This judgment text has undergone conversion so that it is mobile and web-friendly. This may have created formatting or alignment issues. Please refer to the PDF copy for a print-friendly version.
Sakthivel Punithavathi v Public Prosecutor
[2007] SGHC 54
| Case Number | : | |
| Decision Date | : | 18 April 2007 |
| Tribunal/Court | : | |
| Coram | : | V K Rajah JA |
| Counsel Name(s) | : | Subhas Anandan and Sunil Sudheesan (Harry Elias Partnership) for the appellant; Janet Wang (Deputy Public Prosecutor) for the respondent |
| Parties | : | Sakthivel Punithavathi — Public Prosecutor |
Evidence – Principles – Expert evidence – Whether conflicting expert evidence per se giving rise to reasonable doubt – Whether number of experts testifying on same point amounting to crucial indication of whether such testimony to be preferred over expert evidence to the contrary
Evidence – Principles – Expert evidence – Whether court should scrutinise credentials and relevant experience of expert in deciding weight to be given to expert's evidence
Evidence – Proof of evidence – Onus of proof – Standard of proof – Whether Prosecution proving case beyond reasonable doubt – Meaning of "beyond reasonable doubt"
|
18 April 2007 |
Judgment reserved. |
V K Rajah JA:
1 The appellant was tried and convicted in a district court on a charge of voluntarily causing grievous hurt to her maid pursuant to s 326 of the Penal Code (Cap 224, 1985 Rev Ed) (“the Penal Code”). On being sentenced to 15 months’ imprisonment, she appealed against her conviction and sentence. On 9 February 2007, after considering the evidence and submissions, I set aside her conviction. I now set out my grounds of decision.
Factual matrix
Undisputed facts
2 The appellant is a 31-year-old mother of two young boys, aged six and four. Born and educated in India, she arrived in Singapore in 1996 after marrying her Singaporean husband. She is now a Singapore citizen.
3 The complainant, Anbarasu Malarkodi (“Malarkodi”) is a 40-year-old Indian national. The appellant’s parents had recruited, trained and arranged for Malarkodi to work as a domestic maid for the appellant in Singapore. Malarkodi arrived in Singapore on 8 February 2004 and immediately commenced working for the appellant and her family at their residence, a flat in Jurong West (“the flat”).
4 On 10 March 2004, at or about 2.30pm, whilst in the kitchen, Malarkodi sustained serious injuries to the last three fingers of her right hand (“the incident”). The injuries consisted of:
(a) Two clean superficial incised wounds approximately 3mm each over the dorsum of the proximal phalanx of the ring finger and middle finger;
(b) Three deep multiple lacerations very close together over the dorsum of the little finger exposing the proximal phalanx;
(c) Three deep lacerations over the ring finger;
(d) Severed extensor tendons at two to three places on the ring and little fingers;
(e) Fracture of the ring finger at one place; and
(f) Fracture of the little finger with comminution at the site of the fracture (ie, the bone was broken into more than two pieces).
All these injuries were not only in very close proximity to each other but also virtually parallel to each other. Malarkodi sustained no other injuries.
5 At the material time, only Malarkodi, the appellant and the appellant’s two young sons (then aged three years and one year) were present in the flat.
Malarkodi’s testimony
6 Malarkodi testified that she was initially treated well by the appellant. However, soon after the appellant’s husband left on an overseas work trip, the appellant began criticising her work and repeatedly insisted on the reimbursement of 80,000 Indian Rupees as repatriation expenses. Malarkodi claims to have responded indignantly, stating that she was not in a position to fork out such a large amount of money. She countered with a request that the appellant permit her to work in another household, thereby allowing her a reasonable period of time to repay the money. Her request was turned down. Despite the palpable tension in the household, Malarkodi acknowledged that the appellant had never physically harmed or so much as laid a finger on her prior to the alleged incident.
7 Malarkodi stated that there was only one incident prior to 10 March 2004 when they had a heated quarrel. On that occasion, Malarkodi was about to have her dinner around midnight when the appellant’s younger child placed his hand in her food. When she lifted him with the intention of handing him over to the appellant, the child cried loudly. The appellant immediately scolded Malarkodi for betraying the trust that the appellant’s mother had reposed in her and berated her for her unsatisfactory work. According to Malarkodi, unlike all previous occasions when she had tolerated being told off, she feistily retorted that the appellant was denying her a chance to eat, and unreasonably scolding her. The appellant then reprimanded Malarkodi for her impertinence.
The incident
8 On 10 March 2004, at about 1.00pm, while the appellant was feeding her older son, Malarkodi brought the appellant’s younger son into the master bedroom. After the younger child fell asleep, Malarkodi proceeded to clean the laundry in the bathroom. She had, however, forgotten to turn on the fan in the bedroom. The heat in the bedroom caused the child to awaken suddenly and cry. She was instantly sternly reprimanded by the appellant for her carelessness. Malarkodi apologised and returned to the laundry tasks. Soon after, Malarkodi heard the appellant hitting the younger child who proceeded to cry even more loudly. Malarkodi claimed that at this juncture she burst into tears too, “thinking about my situation, getting scolded by her and suffering here.” [emphasis added]
9 Having finished with the laundry, Malarkodi proceeded to the kitchen to wash the dishes. The appellant in the meantime continued scolding her younger son. Suddenly, the appellant entered the kitchen behind Malarkodi. Malarkodi turned to face the appellant as she approached her; Malarkodi claims, however, that her hands were still resting on the edge of the sink when she noticed the appellant removing a chopper from the cupboard under the stove beside the sink, on her right. Malarkodi continued to stand still with her hands placed on the edge of the sink as she expected the appellant to resume berating her.
10 Instead, the appellant in a sudden unexpected movement raised the chopper and lowered it on Malarkodi’s right hand, resting on the side of the sink. Malarkodi states that she was not aware of what transpired thereafter as she fainted immediately. She recollects the appellant striking her once. She shouted “amma” (‘mother’ in Tamil) prior to losing consciousness. Throughout cross-examination, Malarkodi maintained that she was unaware whether the appellant had attacked her again as she was conscious only during the first chopper blow. She could not recall whether the appellant had grabbed her hand to cut her fingers. When asked by Defence counsel whether she had informed the doctors who attended to her injuries at the National University Hospital (“NUH”), that the appellant had cut her fingers repeatedly (see [42] infra), Malarkodi steadfastly maintained that she had not made such a statement.
Events after the “attack”
11 Malarkodi assumes she fell to the ground after fainting. Upon recovering, however, she found herself on the floor with her back against the stove cabinet; a female Chinese neighbour, Fong Meau Chen (“Madam Fong”), was applying ointment to her forehead. When Malarkodi noticed the appellant, she immediately clutched the appellant’s legs and pleaded for forgiveness; she claimed that at that juncture, she had forgotten about the incident and did not feel any pain in her fingers. When asked to explain her response, Malarkodi elaborated that she had no recollection of the incident and had begged for forgiveness because of the obvious inconvenience she had caused. Soon after, Malarkodi blacked out once again, regaining consciousness only when the ambulance officers arrived.
12 As Malarkodi was carried on a stretcher by the ambulance officers, the appellant entreated Malarkodi to claim that she had hurt herself. The appellant pleaded that the appellant’s husband would lose his job and the family their home if the truth surfaced.
13 In the hospital, a Tamil speaking male questioned her, enquiring how the injuries were sustained. Malarkodi initially replied that she had cut herself. She maintained her stance when questioned later by a Tamil speaking doctor. Malarkodi explained that she did this in an attempt to protect the appellant who had young children out of sympathy for all of them. However, when the doctor persisted and informed Malarkodi that he could help her only if she spoke the truth, she finally “revealed” that she had been hurt by the appellant. She explained that she had disclosed the truth because of the doctor’s ultimatum and because she needed “protection”. In the course of cross-examination, Malarkodi was questioned about a police report made on 10 March 2004, in which she asserted that she had accidentally cut her right fingers with a chopper while cutting vegetables. Malarkodi claimed that such an explanation was inspired purely by the appellant’s earlier plea to her.
14 On the following day, the appellant visited Malarkodi at the hospital and asked her what she had disclosed. Malarkodi informed her that she had divulged the truth - that the appellant had cut her fingers. The next time the appellant visited Malarkodi again just before her discharge, she once more implored that Malarkodi claim she had cut herself; Malarkodi however retorted that the truth could no longer be suppressed.
15 After the incident, Malarkodi has been unable to use her injured hand with her earlier dexterity. She is unable to lift heavy objects and has difficulty bending her right ring and little fingers. Malarkodi has since instructed a solicitor to claim damages from the appellant.
The appellant’s testimony
16 The appellant holds a Bachelor of Science (Mathematics) degree as well as a Post-graduate Diploma in Computer Applications. After settling down in Singapore, she worked initially for three months as a sales coordinator and then for two and a half years as a teacher in a childcare centre. She stopped working in 2004 to look after the children. She required a maid because her husband frequently travelled on work. It had become increasingly difficult for her to manage the household chores while simultaneously caring for her two young children, particularly because the younger child fell sick quite frequently. She did however intend to return to work eventually on a part-time basis once the domestic situation was satisfactorily resolved.
17 When Malarkodi arrived, the appellant addressed her as ‘mother’ and asked her to care for the two children as if they were her own grandchildren. Malarkodi was assigned to cook for the adults, clean the house, attend to the laundry and periodically take care of the children.
18 The appellant found Malarkodi to be somewhat ‘emotional’. Malarkodi also had difficulty coping with her work; the appellant claims, in addition, that Malarkodi neither knew how to feed the children nor how to operate electrical appliances. As a result, the appellant attests she had no alternative but to operate the electrical equipment herself while Malarkodi attended mainly to the cooking. She found Malarkodi to be a slow learner who often made errors. The appellant was prompted as a result to repeatedly remind her of her duties.
19 Malarkodi asked for a transfer about 10 days after her arrival. When she asked the appellant if she was satisfied with her work, the latter responded by saying she intended to retain Malarkodi. She stated that while Malarkodi could take her time to learn the work, the children’s welfare always took priority over the housework. The appellant claims that after this exchange there was no further confrontation between them. Their relationship was once again on an even keel.
20 Prior to 10 March 2004, the appellant had raised her voice at Malarkodi on only two occasions. On the first occasion, Malarkodi had left the gate open upon leaving the flat. The appellant’s younger child had as a result wandered out of the flat. The appellant reprimanded Malarkodi ‘in a loud voice’ warning her not to repeat the mistake. The appellant claims that she was not really angry with Malarkodi at that point.
21 On the second occasion, Malarkodi had fallen asleep while water was boiling in a pot. The appellant warned her not to repeat the error as it could lead to disastrous consequences. She was upset with Malarkodi over that incident. In addition, the appellant was forced to consistently remind Malarkodi to exercise patience and caution in supervising the children.
22 When asked for her overall assessment of Malarkodi, the appellant conceded, albeit in a rather guarded fashion, that she was dissatisfied with Malarkodi’s performance. She expressed her apprehension over Malarkodi’s obvious inability to deal with both the children and the housework simultaneously. In addition, the appellant’s elder child was distinctly uncomfortable with Malarkodi. The appellant maintains that Malarkodi was temperamentally ‘emotional’ and curt. Malarkodi would retort angrily when asked to refrain from certain acts. She was careless, and showed neither initiative nor diligence in fulfilling her duties. The appellant also found Malarkodi to be volatile and easily confused. The appellant testified that she would not have been comfortable in turning over the running of the household to Malarkodi as of the date of the incident.
23 Despite Malarkodi’s less than adequate performance, the appellant remained optimistic that things would improve given time. The appellant felt that Malarkodi would eventually be able to take care of the children when they grew accustomed to her. Malarkodi sometimes commiserated with the appellant when the latter was obviously distraught over her children’s illnesses.
The incident
24 On the morning of 10 March 2004, the appellant’s younger son accidentally broke a glass medicine bottle in the playroom. The appellant entreated Malarkodi to clean up the mess after removing her son from the premises. She also instructed Malarkodi to clean the children’s toys. Malarkodi completed the second task but forgot the first. Though the appellant reminded her, the mess of broken glass and spilled medicine remained even after the appellant returned home after picking up her elder son. It was only much later, after yet another reminder that Malarkodi eventually attended to the task, while the appellant was watching television with the children.
25 At about 1.15pm, the appellant fed her children in the hall while Malarkodi remained in the kitchen. The children finished their meals about an hour later. The appellant then had her lunch while instructing Malarkodi to have hers.
26 After her lunch the appellant returned to the master bedroom with the children while Malarkodi attended to the laundry. The cooking utensils and dishes used during lunch had been placed in the sink as it was Malarkodi’s custom to wash them later. Suddenly, as the appellant was attending to her elder son in the toilet, she heard a heavy thud. On hearing the sound again, she left the toilet to investigate. As she proceeded to the kitchen, she heard the sound of a metallic object falling into the sink. As she stood outside the kitchen, she observed Malarkodi standing in front of the sink, with her right lower arm along the edge of the sink and her left palm resting on the edge of the sink. Malarkodi’s body was partially turned to her left as she faced the appellant.
27 Malarkodi shrieked when she saw the appellant and then promptly crumpled to the floor with her right hand stretched outwards. The appellant tapped her shoulder repeatedly while calling her name but to no avail. The appellant panicked when she noticed blood on Malarkodi’s right hand. In distress, she ran out of the flat and repeatedly hammered on her neighbour’s flat door shouting ‘aunty’. Madam Fong hurriedly accompanied the appellant back to her flat. After assessing the seriousness of the situation, Madam Fong dialled the public emergency hotline number while the appellant continued with her attempts to revive Malarkodi.
28 Later, another female neighbour arrived to lend a hand. Both neighbours propped the maid up against the stove cabinet and applied medicated oil to her forehead. Malarkodi then opened her eyes.
29 When the ambulance officers arrived, Malarkodi begged the appellant to forgive her, claiming that she had acted rashly. The appellant then noticed the chopper in the left basin of the sink. She did not accompany Malarkodi to the hospital as she had to remain at home to attend to her children.
After the incident
30 After the ambulance officers left, the appellant contacted her sister-in-law seeking assistance. Later, while the children slept, she cleaned the kitchen. She washed the chopper in the sink and placed it in the stove cabinet. The appellant testified that the chopper in question was usually used for cutting meat and fish and normally placed in an overhead cabinet. She had instructed Malarkodi to keep all cutting implements in overhead cabinets as her elder child habitually opened the lower kitchen cabinet doors. The appellant insisted that she did not normally store the chopper in the stove cabinet and that she had never noticed Malarkodi do so. She was uncertain where the chopper had been stored immediately prior to the incident. When asked why she had placed the chopper in the stove cabinet after the incident, the appellant explained that as she had to hurriedly leave the flat with her sister-in-law and as the chopper was still wet, with no storage space in the dish rack, she left it in the stove cabinet which was lined with plastic sheets unlike the overhead cabinets. After the incident, she hastily left the flat with her children and temporarily stayed with her sister-in-law while she conferred with her husband about sorting out the domestic disarray.
31 The appellant visited Malarkodi at NUH on 11, 12 and 15 March 2004. During the first two visits, the appellant enquired about Malarkodi’s well being. She also brought fruits for Malarkodi on 12 March 2004. On 15 March 2004, as Malarkodi had a police officer in attendance, the appellant did not talk to her. During cross-examination, the appellant denied having either questioned Malarkodi about what had been disclosed to the police about her injuries or why the latter had not lied to protect her.
Medical evidence
32 The Prosecution adduced evidence from two medical experts: Dr Sandeep Jacob Sebastin (“Dr Sandeep”) and Dr David Su Hsien Ching (“Dr Su”). The Defence relied on one medical expert: Associate Professor Lim Beng Hai (“Prof Lim”). The general thrust of the Prosecution’s expert witness testimony is that the injuries suffered by Malarkodi are highly unlikely to be self-inflicted. On the other hand, Prof Lim is firmly of the view that the nature of the injuries strongly suggests that they were self-inflicted.
The Prosecution’s witnesses
Dr Sandeep
33 Dr Sandeep is a plastic surgeon, who had been attached to the Department of Hand and Reconstructive Microsurgery, NUH (“Hand Department”) for about two years when he testified. He secured his basic medical qualifications in India in 1994. He did not at any point treat or conduct a medical examination of Malarkodi. His two medical reports were premised entirely and exclusively on the available documentation, namely, the medical case notes, photographs and X-rays.
34 Dr Sandeep is persuaded that it is highly unlikely that the injuries suffered by Malarkodi were self-inflicted for the following reasons:
(a) Most self-inflicted injuries involve the non-dominant hand (Malarkodi’s dominant hand was her right hand) and the volar aspect;
(b) Most self-inflicted injuries are committed to elicit a reaction and do not usually involve the deeper body or bone structures. Malarkodi had fractures of two bones, which are not typical of a self-inflicted injury; and
(c) The direction of lacerations over the right little and ring fingers (the laceration on the little finger was more proximal to the level of the laceration on the ring finger) suggested that the injury was sustained by a knife coming from the right side of the patient.
35 In the course of his testimony, Dr Sandeep clarified that it was likely that more force was applied from the patient’s right to the little finger with less force applied or received by the ring finger. The injury to the little finger was greater and that injury was more proximal to the arm of the patient. Further, the obliquity of the fracture to the ring finger showed that the fracture originated further from the finger tip and was directed towards the finger tip. This was indicative of a sharp object approaching from the right side. If the cut was inflicted from the left side, the side-view obliquity would be in the opposite direction - originating from closer to the finger tip and in the direction away from the tip.
36 Dr Sandeep was of the view that both fingers were injured simultaneously and on multiple occasions. A single blow by the chopper would not cause multiple lacerations but would have cut the skin, tendon and bone at a single spot. Dr Sandeep opined that the injuries must have been caused by several blows, and concluded that there would have been at least four separate blows to the fingers.
37 He also asserted that if the injuries were self-inflicted with the left hand, it was more than likely that the injuries would be slanting the other way; the lacerations would slant from the ring finger closer to the arm and move outwards as it reached the little finger. If a person were minded to cut the fingers with the left hand with the cutting instrument slanted the other way to produce the actual lacerations, more severe injuries to the ring finger than the little finger could be expected and the obliquity of the fractures would be in the opposite direction from what had been sustained. During cross-examination, he conceded that though it was highly unlikely for a right handed person to use the left hand to injure the right hand, it was “not impossible”. While it would have been extremely difficult and the chopper would have been held awkwardly, this nevertheless constituted a possibility. Such a person would have to have been meticulous in planning the incident. It is significant that Dr Sandeep was unable to completely rule out the possibility that the injuries might have been self-inflicted.
38 Dr Sandeep testified that in all of the 10 to 15 cases he had attended to in relation to attempted suicide, he had never observed any incidents involving injury to the dominant hand. He also commented that while suicide cases usually involved hesitation cuts before the fatal cut was made, such a succession of cuts did not generally figure in what he considered non-suicidal cases such as the present. Dr Sandeep explained that in the latter cases, hesitation cuts were more than enough to get attention and there was no need for that ultimate cut which in this case nearly amputated the finger.
39 Further, according to Dr Sandeep, the degree of pain felt by a victim sustaining similar injuries would vary from one individual to another. Generally, there would be considerable pain when the injury was first sustained, followed by a dull, aching, constant pain that would be aggravated if the injured fingers were moved.
40 Dr Sandeep thought it possible that Malarkodi fainted immediately after the first blow. A vaso-vagal shock could have caused the blood vessels to suddenly dilate in response to the stress. As there could have been insufficient blood flowing to the brain, Malarkodi may have lost consciousness and collapsed to the ground. Drawing from personal experience, Dr Sandeep testified that it was possible that Malarkodi initially had no recollection of events that occurred just prior to the vaso-vagal attack.
Dr Su
41 Dr Su is also a medical officer attached to the Hand Department. Prior to testifying, he had five years of general medical experience and two years of training/practice in the Hand Department.
42 Dr Su’s report stated inter alia that Malarkodi claimed to have been tortured by her employers daily. On admission to NUH, Malarkodi had also complained that her employer attacked her with a knife at about 4.00pm, repeatedly cutting her fingers. When crossed-examined, Dr Su stated that this information must have been disclosed to either one of the attending doctors, Dr Shenthil Kumar or Dr Rowena Ng. The statement had been recorded in the case notes on 10 March 2004, between 9.45pm to 10.00pm. Dr Su could not be sure which of the two doctors had recorded this.
43 Dr Su agreed with his colleague, Dr Sandeep, that it is highly unlikely that the injuries suffered by Malarkodi were self-inflicted. He emphasised that as the injuries were on the patient’s dominant hand, it is unlikely that Malarkodi could have employed her non-dominant hand to injure herself. In addition, it would have been difficult to inflict such injuries as the direction of the injury was in line with the leading edge being from the ring finger middle phalanx towards the little finger proximal phalanx. Dr Su was of the view that the cut probably started from the ring finger and then extended to the little finger; otherwise the middle finger would probably have been cut as well. Such a direction would have been unlikely if the cuts were self-inflicted. When re-examined, Dr Su explained that as the middle finger was the largest and most prominent finger, it would have borne the brunt of the injuries if the knife had been held in Malarkodi’s left hand. In addition, he would have expected more severe injuries to the ring finger rather than to the little finger. He, however, acknowledged that he had neither seen nor examined the chopper.
44 When Dr Su’s attention was drawn to the superficial lacerations over the proximal phalanx of Malarkodi’s middle finger and over the base of her ring finger, Dr Su explained that those cuts were likely to have been caused by pulling the chopper from the middle finger to the ring finger base, thereby causing the injury on the ring finger base to be deeper. He opined that if the two injuries were caused by bringing down the chopper on the two fingers, it would have required only a very gentle blow. Such a blow was likely to be less forceful than the blows to the little finger.
The Defence’s medical expert
Prof Lim
45 Prof Lim qualified as a doctor in 1985 and became a consultant in Singapore General Hospital in 1995. In 2000, he became chief of hand surgery at the Hand Department. He was later appointed an Associate Professor in the Faculty of Medicine, National University of Singapore (“NUS”). Although Prof Lim left to start his own private practice in 2004, he continues to teach at NUS. He has on several previous occasions acted as an expert medical witness in criminal matters on behalf of the Prosecution. His area of expertise is micro-hand surgery.
46 Prof Lim’s opinion and report were both based upon the medical reports of Dr Sandeep and Dr Su, as well as the photographs and X-rays of the injuries. In stark contrast to the two doctors who testified for the Prosecution, he is of the view that the injuries had indeed been self-inflicted.
47 First, Prof Lim pointed out that there were three dents on the chopper. The dent at the tip appeared to be blunt and old. The two dents about 3cm away from the handle were still sharp and therefore relatively fresh. Based on the concave arch of the latter two dents, Prof Lim opined that each was consistent with hitting a cylindrical object of about 1cm or less in diameter, quite possibly an object such as the phalange of the hand. Prof Lim is of the view that if one were to match the two dents to the injuries found on Malarkodi’s ring and little fingers, it may be surmised that the chopper was placed with the handle distally and the tip proximally (with the handle to Malarkodi’s right). This is because if the chopper had injured the hand proximally, there would have been concomitant injuries to Malarkodi’s middle finger, since in order to sustain a fracture of the middle phalanx of the ring finger, the chopper would have to descend 0.75cm or more and the leading edge of the chopper would have cut the middle finger too. In view of the position and the direction of the chopper, Prof Lim stated that the injuries sustained would have to be committed by either a right handed person facing the victim, or by the victim herself using her left hand. Given the position of the kitchen sink in front of Malarkodi, a right handed person could not have faced Malarkodi and cut her hand. Pertinently, even Malarkodi did not assert that the appellant had faced her. The cuts were therefore more likely to be self-inflicted by the victim using her left hand than caused by an assailant.
48 Secondly, Prof Lim noted the presence of ‘classic hesitation cuts’ of varying degrees on Malarkodi’s fingers, starting with superficial cuts at the base of the ring finger and on the middle finger, followed by a further cut on the ring and little fingers, and a second cut on the little finger. The final deep cut was the one closest to the finger tips, and could be seen close to the joint of the ring finger and on the little finger. Prof Lim explained that hesitation cuts are typically prevalent when there are self-inflicted injuries. A person who intends to injure himself would frequently make an initial tentative attempt, assessing his own strength and the sharpness of the weapon. Some inhibition caused by the fear of pain would also prevail initially. As such, initial cuts are usually superficial. This is usually followed by a resolve to inflict the injury of the desired intensity and the subsequent cuts would accordingly become more severe. Taking this into account, the injuries on Malarkodi’s hand appeared to have been self-inflicted rather than effected by another person.
49 Thirdly, Prof Lim noted that there was no evidence of withdrawal wounds that would normally accompany an attack. ‘Withdrawal wounds’ are protective wounds sustained when the victim tries to retract her hand or fingers. A ‘shaving injury’ would occur where the skin is sliced. No such injuries figured in the photographs of Malarkodi’s injured hand. The multiple lacerations present also indicated that the hand had to be resting on the palm for a while to allow the multiple cuts to be effected. This was decidedly not a likely scenario if Malarkodi was being attacked as she would obviously not have willingly and passively succumbed to such a sudden and vicious external force. Given the injuries sustained and the implement used, Prof Lim opined that it was not possible for the assailant to have held Malarkodi’s hand and cut her. The assailant would have had to be very strong as the chopper would have been held in the right hand and the non-dominant hand of the assailant would have been used to hold down Malarkodi’s dominant hand while the injuries were inflicted. The average grip strength in women would usually be insufficient to inflict this type of injury in the manner postulated by Malarkodi. An intense struggle would almost inevitably have ensued. There was no evidence of a struggle or of withdrawal/protective wounds. Prof Lim therefore concluded that the injuries could not conceivably have been committed by another female of similar build. When cross-examined, Prof Lim emphasised that even if Malarkodi had fainted as a consequence of a vaso-vagal attack, she would have been immediately aroused and would inevitably have struggled if further injury was inflicted. Further, it was not possible to hold down Malarkodi’s wrist or maintain its position on the edge of the sink if she had fainted after the first blow. Malarkodi would have fallen to the ground before the second blow could be struck. Prof Lim, however, also accepted that it was possible that Malarkodi did not resist the second blow if she was in a state of shock after the first blow had been struck.
50 Prof Lim concluded that Malarkodi had premeditated and planned the incident to make it appear that the injuries were caused by a person emerging from her right. She was right handed and in normal circumstances, she would have used her right hand to injure her left. Instead, she used her left hand, and had cut herself at an angle that pointed to an assailant attacking her from the right. He stated unequivocally: “Because we see no injury on the middle finger, it is conclusive that the cleaver could not have come from the right”.
51 He further testified that an attacker would not have rested her elbow on a stable pivot point in order to chop at Malarkodi’s fingers. Instead, one would expect an attacker to swing the chopper up and down, from a distance of at least two feet away. From that distance, it was not possible to inflict cuts in such close proximity as reflected in the injuries sustained by Malarkodi. Prof Lim further opined that it was unlikely that the repeated cuts within close proximity on Malarkodi’s hand were caused by blows by the assailant wielding the chopper with great speed even if Malarkodi had lost her consciousness. Instead, it was more likely that the cuts were self-inflicted, consciously controlled and effected by Malarkodi’s left hand which had less strength. Further, the obliquity of the fractures on both fingers did not follow the same direction.
52 Assuming that the two fresh dents on the chopper had been caused concurrently, Prof Lim was of the view that the fractures to both Malarkodi’s ring and little fingers were caused by the same blow. Although strenuously challenged by the Prosecution, Prof Lim maintained his stance. The Prosecution then proposed that Prof Lim continue with his testimony after examining Malarkodi.
53 After examining Malarkodi, Prof Lim testified that he was able to draw a straight line from the most distal cut on the ring finger and the second most distal cut on the little finger when the two fingers were splayed apart in a V-shape. Prof Lim maintained that it was very unlikely that the blows were aimed from Malarkodi’s right. He did, however, concede that if the dents on the chopper did not match to the injuries, it was possible for an assailant to have attacked the victim from the right. Prof Lim also agreed that a single blow from the chopper, causing the fractures on Malarkodi’s ring and little fingers if they were splayed apart, would not have caused the two dents on the chopper which were very close to each other. Prof Lim then raised yet another possibility – that the most proximal dent closest to the handle caused the fracture of the little finger and the second dent away from the handle could have been a dent caused by the kitchen sink. In such a scenario, the fracture to Malarkodi’s ring finger was caused by the sharp tip of the chopper near the handle and the impact did not dent the chopper. He asserted that it was still likely that the same blow caused the fractures because the victim was likely to react immediately after one fracture thereby making it difficult to inflict further blows on the same area.
54 During re-examination, Prof Lim maintained that even in the absence of the two dents in the chopper, Malarkodi’s injuries could still be self-inflicted. He added that the positioning of the victim’s right hand on a particular surface would affect the nature and extent of the injuries, regardless of whether they were self-inflicted or not.
The trial court’s decision
55 In assessing the evidence, the trial judge was mindful that this matter consisted essentially of one person’s word against another’s. Under these circumstances, the trial judge correctly noted that Malarkodi’s testimony had to be so compelling that a conviction could be safely based solely on it. She acknowledged the danger of convicting the appellant on the basis of Malarkodi’s testimony alone.
56 In her grounds of decision in PP v Sakthivel Punithavathi
57 The trial judge also noted approvingly that Malarkodi testified in a ‘forthright manner, answering questions straight to the point’: GD at [88]. Notwithstanding various challenges raised by the Defence, the trial judge found Malarkodi’s evidence on the material issues to be compelling and largely unshaken. Inter alia, the trial judge was of the view that there was “a good likelihood” that Malarkodi was not fully conscious when the further blows were struck, particularly if they were inflicted in quick succession; and Malarkodi’s inability to clarify this was not as dubious or disturbing as the Defence had proposed (GD at [91] and [92]). She also stated that Malarkodi’s plea for forgiveness did not amount to an acknowledgment that she had been guilty of injuring herself. As Malarkodi had been repeatedly reprimanded by the appellant, it was not surprising that the first thing that should have occurred to her was that she was once again in serious trouble, prompting her to instinctively ask for forgiveness (GD at [96]).
58 The trial judge was entirely dismissive of the evidence of Station Inspector Charles Soon Fook Kong (“SI Soon”), the Prosecution’s third witness, which brought into sharp relief glaring inconsistencies between Malarkodi’s initial clarifications and her eventual testimony in court. SI Soon testified that Malarkodi had originally stated that the appellant had held her hand and cut her. However, in court, Malarkodi testified that she could not remember whether the appellant had held her hand while cutting her fingers. As Defence counsel had not asked that the statement recorded by SI Soon be produced, the trial judge decided that the context in which Malarkodi made that statement was unclear. Further, given that the Defence had not cross-examined Malarkodi about this statement, Malarkodi had not been offered an opportunity to explain the apparent contradiction. The trial judge refused on that basis to construe the statement to SI Soon as a material discrepancy affecting Malarkodi’s credibility: GD at [99] and [100].
59 In addressing Dr Su’s medical report containing Malarkodi’s claim that she had been “tortured by her employers (sic) daily” in addition to having her fingers “repeatedly” cut by the appellant, the trial judge again noted that the context in which those statements had been made was unclear. In addition, the Defence had not applied for the attending doctors to verify what she had told them. Thus, despite the fact that Malarkodi had made no reference in court to being ‘tortured’ and had point-blank denied during cross-examination to having told the doctors that she had been cut repeatedly, the trial judge was not persuaded that the contrary claims in the medical report had any bearing on Malarkodi’s credibility: GD at [103] and, more particularly, at [104]:
104 … [T]he context and circumstances under which Malarkodi purportedly made these statements were unclear. The defence also did not summon the doctors who first attended to Malarkodi to verify what she had told them. As such, I did not think that these statements in the medical report have any bearing on Malarkodi’s credibility or serve to cast doubt on the material parts of her testimony.
[emphasis added]
60 Having categorically rejected all the Defence arguments challenging Malarkodi’s credibility, the trial judge gave, in her own words, “full weight” to Malarkodi’s testimony, entirely disregarding the earlier inconsistent statements made to both the doctors and the police.
61 The judge sensed, on the other hand, manoeuvres on the appellant’s part to carefully steer clear of all facts suggesting any motive for resenting Malarkodi or fuelling anger against her. She determined that the appellant appeared vague, ambivalent and even contradictory about her relationship with Malarkodi. The appellant conceded that she was dissatisfied with Malarkodi’s performance, insisting nevertheless that she was confident that Malarkodi would improve with time. Further, the appellant attempted to portray herself as a patient and understanding employer as well as a capable mother-of-two, quite unfazed by her husband’s constant absence. The trial judge found it difficult to believe that the appellant was unruffled by the domestic strains engendered by two young children and a maid who seemed incapable of coping. After all, she extrapolated, the appellant seemed to be easily affected by stress and strain and would cry when her children were ill.
62 The trial judge was also of the view that the appellant’s testimony constantly shifted ground as the trial progressed; this was amply borne out by evidence relating to the storage of the chopper involved in the incident. Malarkodi had alleged that the appellant retrieved the chopper from the stove cupboard prior to the incident. The Defence maintained that the appellant had given specific instructions that the chopper be tucked away in an overhead cabinet and to that extent, should have been entirely unaware that it was actually stored in the stove cabinet. However, the appellant acknowledged that after the incident, she had personally replaced the chopper in the stove cabinet. The trial judge considered that the reasons given – that she was in a rush and that the dish rack was full and the chopper wet – were ‘clearly afterthoughts’; it was no coincidence that the appellant had replaced the chopper in precisely the same cabinet that Malarkodi claims the appellant had taken it from: GD at [113]. In the trial judge’s view, this further reinforced the strength and cogency of Malarkodi’s evidence.
63 With respect to the medical evidence, the trial judge found Dr Sandeep’s and Dr Su’s medical opinions to be inherently and logically consistent. Their testimonies were also assessed to have survived the test of vigorous cross-examination. The trial judge chose therefore to accept their opinions that the injuries were unlikely to be self-inflicted.
64 With respect to Prof Lim’s medical evidence, the trial judge found that Prof Lim had ventured beyond his area of expertise in his analysis of the two dents on the chopper, and that was the ‘weakest link’ in his report: GD at [116]. The trial judge also rejected Prof Lim’s view that the injuries found on Malarkodi’s hand were consistent with a case of self-infliction. She determined that if there were indeed classic hesitation cuts, “it would be surprising for both Dr Sandeep and Dr Su to have missed them”: GD at [119]. She chose instead to accept Dr Sandeep’s opinion that in non-suicide self-mutilation cases, there was no necessity to progress from hesitation cuts to a final cut. Finally, the trial judge surmised that even if she were to accept that there were no protective wounds on Malarkodi’s right hand, it was not an inevitable corollary that Malarkodi had lied about the incident. Malarkodi may simply have been much too shocked to put up any resistance during the sudden attack.
65 After evaluating all the evidence, the trial judge rejected the appellant’s version of facts and accepted Malarkodi’s testimony in its entirety, without any qualification whatsoever. She found that the appellant had indeed attacked Malarkodi with the chopper and hurt her.
The basis for appellate intervention
66 Right at the outset, the limited nature of review afforded to an appellate court must be emphasised. In Jagatheesan s/o Krishnasamy v PP
34 … In Yap Giau Beng Terence v PP
It is trite law that an appellate court should be slow to overturn the trial judge’s findings of fact, especially where they hinge on the trial judge’s assessment of credibility and veracity of witnesses, unless they can be shown to be plainly wrong or against the weight of the evidence. [emphasis added]
…
37 That said, it must be noted that the position apropos the proper inferences to be drawn from findings of fact is quite different. Yong Pung How CJ in Terence Yap observed in this context (at [24]):
[W]hen it comes to inferences of facts to be drawn from the actual findings which have been ascertained, a different approach will be taken. In such cases, it is again trite law that an appellate judge is as competent as any trial judge to draw any necessary inferences of fact from the circumstances of the case.
38 In short, intervention by an appellate court is justified when the inferences drawn by a trial district judge are not supported by the primary or objective evidence on record…
[emphasis added]
67 For a verdict to be assessed as going ‘against the weight of evidence’, the appellant must be able to show that any advantage enjoyed by the trial judge, afforded by having seen and heard the witnesses first hand, is not sufficient to explain and justify the trial judge’s conclusions on credibility: Benmax v Austin Motor Co Ltd [1955] AC 370 at p 375. Professor Tan Yock Lin (“Prof Tan Y L”) postulates two helpful guiding considerations in this respect (see Tan Yock Lin, Criminal Procedure (Butterworths, 2006) (“Criminal Procedure”) at XIX 153-211A). He states, first at [1004], that:
… to rely upon demeanour alone is to seize upon a most unsafe guide. The appellate court may be expected therefore to be vigilant against unwarranted reliance on demeanour alone… If there is also some other evidence upon the point in respect of which two key witnesses conflict, the trial judge must not purport to resolve the conflict by demeanour evidence alone. [emphasis added]
The second guiding consideration is stated at [1052] as follows:
Where the magistrate does not also give reasons for his belief that a witness is a witness of truth, the appellate court will be readier to disregard his findings. If he rejects the evidence of the accused, he must not simply state that he does that because he does not believe the accused. If he merely refers generally to demeanour of the witnesses, without condescending to particulars, he may invite suspicion that an attempt is being made to bolster up a verdict which is contrary not only to the weight of evidence but to the probabilities and which could not be supported on a detailed examination of the evidence.
Appellate intervention in general
68 Kirby J’s masterful and penetrating conspectus in State Rail Authority (NSW) v Earthline Constructions Pty Ltd (in liq) (1999) 73 ALJR 306 at 331-332, [93] of the broad heads and bases of appellate intervention is pertinent to both civil and criminal appeals. Appellate intervention is justified in the following situations (as summarised in Ian Freckelton and Hugh Selby, Expert Evidence – Law, Practice, Procedure and Advocacy (Lawbook Co, 3rd Ed, 2005) at p 336:
• cases where the primary judge’s conclusion, although expressed in terms of credibility was “plainly wrong” as demonstrated by incontrovertible facts or uncontested testimony;
• where the conclusion was based on evidence wrongly admitted, occasioning a substantial miscarriage of the trial;
• where reasons, going beyond credibility, indicated a consideration at trial of irrelevant matters or a failure to weigh all relevant issues;
• where the circumstances in which evidence was given, relevant to credibility, were unsatisfactory;
• where the primary judge had made it plain that credibility considerations or impressions were not determinative for the judgment in question;
• where the credibility determination leaves untouched other evidence which requires separate evaluation with no obstacle of a credibility finding; and
• where, notwithstanding the credibility finding, the “extreme and overwhelming pressure” of the rest of the evidence at the trail is such as to render the conclusion expressed at first instance so “glaringly improbable” or “contrary to compelling inferences” of the case that it justifies and authorises appellate disturbance of the conclusion reached at trial and the judgment giving it effect.
(see also Kirby J’s restatement of these principles in Whisprun Pty Ltd v Dixon (2003) 77 ALJR 1598 at [98] and [99]).
69 Our courts flinch from imposing absolute or mechanical rules. Suppleness in addressing anomalous cases is essential, given that unyielding rules may lead to injustice. The requirement that restraint be exercised at the appellate level in relation to the assessment of credibility must not be allowed to become a stumbling block for justice. In short, if the appellate court forms the view that the trial judge is relying on ‘demeanour’ as a crutch for a decision that cannot stand on terra firma, the court is entitled to carefully sift and sieve through the evidence, analysing and assessing it with reference to established objective facts, along with logic and ordinary commonsense.
Witness credibility
70 In Jagatheesan, I expressed the view that the same restraint governing appellate review in respect of findings of fact applies to a trial judge’s assessment of a witness’s credibility: at p 56, [40]. In point of fact, an appellate court should generally be even more restrained in such circumstances, as the trial judge has had the benefit of directly observing and assessing the witnesses in court. However, these statements should be construed in context: they were made against the backdrop of Yong CJ’s observations in Farida Begam d/o Mohd Artham v PP
A judge can make a finding on the credibility of a witness based on some or all of the following:
(1) His demeanour.
(2) The internal consistency (or lack thereof) in the content of his evidence.
(3) The external consistency (or lack thereof) between the content of his evidence and extrinsic evidence (for example, the evidence of other witnesses, documentary evidence or exhibits).
71 There is decidedly a yawning chasm between the assessment of a witness’s credibility based entirely on his demeanour, on the one hand, as opposed to his credibility based on inferences drawn from the internal consistency in evidence and/or the external consistency between his evidence and the extrinsic evidence, on the other hand. Where the veracity of a witness’s testimony was determined at first instance largely upon inferences made from the content of that witness’s evidence, the appellate court is in no worse position than the trial court to assess the same material: PP v Choo Thiam Hock & Ors
72 It should be noted that Prof Tan Y L’s two guiding considerations (discussed above at [67]) apply with equal force to an evaluation of a trial judge’s findings on witness credibility. The first consideration – that reliance on demeanour alone is generally unsafe – is also illustrated in the Court of Appeal’s decision in PP v Victor Rajoo
The learned trial judge’s acceptance of the accused’s evidence was based mainly on his impression of AB and the accused and the manner in which AB and the accused gave evidence. These factors are of course important and play a vital role in the determination of the veracity and credibility of their evidence. However, it is equally important to test their evidence against some objective facts and independent evidence. In PP v Yeo Choon Poh at p 878 Yong Pung How CJ delivering the judgment of this court said:
As was held by Spenser-Wilkinson J in Tara Singh & Ors v PP
It is also helpful to remind ourselves of what Ong Hock Thye FJ said in Ah Mee v PP, at p 223:
To avoid undue emphasis on demeanour, it may be well to remember what was said by Lord Wright, and often quoted, from his judgment in Powell & Anor v Streatham Manor Nursing Home [1935] AC 243, at p 267 of the possibility of judges being deceived by adroit or plausible knaves or by apparent innocence.
[emphases added]
The second consideration – that a trial judge should give detailed and reasoned grounds for his findings to avoid suspicions of an erroneous verdict – is aimed at ensuring that the trial judge demonstrates an appreciation of the inherent inconsistencies, contradictions and improbabilities in the case; any failure to do so can undermine the basis for any proper finding of credibility: Kuek Ah Lek v PP
73 It is critical to bear in mind that in reviewing a trial judge’s findings of fact and his assessment of witness credibility, the appropriate balance must always be struck between recognising the advantages of exclusive access to witnesses available only to the trial court on the one hand and the concomitant need for an appellate court to discharge its constitutional duty to ensure that a conviction is warranted and safe on the other: see Jagatheesan at p 57, [43]. It would be apt to conclude by reprising Atkin LJ’s famous and succinct observation in this exact context in Societe d’Avances Commerciales (Societe Anonyme Egyptienne) v Merchants’ Marine Insurance Co (“The Palitana”) (1924) 20 Lloyd’s L Rep 140 at 152:
[A]n ounce of intrinsic merit or demerit in the evidence, that is to say, the value of the comparison of evidence with known facts, is worth pounds of demeanour.
Expert Evidence
74 An appellate court will be slow to criticise without good reason a trial court’s findings on expert evidence; see the Privy Council case of Antonio Dias v Frederick Augustus AIR 1936 PC 154 at 155, 157 and 158, applied by the Court of Appeal in Muhammad Jeffry v PP
75 Where there is conflicting evidence between experts it will not be the sheer number of experts articulating a particular opinion or view that matters, but rather the consistency and logic of the preferred evidence that is paramount. Generally speaking, the court should also scrutinise the credentials and relevant experience of the experts in their professed and acknowledged areas of expertise. Not all experts are of equal authority and/or reliability. In so far as medical evidence is concerned, an expert with greater relevant clinical experience may often prove to be more credible and reliable on ‘hands-on’ issues although this is not an inevitable rule of thumb. Having said that, there is no precise pecking order or hierarchy relating to expert evidence. Experts may sometimes be abundantly eminent while lacking credibility in a particular matter.
76 What is axiomatic is that a judge is not entitled to substitute his own views for those of an uncontradicted expert’s: Saeng-Un Udom v PP
Burden of proof
77 Generally speaking, in criminal cases the legal or persuasive burden rests on and remains with the Prosecution throughout the proceedings. I find the following observations in a leading treatise by Tristram Hodgkinson and Mark James, Expert Evidence: Law and Practice (Sweet & Maxwell London (Litigation Library), 2nd Ed, 2007) at para 10-003 particularly pertinent:
There is no direct method of correlating the standard of proof in criminal and civil cases to the varying degrees of certainty which scientists and other experts ascribe to their findings. The criminal standard is susceptible of no precise mathematical description, such as a 99 per cent probability, and although the civil standard can be meaningfully described as 51 per cent probability this is often difficult to translate into terms equivalent to those of scientific results or estimates. In most cases, of course, the expert evidence, whether it be in the form of a generally expressed opinion or of a numerically represented set of scientific results, must be taken together with the factual evidence in the case, to which it is meaningless to ascribe precise numerical probabilities. In R. v Bracewell, Ormrod L.J. drew the distinction between scientific proof and legal proof. The fact that a proposition is not absolutely certain in scientific terms need not entail that there is a doubt in respect of the criminal standard of proof:
“the available date may be inadequate to prove scientifically that the alternative hypothesis is false, so the scientific witness will answer ‘No. I cannot exclude it,’ though the effect of his evidence as a whole can be expressed in terms such as ‘But for all practical purposes’ (including the jury’s) it is so unlikely that it can safely be ignored.”
In the great majority of such cases subject to the criminal standard of proof it will be the combination of the scientific with other factual evidence which permits a finding beyond reasonable doubt. Where there is only scientific evidence, represented in probabilistic terms, the court should not “transmute a mathematical probability into a forensic certainty.” Furthermore, it should not be regarded as inevitable that there is a doubt, upon the criminal standard of proof, merely, because two experts disagree.
[emphasis added]
If, in the final analysis, the court is unable to settle on a preferred view the matter must be determined on the basis of the burden of proof; has a reasonable doubt been raised? It would, however, be a legal heresy to suggest that a reasonable doubt is inexorably raised in all cases where experts differ. What nevertheless may plausibly be suggested is that genuine and irreconcilable differences between experts of comparable standing and credibility can create a reasonable doubt.
Proof of guilt beyond reasonable doubt
78 Whatever is thought about the myriad objectives of criminal punishment, one fundamental principle has been hailed as a cornerstone both at common law and in the Evidence Act (Cap 97, 1997 Rev Ed): before an accused person can be convicted of a crime, his guilt must be proved beyond a reasonable doubt. This bedrock principle is sacrosanct in our criminal justice system and constitutes a fundamental right that the courts in Singapore have consistently emphasised and upheld as a necessary prerequisite for any legitimate and sustainable conviction: see, for example, Jagatheesan ([66] supra); Teo Keng Pong v PP
79 A coherent and workable definition of ‘reasonable doubt’ is indispensable for the proper and consistent application of the criminal evidential burden of proof. In Jagatheesan, I endorsed the use of a working definition that described ‘reasonable doubt’ as ‘reasoned doubt’: at [55]. This definition mandates that all doubt, for which there is a reason related to and supported by the evidence presented, must be excluded. Reasonable doubt might also arise by virtue of the lack of evidence submitted, if such evidence is necessary to support the Prosecution’s theory of guilt: Jagatheesan at [61]. I am of the view that this particular formulation of reasonable doubt correctly shifts the focus from what could potentially be a purely subjective call on the part of the trial judge to a more objective one, requiring the trial judge to reason strictly in accordance with the evidence. The trial judge must be able to say precisely why and how the evidence supports the Prosecution’s theory of the accused’s guilt. This effectively inhibits and constrains the subjectivity of the trial judge’s fact-finding mission. I have adopted the same approach in the present case. However, I pause here to caution, that requiring a trial judge to furnish the reasons for his decision does not require or compel him to seek or extract those reasons purely or mainly from arguments or testimony from the Defence. The burden of proof invariably falls (subject to statutory adjustments) on the Prosecution, and the requirement of reasoned justice does not and cannot shift that burden. In this context, it will also be helpful to bear in mind that legal concepts of certainty and doubt should not be perceived or assessed as mathematical or scientific certainties; see [77] above.
80 The doctrine mandating that the Prosecution prove its case beyond reasonable doubt embodies two important societal values: first, it buttresses the presumption of innocence; secondly, it connotes and conveys the gravity that society equates with punishment. The link between the proof of an accused’s guilt beyond reasonable doubt and the presumption of innocence is that the former constitutes the threshold or the point at which society is prepared to contemplate a conviction thereby allowing for the latter to be displaced. The rationale for the second societal value, is aptly summed up by the following remark by Brennan J in the United States Supreme Court decision in In re Winship 397 US 358 (1970), at p 364:
It is critical that the moral force of the criminal law not be diluted by a standard of proof that leaves people in doubt whether innocent men are being condemned.
It would be a travesty of justice to visit the indignity and pain of punishment upon a person unless and until the Prosecution is able to dispel all reasonable doubts that the evidence (or lack thereof) may reveal.
81 As such, I think it is a matter of considerable significance, in a case such as this, to emphasise and ensure that the criterion of proof of guilt beyond reasonable doubt prohibits the trial judge from filling in the gaps in the Prosecution’s case on her own initiative and through conjecture or supposition; see also Jagatheesan at p 62, [59]-[60]:
59 … [T]he trial judge should not supplement gaps in the Prosecution’s case. If indeed 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 legally guilty. In short, the presumption of innocence has not been displaced.
60 … [I]t is critical that trial judges appreciate that inasmuch as fanciful conspiracy theories, often pleaded by the Defence, will not suffice to establish reasonable doubt, the Prosecution’s theory of guilt must be supportable by reference to the evidence alone and not mere conjecture that seeks to explain away gaps in the evidence. Suspicion and conjecture can never replace proof.
[emphasis added]
82 With these critical considerations in mind, I turn now to evaluate the findings of the trial judge.
Evaluation of trial judge’s findings
Should “full weight” have been accorded to Malarkodi’s evidence?
83 Contrary to the trial judge’s finding, it appears that Malarkodi’s version of events is prima facie incompatible with a number of critical objective facts as well as inherently inconsistent. It is common ground that multiple cuts were inflicted by the chopper. Dr Sandeep accepted that the injuries on Malarkodi’s hand resulted from at least four blows. Inexplicably, as Prof Lim rightly emphasised, no obvious defensive injuries were sustained by Malarkodi. I find it highly unlikely that Malarkodi could have or would have remained passive in the face of such a vicious assault, as she alleges. She would have seen the appellant raising the chopper and taking aim at her fingers. In such a scenario it is highly improbable that she would have continued to passively rest her hands on the edge of the kitchen sink; instead she would have instantly and instinctively attempted to move her hands out of danger’s way.
84 Malarkodi claims that she only remembers the appellant striking her once, whereupon she immediately fainted. This claim is in itself inherently improbable for two reasons: first, if Malarkodi had indeed lost consciousness as a result of a ‘vaso-vagal shock’, her hand must have slipped off the edge of the kitchen sink and it would have been virtually impossible for the appellant to have continued “raining” at least three other blows concentrated on a distinctly small and precise area on Malarkodi’s ring and little fingers; secondly, if Malarkodi fainted, she was likely to have regained consciousness if she was cut again, as astutely pointed out by Prof Lim. Such an assertion was neither disputed nor contradicted in any way by either the Prosecution or its experts. As such, it is exceedingly difficult to reconcile Malarkodi’s version of events with either logic on the one hand or the objective facts on the other. Instead of recognising such a glaring improbability in Malarkodi’s testimony, the trial judge was, on the contrary, persuaded by what she perceived as candour, that Malarkodi was only aware of being struck once by the appellant. In her GD (at [87]) the trial judge noted with obvious and extraordinary approval:
Overall, I must say that Malarkodi impressed me as a reliable witness who did not seek to embellish her testimony or deliberately slant her evidence against the Accused. This was evident when she candidly stated that she was only aware of being struck once by the Accussed when it must have been clear, even to her, that a singular blow could not have resulted in the numerous lacerations found on her right hand.
[emphasis added]
85 It is axiomatic, as the trial judge herself concedes, that a single blow could not have accounted for the numerous lacerations concentrated so remarkably on a miniscule area of Malarkodi’s right hand. How then could the trial judge infer from Malarkodi’s attestation to a single blow, “evidence” that she was a reliable witness? (see above at [56]). With respect, the learned trial judge erred in coming to this rather hasty and entirely unfounded conclusion. The truth of the matter appears to be that Malarkodi had no logical or rational explanation for the other wounds on her fingers – wounds that Prof Lim considered to be hesitant “cuts”. The judge failed to adequately address or evaluate whether Malarkodi’s account was in fact a thinly disguised attempt to feign ignorance over the “hesitation cuts”. Instead, the learned trial judge was content to conclude (GD at [92]):
… [T]o my mind, there was a good likelihood that Malarkodi was not fully conscious when the subsequent blows were struck, especially if they were inflicted in quick succession. [emphasis added]
It is not clear on what basis the trial judge settled on this startling conjecture. It must be reiterated that all the incisions were literally within millimetres of as well as virtually parallel to one another. It is inconceivable that “blows” could have been struck in “quick succession” after Malarkodi lost consciousness and had either fallen or was falling to the ground, and even more incredibly, that they could have been delivered with such prowess that they landed, virtually parallel, within millimetres of one another on the space of two fingers.
86 The trial judge also erred in finding that Malarkodi was ‘a reliable witness who did not seek to embellish her testimony or deliberately slant her evidence against the [appellant]’(see [56] above; GD at [87]). How could that be when she had glibly glossed over disturbing differences between various earlier statements and her court testimony? I refer in particular to Malarkodi’s initial statement to the doctors that she had been tortured daily and that she was cut repeatedly during the incident; see [42] above. Malarkodi testified in court, however, that the appellant had never so much as touched her prior to the incident. In addition, while maintaining she was only aware of being struck once by the appellant with the chopper; she adamantly denied, during cross-examination, that she had told the doctors she was cut repeatedly. These blatant inconsistencies were cursorily swept aside by the trial judge who was content to ignore and bury them under a fixation affirming Malarkodi’s unassailable ‘consistency” and favourable demeanour. She also formed the view that Malarkodi’s statement to the doctors had no bearing whatsoever on her credibility simply because the Defence failed on its part to summon as witnesses, the doctors to whom the statement had been made. Even after acknowledging that the Defence omitted to delve further into this issue, I find it rather incredible that the trial judge should confidently and unequivocally conclude that Malarkodi was not capable of embellishing her testimony. On the contrary, Malarkodi seemed perfectly capable of concocting patent falsehoods as amply evidenced by her assertion to the examining doctor that she had been tortured daily. More disturbingly, the trial judge had abruptly dismissed such an inconsistency by dubiously shifting the burden of proof onto the Defence whom she deemed most remiss for not summoning the relevant examining doctors as witnesses. With respect, she failed to appreciate that the damage wrought by these inconsistencies was inherent within the Prosecution’s very own evidence. It was clearly incumbent on the Prosecution, not the Defence, to resolve such glaring aberrations and inconsistencies by calling Dr Shenthil Kumar and Dr Rowena Ng, the attending physicians. Why did the Prosecution neglect to do so?
87 In seeking to explain the discrepancy in her differing accounts, (see [13]) Malarkodi testified that “the doctor also said that he could help me only if I spoke the truth”. I find any allegation that a doctor could have told her she would only be helped if she spoke the truth far-fetched, to say the least. Unfortunately, the judge once again omitted to scrutinise Malarkodi’s evidence for logic or consistency.
88 Another instance of a critical doubt being dubiously resolved in the Prosecution’s favour is the manner in which the trial judge dealt with the statement supplied by Malarkodi to SI Soon, asserting that the appellant held her hands while cutting her, see [58]. This assertion again runs counter to and is in fact entirely irreconcilable with Malarkodi’s testimony in court. Instead of viewing this as yet another blatant discrepancy which should have seriously dented Malarkodi’s overall credibility, the judge dismissed this inconsistency with startling alacrity without so much as a direction that further evidence be led by the Prosecution to clarify the matter, failing which adverse inferences would be drawn: see [58]. This stance was diffidently justified on the basis that the Defence had neither asked for the relevant statement recorded by SI Soon to be tendered in evidence, nor queried Malarkodi about the statement: see [58] above; GD at [99] and [100]. Once again, the trial judge incorrectly transposed the burden of proof onto the Defence. In addition, it is pertinent to point out that SI Soon gave his testimony only after Malarkodi had completed hers. How could Malarkodi be queried when the relevant facts emerged only during SI Soon’s testimony? Was the Defence legally obliged to apply for Malarkodi to be recalled to the stand?
89 Last but not least, with respect to Malarkodi’s gesture of pleading for forgiveness upon regaining consciousness, the trial judge accepted Malarkodi’s explanation that she had been repeatedly reprimanded by the appellant, and inferred that it was hardly surprising that upon coming to, Malarkodi should assume she was in serious trouble again and therefore instinctively ask for forgiveness; in the trial judge’s estimation, this did not amount to an acknowledgement of guilt: see [57] above; GD at [96]. It is equally plausible to surmise that Malarkodi might have regretted her actions immediately upon recovering consciousness, only to change her mind in hospital. It appears to me that too much effort has been expended by the trial judge in rationalising and ironing out each and every awkward crease in Malarkodi’s testimony individually when the more prudent course of action would simply have been to review her entire testimony objectively and holistically.
90 There is one further crucial point. Not one of the doctors who gave evidence on Malarkodi’s injuries had any particular or specific expertise in the area of vaso-vagal attacks or vasodilation. A cardiologist or neurologist would have been in a far better position to proffer more reliable and specialised knowledge on this particular issue.
91 In the final analysis, I have no choice but to dismiss Malarkodi’s narration of events as dubious and inconsistent both with numerous objective facts as well as the injuries sustained by her. The trial judge erred in according “full weight” to Malarkodi’s evidence. She failed to exercise the necessary caution and discretion in scrutinising and evaluating the evidence for both internal and external consistency: see [72].
Should the appellant’s version of facts be rejected in favour of Malarkodi’s version?
92 I agree with the trial judge that there are gaps and omissions in the evidence of the appellant as well. In particular, she has clearly sought to draw attention away from the tension prevailing in her household at the material time. After sifting through both testimonies, one cannot ignore that there was palpable tension in the household at the material juncture. If Malarkodi’s injuries were indeed self-inflicted as the Defence postulates, she must have been reduced to a very unstable and fragile bundle of nerves just prior to the incident. Such a scenario is somewhat at odds with the picture the appellant has painstakingly painted of a tense albeit promising relationship with Malarkodi. The trial judge quite accurately detected an attempt on the appellant’s part to steer clear of all facts that might suggest a basis for deeply rooted unhappiness and/or resentment festering between the protagonists.
93 Nevertheless, as I mentioned in Jagatheesan ([69] supra) at p 63, [61]:
… [T]he Prosecution bears the burden of proving its case beyond reasonable doubt. While this does not mean that the Prosecution has to dispel all conceivable doubts, the doctrine mandates that, at the very least, those doubts for which there is a reason that is, in turn, relatable to and supported by the evidence presented, must be excluded. 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. … A trial judge must also bear in mind that the starting point of the analysis is not neutral. An accused is presumed innocent and this presumption is not displaced until the Prosecution has discharged its burden of proof. Therefore, if the evidence throws up a reasonable doubt, it is not so much that the accused should be given the benefit of doubt as much as the Prosecution’s case simply not being proved…
[emphasis added]
To that extent, the unsatisfactory wrinkles in the appellant’s testimony alone cannot ground a determination that the Prosecution’s case has been established beyond a reasonable doubt. Particular and/or isolated discrepancies in a witness’s testimony are not as critical as the relevance or significance of such discrepancies in relation to the actual issues to be determined. It is not every lie that undermines a party’s entire credibility and case theory. Whether a witness should lie about a peripheral issue or even a relevant one, it does not inevitably follow that his entire testimony should be rejected. Unsatisfactory evidence must be tested holistically in the crucible of reason and logic in the context of all the relevant objective facts and the respective case theories. It is imperative in the final analysis, that the Prosecution is able to adduce sufficient evidence to support its theory of guilt and to dispel any reasonable doubt, thereby properly discharging its burden of proof.
94 In so far as the probabilities raised or thrown up by the evidence are concerned, Prof Tan Y L’s erudite discussion in Criminal Procedure ([70] supra) on inherent probabilities at XIX 211A-211B is most pertinent. He states:
It commonly happens that the trial judge’s findings based on credibility are disturbed because the advantage of having seen and heard the witnesses is nullified by the inherent probabilities of the case…
A definite statement of what the inherent probabilities are is hard to give. No court has attempted to define the notion but it is recognised that it may refer to a particular account or to the totality of the case. It may happen that the account in conflict is inherently probable or that, although both accounts are probable, one is against the inherent probabilities of the case seen in its totality.
[emphasis added]
This careful evaluative approach is illustrated by PP v Lim Kuan Hock
95 Similarly in this case, quite apart from the trial judge’s assessment of the witness’s credibility based on demeanour alone, it is of vital importance that she should evaluate the inherent probabilities of the witness’s accounts as well. Granting that both Malarkodi’s and the appellant’s versions of fact may theoretically be possible, I find that Malarkodi’s account is intrinsically incompatible with the inherent probabilities of the case, viewed in its totality. For instance, if the appellant had indeed lost complete control of her emotions and conduct on that fateful afternoon as Malarkodi portrayed, how could she have recovered her senses and equanimity so promptly as to immediately seek help from her neighbour? Significantly, the appellant did not from the outset seek to conceal the incident from third parties. Madam Fong’s evidence corroborated the appellant’s testimony that the latter had very promptly sought her assistance upon discovery of Malarkodi’s injuries. Further, as already amply discussed above at [83] to [85], it appears most incredible that Malarkodi had passively allowed four blows to be struck on her hand, or for that matter that the appellant had been able to cut Malarkodi’s hand at least four times in one very specific area if Malarkodi had indeed lost consciousness and collapsed in the manner she recounted. Malarkodi’s explanation that the appellant had proceeded to chop her fingers simply because she had forgotten to turn on the fan for the appellant’s son appears to defy both logic and commonsense. While I cannot completely ignore the possibility that fact may on occasion prove to be stranger than fiction warranting that disbelief be suspended, the instant scenario does not appear to be such an occasion. There is no suggestion that the appellant had, prior to the alleged incident, physically abused Malarkodi or any other party or indeed that she was even capable of such wanton abuse. It is one thing to verbally abuse or to raise one’s hands in anger or frustration; it is quite another to consciously snatch a chopper and viciously rain a series of blows on the hand of a ‘helpless’ victim. I also note that Malarkodi had testified that knives other than the chopper, were collectively stored on a dish rack by the side of the kitchen sink. If the appellant had in fact lost control of herself as alleged, it is wholly illogical for her to have deliberately stooped to retrieve a chopper from the stove cabinet, when there were other knives far more accessible and immediately available. Even assuming she had used the chopper, how did the appellant manage to aim and inflict all four ‘parallel’ blows on one very limited area on Malarkodi’s right hand? Why were there no clear-cut defensive injuries or withdrawal wounds?
96 I am persuaded that the appellant’s version of events, albeit not entirely flawless, is more inherently probable than Malarkodi’s. To that extent, it should not have been so quickly dispensed with . Conversely, Malarkodi’s account of the facts is intrinsically improbable and should not have been so readily accorded “full weight” by the judge, largely on the basis of what struck her as “favourable” demeanour.
Does the medical evidence indeed justify the conclusion that the injuries were not self-inflicted?
97 It is undisputed that multiple cuts were inflicted by the chopper and that the cuts on Malarkodi’s fingers were of varying degrees of severity ranging from mere superficial cuts to deep incisions severing external tendons and fracturing two finger bones. Each of these distinct injuries was literally within millimetres of one another on two adjacent finger bones. I meticulously evaluated the photographs and X-rays of the injuries to Malarkodi’s hands against the medical evidence proffered. The shallower cuts clearly dovetailed with Prof Lim’s opinion that they were hesitation cuts therefore constituting cogent evidence of self-inflicted injuries. It seems to me entirely inconceivable that an assailant could have inflicted one solitary blow to a victim’s fingers, sufficient to induce in the victim a vaso-vagal shock, and then proceed to lightly lacerate the same victim’s fingers in several virtually parallel spots within, almost literally, a hair’s breadth of the first injury. It is equally bizarre that a victim (let alone one with Malarkodi’s feisty disposition) would have passively allowed her assailant to cut her fingers superficially a few times before dealing one or more heavy blows. The only logical explanation accounting for the presence of the superficial cuts is the one offered by Prof Lim ie, that Malarkodi inflicted hesitation cuts on her fingers before plucking up the courage to deliver the final coup de grace.
98 Dr Sandeep was of the view that the superficial cuts in this case were not a testament to hesitation cuts because in his estimation hesitation cuts sufficed to get the attention sought in non-suicide cases, rendering no further need for an ultimate cut that nearly amputated the finger. Dr Sandeep’s opinion was premised entirely and exclusively on the assumption that if Malarkodi had inflicted her own injuries, it would have been for the sole purpose of obtaining attention; see [38]. He has missed the point altogether. With respect, such an assumption is not founded or substantiated by any objective evidence or any psychiatric assessment of Malarkodi; it is pure conjecture to say the least. Malarkodi could very well have been motivated by anger or resentment towards the appellant which escalated and exploded in an irrational urge to retaliate by blaming the appellant for the injuries she inflicted on herself. In the final analysis, the Prosecution’s medical experts, who were considerably less experienced and thorough than Prof Lim, could not convincingly account for the superficial cuts on her fingers. Their relatively limited experience with hand injuries coupled with references to suicide cases (see [38]) was neither impressive nor persuasive. It appears from the subtext of the trial judge’s grounds of decision that she was unduly impressed by the Prosecution’s reliance on two medical experts, in contradistinction to the sole expert the Defence had recourse to: (see [75] above) and GD at [119]. It would also appear that she failed to accord proper significance either to Prof Lim’s substantial clinical experience or to his standing as an expert. I have noted that he has appeared on many occasions as an expert witness for the Public Prosecutor in diverse criminal matters. It is evident from the record that both Dr Sandeep and Dr Su lack similar expertise and credentials. While this is by no means conclusive, it is not entirely insignificant, in the context of this case and in light of the conspicuously disparate views involved. Prof Lim has offered by far the most thoughtful, thorough and comprehensive account of how the injuries might have been sustained. Even if one aspect of his evidence in relation to the dents on the chopper might possibly invite criticism, it does not detract from the coherence and logic of his testimony in relation to the other issues. Prof Lim has correctly emphasised the peculiarities of Malarkodi’s injuries, the very close proximity and parallel positioning of the injuries in relation to one another, the absence of withdrawal wounds and in the same context the absence of evidence pointing to any defensive reaction from Malarkodi. On the contrary, Dr Sandeep had to be recalled to explain several of these issues, having failed to address them initially. Given their lack of thoroughness and inadequate knowledge relating to several controversies, I cannot fathom why the trial judge found it “surprising” that Dr Sandeep and Dr Su should have missed the presence of hesitation cuts, if any, see [64] above.
99 On a separate note, I find it somewhat troubling that no psychiatric evaluation was conducted by the investigators of either Malarkodi or the appellant. In cases where it is contended that injuries have been self-inflicted, it would usually be very helpful, if not crucial, to conduct a psychiatric assessment of both the victim and even, perhaps, the alleged assailant. Self-inflicted injuries are invariably a manifestation of a troubled mental and emotional state and it is almost imperative in such cases that the Prosecution arrange for such ‘victims’ to undergo a psychiatric assessment to explore and evaluate the thrust and essence of any allegation relating to self-inflicted injury.
100 The trial judge had unequivocally accepted Dr Sandeep’s and Dr Su’s opinions that the injuries were unlikely to be self-inflicted, on the basis that the cuts were on Malarkodi’s dominant hand, with the direction of the cuts indicating that self infliction would have been both difficult and awkward. However, the Prosecution’s medical experts failed to appropriately or adequately consider the wider spectrum of possibilities as to how the hand might have been positioned. Counsel for the appellant had suggested the possibility of a rotated right hand. I figure also that it would have been possible for the little finger to have been positioned slightly above the ring finger, instead of having all the fingers evenly splayed on a surface as envisioned by the Prosecution’s experts, who were incidentally unduly preoccupied in their quest to understand and correlate the incident with their experience of suicide cases. Such positioning might possibly explain the deeper cuts on the little finger as well as the alignment of the fractures on the ring and little fingers. There are many other possible permutations of how Malarkodi’s right hand might have been positioned. Unfortunately, none of these had been adequately explored and analysed, particularly by the Prosecution’s medical experts. The crucial point to note, however, is that the conclusion drawn by Dr Sandeep and Dr Su, that self-infliction is unlikely is founded on and inextricably tied to their notion that Malarkodi’s hand had been placed in a particular position. Such a conclusion must necessarily be re-evaluated when a wider range of other possible hand positions is brought into the picture. I am not persuaded objectively that Malarkodi’s injuries were necessarily caused by another person. I further note that both Dr Su and Dr Sandeep were unable to categorically rule out the possibility of self-infliction.
101 Based on an objective evaluation of both the medical and factual evidence, there is decidedly more than an element of reasonable doubt that the injuries on Malarkodi’s hand were not caused by the appellant. The possibility that the injuries were self-inflicted is both real and tangible and should not have been so readily dismissed by the trial judge.
Conclusion
102 The legitimacy of our criminal justice system is founded on the bedrock principle articulating and mandating that proof of guilt be established beyond reasonable doubt. Such a doctrine is two-fold: first, the burden of proof lies with the Prosecution; second, the standard of proof required is “beyond a reasonable doubt”. The first concept requires the Prosecution to adduce sufficient evidence not only to support its theory of guilt but to categorically dispel any reasonable doubts that may arise on the evidence presented. Unfortunately, the trial judge seems to have overlooked this when she inexplicably placed the burden on the Defence to clarify the inconsistencies in the Prosecution’s own evidence and then proceeded on her own initiative to resolve major doubts in favour of the Prosecution. I am also puzzled why the trial judge chose not to direct, at the very least, that the issues be clarified if they continued to cause her even the slightest anxiety at the end of the testimonial phase of the proceedings. It is apparent that she had noted the inconsistencies between Malarkodi’s statements to the examining doctors and SI Soon on the one hand and her testimony to the court on the other. Contrary to all expectations, she did not consider it appropriate to delve further into these inconsistencies, despite the fact that it was well within her constitutional and statutory remit to do so. Instead, she summarily and incorrectly concluded that the burden was on the Defence to clarify these issues notwithstanding that the doubts in question surfaced in the course of testimonies given by the Prosecution’s witnesses. Such an approach is entirely incongruous with the axiom that the burden of proof lies with the Prosecution (unless otherwise provided statutorily).
103 The necessity for the Prosecution to prove its case beyond reasonable doubt must be considered in light of the entirety of the evidence in the proceedings. Judges must bear in mind that when both the Prosecution’s and Defence’s case theories seem improbable, it does not inevitably mean that the court must choose between them. In such cases, the judge should simply rule that the burden of proof has not been properly discharged. It is most unfortunate that the trial judge chose to rely so heavily and unduly on her impressions of Malarkodi’s and the appellant’s credibility; in the process, she failed to objectively accord sufficient weight to the evidence and to properly evaluate the inherent probabilities of the Prosecution’s case theory, tested against logic and commonsense. I find that the Prosecution has failed to discharge the burden of proving the appellant’s guilt beyond any reasonable doubt. There remain several unanswered and troubling questions, not to mention numerous unexplained inconsistencies when the sum total of the various statements given by Malarkodi is carefully scrutinised. The trial judge’s findings of credibility apropos Malarkodi cannot be sustained and are most certainly susceptible to review. Our courts have always been astute in seeking to ensure that an assessment of credibility does not preclude an appellate court from review if it is apparent that such a finding could create serious injustice.
104 The Prosecution has failed to appropriately and adequately dispel the many doubts clouding and confounding its case theory. Further, there are serious gaps in the Prosecution’s case, which the trial judge should not have sought to supplement by resorting to and relying on unverifiable inferences and suppositions. Findings on these issues should be supported and substantiated by wholly verifiable facts and inferences, rather than conjecture, and are fundamental in assessing whether the Prosecution’s burden of proof has been properly discharged. As such, there remain several unresolved reasonable doubts, to say the least, in relation to the appellant’s guilt.
105 After carefully reviewing the evidence, I have concluded that the trial judge seriously erred in convicting the appellant on the basis of the evidence adduced. In the result, the conviction has been set aside and the appellant acquitted.
Copyright © Government of Singapore.
This judgment text has undergone conversion so that it is mobile and web-friendly. This may have created formatting or alignment issues. Please refer to the PDF copy for a print-friendly version.
Version No 0: 18 Apr 2007 (00:00 hrs)