1 This is a tragic case. Over the course of a week, the respondents poured very hot water on their young son on four occasions and it ended in his death. The respondents had also cruelly abused the child in many other ways in the three months prior to that fatal week. The respondents are Azlin binte Arujunah (“Azlin”) and Ridzuan bin Mega Abdul Rahman (“Ridzuan”). They were jointly tried before the High Court judge (the “Judge”) on six and nine charges respectively of offences involving various acts of physical abuse they committed against their son (the “Deceased”) from July until October 2016. These included one charge of murder under s 300(c) read with s 34 and punishable under s 302(2) of the Penal Code (Cap 224, 2008 Rev Ed) (“Penal Code”) brought against each of them (“Murder Charges”). The Murder Charges arose out of the aforementioned four occasions when Azlin and/or Ridzuan intentionally inflicted severe scalding injuries on the Deceased by pouring very hot water on him. We refer to these scalding incidents as Incidents 1 to 4 respectively.

2 Azlin was solely responsible for Incidents 1 and 3, while Incidents 2 and 4 were carried out by her acting jointly with Ridzuan. Specifically, the Judge found that Azlin and Ridzuan both scalded the Deceased in Incident 2, while Ridzuan was the only one who physically committed the acts in question in Incident 4 (though these acts were intended by Azlin who had instigated Ridzuan). It is undisputed that it was the cumulative scald injury caused by the collective acts of scalding carried out by Azlin and Ridzuan over the four incidents (“Cumulative Scald Injury”) that killed the Deceased. The hot water that the respondents poured on the Deceased was between 70 and 90.5℃, and the undisputed medical evidence was that water hotter than 70℃ would cause mid to deep thermal burns even with minimal contact.

3 The Judge acquitted Azlin and Ridzuan of their respective Murder Charges primarily because she considered that there was insufficient evidence to infer that the respondents intended specifically to inflict what was referred to as a “s 300(c) injury”. By this, the Judge meant a bodily injury which is sufficient in the ordinary course of nature to cause death. The Judge thought that this had to be shown when a conviction was sought in the context of acts done pursuant to a common intention under s 34 of the Penal Code. The Prosecution then sought the conviction of Azlin alone on the following amended charge under s 300(c) of the Penal Code (“alternative s 300(c) charge”):

4 What was somewhat unusual about the alternative s 300(c) charge was that it sought to employ s 34 of the Penal Code not to impose constructive liability for the entire “criminal act” giving rise to the offence in question (which encompassed all of Incidents 1 to 4), but to attribute liability to Azlin for two discrete components (Incidents 2 and 4) that had been carried out by Ridzuan and treating these as part of the entire criminal act (Incidents 1 to 4) that is charged against Azlin.

5 The Judge rejected the alternative s 300(c) charge for two broad reasons: see Public Prosecutor v Azlin bte Arujunah and another [2020] SGHC 168 (“GD”). The first main reason was her view that s 34 is not a “free-standing principle of attribution” that allows the court to attribute liability for acts done by another that forms a part of the “criminal act” that is the subject of the charge against the accused person (GD at [121]). The second reason was that, in the Judge’s view, for Ridzuan’s acts in Incidents 2 and 4 “to be attributed to Azlin for the purposes of liability under s 300(c) of the Penal Code”, the “common intention they needed to share” was the common intention to inflict a s 300(c) injury (GD at [121]). In coming to the latter view, the Judge relied on what she understood to be this court’s ruling in Daniel Vijay s/o Katherasan and others v Public Prosecutor [2010] 4 SLR 1119 (“Daniel Vijay”) to the effect that, if two offenders, A and B, intend to commit a certain offence, say robbery, but in the course of carrying out that intention, one of the offenders, A, commits the offence of murder under s 300(c) of the Penal Code, then B can only be held jointly liable for murder under s 300(c) read with s 34 of the Penal Code if B intended specifically that A should inflict a s 300(c) injury, meaning that B must have intended that an injury that is sufficient in the ordinary course of nature to cause death is inflicted (“Daniel Vijay test”) (see GD at [97]). The Judge found that the Prosecution was not able to prove a common intention to inflict the s 300(c) injury beyond reasonable doubt in this case, and so she held that the alternative s 300(c) charge was not made out (GD at [110] and [121]).

6 However, Daniel Vijay concerned a “dual crime” scenario (or what was referred to as a “twin crime” scenario in that judgment). This is where multiple offenders commonly intend to commit a primary offence (such as robbery), but one of the offenders (the “primary offender”) then commits an offence that was not part of the common venture (such as murder under s 300(c) of the Penal Code (“s 300(c) murder”)) in the course of committing the primary offence. The Daniel Vijay test was developed to answer the question whether the other offenders (the “secondary offenders”) in such a “dual crime” situation can be held liable for the collateral offence. On the other hand, the present case does not concern such a “dual crime” scenario because only one offence – the murder of the Deceased – had allegedly been committed pursuant to Azlin’s intention, and the alternative s 300(c) charge seeks to hold Azlin liable for that very offence, rather than some other “collateral offence” that had been committed by Ridzuan and that went beyond the scope of Azlin and Ridzuan’s original common intention. However, neither is the present case a “single crime” scenario, since Azlin and Ridzuan did not commonly intend to commit all four scalding incidents. This raises the question of whether there is a difference between “dual crime” and “single crime” scenarios when considering constructive liability under s 34 of the Penal Code, particularly in the context of s 300(c) murder, and whether s 34 can be applied in the present case given that it does not fit neatly into either scenario.

7 In the event, the Judge amended the Murder Charges to charges of voluntarily causing grievous hurt by means of a heated substance under s 326 of the Penal Code and sentenced Azlin to an aggregate sentence of 27 years’ imprisonment and an additional 12 months’ imprisonment in lieu of caning, and Ridzuan to an aggregate sentence of 27 years’ imprisonment and 24 strokes of the cane. CA/CCA 17/2020 (“CCA 17”) is the Prosecution’s appeal against the Judge’s decision not to amend the Murder Charge against Azlin to the alternative s 300(c) charge. CA/CCA 24/2020 (“CCA 24”) and CA/CCA 25/2020 (“CCA 25”) are the Prosecution’s appeals against the Judge’s decision not to sentence Ridzuan and Azlin respectively to life imprisonment for the amended s 326 charges. This case presents us with the opportunity to clarify the principles and operation of s 34 of the Penal Code and in particular, its operation in the context of murder under s 300(c).

8 Given the number of issues involved in the present judgment, it is helpful to set out a table of contents for reference:

9 Azlin and Ridzuan are both Singaporeans and were 24 years’ old at the time of the offences.

10 Azlin faced the following six charges in the joint trial:

(a) the Murder Charge against her (charge marked “C1A”);

(b) two charges of ill-treating a child, an offence under s 5(1) punishable under s 5(5)(b) of the Children and Young Persons Act (Cap 38, 2001 Rev Ed) (“CYPA”), by hitting the Deceased on his body, back and legs with a broom (charge marked “C2”), and pushing the Deceased on the left shoulder, causing him to fall sideways (charge marked “C3”), both of which were committed in August 2016;

(c) one charge of abetment by instigating Ridzuan to voluntarily cause hurt to the Deceased by means of a heated substance, an offence under s 324 read with s 109 of the Penal Code, by using a heated metal spoon to burn the Deceased’s right palm, which caused a blister on his palm, sometime between end-August and early-September 2016 (charge marked “C4”); and

(d) two charges of ill-treating a child pursuant to a common intention with Ridzuan, an offence under s 5(1) punishable under s 5(5)(b) of the CYPA read with s 34 of the Penal Code, “by pushing [the Deceased], causing his head to hit the wall and punching him on his face” sometime in October 2016, causing a laceration on his head and comminuted fractures of his nasal bone (charge marked “C5A”), and by confining the Deceased in a cage sometime between 21 and 22 October 2016 (charge marked “C6”). The act in charge C5A was originally framed by the Prosecution in a charge marked “C5” as “pushing [the Deceased’s] head against the wall and punching him on his face”, but this was amended by the Judge after trial into the act as it is now formulated in the amended charge C5A, as just stated (see GD at [44]).

11 Ridzuan faced the following nine charges in the joint trial:

(a) the Murder Charge against him (charge marked “D1A”);

(b) three charges of ill-treating a child under s 5(1) punishable under s 5(5)(b) of the CYPA by using a pair of pliers to pinch the Deceased’s buttocks sometime in July 2016 (charge marked “D2”); using a pair of pliers to pinch the back of the Deceased’s thighs sometime in July 2016 (charge marked “D3”); and flicking ashes from a lighted cigarette on the Deceased’s arms and using a hanger to hit him on the palm sometime in October 2016 (charge marked “D6”);

(c) three charges of voluntarily causing hurt by means of a heated substance, an offence under s 324 of the Penal Code, by using a heated metal spoon to burn the Deceased’s right palm, which caused a blister on his palm, on three occasions: sometime between end-August 2016 and early-September 2016 (charge marked “D4”), sometime in early-October 2016 (charge marked “D5”), and sometime between 18 and 19 October 2016 (charge marked “D8”);

(d) two charges of ill-treating a child in furtherance of the common intention of both Azlin and Ridzuan, an offence under s 5(1) punishable under s 5(5)(b) of the CYPA read with s 34 of the Penal Code, for the same acts stated at [10(d)] above (charges marked “D7A” and “D9”). The act in charge D7A was similarly originally framed by the Prosecution in a charge marked “D7” as “pushing [the Deceased’s] head against the wall and punching him on his face”, but this was amended by the Judge after trial into the act of “pushing [the Deceased], causing his head to hit the wall and punching him on his face” as it is now formulated in charge D7A (see GD at [44]). Charges D7A and D9 correspond to charges C5A and C6.

12 All the foregoing offences were committed in Azlin and Ridzuan’s home. We shall refer to the charges against Azlin and Ridzuan, besides the Murder Charges, collectively as the “Abuse Charges”. It is undisputed that, prior to the offences, the Deceased had lived with a friend of Azlin’s, [Z], since March 2011, when the Deceased was an infant.

13 The Abuse Charges occurred from July 2016 until the time covered by the Murder Charges between 15 and 22 October 2016. The Judge acquitted Azlin and Ridzuan of charges C4 and D4. The Judge convicted Azlin and Ridzuan of the remaining Abuse Charges. The convictions, sentences, and acquittals of the Abuse Charges are not in issue in these appeals.

14 As for the Murder Charges, the roles played by Azlin and Ridzuan in the four scalding incidents were largely not disputed (see GD at [61]). As both parties elected to remain silent and not give evidence in court, the primary source of evidence for the scalding incidents was what they disclosed in their respective investigative statements. The sequence of events involving the Abuse and Murder Charges unfolded as follows.

15 The abusive acts began in July 2016. Ridzuan first used pliers to hurt the Deceased twice in July 2016 (charges D2 and D3). This was followed in August 2016 by Azlin hitting the Deceased with a broomstick so hard that he was limping thereafter (charge C2). Later that same month, Azlin pushed the Deceased so hard that he fell, hitting his head on the edge of a pillar. As a result, he bled from the head (charge C3).

16 In October 2016, Ridzuan used a heated spoon on the Deceased’s palm (charge D5), flicked ash from a lighted cigarette on him, and hit him with a hanger (charge D6).

17 Incident 1 was committed by Azlin alone and occurred sometime between 15 and 17 October 2016. It was around noon when Azlin was in the kitchen and noticed that their milk powder had dwindled in quantity. Azlin then called the Deceased to the kitchen. Right after the Deceased arrived in the kitchen, Azlin grabbed him by his right ankle. While still holding onto him, Azlin filled a glass mug to around one-quarter full with hot water from the water dispenser and poured it onto his right leg. The Deceased started crying and Azlin repeated this two or three times before letting go of the Deceased. The Deceased then ran into the toilet, and Azlin questioned him over her suspicion that he had consumed some milk powder but he denied this. Azlin then took hold of him, refilled the mug with hot water and poured it on the Deceased’s hand four or five more times. When the Deceased got free of Azlin’s grip, she refilled the mug and splashed it over his left arm, and some also splashed onto his chest. She stopped when Ridzuan woke up and shouted for them to keep quiet.

18 Following that incident, Azlin saw that the Deceased was limping and that skin was peeling from his hands, arms and chest. The Deceased also told Azlin that he was in pain. She purchased some cream for the skin injuries, and thought the Deceased was walking normally by the next day. Ridzuan, on the other hand, claimed that he did not observe any peeling skin, but that the Deceased’s skin was “reddish”, and that the Deceased was able to “walk normally and run and play with his brother.”

19 Incident 2 was committed by Azlin and Ridzuan jointly. Sometime between 17 and 19 October 2016, Azlin splashed the Deceased’s body with hot water. Azlin stated in her investigative statement that she could not remember why she did so. In response, the Deceased shouted, “Kau gila ke apa” (translated by Azlin in her statement as “Are you crazy or what”). Azlin became angry and re-filled the glass mug with hot water and splashed the Deceased on his face. She then re-filled the glass mug and splashed the Deceased at least five and up to seven times, on his face, body, arms, and legs. Ridzuan also participated in this incident. When he heard the Deceased shout at Azlin, Ridzuan picked up a green mug and splashed hot water at the Deceased, and it landed on the Deceased’s face and body. On Ridzuan’s account, there was more splashing of the Deceased with hot water after the Deceased had bathed, but, according to Azlin, at some point, Ridzuan told Azlin “to stop and cool down”.

20 After Incident 2, both Azlin and Ridzuan noticed that the Deceased suffered significant injuries. This included white patches of raw skin that appeared on his face and chin, on his stomach and body, and on his left shoulder; pus was oozing from his forehead, and from his back and left shoulder; and skin was peeling from his back, face, hands, thighs and legs. Azlin also said she “could see the whitish flesh” [emphasis added] below the outer skin surface. The Deceased also became noticeably “weak” after Incident 2, and was “not able to move [as] usual”, needing assistance from his brother even to get food to eat.

21 In another incident, seemingly after Incident 2, Azlin pushed the Deceased, causing him to hit his head against the wall, and Ridzuan punched the Deceased on the face so hard that his nasal bone was fractured (charges C5A and D7A). Although this happened during the fateful week in question, the Prosecution does not rely on this incident in connection with the alternative s 300(c) charge.

22 Incident 3 was committed by Azlin alone. On 21 October 2016 at about 9pm, Azlin became angry with the Deceased when he kept asking for a drink. Azlin splashed the Deceased with a glass mug filled with hot water. In all, she splashed water at the Deceased nine or ten times, though on some of these attempts, she missed the Deceased. Azlin subsequently went to sleep.

23 On 21 and 22 October 2016, the respondents locked the Deceased in a cat cage (charges C6 and D9). He was only let out of the cage to be fed. The cat cage measured 0.91m in length, 0.58m in width, and 0.70m in height, while the Deceased was 1.05m tall at the material time. The cat cage was made of metallic bars, and Dr Chan Shijia, who performed the autopsy on 24 October 2016, testified that it was possible that the lacerations on the Deceased’s face and scalp might have been a result of being confined in the cage and being scratched by the sharper metallic parts when the Deceased tried to move in the cage (see GD at [49]). Based on the photograph of the cat cage, it does not appear that there was any mattress or soft padding in the cat cage. The Deceased was in the cage from around 7pm until about 10pm on 21 October 2016, and from around 4am until about noon on 22 October 2016. By this time, the Deceased was clearly unwell, with a fever and with skin peeling off his face, hands, back, thighs and the back of his legs (see GD at [48] to [49]). Again, this incident is not relied on by the Prosecution in connection with the alternative s 300(c) charge.

24 Incident 4 was committed by Azlin and Ridzuan jointly on 22 October 2016. Azlin instructed the Deceased to bathe, but he had not removed his shorts when he came to the kitchen. Azlin got upset and woke Ridzuan up and, as the Judge had found, told him to deal with the Deceased (GD at [128]). She then started bathing her two daughters. Ridzuan asked the Deceased to remove his shorts and when he refused, Ridzuan used the handle of a broom to beat the Deceased two or three times on his legs. Both Azlin and Ridzuan again asked the Deceased to remove his shorts. Ridzuan then filled half a glass mug with hot water from the dispenser and threw the hot water on the floor beside the Deceased as a warning. Some of the water touched the Deceased’s leg. The situation escalated with more scolding before Ridzuan refilled the mug with hot water and splashed the Deceased on the left side of his body. When the Deceased again refused to remove his shorts, Ridzuan refilled half the mug with hot water yet again and poured this on the Deceased’s back. Ridzuan then refilled the mug a fourth time and splashed hot water on one or both of the Deceased’s calves. The Prosecution’s case was that Azlin was present throughout the incident. Ridzuan’s account in his investigative statement is that Azlin was “shouting” at the Deceased and was “beside” Ridzuan during this incident. Azlin on the other hand asserted that she was busy with her daughters. The Judge found that, while Azlin was not beside Ridzuan throughout this incident, her investigative statement made it clear that she saw and acquiesced in Ridzuan’s actions, including his repeatedly splashing the Deceased with hot water (GD at [67]). The Deceased finally fell and lay on his side. Ridzuan then summoned Azlin and together they rinsed the Deceased with cold water.

25 After the Deceased collapsed, Ridzuan carried him into the bedroom and laid him on the floor. The skin on the Deceased’s face, trunk, arms and legs was peeling badly, and some areas had turned white. His eyes were open, but he was weak and, according to Azlin, was unable to move. The Deceased also complained that he felt cold. Seeing the Deceased in this state, both Azlin and Ridzuan became “very scared”, as stated in Ridzuan’s investigative statement. Ridzuan contemplated sending the Deceased to the hospital, but did not want to call the ambulance because he was afraid the police would come as well. Azlin suggested that they wait to see if the Deceased’s condition would improve.

26 At around 6pm that evening, Azlin and Ridzuan left the Deceased alone in the flat. Accompanied by their other children, they went to fetch Ridzuan’s aunt, Kasmah binte Latiff (“Kasmah”), from her home. Prior to going to Kasmah’s house to seek help, Ridzuan told Azlin to lie about the injuries and to say that the Deceased had accidentally pulled the electrical cord of the kettle and the water had splashed on him as a result. They then returned to the flat, and brought the Deceased to KK Women’s and Children’s Hospital (“KKWCH”). The Deceased was admitted to the emergency department on the same day (22 October 2016) at around 7.57pm. At the emergency room, Ridzuan told the nursing staff that he was disciplining the Deceased when the Deceased accidentally pulled on the kettle, splashing hot water on himself. Ridzuan repeated this story to the police officers who first spoke with him.

27 The Deceased received emergency intensive care, but was pronounced dead on 23 October 2016 at 9.13am. The extent of the Deceased’s total body surface area (“TBSA”) covered by burns was estimated by Dr Gavin Kang Chun-Wui, the burn specialist who performed debridement to clean the Deceased’s wounds on the evening of 22 October 2016, at 67%, and by Dr Chan Shijia, who performed the autopsy on 24 October 2016, at 75% after debridement; this consisted of mid and deep dermal burns (see GD at [72]) and included sensitive parts of the Deceased’s body, including his face and genital area.

28 The Judge found that the Cumulative Scald Injury, caused by the four incidents, was the cause of death (GD at [60] and [78]). The Judge also found that, while the medical evidence was not able to show the extent of burns caused by each scalding incident or how each incident contributed to the Deceased’s death (GD at [87]), it did establish that the Cumulative Scald Injury was sufficient in the ordinary course of nature to cause death (GD at [60], [86] and [87]). These findings are not challenged in these appeals.

29 As mentioned at [3] above, the Judge acquitted Azlin and Ridzuan of the Murder Charges. The Judge, relying primarily on Daniel Vijay, held that, for joint liability to be imposed under s 300(c) read with s 34 of the Penal Code, the joint offenders must have had the common intention to cause what was referred to as a s 300(c) injury, and “not any other type of injury” (see GD at [97]). By a s 300(c) injury, the Judge meant a bodily injury which was sufficient in the ordinary course of nature to cause death (see [3] above; see GD at [92]). The Judge found that there was insufficient evidence to infer that Azlin and Ridzuan intended to inflict a s 300(c) injury (GD at [110]). There was no evidence of any pre-arranged plan between the respondents to inflict any particular injury, and there was no evidence of an intention to cause any particular aggregate injury or to continue scalding the Deceased to the point where it amounted to a s 300(c) injury. Therefore, the Judge acquitted Azlin and Ridzuan of the Murder Charges.

30 The Judge then invited views from the parties on the alternative charges that could be framed against the respondents under s 128 of the Criminal Procedure Code (Cap 68, 2012 Rev Ed) (“CPC”). The Prosecution made two alternative proposals in respect of Azlin. The first was that the Murder Charge should be amended to the alternative s 300(c) charge, as extracted at [3] above. In the alternative, the Prosecution proposed that Azlin should face four charges under s 326 of the Penal Code for voluntarily causing grievous hurt by dangerous means in respect of each of the four scalding incidents respectively, with two of those charges to be read with s 34 of the Penal Code to reflect a common intention shared with Ridzuan to commit Incidents 2 and 4 respectively. Under this alternative proposal, Ridzuan would be charged with two charges of s 326 read with s 34 of the Penal Code for Incidents 2 and 4 respectively. Azlin submitted that the appropriate amended charges should be under s 326 of the Penal Code, while Ridzuan submitted that alternative charges for Incidents 2 and 4 under s 324 of the Penal Code (voluntarily causing hurt by a heated substance rather than grievous hurt) would be more appropriate.

31 As we alluded to in the introduction at [5] above, the Judge rejected the alternative s 300(c) charge for two primary reasons. These reasons turned on the Judge’s interpretation of s 34 of the Penal Code which we reproduce as follows (GD at [121]):

32 The Judge further elaborated at [124] of the GD that Azlin and Ridzuan had to share a common intention to commit Incidents 1 to 4 in order for the alternative s 300(c) charge to be permissible:

33 In the premises, the Judge accepted the Prosecution’s alternative submission and exercised her power under s 128 of the CPC to amend the Murder Charges to charges under s 326 of the Penal Code (GD at [125] to [126]). The Judge framed the grievous hurt in these charges as “hurt which endangered life” (under s 320(h), Penal Code) rather than “death” (under s 320(aa), Penal Code). This was also the Prosecution’s position.

34 Azlin pleaded guilty to three of the amended s 326 charges and claimed trial to the amended s 326 charge concerning Incident 4. Ridzuan pleaded guilty to the amended s 326 charges. The Judge convicted Azlin and Ridzuan of all the amended s 326 charges (GD at [172] to [173]):

(a) For Incident 1, the amended charge was framed against Azlin as follows (charge marked “C1B2”):

(b) For Incident 2, the following amended charge was framed against Azlin and Ridzuan (charges marked “C1B3” and “D1B2” respectively”):

(c) For Incident 3, the following amended charge was framed against Azlin as follows (charge marked “C1B4”):

(d) For Incident 4, the following amended charge was framed against Azlin and Ridzuan (charges marked “C1B1” and “D1B1” respectively”):

35 The sentences that were imposed on Ridzuan for the amended s 326 charge concerning Incident 4 (charge “D1B1”) and Azlin for the amended s 326 charge concerning Incident 2 (charge “C1B3”) are the subject of the appeals in CCA 24 and CCA 25 respectively.

36 On sentence, the Prosecution sought life imprisonment for Azlin for charge C1B3 and for Ridzuan for charge D1B1. This was because they had personally and respectively participated in scalding the Deceased in those incidents.

37 The Judge rejected the Prosecution’s primary position that Azlin and Ridzuan should be sentenced to life imprisonment. The Judge found that life imprisonment was not appropriate for either Azlin or Ridzuan because this was not the “worst case” of offending under s 326 of the Penal Code. In the Judge’s view, they did not “entirely comprehend the likelihood of death resulting from their actions” (GD at [191] to [192]).

38 The Judge sentenced Azlin to an aggregate sentence of 27 years’ imprisonment and an additional 12 months’ imprisonment in lieu of caning, and Ridzuan to an aggregate sentence of 27 years’ imprisonment and 24 strokes of the cane. The breakdown of the sentences is as follows, with the sentences for which the Prosecution sought life imprisonment emphasised in bold:

(a) For Azlin:

(b) For Ridzuan:

39 The Prosecution originally stated in its Notice of Appeal in CCA 17 that it was appealing against the Judge’s decision to acquit Azlin of the Murder Charge. The Prosecution also filed an appeal against the Judge’s decision to acquit Ridzuan of the Murder Charge against him (CA/CCA 18/2020 (“CCA 18”)). However, it subsequently withdrew its appeal in CCA 18 regarding Ridzuan’s Murder Charge. In the Prosecution’s Petition of Appeal in CCA 17, the Prosecution also confined its appeal to the Judge’s decision not to amend Azlin’s Murder Charge to the alternative s 300(c) charge. This position was subsequently confirmed in the Prosecution’s written appeal submissions.

40 In CCA 17, the Prosecution submits that the Judge erred in not amending Azlin’s Murder Charge to the alternative s 300(c) charge. The Prosecution’s key submissions may be summarised as follows.

(a) First, the pith of the Prosecution’s submission is that the Judge erred in her reading and application of Daniel Vijay. In particular, the Prosecution challenges the Judge’s conclusion that the Daniel Vijay test – the requirement for an intention to cause a s 300(c) injury – applies in a case such as the present (see [31] above). Rather, the Prosecution submits that, where only a single crime has been jointly committed and that happens to be murder under s 300(c), the secondary offender – in this context, Azlin – is equally responsible for the acts of the primary offender as long as those acts are jointly intended and there is no need to prove separately any specific intention to cause a s 300(c) injury.

(b) Second, the Prosecution also challenges the Judge’s finding that there had to be a common intention between Azlin and Ridzuan to commit all four scalding incidents (see [32] above). Instead, the Prosecution submits that there is no principled reason why Azlin cannot be held liable for s 300(c) murder through a combination of her direct liability for committing Incidents 1 and 3 and her constructive liability for jointly committing Incidents 2 and 4 with Ridzuan.

(c) Third, the Prosecution also challenges the Judge’s finding that “s 34 of the Penal Code does not enable the proof of common intention only of component offences of a ‘criminal act’” (see [31] above). The Prosecution submits that there is nothing in the text or purpose of s 34 that prevents s 34 from being employed in this manner.

41 In CCA 24 and CCA 25, the Prosecution submits that the Judge erred in not imposing life imprisonment on Ridzuan and Azlin for the amended D1B1 and C1B3 charges respectively. It contends that Azlin’s high culpability in Incident 2 and the aggravated nature of Ridzuan’s conduct in Incident 4 justify a sentence of life imprisonment for these charges. The Judge also failed to appreciate that the overall criminality and consequence of the actions of both respondents can and should be taken into consideration in sentencing. The Prosecution submits that the multiple aggravating factors here renders this case one of the worst types of offending under s 326 and life imprisonment ought therefore to be imposed.

42 For the appeal in CCA 17, Azlin submits that the alternative s 300(c) charge is not permissible because the Prosecution is required to prove a common intention specifically to cause a s 300(c) injury, and this has not been shown because Ridzuan was only involved in two of the four scalding incidents and there was no evidence that any of these incidents were sufficient in themselves to constitute a s 300(c) injury. Azlin also submits that the Prosecution must prove the existence of a common intention to cause the particular bodily injury in question (in this case, the Cumulative Scald Injury), and there is no evidence to suggest that Azlin had any such intention at any material time. Azlin and Ridzuan only had an intention to discipline the Deceased.

43 For the appeal against sentence in CCA 25, Azlin submits that life imprisonment is much “harsher” for a young offender like her, given that she is only 30 years’ old now. Azlin submits that the Judge correctly determined that Azlin did not entirely comprehend the likelihood of death that resulted from her actions, and that the Judge had correctly given due weight to the aggravating and mitigating factors. In sum, Azlin submits that her case is not one of the worst types of cases under s 326 of the Penal Code and her sentence of 27 years’ imprisonment and an additional 12 months’ imprisonment in lieu of caning is not manifestly inadequate.

44 In CCA 24, Ridzuan submits that the sentence of 27 years’ imprisonment and 24 strokes of the cane is a sufficiently heavy sentence. Life imprisonment is not appropriate in this case because he is not at any risk of re-offending in a similar manner. Charge D1B1 is also not the “worst example” of offending under s 326 because the death of the Deceased cannot be attributed solely to charge D1B1 and Ridzuan did not know how ill the Deceased was at the time he committed the acts that were captured in charge D1B1.

45 It is clear that Azlin committed Incidents 1 and 3 herself and that Azlin and Ridzuan both intended and carried out all or parts of Incidents 2 and 4. Indeed, the basis for Azlin’s convictions on charges C1B3 and C1B1, and Ridzuan’s convictions on charges D1B2 and D1B1, was that Azlin and Ridzuan commonly intended to commit Incidents 2 and 4 respectively, and neither Azlin nor Ridzuan has appealed against their convictions for those charges. Further, as we have already noted, Azlin and Ridzuan do not challenge the Judge’s findings that the Cumulative Scald Injury caused the Deceased’s death and was sufficient in the ordinary course of nature to cause death. This, in our judgment, is correct because the Judge reached her findings on this issue after a careful and appropriate analysis and evaluation of the evidence.

46 The Prosecution is not appealing against the Judge’s decision to acquit Azlin and Ridzuan of their Murder Charges. The Prosecution is also not appealing against the Judge’s decision not to amend Ridzuan’s Murder Charge to the alternative s 300(c) charge.

47 There are therefore three main issues in CCA 17.

(a) The first issue arises from the Prosecution’s submission that the Judge erred in finding that Azlin and Ridzuan both had to share a common intention specifically to inflict a s 300(c) injury in order for Azlin to be convicted of the alternative s 300(c) charge (see [40] above). The Judge applied the Daniel Vijay test to determine whether Azlin could be convicted of the alternative s 300(c) charge, even though Daniel Vijay concerned a “dual crime” situation in which murder under s 300(c) had been committed by one of a group of offenders as the collateral offence in the course of jointly setting out to commit a different offence. The present case does not concern such a “dual crime” scenario (as mentioned at [6] above). This raises the question of whether the test for constructive liability under s 34 of the Penal Code differs depending on whether the court is faced with a “dual crime” or a “single crime” scenario, and particularly whether this is so in the context of s 300(c) murder. Even more specifically, when s 300(c) murder is the only offence allegedly committed by joint offenders, is it necessary for the secondary offender, who is not the person who physically committed the criminal act that caused the death of the victim, to have commonly and specifically intended to cause a s 300(c) injury (meaning a bodily injury that is sufficient in the ordinary course of nature to cause death), or is it sufficient that the secondary offender only commonly intended to cause the actual injury inflicted? Although the Prosecution has devoted the bulk of its attention to this issue, and although we address this, we observe that this issue is not ultimately necessary to resolve the present appeals, as we explain below.

(b) The second issue pertains to the Judge’s findings on the requirements of the alternative s 300(c) charge. The Judge reasoned that the charge required Azlin and Ridzuan to have the common intention to commit all four scalding incidents if they were to be found to have intended to inflict a s 300(c) injury (GD at [121] and [124]; see [31] and [32] above). The question is whether the Judge was correct in this reasoning; if not, what are the actual requirements of the alternative s 300(c) charge?

(c) The final issue concerns the nature and scope of s 34 of the Penal Code and arises from the Judge’s reasoning at [121] of the GD (see [31] above). There are two sub-issues that flow from this.

(i) Regarding the actus reus of the alternative s 300(c) charge, can s 34 be used to attribute liability for component acts committed by another person (Incidents 2 and 4 committed by Ridzuan in this case) to the offender (Azlin) so as to aggregate those component acts with other acts personally committed by the offender (Incidents 1 and 3 committed by Azlin) to form a “larger” criminal act (the four scalding incidents) that is the actual basis of the offence charged (the alternative s 300(c) charge)? We shall refer to this as the “expanded interpretation” of s 34 in this judgment.

(ii) Regarding the mens rea for the alternative s 300(c) charge, if the Judge was wrong to find that the alternative s 300(c) charge requires Azlin to have commonly and specifically intended to inflict a s 300(c) injury by the four scalding incidents, then is the mens rea requirement for the alternative s 300(c) charge satisfied by the “aggregation” of Azlin’s direct intention to commit Incidents 1 and 3 with Azlin’s common intention with Ridzuan to commit Incidents 2 and 4?

48 To assist the court in its determination of these issues in CCA 17, we appointed Professor Goh Yihan SC (“Prof Goh”) as amicus curiae to address the following list of issues:

(a) “Question (i)”: Are the Judge’s comments at [121] of the GD an accurate view of the current state of the law on s 34 of the Penal Code?

(b) “Question (ii)”: If the answer to Question (i) is in the affirmative, can and should the interpretation of s 34 of the Penal Code be developed and expanded such that s 34 would allow the attribution of a specific act committed by a “principal offender” (ie, the person who directly committed the act) to a “secondary offender” (ie, the person who did not directly commit the act, but who participated in the criminal act and who commonly intended it), where only one crime is jointly committed by the principal and secondary offender (commonly referred to as a “single crime” situation, as in the present case)?

(c) “Question (iii)”: Is the alternative s 300(c) murder charge as proposed by the Prosecution permissible under the current state of the law on s 34 of the Penal Code?

(d) “Question (iv)”: If the answer to Question (iii) is in the negative and the answer to Question (ii) is in the affirmative, would the alternative s 300(c) murder charge be permissible under the expanded interpretation of s 34 of the Penal Code?

(e) “Question (v)”: Where a principal and secondary offender jointly commit a single offence of murder under s 300(c) of the Penal Code (ie, there is no other collateral offence committed), does the current state of the law require proof that the secondary offender intended specifically to inflict an injury that would be sufficient in the ordinary course of nature to cause death or is it sufficient that it be proved that the secondary offender intended to inflict the actual injury that was inflicted and separately that such injury was sufficient in the ordinary course of nature to cause death, as set out in Virsa Singh v State of Punjab AIR 1958 SC 465 (“Virsa Singh”)?

(f) “Question (vi)”: If the answer to Question (v) is the former, can and should the law on s 34 of the Penal Code be developed to cover the latter position?

49 We begin with a brief synopsis of Prof Goh’s views, which we will set out more fully at appropriate points in our analysis below. Regarding Questions (i) to (iv), Prof Goh submits that the Judge was correct to find that the current state of the law does not allow s 34 to be invoked in order to attribute liability for acts carried out by one offender to another offender such that, taken together with other acts that the latter has committed, the latter may be held liable for a broader “criminal act”. Prof Goh also submits that the interpretation of s 34 should not be expanded in this way. Azlin essentially aligns herself with Prof Goh’s submissions. On the other hand, the Prosecution submits that the expanded interpretation of s 34 is permissible even under the current state of the law and, in any event, that this is the correct interpretation of s 34.

50 Regarding Questions (v) and (vi), Prof Goh submits that, under the current law, the Daniel Vijay test does apply even when s 300(c) murder has been jointly committed as a single crime. On this issue, the respondents similarly support Prof Goh’s submissions. However, Prof Goh submits that the law should be developed such that the Daniel Vijay test should be departed from. On the other hand, the Prosecution submits that the current state of the law is that the Daniel Vijay test does not apply when s 300(c) murder is jointly committed as a single crime; Daniel Vijay only applies when s 300(c) murder is committed in a “dual crime” situation.

51 Prof Goh also submits that the Daniel Vijay test does not apply to the present case because the alternative s 300(c) charge is not a charge under s 300(c) read with s 34 of the Penal Code (a “s 300(c) common intention murder charge”) committed as a collateral offence in a dual crime scenario. Rather, Prof Goh submits that the mens rea test for the alternative s 300(c) charge is the test set out in Virsa Singh, which is an intention to cause the particular injury that had in fact been inflicted on the Deceased. Prof Goh submits that, in this case, that would be an intention to cause the Cumulative Scald Injury. However, Prof Goh questions whether the Prosecution has successfully proven such an intention on the facts of this case, though Prof Goh was not invited to and so did not analyse the evidence in detail. Azlin disagrees with Prof Goh’s submissions on this issue. Instead, Azlin supports the Judge’s finding that the Daniel Vijay test applies to this case. The Prosecution also hesitates to agree with Prof Goh that the Daniel Vijay test is not relevant to these appeals.

52 The statutorily prescribed minimum sentence for the alternative s 300(c) charges is life imprisonment. Hence, if the appeal in CCA 17 is allowed, the Prosecution’s appeal on sentence in CCA 25 would be moot. Consequently, we shall first address CCA 17 and the main issues outlined at [45] above in turn before we turn to the appeals in CCA 25 and 24 on Azlin’s and Ridzuan’s respective sentences.

53 The first issue concerns the Judge’s findings on the Daniel Vijay test at [121] and [124] of the GD, as highlighted at [31] and [32] above. There are two key points made by the Judge at [121] and [124] of the GD which are relevant here.

(a) Common intention to inflict s 300(c) injury: First, the Judge held that, “in order for Ridzuan’s acts to be attributed to Azlin for the purposes of liability under s 300(c) of the Penal Code, the common intention they needed to share would be the common intention to inflict s 300(c) injury” [emphasis added]. Since this common intention was not proved beyond reasonable doubt, the alternative s 300(c) charge was not made out (GD at [121]).

(b) Common intention to commit Incidents 1 to 4: Second, the Judge held that “[w]hat Daniel Vijay ... makes clear is that the unity of common intention must exist in relation to the ‘very criminal act’ for which the offender is charged” [emphasis added]. “In the present case, the physical components that led to the Cumulative Scald Injury were the collective result of the actions of both Azlin and Ridzuan.” Thus, “‘the very criminal act’ comprised four incidents, and its component parts were the actions resulting from two [actors], acting at different points in time.” The “logic of Daniel Vijay applied to require common intention in order to bind both these principals to the very criminal act of the offence which the four acts comprise” [emphasis added] (GD at [124]).

54 In sum, the Judge reasoned that, applying Daniel Vijay, for Azlin to be convicted of the alternative s 300(c) charge, there had to be (a) a common intention between Azlin and Ridzuan to cause a s 300(c) injury (that is, an injury that is sufficient in the ordinary course of nature to cause death) and (b) a common intention between Azlin and Ridzuan to commit Incidents 1 to 4.

55 In addressing Question (v) (see [48(e)] above), Prof Goh submits that, under the existing law, the Daniel Vijay test does apply even when s 300(c) murder is the only offence which has been committed. In other words, Prof Goh submits that the offender charged with a s 300(c) common intention murder charge must have intended specifically to inflict a s 300(c) injury (that is, an injury that would be sufficient in the ordinary course of nature to cause death), even when s 300(c) murder is the only offence that has been committed.

56 Prof Goh’s reasons for coming to this view may be summarised as follows.

(a) He submits that the relevant portions of the judgment in Daniel Vijay that set out the Daniel Vijay test do not clearly set out whether it applies to a “single crime” situation as well.

(b) Nevertheless, Prof Goh submits that, although this court’s reasoning in Daniel Vijay was undertaken in relation to a “dual crime” situation, the underlying reasons in support of that analysis apply equally to a “single crime” situation. This is because the primary reason underlying the Daniel Vijay test is that it would be unjust to hold a secondary offender constructively liable if that offender has “no intention to do the specific criminal act done by the actual [actor] which gave rise to the offence of s 300(c) murder” (Daniel Vijay at [76]).

(c) Prof Goh also cites Daniel Vijay (at [168(b)]) to infer that the Daniel Vijay test does apply to “single crime” situations:

Prof Goh submits that, in this “single crime” example, the court was quite clear that the common intention must be to cause a “s 300(c) injury” rather than a bodily injury that is later shown to be sufficient in the ordinary course of nature to cause death. Prof Goh highlights that the term “s 300(c) injury” is defined in Daniel Vijay at [146] as requiring “a common intention to cause death by the infliction of the specific injury which was in fact caused to the victim”. Thus, Prof Goh submits that the Daniel Vijay test applies to both “single crime” and “dual crime” situations.

57 That said, Prof Goh does not however maintain that the foregoing applies to the present case. On the contrary, he submits that the Daniel Vijay test does not apply to the alternative s 300(c) charge because the alternative s 300(c) charge is, on Prof Goh’s submission, not a s 300(c) common intention murder charge (that is, a charge under s 300(c) read with s 34 of the Penal Code). Rather, Prof Goh submits that the alternative s 300(c) charge is what he terms a s 300(c) charge “simpliciter”.

(a) Prof Goh submits that the alternative s 300(c) charge is not a s 300(c) common intention murder charge, as that charge is not based on a common intention between Azlin and Ridzuan to commit s 300(c) murder. If Prof Goh is correct on this submission, this would mean that the Judge was incorrect to apply Daniel Vijay to reach her findings summarised at [54] above, and the entire controversy surrounding the application of Daniel Vijay in these appeals would be moot, because that decision unquestionably concerned a s 300(c) common intention murder charge. We will consider this issue below.

(b) Prof Goh submits that the alternative s 300(c) charge is in truth a s 300(c) charge “simpliciter” against Azlin. It is not a joint crime situation at all where two or more participants shared the common intention to commit the offence concerned (in this case, s 300(c) murder), pursuant to which only one or more of them physically carried out the offence itself. Rather, the alternative s 300(c) charge explicitly alleges that only Azlin had the intention to commit s 300(c) murder and therefore that she alone is liable for s 300(c) murder. Thus, the alternative s 300(c) charge is not a charge under s 300(c) that is being read with s 34. Rather, the alternative s 300(c) charge is a charge under s 300(c) only. This is what Prof Goh means when he terms the charge a s 300(c) charge “simpliciter”. In the alternative s 300(c) charge, s 34 is only employed in an attempt to establish part of the actus reus of the offence (Incidents 2 and 4).

58 Prof Goh submits on this basis that the Prosecution does not need to prove a common intention between Azlin and Ridzuan to commit s 300(c) murder, or even a common intention to cause s 300(c) injury. To satisfy the alternative s 300(c) charge, the Prosecution only needs to satisfy the traditional requirements under s 300(c) murder. The well-established mens rea requirement under s 300(c) murder is the test as laid down in Virsa Singh: an intention of causing the bodily injury which was inflicted on the victim (see [71] below).

59 However, Prof Goh highlights that it is uncertain whether s 300(c) requires the Prosecution to prove that the accused person intended to inflict the particular injury which was in fact inflicted on the victim, or if it is sufficient that the Prosecution proves that the accused person intended to inflict only some bodily injury, and that the injury in fact inflicted was sufficient in the ordinary course of nature to cause death. Prof Goh highlights that our courts have generally adopted the former approach (see for instance Public Prosecutor v Lim Poh Lye and another [2005] 4 SLR(R) 582 (“Lim Poh Lye”) at [22] and [25]; Public Prosecutor v AFR [2011] 3 SLR 653). On the assumption that the former approach applies, Prof Goh submits that the Prosecution would need to prove that Azlin intended to inflict the particular injury which was in fact inflicted on the Deceased. Prof Goh submits that, in the present case, that would be the intention to cause the Cumulative Scald Injury.

60 However, Prof Goh submits that the Prosecution’s difficulty in this case lies not in having to prove a “common intention”, but rather in the fact that the present case is a situation where multiple injuries were inflicted over an extended period of time. Prof Goh refers to this as a “multiple acts situation”, and we shall adopt the same term for convenience. Prof Goh submits that it is unclear whether the Prosecution can prove that Azlin intended the particular injury caused – the Cumulative Scald Injury – in the present “multiple acts” situation.

(a) Prof Goh reasons that, in “most cases”, an intention to inflict the “particular injury” that was in fact inflicted on the victim in a multiple acts situation would refer to the cumulative injury that the victim suffered at the end of all the acts. In this case, that would mean that the Prosecution would have to prove that Azlin intended to inflict the Cumulative Scald Injury. The intention to inflict the cumulative injury may be proved either by (i) showing that the accused person intended to inflict each individual injury that made up the cumulative injury, or (ii) showing that the accused person intended to inflict the cumulative injury by way of a pre-arranged plan that was formed.

(b) It would be easier to prove that the accused person intended to inflict the cumulative injury by proving his intention to inflict each individual injury that made up the cumulative injury, if each individual injury is consistent in nature and/or the accused person is able to observe the outward deterioration in the victim’s condition. In such cases, the accused person would likely have treated each injury as accumulating in effect.

(c) Conversely, it may be more difficult to prove that the accused person intended to inflict the cumulative injury by proving his intention to inflict each individual injury that made up the cumulative injury, if each individual injury is inconsistent in nature and/or the accused person is not able to observe the outward deterioration in the victim’s condition. In these circumstances, the accused person would likely have treated each injury as isolated incidents that do not accumulate in effect.

(d) Prof Goh however does not arrive at a firm conclusion whether, on these facts, the Prosecution had proven beyond reasonable doubt that Azlin intended to inflict the Cumulative Scald Injury.

61 In sum, Prof Goh submits that, while the Daniel Vijay test applies to “single crime” cases when s 300(c) murder is the only offence that had been jointly committed by multiple offenders, the present case is not such a situation because the alternative s 300(c) charge only entails one offender – Azlin – who is charged with having committed s 300(c) murder; the s 300(c) murder under the alternative s 300(c) charge does not entail two or more offenders who had jointly committed the criminal act constituting the s 300(c) murder (that is, all four scalding incidents). Thus, the Daniel Vijay test does not apply because the Daniel Vijay test applies to s 300(c) common intention murder charges in which s 300(c) murder had been jointly committed by multiple offenders. Hence, Prof Goh submits that the mens rea test for the alternative s 300(c) charge is only an intention to cause the particular injury that had in fact been inflicted on the Deceased (the Cumulative Scald Injury), though Prof Goh questions whether the Prosecution has successfully proved such an intention on the facts of this case.

62 The Prosecution challenges both the main findings made by the Judge that are summarised at [54] above. The Prosecution submits that the Judge misunderstood Daniel Vijay as establishing a requirement that there must be a common intention to inflict a s 300(c) injury in all cases where an accused person is charged with a s 300(c) common intention murder charge. The Prosecution submits that, while the Judge appreciated the fundamental distinction between the present case (namely a “single crime” scenario) and the circumstances in Daniel Vijay (a “dual crime” scenario), the Judge erroneously characterised the distinction as just a factual difference rather than a conceptual distinction. Contrary to the Judge’s finding at [124] of the GD, the “logic of Daniel Vijay” is not that it enunciates any freestanding principle that the Prosecution will always have to prove that offenders charged with a s 300(c) common intention murder charge specifically intended to commit a s 300(c) injury. Instead, what Daniel Vijay introduced was an additional evidential requirement – the need to prove intent to cause a s 300(c) injury – for the secondary offender in “dual crime” cases when s 300(c) murder has been committed as the collateral offence. It is clear from Daniel Vijay at [42] that the imposition of such an additional mens rea requirement is specific to the particular context of a “dual crime” case, and is an evidential proxy devised to ensure that the secondary offender is only liable for the consequences of acts that had been specifically intended. This ensures that there is concurrence of moral blameworthiness and criminal responsibility in cases of constructive liability.

63 The Prosecution submits that the latter concern does not arise at all when s 300(c) murder is the only offence that has been jointly committed by different offenders. The Prosecution highlights that, in the context of s 300(c) murder, a “single crime” case is one where the accused persons share a common intention to commit the particular criminal act which caused death and which act was sufficient in the ordinary course of nature to cause death. This would include situations where:

(a) the accused persons act in concert in assaulting the deceased as joint or co-principals and their actions collectively caused death; or

(b) one of the accused persons aids the actual actor to perform the criminal acts intended by both of them, for example by handing the actual actor the murder weapon or by restraining the deceased, while the actual actor fatally assaults the deceased (as was the fact pattern in Public Prosecutor v Chia Kee Chen and another appeal [2018] 2 SLR 249 (“Chia Kee Chen”)).

64 The Prosecution submits that where s 300(c) murder is the only offence that has been jointly committed by multiple offenders, the concern in Daniel Vijay that the secondary offender will be unfairly held liable for an act that he did not himself do or did not specifically intend would not arise because all the offenders are acting in concert to inflict the particular injuries which form the actus reus of the s 300(c) murder charge. There is therefore no principled basis for departing from the general mens rea requirement for s 300(c) set out in Virsa Singh just because in a given setting that provision is being applied with s 34 of the Penal Code. The Prosecution submits that, “properly understood, all Daniel Vijay highlights is that an individual must intend the very criminal act he is being charged for.” Thus, all that is required is for the Prosecution to prove that Azlin intended the various scalding injuries that together constitute the actus reus of the offence under the alternative s 300(c) charge.

65 Next, on the Judge’s finding that Azlin and Ridzuan had to commonly intend to commit Incidents 1 to 4 (see [54] above), the Prosecution submits that the Judge erred in being “influenced” by observations made in Daniel Vijay. The Prosecution highlights that the scope of the requisite common intention is a distinct issue from the question whether an accused person can be held liable through a combination of direct and constructive liability. The Prosecution submits that there is no principled reason why this cannot be done. Azlin was the sole actor for Incidents 1 and 3, so there is no need to employ s 34 to hold her liable for those incidents. For Incidents 2 and 4, there is nothing in the language or object of s 34 that precludes the provision from being invoked as a basis for holding Azlin liable for the criminal acts done by Ridzuan so long as she intended those criminal acts (that is, Azlin intended that Ridzuan scald the Deceased with hot water on those two occasions). Such an approach does not lead to any injustice or improper extension of legal liability. In fact, it is entirely in line with the underlying rationale of s 34, which “embodies the commonsense principle that if two or more persons intentionally do a thing jointly it is just the same as if each of them had done it individually” (Ratanlal & Dhirajlal’s Law of Crimes: A Commentary on the Indian Penal Code 1860 vol 1 (CK Thakker & M C Thakker eds) (Bharat Law House, 27th Ed, 2013) (“Ratanlal”) at p 113). The Prosecution therefore submits that there is no need for there to be a common intention between Azlin and Ridzuan to commit all four scalding incidents.

66 Finally, regarding Prof Goh’s submission that the alternative s 300(c) charge is a s 300(c) murder charge simpliciter (see [57] above), the Prosecution accepts that the alternative s 300(c) charge is not the archetypal common intention charge in that it seeks to establish liability through a combination of direct and constructive liability. However, given that the alternative s 300(c) charge partially invokes s 34, the Prosecution submits that it may not be accurate to characterise the charge as a s 300(c) charge simpliciter and to ignore Daniel Vijay altogether. The Prosecution also notes that Daniel Vijay is undoubtedly relevant to these appeals because the Judge considered it in arriving at her conclusions, and the Prosecution is challenging some of those conclusions. The Prosecution nevertheless agrees with Prof Goh that the true legal question in relation to the actus reus is whether s 34 can operate within s 300(c) to affix liability on Azlin through a combination of direct liability (for Incidents 1 and 3) and constructive liability (for Incidents 2 and 4). As aforementioned, the Prosecution submits that there is nothing in the language or legislative purpose of s 34 that precludes the provision from being utilised in this manner.

67 Azlin purports to “agree” with Prof Goh that where the offence is committed by a combination of multiple acts, the Prosecution would have to prove that the accused person had intended to inflict the cumulative injury, and not just the last of the discrete injuries or even each of the discrete injuries. As the Prosecution’s case has been run on the basis that the Prosecution need only prove an intention to cause each of scalding incidents, and not the cumulative injury, Azlin submits that the alternative s 300(c) charge is “legally deficient”.

68 We observe that Azlin has misunderstood Prof Goh’s submissions on this point. Prof Goh does not contend that there is a particular burden on the Prosecution in all multiple acts situations. Rather, his submission is more nuanced: he submits that whether the mens rea requirement is satisfied in a multiple acts situation has to be assessed in the context of the facts of each case. Prof Goh’s submission is that, in “most cases”, the Prosecution may have to prove that the accused person intended to inflict the cumulative injury. In some cases, however, he accepts that it would be sufficient for the Prosecution to prove that the accused person intended to inflict the individual injuries (see [60(a)] to [60(c)] above).

69 However, Azlin disagrees with Prof Goh that the alternative s 300(c) charge is not a s 300(c) common intention murder charge and that the alternative s 300(c) charge does not require an intention to cause a s 300(c) injury. Instead, Azlin submits that, insofar as s 34 is being employed in the alternative s 300(c) charge to constructively attribute liability for a part of a criminal offence on Azlin, she should be considered a secondary offender, and the charge in question should be considered a “common intention” charge. Azlin highlights that this is in line with the Judge’s finding at [124] of the GD that, where there is no single actor for the whole criminal act, a finding of a common intention is necessary before constructive liability can be imposed (see [32] and [53(b)] above).

70 To properly understand the scope of the contested issues in this appeal, we first outline the established principles under s 300(c) of the Penal Code. Section 300(c) of the Penal Code provides:

71 The requirements of s 300(c) murder are well established: (a) death must have been caused by the acts of the accused person; (b) the bodily injury inflicted by those acts must be sufficient in the ordinary course of nature to cause death; and (c) the act resulting in bodily injury must have been done with the intention of causing that bodily injury that was inflicted on the victim: Wang Wenfeng v Public Prosecutor [2012] 4 SLR 590 at [32], endorsing Virsa Singh; see also Daniel Vijay at [167]. Element (b) is determined objectively while element (c) is subjective: see Kho Jabing and another v Public Prosecutor [2011] 3 SLR 634 (“Kho Jabing”) at [22]. We shall refer to element (c) as the “Virsa Singh test”.

72 As for element (c), it is important to bear in mind that the sole question under the Virsa Singh test is whether the accused person intended to inflict the specific bodily injury caused, and not whether the accused person intended to inflict a serious injury or an injury that is sufficient in the ordinary course of nature to cause death. This well-established requirement has been clearly explained in Virsa Singh at [27] and [32]:

73 Therefore, in Virsa Singh, the appellant was convicted of murder under s 300(c) of the Indian Penal Code 1860 (Act No 45 of 1860) because he intentionally thrust a spear into the abdomen of the deceased victim, and the medical evidence showed that the injury caused was sufficient in the ordinary course of nature to cause death. The Supreme Court of India rejected the appellant’s submission on appeal that the Prosecution had to prove an intention to inflict a bodily injury that was sufficient in the ordinary course of nature to cause death. As Bose J (delivering the judgment of the Supreme Court of India) put it (in Virsa Singh at [17]):

74 Bose J then explained that the mens rea test is a subjective intention to “cause the bodily injury that is found to be present” (in Virsa Singh at [19]):

75 It is critical to flesh out what the requirement “the bodily injury that is found to be present” means, as Prof Goh suggested that it is uncertain whether proof of s 300(c) murder requires proof of an intention to cause the particular injury that was in fact inflicted on the victim, or if it suffices that the accused person intended to cause any bodily injury (see [59] above). In our judgment, it is clear that the former is the proper test to be applied. In Virsa Singh itself, the Supreme Court of India had already phrased the requirement in these terms (at [24]):

76 Bose J further explained in Virsa Singh at [21] that:

77 Thus, on the facts of Virsa Singh, all the Prosecution was required to prove was that the appellant intended to stab the deceased victim’s abdomen with the spear. The appellant did not need to have intended all the specific consequences that flowed from the spear thrust (see Virsa Singh at [31]):

78 Consequently, the appellant’s conviction was upheld on appeal in Virsa Singh.

79 In our own caselaw, Lim Poh Lye is instructive. In that case, the accused persons intended to rob the victim, but ended up stabbing the victim as well. One of the stab wounds inflicted on the victim’s right leg was eight to ten centimetres in depth and was sufficient in the ordinary course of nature to cause death because it severed a major blood vessel, the right femoral vein, which caused uncontrolled and continuous bleeding that caused the victim’s death. The trial judge accepted that the accused persons intended to stab the victim, but found that the evidence was not clear as to which particular person had inflicted the fatal stab wound. The accused persons also did not know, at the time of the stabbing, that they would sever a major femoral vein. The trial judge relied on this fact to find that the severing of the victim’s femoral vein was not intentional. The trial judge thus convicted the accused persons of charges of robbery instead of the murder charges under s 300(c) read with s 34.

80 This court reversed the trial judge’s decision on the basis that s 300(c) murder did not require an intention on the accused persons’ part to cut the victim’s right femoral vein. Rather, all that was required was an intention to cause that stab wound to the victim (see Lim Poh Lye at [24]–[25]):

81 The emphasised portions of the foregoing extract from Lim Poh Lye spell out clearly that what is required under s 300(c) murder is that the accused person intended to cause the particular injury that was in fact inflicted on the victim, rather than any bodily injury. This was reiterated in Chia Kee Chen at [88]:

82 In addition, this court also emphasised in Lim Poh Lye at [41]–[47] that the accused does not need to intend all the specific consequences that flow from the injury that he intended to inflict. The relevant paragraphs of the judgment in Lim Poh Lye on this issue explain the point clearly and bears replicating in full:

83 Therefore, the accused in Lim Poh Lye did not need to intend that the stab wound would cut the victim’s right femoral vein, and so cause or bring about the effect of uncontrolled bleeding that in turn leads to death. This would amount to an intention to cause the consequences flowing from the injury, which is not required under the Virsa Singh test. Rather, all that is required under the Virsa Singh test is an intention to cause the stab wound to the victim’s right leg.

84 We next outline the requirements of s 34. That provision provides:

85 The core principles of s 34 were clarified by this court in Daniel Vijay, and were later reaffirmed by this court in Chia Kee Chen and Public Prosecutor v Aishamudin bin Jamaludin [2020] 2 SLR 769 (“Aishamudin”), though Aishamudin was not cited by the Judge. The general principles governing s 34 may be summarised as follows.

(a) Three elements are required to establish joint liability pursuant to s 34: (i) there must be a “criminal act” done by several persons (the criminal act element); (ii) that act must have been done “in furtherance of the common intention of all” (the common intention element); and (iii) the offender must have participated in the criminal act (the participation element): Daniel Vijay at [91]; Aishamudin at [49].

(b) As for the criminal act element, a “criminal act” has been interpreted to refer to “the aggregate of all the diverse acts done by the actual [actor] and the secondary offenders, which diverse acts collectively give rise to the offence or offences that the actual [actor] and the secondary offenders are charged with”: Daniel Vijay at [92]. In the decision of the Privy Council, delivered by Lord Sumner, in Barendra Kumar Ghosh v The King-Emperor (1925) 1 MLJ 543 (“Barendra”) at 552, 554–555 and 559, the Privy Council held that the term “a criminal act” means “that unity of criminal behaviour, which results in something, for which an individual would be punishable, if it were all done by himself alone, that is, in a criminal offence”:

(c) Based on the foregoing emphasised portion of that extract from Barendra, this court further explained in Lee Chez Kee v Public Prosecutor [2008] 3 SLR(R) 447 (“Lee Chez Kee”) at [137] that the term “criminal act” … “refers to all the acts done by the persons involved which cumulatively result in the criminal offence in question”.

(d) This was affirmed in Daniel Vijay at [95], where this court similarly cited the foregoing emphasised portion of the extract of Barendra to explain that:

(e) And in Aishamudin at [49(a)], we again endorsed the holding in Barendra that:

(f) It follows from the foregoing that a “criminal act” does not refer to the offence that the individuals concerned plan or carry out, but to an act or a continuum of acts: Aishamudin at [49(a)]. Thus, a single “criminal act” may involve and give rise to several “offences”: Aishamudin at [44], affirming Lee Chez Kee at [136].

(g) As for the common intention element, a common intention refers to a “common design” or plan, which might either have been pre-arranged or formed spontaneously at the scene of the criminal act: Aishamudin at [49(b)]; Lee Chez Kee at [158] and [161]. The common intention, strictly speaking, refers not to the intention to commit the offence which is the subject of the charge, but to the intention to do the “criminal act” (even if, in many cases, the two will overlap): Aishamudin at [49(b)].

(h) As for the participation element, participation may take many forms and degrees, and whether the element is satisfied is a question of fact. There is no requirement for an accused person to be physically present at the scene of the criminal act: Aishamudin at [49(c)], affirming Lee Chez Kee at [146].

(i) Section 34 is a principle of joint liability for the commission of a criminal act. Section 34 imposes constructive liability on joint offenders where the criminal act is done by one or more of them in furtherance of the common intention of all: see Daniel Vijay at [97]; Chia Kee Chen at [88].

(j) Section 34 does not create a substantive offence. Rather, it lays down a “principle of liability”: see Daniel Vijay at [75]; Aishamudin at [43]. It is inaccurate and potentially confusing to label s 34 as a “rule of evidence”: Aishamudin at [43]. The effect of s 34 “is to make an offender liable even for acts carried out by others pursuant to a shared common intention, as if those acts had been carried out by himself”: Aishamudin at [44].

86 To further understand the context underlying the issues in this appeal, it is also helpful to distinguish among the three types of situations where s 34 could potentially be employed. The first two situations – the “dual crime” and “single crime” scenarios – are well established in the caselaw, while the third situation is the relatively novel scenario that we are presently faced with. Somewhat surprisingly, the parties dispute the requirements of s 34 of the Penal Code not only in respect of the third situation that is facing this court, but also in respect of the “single crime” situation, as outlined above. As such, we will analyse the requirements of s 34 in respect of “single crime” and “dual crime” situations first before turning to the present circumstances.

87 We begin with the “dual crime” scenario, because the requirements of s 34 in respect of this scenario are not disputed among the parties and Prof Goh. This is where the offenders commonly intend to commit a “primary criminal act” but, in the course of carrying out that primary criminal act, one of the offenders – the “primary offender” – commits an additional “collateral criminal act”. The primary offender is the person who directly and physically committed the collateral criminal act. The question is whether the co-offenders can be held liable for the collateral criminal act. An illustration of this can be found in Daniel Vijay itself, where the original intention of the three offenders (Daniel, Christopher and Bala) was to commit robbery (the primary criminal act), but one of the three offenders (Bala) then committed murder under s 300(c) (the collateral criminal act) in the course of the robbery by hitting the victim repeatedly on his head and other parts of the body with a baseball bat, which led to injuries that caused the victim’s death and were sufficient in the ordinary course of nature to cause death.

88 In a typical “dual crime” case, the offenders’ liability for the primary criminal act (robbery in Daniel Vijay) is usually not at issue, because the offenders would have commonly intended to commit, and participated in the commission of, the primary criminal act. The primary offender’s liability for the collateral criminal act (s 300(c) murder in Daniel Vijay) is also usually not the key issue because the primary offender would have been the person who intended to and did commit the collateral criminal act. Indeed, the liability of the primary offender will typically be such that s 34 of the Penal Code need not be employed for that purpose. In the circumstances of Daniel Vijay, for example, the primary offender – the actual actor – of s 300(c) murder could have been charged with an offence of s 300(c) murder without it being read with s 34 of the Penal Code. This is what Prof Goh described as a s 300(c) murder charge “simpliciter”.

89 In such circumstances, in relation to the primary offender’s liability for s 300(c) murder, the Prosecution would typically need only to satisfy the established requirements of s 300(c) murder, as outlined at [71] above: namely that he intentionally caused the particular bodily injury that was inflicted on the victim; the bodily injury must have caused the victim’s death; and the bodily injury must be sufficient in the ordinary course of nature to cause death. The Daniel Vijay test, which applies in the context of extending liability for the collateral criminal act to the other co-offenders, prescribes an intention to cause an injury that is sufficient in the ordinary course of nature to cause death (or, in other words, a s 300(c) injury), but this is irrelevant when it comes to establishing the guilt of the primary offender in relation to s 300(c) murder. This much is uncontroversial, and was made explicit in Daniel Vijay at [167]:

90 As evident from that passage, the key issue that usually presents itself in “dual crime” cases is whether the secondary offender – that is, the offender who did not personally commit the collateral criminal act – can be held constructively liable for the collateral criminal act (committed by the primary offender) pursuant to s 34 of the Penal Code. Such liability would be “constructive” liability, rather than direct liability. In Daniel Vijay, for instance, Daniel and Christopher were not involved in hitting the victim with the baseball bat that resulted in the injuries which caused his death.

91 The critical question for the imposition of such constructive liability, in accordance with the text of s 34, is whether the collateral criminal act had been committed “in furtherance of” all the offenders’ “common intention” (see [84] above). The law on what is required to fulfil this test has developed considerably in our jurisprudence over the years.

92 The first significant case is the 1972 decision of this court in Wong Mimi and another v Public Prosecutor [1971–1973] SLR(R) 412 (“Mimi Wong”). While that case concerned a “single crime” situation where the co-accused persons had acted together to murder the victim, we highlight it here to provide context to the subsequent caselaw concerning “dual crime” situations, because Mimi Wong laid down the parameters of s 34 in Singapore, which were then considered in the subsequent cases. There, the second appellant had thrown detergent into the victim’s eyes before the first appellant stabbed the victim in her neck and abdomen, causing her to bleed to death. This court held that the intention of the actual actor (in that case, the first appellant who stabbed the victim) had to be distinguished from the common intention of all the offenders (Mimi Wong at [25]). The actual actor’s intention may or may not be identical with the common intention of all the offenders. Where the intention of the actual actor of the offence was not identical with the common intention of all the offenders, the test to determine whether the criminal act was done “in furtherance of” the parties’ “common intention” under s 34 is to determine whether the actual actor’s intention in carrying out the offence was “consistent with the carrying out of the common intention” [emphasis added]. If so, the criminal act done by the actual actor would be “in furtherance of the common intention” of the parties, such that the other offenders could thereby be constructively liable for the offence under s 34. On the facts of Mimi Wong, this court upheld the trial judge’s finding that the second appellant shared a common intention with the first appellant to cause bodily injury to the victim with a knife (Mimi Wong at [26]). Consequently, the convictions of both appellants of s 300(c) murder were upheld.

93 The next important decision is this court’s 2008 decision in Lee Chez Kee. That concerned a “dual crime” situation in which the common intention of three offenders was to rob the victim by tying him up and threatening him with a knife. However, one of the three offenders – the appellant – punched and stabbed the victim with a knife, and the victim later died. This court reaffirmed the approach in Mimi Wong and held in respect of the expression “in furtherance of the common intention” in s 34 that there was no need for the common intention of the parties to specifically be to commit the precise collateral criminal act in a “dual crime” situation.

94 However, in Lee Chez Kee we also held that an additional requirement was needed in order to impose constructive liability under s 34 on a secondary offender for a collateral criminal act committed by a primary offender. This was that the secondary offender had to “subjectively know that one in his party may likely commit” the collateral criminal act “in furtherance of the common intention of carrying out” the primary criminal act [emphasis in original in italics; emphasis added in bold italics] (Lee Chez Kee at [236] and [253(d)] (“Lee Chez Kee test”). On this basis, the majority of this court upheld the appellant’s conviction of murder under s 300(c) read with s 34 of the Penal Code, because the evidence showed that the appellant knew that either one of his co-offenders or he himself would have seriously harmed the deceased if the deceased had struggled or retaliated, and the appellant also appreciated the fact that the deceased would have to be killed to protect their identities in the light of the harm they had inflicted on him (Lee Chez Kee at [262]).

95 The third case in this series is our decision in Daniel Vijay. The brief facts of that case have already been summarised at [87] above. Essentially, three offenders – Bala, Daniel and Christopher – had set out to commit robbery, but Bala committed s 300(c) murder in the course of the robbery. As for Bala, this court dismissed his appeal against his conviction on s 300(c) murder, but amended his charge from one under s 300(c) read with s 34 of the Penal Code to a s 300(c) charge simpliciter, on the basis that he was the primary offender – the actual actor – of the s 300(c) murder, and was thus directly liable for s 300(c) murder (see [167] of Daniel Vijay extracted at [89] above).

96 As for Daniel and Christopher, this court allowed their appeals against their convictions on the offence of murder under s 300(c) read with s 34 of the Penal Code, and convicted them instead of the offence of robbery with hurt under s 394 read with s 34 of the Penal Code. This court made three important points on s 34.

(a) First, relying on Barendra, we held that the collateral criminal act done by the primary offender which resulted in the offence charged (s 300(c) murder committed by Bala) would only be considered to be done in furtherance of the common intention of all the offenders if that common intention included an intention to commit “the very criminal act” done by the actual actor (Daniel Vijay at [107], [143] and [166]). We refer to this as the “Barendra test”.

(b) Second, for the secondary offender (in that case Daniel and Christopher) to be constructively liable for the collateral criminal act (s 300(c) murder) pursuant to s 34 of the Penal Code, the secondary offender had to share a common intention with the primary offender to commit the collateral criminal act. The court thus departed from the Lee Chez Kee test such that it would no longer be sufficient that the “secondary offender” subjectively knew that one in their party might likely commit s 300(c) murder in furtherance of their common intention to commit the primary criminal act (robbery in that case) (Daniel Vijay at [87]).

(c) Third, where s 300(c) murder is the collateral criminal act committed by the primary offender in a “dual crime” situation, the person charged with the secondary offence of s 300(c) murder can only be held constructively liable for it, if he shared the common intention with the primary offender to cause a “s 300(c) injury” (that is, an injury which is sufficient in the ordinary course of nature to cause death). This is the Daniel Vijay test referenced in the introduction at [5] above. Thus, the Virsa Singh test for s 300(c) murder (see [71] above) does not apply to determine the liability of the secondary offender for s 300(c) murder committed by the primary offender as a collateral criminal act in a “dual crime” scenario.

97 As to the meaning of a “s 300(c) injury” under the Daniel Vijay test, Prof Goh highlights that three different interpretations of the term have emerged from the caselaw: (a) the specific injury that was actually inflicted on the deceased and that in fact caused his death (Public Prosecutor v Ellarry bin Puling and another [2011] SGHC 214; Chia Kee Chen); (b) an injury that is sufficiently serious that may result in an injury sufficient in the ordinary course of nature to cause death (Kho Jabing; Muhammad bin Kadar v Public Prosecutor [2011] 3 SLR 1205); or (c) an injury inflicted with the intention to cause death (GD at [97]).

98 We first note our agreement with Prof Goh that the present case is not a “dual crime” case. Azlin and Ridzuan did not commonly intend to commit some other primary criminal act in the course of which a collateral criminal act was committed by Ridzuan for which Azlin is sought to be made liable. Second, the difference between the three possible meanings of a “s 300(c) injury” under the Daniel Vijay test is slender and not likely to be material in most cases. Nevertheless, the judgment in Daniel Vijay uses the term “s 300(c) injury” to mean a “bodily injury which was sufficient in the ordinary course of nature to cause death”. This seems to us to be the second of the three meanings identified at [97] above though as will be seen later at [117(a)] and [117(b)], this can shade into or approach the third meaning. In any case, it was stated at [49] of the judgment in Daniel Vijay, and held at [145] that a secondary offender of s 300(c) murder in a “dual crime” situation must intend to cause a “s 300(c) injury”:

99 Therefore, in Daniel Vijay, when we held that an intention to inflict a s 300(c) injury must be established to make a secondary offender constructively liable for a s 300(c) murder committed by the primary offender, what was required was a common intention to cause a “bodily injury which was sufficient in the ordinary course of nature to cause death”. This was specifically contrasted with the Virsa Singh test in the extract from Daniel Vijay at [145] that we have reproduced in the previous paragraph. This was further clarified in Chia Kee Chen at [88] to mean that it must not merely be any type of bodily injury which was sufficient in the ordinary course of nature to cause death, but “the particular s 300(c) injury or injuries on the victim, the actual infliction of such injury being the criminal act which gives rise to the offence of s 300(c) murder” [emphasis added].

100 The foregoing principles represent the current state of the law on s 34 when applied to “dual crime” scenarios, and this much is undisputed among the parties and Prof Goh. However, as we have noted, the present case is not such a case. Consequently, the foregoing principles concerning s 34 in a “dual crime” scenario do not directly apply to the present case, though these principles provide the context against which we turn to the issues in this appeal.

101 We next consider the legal principles concerning s 34 when applied to the “single crime” scenario. This is when one criminal act is commonly intended by all the offenders, and it is carried out through a variety of different constituent parts by a variety of actors. Despite the number of constituent parts and actors, the criminal act as a whole only gives rise to one offence which all the offenders are charged with. Such “single crime” cases can present themselves in at least two possible configurations, as highlighted by the Prosecution at [63(a)] and [63(b)] above.

102 The first is where there are multiple offenders involved in the criminal venture, but only one of the offenders has directly committed the criminal act giving rise to the offence charged. A simple illustration of this is where A, B and C commonly intend to stab a victim, and, pursuant to this they agree that A will act as a lookout, B as the driver of the getaway vehicle, while C will carry out the act of stabbing the victim. The injuries caused by the stabbing are shown to be sufficient in the ordinary course of nature to cause death.

(a) In this scenario, C’s acts alone are sufficient to constitute the offence of s 300(c) murder, since he alone intentionally stabbed the victim, and the bodily injury caused by the stabbing was sufficient in the ordinary course of nature to cause death. Thus, C is directly liable for the offence of s 300(c) murder and can be charged with the offence of s 300(c) murder without reliance on s 34 of the Penal Code (see [88] above). This is consistent with our holding in Daniel Vijay at [167] (though in the context of Daniel Vijay, that was a “dual crime” case), extracted at [89] above.

(b) On the other hand, while A and B commonly intended to commit the criminal act of stabbing, neither of them directly committed the act of stabbing. A and B’s actions are not, strictly speaking, required to constitute the offence of s 300(c) murder; nor is their intention alone sufficient to make them liable for the acts of C. Consequently, to hold A and B liable for the offence of s 300(c) murder, the Prosecution would have to invoke s 34 of the Penal Code (as outlined at [85(a)] above) so that A and B may be made constructively liable for it. Thus, A and B would have to be charged with an offence under s 300(c) read with s 34 of the Penal Code. This is what Prof Goh terms a “s 300(c) common intention murder charge” (see [57] above).

103 The second possible configuration in a “single crime” case is where there are multiple offenders involved in the commission of the criminal act, and all the offenders’ actions are required to make out the offence that arises from the criminal act as a whole. This is unlike the first configuration above where the actions of a single actor suffice to make out the offence. A simple illustration of the second configuration is where A, B and C together punch and kick a victim pursuant to their common intention, and the victim dies as a result. Assume that the medical evidence shows that the collective punches and kicks of A, B and C are sufficient in the ordinary course of nature to cause death, while there is no evidence that any of the individual punches and kicks would have sufficed in itself.

104 In this example, none of the offenders – A, B or C – can be held liable for s 300(c) murder on their own because none of their acts would in themselves satisfy the elements of the offence. To hold A, B and C liable for the offence of s 300(c) murder, s 34 of the Penal Code would have to be employed such that each of them would have to be charged with an offence under s 300(c) read with s 34 of the Penal Code. In this way, each of them may be made both directly liable for their own actions and constructively liable for the acts of the other offenders in punching and kicking the victim pursuant to the common intention of all of them (see [71] above for the elements of s 300(c) murder).

105 In both configurations of “single crime” scenarios, the important common factor is that all the offenders (A, B and C) share the common intention to carry out the criminal act which is then committed by the various actors.

106 The mens rea test for the offence of s 300(c) murder is that the offender must have intended to cause the particular injury that was inflicted on the deceased victim (the Virsa Singh test: see [72] to [83] above). However, the Judge held that the Daniel Vijay test applied in the present case such that, for Azlin and Ridzuan to be convicted of the Murder Charges, Azlin and Ridzuan had to have commonly intended to cause a s 300(c) injury (meaning an injury that is sufficient in the ordinary course of nature to cause death) (see [29] above). In line with this, the Judge also reasoned that, to be convicted of the alternative s 300(c) charge, Azlin had to have commonly intended to cause s 300(c) injury, and not just the particular injury (the Cumulative Scald Injury) that was actually inflicted on the Deceased (see [31] above). As highlighted at [40] above, the Prosecution challenges this finding in these appeals.

107 Consequently, the question that is presented is what is the applicable mens rea where s 300(c) murder is jointly committed as a single crime pursuant to the common intention of multiple offenders? Is it the classical test in Virsa Singh? After all, why should the mens rea be different just because the act is carried out by several people and not just by one? Or is it the Daniel Vijay test? And is there a difference between the two configurations of the “single crime” scenario? Prof Goh notes that there is some degree of confusion and uncertainty over whether Daniel Vijay extends to single crime scenarios. The Prosecution similarly submits that “clarity is needed on this issue”.

108 The Prosecution submits that the Judge erred in finding that the Daniel Vijay test applies in this context, and it contends that the Virsa Singh test should instead apply. Prof Goh on the other hand submits that the current state of the law is such that the Daniel Vijay test applies, and Azlin essentially aligns herself with this aspect of Prof Goh’s submission. Prof Goh, however, submits that, the Daniel Vijay test should be abandoned and that the Lee Chez Kee test (defined at [94] above) should apply instead.

109 In our judgment, the Judge erred in finding that the Daniel Vijay test prescribes the applicable mens rea test to determine if an offender charged with a s 300(c) common intention murder charge can be held constructively liable for s 300(c) murder in a “single crime” setting. We agree with the Prosecution that the Daniel Vijay test is confined to a “dual crime” situation (see [62] above). As already explained at [87] above, Daniel Vijay was a “dual crime” case and this court did not purport to lay down any test in Daniel Vijay that was meant to apply to a “single crime” situation when s 300(c) murder is the only offence that was commonly intended and then committed by co-offenders. A careful reading of Daniel Vijay would show that it only dealt with and changed the law on the s 34 requirements for a secondary offender of s 300(c) murder when it is committed by a primary offender as a collateral criminal act in a “dual crime” situation.

110 The clearest indication of this is in [41] of that judgment, where this court explained that, in a “single crime” situation, there is no controversy about the requirement of common intention because the co-offenders would have commonly intended to commit the criminal act:

111 After setting out the foregoing “dual crime” context and explaining the terms “actual doer” and “secondary offender” (at [41]), this court went on to lay down the Daniel Vijay test at [74] to [76] and [145] to [146]. The significance of the Daniel Vijay test was that it departed from the Lee Chez Kee test that it would be sufficient for the secondary offender to “subjectively know that one in his party may likely commit the criminal act constituting the collateral offence” [emphasis in original omitted] (see [94] above). As the Prosecution rightly note, this was meant to address the potential injustice of holding a secondary offender liable for a collateral offence in a “dual crime” situation which he did not intend, as explained by this court in Daniel Vijay at [76]; this has no application to “single crime” cases of s 300(c) common intention murder charges:

112 Furthermore, we disagree with Prof Goh that the hypothetical example highlighted at [168(b)] of Daniel Vijay indicates that the Daniel Vijay test applies to “single crime” cases of s 300(c) murder (see [56(c)] above). That example explicitly assumes that the joint offenders shared a common intention to cause a s 300(c) injury and so it cannot stand for the proposition that this is a pre-requisite in a “single crime” case.

113 It follows that the different circumstances between “dual crime” and “single crime” scenarios will have to be firmly borne in mind because these are material in developing and applying the correct analytical framework. To summarise what we have set out thus far:

(a) In a “dual crime” scenario, the question is whether a secondary offender, who did not intend to commit the collateral criminal act (committed by the primary offender), should nevertheless be constructively liable for it.

(b) The “single crime” scenario is fundamentally different from the “dual crime” scenario because, in both possible configurations of the “single crime” scenario as outlined at [102] to [105] above, all the offenders would have commonly intended to commit the criminal act that has in fact been committed, as emphasised at [105] above. For instance, in the first configuration where the acts of just one of the co-offenders would suffice to make out the elements of the s 300(c) murder offence charged, the other offenders (A and B) are nonetheless liable because the acts were carried out pursuant to the intention that they also shared, to cause the victim the stabbing injuries in question (see [102(b)] above). And in the second configuration in which all the co-offenders’ separate actions are required to make out the elements of the s 300(c) murder offence charged, it is also clear that all the offenders in that example would have commonly intended the collective injuries caused to the victim. We therefore agree with the Prosecution that the underlying premise and logic of Daniel Vijay (extracted at [111] above) – that it might be potentially unjust to hold a secondary offender liable for a collateral offence in a “dual crime” situation which he did not intend – simply does not apply in “single crime” cases. And where s 300(c) murder has been committed as a single crime, it does not make any sense to apply the Daniel Vijay test, even if the offender has been charged with a s 300(c) common intention murder charge, because the Daniel Vijay test imposes a stricter mens rea test that was developed for the very different “dual crime” situation. This is consistent with the position taken in Stanley Yeo, Neil Morgan & Chan Wing Cheong, Criminal Law in Malaysia and Singapore (LexisNexis, 3rd Ed, 2018) at para 35.33:

114 The inappropriateness of applying a stricter mens rea test to an offender charged with a s 300(c) common intention murder charge in a “single crime” scenario as compared to an offender charged with a s 300(c) murder simpliciter charge becomes even more evident when it is recalled that the fundamental purpose of s 34 is to deter group crimes. This is done by expanding, rather than restricting, the scope of liability of those who commonly intend and participate in group crimes beyond the specific actions personally committed by the offender (see also Ratanlal at 111, extracted at [160] below). Thus, as explained by this court in Aishamudin at [44] (as highlighted at [85(j)] above), the entire purpose of s 34 “is to make an offender liable even for acts carried out by others pursuant to a shared common intention, as if those acts had been carried out by himself” [emphasis added]. On this basis, it would undermine the purpose of s 34 if the term “in furtherance of the common intention of all” in that provision were to be interpreted such that, even when only a single criminal act is commonly intended by multiple offenders, a stricter mens rea test were imposed to determine whether the offender who did not personally commit the criminal act can be constructively liable for the offence arising from that criminal act.

115 Both Prof Goh and the Judge relied on Chia Kee Chen for the position that the Daniel Vijay test applies even when s 300(c) murder is committed as a single crime (see GD at [100] to [102]). With respect, we disagree. While in Chia Kee Chen we did cite the Daniel Vijay test (see Chia Kee Chen (at [46]), and Chia Kee Chen did involve s 300(c) murder being the only crime that was jointly committed, that does not mean that in Chia Kee Chen we had endorsed the application of the Daniel Vijay test in the “single crime” scenario. The issue in Chia Kee Chen arose in the context of determining whether the accused person could be held responsible for the mortal blow that was inflicted on the victim, if these had been administered by those he had recruited for the purpose of attacking the victim. The court’s analysis on the common intention was entirely focused on whether he intended to inflict the specific injury inflicted (the craniofacial injuries) such that he should be held responsible for it, not whether he had intended to inflict a s 300(c) injury (see Chia Kee Chen at [90] to [95]).

116 This is also consistent with our caselaw, which is to the effect that when murder under s 300(c) is the only offence that has been committed by joint offenders, the secondary offender need not know or intend to cause a bodily injury of such seriousness that it would be sufficient in the ordinary course of nature cause death. The best example of this is our decision in Lim Poh Lye, as explained at [79] to [83] above. To reiterate, Lim Poh Lye stands for the important principle that, even in “single crime” cases of s 300(c) common intention murder charges, the accused person does not need to intend to cause the specific consequences that flow from the injury that was actually inflicted on the victim. It follows that, in such “single crime” scenarios, it should also not be required that the accused person intended to cause a bodily injury that is sufficient in the ordinary course of nature to cause death, as that would, in substance, be tantamount to requiring an intention to inflict the consequences of the injury. This is consistent with Virsa Singh at [27] and [32], highlighted at [72] above. While Lim Poh Lye pre-dated Lee Chez Kee and Daniel Vijay, the latter two cases concern the ambit of s 34 of the Penal Code when s 300(c) murder has been committed in a “dual crime” situation, and the important finding in Lim Poh Lye regarding the necessary mens rea for s 300(c) murder when it is a single crime committed by joint offenders remains applicable today.

117 We also agree with the Prosecution that there are sound reasons why the Daniel Vijay test should not apply when s 300(c) murder is the only crime that has been jointly committed by co-offenders.

(a) Applying the Daniel Vijay test to s 300(c) murder when it is a single crime that has been jointly committed would result in the perverse outcome that concerted group attacks that cause fatal injuries would impose a higher burden on the Prosecution and so would often attract less serious charges than the very same attack involving a single individual. This is because it is uncontroversial that an intention to cause a s 300(c) injury is an intention with a higher threshold to prove because it is “substantially the same as a common intention to cause death” (Daniel Vijay at [146]).

(b) The Daniel Vijay test effectively conflates the mens rea for s 300(c) with that of s 300(a) of the Penal Code, and severely undermines the purpose and intent of s 300(c), which is that “[n]o one has a licence to run around inflicting injuries that are sufficient to cause death in the ordinary course of nature and claim that they are not guilty of murder” (Virsa Singh at [27]).

(c) Section 34 “embodies the commonsense principle that if two or more persons intentionally do a thing jointly it is just the same as if each of them had done it individually”: see Ratanlal at 113. By requiring an intention to cause s 300(c) injury rather than the actual injury inflicted, the Daniel Vijay test effectively raises the mens rea requirement for s 300(c) murder, even though the offender may well be equally, if not more, culpable than the person who physically committed the s 300(c) murder.

(d) The seeming concern in requiring only an intention to cause the actual injury inflicted, rather than a s 300(c) injury, stems not from s 34 but from the terms of s 300(c) itself, in that that imposes criminal liability for murder without an intention that is “substantially the same as a common intention to cause death”. However, that is the position explicitly laid down in the text of s 300(c). As the Penal Code currently stands, there is no principled reason to read into the text of either s 300(c) or s 34 a requirement that the Daniel Vijay test will apply in circumstances where s 300(c) has been jointly committed as a single crime.

118 And as we have already noted, we consider that this is so even under the existing law. In such circumstances, the Virsa Singh test applies.

119 Prof Goh submits in any event that we should depart from the Daniel Vijay test and revert to the Lee Chez Kee test instead, such that it would be sufficient for the secondary offender to “subjectively know that one in his party may likely commit the criminal act constituting the collateral offence” [emphasis in original omitted]. Prof Goh emphasises that s 34 states that the criminal act is done “in furtherance of the common intention of all” [emphasis added], rather than simply “with the common intention of all”. Therefore, for a criminal act to be done “in furtherance of” the offenders’ common intention, there should not be a need to specifically intend the criminal act making up the collateral offence. Rather, Prof Goh argues, the Lee Chez Kee threshold of a subjective awareness that the criminal act might likely occur should suffice. Prof Goh also submits that the Lee Chez Kee test is more consistent with the historical genesis of s 34. Prof Goh appears to take the position that this should be so not only for “single crime” situations but also for “dual crime” situations where s 300(c) murder is committed as a collateral offence. The Prosecution aligns itself with Prof Goh’s position on this.

120 We have already held that the current state of the law on s 34 and its relationship with s 300(c) murder is such that, for “single crime” cases, the mens rea is the same as that which applies where an offender is charged with a s 300(c) murder simpliciter charge, namely, it is the well-established Virsa Singh test. Therefore, to the extent that Prof Goh is submitting that the Lee Chez Kee test should apply when s 300(c) murder has been jointly committed as a single crime, we disagree for the reasons we have set out above. As for whether the Lee Chez Kee test should apply when s 300(c) murder has been committed as a collateral offence in a “dual crime” scenario, we leave this question for determination in an appropriate case in the future, because the present case does not present such a scenario, as explained at [98] above, and so this issue is not relevant for the present appeals.

121 In sum, when s 300(c) murder is the sole offence that has been jointly committed by co-offenders, the following principles apply.

(a) The elements of the offence for the offender charged with a s 300(c) murder charge simpliciter (meaning a charge that does not employ s 34 of the Penal Code) are the three well-established requirements for s 300(c) murder outlined at [71] above. The mens rea is that embodied in the Virsa Singh test, so that it need only to be established that the offender intended to inflict the particular injury that was actually inflicted on the deceased victim, and it need not be shown that he did so intending or even knowing that it was sufficient to cause death.

(b) The elements of the offence where the offender is charged with a s 300(c) common intention murder charge in a “single crime” setting are the same three well-established requirements for s 300(c) murder outlined at [71] above. In addition, the offender must satisfy the three elements required to establish joint liability pursuant to s 34 (as outlined at [85(a)] above): the criminal act element, the common intention element, and the participation element. The common intention element is satisfied by the Virsa Singh test, not the Daniel Vijay test.

122 As such, we provide our answers to Questions (v) and (vi) (highlighted at [48] above) as follows.

(a) We answer Question (v) in the negative. Where multiple offenders jointly commit a single offence of s 300(c) murder, the current state of the law is such that the Daniel Vijay test does not apply, and there is no need for the offender who is charged with a s 300(c) common intention murder charge to have intended to inflict an injury that would be sufficient in the ordinary course of nature to cause death. Instead, the Virsa Singh test applies such that it is sufficient that the said offender intended to cause the actual injury that was inflicted on the victim.

(b) It follows that Question (vi) is moot, because that concerns whether the law on s 34 of the Penal Code should be developed if the current state of the law is such that the Daniel Vijay test applies when co-offenders jointly commit a single offence of s 300(c) murder.

123 We have summarised the applicable legal principles concerning s 34 of the Penal Code when applied to “dual crime” and “single crime” scenarios, particularly when the offence of s 300(c) murder has been committed in both scenarios. We end this section by returning to the present case. This case does not neatly fit into the fact pattern of the typical “single crime” case, although there is only one charge in the present case which has allegedly been committed by Azlin (the alternative s 300(c) charge). This is because Azlin and Ridzuan did not commonly intend to commit the entire criminal act which is the subject matter of the alternative s 300(c) charge (all four scalding incidents). Rather, Ridzuan only shared a common intention to commit a part of that criminal act (Incidents 2 and 4). The present case thus presents us with a third type of situation to which s 34 could potentially be applied and it is to this we now turn.

124 The third type of situation where s 34 might potentially be applicable is the fact pattern that we face in these appeals. This is where there is a variety of acts committed by multiple offenders, and each act could potentially form a distinct offence because the offenders’ intentions in respect of the aggregate of the acts may be different even if they might share the intention to commit some of the acts. Moreover, these acts, when aggregated, potentially form a different offence.

125 A simple example concerning the offence of possession of controlled drugs for the purpose of trafficking under s 5(1)(a) read with s 5(2) of the Misuse of Drugs Act 1973 (2020 Rev Ed) (“MDA”), and ignoring s 18(4) of the MDA for the moment, will help illustrate the point. The elements of this offence include possession of the drug, knowledge of the nature of the drug, and the requirement that the drug was possessed for the purpose of trafficking: Masoud Rahimi bin Mehrzad v Public Prosecutor and another appeal [2017] 1 SLR 257 at [28]. A and B jointly decide to purchase some drugs. A thinks this is for their own consumption while B intends to traffic in a portion of the drugs. A takes possession and is arrested before B has obtained his share. A testifies that he took possession of the drugs intending to meet B and then consume them with B. B admits that unbeknownst to A, B intended to sell some of the drugs to C. There is no dispute that A was in possession of the drugs for both of them pursuant to their common intention. The question is whether A’s physical possession of the drugs can be attributed to B by virtue of s 34 so that B can be made liable for the offence of drug possession for the purpose of trafficking.

126 Such a case does not neatly fall within either of the archetypal “single crime” or “dual crime” scenarios.

(a) It is not the conventional “single crime” scenario because there is more than one crime involved. The acts intended by A and B are somewhat different and give rise to different offences. A did not share B’s intention to sell the drugs to C. While they both intended A to possess the drugs, they had different intentions as to what they were going to do with the drugs. And this gives rise not to one offence but potentially to two.

(b) However, it is also not the conventional “dual crime” situation because the acts of B are not collateral to the primary act of A. Rather, those are acts that B alone intended from the outset. It does not seem possible or satisfactory to suggest that A can be held liable for B’s acts by virtue of s 34. And the real question is not whether A, a secondary offender (who did not personally commit the “collateral” offence) can be held liable for it, but whether B, can be held liable for the offence that B intended, even though part of the offence was not committed by B but by another offender pursuant to their common intention. That is the present scenario facing this court: the question is not whether, having commonly intended to commit Incidents 2 and 4, Ridzuan can also be held liable for Incidents 1 and 3 that were committed by Azlin. Rather, the question is whether Azlin can be held liable for all four scalding incidents which she intended to cause, even though some of the incidents – Incidents 2 and 4 – were committed by Ridzuan and not by her, albeit pursuant to their common intention.

127 In that light, we turn to the other major part of Prof Goh’s submission, which is that, on the basis of the alternative s 300(c) charge being a s 300(c) murder charge simpliciter, the Daniel Vijay test does not apply in the present case (see [61] above). Azlin disagrees (see [69] above), while the Prosecution also hesitates to agree principally because it contends that s 34 continues to be relevant and therefore, it does not accept this is a s 300(c) murder charger simpliciter (see [66] above).

128 We agree with Prof Goh that the alternative s 300(c) charge is not a s 300(c) common intention murder charge. The text of the alternative s 300(c) charge stipulates that, by committing Incidents 1 and 3 herself and by committing Incidents 2 and 4 “together with Ridzuan … and in furtherance of the common intention of [them] both”, Azlin had “thereby committed an offence under s 300(c) read with s 34 in respect of [Incidents 2 and 4], and punishable under s 302(2) of the Penal Code” [emphasis added] (see [3] above).

129 As such, as Prof Goh rightly points out, s 34 is only being employed in this case to satisfy part of the criminal act forming the actus reus of s 300(c) murder – namely the commission of Incidents 2 and 4. This is not how s 34 is conventionally used. When s 34 is used in that conventional sense, all the offenders are liable for all the elements of the offence once the requirements of s 34 are satisfied (these being the criminal act element, the common intention element, and the participation element: see [85(a)] above). Each offender may be liable in such circumstances as if all the acts were done by that offender even if some or all of the acts were in fact done by another.

130 Thus, for instance, in the hypothetical example at [102] above, A and B may be held constructively liable for the offence of s 300(c) murder, even though they did not personally carry out the acts of stabbing of the victim, as long as A and B satisfy the requirements of s 34 in that case. There is no need for A and B to have been the ones to cause the particular injury inflicted on the victim (the actus reus of the offence of s 300(c) murder).

131 The Daniel Vijay test (that there must be a common intention to cause s 300(c) injury) is a test going to the common intention element of s 34 to ensure that the secondary offender who is charged with the collateral offence of s 300(c) murder satisfies the requirement under s 34 that murder was done “in furtherance of the common intention” including of the secondary offender. In this case, the alternative s 300(c) charge does not even allege that the entire criminal act forming the basis of the charge – Incidents 1 to 4 – were done by several persons in furtherance of their common intention.

132 The Judge’s view that, to enable the Prosecution to invoke s 34, Azlin had to share a common intention with Ridzuan to commit all four scalding incidents is, with respect, mistaken because it does not follow from what the alternative s 300(c) charge itself requires. Instead, the charge only seeks to employ s 34 to attribute liability for Incidents 2 and 4 to Azlin. The Judge also erred when she held that there had to be a common intention between Azlin and Ridzuan to inflict a s 300(c) injury. In fairness to the Judge, it should be noted that this reasoning seemed to us to stem from her view of how s 34 could be invoked, and given that the charge mentions s 34 in some parts, the Judge seemed to think that, by virtue of that reference, the aforementioned consequences would flow.

133 Based on the elements of the alternative s 300(c) charge and the requirements to establish murder under s 300(c), the Prosecution would have to establish the following.

(a) The first element relates to the cause of death – death must have been caused by Azlin as a result of Incidents 1 to 4. However, Azlin did not personally commit all the acts of scalding in Incidents 2 and 4. The question then is whether Ridzuan’s acts in Incidents 2 and 4 can be attributed to Azlin pursuant to s 34. To determine this question, Azlin would have to satisfy the requirements of s 34 (the participation, criminal act, and common intention elements) in relation to Incidents 2 and 4. Specifically, the question may be framed thus: was there a criminal act (Incidents 2 and 4) done by several persons (Azlin and Ridzuan) in furtherance of their common intention, and did Azlin participate in that criminal act? If Azlin can be liable for Incidents 2 and 4 pursuant to s 34, the next question is whether the aggregation of Azlin’s direct liability for Incidents 1 and 3 with Azlin’s constructive liability for Incidents 2 and 4 would amount to the commission of all four scalding incidents to cause the Cumulative Scald Injury.

(b) The second element relates to the intention to cause the injury, which is a subjective inquiry pursuant to the well-established test laid down in Virsa Singh – did Azlin intend to cause the specific injury that was in fact inflicted on the Deceased, which is the Cumulative Scald Injury? This enquiry requires the court to determine if the aggregation of Azlin’s intention to commit Incidents 1 and 3 with her intention to commit Incidents 2 and 4 would amount to an intention to cause the Cumulative Scald Injury.

(c) The third element relates to the consequences of the injury, which is an objective inquiry – was the bodily injury inflicted by Incidents 1 to 4 – which is the Cumulative Scald Injury – sufficient in the ordinary course of nature to cause death. It is undisputed that this requirement is satisfied in this case.

134 We therefore agree with Prof Goh that the Daniel Vijay test is irrelevant to the alternative s 300(c) charge. The remaining question is whether s 34 can be employed in the manner described at [129] and [133(a)] above: can s 34 be employed to attribute liability for component acts committed by another person (in this case Incidents 2 and 4 that were committed by Ridzuan) to the offender (Azlin) so as to aggregate those component acts with other acts personally committed by the offender (Incidents 1 and 3) to form a “larger” criminal act (the four scalding incidents) that is the actual basis of the offence charged (the alternative s 300(c) charge) (what we have referred to as the “expanded interpretation” of s 34 at [47(c)(i)] above)? It is this question to which we now turn.

135 The first question is whether this “expanded interpretation” of s 34 represents the current law on that provision. To recapitulate, the Judge held at [121] of the GD that the expanded interpretation is not permissible under the current state of the law on s 34 because s 34 “is not a free-standing principle of attribution” and “does not enable the proof of common intention only of component offences of a ‘criminal act’” (see [31] above). The Prosecution submits that the Judge erred in this regard and that there is nothing in principle that prevents the “expanded interpretation” of s 34. Prof Goh submits that the Judge is correct in this finding, and Azlin aligns herself with Prof Goh’s submissions.

136 Prof Goh’s submissions on this issue may be summarised as follows.

(a) Text of s 34:

(i) Section 34 provides for a “criminal act” as opposed to merely an “act” [emphasis added], so the term “criminal act” under s 34 cannot simply be a reference to the individual acts that cumulatively form the “criminal act” that is the subject matter of the charge. Rather, Prof Goh submits that an act would only be “criminal” if the act, “with the requisite mens rea, is an offence under the Penal Code or other written law”.

(ii) Prof Goh also placed emphasis on the holding in Barendra that a “criminal act” means “that unity of criminal behaviour, which results in something, for which an individual would be punishable, if it were all done by himself alone, that is, in a criminal offence” [emphasis added] (see [85(b)] above). Prof Goh submits that, in the context of s 34, that “something” must refer to an offence that is fully constituted from the criminal act and the common intention. Prof Goh also highlights this court’s finding in Lee Chez Kee at [137] that the term “criminal act” “refers to all the acts done by the persons involved which cumulatively result in the criminal offence in question” [emphasis added]. Therefore, Prof Goh submits, a “criminal act” cannot be a component part of a larger “criminal act” forming the basis of an offence. Rather, a “criminal act” must fully constitute the actus reus of the offence charged.

(iii) Prof Goh also submits that the term “liable for that act” in s 34 means “liable to be punished for ... an offence that is fully constituted by the act and the common intention”. This is because, in almost all the instances that the word “liable” is used in the Penal Code, “it is used in the sense of being liable for punishment”. One example highlighted by Prof Goh is s 53 of the Penal Code, which provides that “[t]he punishments to which offenders are liable under the provisions of this Code are (a) death; (b) imprisonment; (c) forfeiture of property; (d) fine; (e) caning”. Thus, in the context of the expression “liable for that act”, Prof Goh submits that an act attracts punishment only when it fully satisfies the actus reus of an offence prescribed by the Penal Code.

(b) Purpose of s 34: Prof Goh’s next submission is premised on a purposive interpretation of s 34. Prof Goh highlights that this court pointed out in Lee Chez Kee at [194] that s 34 of the Penal Code was amended in 1870 – when the clause “in furtherance of the common intention of all” was added – in order to bring the concept of complicity under s 34 in line with the pre-existing English doctrine of common purpose. The doctrine of common purpose was concerned with the question of whether A can be held liable for B’s further collateral offence (in “dual crime” situations), not whether A can be held liable for B’s acts so as to satisfy the elements of a different offence.

(c) Theoretical foundation of s 34: Prof Goh next submits that the theoretical explanations for the doctrine of common purpose do not support the view that it may be used to impute secondary liability for specific acts insofar as they constitute components of a larger “criminal act”. For instance, equivalence theories, which seek an overall equivalence of culpability between the primary and secondary offender, will only make sense if secondary liability is attached for an offence that emanates from a common purpose. A second major theory to explain the complicity doctrine is to regard the perpetrators of crimes as agents of accessories. Prof Goh submits that this theory only makes sense if the act concerned fully constitutes an offence, because the theory is based on the agent being authorised to carry out an offence on behalf of the accessory, such that both are liable in respect of the offence.

(d) Concurrence principle: Prof Goh also submits that the expanded interpretation of s 34 would offend the fundamental principle that there must be a concurrence of actus reus and mens rea for any offence (see Wang Wenfeng at [45]). This is because the use of s 34 in the way urged by the Prosecution would permit it to attribute constructive liability for acts done pursuant to a common intention, to be used to satisfy the actus reus of another offence, but for which the mens rea may not be satisfied by that common intention.

(e) Indian caselaw: Prof Goh next submits that the expanded interpretation of s 34 would be inconsistent with the Indian case of The Empress v Jhubboo Mahton and others (1882) ILR 8 Cal 739 (“Jhubboo”), which has rejected such an interpretation of s 34.

(f) Principle of doubtful penalisation: Finally, Prof Goh submits that, even if the meaning of s 34 is ambiguous in the present context, the principle of doubtful penalisation should apply so that s 34 should be interpreted in the way most favourable to Azlin, such that s 34 should be restrictively interpreted.

137 For the foregoing reasons, Prof Goh submits that the Judge’s observations at [121] of the GD reflect an accurate view of the current state of the law on s 34 of the Penal Code. Prof Goh further submits that s 34 should not be developed and expanded to allow the Prosecution to do the aforesaid. Prof Goh therefore answers Question (i) in the affirmative; and Questions (ii) and (iii) in the negative. Following this, Question (iv) is moot.

138 We accept that the current understanding of s 34 of the Penal Code is that s 34 is not a “free-standing principle of attribution” to attribute liability for component parts of a “criminal act”, as reasoned by the Judge in the GD at [121] (see [31] above). There are two points which show that this is so.

139 First, the term “criminal act” in s 34 has thus far only been interpreted to refer to the entirety of the criminal act that gives rise to the offence charged, rather than any criminal act that could form a component part of the larger “criminal act” that is the subject matter of the offence charged. This is evident from Barendra at 559, where Lord Sumner famously held that the term “criminal act” refers to “that unity of criminal behaviour, which results in something, for which an individual would be punishable, if it were all done by himself alone” [emphasis added] (see [85(b)] above). As we have already noted, this was followed by this court in Daniel Vijay at [92] and Aishamudin at [49(a)]. In Daniel Vijay, this court further described the term “criminal act” as referring to the “aggregate of all the diverse acts done by the actual [actor] and the secondary offenders, which diverse acts collectively give rise to the offence or offences that the actual [actor] and the secondary offenders are charged with” [emphasis added] (Daniel Vijay at [92]). In Lee Chez Kee at [136], this court also explained the term “criminal act” as meaning “the whole of the criminal transaction in which the co-offenders engage themselves by virtue of their common design and not any particular offence or offences that may be committed in the course of such a transaction” [emphasis added]. This holding was cited and applied by this court in Aishamudin at [44].

140 Second, there have also been observations in the caselaw to the effect that this “unity” of criminal behaviour or acts that constitutes a “criminal act” under s 34 must result in an offence. For instance, in Barendra, Lord Sumner stated that a “criminal act” under s 34 is that “unity of criminal behaviour, which results in something, for which an individual would be punishable, if it were all done by himself alone” [emphasis added] (see [85(b)] above). In Lee Chez Kee at [137], this court cited those observations by Lord Sumner in Barendra and stated that the term “criminal act” “refers to all the acts done by the persons involved which cumulatively result in the criminal offence in question” [emphasis added]. Similarly, in Daniel Vijay at [92], this court explained that the term “criminal act” refers to the “aggregate of all the diverse acts done by the actual actor and the secondary offenders, which diverse acts collectively give rise to the offence or offences that the actual actor and the secondary offenders are charged with” [emphasis added in italics and bold italics]. At [95] of Daniel Vijay, this court further stated that:

141 Therefore, the existing interpretations of s 34 by the courts have only explained the term “criminal act” under s 34 to mean the entirety of the criminal endeavour undertaken by multiple persons which results in a criminal offence for which they are held liable. Consequently, the caselaw has only gone so far as to hold that s 34 can be employed to make an actor responsible for the entirety of a “criminal act” committed by the group of actors if that entire criminal act is done pursuant to their common intention.

142 Prof Goh relies on some of the foregoing interpretations of the term “criminal act” in the caselaw to submit that the proposed expanded interpretation of s 34 is impermissible. This is because, Prof Goh submits, if the term “criminal act” refers to the entirety of the criminal act that must result in a criminal offence, that would mean that the “criminal act” under s 34 cannot be a component part of the larger “criminal act” that is the subject matter of the charge. In Prof Goh’s words, the “criminal act” referred to in s 34 “must fully constitute the actus reus of an offence the accused person is charged with by aid of [s 34]”, and the criminal act must, “together with the common intention, fully form the basis of the offence alleged to be committed” [emphasis in original]. This was also essentially the Judge’s reasoning in the GD at [121] (see [31] above). The Prosecution, as we have highlighted at [40(c)] above, submits that there is nothing in the text of s 34 that precludes its proposed interpretation of s 34.

143 In our judgment, while we agree with Prof Goh’s summary of the current view of s 34, with respect, we disagree that the observations reflected in the caselaw necessarily limit the application of s 34 in that way. The statements cited at [139] and [140] above do not state that the term “criminal act” under s 34 can or must only refer to the entire criminal act that is the subject matter of the charge. In other words, those observations do not go so far as to state that the term “criminal act” under s 34 cannot refer to a component part of the criminal act that is the subject matter of the charge.

144 In our judgment, the crucial fact that must be noted is that the specific issue that we are concerned with did not arise in the cited cases that have interpreted s 34, all of which dealt with either a “dual crime” or “single crime” situation. As we have explained at [98] and [123] above, the present case is not the typical “dual crime” or “single crime” case; the present case falls within a third type of situation, as explained at [126(b)] above.

145 There has been no case where the court has been presented with the issue that this court is faced with, and Prof Goh accepted this at the hearing before us. The Indian case which Prof Goh had cited – Jhubboo (at [136(e)] above) – is not analogous to the present case, and the judgment in that case also did not deal with the present issue, as explained at [173] to [178] below. In short, one reason why s 34 has not been given the expanded interpretation is that no court has explicitly been asked to consider doing so. Therefore, the prevailing interpretation of s 34 is not dispositive of the question whether s 34 of the Penal Code can, in principle, be given the expanded interpretation, and that is a matter that falls on us to decide and we approach it from first principles.

146 In our judgment, the proposed interpretation of s 34 is permissible. There are two main reasons why this is so: the text of s 34 supports the expanded interpretation of that provision, and the expanded interpretation of s 34 also furthers the purpose of the provision.

147 As stated in Aishamudin at [40], “the text of s 34 is of critical importance and anchors” [emphasis in original omitted] any analysis on the ambit of the provision. Thus, we first turn to the text of s 34, which we set out again here for convenience:

148 In our judgment, a careful scrutiny of the text of s 34 suggests that it is capable of supporting the expanded interpretation such that the alternative s 300(c) charge is permissible. Section 34 states that, “[w]hen a criminal act is done by several persons, in furtherance of the common intention of all, each of such persons is liable for that act in the same manner as if the act were done by him alone” [emphasis added]. The important phrases to note here are the terms that have just been emphasised: s 34 renders an offender “liable” for a “criminal act” done by several persons in furtherance of their common intention “as if the act were done by him alone”. Section 34 does not state that the offender is “guilty of” or to be “punished” for an “offence” committed by several persons in furtherance of their common intention as if “the offence” were done by him alone. Nor does it say that each of the co-offenders is only to be liable for the same offence as every other co-offender. The choice of words used in s 34 is telling because at least some of the alternative possibilities just mentioned explicitly feature in other provisions of the Penal Code. Some notable examples are as follows.

(a) The term “offence” is explicitly defined under s 40(1) of the Penal Code to denote “a thing made punishable by this Code”. Numerous offence-creating provisions in the Penal Code use the phrase “shall be guilty of an offence” [emphasis added] to denote that a certain act would be an offence. For instance, s 375 states that “[a]ny man who penetrates the vagina of a woman with his penis (a) without her consent; or (b) with or without her consent, when she is under 14 years of age, shall be guilty of an offence” [emphasis added]. Therefore, it is clear that the term “criminal act” cannot be restricted to refer to the “offence” which the offenders plan to carry out. The distinction between “offence” and “criminal act” was also reiterated in Aishamudin at [49(a)].

(b) The term “liable” in s 34, as opposed to “guilty” or “punished”, is also notable. In other provisions of the Penal Code, the term “guilty” is used to denote that an offender can be guilty of an offence if the offender commits the acts stated under that provision of the Penal Code (see for example, s 375 of the Penal Code). The term “punished” is similarly used in other provisions of the Penal Code to denote the punishment range that an offender who is guilty of an offence could be sentenced to (see for example, s 323 of the Penal Code).

149 In the final analysis, we respectfully decline to accept Prof Goh’s submission on this because it seems to us that it would have the effect of altering the meaning of the terms that are in fact used in s 34.

150 First, Prof Goh’s interpretation of “criminal act” (see [136(a)(i)] above) would effectively equate the meaning of the term “criminal act” with the term “offence”. This would erode the distinction between a “criminal act” and an “offence”. The term “offence” is even explicitly defined under s 40(1) of the Penal Code to denote “a thing made punishable by this Code”. The distinction between a “criminal act” and an “offence” under the Penal Code is important because any given act may amount to different “offences” under the Penal Code (see [85(f)] above).

151 The term “act” is defined under s 33(1) of the Penal Code to denote “as well a series of acts as a single act”. The term “criminal” is not defined in the Penal Code, but it appears multiple times throughout the Penal Code (for example, under s 35 as “Whenever an act, which is criminal only by reason of its being done with a criminal knowledge or intention”; under s 120A as “criminal conspiracy”; under s 350 as “[c]riminal force”; under s 405 as “[c]riminal breach of trust”; under s 441 as “[c]riminal trespass” [emphasis added]). It is evident from these various uses of the word “criminal” in the Penal Code that the word is simply meant to denote that what would otherwise be a non-criminal act or matter is made “criminal” by way of that particular provision of the Penal Code.

152 As for Prof Goh’s reliance on the dicta in the caselaw explaining the term “criminal act” as referring to the collective acts done resulting in a criminal offence charged (see [136(a)(ii)] above), this submission brings us back to the preliminary point we began with at [145] above: while we accept that the caselaw has hitherto interpreted the term “criminal act” to mean the “unity of criminal behaviour” among the co-offenders which forms the basis of the offence charged (see [139] to [141] above), that does not in and of itself explain why, in principle, the text in s 34 is limited to that meaning.

153 Prof Goh also submits that the term “liable for that act” in s 34 means “liable to be punished for an offence that is fully constituted by the act and the common intention”. With respect, we disagree with this. By convicting Azlin of the alternative s 300(c) charge, s 34 would be rendering her “liable for” Incidents 2 and 4. What Prof Goh’s submission achieves instead is that it limits the acts that an offender can be made “liable for” to those that are offences. However, this would, again, entail eroding the distinction between a “criminal act” and an “offence”.

154 It bears highlighting that Azlin can be liable for Incidents 2 and 4 pursuant to s 34 even though all or part of the acts concerned were done by Ridzuan. This in fact was the very result of the proceedings below: the Judge convicted Azlin of two alternative charges under s 326 read with s 34 in respect of Incidents 2 and 4 respectively (see [34(b)] and [34(d)] above). Moreover, it seems offensive to common sense to hold that Azlin should not be liable for an act done by Ridzuan at her urging or with her agreement because that act, taken with other acts she herself did, expose her to a more serious penalty.

155 The true question, thus, concerns the question of the permissibility of amalgamating discrete acts to form the larger criminal act that forms the basis of the offence charged: whether Azlin’s liability for Ridzuan’s acts in Incidents 2 and 4 – imposed constructively by way of s 34 – can be aggregated with her own acts in Incidents 1 and 3 – for which she is directly liable – to form a “larger” criminal act (the Cumulative Scald Injury from Incidents 1 to 4) that is the basis of the alternative s 300(c) charge.

156 We agree with the Prosecution that there is no reason in principle why this should be impermissible, since neither the text of s 300(c) nor that of s 34 prevents this in any way. Nor does such an amalgamation pursuant to the expanded interpretation of s 34 offend s 132 of the CPC, which provides that, “[f]or every distinct offence of which any person is accused, there must be a separate charge” [emphasis added]. Aggregating Azlin’s direct liability for Incidents 1 and 3 with her constructive liability for Incidents 2 and 4 to form a “larger” criminal act forming the basis of the charge leads to only a single offence arising from that larger criminal act that is stated in the form of the alternative s 300(c) charge – s 300(c) murder.

157 Aside from the absence of any reason in principle why such an aggregation should be impermissible, in our judgment, whether or not it can be invoked in any given case will largely be an evidential question of whether the actus reus and mens rea of the ultimate charge in that case can be established. In the final analysis, the issue is fact-specific and would have to be resolved on a case-by-case basis.

158 We accordingly find that the text of s 34 does permit the expanded interpretation such that the alternative s 300(c) charge is permissible.

159 We next turn to the purpose of s 34. The three-step approach to purposive interpretation is well-established. First, a court should ascertain the possible interpretations of the provision in question, by determining the ordinary meaning of the words in the provision, aided by rules and canons of statutory construction (Tan Cheng Bock v Attorney-General [2017] 2 SLR 850 (“Tan Cheng Bock”) at [38]). Second, a court should then ascertain the legislative purpose of the provision and the part of the statute in which the provision is situated. Third, a court should compare the possible interpretations of the provision against the purpose of the relevant provision and prefer the interpretation which furthers the purpose of the written text (Tan Cheng Bock at [54(c)]).

160 It is uncontroversial that the fundamental purpose of s 34, as also highlighted by the Prosecution, is to deter group crimes. This is clear from Ratanlal at 111:

161 The foregoing extract from Ratanlal clearly shows that s 34 of the Penal Code is meant to expand, rather than restrict, the criminal liability of those who commonly intend and participate in group crimes beyond the specific actions personally committed by the offender. This is because “the presence of accomplices gives encouragement, support and protection to the person actually committing the act”. This was the precise situation mirrored by Incident 4, where it was Azlin who told Ridzuan to deal with the Deceased, knowing full well, based on their past practice, that Ridzuan would pour hot water on the Deceased to scald him (see GD at [128]; see [24] above). This was also the situation in Incident 2, where both respondents went after the Deceased to scald him together. Section 34 also seeks to overcome the difficulty in “distinguish[ing] between the acts of individual members of a party” and “to prove exactly what part was taken by each of them”. This was the precise difficulty that could otherwise have arisen from Incident 2, as both Azlin and Ridzuan pursued the Deceased around the room while taking turns to pour hot water on the Deceased (see [19] above).

162 It is also important to bear in mind our holding in Tan Cheng Bock at [43] that, in ascertaining the legislative purpose behind a statutory provision, while extraneous material may be a useful aid to interpretation, primacy should be accorded to the text of the provision and its statutory context. In this regard, the critical part of the text of s 34 are the words that, when a criminal act is done by several persons in furtherance of their common intention, each of these persons is liable for that act “in the same manner as if the act were done by him alone” [emphasis added]. These words suggest that the fundamental purpose of s 34 is to ensure that, when A intends to commit a criminal venture, A should be liable for that venture, including being liable for acts committed by others pursuant to their common intention. This is because A should be liable for the criminal acts he intended to be committed, and did bring about by means of the joint acts of himself and his co-offenders, “as if the act were done by [A] alone”. The expanded interpretation of s 34 would further this purpose because it would hold Azlin liable for all four scalding incidents as if all four incidents were done by her alone when she was the one who intended to commit all four scalding incidents, and two of those incidents were committed by Ridzuan pursuant to a common intention shared with Azlin. It would be illogical and would undermine the purpose of s 34 if the expanded interpretation were impermissible.

163 We agree with the Deputy Public Prosecutor (“DPP”), Mr Mohamed Faizal SC, that a purposive reading of s 34 should not result in an absurd or unreasonable outcome and this weighs against excluding the expanded interpretation of s 34.

164 We acknowledge that it is also an important general principle of the criminal law that an offender should not be punished beyond his or her personal culpability. This may at times seem to pull in the opposite direction from the purpose of deterring group crimes. The question for the court in such circumstances is how to strike the right balance between the foregoing two principles.

165 However, in the present case, the proposed expanded interpretation of s 34 would further both the foregoing principles. The expanded interpretation of s 34 would both deter group crimes and ensure that the primary perpetrator behind the aggregated criminal act, Azlin in this case, is charged with an offence that reflects her full culpability. The point can also be illustrated by the following example: suppose A intends to import 20g of diamorphine into Singapore. A knows the threshold for capital punishment is 15g. He therefore engages a co-offender, B, to transport 8g into Singapore, without informing the co-offender that he will be bringing the remaining 12g. There is no doubt at all that (a) A would be jointly liable with B for importing the 8g; and (b) B would not be jointly liable with A for importing the 12g. Yet, it seems implausibly illogical that A could not be held liable under s 34 for precisely the offence he intended to and did commit, which is to import 20g. Thus, the expanded interpretation of s 34 would not undermine the principle that an offender should not be punished beyond his or her personal culpability. While s 18(4) of the MDA might provide another solution to deem the drug to be in A’s possession, as B would have possessed the drug “with the knowledge and consent” of A, the point is that the expanded interpretation of s 34 is not inconsistent with the general principle that an offender should not be punished beyond his or her personal culpability.

166 Prof Goh’s submission that s 34 was meant to be aligned with the doctrine of common purpose in English law, which was concerned with “dual crime” situations (see [136(b)] above), does not deal with the fact s 34 also avails in a “single crime” situation. Section 34 does not differentiate between “single crime” and “dual crime” situations.

167 For these reasons, we hold that the expanded interpretation of s 34 is permissible because it would further the purpose of s 34 of deterring group crimes and hold those who are the most culpable liable for the full extent of their intended acts.

168 The remaining points raised by Prof Goh are, in our judgment, more straightforward and can be dealt with briefly. We respectfully disagree with Prof Goh on his submissions that the expanded interpretation of s 34 would be inconsistent with the theories underlying s 34 (see [136(c)] above). We first observe that these theoretical foundations of the doctrine of common purpose may not be regarded as settled. This is explicitly stated in the text cited by Prof Goh, which caveats the entire discussion on the theories of complicity as “a number of apparently feasible, if not ultimately convincing, theories” [emphasis added] (K J M Smith, A Modern Treatise on the Law of Criminal Complicity (Clarendon Press, 1991) at p 5). Neither Prof Goh nor any of the parties have pointed us to a case that has authoritatively accepted any of these theories as the proper theoretical foundation underlying s 34.

169 Further, the expanded interpretation of s 34 would in fact be consistent with the two theories highlighted by Prof Goh. It would better give effect to the agency theory because it would capture offenders who intentionally arrange matters such that other persons commit component parts of the larger “criminal act” forming the basis of the charge for the offender. This would be the case for Azlin in this case – Ridzuan was effectively acting as Azlin’s “agent” when he was carrying out his acts of scalding in Incidents 2 and 4. As for the equivalence theory, none of the materials cited by Prof Goh show that this theory seeks to limit the principal offender’s liability, even if the principal offender is more culpable than the secondary offender. This would be the case for Azlin in relation to Ridzuan in this case. We therefore do not accept Prof Goh’s submissions on this.

170 Prof Goh also submits that the expanded interpretation of s 34 would offend the fundamental principle that there must be a concurrence of actus reus and mens rea for any offence (see [136(d)] above).

171 As against this, the learned DPP submits that the concurrence principle poses no difficulties because, if in a given case, there is in fact no coincidence of the actus reus with the mens rea where the expanded interpretation of s 34 is invoked, then the offence would simply not be made out, and the charge would therefore not be proven. At the hearing before us, Prof Goh accepted that this would follow. In the present case, there is a coincidence of the actus reus and mens rea, as explained at [182] below and so the issue simply does not arise.

172 We appreciate and accept Prof Goh’s broader point that it may be possible that, in some instances, the aggregation of the component acts and intentions would not suffice to achieve the concurrence of the actus reus and mens rea of the offence charged. While no example of this was forthcoming, the question in every case would turn on whether, in that particular case, by reason of the aggregation of the component acts and intentions, there is or is not a concurrence of the actus reus and mens rea of the offence charged.

173 We turn to Prof Goh’s reliance on Jhubboo to submit that the expanded interpretation of s 34 is not permitted. We again respectfully disagree.

174 In that case, Jhubboo and seven other accused persons were convicted of murder under s 302 read with s 149 of the Indian Penal Code. Under s 149 of the Indian Penal Code, if an offence is committed by any member of an unlawful assembly in pursuit of the common object of that assembly, or such as the members of that assembly knew to be likely to be committed in that context, every person who, at the time of the committing of that offence, is a member of the same assembly, is guilty of that offence. In Jhubboo, the charge against the co-accused persons was that Jhubboo had committed murder and that the co-accused persons were, with Jhubboo, members of an unlawful assembly and were therefore, by virtue of s 149, guilty of murder because they knew it to be likely that murder would be committed in the course of prosecuting the common object of the unlawful assembly. The evidence showed that the injuries to the deceased included injuries to the head and to the small intestines, and a cut on the arm inflicted by means of a sword. The medical evidence was that death was caused by shock following the injuries to the small intestines and the wound to the arm. The jury found that Jhubboo had not committed murder, but convicted the seven other accused persons under s 302 read with s 149.

175 Field J faulted the Sessions Judge for not giving proper directions to the jury as to what might constitute murder. The issue arose because the jury apparently did not believe that Jhubboo had caused the injuries to the small intestines, and so considered it questionable whether there was sufficient evidence to hold Jhubboo guilty of murder if he were found to have only inflicted the wound to the arm.

176 Prof Goh placed emphasis on Field J’s observations in Jhubboo to the following effect at 751–752:

177 In the portion of the foregoing extract that has not been emphasised, Field J considered that, assuming the evidence was that Jhubboo only inflicted one of the two injuries that cumulatively resulted in death, Jhubboo could nonetheless be convicted of murder by the employment of s 34. Prof Goh highlighted the italicised portion of the foregoing extract to submit that Field J had further gone on to doubt “whether Jhubboo’s constructive guilt [for] murder could then be used to satisfy the elements of another offence, that is s 149, so as to make him and the other seven accused persons guilty of that offence [under s 149]” (as quoted from Prof Goh’s submissions). Prof Goh submits that, just as Field J expressed an intuitive hesitation in the use of s 34 to convict others for another offence by way of a “double construction”, so should s 34 not be used with the expanded interpretation.

178 In our judgment, Field J was making a different point. Section 149 is not in and of itself an offence-creating provision; s 149 imposes liability on a member of an unlawful assembly for an offence not committed by that member if it is “committed by” another member of the unlawful assembly. Just prior to making that observation extracted at [176] above, Field J highlighted that the “first essential question was, whether murder had been committed by Jhubboo” [emphasis added]. It was in this context that Field J questioned whether Jhubboo can be said to have “committed” murder within the meaning of s 149 if he had not directly committed murder himself, but had only been constructively liable for it owing to the combined acts of himself and others (see the bolded portion of the extract at [176] above). Field J was in fact making an observation concerning the possible danger of imposing constructive liability on other offenders under s 149 for an offence for which the primary offender was only liable under another basis for imposing constructive liability, namely, under s 34. It does not, with respect, seem to us that Field J was commenting on the operation or scope of s 34 itself. Thus, in our view, what Field J had doubted was whether Jhubboo’s (hypothetical) constructive guilt for murder by means of s 34 could be used to satisfy the elements of another offence, specifically murder under s 302 read with s 149, committed by seven other people so as to make not Jhubboo but these other seven persons constructively guilty of murder by means of s 149. We think that is a different situation altogether and we therefore do not find that Jhubboo assists us in this case.

179 Finally, we turn to Prof Goh’s submission that, since the meaning of s 34 is ambiguous, s 34 should be restrictively interpreted in a way more favourable to Azlin. The difficulty with this is that the text of s 34 itself is not ambiguous, and the purpose of s 34 is also clear. Section 34 (a) uses the term “criminal act” instead of “offence”; (b) s 34 refers to only a criminal act that is done “in furtherance of the common intention of all”; (c) s 34 states that a person would be “liable for” that criminal act, not that the person would be “punished” or “liable to be punished” for that criminal act or offence; and (d) s 34 states that the offender would be liable for the criminal act done by several persons and commonly intended by them “as if the act were done by him alone”. In other words, the text of s 34 itself is fully capable of the proposed expanded interpretation, and the expanded interpretation would also further the purpose of s 34. In this situation, there is simply no basis for the principle of doubtful penalisation to preclude the adoption of the expanded interpretation of s 34.

180 Accordingly, we summarise our findings on the applicable legal principles, and answer Questions (i) to (vi) (outlined at [48] above), as follows.

(a) It is uncontroversial that, under the existing law, where s 300(c) murder has been committed as the collateral criminal act in a “dual crime” scenario, the Daniel Vijay test applies to determine if the secondary offender charged with a s 300(c) common intention murder charge should be constructively liable for the s 300(c) murder (see [87] to [100] above).

(b) On the other hand, where s 300(c) murder has been jointly committed in a “single crime” scenario (that is, multiple offenders jointly commit a single offence of s 300(c) murder), the current state of the law is such that the Daniel Vijay test does not apply. There is therefore no need for the Prosecution to prove that the offender who is charged with a s 300(c) common intention murder charge intended to inflict a s 300(c) injury (that is, an injury that would be sufficient in the ordinary course of nature to cause death). Instead, the Virsa Singh test applies such that it is sufficient that the said offender intended to cause the actual injury that was inflicted on the victim (see [106] to [118] above). Therefore, we answer Question (v) in the negative. It follows that Question (vi) is moot.

(c) However, the present case is neither a “single crime” nor a “dual crime” scenario (see [98] and [123] above). Instead, the present case presents a novel third type of situation where s 34 might potentially be applicable (see [124] above). Furthermore, we also agree with Prof Goh that the alternative s 300(c) charge is not a s 300(c) common intention murder charge. This is because s 34 is only being employed in the alternative s 300(c) charge to satisfy part of the criminal act forming the actus reus of s 300(c) murder. On the other hand, s 34 is conventionally used to render an offender liable for all the elements of the offence once the requirements of s 34 are satisfied. We therefore agree with Prof Goh that the Daniel Vijay test is irrelevant to the alternative s 300(c) charge (see [128] to [134] above). The question then is whether s 34 can be employed in the manner envisaged under the alternative s 300(c) charge, as described at [129], [133(a)] and [134] above (and what we have referred to as the “expanded interpretation” of s 34 at [47(c)(i)] above).

(d) We agree with Prof Goh that Question (i) should be answered in the affirmative. The Judge’s comments at [121] of the GD that s 34 is not a “free-standing principle of attribution” to attribute liability for component parts of the “criminal act” accurately reflect the state of the law on s 34 of the Penal Code as it was at the time of the judgment (see [138] to [141] above). However, one reason why s 34 has not been given the expanded interpretation is that no court has explicitly been asked to consider doing so. Therefore, the prevailing interpretation of s 34 is not dispositive of the question whether s 34 of the Penal Code can, in principle, be given the expanded interpretation (see [144] to [145] above)

(e) We respectfully disagree with Prof Goh on Question (ii). In our judgment, Question (ii) should also be answered in the affirmative, such that, in the third type of situation highlighted at [124] above, s 34 may be employed to attribute liability for component acts committed by another person (Incidents 2 and 4 committed by Ridzuan in this case) to the offender (Azlin) so as to aggregate those component acts with other acts personally committed by the offender (Incidents 1 and 3 committed by Azlin) to form a “larger” criminal act (the four scalding incidents cumulatively) that is the actual basis of the offence charged (the alternative s 300(c) charge). The text of s 34 permits this (see [147] to [158] above), and this interpretation of s 34 would also further its purpose, which is to deter group crimes and expand the criminal liability of those who commonly intend and participate in group crimes beyond the specific actions personally committed by the offender (see [159] to [167] above). When considering whether s 34 is satisfied when it is employed in this manner, the traditional elements of s 34 – the elements of participation, criminal act, and common intention – should be applied in relation to the relevant component acts. It is also important to consider, in each case, whether the aggregation of the component acts and intentions would achieve the concurrence of the actus reus and mens rea of the offence charged (see [172] above).

(f) We agree with Prof Goh that Question (iii) should be answered in the negative, in that the alternative s 300(c) charge would not have been permissible under the hitherto existing caselaw on s 34 of the Penal Code, but only because the point had not squarely been considered.

(g) We respectfully disagree with Prof Goh on Question (iv). In our judgment, Question (iv) should be answered in the affirmative, such that the alternative s 300(c) charge should be permissible under the interpretation of s 34 of the Penal Code that we have arrived at.

181 We turn to the elements of the alternative s 300(c) charge, which have been outlined at [133] above. The remaining questions are (a) whether the requirements of s 34 are satisfied to attribute Ridzuan’s acts in Incidents 2 and 4 to Azlin, and (b) whether the aggregation of Azlin’s direct liability for Incidents 1 and 3 with her constructive liability for Incidents 2 and 4 would satisfy the actus reus and mens rea requirements of the alternative s 300(c) charge. In our judgment, the alternative s 300(c) charge is proved beyond reasonable doubt.

182 First, it is clear that the requirements of s 34 are satisfied to attribute Ridzuan’s acts in Incidents 2 and 4 to Azlin. The criminal act element is satisfied because there are criminal acts (Incidents 2 and 4) which were done by several persons (Azlin and Ridzuan). The participation and common intention elements are also satisfied, as follows. Indeed, this is uncontroversial given that Azlin is not contesting her conviction on charges C1B3 and C1B1 under s 326 read with s 34 of the Penal Code for voluntarily causing Incidents 2 and 4 in furtherance of a common intention shared with Ridzuan.

(a) For Incident 2, it was Azlin who first became angry at the Deceased and splashed hot water on the Deceased repeatedly all over his body. When Ridzuan later also splashed hot water on the Deceased together with Azlin, Azlin not only did not stop Ridzuan but pursued the Deceased around the house and splashed hot water on him repeatedly as well. We agree with the Judge that Azlin was acting in implicit agreement with Ridzuan for them to splash hot water on the Deceased together. In short, Azlin intended to inflict not only her own acts of scalding but also their combined acts of scalding on the Deceased. Thus, it is clear that the acts of scalding in Incident 2 were done in furtherance of Azlin and Ridzuan’s common intention, and it is clear that Azlin participated in this criminal act.

(b) For Incident 4, it was Azlin who woke Ridzuan up when the Deceased refused to bathe and who asked him to deal with the Deceased. Ridzuan went to the toilet, beat the Deceased’s legs with a broomstick, and then started splashing hot water at him while standing at the entrance of the toilet. During this time, Azlin, together with Ridzuan, continued to shout at the Deceased to remove his shorts. Azlin clearly approved of Ridzuan’s scalding of the Deceased (GD at [67]). There is no doubt that she intended what happened. As the Judge observed, Azlin “was the one who had asked Ridzuan to deal with the situation, well knowing how he would proceed. She saw and acquiesced, in any event, in his actions” (GD at [128]). Thus, it is also clear that the acts of scalding in Incident 4 were done in furtherance of Azlin and Ridzuan’s common intention, and Azlin also participated in this criminal act by being the one who instigated Ridzuan to scald the Deceased, and encouraged him throughout the process by, according to Ridzuan, shouting at the Deceased while standing next to Ridzuan when he was scalding the Deceased (see [24] above).

183 We are also satisfied that the “aggregation” of Azlin’s acts and intentions in Incidents 2 and 4 with her acts and intentions in Incidents 1 and 3 would satisfy the actus reus (causation of the Cumulative Scald Injury) and mens rea (intention to cause the Cumulative Scald Injury) requirements of the alternative s 300(c) charge.

(a) Azlin’s commission of Incidents 1 and 3, when combined with her joint commission of Incidents 2 and 4 with Ridzuan, gave rise to the commission of Incidents 1 to 4 which is what caused the Cumulative Scald Injury. It is not disputed that this is sufficient in the ordinary course of nature to cause death, and did cause the death of the Deceased. This satisfies the actus reus for the alternative s 300(c) charge.

(b) It is undisputed that Azlin intended to commit Incidents 1 and 3 and carried out these incidents herself. It is also undisputed in these appeals that Azlin intended to commit Incidents 2 and 4. This is because the very basis for the convictions of charges C1B3 and C1B1 is that Azlin commonly intended to commit Incidents 2 and 4 with Ridzuan, and Azlin has not appealed against her conviction for charges C1B3 and C1B1. Although those are different offences, the acts in question are the very ones we are concerned with. The aggregation of Azlin’s intention to commit Incidents 1 and 3 with her intention to commit Incidents 2 and 4 with Ridzuan amounts to an intention to commit all four incidents of scalding to cause the Cumulative Scald Injury. This satisfies the mens rea requirement for the alternative s 300(c) charge, which is the intention to cause the particular injury caused (the Cumulative Scald Injury). The fact that Azlin’s intention to commit Incidents 2 and 4 was an intention she shared with Ridzuan is immaterial.

184 Consequently, while we appreciate Prof Goh’s broader point that an aggregation of component intentions may not, in some cases of such “multiple acts situations”, suffice to satisfy the mens rea of the ultimate offence charged (see [60] above), we are satisfied in this case that Azlin did intend the Cumulative Scald Injury as it is clear beyond reasonable doubt that she intended to cause all four scalding incidents. We therefore find that Azlin is guilty of the alternative s 300(c) charge beyond reasonable doubt. Accordingly, we allow the appeal in CCA 17 and substitute Azlin’s conviction on the four s 326 charges (charges C1B1 to C1B4) with the alternative s 300(c) charge.

185 For completeness, we make a few observations regarding some other issues arising in this case.

186 We agree with the Judge that, based on the evidence as it was adduced at the trial below, Ridzuan would not be guilty of the Murder Charge. This is because he only participated in Incidents 2 and 4, and the medical evidence adduced by the Prosecution could not establish the extent of the burns or injury caused by each particular scalding incident standing in isolation or how each contributed to the death. The Prosecution itself accepted and ran its case on the basis that the “criminal act” forming the basis of the Murder Charge was the Cumulative Scald Injury caused by the collective acts of scalding from all four incidents. Hence, there was no evidentiary basis for concluding that the injuries inflicted in Incidents 2 and 4 were sufficient in the ordinary course of nature to cause death.

187 However, we question if that result is possibly a consequence of the way the Murder Charge was framed and the way the Prosecution’s case was run at the trial below. In our judgment, it was completely artificial to analyse the injuries on the footing that each burn injury caused by each scalding incident should stand alone, as though none of the other prior injuries had occurred.

188 The question of whether a bodily injury was sufficient in the ordinary course of nature to cause death cannot be assessed in a vacuum or in the abstract. Rather, this must depend on whether that bodily injury is sufficient in the ordinary course of nature to cause death to that specific victim (in this case, the Deceased). By way of example, a hard punch may well be insufficient in the ordinary course of nature to cause death to a well-built adult man, but it may well be otherwise if the hard punch was intentionally inflicted on a one-month-old baby or a frail 100-year-old person. It follows that in considering this case, it was necessary to first consider what physical condition the Deceased was in just prior to Incident 4. Once that is done, it seems to us that the injury caused by Incident 4 would likely be regarded as sufficient in the ordinary course to cause death, because that is what completed the Cumulative Scald Injury. And it was common ground that this was sufficient to cause death. In any event, the point was not fully explored and we say no more on this.

189 We turn to Azlin’s sentence. Under s 302(2) of the Penal Code, the statutorily prescribed minimum punishment for s 300(c) murder is life imprisonment:

190 In light of our finding in CCA 17, the appeal in CCA 25 is dismissed as it is moot. We direct that the matter be adjourned for further submissions on sentence pending the intimation of the Prosecution’s position on sentencing.

191 We turn to the appeal against Ridzuan’s sentence for charge D1B1 for the offence of voluntarily causing grievous hurt by heated means for Incident 4 under s 326 of the Penal Code. The Judge sentenced Ridzuan to 14 years’ imprisonment and 12 strokes of the cane for charge D1B1, and a global sentence of 27 years’ imprisonment and 24 strokes of the cane (see [38(b)] above).

192 The Prosecution has not appealed against the dismissal of Ridzuan’s Murder Charge. At the same time, we note the Judge’s observations on the relative positions of Azlin and Ridzuan (GD at [194]):

193 The Prosecution sought life imprisonment against Ridzuan and the Judge rejected this for various reasons. She emphasised that a “critical distinction” between s 326 and culpable homicide under s 304(a) of the Penal Code is that the former “operates within a less culpable range of intention” such that the mens rea for s 326 “is satisfied so long as the offender knows himself to be likely to cause grievous hurt”, whereas, under s 304(a), the accused person would “minimally” have the intent to “cause bodily injury that is likely to cause death” (GD at [186]). Hence, in considering a sentence of life imprisonment under s 326, she thought two additional factors were important: the dangerous weapon or means used and the level of intention or knowledge that the offender had in using the particular dangerous means in inflicting the particular grievous hurt (GD at [187]).

194 The Judge noted that, in the instant case, there were “multiple individual charges” unlike the case of culpable homicide where the offence is encapsulated in a single charge. The charges for which life imprisonment were sought – charges C1B3 and D1B1 – involve “hurt which endangered life” rather than “death” because the medical evidence could not pinpoint which of the four scalding incidents caused death. Thus, the Judge reasoned that, “to address the consequence of all four incidents in the sentence on one offence could be an excessive sentence for the particular charge” (GD at [189]). The fact that Azlin and Ridzuan did not entirely comprehend the likelihood of death, coupled with Azlin’s adjustment disorder and Ridzuan’s low intelligence, led her to conclude that this is not the “worst case” under s 326 so as to warrant the imposition of life imprisonment.

195 Turning to the individual sentences, adapting from the sentencing framework in Ng Soon Kim v Public Prosecutor [2020] 3 SLR 1097 (“Ng Soon Kim”), which pertained to s 324 of the Penal Code (voluntarily causing hurt by dangerous weapons or means), the Judge first considered what the appropriate sentence would be if the charge had been one under s 325 of the Penal Code (voluntarily causing grievous hurt). The Judge considered that this would be, for Ridzuan, nine years’ imprisonment and 12 strokes of the cane for charge D1B1 (Incident 4) (GD at [203]).

196 Second, to account for the dangerous means used, the Judge applied an uplift of two years’ imprisonment, bearing in mind the “exceptionally cruel and painful use of a dangerous means” in this case (GD at [206]).

197 Third, the Judge considered the aggravating and mitigating factors and imposed a further uplift of three years’ imprisonment for charge D1B1 (GD at [211]).

(a) The Judge, primarily relying on Public Prosecutor v AFR [2011] 3 SLR 833 (“AFR (CA)”) at [20], accepted that there was a need for deterrence and retribution in this case, as society has “a special interest in protecting the young from physical abuse” (GD at [177]–[179]).

(b) The Judge took into consideration the joint action of Azlin and Ridzuan in assaulting the young Deceased as an aggravating factor for both Incidents 2 and 4 (GD at [208]).

(c) The Judge also noted the mutual prevarication in seeking medical attention, and the jointly fabricated narrative of the kettle accident used at the hospital (see [26] above), as a further aggravating factor for Incident 4. This justified a higher uplift for Incident 4 (GD at [209] and [211]).

(d) On the other hand, the Judge took into account the mitigating factor that Ridzuan pleaded guilty to charge D1B1 (GD at [210]).

(e) The Judge placed “limited” mitigating weight on Ridzuan’s cooperation with the police (GD at [180]).

(f) The Judge did not give any weight to the contentions of psychiatric conditions made by both offenders, as she found that neither Azlin nor Ridzuan had any mental disorder which would diminish culpability or would prevent the need for deterrence and retribution from being given full effect (see GD at [181]–[182]).

(g) The Judge also rejected the respondents’ purported “difficulties” and “stressors”, such as financial difficulty, as a mitigating factor (GD at [183]).

198 Consequently, for charge D1B1 (for Incident 4), the Judge imposed 14 years’ imprisonment and 12 strokes of the cane on Ridzuan.

199 The sentencing approach where an accused person commits multiple offences has been clarified and summarised in Gan Chai Bee Anne v Public Prosecutor [2019] 4 SLR 838 (“Anne Gan”) at [19]–[22]. The principles may be summarised as follows.

(a) Sentencing for multiple offences entails two distinct steps which should be taken in sequence. First, the court should reach a provisional view of the individual sentence for each offence. Second, the court has to determine the overall sentence to be imposed.

(b) The second step concerns not only the issue of how the sentences ought to be run, but also whether the totality of the offender’s conduct justified an adjustment, whether upwards or downwards, in the individual sentences decided at the first step. The basis of this adjustment is the totality principle, which not only possesses a limiting function, in guarding against an excessive overall sentence, but also a boosting effect on individual sentences where they would otherwise result in a manifestly inadequate overall sentence.

(c) Therefore, at the second step, the court ought to consider whether the existence of any cumulative aggravating factors justifies recalibrating the individual sentences upwards and/or running those recalibrated sentences consecutively.

200 The principles governing when a maximum prescribed sentence of life imprisonment is suitable were outlined by this court in Public Prosecutor v P Mageswaran and another appeal [2019] 1 SLR 1253 (“Mageswaran”) at [43], [45] to [46], and [49]. While that case dealt with culpable homicide under s 304(a) of the Penal Code, the principles can also be applied to s 326. This is because s 304(a) of the Penal Code also provides for a broadly similar sentencing band as s 326 (in both cases, the person convicted shall be punished with (a) imprisonment for life, and shall also be liable to caning; or (b) imprisonment for a term which may extend to 20 years in the case of s 304(a) and up to 15 years in the case of s 326, and also be liable to fine or to caning). The principles may be summarised as follows.

(a) The maximum sentence is not reserved for the worst offence of the kind dealt with that can be imagined; instead, it “should be reserved for the worst type of cases falling within the prohibition” [emphasis in original] (Mageswaran at [45]).

(b) To determine if a particular case is one of the worst type of cases of culpable homicide, the court should identify a range of conduct which characterises the most serious instances of the offence in question, taking into account both the nature of the crime and the circumstances of the criminal (Mageswaran at [45]).

(c) As the range of circumstances in which the offence is committed will be extremely varied, it will not be possible to lay down concrete guidelines or rules as to when a case becomes one of the worst type of its offence (Mageswaran at [46]).

(d) For a case to be “one of the worst type of cases” would take an “exceptional case, devoid of any mitigating circumstances” [emphasis in original] (Mageswaran at [49]).

(e) The burden is on the Prosecution to demonstrate that the particular case in question is one of the worst type of cases under s 326 of the Penal Code (Mageswaran at [43]).

201 We also consider that it is relevant in the context of violent crimes to have regard to the High Court’s observations in Public Prosecutor v Aniza bte Essa [2008] 3 SLR(R) 832 at [47] (made in the context of s 304(a) of the Penal Code):

202 An appellate court will not ordinarily disturb the sentence imposed by the lower court, except where it is satisfied that: (a) the sentencing judge erred with respect to the proper factual basis for sentencing; (b) the trial judge failed to appreciate the materials placed before him; (c) the sentencing was wrong in principle; and/or (d) the sentence was manifestly excessive or manifestly inadequate: see for instance ADF v Public Prosecutor and another appeal [2010] 1 SLR 874 at [17]. The Prosecution submits that the sentence for Ridzuan in manifestly inadequate. The Prosecution submits that the Judge erred in both steps of the Anne Gan framework either because the charge D1B1 in itself warrants life imprisonment or, alternatively, the sentence for this charge should have been increased to life imprisonment at the second step of the Anne Gan framework.

203 The Judge did not cite Anne Gan in the GD, but she did correctly identify the broad two-step framework outlined at [199(a)] above (see GD at [188]). In that respect, the Judge rightly found it important to “first ensure that each offence is addressed with an appropriate sentence” before considering the “overall criminality … in the context of the offences to arrive at a global sentence.”

204 We do not accept the Prosecution’s submission that Incident 4 alone can be viewed as one of the worst type of cases of s 326 which in itself would warrant a sentence of life imprisonment. Incident 4 is ultimately a single occasion of scalding that occurred within a night, and the grievous hurt caused by Incident 4, as framed in charge D1B1 and as brought out in the evidence in the trial below, was only hurt which endangered life, which is not the most serious form of grievous hurt under s 326 (which would be death (see s 320(aa), Penal Code)).

205 However, we agree with the Prosecution that the Judge seemed to have overlooked the application of the second step of the Anne Gan framework. The fact that the totality principle may have a boosting effect on individual sentences where they would otherwise result in a manifestly inadequate overall sentence (see [199(b)] and [199(c)] above) is a facet of the totality principle which, with respect, the Judge did not seem to consider. The Judge could, pursuant to the first step of the Anne Gan framework, have considered the sentence of 14 years’ imprisonment and 12 strokes of the cane as an appropriate provisional sentence for charge D1B1, but she should then have considered at the second step of the Anne Gan framework, whether there were any cumulative aggravating or mitigating factors that justified calibrating any of Ridzuan’s individual sentences upwards or downwards.

206 At the second step of the Anne Gan framework, it was incumbent on the Judge to have considered the totality of Ridzuan’s criminal wrongdoing. This would have entailed consideration of not just the two substituted s 326 charges, but the entire range of conduct that Ridzuan had been convicted of, including the Abuse Charges in order to correctly contextualise this offence and gauge the overall criminality it entailed and the appropriate sentence. In our judgment, in failing to do this, the Judge failed to consider the multiple cumulative aggravating factors in this case, and the fact that there are no material mitigating factors, and that this combination did make this one of the worst type of cases under s 326 of the Penal Code which justifies the maximum sentence of life imprisonment.

207 There are three critical cumulative aggravating factors in this case that make the nature of the crime “so cruel and inhumane” that it does come within the worst type of cases under s 326 of the Penal Code.

208 The first is the fact that there was a prolonged period of escalating abuse. The Deceased was subjected to a very slow and painful death at the hands of the respondents who burnt the victim to death by hot water over four cumulative incidents, in addition to the other painful and humiliating abusive acts, including treating the Deceased, their own child, like an animal by confining him in a cat cage. It is also undisputed that the Deceased did not receive any professional medical treatment until 22 October 2016, which is one week after Incident 1 and over three months after the first act of abuse (the Deceased was pinched with pliers) which took place in July 2016 (see [15] above).

209 The Judge reasoned that, to address the consequence of all four scalding incidents in the sentence on a charge for only one such incident “could be an excessive sentence for the particular charge” (GD at [189]; see [194] above). While this reasoning would be apt at the first step of the Anne Gan framework, the inquiry is different at the second step which assesses the totality of the offender’s culpability and criminality in order to properly contextualise the offence and its gravity. The punishment we are concerned with here is that for the charge D1B1 but to properly understand that charge, it had to be seen as coming at the end of a sustained period of cruelty and violence directed at the Deceased.

210 On the Judge’s reasoning as outlined in the previous paragraph, a “single act” causing death might justify a higher global sentence because the knowledge of the likelihood of death would presumably be more readily inferable by such a single act (such as by pouring boiling water on the Deceased ceaselessly in one incident until the Deceased dies). However, the Judge failed to appreciate that, at step two of the Anne Gan framework, it makes it worse that the abusive acts took place over a prolonged period of time rather than as a single incident. This is analogous to the observation in Public Prosecutor v Fernando Payagala Waduge Malitha Kumar [2007] 2 SLR(R) 334 at [48] that offenders who engage in a spree of offences could face “significantly higher” punishment than offenders who cheat a victim of the same cumulative sum in a one-off offence, as the offender in the former situation is hard put to credibly submit that his conduct was the result of a momentary indiscretion.

211 Second, in addition to the prolonged duration of the abuse, we agree with the Prosecution that the manner in which the offence was carried out was particularly cruel, as the Deceased was burnt extensively over his entire body, including sensitive parts of his body such as his face and genital area. It is damning that 75% of the Deceased’s body had been burnt by the end of Incident 4 (see [27] above). To put this in context, this is similar to the area of the victim’s body that had been burnt in Lim Ghim Peow v Public Prosecutor [2014] 4 SLR 1287 (“Lim Ghim Peow”), where the victim was doused in petrol and set ablaze. In this case, the manner of inflicting injury was perhaps even more cruel because it occurred over the course of a week, and each time entailed fresh wounds and injuries being added to what had already been endured.

212 As Associate Prof Loh Tsee Foong (“Assoc Prof Loh”; a senior consultant of KKWCH who was a member of the team who first treated the Deceased on 22 October 2016) observed, the Deceased would have been in severe distress from the time he was scalded until pain relief was administered to him in hospital. This can only be appreciated once all the acts are considered collectively. The final incident of scalding, for instance, was particularly cruel because it was done not for the first time on unblemished skin, but on skin that had already been repeatedly and brutally injured over three previous scalding incidents. Significantly, the Judge accepted on the medical evidence that, after Incident 2, the burns had left the Deceased’s nerves intact, which allowed the Deceased to “fully experience pain and suffering”, thereby causing the Deceased “intense pain” (GD at [201]). This means that the scalding in Incident 4 would have been particularly painful and distressing to the Deceased, once Incident 4 is viewed in the broader context of all the charges at step two of the Anne Gan framework.

213 Third, the offences were committed by the Deceased’s own parents against their young child, the Deceased. As correctly noted by the Judge, any parent or caregiver who breaches the trust and confidence reposed in him by abusing his child or ward will face the most severe condemnation of the law: AFR (CA) at [12]. Egregiously, both respondents knew that they were abusing their child, as they admitted in their statements. The offences were also committed in the confines of the Deceased’s own home, where outside detection would be difficult. This would also have had the effect of aggravating the fear felt by the Deceased, who had little means of escaping the abuse: see for example Public Prosecutor v Luan Yuanxin [2002] 1 SLR(R) 613 at [9].

214 Again, the full gravity of this factor can only be appreciated at step two of the Anne Gan framework. It is because it was the Deceased’s parents who were committing the offences against a defenceless young child in their own home that the abuse was able to continue for four whole months in an escalating fashion from July until it culminated in the final incident in October 2016.

215 At the hearing before us, counsel for Ridzuan, Mr Eugene Thuraisingam, made only three points to defend the sentence imposed on Ridzuan by the Judge: Ridzuan was young when the offences occurred, he had low adaptive functioning, and he is remorseful. We are unable to accept that any of these reasons constitute a material mitigating factor in this case. In our judgment, Ridzuan’s circumstances justify the maximum sentence of life imprisonment, because his case is devoid of any material mitigating factors.

216 First, Ridzuan was already a fully grown working adult aged 24 years when he committed the offences. Therefore, the argument that Ridzuan was “young” when he committed the offences does not move us.

217 Second, it appears from Ridzuan’s investigative statements that he was not even truly remorseful for the actions that the Deceased suffered.

(a) In Ridzuan’s cautioned statement to the Murder Charge, he sought to blame the Deceased for his acts, stating, “I do not know why my son’s behaviour is different from the rest. He likes to go against me.”

(b) In Ridzuan’s investigative long statement dated 25 October 2016 at 3.32pm, Ridzuan even stated that, when he poured hot water on the Deceased’s back, he “did not know if [the Deceased] was pretending to be in pain or not” [emphasis added].

218 In their investigative statements, both respondents attempted to justify their acts of extreme violence against the Deceased by contending that he was “stubborn” and “naughty”, and behaved differently from their other children (whom they did not abuse). The respondents even sought to blame the Deceased for their actions, saying that the Deceased was “stealing” food in their house (such as milk powder, milo powder and snacks). Not only is this an absurd excuse, considering that the Deceased was their young child, but it also ran counter to the evidence, in that Assoc Prof Loh testified that the Deceased was slightly malnourished when he was admitted to hospital after Incident 4.

219 Furthermore, Ridzuan cannot contend that he acted out of impulse or rashness. As explained at [208] above, the offences took place over a prolonged period of over some months, and the scalding incidents spanned a whole week. The respondents had many opportunities to consider and reconsider their abusive acts. Even when they knew that the Deceased was in a dire state of distress, they did not stop and take the Deceased for professional medical treatment, but continued to inflict torturous hurt on him.

220 To exacerbate matters, the respondents knew that there was an available and willing alternative caretaker for the Deceased – [Z] (see [12] above). As the Prosecution points out, [Z] wanted to continue caring for the Deceased and had offered to assume guardianship.

221 We also agree with the Prosecution that the Judge erred when she reasoned that the respondents’ “wholly inappropriate” responses after each scalding incident (such as applying medicated oil and baby powder on the Deceased) supported an inference that they did not fully comprehend the likelihood of death. In fact, it is evident from Ridzuan’s investigative statements that he knew the Deceased was already in a very bad state after Incident 2: he admitted that, after Incident 2, he had “told [the Deceased] that he was already in a bad condition” [emphasis added], and that he chose not to send the Deceased to the hospital, even after the Deceased’s nose was bleeding profusely after he had been punched by Ridzuan, because he “was afraid that [he] would be charged for child abuse” [emphasis added]. This suggests that the wholly inadequate attempts at treating the Deceased’s injuries stemmed from a concern with self-preservation, because he was concerned that sending the Deceased to a hospital would result in him being punished for his cruel abuse of his own child. In other words, Ridzuan was choosing to place his own self-interests above the Deceased’s.

222 We also disagree with the Judge’s view that there was some degree of remorse on the respondents’ part in choosing to send the Deceased to the hospital even though they knew that serious criminal consequences would follow for them (GD at [191]). We agree with the Prosecution that no weight should be placed on this, because the respondents had no choice, by that stage, but to send the Deceased to hospital, in a final desperate hope that the Deceased would not die. It bears emphasis that, even after the Deceased had collapsed after Incident 4, the respondents delayed sending the Deceased to hospital, and Ridzuan did not even want to call the ambulance initially because he was worried that the police would come as well (see [25] to [26] above). As the Prosecution points out, had the respondents chosen to prioritise the Deceased’s recovery and health, rather than their own self-preservation and interests, they would have taken the Deceased to the doctor much earlier. Further, when they did take him to the hospital, they went armed with a concocted tale of how this had resulted from an unfortunate accident.

223 Finally, Ridzuan’s reliance on his low adaptive functioning also holds no water, and we agree with the Prosecution that the Judge’s reliance on Ridzuan’s low intelligence to infer that he did not fully comprehend the likelihood of death is without basis (see GD at [191]). It is trite that mitigating value may only be attributed to an offender’s mental condition if the evidence establishes that the offender’s mental responsibility for his criminal act was substantially diminished at the time of the offence by reason of his mental condition. If the offender’s mental condition is not serious or is not causally related to the commission of the offence and the offence is a serious one, the sentencing principle of general deterrence may be accorded full weight (Public Prosecutor v BDB [2018] 1 SLR 127 at [72]; see also Public Prosecutor v Kong Peng Yee [2018] 2 SLR 295 at [65] to [72]).

224 Adaptive functioning refers to one’s ability to take care of himself or herself in daily life. The test which was administered on Ridzuan to test his adaptive functioning involved a series of questions, covering 10 domains (including communication, health and safety, and self-care), in which Ridzuan had to rate, on a scale of 0 to 3, his ability and initiative in performing various activities. In this case, as the Judge herself had found, while Ridzuan’s test results showed that he had “extremely low to low average” adaptive functioning (and his cognitive functioning was “at the borderline to low average”), Ridzuan was assessed not to meet the criteria for intellectual disability because he had the “ability to communicate, socialise, hold down various jobs and perform daily living skills” (GD at [153]). Indeed, the psychologist who conducted the intellectual assessment test on Ridzuan, Ms Leung Hoi Ting (“Ms Leung”), clarified on the stand that Ridzuan’s adaptive functioning is actually “adequate” and “proportionate to that of his age-matched peers”. In other words, Ridzuan’s adaptive functioning is higher than what his test scores suggested. Ms Leung reached this conclusion because, after conducting a clinical interview of Ridzuan after the intellectual assessment test was administered, it emerged that Ridzuan actually has the ability to perform these tasks, but he simply chooses not to do so and to rely on others instead:

225 In other words, while Ridzuan’s test score showed an extremely low to low average adaptive functioning, that was essentially because Ridzuan self-reported his actions in a way that did not accurately reflect his actual adaptive functioning in reality. There was no other expert evidence to support an inference that Ridzuan was unable to appreciate the consequences of his acts due to his purported low adaptive functioning or low intelligence.

226 In the circumstances, we are satisfied that the aggregate sentence imposed on Ridzuan is manifestly inadequate, and that the Judge failed to fully appreciate the evidence placed before her in their proper context such that the sentence imposed was also wrong in principle. Therefore, we allow the appeal in CCA 25 and amend Ridzuan’s sentence for charge D1B1 to life imprisonment. The sentences for the other charges which Ridzuan has been convicted on are to run concurrently with the sentence of life imprisonment for charge D1B1, pursuant to s 307(2) of the CPC.

227 There is one final point which the parties had initially not addressed us on. Ridzuan’s sentence for charge D1B2 (Incident 2) includes 12 strokes of the cane. It is well-established that a sentence of caning cannot be run “concurrently”: see Public Prosecutor v Poopathi Chinaiyah s/o Paliandi [2020] 5 SLR 734 at [42], affirming Public Prosecutor v Chan Chuan and another [1991] 1 SLR(R) 14 at [41]. This is because s 306(2) of the CPC, which empowers the court to run sentences concurrently, only applies to the sentence of imprisonment:

228 In this case, the Prosecution only submitted for an aggregate sentence of life imprisonment, both at the trial below and in this appeal, but the Prosecution has not sought a reduction of Ridzuan’s sentence for charge D1B2 to remove the 12 strokes of the cane imposed for that charge. Given the lack of clarity, the parties are to address us on this by way of further submissions (see [230] below).

229 In sum, we make the following orders with respect to the present appeals:

(a) the appeal in CCA 17 is allowed such that Azlin’s conviction on the four s 326 charges (charges C1B1 to C1B4) is replaced by her conviction on the alternative s 300(c) charge and the question of sentencing is adjourned for further submissions;

(b) the appeal in CCA 24 is allowed, but with the parties’ further submissions to be made on the sentence of caning; and

(c) the appeal in CCA 25 is dismissed.

230 Directions will be issued through the Registry for the further conduct of these appeals in respect of the remaining issues pertaining to sentencing.

231 Finally, we express our deep gratitude to Prof Goh for his assistance and his comprehensive submissions in this matter. This case raised some extremely difficult questions and we benefitted immensely from the characteristically careful and thorough submissions that were made by Prof Goh.