Lim Boon Keong v Public Prosecutor

Steven Chong J:

[1] This is an appeal against the decision of the district judge in [2009] SGDC 511 convicting the appellant of one charge of consuming a controlled drug, viz norketamine, in contravention of s 8(b)(i) of the Misuse of Drugs Act (Cap 185, 2008 Rev Ed) (“the Act”).

[2] In the course of this appeal, the parties had initially raised a number of points of law of general importance which had hitherto not been considered in a reported decision of our courts. First, whether the testing of the appellant’s urine samples, as described in the trial below, complied with the requirements laid down in s 31(4)(b) of the Act, so as to give rise to the presumption of consumption in s 22. Second, whether the presumption in s 16 relating to analysts’ certificates could apply to prove that the appellant’s urine samples contained norketamine if the s 22 presumption did not apply (because the s 31(4)(b) requirements were not complied with). Third, whether the appellant’s confession per se was of any evidential value in deciding whether he in fact took norketamine.

[3] In light of the prosecution’s confinement of its case, which I will elaborate upon below, it has become unnecessary for me to make findings on all these issues. However, it is plain that the above issues are pertinent to the proper administration of our drug laws and the investigation of drug offences. I would therefore deal fully with all the points of law which have been raised, in addition to making the findings necessary to dispose of this appeal.

[4] The appellant, Mr Lim Boon Keong, claimed trial to the following charge:

[5] On 4 February 2008, at about 11.50 am, the appellant was arrested at No 4 Lorong 22 Geylang in the course of a raid by a group of officers from the Criminal Investigations Department of the Singapore Police Force. He was brought to the Police Cantonment Complex for further investigations, and a specimen of urine was taken from him, which was divided into three bottles. The urine in one of the bottles was tested by the Instant Urine Test Machine. The other two bottles were sealed in the presence of the appellant and sent to the Health Sciences Authority (“HSA”) for further testing. After the testing and further investigations, the above charge was preferred against the appellant on 20 March 2008.

[6] At the trial below, the prosecution produced two certificates purported to be signed by two analysts from the HSA. The first certificate was dated 25 February 2008 and was purported to be signed by one Kuan Soo Yan (“Ms Kuan”). It stated that, on analysis, the urine sample of the appellant was found to contain norketamine at a concentration of 7640 nanograms per millilitre. The second certificate was dated 27 February 2008 and was signed by one Tan Joo Chin (“Ms Tan”). It stated that, on analysis, the urine sample of the appellant was found to contain norketamine at a concentration of 6630 nanograms per millilitre. Both Ms Kuan and Ms Tan were analysts with the Illicit Drugs and Toxicology Division of the HSA at the material time. A large part of the trial concerned whether the appellant’s urine samples were tested in accordance with s 31(4)(b) of the Act. In this connection, the prosecution called Ms Tan to testify; it was unable to call Ms Kuan because she had emigrated to the United Kingdom. The prosecution also called Dr Lui Chi Pang (“Dr Lui”), the director of the Illicit Drugs and Toxicology Division, to testify generally on the urine testing procedures employed by the HSA. I note here that the certificates used “sample” while s 31(4) refers to “specimen”. Nothing turns on the difference in wording, but for convenience I will use sample since this was the term used by the witnesses in giving evidence.

[7] The prosecution also produced the statement made by the appellant on 20 March 2008 after the notice required by s 122(6) of the Criminal Procedure Code (Cap 68, 1985 Rev Ed) was served on him. The appellant’s cautioned statement read in its entirety as follows:

The caution was made and the cautioned statement recorded by Sergeant Amos Yap Hon Chian (“Sgt Yap”), the investigating officer for the appellant’s case, who was called as a prosecution witness. It was not in dispute that, prior to recording the appellant’s cautioned statement, Sgt Yap had recorded a “long statement” from the appellant pursuant to s 121 of the Criminal Procedure Code. It was also not in dispute that the long statement recorded the appellant’s purported explanation that the norketamine was found in his urine because he mistakenly took some sips of a drink from a glass placed on the table at the Geylang premises where he was arrested. The appellant’s counsel, Mr S K Kumar (“Mr Kumar”), applied for the long statement but the district judge ruled that the appellant was not entitled to it. The district judge also ruled after a voir dire that the appellant’s cautioned statement, quoted above, was admissible.

[8] When called upon by the district judge to enter his defence at the close of the prosecution’s case, the appellant elected to remain silent and did not call any witness in his defence.

[9] After considering the evidence, the district judge determined that norketamine was found in the appellant’s urine as a result of both urine tests conducted under s 31(4)(b) of the Act, thus giving rise to the presumption under s 22 that the appellant had consumed the norketamine in contravention of s 8(b). In the alternative, the district judge admitted Ms Kuan’s certificate as proof of its contents, ie that the appellant’s urine sample contained norketamine, via s 16 of the Act. Separately, the district judge found that appellant’s refusal to give evidence in his defence gave rise to an adverse inference that the confession in the appellant’s cautioned statement was true and reliable. In the circumstances, the district judge found the appellant guilty of the charge and convicted him accordingly. As the appellant had a previous conviction under s 8(b)(i) of the Act, he was liable for enhanced punishment under s 33(4), and the district judge accordingly sentenced him to the minimum three years’ imprisonment.

[10] Specifically, the district judge found that the tests of the appellant’s urine samples complied with s 31(4)(b) because:

(a) Both Ms Tan and Ms Kuan had independently reviewed and certified the test results for each of the appellant’s urine samples. They were also personally responsible for their respective certificates. They can therefore be said to have had conduct of the test of the respective urine samples of the appellant.

(b) Regulation 5(2) of the Misuse of Drugs (Urine Specimens and Urine Tests) Regulations (Cap 185, Rg 6, 1999 Rev Ed) (“the Regulations”) has been satisfied because the two urine samples of the appellant were tested by different laboratory officers.

(c) It was immaterial that analysts and laboratory officers, including Ms Tan and Ms Kuan, had supervised the testing of the two urine samples interchangeably. Section 31(4)(b) only required an independent review of the results, and this was satisfied here.

[11] In the same proceedings, the appellant also pleaded guilty to a charge under s 7 of the Common Gaming Houses Act (Cap 49, 1985 Rev Ed), and a charge under s 267B of the Penal Code (Cap 224, 2008 Rev Ed). Mr Kumar informed me that fines were imposed in respect of these charges and have since been paid.

[12] Before going into the substance, it is first necessary to set out in some detail the rather curious course which this appeal has taken.

[13] When this appeal first came before me, the following issues were presented for decision:

(a) Whether norketamine was found in the appellant’s urine samples as a result of both urine tests conducted under section 31(4)(b) of the Act, thus giving rise to the presumption in s 22;

(b) If s 22 does not apply, whether the presumption in s 16 can be relied upon to prove that the appellant’s urine contained norketamine;

(c) Whether the appellant can be convicted on the basis of his confession and/or the drawing of an adverse inference from his silence at trial.

As can readily be seen, this involved a reconsideration of all aspects of the district judge’s judgment.

[14] After hearing the parties’ oral submissions on 12 March 2010, I directed the parties to address me on four specific points:

(a) Whether reg 5(2) of the Regulations is ultra vires s 31(4)(b) of the Act;

(b) Whether the presumption under s 16 can arise if the presumption under s 22 does not arise as a result of non-compliance with s 31(4)(b) (this was an issue from the beginning but in my view the parties had not addressed it satisfactorily);

(c) Whether there exist precedents where an accused person had been convicted of an offence of consumption solely on the strength of his confession (and in particular without any certification by a HSA analyst);

(d) The differences in qualification between a HSA analyst and a HSA laboratory officer.

[15] On 4 May 2010 following receipt of the further submissions, I directed the parties to address me on four additional points:

(b) Can the nature of a controlled drug be proven on the basis of a bare admission alone? If not, what further proof is required?

(c) Who bears the burden of showing that an accused person’s admission as to the nature of a controlled drug is not sufficiently probative?

(d) Is the appellant precluded by his stance in the proceedings below (both in the voir dire and the main trial) from arguing now that he was not familiar with the controlled drug norketamine?

I also referred the parties to some cases in this area which I had identified through my own research.

[16] The parties addressed these points fully in their further oral submissions on 6 May 2010. On 10 June 2010, more than a month after the parties made their further oral submissions, the Public Prosecutor sent a letter to the Registry of the Supreme Court, the relevant parts of which read as follows:

[17] On receipt of the Public Prosecutor’s letter, I directed the parties to appear before me to seek further clarifications arising from the letter. On 15 June 2010, I heard the parties during which the prosecution clarified the following points:

(a) The prosecution confirmed that it was withdrawing its reliance on the presumptions because the evidence adduced below for this case was insufficient for the presumptions to apply.

(b) The withdrawal applies to both presumptions under s 16 and s 22 which were relied on by the prosecution and accepted by the district judge below.

(c) The prosecution is relying on the appellant’s confession only for the purpose of establishing the mens rea and not for the actus reus of consumption.

(d) The prosecution will not be making further arguments on the remaining live issues and will rest its case on the submissions already made.

[18] I must say that this is a most extraordinary turn of events. With all due respect, this is not an entirely satisfactory manner of managing a case which raises points of such general importance. In any case, given the prosecution’s concession that the evidence adduced below was insufficient to give rise to the presumptions and the consequent withdrawal of its reliance on the presumption in s 22, it is strictly unnecessary for me to make specific findings as to whether the appellant’s urine samples were tested in accordance with s 31(4)(b). For obvious reasons, Mr Kumar confirmed that the appellant had no objection to the prosecution’s change of position.

[19] For the avoidance of doubt, in the light of this development, the district judge’s decision is not to be regarded as authority for the proposition that the HSA’s urine testing methodology, as described in the trial below, is in compliance with s 31(4)(b). The point is to be regarded as undecided.

[20] However, since the parties had fully argued the relevant points of law, which are of general importance, I shall express my views on them, in addition to addressing the issues necessary to dispose of this appeal, as they stand after the letter from the Public Prosecutor and the clarifications made on 15 June 2010.

[21] I first consider the urine testing procedure in s 31(4)(b) and the presumption in s 22 which it is related to. Section 22 of the Act provides that:

Section 31(4) of the Act provides that:

[22] In order to understand the requirements of s 31(4)(b) and its relation to s 22, it is essential to examine the relevant history of the Act. For convenience, I will refer to the various Misuse of Drugs (Amendment) Acts simply by their Act number and year.

[23] There was no presumption of consumption or prescribed urine testing procedure when the Act was first enacted as Act 5 of 1973. The presumption of consumption was first introduced as s 19A by Act 49 of 1975. Section 19A, which is the progenitor of s 22, provided as follows:

[24] While s 19A referred to a urine test, there was no procedure prescribed in the Act on how such a test was to be carried out. A procedure was first introduced later, as new subsections (4) and (5) to s 28, by Act 12 of 1977. Those subsections provided as follows:

[25] It is pertinent to note that s 28(4) and (5) were amendments introduced by the Minister at the committee stage of the Bill, as a result of “a further study of the Bill [which] was made with a view to providing safeguards for any possible abuse or errors by enforcement and other officers in the taking and handling of urine samples and in their analysis”. Specifically, s 28(4) and (5) were introduced to (Singapore Parliamentary Debates, Official Report (9 November 1997) vol 37 at col 171):

It should be emphasised that, under the 1977 amendments, the second test is optional, and need not be performed by a different person.

[26] The safeguards provided by s 28(4) and (5) were substantially fortified by Act 38 of 1989, which deleted the subsections (by then renumbered as s 31(4) and (5) in the 1985 Reprint) as they were enacted in 1977 and replaced them with an entirely new procedure, as follows:

The explanatory statement to the Bill explained that the new procedure:

[27] In the Second Reading of the Bill, the Minister for Home Affairs similarly explained (Singapore Parliamentary Debates, Official Report (30 November 1989) vol 54 at col 865) that:

[28] It is apparent that the 1989 procedure introduced for the first time:

It is also pertinent to note that, as a result of this new regime, a consequential amendment was made to the presumption of consumption in s 19A (renumbered as s 22 in the 1985 Reprint). The amended presumption read as follows:

[29] The 1989 procedure was further augmented by Act 2 of 2006 which introduced the preliminary urine test in s 31(4)(a) in addition to the two substantive tests already required. The preliminary urine test does not by itself trigger any legal consequences, and nothing more need be said about it. After the enactment of the 2006 amendments, s 31(4) attained its present form.

[30] It is clear from this survey of legislative history that Parliament had, since 1977, laid down specific criteria in relation to the testing of urine samples for controlled drugs with the intention that they should operate as safeguards against error. It was in furtherance of this policy that the 1989 amendments, prescribing an approach where urine samples will be independently tested by two different chemists, were enacted. The importance of the safeguards introduced by the 1989 amendments is reflected in the fact that the bar for triggering the presumption of consumption in s 22 was amended to peg it to the new procedure. If I may say so, the safeguards which have been continually reinforced by Parliament are a very necessary part of the Act, given the serious consequences which follow from positive urine tests done in accordance with s 31(4)(b). In this regard, it is to be pointed out that positive tests not only trigger the s 22 presumption of consumption; separately, they also empower the Director of the Central Narcotics Bureau to detain the person supplying the urine specimen at an approved institution for treatment and/or rehabilitation, without first having to obtain a court order: see s 34 of the Act. It is also pertinent to refer to Mr Amarjit Singh’s note “Legal Aspects Of Substance Abuse” [1989] 1 SAcLJ 36, reproducing his talk by the same title at the Medico Legal Seminar on Scientific Evidence in Drugs and other Substances of Abuse. Mr Singh pointed out some of the documented errors in urine testing, including:

[31] The source cited by Mr Singh for these errors was dated at the time of his note, and Mr Singh’s note is itself dated. But the safeguards introduced in 1989 remain the basis of today’s s 31(4)(b) in an area where the law is frequently updated. This no doubt reflects Parliament’s view that they remain vital and necessary. It is in this light that s 31(4)(b) must be interpreted.

[32] It is next necessary to examine the legislative history of the Act to understand how Parliament identifies the persons it considers qualified to conduct the urine tests.

[33] The 1989 procedure, quoted above, required the urine tests to be conducted by “Government chemists”. That designation was changed to “a Scientific Officer in the Department of Scientific Services (including the Director and Deputy Director of Scientific Services)” by Act 20 of 1998. The explanatory statement to the Bill describes the change as a “technical amendment” to reflect the changed titles of Government chemists. In order to emphasise the technical nature of the amendment, s 26(2) of Act 20 of 1998 provided that:

[34] The designation of Scientific Officer in the Department of Scientific Services was in turn replaced with “an analyst employed by the Health Sciences Authority” as a result of consequential amendments made when the Health Sciences Authority Act (Act 4 of 2001; now Cap 122C, 2002 Rev Ed) was enacted. As the Minister for Health explained in the Second Reading of the Bill, the purpose of the Heath Sciences Authority Act was to integrate a number of departments in the Ministry of Health which provide various health sciences expertise and services. The integrated departments included the Forensic Science Division of the Department of Scientific Services (in turn part of the Institute of Science and Forensic Management), which provided a range of scientific expertise to the police for the investigation of crimes, eg toxicology. See Singapore Parliamentary Debates, Official Report (22 February 2001) vol 72 at cols 1487-1488. The technical nature of the re-designation was again emphasised by a savings clause, viz s 42 of the Health Sciences Authority Act:

[35] Finally, Act 2 of 2006 further amended s 31(4)(b) to provide that, in addition to an analyst of the HSA, the urine tests may be conducted by any other person as the Minister may, by notification in the Gazette, appoint for such purpose. The Deputy Prime Minister and Minister for Home Affairs explained in the Second Reading of the Bill that this amendment was to give the Central Narcotics Bureau the option of utilising laboratories other than the Health Sciences Authority for drug analyses: Singapore Parliamentary Debates, Official Report (16 January 2006) vol 80 at cols 2100-2101.

[36] It seems, therefore, that the Act has consistently referred to the titles and designations of the relevant government chemists, as they were known from time to time, in describing the person authorised to perform the urine tests for the purpose of s 31(4)(b) and its predecessors.

[37] In light of the foregoing survey of the legislative history, and also in light of the actual wording of s 31(4)(b), the following requirements must in my judgment be satisfied before it can be presumed that a controlled drug is found in the urine of an accused person as a result of both urine tests conducted under s 31(4)(b).

[38] First, both urine tests must be conducted by persons authorised by the Act, ie HSA analysts or gazetted persons. Tests conducted by any other person will not give rise to the s 22 presumption. For convenience I will express the following requirements with respect to the conduct of the tests by a HSA analyst, but the same requirements would apply to any person gazetted by the Minister for the purpose of s 31(4)(b).

[39] Secondly, both urine tests must be conducted by the HSA analysts who eventually certify the presence of the controlled drug. It is obvious that, if the HSA analyst did not conduct the urine test which found the controlled drug, he or she cannot certify its presence. This requirement was not disputed by the parties; however, they disputed the meaning of the word “conduct”. On this point I am in general agreement with the district judge that it is both necessary and sufficient for the analyst to supervise the testing process. It is strictly not necessary for the analyst to physically conduct the actual tests. After some argument Mr Kumar also conceded this point. However, I cannot agree with the district judge’s holding that an analyst can be said to have conduct of the test if he or she only reviews the test results, as opposed to actually supervising the test process. The district judge appears to have reached this view after referring to the definition of the word “conduct” in the Reader’s Digest publication, Use The Right Word: Modern Guide To Synonyms And Related Words (1979) at p 428, as follows:

With respect, there seems nothing in this definition that supports the district judge’s finding that review qualifies as conduct. In any event there is in my view a vital difference between review and supervision. In reviewing the test results and related documentation, the analyst does not directly perform the tests, or even observe the performance of the tests. He or she is wholly dependent on what is recorded by the persons who actually performed or supervised the testing process. He or she is therefore unable to detect any error which is not recorded or which cannot be detected from such records. This, in my view, would be plainly insufficient to constitute compliance with s 31(4)(b), in light of its purpose as a safeguard against error. The analyst must actually supervise the test process before he or she can be said to have conducted it. I would add that, as a general rule, the degree of supervision must be such that the analyst is able to claim responsibility for the whole testing process and authorship of the certificate consequently issued. In more practical terms, I would say that, at the minimum, the parts of the testing process which are currently supervised ought to be supervised by an analyst, and in particular the analyst who eventually certifies the presence of the controlled drug.

[40] Thirdly, the entire conduct of both urine tests must be done independently of each other. This means that the personnel involved in the testing of one urine sample cannot be involved in any way at all in the testing of the other urine sample. This applies equally to the actual physical testing as well as to supervision and review. I should emphasise that in applying this requirement the court will not be concerned with nice arguments about whether the personnel involved in the testing of one sample actually relied upon or were influenced by the personnel involved in the testing of the other sample – as amply shown by the survey of the legislative history, the absolute independence of the two tests from each other is the cornerstone of the regime in s 31(4)(b). Separately, I should refer to reg 5 of the Regulations:

I would observe that reg 5(2) refers to “officer” and not “analyst”, which is the designation referred to in s 31(4)(b). As mentioned, I invited submissions from the parties whether reg 5(2) was ultra vires s 31(4)(b) of the Act. In this regard, I agreed with Mr David Khoo (“Mr Khoo”) for the prosecution that the status of reg 5(2) is immaterial if ultimately s 31(4)(b) is complied with. However, I would point out that, to the extent that “officer” is not the same as “analyst”, compliance with reg 5(2) will not necessarily translate into compliance with s 31(4)(b). The choice of the word “officer” in reg 5(2) is unfortunate as it may lead to confusion as to the correct designation that is authorised under the Act to conduct the tests. In the course of hearing the appeal, I requested the parties to find out the qualifications of an analyst and a laboratory officer. They informed me that the current entry requirement for an analyst is a Bachelor of Science degree in chemistry, biochemistry or related fields, with Second Class (Upper Division) Honours. A laboratory officer would need to have a diploma in chemical process technology or equivalent. The material difference in the entry qualification of an analyst as compared to that of a laboratory officer is consistent with Parliament’s objective in assigning the task for the conducting of the urine tests only to analysts or other persons gazetted for such purpose.

[41] I take note of Mr Khoo’s submission that there are only nine analysts and eighteen laboratory officers in Illicit Drugs and Toxicology Division of the HSA. I have no doubt that they perform an important, demanding and quite possibly thankless role in the war against drugs. However, in the final analysis the court’s role is to apply the requirements of s 31(4)(b) strictly, and in this it cannot and will not be deflected by considerations of what is convenient or expedient. Those considerations can only be addressed by the Legislature or the Executive – the Minister may, for example, gazette other qualified persons, in addition to HSA analysts, to conduct the testing of urine samples.

[42] I should also remark that the HSA is not only a scientific agency; it is a statutory body with legal responsibilities, and in particular its expert opinion is intended by Parliament to carry weight in a court of law. In light of its statutory responsibilities, the HSA should continually review its procedures, in particular the protocol set out in its Narcotics II Laboratory Quality Manual (the “HSA manual”), parts of which were placed in evidence in the trial below, to ensure that they accurately reflect the legal regime under which it operates. It cannot be over-emphasised that whether the appellant’s urine samples are tested in accordance with s 31(4)(b) is to be determined with reference to the language, purpose and history of the subsection, and not with reference to the HSA manual. Similarly, the fact that HSA’s testing protocols comply with international scientific standards, a point which was raised by the prosecution both here and below, does not mean that they comply with s 31(4)(b).

[43] I would reiterate that nothing which I have said is to be taken as deciding whether or not the HSA’s urine testing methodology, as described in the court below, complies with s 31(4)(b).

[44] I turn next to consider the issue of whether, in cases where the procedure in s 31(4)(b) has been shown (or, as here, conceded) not to have been complied with, the presumption in s 16 can nonetheless apply to prove that a controlled drug is found in the urine samples of an accused person.

[45] I would reject at the outset Mr Khoo’s submission that s 16 is a provision on admissibility only. This is plainly incorrect. Section 16, besides providing for the admissibility of the certificate of the HSA analyst, further goes on to say that such a certificate, “on its production by the prosecution and without proof of signature and, until the contrary is proved, shall be proof of all matters contained therein.”

[46] Mr Khoo further submitted that there was no inconsistency between the s 16 and s 22 presumptions because they relate to separate matters. Section 22 creates a presumption that the accused person consumed the controlled drug found in his urine in contravention of s 8(b). This presumption extends to the fact of consumption, and the mental element of intention to consume. Section 16 only creates a presumption as to the truth of the contents of the analyst’s certificate, and does not expressly touch on whether the accused person consumed the controlled drug found in his urine, or did so intentionally.

[47] This is a nice argument but must ultimately fail. The presumptions in ss 16 and 22 are admittedly different in scope. However, they overlap substantially with each other in the sense that they are both capable of leading to the conclusion that an accused person committed the actus reus of consuming the controlled drug found in his urine. Section 22 does so by directly presuming the commission of the offence of consumption. Section 16 does not presume any element of the offence but comes very close to doing so, in the sense that a presumption that a controlled drug is found in the urine sample of an accused person would, bar cases of alleged contamination or tampering, almost inescapably lead to an inference that the actus reus of consumption is made out. At the same time, the conditions triggering the presumption contained in each section are quite different: s 22 requires the controlled drug to be found as a result of both urine tests conducted under s 31(4)(b), while s 16 only requires an analyst’s certificate that a controlled drug is found in the accused person’s urine. Given this functional overlap between the two presumptions, if the court accepts that the presence of a controlled drug in a urine sample can be prima facie proven by a s 16 certificate in circumstances where s 31(4)(b) had not been complied with, the safeguards in the latter provision will be altogether obviated. Given that (as Mr Khoo has acknowledged) there are other applications of s 16 outside of urine testing, I am therefore of the view that a court should not accept a s 16 certificate as presumptive proof that a controlled drug is found in an accused person’s urine if the urine tests were not carried out in compliance with s 31(4)(b).

[48] Before moving on, I should reiterate that, in this case, there was evidence led as to whether s 31(4)(b) was complied with, and in the end the prosecution conceded that the evidence led was insufficient to prove compliance. In the circumstances, there was quite naturally no arguments made as to whether, in the absence of any actual proof either way, s 16 can be used to certify compliance with s 31(4)(b) and thereby trigger the presumption in s 22. There are arguments for and against such an interpretation. It could be argued that, if an analyst’s certificate can prove the presence of a controlled drug in a urine sample, it must also be able to prove the method by which the controlled drug is found. On the other hand, there is the judicial reluctance to read presumptions cumulatively in criminal cases. Given that the issue was not raised before me, it would be inappropriate for me to express my views on the matter, save to note that this is an issue which remains to be resolved, either in a future case or by Act of Parliament.

[49] I turn now to consider the remaining live issues after the prosecution’s withdrawal of its reliance on the presumptions in ss 16 and 22. The non-applicability of the presumptions does not ipso facto mean that the offence has not been proven – it is still open to the prosecution to rely on other evidence led at trial to show that norketamine was found in the appellant’s urine sample as a result of a process that is scientifically reliable. Of course, since the prosecution does not rely on any presumption, it will have to prove its case beyond a reasonable doubt. Although strictly speaking such a route to secure conviction is possible, I was informed by the prosecution that there is no known case to the best of their knowledge of an accused person being convicted for a consumption charge in the absence of certificates issued by analysts from the HSA.

[50] On the facts, not much evidence appears to have been adduced as to the presence of norketamine in the appellant’s urine sample other than the two certificates issued by the two analysts. The testimony of Ms Tan, the only witness before the court who actually involved in the testing of the appellant’s urine, was focussed on clarifying the urine testing methodology employed by the HSA generally and in this case specifically. There was no testimony as to how the methodology led to a conclusion that the appellant’s urine samples contained norketamine, which is the fact the prosecution needed to prove. This was perhaps unsurprising since the focus on the trial was on the applicability of the presumptions. The only scientific evidence apart from that led in connection with the presumptions was two reports entitled “NARCOTICS II LABORATORY, Quantitation of KETAMINE in urine”. These reports were produced before the district judge in the course of Dr Lui’s re-examination, but their contents were not explained by Dr Lui, who was in any case not in a position to do so since they were not prepared by him. At that point, Ms Tan had already been released as a witness, having completed her evidence. The reports formed part of the record which was transmitted to me on appeal, but they are highly technical in nature and I am unable, without expert assistance, to derive any assistance from them. During the hearing of the appeal, I sought clarification from the prosecution on the reports because counsel for the appellant had accepted their authenticity and accuracy in the court below. Before me, Mr Khoo confirmed, very properly in the circumstances, that the prosecution was not relying on the two reports as proof that the appellant had consumed norketamine. This position was impliedly reaffirmed in para 4(a) of the Public Prosecutor’s letter, which only referred to the evidence of Ms Tan and Dr Lui in support of its case on the actus reus of the charge.

[51] I am therefore unable to say that there was scientific evidence which proved beyond a reasonable doubt that norketamine was found in the appellant’s urine samples. Given that the prosecution had decided to confine its case on the actus reus of the charge to the scientific evidence, this would mean that it has failed to make out an essential element of the charge.

[52] For completeness, I would also say that the prosecution would have failed to make out the actus reus even if it had maintained its original reliance on the confession in the appellant’s cautioned statement and his silence at trial to prove that he had in fact consumed norketamine. For convenience I set out the very short confession again:

As mentioned, the district judge had ruled that the prosecution need not produce the appellant’s long statement. This ruling was not challenged before me.

[53] In assessing the evidential value of the appellant’s confession, I begin with the statement in Ratanlal & Dhirajlal‘s The Law of Evidence (Wadhwa and Co, 22 Ed, 2006) at p 322 that:

[54] In support of this proposition, the learned authors referred to Comptroller of Customs v Western Lectric Co Ltd [1966] AC 367, where the Privy Council held that an admission as to the origin of certain goods, made upon reading of marks and labels on those goods, was of no more evidential value than those marks and labels themselves.

[55] More on point with the present facts is R v Chatwood [1980] 1 WLR 874, a decision of the English Court of Appeal. In Chatwood the accused persons were convicted for possessing controlled drugs when the only evidence as to the accused persons’ consumption of heroin was their admission after caution to that effect. They appealed on the ground that this did not constitute sufficient prima facie evidence of the nature of the substances involved, and therefore should not have been sent to the jury. The judgment of the court referred to R v Wells [1976] Crim LR 518, a decision of the English Court of Appeal, and Bird v Adams [1972] Crim LR 174, a decision of the Queen’s Bench Divisional Court. The actual judgments are not reported and the original transcripts are no longer available. I am also unwilling to rely on the summaries in the Criminal Law Review, especially since it was pointed out in Chatwood itself that some of these summaries were inaccurate. But the court in Chatwood appears to have had access to the relevant transcripts, and I think it is useful to refer to some of the passages quoted in that case. The first passage I would refer to is that by Ormrod LJ in R v Wells, quoted in Chatwood at 879:

[56] The second passage is from Bird v Adams. According to the court in Chatwood (at 877), the appellant in that case had been charged with unlawful possession of a controlled drug. The only evidence as to the nature of the substance which he was found to possess was his own admission that it was LSD. The court in Chatwood quoted (at 877–878) and approved (at 879) the following passage from the judgment of Lord Widgery CJ in Bird v Adams:

[57] On the facts (which were not discussed at length), the court in Chatwood found (at 879) that the accused persons “were expressing an opinion, and an informed opinion, that, having used the substance which they did use, it was indeed heroin, because they were experienced in the effects of heroin” (emphasis added). It then held that (at 880) that the statements of the accused persons were sufficient prima facie evidence of the identity of the substances involved, and accordingly their convictions were upheld.

[58] The facts in the above cases can be contrasted to those in R v Dillon (1983) 2 Qd R 627, a decision of the Court of Criminal Appeal of Queensland. In that case, there were admissions made to a police officer and evidence given at trial which showed that the accused person was supplying what he thought to be marijuana (spelt as marihuana in the judgment), but evidence was not led to show the degree of familiarity which he might have had with the drug. Matthews J (with whom Kneipp and Derrington JJ concurred) approved the test laid down by Neasey J in the unreported case of Harris v Pandava (10 June 1975):

Applying this test, Matthews J found that there was simply no evidence to support a finding that the accused person was sufficiently familiar with marijuana to enable him (and a witness) to say to the required satisfaction of the magistrate below that indeed he had been dealing with marijuana. Accordingly his conviction was set aside.

[59] In addition to these cases, there are also a handful of Commonwealth authorities which are helpfully referred to in the judgment of Phillip J, sitting in the Court of Criminal Appeal of Victoria, in Reardon v Baker [1987] VR 887 at 892–893. They include Anglim and Cooke v Thomas [1974] VR 363; Police v Coward [1976] 2 NZLR 86; and Parks v Bullock [1982] VR 258. The facts are various but the underlying principle is the same. The nature of a substance can in principle be proved by the admission of the accused person, even though he is not an expert as such. However, before an accused person’s admission as to the nature of a substance can be of evidential value, it must be shown that the accused person has sufficient knowledge of or familiarity with the substance which he claims to identify. Absent such proof, the admission would be worthless. This general proposition was not challenged by Mr Khoo.

[60] A stricter approach seems to have been taken in R v Lang and Evans (Inspector of Police) [1977] Crim LR 286. According to the summary in the Criminal Law Review (at 287), Judge Robin David QC held in that case that “In future it will now only be possible to secure a conviction for possessing cannabis by an analyst’s report.” In the absence of any report of the learned judge’s actual reasoning, I am with respect unable to treat the case as persuasive authority. The holding of the learned judge, as stated in the Criminal Law Review, also goes against the authorities I have just reviewed, and in principle I cannot see why the nature of a drug cannot be proved by the admission of an accused person who is proven to have sufficient knowledge of or familiarity with the drug. Of course, whether there is sufficient knowledge or familiarity is a question of fact to be decided with reference to the particular accused person and controlled drug in question.

[61] It is also clear to me that the prosecution bears the legal burden of proving that the accused person in making the admission had the requisite personal knowledge or familiarity with the drug in question. This is entirely in line with the prosecution’s burden to prove its case beyond any reasonable doubt.

[62] On the facts, I can readily accept that the appellant’s confession, together with his refusal to give evidence in his own defence, may prove a number of things that can ordinarily be expected to be within his knowledge. It may prove that the appellant accepted a substance called norketamine. It may prove that the appellant did not doubt that the substance is norketamine. It may prove that the appellant intentionally consumed the substance thinking it was norketamine. All this would go a long way, if not the full distance, towards proving the mens rea for s 8(b).

[63] However, there are considerable difficulties in saying that the appellant’s confession and his silence at trial proved beyond all reasonable doubt that the substance consumed by him was, in fact, norketamine. There was no evidence whatsoever that the appellant was capable of identifying norketamine, an ability which thankfully is not so ordinary that it can be assumed without specific proof. His previous convictions related to ketamine or nimetazepam, not norketamine, and no evidence was led by the prosecution that ketamine and nimetazepan are similar to norketamine. Moreover, it is not disputed that, during the taking of his long statement, the appellant had stated that he mistakenly took a few sips of water from a glass on the table in the Geylang premises where he was arrested. The appellant only confessed when the investigating officer, Sgt Yap, informed him that norketamine had been found in his urine after tests by the HSA. Here I refer to Sgt Yap’s cross-examination:

It is clear from this exchange that Sgt Yap must have at some point prior to the taking of the cautioned statement showed the appellant his positive urine test results, and reiterated the fact when taking the cautioned statement. I also find it curious that the appellant confessed his guilt in the course of being cautioned, the whole purpose of which is for him to state any defence he might have. In these circumstances, I do not think that an adverse inference as to the appellant’s knowledge of or familiarity with norketamine can safely be drawn from his silence at trial, and more generally I am not able to say that the appellant had sufficient knowledge of and familiarity with norketamine such that his confession is proof beyond reasonable doubt that he had in fact consumed norketamine. Here I note that, in abandoning its original reliance on the confession as proof of the actus reus, the prosecution appears to have arrived at the same conclusion.

[64] Mr Khoo had argued that if an accused person could plead guilty to the offence he could equally confess to it. The short answer to this argument is that it places the cart before the horse – the court would not accept a plea of guilty if it was not sure that the accused person was capable of knowing what he was pleading to.

[65] Before concluding I should say that, while it is in principle possible for the nature of a drug to be proven by admission alone, it is prudent in my view for the prosecution of a consumption offence to always be supported by expert evidence or certification, since an expert opinion would be the best evidence of the nature of the substance involved, and would obviate the need for an extended inquiry into the accused person’s knowledge of or familiarity with the relevant drug. Therefore, if the adduction of expert evidence or certification is already the practice – and to the credit of the prosecution and the police this appears to be the case – it is most desirable that it should continue to be so. Of course, there is the all-important proviso that any certification must be in strict compliance with the requirements imposed by the Act.

[66] For all these reasons, I would allow the appeal and set aside the appellant’s conviction under s 8(b)(i) of the Act. Notwithstanding the late change of position by the prosecution, I should record that I derived considerable assistance from the able submissions made by Mr Khoo and Mr Kumar. I am especially grateful that they confined themselves to the crux of the dispute and did not pursue frivolous points.

[67] Given that the prosecution’s conduct of this case has been less than perfect, some might take the view that the appellant was “in fact” guilty but has “escaped” on “technicalities”. In response to those who might hold such views, I can do no better than to repeat what was said by V K Rajah JA in XP v Public Prosecutor [2008] 4 SLR(R) 686 at [98]:

While Rajah JA was writing in a context where no presumptions were applicable, his observations apply with equal force to cases where the prosecution either fails to prove the conditions precedent for the presumptions on which it could have relied, or where, as here, it elects to withdraw its reliance on such presumptions altogether.