Lee Chez Kee v Public Prosecutor
[2008] SGCA 20

Case Number	:	Cr App No 9/2006
Decision Date	:	12 May 2008
Tribunal/Court	:	Court of Appeal
Coram	:	Choo Han Teck J; V K Rajah JA; Woo Bih Li J
Counsel Name(s)	:	Rupert Seah (Rupert Seah & Co) and B Uthayachanran (B Uthayachanran & Co) for the appellant; Lau Wing Yum, Vincent Leow and Tan Wee Soon (Attorney-General's Chambers) for the respondent
Parties	:	Lee Chez Kee — Public Prosecutor

V K Rajah JA:

Introduction

1          This is the appeal of Lee Chez Kee (“the appellant”) against his conviction of a charge of murder with common intention under s 302 read with s 34 of the Penal Code (Cap 224, 1985 Rev Ed) (“Penal Code”). This appeal raises difficult issues of law relating broadly to the areas of evidence and criminal common intention. While this appeal affords the opportunity to clarify these areas of law, I am fully aware that, at the end of the day, it is the result of the appeal which must remain at the forefront of the decision, given the nature of the offence and, more importantly, the irreversible punishment that will be visited upon the appellant should his appeal be dismissed. To facilitate understanding, I first set out the schematic arrangement of the contents of this judgment followed immediately by a brief background of the facts including the trial proceedings below:

INTRODUCTION……………………………………………………………......	1
BACKGROUND FACTS……………………………………………………......	4
The parties’ cases at the trial below……………………………………..........	5
The Prosecution’s case below…………………………………............	5
(1) The discovery of the deceased’s body.....................................	5
(2) Events before and after the robbery.........................................	7
(3) The appellant’s statements......................................................	10
(4) Too’s statements....................................................................	16
The appellant’s case below.....................................................................	17
The decision of the trial judge...........................................................................	20
The involvement of the appellant in the deceased’s murder......................	21
The common intention of the parties........................................................	24
Summary of the trial judge’s reasoning....................................................	26
ISSUES ON APPEAL............................................................................................	27
ADMISSIBILITY OF TOO’S STATEMENTS...................................................	28
Introduction and the proceedings below............................................................	28
The trial judge’s reasoning......................................................................	29
(1) General propositions...............................................................	29
(2) The scope of section 378(1) of the CPC.................................	30
(a) The legislative intent behind section 378(1) of the CPC..	31
(b) Contrary interpretation would lead to absurdity and                         inconsistency.................................................................	37
Summary of the trial judge’s reasoning....................................................	37
The parties’ submissions in the present appeal...................................................	38
The appellant’s submissions....................................................................	38
The Prosecution’s submissions................................................................	38
Analysis and discussion....................................................................................	39
The hearsay rule under the EA................................................................	40
(1) Conceptual basis....................................................................	40
(2) The prevailing judicial approaches...........................................	42
(a) Reference to the hearsay rule without regard to the EA..	42
(b) Implicit reference to the hearsay rule in the EA...............	43
(3) Problems with the existing judicial views and the way forward..	45
The hearsay rule under the CPC.............................................................	45
The scope of section 378(1)(b)(i) of the CPC.........................................	46
(1) Preliminary observations.........................................................	46
(2) Are the provisions in the UK Bill the exact equivalent of the                CPC amendments in 1976?.....................................................	48
(a) Amendments by the Select Committee showing                           difference between the CPC Bill and the UK Bill............	49
(b) Absence of provision in the UK Bill relating to co-                          accused’s confession in the CPC Bill.............................	52
(3) The legislative intent behind section 378(1)(b)(i) of the CPC....	56
(a) Relevant principles of intepretation.................................	56
(b) The rules of law governing the admissibility of                           confessions...................................................................	57
(c) Section 378(1) of the CPC subject to provisions on                           confessions in the EA....................................................	58
Does this interpretation of section 378(1)(b)(i) of the CPC lead to absurdity and inconsistency?...................................................................	60
Does section 378(1)(b)(i) of the CPC apply to admit Too’s statements in the present case?....................................................................................	61
(1) Section 30 of the EA..............................................................	62
(2) The rationale behind section 30 of the EA................................	64
(3) Can the rationale behind section 30 of the EA be extended to                  cover Too’s statements?.........................................................	67
Conclusion in relation to the admissibility of Too’s statements............................	68
INTERPRETATION OF SECTION 34 OF THE PENAL CODE......................	69
Introduction and preliminary observations.........................................................	69
Background on the Indian Penal Code..............................................................	73
The enactment of the Indian Penal Code.................................................	73
The interpretation of the Indian Penal Code.............................................	75
The correct interpretation of section 34 of the Penal Code................................	77
Typical requirements..............................................................................	77
Criminal act............................................................................................	77
Participation...........................................................................................	78
(1) The need for presence?...........................................................	79
(a) The Singapore courts’ present requirement....................	80
(b) Restatement of the requirement.....................................	82
(2) Participation in “twin crime” situations......................................	84
(a) The Singapore courts’ present requirement....................	84
(b) Restatement of the requirement.....................................	88
Proving the common intention.................................................................	88
(1) The Singapore courts’ present requirement..............................	88
(2) Restatement of the requirement...............................................	89
Common intention in “twin crime” situations............................................	90
(1) The Singapore courts’ present requirement..............................	90
(a) Intention to commit the offence committed.....................	91
(b) Common intention to commit the primary offence...........	93
(c) Other decisions.............................................................	95
(2) Restatement of the requirement...............................................	97
(a) The argument for a narrower reading of section 34 of                            the Penal Code............................................................	97
(b) The approach in other jurisdictions................................	101
(c) Analysis: The historical underpinnings of section 34 of                           the Penal Code.............................................................	103
(d) Analysis: The doctrine of common purpose and section                           34 of the Penal Code....................................................	108
(d) The relevance of sections 35 and 38 of the Penal Code.	118
(e) Analysis: Abetment provisions under sections 111 and                           113 of the Penal Code..................................................	120
In furtherance of the common intention....................................................	122
(1) The Singapore courts’ present requirement..............................	124
(a) Subjective knowledge...................................................	125
(b) Objective foreseeability.................................................	126
(c) Strict liability plus intention of actual doer to further                           common intention..........................................................	128
(d) Strict liability per se.......................................................	128
(2) Restatement of the requirement...............................................	130
(a) Requirement of English common law..............................	131
(b) Conformity with the abetment provisions of the Penal                           Code............................................................................	131
(c) Conformity with the common object provision in section                          149 of the Penal Code..................................................	133
(d) Subjective knowledge approach consistent with other                            jurisdictions and with universal principles.......................	135
The relationship of section 34 of the Penal Code with the other provisions.........	137
Summary of the correct interpretation of section 34 of the Penal Code..............	138
WHETHER THE APPELLANT WAS RIGHTLY CONVICTED OF THE CHARGE................................................................................................................	140
Link between the appellant and the events surrounding the deceased’s death.....	141
Whether the appellant knew that the deceased may likely be killed....................	142
Events before the robbery......................................................................	142
Events after the robbery.........................................................................	143
Events during the robbery.......................................................................	144
Section 111 or section 113 of the Penal Code..................................................	147
CONCLUSION......................................................................................................	148

A more detailed exposition of the relevant facts and issues now follows.

Background facts

2          More than a decade ago on the morning of 14 December 1993, two police officers found Prof Lee Kok Cheong (“the deceased”) lying dead in the master bedroom of his house at 20 Greenleaf Place, Singapore. Police investigations revealed that three persons, one of whom was the appellant, were involved in the death of the deceased. However, the appellant remained at large for almost 13 years and was apprehended in Malaysia only in 2006. Prior to the appellant’s arrest, his accomplices, Too Yin Sheong (“Too”) and Ng Chek Siong (“Ng”), were convicted and sentenced in 1998 for their involvement in the deceased’s death. Too was convicted of murder and sentenced to suffer death, whereas Ng was convicted and sentenced to a total of eight years’ imprisonment and ten strokes of the cane for one count of robbery, five counts of theft and 11 counts of cheating, all with common intention.

The parties’ cases at the trial below 

3          After the appellant’s arrest in 2006, he was extradited to Singapore to stand trial in the High Court on the following charge (“the Charge”):

That you …

between 12.00 p.m. on the 12th day of December 1993 and 7.03 a.m. on the 14th day of December 1993, at 20 Greenleaf Place, Singapore, together with one Too Yin Sheong and one Ng Chek Siong, in furtherance of the common intention of you all, did commit murder by causing the death of [the deceased], male then 54 years old, and you have thereby committed an offence punishable under section 302 read with section 34 of the Penal Code, Chapter 224.

4          During the course of the trial, the trial judge heard full and comprehensive arguments from both sides. As it is trite law that an appellate court should not disturb the findings of fact of a trial court without clear evidence that such findings are unsupported by the evidence, I propose to set out the primary contentions advanced by either party before the trial judge and the reasons for his decision. I should also mention that as the trial judge has ably summarised the contentions presented in his grounds of decision (see PP v Lee Chez Kee [2007] 1 SLR 1142 (“Lee Chez Kee”)), I will gratefully adopt, with minor modifications, his summaries.

The Prosecution’s case below

(1)        The discovery of the deceased’s body

5          On the morning of 14 December 1993, two police officers (“the police officers”) found the deceased’s body in the master bedroom of his house. The police officers had been instructed by their operations room staff to proceed to the deceased’s house following a call received from the deceased’s neighbour at about 7.00am that day. The neighbour had contacted the police when she noticed that something was amiss at the deceased’s house. The lights on the ground floor remained switched on, the main sliding door and main gate of the house were ajar and the interior light of the car parked at the porch had also been left on.

6          According to the police officers, they noticed, on their way into the house, that the glove compartment of the deceased’s car, a red Honda Concerto, had been opened and appeared to have been rummaged. Upon entering the deceased’s house, they also observed that the hall on the first storey and two rooms on the second storey had been ransacked. The deceased’s body was found in the third and last room on the second storey in a supine position with a pillow placed over his face. His hands were above his head and his wrists tied together with a white electrical cord. His feet were bound at the ankles with a black belt. When the pillow was lifted up, the police officers observed that the deceased’s denture protruded from his mouth.

7          Deputy Superintendent Low Hock Peng (“DSP Low”), who was then a senior investigating officer attached to the Criminal Investigation Department (“CID”), attended at the scene shortly thereafter. DSP Low found a bent knife, with a length of about 18cm from its tip to its hilt, beneath the deceased’s body. A chopper was also found underneath some papers on the study table in the first storey hall.

8          The pathologist, Dr Paul Chui, gave evidence that a black electrical cord was found across the front of the deceased’s neck. The cord had not been knotted and was not wound completely around the back of his neck. A stab wound was observed at the left side of the deceased’s neck. Bruising to the right side of the deceased’s face was noted. More specifically, in his report, Dr Chui noted some 18 external injuries caused to the deceased on his head, neck, upper limbs, abdomen, back and lower limbs. It is without any doubt that the deceased was subjected to callous abuse and brutish violence before he met his death.

9          Dr Chui estimated that death had occurred about one to two days before 14 December 1993, when the deceased’s body was found. The post-mortem examination revealed that the cause of death was asphyxia due to strangulation, which had been effected using the black electrical cord found around the deceased’s neck. The stab wound sustained at the deceased’s left anterior neck was not an acutely fatal injury, and was more consistent with an injury caused by the knife, which was smaller and narrower than the chopper. Though the knife was bent when it was found, Dr Chui’s evidence was that he would not have expected this bending to have occurred in the course of causing the stab wound to the deceased’s neck. The force required to bend the knife would have been considerably greater than the force that would have been required to inflict the stab wound that was found. According to Dr Chui, it was possible for the stab and strangulation wounds to have been inflicted by one person.

(2)        Events before and after the robbery

10        The deceased’s brother, one Lee Kok Fatt (“Lee”), gave evidence that he and his family had gone to the deceased’s house on the morning of 12 December 1993 (“the material date”), at about 10.00am. The purpose of their visit was to collect Chinese New Year red packets from the deceased to distribute these on his behalf to Lee’s children and their other relatives during the Chinese New Year as the deceased would be away in England then.

11        According to Lee, the deceased had specifically instructed Lee’s wife to visit him between 10.00am and 11.00am on the material date as he was expecting friends later that evening. When they met on the morning of the material date, Lee asked the deceased which friends he was expecting, to which the deceased replied that it would be pointless to tell Lee who “his two friends” were since Lee would not know them. Lee and his family left the deceased’s house sometime before 12.00pm. That was the last time Lee saw or spoke to his brother.

12        On the next day, ie, 13 December 1993, a number of Network Electronic Transfer System (“NETS”) transactions were executed using the deceased’s Cash-On-Line (“COL”) card. The appellant was linked to a number of these transactions.

13        The deceased’s COL card was first used to purchase a number of items at Parkway Parade Shopping Centre (“Parkway”). The first transaction took place at Jay Gee Enterprise Pte Ltd (“Jay Gee Enterprise”), where the deceased’s card was used to purchase three pairs of “Levi’s” jeans, a T-shirt and a belt. One of the sales assistants then on duty at Jay Gee Enterprise (“Ms Lim”) gave evidence in court that these purchases were made by three male Chinese. From their use of language, Ms Lim assumed that the three males were Malaysians. The store records also documented that a sale had been made to a male Malaysian of about 20 years of age.

14        Ms Lim had, in March 1994, positively identified a picture of the appellant (“picture B”) as one of the three men in question. However, she had been unable to remember if Too or Ng were the other two persons present with the appellant on that day. According to Ms Lim, though she had spoken to all three persons, she remembered the person in picture B, ie, the appellant, particularly vividly as she had spoken “quite a lot” with him. Ms Lim also gave evidence that the male in picture B was the person who had handed her the COL card with which the purchases were made and was also the person who had keyed in the personal identification number (“PIN”) for the card.

15        Ms Lim additionally testified that the person shown in picture B and the two other accompanying Chinese males returned to the store again the next day, ie, 14 December 1993, to exchange some of the purchases they had made. Though Ms Lim admitted that she was no longer able to affirmatively identify any of the three Chinese men who had been present at the store on those two occasions, she affirmed on re-examination that on the date when she identified the appellant in 1994, she clearly remembered that the Chinese male she identified had been present at the shop on two dates, namely, 13 and 14 December 1993. Ms Lim further testified that she had, in the course of identifying the photographs in 1994, informed the investigating officer, DSP Low, that the appellant and his two companions had returned to the shop again on 14 December 1993.

16        Ms Lim’s evidence regarding the appellant’s return to the shop on 14 December 1993 was not confirmed by the evidence of DSP Low. According to DSP Low, Ms Lim did not mention this additional fact when she identified the appellant as the user of the deceased’s COL card on 13 December 1993. Apart from Ms Lim, the Prosecution also called one of her former colleagues, See Ching Li Veronica, who had also attended to the three Chinese customers on 13 December 1993. This witness was unable to recall the identities or the number of persons who were present when those purchases were made.

17        Apart from the purchases made at Ms Lim’s shop, one pair of black men’s “Balene” brand shoes and three pairs of “Dr Marten” brand shoes were later purchased from two other shops in Parkway. The sales assistant from the shop where the “Dr Marten” shoes were purchased (“Yeo”) gave evidence that three Chinese men had entered the shop and had tried on each of the three pairs of shoes. Payment had been made by a male Chinese whom Yeo subsequently identified as Too. Yeo was unable to identify the other two Chinese men who were together with Too. Apart from the purchases at Parkway, the deceased’s COL card was also used to purchase a pair of “Reebok” sports shoes and three pairs of socks at a store in City Plaza. The shop attendant was unable to remember who had effected the NETS transaction or whether that person had been alone or, if accompanied, with how many others.

18        On the following day, ie, 14 December 1993, further purchases were made using the deceased’s COL card. Four T-shirts were purchased from a shop in People’s Park Complex. The shop assistant (“Irene Tan”) gave evidence that two male Chinese had entered the shop, and that one of the male Chinese had picked out the relevant items and paid for them by a NETS transaction using the deceased’s COL card. Irene Tan, in June 1994, identified Too as the person who had made the payment and Ng as Too’s companion on that day.

19        Though the records of the deceased’s bank accounts documented a number of other NETS transactions that were executed on 13 and 14 December 1993, no witnesses were called to give evidence regarding those transactions. The Prosecution informed the court in the course of the trial that it would not be seeking to rely on those transactions. The trial judge therefore paid no regard to the additional NETS transactions exhibited in the statement for the deceased’s Post Office Savings Bank account during the course of those two days.

(3)        The appellant’s statements

20        Following his arrest in Malaysia and extradition to Singapore, the appellant made a total of four written statements (under the Criminal Procedure Code (Cap 68, 1985 Rev Ed) (“CPC”)). During the trial, the Prosecution sought to introduce these statements in evidence. No challenge to admissibility was mounted by the appellant and the trial judge accordingly admitted these four statements, which were as follows:

(a)        a statement made under s 122(6) of the CPC dated 18 February 2006 (“the appellant’s cautioned statement”);

(b)        a statement made under s 121 of the CPC dated 21 February 2006 (“the appellant’s first long statement”);

(c)        a second statement made under s 121 of the CPC dated 22 February 2006 (“the appellant’s second long statement”); and

(d)        a third statement made under s 121 of the CPC dated 27 February 2006 (“the appellant’s third long statement”).

21        In some of the material portions of these statements, Too is referred to as “Nelson” and Ng as “Koo Neng” or “Koo Nerng” (tortoise egg):

The appellant’s cautioned statement

I went there [the deceased’s house] with Nelson and Koo Neng. I was asked by Nelson to join him to rob his relative. I did not know what place was that. Nelson and I tied up a man on the second floor. After that, Nelson and I searched the house for valuables. I cannot recall what did I take. After taking the valuables, I went to the hall on the ground floor to wait for Nelson. I did not know the man died. I also cannot recall what I was given after the robbery.

The appellant’s first long statement

Question:    What do you know about the death of Prof Lee Kok Cheong, which occurred between 3.00 pm on 12 Dec 1993 and 7.03 am on 14 Dec 1993 at No. 20 Greenleaf Place, Singapore?

Answer:

            At the beginning of 1993, I came to know the two friends, ‘Koo Nerng’ and ‘Nelson’. These two came to visit me before the present case took place. We went for tea, during which time ‘Nelson’ told me that he was going to a ‘relative’s place to move something. Actually, ‘Nelson’ revealed that he was going to rob this ‘relative’.

2          The very next day, ‘Koo Nerng’, ‘Nelson’ and myself went to the place which I was not familiar with. I am not sure of the date now. On arrival, ‘Nelson’ and I entered a terrace house. It was a residence of ‘Nelson’s so-called ‘relative’. We went in and chatted. The victim served us tea. During the chat, ‘Nelson’ told me that the victim had a lot of antiques. Thereafter, I requested the victim to take me upstairs to view his stuff. However, he told me to wait downstairs and did not bring me up. Earlier on, before my request and while we were still chatting, ‘Nelson’ had gone to the kitchen to get hold of a knife. ‘Nelson’ was more familiar with the place and when he emerged from the kitchen, he went to confront the victim with a knife. The victim was taken up to the second storey. Together with ‘Nelson’, the two of us took him upstairs. He was rather big sized. Next, I bound his hands with something. I cannot remember what I had used to bind him. After this, ‘Nelson’ and myself searched his bedroom. After searching the premises upstairs, I went downstairs and searched the place downstairs. After I had finished, I called out [to] ‘Nelson’ to leave. He was still upstairs. He came downstairs and we left together.

3          From there, we went straight to Katong, at which time, ‘Nelson’ suddenly produced a cash withdrawal card. There were two automatic cash machines at that place outside the Katong Shopping Complex. ‘Nelson’ slotted in the card he had shown me and he proceeded to withdraw a sum of either $1000/- or $200/-. The cash was withdrawn successfully. When he was pressing the buttons at a machine, I asked him what he was doing. He replied that the card belonged to the victim and he was withdrawing money with the card. After the cash withdrawal, the money was split among the three of us, namely, ‘Nelson’, myself and ‘Koo Nerng’. I think the money was divided equally among us. After this, I went back to my residence in Joo Chiat area. The other two went back to their respective residences.

4          The following day, I telephoned ‘Nelson’ and asked him to meet me. ‘Nelson’ and ‘Koo Nerng’ came and met me at Joo Chiat. After this, we went shopping using the card which was still with ‘Nelson’. We went to this shopping place in Marine Parade. I cannot remember what the place is called. We bought garments and shoes. After the shopping, we split the stuff and went separate ways. ... After we parted ways, I did not make contact with them. I continued my work as usual in the days that followed. …

The appellant’s second long statement

...

Q3:      You are now shown two photographs numbered ‘1’ and ‘2’ of male persons. Do you know any of them?

A3:       Photograph number 1 shows ‘Nelson’ (Recorder’s note: refers to Too Yin Sheong). The person in photograph 2 is ‘Koo Nerng’ (Recorder’s note: refers to Ng Chek Siong).

Q4:      You have mentioned in paragraph 1 of your statement recorded on 21 Feb 2006 that ‘Nelson’ revealed that he was going to rob his ‘relative’. How long have you known ‘Nelson’ already when he revealed that to you and did he reveal to you how he was going to carry out the robbery?

A4:       I have known him for a few months then. During our discussion about the robbery, I said that we would tie up the victim before getting his things.

…

Q10:    You have mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that you requested the victim to take you upstairs to view his stuff. Can you elaborate why and what you meant by ‘stuff’?

A10:     I meant his antiques. The intention was to bind him or tie him up when he [went] upstairs so that we could rob him.

Q11:    You have mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that ‘Nelson’ had gone to the kitchen to get hold of a knife. Can you describe the knife?

A11:     It is a fruit knife. The knife is small, about half a foot long. It has a small handle.

…

Q13:    You have mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that ‘Nelson’ and you searched the victim’s bedroom and the premises upstairs. Can you elaborate [which] were the premises that you and ‘Nelson’ searched and was there anything taken either by you or ‘Nelson’?

A13:     Besides the bedroom, ‘Nelson’ and I [had] also searched another bedroom upstairs. We took some valuables from both the bedrooms. I cannot recall what the valuables [were].

Q14:    You have mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that after searching the premises upstairs, you went downstairs and searched the place downstairs. Can you elaborate on the victim’s condition when you left?

A14:     He was still moving when I came downstairs. ‘Nelson’ had asked him questions in English and he [the deceased] had replied to him [“Nelson”].

…

Q17:    You have mentioned in paragraph 3 of your statement recorded on 21 Feb 2006 that ‘Nelson’ withdrew some money from the automatic cash machines at Katong using a cash withdrawal card belonging to the victim. Can you elaborate how many times the victim’s cash withdrawal card was used to withdraw money and how did you all know the card PIN number?

A17:     All in, I think the card was used to withdraw cash three to four times. ‘Nelson’ had asked the victim for his PIN number. I had not known that ‘Nelson’ [had] got hold of the victim’s card until then.

Q18:    You have mentioned in paragraph 4 of your statement recorded on 21 Feb 2006 that you telephoned ‘Nelson’ the following day and asked him to meet you. Can you elaborate why you asked ‘Nelson’ to meet you?

A18:     I had asked him to meet me for tea at Joo Chiat, the place where I was residing. My idea was to have tea with him and during the chat, I ... asked him whether the victim’s card could still be used. I think he said that he was not sure but then he suggested that we could try it out. I do not remember much about it but we later went shopping. We managed to buy things with the card.

…

Q21:    Besides ‘Nelson’ who got hold of a knife in the kitchen, I put it to you that you also got hold of a knife from the kitchen. What have you got to say?

A21:     Actually, soon after ‘Nelson’ got hold of a knife from the kitchen, he had gone into the hall and passed the knife to me. With this knife, I confronted the victim with ‘Nelson’ looking on close by.

Q22:    You have mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that ‘Nelson’ confronted the victim with a knife when he emerged from the kitchen. Your answer for Q21 states that you confronted the victim with ‘Nelson’ looking on close by. Can you explain the difference?

A22:     Actually, after I took the knife from ‘Nelson’, I ... pointed it at the victim’s stomach region. I pushed him and told him to go upstairs. ‘Nelson’ was looking on then.

Q23:    Can you explain how the victim sustained the injuries on his body?

A23:     I ... landed a few punches on him. I cannot remember exactly on which part of his body I had hit. When we were downstairs, that was what I did. ‘Nelson’ did not hit him then. When we were upstairs, I bound his hands. We did not hit him when we were upstairs.

…

Q25:    Did the victim put up any struggle during the hold-up?

A25:     Yes. He did. He put up a struggle before and after I tied him up.

The appellant’s third long statement

…

Q28:    Your answer for Q6 relates to the arrangement made between ‘Nelson’, ‘Koo Nerng’ and you to rob ‘Nelson’s relative. You have also mentioned in paragraph 2 of your statement recorded on 21 Feb 2006 that ‘Nelson’ had gone to the kitchen to get hold of a knife. Was this also part of the arrangement?

A28:     Yes. I believe so. I think ‘Nelson’ mentioned about getting a knife. We had gone there empty-handed. Either ‘Nelson’ or myself have mentioned about threatening the victim with a knife.

…

Q30:    Your answer for Q25[in the appellant’s second long statement] states that the victim put up a struggle before and after you tied him up. What did you and ‘Nelson’ do when the victim was struggling?

A30:     When he was struggling, he was also talking to ‘Nelson’ in English. I told him to keep quiet and to stop moving. ‘Nelson’ was beside us at that time. I cannot recall what ‘Nelson’ was doing then.

Q31:    You have mentioned in your statement recorded on 21 Feb 2006 that you bound the victim’s hands with something. Besides the hands, did you or ‘Nelson’ tie up other parts of the victim’s body?

A31:     Perhaps the legs. I cannot remember what was used by ‘Nelson’. After I had tied the victim’s hands, I put him on the bed and I proceeded to search the bedroom. ‘Nelson’ also searched the room. After the search, I went downstairs.

…

Q34:    Do you have anything else to say?

A34.     I now recall that ‘Nelson’ seemed to have taken something to cover the victim’s face. It was during the search of the victim’s bedroom that ‘Nelson’ did so. No force was exerted. The thing was merely put on his face. It was something taken from the bed itself, something that was light. After the search, I went downstairs. When I was downstairs, I think I [had] called out to ‘Koo Nerng’ and asked him to enter the house. I do remember now that I have indeed asked ‘Koo Nerng’ to enter the premises. I told him to help in searching the premises downstairs and to move the things. ‘Koo Nerng’ came in for a short time. He stayed downstairs and did not go up. He looked around the place and asked where the victim was. I then called out to ‘Nelson’ to leave. That was roughly what happened and then we left. That’s all.

Q35:    You have mentioned in A34 that ‘Nelson’ covered the victim’s face with something light taken from the bed. Was this part of the arrangement?

A35.     This was not part of the arrangement. I do not know why ‘Nelson’ had done this. Perhaps because the victim kept talking and it disturbed him.

[emphasis added in bold italics]

(4)        Too’s statements

22        It is also significant to note that (for reasons to follow) in the proceedings below, as part of its case, the Prosecution additionally sought to admit four written and two oral statements of Too, all of which were made before his execution (collectively, “Too’s statements”). The Prosecution had submitted before the trial judge that these statements were admissible under s 378(1)(b)(i) of the CPC, notwithstanding the fact that Too was an accomplice since s 378(1)(b)(i) provided for the admission of out-of-court statements made by a dead person. In the course of his written grounds of decision, the trial judge acknowledged the “somewhat unusual prospect” of trying the appellant for a charge of common intention when both Too and Ng, his accomplices, were not available as witnesses (see Lee Chez Kee ([4] supra) at [3]). Too had been executed in April 1999 and Ng could not be located after having been repatriated to Malaysia in October 2003. It is only necessary at this juncture to mention that the trial judge admitted Too’s statements.

23        The above formed the gist of the Prosecution’s case before the trial judge.

The appellant’s case below

24        As the trial judge was satisfied that the evidence adduced by the Prosecution had established a prima facie case against the appellant, the trial judge called upon the appellant to enter his defence. The appellant elected to give evidence, and was the defence’s only witness. Notably, he did not, in the course of his oral testimony, dispute his presence at the deceased’s house on 12 December 1993, and admitted that he had gone to the deceased’s house with the intention of committing robbery. The main bone of contention surrounded his specific involvement in the stabbing and asphyxiation of the deceased.

25        According to the appellant, he had first become acquainted with Too when he was working at Apollo Nightclub in or about 1990. He had in turn met Ng when they sold durians together. The appellant claimed that Too and Ng were closer friends as they often spent time together whilst the appellant spent most of his time with his girlfriend.

26        According to the appellant, Too had asked him out to a coffee shop near the appellant’s house and had asked if he was interested in robbing Too’s wealthy relative. In the course of giving his testimony, the appellant clarified that whilst he had initially stated in his first long statement that Too had asked him out to the coffee shop on the day preceding the robbery, Too’s invitation had in fact occurred on the same day as the robbery. During examination by his counsel, the appellant initially claimed that no details or plans about the robbery had been discussed, and that Too had merely mentioned that they would rob his relative. However, he subsequently conceded during cross-examination that he had in fact asked Too and Ng about his role in the robbery. Nevertheless, he insisted that they had not talked in detail as to how they were going to rob the deceased. Again, while he initially testified during cross-examination that there was no discussion about either tying the deceased up or using a knife to threaten the deceased, or even about what they were to do if the deceased resisted or retaliated, he subsequently conceded, upon being shown his first long statement, that there was indeed a pre-arranged plan to tie up the deceased and to threaten him with a knife.

27        The appellant acknowledged that he had also asked Too whether he was worried that his “relative” would be able to recognise him. Too had supposedly been very confident, and had told the appellant “to believe him”. The threesome then got into a car driven by Ng and proceeded to the deceased’s house. The appellant affirmed that he had quarrelled with Too during the journey to the deceased’s house. The quarrel had probably arisen because he, rather than Too, was afraid of being recognised. However, as Too was confident, the appellant trusted him. During this quarrel, Ng mentioned that he did not want to go in. They did not bring any weapons with them in the car.

28        When they arrived at the deceased’s house, it was still sometime in the afternoon. Ng dropped the appellant and Too off a few houses away from the deceased’s house, and remained in the car. When they arrived at the front gate of the deceased’s house, Too rang the doorbell. The deceased came out to meet them, following which Too conversed with the deceased in English before they all proceeded into the house. It was only after they entered the house that Too introduced the appellant to the deceased. The appellant claimed that he was unable to understand how Too had introduced him as the entire introduction had taken place in English, a language which the appellant had only a rather rudimentary grasp of.

29        The deceased then invited Too and the appellant into the living room and served them drinks. The appellant claimed that Too later passed him a knife which Too had taken from the kitchen, whereupon the appellant then employed it to threaten the deceased. Upon hearing that he was being robbed, the deceased put up a struggle. On the appellant’s own account, he then stabbed the deceased’s left abdominal region twice, but the knife did not penetrate the deceased’s body. He also rained a few blows on the deceased. The appellant confirmed that the knife he had used was the bent fruit knife which was subsequently found under the deceased’s body.

30        After that, the appellant and Too led the deceased upstairs into the master bedroom. Too then tied the deceased’s legs while the appellant tied his hands. According to the appellant, the deceased became “very obedient” once he had been tied up. The appellant then left the deceased lying on the bed and proceeded to search the house for valuables. Too remained in the vicinity of the room and conversed with the deceased in English.

31        After searching upstairs for what the appellant claimed was a “few minutes”, he went downstairs. On his way downstairs, the appellant saw the deceased for the last time before he left the house. According to the appellant, the deceased was still alive at that time. He did, however, recall seeing Too covering the deceased’s face with a pillow when he was going downstairs. The appellant then summoned Ng into the house to help in the search for valuables. After they searched the deceased’s car, they called for Too and left the deceased’s house. Too was alone upstairs throughout this period.

32        According to the appellant, once they left the deceased’s house, Ng drove them back to Katong. Somewhere along the way Too alighted by an automated teller machine. It was only when the appellant followed Too and Ng that he realised that Too had taken the deceased’s COL card. Too used the deceased’s COL card to make a withdrawal, and the three of them divided the money among themselves. On the next day, the appellant called Too to ask him whether the deceased’s COL card could still be used. The three of them then met up and made some purchases using the deceased’s COL card. The appellant did not see either Too or Ng again. According to the appellant, he continued travelling between Singapore and Malaysia using his passport until he learnt that Too had accused him of killing the deceased.

33        Significantly, though the appellant claimed that he had thought that the deceased was still alive, he conceded during cross-examination that he had not, at any point after leaving the deceased’s house, asked Too what had happened to the deceased, nor had they discussed what they would do if the deceased freed himself and proceeded to make a police report. Notwithstanding his alleged belief that the deceased was still alive, it had not occurred to the appellant that the deceased might have, between the material date and the next day, reported the theft of his COL card to the authorities. According to the appellant, it had not occurred to him, after they had left, that the deceased could have escaped from his house or shouted for help since his mouth was not gagged, or that he could have made his way downstairs even though his legs and hands were tied.

The decision of the trial judge 

34        The appellant was convicted of murder and accordingly sentenced to the mandatory death penalty. The trial judge’s reasoning can be shortly summarised. First, the trial judge stated that the evidence, considered in its totality, gave rise to the irresistible inference that the deceased had been murdered in furtherance of the common intention among Too, the accused and Ng to commit robbery (see Lee Chez Kee ([4] supra) at [59]). To begin with, the Prosecution had established beyond a reasonable doubt that the deceased had died as a result of asphyxiation with the black cord and, further, that the event of strangulation had occurred in the course of the robbery committed on the material date.

The involvement of the appellant in the deceased’s murder

35        The trial judge next found that the material circumstances surrounding the robbery on the material date were capable of establishing a coherent and irresistible inference that the appellant in question was guilty as charged. The trial judge placed considerable emphasis on the events which had occurred after the robbery in having “created an indelible link between the [appellant] and the tragic events which had occurred at the deceased’s house” (at [61]). Apart from the appellant’s own admission that he had shared in the spoils of the robbery, the trial judge accepted that the independent evidence adduced by the Prosecution also identified the appellant as having been party to the subsequent usage of the deceased’s COL card. Further, the trial judge also accepted Ms Lim’s evidence that the appellant had in fact been present at her store on 13 December 1993 and had used the deceased’s COL card to make the relevant purchases. Accordingly, the trial judge found that the evidence linking the appellant with the robbery at the deceased’s house, coupled with his concordant finding that the deceased’s death had occurred in the course of this robbery, gave rise to a prima facie inference that the appellant “had been involved in the deceased’s death” (at [62]). I pause to note that the degree of involvement by the appellant was never quite clearly explained clearly by the trial judge at this point. In any case, according to the trial judge, this inference was further supported by s 116 of the Evidence Act (Cap 97, 1997 Rev Ed) (“EA”), in that the appellant’s possession of the deceased’s items after his robbery and death gave rise to the presumption that the appellant was “involved in the deceased’s death”.

36        In the trial judge’s view, this inference was also supported by the events which had taken place before the robbery (at [65]):

On the accused’s own admission, he was, prior to the robbery, afraid of being recognised by the deceased. After all, he was with Too and Too was well known to the deceased. Identification of Too would inevitably lead to the identification of the accused. These fears were sufficiently compelling to cause the accused to get into a quarrel with Too during the journey to the deceased’s house. Whilst the accused contended that his fears were quelled by Too’s confidence, I found this assertion to be little more than an artificial construct manufactured by the accused to remove his otherwise undeniable motive for killing the deceased. On the accused’s own admission, he was not particularly well acquainted with Too or Ng. His seeming trust in Too, and his sanguine expectation that Too would not implicate him if arrested, was wholly inconsistent with the admittedly superficial nature of their acquaintance.

Making reference once again to the appellant’s conduct after the robbery, the trial judge found that such conduct again supported his earlier finding that the appellant had been party to the fatal injuries (note the plurality employed) inflicted on the deceased (at [66]):

In my view, the accused’s conduct following the robbery augmented this compelling inference that he had been party to the fatal injuries afflicted on the deceased. The complete lack of any discussion regarding what would happen when the deceased was freed, coupled with the calm and calculated manner in which they went about repeatedly exploiting the deceased’s COL card on the day following the robbery and the accused’s continued travel between Singapore and Malaysia in the years following the robbery, could only support the conclusion that the accused and his accomplices were not worried about being identified because they knew that the deceased was dead. Without the knowledge that the deceased had died, the likelihood of their identification and arrest would have appeared particularly imminent since the deceased knew Too personally and would, hence, have been able to positively identify him. Given these circumstances, it was wholly incredible that the accused would have had no regard whatsoever to the possibility – or indeed, the certainty – of the deceased making a report against Too when he was freed. The accused’s attempt to portray himself as a passive follower of Too was directly rebutted by his own admission that he had argued with Too because of his fear of being recognised. This, coupled with the accused’s proactive involvement in clarifying the modus operandi of the robbery prior to its occurrence, could only lead one to conclude that the accused was not a mere robber but someone prepared to annihilate his victim.

37        Once again, the trial judge found that the appellant was not a mere robber but someone prepared to annihilate the deceased; he, however, did not yet find that the appellant had in fact inflicted the fatal strangulation on the deceased. The trial judge also apparently discounted the appellant’s evidence in relation to his involvement with the events which had taken place upstairs in the deceased’s house prior to his murder. According to the trial judge, the appellant’s evidence at trial that he had merely observed Too placing the pillow on the deceased’s face from a distance while he was going downstairs was clearly inconsistent with his prior evidence in his third long statement. According to the appellant’s third long statement, he had only left the master bedroom and gone on to search the other rooms after Too had done so. Based on Dr Chui’s expert evidence, the act of placing the pillow on the deceased’s face was an event which occurred after the deceased had been strangled to death. Thus, according to the trial judge, the appellant’s provision of this information was, hence, further evidence that he had been present in the room when the deceased’s life was brutally ended.

38        Finally, the trial judge held that the appellant’s involvement in the deceased’s death was further supported by Too’s statements. In his view, Too’s statements, when considered together with the rest of the independent evidence which the Prosecution had adduced, served the limited function of reinforcing the already compelling inference which the latter had given rise to, which was that the appellant was “inextricably involved in the deceased’s death” (see Lee Chez Kee at [68]). At this point, thus, it appears that the trial judge was wholly convinced that the appellant was “involved” in the murder in that the appellant had actually done the act which killed the deceased. However, the trial judge, with respect, does not quite clearly state this conclusion, making allusions instead to the appellant’s supposed “involvement” in the murder without quite explaining explicitly the degree of such involvement.

The common intention of the parties

39        In what seems to me as almost a secondary point, the trial judge next turned to the common intention of the parties pursuant to s 34 of the Penal Code. After considering the judicial interpretation of s 34 of the Penal Code, the trial judge stated that, in his view, it was not necessary to establish the identity of the person who actually strangled the deceased as the requirements of s 34 were satisfied on the facts. These requirements were: (a) the existence of a common intention among all the persons who committed the criminal act; (b) the furtherance of the common intention by the criminal act; and (c) the participation in the criminal act by those sharing the common intention. On the facts, the trial judge found these requirements to be fulfilled. First, there was evidence of the existence of a common intention between the parties to rob. Secondly, the murder was apparently in furtherance of such a common intention, although this point was dealt with only at a later paragraph of the trial judge’s grounds of decision. Finally, the requirement of participation was sufficiently made out given the appellant’s presence at the deceased’s house when the murder occurred. By these findings, the trial judge held that he would have convicted the appellant of the Charge (see [3] above).

40        In any event, the trial judge also opined that even if one were to accept that it was Too, and not the appellant, who was solely responsible for the deceased’s death, the rest of the evidence showed that Too’s conduct to this effect was in furtherance of their common intention to rob the deceased. This, as I mentioned in the preceding paragraph, appeared to go towards fulfilling the second requirement under s 34 of the Penal Code. It is worthwhile to reproduce what the trial judge had written (Lee Chez Kee at [72]):

Finally, even if one were to accept – which I did not – that it was Too, and not the accused, who was solely responsible for the deceased’s death, the rest of the evidence showed that Too’s conduct to this effect was in furtherance of their common intention to rob the deceased. The accused himself conceded that he had, when the deceased initially protested downstairs, attempted to stab the deceased in order to quell his acts of resistance. According to the accused, at the time when he used the knife to threaten the deceased, he had been prepared to use the knife on the deceased if the latter had struggled or retaliated. This admission by the accused himself amounted to positive evidence that it would not have been inconsistent with or extraneous to his common intention with Too to cause harm to the deceased. [emphasis added in bold italics]

It would thus appear that the “common intention” here has suddenly transformed from the common intention to rob to the common intention to cause harm to the deceased. So did the trial judge ultimately find that the murder of the deceased was in furtherance of the parties’ common intention to rob the deceased? With respect, I am unable to say with certainty what the trial judge found in relation to this second requirement of s 34 of the Penal Code, unless one were to conclude that the act of strangulating the deceased was in furtherance of the common intention to cause harm to the deceased, which was in turn was in furtherance of the common intention to rob.

41        Ultimately, it appeared that the use of s 34 of the Penal Code was not crucial to the trial judge’s decision. The trial judge was wholly convinced of the appellant’s involvement in the infliction of the injuries that resulted in the demise of the deceased, as the penultimate paragraph in his grounds of decision showed (at [73]):

In the light of the findings made above, the evidence before me in fact went much further and established the accused’s involvement in the infliction of the injuries which subsequently led to the deceased’s death. Considered against all the other facts, the accused’s averment that he was wholly unaware of the deceased’s demise when he left the latter’s house was inherently unbelievable. While a conviction would have been warranted even if the accused did not know of the deceased’s demise, the added factor that he was complicit, at least to some degree, in the stabbing and strangulation of the deceased gave further impetus to the Prosecution’s case against him. [emphasis added]

Of course, given Dr Chiu’s expert evidence that it was the strangulation with the black cord which had led to the deceased’s death, it was not entirely accurate to say that the appellant was involved in the infliction of the injuries which led to the deceased’s death (see also [37] above). It seems more the case that the trial judge was certain that the appellant had strangled the deceased, given that he wrote that the appellant was “complicit … in the … strangulation of the deceased” (Lee Chez Kee at [73]). This conclusion, as the trial judge stated (at [69]), was only explicitly established by Too’s statements. However, this in no way precluded the trial judge from inferring that the appellant had strangled the deceased, as he in fact did.

Summary of the trial judge’s reasoning 

42        In summary, the trial judge’s reasoning can be stated as such. First, the evidence, including Too’s statements, proved beyond a reasonable doubt that the appellant actually strangled the deceased to death. By this, it was not necessary to rely on s 34 of the Penal Code to convict the appellant of murder since he was primarily responsible for the offence. However, and secondarily, if this finding was incorrect, the application of s 34 meant that it was not necessary to establish whether it was the appellant (or indeed, any of the other parties) who strangled the deceased to death. By this secondary reason, so long as there existed a common intention between the parties to rob the deceased, and the criminal act of strangulating him was then found to be in furtherance of this common intention (to rob), the appellant would be guilty of the Charge, even if it could not be established that he was the actual person who strangled the deceased to death. On this point, I would repeat my observation that the trial judge had failed to make any express finding to the effect that the act of strangulating the deceased was in furtherance of the common intention of the parties to rob the deceased (see [40] above).

Issues on appeal 

43        The appellant has, in the present appeal, appealed against the trial judge’s decision on both conviction and sentence. However, as the sentence of death is a mandatory one in relation to the charge, and the appellant has not challenged, for example, the constitutionality of the mandatory death sentence, the outcome in the present proceedings, in effect, rests entirely on the result of the appeal against his conviction. In relation to this appeal then, the appellant’s petition of appeal lists four grounds of appeal:

(a)        The trial judge had wrongly admitted Too’s statements sought to be admitted by the Prosecution at the trial.

(b)        Having admitted Too’s statements, the trial judge had failed to exercise due caution in giving weight to Too’s statements and had also failed to give due consideration to the prejudicial effect they had on the appellant’s case.

(c)        The trial judge erred in finding that the Prosecution had proved the Charge beyond a reasonable doubt as this was against the weight of the evidence adduced.

(d)        The trial judge erred in finding that the weight of the evidence was sufficient to establish the guilt of the appellant for murder under s 34 of the Penal Code.

44        In my view, these four grounds can be broken down essentially into two broad issues, which I will address fully in this judgment: (a) whether the trial judge erred in admitting Too’s statements; and (b) whether, in the totality of the evidence (which content is dependent on the answer to the preceding issue), the trial judge erred in convicting the appellant of the Charge. This latter issue also necessitates a discussion of the law in relation to the essential ingredients of the Charge. I now deal with the legal issues relating to the admissibility of Too’s statements and the ingredients of the Charge before applying the applicable law to the facts of the present case.

Admissibility of Too’s statements

Introduction and the proceedings below

45        As I mentioned above (at [22]), in the proceedings below, as part of its case, the Prosecution additionally sought to admit Too’s statements into the evidence. The Prosecution had submitted to the trial judge that Too’s statements were admissible under s 378(1)(b)(i) of the CPC. This particular subsection stipulates:

Admissibility of out-of-court statements as evidence of facts stated.

378.—(1) In any criminal proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this section and section 379 and to the rules of law governing the admissibility of confessions, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(b)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness;

…

46        Unsurprisingly, the appellant vigorously took issue with the Prosecution’s attempt to admit Too’s statements in the proceedings below. However, the trial judge, after an extensive analysis of the legislative history of s 378(1)(b)(i) of the CPC, decided that Too’s statements were admissible. In view of the undeniable importance of achieving the correct interpretation of s 378(1)(b)(i), it would be apposite at this juncture to revisit, albeit in brief, the trial judge’s reasoning for his decision in this regard.

The trial judge’s reasoning

(1)        General propositions

47        The trial judge first determined that Too’s statements, being made out of the present proceedings, fell within the general purview of the hearsay rule. In this regard, the trial judge characterised the hearsay rule as being “exclusionary” in nature (see Lee Chez Kee at [27]) and quoted Phipson on Evidence (Sweet & Maxwell, 14th Ed, 1990), a textbook on the English law of evidence, in support of this proposition. However, after asserting that the hearsay rule was an exclusionary one, the trial judge then went on to say that this rule had been restated statutorily, in the context of the CPC, in an “inclusionary” form by s 377 (at [28]). For completeness, s 377 of the CPC provides as follows:

Hearsay evidence to be admissible only by virtue of this Code and other written law.

377.     In any criminal proceedings a statement other than one made by a person while giving oral evidence in those proceedings shall be admissible as evidence of any fact stated therein to the extent that it is so admissible by virtue of any provision of this Code or any other written law, but not otherwise.

48        The trial judge then held that, given Too’s absence from the proceedings before him, his statements clearly fell within the purview of s 377 of the CPC. Accordingly, the trial judge decided that Too’s statements would be only admissible if “one of the statutory exceptions under either the CPC or the Evidence Act could be shown to apply” [emphasis added] (at [28]).

49        Having stated the above proposition, the trial judge came to the conclusion (at [29]) that none of the “potentially relevant statutory exceptions” to the hearsay rule applied to Too’s statements. In fact, the trial judge readily acknowledged, or at least accepted the Prosecution’s own concessions in relation to s 10 and 30 of the EA, that none of the “exceptions” embodied in s 10, s 30 or s 32 of the EA applied. In order to fortify his conclusion, the trial judge drew attention to some English authorities which, in his view, “mirror the position generally obtaining under Singapore law” (at [32]). The trial judge went on to state (ibid):

Notwithstanding the admissibility of out-of-court confessions made by an accused to the proceedings (see s 122(5) of the CPC), there is no similar exception under either the CPC or the Evidence Act which applies to confessions made by other persons. The [appellant] was therefore correct in so far as it had submitted that confessions made by a so-called “co-accused” who is not party to the proceedings are not generally the subject of any exception to the hearsay rule. … Where a confession by a person other than an accused is in contention, the party seeking to rely on the statement should rightly admit this evidence by calling the maker of the confession as a witness. [emphasis added]

50        However, having concluded that confessions made by a co-accused who was not a party to the proceedings were not generally the subject of any exception to the hearsay rule, the trial judge then went on to state (at [33]) that the exception enshrined in s 378(1)(b)(i) of the CPC nonetheless “represented the only avenue for admitting Too’s statements”.

(2)        The scope of section 378(1) of the CPC

51        Having decided that s 378(1)(b)(i) of the CPC represented the “only avenue” for admitting Too’s statements, the trial judge then went on to consider the scope of this subsection. In particular, in the trial judge’s view, the applicability of s 378(1)(b)(i) hinged upon the proper construction of the phrase “subject … to the rules of law governing the admissibility of confessions”, which I shall for convenience term, as the trial judge did, “the qualifying phrase” (Lee Chez Kee at [38]). As could be expected, both the Prosecution and the appellant tendered diametrically opposing interpretations of s 378(1)(b)(i) to the trial judge. According to the Prosecution, the qualifying phrase merely had the effect of importing the requirements of voluntariness. As such, since there was no challenge on the voluntariness of Too’s statements, these statements were admissible under the exception contained in s 378(1)(b)(i) of the CPC. On the other hand, the appellant contended before the trial judge that the qualifying phrase imported the general “common law prohibition against confessions”, and therefore rendered s 378(1) incapable of rendering admissible those confessions that were otherwise inadmissible.

52        Having heard the arguments of both parties, the trial judge accepted the Prosecution’s submission that the qualifying phrase merely had the effect of importing the requirements of voluntariness. In coming to this decision, he relied primarily on the legislative intent behind s 378(1) itself and secondarily on his belief that a contrary interpretation would lead to manifest absurdity in the application of the particular subsection.

^{(A)        THE LEGISLATIVE INTENT BEHIND SECTION 378(1) OF THE CPC}

53        Turning first to the supposed legislative intent behind s 378(1) of the CPC, the trial judge pointed out that s 378(1) of the CPC first appeared as s 371C(1)(b)(i) within cl 23 of the Criminal Procedure Code (Amendment) Bill (Bill No 35 of 1975) (“the CPC Bill”). The proposed s 371C(1)(b)(i) was in turn adapted from cl 31(1) of the draft Criminal Evidence Bill (“the UK Bill”) proposed by the UK Criminal Law Revision Committee (“the UK Committee”): see the comparative table to the CPC Bill; see also UK Criminal Law Revision Committee, Eleventh Report: Evidence (General) (Cmnd 4991, 1972) (Chairman: Edmund Davies LJ) (“Eleventh Report”) at p 190.

54        The trial judge stated (Lee Chez Kee at [37]) that cl 31(1) of the UK Bill provided as follows:

Admissibility of out-of-court statements as evidence of facts stated.

31.—(1) In any proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this and the next following section and to section 2 of this Act, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(c)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness ...

[emphasis added by the trial judge]

For convenience, I shall also at this point reproduce s 378(1)(b)(i) of the CPC, although this has already been produced earlier:

Admissibility of out-of-court statements as evidence of facts stated.

378.—(1) In any criminal proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this section and section 379 and to the rules of law governing the admissibility of confessions, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(b)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness;

…

[emphasis added]

55        Subsequently, the trial judge went on to observe that the qualifying phrase that is currently present in s 378(1) of the CPC and which was also present in the proposed s 371C(1)(b)(i) of the CPC Bill was absent from cl 31(1) of the UK Bill. In place of the qualifying phrase, there appeared the words “subject … to section 2 of this Act”. This will be clear from an examination of the provisions reproduced in the preceding paragraph. Clause 2 of the UK Bill (ie, “section 2 of this Act”) provided as follows:

Confessions.

2.—(1) In any proceedings a confession made by the accused may be given in evidence by the prosecution in so far as it is relevant to any matter in issue in the proceedings and is not excluded by the court in pursuance of subsection (2) or (3) below.

(2)        If, in any proceedings where the prosecution proposes to give in evidence a confession made by the accused, it is represented to the court that the confession was or may have been made in consequence of oppressive treatment of the accused or in consequence of any threat or inducement, the court shall not allow the confession to be given in evidence by the prosecution (whether by virtue of this section or otherwise) except in so far as the prosecution proves to the court beyond reasonable doubt that the confession (notwithstanding that it may be true) —

(a)        was not obtained by oppressive treatment of the accused; and

(b)        was not made in consequence of any threat or inducement of a sort likely, in the circumstances existing at the time, to render unreliable any confession which might be made by the accused in consequence thereof.

(3)        In any proceedings where the prosecution proposes to give in evidence a confession made by the accused, the court may of its own motion require the prosecution, as a condition of allowing it to do so, to prove with respect to the confession the matters mentioned in paragraphs (a) and (b) of subsection (2) above.

(4)        Where in any proceedings a confession is received in evidence by virtue of the foregoing provisions of this section, it shall by virtue of this subsection be admissible as evidence of any fact stated therein and of any matter dealt with in any opinion expressed therein, including any fact or matter favourable to the accused:

Provided that at the trial of any person for an offence the court shall not be required to treat an issue as having been raised with respect to any matter by reason only of evidence favourable to the accused which is admissible by virtue of this subsection.

(5)        The fact that a confession is wholly or partly excluded in pursuance of subsection (2) or (3) above shall not affect the admissibility in evidence —

(a)        of any facts discovered as a result of the confession; or

(b)        as regards any fact so discovered, of the fact that it was discovered as a result of a statement made by the accused; or

(c)        where the confession is relevant as showing that the accused speaks, writes or expresses himself in a particular way, of so much of the confession as is necessary to show this about him.

(6)        In this section “confession” includes any statement wholly or partly adverse to the accused, whether made to a person in authority or not and whether made in words or otherwise.

[emphasis added]

56        The trial judge then essentially embarked on an exercise of importing the legislative intent behind the UK Bill to the CPC on the basis of the historical linkage between the present s 378(1) of the CPC with cl 31(1) of the UK Bill. This exercise first rested on the “close coincidence” between s 378(1) of the CPC and cl 31(1) of the UK Bill (at [39]). In the trial judge’s view, it was significant that “the other provisos to s 378(1) of the CPC appear to mirror those in cl 31(1) of the UK Bill” (ibid). The trial judge then continued (ibid):

Apart from the qualifying phrase, s 378(1) additionally provides that it is “subject to this section and section 379” [emphasis added]. Clause 31(1) was in turn expressed as being “subject to this and the next following section” [emphasis added]. Notwithstanding the apparent disparity in language, “the next following section” to s 378 of the CPC, ie, s 379, was in fact adopted from cl 32 of the UK Bill, which was itself “the next following section” to cl 31: see the comparative table to the CPC Bill. This general coincidence of the respective provisos to s 378(1) of the CPC and cl 31(1) of the UK Bill leads clearly to the conclusion that the current qualifying phrase in the former was intended to correspond with the reference to cl 2 in the latter. [emphasis in original]

57        Accordingly, having been convinced that the qualifying phrase was intended to correspond with the reference to cl 2 of the UK Bill by virtue of the historical nexus between s 378(1) of the CPC and cl 31(1) of the UK Bill and their general similarity, the trial judge then proceeded to the second stage of his reasoning. In this connection, he noted that the ambit of cl 2 of the UK Bill was confined solely to the requirement of voluntariness in confessions. Accordingly, he came to the unequivocal conclusion that the qualifying phrase in s 378(1) of the CPC was intended to encompass only what cl 2 in the UK Bill referred to, and this was the requirement of voluntariness.

58        Finally, to further bolster this conclusion, the trial judge remarked that this interpretation of the qualifying phrase was augmented by the UK Committee’s commentary to cl 31(1) of the UK Bill. The trial judge emphasised that, according to the UK Committee (see Eleventh Report at p 236), admissibility pursuant to cl 31(1) of the UK Bill was to be subject to the provisions in the later subsections of the clause, to the provisions of cl 2 preventing the Prosecution from giving in evidence a confession obtained in the ways mentioned in that clause and to the restrictions imposed by cl 32. In essence, the trial judge alluded once again to the fact that admissibility pursuant to cl 31(1) of the UK Bill was subject only to, inter alia, the voluntariness requirement as made clear by cl 2. This in reality merely reiterated what the trial judge had already discussed earlier in his grounds of decision.

59        Apart from referring to the historical connection and general coincidence between s 378(1) of the CPC and cl 31(1) of the UK Bill, the trial judge also thought that his conclusion in relation to the ambit of the qualifying phrase (and hence the scope of s 378(1)(b)(i)) was consistent with the legislative intent behind s 378 of the CPC and its related provisions as expressed by our Parliament. In this respect, the trial judge pointed out (at [42]) that, according to the then Minister for Law and the Environment, Mr E W Barker (see Singapore Parliamentary Debates, Official Report (19 August 1975) vol 34 at cols 1222–1223):

Clause 23 of the [CPC] Bill [of which the current ss 378 and 379 formed a part] seeks to make radical changes to the rule against hearsay evidence. ...

...

The present law has caused a great deal of trouble owing to the difficulty of deciding whether a statement is hearsay, and, if so, whether it is admissible under any of the exceptions to the rule. The difficulty is further compounded by the differences between the English law of evidence and our law which is based on the Indian Evidence Act.

The scheme that the Bill proposes has the following purposes:-

(1)        to admit all hearsay evidence likely to be valuable to the greatest extent possible without undue complication or delay to the proceedings;

(2)        to ensure that evidence should continue to be given for the most part orally by allowing hearsay evidence only if the maker of the statement cannot be called or it is desirable to supplement his oral evidence; and

(3)        to include necessary safeguards against the danger of manufactured hearsay evidence.

[emphasis added by the trial judge]

As such, the trial judge was of the opinion (at [43]) that “Parliament’s intention to ‘admit all hearsay evidence … to the greatest extent possible’ … clearly accords with a more limited interpretation of the qualifying phrase” [emphasis added by the trial judge]. For good measure, the trial judge added that the dangers of manufactured or unreliable out-of-court confessions being admitted under s 378(1) would also be sufficiently safeguarded against by the requirement of voluntariness.

^{(B)       CONTRARY INTERPRETATION WOULD LEAD TO ABSURDITY AND INCONSISTENCY}

60        Turning now to the trial judge’s secondary ground on which his interpretation of s 378(1)(b)(i) of the CPC was based, it bears repeating that the trial judge thought that a contrary interpretation would lead to manifest absurdity and inconsistency in the application of the particular subsection. In his view, the appellant’s interpretation of the qualifying phrase “would turn the concern regarding manufactured or unreliable hearsay evidence on its head” (at [44]). This was because inculpatory out-of-court statements would be excluded from the purview of s 378(1), whilst exculpatory out-of-court statements would remain potentially admissible under the exceptions to hearsay enshrined therein. According to the trial judge, this result would be unjustifiable given that confessions, which implicate their makers, are generally regarded as being more reliable since they are against the makers’ interests. In contrast, statements which purport to exculpate their makers would perceivably be less reliable since they would be more likely to be manufactured evidence. To permit the admission of the latter but not the former through the exceptions to hearsay under s 378(1) would, therefore, be absurd and irrational, to say the least (ibid).

Summary of the trial judge’s reasoning

61        To recapitulate, the trial judge’s interpretation of the qualifying phrase was essentially premised on what he believed to be the legislative intent behind s 378(1) of the CPC as evidenced by the similarity between the wording of the subsection with cl 2 of the UK Bill. This legislative intent, supported in turn by other reference material, showed that the qualifying phrase in s 378(1) of the CPC referred merely to the requirement of voluntariness. Accordingly, since Too’s statements were never shown to have been made involuntarily, they were admissible by virtue of s 378(1)(b)(i) since Too was now dead and the requirement of voluntariness had been fulfilled.

The parties’ submissions in the present appeal

The appellant’s submissions

62        The appellant contended that the trial judge’s interpretation of the qualifying phrase was wrong. His argument was premised on four points, namely:

(a)        Too’s statements were inadmissible not only under the hearsay rule but were also not admissible under any of the exceptions to the hearsay rule, and, therefore, they should not become admissible under s 378(1) of the CPC.

(b)        The UK Bill was not intended by Parliament to be adopted in toto but rather was a guide for adaptation to amend the CPC and, as such, the qualifying phrase was not restricted to the trial judge’s interpretation.

(c)        While Parliament intended to widen the scope of admissibility of hearsay evidence, it also expressed the need to have safeguards, hence the insertion of the qualifying phrase.

(d)        The distinction between inculpatory and exculpatory hearsay confessions utilised by the trial judge at the trial to arrive at his interpretation was mistaken since it was irrelevant.

The Prosecution’s submissions

63        On the other hand, the Prosecution’s submissions were:

(a)        Section 378 of the CPC allowed for the admission of hearsay evidence when certain specified requirements were met.

(b)        One of the requirements was that the hearsay evidence must comply with the rules of law governing the admissibility of confessions.

(c)        The rules of law governing the admissibility of confessions essentially referred to the requirement of voluntariness.

(d)        The interpretation adopted by the trial judge was obvious on a plain reading of s 378.

(e)        This interpretation would give effect to the intention of Parliament.

(f)         This interpretation was consistent with the existing body of local cases which had applied s 378.

(e)        This was also the approach accepted by local commentators.

Analysis and discussion

64        A few preliminary observations are apposite at this point. It is undisputed by either party that Too’s statements amounted to confessions. It is also undisputed that Too’s statements are hearsay pursuant to the common law definition, viz, they are statements made out of court adduced to prove the facts contained therein. Before the trial judge, the Prosecution conceded that these statements did not come within the “statutory exceptions” to the hearsay rule contained in ss 10 and 30 of the EA. The trial judge also found that s 32 of the EA did not operate to admit the statements and, in any event, the Prosecution had not additionally sought to rely on s 32. Thus, Too’s statements could not be admitted on the basis of these sections. The admissibility of Too statements therefore centred on the interpretation of s 378(1)(b)(i) of the CPC.

65        In my view, the correct interpretation of s 378(1)(b)(i) of the CPC boils down to the meaning to be ascribed to the qualifying phrase. Was the trial judge correct in equating s 378(1) of the CPC with cl 31(1) of the UK Bill and importing the requirements of cl 2 of the UK Bill to s 378(1) such that only the requirement of voluntariness was read into the qualifying phrase? However, before answering this question, I should also point out that given the dearth of local case law dealing with the difficult area of hearsay evidence, this case affords a good opportunity to discuss the conceptual basis of hearsay in the EA and the CPC, given that the parties, and indeed the trial judge, have seemingly characterised the hearsay rule as an exclusionary one even in the context of our statutory regime. It is to this preliminary question that I first turn.

The hearsay rule under the EA

(1)        Conceptual basis

66        In my view, the judicial interpretation of the EA in relation to hearsay evidence has clouded the co-relation between the statutory provisions and the common law to such an extent that any number of conflicting authorities and principles are now available to one who is considering the problem. For a start, the characterisation of the hearsay rule as “exclusionary” by both parties and even the trial judge in apparent connection with the EA is not strictly accurate given the backdrop of the EA. The scheme of the EA, as the draftsman of its precursor, Sir James Stephen, intended it to be, is inclusionary and not exclusionary. Therefore, to say that the hearsay “exclusionary” rule exists either within or independently of the EA is to imply the existence of something which is, strictly speaking, beyond the intention of Sir James Stephen.

67        It must be clarified that the EA does not contain an express definition of hearsay. Instead, the EA contains an implicit acknowledgement of the rule. As Prof Tan Yock Lin perceptively notes in his seminal work, Criminal Procedure (LexisNexis, 2007) vol 2 at ch XVI para [3], hearsay in the EA is perceived as being a statement of relevant facts and as such is an irrelevant fact as opposed to a statement which is itself declared by the EA to be a relevant fact. This is assured by the absence of any general provisions making statements of relevant facts themselves relevant facts. Where exceptions to the hearsay rule are intended to be relevant, they are rendered specifically as relevant facts. There are thus no real “exceptions” in the EA; more accurately, the EA gives effect to these common law exceptions to the hearsay rule.

68        Similarly, Assoc Prof Chin Tet Yung in his article, “Hearsay – A Doctrine in Retreat?” (1990) 32 Mal LR 239, incisively notes at 240 that the admissibility of hearsay evidence is achieved in a “circuitous way” in the EA: statements containing relevant facts or facts in issue also have to be legally relevant under the EA. The legal relevancy sections concerning statements are ss 17 to 40 and are similar to, but not identical with, the English common law exceptions at the time of about 1872. These sections made certain statements relevant which also means that they are admissible. Assoc Prof Chin further notes that the absence of the hearsay rule in its exclusionary form does not mean that hearsay statements may be admitted “willy-nilly” (at 240), presumably by virtue of the implicit acknowledgement of the hearsay rule in the EA itself.

69        In other words, as succinctly summarised by the High Court in Roy S Selvarajah v PP [1998] 3 SLR 517 at [40], the EA does not formulate the rule against hearsay evidence. Rather, it adopts an inclusionary approach, stating what may be admitted in evidence. The important question is thus whether the statement to be admitted satisfies any of the definitions of legal relevancy (which is a separate issue from whether the statement is logically relevant) in the EA. If so, it is relevant, and is made admissible by s 5 of the EA and that is the end of the enquiry. It does not matter whether evidence of the relevant fact thus established matches that which the common law denotes as being original evidence or as being hearsay evidence admissible under an exception to the hearsay rule.

70        As such, hearsay in the EA is not determined in precisely the same way as at common law. At common law, the hearsay issue is usually resolved by asking whether one’s interest in the statement is in its existence or in its truth. This follows from the exclusionary characterisation of the hearsay rule at common law. For example, Prof Jeffrey Pinsler SC in Evidence, Advocacy and the Litigation Process (Butterworths Asia, 1992) defines “hearsay” as such (at p 64):

[T]he assertions of persons made out of court whether orally or in documentary form or in the form of conduct tendered to prove the facts which they refer to (ie facts in issue and relevant facts) are inadmissible unless they fall within the scope of the established exceptions.

Therefore, at common law, if a statement falls within this definition and is hearsay, it is excluded unless it falls within an established common law exception to hearsay. This is rather different from the admission of hearsay evidence in the EA.

(2)        The prevailing judicial approaches

^{(A)        REFERENCE TO THE HEARSAY RULE WITHOUT REGARD TO THE EA} 

71        However, one of the popular judicial views, contrary to the tone of the EA, adopts precisely the characterisation of the hearsay rule as one that makes no reference to the EA and instead utilises the common law definition. In a Privy Council case on appeal from Malaya, Subramaniam v PP [1956] MLJ 220, Mr L M D de Silva characterised the hearsay rule as such (at 222):

Evidence of a statement made to a witness by a person who is not himself called as a witness may or may not be hearsay. It is hearsay and inadmissible when the object of the evidence is to establish the truth of what is contained in the statement. It is not hearsay and is admissible when it is proposed to establish by the evidence, not the truth of the statement, but the fact that it was made.

As will be appreciated, this definition is reminiscent of the common law definition and characterises the hearsay rule as an exclusionary one notwithstanding the scheme of the EA. By this approach, one does not even have to look to the sections in the EA defining legal relevancy; all that one needs to do is to ascertain whether the evidence falls within the hearsay rule as defined at common law.

^{(B)        IMPLICIT REFERENCE TO THE HEARSAY RULE IN THE EA}

72        Yet another judicial view, as Prof Tan additionally notes in Criminal Procedure ([67] supra) at ch XVI para [52], is that provisions in the EA embodying the direct evidence rule are an implicit reference to the hearsay rule (see further an article by the same author, “Stephen’s Hearsay – Does it Matter?” (1991) 12 Sing LR 128). Thus, in Soon Peck Wah v Woon Che Chye [1998] 1 SLR 234 (“Soon Peck Wah”), this court emphatically declared that, in Singapore, the rule against hearsay was reflected in s 62 of the EA (see also Wong Kok Keong v Regina [1955] MLJ 13 at 14 and Lim Ah Oh v Rex [1950] MLJ 269 at 270). Soon Peck Wah was recently cited by this court in Jet Holding Ltd v Cooper Cameron (Singapore) Pte Ltd [2006] 3 SLR 769, although this court in that case had not referred to s 62 of the EA as forming the foundation for the hearsay rule in the EA. For completeness, I should mention that s 62(1) of the EA provides as follows:

Oral evidence must be direct

62.—(1) Oral evidence must in all cases whatever be direct —

(a)        if it refers to a fact which could be seen, it must be the evidence of a witness who says he saw that fact;

(b)        if it refers to a fact which could be heard, it must be the evidence of a witness who says he heard that fact;

(c)        if it refers to a fact which could be perceived by any other sense or in any other manner, it must be the evidence of a witness who says he perceived that fact by that sense or in that manner;

(d)        if it refers to an opinion or to the grounds on which that opinion is held, it must be the evidence of the person who holds that opinion on those grounds.

This court also further stated in Soon Peck Wah (at [34]) that, by virtue of s 2(2) of the EA, the common law exceptions to the hearsay rule had also been incorporated into our law of evidence.

73        There appears to be two problems with this judicial view as set out in Soon Peck Wah. First is the problem with the proposition that the hearsay rule finds implicit expression in s 62 of the EA. As Prof Tan correctly notes in Criminal Procedure at ch XVI para [52], this view continues to mistake a prohibition on the use of indirect evidence for a prohibition on the use of hearsay evidence. To say that s 62 imports the hearsay rule is to confuse a description of the mode of proof with the type of proof. Section 62 is not concerned with relevancy; it simply tells us how to prove facts which already have been found to be relevant by the definition of relevancy in the earlier parts of the EA.

74        The second problem with this judicial view is the reference to the applicability of the common law exceptions to the hearsay rule. Prof Pinsler in “Approaches to the Evidence Act: The Judicial Development of a Code” (2002) 14 SAcLJ 365 at 382 neatly summarises the many facets of this problem. On one level, there is no question that there are a number of exceptions in English law which are not recognised or only acknowledged in modified form by the EA. Their application would thus be inconsistent with the EA. Further, Sir James Stephen intended to comprehensively formulate the traditional exceptions to the hearsay rule in ss 17 to 41 of the EA. The application of all the common law exceptions without discrimination would dislocate this scheme.

(3)        Problems with the existing judicial views and the way forward

75        As can be seen, the popular judicial views in relation to the admissibility of hearsay evidence in Singapore can be faulted on two grounds. First, it can be said that the courts have not made reference to the EA but have consistently restated the common law formulation of the hearsay rule, even though there is an inconsistency between the conceptual bases of the common law and the statutory approaches. Strictly speaking, such an approach is forbidden by s 2(2) of the EA, which repeals all rules which are not saved by statute and which are inconsistent with the provisions of the EA. Secondly, where the courts have sought to link the common law approach with the EA, the reason given, ie, that s 62 reflects the common law hearsay rule (via s 62 of the EA), is not convincing and is, at its heart, conceptually erroneous. In my view, these judicial views, epitomised by the dictum of this court in Soon Peck Wah, ought not to be followed. While it is true that both the common law and the EA ways of identifying hearsay will always nearly provide the same result, and I certainly do not believe that any injustice has been caused by the different approaches, it is just as important to be conceptually clear about the admissibility of hearsay evidence in the context of our statutory regime.

The hearsay rule under the CPC

76        This is not the end of the enquiry. Against this notion of hearsay in the EA, the CPC’s evidentiary provisions can be seen as extending the ambit of the admissibility hearsay evidence in Singapore. However, as Prof Tan notes in Criminal Procedure at ch XVI para [1202], in widening the categories of admissible hearsay, the draftsman of the CPC provisions was apparently unaware of the notion of hearsay in the EA. Instead, he employed the common law conception and so, for instance, in s 378 of the CPC, the conception of hearsay as a statement to be used as evidence of the truth of facts stated therein is evident. Similarly, Prof Jeffrey Pinsler in “Statements of Witnesses to the Police: A Story of Strange Bedfellows in the Criminal Procedure Code and Evidence Act” [2001] Sing JLS 53 (“Prof Pinsler’s article”) also laments about the lack of symbiosis among the provisions within the CPC and between that statute and the EA concerning the admissibility at trial of statements of witnesses to the police. This criticism extends in some measure to the different conceptions of hearsay in the present-day EA and the CPC as well.

77        I agree that the present statutory framework is not satisfactory. Indeed, quite apart from the different conceptual bases of the admissibility of hearsay evidence in the two Acts, there are several other inconsistencies and problems which arise but do not present themselves for mention before this court in this appeal. Much of the difficulty, as rightly pointed out in Prof Pinsler’s article at 79–80, stems from the manner in which statutory provisions were incorporated in 1976 without careful consideration of the pre-existing legislation in this area. The way forward must surely involve a reconsideration of these principles and their appropriate statutory reformulation. However, until such reformulation is actually realised, the courts will do well to be simply aware of the different conceptual bases underpinning the admissibility of hearsay evidence in both the EA and the CPC, and be equally alive to the problems which might arise as a result.

The scope of section 378(1)(b)(i) of the CPC

(1)        Preliminary observations

78        The above exposition on the conceptual differences in the admissibility of hearsay evidence in both the EA and the CPC is not without direct relevance in this appeal. In essence, it shows that quite apart from the CPC, there is an inextricably related statute in the admission of evidence, including confessions, in the local evidential statutory regime, viz, the EA. As will be recounted, the trial judge, in interpreting s 378(1)(b)(i) of the CPC, did so on the basis of the CPC alone. He had equated the relevant clause in the UK Bill with s 378(1)(b)(i) in coming to his conclusion that the legislative intent behind the qualifying phrase referred only to the requirement of voluntariness.

79        A few preliminary observations can be made in relation to this approach. First, in my view, the trial judge was entirely correct in seeking the legislative intent behind s 378(1)(b)(i) of the CPC when interpreting the section. It is trite law that a court should give effect to the legislative purpose when interpreting an Act of Parliament: see, for example, Donald McArthyTrading Pte Ltd v Pankaj s/o Dhirajlal [2007] 2 SLR 321 at [6] and PP v Low Kok Heng [2007] 4 SLR 183 at [39]–[49]. This requirement is also statutorily enshrined in s 9A(1) of the Interpretation Act (Cap 1, 2002 Rev Ed), which mandates the preference for “an interpretation that would promote the purpose or object underlying the written law (whether that purpose or object is expressly stated in the written law or not)”.

80        However, and this leads to the second preliminary observation, the trial judge’s approach in seeking the legislative intent is not entirely satisfactory. His approach of embarking on a rudimentary comparison of the language between the relevant provisions of the UK Bill and the CPC, with respect, ignores the important fact that the local law in relation to the admissibility of confessions is not on all fours with that contained in the UK Bill.

81        It bears repeating at this point that the applicability of s 378(1)(b)(i) of the CPC hinges upon the proper construction of the phrase “subject … to the rules of law governing the admissibility of confessions”. Given that Too’s statements were regarded by both parties as amounting to confessions and that ss 10, 30 and 32 of the EA did not operate to render these statements legally relevant (and hence admissible), s 378(1)(b)(i) of the CPC remained the only possible avenue by which these statements could be admitted. To ascertain the scope of s 378(1)(b)(i) in turn, the question to be answered is simply this: What are the rules of law governing the admissibility of confessions? Was the trial judge correct in equating the rules of law to the requirement contained in cl 2 of the UK Bill simply by reason of the similarity in structure and historical connection between the UK Bill and the CPC? To answer these questions, one must first address the threshold question of whether the UK Bill can, and should, be regarded as the equivalent, and therefore the historical predecessor, of the sections of the CPC as amended in 1976.

(2)        Are the provisions in the UK Bill the exact equivalent of the CPC amendments in 1976?

82        There is no doubt that the CPC was based on the UK Bill. This much is clear from a perusal of the parliamentary debates at the second reading of the CPC Bill in 1975. Following from the passage cited by the trial judge (see [59] above), it bears emphasis that Mr Barker also said (see Singapore Parliamentary Debates, Official Report (19 August 1975) vol 34 at cols 1223–1224):

The proposals in this Bill relating to hearsay evidence adopt many of the recommendations contained in the Criminal Law Revision Committee’s Eleventh Report. Certain modifications have been made to these proposals after considering the views expressed by the General Council of the Bar of England and Wales and the Council of the Law Society of England in their memoranda.

The opportunity has been taken to amend other sections of the Code to correct existing mistakes and anomalies as well as to expedite the administration of justice as when an accused wishes to plead guilty in the High Court.

It is proposed to refer the Bill to Select Committee where all representations will be given careful consideration.

[emphasis added]

It can therefore be seen clearly that the amendments to the CPC in 1976 were modelled on the UK Bill. Specifically, it is evident that the eventual amendments to the CPC relating to hearsay were based substantially on the UK Bill. However, this did not mean that the amendments to the CPC were an exact mirror of the UK Bill. Indeed, the trial judge had failed to note that Parliament, after the second reading of the CPC Bill, had then referred the Bill to a select committee (“the Select Committee”) for further and more careful consideration. Reference to the report of the Select Committee will therefore be vital in understanding whether s 378(1) of the CPC, based as it was on s 371(1) under cl 23 of the CPC Bill, was in turn based on cl 31(1) of the UK Bill.

^{(A)        AMENDMENTS BY THE SELECT COMMITTEE SHOWING DIFFERENCE BETWEEN THE CPC BILL AND THE UK BILL} 

83        The report of the Select Committee (see Report of the Select Committee on the Criminal Procedure Code (Amendment) Bill (Parl 4 of 1976, 24 June 1976) (Chairman: Dr Yeoh Ghim Seng) (“Report of the Select Committee”) was presented to Parliament on 24 June 1976. The Select Committee received seven written representations and heard oral evidence from the authors of four of the seven written representations. Subsequently, the Select Committee made certain amendments to the CPC Bill which were then accepted by Parliament at its third reading (see Singapore Parliamentary Debates, Official Report (23 July 1976) vol 35 at cols 993–995).

84        One of the amendments made by the Select Committee related to s 371C(1) under cl 23 of the CPC Bill (see Report of the Select Committee at Appendix V, p D3, col 6). In moving for the amendment to s 371C(1), Mr Barker had told the Select Committee that (ibid):

Various representors spoke of the dangers of admitting hearsay evidence and suggested further safeguards. This amendment seeks to delete sub-paragraph (iv) of paragraph (b) of subsection (1) of the new section 371cbecause the provision therein making admissible a statement made by a person whose identity is known but who cannot be found can lead to abuse and the fabrication of evidence. [emphasis added]

It will be noted that in deleting s 371C(1)(b)(iv) of the CPC Bill, the Select Committee had declined to completely model the eventual s 378(1) of the CPC on cl 31(1) of the UK Bill. Indeed, the present s 378(1)(b) does not contain a sub-paragraph dealing with a statement made by a person whose identity is known but who cannot be found. This is not only a superficial change, for it more than hints at a crucial difference in opinion between the UK Committee and the Select Committee in formulating the UK Bill and the CPC Bill respectively. In other words, this is evidence that the UK Bill cannot be taken as the equivalent of the CPC amendments in 1976.

85        To elaborate, it must first be mentioned that cl 31(1) of the UK Bill also contained a sub-para (c)(v) which provided as follows:

Admissibility of out-of-court statements as evidence of facts stated.

31.—(1) In any proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this and the next following section and to section 2 of this Act, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(c)        it is shown with respect to him —

…

(v)        that, his identity being known, all reasonable steps have been taken to find him, but that he cannot be found.

[emphasis added]

In its Eleventh Report, the UK Committee acknowledged (at para 240) the danger of manufactured evidence and had considered (at para 242) excluding altogether a statement made by an unidentified person or by a person who, although identified, could not be found. Eventually, the UK Committee decided against excluding these statements and gave its reasons in the following manner (ibid):

But although to make these further restrictions [ie, the exclusions considered by the UK Committee] would decrease the danger of manufactured evidence, we do not think that they would be justified. ... In the case of an identified person who cannot be found there may be a stronger argument because of the danger of his being kept out of the way. But we do not think it necessary to make even this restriction, especially as the person concerned might have given his statement to the prosecution and the defence might be keeping him out of the way. [emphasis added]

86        As is immediately apparent, the Select Committee did not adopt the UK Committee’s Eleventh Report fully and unquestionably in the lead-up to the eventual amendments to the CPC in 1976. In contrast, as the above examination of the different opinions expressed by the two committees in relation to the admissibility of statements by an identified person who could not be found shows, the Select Committee was probably more visibly concerned about the dangers of manufactured evidence than its UK counterpart. Indeed, while the UK Committee had put faith in its system of notice and counter-notice in guarding against manufactured evidence, the Select Committee, despite adopting the same notice system in the CPC Bill, took the more cautious approach of deleting s 371C(1)(b)(iv) of the CPC Bill completely. From a wider perspective, this shows the danger of equating the UK Bill wholly with the CPC Bill.

87        As such, I am unable to accept the trial judge’s conclusion, based simply on a general examination of the wordings of the relevant sections, that the general coincidence of the respective provisos to s 378(1) of the CPC and cl 31(1) of the UK Bill leads clearly to the conclusion that the current qualifying phrase in the former was intended to correspond with the reference to cl 2 in the latter. Certainly, a more in-depth analysis needs to be undertaken.

^{(B)        ABSENCE OF PROVISION IN THE UK BILL RELATING TO CO-ACCUSED'S CONFESSION IN THE CPC BILL} 

88        There is another important difference between cl 31(1) the UK Bill and s 378(1) of the CPC which detracts from the direct comparison of the respective provisos in the said subsections. In his grounds of decision (at [37]), the trial judge placed considerable emphasis on cl 31(1) being expressed as being “subject to this and the next following section” [emphasis added] (see [54] above). The trial judge had reasoned that since the “next following section”, viz, cl 32 of the UK Bill, was substantially similar to s 379 of the CPC, and s 378(1) additionally provided that it was “subject to this section and section 379” [emphasis added by the trial judge], it must follow that the qualifying phrase in s 378(1) must thereby correspond to its equivalent phrase in cl 31(1), ie, the reference to cl 2 in the UK Bill. Attractive as this reasoning may seem at first sight, it neglects to mention that the references to “this … section” in the respective subsections, viz, cl 31(1) and s 378(1), refer to very different sub-provisions. This is because cl 31(1) of the UK Bill contains a sub-s (2) touching on the confessions of a co-accused. In contrast, s 378(1) of the CPC (and its predecessor s 371C(1) under cl 23 of the CPC Bill) contains absolutely nothing on the confessions of a co-accused. The section which deals with this matter is s 30 of the EA, which provides as follows:

Consideration of proved confession affecting person making it and others jointly under trial for same offence

30.       When more persons than one are being tried jointly for the same offence, and a confession made by one of such persons affecting himself and some other of such persons is proved, the court may take into consideration the confession as against the other person as well as against the person who makes the confession.

Explanation.—“Offence” as used in this section includes the abetment of or attempt to commit the offence.

(a)        A and B are jointly tried for the murder of C. It is proved that A said “B and I murdered C”. The court may consider the effect of this confession as against B.

(b)        A is on his trial for the murder of C. There is evidence to show that C was murdered by A and B and that B said: “A and I murdered C”.

This statement may not be taken into consideration by the court against Aas Bis not being jointly tried.

[emphasis added]

89        It would perhaps also be useful to set out the relevant parts of cl 31(1) of the UK Bill, along with cl 31(2), at this juncture:

Admissibility of out-of-court statements as evidence of facts stated.

31.—(1) In any proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this and the next following section and to section 2 of this Act, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(c)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness ...

(2)        Subject to section 2 of this Act, in any proceedings in which two or more persons are jointly charged, a statement made, whether orally or in a document or otherwise, by any of the accused may be given in evidence by the prosecution as evidence of any fact stated therein of which direct oral evidence by the maker would be admissible, notwithstanding that the maker has not been and is not to be called as a witness in the proceedings; and any statement given in evidence by virtue of this subsection shall be admissible as evidence of any such fact in relation to each of the accused.

[emphasis added]

90        Clause 31(2) of the UK Bill enables the Prosecution to adduce as evidence a statement made by one accused (A) implicating another (B) who is being charged jointly with him. The UK Committee regarded (at para 251 of the Eleventh Report) that a special provision was necessary for this purpose as a matter of policy as there were many cases where the interests of justice required that what any of the accused had said out of court about the part played by the others in the event in question should be placed before the court. The UK Committee opined that, for example, it often happened that, when A and B were jointly charged with an offence, A had made a statement implicating them both and B had made no statement; or again each might have made a statement seeking to throw the blame on each other. In the latter kind of case, the UK Committee thought that it was particularly desirable that the out-of-court statements should be admissible in evidence in order that they might be properly evaluated. Although the UK Committee stated further (ibid) that if the maker of the statement in question should have died or become unavailable to give evidence in one of the ways mentioned in cl 31(1) the statement would be admissible in any event, this statement might not be readily rationalised with the strict requirement of a joint trial in cl 31(2). In this regard, I should also make a preliminary observation that cl 31(2) is expressed to be in relation to the admissibility of statements in respect of persons who are jointly charged but not jointly tried, as is the case in s 30 of the EA. This, to my mind, is a material difference. While it might be argued that there is express language by the UK Committee to admit a dead co-accused’s statements even in the event of merely a joint charge (as opposed to a joint trial where the trial judge can at least himself appraise the demeanour of the co-accused in court), there is no such express statement in the local context, particularly since s 30 of the EA is both expressed in the form of a joint trial and contains an express prohibition in its illustration for precisely the situation contemplated in the present appeal. I shall elaborate on this in due course. In any event, as has been discussed, I do not think that it is correct to accept the UK Committee’s comments in this regard as fully correct in the local context, given the differences across the statutory language and, more importantly, the apparent disparity between the legislative intent in both countries.

91        In contrast to cl 31, s 378 of the CPC does not contain any provision or express reference whatsoever in relation to the confession of a co-accused. Such confessions, as mentioned above, are dealt with by s 30 of the EA. Section 30, while similar to cl 31(2), is not identical to the English clause. Thus, while the trial judge was correct in saying that the references to “the next following section” in the UK Bill and to “section 379” in the CPC pointed to broadly similar sections, he had, with respect, neglected to have regard to the fact that the references to “this … section” [emphasis added] in both the UK Bill and the CPC did not likewise point to broadly similar sections. The important omission of cl 31(2) of the UK Bill (which touched on the confession of a co-accused) from s 378 of the CPC means that the qualifying phrase could not have been intended to correspond entirely to cl 2 of the UK Bill. In other words, one will have to consider s 30 of the EA as well, since this section is made applicable by the reference to “the rules of law governing the admissibility of confessions” in s 378(1) of the CPC.

92        Having concluded that it would not be correct to regard the UK Bill as exactly equivalent to the CPC, the deliberations behind the UK Bill, similarly, cannot be taken as identical expressions of the legislative intent behind the CPC. As such, it now remains to ascertain the legislative intent behind (and hence the scope of) s 378(1)(b)(i) of the CPC.

(3)        The legislative intent behind section 378(1)(b)(i) of the CPC

^{(A)        RELEVANT PRINCIPLES OF INTEPRETATION}

93        A relevant principle of interpretation would be the principle that the law should be coherent and self-consistent (see F A R Bennion, Statutory Interpretation (Butterworths, 4th Ed, 2002) at p 690). It stands as a principle of interpretation that the court, when considering, in relation to the facts of the instant case, which of the opposing constructions of the enactment would give effect to the legislative intent, should presume that the legislator intended to observe this principle of coherence and self-consistency. The court should therefore strive to avoid adopting a construction which involves accepting that the law is not coherent and self-consistent on the point in question.

94        Applying this principle to the context of the present case, it is clear that in interpreting s 378(1)(b)(i) of the CPC, this court should not adopt an interpretation which would render another statutory provision otiose. The relevant statutes will be the CPC and the EA. I acknowledge the difficulty involved in reconciling the conceptual bases of both the CPC and the EA, especially in relation to hearsay evidence, of which Too’s statements surely are. However, notwithstanding the fact that the conceptual bases may be different, I do not think that this precludes me from attempting to reconcile the substantive requirements of both statutes so as to achieve coherence and self-consistency.

95        Turning now to s 378(1)(b)(i) of the CPC and, in particular, the qualifying phrase, one must ask if the reference to the “rules of law governing the admissibility of confessions” refers merely to the requirement of voluntariness or includes other rules as well. To answer this question requires me to consider the content of the rules of law governing the admissibility of confessions. If confessions are admitted solely on the basis of voluntariness alone, then it is clear that the trial judge’s interpretation would be correct. However, if confessions are governed by other admissibility requirements in the EA apart from the requirement of voluntariness, then an interpretation of the qualifying phrase which only imports the requirement of voluntariness would render such other admissibility requirements otiose and would in turn offend the requirement of coherence and self-consistency.

^{(B)        THE RULES OF LAW GOVERNING THE ADMISSIBILITY OF CONFESSIONS}

96        The CPC and the EA contain the rules of law governing the admissibility of confessions. In so far as the EA is concerned, its statutory arrangement and overall scheme tells us which sections are related to the rules of law governing the admissibility of confessions. First, as I have stated above, the EA adopts an inclusionary scheme. By this scheme, any fact deemed to be legally relevant by the EA can be given in evidence. It is provided by s 5 of the EA that the definitions of legal relevancy are exhaustive within the confines of the EA:

Evidence may be given of facts in issue and relevant facts

5.         Evidence may be given in any suit or proceeding of the existence or non-existence of every fact in issue and of such other facts as are hereinafter declared to be relevant, and of no others.

Explanation.—This section shall not enable any person to give evidence of a fact which he is disentitled to prove by any provision of the law for the time being in force relating to civil procedure.

[emphasis added]

97        Secondly, the EA, from ss 17 to 33, declares when admissions and confessions are deemed to be relevant. The provisions relating to confessions are covered by ss 24 to 30. Most importantly, in cases involving two or more offenders jointly tried for the same offence, the confession of one offender may only be taken into consideration against the other offenders if s 30 of the EA is satisfied.

^{(C)        SECTION 378(1) OF  THE CPC SUBJECT TO PROVISIONS ON CONFESSIONS IN THE EA}

98        As such, from a broader perspective, it is undoubtedly the case that “the rules of law governing the admissibility of confessions” include ss 17 to 33 of the EA. To interpret the qualifying phrase otherwise would be to uncouple the EA from the CPC, and it is not at all apparent that this represents the legislative intent. Indeed, s 384 of the CPC expressly preserves the independence (and hence relevance) of the EA in respect of the admissibility in any criminal proceedings of any statement which would by virtue of the EA be admissible as evidence of any fact stated therein. In addition, considering our earlier conclusion that the UK Bill is not to be taken as representing the ambit and scope of the CPC, I feel that such an interpretation of s 378(1) of the CPC would best accord with the need for coherence and self-consistency.

99        Such an interpretation is consistent with the sentiments expressed by the Minister in Parliament before and after the Select Committee was formed to consider the CPC Bill. Before the Select Committee was formed, and as the trial judge has mentioned, Mr Barker had said (see Singapore Parliamentary Debates, Official Report (19 August 1975) vol 34 at col 1223):

The scheme that the Bill proposes has the following purposes:-

(1)        to admit all hearsay evidence likely to be valuable to the greatest extent possible without undue complication or delay to the proceedings;

(2)        to ensure that evidence should continue to be given for the most part orally by allowing hearsay evidence only if the maker of the statement cannot be called or it is desirable to supplement his oral evidence; and

(3)        to include necessary safeguards against the danger of manufactured hearsay evidence.

[emphasis added]

In my view, the trial judge placed too much emphasis on points (1) and (2) without considering the desire of Parliament to also include all necessary safeguards against the danger of manufactured hearsay evidence.

100      Indeed, this concern manifested itself to a greater extent in Singapore since the Select Committee expressly opted to delete certain provisions deemed to be “safe” by the UK Committee (see [85] and [86] above). This shows quite plainly that the legislative intent was to include even more safeguards than those formulated by the UK Committee in relation to hearsay evidence and its associated dangers. This must include the safeguards against the dangers associated with confessions. I would hasten to add that even if I am wrong in concluding that the legislative intent points towards a clear aversion from the dangers of hearsay and positively supports our preferred interpretation of s 378, the fact remains that the legislative intent behind s 378 is, at the very least, not apparently or schematically supportive of the stance adopted by the trial judge. In the absence of limpid clarity, I am reluctant to take a similar position given the adverse and unfair consequences that can be visited on an accused like the appellant. In summary, therefore, if it is accepted that cl 31 of the UK Bill and s 378 of the CPC are not identical, and it is further accepted that our Select Committee declined to adopt cl 31(1)(c)(v) (see [86] above), then it must mean that it is not at all evident that our Parliament accepted the English objectives underpinning the UK Bill entirely. Thus, if there is no suggestion in Singapore (as evidenced by the parliamentary debates) contrary to this conclusion, and yet the prima facie evidence is that the Select Committee did not intend to adopt cl 31 in its entirety, it must then tilt the supposition in favour of the view that our Parliament did not accept the English objectives entirely. Even if this is not evident (and there admittedly is no material explicitly stating this to be the case), the natural construction would be to presume this to be case owing to what is evident from the incomplete adoption of cl 31.

Does this interpretation of section 378(1)(b)(i) of the CPC lead to absurdity and inconsistency?

101      It will be recalled that the trial judge thought that the very interpretation I have adopted would lead to absurdity and inconsistency (see [60] above). This is apparently because by this interpretation, inculpatory confessions would be excluded from s 378(1) of the CPC whereas exculpatory statements would remain potentially admissible. By the trial judge’s reasoning, because inculpatory statements, being made against the maker’s own interest, were more reliable than exculpatory statements, it would be anomalous to admit the latter and not the former.

102      The trial judge’s concern is, with respect, misplaced. It confuses the characterisation of the statement sought to be admitted with the effect of the statement on the person it is admitted against. To analyse statements as inculpatory or exculpatory ignores the fundamental reason why the former is regarded as more reliable. I agree that a confession is inculpatory in nature. Its admissibility is premised on the fact that it is a statement made against the interest of its maker and hence inherently more reliable. However, that is only in relation to its maker. It does not follow that the entirety of a co-accused’s confession is against the interest of its maker. In the case of a co-accused’s confession, the part implicating another accused is most definitely in the interests of its maker. In this respect, this part of the confession may not be reliable vis-à-vis the other accused against whom the confession is admitted against. The cases of Tee Chu Feng v PP [2005] SGHC 181 and Gurbak Singh v PP [1998] 2 SLR 378, relied on and cited by the Prosecution, thus do not assist. Further, these cases do not appear to have considered in full the genesis of s 378 of the CPC. The dangers of unreliability remain and this is exactly what is addressed by the rules governing the admissibility of such confessions. The admissibility of such confessions is limited to the ambit of s 30 of the EA. There is no absurdity or inconsistency with such an interpretation.

Does section 378(1)(b)(i) of the CPC apply to admit Too’s statements in the present case?

103      Having concluded that the qualifying phrase includes those sections in the EA concerning the admissibility of confessions, it remains to be considered if s 378(1)(b)(i) of the CPC can be construed to admit Too’s statements in the present case. Too is dead and hence Too’s statements are prima facie admissible under s 378(1)(b)(i) unless the rules governing admissibility disallow them. In this respect, I note that the Prosecution has conceded before the trial judge that ss 10 and 30 of the EA do not allow for the admission of Too’s statements, and that the Prosecution has additionally not relied on s 32 of the EA before this court. Given this to be the case, and given also that no other section in the CPC can independently admit Too’s statements, it must follow that Too’s statements are inadmissible via s 378(1)(b)(i) of the CPC since the application of this section is subject to ss 10, 30 and 32 of the EA, which the Prosecution quite correctly has conceded does not allow for the admission of Too’s statements. Indeed, as noted above (at [90]), I should point out that illus (b) of s 30 of the EA contains a positive prohibition against the admission of statements in the nature of Too’s statements (as will be seen below). Since s 378(1)(b)(i) of the CPC is expressed to be subject to s 30 of the EA, Too’s statements are not admissible. In essence, this is because as Too’s statements are confessions, they are admissible only by the rules governing the admissibility of confessions, and if such rules do not allow for the admission of Too’s statements, they cannot be so admitted, notwithstanding the fact that Too is dead.

104      Nonetheless, I should point out that s 30 of the EA concerns the confession of a co-accused and, despite the Prosecution’s ready concession that it does not apply in the present case, there may be some value in examining its ambit a little more closely.

(1)        Section 30 of the EA

105      Section 30 of the EA applies only where there is a joint trial involving the offenders for the same offence; it does not apply otherwise. Section 30 also implies that where offenders are tried separately, a confession of one is not to be taken into consideration by the court against another offender in a separate trial. So much is clear from illus (b) to s 30, which provides as follows:

A is on his trial for the murder of C. There is evidence to show that C was murdered by A and B and that B said: “A and I murdered C”.

This statement may not be taken into consideration by the court against A as B is not being jointly tried.

106      While it is certainly true that an accomplice may be called by the Prosecution as a witness in a separate trial against the accused, this is a wholly different matter where the confession of an accused in a separate trial is tendered in evidence against another accused in another trial. In the first case, s 135 of the EA provides that an accomplice is a competent witness against an accused person. Being competent, he is also a compellable witness. Thus, where he is called to testify against the accused in a separate trial, the accused is afforded the opportunity to cross-examine the accomplice. Indeed, it seems that the High Court in Roy S Selvarajah v PP ([69] supra) at [59] has even accepted the discretion founded at common law to exclude accomplice evidence if there was an obvious and powerful inducement for him to ingratiate himself with the Prosecution. I do not propose to comment fully on the validity of any such discretion, save to say that the court of three judges has recently in Law Society of Singapore v Tan Guat Neo Phyllis [2008] 2 SLR 239 (“Phyllis Tan”) persuasively ruled that apart from the confines of the EA, there is no residual discretion to exclude evidence which is otherwise rendered legally relevant by the EA. The point which I would emphasise is that in cases involving accomplice evidence, there is nevertheless scope for cross-examination and the accomplice’s version of the events may be subject to adequate testing by the accused himself.

107      However, such an opportunity does not present itself when the confessor-accomplice is being tried in a separate trial and his confession is being relied upon against the accused. It is trite law that an accused is competent to give evidence on behalf of himself but is not compellable to do so (see s 122(3) of the EA). Thus, where the accomplice confesses in a separate trial but is not called to testify in the accused’s trial, there is no chance for the accused to test the accomplice’s version of events as contained in his confession in the separate trial. In fact, there is no assurance that even the accomplice-confessor’s version of events would be inevitably tested in his own trial by way of cross-examination. Thus, the accomplice-confessor’s confession obtained in a separate trial ought, in principle, to be inadmissible against the accused at the latter’s trial. This is precisely the situation contemplated in illus (b) to s 30 of the EA since, using the terms of the illustration, when B says elsewhere (presumably in a separate trial) that he and A murdered C, that statement amounts to a confession on his part. In fact, P S Atchuthen Pillai in The Law of Evidence (N M Tripathi Private Limited, 3rd Ed, 1984) notes at p 130 that in 1873, two persons were concerned in the murder of a third person; the first absconded, but the second was arrested and tried in the same year. The second person, during his defence at the sessions trial, made a statement that the first person had admitted to him to having committed the murder. In 1877, the first person was arrested and, at his trial (the second person having meanwhile died), the court of sessions used the statement which the second person had made at his trial in 1873 as evidence against the first person. It was held that the second person’s statement was not admissible against the first, under ss 30 and 133 of the Indian Evidence Act (Act No 1 of 1872): Crown v Jhaba (1878) 13 Punjab Record 34.

108      At this point, I readily acknowledge that s 30 of the EA is seemingly an aberration of the preceding paragraph given that in the situation contemplated under s 30, there is similarly no opportunity for the co-accused to cross-examine his fellow confessor-accused in the same trial because the confessor-accused is, pursuant to s 122(3) of the EA, not compellable to give evidence on behalf of himself in the joint trial. Yet, in this case, the confession is admissible and can be taken into consideration by the court whereas in cases involving the separate trial of the accomplice and the accused, the accomplice’s confession is inadmissible at the accused’s trial. It is important, therefore, to clarify and rationalise the relationship between s 30 of the EA with situations such as the latter.

(2)        The rationale behind section 30 of the EA 

109      Prof Michael Hor has in “The Confession of a Co-Accused” (1994) SAcLJ 366 at 380 suggested that the rationale behind s 30 of the EA could be that the confession of a co-accused was hearsay and therefore fraught with the traditional hearsay dangers. There being no convincing countervailing guarantee of reliability, the reason for its admissibility had to rest on the unavailability of the confessor, which was the result of his privilege against self-incrimination. However, as Prof Hor readily admits, if unavailability was the basis of s 30, a number of restrictions in the section were unnecessary. Fundamentally, it should apply to all statements which were sought to be adduced at the trial, and not just to confessions. Also, the accused and the co-accused need not be tried “for the same offence”. As such, it appeared that unavailability was not a satisfactory explanation behind the rationale of s 30.

110      Perhaps a more plausible rationale can be found in the need for a joint trial. Pillai in The Law of Evidence ([107] supra) at p 129 quotes from one of the earliest commentaries on the Indian Evidence Act by Sir Henry Cunningham, who made the following attempt:

[J]udges are relieved from attempt to perform an intellectual impossibility by a provision that, when more persons than one are tried for an offence, and one of them makes a confession affecting himself and any other of the accused, the confession may be taken into consideration against such other person as well as against the person making it. Such a confession is, of course, in the highest degree suspicious; it deserves ordinarily very little reliance; but nevertheless it is impossible for a judge to ignore it, and under the Indian Evidence Act, he need no longer pretend to do so.

It bears emphasis, however, that these comments are made in relation to a joint trial and not merely instances of joint charges which can conceivably take place in different proceedings before separate judges.

111      Indeed, the UK Committee in its Eleventh Report (at para 251) pointed to the same rationale behind cl 31(2) of the UK Bill, which achieves substantially the same effect as s 30 of the EA in joint trials. The UK Committee noted that this clause, in making the co-accused’s statement admissible against another co-accused, got rid of the “absurd situation” which occurred under the prevailing law then that, when co-accused A had made a statement implicating himself and co-accused B, it was necessary to direct the jury that the statement was admissible in evidence against A but not against B. The UK Committee concluded (ibid) that “[t]his is a subtlety that must be confusing to juries, and in reality they will inevitably take the statement into account against both accused”. However, given my observation in the preceding paragraph that cl 31(2) applies merely to joint charges and not joint trials, such a blanket conclusion may not be entirely accurate.

112      Thus, it appears that s 30 of the EA was designed to avoid a situation in a joint trial whereby one of the co-accused had confessed to the charge, and yet, the court was being asked to perform the intellectually difficult task of excluding this evidence against the other co-accused. This means that the court should still be cautious of the dangers of a co-accused’s statements, given the potential lack of an opportunity to cross-examine its maker. But rather than being forced to pretend the irrelevance of such confessions in relation to another co-accused, s 30 of the EA removes the need for such pretence and admits such confessions, but only in the limited circumstances as prescribed under the section. The law has not seen it fit to entirely discard its concerns with the unreliability of such confessions; indeed, apart from this limited circumstance, there is no admissibility of such confessions. In only the limited circumstance of joint trials has the law struck the balance between the prevention of unreliability and the prevention of an impossible intellectual exercise in favour of the latter.

113      Before leaving the subject, I acknowledge that this court has, in a series of cases since Chin Seow Noi v PP [1994] 1 SLR 135 (“Chin Seow Noi”), ruled that a conviction may be founded entirely on the confession of a co-accused alone under s 30 of the EA. Given our analysis above, and given the law’s seeming concern with the unreliability of a co-accused’s confession in situations apart from s 30 (to the extent of making such confessions inadmissible against another accused), it does seem a bit out of the ordinary for a co-accused’s confession admitted under s 30 to be attributed so much weight to the extent of it being able to secure a conviction on its own. The need to reconsider this decision may come in the future.

(3)        Can the rationale behind section 30 of the EA be extended to cover Too’s statements?

114      Returning to the present case, it follows that if the co-accused is dead and is no longer involved in a joint trial with the accused, there is no need to avoid the intellectual difficulty which could plague the court hearing the joint trial. The rationale behind s 30 of the EA stands on its own and is confined only to the limited situations of joint trials; it does not extend beyond these boundaries. Accordingly, that rationale cannot be extended to admit Too’s statements.

115      More fundamentally, once outside the ambit of s 30 of the EA, the law, as I have stated, places considerable emphasis on the danger of unreliability of such co-accused confessions. The fact that its maker has died after making such confessions does not increase its reliability in the slightest extent; there is no causal connection between the survival of the confessor and its reliability if it is made in circumstances which would be conducive to fabrication. As such, I am unable to see how Too’s statements can in effect be re-characterised as a witness’s statement and admitted by s 378(1)(b)(i) of the CPC simply because he has now been executed for the very offence to which he confessed as well as for which he inculpated the accused. The unreliability remains. Too may now be dead but the dangers of unreliability remain very much alive. The inadmissibility must thereby remain similarly undissipated because s 378(1)(b)(i) is expressed to be subject to the rules governing the admissibility of confessions, and such rules make it very clear that, apart from the limited situations of joint trials (and other attendant conditions in s 30 of the EA), such confessions are inadmissible.

116      Finally, to prevent any argument that a “facilitative” approach ought to be taken in respect of s 30 of the EA to permit the admissibility of Too’s statements, I ought only to state that the court of three judges in Phyllis Tan ([106] supra) has, once again, very persuasively (albeit implicitly) held that the “facilitative” characterisation of the EA cannot extend the ambit of the EA to cover common law developments which are inconsistent with it. Given that s 5 of the EA has exhaustively defined all the facts which are legally relevant (and hence admissible) and s 30 has in turn expressly provided that an accomplice’s confession in a separate trial is inadmissible against another accused in another trial, it is impossible to construe s 30 in any other way. In my view, the EA is not a facilitative statute in the sense of assisting in the application of new evidentiary rules. It cannot facilitate the application of common law rules if these rules are inconsistent with the will and intent of Parliament. However, the EA does have the opposite facilitative role of enabling the decision maker to identify the applicable rules of evidence by simply referring to the EA.

Conclusion in relation to the admissibility of Too’s statements

117      For the above reasons, I am of the view that Too’s statements cannot be admitted pursuant to s 378(1)(b)(i) of the CPC. It is clear from a perusal of both the CPC and the EA that s 378(1)(b)(i) needs to be read subject to the admissibility provisions relating to confessions in the EA. Having considered that none of these provisions allow for the admission of Too’s statements, it is abundantly clear that no reasonable construction of s 378(1)(b)(i) of the CPC would allow for their admission. Accordingly, I find that the trial judge was incorrect in admitting Too’s statements and these statements will be disregarded when I consider whether the appellant’s conviction ought to stand. Before I do that, I still need to resolve the second legal question concerning s 34 of the Penal Code since it is the ambit of s 34 which will determine, to some extent (given the trial judge’s primary reliance on his finding that it was the appellant who actually strangled the deceased to death), the appellant’s conviction of the Charge.

Interpretation of section 34 of the Penal Code

Introduction and preliminary observations 

118      The interpretation of s 34 of the Penal Code was not dealt with at any great length by the trial judge. He was apparently satisfied that the interpretation of the section had been conclusively resolved by a number of decisions emanating from this court. In fairness to him, it ought to be pointed out that the conflicting considerations were not adequately developed by counsel. In my view, the interpretation of s 34 is far from clear, although I should make clear that this should not be read as pinning any blame whatsoever on the trial judge in not considering the interpretation further. This is, in part, because of the doctrine of stare decisis which binds the trial judge to the decisions of this court. Given the importance of achieving a coherent and consistent interpretation of s 34 of the Penal Code, I propose to deal with this issue in detail here. In this connection, this court had invited both parties to submit additional arguments on the interpretation of s 34, and we heard their oral arguments on 4 October 2007.

119      Section 34 of the Penal Code provides as follows:

Each of several persons liable for an act done by all, in like manner as if done by him alone.

34.       When 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.

In just 39 plain words, s 34 purports to lay down a principle of general application in such unvarnished language that it belies the inherent difficulties which are hidden beneath. Section 34 concerns an area of criminal law which is deceiving in its apparent simplicity. The provision is regarded as having been crafted to address situations where it is difficult to prove the exact role of each of the wrongdoers and draw distinctions between all those involved in acts in furtherance of a common intention. This encompasses complicity in the criminal law. As K J M Smith stated in his treatise, A Modern Treatise on the Law of Criminal Complicity (Clarendon Press, 1991) at p 1, “[c]omplicity’s function is to determine the circumstances when one party … by virtue of prior or simultaneous activity or association will be held criminally responsible for another’s … wrongful behaviour”. Because of its general function throughout the criminal law, the opportunities for extreme confusion and complexity have been considerable, and the courts have certainly contributed in no uncertain way to the growing confusion and complexity. In this context, the interpretation of s 34 of the Penal Code has vexed the highest courts of countries which have adopted its phraseology in their criminal codes. This is true even in distant jurisdictions such as Sudan and Northern Nigeria (see generally Alan Gledhill, The Penal Codes of Northern Nigeria and the Sudan (Sweet & Maxwell, 1963)). We here in Singapore should not pretend to be free from such difficulties.

120      The particular scenario involving s 34 of the Penal Code which troubles the courts is that involving a “twin crime” situation. There is relatively little difficulty in relation to a “single crime” situation. In such a situation, all the participants to the criminal enterprise share the intention to commit the criminal act eventually perpetuated even though only one or more of them may have physically perpetuated the offence itself. Here, s 34 of the Penal Code clearly applies to render all the participants criminally liable for the offence which was the result of the criminal act perpetuated if it could be said that the remaining participants somehow participated in the criminal act. In contrast, in a “twin crime” situation, what happens is that a group of participants in a criminal enterprise agree on the main goal to commit a primary criminal act but did not share in the intention of one or more unidentified members of the group to also commit a collateral criminal act which is incidental or collateral to the main goal of the group. For convenience, I shall term the offender who actually committed both the primary and collateral criminal acts as the “primary offender” and the offender who merely committed the primary criminal act but is sought to be made liable for the collateral criminal act by virtue of s 34 as the “secondary offender”. The issue which is in need of clarification is the fault element required of the parties under s 34 of the Penal Code in order to hold them liable for the offence which results from the collateral criminal act committed by one or more unidentified members of the group.

121      In this regard, I note that an eminent academic, Prof M Sornarajah, has commented that the present interpretation of s 34 of the Penal Code, if indeed such an interpretation can be stated to exist with any confident precision, has taken so deep a hold in the law that it is too late in the day to reopen these issues of interpretation (see M Sornarajah, “Common Intention and Murder under the Penal Codes” [1995] Sing JLS 29 at 31).

122      With respect, I disagree with this view of the learned professor. In my opinion, the courts should never shy away from re-examining the interpretation of any statutory provision (or any principle of common law) if it is found that the existing interpretation is not satisfactory or is plainly wrong. The fact that the existing interpretation has been around for a long time does not preclude the courts from re-examining such interpretation, if the doctrine of stare decisis is not offended. This court, which stands at the apex of the judicial system in Singapore (and is hence spared from the constraints of stare decisis: see Practice Statement (Judicial Precedent) of the Court of Appeal [1994] 2 SLR 689) is constitutionally charged with the responsibility of departing from any existing interpretation if such interpretation no longer assumes the currency of accuracy or correctness. There are countless examples where this court has seen fit to depart from an existing interpretation of a statutory provision or a principle of common law. It suffices only to list some of these.

123      First, in the criminal arena, as I have referred to above, this court in Chin Seow Noi ([113] supra) re-examined the application of s 30 of the EA after a long period of sustained and consistent interpretation by other courts. In the civil arena, albeit not a decision emanating from this court, one need not look further than the oft-cited case decided by Lai Kew Chai J in Sumitomo Bank Ltd v Kartika Ratna Thahir [1993] 1 SLR 735, which declined to follow the antiquated English case of Lister & Co v Stubbs (1890) 45 Ch D 1, a decision of the English Court of Appeal. The Privy Council in Attorney-General for Hong Kong v Charles Warwick Reid [1994] 1 AC 324 later endorsed the Singapore decision and held that Lister & Co v Stubbs was wrongly decided. More recently, in Ho Soo Fong v Standard Chartered Bank [2007] 2 SLR 181, this court declined to follow its own decision in Khushvinder Singh Chopra v Mookka Pillai Rajagopal [1999] 1 SLR 589, which in turn followed the principle in Owners of Dredger Liesbosch v Owners of Steamship Edison [1933] AC 449, a case decided by the House of Lords. Finally, in Spandeck Engineering (S) Pte Ltd v Defence Science & Technology Agency [2007] 4 SLR 100, this court re-examined the applicable principles relating to a claim in negligence for pure economic loss and departed, to some extent, from existing case law.

124      These examples serve to remind that it is only through the diligence and boldness of courts to re-examine and, in appropriate cases, depart, from existing interpretations of statutory provisions or principles of common law that the law can maintain its relevance and coherence. Judges should not be prepared to condone mistakes (if so shown) only because such law has been promulgated for time eternal. The law as we all know, being primarily formulated by the imperfections of the pen wielded by legislators and interpreted by the equally not invariably infallible minds of the judicial officer, cannot unfailingly paint a picture of perfection, nor can it pretend to be. However, through the willingness and openness of the courts in re-examining the legal principles and precedents from time to time, such imperfections can be limited and, in appropriate circumstances, re-moulded and corrected. It is also appropriate to mention at this juncture that judges, in re-examining the law, do not and cannot function alone without the assistance of counsel. It is in acknowledgement of this that I should record our appreciation to counsel who responded to our invitation to tender additional arguments, both written and oral, to this court on the interpretation of s 34 of the Penal Code. Not surprisingly, the parties urged this court to take conflicting interpretations of s 34. The controversy centred on two well-known local cases, viz, Rex v Vincent Banka [1936] MLJ 53 (“Vincent Banka”) and Wong Mimi v PP [1972-1974] SLR 73 (“Mimi Wong”), and on which of their interpretations was correct. I shall consider this below. It would be churlish if I do not also record our indebtedness to the academics whose articles I shall refer to in great detail below.

125      With these preliminary observations, I now turn to re-examine the interpretation of s 34 of the Penal Code.

Background on the Indian Penal Code

The enactment of the Indian Penal Code

126      Returning to the interpretation of s 34, I propose to first provide a brief summary of the history behind the enactment of the Indian Penal Code itself. Such a summary will provide the backdrop from which to understand the legislative changes to s 34 and to further decipher its meaning. In this connection, I can do no better than to refer primarily to a comprehensive article (along with the sources cited within) written by my judicial colleague, Andrew Phang Boon Leong JA, when he was still in legal academia, viz, “Of Codes and Ideology: Some Notes on the Origins of the Major Criminal Enactments of Singapore” (1989) 31 Mal LR 46.

127      I note that the Indian Law Commission prepared the draft Indian Penal Code and submitted it to the Governor-General of India in Council on 14 October 1837 (see Thomas Macaulay, Indian Penal Code (Reprinted: The Lawbook Exchange, Ltd, 2002) (“Macaulay”) at p viii). A revised Indian Penal Code was prepared and introduced by Barnes P Peacock (an individual who will feature later in the discussion of s 34), Sir James William Colvile (another individual who will similarly feature in relation to s 34), J P Grant, D Eliott and Sir Arthur Buller. It was read for the first time on 28 December 1856. The Indian Penal Code Bill was read a second time on 3 January 1857 and was referred to a select committee which was to report thereon after 21 April 1857. The Supplement to the Calcutta Gazette of 21, 24 and 28 January 1857 published the Indian Penal Code Bill after its second reading. It was eventually passed by the Legislative Council of India and received the assent of the Governor-General on 6 October 1860 (see A C Patra, “Historical Introduction to the Indian Penal Code” in Essays on the Indian Penal Code (N M Tripathi Private Ltd, 1962) at p 41).

128      Thus, after subsequent revisions, the Indian Penal Code (Act No 45 of 1860) was enacted as the general criminal law of India in 1860. It retained the essential substance and form of Lord Macaulay’s draft (see Phang JA’s “Of Codes and Ideology” at 54, especially fn 57 and the authorities cited therein). The Indian Penal Code was eventually introduced to the Straits Settlements in 1872, and continued, with relatively few substantive amendments, even up to today (see also Shaiful Edham bin Adam v PP [1999] 2 SLR 57 (“Shaiful”) at [50]). This short historical exposition allows me to next consider how the court should interpret the Indian Penal Code, particularly if recourse to the English common law at the time of its enactment is permissible.

The interpretation of the Indian Penal Code

129      As will later be seen, a key consideration is whether it is permissible to refer to the pre-existing English criminal law at the time the Indian Penal Code was enacted to aid in the latter’s interpretation. In this area, I find some contradictory views. Whitley Stokes in The Anglo-Indian Codes (Clarendon Press, 1887) describes the Indian Penal Code as being based upon “the law of England, stript of technicality and local peculiarities, shortened, simplified, made intelligible and precise; but suggestions were derived from the French Code Penal and from Livingston’s Code of Louisiana” [emphasis added]. However, when one examines the comments of the Indian Penal Code’s principal drafter, Lord Macaulay, then one would unmistakably find that he does not overtly acknowledge that the Code was based on the existing English common law. In fact, Lord Macaulay seems to purposely distance the Indian Penal Code from any suggestion that it was based on the English common law then in existence. In his report to the Governor-General of India in Council, he stated (see Macaulay at p i):

Your Lordship in Council will perceive that the system of penal law which we propose is not a digest of any existing system, and that no existing system has furnished us even with a ground work. [emphasis added]

130      Similarly, as Prof Eric Stokes in The English Utilitarians and India (Clarendon Press, 1959) at p 226 pertinently pointed out, the Indian Penal Code was not “merely an attempt to apply a reformed English law to Indian conditions”. Indeed, Prof Stokes further commented that “Macaulay’s aim was a code that was not derivative from the laws of any creed or country but sprang from the universal science of jurisprudence” (at p 227).

131      However, later judicial decisions have found it artificial to regard Lord Macaulay as completely disregarding English law in drafting the Indian Penal Code. In the Privy Council decision of Barendra Kumar Ghosh v Emperor AIR 1925 PC 1 (“Barendra Kumar”), Lord Sumner said (at 8):

That the criminal law of India is prescribed by and, so far as it goes, is contained in the Indian Penal Code, that accordingly (as the Code itself shows) the criminal law of India and that of England differ in sundry respects, and that the Code has first of all to be construed in accordance with its natural meaning and irrespective of any assumed intention on the part of its framers to leave unaltered the law as it existed before, are though common-places, considerations which it is important never to forget. It is, however, equally true that the Code must not be assumed to have sought to introduce differences from the prior law. [emphasis added]

132      Indeed, the better view, as stated by W W Chitaley and V B Bakhale in The Indian Penal Code (XLV of 1860) (The All India Reporter Ltd, 3rd Ed, 1980) vol 1 at p 6, is that the Indian Penal Code is meant to be a codifying statute and is intended to be complete in itself with regard to the subject-matter with which it deals. Hence, it is not permissible to refer to the previous law on the subjects dealt with by the Code, for the purpose of adding to the provisions of the Code. However, reference to previous case law, as the Privy Council implied in Barendra Kumar, is not precluded for the purpose of throwing light on the true interpretation of the words of the statute, where the meaning is ambiguous or doubtful. This is based, in part, on the assumption that the Indian Penal Code, as innovative as it is, was nonetheless based on some part of the existing body of English law at the time of its enactment.

133      Indeed, the same point is made specifically in relation to s 34 of the Indian Penal Code. In Ibra Akanda v Emperor AIR 1944 Cal 339 at 359, Khundkar J said:

It will now be clear that the expression “common intention” in S. 34 is not free from ambiguity, and therefore in order to ascertain its real meaning it is permissible, and indeed necessary, to consider what are the principles of joint criminal liability in the English Common Law.

Similarly, in the later case of Bashir v State AIR 1953 All 668 (“Bashir”), I find the following expression (at [24]):

Section 34 has to be interpreted in conformity with the English law … The interpretation that we have placed upon the various expressions used in the section is in conformity with the English law.

134      With this general background in relation to the Indian Penal Code in mind, it is now apposite to evaluate the Singapore courts’ existing interpretation of s 34 of the Penal Code.

The correct interpretation of section 34 of the Penal Code

Typical requirements

135      It is generally agreed that four elements are required to make s 34 of the Penal Code applicable: there must be (a) a criminal act; (b) participation in the doing of the act; (c) a common intention between the parties; and (d) an act done in furtherance of the common intention of the parties. I will now proceed to examine these elements as explained by our courts in a “twin crime” situation.

Criminal act

136      There is usually no problem with the requirement of a “criminal act”, save that some decisions have perhaps been a bit inaccurate in stating that the common intention must be to commit a particular “offence”. In this regard, it is important to bear in mind that the “criminal act” that is “done by several persons” in s 34 does not refer to the actual crime done only. It is essential to realise that the expression “criminal act” is not synonymous with “offence” as defined in s 40 of the Penal Code, which provides as follows:

“Offence”.

40.—(1) Except in the Chapters and sections mentioned in subsections (2) and (3), “offence” denotes a thing made punishable by this Code.

Thus, a single criminal act may involve and give rise to several “offences”. In other words, as the learned authors of The Indian Penal Code (XLV of 1860) point out at vol 1 p 160, the expression “criminal act” in s 34 means 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.

137      Similarly, in Barendra Kumar ([131] supra), Lord Sumner said (at 9) that “a criminal act” means “that unity of criminal behaviour, which results in something, for which an individual would be punishable, if it were done by him alone”. Thus, the “criminal act” refers to all the acts done by the persons involved which cumulatively result in the criminal offence in question. The acts committed by the different parties in the criminal action may be different but all must in one way or another participate and engage in the criminal enterprise. This shows that the expression “criminal act” and “offence” do not mean the same thing for the purpose of s 34 of the Penal Code. When it is said that several persons do “a criminal act” in furtherance of their common intention, the meaning is not that they commit an “offence” in furtherance of their common intention. It is not the offence that they plan or carry out but it is an act or a continuum of acts. In short, it is a criminal design. The offence or offences are committed in the course of their carrying out their criminal design.

Participation

138      Continuing from the above, s 34 of the Penal Code does not apply unless such a “criminal act” has been “done by several persons”. Participation in the criminal act is the main feature of s 34 of the Penal Code and it is this which explains why the persons involved are made to share in the criminal liability for the offence jointly. It necessarily follows from this that a person cannot be made liable for an offence with the help of s 34 unless he has actually participated in the commission of the crime. In other words, the mere agreement between a number of persons to commit a certain crime is not enough for the purpose of this section. Such persons could be committing the offence of a criminal conspiracy, but they would not fall foul of s 34.

139      The more pertinent question is what can constitute “participation”. The cases have been divided as to the extent of such participation. In some cases, it has been held that participation can be active or passive. Where the participation is passive, it has further been held that mere presence can be participation. In other words, presence is evidence of participation, although the lack of presence does not necessarily mean that there is no participation. An early statement of the law can be found in the High Court case of Chew Cheng Lye v Reg [1956] MLJ 240, where Whitton J said at 241:

In my opinion before the section [ie, s 34 of the Penal Code] applies there must on the construction of its language be more than a joint venture to commit crime. I would say it must be established that more than one person participated in the commission of the criminal act which forms the subject of the charge, although in certain circumstances mere presence may constitute participation. In this case I do not think participation on the part of the appellant has been proved. [emphasis added]

(1)        The need for presence? 

140      Regrettably, over the years, the view that presence could be merely indicative of participation has evolved to the proposition that presence is the only possible indication of participation. In other words, presence is a requirement for there to be participation. It is therefore necessary to reconsider whether presence is better viewed as being indicative of participation or whether presence must be insisted upon to establish participation. The issue of whether physical presence is required at the scene of the criminal act is one which has engendered sharp disagreement in both India and in Singapore.

^{(A)        THE SINGAPORE COURTS' PRESENT REQUIREMENT} 

141      In PP v Gerardine Andrew [1998] 3 SLR 736 (“Gerardine Andrew”), the requirement of presence at the scene of the criminal act was laid down by this court. Without presence, it was said there could be no participation in the criminal act and hence no liability under s 34 of the Penal Code. In that case, the appellant and three others were charged with causing the death of the appellant’s landlady under s 302 read with s 34 of the Penal Code. One of the co-accused pleaded guilty to a charge of culpable homicide not amounting to murder under s 304(a) of the Penal Code and was sentenced to seven years’ imprisonment, leaving the appellant and two others to be tried in the High Court for murder. The appellant was convicted of culpable homicide not amounting to murder under s 304(a) read with s 34 of the Penal Code and sentenced to eight years’ imprisonment while the other two accused were convicted of murder under s 302 read with s 34 of the Penal Code and sentenced to death. The Prosecution appealed for the appellant’s conviction under s 304(a) read with s 34 of the Penal Code to be set aside and that she be convicted of the original charge of murder under s 302 read with s 34 of the Penal Code.

142      This court allowed the Prosecution’s appeal. It reiterated that to establish liability under s 34 of the Penal Code it had to be clear an accused had participated either actively or passively in the criminal act. Accordingly, this court held that the proper issue before it was not whether participation in the criminal act was a necessary element of s 34 liability, but whether there could be participation in the criminal act notwithstanding the fact that an accused was not physically present when the act constituting the offence was committed. After examining a long line of Indian cases, this court opined that the differing positions taken by the courts indicated the difficulty in balancing the proper use of the doctrine of common intention as found in s 34. In the event, it decided that presence at the scene of the offence was necessary (at [34]):

Section 34 operates on the basis of joint liability and its essence lies in the doing of a criminal act which furthers the common intention of the participants. The participants are regarded by virtue of this joint liability as principals in the crime. In abetment on the other hand, the actual offence is committed by someone else other than the abettor. The very nature of s 34 demands a closer association with the actual commission of the offence, as compared to abetment where the person is punished for aiding or abetting the principal. There is therefore no requirement that an abettor must be present at the immediate scene of the crime in order for there to be liability for abetment. In our view, to hold that an accused can be liable under s 34 despite being absent when the commission of the offence occurred would render much of the abetment provisions in the Penal Code otiose. ... In our view, therefore, in order for an accused person to be liable under s 34, there must be a requirement that he was physically present when the commission of the offence took place. [emphasis added]

143      On the facts, this court held that although the appellant was not in the flat when the stabbing took place, she was at the staircase landing near the flat. As such, it could not be said that she was in a place remote from the actual occurrence of the offence and the clear inference from the whole of her behaviour was that, if she did see anyone coming, she would have warned her companions who were in the flat. This court held, accordingly, that her behaviour satisfied the requirements of s 34 and her conviction of culpable homicide not amounting to murder was accordingly substituted by a conviction of murder. As such, Gerardine Andrew stands for the proposition that the secondary offender must not only be present at the primary offence, he must also be present at the collateral offence in order to fulfil the requirement of “participation”. I will deal with the issue of whether participation in the collateral criminal act is required later; for now, I am concerned with the broader ruling in Gerardine Andrew that presence at the scene of the criminal act, primary or collateral, is necessary for participation.

^{(B)        RESTATEMENT OF THE REQUIREMENT}

144      I propose to first examine the arguments for and against the need for physical presence at the scene of the criminal act before s 34 of the Penal Code can be attracted. In Criminal Law in Malaysia and Singapore (LexisNexis, 2007), Stanley Yeo, Neil Morgan and Chan Wing Cheong point out (at para 35.30) that the arguments in favour of presence at the scene of the criminal act are said to be as follows. First, s 34 of the Penal Code refers to the criminal act being “done by several persons”. This means that the accused must actually join in the execution of the act and not merely be involved in its planning. Second, a distinction is made between joint liability under s 34 and abetment where presence at the offence abetted is not required. Presence at the offence is therefore a justification for treating a secondary offender in the same league as a primary offender and not just as an abettor. Finally, s 37 of the Penal Code already caters to situations where the acts of the participants are committed at different times and places and this indicates that s 34 should be interpreted such that actual presence of the participants at the criminal act is required. On the other hand, it is the contrary view that proof of participation is all that is needed for s 34. Hence, there can be sufficient participation for the purposes of s 34 even if the accused is not physically present when the act constituting the offence took place.

145      That these opposing views are attractive in their own way can be seen from the contradictory Indian cases on the subject, for example, Shreekantiah Ramayya Munipalli v State of Bombay AIR 1955 SC 287, which insisted that physical presence at the scene must be proved, whereas Jaikrishnadas Manohardas Desai v State of Bombay AIR 1960 SC 889 held that physical presence, though evidential of participation, was not invariably required.

146      In my view, it is a better approach to view participation, not presence, as the key ingredient in imposing liability under s 34 of the Penal Code. It should be a question of fact in each case whether the accused had participated to a sufficient degree such that he is deemed to be as blameworthy as the primary offender. Participation, however, need not in all cases be by physical presence. As the learned authors of Criminal Law in Malaysia and Singapore state at para 35.36, this is particularly so in view of modern technological advances where assistance in committing an offence can be given from afar.

147      As has been noted, in this regard, the Malaysian position is much clearer because the requirement of presence has not been imposed. In the Malaysian Court of Appeal case of Sabarudin bin Non v PP [2005] 4 MLJ 37, Gopal Sri Ram JCA said (at [31]):

In our judgment, presence in every case is not necessary for s 34 to apply. In our judgment, s 34 should be interpreted having regard to modern technological advances. The early decisions on the section, admittedly by the Privy Council, that held presence to be essential for s 34 to bite were handed down at a time when modes of communication were not as advanced as today. It would, in our judgment, be a perversion of justice if we are required to cling on to an interpretation of the section made at a time when science was at a very early stage of development.

I respectfully accept the wisdom behind the pronouncement and hold that presence at the scene of the criminal act, primary or collateral, need no longer be rigidly insisted on for s 34 of the Penal Code to apply. I repeat that the crux of the section is participation, and presence may or may not provide evidence of participation; this is a question of fact to be decided in each case.

(2)        Participation in “twin crime” situations

148      One major controversy which has arisen in relation to the “twin crime” situation is whether there is a need for the secondary offender to participate in the collateral criminal act as well as the primary criminal act. It must be said that, as Prof Michael Hor points out in “Common Intention and the Enterprise of Constructing Criminal Liability” [1999] Sing JLS 494 at 504, for a long time the Privy Council decision of Barendra Kumar ([131] supra) was thought to have settled the matter in favour of the view that participation in the primary criminal act which constituted robbery was enough. Indeed, it was in this context that Lord Sumner pointed out that participation in the criminal act constituting the robbery was a participation in the unity of criminal behaviour, which included the murder eventually committed.

^{(A)        THE SINGAPORE COURTS' PRESENT REQUIREMENT} 

149      As Prof Hor continues at 504 of his article, this state of affairs was thrown into doubt by the decision of the High Court in PP v Tan Joo Cheng [1990] SLR 743 (“Tan Joo Cheng”). In this case, the three accused were charged with murdering one Lee under s 302 read with s 34 of the Penal Code. The Prosecution failed to make out a case of murder against the second accused who instead pleaded guilty to an amended charge of attempted robbery. The Prosecution relied on the provisions of ss 300(c) and 34 against the other two accused alleging that: (a) there was a common intention to commit robbery on Lee and to use the knife to intimidate, counter or overcome any resistance put up by Lee; and (b) in the course of overcoming this resistance, the first accused inflicted the fatal wound. The first accused’s defence was that the stab wound was inflicted in the course of a struggle and that there was no intention on his part to stab Lee.

150      The High Court convicted the first accused of murder and acquitted the third accused. In acquitting the third accused, the High Court held that when Lee was confronted, the third accused was not present, nor did he participate either actively or passively in the holding up of Lee. The High Court further held that for s 34 of the Penal Code to apply, there had to be some form of participation, whether actively or passively, in the act constituting the offence. Accordingly, the High Court ruled that as the third accused was not present when Lee was murdered and did not participate in the stabbing of Lee, he was acquitted of murder and convicted of attempted robbery.

151      In the course of its decision, the High Court in Tan Joo Cheng had evidently found that as the third accused was not present at the scene of Lee’s murder, nor did he hold up Lee, he did not participate in the said murder. This seems to imply that not only need the secondary offender participate in the primary criminal act constituting robbery, he also needs to participate in the collateral criminal act constituting murder in a classic robbery-murder “twin crime” situation. Indeed, according to the analysis of Tan Joo Cheng by Prof Koh Kheng Lian in “Penal Code: Section 34 and Participation” [1992] Sing JLS 232 at 235:

The sort of participation required to bring an accused within the ambit of section 34, as envisaged by the Court, is some participation in the “act constituting the offence” – in the instant case, the actual killing itself. This would then appear to require the accused to be present (i.e., “face to face”) at the killing.

152      The case of Tan Joo Cheng was purportedly followed without further analysis by the decision of this court in Ibrahim bin Masod v PP [1993] 3 SLR 873 (“Ibrahim bin Masod”), which seemed to contradict the requirement of participation in the collateral criminal act as laid down by the High Court in Tan Joo Cheng. In that case, the two appellants were charged with having murdered one Phang between 11 April 1989 and 14 April 1989, in furtherance of the common intention of them both. It was admitted in the court below that the two appellants had met with Phang on 11 April 1989 and had taken him to the second appellant’s flat. It was also not denied that Phang had subsequently been deliberately strangled to death during the period of time when he was held by the two appellants in the said flat. The two appellants had also made various telephone calls to Phang’s family demanding a ransom for his release. Most pertinently, it appeared that the second appellant had strangled Phang while the first appellant had been away from the flat, but the latter assisted the second appellant in disposing of Phang’s body and personal effects. In other words, the first appellant was not present at the scene of the murder.

153      In the event, both appellants were found guilty of murder in furtherance of a common intention. They appealed against conviction and sentence. The second appellant died before his appeal was heard and the appeal was accordingly held to have abated. This court then heard the appeal of the first appellant. It was argued for the first appellant that the evidence at the close of the prosecution case failed to establish a common intention between the two appellants to murder Phang, and further that the first appellant had not been present at the time Phang was strangled to death.

154      In dismissing the first appellant’s appeal, this court addressed the argument that the first appellant had not been present at the time of Phang’s murder in the following manner (at 882, [40]):

The only point of substance in this appeal was that it appears from the evidence that at the moment Liow [the second appellant] wilfully and intentionally strangled Phang to death, Ibrahim [the first appellant] was far removed from the scene selling Phang’s watch in Geylang. This, it was submitted, negated any common intention and reliance was placed on the High Court case of PP v Tan Joo Cheng & Ors where it was said that ‘A person sought to be made liable by s 34 must in some manner participate — whether actively or passively in the act constituting the offence.’ Each case must depend on its own particular facts. In Tan Joo Cheng’s case the learned judges of the High Court found on the evidence that the accused did not actively or passively participate in the commission of the offence. That is not the case here. There is ample evidence as found by the learned judge, with which we agree, that Ibrahim had passively if not actively participated in the killing of Phang, notwithstanding his absence when the actual act of strangling Phang was committed by Liow. [emphasis added]

155      As Prof Hor rightly asks in “Common Intention and the Enterprise of Constructing Criminal Liability” at 505, how could the first appellant in Ibrahim bin Masod, many kilometres away, be said to have participated in the murder without reducing the meaning of “participation” to vanishing point, if one were to assume that the High Court in Tan Joo Cheng was correct to hold that participation in the collateral criminal act was required in a “twin crime” situation? However, Prof Hor has suggested (at 505) a way of reconciling Tan Joo Cheng with Ibrahim bin Masod: If the common intention contains a settled plan to kill (as was found in Ibrahim bin Masod), the act of killing starts with the primary criminal act (of kidnap in Ibrahim bin Masod) and hence participation in the primary criminal act extends to the collateral criminal act to satisfy the requirement of participation. On the other hand, if the common intention contains only a conditional intent, the killing does not start with the primary criminal act, and participation in the killing at a later stage becomes necessary (as was the case in Tan Joo Cheng). In other words, participation in the collateral criminal act is still necessary, although the time when such participation can be said to begin departs on the scope of the common intention to begin with.

156      The requirement of participation in the collateral criminal act was finally concretised in the decision of this court in Gerardine Andrew ([141] supra). I had earlier referred to the facts of this case and noted that this court held that participation (and, in accordance with the reasoning, presence at the scene) of the collateral criminal act was necessary for “twin crime” situations.

^{(B)        RESTATEMENT OF THE REQUIREMENT}

157      In my view, this court in Gerardine Andrew wrongly decided that there was a need for participation in the collateral criminal act as well as in the primary criminal act. Indeed, if participation were necessary in the collateral criminal act, then it would be artificial to still regard the common intention as having not been extended to the collateral criminal act. In other words, an insistence on participation in the collateral criminal act is likely to mean that the common intenders all intended the commission of the collateral criminal act in the first place. Participation in the primary criminal act would be sufficient for liability to fix.

Proving the common intention

158      Common intention refers to the common design of two or more persons acting together. It is the reason or object for doing the acts forming the criminal act. This is different from the intention to commit the offence which is the result of the criminal act committed.

(1)        The Singapore courts’ present requirement 

159      A long line of cases in Singapore have followed the Privy Council case of Mahbub Shah v Emperor AIR 1945 PC 118 (“Mahbub Shah”) which imposed the requirement of proof that the criminal act was done pursuant to a pre-arranged plan before a common intention could be inferred. However, the requirement of a pre-arranged plan has been substantially qualified. It has also been said that a common intention can be formed only a moment before the commission of the offence. It may develop on the spot or during the course of the commission of the offence so long as the evidence shows that the parties were acting in concert.

160      This court in Asogan Ramesh s/o Ramachandren v PP [1998] 1 SLR 286 (“Asogan Ramesh”) (discussed further at [231]–[232] below), for example, explained that where the three appellants were scolded by the victim, and the three appellants developed a common intention on the spot to assault the victim for what he had said, the murder of the victim was committed in furtherance of the common intention to assault him. This follows the decision of the High Court of Allahabad in Bashir ([133] supra), where Desai J said at [13]:

In [Mahbub Shah], “common intention” was held to imply a “pre-arranged plan”. This does not mean either that there should be confabulation, discussion and agreement in writing or by word, nor that the plan should be arranged for a considerable time before the doing of the criminal act. The Judicial Committee in [Mahbub Shah], did not lay down that a certain interval should elapse between the formation of a pre-arranged plan and the doing of the criminal act and did not negative the formation of a pre-arranged plan just a moment before the doing of the criminal act.

(2)        Restatement of the requirement

161      In my view, these existing requirements should be retained. However, I would add that, in most situations, it is virtually impossible to directly prove a pre-arranged plan between the parties. I would stress that common intention to bring about a particular result may well develop on the spot all of a sudden, and may be established from proof of circumstances showing that after some persons gathered at the scene of occurrence they then developed a sudden consensus of their minds to bring about a particular result. The circumstances which are relevant for the purpose of establishing common intention have been stated by the Indian Supreme Court in Pandurang v State of Hyderabad AIR 1955 SC 216 as being:

(a)        common motive or enmity or ill will which is shared by all the accused, or the fact that they all belong to a faction or family which is on terms of rivalry with the faction or family to which the victim belonged; or

(b)        previous conduct of the assailants immediately before the occurrence, such as, that they had met together and then came in a body to the spot where the incident occurred; or

(c)        subsequent conduct of the assailants, such as, that they all ran away after the attack in a body or in the same direction, or that they were all absconding after the happening of the incident.

These non-exhaustive circumstances can lead to the inference of a common intention between the parties. Following the establishment of a common intention, inferences can next be made from the circumstances of the case to show that the criminal act was committed in furtherance of this common intention, such as the conduct of the parties, the weapons used and the nature of the wounds inflicted. Indeed, I would repeat the warning given by the Privy Council in Mahbub Shah that such inferences should never be made unless it was “a necessary inference deducible from the circumstances of the case” (at 121).

Common intention in “twin crime” situations

(1)        The Singapore courts’ present requirement 

162      An issue which has troubled the courts is the mental state of the secondary offender in a “twin crime” situation. In the typical robbery-murder scenario, an early view in the local courts was that a person must have the intention to commit the criminal act constituting murder in common with the person or persons who actually committed the murder. Unless this was proved, he or she could not be liable for it even though murder was committed in order to accomplish some other intention shared in common.

^{(A)        INTENTION TO COMMIT THE OFFENCE COMMITTED} 

163      Thus, in Vincent Banka ([124] supra), described by Myint Soe in “Some Aspects of Common Intention in the Penal Code of Singapore and West Malaysia” (1972) 14 Mal LR 163 at 163 as “[t]he first important reported ruling in Singapore relating to common intention”, it was held by the Court of Criminal Appeal of the Straits Settlements (Vincent Banka at 56) that “there must exist a common intention to commit the crime actually committed, and it is not sufficient that there should be merely a common intention to ‘behave criminally.’” In this case, the two accused had set out to commit robbery and in the process murder was committed. The evidence was inconclusive as to which of them inflicted the fatal wound or who carried the knife used to inflict the wound. It was held that there must be a common intention between the robbers not merely to commit robbery but, if necessary, to kill the victim. Since there was no express agreement between them that a knife should be carried or that the victim should be stabbed, the court convicted them of robbery and acquitted them of murder. The effect of this approach was that the Prosecution was required to prove that the secondary party possessed the fault element for the actual offence charged.

164      In acquitting the two accused and stating the law in relation to s 34 of the then Penal Code, Huggard CJ relied on the Burmese case of Emperor v Nga Aung Thein AIR 1935 Rang 89 (“Nga Aung Thein”). The question before the Full Bench of the Rangoon High Court in that case was whether, when less than five persons go out armed to commit robbery without any pre-arranged intention to commit murder but in the course of the robbery one of the robbers does commit murder, all the robbers were liable to be convicted under s 302 read with s 34 of the Penal Code. The Full Bench answered this question by saying that in effect no hard and fast rule could be laid down, as the existence of the common intention was a question of fact to be determined on a consideration of the facts of the case. After considering Nga Aung Thein and some other cases, Huggard CJ in Vincent Banka observed as follows (at 55):

[I]t is necessary to remember that there is a very important difference between the law as enacted in section 34 of our Penal Code and the Common Law of England as to the evidence necessary to establish a common intention. Under our Code it is essential, that there should be evidence of a common intention, or evidence from which such a [common] intention can properly be inferred, to commit the act actually committed. In England that is not essential. [emphasis added]

Thus, after stating that he was in full agreement with the Rangoon High Court in Nga Aung Thein, Huggard CJ further said (ibid):

[I]t follows that it is the duty of the trial Judge, in cases where section 34 of the Penal Code is relied on, to direct the attention of the jury to any evidence from which they may legitimately infer the existence of a common intention to commit the criminal act actually committed … [emphasis added]

165      Myint Soe in his article suggests that the above view expressed by the Straits Settlements Court of Criminal Appeal in Vincent Banka was probably influenced by the case of Nga Aung Thein and the other Burmese cases quoted therein. Page CJ in Nga Aung Thein had said (at 90):

[I]f the persons by whom the criminal act was done had expressly agreed beforehand that they would endeavour to commit the offence, that, no doubt would be cogent evidence that the act, if committed[,] was done in furtherance of the common intention of the conspirators.

166      A similar view was borne out in Nga E v Emperor AIR 1931 Rang 1. In this case, Carr J said (at 6):

The general effect of the cases discussed is that the common intention referred to in S. 34, [Indian Penal Code], is an intention to commit the crime actually committed and that each accused person can be convicted of that crime only if he has participated in that common intention. [emphasis added]

167      Whatever the motivations behind Vincent Banka, in the later case of Rex v Chhui Yi [1936] MLJ 142, the Court of Criminal Appeal of the Straits Settlements interpreted s 34 of the then Penal Code slightly wider. Using murder as an example, it was held (at 144) that this did not mean that:

… there need have been a common intention actually to kill; but there must have been a common intention to do any of the acts which are described in sections 299 and 300 of the Penal Code and the doing of which, if death results, amounts to murder.

Hence, an accused could be held liable for the murder committed by his companion even though the accused did not intend the consequences, provided that he possessed the requisite mens rea for murder under the Penal Code. This included, for example, the situation where such a secondary party knew that the act was imminently dangerous and would in all probability cause death.

^{(B)        COMMON INTENTION TO COMMIT THE PRIMARY OFFENCE}

168      This view changed with the case of Mimi Wong ([124] supra). In this case, the first appellant, Mimi Wong, who was a waitress, came to know a Japanese engineer called Mr Watanabe, and in due course they became intimate and lived together. Mimi Wong had a husband, the second appellant, who was apparently quite fond of her and the children of their marriage. Subsequently, Mrs Watanabe arrived from Japan. Two weeks after her arrival, Mrs Watanabe was stabbed and killed at her home in Singapore. It was later established that the stabbing was done by Mimi Wong, and that she was helped in the crime by her husband. Before she was stabbed, detergent was thrown in Mrs Watanabe’s eyes. The question of common intention which arose with regard to her husband was whether he could be guilty under s 302 read with s 34 of the Penal Code (Cap 119, 1955 Rev Ed).

169      At the trial court, it was found as a fact that the idea of throwing detergent at Mrs Watanabe before stabbing her was from the second appellant. There was thus a common intention to cause death on the part of the second appellant. On appeal, this court affirmed the second appellant’s conviction but made certain remarks about the law in relation to common intention which were to change its course in Singapore. It said that the parties were required only to have the common intention to do a particular criminal act which ultimately led to an offence being committed in furtherance of that common intention. In this case, Wee Chong Jin CJ said (at 80, [25]) that:

If the nature of the offence depends on a particular intention the intention of the actual doer of the criminal act has to be considered. What this intention is will decide the offence committed by him and then s 34 applies to make the others vicariously or collectively liable for the same offence. The intention that is an ingredient of the offence constituted by the criminal act is the intention of the actual doer and must be distinguished from the common intention of the doer and his confederates. It may be identical with the common intention or it may not. Where it is not identical with the common intention, it must nevertheless be consistent with the carrying out of the common intention, otherwise the criminal act done by the actual doer would not be in furtherance of the common intention. [emphasis added]

170      As Myint Soe notes drily in his article ([163] supra) at 171, this statement by Wee CJ bears a striking resemblance to the following passages from the Indian case of Bashir ([133] supra) at [21]:

If the nature of the offence depends on a particular intention or knowledge, the intention or knowledge of the actual doer of the criminal act is to be taken into account. What intention or knowledge will decide the nature of the offence committed by him and the others will be convicted of the same offence because, as pointed out above, they cannot be convicted of a different offence. The intention of the actual doer must be distinguished from the common intention as already pointed out. It is an ingredient of the offence said to be constituted by the criminal act. … Though the intention of the actual doer is to be distinguished from the common intention, the intention must not be foreign to or inconsistent with the common intention. It must be consistent with the carrying out of the common intention, otherwise the criminal act done will not be in furtherance of the common intention.

171      Therefore, after Mimi Wong, there was no longer a need, in a “twin crime” situation, for the common intention of the secondary offenders to be to commit the criminal act which eventually constituted the offence which the primary offender was charged with. It suffices that the intention of the primary offender was “consistent” with the common intention of the secondary offenders. If this is so, all will be liable for the eventual criminal act committed. And so Mimi Wong stood for, and still stands, for the wider interpretation of s 34 of the Penal Code in Singapore (henceforth “the Mimi Wong approach”).

^{(C)        OTHER DECISIONS}

172      While all this was going on, Prof Hor notes in “Common Intention and the Enterprise of Constructing Criminal Liability” at 499 that there remains a persistent number of cases which, notwithstanding Mimi Wong, continued to use the Vincent Banka formula. For example, in PP v Tan Bee Hock [1993] SGHC 115, which concerned a murder committed in the course of an armed robbery by the accused and an unknown person, the High Court held that, in addition to the mens rea to commit the primary offence of armed robbery, there must also be a common intention to use the weapons brought along to overcome resistance. In other words, there must have been a common intention to kill should the need arise. The court found that this requirement was satisfied (at [56]–[57]):

The paraphernalia which accompanied the accused and his companion and their deliberateness of purpose on the evening in question left no room for any doubt that there was not only the common intention to commit armed robbery but also to use the knives they had brought along to overcome any resistance they might encounter. As for the accused, the threatening words uttered by him to [the victim’s fiancée] that he would stab her if she made a move and his subsequent conduct in stabbing the [victim], fortified my conclusion that those words were not mere sabre- rattling but that he was intent on using the knife [on] that occasion. Hence when the [victim] resisted he was overpowered and the fatal stabbing took place. There was in this case overwhelming evidence that there existed a common intention between the accused and his accomplice to commit armed robbery and the act of murder was committed in the course of the robbery.

In the circumstances, even if the fatal injuries found on the [victim] were inflicted by [the accused’s companion], the accused would still be guilty of the offence of murder by virtue of the operation of s 34 of the Penal Code.

173      Similarly, in PP v L Hassan [1998] SGHC 357 (“L Hassan”), which was again a robbery-murder case, the High Court required the common intention to be to commit murder, the very offence the accused were charged with committing. At [143]–[144], the court ruled as follows:

The next issue considered by the court concerned common intention. I must observe presently that having regard to the contents of the statements of the first and the second accused, I came to conclude that the prosecution had not to my satisfaction established common intention on the part of the first and the second accused to commit murder on that day. Though common intention could be inferred from the circumstances and could have been formed at the time of the commission of the offence, I gave the benefit of the doubt to the first accused as claimed in his statement that he was unaware at the material time that the third person would hit the victim on his head and inflict the fatal injury. Correspondingly, there was also no evidence to support any inference that the second accused whom I found was one of the persons involved in the robbery on that date, had any common intention to commit murder.

The first and the second accused mounted a vehement denial of their involvement even with the offence of robbery. Though lies can in certain circumstances amount to corroboration because it indicates a consciousness of guilt (see the observations of Lord Lane CJ in R v Lucas [1981] QB 720 at 724 approved in Singapore in Tan Pin Seng v Public Prosecutor [1998] 1 SLR 418), my finding at the end was that the prosecution whilst it had proved beyond a reasonable doubt that the first and second accused had a common intention to commit robbery, had not satisfied me beyond a reasonable doubt that the first and second accused had the requisite common intention to commit murder.

[emphasis added]

In the event, the accused were convicted of robbery with hurt instead.

174      Finally, in Tan Joo Cheng ([149] supra), which I referred to above, the High Court thought that it was necessary to decide whether it was commonly intended that the knife brought along by the accused was to be used to overcome resistance on the part of the robbery victim.

(2)        Restatement of the requirement

175      In my view, the clear pattern of the more recent authorities in Singapore, notwithstanding decisions like L Hassan, has been inclined towards the Mimi Wong approach, that is, the common intention need not be the intention to commit the criminal act constituting the offence actually committed. This position has remained relatively stable in local jurisprudence, and the real problem is in identifying when it can be said that the collateral offence is in furtherance of the common intention, ie, if the mens rea required of the secondary offenders is not that of the collateral offence, what is the additional mens rea that is required then, apart from the mens rea to commit the criminal act constituting the primary offence? I will deal with this issue below.

176      For now, I propose to re-examine the Mimi Wong approach to see if it should still remain. While the requirement which Mimi Wong laid down has been almost invariably followed by the courts ever since, later courts have not really bothered to undertake an extensive analysis of whether the Mimi Wong approach is correct. I feel that this led in part to the slight deviation from the Mimi Wong approach and it would therefore be beneficial to undertake an extensive analysis in this case to lay the matter to rest one way or the other.

^{(A)        THE ARGUMENT FOR A NARROWER READING OF SECTION 34 OF THE PENAL CODE} 

177      I first turn to the justifications behind a “narrower” reading of s 34 of the Penal Code, ie, the approach taken in Vincent Banka. In this respect, Prof Hor in “Common Intention and the Enterprise of Constructing Criminal Liability” at 509 has put forward an argument which calls for a narrower reading of s 34, taking into account the presence of s 35 of the Penal Code and the existence of elaborate provisions which deal with abetment in ch V of the Penal Code. I note that a similar argument was also put forward many years earlier by Gillian Douglas in “Joint Liability in the Penal Code” (1983) 25 Mal LR 259. By this argument, s 35 of the Penal Code must be read together with s 34. Section 35 provides as follows:

When such an act is criminal by reason of its being done with a criminal knowledge or intention.

35.       Whenever an act, which is criminal only by reason of its being done with a criminal knowledge or intention, is done by several persons, each of such persons who joins in the act with such knowledge or intention, is liable for the act in the same manner as if the act were done by him alone with that knowledge or intention.

178      Thus, by this argument, if s 34 of the Penal Code means what the cases taking the Mimi Wong approach say it means, ie, that the common intender is liable for murder although he does not possess the mens rea for murder, then s 35 contradicts this reading of s 34. In this regard, s 35 provides that the common intender is liable only if he possessed “such knowledge or intention” as would qualify the common intender for murder. It is thereby said that s 35 specifies that there is to be no reduction of or construction for the normal requirement of the mens rea. A particular member is liable “in the same manner” only if he had “such [criminal] knowledge or intention” [emphasis added] as would be necessary for the primary or collateral offence.

179      Following from the above, Prof Hor in his article further points out that the elaborate provisions in relation to abetment in the Penal Code contemplate the very “twin crime” situations which s 34 has been concerned with. He refers specifically to ss 111 and 113 of the Penal Code, which I now also set out with the relevant illustrations:

Liability of abettor when one act is abetted and a different act is done.

111.     When an act is abetted and a different act is done, the abettor is liable for the act done, in the same manner, and to the same extent, as if he had directly abetted it:

Provided the act done was a probable consequence of the abetment, and was committed under the influence of the instigation, or with the aid or in pursuance of the conspiracy which constituted the abetment.

Illustrations

(a)        A instigates a child to put poison into the food of Z, and gives him poison for that purpose. The child, in consequence of the instigation, by mistake puts the poison into the food of Y, which is by the side of that of Z. Here, if the child was acting under the influence of A’s instigation, and the act done was under the circumstances a probable consequence of the abetment, A is liable in the same manner, and to the same extent, as if he had instigated the child to put the poison into the food of Y.

(b)        A instigates B to burn Z’s house. B sets fire to the house, and at the same time commits theft of property there. A, though guilty of abetting the burning of the house, is not guilty of abetting the theft; for the theft was a distinct act, and not a probable consequence of the burning.

(c)        A instigates B and C to break into an inhabited house at midnight for the purpose of robbery, and provides them with arms for that purpose. B and C break into the house, and being resisted by Z, one of the inmates, murder Z. Here, if that murder was the probable consequence of the abetment, A is liable to the punishment provided for murder.

Section 113 reads:

Liability of abettor for an offence caused by the act abetted different from that intended by the abettor.

113.     When an act is abetted with the intention on the part of the abettor of causing a particular effect, and an act for which the abettor is liable in consequence of the abetment causes a different effect from that intended by the abettor, the abettor is liable for the effect caused, in the same manner, and to the same extent, as if he had abetted the act with the intention of causing that effect, provided he knew that the act abetted was likely to cause that effect.

Illustration

A instigates B to cause grievous hurt to Z. B, in consequence of the instigation, causes grievous hurt to Z. Z dies in consequence. Here, if A knew that the grievous hurt abetted was likely to cause death, A is liable to be punished with the punishment provided for murder.

180      According to Prof Hor, these two sections are concerned with the Mimi Wong approach to the interpretation of s 34. He points out that, comparing the pre-conditions, the existence of a common intention to commit a primary offence is at least abetment by conspiracy or intentional aiding. Secondly, if the mens rea required of the secondary offenders to hold them liable for the collateral offence is either negligence or recklessness, then this is also aptly covered by ss 111 and 113. Thirdly, if the actus reus required (as held in Gerardine Andrew; however, on this, see [146] above) is for participation in the collateral offence, then the abetment provisions will once again be satisfied since participation in the collateral offence must be abetment of the collateral offence. In fact, s 114 of the Penal Code would then operate to deem the secondary offender who was present at the collateral offence as having committed the collateral offence itself. As such, Prof Hor concludes that the Mimi Wong approach to the interpretation of s 34 renders much of ss 111 and 113 otiose, and that such an interpretation is thereby undesirable. On the other hand, Prof Hor suggests that the Vincent Banka interpretation of s 34 yields a much more convincing account for its conceptual reach which will not then extend into ss 111 and 113.

181      Before stating my views on the learned professor’s suggested interpretation of s 34, I propose to first examine the approach of the other jurisdictions which possess a similar Penal Code (and, more importantly, a similar section as our s 34) to see if one can derive any additional assistance there.

^{(B)        THE APPROACH IN OTHER JURISDICTIONS} 

182      In India, the interpretation of s 34 has been a vexed issue. In the leading textbooks, I find that it is typical for the textbooks to articulate a certain interpretation of s 34 with seeming clarity, and then for this interpretation to be later qualified, or even then contradicted. In saying this, I acknowledge Assoc Prof Kumaralingam Amirthalingam’s view in “Common Purpose and Homicide” (2007) 123 LQR 369 (at 371) that the Indian Supreme Court has “consistently required nothing less than an intention to commit the [collateral] offence as the necessary mens rea”. However, as will be seen, notwithstanding the seeming fortitude of the Indian Supreme Court, the lower Indian courts have not been so consistent, even in the face of apparently binding authority on such a requirement. In view of this, I cannot take the Indian position to be anywhere near settled, and must view it with a degree of respectful reservation.

183      To start off, the Supreme Court of India in Hardev Singh v The State of Punjab AIR 1975 SC 179 (“Hardev Singh”) (at [9]) noted as follows:

The view of the High Court that even the person not committing the particular crime could be held guilty of that crime with the aid of Section 34 of the Penal Code if the commission of the act was such as could be shown to be in furtherance of the common intention not necessarily intended by every one of the participants, is not correct. The common intention must be to commit the particular crime, although the actual crime may be committed by any one sharing the common intention. Then only others can be held to be guilty. In this case assault on Tej Kaur by appellant Hardev Singh was his individual act. There was no common intention to commit the murder or cause grievous hurt to anybody. Circumstances are completely lacking to lead us to any such inference.

This seems to be in line with the Vincent Banka approach.

184      However, there are several other cases which did not insist on actual intention; in these cases, mere “awareness” on the part of the secondary offender of the intention of the primary offender was sufficient. In fact, the authors of The Indian Penal Code (XLV of 1860) ([132] supra) vol 1 at p 171 take the view that Hardev Singh does not in fact support a conclusion that the expression “common intention” refers to the actual offence committed. In any event, the learned authors also opine that this view, even if it was made in Hardev Singh, “is not correct” (ibid). Thus in Nazir v Emperor AIR 1948 All 229, where the common intention of the three accused was to give someone a beating so as to disable him from successfully obstructing the abduction of his wife, which had been planned by the accused, and there was no clear conception as to how such a beating would result, it was held that the accused were aware of the likelihood that death might be caused, and all of them could be held guilty under s 302 read with s 34 of the Indian Penal Code, if the husband was in fact killed. Similarly, in The State v Hira Dubey AIR 1952 Pat 135, Shearer J said that if a number of men set out to burgle a house and one of them was given a revolver with instructions not to use it except under extreme necessity, if that man were to lose his head in the middle of the adventure and shoot and kill an inmate of the house contrary to instructions, his companions would be liable for the murder by virtue of s 34 equally with him, the reason being that under the circumstances of the case all the accused must be held to have been “aware” of the possibility of murder being committed in the course of the adventure. These cases imposed a lower mens rea requirement on the part of the secondary offender; unlike in Hardev Singh, he did not need to have intended the actual crime committed but he nonetheless had to be “aware” of the likelihood of that crime being committed.

185      Then, on the other hand, we have cases such as Bashir ([133] supra), in which Desai J said (at [15]):

These words [“in furtherance of the common intention”] were added by the legislature in 1870 and must have been added for a purpose. That purpose could be none other than to make persons, acting in concert, liable for an act, which is not exactly the act jointly intended by them, but has been done in furtherance of their common intention. The words would not have been required at all if the common intention implied an intention to do the very criminal act done.

On this reading of s 34, an act done in furtherance of a common intention need not be the very act which is intended by the criminals. It may be a preliminary act necessary to be done before achieving the common intention or it may be an act necessary after achieving the common intention, or it may be an act necessary to be done while achieving the common intention. In short, notwithstanding the apparent uniformity in the Indian Supreme Court decisions, the contradicting and inconsistent lower court decisions lead me to the view that the Indian cases do not offer me much assistance.

186      It has similarly been noted by Assoc Prof Amirthalingam in “Common Purpose and Homicide” ([182] supra), citing the cases of Namasiyiam v PP [1987] 2 MLJ 336 and PP v Low Kian Boon [2006] 6 MLJ 254, that the Malaysian courts, in similar fashion to the Indian Supreme Court, adopt the view that the necessary mens rea on the part of the secondary offender is to commit the collateral offence itself. While it is certainly true that this is the view adopted in these cases, it is equally true that there are other cases that purport to follow other interpretations of s 34. For example, in PP v Krishna Rao a/l Gurumurthi [2000] 1 MLJ 274, the court therein in fact cited with approval a Singapore case which supported the Mimi Wong approach to the interpretation of s 34.

187      Thus seen, the Malaysian cases similarly do not yield a convincing account, one way or the other, of the correct interpretation to take of s 34 of the Penal Code.

^{(C)        ANALYSIS: THE HISTORICAL UNDERPINNINGS OF SECTION 34 OF THE PENAL CODE}

188      Having seen that the interpretation of the equivalent of s 34 in other jurisdictions, in particular India, is equally, if not more, unsettled, I feel that it is best to turn to first principles in interpreting s 34 of the Penal Code. On this note, I acknowledge the conceptual attractiveness of Prof Hor’s suggested interpretation, premised as it is on the relevance of s 35 and the abetment provisions. However, in my view, a more convincing account of s 34 can be given if one were to go right down to its origin to assess its purpose in the taxonomy of the Penal Code. It is only with this information that we can properly assess whether Prof Hor’s suggested interpretation based on Vincent Banka or the Mimi Wong approach is more preferable.

189      On this note, it is interesting to note that in Lord Macaulay’s original draft Indian Penal Code, there was no equivalent of s 34 as it exists today (see also “Common Intention and the Enterprise of Constructing Criminal Liability” ([148] supra) at 508, fn 54). In its place was a provision very similar to the present s 35, and this was cl 3 (see Macaulay ([127] supra) at p 1), which provided as follows:

Wherever the causing of a certain effect with a certain intention, or with a knowledge of certain circumstances, is an offence, it is to be understood that if more persons than one jointly cause that effect, every one of them who has that intention, or that knowledge, commits that offence.

Illustrations.

(a)        A digs a pit, intending or knowing it to be likely that he may thereby cause a person’s death. B puts turf over the mouth of the pit, intending or knowing it to be likely that he may thereby cause a person’s death. Here, if Z falls in and is killed, both A and B have committed voluntary culpable homicide.

(b)        A and B are joint gaolers, and as such have the charge of Z, alternately, for six hours at a time. Each of them, during his time of attendance, illegally omits to furnish Z with food, intending or knowing it to be likely that he may thereby cause Z’s death. Z dies of hunger. Both A and B have committed voluntary culpable homicide.

190      It can be seen that there was no contemplation of a common intention in cl 3, which read similarly to the present-day s 35. In particular, the illustrations are clear that the separate parties involved have committed the offence in question separately with the same intention, but not the common intention. This, as we shall see later, shines considerable light on how s 35 is to be interpreted.

191      Moving back to s 34, it was only after repeated revisions to Lord Macaulay’s draft Indian Penal Code that a slightly different version of s 34 appeared in the final version that was passed into law in 1860. The then s 34 did not contain the expression “in furtherance of the common intention of all” but instead read as follows:

When a criminal act is done by several persons, each of such persons is liable for that act in the same manner as if the act were done by him alone.

It was in fact after the    Indian Penal Code Amendment Act 1870 (Act 27 of 1870) was passed that s 1 of the same Act amended s 34 to its present form with the addition of the expression “in furtherance of the common intention of all”. Hence, whatever Lord Macaulay’s original intentions were when he drafted the Indian Penal Code, these were evidently papered over when the final Penal Code was enacted in 1860 because, by that time, he had passed away.

192      As for the 1870 amendment to s 34, according to various textbooks (see, for example, Surya Narayan Misra & Rakesh Kumar Misra, The Indian Penal Code (Central Law Publications, 7th Ed, 1995) (“Misra”) at p 94), the expression was added in consequence of either one or both of two particular Indian cases which were decided prior to 1870. The first case cited by these textbooks is The Queen v Gorachand Gope (1866) Bengal LR Supp 443 (“Gorachand Gope”), in which Sir Barnes Peacock CJ, who I noted earlier was party to the drafting of the Indian Penal Code (see [127] above), stated (at 456–457):

[W]hen several persons are in company together engaged in one common purpose, lawful or unlawful, and one of them, without the knowledge or consent of the others commits an offence, the others will not be involved in the guilt, unless the act done was in some manner in furtherance of the common intention. It is also said, although a man is present when a felony is committed, if he take no part in it, and do not act in concert with those who commit it, he will not be a principal merely because he did not endeavour to prevent it or to apprehend the felon. But if several persons go out together for the purpose of apprehending a man and taking him to the thannah [police station] on a charge of theft, and some of the party in the presence of the others beat and ill-treat the man in a cruel and violent manner, and the others stand by and look on without endeavouring to dissuade them from their cruel and violent conduct, it appears to me that those who have to deal with the facts might very properly infer that they were all assenting parties and acting in concert, and that the beating was in furtherance of a common design. [emphasis added]

193      The second case which is said to have led to the 1870 amendment to s 34 of the Indian Penal Code is the decision of the Privy Council in Ganesh Sing v Ram Raja (1869) 3 Bengal LR 44 (“Ganesh Sing”). The decision of the Board was given by Sir James W Colvile (who, as I noted above, was involved with Peacock CJ in the refinement of the Indian Penal Code), Sir Joseph Napier, Giffard LJ and Sir Lawrence Peel. This case involved a suit to recover damages caused by the defendants’ plundering the house of the plaintiff. The court of first instance ruled in favour of the plaintiff but this decision was reversed by the judges of the Sudder Dewanny Adawlut of Agra on the basis that the fact of plunder was not proved. On appeal to the Privy Council, one of the issues was whether there had been a plunder of the plaintiff’s property to some extent. In this respect, the Privy Council held (at 45–46):

Their Lordships cannot entertain a reasonable doubt, on the whole of the evidence, that there had been a plunder of the plaintiff’s property to some extent, and that it was a joint transaction. During the time of the mutiny the chiefs of some villages collected people together with a preconcerted purpose of plundering the plaintiff’s property, and it is quite plain upon the evidence that all were acting with a common purpose of plunder, that they went to the plaintiff’s house for the purpose of plundering, and each co-operated more or less; and where parties go with a common purpose to execute a common object, each and every one becomes responsible for the acts of each and every other in execution and furtherance of their common purpose; as the purpose is common, so must be the responsibility. [emphasis added]

194      It has been said (see Misra at p 94) that before the 1870 amendments there was some difference between the English law and Indian law, on the subject, for in English law a person not a sharer in the intentions of his companions to commit murder could not be held liable for murder. But under Indian law, a person, even though he was not a party to the evil intent of his companions, could be held liable for murder. Similarly, Sir Hari Singh Gour in The Penal Law of India (Law Publishers (India) Pvt Ltd, 11th Ed, 2000) vol 1 states at p 260 that the words “in furtherance of the common intention of all” make all the difference between the old section and the new, for without those words, the Penal Code would be widely at variance with the English law, where a person not cognisant of the intention of his companions to commit, for example, murder, was never held liable, though he might have joined their company to commit an unlawful act. However, “cognisance” has many levels, and I need to see what exactly English law required, in order to ascertain what s 34 in turn requires.

195      At this juncture, I reiterate that it is important to bear in mind, as the Privy Council said in Barendra Kumar ([131] supra), that it cannot be assumed that the Indian Penal Code (given its evolutionary progress) was intended to differ from the existing English common law at the time of its enactment. More pertinently, it is permissible to refer to the common law to shed light on the interpretation on the Penal Code itself, although this should not be taken as being absolutely conclusive (see [132] above). With this in mind, I think that s 34 of the Penal Code was enacted to put into the Penal Code the common law doctrine of common purpose. If the addition of the words “in furtherance of the common intention of all” was influenced by the Privy Council’s decision in Ganesh Sing, then the use of the expression “common purpose” must give rise to the preliminary hypothesis that s 34 was intended, perhaps not at the outset of its enactment but definitely by the time of its amendment in 1870, to embody the doctrine of common purpose. That one of the drafters of the Indian Penal Code sat on the Board in Ganesh Sing, which itself originated from India, further aids in this hypothesis. Indeed, M A Rabie in “The Doctrine of Common Purpose in Criminal Law” (1971) 88 S African LJ 227 at 229 opines that the common law doctrine of common purpose was given effect to in s 5 of the Native Territories’ Penal Code (Act 24 of 1886) (South Africa), which is substantively similar to our s 34. The wording of the section, viz, the addition of the words “in furtherance of the common intention of all” [emphasis added] seemingly gives strength to this preliminary conclusion.

^{(D)       ANALYSIS: THE DOCTRINE OF COMMON PURPOSE AND SECTION 34 OF THE PENAL CODE}

196      If s 34 was amended in 1870 to make it in line with the prevailing English approach, viz, the doctrine of common purpose, the question which then arises is: What was the doctrine of common purpose at the time of the enactment (and subsequent amendment in 1870) of the Indian Penal Code? I first reiterate that reference to English law is not only permissible, it has been regarded as being necessary in some cases (see [133] above).

197      As Smith notes in A Modern Treatise on the Law of Criminal Complicity ([119] supra) at p 209, a standard formulation of the broad theory of the doctrine of common purpose is provided by Alderson B in the mid-19th century case of Macklin, Murphy, and Others’ Case (1838) 2 Lewin 225; 168 ER 1136 (“Macklin”) as follows (at 226; 1136):

[I]t is a principle of law, that if several persons act together in pursuance of a common intent, every act done in furtherance of such intent by each of them is, in law, done by all. The act, however, must be in pursuance of the common intent. [emphasis added]

This was followed in Art 38 of Sir James Stephen, A Digest of the Criminal Law (Macmillan & Co, 1877), which purported to state the common law at that time. Article 38 provided as follows:

Article 38. Common Purpose.

When several persons take part in the execution of a common criminal purpose, each is a principal in the second degree, in respect of every crime committed by any one of them in the execution of that purpose.

If any of the offenders commits a crime foreign to the common criminal purpose, the others are neither principals in the second degree, not accessories unless they actually instigate or assist in its commission.

198      Interestingly, according to Smith in A Modern Treatise on the Law of Criminal Complicity (at pp 209–210), resort to the terminology of common purpose occurs most frequently where the issue is the liability of an accessory for an offence additional or collateral to the primary principal offence. He writes that:

Most typically, during the course of a robbery an intervening third party is killed by one of the robbers. Here the primary offence or objective of the common purpose is robbery, the killing being a collateral or additional degree of criminality. When determining the liability for collateral offences, the central question has been whether, if at all, the basis of responsibility is different from that used in settling liability for the primary offence. More particularly, in relation to specificity of knowledge, does the mens rea necessary for complicity in the primary offence in any way act as a substitute for or dilute the full accessorial mens rea that would normally be required for the collateral offence?

This sounds very similar to the question which stands at the heart of an assessment of the Mimi Wong approach to the interpretation of s 34. Similarly, Peter Gillies in The Law of Criminal Complicity (The Law Book Company Limited, 1980) notes at p 90 that in practice the doctrine of common purpose has generally been employed to resolve issues of liability in a relatively specialised fact situation, ie, where A and B jointly agree to commit crime X in the course of which B commits a further crime. The issue then is that of A’s liability for this incidental crime which B committed.

199      Returning to the analysis of the doctrine of common purpose, Smith notes in A Modern Treatise on the Law of Criminal Complicity at p 210 that over the 300 years or so, running from early in the 16th century to the Victorian period, in line with the gradual general subjectivising of criminal responsibility, the doctrine of common purpose developed with increasing particularity the enquiry into how much the accessory knew of (or should have known of) how the principal might act. This is in line with what Gillies suggests in The Law of Criminal Complicity at pp 92–93, where he notes emphatically that one of the two basic features of the doctrine is that the accessory’s liability is to be evaluated on a subjective basis.

200      However, both authors agree that the requirement of subjective contemplation on the part of the accessory has not always prevailed. Prior to 1700, a number of English judges took the view that a person who had conspired with another for the commission of crime X could be inculpated as an accessory to the latter’s commission of crime Y, provided that Y was objectively incidental to their joint commission of X. This was an application of an objective probable consequences test, following which what the accessory knew did not matter; what mattered was that the ultimate crime which occurred was a probable consequence of the primary crime. For example, in the Trial of Lord Mohun (1692) Holt KB 479; 90 ER 1164, the House of Lords proposed seven questions to the judges, which were answered by them. Only two of these are relevant to our present purposes, and I reproduce them as follows (at 480–481; 1164–1165):

5.         Whether a person, knowing of the design of another to lie in wait to assault a third person, and accompanying him in that design, if it shall happen that the third person be killed at that time, in the presence of him who knew of that design, and accompanied the other in it, be guilty in law of the same crime with the party who had that design, and killed him, though he had no actual hand in his death?

Ans.     If a person is privy to a felonious design, or to a design of committing any personal violence, and accompanieth the party in putting that design in execution, though he may think it will not extend so far as death, but only beating, and hath no personal hand, or doth otherwise contribute to it than by his being with the other person, when he executeth his design of assaulting the party, if the party dieth, they are both guilty of murder. For by his accompanying him in the design, he shews his approbation of it, and gives the party more courage to put it in execution; which is an aiding, abetting, assisting and comforting of him, as laid in the indictment.

…

7.         If A. accompanieth B. in an unlawful action, in which C. is not concerned, and C. happeneth to come in the way of B. after the first action is wholly over, and happeneth to be killed by B. without the assistance of A. whether A. is guilty of that man’s murder?

Ans.     As this case is stated, A. is not guilty of murder.

Similarly, in Regina v Wallis (1703) 1 Salk 334; 91 ER 294, it was laid down by Holt CJ (at 335; 295) that if a man began a riot, as was the case there, and the same riot continued, and an officer was killed, he that began the riot was a principal murderer, although he did not do the act of killing. Finally, in a case involving an indictment for libel against the King, it was held by Lord Denman CJ in The Queen v Cooper (1846) 8 QB 533 at 536; 115 ER 976 at 977 that “[i]f a man request another generally to write a libel, he must be answerable for any libel written in pursuance of his request: he contributes to a misdemeanor and is therefore responsible as a principal”.

201      In the cases involving the typical robbery-murder scenario, Regina v Bowen (1841) Car & M 149; 174 ER 448 stands as a good example of an objective approach. In this case, the indictment charged the prisoner with having maliciously wounded one John Bailey, with intent to disable and disfigure him, and with intent to cause him some grievous bodily harm. It later emerged that the prisoner had intended to rob John Bailey with another person but there was no evidence as to which of the two actually struck the blows. Coleridge J, in summing up the law to the jury, said (at 151; 449):

It has also been contended, that there is no evidence to shew that the prisoner inflicted the wound; but I am of opinion, that, if the prisoner did not with his own hand inflict the wound, he may be convicted upon this indictment, if you are satisfied that the prisoner and the other person were engaged in a common purpose of robbing the prosecutor, and that the other person’s was the hand that inflicted the wound.

This was clearly an objective approach, for it mattered not what the prisoner intended apart from the common intention (or purpose) to rob. He was liable for causing grievous hurt to John Bailey simply by virtue of his involvement in the robbery, presumably because the beating up was a probable consequence of the robbery. However, eventually, the jury found that the prisoner had actual intent to cause some grievous bodily harm, although this does not change the substance of Coleridge J’s summing up.

202      However, other cases contemporaneous with these conform to the requirement of subjective knowledge or even a need for the secondary offender   to have the intention to commit the collateral offence committed. I agree with Smith in A Modern Treatise on the Law of Criminal Complicity at p 211 fn 6 that the exact nature of the subjective element is often far from clear.

203      The need for a subjective knowledge of the collateral offence being committed can arguably be found in some cases. In The King v Hodgson (1730) 1 Leach 6; 168 ER 105, the prisoners, together with several others, were hired by one JS to assist him in carrying away his household furniture, in order to avoid its being removed for rent. They accordingly assembled for this purpose, armed with bludgeons and other offensive weapons. The landlord of the house, accompanied on his part by another set of men, came to prevent the removal of the goods, and a violent affray ensued. The constable was called it but he could not disperse the mob. While they were fighting in the street, one of the prisoners, whose identity is unknown, killed a boy who was totally unconcerned in the affray. The question before the court was whether this was murder on the part of all.

204      The minority adopted a decidedly objective analysis (at 6; 106), holding that it was murder for all concerned, because they were involved in an unlawful act which ultimately led to the killing of a boy. However, the majority disagreed and held (at 6–7; 106) that:

[A]s the boy was found to be unconcerned in the affray, his having been killed by one of the company could not possibly affect the rest; for the homicide did not happen in prosecution of the illegal act …, and therefore the persons, though constructively present, could not be said to be aiding and abetting the death of one who was totally unconcerned in the design for which the parties had assembled.

What is clear from this exposition of the law is that the majority thought that the prisoners could not possibly have known that their affray would lead to the death of one unconcerned with their common purpose.

205      In Reg v Price (1858) 8 Cox CC 96, the prisoners chased a German sailor belonging to another ship and as he took refuge from their attack against a railing, he was stabbed by one of them with a knife and died. All six men who chased him were indicted for murder. Byles J, in summing up, said, inter alia (at 97), that if the jury did not think that there was a common intention and design to kill, the secondary offenders would nonetheless be guilty of murder if the knife was used in pursuance of the common design to use it, because the hand that used the knife was the hand of all of them. Presumably this meant that even if the prisoners had no actual intention to kill, they could be guilty of murder if they had commonly intended the knife to be used, and knew that it was likely that it could be used to inflict death on the deceased German sailor.

206      On the other hand, some cases insisted on there having been the intention to commit the very collateral offence committed. In Rex v John Leonard White and John Robinson (1806) Russ & Ry 99; 168 ER 704, it was laid down that if two persons were out for the purpose of committing a felony, and upon an alarm, ran in different ways with the result that one of them maimed a pursuer to avoid being taken, the other was not to be considered a principal in such an act. The judge directed the jury that if the two prisoners “came with the same illegal purpose, and both determined to resist, the act of one would fix guilt upon both” [emphasis added]. This seems to require the common intention to commit the very offence charged with. Such was the case in Rex v Collison (1831) 4 Car & P 565; 172 ER 827, in which Garrow B said (at 566; 828):

To make the prisoner a principal, the Jury must be satisfied, that, when he and his companion went out with a common illegal purpose of committing the felony of stealing apples, they also entertained the common guilty purpose of resisting to death, or with extreme violence, any persons who might endeavour to apprehend them; but if they had only the common purpose of stealing applies, and the violence of the prisoner’s companion was merely the result of the situation in which he found himself, and proceeded from the impulse of the moment, without any previous concert, the prisoner will be entitled to an acquittal. [emphasis added]

207      Thus, at the time the Indian Penal Code was drafted in the early 1800s, and later enacted in 1860, no tolerably clear authoritative principle emerged from the case law in England. The English courts were faced with choosing between the approaches of (a) objective foresight; (b) subjective knowledge; or (c) actual intention to hold the secondary offender liable in a “twin crime” situation, with each approach seemingly supported by authority. In the context of the Indian Penal Code, which I view to encompass the doctrine of common purpose, what had the drafters intended with respect to the mens rea required on the part of the secondary offender who had not committed the collateral offence?

208      One of the most valuable clues can be found in cl 98 of the draft Indian Penal Code prepared by Lord Macaulay. This provides as follows (see Macaulay ([127] supra) at p 13):

Wherever in an attempt to commit an offence, or in the commission of an offence, or in consequence of the commission of an offence, a different offence is committed, then whoever by instigation, conspiracy, or aid, was a previous abettor of the first mentioned offence shall be liable to the punishment of the last mentioned offence, if the last mentioned offence were such as the said abettor knew to be likely to be committed in the attempt to commit the first mentioned offence, or in the commission of the first mentioned offence, or in consequence of the commission of the first-mentioned offence; and if both offences be actually committed, and the person who has committed them be liable to cumulative punishment, the abettor shall also be liable to cumulative punishment.

Illustrations.

(a)        B, with arms, breaks into an inhabited house at midnight, for the purpose of robbery. A watches at the door. B being resisted by Z, one of the inmates, murders Z. Here, if A considered murder as likely to be committed by B in the attempt to rob the house, or in the robbing of the house, or in consequence of the robbing of the house, A is liable to the punishment provided for murder.

(b)        A instigates B to resist a distress. B, in consequence, resists that distress. In offering the resistance B voluntarily causes grievous hurt to the officer executing the distress. As B has committed both the offence of resisting legal process and the offence of voluntary causing grievous hurt, B is liable to cumulative punishment for these offences; and if A knew that B was likely voluntarily to cause grievous hurt in resisting the distress, A will also be liable to cumulative punishment.

[emphasis added]

209      I think that it was likely that, in its original form, the doctrine of common purpose (in a “twin crime” situation in which the collateral offence committed was not intended by all the parties) was intended to be incorporated in cl 98 of the draft Indian Penal Code. It is pertinent to note that there was no equivalent of s 34 in this draft – only a clause (cl 3) which is similar to the present-day s 35 and which does not contemplate the existence of any common intention. It was only in the final version of the Indian Penal Code that s 34 appeared, and the present-day s 113 seemingly replaced cl 98. In my view, although the doctrine of common purpose was originally intended to be encompassed by cl 98, subsequent legislators (who by this time did not include Lord Macaulay but included individuals such as Peacock CJ and Sir James Colvile, who were later involved in the stages leading up to the enactment of the Indian Penal Code in 1860 and its later amendment in 1870) thought that s 34 was to encompass the doctrine of common purpose, perhaps in addition to cl 98 and the abetment provisions. The addition of the words “in furtherance of the common intention of all” bolsters this view, for the striking similarity between the words and the doctrine of common purpose as enunciated by the English courts (see, especially, Macklin ([197] supra) meant that the doctrine was always intended to be covered by s 34, perhaps in addition to the abetment provisions.

210      By this view, the mens rea requirement of the secondary offender in a “twin crime” situation is certainly not the actual intention to commit the collateral offence, or, more accurately, the criminal act which constitutes the collateral offence. Clause 98 very clearly spells out the subjective knowledge approach and if that was intended to encompass the doctrine of common purpose in the first place, then its use of the subjective knowledge test could conceivably have been transferred to s 34 when it was enacted and when it was subsequently amended in 1870. I acknowledge that by 1870, ss 111 and 113 of the Indian Penal Code had also been enacted, with the former embodying an objective foresight test and the latter, a subjective knowledge test. However, in view of the very clear intention on the part of the drafter to impute only a subjective knowledge test to the Indian Penal Code for the doctrine of common purpose by way of cl 98, I feel that ss 111 and 113 can be reconciled if one were to interpret both sections subjectively as well. Indeed, also of interest is the English Report of the Royal Commission Appointed to Consider the Law Relating to Indictable Offences: With an Appendix containing a Draft Code embodying the Suggestions of the Commissioners (C 2345, 1879) leading up to the Criminal Code Bill of 1880 (UK). The final provision which appeared in the Criminal Code Bill read as follows:

72.        PARTIES TO OFFENCES

Every one is a party to and guilty of an offence who

(a)        Actually commits it;

(b)        Does or omits an act for the purpose of aiding any person to commit the offence;

(c)        Abets any person in the commission of the offence;

(d)        Counsels or procures any person to commit the offence.

If several persons form a common intention to prosecute any unlawful purpose, and to assist each other therein, each of them is a party to every offence committed by any one of them in the prosecution of such common purpose, the commission of which offence was or ought to have been known to be a probable consequence of the prosecution of such common purpose.

I will say more about this later. For now, I hold that the mens rea is instead subjective knowledge on the part of the secondary offender of the likelihood of the collateral offence happening. This shall be elaborated on this in due course.

211      This reading of s 34 is further supported by Peacock CJ’s rendition of the principle in Gorachand Gope (see [192] above) at 456, which I reproduce once again:

[W]hen several persons are in company together engaged in one common purpose, lawful or unlawful, and one of them, without the knowledge or consent of the others commits an offence, the others will not be involved in the guilt, unless the act done was in some manner in furtherance of the common intention. [emphasis added]

From this statement, which is said to have influenced the addition of the words “in furtherance of the common intention of all” in s 34, it is clear that the common intention of the secondary offenders need not be the same as the primary offender in a “twin crime” situation since Peacock CJ contemplated that the common purpose could be “legal or not legal”, whereas the act which was committed in furtherance of the common intention was an “offence”. If the common intention of all was “legal”, how then could they also commonly intend to commit an “offence” or “criminal act”? This must mean that the intention behind the primary act and the collateral act cannot be the same. Peacock CJ also spoke of “knowledge” and “consent”, which are in the language of subjective knowledge. Certainly, assent can be said to require the actual intention to commit the criminal act constituting the offence actually committed, but it goes without saying that if the secondary offenders had actually changed their common intention to that of the offence eventually committed, then the situation would in effect be a “single crime” as opposed to a “twin crime” situation. If they had intended to commit the collateral offence then obviously they would be liable. On the other hand, the use of the word “knowledge” retains the subjective knowledge requirement in s 34.

212      Accordingly, I hold that the Mimi Wong approach is thereby justified by the historical underpinnings of the Indian Penal Code and the doctrine of common purpose in English law. There is no need for the parties to possess a common intention to commit the offence actually committed in a “twin crime” situation. I note that there are problems with determining exactly what the additional mens rea is then to be, and although I have already hinted at the answer through the earlier analysis, I shall leave this to be dealt with later when I analyse the local cases following Mimi Wong. For now, I propose to deal with Prof Hor’s proposal that the Vincent Banka interpretation should remain, particularly his argument that the Penal Code would be conceptually sounder if this were so.

^{(D)       THE RELEVANCE OF SECTIONS 35 AND 38 OF THE PENAL CODE} 

213      As I have mentioned, although I have come to the conclusion that the historical underpinnings of s 34 incline towards the Mimi Wong approach, I also acknowledge the conceptual force of Prof Hor’s arguments. In particular, his reference to s 35 as undercutting the interpretation proffered in Mimi Wong cannot be brushed aside easily. I further observe that V B Raju has in Commentaries on Indian Penal Code (Eastern Book Company, 4th Ed, 1982) noted at vol 1 p 85 that the marginal note to s 35 states, “When such an act is criminal by reason of its being done with a criminal knowledge or intention” [emphasis added], and that the word “such” must refer to the “act” in s 34, thereby lending support to the Vincent Banka interpretation which requires the intention to be to commit the actual offence committed. In fact, even the Law Commission of India in their Forty-Second Report: Indian Penal Code (June 1971) (Chairman: Mr K V K Sundaram) pointed out at para 2.64 that s 35 “is complementary to the main rule in section 34, and the two have to be read together”.

214      Although the words in ss 34 and 35 read similarly, and the marginal note to s 35 seemingly refers to the criminal acts contemplated in s 34, the historical underpinnings of s 34 mean that the better reading of the two sections is that s 35 applies to a situation where there is no common intention. I think that somewhere along the way, s 34 gained dominance over s 35 in laying down the common law doctrine of common purpose in the Penal Code, and this has been implicitly acknowledged by the Indian courts (see [192] above) before s 34 was amended in 1870. I feel that the addition of the words “in furtherance of the common intention of all” showed the legislative intent to use s 34 to give expression to the doctrine of common purpose in the Penal Code.

215      In any event, the predecessor to s 35, viz, cl 3 of the draft Indian Penal Code, clearly contemplated that the provision only applied in the absence of a common intention. This gives strength to the analysis above.

^{(E)        ANALYSIS: ABETMENT PROVISIONS UNDER SECTIONS 111 AND 113 OF THE PENAL CODE}

216      I agree with Prof Hor that the abetment provisions seemingly provide for the same situations involving unintended consequences as those arising from the Mimi Wong approach to the interpretation of s 34. There is, of course, the concern that in so far as the abetment provisions are concerned, there is the need to identify the actual person who committed the collateral offence (ie, murder in a robbery-murder situation), whereas under s 34, there is no such need. This may therefore lead to the argument that s 34 has different requirements for application from ss 111 and 113. This may be so under the traditional common law analysis of abetment which requires a principal and an abettor – there must be at least one of each and each must be clearly identified. However, the Penal Code includes within its definition of abetment a conspiracy as well. Specifically, s 107 provides as follows:

Abetment of the doing of a thing.

107.     A person abets the doing of a thing who —

(a)        instigates any person to do that thing;

(b)        engages with one or more other person or persons in any conspiracy for the doing of that thing, if an act or illegal omission takes place in pursuance of that conspiracy, and in order to the doing of that thing; or

(c)        intentionally aids, by any act or illegal omission, the doing of that thing.

Plainly, s 107(b) contemplates an abetment wherein the principal and the abettor are not clearly defined. Transposed to ss 111 and 113, this is broadly the same kind of situation s 34 is concerned about, ie, where it is not possible to ascertain the actual perpetrator of the collateral offence.

217      Thus, I am of the view that there is no injustice in interpreting s 34 as requiring no less than ss 111 and 113, and that both provisions, in one way or another, give effect to the doctrine of common purpose. In fact, Prof Hor himself opines in “Common Intention and the Enterprise of Constructing Criminal Liability” ([148] supra) at 512 that “no harm is done if section 34 is consistently interpreted to require no less than [the] abetment provisions”. Further, Prof Hor himself points out in the same article at 511 that abetment at times does not involve common intention. This occurs when, for example, the abettor intentionally aids a principal who does not know that the abettor is doing so. In my view, this is precisely where the abetment provisions come into their own and supplement s 34 of the Penal Code. I shall have more to say about the abetment provisions after I have next analysed the precise mens rea required of the secondary offender in s 34.

218      Finally, even the abetment provisions themselves provide for the circumstances in which the consequence was intended, in other words, where the act abetted is committed in consequence. This is provided by s 109 of the Penal Code:

Punishment of abetment if the act abetted is committed in consequence, and where no express provision is made for its punishment.

109.     Whoever abets any offence shall, if the act abetted is committed in consequence of the abetment, and no express provision is made by this Code for the punishment of such abetment, be punished with the punishment provided for the offence.

Explanation.

An act or offence is said to be committed in consequence of abetment, when it is committed in consequence of the instigation, or in pursuance of the conspiracy, or with the aid which constitutes the abetment.

Illustrations

…

(c)        A and B conspire to poison Z. A, in pursuance of the conspiracy, procures the poison and delivers it to B, in order that he may administer it to Z. B, in pursuance of the conspiracy, administers the poison to Z, in A’s absence and thereby causes Z’s death. Here B is guilty of murder. A is guilty of abetting that offence by conspiracy, and is liable to the punishment for murder.

If Prof Hor is saying that s 34 should be reserved for the domain of intended consequences and the abetment provisions for unintended consequences, then s 109 flatly contradicts this suggestion. Section 109 evidently covers a situation contemplated by the Vincent Banka interpretation of s 34, viz, that there must be the intention to commit the very offence committed. Thus, if the abetment provisions themselves admit of a situation involving intended consequences, I do not think that it is correct to say that s 34 is reserved for such situations. Accordingly, I see nothing wrong with the Mimi Wong approach to the interpretation of s 34.

In furtherance of the common intention

219      I come now to the next major problem in relation to s 34 of the Penal Code. As I have concluded, the phrase “in furtherance of the common intention” enables criminal liability for collateral offences to be imputed to persons acting in concert even though that precise act may not have been jointly intended by the parties. This of course follows only if the common intention required of the parties in a “twin crime” situation is not of the actual offence committed; if so, then the preceding sentence would not be correct. However, the question that then arises is what is “in furtherance” of the common intention?

220      Courts have at times resorted to pointing out that the collateral offence must be “consistent” with the carrying out of the common intention of the parties. In Mimi Wong ([124] supra), this court held that the intention of the primary offender must be distinguished from the common intention of all. In this regard, the two intentions need not be the same, but they must be “consistent”. The context of this was as follows (at 80, [25]):

The intention that is an ingredient of the offence constituted by the criminal act is the intention of the actual doer and must be distinguished from the common intention of the doer and his confederates. It may be identical with the common intention or it may not. Where it is not identical with the common intention, it must nevertheless be consistent with the carrying out of the common intention …

However, as Prof Hor astutely notes in “Common Intention and the Enterprise of Constructing Criminal Liability” at 497, what is clear is that the Mimi Wong approach does not require the secondary offenders to possess any of the mens rea of the collateral offence. What is not clear is the precise requirement that has taken its place, and the word “consistent” fails to give any guidance on how the matter should be decided. What followed Mimi Wong was a series of decisions which seemed to adopt varying notions of “consistency” ranging from recklessness or negligence to strict liability, but without any real attempt at resolving the inherent ambiguity of the “consistency” formula given by this court in Mimi Wong.

221      Since I have earlier decided above that the intention of the secondary offender need not be the same as the common intention to give effect to the words “in furtherance of the common intention of all” [emphasis added], I now need to decide just when the criminal act physically perpetrated by the primary offender can be said to be “in furtherance of the common intention” of all the offenders where the intention of the primary offender in doing the criminal act does not coincide squarely with the common intention of the secondary offenders. Unfortunately, the definition of the word “furtherance” in The Oxford English Dictionary (Clarendon Press, 2nd Ed, 1989) vol VI at p 285, viz, “[t]he fact or state of being furthered or helped forward; the action of helping forward; advancement, aid, assistance”, does not lend much aid to the subject. I agree with Prof Hor and acknowledge that the decisions on this point have not conduced toward a model of clarity, although this is not to say that I believe that any injustice has been engendered along the way. Before laying down what I hope will be a clear pronouncement on what the requisite mens rea of the secondary offenders should be in such a context, I propose to first elucidate on the present state of the law.

(1)        The Singapore courts’ present requirement

222      Prof Hor in “Common Intention and the Enterprise of Constructing Criminal Liability” has helpfully listed out the various types of mens rea required of the secondary offender and I gratefully adopt these categories in our analysis of the present law. These types of mens rea said to be required of the secondary offender are as follows:

(a)        subjective knowledge of the likelihood of the collateral act being committed;

(b)        objective foreseeability of the likelihood of the collateral act being committed;

(c)        strict liability plus intention of the actual doer to further the common intention; and

(d)        strict liability per se.

223      Before I elaborate on each of these types of mens rea in turn, I acknowledge that there may be other variables involved in the mental element required. What is the particular harm needed to be satisfied by this mental element? What is the probability of it happening? For now, I am content to settle these additional variables as the likelihood of the collateral act being committed.

^{(A)        SUBJECTIVE KNOWLEDGE}

224      In Shaiful ([128] supra), this court reiterated the view that all that the Prosecution needed to prove was that there was in existence a common intention between all the persons involved to commit a criminal act and that the act which constituted the offence charged (the “criminal act” referred to in s 34 of the Penal Code) was committed in furtherance of that criminal act. However, it imposed a rider to this (at [57]):

The rider to this is that the participants must have some knowledge that an act may be committed which is consistent with or would be in furtherance of, the common intention. [emphasis added]

225      This court went on to elaborate, adopting the views of Ratanlal & Dhirajlal’s Law of Crimes (Bharat Law House, 24th Ed, 1997) vol 1 at p 122, that acts done “in furtherance of a common intention” fell into three categories: (a) acts which were directly intended by all the confederates; (b) acts which in the circumstances were undoubtedly to be taken as included in the common intention, although they were not directly intended by all the confederates; and (c) acts which were committed by any of the confederates to avoid or remove any obstruction or resistance put up against the proper execution of the common intention. The latest edition of Ratanlal & Dhirajlal’s Law of Crimes (Bharat Law House, 26th Ed, 2007) vol 1 at p 143 bears a similar passage.

226      I should clarify that this court in Shaiful had not intended that categories (b) and (c) should apply concurrently, since (b) is quite clearly based on subjective knowledge, whereas (c) is based on objective foresight. I think that it is best that only one of these bases apply to fix liability, and I shall elaborate on this in due course.

^{(B)        OBJECTIVE FORESEEABILITY}

227      On the other hand, the language of objective foreseeability was favoured in PP v Tan Lay Heong [1996] 2 SLR 150. The High Court in that case emphasised, in acquitting the alleged common intender, that the collateral crime was “not something that was either contemplated or done ordinarily in furtherance of a common intention” [emphasis added] to commit the primary crime (at 161, [46]).

228      Similarly, in PP v Too Yin Sheong [1998] SGHC 286 (“Too Yin Sheong”), the High Court seemingly advanced a theory of objective foreseeability. After rejecting the theory of subjective knowledge adopted by the English courts in relation to the common law dealing with a joint unlawful enterprise as being inapplicable to s 34 of the Penal Code (although without explaining why), the High Court said (at [119]) that:

If the acts of the main offender could reasonably be foreseen or contemplated by the secondary party to be in furtherance of their common intention and the acts were in fact committed by the main offender with the intention to further their shared common intention, then obviously the secondary party will similarly be convicted of the offence committed by the principal offender by virtue of s 34. [emphasis added in bold italics]

The court further emphasised later (at [121]):

Unless the foreseeability test is narrowed to “foreseeability of the act as a natural or likely incident of carrying out the common intention”, then the difference between s 34 and the law in England relating to joint enterprise will be less. But a major difference remains that in England, “forseeability” [sic] is taken from the subjective view point or knowledge of the secondary offender.Whereas the act referred to in s 34 must be an act done with the subjective intention of the actual doer, to further their shared common intention. [emphasis in original]

229      However, curiously, and as Prof Hor has pointed out in “Common Intention and the Enterprise of Constructing Criminal Liability” ([148] supra) at 499, the court then veered to a strict liability theory. The court said (at [132]):

It is immaterial therefore whether the secondary offenders, up to the time of the commission of the act by the principal offender, actually knew, expected, contemplated or even [foresaw] that act as a real possibility or not. It is not an essential ingredient in the application of s 34 that the prosecution must further prove that each of the individual accused persons, had known, expected, contemplated or foreseen the actual doer’s acts or that such an act, if carried out, would be in furtherance of their common intention. It is immaterial whether the secondary offenders had earlier told the actual doer that he ought not to do the act in question during the carrying out of their criminal venture or enterprise planned or agreed to by them. Neither is it immaterial that the secondary offenders never intended the doer’s act to take place. I reiterate that so long as the doer of the act had done the act in furtherance of the common intention of all of them, then the liability for that act automatically extends to the rest of the secondary offenders by virtue of s 34 of the Penal Code. It is irrelevant if the secondary offenders were not physically present at the time the doer was doing the act in question e.g. the look-out stationed outside the house or the robber searching downstairs not realising or knowing that the third robber had killed the victim upstairs when the victim suddenly resisted and wanted to escape. [emphasis added]

Thus, the court declared, in apparent antithesis to its earlier conclusion that a theory of reasonable foreseeability should apply in relation to the common intenders, that it was immaterial what the secondary offenders knew, expected, contemplated or foresaw. What mattered was the “subjective intention of the actual doer to further” the common intention, whatever this means in the context of the reasoning which had preceded it. In short, there is no additional mens rea requirement for the common intenders. All the mens rea they needed to have was the mens rea for the criminal act constituting the primary offence and the intention of the primary offender took over from there.

^{(C)        STRICT LIABILITY PLUS INTENTION OF ACTUAL DOER TO FURTHER COMMON INTENTION}

230      As I mentioned above, in agreement with the views of Prof Hor, the approach taken by the High Court in Too Yin Sheong appears to be best characterised as strict liability conditioned on the intention of the actual doer to further the common intention. It would perhaps be appropriate now to focus our attention on the theory of strict liability per se, which I now discuss.

^{(D)       STRICT LIABILITY PER SE}

231      In Asogan Ramesh ([160] supra), the three appellants and a mutual friend, Vijay, were walking home after a night of drinking when they spotted the victim walking towards them. The first appellant had several previous confrontations with the victim and the third appellant called out to the victim to talk to them. The victim responded by scolding them with vulgarities and then fled. The three appellants gave chase while Vijay was busy relieving himself nearby. When the second appellant caught up with the victim, he pulled out a knife. A fight ensued and the victim was eventually stabbed to death. The three appellants were convicted of committing murder in furtherance of a common intention under s 302, read with s 34 of the Penal Code. They appealed to this court against conviction on grounds that the judge erred in finding that there was a common intention between them and in rejecting their defence of sudden fight.

232      In dismissing the appeal, this court held (at [35]) that it was clear from the surrounding facts and circumstances that the three appellants had the common intention to assault the victim and it was the commonly intended assault which ultimately led to the victim’s death. This court continued (ibid):

The appellants intended to beat up the [victim] together, and it was in furtherance of carrying out this common intention that the [victim] was killed. In our view, the actions of the three appellants, in stabbing [victim] with the knife or by using a chair on him did not, contrary to counsel’s argument, … indicate to us actions consistent with the protection of themselves. In fact, in our opinion, these actions were the prolonged result of the unified intent of the three appellants to assault the [victim]. This subsequently led to the death of the [victim]. [emphasis added]

233      According to Prof Hor in “Common Intention and the Enterprise of Constructing Criminal Liability” at 498, and with whom I agree, the language used by the court in Asogan Ramesh is the language of causation and strict liability. It is irrelevant what was known, or what could have been known, so long as the commonly intended criminal activity led to or resulted in the collateral crime. However, the facts were such that it was unnecessary to use a strict liability theory to implicate the common intenders. The common intenders in the case had ganged up to beat and stab the deceased, which they did, and they would have clearly been guilty of murder even under a theory of subjective knowledge.

234      Yet another example of a case applying a strict liability requirement is Ng Beng Kiat v PP [1995] 3 SLR 335. In this case, the appellant was charged, together with two other persons, of murdering one Chia in furtherance of their common intention under s 300(c) of the Penal Code. The appellant and Chia were from rival gangs. On 5 January 1995, Chia was walking home alone when the appellant his two friends (the assailants) chased him, armed with wooden sticks, cornered him in a lift and attacked him till he collapsed. Subsequently, he died of head injuries inflicted by the assailants, which were sufficient in the ordinary course of nature to cause death. The appellant argued that he did not have the intention to kill Chia, but only wanted to teach him a lesson. The trial judge convicted the appellant who, because he was 17 years old at the time of the offence, was then detained at the President’s pleasure. On appeal, the appellant argued that: (a) the assailants did not have the intention to inflict the fatal injuries; and (b) alternatively, the intention to cause the fatal injuries was not shared by, and common to, all the assailants.

235      In dismissing the appeal, this court held that the common intention relied on by the Prosecution was the intention of the appellant and his two friends to cause Chia bodily harm. The appellant admitted that he wanted to beat Chia up to teach him a lesson. From the facts and circumstances of the case as stated above, it was clear to this court that this intention was shared by the two friends and it was in furtherance of the common intention to cause hurt that the fatal injuries were inflicted which led to Chia’s death. The secondary offenders simply needed to have intended to cause bodily harm to the victim; there was no additional mens rea required in relation to the eventual murder. This is, once again, based on a theory of strict liability.

(2)        Restatement of the requirement

236      In view of the varied approaches adopted by the local courts, I propose to lay down with this judgment a determinative pronouncement on the additional mens rea required of the secondary offender for him to be liable for the collateral offence which was eventually committed. As our analysis above in relation to the Mimi Wong approach shows, the additional mens rea required is that of a subjective knowledge on the part of the secondary offender in relation to the collateral offence likely happening. To be more precise, the secondary offender must subjectively know that one in his party may likely commit the criminal act constituting the collateral offence in furtherance of the common intention of carrying out the primary offence. In this regard, in connection with the expression “criminal act”, I do not think it is necessary for the actual method of execution (in murder) to have been known by the secondary offender. The expression “criminal act” is to be given a wide interpretation and I think that it is sufficient, in the case of murder, that the secondary offender knew that one in his party might inflict a bodily injury which was sufficient in the ordinary course of nature to cause death. It bears summarising our conclusions above at this point.

^{(A)        REQUIREMENT OF ENGLISH COMMON LAW}

237      First, I note that the requirement of the doctrine of common purpose at the time of the enactment of the Indian Penal Code imposed liability on a secondary offender for a collateral offence if such an offence was, inter alia, subjectively contemplated by the secondary offender. This is the standard adopted in the Penal Code because cl 98 of the draft Indian Penal Code, which I think imported the doctrine of common purpose into the Penal Code, clearly showed a subjective knowledge approach. Further, when s 34 was enacted and subsequently amended in 1870, the dictum of Peacock CJ in Gorachand Gope ([192] supra), which was said to have influenced the amendments, spoke of assent and knowledge (see [192] above). The usage of the term “knowledge” supports the use of the subjective knowledge analysis. This is also broadly in line with the prevailing English law at that time, as expressed by Sir James Stephen and in the Criminal Code Bill of 1880 (see [210] above), of which the latter, while seemingly embodying both a subjective and objective knowledge approach, should be read subject to the countervailing views of those involved in the framing of the Indian Penal Code, which may not have been identical.

^{(B)        CONFORMITY WITH THE ABETMENT PROVISIONS OF THE PENAL CODE}

238      Next, I further note that the requirement of subjective knowledge under s 34 would bring it into conformity with the abetment provisions where the offence which was actually committed was not that intended by the abettor. It bears setting out again the relevant provisions of the Penal Code, viz, ss 111 and 113:

Liability of abettor when one act is abetted and a different act is done.

111.     When an act is abetted and a different act is done, the abettor is liable for the act done, in the same manner, and to the same extent, as if he had directly abetted it:

Provided the act done was a probable consequence of the abetment, and was committed under the influence of the instigation, or with the aid or in pursuance of the conspiracy which constituted the abetment.

Liability of abettor for an offence caused by the act abetted different from that intended by the abettor.

113.     When an act is abetted with the intention on the part of the abettor of causing a particular effect, and an act for which the abettor is liable in consequence of the abetment causes a different effect from that intended by the abettor, the abettor is liable for the effect caused, in the same manner, and to the same extent, as if he had abetted the act with the intention of causing that effect, provided he knew that the act abetted was likely to cause that effect.

[emphasis added]

239      I acknowledge that there is Indian authority suggesting that the phrase “probable consequence” in s 111 of the Penal Code is to be given an objective reading: see, for example, Queen-Empress v Mathura Das (1884) ILR 6 All 491; Girja Prasad Singh v Emperor AIR 1935 All 346(2); and In re Irala Palle Ramiah AIR 1957 AP 231. In Singapore, however, there is some uncertainty as to the meaning of the phrase “probable consequence” as it is used in s 111.

240      In Mok Swee Kok v PP [1994] 3 SLR 140, this court was faced with an interpretation of s 111 of the Penal Code. In that case, three friends had hatched a plan with another friend by the name of Tang to enter Tang’s house using his key while no one was in, for the purpose of stealing from Tang’s parents. The maid who worked there returned home unexpectedly and was strangled to death by one of the parties. The appellant who had acted as a look-out was charged with abetting the robbery with hurt on the basis that the “hurt” was a probable consequence of the abetment to rob. This court declined to rule on whether an objective or subjective approach should be applied to the phrase “probable consequence” in s 111. It was content to hold that whichever approach was adopted, the strangulation of the maid was not a probable consequence of the appellant’s abetting of the original offence of house-breaking in order to commit theft since the plan was to go unarmed to the house when no one was at home to steal valuables. Similarly, the High Court in Ang Ser Kuang v PP [1998] 3 SLR 909 once again declined to rule conclusively on the approach to be taken (ie, objective or subjective) to be taken with respect to s 111 of the Penal Code.

241      The time to end the uncertainty has come. I agree with the learned authors of Criminal Law in Malaysia and Singapore ([144] supra) at para 34.44 that ss 111 and 113 of the Penal Code should be read similarly to embody a subjective knowledge approach. I think that the dichotomy between “act” (in s 111) and “effect” (in s 113) is much too close to justify the application of an objective test to one, and a subjective test to the other. Indeed, it bears repeating that in Lord Macaulay’s original cl 98, which is similar to the present-day s 113 of the Penal Code, only a subjective knowledge test was envisaged (see [208] above).

242      Thus, I hold that the words “probable consequence” in s 111 of the Penal Code should be taken to embody a subjective knowledge test similar to that which applies to s 34, viz, the abettor would only be liable for the unintended act if he subjectively knew of the likelihood of the act happening.

^{(C)        CONFORMITY WITH THE COMMON OBJECT PROVISION IN SECTION 149 OF THE PENAL CODE}

243      Also, a subjective knowledge approach would also bring consistency with our courts’ current interpretation of s 149 of the Penal Code. This section provides as follows:

Every member of an unlawful assembly to be deemed guilty of any offence committed in prosecution of common object.

149.     If an offence is committed by any member of an unlawful assembly in prosecution of the common object of that assembly, or such as the members of that assembly knew to be likely to be committed in prosecution of that object, every person who, at the time of the committing of that offence, is a member of the same assembly, is guilty of that offence.

244      In PP v Fazely bin Rahmat [2002] 4 SLR 655, the two accused were members of a gang who assaulted one “Sulaiman”, whom they mistook for a rival gang member. During the assault, knives were drawn and used to stab Sulaiman, who later died. Both accused were charged with being members of an unlawful assembly whose common object was to cause hurt with dangerous weapons. They did not deny that they were part of the gang that attacked Sulaiman, but denied stabbing him and only admitted to punching and kicking him. The two accused had stated in their written statements that they had kicked Sulaiman after he was stabbed. The Prosecution urged the court to draw the inference that this established the common object. However, the two accused testified that they assaulted Sulaiman before he was stabbed. The defence was that others in the gang stabbed Sulaiman, and that the accused did not know that knives were being carried by the others.

245      The High Court found the two accused not guilty of murder but guilty of rioting. It held that the Prosecution had established that the common object of the gang was to look for rival gang members which was understood to mean beating them up. There was no evidence that weapons of any kind were discussed or contemplated. The Prosecution relied on the first limb of s 149 that the murder was committed by one of the members of the unlawful assembly “in prosecution of the common object of that assembly”. The High Court interpreted the limb to require proof that the accused knew of the gang’s common object to cause hurt by dangerous weapons. This decision was upheld by this court in PP v Fazely bin Rahmat [2003] 2 SLR 184.

246      The imputation of a subjective knowledge test to s 34 would hence bring it into conformity with the approach already adopted by the courts in relation to the substantively similar s 149 limb of “in prosecution of the common object”. In my view, the expressions “common object” and “common intention” should be interpreted as meaning the same thing. Indeed, in our analysis of the old English cases above, the two expressions are used interchangeably with such other expressions as “common purpose”. These all mean the same thing. I should not, as a line of decisions appear to have done, attempt to draw fine distinctions where such distinctions do not exist. Accordingly, given the similarity between the two sections, a subjective knowledge approach to s 34 would bring it into conformity with s 149.

247      Ultimately, this would have the effect of harmonising ss 34, 111, 113 and 149 of the Penal Code, which undoubtedly overlap to some extent, as requiring subjective knowledge to affix liability for unintended consequences.

^{(D)       SUBJECTIVE KNOWLEDGE APPROACH CONSISTENT WITH OTHER JURISDICTIONS AND WITH UNIVERSAL PRINCIPLES}

248      Finally, I note that a subjective knowledge approach would bring our law in line with those of other jurisdictions in the Commonwealth. While it has often been said that the local criminal law should develop on its own to cater for our own societal wants, it is comforting also to see that the result which I have reached, after interpreting the relevant historical materials and case law, matches that of other jurisdictions. For example, in the House of Lords case of Regina v Powell (Anthony) [1999] 1 AC 1 at 27, Lord Hutton opined that it was sufficient to found a conviction for murder for a secondary party to have realised that in the course of the joint enterprise the primary party might kill with intent to do so or with intent to cause grievous bodily hurt. This test has been reaffirmed in England in Regina v Rahman [2007] 1 WLR 2191 and by the Privy Council on appeal from the Bahamas in Simmons v The Queen [2006] UKPC 19. The position in Australia is similar: see Sean Patrick McAuliffe v The Queen (1995) 183 CLR 108 (“McAuliffe”). McAuliffe has in fact been recently affirmed by the Australian High Court in Clayton v R (2006) 231 ALR 500 by a majority of six to one in a relatively short, unanimous judgment which held that an accused who foresaw that a member of the joint criminal enterprise might commit murder and continued to participate in the enterprise had the requisite mens rea to be held liable for murder. The court there also noted that the McAuliffe approach had not caused injustice and was in line with the general jurisprudence on complicity in the major jurisdictions.

249      Moreover, I note that the UK Law Commission in its recent report, Participating in Crime (Law Com No 305, Cm 7084, 2007), proposed a similar test in relation to joint criminal ventures. At para 3.151, the Law Commission made the following recommendation:

We recommend that, if P [the principal offender] and D [another person] are parties to a joint criminal venture, D satisfies the fault required in relation to the conduct element of the principal offence committed by P if:

(1)        D intended that P (or another party to the venture) should commit the conduct element;

(2)        D believed that P (or another party to the venture) would commit the conduct element; or

(3)        D believed that P (or another party to the venture) might commit the conduct element.

250      Ultimately, I believe that the convergence of my conclusion about the approaches of other jurisdictions is no mere coincidence. It bears testimony to the universal truth and principles that stand independent from the static hard law which is etched onto our statutory material. I agree with Prof A P Simester’s compelling analysis in “The Mental Element in Complicity” (2006) 122 LQR 578 that, by forming a joint enterprise, the accessory perpetrates an independent and discrete wrong, and the collusion justifies the extension of liability of the principal’s crime on the accessory for what is said to be a reduced mens rea of subjective knowledge. Indeed, the learned professor’s views as summed up at 599 of the article should be cited for its clarity and logic in its analysis:

Joint enterprise doctrines come into their own, however, in respect of offences incidental to the common purpose. By forming a joint enterprise, S [the secondary offender] signs up to its goal. In so doing, she accepts responsibility for the wrongs perpetrated in realising that goal, even though they be done by someone else. Her joining with P [the primary offender] in a common purpose means that she is no longer fully in command of how the purpose is achieved. Given that P is an autonomous agent, S cannot control the precise manner in which P acts. Yet her commitment to the common purpose implies an acceptance of the choices and actions that are taken by P in the course of realising that purpose. Her responsibility for incidental offences is not unlimited: S cannot be said to accept the risk of wrongs by P that she does not foresee, or which depart radically from their shared enterprise, and joint enterprise liability rightly does not extend to such cases. Within these limitations, however, the execution of the common purpose – including its foreseen attendant risks – is a package deal. Just as risks attend the pursuit of the common purpose, an assumption of those risks flows from S’s subscription to that purpose. [emphasis added]

All of these supporting authorities lend credence to our interpretation of s 34 of the Penal Code as applied in Singapore.

The relationship of section 34 of the Penal Code with the other provisions

251      I had earlier discussed the relationship between s 34 and the other relevant provisions of the Penal Code. In my view, our present interpretation of s 34, following the Mimi Wong approach, brings it in line with the abetment provisions and s 149. Although s 149 creates a distinct offence, it is conceptually similar to s 34. Sections 111 and 113 are, in my view, another way of expressing the doctrine of common purpose embodied in s 34. There will be instances where a situation captured by ss 111 and 113 do not fall within s 34, and hence ss 111 and 113 retain a practical significance notwithstanding the similarity in interpretation. On this note, the Prosecution may wish to consider analysing s 34 cases under ss 111 and 113 in the future as well.

252      There is one final point and this concerns s 396 of the Penal Code, which provides as follows:

Gang-robbery with murder.

396.     If any one of 5 or more persons who are conjointly committing gang-robbery, commits murder in so committing gang-robbery, every one of those persons shall be punished with death or imprisonment for life, and if he is not sentenced to death, shall also be punished with caning with not less than 12 strokes.

It can be seen that in this case, there is no need for a subjective knowledge on the part of the common intenders to the robbery to be liable for murder. In my view, this can be explained on the basis that the Legislature wanted to create a more serious offence for gang robberies with murder. Indeed, Lord Macaulay in his notes to the draft Indian Penal Code ([127] supra) noted at p 121 that he and his commissioners had provided punishment of “extreme severity for that atrocious crime” by the name of “dacoity”, which was what gang robbery was called (and is still known as) in India. In any event, there is discretion on punishment (ie, death or life imprisonment) to mitigate against the harshness of the rule.

Summary of the correct interpretation of section 34 of the Penal Code 

253      Thus, I can now summarise what I regard as the correct interpretation of s 34 of the Penal Code, taking into account the typical requirements to make s 34 of the Penal Code applicable, viz, (a) a criminal act; (b) participation in the doing of the act; (c) a common intention between the parties; and (d) an act done in furtherance of the common intention of the parties:

(a)        Criminal act: Section 34 does not refer to the actual crime committed only. It is essential to realise that the expression “criminal act” is not synonymous with “offence” as defined in s 40 of the Penal Code.

(b)        Participation: Presence at the scene of the criminal act, primary or collateral, need no longer be rigidly insisted on for s 34 of the Penal Code to apply. In a “twin crime” situation, there is no need for participation in the collateral criminal act as well as the primary criminal act; participation in the primary criminal act would be sufficient for liability to fix on all subsequent secondary offenders. The crux of the section is participation, and presence may or may not provide the evidence for participation; this is a question of fact to be decided in each case.

(c)        Proving the common intention: To prove the common intention between the parties, inferences must be made from the circumstances of the case to show that the criminal act was committed in furtherance of a pre-arranged plan such as the conduct of the parties, the weapons used and the nature of the wounds inflicted. However, such inferences should never be made unless it was a necessary inference deducible from the circumstances of the case. All the circumstances, including antecedent and subsequent conduct, are relevant in inferring the common intention of all involved.

(d)        In furtherance of the common intention: There is no need for the common intention of the parties to be to commit the offence actually committed in a “twin crime” situation, otherwise the words “in furtherance” would be superfluous. The Mimi Wong approach to the interpretation of s 34 of the Penal Code is justified by the historical underpinnings of the Indian Penal Code and the doctrine of common purpose in English law. The additional mens rea required of the secondary offenders is that of a subjective knowledge on the part of the secondary offender in relation to the likelihood of the collateral offence happening. To be more precise, the secondary offender must subjectively know that one in his party may likely commit the criminal act constituting the collateral offence in furtherance of the common intention of carrying out the primary offence. There is no need to have known of the actual method of execution in a murder situation.

It is hoped this will go some way towards resolving the confusion surrounding s 34 of the Penal Code.

Whether the appellant was rightly convicted of the Charge

254      I now come to consider whether the appellant was rightly convicted of the Charge. As I have alluded to above, Too’s statements should not have been considered by the trial judge. The question then is whether, with the remaining evidence and with the interpretation of s 34 of the Penal Code which I have elucidated in this judgment, the appellant was nonetheless rightly convicted of the charge. To be rightly convicted under s 34, the evidence would need to show that the appellant subjectively knew that the deceased may likely be killed in furtherance of the robbery at some point of time before the actual act of killing. However, a few preliminary remarks should be made at this point about the evidence required to establish subjective knowledge. Very often, it will not be the case that the accused states that he had a particular state of knowledge. The existence of a state of knowledge is therefore to be carefully inferred from the surrounding evidence. This is not to say that the courts should “objectivise” subjective knowledge with what they think the accused ought to have known; what this simply requires is for a careful evaluation of the evidence to disclose what the accused actually knew but had not stated explicitly. Indeed, this is the entire nature of circumstantial evidence. It would be useful to recount the remarks made by the High Court in PP v Chee Cheong Hin Constance [2006] 2 SLR 24 at [77]:

Often perpetrators take pains to conceal their crime. Direct evidence of the precise circumstances preceding a homicide will usually be unavailable. In such instances, compelling circumstantial evidence may be relied on to infer guilt. The Prosecution’s case against the accused, not surprisingly, is premised entirely on circumstantial evidence. It is trite law that the circumstantial evidence on which the prosecution relies must in the final analysis “inevitably and inexorably” lead the court to a single conclusion of the accused’s guilt: see Ang Sunny v PP [1965-1968] SLR 67 at 72, [14]. In PP v Oh Laye Koh [1994] 2 SLR 385, the Court of Appeal emphasised that the Prosecution did not carry a higher burden in the final evaluation of a case predicated upon circumstantial evidence as opposed to one based on direct evidence. The court declared at 392, [19]:

There is one and only one principle at the close of the trial, that of guilt beyond reasonable doubt, and this principle applies equally to cases where the prosecution evidence is wholly circumstantial as it does to those where direct evidence is adduced.

Having clarified the nature of the evidence which I will be considering, I propose to deal with this in a systematic manner and, it bears repeating, without considering Too’s statements.

Link between the appellant and the events surrounding the deceased’s death

255      It is firstly imperative to establish a link between the appellant and the murder of the deceased. In my view, and as the trial judge had found, the events which occurred after the robbery created an indelible link between the appellant and the tragic events which had occurred at the deceased’s house. As the trial judge rightfully noted, apart from the appellant’s own admission that he had shared in the spoils of the robbery, the independent evidence adduced by the Prosecution also identified the appellant as having been party to the subsequent usage of the deceased’s COL card. In this regard, I accept the trial judge’s assessment of Ms Lim who gave evidence to this effect, notwithstanding some inconsistencies in her testimony (see [13]–[16] above).

256      Indeed, the appellant himself conceded that he had, when the deceased initially protested downstairs, attempted to stab the deceased as well as punched him in order to quell his acts of resistance. According to the appellant, at the time when he used the knife to threaten the deceased, he had been prepared to use the knife on the deceased if the latter had further struggled or retaliated. This thus established that the appellant was in some way connected with the events surrounding the violence and injuries inflicted on the deceased. However, this is not enough to affix liability on the appellant on the charge. Next to be considered is whether the appellant had the requisite mens rea, ie, that he subjectively knew that the deceased may likely be killed in furtherance of the robbery.

257      On this note, I must point out that the deceased was strangled to death; he did not die from knife wounds caused by the stabbing. However, I repeat my observations above (at [236]) that it is not necessary for the actual method of execution to have been subjectively known by the appellant; all that is required is for the likelihood of the criminal act, ie, the infliction of a bodily injury sufficient in the ordinary course of nature to cause death, to be known to the appellant.

Whether the appellant knew that the deceased may likely be killed

Events before the robbery 

258      An important fact was what had happened before the robbery. In this connection, on the appellant’s own admission, he was, prior to the robbery, afraid of being recognised by the deceased. After all, he was with Too and Too was well known to the deceased. Identification of Too would inevitably lead to the identification of the appellant. These fears were sufficiently compelling to cause the appellant to quarrel with Too during the journey to the deceased’s house (see the record of proceedings at p 596). Whilst the appellant contended that his fears were quelled by Too’s confidence, I agree with the trial judge that this assertion was little more than an artificial construct manufactured by the accused to remove his otherwise undeniable motive for killing the deceased. As the trial judge noted, on the appellant’s own admission, he was not particularly well acquainted with Too or Ng. His seeming trust in Too, and his expectation that Too would not implicate him if arrested, were wholly inconsistent with the admittedly superficial nature of their acquaintance. These facts lead to the inescapable inference that the appellant must have been privy to an intention to get rid of the deceased, if not by himself, then by way of Too.

Events after the robbery 

259      This inference is strengthened by the appellant’s conduct following the robbery. The compelling inference is that he knew about the fate of the deceased and had been party to the fatal injuries inflicted on the deceased. The complete lack of any discussion regarding what would happen when the deceased was freed (see, for example, the record of proceedings at p 653), coupled with the calm and calculated, indeed nonchalant, manner in which they went about repeatedly exploiting the deceased’s COL card on the day following the robbery (see, for example, the record of proceedings at pp 658–659) and the appellant’s continued travel between Singapore and Malaysia in the years following the robbery (see, for example, the record of proceedings at pp 573 and 582), could only support the conclusion that the appellant and his accomplices were not worried about being identified because they knew that the deceased was dead.

260      Without the knowledge that the deceased had died, the likelihood of their identification and arrest would have appeared particularly imminent since the deceased knew Too personally and would, hence, have been able to positively identify him. Given these circumstances, it was wholly incredible that the appellant would have had no regard whatsoever to the possibility – or indeed, the certainty – of the deceased making a report against Too when he was freed. The appellant’s attempt to portray himself as a passive follower of Too was directly rebutted by his own admission in cross-examination that he had argued with Too because of his fear of being recognised. This, coupled with the appellant’s proactive involvement in clarifying the modus operandi of the robbery prior to its occurrence, could only lead one to ineluctably conclude that the appellant was not a mere robber but someone who was at least privy to a plan to conceal his involvement in the robbery.

261      Our conclusion that the appellant had the requisite mens rea is finally fortified by the appellant’s own admission as regards the events during the robbery itself.

Events during the robbery

262      It bears repeating that the appellant himself conceded that he had, when the deceased initially protested downstairs, stabbed the deceased in order to quell his acts of resistance. According to the appellant, at the time when he used the knife to threaten the deceased, he had been prepared to use the knife on the deceased if the latter had struggled or retaliated. This admission by the appellant himself amounted to positive evidence that he knew that he or Too would have inflicted serious harm on the deceased if he had struggled or retaliated. Coupled with the events before and after the robbery as well as the gratuitous and callous violence inflicted on the deceased, I feel compelled to conclude that the appellant knew further, in that not only did he know that Too (or himself) would have seriously harmed the deceased if the deceased had struggled or retaliated, but he must also have appreciated that the deceased would have to be killed to protect their identities in the light of the harm they had inflicted on him.

263      The primary root of this conclusion begins with the appellant’s fear of being recognised at the outset of the robbery. Indeed, given that the appellant was extremely perturbed about being identified by the deceased and his later apparent calmness, I think that he must have known there that the deceased was likely to be killed during the course of the robbery. This conclusion is bolstered by the fact that, during the course of the robbery, the appellant in fact subjected the deceased to some degree of violence (contributing to some of the 18 external injuries (see [8] above)), even to the extent of stabbing the deceased with a knife. In fact, the appellant himself testified that he had stabbed the deceased because the deceased did not keep quiet as instructed and the appellant had done this to prevent him from raising an alarm which might be heard by the neighbours (see the record of proceedings at pp 670–671). On top of this, the deceased was also bound up against his will in the course of the robbery to prevent him from resisting and/or escaping. These cumulative facts showed how complicit the appellant was in the violence, the efforts to conceal their involvement in the course of the robbery and the inescapable conclusion that it must have occurred to both the perpetrators that they would need to protect their identities after they made their getaway. The appellant’s assertions, in the course of his examination-in-chief, that he was not afraid of being recognised as the deceased only knew Too and if it was anyone who the deceased should have recognised it ought to have been only Too, is implausible to say the least (see the record of proceedings at pp 567–568). In any event, I do not think that it escaped the appellant that, as Too was known to the deceased, he too would have been similarly implicated notwithstanding the fact that he himself was not personally known to the deceased. It would also not be a stretch of logic to conclude that the deceased would also have implicated the appellant by way of a description of the appellant’s appearance to the police, had he escaped from the ordeal alive.

264      These aforementioned acts of remorseless violence against the deceased, coupled with (a) the appellant’s deep-seated and enduring concern about being identified, and (b) the deceased’s familiarity with Too, means that it requires a complete suspension of disbelief to accept the appellant’s persistent disavowal of his knowledge of the circumstances preceding the actual murder. The brazen and callous manner in which Too and the appellant conducted themselves during and after the killing speaks for itself. Indeed, the appellant himself admitted in the course of cross-examination that he was prepared to use the knife “on the deceased” if he struggled or retaliated (see the record of proceedings at p 668).

265      Thus, even if the appellant himself did not have the prior intention to kill the deceased, it is sufficient, under s 34, that he must have himself known that Too may likely have done the same (ie, to kill), and since that happened in furtherance of the robbery, the appellant is therefore liable. Indeed, having in fact subjected the deceased to some degree of violence during the course of the robbery (resulting, it bears emphasising, in 18 external injuries), it would be abundantly clear that the appellant’s own fear of being recognised by the deceased, should the latter somehow escape from the scene of the robbery, would be even more elevated. This reinforces the conclusion that the appellant knew that either Too or he himself would have killed the deceased in furtherance of the robbery. In fact, it could even be said that the appellant knew, as the robbery progressed, that either Too or he himself would have to kill the deceased in furtherance of their common intention to conceal their involvement in the commission of the robbery or (by the appellant’s own admission) to use the knife to inflict harm on the deceased (see the record of proceedings at p 625). Indeed, on this analysis, having formed the common intention to use the knife to threaten the deceased, the appellant must surely have known that either Too or he himself may likely kill the deceased.

266      Ultimately, it bears mention that the trial judge disbelieved the appellant’s version of events that desperately sought to minimise his role in the actual killing. It is plain that the appellant’s testimony and statements were irreparably riddled with inconsistencies which inexorably pointed to his actual knowledge of how and why the deceased was killed. In the final analysis, after very careful consideration, I therefore also accept the trial judge’s conclusion that, in the light of the findings made above, the evidence in fact went much further and established the appellant’s involvement in the infliction of the injury which subsequently led to the deceased’s death. For these reasons, I have no hesitation in affirming the trial judge’s conviction of the appellant in respect of the Charge, ie, that of murder punishable under s 302 read with s 34 of the Penal Code.

Section 111 or section 113 of the Penal Code

269      Finally, it must be noted that even if the appellant is not guilty under s 302 read with s 34 of the Penal Code, the evidence adduced clearly makes him guilty under s 111 or s 113 of the Penal Code. Even though the charge preferred against the appellant was not one of abetment, it would not have precluded this court from amending the charge against the appellant if a more restrictive reading of s 34 were the result of our review of underlying legal principles. Indeed, a reference to illus (c) of s 111 would make this point clear:

A instigates B and C to break into an inhabited house at midnight for the purpose of robbery, and provides them with arms for that purpose. B and C break into the house, and being resisted by Z, one of the inmates, murder Z. Here, if that murder was the probable consequence of the abetment, A is liable to the punishment provided for murder.

270      Therefore, even if s 34 of the Penal Code required the appellant to have the actual intention to murder the deceased, ss 111 and 113 (as I have interpreted above) would apply to require the appellant to only have subjective knowledge that the act of murder would likely take place. Revisiting the facts once again in the present case, it bears mention that the deceased was not only subjected to acts of violence (which included stabbing) by the appellant, but he was also tied up later on. Transposing these facts and the appellant’s fear of being recognised for his role in the robbery onto illus (c) of s 111, specifically the provision of arms in the said illustration, I come unhesitatingly to the conclusion that the appellant subjectively knew, as evidenced by the facts mentioned, that the deceased may likely have been killed in furtherance of the robbery. The analysis therefore is no different from that under my interpretation of s 34 of the Penal Code, with the identical result that the appellant remains equally liable for the murder of the deceased.

Conclusion

271      For the reasons above, I would dismiss the appellant’s appeal even though I am of the view that Too’s statements should not have been admitted by the trial judge.

V K Rajah
Judge of Appeal

Choo Han Teck J (delivering the dissenting judgment):

272      I concur with the grounds and reasoning of V K Rajah JA in respect of the legal issues concerning s 378(1)(b) of the Criminal Procedure Code (Cap 68, 1985 Rev Ed) and s 34 of the Penal Code (Cap 224, 1985 Rev Ed). The appellant was convicted by the trial judge on an entirely different interpretation of those two provisions and the findings of fact had thus been made under a different understanding of the legal requirements. In the circumstances, I disagree as to the final order dismissing the appeal and am of the view that the appellant ought to be retried.

273      First, the wrongful admission of Too’s statements alone would warrant a retrial because the danger of the findings of fact being tainted by those untested and prejudicial statements are too great to be allowed to stand.

274      Second, in holding that the correct interpretation of s 34 of the Penal Code requires a finding of fact, that the appellant had the subjective knowledge at the time of the commission of the primary offence of robbery that culpable homicide was likely to be committed, the court requires a stricter and more specific finding of fact than was required under the previous position in law. The trial judge would be required to address his mind directly on the law as we have so interpreted in the judgment of Rajah JA.

275      This court, as an appellate court, having to rely on the facts found by the court of first instance, should not, in my view, endorse facts found by the court below on a different interpretation of the law. The present interpretation of s 34 of the Penal Code no longer condemns an accused for other crimes committed by his accomplice (or accomplices) merely by his agreement to commit the primary crime. We have found it fairer that the accused be guilty of secondary offences committed in furtherance of a common intention to commit the primary offence only if he shared in the common intention to commit those secondary offences, or that he knew that those offences were likely to be committed. The trial judge must address his mind to this issue and make his decision on the whole of the evidence – the faces of the witnesses; their voices; the statements in print; and, finally, the piecing together of all that he has heard and observed in the course of the trial. The appellate court is not in a position to undertake all these functions.

Choo Han Teck
Judge

Woo Bih Li J:

Introduction

276      The circumstances leading to this appeal by the accused person, Lee Chez Kee (“Lee”), have been set out in the judgment of V K Rajah JA which I have read. I agree with Justice Rajah’s views on s 34 of the Penal Code (Cap 224, 1985 Rev Ed) and the decision to dismiss Lee’s appeal.

277      However, as regards the question whether the statements of an accomplice by the name of Too Yin Sheong (“Too”) should have been admitted under s 378(1)(b)(i) Criminal Procedure Code (Cap 68, 1985 Rev Ed) (“CPC”), I have a different view from Justice Rajah. Justice Choo Han Teck agrees with Justice Rajah on s 378(1)(b)(i) CPC and s 34 of the Penal Code, although not on the final outcome of the appeal. I am of the view that the trial judge was correct in admitting Too’s statements but, unlike the trial judge, I would not have given any weight to Too’s statements in so far as those statements sought to blame Lee for the actual act of strangulation resulting in the murder of the victim because they are self-serving and not tested by cross-examination. However, this does not necessarily mean that in every case such statements will not be given any weight, as I shall elaborate later. I would add that the trial judge had specifically said that he had borne in mind the potential danger of placing too much weight on Too’s statements and he was of the view that such statements “served the limited function of reinforcing the already compelling inference” of the independent evidence (see PP v Lee Chez Kee [2007] 1 SLR 1142 at [68]).

278      In any event, as there was sufficient evidence to support a verdict of guilty against Lee for the charge of murder as set out in Justice Rajah’s judgment, without having to rely on Too’s statements, I am of the view that Lee’s appeal should be dismissed.

279      The scope of s 378 CPC and, more specifically, s 378(1)(b)(i) CPC, is of wider significance. For easy reference, I set out s 378(1)(b)(i) CPC below:

378.—(1) In any criminal proceedings a statement made whether orally or in a document or otherwise, by any person shall, subject to this section and section 379 and to the rules of law governing the admissibility of confessions, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(b)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness;

…

280      I should also mention that in Gurbak Singh v PP [1998] 2 SLR 378 two statements of a friend of an accused person were admitted under s 378(1)(b)(i) CPC. In Tee Chu Feng v PP [2005] SGHC 181, two statements of the purchaser of Ecstasy drugs were admitted also under s 378(1)(b)(i) against the vendor. However, there was no argument in either case as to whether a confession of an accomplice should be excluded in view of s 30 of the Evidence Act (Cap 97, 1997 Rev Ed) (“EA”) which is the very question we have to consider.

281      As regards the phrase “subject … to the rules of law governing the admissibility of confessions” (“the qualifying phrase”) in s 378(1)(b)(i), I accept that such rules would include s 30 EA which states:

When more persons than one are being tried jointly for the same offence, and a confession made by one of such persons affecting himself and some other of such persons is proved, the court may take into consideration the confession as against the other person as well as against the person who makes the confession.

282      I also accept that based on a literal interpretation of s 378(1), the admissibility of a statement thereunder would be subject to s 30 EA. This would in turn suggest that even though Too is dead, his statements, which amount to confessions, cannot be admitted in the trial against Lee because of s 30 EA, as Too is not jointly tried with Lee. The question then is whether that was the intention of Parliament. The majority of this court have concluded that that is the position.

283      The trial judge pointed out that s 378(1)(b)(i) CPC first appeared as s 371C(1)(b)(i) within cl 23 of the Criminal Procedure Code (Amendment) Bill (Bill 35 of 1975) (“the CPC Bill”). This provision was in turn adopted from cl 31(1)(c)(i) of the Criminal Evidence Bill (“the UK Bill”) proposed by the UK Criminal Law Revision Committee (“the UK Committee”) (see the Comparative Table to the CPC Bill and also the Criminal Law Revision Committee, Eleventh Report: Evidence (General) (Cmnd 4991, 1972) (Chairman: Edmund Davies LJ) (“the Eleventh Report”).

284      I set out below s 31(1)(c) of the UK Bill as well as s 31(2) of the UK Bill as the latter is similar, although not identical in terms, with s 30 EA:

31.—(1) In any proceedings a statement made, whether orally or in a document or otherwise, by any person shall, subject to this and the next following section and to section 2 of this Act, be admissible as evidence of any fact stated therein of which direct oral evidence by him would be admissible, if —

…

(c)        it is shown with respect to him —

(i)         that he is dead, or is unfit by reason of his bodily or mental condition to attend as a witness; …

(2)        Subject to section 2 of this Act, in any proceedings in which two or more persons are jointly charged, a statement made, whether orally or in a document or otherwise, by any of the accused may be given in evidence by the prosecution as evidence of any fact stated therein of which direct oral evidence by the maker would be admissible, notwithstanding that the maker has not been and is not to be called as a witness in the proceedings; and any statement given in evidence by virtue of this subsection shall be admissible as evidence of any such fact in relation to each of the accused.

285      The trial judge matched the words in s 378(1) CPC with those in cl 31(1) of the UK Bill. So, “subject to this section and section 379” in s 378(1) CPC was matched with “subject to this and the next following section” in cl 31(1) of the UK Bill. It is not disputed that s 379 CPC is the equivalent of “the next following section” in the UK Bill.

286      As for cl 31(1) of the UK Bill being subject to cl 2 of the UK Bill, the latter clause contained provisions to ensure that the confessions were given voluntarily before they were admitted in evidence. The trial judge was of the view that the qualifying phrase had adopted this concept only, ie, of the concept of voluntariness, from the UK. The CPC Bill did not set out a provision similar to cl 2 of the UK Bill because Singapore already had similar provisions to ensure the voluntariness of confessions, although such provisions are not identical in terms to cl 2 of the UK Bill. In Singapore, such provisions are s 122(5) CPC and s 24 EA. It is not necessary for me to set out these provisions or cl 2 of the UK Bill.

287      I am of the view that there was some force in the trial judge’s matching of the words in s 378(1) CPC with cl 31(1) of the UK Bill.

288      It also seems clear to me, that, generally speaking, the scheme under s 378(1) CPC followed that under the UK Bill. In the speech of the Minister for Law and the Environment, Mr E W Barker, on the CPC Bill on 19 August 1975, Mr Barker said (see Singapore Parliamentary Debates, Official Report (19 August 1975) vol 34 at cols 1223–1224):

The scheme that the Bill proposes has the following purposes:-

(1)        to admit all hearsay evidence likely to be valuable to the greatest extent possible without undue complication or delay to the proceedings;

(2)        to ensure that evidence should continue to be given for the most part orally by allowing hearsay evidence only if the maker of the statement cannot be called or it is desirable to supplement his oral evidence; and

(3)        to include necessary safeguards against the danger of manufactured hearsay evidence.

The proposals in this Bill relating to hearsay evidence adopt many of the recommendations contained in the Criminal Law Revision Committee’s Eleventh Report. Certain modifications have been made to these proposals after considering the views expressed by the General Council of the Bar of England and Wales and the Council of the Law Society of England in their memoranda.

The opportunity has been taken to amend other sections of the Code to correct existing mistakes and anomalies as well as to expedite the administration of justice as when an accused wishes to plead guilty in the High Court.

It is proposed to refer the Bill to Select Committee where all representations will be given careful consideration.

These three purposes were also three of the main purposes of the UK Bill.

289      What then was the position under the UK Bill as regards the interplay between cl 31(2) (which is similar to s 30 EA) and cl 31(1) (which is similar to s 378(1) CPC)? I note that when the UK Committee was discussing the admissibility of confessions of an accomplice under cl 31(2) of the UK Bill, the Eleventh Report stated at p 146 thereof that such a confession would be admissible “[i]n any event” under cl 31(1) if the conditions therein and in cl 2 were met. In my view, this demonstrated that although cl 31(1) was stated to be subject to “this … section”, which included cl 31(2), cl 31(2) was not intended to override cl 31(1), as the literal interpretation would otherwise have suggested. This is another argument against the literal interpretation of the qualifying phrase.

290      Another reason which supports the trial judge’s view is that if an accomplice’s confession is not admissible under s 378(1)(b)(i) CPC even though he is dead, then no confession would be admissible at all. Let me elaborate. By definition, a confession incriminates the maker. Therefore, the qualifying phrase must refer to the confession of an accomplice who is dead but not the confession of the accused person himself, because if the accused person was dead, then there would no longer be a trial of that accused person. The confession of an accomplice may incriminate or exonerate the accused person of the secondary offence (as defined in Justice Rajah’s judgment). If it exonerates the accused person, the confession is apparently in any event admissible under the exceptions to the hearsay rule and s 378(1)(b)(i) CPC is not required. If it incriminates the accused person, it will never be admissible under s 378(1)(b)(i) CPC, if the literal interpretation prevails, because there is no joint trial. The question of voluntariness of the confession is therefore academic. Effectively, the literal interpretation would mean that no confession of an accomplice would ever be admitted under it. Yet, instead of suggesting that no confession of an accomplice is admissible under s 378(1) CPC, the tenor of the qualifying phrase is that such a confession is admissible provided certain conditions are fulfilled.

291      Clause 31(2) of the UK Bill was not repeated in s 378 CPC because Singapore already had s 30 EA before the UK Bill was introduced. Although s 30 EA is not identical in terms with cl 31(2) of the UK Bill, they have the same effect in joint trials as I understand Justice Rajah to accept. The effect is not to allow the confession of an accomplice to be introduced at the trial of an accused person unless the accomplice is jointly tried with the accused person.

292      I am aware that the select committee to which the CPC Bill was referred had suggested that Singapore should not adopt a particular provision which was found in cl 31(1) of the UK Bill. This was sub-para (v) of cl 31(1)(c) of the UK Bill which would have allowed the admission of statements by a person whose identity was known but could not be found after all reasonable steps had been taken to find him. While I accept this difference, I am of the view that it does not mean that Singapore also intended to depart from the position in the UK as regards the qualifying phrase. In my view, this difference is neutral as to what meaning is to be given to the qualifying phrase and there is no suggestion that, as regards cl 31(9) of the UK Bill, Singapore intended to depart from the position in the UK.

293      I also do not think that adopting the interpretation of the trial judge would render provisions on admissibility of confessions, like s 30 EA, otiose. I am of the view that the trial judge’s interpretation would only mean that s 30 will no longer apply when the conditions in s 378(1) CPC are met. In other conditions, s 30 EA continues to apply.

294      Was the primary reason for s 30 EA the exclusion of hearsay evidence? If so, that is not a valid reason for adopting the literal interpretation for the qualifying phrase because s 378 CPC was meant “to admit all hearsay evidence likely to be valuable to the greatest extent possible without undue complication or delay to the proceedings”. This was the first of the three purposes of the CPC Bill mentioned by Mr Barker in his speech to Parliament (see [288] above).

295      Justice Rajah notes a more plausible rationale for s 30 EA. He said at [110] of his judgment in the present case, and I reiterate:

Perhaps a more plausible rationale can be found in the need for a joint trial. Pillai in The Law of Evidence … at p 129 quotes from one of the earliest commentaries on the Indian Evidence Act by Sir Henry Cunningham, who made the following attempt:

[J]udges are relieved from attempt to perform an intellectual impossibility by a provision that, when more persons than one are tried for an offence, and one of them makes a confession affecting himself and any other of the accused, the confession may be taken into consideration against such other person as well as against the person making it. Such a confession is, of course, in the highest degree suspicious; it deserves ordinarily very little reliance; but nevertheless it is impossible for a judge to ignore it, and under the Indian Evidence Act, he need no longer pretend to do so.

296      If that is the more plausible rationale, it is no reason why s 30 EA should continue to apply to exclude the admissibility of the confession of a dead accomplice under s 378(1)(b)(i) CPC. The weight to be given to such a confession is another matter. Indeed, for the present case, I have concluded that I will give no weight to Too’s statements. However that does not mean that the confessions of a dead accomplice which incriminate an accused person will always be given no weight. It depends on the facts in each case. For example, such a confession may be useful in considering the weight to be given to another confession from the same accomplice which exonerates the accused person.

297      In the circumstances, I am of the view that the intention of Parliament was to admit the confession of an accomplice under s 378(1)(b)(i) CPC.