1 In cases involving an allegation of sexual assault, it is possible for a conviction to be secured even if the only substantive evidence against the accused is the uncorroborated evidence of the complainant. This is so even though the evidential threshold that the Prosecution must meet to secure such a conviction is that the relevant facts are established beyond a reasonable doubt and an accused person is presumed to be innocent until the Prosecution is able to meet that legal standard. That threshold may be met if the uncorroborated evidence of the complainant is so “unusually convincing” as to overcome any doubts that may arise from the lack of corroboration.

2 Yet, when determining if the “unusually convincing” threshold has been reached, it is incumbent upon the court, as the last line of defence against injustice, to apply the standard of proof beyond reasonable doubt with the utmost rigour and precision. This is all the more so when, as in the present case, both the accused person and the complainant had given diametrically opposing accounts of what had happened and there is little to no objective corroborative evidence. Such a situation necessitates a descent into the granularity of the complainant’s testimony and the available evidence. In this case, the issue of whether the complainant’s evidence had met the high threshold of being “unusually convincing” would prove to be determinative of the question of whether it was safe to convict the accused of the sexual assault he was alleged to have committed.

3 The accused, Mr Lev Panfilov (“Mr Panfilov”), and the complainant met online on the dating application Tinder. A few days later, on 12 January 2021, they proceeded to meet in person at a restaurant. From the restaurant, the complainant followed Mr Panfilov back to his apartment where they engaged in a number of sexual acts. On 18 January 2021, the complainant lodged a police report alleging that she was raped and sexually assaulted by Mr Panfilov. Mr Panfilov claimed trial to the following four charges in the court below: (a) one charge of outrage of modesty under s 354(1) of the Penal Code (Cap 224, 2008 Rev Ed) (“PC”) (“the First Charge”); (b) one charge of sexual assault by penetration under s 376(2)(a) punishable under s 376(3) of the PC ( “the Second Charge”); (c) one charge of rape (penile-oral) under s 375(1A)(a) punishable under s 375(2) of the PC (“the Third Charge”); and (d) one charge of rape (penile-vaginal) under s 375(1)(a) punishable under s 375(2) of the PC ( “the Fourth Charge”).

4 At trial, Mr Panfilov and the complainant gave divergent accounts of what transpired in the days leading up to, and during, their sexual encounter. Mr Panfilov characterised his meet-up with the complainant at the restaurant as a date. The complainant testified that she met up with Mr Panfilov for purely professional reasons, namely, to obtain advice for a script she was working on. Mr Panfilov accepted that he committed the sexual acts underlying the charges (save for the act of digitally penetrating the complainant’s vagina, which he denied), but maintained that his sexual encounter with the complainant was entirely consensual. The complainant denied that she had consented to any of the sexual acts. It was her position that she was the victim of an exceptionally violent assault by Mr Panfilov, one which left her with significant physical and emotional trauma. Mr Panfilov’s case theory was that the complainant reported that she had been raped because she was not happy that he subsequently ghosted her. (As we understood it, the term “ghosted”, as used in the modern dating lexicon, refers to a situation in which one party abruptly ceases all communication with another party.)

5 After a trial lasting 27 days, Mr Panfilov was convicted on all four charges by a Judge of the High Court (“the Judge”) and sentenced to 11 years and six months’ imprisonment and 12 strokes of the cane. Mr Panfilov appealed against his conviction and sentence in CA/CCA 12/2025 (“CCA 12”). The Prosecution appealed against the sentence in CA/CCA 13/2025 (“CCA 13”).

6 In the present case, the learned Judge considered the Prosecution to have met the legal standard of proof beyond reasonable doubt in relation to the charges faced by Mr Panfilov. Having deliberated on the evidence and the parties’ arguments, we found, with the greatest of respect, that the Judge had erred in coming to this conclusion. In arriving at our decision, we came to the view that the evidence of the complainant did not meet the (high) standard of being “unusually convincing”. As such, we allowed Mr Panfilov’s appeal in CCA 12 and acquitted him of all charges. Consequently, we dismissed the Prosecution’s appeal against sentence in CCA 13. These are the written grounds of our decision, which elaborate on and supplement the brief oral grounds which we delivered on 2 September 2026.

7 We first begin by summarising the salient background facts, which can substantially be found in the grounds issued by the Judge in Public Prosecutor v Lev Panfilov [2025] SGHC 249 (“the GD”).

8 At the material time, Mr Panfilov was working as an actor in a comedic YouTube channel. He was also involved in comedic scriptwriting as part of his role.

9 In early January 2021, Mr Panfilov and the complainant “matched” on Tinder, an online dating application. The complainant, who had worked as a model and actress for some time, was trying to branch into comedy and learn comedic scriptwriting on her own at the time because she had developed a medical condition which affected her facial muscles and limited her prospects of continuing with an acting career. After Mr Panfilov and the complainant “matched” on Tinder, they proceeded to chat with each other on Tinder and, subsequently, on WhatsApp. A few days later, they agreed to meet at Wine Connection at Robertson Quay on 12 January 2021.

10 At this juncture, and as briefly alluded to above (at [4]), it should be noted that Mr Panfilov and the complainant offered fundamentally different characterisations of their interactions with one another on Tinder and WhatsApp as well as their meeting at Wine Connection. Mr Panfilov testified that he was “swiping through Tinder” to look for a date when he came across the complainant’s profile, saw that the complainant had already “liked” him, felt that the complainant’s photo “seemed attractive to [him]” and, hence, swiped right on the complainant’s profile and “matched” with her. According to Mr Panfilov, he and the complainant “agreed to meet up for a date”. The complainant, on the other hand, explained that she was using Tinder to “meet people”, “broaden [her] social circle” and “share knowledge with each other”. According to the complainant, she only “matched” with Mr Panfilov on Tinder as his profile indicated that he was either a writer or a scriptwriter and she was hoping to get some advice on scriptwriting from him. The complainant denied that the meeting at Wine Connection was a date. In other words, while Mr Panfilov characterised his interactions with the complainant prior to and at Wine Connection as having romantic undertones and encompassing some degree of mutual attraction, the complainant’s account was devoid of these aspects and was centred around only the specific aim of improving her scriptwriting.

11 Leaving that aside, it was undisputed that Mr Panfilov and the complainant met at Wine Connection, at around 8pm on 12 January 2021, where they exchanged small talk and worked together on a script which the complainant was in the process of writing. Around one hour and 40 minutes later, when Wine Connection was closing, Mr Panfilov and the complainant agreed to go to Mr Panfilov’s home (“the Flat”).

12 Mr Panfilov and the complainant made their own separate ways to the Flat from Wine Connection. When the complainant arrived at the Flat, Mr Panfilov ushered the complainant to his bedroom. There, they sat on Mr Panfilov’s bed to work on the complainant’s script. For around one-and-a-half hours, Mr Panfilov and the complainant talked about writing characters and watched YouTube videos which each of them found funny. At some point, Mr Panfilov started kissing the complainant. It was from this point onwards that the accounts of Mr Panfilov and the complainant diverged. As such, we will, in the following section, summarise the Prosecution’s and the Defence’s respective cases in the court below.

13 The Prosecution’s case was that the complainant did not consent to Mr Panfilov’s kiss mentioned in the preceding paragraph and the various sexual acts that followed in the Flat.

14 According to the complainant, while she and Mr Panfilov were watching comedians on YouTube, Mr Panfilov suddenly leaned in and kissed her on the lips. She was shocked and told him “no”, but he continued to kiss her again on the lips, used both of his hands to touch her breasts, and put her hand over his erection. When she tried to gather her belongings from a table in Mr Panfilov’s room so that she could leave the room, Mr Panfilov hugged her from behind and rubbed his body against her such that she could feel his erection against her buttocks. At this point in time, the complainant tried to move her body but could not break free. While Mr Panfilov was hugging the complainant, he slipped his hands underneath her dress and grabbed her breasts. This caused the complainant to feel pain, and she told him “no” again. Mr Panfilov’s acts of kissing the complainant on her lips, hugging her, and grabbing her breasts, formed the basis of the First Charge.

15 Mr Panfilov continued to grab the complainant’s breasts. He then removed one hand from the complainant’s breasts, inserted that hand underneath her dress and into her panties, and then inserted his fingers into her vagina. Mr Panfilov moved his fingers in and out of the complainant’s vagina, causing her severe pain. Mr Panfilov’s alleged act of digital penetration of the complainant’s vagina formed the basis of the Second Charge.

16 After some time, Mr Panfilov withdrew his hands from the complainant’s body, grabbed the complainant’s arm, and turned the complainant to face him. He then pulled down his shorts to expose his genitalia and pushed the complainant’s head down. Mr Panfilov forced the complainant to assume a kneeling position and to perform fellatio on him. While the complainant tried to close her mouth to prevent this, she was unsuccessful in doing so as Mr Panfilov “kept pushing” his penis in. The complainant was gagging and choking during this act. Mr Panfilov’s act of penetrating the complainant’s mouth with his penis formed the basis of the Third Charge.

17 At some point, Mr Panfilov withdrew his penis from the complainant’s mouth, pulled the complainant up by her arm, pulled her to the side of the bed, and then pushed her face down onto the bed. Mr Panfilov then pulled down the complainant’s underwear and flipped her dress up to expose her vagina and buttocks. He proceeded to insert his penis into the complainant’s vagina from the back, which caused her so much pain that she wanted to scream and she kept flinching. The complainant felt him put a hand on the back of her neck and a pillow over the back of her head while he was penetrating her. According to the complainant, she felt suffocated and wanted to scream but could not do so. Mr Panfilov then flipped the complainant over and continued penetrating her vagina with his penis from the front. Mr Panfilov’s acts of penetrating the complainant’s vagina with his penis from the back and from the front formed the basis of the Fourth Charge. It was undisputed that Mr Panfilov had put a condom on at some point in time before he penetrated the complainant’s vagina with his penis, though the complainant did not testify as to when exactly Mr Panfilov had worn the condom (see [122]–[124] below).

18 After some time, Mr Panfilov withdrew his penis from the complainant’s vagina and released his grip on her neck and waist. The complainant saw “white stuff” at the tip of the condom that Mr Panfilov wore. The complainant gathered up her strength to put on her underwear and get her bag, following which Mr Panfilov told her that she could go. She then attempted to book a Grab car while standing at a corner of the bedroom. However, she was unable to do so. Mr Panfilov then asked for her address and booked her a Gojek car to go home.

19 We will, in the course of these written grounds, refer to the various acts underlying the four charges against Mr Panfilov collectively as the “Incident”.

20 The complainant recalled replying to Mr Panfilov’s messages asking her if she had reached her home. According to her, she was still experiencing moderate pain in her vaginal and abdominal areas after returning home. She went back to her room and, when she took off her underwear in the toilet, noticed that there was blood on her underwear (the “blood-stained underwear”). She broke down in tears, and then placed her underwear and the dress which she was wearing in her laundry basket. The significance of the complainant’s blood-stained underwear will become apparent below.

21 At some point on 13 January 2021 (ie, one day after the Incident), the complainant “unmatched” Mr Panfilov on Tinder and deleted the messages which they exchanged on WhatsApp as she felt disgusted about what happened and “[did not] want anything to do with [Mr Panfilov] anymore”.

22 On 16 January 2021, the complainant continued to feel pain in her lower abdominal and vaginal area and was worried about whether she had contracted a sexually transmitted infection or become pregnant due to her sexual encounter with Mr Panfilov. She visited the Clementi Family and Aesthetic Clinic (“Clementi Clinic”) and informed PW3 Dr Foo Minli Joanna (“Dr Foo”) that she was raped. Dr Foo explained that she could not examine the complainant or otherwise help her until a police report was lodged. Accordingly, the complainant proceeded to the Clementi Neighbourhood Police Centre (“Clementi NPC”) on the same day.

23 At Clementi NPC, the complainant spoke to several police officers, including PW9 Station Inspector Lim Chew Boon (“SI Lim”), but chose not to file a police report as she: (a) felt that she had no evidence; (b) was worried about causing more stress to her mother if her mother were to find out about the Incident through the police; (c) was worried about students whom she was teaching at the time finding out about the Incident; (d) was worried that Mr Panfilov may sue her if she lodged a police report; and (e) felt traumatised at having to explain what happened to a few different officers.

24 Around midnight on 18 January 2021, however, the complainant told her mother, PW12, about the Incident. The complainant’s mother brought the complainant to KK Women’s and Children’s Hospital (“KKH”), where a police report was lodged. The complainant brought the underwear and dress she wore during the Incident to KKH. At KKH, the complainant was also examined by PW8 Dr Geetha Visvalingam (“Dr Geetha”).

25 Mr Panfilov admitted that he had kissed the complainant, touched her breasts, and penetrated her mouth and vagina with his penis (ie, he admitted to having done the acts underlying the First, Third and Fourth Charges). However, he denied inserting his fingers into the complainant’s vagina (ie, he denied having done the act underlying the Second Charge). According to him, “putting dirty fingers into vaginas [was] not really his style” because it was “not as hygienic” (as having sex with a condom). Mr Panfilov’s position was that the complainant consented to the acts underlying the First, Third and Fourth Charges.

26 According to Mr Panfilov, it was the complainant who first made advances on him at the Flat by laughing at his jokes, tapping on his leg and stroking his thigh. He then moved in to kiss the complainant, which she reciprocated. He then undressed the complainant, following which the complainant fellated him. After this act of fellatio, Mr Panfilov went to retrieve a condom from his bathroom, returned to his bed where he proceeded to remove the complainant’s underwear and struggled to put on the condom. After he managed to put the condom on, Mr Panfilov and the complainant then engaged in sexual intercourse on his bed. After penetrating the complainant from both the front and rear, he withdrew his penis from her vagina and ejaculated into the condom which he was wearing.

27 After the sexual encounter, the complainant dressed herself and used the toilet in his room. They then shared some jokes, including a joke from the complainant which Mr Panfilov found morbid and off-putting. Mr Panfilov then asked the complainant if he could call her a cab, which the complainant accepted. They continued engaging in a conversation about Mr Panfilov’s experiences performing stand-up comedy while waiting for the Gojek car to arrive.

28 While Mr Panfilov was not sure if the complainant had messaged him after 12 January 2021, he testified that he did not feel like chatting with the complainant any longer as he found her jokes and personality to be off-putting.

29 On 18 January 2021, Mr Panfilov received a call from PW1 ASP Dan Dunstan Cheang (“ASP Cheang”). He met ASP Cheang, along with PW13 ASP Ho Baoxin (“ASP Ho”) and PW14 DSP Muhammad Jamil bin Agus Rizal (“DSP Jamil”), at his workplace, following which he brought the three police officers to the Flat on their instructions. He was then escorted to Police Cantonment Complex (“PCC”), where a video recorded interview (“the VRI”) was conducted.

30 As the only witnesses to the alleged offences were the complainant and Mr Panfilov themselves, the Judge considered whether the complainant’s testimony was “unusually convincing” with reference to the considerations of internal consistency, external consistency and demeanour.

31 The Judge was of the view that the complainant provided a “textured, vivid and consistent” testimony of the assault which remained unchanged despite having undergone a long and difficult cross-examination that lasted over 13 non-consecutive trial days spanning a period of six months. Moreover, the Defence did not point the court to any internal inconsistencies in the complainant’s testimony. The Judge thus concluded that the complainant’s testimony was internally consistent.

32 The Judge then proceeded to evaluate the external consistency of the complainant’s testimony against her out of court statements to the other witnesses and the evidence surrounding the blood-stained underwear. In this regard, the Judge considered the complainant’s testimony to be generally consistent with the statements that she made outside of court to Dr Foo, SI Lim and Dr Geetha:

(a) Dr Foo’s account of what the complainant told her at the Clementi Clinic was largely consistent with the complainant’s testimony, except for a part about the complainant asking for a memo stating the extent of her injuries and what the probable cause of those injuries might have been with a view to potentially approaching Mr Panfilov’s family, which was missing from the complainant’s testimony. The Judge was of the view that the complainant’s inability to remember that she had asked Dr Foo for a memo was merely an omission which should not be treated as an inconsistency unless the court believed that the complainant was being evasive and was trying to hide an unfavourable fact, and that it was not necessarily inconsistent for the complainant to have wanted nothing to do with Mr Panfilov after the attack and yet want a memo to bring to his family. Furthermore, the complainant’s omission to mention to Dr Foo that she bled as a result of the attack was not a significant one as the consultation was rather short, she was still traumatised, stressed, and confused, and the consultation itself was not focused on her injuries.

(b) The complainant’s account of the assault to SI Lim at Clementi NPC, and her insistence on not wanting to lodge a report as well as the reasons she gave for not wanting to do so, as recounted by SI Lim in his conditioned statement, was consistent with the case advanced by the complainant at trial.

(c) While the complainant’s story before Dr Geetha was not entirely consistent with the evidence that she gave on the stand, the Judge did not think that this inconsistency was material. Essentially, the complainant’s account to Dr Geetha at KKH, as recorded in Dr Geetha’s medical report, was that she screamed during the assault and Mr Panfilov tried to choke her and put a pillow over her head to silence her whereas her account on the stand was that she wanted to scream but did not scream. The Judge found the complainant’s explanation for this inconsistency, namely that Dr Geetha may have misunderstood her because she was very tired, had not been sleeping well, and was still in pain when she met Dr Geetha, to be credible, because the evidence surrounding the complainant’s state of mind at the time suggested that she was “confused, unsure, and still traumatised”.

33 As regards the blood-stained underwear, the Judge was confronted with the opinions of two experts, Dr Geetha and DW3 Dr Harvard Lin Zhenjia (“Dr Lin”), who was the medical expert put forward by the Defence, on the possible sources of the blood on the underwear (which the complainant attributed to Mr Panfilov’s assault on her – see [20] above). Apart from finding an old hymen tear when examining the complainant at KKH (which was likely to have been sustained from previous sexual experience), Dr Geetha only found a “very small” 1 cm x 1cm abrasion just to the left of the complainant’s vaginal fourchette which was tender to the touch and caused her pain when pressure was exerted on it. It was common ground between Dr Geetha and Dr Lin that the amount of bleeding seen on the blood-stained underwear could not have been caused by the 1cm x 1cm abrasion but could have been caused if there had been a laceration of the mucosal tissue within the complainant’s vagina. However, the experts differed on whether such a laceration could have healed completely by the time of Dr Geetha’s examination of the complainant on 18 January 2021. In the Judge’s view, the root cause of this difference in opinion lay in the experts’ differing factual observations concerning the amount of blood on the underwear, which led to differing views concerning the size of the laceration needed to produce such amount of blood. The size of a laceration would, in turn, affect how long it takes to heal. Dr Lin had overstated the amount of blood seen on the underwear and, thus, his opinion was based on wrong factual observations. The Judge therefore found Dr Geetha’s opinion more persuasive and adopted her views, with the consequence that there was no external inconsistency between the complainant’s testimony concerning the blood-stained underwear and the expert medical evidence.

34 Apart from the purported internal and external inconsistencies alluded to above, the Judge also dealt with the following points raised by the Defence as part of its submissions on the internal and external inconsistencies in the complainant’s testimony even though, in the Judge’s view, it was arguable that some of these arguments did not directly concern internal or external consistency:

(a) The Defence contended that the complainant deleted her WhatsApp messages with Mr Panfilov and unmatched him on Tinder to support her contrived allegations against him, and that she would not have kept the blood-stained underwear if she was so disgusted to the point of deleting all her messages with him. However, there was nothing surprising or suspicious about the complainant’s decision to delete her WhatsApp conversation with Mr Panfilov after the Incident as, based on her evidence and observations of her demeanour by the other witnesses, she was traumatised and scared after the attack. The complainant’s retention of the blood-stained underwear was also not inconsistent with her decision to delete her WhatsApp conversation with Mr Panfilov as the evidence suggested that she dropped the clothes she was wearing on the night of the Incident in her laundry basket and thought no further about them until her mother reminded her on 18 January 2021 to bring those clothes to KKH.

(b) The Defence submitted that, as Tinder was a dating application, the complainant’s explanations on why she was using Tinder and why she had “matched” with Mr Panfilov (see [10] above) was rife with inconsistencies and ought not to be believed as she met exclusively with men on Tinder and told Dr Foo that she was on a date with Mr Panfilov. The Judge was of the view that, although Tinder was designed as a dating application, there was no reason why a person could not use it for the more general purpose of meeting people or broadening one’s social circle. Since Tinder was originally intended to be a dating application, it was not surprising that the application would match the complainant only with members of the opposite sex. Moreover, the Defence had not summarised Dr Foo’s evidence correctly as Dr Foo did not actually say that the complainant told her that she was on a date with Mr Panfilov.

(c) The Defence argued that the complainant’s lack of explanation for when Mr Panfilov’s condom was worn was a glaring omission as the time it would have taken for him to go to the bathroom to retrieve the condom was inconsistent with the complainant’s account which gave the impression that he penetrated her vagina from behind immediately after he had pushed her onto the mattress (see [17] above). The Judge did not think that the complainant would have registered or been able to know when exactly Mr Panfilov put on the condom as, based on her testimony, her eyes were still closed and she was trying to fight back her vomit when she was pushed onto the mattress faced down. As such, the complainant’s lack of explanation for when exactly the condom was worn did not constitute an inconsistency that affected her credibility.

35 Additionally, the Judge found that the complainant’s demeanour supported a finding that she was credible witness. In this regard, the complainant’s “instinctual reactions”, such as her physically gagging in court in response to recalling the act of her fellating Mr Panfilov, suggested that she was indeed attacked on the night of the Incident. While the Judge acknowledged that the complainant appeared to be “cagey” at some points during cross-examination, such as by requesting counsel to define mundane words, it was a natural corollary of her worries that she might not be believed and about the quality of the evidence that she was giving, as well as Defence counsel’s accusations of her fabricating her evidence in court, that she was careful – even hypervigilant – of the answers that she was giving during cross-examination. Furthermore, the complainant was forthcoming where doubts in the evidence could be resolved in Mr Panfilov’s favour.

36 In the round, the Judge concluded that the complainant’s testimony met the “unusually convincing” standard.

37 The Judge then went on to consider Mr Panfilov’s testimony and concluded that it did not raise reasonable doubts on the complainant’s evidence. While Mr Panfilov’s account of the actual sexual encounter was consistent both at trial and in the VRI, there were significant inconsistencies between the VRI and his evidence-in-chief (“EIC”) at trial concerning events leading up to the sexual acts and points where the position he took initially in the VRI supported the complainant’s case but not his own at trial:

(a) First, Mr Panfilov admitted that he had lied to ASP Cheang, ASP Ho and DSP Jamil on 18 January 2021 (see [29] above) that he had not engaged in sexual intercourse with the complainant. It was only shortly before the VRI at PCC that he confessed to having sex with the complainant but said that it was consensual.

(b) Second, there was an inconsistency in Mr Panfilov’s account of how he introduced himself to the complainant and whether he discussed scriptwriting with the complainant prior to meeting her. Mr Panfilov’s introduction of himself to the police as a scriptwriter in the VRI was inconsistent with his position at trial that he would “most likely not” have introduced himself to anyone as a scriptwriter during the period of time he met the complainant on Tinder. Moreover, Mr Panfilov’s position in the VRI, that there was nothing strange about him and the complainant working on the complainant’s scripts together at Wine Connection as they were both interested in writing and he wanted to help her write, as well as his explanation in the VRI that he was working as a scriptwriter and that he and the complainant talked about her scripts on WhatsApp, supported and corroborated the complainant’s account.

(c) Third, Mr Panfilov presented inconsistent versions of the sequence of events leading up to the sexual encounter with the complainant. Mr Panfilov’s account in his EIC suggested that he and the complainant were mutually interested in each other, while his account in the VRI suggested that he was a more disinterested and reluctant participant in the date.

38 Finally, the Judge dealt with the following remaining issues:

(a) The Defence highlighted that, notwithstanding the complainant’s allegations of the various violent, non-consensual sexual acts committed by Mr Panfilov, she voluntarily gave him her home address for him to book a Gojek car for her to go home. The Judge did not think that giving Mr Panfilov her home address was inconsistent with the complainant’s allegation of sexual assault as there was some force in the complainant’s explanation that what happened to her was not normal. The evidence also showed that the complainant was likely suffering from some form of “tonic immobility” as she testified that her body went limp during the attack and that she was gagging, scared and traumatised. Thus, the complainant’s decision to give Mr Panfilov her home address so that she could get a ride home was explicable.

(b) The Defence submitted that the complainant’s act of messaging Mr Panfilov to inform him that she had arrived home (the complainant testified that she recalled replying to Mr Panfilov’s message checking if she had gotten home safely) undermined her allegation that she had been violently raped by him. The Judge found the complainant’s explanation for her behaviour, namely that she was too scared not to respond to the Mr Panfilov’s message as he had her address, to be adequate. Therefore, the Judge did not agree with the Defence that this was behaviour which undermined the complainant’s credibility.

(c) While the Judge observed that the complainant’s late filing of the police report called for some examination, he was ultimately of the view that the complainant had given reasonable and cogent explanations for her delay having regard to the circumstances surrounding the offence (ie, there being no witnesses), the family situation (ie, her mother’s heart condition) and her own worries and concerns (see [23] above). The Judge thus did not regard the late reporting as having the effect of casting reasonable doubts on the Prosecution’s case.

39 As the Defence did not succeed in raising any reasonable doubts in the Prosecution’s case, the Judge accepted the complainant’s account, held that the Prosecution had proven all four charges beyond reasonable doubt, and found Mr Panfilov guilty on all four charges.

40 After hearing the parties’ arguments on sentence, the Judge imposed the following sentences on Mr Panfilov:

41 The imprisonment terms for the First and Fourth Charges were ordered to run consecutively, resulting in a global sentence of 11 years and six months’ imprisonment and 12 strokes of the cane.

42 The Defence submitted that the Judge erred in finding that the complainant’s testimony met the “unusually convincing” standard and that Mr Panfilov’s testimony did not raise a reasonable doubt in the Prosecution’s case. As such, the Judge’s conviction on all four charges was unsafe. The Defence maintained that Mr Panfilov’s sexual encounter with the complainant on 12 January 2021 was consensual and that the complainant’s claim of a forced and violent sexual encounter was riddled with numerous material inconsistencies.

43 The Defence contended that the Judge erred in finding that the complainant’s evidence was unusually convincing because her account was marked by numerous material inconsistencies that she was unable to explain credibly or logically. The Defence advanced the following arguments in support of its position that the Judge erred in this regard:

(a) The inconsistency pertaining to whether the complainant screamed during the alleged assault went to the core of the complainant’s account of what happened in Mr Panfilov’s bedroom on the night of the Incident. The Judge’s finding that the complainant was probably confused and unsure (when she was being examined by Dr Geetha) went against the weight of the complainant’s evidence as, on her own case, she was not unsure of whether she had screamed.

(b) The undisputed fact, that Mr Panfilov went to the bathroom to retrieve a condom and took quite some time to put it on after the complainant fellated him and before he had penetrative vaginal sex with the complainant, exposed an unbridgeable gap in the Prosecution’s case which implied a near-seamless continuity and did not leave room for any gap in the sequence of events for Mr Panfilov to put on a condom. The Judge’s dismissal of the Defence’s submission on this point on the basis that the complainant had her eyes closed, was scared and in a lot of pain, and was focused on breathing, at most explained why the complainant may not have been able to know when exactly Mr Panfilov put on the condom but did not address the existence of an interval itself. Moreover, the Judge filled this material gap in the Prosecution’s case by relying on his own finding of “tonic immobility” which was unsupported by any expert evidence.

(c) As regards the blood-stained underwear, the Judge: failed to address whether this was in fact the underwear the complainant wore on the night of the Incident; improperly substituted his own lay assessment for the expert evidence on the source of the blood (as the Judge treated the general medical possibility of the source of the blood being an unobserved laceration as being more probable than the alternative possibility that the blood was caused by pre-menstrual spotting without any concrete reasons for doing so); and failed to consider pre-menstrual spotting as a possible explanation for the blood on the underwear. Therefore, the Judge erred in treating the blood-stained underwear as corroborative of the complainant’s account that the sexual encounter was non-consensual.

(d) While the Judge explained that it was not necessarily inconsistent for the complainant to have wanted nothing further to do with Mr Panfilov while at the same time wishing to obtain a memo from Dr Foo to show his family, this did not answer the question of whether the memo request, which Dr Foo testified was one of the complainant’s reasons for visiting the Clementi Clinic, was inconsistent with the complainant’s own account of why she attended the Clementi Clinic, which was that she attended because she remained in pain and wanted to get herself checked. This mattered because the Judge relied on this episode in rejecting the Defence’s challenge to the complainant’s credibility. Moreover, the Judge failed to address his mind to the complainant’s request that Dr Foo cancel the consultation and erase the evidence of that consultation from the clinic records (which Dr Foo recollected but the complainant could not remember).

(e) The Judge’s finding that the complainant’s testimony was internally consistent appeared to have been reached without any affirmative scrutiny of the evidence. The “unusually convincing” standard required more than the mere absence of identified inconsistencies and required the court to actively assess whether the testimony had the “ring of truth”.

(f) In relation to demeanour, the Judge erred in placing significant weight on the complainant’s physical gagging in court during her EIC as positive evidence of the truth of her account.

(g) While the Judge correctly acknowledged that victims of sexual offences cannot be straightjacketed in their expected reactions, he failed to give adequate weight to the cumulative significance of several aspects of the complainant’s post-incident conduct, each of which sat more comfortably with the sexual encounter having been consensual rather than non-consensual.

44 In the round, the Defence advanced the following case theory on what changed between the night of 12 January 2021 (ie, when, according to the Defence, Mr Panfilov and the complainant had consensual sex) and the morning of 16 January 2021 (ie, when the complainant visited the Clementi Clinic to seek a memo from Dr Foo):

(a) The most plausible explanation for what changed was that the complainant felt that she had been ghosted or ignored by Mr Panfilov.

(b) The complainant visited Dr Foo’s clinic on 16 January 2021 with the intention of obtaining a formal memorandum documenting injuries from a sexual assault so that she could confront Mr Panfilov’s family.

(c) When Dr Foo declined to examine the complainant or provide any documentation until a police report was made, the complainant was confronted with the practical problem that she had no evidence of the alleged assault. The complainant knew that there was no blood on the underwear that she wore on the night of 12 January 2021. The blood on the blood-stained underwear was most likely menstrual blood. This was why the complainant told SI Lim that she had no evidence despite the fact that the blood-stained underwear was available to her.

(d) The complainant then told Dr Foo to cancel her consultation and registration at the Clementi Clinic as she wanted to remove traces of her visit since the visit would reveal her motive of confronting Mr Panfilov’s family with allegations of rape.

(e) The complainant changed her account, that she had screamed during the alleged assault, to saying that she wanted to scream but did not manage to do so because, when preparing for trial, she belatedly realised that this account was untenable due to the presence of Mr Panfilov’s flatmates.

45 The Defence also argued that the Judge erred in finding that Mr Panfilov was not a credible witness, and therefore that his evidence did not raise a reasonable doubt, because the inconsistencies in Mr Panfilov’s testimony were wholly tangential to the issues in the case and did not go to the core elements of the offences that he was being charged with.

46 Finally, the Defence submitted, in the event that its appeal against Mr Panfilov’s conviction failed, that the sentence imposed on him was manifestly excessive.

47 The Prosecution submitted that the Defence’s appeal against conviction should be dismissed. The Prosecution contended that the conviction was well-founded on the complainant’s testimony, which was internally and externally consistent. In stark contrast, Mr Panfilov’s testimony was inconsistent and demonstrably self-serving, and he was an unreliable witness. The Judge thus rightly held that there was nothing in his account that had the effect of raising reasonable doubts in the complainant’s evidence. As regards the complainant’s testimony, demeanour and conduct, the Prosecution made the following submissions:

(a) The complainant’s account was internally consistent and the complete absence of any internal inconsistencies in her testimony was especially significant given the Defence’s “prolonged and ferocious” cross-examination of her.

(b) The complainant’s account of the sexual assault was coherent and logical, and the Defence’s submission that the complainant’s account was incomplete as she failed to explain when Mr Panfilov put on the condom should be rejected. There was no evidence to support any finding that the interval during which Mr Panfilov put on the condom was a long one, and the Defence’s submission that the complainant ought to have attempted to escape or call out for help was completely inconsonant with established legal authority. Moreover, the Judge was well within his rights to make a casual reference to the established concept of “tonic immobility”.

(c) The complainant’s demeanour while testifying showed that she was a credible witness. The contextual evidence, in addition to the complainant’s direct testimony, clearly showed an irrefutable link between her gagging reaction on the stand and the traumatic incident that she was recounting at the time.

(d) The complainant’s account was externally consistent with her out-of-court statements to other witnesses such as Dr Foo and Dr Geetha:

(i) The Defence had overstated the import of a minor difference in recollection between the complainant and Dr Foo. The complainant was not defensive or evasive about whether or not she had asked Dr Foo for a memo. She simply stated her inability to recall this fact in a forthright manner. Moreover, the complainant’s request for a memo would not have been incongruent with her previous action of cutting off contact with Mr Panfilov as the focus of her attention and anxiety justifiably shifted from one area to another in the aftermath of the Incident. Additionally, the complainant’s and Dr Foo’s differing accounts of how the consultation ended (ie, whether the complainant had asked for records of her visit to be erased) were a minor and explicable inconsistency as it was plausible that a miscommunication occurred.

(ii) As regards the screaming inconsistency between the complainant’s testimony at trial and her account to Dr Geetha, the Judge rightly accorded minimal weight to this minor inconsistency which was eminently explicable. It would be unwise to place excessive weight on a single line in Dr Geetha’s report (ie, the line recording that the complainant screamed) given that its own author would have considered this to be ancillary information (because Dr Geetha’s focus would have been on what kind of penetration took place and any other physical assaults).

(e) The complainant’s account was externally consistent with the medical evidence and the blood-stained underwear. The Judge was right to prefer the opinion of Dr Geetha on the source of the blood over the opinion of Dr Lin, and rightly disregarded the remote possibility that the blood on the underwear was menstrual blood. Moreover, the Judge was right to disregard Mr Panfilov’s unsubstantiated assertions that the blood-stained underwear was not the underwear which the complainant wore on the night of the Incident as Mr Panfilov was a demonstrably unreliable witness when it came to describing the complainant’s attire.

(f) The Judge rightly accepted the complainant’s honest and convincing explanations for the actions she took in the aftermath of the Incident.

48 The Prosecution also submitted that the Judge rightly found that Mr Panfilov was not a reliable witness given the numerous inconsistencies that plagued his account of events. In this regard, the Prosecution highlighted that: (a) Mr Panfilov provided inconsistent descriptions of his occupation; (b) Mr Panfilov provided inconsistent descriptions of the complainant’s attire on 12 January 2021; (c) Mr Panfilov presented two different versions of the sequence of events leading up to the sexual acts with the complainant; (d) Mr Panfilov gave inconsistent descriptions of the events surrounding the Third Charge; (e) Mr Panfilov initially lied about whether he had sex with the complainant; and (f) there was no reason for the complainant to frame Mr Panfilov.

49 Finally, the Prosecution submitted that the imprisonment terms imposed on Mr Panfilov for the Second, Third and Fourth Charges were manifestly inadequate.

50 Broadly speaking, the issues that arose for this court’s consideration could be distilled into the following:

51 Mr Panfilov was convicted on: (a) one charge of outrage of modesty under s 354(1) of the PC (the First Charge); (b) one charge of sexual assault by penetration under s 376(2)(a) punishable under s 376(3) of the PC (the Second Charge); (c) one charge of rape (penile-oral) under s 375(1A)(a) punishable under s 375(2) of the PC (the Third Charge); and (d) one charge of rape (penile-vaginal) under s 375(1)(a) punishable under s 375(2) of the PC (the Fourth Charge). The elements of these offences, as at the time of the alleged offending, were as follows:

52 As mentioned above (at [25]), Mr Panfilov admitted to doing the physical acts underlying the First, Third and Fourth Charges but maintained that the complainant had consented to them. The conviction for these charges thus turned on whether the complainant consented to Mr Panfilov doing those acts. Mr Panfilov denied digitally penetrating the complainant’s vagina (ie, the physical act underlying the Second Charge).

53 Since the acts underlying the four charges were alleged to have taken place in close succession, we considered the evidence for these four charges together.

54 Before delving into the issues proper, we considered the following principles to be of relevance for the determination of this appeal. These principles relate to: (a) the presumption of innocence and the Prosecution’s burden of proof; (b) the unusually convincing standard; and (c) the threshold for appellate intervention.

55 First, it has been said that the presumption of innocence is a “central and fundamental moral assumption” in criminal law and it cannot be assumed that an individual is guilty by mere dint of the fact that he has been accused of an offence unless and until the Prosecution adduces sufficient evidence to displace this presumption of innocence (Jagatheesan s/o Krishnasamy v Public Prosecutor [2006] 4 SLR(R) 45 (“Jagatheesan”) at [59]). In discharging its burden of proof, it is incumbent upon the Prosecution to prove its case against the accused beyond reasonable doubt (Jagatheesan at [46]). This means that those doubts for which there is a reason that is, in turn, relatable to and supported by the evidence presented, must be excluded. Fundamentally, a trial judge must bear in mind that the starting point of the analysis is not neutral in that the accused is presumed innocent until proven otherwise (Jagatheesan at [61]). The trial judge should not supplement gaps in the Prosecution’s case and, if gaps in the evidence should prevail so that the trial judge feels it is necessary to fill them to satisfy himself that the Prosecution’s burden of proof has been met, then the accused simply cannot be found guilty (Jagatheesan at [59]).

56 Second, while a reasonable doubt might also arise by virtue of the lack of evidence submitted when such evidence is necessary to support the Prosecution’s theory of guilt (Jagatheesan at [61]), a complainant’s testimony can constitute proof beyond reasonable doubt even where no other evidence is available (AOF v Public Prosecutor [2012] 3 SLR 34 (“AOF”) at [111]). However, this can only be so when the complainant’s testimony is so “unusually convincing” as to overcome any doubts that might arise from the lack of corroboration. It bears noting that the phrase “unusually convincing” is not a term of art. It does not automatically entail a guilty verdict and cannot dispense with the need to consider the other evidence and the factual circumstances peculiar to each case, nor does it dispense with having to assess the complainant’s testimony against that of the accused where the case turns on one person’s word against the other’s (AOF at [114(d)]). The present case was precisely one such case. Moreover, it must be emphasised that the “unusually convincing” standard is meant to be a cautionary reminder to the court of the high threshold that the Prosecution must meet in order to secure a conviction and of the anxious scrutiny that is required because of the severe consequences that will follow from a conviction (Public Prosecutor v GCK [2020] 1 SLR 486 at [91]). These considerations, which are engaged in any criminal case, are particularly acute where the Prosecution’s entire case rests on the evidence of a sole witness whose evidence is uncorroborated and, to the contrary, is controverted by the accused. We will, in the course of these written grounds, expound on what the “unusually convincing” standard entails and how it ought to be applied to determine if the Prosecution has proven its case beyond reasonable doubt.

57 Third, while the abovementioned principles are intended to guide trial judges in their determination of the accused’s guilt, the role of an appellate court is not to re-assess the evidence in the same way a trial judge would (Haliffie bin Mamat v Public Prosecutor [2016] 5 SLR 636 at [31]). Rather, it is well-established that an appellate court has a limited role when it is asked to assess findings of fact made by the trial court in an appeal against conviction. As summarised by this court in ADF v Public Prosecutor [2010] 1 SLR 874 (at [16]):

58 With these principles in mind, we turn to explain our decision in relation to the substantive issues proper.

59 As this was a case in which the Prosecution sought a conviction on the basis of the complainant’s uncorroborated evidence, the Judge rightly applied the “usually convincing” standard to the complainant’s evidence (see [30] and [56] above). In this regard, a witness’s testimony may only be found to be “unusually convincing” by weighing the demeanour of the witness alongside both the internal and external consistencies found in that witness’s testimony (AOF at [115]). Demeanour, however, is not invariably determinative. Contrary evidence by other witnesses must be given due weight and, if the witness fails to recall or satisfactorily explain material facts and assertions, her credible demeanour cannot overcome such deficiencies (XP v PP [2008] 4 SLR(R) 686 (“XP”) at [71], cited in AOF at [115]). In relation to how an appellate court ought to review a trial judge’s performance of such an exercise, it should be noted that the trial judge’s advantage in having studied the witness is not critical as regards the inferences drawn from the internal consistency and external consistency of a witness’s testimony because “the appellate court has access to the same material and is accordingly in an equal position to assess the veracity of the witness’s evidence” (XP at [72], cited in AOF at [115]).

60 In the present case, the Judge arrived at the conclusion that the complainant was an “unusually convincing” witness after separately assessing the internal consistency and external consistency of her testimony as well as her demeanour. We deal with each of these aspects in turn.

61 As mentioned above (at [31]), the Judge found that the complainant’s testimony was unusually convincing as the complainant provided a “textured, vivid and consistent” testimony of the assault which remained unchanged while she was on the stand and the Defence did not point the court to any internal inconsistencies in the complainant’s testimony (GD at [36]–[37]). The Defence characterised this as a “by-default” finding of internal consistency which “appears to have been reached without any affirmative scrutiny of the evidence”, and submitted that this was insufficient to establish that the complainant’s testimony was internally consistent.

62 In our view, it did not lie in the mouth of the Defence to make this submission when it had not been able to point to any internal inconsistencies in the complainant’s testimony both in the court below and in the present appeal. All the Defence was able to take issue with was the purported lack of details which the complainant was able to provide in relation to the gap in time between the acts underlying the Third and Fourth Charges (during which Mr Panfilov allegedly went to put on a condom in the bathroom). This, however, did not rise to the level of being an inconsistency. Moreover, it was not accurate for the Defence to refer to the “internal consistency between the Complainant’s trial evidence and her contemporaneous account to Dr Geetha” (in relation to whether she had screamed during the alleged assault). As the Prosecution rightly pointed out, this related to the external consistency of the complainant’s testimony instead. Internal consistency relates to the content within a witness’s testimony (while on the stand) whereas external consistency relates to the consistency between a witness’s evidence and the extrinsic evidence (Public Prosecutor v Tan Chee Beng [2023] SGHC 93 at [23]). We thus dealt with this alleged inconsistency when assessing the external consistency of the complainant’s testimony.

63 While the Defence was right in saying that the “unusually convincing” standard required more than the mere absence of identified inconsistencies and required the court to actively assess whether the witness’s testimony has the “ring of truth”, this was different from saying that the assessment of internal consistency required more than the absence of identified inconsistencies. Since the Defence failed to identify any internal inconsistencies within the content of the complainant’s testimony, there was no reason to overturn the Judge’s finding that the complainant’s testimony was internally consistent.

64 As alluded to above (at [43]), the Defence raised the following inconsistencies in the complainant’s testimony as militating against a finding that the complainant’s evidence was “unusually convincing”:

(a) the inconsistency between the complainant’s testimony and her account to Dr Geetha as to whether she screamed during the alleged assault (“the screaming inconsistency”);

(b) the inconsistency between the “near-seamless continuity” in the complainant’s account of events and the undisputed fact that Mr Panfilov went to the bathroom to retrieve a condom and took some time to put it on, which the Defence referred to as an “unbridgeable gap” in the Prosecution’s case;

(c) the inconsistency between the complainant’s testimony on the blood-stained underwear and the medical evidence;

(d) the inconsistency pertaining to the complainant’s alleged purpose of visiting Dr Foo to obtain a memo to confront Mr Panfilov’s family (“the memo inconsistency”); and

(e) the inconsistencies between the complainant’s testimony and several aspects of her post-incident conduct.

65 Having reviewed the evidence, we were of the view that a number of external inconsistencies arose in the course of the complainant’s testimony. These inconsistencies, when viewed in isolation, might not have stood in the way of the complainant’s testimony meeting the “unusually convincing” standard. However, we came to the view that, when taken collectively, they gave rise to a reasonable doubt in the Prosecution’s case. In addition, a consideration of the physical evidence relating to this case, against the findings reached by the learned Judge in coming to the conclusions he did, raised a number of unresolved questions.

66 During her EIC, the complainant testified that, when Mr Panfilov digitally penetrated her vagina, she was “just frozen” and did not dare to scream or do anything because she was afraid that he would escalate the violence if she did. She also said that she did not dare to scream or shout or move her body, and no longer tried to break free, because of the pain she was experiencing. The complainant also explained that, although she felt so much pain that she wanted to scream when Mr Panfilov penetrated her vagina with his penis from the back, she could not scream because he had placed a pillow over her head:

67 The complainant also maintained that she did not scream even after Mr Panfilov had flipped her over to penetrate her vagina with his penis from the front, as she was in total shock and a lot of pain and she was scared of what Mr Panfilov would do to her if she screamed:

68 The complainant’s abovementioned testimony that she did not scream, however, was contradicted by the evidence of Dr Geetha. In the medical report which Dr Geetha prepared in relation to her examination of the complainant on 18 January 2021 at KKH, she recorded that, after Mr Panfilov pushed the complainant onto the bed from behind and penetrated her (vagina), “[the complainant] screamed and [Mr Panfilov] allegedly tried to choke her with his hands and pressed a pillow over her head to silence her” [emphasis added].

69 When the complainant was confronted with this apparent discrepancy between her testimony in court and what Dr Geetha recorded in her medical report during cross-examination, she attributed it to a misunderstanding or miscommunication:

70  While the Judge accepted that “the Complainant’s story before PW8 Dr Geetha was not entirely consistent with the evidence that she gave on the stand”, he found the complainant’s explanation for the discrepancy to be credible as the evidence surrounding the complainant’s state of mind at the time of her medical examination suggested that she was “confused, unsure, and still traumatised” (GD at [52] and [56]). Thus, it was “not surprising that there might have been some miscommunication on matters of detail” as the stress of the attack and having to recount it at the hospital “could have affected the Complainant’s ability to recall specific episodic information during subsequent recall attempts” (GD at [57]).

71 To begin with, we agreed with the Defence that the Judge might have conflated Dr Geetha potentially misunderstanding what the complainant had said to her and the complainant being unable to recall what had happened that day. The Judge’s acceptance of the complainant’s explanation for the inconsistency was premised on his conclusion that the complainant’s state of mind during the medical examination would have affected her ability to recall what had happened during the Incident accurately. However, it was not the complainant’s evidence that she had difficulty recalling whether she had screamed or not during the medical examination or that she might have recounted such details wrongly due to her state of mind. Rather, she was sure that she “always said [she] wanted to scream” but maintained that Dr Geetha might have misunderstood her (see [69] above). In other words, the complainant’s account was that her story before Dr Geetha was entirely consistent with her testimony in court in respect of her not screaming but that Dr Geetha had inadvertently recorded a different story in the medical report. Thus, while the complainant’s state of mind during the medical examination might have been a credible explanation for the narrower question of why her story before Dr Geetha was not consistent with her testimony in court, whether it was a credible explanation for the broader question of why the account recorded in the medical report itself was not consistent with the complainant’s testimony (which could either have been due to the complainant’s faulty recollection during the medical examination or Dr Geetha’s misunderstanding of the complainant) was a different question altogether.

72 In this regard, the Judge might have unduly narrowed the scope of his inquiry in his assessment of the complainant’s explanation as, on our reading of the GD, he appeared to have only considered the complainant’s explanation in relation to the narrower question (see [70] above). A consideration of the broader question, on the other hand, would have entailed testing the complainant’s explanation against Dr Geetha’s account in relation to whether she might have misunderstood the complainant so as to answer the anterior question of whether Dr Geetha’s medical report provided an accurate account of what the complainant had told her during the medical examination. Hence, the Defence’s submission that the Judge did not adequately engage with Dr Geetha’s evidence was not without basis.

73 When the complainant’s explanation that Dr Geetha might have misunderstood her was put to Dr Geetha on the stand, Dr Geetha was adamant that she did not misunderstand the complainant. In fact, Dr Geetha testified that she knew that there was a big difference between trying to scream and actually screaming, she recorded her notes from the complainant’s account contemporaneously, and she had given the complainant ample time to recollect what happened during the Incident:

74 Essentially, when examining the complainant, Dr Geetha first recorded the complainant’s account of the Incident on a “Medical Form for Complaint of Sexual Offences” (ie, the form which counsel referred to as the “pink form” in the extract quoted above) given by the police to the complainant in real time before using that form as the basis for preparing her medical report subsequently.

75 In our view, the complainant’s explanation for the inconsistency, namely that Dr Geetha might have misunderstood her, did not withstand scrutiny when tested against Dr Geetha’s evidence. The box describing the “Events leading to and during Assault” in the pink form contained Dr Geetha’s handwritten notes during the medical examination, including her notes that: “Pt [ie, patient] screamed. He [ie, Mr Panfilov] tried to choke her with his hands & pressed a pillow over her head to silence her”. There was therefore no doubt that the medical report accurately reflected Dr Geetha’s handwritten notes in the history-taking box in the pink form which, according to her, was “all what the patient narrates”. We should add as well that, even prior to Dr Geetha being confronted on the stand with the possibility that she might have misunderstood the complainant, Dr Geetha was asked if she recalled what the complainant had told her in relation to her recording of “Pt screamed” on the pink form and her response was that the complainant said “I screamed”. Given Dr Geetha’s account that she recorded these notes contemporaneously, as well as her avowed sensitivity to the difference between trying to scream and screaming due to her prior experience testifying in court, we considered it unlikely that Dr Geetha would have misunderstood the complainant if the complainant had indeed told Dr Geetha that she wanted to scream but did not manage to do so.

76 We were not persuaded by the Prosecution’s submission that “[i]t would be unwise to place excessive weight on a single line in Exhibit P5 [ie, Dr Geetha’s medical report], given that its own author would have considered this to be ancillary information”. The Prosecution submitted that Dr Geetha’s purpose of “taking a brief history of the assault” was to help tailor the medical examination that would come after and, consequently, her focus on obtaining this history from the complainant would have been on the kind of penetration that took place as well as other physical assaults. The Prosecution’s argument was not convincing as its attempt to downplay the attention which Dr Geetha paid to taking a “brief” history of the assault did not square with Dr Geetha’s own account that the majority of her one-and-a-half hour long consultation with the complainant was spent on history-taking and that she had given the complainant longer than usual to recollect the details of the Incident (see [73] above).

77 Furthermore, any detail pertaining to the complainant screaming during the Incident would have been tightly connected to her account of Mr Panfilov’s alleged physical assault on her since it was recorded on the pink form that he tried to silence her screaming by physically choking her. In fact, Dr Geetha testified that a bruise which she observed on the complainant’s neck during her physical examination of the complainant was significant to her because of the complainant’s account during the history-taking part of the examination that Mr Panfilov tried to choke her. This is why she specifically enquired if the complainant sustained this bruise from the assault, to which the complainant replied that she was not sure if that bruise was sustained from the assault or from her icing habits for fat loss. Dr Geetha even said that, for a “majority” of the history-taking portion of the medical examination, she and the complainant were “stuck on” the bruise on the complainant’s neck since the complainant was not sure about how it was sustained. Hence, we had difficulty accepting the Prosecution’s contention that Dr Geetha would have considered the complainant’s account of screaming, which would have been accompanied by her account of the accused choking her, to be merely “ancillary information”.

78 In the round, we accepted the accuracy of Dr Geetha’s contemporaneous record as we saw no reason to hold otherwise. Indeed, Mr Han Ming Kuang (“Mr Han”), who appeared on behalf of the Prosecution at the hearing of the present appeals, said that the Prosecution accepted Dr Geetha’s evidence. Thus, we proceeded on the basis that the complainant told Dr Geetha that she screamed and was silenced by Mr Panfilov. This meant that an inconsistency arose between the complainant’s account to Dr Geetha and her testimony at trial that she wanted to scream but did not end up screaming. The question then was whether there was a credible explanation for this inconsistency.

79 On the evidence before us, we were not satisfied that there was a credible explanation for this inconsistency. Once it was accepted that Dr Geetha’s contemporaneous record was accurate, all the Prosecution could suggest was that the complainant had most likely misspoken to Dr Geetha. However, as we pointed out to the Prosecution at the hearing, this was not the complainant’s evidence. As recounted above (at [69]), the complainant’s evidence was that it was Dr Geetha who misunderstood her. It was not the complainant’s evidence that she intended to tell Dr Geetha she did not scream but mistakenly said that she screamed instead. When this was pointed out to the Prosecution, Mr Han then submitted that, when the complainant testified in court, she would have thought that she told Dr Geetha she did not scream. As Mr Han explained:

80 Mr Han was essentially suggesting that the complainant might have testified at trial that she told Dr Geetha she did not scream and Dr Geetha misunderstood her, as opposed to simply saying that she misspoke during the medical examination, because she genuinely held the mistaken belief that she told Dr Geetha she did not scream even though this might not have been the case. Quite apart from the fact that this was a speculative submission, this was only one plausible possibility as to why the complainant’s account at trial was different from what she told Dr Geetha. The other possibility was the Defence’s suggestion that the complainant abandoned her original account that she screamed because of a belated realisation that this account could not be reconciled with the evidence of others in the Flat on the night of the Incident, which is elaborated on hereafter at [82]. For the reasons given below, this was certainly a plausible reason for the discrepancy (and we put it no higher than that) which could not be discounted. This was also why we did not agree with the Judge’s conclusion that the screaming inconsistency was not material.

81 In our view, the screaming inconsistency was a material one for a few reasons. While the difference between “screaming” and “wanting to scream” might not have been significant per se, the two accounts provided by the complainant conveyed different impressions of her state of mind during the alleged assault. The complainant’s in-court testimony implied that she made a deliberate choice not to scream even after Mr Panfilov stopped covering her head with a pillow because, amongst other reasons, she was fearful of him, whereas the account conveyed to Dr Geetha implied that she actually screamed and only stopped screaming because Mr Panfilov forcefully prevented her from screaming further. Thus, we did not accept the Prosecution’s characterisation of this inconsistency as a “minor inconsistency”.

82 Furthermore, the question of whether the complainant had actually screamed or not during the Incident assumed heightened significance in the light of the evidence of Mr Panfilov’s flatmates, Ms Go Jinju (“Ms Go”) and Mr Justin Wayne Bratton (“Mr Bratton”). It was an uncontroverted fact that there was at least one other person present in the Flat on the night of the Incident. In his conditioned statement, Mr Bratton stated that he was in his bedroom at the time of the Incident and went to sleep at around 11pm that night. For reference, the Incident happened sometime between 10pm and 11.55pm (as the charges recorded that the Incident happened sometime after 10pm and the Gojek ledger from Mr Panfilov showed that a Gojek booking with the complainant’s home as the drop off location was made at 11.55pm). Mr Bratton stated in no unclear terms that he “did not hear any commotion or dispute in the apartment on the night of 12 January 2021”. As for Ms Go, she stated in her conditioned statement that she could not remember if she was home on the night of the Incident but noted that, on most days, she would go home immediately after work ended at 7pm. Nonetheless, Ms Go likewise could not remember hearing any commotion coming from Mr Panfilov’s room on the night of 12 January 2021. If the complainant had indeed screamed so loudly on the night of the Incident that Mr Panfilov had to forcefully stop her from screaming further by choking her and pressing a pillow over her head to silence her, this would stand in stark contradistinction to the accounts of Mr Bratton and Ms Go that they could not recall hearing anything unusual coming from Mr Panfilov’s bedroom at the time of the Incident.

83 Crucially, the complainant knew that at least one of Mr Panfilov’s flatmates was in the Flat at the material time or, at minimum, that there was at least one other person in the Flat at the material time. This was apparent from the complainant’s EIC:

84 We noted that the complainant later said during cross-examination that she did not know that there was someone else (besides her and Mr Panfilov) in the Flat at the material time:

85 Despite this apparent contradiction, it was clear to us that the complainant knew there was at least one other person present in the Flat at the material time because she volunteered this information herself in her EIC. The Prosecution also acknowledged that the complainant was “aware of this from the beginning”, since it was recorded in Dr Geetha’s medical report that the complainant “noted there was another person in [the Flat] using the bathroom”. By the time the complainant took the stand, the police would presumably have interviewed Mr Panfilov’s flatmates. When viewed in the light of the circumstances we have highlighted above (in particular, the accounts of Mr Panfilov’s flatmates that they did not recall hearing anything unusual on the night of the Incident), the complainant’s change in position, from telling Dr Geetha that she screamed to testifying in court that she wanted to scream but did not scream, undermined her evidence. It suggested that the complainant may have embellished the facts relating to her having screamed when she described the assault to Dr Geetha and gave force to the Defence’s theory as to why the complainant might have changed her account (see [80] above). At the very least, this theory was no less plausible than the Prosecution’s theory that the complainant might have misspoken to Dr Geetha. When faced with two different possibilities, neither of which was more plausible than the other, it was only right for this court to resolve such a conflict in favour of the accused. After all, even if the court finds the complainant’s story to be more probable than the accused person’s version in a criminal case, all the standard of proof beyond reasonable doubt requires is the establishment of a reasonable possibility that the accused person’s story could be true (Public Prosecutor v Mohammed Liton Mohammed Syeed Mallik [2008] 1 SLR(R) 601 at [35], citing Public Prosecutor v Mohammed Liton Mohammed Syeed Mallik [2007] SGHC 47 at [4]). The possibility that the complainant may have embellished the facts is certainly a reasonable one, and this was all that was needed for us to resolve the doubt raised by this possibility in favour of Mr Panfilov.

86 As such, we were of the view that the screaming inconsistency was a material inconsistency for which the complainant was not able to offer a credible explanation.

87 The complainant’s evidence was that, when she returned home after the Incident, she took off her underwear and noticed that there was blood on it. She testified that her vagina bled because of the sexual assault. This underwear was eventually admitted into evidence as Exhibit P7. The blood-stained underwear was thus the only physical evidence of the sexual encounter offered by the Prosecution. It was advanced to corroborate the complainant’s account that the sexual encounter was not consensual and that it was stained by blood from the complainant’s vagina during the alleged assault. The question which arose in the court below, and which was dealt with by the Judge, was whether the complainant’s attribution of the blood on the blood-stained underwear to the sexual assault was consistent with the expert evidence of Dr Geetha and Dr Lin. However, in our view, an anterior question which had to be answered first was whether the blood-stained underwear was even the underwear which the complainant wore on the night of the Incident.

88 On the complainant’s own account, the blood-stained underwear was dug out of her laundry basket six days after the Incident (see [20]–[24] above). According to her, the clothes which she wore on the night of the Incident, including the underwear which she wore, were at the bottom of her laundry basket and she had to “dig through and find [them]” before bringing them along with her to make a police report six days after the Incident. The complainant’s mother also testified that the complainant would only wash her clothes “2, 3 weeks later” and agreed with the suggestion that the complainant “accumulates clothes for 2 or 3 weeks”. In other words, it was entirely plausible that, apart from the clothes which the complainant wore on the night of the Incident, there might have been up to two to three weeks’ worth of clothes sitting in her laundry basket on 18 January 2021 when she dug through it. It was undisputed that, during the six-day period from 12 January 2021 (ie, the date of the Incident) to 18 January 2021, the complainant had her period on 14 January 2021. This was recorded by Dr Geetha in her medical report and the complainant agreed on the stand that she provided this information to Dr Geetha.

89 Moreover, there was no evidence that the blood on the blood-stained underwear was not menstrual blood. According to PW15 Ms Tan Ying Ying Jolena (“Ms Tan”) from the Health Sciences Authority (“HSA”), who was asked to examine the blood-stained underwear for the presence of blood, semen and DNA, HSA’s laboratory did not have any tests to differentiate menstrual blood and blood from a wound. Ms Tan acknowledged that there were tests available to test for menstrual blood but explained that they were “not routinely used in the DNA forensic community”. When we questioned the Prosecution during the hearing of the appeals on why a test to determine menstrual blood was not carried out, the Prosecution’s response was that this was because the HSA did not have such a test. The Prosecution also acknowledged that it did not explore other areas. The upshot of all of this was that there was a very real possibility that the blood from the blood-stained underwear might have been menstrual blood and, concomitantly, that the blood-stained underwear was not the underwear which the complainant wore on the night of the Incident.

90 This possibility was strengthened by the fact that the description which Mr Panfilov gave of the underwear which the complainant wore on the night of the Incident was materially different from what could be seen from the photograph of the blood-stained underwear admitted into evidence. During his EIC, Mr Panfilov described the underwear which the complainant was wearing at the time of the Incident as being lacy, pinkish red, with a narrower area surrounding the buttocks and, most importantly, without any white padding:

The white padding (or gusset) was an important feature of the blood-stained underwear because it was the main part of the underwear which was stained by blood (see GD at [70]). It could also be seen from the above extract that Mr Panfilov expressly stated that the blood-stained underwear as seen from the photographs was not the underwear which he saw the complainant wearing on the night of the Incident.

91 Crucially, it was not put to Mr Panfilov that his description of the underwear which the complainant wore on the night of the Incident was false. As the Prosecution acknowledged, the extent to which Mr Panfilov was cross-examined on his description of the underwear was as follows:

As can be seen from this extract, Mr Panfilov was mainly questioned on why he had omitted to mention certain details about the complainant’s underwear in the VRI and only mentioned them during the trial. In our view, this was not the “huge difference” which the Prosecution made it out to be. Mr Panfilov had simply offered more details in his EIC than during the VRI and, to our minds, had a reasonable explanation (ie, that he was never asked to describe the underwear in detail during the VRI) for so doing. It could also be seen that the Prosecution had only put to Mr Panfilov that the photograph of the blood-stained underwear showed a pink underwear. In other words, only Mr Panfilov’s description of the picture of the blood-stained underwear was challenged. He was not seriously challenged on his own description of the underwear he perceived the complainant to be wearing on the night of the Incident, much less his broader case that the blood-stained underwear was not the underwear which the complainant wore that night.

92 Taking this together with the uncertainty generated by the fact that the blood-stained underwear was dug out of the complainant’s laundry basket, which would in all likelihood have contained clothes worn by the complainant during a length of time which coincided with her period (see [88]–[89]), we had no hesitation finding that a reasonable doubt arose in relation to whether the blood-stained underwear was even the underwear which the complainant wore on the night of the Incident.

93 Even if we proceeded on the assumption that the blood-stained underwear was the underwear which the complainant wore on the night of the Incident, we did not think that there was enough evidence from which a finding could be made that the blood on the blood-stained underwear had arisen from an injury sustained during the alleged assault as neither Dr Geetha nor Dr Lin could definitively take a conclusive position on this.

94 As mentioned above (at [33]), both Dr Geetha and Dr Lin were in agreement that the 1cm x 1cm abrasion which Dr Geetha observed during her physical examination of the complainant could not have caused the amount of blood seen on the blood-stained underwear. Instead, the experts differed in their opinion on whether there could have been a larger tear or laceration inside the complainant’s vagina which had caused the amount of blood seen on the underwear and, at the same time, healed between the time of the Incident on 12 January 2021 and Dr Geetha’s medical examination on 18 January 2021. If the complainant’s account that the blood on the blood-stained underwear was sustained during the Incident was true, then any tear or laceration which had caused the bleeding during the Incident must have healed by the time of Dr Geetha’s medical examination because Dr Geetha did not observe any such injuries inside the complainant’s vagina:

95 As can be seen from the extract quoted above, Dr Geetha was of the view that it was possible for the bleeding to have been sustained from either (a) a bigger abrasion (than the 1cm x 1cm abrasion) or (b) an internal laceration inside the vagina. In relation to (a), Dr Geetha testified that she wondered if the 1cm x 1cm abrasion was “a bigger lesion that [was] … possibly sustained from the assault and then was smaller at that point” but noted that it was “hard for [her] to conclude” [emphasis added]. When asked about the significance of the time that had lapsed between the Incident and her medical examination in relation to the injuries sustained by the complainant, Dr Geetha explained that “any injuries sustained to that region [ie, the vaginal region] heals very fast, because they’re very vascular, and they heal very fast, so it’s hard to correlate”. Hence, Dr Geetha could only conclude that it was a possibility, and could not say with certainty, that the 1cm x 1cm abrasion itself was sustained from an aggressive sexual act during the Incident. As for (b), Dr Geetha’s evidence was that “small, tiny lacerations could give you---because they’re so vascular, could give you that stain, but then they close up and they heal off very fast” but, when asked if such a laceration could heal within six days, she said that she “would not be able to say it”. When subsequently asked by the court if it was “likely that a tear that’s large enough to produce that amount of blood could heal completely in 6 days”, Dr Geetha responded “Yah. I mean, le---vaginal mucosal, yes, yes, yah”.

96 On the other hand, Dr Lin opined in his report, which was based on his interpretation of Dr Geetha’s medical examination, that a possibility of the bleeding was a “tear or laceration inside the vagina” but that “this should have been noted on the speculum examination performed 6 days later”. As Dr Lin explained during his EIC, he believed that Dr Geetha would have been able to see a scar or at least some form of granulation tissue around the location of the laceration if the bleeding was indeed caused by a deeper laceration inside the complainant’s vagina. When asked about the length of time which it would take for such a laceration to heal, Dr Lin responded that it could take up to two weeks or more. He also elaborated upon what the observable remnants of such a laceration after one week would have been:

97 As regards the possibility raised by Dr Geetha that the 1cm x 1cm abrasion could have been larger and hence caused the amount of blood seen on the blood-stained underwear at the time of the Incident, Dr Lin acknowledged that the 1cm x 1cm abrasion could have been larger six days prior to Dr Geetha’s medical examination but was of the view that it “wouldn’t have been larger by a lot”. Dr Lin also testified that, in arriving at his opinion that the bleeding could not have been caused by the 1cm x 1cm abrasion, he had taken into account that the abrasion “could have been slightly bigger, because 6 days later, … it could have shrunk a bit”.

98 The Judge preferred Dr Geetha’s evidence over Dr Lin’s evidence in relation to the question of whether a laceration inside the complainant’s vagina which caused the amount of blood seen on the blood-stained underwear could have healed within six days as he was of the view that Dr Lin’s opinion was “based on wrong factual observations concerning the amount of blood found on the underwear and thus could not be relied upon by the court” (GD at [71]–[72]). These “wrong factual observations” concerned Dr Lin’s characterisation of the proportion of the underwear which was stained by blood as well as of the underwear being soaked through by the blood (GD at [71]). The Judge observed that: (a) from his own visual examination of the photograph of the underwear (ie, Exhibit P1-5), the blood stain covered no more than 20% of the gusset, which was a far cry from Dr Lin’s estimation of “three-quarters”; and (b) although some blood stain was visible on the other side of the underwear, it would be an exaggeration to use the term “soak through”. In the Judge’s view, Dr Lin “had a tendency to exaggerate his factual description of the blood stain on the underwear in order to support an opinion which could not otherwise be supported by an objective and accurate factual observation of the physical evidence”.

99 In our view, the Judge had erred in characterising Dr Lin’s opinion as being based on “wrong factual observations”. Regardless of the words which Dr Lin chose to characterise the amount of staining on the underwear, the fact was that his opinions were based on a visual examination of images of the underwear. In particular, it should be noted that Dr Lin did not characterise the underwear as being three-quarters soaked in his report. Instead, when he first characterised the gusset of the underwear as being three-quarters soaked during his EIC, he was recollecting from memory what the picture of the underwear showed. When Dr Lin was actually shown a picture of the underwear during cross-examination, he conceded that he “might have given the proportions a bit mistakenly” and that “it may not have been three-quarters” but maintained that “it was significant bleeding because it did soak through the---to the other side”. Hence, it was clear that, in characterising the underwear as being three-quarters soaked during his EIC, Dr Lin simply intended to convey that he recollected there being significant bleeding on the underwear. This did not mean that, at the time he prepared his report, Dr Lin based his opinions on a mistaken observation that the underwear was three-quarters soaked. As such, the Judge ought not to have rejected Dr Lin’s expert evidence in its entirety on the basis that it was grounded on the wrong factual observations.

100 This left the issue of whether the blood could have been caused by a deeper laceration inside the complainant’s vagina which healed completely by the time of Dr Geetha’s medical examination unresolved. It bore noting that, even on Dr Geetha’s evidence, she could only, at most, allude to the possibility that such a laceration could have healed within six days.

101 Furthermore, Dr Lin also raised the possibility that the bleeding could have been caused by “early spotting prior to menstrual flow” since the complainant’s menses came two days after the Incident on 14 January 2021. This possibility was not considered by the Judge. While the Prosecution contended that Dr Geetha “ruled out this possibility”, we did not find this to be the case. In her medical report, Dr Geetha wrote that “[the complainant] reported that she had per vaginal bleeding [ie, bleeding coming from the vagina] post incident and that this was not her period, as it just stained her underwear”. When she was on the stand, Dr Geetha explained that she recorded that it was not the complainant’s period as this was what the complainant had told her after she asked the complainant if it was her period and the statement on the blood just staining the underwear (ie, a one-off stain as opposed to a continuous flow of blood) was the complainant’s reason for thinking that the blood stain was not caused by her period. In other words, this was simply Dr Geetha recording what the complainant had told her as part of the history-taking exercise during the medical examination. Hence, we did not think that Dr Geetha could be said to have conclusively ruled out the possibility that the blood on the blood-stained underwear was caused by pre-menstrual spotting.

102 Moreover, Dr Geetha’s evidence, that “patients who start period will start off with light spotting and then progresses to a period naturally”, was not inconsistent with Dr Lin’s evidence that pre-menstrual spotting could be “on and off bleeding” (ie, it need not be a continuous flow of blood). We also did not think that the Prosecution was right in saying that Dr Lin resiled from his earlier position on the stand that, if the blood on the underwear was caused by pre-menstrual spotting on 12 January 2021, it would have continued up until the actual menstruation on 14 January 2021. Dr Lin was simply explaining that spotting was “on and off bleeding” which, to our minds, must be the case because one could not expect pre-menstrual spotting to encompass a continuous and uninterrupted flow of blood in the days leading up to the actual menstruation. In this vein, we found the Prosecution’s submission, that the possibility that the blood was caused by pre-menstrual spotting was ruled out by the expert evidence and was tenuous because there was a “distinct break” and “clear temporal divide” between 12 and 14 January 2021, to be contrived. Given that the Incident happened close to midnight on 12 January, we did not think the divide was so distinct as to rule out the possibility of pre-menstrual spotting.

103 Another important point was that it was Dr Lin’s evidence that there should have been blood on the condom worn by Mr Panfilov, or on the bed where the penile-vaginal penetration took place, if indeed the blood on the blood-stained underwear had been caused by an injury sustained during the alleged assault. This was not challenged by the Prosecution during cross-examination, even though the Prosecution, along with the Defence, was specifically invited by the Judge to ask further follow-up questions after Dr Lin elaborated on this point in response to questions asked by the Judge. It was not disputed that there was no evidence of any blood being found on the condom or the bed at the material time.

104 In the round, neither Dr Geetha’s nor Dr Lin’s expert evidence could corroborate the complainant’s account that the blood on her underwear was caused by the alleged assault during the Incident. If anything, they raised a reasonable doubt in the complainant’s account on the provenance of the blood on the blood-stained underwear. They also contributed to the uncertainty surrounding whether the blood-stained underwear was the underwear which the complainant wore on the night of the Incident in the first place.

105 These inconsistencies and unresolved questions pertaining to the blood-stained underwear were material because they struck at the very heart of the Prosecution’s case. The Prosecution’s case was that the blood-stained underwear was physical evidence of a violent assault by Mr Panfilov because the assault had caused the complainant to sustain an injury which, in turn, had caused the amount of blood seen on the blood-stained underwear. The blood-stained underwear was offered as physical corroboration of an unwanted and forced encounter by the complainant herself. However, this was not made out because we were none the wiser as to whether the blood-stained underwear was the underwear which the complainant wore on the night of the Incident and whether the blood on the blood-stained underwear had arisen from an injury sustained during the alleged assault.

106 According to Dr Foo, the complainant approached her at the Clementi Clinic and begun by saying that she wanted to inquire about a sexual assault. The complainant said that she was unsure of what to do and did not want to identify the attacker. She then asked Dr Foo for a memo stating her injuries and what the probable cause of those injuries might have been because she wanted to potentially approach Mr Panfilov’s family with the memo and tell them what happened. Dr Foo then advised the complainant that she could not provide such a memo as the complainant would need to “go through the legal process”. Dr Foo’s impression was that the complainant “seemed worried about formal reporting”. Nevertheless, the complainant made the eventual decision to visit Clementi NPC on the same day.

107 The complainant, on the other hand, omitted mentioning that she had asked Dr Foo for a memo during her testimony. When the complainant was asked during her EIC about what prompted her to visit the Clementi Clinic on 16 January 2021, she said that she was still feeling pain in her vaginal and abdominal area and “wanted to get it checked out and make sure if everything is okay”. When asked to recount what happened at the clinic, she mentioned that she told the doctor she had been raped and she remembered the doctor saying that the doctor could not examine her and could not help her unless she made a police report first. After the consultation, she was worried that she might get pregnant or that she might get an infection or sexually-transmitted disease and, since she surmised that any other doctor would say the same thing to her (ie, that she needed to make a police report first), she decided to “brave up some courage and … walk over to the police station at Clementi”.

108 The Judge characterised this as “merely an omission which should not be treated as an inconsistency” unless the court believed that the complainant was being evasive and trying to hide an unfavourable fact (GD at [44]). In our respectful view, this characterisation was not quite accurate. The complainant’s alleged request for a memo was not merely presented by Dr Foo as a discrete and/or minor request in the series of events which took place at the Clementi Clinic. It was reasonably clear from Dr Foo’s evidence that one of the reasons, if not the primary reason, for the complainant’s visit to the Clementi Clinic was to request for such a memo. In other words, Dr Foo’s account was that the complainant had made a specific request for a specific purpose. Moreover, on Dr Foo’s account, it was the complainant’s request for a memo stating the causes of her injuries which led to Dr Foo advising her that she had to make a police report first as such a memo could not be provided without the complainant going through the formal legal process. In other words, according to Dr Foo, it was because of the complainant’s request for a memo that she advised the complainant to lodge a police report. Hence, the request for a memo was pivotal in shaping what Dr Foo and the complainant discussed during the consultation. The complainant, however, characterised the purpose of her visit to the Clementi Clinic as being only to seek medical attention for her pain and because she needed to see a doctor to address her fears of becoming pregnant and/or contracting a sexually transmitted disease. Thus, the complainant’s omission to mention requesting for a memo from Dr Foo was not just a minor omission. It was central to a fundamentally different account of what transpired at the Clementi Clinic.

109 Moreover, when the claimant was cross-examined on her alleged request for a memo from Dr Foo, her position was that she could not remember making such a request:

110 We found it difficult to accept the complainant’s account that she was not able to recollect this fact which went towards the very purpose of her visit to the Clementi Clinic. It was also perplexing that the complainant could not remember asking Dr Foo for a memo when, on her own account, Dr Foo said that she could not help her until a police report was made first. Dr Foo’s refusal to help the complainant would have only made sense in the context of the complainant requesting for a memo documenting the causes of her injuries. If, as the complainant testified, her sole reason for visiting the Clementi Clinic was to seek medical attention for her pain and/or to assuage her fears of pregnancy and contracting a sexually transmitted disease, there was no reason why Dr Foo, or any other doctor, could not have offered that to her without a police report having to be lodged. In other words, it was unlikely that the failure to make a police report would have stood in the way of the complainant’s professed reasons for visiting a doctor. We therefore found it strange that the complainant could remember Dr Foo’s refusal to provide her with any help unless and until she made a police report but could not remember requesting for a memo from Dr Foo even when this was put to her during cross-examination.

111 Hence, while the Judge was right in saying that it was not necessarily inconsistent for the complainant to have wanted nothing to do with Mr Panfilov after the attack and yet want a memo to bring to his family (GD at [45]), we agreed with the Defence that the Judge did not address the question of whether the memo request was inconsistent with the complainant’s account of why she visited the Clementi Clinic. For this reason, it was necessary to reconsider the Judge’s conclusion that the complainant was not being evasive and trying to hide an unfavourable fact. The Judge did not regard the memo request to be a fact unfavourable to the complainant only because he did not find it to be inconsistent with the complainant wanting nothing to do with Mr Panfilov after the attack. For the reasons given above, however, the memo request was inconsistent with the complainant’s account of her visit to the Clementi Clinic in relation to her purpose for visiting the clinic. Thus, there was reason to characterise the memo request as a fact unfavourable to the complainant and, therefore, for the court to believe that the complainant was trying to hide an unfavourable fact in omitting to mention the memo request and maintaining during cross-examination that she could not remember it.

112 Apart from the memo request itself, there was another aspect of the complainant’s testimony which was inconsistent with Dr Foo’s account of the complainant’s visit. According to Dr Foo, after the consultation ended, the complainant initially requested if she could “cancel the consultation and registration at the clinic”. However, Dr Foo told the complainant that she would not be able to cancel the consultation in view of medical and legal concerns but that she would waive the consultation fee as a goodwill gesture since no physical examination was performed. This was yet another omission from the complainant’s account as the complainant only testified to having wanting to pay Dr Foo for the consultation but not being charged after the consultation ended. The Prosecution submitted that it was not hard to conceive of how a discussion about cancelling or waiving the consultation fee could be misunderstood as being about cancelling the consultation entirely, especially given the poor mental and physical condition of the complainant at the time. However, quite apart from the fact that it was not suggested to Dr Foo that she might have misunderstood the complainant and thus Dr Foo was not given the opportunity to comment on this possibility, as the Prosecution acknowledged, it was tolerably clear from Dr Foo’s testimony that the discussion on the cancellation of the consultation and on the cancellation of the fees proceeded in two different parts:

113 It was difficult to imagine how a discussion on the cancellation of the consultation fees could have been misunderstood as being about cancelling the records of the consultation entirely since, on both the complainant’s and Dr Foo’s accounts, it was the complainant who wanted to pay the consultation fees and Dr Foo who wanted to waive it, whereas, on Dr Foo’s account, it was the complainant who wanted to cancel the records of the consultation and Dr Foo who declined to do so. Moreover, it was Dr Foo’s evidence, when asked if the complainant expressed to her why she wanted the consultation cancelled, that “[the complainant] seemed fearful about formal reporting or that [Dr Foo] may lodge a police report on [the complainant’s] behalf”. Hence, the possibility of any misunderstanding having arisen was, in our view, low and, as a result, there was no credible explanation for this omission in the complainant’s testimony.

114 In this regard, we did not agree with the Prosecution that this was an “insignificant inconsistency” which the Judge was right to disregard. This inconsistency, taken together with the inconsistency pertaining to the memo request, lent some credence to the Defence’s case that the complainant wanted to obtain a memo to confront Mr Panfilov’s family and, when she failed to do so, wanted to erase evidence of the consultation from the clinic records. At the very least, they put a dent in the complainant’s overall credibility.

115 At this juncture, we should mention that SI Lim’s account was that, after he explained the investigative procedure which would follow the lodging of a police report to the complainant at Clementi NPC, the complainant informed him that she “wanted to consult her lawyer first before deciding whether to lodge a police report”. As SI Lim put it while he was on the stand, the complainant “still insist[ed] that she need[ed] to consult her lawyer before deciding whether to lodge a report or not” even after he encouraged her to lodge a report by telling her that the police would investigate and try to find the evidence. The complainant testified that she did not have a lawyer at the time but wanted to consult a lawyer because she thought that there was a chance that Mr Panfilov might sue her if the police investigations “turn[ed] [Mr Panfilov’s] life upside down”. We only observe that the complainant’s remark that she would consult a lawyer first would also cohere with any motive in obtaining a report to confront Mr Panfilov’s family (for whatever reason) and render it less likely that the complainant would completely forget asking for a memo.

116 This brings us to a related point pertaining to what the complainant told SI Lim at Clementi NPC.

117 We found it puzzling that the complainant would have initially hesitated to make a police report at Clementi NPC when she would have known about the existence of the blood-stained underwear which was, according to her, physical evidence of the alleged assault. To be clear, we were not taking issue with the complainant’s hesitance to make a police report per se. As Aidan Xu J noted in Public Prosecutor v Yue Roger Jr [2019] 3 SLR 749 (“Yue Roger”) (at [30]), there is no general rule requiring victims of sexual offences to report the offences immediately or in a timely fashion and, instead, the explanation for any such delay in reporting is to be considered and assessed by the court on a case-by-case basis. In the present case, the complainant had given a number of reasons for her delay in reporting, namely that: (a) she was worried that she had no evidence; (b) she was worried about causing stress to her mother; (c) she was worried about her students finding out (the complainant taught art lessons); (d) she was worried of being sued by Mr Panfilov; and (e) she felt traumatised by the questions posed to her by the police officers (at Clementi NPC).

118 We were of the view that there was an inconsistency between the complainant’s account to the police at Clementi NPC, that she felt she had no evidence of the alleged assault (ie, reason (a) above), and her account at trial that the blood-stained underwear had been sitting in her laundry basket all along. On the complainant’s own account, she had taken the time and trouble to read articles on sexual assault during the two days following the incident. This was because she “wanted to know what to do” and was concerned about there being no evidence since the alleged attack happened behind closed doors:

119 When the complainant visited Clementi NPC four days after the incident, on 16 January 2021, she would have known that the only piece of physical evidence she had of the alleged assault was the blood-stained underwear sitting in her laundry basket. It was therefore unusual that she chose to take the position that she had no evidence of the assault at Clementi NPC (see [23] above). The Prosecution submitted that the complainant was only reminded about the existence of the blood-stained underwear on 18 January 2021, when her mother told her to bring the clothes she was wearing on the night of the Incident to make a police report. However, this did not square with the complainant’s vivid testimony, which suggested that the blood-stained underwear had specifically caused her to break down in tears shortly after the alleged assault:

The extract quoted above gave rise to a reasonable inference that, when the complainant supposedly saw that her underwear was bloodied, it was something that had specifically hit her, and hit her hard, at that point in time immediately after the Incident. It was therefore somewhat difficult to understand how the complainant could have completely forgotten about the existence of the very thing that had caused her to break down just barely three to four days later.

120 Apart from raising doubts over the complainant’s credibility, this inconsistency also casted doubt on whether Exhibit P7 was the underwear which the complainant wore on the night of the incident, a point we have discussed above (at [88]–[92]).

121 In addition to the inconsistencies highlighted above, there were also other issues which troubled us.

122 It was not disputed that Mr Panfilov wore a condom while penetrating the complainant’s vagina with his penis as the complainant herself testified that she saw “white stuff” at the tip of the condom that Mr Panfilov wore after he withdrew his penis. This meant that Mr Panfilov must have worn the condom sometime before the commission of the acts underlying the Fourth Charge. It was also not disputed that there was an interval separating the commission of the acts underlying the Third and Fourth Charges (ie, the penile-oral rape and penile-vaginal rape charges respectively). Mr Panfilov’s rather textured account in his EIC was that, after the complainant fellated him, he retrieved a condom from his bathroom and struggled to put on the condom before he had consensual penile-vaginal sex with the complainant:

123 On the other hand, the complainant’s account in her EIC of what happened between the acts underlying the Third and Fourth Charges was that she was gagging and choking while Mr Panfilov forced her to fellate him and when he pushed her face down on the mattress after he stopped inserting his penis into her mouth. According to the complainant, at this point in time, she was still trying to fight back her urge to vomit and she did not know how long she was on the mattress for before Mr Panfilov penetrated her vagina from behind:

124 The complainant’s only mention of Mr Panfilov’s condom in her EIC was her account of how she saw that Mr Panfilov was wearing a condom after he finished penetrating her vagina with his penis and there was “some white stuff” at the tip of the condom. During cross-examination, the complainant agreed that Mr Panfilov did not have a condom on at or before the time he inserted his penis into her mouth and that he must have had the condom on sometime after the end of the penile-oral penetration and before he penetrated her vagina with his penis from the front. However, she disagreed with the suggestion that her description of the various sexual acts during the Incident left no room for Mr Panfilov to put on a condom. She also maintained that she did not know when exactly Mr Panfilov put the condom on, that she only remembered that she was gagging and choking while on the bed, and that she was not keeping track of the time during this interval.

125 Given the complainant’s account that she was scared and in a lot of pain, with her eyes closed, the Judge did not think that she would have registered or been able to know when exactly Mr Panfilov put on the condom (GD at [82]). Hence, the Judge did not think that the complainant’s lack of explanation for when exactly the condom was worn constituted an inconsistency that affected her credibility (GD at [83]).

126 The Defence submitted that the Judge’s reasoning explained, at most, why the complainant may not have been able to know when exactly Mr Panfilov had put on the condom but that it did not address the existence of an interval itself. In this regard, the Defence argued that the Prosecution’s narrative “implie[d] a near-seamless continuity and d[id] not leave room for any gap in the sequence of events for [Mr Panfilov] to put on a condom” [emphasis added]. In our view, this gap in the Prosecution’s case pertaining to the interval in which Mr Panfilov had put on a condom was something that could cause reasonable concern for the court, both from the perspective of the complainant and Mr Panfilov.

127 According to the complainant, she had tried to leave the room and/or break free from Mr Panfilov at least twice before she was pushed onto the bed (see [14] above). If there was indeed an interval during which the complainant remained on the bed and had no contact with Mr Panfilov, this would have afforded her the perfect opportunity to leave. It was the evidence of the complainant that she was, at this point in time, fully dressed (since, on her own account, Mr Panfilov had only pulled down her underwear and flipped her dress up to expose her vagina and buttocks just prior to penetrating her vagina with his penis from behind – see [17] above). It was therefore open to the complainant to stand up and walk out of the bedroom at this juncture.

128 In this vein, we were alive to the reality that victims of sexual assault can have a variety of reactions to the assault, and that some may quite understandably “freeze” in such a situation. Indeed, it has been emphasised before that victims of sexual crimes cannot be straightjacketed in the expectation that they must act or react in a certain manner (GBR v Public Prosecutor [2018] 3 SLR 1048 (“GBR”) at [20]). And as Sundaresh Menon CJ outlined in Thangarajan Elanchezhian v Public Prosecutor [2024] 6 SLR 507 at [72(b)] (citing GBR at [20]), albeit in the context of discussing examples of “harmful stereotypes” which ought not to be relied upon in cross-examination, it is now widely accepted that there is no one “typical” emotional or behavioural reaction that is expected to be exhibited by victims of sexual offences. As an example, rape victims “display a diverse range of reactions in the immediate aftermath of the offence: some display great distress while others are quiet and controlled”. Hence, it is often not helpful to draw any conclusion on the reasonableness of a complainant’s reaction to the alleged assault from such reaction alone. However, this does not mean that the reasonableness of a complainant’s reaction cannot be assessed at all when there are barometers against which such an assessment can take place. Such barometers may include other aspects of the complainant’s testimony. In this case, even on the complainant’s account, Mr Panfilov must have taken some time to retrieve and put on the condom. The complainant said she did not know when this took place, or how long this would have taken (see [123]–[124] above), but it was apparent from her own testimony that there was a gap in time during which she did not perceive Mr Panfilov to be doing anything to her. Seen in this light, the complainant’s account of remaining on the bed for a certain amount of time, when Mr Panfilov left the scene, was somewhat at odds with her account that she had previously attempted to put up a resistance and escape just a few moments earlier. We therefore disagreed with the Prosecution’s suggestion that “it [was] absurd to question why the [complainant] failed to seize a fleeting opportunity to save herself”.

129 This interval also raised doubts in the Prosecution’s case when viewed from the perspective of Mr Panfilov. It was questionable that Mr Panfilov, knowing that the complainant had attempted to leave the room and/or break free on more than one occasion during the alleged assault up to that point, would simply just proceed into the bathroom, return to the bed where the complainant was positioned on, and struggle to put on the condom, before continuing with his attack. Objectively speaking, this gap in time was more consistent with the sexual encounter being consensual rather than being the non-consensual, sustained and exceptionally violent attack recounted by the complainant.

130 The Defence contended that the Judge failed to give adequate weight to the cumulative significance of several aspects of the complainant’s post-incident conduct, each of which sat more comfortably with the sexual encounter having been consensual rather than non-consensual. The aspects of the complainant’s post-incident conduct which the Defence highlighted were as follows:

131 We noted that the Judge did not appear to have dealt with (b) in the GD. As for (a), (c) and (d), as mentioned above at [38], the Judge dealt with them in the following manner:

(a) The Judge did not think that giving Mr Panfilov her home address was inconsistent with the complainant’s allegation of sexual assault as there was some force in the complainant’s explanation that what happened to her was not normal (GD at [114] and [116]). The evidence also showed that the complainant was likely suffering from some form of “tonic immobility” as she testified that her body went limp during the attack and that she was gagging, scared and traumatised (GD at [116]). Thus, the complainant’s decision to give Mr Panfilov her home address so that she could get a ride home was explicable.

(b) The Judge found the complainant’s explanation for her behaviour, namely that she was too scared not to respond to Mr Panfilov’s message as he had her address, to be adequate (GD at [118]). Therefore, the Judge did not agree with the Defence that this was behaviour which undermined the complainant’s credibility.

(c) While the Judge observed that the complainant’s late filing of the police report called for some examination, he was ultimately of the view that the complainant had given reasonable and cogent explanations for her delay having regard to the circumstances surrounding the offence (ie, there being no witnesses), the family situation (ie, the complainant’s mother’s heart condition) and her own worries and concerns (see [23] above). The Judge thus did not regard the late reporting as having the effect of casting reasonable doubts on the Prosecution’s case.

132 As mentioned above (at [128]), we were alive to the caution in GBR (at [20]) that victims of sexual crimes cannot be straightjacketed in the expectation that they must act or react in a certain manner. However, this, in our view, did not necessarily mean that a complainant’s post-incident conduct has zero probative value. Bearing in mind the abovementioned proposition when assessing a complainant’s post-incident conduct simply means that it is not necessary for a complainant to be distraught (or, for that matter, to have had any particular reaction) to be believed. This proposition could co-exist comfortably with the notion that “[a] calm, undisturbed disposition may generally incline the court to conclude that no wrong was committed” (see Yue Roger at [34]). In other words, while the presence of distress is not a necessary condition for a complainant’s testimony to be believed, it does not follow that the absence of distress can never be a sufficient condition (or a relevant consideration) for a complainant’s testimony to be disbelieved. This is especially so when the complainant’s post-incident conduct is inconsistent with other aspects of her testimony.

133 In this regard, there were at least two aspects of the complainant’s post-incident conduct which raised questions as to the veracity of her account.

134 First, we did not think that the complainant’s decision to give Mr Panfilov her home address after the alleged assault was explicable. According to the complainant, after retrieving her bag from the table in Mr Panfilov’s room, she saw him sitting at the corner of the bed. He then said that she could go now. However, she was “scared to go past [Mr Panfilov]” and tried to book a ride home at the corner of the table where she retrieved her bag. She remembered that Mr Panfilov asked her for her address so that he could arrange a ride home for her and, thus, she gave him her address. When the complainant was asked what she was thinking about at the point when Mr Panfilov asked for her address, she said that “[she] just wanted to go home and [she] was terrified of [Mr Panfilov], [she] would have given him anything he wanted. As long as he lets [her] go”.

135 The Judge understood the complainant’s testimony to mean that she was still afraid of what Mr Panfilov could do to her since he had just raped her and was still physically positioned near the exit after the intercourse had occurred (GD at [115]). As the complainant was still afraid of Mr Panfilov and in shock, and based on how she had frozen during the rape itself, it was not surprising that she had waited for him to let her go rather than running away herself.

136 In our view, this rationalisation of the complainant’s act of giving Mr Panfilov her home address would only make sense if the complainant subjectively perceived that she was being held in Mr Panfilov’s bedroom against her will and/or otherwise prevented from leaving. However, it was the complainant’s own account that her decision to give Mr Panfilov her home address came after he told her to leave. In fact, it was precisely because Mr Panfilov told her to leave that she attempted to book a ride home. In other words, in the complainant’s own mind, she was free to “run away”. Hence, we did not think that the Judge’s rationalisation of the complainant’s act was justified.

137 We noted that the Judge also considered that, while the intuitive reaction was to run out of the bedroom once the attack had stopped instead of waiting for Mr Panfilov to book her a ride and allow her to leave, the evidence showed that the complainant was “likely suffering from some form of tonic immobility” as she testified that her body went limp during the attack and she was gagging, scared and traumatised (GD at [116]). We did not agree with the Defence’s submission that, in using the term “tonic immobility”, the Judge had provided a “provisional medical opinion” which was not founded on any medical evidence or assertion by either party. There was nothing to indicate that the Judge intended to use the term “tonic immobility” as a medical term of art. We therefore viewed the Judge as simply having used that term in a descriptive sense and the words which the Judge chose to use to describe the complainant’s physical state was founded (whether rightly or wrongly) on the complainant’s testimony that her body went limp during the purported attack. We thus did not find the Judge’s use of the term “tonic immobility”, in and of itself, to be objectionable. However, we agreed with the Defence that a finding that the complainant was likely suffering from some form of tonic immobility did not explain the complainant’s deliberate choice to give her address to Mr Panfilov. Hence, we did not think that the Judge was right to have relied on the complainant suffering from tonic immobility to arrive at the conclusion that her decision to give Mr Panfilov her home address was explicable.

138 On a related note, we also found it difficult to understand how Mr Panfilov could have quite casually let the complainant leave, and even go so far as to assist her in getting home, immediately after subjecting her to a violent and sustained sexual assault (if that was indeed the case).

139 Second, we did not think the fact that the complainant replied to the messages sent by Mr Panfilov after the alleged assault was explicable either. It was not disputed that the complainant responded to Mr Panfilov’s messages asking her if she had arrived home safely after the Incident with a message saying something along the lines of “yes”. The Judge accepted the complainant’s explanation that she was too scared not to respond to Mr Panfilov as he had her address and she was afraid of what he would do if she did not respond, and thus found that the complainant’s act of replying to the messages after the Incident was not behaviour which undermined her credibility (GD at [118]). While there was nothing inherently illogical about the complainant’s explanation in and of itself, we were of the view that this explanation did not stack up when viewed against the fact that: (a) the complainant voluntarily gave Mr Panfilov her home address; and (b) as explained above (at [136]–[137]), the complainant’s explanation for her decision to give Mr Panfilov her home address was hard to understand.

140 Lastly, we noted the video evidence showing the complainant walking out from the elevator of the condominium to the car which Mr Panfilov had booked for her. The Prosecution acknowledged that the CCTV footage, which was admitted into evidence as part of Exhibit D1, showed the complainant “walking normally” while leaving the condominium. The Defence suggested that nothing in the complainant’s appearance suggested the violence of the alleged sexual encounter she described happened to her. The Defence also contended that the complainant’s neat appearance suggested that she had tidied herself up before leaving the Flat, which was consistent with Mr Panfilov’s testimony that she had used the bathroom for a short while after the Incident and before leaving the Flat.

141 When we questioned counsel for the Defence, Mr Johannes Hadi (“Mr Hadi”), about the CCTV footage at the hearing of the appeals, Mr Hadi responded that the Defence was not placing any reliance on it for the hearing. Mr Hadi said that the Defence stood by its submission that the CCTV footage showed that the complainant was not visibly in distress, but appeared to be reluctant to push the point further because of a concern that there would be a pushback from the court on the basis that the Defence was trying to stereotype the reaction of a purported victim of sexual assault (in line with the caution in GBR as mentioned at [128] above). However, as we indicated at the hearing, the thrust of our concern over the CCTV footage was directed at the physical evidence as opposed to a victim’s emotional reaction to the trauma of a sexual assault. The caution in GBR that victims of sexual assault ought not to be straightjacketed in the expectation that they must act or react in a certain manner was directed at the emotional state of a victim as externalised in his or her reactions. This would not preclude a court from analysing whether the physical evidence, as manifested in a purported victim’s physical appearance and demeanour, was consistent with other aspects of his or her testimony.

142 For instance, in the present case, the complainant testified to being in a state of shock, fear and extreme pain, throughout the alleged assault. She described the amount of pain she was feeling during the assault as being ten on a scale of ten at one point in time. While that level of pain had died down after the Incident, she testified that she was still in pain when she left the Flat and when she was looking for the car which Mr Panfilov had booked for her at the ground floor of the condominium. All in all, the picture which the complainant painted of the Incident was that she was subjected to a violent and sustained sexual assault which involved multiple acts of forced penetration. Following from such an assault and the resultant pain, there would have been some force in the argument that one would have expected her physical appearance and movement to reflect that and, if they did not, for there to be a credible explanation as to why this was the case. Having reviewed the CCTV footage, we ultimately declined to place any weight on it as we did not think that the footage of the complainant, which lasted around 40 seconds, would give rise to evidence which would allow the court to make a helpful determination one way or the other on whether this added to or detracted from the complainant’s credibility. Nevertheless, we leave it open for a court to take into account physical evidence of this sort for the reasons given above in a future case where the evidence possesses a higher degree of clarity.

143 Finally, as detailed above (at [117]–[120]), we did not see any issue in the complainant’s late reporting of the alleged assault per se. Rather, we were of the view that the explanation given by the complainant for her initial reluctance to make a report at Clementi NPC, insofar as it was based on her position that she had no evidence of the assault, was inconsistent with the existence of the blood-stained underwear.

144 For the reasons given above, we were of the view that the Judge erred in finding that the complainant’s evidence was “unusually convincing”. It bears reiterating that the word “unusually” in the “unusually convincing” standard implies that it is not sufficient for the complainant’s testimony to be merely convincing and there must be something more in the testimony to bring it over the threshold (Yue Roger at [37]). If there are inconsistencies in a complainant’s testimony, the question of whether these inconsistencies give rise to a reasonable doubt would depend on whether they go beyond possible errors in recalling insignificant details and, taken collectively, undermine the core elements of the complainant’s evidence (JDA v Public Prosecutor [2025] SGHC 157 (“JDA”) at [110], referring to GII v Public Prosecutor [2025] 3 SLR 578 at [39] and [43]). In this regard, it is relevant to consider the complainant’s explanations for the inconsistencies (JDA at [111(d)]). In the present case, there were a number of significant external inconsistencies in the complainant’s testimony which related to the core elements of the complainant’s evidence. These inconsistencies went beyond possible errors in recalling insignificant details as they related to key aspects of the complainant’s account that the accused had perpetrated a violent assault on her. The complainant was not able to offer credible explanations for these inconsistencies. Additionally, there were other aspects of the complainant’s testimony which further dented her credibility.

145 For completeness, we did not consider it appropriate to review the Judge’s assessment of the complainant’s demeanour since much of this was partly based on his first-hand observations of the complainant on the stand. To the extent that the Judge had made certain observations on the manner in which the complainant responded to questions during cross-examination, which we could glean for ourselves in the relevant transcripts of the trial, and they impinged on this court’s assessment of Mr Panfilov’s guilt, we have dealt with these observations in other sections of these written grounds. In any case, as mentioned at [59] above, if a witness fails to recall or satisfactorily explain material facts and assertions, her credible demeanour cannot overcome such deficiencies.

146 All in all, we found that a reasonable doubt arose within the Prosecution’s case. As this court had explained in AOF (at [114(d)]), a finding that the complainant’s evidence is “unusually convincing” does not automatically entail a guilty verdict and such a finding does not dispense with the need to assess the complainant’s testimony against that of the accused where the case turns on one person’s word against another. However, given our finding that the complainant was not an “unusually convincing” witness, it was not necessary for us to substantively assess Mr Panfilov’s testimony and deal with the alleged inconsistencies therein to determine if a reasonable doubt arose on the totality of the evidence. The external evidence which had established the external inconsistencies in the complainant’s testimony was, in the main, evidence led by the Prosecution and separate from Mr Panfilov’s testimony. To the extent that we had considered certain aspects of Mr Panfilov’s testimony, such as his account of the underwear which the complainant wore on the night of the Incident, these were aspects which were not seriously contradicted.

147 In summary, due to the unresolved inconsistencies in the complainant’s evidence, which arose from a testing of the complainant’s evidence against the objective facts and the evidence of the other witnesses, there was reason to doubt the complainant’s account that she was subject to a non-consensual and violent sexual assault on the night of 12 January 2021. As such, we did not think that the complainant met the threshold of being an “unusually convincing” witness and a reasonable doubt arose within the Prosecution’s case. At the very least, given the state of the evidence, we did not think that either the complainant’s account of the sexual encounter being non-consensual or Mr Panfilov’s account of the sexual encounter being consensual could be said to be more believable than the other. In such a situation, the court could only be guided by the “cornerstone of our criminal jurisprudence” that, where the evidence may lead to two equally reasonable results, that suffices to establish reasonable doubt (Jagatheesan at [90]).

148 In the course of its submissions, the Prosecution repeatedly alluded to the trauma which the complainant purportedly endured as a result of the alleged assault to explain the inconsistencies in her actions in the days following the Incident. Philosophically, it is not inconceivable that a complainant could get it all wrong in the face of such trauma. We were acutely aware of the need to give adequate allowance to victims of sexual assault who might, quite understandably, experience a wide range of emotions and react in a multitude of different ways which those of us who are not in their shoes can only begin to comprehend. It is only right for a court to view the evidence given by complainants of sexual assault with a healthy dose of empathy and compassion. Nevertheless, it falls upon the court to deal with such evidence in a manner which is consistent with the evidential standard of proof beyond reasonable doubt as well as the “unusually convincing” standard when the Prosecution’s case is reliant on the complainant’s uncorroborated evidence. In undertaking such an exercise, the court has to strike an appropriate balance between affording the complainant adequate allowance for the trauma she may have suffered on the one hand and the fundamental principle that any reasonable doubt ought to be resolved in favour of the accused on the other. To borrow from Dedar Singh Gill J in JDA (at [77]), an allowance is not a blank cheque and the argument that allowance ought to be given to the complainant on account of the possible trauma she might have suffered can only be taken to a certain mileage. Such allowance cannot operate to explain away any and every inconsistency in the complainant’s testimony, especially when that inconsistency is so material that it casts a reasonable doubt on whether that testimony can prove what it is intended to prove.

149 Even if one accepted the Prosecution’s arguments that the complainant’s actions and interactions with various third party witnesses such as Dr Geetha and Dr Foo in the days following the Incident could be explained by her emotional state and the trauma she might have been experiencing in the aftermath of the alleged assault, one would have to assume that her trauma and resultant inability to recollect material facts had extended for at least six days after the Incident, which is a not inconsiderable period of time. And even if one assumed this to be the case, the court has to come to a landing on the evidence it is presented with. The court cannot lean in favour of the complainant just because there may be explanations for her inconsistencies. There would be a point at which the inconsistencies in the complainant’s evidence result in such evidence falling short of the “unusually convincing” standard, notwithstanding that explanations have been provided. In addition, any explanations given for the inconsistencies in the complainant’s evidence fall to be scrutinised and, if they do not stack up when viewed against the objective evidence on the record, then the court must take the view that the high threshold which has to be crossed before an accused person can be convicted is simply not met. It thus cannot be the case that the “unusually convincing” standard is met so long as possible explanations for the inconsistencies in the complainant’s evidence are provided, without any consideration of how plausible and credible those explanations are. In the present case, because the inconsistencies in the complainant’s evidence were material, and the explanations provided for those inconsistencies were either insufficient or lacking in credibility in the light of the evidence, the “unusually convincing” threshold, and thus the Prosecution’s burden of proving its case beyond a reasonable doubt, had not been met.

150 Accordingly, we allowed Mr Panfilov’s appeal against his conviction in CCA 12 and dismissed the Prosecution’s appeal against sentence in CCA 13.