Published date01 December 1996
Citation(1996) 8 SAcLJ 323
Date01 December 1996

It has been almost three years since the Court of Appeal in Chin Seow Noi reinterpreted section 30 of the Evidence Act to allow convictions on the sole evidence of the confession of a co-accused.1 A number of subsequent decisions have used this new doctrine and the time is perhaps ripe to assess how it has worked out in practice. I shall not belabour the concern expressed in a previous discussion that this was a risky course to take as such confessions are tainted with the triple dangers of hearsay, of accomplice evidence and of statements obtained in custody.2 The Attorney-General, speaking at the Opening of the Legal Year 1995, took quite a different tack. Although he did not attack this critique head-on, he made the powerful point that Chin Seow Noi was not really about reliability but about the erosion of an undeserved privilege against self incrimination:3

The decision in Chin Seow Noi ultimately impinges on the right to silence…there is nothing to prevent the other accused from testifying in his own defence, in which event and if he is innocent, he should have no trouble in rebutting the probative value of a statement which is both unsworn and untested by cross-examination.

Chin Seow Noi has significant implications at two crucial points in the criminal trial and they should be examined in turn. First, it allows the prosecution, at the end of its case, to cross the evidential threshold required for the court to call on the accused to make his defence on the mere adduction of the confession of a co-accused.4 That this has severe consequences for the accused is clear — he must either testify and subject himself to potentially damaging cross-examination or remain silent and have adverse inferences drawn against him.5 The practical burden, once he is called upon to make his defence, is on the accused to provide an innocent explanation of himself.

The imposition of this burden on the accused has (at least) two dimensions. There is first the constitutional dimension of the balance between the power of the state to convict the guilty and the right of the citizen not to be harassed by insufficiently substantiated charges. The very existence of an evidential barrier at the close of the prosecution’s case is reflective of

a social decision that no one should be made to explain himself unless there is evidence of guilt sufficient to justify placing such a burden (of explanation) on him. Exactly how much evidence there must be at this stage has been a matter of some debate. Although the Court of Appeal,6 unlike the Federal Court of Malaysia,7 is firmly on the side of “minimum evaluation”, there is the more specific question of whether the confession of a co-accused should trigger the burden of self explanation. Structurally, the criminal process should have in place an increasing gradation of evidential barriers. At one extreme is the low threshold “reasonable suspicion” standard for arrest which is the least that is required to enable the state to place restraints on the liberty of a citizen.8 At the other is the highest standard ever required by the criminal law, proof beyond reasonable doubt, which entitles the state to punish the citizen.9 Somewhere in between are two further barriers, that of evidential sufficiency to justify the exercise of prosecutorial discretion,10 and between this and proof beyond reasonable doubt, prima facie evidence to justify calling the accused to make his defence.11 As to what should constitute prima facie evidence is ultimately a question of political philosophy. The former law was that the confession of a co-accused was not enough. That has changed and the fear is that the right of the citizen not to be harassed by insufficiently substantiated charges is sliding ominously down the proverbial slippery slope. What next? Is the accused to be called to defend himself on only a reasonable suspicion that he has committed a crime, or worse, on the mere preferment of a charge? While it is not suggested that it has or ever will be taken this far, the progressive dilution of the standard of proof at the close of the prosecution’s case does legitimately give rise to such fears.

The Attorney-General, however, chose to proceed on the other dimension — the more obvious concern of criminal justice that the guilty be convicted and the innocent exonerated. If the accused is really innocent, he will encounter no problem in demonstrating his innocence, it is said. So long as the correct result is reached, the procedural rules are justified.12 This

brings us to the other point of the criminal trial at which Chin Seow Noi has had a significant impact — at the end of the day when all the evidence is in and the court has to make a decision to convict or acquit. The question is whether the innocent accused will always be able to challenge the confession of the co-accused effectively. Occasionally, the co-accused does take the stand and repeats the incrimination. This situation is more satisfactory — the sworn testimony of the co-accused takes the place of the out-of-court confession and, more crucially, the accused may subject the co-accused to cross-examination to test his veracity. The court is then in a better position to decide who to believe. But the accused is also faced with the possibility that the co-accused either chooses not to testify or to take the stand but retracts his incrimination of the accused. Normally, all the court is left with is the out-of-court confession of the co-accused and the contradictory testimony of the accused. The raw material for decision making is scarce and probative value exceedingly difficult to assess. It was probably a scenario like this that the older cases sought to avoid. After Chin Seow Noi the courts have had no choice but to deal with it. We shall have the opportunity to look at this major difficulty of probative value, but first, this discussion must analyse the other problem, which is of no less importance — that of the interpretation of the pre-conditions to the admissibility of the confession of the co-accused.


Theoretically, the pre-conditions to the operation of section 30 have not changed.13 Whatever interpretative problems the courts now encounter with words like “confession” and “offence”, they have had before. Nevertheless, because of the magnified significance of the admissibility of section 30 confessions, a great deal of pressure has been put on the construction and application of these requirements.

(a) “Confession”

Without a doubt, the greatest difficulty the courts have had has been with the requirement that the statement must first of all be a “confession” — that is, a confession vis-à-vis the co-accused who made the statement. “Confession” is defined in section 17(2) of the Evidence Act:14

A confession is an admission made at any time by a person accused of an offence stating or suggesting the inference that he committed the offence.

The definition is in the alternative, the first limb being that the declarant states that he committed the offence. This covers the situation where the co-accused makes express admissions as to all the elements of the crime he is charged with. The second limb has however been the source of what appears to be endless litigation. The situation which has caused difficulty is one where the declarant makes admissions as to some of the elements of the crime, but makes no express admissions as to others. The problem is simply that the definition contains no prescription as the strength of circumstantial inference which is required before it can be said that the statement suggests the inference that he committed the offence. The unchallenged exposition of this ambiguous phrase is contained in the Privy Council decision of Anandagoda,15 a case which has been approved on numerous occasions in Singapore. The Privy Council said:16

The appropriate test in deciding whether a particular statement is a confession is whether the words of admission in the context expressly or substantially admit guilt, or do they taken together in the context inferentially admit guilt.

However, this, in itself, does not take us very far. Again there is no indication of the strength of inference required. Reading the judgement more closely, we get an indication that the Privy Council had in mind a fairly strict idea of an inferential admission of guilt. Significantly, the Privy Council quotes with approval this passage from Professor Wigmore’s famous treatise:17

we conclude that whenever the statements or declarations of the accused…are of such facts as involve necessarily the commission of a crime, or in themselves constitute a crime, then the facts admitted import guilt, and such admissions may properly be denominated confessions.

Thus, where the facts admitted are not in themselves sufficient to import guilt, the facts which are admitted must by inference necessarily amount to guilt. In other words, guilt must be the only reasonable inference. To say it in yet another way, the facts which are admitted must prove the existence of the remaining elements of the offence beyond reasonable doubt.

Unfortunately, Wigmore’s idea of inferential admission of guilt seems to have escaped the attention of a number of important decisions. The result is that statements have been held to be confessions although guilt was not the only reasonable inference from the facts admitted. This happened in Tan Aik Heng.18 A summary of the facts is necessary. It was a classic drug raid. Narcotics officers found illicit drugs laid out on the floor of an

apartment belonging to a man called Foong. Foong was arrested at the lift landing. Another man, Ah Heng, was in the apartment together with a few other people. Foong gave this statement19 to the police:

Ah Heng brought the heroine to my flat for packing. The heroin does not belong to me or my wife.

Both Foong and Ah Heng were...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT