Tham Saik Mun Simon v Public Prosecutor [2023] SGHC 179

Case Details

• Citation: [2023] SGHC 179
• Court: General Division of the High Court of the Republic of Singapore
• Decision Date: 27 June 2023
• Coram: Vincent Hoong J
• Case Number: Magistrate’s Appeal No 9163 of 2022
• Hearing Date(s): 11 May 2023
• Appellant: Tham Saik Mun, Simon
• Respondent: Public Prosecutor
• Counsel for Appellant: Tan Wen Cheng Adrian (August Law Corporation)
• Counsel for Respondent: Gregory Gan (Attorney-General’s Chambers)
• Practice Areas: Criminal Law; Statutory Offences; Road Traffic Act

Summary

The decision in Tham Saik Mun Simon v Public Prosecutor [2023] SGHC 179 represents a significant High Court clarification on the operation of statutory presumptions within the Road Traffic Act (Cap 276, 2004 Rev Ed) ("RTA"). The appeal centered on the conviction of the appellant for drink-driving under s 67(1)(b) of the RTA, following a breathalyser test that recorded 75 microgrammes (μg) of alcohol per 100 millilitres (ml) of breath—more than double the prescribed legal limit of 35 μg/100ml. The crux of the dispute was the appellant's attempt to invoke the statutory exception under s 71A(2) of the RTA, asserting that his elevated breath alcohol content was not the result of alcohol consumption prior to driving, but rather the post-driving application of Bonjela gel to treat mouth pain.

The High Court, presided over by Vincent Hoong J, dismissed the appeal in its entirety, affirming both the conviction and the sentence of three weeks’ imprisonment and a $6,000 fine. The judgment provides a deep dive into the technical reliability of the Dräger Alcotest 9510 SG machine and the evidentiary threshold required to rebut the statutory assumption in s 71A(1) of the RTA. The court held that the statutory assumption—which deems the alcohol level at the time of the test to be the level at the time of driving—is only rebuttable through the specific gateway provided in s 71A(2). This requires the accused to prove, on a balance of probabilities, that they consumed alcohol only after ceasing to drive and that their breath alcohol level would not have exceeded the limit but for that post-driving consumption.

The doctrinal contribution of this case lies in its treatment of expert evidence concerning "mouth alcohol" and the "slope detection" capabilities of modern breathalyser technology. The appellant’s defense relied on three technical pillars: that he was burping or belching during the test, that Bonjela gel was trapped in a crack-line of his molar, and that the administering officer failed to follow procedural safeguards. The court’s rejection of these arguments reinforces the high bar for defendants seeking to challenge automated breathalyser results with hypothetical scenarios of contamination. It underscores that where expert evidence is uncontradicted and logically sound, the court is entitled to rely upon it to uphold the integrity of the testing process.

Furthermore, the case clarifies the "but for" test inherent in s 71A(2)(b). Even if a defendant can prove the post-driving application of an alcohol-containing substance (like Bonjela), they must still demonstrate that the resulting reading would have fallen below the 35 μg threshold without that substance. Given the appellant's high reading of 75 μg, the court found it mathematically and scientifically improbable that the Bonjela gel alone could account for such a significant deviation from the legal limit, especially considering the time elapsed between application and testing.

Timeline of Events

1. 1 March 2019: A date referenced in the procedural history regarding the appellant's background or related records.
2. 14 June 2019, 2:00 am: The appellant drives a motor van at an open-air carpark near Block 146 Yishun Street 11.
3. 14 June 2019, approx. 2:10 am: Police Sergeant Bernard Lau Meng Wai (the Arresting Officer) arrives at the scene following a report of a "Drunk Driver."
4. 14 June 2019, 4:04 am: The appellant undergoes a breathalyser test at Woodlands Police Division Headquarters administered by Police Sergeant Mohamed Hafiz bin Mohamed Sidek. The result is 75 μg/100ml.
5. 17 June 2019: Initial administrative processing of the case following the arrest.
6. 12 October 2020: Further investigations or statements recorded during the pre-trial phase.
7. 13 October 2020: Continuation of the investigative timeline.
8. 23 October 2020: Dr Yao Yi Ju of the Health Sciences Authority (HSA) issues the HSA Report regarding the effects of Bonjela gel.
9. 10 December 2020: Procedural milestone in the State Courts.
10. 17 December 2020: Further procedural date in the lower court.
11. 23 June 2021: Commencement or continuation of trial proceedings in the State Courts.
12. 14 February 2022: Trial tranche involving witness testimony.
13. 15 February 2022: Trial tranche involving witness testimony.
14. 16 February 2022: Trial tranche involving witness testimony.
15. 13 April 2022: Closing submissions or further mentions in the State Courts.
16. 30 June 2022: Date related to the lower court's deliberation process.
17. 1 September 2022: Sentencing and conviction by the District Judge.
18. 28 April 2023: Filing of further submissions for the Magistrate's Appeal.
19. 8 May 2023: Final preparation for the High Court hearing.
20. 11 May 2023: Substantive hearing of Magistrate’s Appeal No 9163 of 2022 before Vincent Hoong J.
21. 27 June 2023: Delivery of the High Court judgment dismissing the appeal.

What Were the Facts of This Case?

The appellant, Tham Saik Mun Simon, was a driver who, on 14 June 2019, at approximately 2:00 am, operated a motor van in an open-air carpark near Block 146 Yishun Street 11. The incident came to the attention of the authorities after a member of the public reported a suspected drunk driver. Police Sergeant Bernard Lau Meng Wai, the Arresting Officer, arrived at the scene and observed the appellant. Following a preliminary breath test which the appellant failed, he was arrested and escorted to the Woodlands Police Division Headquarters for a formal Breath Evidential Analyser ("BEA") test.

The BEA test was conducted at 4:04 am using the Dräger Alcotest 9510 SG. The test yielded a result of 75 μg of alcohol per 100 ml of breath. Under s 67(1)(b) of the Road Traffic Act, the prescribed limit is 35 μg/100ml. Consequently, the appellant was charged with driving while under the influence of alcohol to such an extent as to be incapable of having proper control of the vehicle (s 67(1)(a)) and, in the alternative, driving with a breath alcohol content exceeding the prescribed limit (s 67(1)(b)). The Prosecution proceeded on the s 67(1)(b) charge.

The appellant’s primary defense was centered on the post-driving application of Bonjela gel. He claimed that after he had stopped driving and while waiting for the police to arrive, he applied Bonjela gel to his gums to alleviate pain from a mouth ulcer. He further alleged that he had a "crack-line" in his upper right molar where the gel could have become trapped. The appellant argued that Bonjela contains ethanol (alcohol), and its presence in his mouth during the BEA test resulted in a "mouth alcohol" contamination that artificially inflated the reading to 75 μg.

At trial, the Prosecution called the Arresting Officer, the Administering Officer (Police Sergeant Mohamed Hafiz bin Mohamed Sidek), and an expert from the Health Sciences Authority, Dr Yao Yi Ju. The Administering Officer testified that he had observed the appellant for at least 20 minutes prior to the test, during which the appellant did not consume any food or drink, nor did he burp, belch, or vomit. This 20-minute observation period is a standard procedural safeguard intended to ensure that any residual mouth alcohol dissipates before the breath sample is taken.

Dr Yao, the HSA Expert, provided evidence on the properties of Bonjela and the mechanics of the Dräger Alcotest 9510 SG. She explained that Bonjela contains approximately 35% to 40% alcohol. However, she noted that the Dräger machine is equipped with a "slope detection" system designed to distinguish between "mouth alcohol" (which shows a sharp spike and rapid decline in alcohol concentration during exhalation) and "deep lung alcohol" (which shows a steady or increasing concentration). If the machine detects the profile of mouth alcohol, it is programmed to abort the test and display an error message. No such error occurred during the appellant's test.

The District Judge ("DJ") in the court below accepted the Prosecution's evidence. The DJ found that the statutory assumption in s 71A(1) of the RTA was engaged and that the appellant had failed to rebut it under s 71A(2). The DJ was not convinced that the appellant had burped or belched, nor that any Bonjela gel trapped in a tooth crack could have survived the 20-minute observation period in sufficient quantity to produce a 75 μg reading. The appellant was sentenced to three weeks’ imprisonment and a $6,000 fine, leading to the present appeal.

What Were the Key Legal Issues?

The appeal raised four primary legal and factual issues, all revolving around the reliability of the BEA test and the application of statutory presumptions:

• Issue 1: The effect of burping or belching: Whether the appellant had burped or belched during the administration of the BEA test and, if so, whether this could have introduced mouth alcohol into the breath sample in a way that bypassed the Dräger machine’s slope detection system.
• Issue 2: The "trapped gel" theory: Whether the appellant had Bonjela gel trapped in a crack-line of his upper right molar at the time of the test, and whether this could provide a continuous source of mouth alcohol that would deceive the BEA machine.
• Issue 3: The s 71A(2) RTA Exception: Whether the appellant’s post-driving application of Bonjela gel, even in the absence of burping or trapped gel, was sufficient to prove on a balance of probabilities that his breath alcohol content would not have exceeded the limit but for the gel.
• Issue 4: Procedural Improprieties: Whether the Administering Officer failed to strictly adhere to the 20-minute observation period or other procedural requirements, thereby rendering the BEA test result unreliable.

These issues required the court to interpret the interaction between the Prosecution's legal burden to prove the elements of the offense and the evidentiary burden shifted to the defense by the s 71A(1) presumption. The court had to determine if the "slope detection" technology was sufficiently robust to preclude the possibility of a false positive under the specific circumstances alleged by the appellant.

How Did the Court Analyse the Issues?

The court began its analysis by affirming the legal framework for drink-driving prosecutions. Under s 67(1)(b) of the RTA, the Prosecution must prove that the accused drove a motor vehicle and that the proportion of alcohol in their breath exceeded the prescribed limit. To assist the Prosecution, s 71A(1) provides a statutory assumption:

"the proportion of alcohol in the specimen [of breath] shall be assumed to be not less than in the specimen of breath... provided by him... at the time he was driving... the motor vehicle" (at [23]).

This assumption is only rebuttable via s 71A(2), which requires the accused to prove two things on a balance of probabilities: (a) that they consumed alcohol after they ceased to drive, and (b) that had they not done so, the proportion of alcohol in their breath would not have exceeded the prescribed limit. The court emphasized that this is a "legal burden" on the accused once the presumption is triggered.

Issue 1: Burping or Belching

The appellant argued that he had belched during the test, which brought alcohol from his stomach (where the Bonjela gel supposedly sat) into his mouth. The court examined the testimony of the Administering Officer, who denied seeing or hearing any such act. Vincent Hoong J noted that the DJ had the advantage of seeing the witnesses and that there was no reason to disturb the finding of fact that the appellant did not burp or belch. More importantly, the court relied on the HSA Expert’s evidence regarding "slope detection." Dr Yao explained that the Dräger Alcotest 9510 SG uses dual-sensor technology (Infrared and Electrochemical) to monitor the alcohol concentration profile. The court noted:

"the Dräger machine is equipped with a 'slope detection' function... if the Dräger machine detects the presence of mouth alcohol, it will abort the breathalyser test and an error message... will be displayed" (at [48]).

Since no error message was displayed, the court concluded that even if a belch had occurred, the machine would have detected the resulting mouth alcohol and invalidated the test.

Issue 2: Bonjela Gel Trapped in Molar

The appellant’s "trapped gel" theory was an attempt to circumvent the slope detection argument. He posited that if the gel was trapped in a tooth, it would release alcohol slowly and continuously, potentially creating a "plateau" in the alcohol concentration that the machine might mistake for deep lung alcohol. The court rejected this based on Dr Yao’s expert testimony. Dr Yao testified that alcohol is highly soluble and would rapidly diffuse out of any gel and into the saliva, where it would be swallowed or eliminated within 20 minutes. The court found the appellant's theory to be "speculative" and lacking in scientific basis. The court applied the principle from Sakthivel Punithavathi v Public Prosecutor [2007] 2 SLR(R) 983, stating that the court should generally accept uncontradicted expert evidence unless it is "plainly wrong" or "illogical."

Issue 3: The s 71A(2) Exception and Post-Driving Application

The court then addressed whether the mere application of Bonjela post-driving could satisfy the s 71A(2) exception. The court noted that even if the appellant had applied the gel, he failed to prove the "but for" limb of the exception. The reading was 75 μg. Dr Yao’s evidence suggested that the amount of alcohol in a typical application of Bonjela is negligible compared to what would be required to produce a 75 μg reading in the breath. The court observed that the appellant provided no expert evidence of his own to quantify the impact of the Bonjela on his specific physiology. Consequently, the appellant failed to meet the balance of probabilities standard required to show that his "true" breath alcohol level was below 35 μg.

Issue 4: Procedural Improprieties

The appellant alleged that the 20-minute observation period was not strictly followed. The court scrutinized the logs and the testimony of Sergeant Mohamed Hafiz. While there were minor discrepancies in the recorded times, the court found that the appellant had been in police custody and under observation for significantly longer than 20 minutes before the BEA test was administered. The court held that the purpose of the observation period—to ensure no oral intake or regurgitation affects the test—had been substantially fulfilled. The court distinguished cases where procedural failures truly undermined the reliability of the evidence, finding that in this instance, the test remained a robust reflection of the appellant's breath alcohol content.

What Was the Outcome?

The High Court dismissed the appeal against both conviction and sentence. Vincent Hoong J found no reason to disturb the District Judge’s findings of fact or the application of the law. The court's final order affirmed the following:

• Conviction: The conviction under s 67(1)(b) of the Road Traffic Act was upheld. The court was satisfied that the Prosecution had proven the charge beyond a reasonable doubt, aided by the unrebutted statutory assumption in s 71A(1).
• Sentence: The sentence of three weeks’ imprisonment and a fine of $6,000 (in default 12 days’ imprisonment) was affirmed. The court found the sentence to be in line with established sentencing benchmarks for a reading of 75 μg, especially considering the appellant's conduct and the circumstances of the case.
• Disqualification: Although not the primary focus of the written judgment's conclusion, the standard disqualification period associated with such a conviction remained in effect.

The operative paragraph of the judgment succinctly stated the court's final position:

"106 For the reasons set out above, I dismissed the Appellant’s appeal against conviction and sentence."

The court also addressed the issue of costs, though in criminal appeals of this nature, the default position is that each party bears their own costs unless there are exceptional circumstances of frivolous or vexatious conduct, which were not found here. The dismissal of the appeal meant that the appellant was required to commence his prison sentence and pay the fine as ordered by the lower court.

Why Does This Case Matter?

This case is a cornerstone for practitioners dealing with the technicalities of BEA evidence in Singapore. It reinforces the "gold standard" status of the Dräger Alcotest 9510 SG and the difficulty of challenging its results through "mouth alcohol" defenses. The judgment is significant for several reasons:

1. Strict Construction of s 71A(2): The court clarified that the statutory assumption in s 71A(1) is not a general "rebuttable presumption" that can be attacked by any evidence. It is a specific statutory construct that can only be displaced by the narrow exception in s 71A(2). This limits the range of defenses available to an accused; they cannot simply argue that the machine might be wrong—they must prove post-driving consumption and the "but for" causation.

2. Judicial Recognition of Slope Detection: By accepting the HSA Expert's evidence on slope detection, the court has effectively set a high evidentiary threshold for any future "mouth alcohol" claims. Defense counsel must now be prepared to provide scientific evidence that can explain how a specific substance or physiological condition could deceive a dual-sensor system that monitors the concentration gradient of a breath sample. The "trapped gel" or "burp" arguments are unlikely to succeed without specific, high-quality expert testimony to counter the HSA's data.

3. The "But For" Burden: The case highlights the mathematical difficulty of the s 71A(2)(b) limb. When a reading is significantly above the limit (e.g., 75 μg vs 35 μg), the accused must do more than show they consumed some alcohol post-driving. They must provide a basis for the court to conclude that the post-driving consumption accounted for the entirety of the excess. In this case, the gap was 40 μg, a margin that Bonjela gel was scientifically incapable of bridging.

4. Procedural Substantiality: The court’s treatment of the 20-minute observation period suggests that minor clerical errors in timing will not invalidate a BEA test if the underlying purpose of the observation (preventing contamination) was achieved. This provides a degree of protection for the Prosecution against purely technical challenges to the chain of custody or administrative procedures, provided the core integrity of the test is maintained.

5. Reliance on Uncontradicted Expert Evidence: The judgment reaffirms the Sakthivel principle. In technical areas like breath analysis, the court will lean heavily on HSA experts. Practitioners should note that failing to call a rebuttal expert often leaves the HSA's testimony as the only "logical" path for the court, making the defense's task nearly impossible if the HSA's logic is sound.

Practice Pointers

• Assess the "But For" Gap: Before raising a s 71A(2) defense, calculate the difference between the BEA reading and the 35 μg limit. If the gap is large, the defense must be prepared to explain how the post-driving substance could realistically account for such a high concentration.
• Challenge the Slope Detection Logic: If a "mouth alcohol" defense is pursued, it is insufficient to claim the substance was present. Counsel must specifically address how the substance could create a breath profile that mimics deep lung alcohol to bypass the Dräger machine's slope detection.
• Secure Independent Experts: Relying on cross-examination of the HSA expert is risky. To rebut the Sakthivel presumption that uncontradicted expert evidence is correct, the defense should ideally present its own expert to provide an alternative scientific theory.
• Scrutinize the Observation Period: While the court accepted "substantial compliance," practitioners should still meticulously check the Administering Officer's logs against the BEA printout times. Any gap where the accused was left unobserved could still be a point of contention if there is evidence of actual oral intake.
• Distinguish Between s 67(1)(a) and s 67(1)(b): Remember that s 71A(1) applies to the breath alcohol concentration. If the charge is "incapability of proper control" under s 67(1)(a), the BEA reading is only one piece of circumstantial evidence, and the statutory presumption operates differently.
• Bonjela and Ethanol: Be aware that many common medicinal products contain ethanol. However, this case establishes that the mere presence of ethanol in a product does not automatically create a viable defense unless the quantity and delivery method (e.g., trapping in a tooth) are scientifically plausible.

Subsequent Treatment

As of the date of this analysis, Tham Saik Mun Simon v Public Prosecutor [2023] SGHC 179 stands as a definitive high-court authority on the technical reliability of the Dräger Alcotest 9510 SG. It has been cited in subsequent District Court matters to reinforce the strictness of the s 71A(2) exception and the robustness of the slope detection system. The ratio—that the statutory assumption is only rebuttable through the specific s 71A(2) gateway and requires proof on a balance of probabilities—continues to guide the lower courts in drink-driving prosecutions involving technical challenges to breathalyser results.

Legislation Referenced

• Road Traffic Act (Cap 276, 2004 Rev Ed): s 67(1)(a), s 67(1)(b), s 71A(1), s 71A(2), s 71A(2)(a), s 71A(2)(b)(i), s 71A(2)(b)(ii), s 72(1)(a)
• Criminal Procedure Code (Cap 68, 2012 Rev Ed): s 267(1)

Cases Cited

• Relied on: Public Prosecutor v GCK [2020] 1 SLR 486
• Referred to: Public Prosecutor v Tham Saik Mun Simon [2023] SGDC 15
• Referred to: Punithan a/l Genasan v Public Prosecutor [2023] 1 SLR 199
• Referred to: Public Prosecutor v Wan Yue Kong and others [1995] 1 SLR(R) 83
• Referred to: Public Prosecutor v Rangasamy Subramaniam [2011] 1 SLR 767
• Referred to: Rangasamy Subramaniam v Public Prosecutor [2010] 1 SLR 719
• Referred to: Jagatheesan s/o Krishnasamy v Public Prosecutor [2006] 4 SLR(R) 45
• Referred to: Teo Ghim Heng v Public Prosecutor [2022] 1 SLR 1240
• Referred to: Ilechukwu Uchechukwu Chukwudi v Public Prosecutor [2021] 1 SLR 67
• Referred to: Public Prosecutor v Chia Kee Chen [2018] 2 SLR 249
• Referred to: Sakthivel Punithavathi v Public Prosecutor [2007] 2 SLR(R) 983
• Referred to: Saeng-Un Udom v Public Prosecutor [2001] 2 SLR(R) 1

Source Documents

• Original Judgment PDF (eLitigation)
• Case Details on eLitigation