accused informed her that the ride had already been paid for, this prompted her to remember that she had seen LYS passing the accused cash at the Bar. ¹⁸⁷ This meant that, firstly, she understood what the accused told her and, secondly, she was able to correlate what the accused said with her own memory of an event she had witnessed earlier (but which had momentarily slipped her mind). As for the complainant not asking the accused to switch on the lights, this did not detract from the fact that she actually asked the accused for help with finding her wallet. She testified that when she said “[my] wallet is over there” instead of taking the wallet herself, it was a request for help in response to the accused asking her whether she needed help. ¹⁸⁸ This meant that she understood what the accused was saying when he offered to help, and she was able to make the decision to ask for help.

115 The next relevant period concerned the complainant’s observed behaviour as she attempted to access the Condominium by the side gate.

116 The Prosecution relied on the complainant’s unsteady gait and feeling “giddy and lightheaded” as evidence of her continued intoxication. ¹⁸⁹ She testified that she was stumbling as she walked to the side gate and was feeling “very tired” at this point of the night. The Condominium’s CCTV footage also captured her movements while waiting at the side gate which she described as “very unstable” and “cannot stand straight”, after dialling the telecom and waiting for someone to pick up. ¹⁹⁰ That the complainant’s motor control continued to be impaired was corroborated by the CCTV footage which also showed that the accused had placed his hands on the complainant’s shoulders to support her as she later walked back to the Car. ¹⁹¹

117 At one point, the complainant squatted down outside the side gate. She explained that she was “a bit tired” and was trying to find her access card with both her hands. ¹⁹² However, she also explained that she had stood a little further away from the gate “because…usually there’s a cat on the top of the gate. So it will jump down any time”. ¹⁹³ Dr Cheok agreed that the complainant’s ability to remember that there might be a cat at the gate, despite not seeing it at that time, ¹⁹⁴ indicated that she had been aware that she was home and that her ability to determine the nature and consequences of her actions was not severely impaired. ¹⁹⁵

118 Overall, I was not convinced that the evidence of the complainant’s behaviour during this period pointed to the conclusion that she was unable to understand the nature and consequences of her actions. While I accepted that her motor impairment and certain actions were indicative of continued intoxication, the evidence on a whole also suggested that she retained the ability to understand and decide.

119 The complainant remembered the accused asking her if she wanted to go back to the Car and take a seat first after she failed to enter the Condominium. She believed that he did so because he saw that she was still crying. ¹⁹⁶ She remembered that she agreed to go back to the Car “because [she] was feeling very tired”. ¹⁹⁷

120 After returning to the Car, the complainant not only continued crying but also knocked her head against the window and repeatedly “thumped” herself on her chest. The Prosecution relied on this evidence of the complainant’s heightened “emotional lability” and “emotional instability” as a “clear and unambiguous sign” that she was severely intoxicated, citing Dr Cheok’s evidence. ¹⁹⁸

121 However, it was also relevant that the complainant had, several times, stopped such behaviour momentarily when the accused verbally urged her to stop. This largely took place in the 14 minutes where there was a gap in the in-car video recording. Hence, the sole account was from the accused, whom I found to be a credible witness and whose evidence I accepted (see [154]–[159] below). I set out the relevant portions of the accused’s evidence as follows: ¹⁹⁹

122 The complainant’s actions thus demonstrated an awareness of what the accused was communicating to her, and her ability to process and accept his suggestion. As the Defence pointed out, the fact that she would resume these actions thereafter was not, in itself, an indicator that she was unable to understand her actions. ²⁰⁰

123 In another brief exchange, the accused asked the complainant whether she was okay. The complainant replied that she was okay, before continuing to cry. At trial, she explained that she had responded in this way because she did not know the accused and thus did not wish to say much to him. Dr Cheok agreed that this explanation showed an awareness of her surroundings (ie, she was aware that she was with a stranger), and of the differences in sharing “personal” information with different people. ²⁰¹

124 I deal with the evidence of the complainant’s behaviour during the sexual activities in greater detail below when considering the issue of consent in fact. What was significant to the question of capacity was that towards the end of the sexual activities, the complainant started saying “no” and pushed the accused’s hands away. This “no” was verbalised as the accused was trying to move his hand up her thigh and towards her vagina (see [145] below). On cross-examination, the complainant accepted that the “no” indicated that she did not want to be touched anymore but was slow to agree that it evidenced her ability to convey such an intention. The basis of her hesitation to this question, however, was that she thought that she might have been asleep at that time and was capable of making noises while asleep. ²⁰² But the evidence showed that she was in fact awake and neither was it the Prosecution’s case that she had been asleep during the sexual activities. By process of elimination, it must be taken that the complainant’s verbalised “no” evidenced an ability to convey the intention that she did not want to be touched anymore. Dr Cheok, too, accepted as much during cross-examination: ²⁰³

125 Similarly, Dr Lim suggested that the complainant’s reaction showed an awareness of what was happening and an ability to give or refuse consent at that point. ²⁰⁴

126 After the sexual activities, the accused asked the complainant if she was okay. She responded by saying yes, before asking him to “just drive” despite the Car having been parked right opposite her Condominium. ²⁰⁵ Dr Lim opined that this showed that she wanted to process something in her head, which substantiated his opinion above that she was able to give consent: ²⁰⁶

127 When asked at trial, Dr Cheok also accepted this to be a reasonable explanation. ²⁰⁷ To my mind, the complainant’s responses reflected an ability to assess whether to leave or to remain in the car with the accused, even after the sexual activities had just taken place.

128 The Prosecution relied on several facts in contending that the complainant continued to exhibit signs of intoxication after the alleged offences. First, the complainant had left the accused’s car without her safety shorts and underwear. According to the Prosecution, this evidenced that her brain function was “considerably impaired”. ²⁰⁸ Second, the complainant testified that she did not have any reason for exiting her Condominium shortly after entering the compound. ²⁰⁹ Third, the complainant continued to act “bizarrely” after exiting the Condominium. She recalled walking along the road in front of the Condominium by herself without her belongings. ²¹⁰ When shown the CCTV footage in court, the complainant commented that she was walking “very aimlessly” and “very unstable” in her movements. ²¹¹ Eventually, she found herself lying in the middle of the road and did not know if she had fainted or laid down on the road on her own, ²¹² feeling “tired” and “no energy”. Her next memory was of the paramedics waking her up. Fourth, the complainant was found to have an “altered mental state” with some impairment in her level of consciousness when the paramedics attended to her at around 4.54am, ²¹³ and was crying. PW29 Zaneta Lee (“Ms Lee”), a paramedic, also observed that the complainant kept struggling and crying, and did not want anyone to touch her.

129  On first impression, these behaviours and the fact that the complainant was later found unconscious in the middle of the road appeared to provide some support for the Prosecution’s case of general lethargy and incapacity. I was, however, slow to rely too heavily on evidence of the complainant’s state after the sexual activities had occurred. The Defence suggested that deterioration in the complainant’s mental state may have been due to fatigue, ²¹⁴ which only became severe enough after the sexual episodes. This was plausible when one considered (a) the time of the night and, consequently, the complainant’s increasing state of sleep deprivation, and (b) the fact that she had just engaged in sexual activities.

130 I was satisfied that there were several points across the night where the complainant demonstrated awareness of her surroundings and the capacity to make decisions or choices (with awareness of their nature and consequences), despite her intoxication. In assessing the evidence, I was mindful that consent is highly contextual. Deciding to, for instance, enter a car and deciding to have sex are qualitatively different choices. Understanding the nature and consequence of these two decisions may call for differing degrees of understanding and cognition. Therefore, the point was not, for example, that simply because the complainant could decide whether or not to accept her friend’s offer to send her home, that she must necessarily have been able to decide whether to engage in sexual activities. In the final analysis, viewing in totality her situational awareness and the decisions she was shown to be capable of making throughout the night, there existed sufficient reasonable doubt over the assertion that she was incapable of consenting to the Sexual Acts.

131 The Prosecution submitted that the evidence supported an inference that the complainant did not consent to sexual activity with the accused. ²¹⁵ There were three key planks to the Prosecution’s submission, namely: One, the accused’s in-car camera did not record any verbalised consent by the complainant to the Sexual Acts. Two, the fact that the complainant said “no” towards the end of the sexual activities indicated that she did not consent to any sexual activity with the accused. Three, the Prosecution relied on evidence of the complainant’s behaviour after the alleged offences, namely, the suggestion by the paramedic, Ms Lee, that the complainant could have been sexually assaulted because she kept crying and struggling when the team of paramedics attended to her.

132 The Defence submitted that the Prosecution has not proven beyond a reasonable doubt that the complainant did not consent. The Defence highlighted that the complainant did not voice any objection or show any resistance throughout the two episodes of sexual activities, and she appeared wholly at ease with the accused.

133 I note as a preliminary point that, as the complainant could not recall whether she gave consent due to her alcohol-induced blackout, the sole account of what happened during the Sexual Acts therefore came from the accused, who was the only other person in the Car with the complainant at the material time. As such, as the Prosecution also accepted during submissions, the assessment of whether there was actual consent in fact turned largely on the assessment of credibility of the accused’s testimony.

134 There were two distinct episodes of the sexual activities. In between the two episodes, there was a brief break in the sexual activities during which the accused drove the Car down the road, made a three-point turn and stopped across the road from the Condominium. ²¹⁶

135 The accused’s account was broadly consistent across his investigative statements and at trial, and was as follows. After the complainant was unable to gain access to her Condominium and returned to the back seat of the Car, the accused initially returned to the driver’s seat and sat down. Over the next few minutes, the complainant continued “sobbing and crying”, ²¹⁷ was “thumping” (ie, pounding) herself on her chest ²¹⁸ and hitting her head on the window of the Car (see [120] above). As a result, the accused was concerned about the complainant and, on certain occasions, went to the back seat to “calm her down” and “stop her from … thumping her chest” by holding her hand, ²¹⁹ and thereafter returned to the driver’s seat when she seemed to calm down. This happened about “three or four times, or maybe two or three times” by his recollection. ²²⁰ The accused’s account was broadly corroborated by the in-car audio recording starting at timestamp 3:10:02, in which the accused could be heard telling the complainant to “don’t think too much” and “just close your eyes” and “just take a nap”. ²²¹ It was also consistent with the accused’s First Long Statement and Second Long Statement, where he described that he had been trying to console her and prevent her from hitting herself.

136 After a period of time, the accused went to the back seat of the Car to look through the complainant’s handbag. He testified that the complainant was still “visibly upset” and he intended to search for her NRIC and handphone in the hope of finding a way to contact someone who could come and get her, or alternatively, to check for her address on her NRIC so that he could obtain a unit number to dial using the Condominium’s intercom system. ²²² He testified that the complainant’s head was leaning on his shoulder as he was searching inside her handbag with one hand. ²²³ The other hand (namely, his left hand) was holding on to the complainant’s right hand. He managed to locate the complainant’s unit number on her NRIC and turned around to look at her.

137 It was then that the complainant initiated the series of acts that turned sexual in nature, beginning by kissing him. To be fair, the accused did not actually tell the court that the complainant initiated the kiss. All that he said was “after I put the items back in the bag, I turned around and then next [thing] I know we end up kissing”. ²²⁴ When asked who initiated the kiss, he answered “not me”. ²²⁵ By the process of elimination, the accused’s position must be taken to be that the complainant had initiated the kiss.

138 However, there is one aspect of the accused’s evidence that calls for some comment. When asked whether his denial of being the one who initiated the kiss meant that the kiss was initiated by the only other person in the Car (ie, the complainant), he replied “I cannot remember but I can say with certainty that it’s not me because I did not initiate the kiss”. ²²⁶ The Prosecution submitted that the accused’s account was “confounding” and “most unbelievable”, and was a disingenuous response borne out of the realisation that “it would be inherently incredible for the victim to ever initiate such sexual contact in such an inebriated state”. ²²⁷ First, given my finding on the complainant’s capacity to consent, I did not agree with the Prosecution that it would be inherently incredible for the complainant to initiate sexual contact. Second, the accused had provided a reasonable explanation for his reticence on this issue in the following exchange: ²²⁸

It would be clear from this exchange that the accused spoke the way he did because he wanted to confine himself to giving evidence only on matters which he could remember as facts, and did not wish to engage in making inferences about things he could not remember, even if such inferences might appear to be logically inevitable to anyone listening to his evidence. While I found the accused’s concern rather strange and unusual, I could understand and appreciate his sentiment and explanation. I therefore did not regard this as a strong reason to doubt the accused’s credibility.

139 Going back to the narrative, the accused testified that sometime during the kissing, they had started “touching” and “fondling” each other: he described that the complainant was stroking his thighs and “actually reaching out to [touch] my thighs, my legs especially, the lower part of my body”. ²²⁹ He testified that she had straddled him and it was around this time that he inserted his right finger into her vagina and was fingering her, which formed the subject of the first charge against the accused. ²³⁰ At this time, the accused was still fully clothed.

140 By his account, the complainant started gyrating her hips and, at some point in time, laid down, spread her legs ²³¹ and subsequently tried to pull the accused towards her. ²³² By this time, the accused had removed his pants and attempted to penetrate the complainant with his penis. This formed the subject of the second charge against him. In this manner, the complainant reciprocated the accused’s kissing and touching. She allegedly fellated him at one point. And there was no evidence of any resistance or objection throughout. The accused interpreted these signs to mean that the complainant wanted sex. ²³³ Throughout the entire period of the first episode of sexual activities, the accused confirmed at trial that the complainant “never [said] anything”. ²³⁴

141 The Defence suggested that the complainant’s moaning sounds, part of which were captured by the in-car audio recordings during this period, did not indicate any resistance to the sexual activities and in fact suggested that she was responding favourably to the physical stimulation. In this sense, they broadly corroborated the accused’s version of events, in that the complainant had been a willing participant and reciprocated his kissing and touching. When listening to the in-car audio recording from this period, the complainant accepted that she was making “loud moaning noises” at various times throughout the first episode of sexual activities. ²³⁵ For instance, the in-car audio recording starting at timestamp 3:32:37 captured the complainant moaning in the first five seconds. The accused then asked her “are you okay” and the complainant appeared to give a short moan in response. ²³⁶

142 According to the accused, he ended the first episode of sexual activities when he became “suddenly … aware” of his surroundings and felt “uncomfortable with having sex in the open”. ²³⁷ Up to this point in time, the Car had been parked at the entrance of the complainant’s Condominium and the place was well lit. ²³⁸ The accused thus decided to shift his car before continuing with the sexual activities. He told the complainant to stop and that “[i]f you want to continue, maybe we go somewhere else”.

143 By the accused’s account, the complainant did not indicate any discomfort, verbally or otherwise, during this brief break in the sexual activities. He testified that the complainant remained in the back seat and said nothing, while he moved the Car to the road opposite the Condominium. ²³⁹ The in-car audio recordings (namely, starting at timestamp 3:43:41, 3:44:03, 3:45:04 and part of 3:46:04) corroborated this account in so far as they did not capture any verbal sounds of discomfort or protest from the complainant.

144 Neither was there any resistance from the complainant when the second episode began shortly after the accused parked his Car. This time, it was the accused who initiated intimacy as he testified that he “wanted to continue to have sex”. ²⁴⁰ Similar to the first episode, the complainant’s moaning continued and was captured by the in-car audio recording. The accused testified that there was some reciprocation of his kissing and touching her. ²⁴¹ It was during this time that the accused inserted his right hand beneath the complainant’s bra, fondled her breasts and sucked on her left nipple, which formed part of the subject of the third charge. He also attempted to insert his finger into her vagina, which formed the remaining part of the subject of the third charge. He testified that he stopped doing so when he sensed that she was “not as involved” and less participative than before. ²⁴²

145 This part of the accused’s testimony was consistent with the account given in his First Long Statement ²⁴³ and Second Long Statement.” ²⁴⁴ It was also corroborated by the in-car audio recording. At trial, the complainant agreed that she heard herself moaning and saying “no” in the in-car audio recording starting at timestamp 3:48:04, and that after she uttered the word “no”, no further moans from the complainant were heard. ²⁴⁵

146 I turn now to consider the Prosecution’s submissions outlined at [131] above, before making my findings on the credibility of the accused’s testimony.

147 First, the Prosecution submitted that the lack of verbalised consent supported an inference that the complainant did not consent. I did not understand the Prosecution to be submitting that consent must always be verbalised and could never be manifested in non-verbal ways. Instead, I believed that the point being made was that, had there been evidence of verbal consent, this would have been a point conclusively in favour of the Defence. However, since there was no verbal consent, it left room for the court to infer from the totality of the evidence that there was no consent. The foregoing understanding was confirmed by counsel for the Prosecution when he agreed with me during oral submissions that the lack of verbal consent was not conclusive because the law does not require consent to be verbal. Understood in this way, I did not find this submission objectionable. However, acceptance of this submission would not automatically lead to the inference that the complainant did not consent. It just meant that the court needed to look to other available evidence to determine the question.

148 Second, the Prosecution relied on the evidence of the paramedic, Ms Lee, who attended to the complainant after the event. Ms Lee stated that she had observed the complainant’s behaviour to be similar to those of other sexual assault patients in her experience. This evidence constituted lay opinion as Ms Lee had gone beyond mere observation of the complainant’s state to an evaluative comparison of (i) the complainant’s behaviour with (ii) those observed by her in past cases. Her evidence also consisted of a conclusion that (i) and (ii) were sufficiently similar as to warrant an inference of sexual assault. It was difficult to see how an assessment of this nature could be reliably conducted without some form of expert scientific evidence.

149  Thus, even if Ms Lee’s lay opinion was admissible, I would not give it much weight. Quite apart from the concerns raised above, her comparisons were drawn from a very small sample size of three or four past cases in her experience of working as a paramedic. ²⁴⁶ When pressed, she provided details on the behaviour and circumstances of those patients which were, at best, vague and sparse, and could not be tested by counsel or the court. This provided a shaky foundation for Ms Lee’s opinion at best.

150 In any case, Dr Lim had also explained that the complainant’s observed reaction might be attributable to confusion arousal and sleep inertia. ²⁴⁷ This offered a reasonable explanation which would be consistent with the facts. The complainant had just been awoken by the paramedics, in an unfamiliar location, with difficulty recalling what had occurred prior because of her alcohol-induced blackout. Accordingly, I decided to give no weight to Ms Lee’s opinion evidence.

151 Third, the Prosecution submitted that the complainant’s utterance of “no” towards the end of the sexual activities was an indication that she had not consented to any sexual activity with the accused. The Prosecution suggested that the complainant did not say “no” earlier only because “her state of intoxication rendered her incapable of doing so”. ²⁴⁸ Framed in this manner, this submission was in substance premised on the complainant not having the capacity to consent to any of the sexual activities which occurred before she uttered the word “no”. Given my earlier finding that the complainant had capacity at the time of the alleged offences, this submission would naturally fall away. In fact, my finding on capacity meant that the complainant’s utterance of “no” was more plausibly construed as a prospective withdrawal of consent which had, up to that point in time, been given.

152 The Prosecution also made a submission concerning the interpretation of the complainant’s moans. From the way this submission appeared within the structure of the Prosecution’s Closing Submission, it would seem that this was not an independent fourth plank of the Prosecution’s submission, but a rebuttal of an anticipated argument which was foreshadowed in the Defence’s earlier submissions for no case to answer. The submission was that the Defence would be wrong to argue that the only reasonable inference to be drawn from the complainant’s moans was that she had consented to the accused’s advances. My first observation would be that the Defence did not make this argument in its closing submission, so there was no argument along these lines for the Prosecution to rebut. My second observation would be that the Prosecution’s rebuttal referred to Dr Cheok’s comment that the complainant’s moaning could be described as “basal human automatic behaviours”. But Dr Cheok’s comment was made in response to a question relating to the capacity to consent and not to the issue of consent in fact. This was the actual exchange: ²⁴⁹

153 Thus the entire premise of this exchange was that the moaning could mean that the complainant was “participative” or even an “active participant” (ie, there was ostensible, non-verbal expressions of consent), but that may not mean she had the capacity to consent. Therefore, the exchange with Dr Cheok concerning “basal human automatic behaviour” did not in any way undermine the Defence’s case on consent in fact.

154 I found the accused to be a credible witness and accepted his version of events. The Prosecution suggested that he had embellished his testimony to portray the complainant as a more active participant during both episodes of the sexual activities. ²⁵⁰ I accepted that the accused went into more details when describing the sexual activities in his testimony at trial than in his investigative statements and that this could potentially be self-serving, but I did not consider the differences to be of such a scale or nature as to call the accused’s credibility into question. The accused explained that he did not go into certain details when describing the sexual activities because he was not sure how much details he was expected to go into, and IO Cheang did not ask for specifics. ²⁵¹ I found this explanation reasonable and credible. Often, the level of detail one goes into when narrating a past event depends on the nature of the questions posed as well as on one’s understanding of both the expectations of the questioner and the purpose of the narration. More importantly, there were also instances within the accused’s testimony which suggested that he did not shy away from evidence that could potentially be construed as inculpatory. One example was the accused’s candid admission, maintained across his investigative statements and at trial, that he had initiated the second episode of sexual activities and that the complainant did not appear as “involved” and “participative” to his advances in the same (see [144] and [145] above), even though she had continued moaning and reciprocating his physical touch.

155 It was also relevant that the accused’s investigative statements were given at a time when he was unaware that the complainant could not recall what happened during the sexual activities and thus would not be able to contradict him by her account. This suggested to me that he was candid and forthcoming. There was no evidence of any attempt by the accused during the investigative process to conceal his actions. He had returned home and changed his clothes without removing his underwear ²⁵² (which would have shown contact) or destroying evidence from the in-car camera recordings. IO Cheang agreed that the accused was “fully cooperative” and was “volunteering information” and did not give the impression “that he was trying to hide something or that he was holding back on anything” in the course of questioning. ²⁵³

156 Despite a handful of minor discrepancies which the Prosecution had highlighted, I found that the accused’s version of events were materially consistent across his investigative statements and at trial (see [135]–[145] above). His account was also externally consistent and positively corroborated by the objective evidence (ie, the in-car audio recordings) where available (see [135], [141], [143], [144] and [145] above). Crucially, the sexual activities ended once the complainant indicated, by her actions, that she was “not as involved” and less “participative” than before and expressly said “no” (see [144] and [145] above). This indicated the withdrawal of the consent which had been in fact provided. The accused recognised this and stopped his actions: ²⁵⁴

157 In support of the Prosecution’s submission on the credibility of the accused’s account, the Prosecution suggested that there was no good reason for the accused to wait around after the complainant had alighted from the Car, bring her back to the Car when she did not enter the Condominium after some time, and ask her to take a rest in the Car. The accused had originally intended to stop work at 3.00am to reach home at 3.30am as he had several appointments the next morning. ²⁵⁵ The Prosecution submitted that this behaviour was part of a sinister plan to monitor the extent of the complainant’s extent of intoxication with the intention of taking advantage of her. ²⁵⁶ I did not find this submission persuasive.

158 Given the lateness of the hour and the fact that the complainant was alone, I considered it entirely reasonable and appropriate for the accused to have waited and watched the complainant enter the Condominium safely before driving off, and to have gone to the complainant’s assistance when he observed her not entering the Condominium after some time. Further, if the accused had planned all along to sexually assault the complainant without her consent, he would likely have driven her to a more secluded area before the first episode of sexual activities, instead of engaging in the Sexual Acts in front of the Condominium’s gate in a brightly lit area. The fact that the accused stopped the first episode of sexual activities after a few minutes, upon realising that the Car was in a rather exposed location, tended to support the accused’s account that the sexual activities were spontaneous and unplanned. It was also relevant to note that the accused had initially declined to ferry the complainant upon being told by LYS that she was “a little drunk” but relented after LYS repeatedly reassured the accused that the complainant was okay. Finally, the accused stopped the sexual activities upon sensing the withdrawal of consent. This would clearly not have been the behaviour of a person who was seeking to take advantage of an intoxicated passenger without her consent.

159 I therefore found the accused’s testimony credible and considered that the accused’s version of events should be accepted in the main.

160 In summary, I accepted the Accused’s version of events and rejected the Prosecution’s submission that the evidence presented was sufficient for the court to infer, beyond reasonable doubt, that the complainant did not consent to the sexual activities. In light of the above, I found that the Prosecution had failed to prove that the Sexual Acts were committed without the complainant’s consent.

161 Given the above findings, it was not strictly necessary for me to pronounce on the defence of mistake. Nonetheless, for completeness, in the event that I was wrong on the question of capacity or consent in fact, I was prepared to find that the defence of mistake had been made out.

162 The Defence contended that the accused had applied due care and attention and mistakenly believed the complainant to have validly consented to engaging in sexual activities with him. ²⁵⁷ The accused believed that the complainant was not so intoxicated as to be unable to validly consent because she had been responding to the things he said to her in a relevant and appropriate manner throughout the night. ²⁵⁸ He also believed the complainant to have consented to sex, as her actions and non-verbal cues suggested that she was aroused and wanted to have sex. ²⁵⁹

163 In response, the Prosecution’s case was that it was not plausible for the accused to have believed in good faith that the complainant had given valid consent. ²⁶⁰ He had known from the outset, ie, from the time he picked her up at the Bar, that she was intoxicated ²⁶¹ and had decided to take advantage of her intoxication. ²⁶² It was also the Prosecution’s case that the complainant could not have participated enthusiastically in the sexual activities as she was too tired and intoxicated to do so. ²⁶³ Furthermore, she did not have any experience in respect of performing fellatio and would not have acquired such knowledge when she was intoxicated. ²⁶⁴

164 As a preliminary observation, it was unclear to me why the Prosecution raised the last two submissions (concerning tiredness and lack of experience with fellatio) in relation to the defence of mistake, as opposed to raising them to challenge the accused’s credibility on the issue of consent in fact. Since the Prosecution has made these submissions in relation to the defence of mistake, I would deal with them as part of my analysis on the defence of mistake. In any event, given the views I have taken on the merits of these two submissions (at [182]–[184] below), it would make no difference to the final outcome of this case whether I dealt with them as submissions raised in relation to the issue of consent in fact or in relation to the defence of mistake.

165 In situations where the accused seeks to argue that he believed that the victim was consenting, it has been held that the proper approach for the court to approach the matter is through the mistake of fact defence under s 79 PC rather than through a mens rea analysis (Public Prosecutor v Teo Eng Chan and others [1987] SLR(R) 567 (“Teo Eng Chan”)). Section 79 PC provides:

166 The burden of proof under this provision is upon the accused and has to be discharged on a balance of probabilities and not beyond reasonable doubt (Teo Eng Chan at [26]). “Good faith”, in turn, is defined in s 26B PC:

167 In Tan Khee Wan Iris v Public Prosecutor [1995] 1 SLR(R) 723, which concerned an offence of providing public entertainment without a valid licence under s 18(1)(a) of the Public Entertainments Act (Cap 257, 1985 Rev Ed), Yong Pung How CJ observed in relation to the accused’s reliance on the defence of mistake of fact under s 79 PC (at [19]):

168 The mistake of fact that the accused raised in the present case was that he believed in good faith that the complainant had the capacity to consent and did in fact consent to the Sexual Acts.

169 On the totality of the evidence before me, I found that the accused had discharged his burden of proof and successfully availed himself of the defence of mistake of fact. All the factors that I earlier listed for finding that the complainant had capacity to consent also constituted external manifestations of the complainant’s demeanour and behaviour that could lead the accused to reasonably believe in good faith that she had the capacity to consent. In a similar way, the complainant’s conduct in initiating and reciprocating the sexual episodes were construed in good faith by the accused as consent to the Sexual Acts.

170 What distinguished this case, among other features, was the complainant’s alcohol-induced blackout (see [64] and [65] above). Dr Cheok and Dr Lim both agreed that individuals in such a state might appear sober or competent to others around them. In other words, those interacting with them might not realise that the individual was in fact intoxicated (or so intoxicated as to experience an alcohol-induced blackout). ²⁶⁵ They might not even develop a reasonable suspicion of this. They might not be able to discover the complainant’s actual state even assuming efforts were diligently made. It was therefore incorrect to simply assume that because the complainant was in fact intoxicated to some extent, she must have appeared to the accused to be so intoxicated as to be incapable of consenting to sexual activities.

171 The issue therefore required a close examination of the facts, directed at how the complainant behaved and appeared to others, as well as what the accused knew of her alcohol consumption that night.

172 As to the latter, the Prosecution’s case was that the accused knew from the outset that the complainant was intoxicated. ²⁶⁶ The Defence’s case was that the accused had thought that she was not so intoxicated as to be unable to validly consent to sex, because she had been responding appropriately and relevantly to him throughout the night. ²⁶⁷ The accused also testified that he had not relied solely on LYS’s and VT’s assurances that the complainant was not too drunk, but had also observed the complainant’s demeanour in coming to his own impression that she was well enough to not make a mess in his car. ²⁶⁸

173 At the outset, I found that LYS and VT had conveyed the impression to the accused that the complainant was not too drunk. LYS initially told the accused that the complainant was “a little drunk”, to which the accused replied “I cannot help you if she’s drunk”. LYS assured him by saying “She’s okay” repeatedly and that “She won’t vomit”. ²⁶⁹ VT’s evidence was that she also added (in Mandarin) that “She’s not drunk”, “She’s okay” and “She won’t vomit”. ²⁷⁰ These were corroborated by the in-car audio recording.

174 The Prosecution relied on the fact that LYS had nonetheless handed the complainant a plastic bag (which might suggest that he had assessed that the complainant would vomit). ²⁷¹ This must, however, be viewed in context: LYS told the accused that he was giving the plastic bag to the complainant “just in case” and had repeatedly told the accused that the complainant would not vomit. ²⁷² Neither did the fact that the accused thereafter told the victim “if you want to puke, just let me know” necessarily suggest that he thought the complainant to be so intoxicated as the Prosecution suggested. The accused testified that he had said this as the exchange with LYS “just happened less than a minute … ago” and was “still fresh on [his] mind”. ²⁷³

175 I thus accepted that the accused had arrived at the opinion that the complainant had a few drinks but was not intoxicated to the extent of being unaware of her surroundings, at the time he ferried her from the Bar. ²⁷⁴

176 It was also relevant that throughout the time the accused was with the complainant on the night in question, there were several points across the night where she responded appropriately and relevantly to the accused’s questions and suggestions as well as to her surroundings (see [105]–[106] and [111]–[127] above). In addition to these exchanges set out above, the complainant had also responded to the accused’s question of “if you want to puke, just let me know” (see [174] above). The in-car audio recording captured her replying “it’s okay, it’s okay”. The complainant agreed that she had sounded “alert” in the in-car audio recording which captured this exchange and that this was an appropriate response to the accused’s question. ²⁷⁵ In another instance, which took place at an early stage during the sexual activities, the accused had told the complainant “I can’t put it in” (by which he meant that he had difficulty inserting his fingers into her vagina) and she had adjusted her position to accommodate him. The accused must have said this to the complainant because he genuinely believed that the complainant would be able to understand what he was saying and react accordingly. Another instance involved the accused saying “quiet” to the complainant when her moaning grew louder. The accused could only have said this if he believed that the complainant would understand and respond accordingly. In the same vein, also relevant was the accused saying “I just take a round, okay?” to the complainant when he took a short drive between the two episodes of sexual activities.

177 While the Prosecution relied on evidence of the complainant’s “hysterical” conduct in thumping her chest and hitting her head on the window, ²⁷⁶ it was also particularly significant that the complainant had momentarily stopped doing so several times when the accused verbally urged her to stop (see [121] and [122] above). I also noted the evidence of Dr Lim, who testified that the thumping of the chest “would not be done by somebody who is significantly intoxicated” and was, if he could speculate, “a gesture that she is very, very frustrated or in distress”. ²⁷⁷

178 These interactions between the accused and complainant satisfied me that the accused had come to a good faith belief that the complainant had capacity to consent after exercising due care and attention.

179 On the totality of the evidence before me, I found that the complainant’s conduct in initiating and reciprocating the sexual activities were construed in good faith by the accused as consent to the Sexual Acts.

180 First, she initiated the series of acts which led to the sexual encounter, beginning by kissing him. The context of this act was crucial, namely, that the accused had gone to the back seat of the Car with the intention of looking through the complainant’s handbag for a way to contact someone to come and get her, and not to make advances on her. Second, as the sexual activities progressed, the complainant reciprocated the accused’s actions by kissing and touching him, straddling him, and gyrating her hips. She spread her legs and subsequently pulled him towards her. When the accused told the complainant, “I can’t put it in”, she responded by changing her position to accommodate him. The accused interpreted these external signs to mean that the complainant wanted sex. Third, there were no signs of resistance from the complainant throughout. Neither was there any sign of resistance or protest from her during the brief break, during which the accused moved his Car to the other side of the road. Sometime during the first sexual episode, the in-car audio recording captured the accused asking the complainant, “are you okay”, and the complainant appeared to give a short moan in response. This suggested that the complainant had responded favourably to the physical stimulation. Fourth, neither was there any external sign of resistance when the second episode of sexual activities began. Similar to the first episode, the complainant’s moaning continued, with some reciprocation of kissing and touching. Eventually, she indicated through non-verbal cues that she did not wish to continue and said the word “no”, upon which the accused stopped touching her and returned to the driver’s seat.

181 I was therefore satisfied that the accused had exercised due care and attention in the course of the sexual episodes.

182 The Prosecution submitted that it was clear from the evidence that the complainant had been too tired to have participated in the manner described by the accused – ie, sitting on his thighs with her legs apart and grinding her crotch against his groin. In my view, this submission has not been made out. Admittedly, the complainant was intoxicated, it was already the wee hours of the morning and there was some evidence which suggested that the complainant was sleepy (see [99] and [108] above). But having seen and heard the complainant’s behaviour from the CCTV footages and in-car audio recordings, I would be unable to agree that the complainant was not in a state to perform the types of activities described above. She was able to get out of the Car on her own and walk unassisted, and had sufficient energy to pound on her chest loudly enough for it to be captured in the in-car audio recording.

183 Neither did the fact that the complainant was later found lying unconscious in the middle of the road support the inference of lethargy to the extent that the Prosecution sought to draw. For the reasons given above at [129], I did not find evidence concerning the complainant’s physical state after the sexual activities had taken place to be probative.

184 Lastly, the Prosecution asserted that the accused’s account of the alleged fellatio was “highly implausible” as the complainant testified that she did not have sexual experience in this regard and did not know the basic mechanics of how fellatio worked. ²⁷⁸ It was unclear how much weight could be safely placed on this part of the complainant’s testimony. The complainant had given inconsistent evidence before, such as her under-reporting of her drinking habits. In any event, it was clear from her evidence that she knew what oral sex was and had heard from people she knew about oral sex being performed. ²⁷⁹ With at least this level of knowledge, it was not beyond belief that she might, on this occasion, have tried carrying out a new sexual act for the first time given the disinhibitory effect of alcohol (of which, see discussion at [84] above). The Defence also raised another area of concern in relation to this submission. IO Cheang had acknowledged that there was an available test that could determine whether saliva was present on the accused’s underwear. However, this test was not done during investigations and, by the time the matter came to trial, it was no longer possible to perform such a test. The Defence therefore submitted that an important opportunity to verify the accused’s account had been missed. Taking the foregoing factors together, I considered that the Prosecution has not sufficiently persuaded me that the accused should be disbelieved on this issue.

185 For the reasons given above, I found that the accused had established on a balance of probabilities that he believed in good faith, after exercising due care and attention, that the complainant had the capacity to consent and did in fact consent to the Sexual Acts. Consequently, the defence of mistake under s 79 PC had been made out.

186 In light of the above, I found that the Prosecution had failed to prove its case beyond a reasonable doubt. I therefore acquitted the accused of all three charges.