This judgment text has undergone conversion so that it is mobile and web-friendly. This may have created formatting or alignment issues. Please refer to the PDF copy for a print-friendly version.
|Decision Date||:||13 August 2004|
|Coram||:||Yong Pung How CJ|
|Counsel Name(s)||:||Steven Chong SC, Lionel Tan, Ian Teo and Loh Wai Yue (Rajah and Tann) for appellants; Hamidul Haq and Hui Choon Kuen (Deputy Public Prosecutors) for respondent|
|Parties||:||Ng Keng Yong — Public Prosecutor|
13 August 2004
Yong Pung How CJ:
1 These were two related appeals by the first appellant, Ng Keng Yong, and the second appellant, Chua Chue Teng, against their respective convictions for causing death by a negligent act not amounting to culpable homicide, an offence under s 304A of the Penal Code (Cap 224, 1985 Rev Ed) (“the PC”). I dismissed both appeals and now give my reasons.
2 Both the appellants are lieutenants in the Republic of Singapore Navy (“RSN”). The charges against them arose out of the tragic collision between the RSN ship RSS Courageous (“the Courageous”) and the merchant vessel ANL Indonesia (“the ANL”) on 3 January 2003, which resulted in the deaths of four RSN servicewomen.
3 The ANL was a Netherlands-registered container ship with a gross tonnage of 51,938. She was 293.5m in length and had a top speed of 24.8 knots (46km/h). The Courageous was a “Fearless Class” anti-submarine patrol vessel which displaced approximately 452 tonnes. At 57.94m, she was less than one-fifth the length of the ANL. She was powered by an advanced twin jet propulsion system, and had a top speed exceeding 20 knots (37km/h).
4 Both ships were equipped with a Gyro compass, radar, Automatic Radar Plotting Aid and other navigational equipment and charts. All the equipment was functioning at the material time.
The sequence of events
5 At about 11.20pm on 3 January 2003, the Courageous and the ANL were both travelling on the eastbound lane of the Traffic Separation Scheme (“TSS”) just off the Horsburgh Lighthouse at Pedra Branca. Both vessels were moving in a north-easterly direction in accordance with the general traffic flow. The ANL was en route from Port Klang, Malaysia, to Busan, South Korea, while the Courageous was some 6 nautical miles (“nm”) ahead of the ANL conducting her regular patrol of the waters.
6 At the material time, the first appellant was the Officer-of-the-Watch (“OOW”) on the Courageous. He was responsible for the safe navigation of the ship. The second appellant was only a trainee OOW, but she had the con, or control of the steering, of the vessel. She made navigational decisions and issued the relevant orders to the rest of the bridge team. As the supervising OOW, the first appellant would only intervene if he was of the view that any of her actions or instructions were wrong.
7 At 11.25pm, the Courageous reached the end of her patrol area. After checking that both sides were clear, the second appellant ordered the Courageous to execute a “U-turn”. This changed her course to a south-westerly direction of 235°, leading her to proceed down the eastbound lane of the TSS against the general flow of traffic.
8 Shortly after, the officer on the bridge of the ANL, Gerrit Easge Botma (“Botma”), observed the red port sidelight of the Courageous on the port side of the ANL. This suggested that the vessels would pass port to port. Meanwhile, on board the Courageous, the Closest Point of Approach (“CPA”) alarm on the radar came on. The CPA alarm was set at three cables (0.3nm), indicating that the closest distance at which the Courageous would pass the target vessel (the ANL) was about three cables.
9 Second Sergeant Goh Chun Yong, the radar plotter on the Courageous, reported the activation of the CPA alarm to the second appellant and informed her that the ANL was at “green 10” (10° on the starboard side). The lookout, First Sergeant Tan Soon Teck, and the petty OOW, Marcus Kuah Yeow Hwee, both confirmed his report. At this time, the vessels were closing in on each other at 31 knots or five cables per minute. Based on the report from the bridge team that the ANL was on the starboard side of the Courageous, the second appellant ordered an alteration to port from 235° to 215° at 11.30pm to increase the CPA. She also ordered a reduction in speed from Step 4 (1,000rpm or 13 knots) to Step 3 (860rpm or 12 knots).
10 Unfortunately, neither the appellants nor the rest of the bridge team realised that the ANL was in fact on the port side of the Courageous. The latter’s alteration to port therefore inadvertently brought the courses of the vessels even closer together. When the CPA did not open up as the second appellant had hoped, she ordered a further alteration to port from 215° to 210° at 11.31pm.
11 A minute later, Botma saw the green starboard light of the Courageous and realised that she had altered to port. As her relative bearing was not changing and the CPA was now zero, Botma alerted the Captain of the ANL, Petrus Paulaus Maria Koop (“Koop”), to the danger of collision. After a quick discussion, they decided to alter the ANL’s course to starboard in accordance with the International Regulations for Preventing Collisions at Sea 1972 (“the Collision Regulations”). The ANL made two alterations to starboard by autopilot between 11.32pm and 11.34pm, bringing her course from 51° to 73°. She also sounded a whistle blast to alert the Courageous.
12 At 11.33pm, the second appellant ordered the Courageous to make yet another alteration to port to 200°. She also ordered the speed to be increased back to Step 4. At 11.34pm, she ordered a final alteration to port to 190°, followed shortly by an order to put the helm hard to port. The first appellant also ordered both engines to “full throttle”.
13 The combination of the ANL’s alterations to starboard and the Courageous’ alterations to port resulted in the vessels colliding at 11.35pm about 1nm northeast of the Horsburgh Lighthouse. The ANL sustained superficial damage to her bow area, and none of her crew of 23 were injured. However, the Courageous sustained heavy losses. Part of her aft section was sheared off and sank. Of her 44 crewmembers, eight were injured and four died.
The inquiry and the charges
14 An inquiry was conducted under s 118 of the Merchant Shipping Act (Cap 179, 1996 Rev Ed) by investigators from the Maritime and Port Authority of Singapore (“MPA”). In their findings, the inquiry team noted that both the ANL and the Courageous “could have exercised better seamanship”. Nonetheless, they found that the collision was caused by “errors in judgment in assessing the situation and the wrongful application of the Collision Regulations” by the Courageous.
15 As the appellants were jointly responsible for negligently navigating the Courageous in an unsafe manner across the path of the ANL, they were subsequently charged under s 304A of the PC for causing the deaths of the four RSN servicewomen. The charges against them were heard together.
The Prosecution’s case
16 The Prosecution alleged that the appellants were negligent in:
(a) navigating the Courageous against the flow of traffic in the TSS;
(b) altering her course in the wrong direction; and
(c) turning to port in a series of small alterations with a decrease in speed.
Navigating against the flow of traffic
17 The 1997 RSN Operations Directive entitled “Guidelines for RSN Craft/Ships Operating in Singapore Territorial Waters” states that RSN vessels such as the Courageous should, as far as possible, comply with the Collision Regulations. At the material time, the Courageous was travelling against the general flow of traffic on the eastbound lane of the TSS. This was a technical breach of r 10(b)(i) of the Collision Regulations, which provides that:
(b) A vessel using a traffic separation scheme shall —
(i) proceed in the appropriate traffic lane in the general direction of traffic flow for that lane …
Altering course in the wrong direction
18 The Prosecution also contended that when the Courageous found herself on a reciprocal or near reciprocal course with the ANL, she ought to have altered her course to starboard in accordance with r 14(a) of the Collision Regulations, which reads:
When two power-driven vessels are meeting on reciprocal or nearly reciprocal courses so as to involve risk of collision each shall alter her course to starboard so that each shall pass on the port side of the other.
19 As the ANL was on the port side of the Courageous, an alteration to starboard in compliance with R 14(a) would have ensured that the vessels passed each other safely. By doing exactly what she should not have done, the Courageous placed herself on a particularly hazardous course.
Turning to port in a series of small alterations with a reduction in speed
20 The Courageous made a total of four alterations to port:
(a) 11.30pm: from 235° to 215° (a 20° alteration);
(b) 11.31pm: from 215° to 210° (a 5° alteration);
(c) 11.33pm: from 210° to 200° (a 10° alteration); and
(d) 11.34pm: from 200° to 190° (a 10° alteration).
21 The first alteration was accompanied by a reduction in speed from Step 4 to Step 3. The Prosecution argued that these actions violated r 8(b) of the Collision Regulations, which states that:
Any alteration of course and/or speed to avoid collision shall, if the circumstances of the case admit, be large enough to be readily apparent to another vessel observing visually or by radar; a succession of small alterations of course and/or speed should be avoided.
The Defence’s case
22 The appellants denied any negligence on their part. Instead, they claimed that the collision was caused by Botma and Koop’s negligent breach of r 8(b) of the Collision Regulations, set out above. Specifically, they argued that the ANL should have made a decisive alteration to starboard by manual steering, instead of two small alterations using her auto-pilot.
Navigating against the flow of traffic
23 The appellants acknowledged that they were technically in breach of r 10(b)(i). However, a minute of 27 January 2003 from the Office of the Chief of Navy to the MPA entitled “Compliance with COLREGS” states clearly that there can be situations where RSN vessels may be compelled to deviate from the Collision Regulations due to the nature and requirements of military missions and operational profiles. In this case, the Courageous was entitled to patrol against the flow of traffic as it placed her in a good position to reinforce the waters off Pedra Branca.
Altering course in the wrong direction
24 The appellants also asserted that r 14(a) was not applicable when they made the alterations to port as there was no risk of collision at the time. The CPA of three cables was safe, and the night order book of the Commanding Officer, Major Tang Yong Yang (“CO”), showed that previous CPAs practiced on board the Courageous were as low as two to three cables.
Turning to port in a series of small alterations with a reduction in speed
25 A Seaman’s Guide to the Rule of the Road (6th Ed, 1995) stipulates that a bold alteration is generally 30° or more. Although the appellants’ alterations fell below this standard, they argued that their initial alteration of 20° made their intention to cross ahead of the ANL readily apparent. The defence expert, Captain John Third (“DE”), also suggested that the two initial alterations should be considered as one, as they occurred within a minute of each other. This meant that the Courageous’ first alteration was 25°, a mere 5° below the standard.
The decision below
26 It is settled law that the standard of care for criminal negligence is the same as that for civil negligence: Lim Poh Eng v PP
Navigating against the flow of traffic
27 The district judge recognised that the Courageous had technically contravened r 10(b)(i) of the Collision Regulations by proceeding against the flow of traffic in the TSS. However, he accepted that this manoeuvre was safe and operationally necessary. The appellants were therefore not negligent in this respect.
Altering course in the wrong direction
28 Although the appellants were not negligent in navigating against the flow of traffic, the onus was on them to take extra precautions to avoid a close-quarters situation with other vessels proceeding in the right direction. The district judge accepted the evidence of the prosecution expert, Captain Hilbert Fernandez (“PE”), that the CPA of three cables was, in the circumstances, unsafe. Given that this was a head-on situation at night with a large vessel, the appellants should have kept a greater buffer to take into account contingencies such as radar time lag, margins of error and the ANL’s size and lack of manoeuvrability. Bearing in mind the conservative approach advocated by the Collision Regulations, the district judge found that there was a risk of collision and r 14(a) therefore applied.
29 The district judge noted that r 14(a) applied to all head-on situations where there was a risk of collision, whether the vessels were passing port to port or starboard to starboard. He found support for this proposition in Nicholas J Healey and Joseph C Sweeney, The Law of Marine Collision (1998) at 184:
[I]n the head-on situation, if any change of course is required to effect a safe passing it must be to starboard so that the vessels will pass port to port. Even if the courses are somewhat starboard of each other, but a safe starboard to starboard passing cannot be made without an alteration of course by either vessel, each must turn to starboard. This is so, even though it may mean that one vessel must cross the other’s bow in order to pass port to port.
30 Therefore, although he accepted that the appellants had perceived the ANL to be on their starboard side, he found that they were nonetheless obliged to alter to starboard. By altering to port instead, they had negligently breached r 14(a) of the Collision Regulations.
31 Turning to the question of causation, the district judge observed that both causation in fact and in law had to be established. “But for” the Courageous turning to port in breach of r 14(a), the collision would have been avoided – causation in fact was proven. The reasonable foreseeability test for causation in law was also satisfied. The district judge therefore held that the collision was caused by the appellants’ negligent breach of r 14(a).
Turning to port in a series of small alterations with a reduction in speed
32 The district judge accepted DE’s opinion that the first two alterations of the Courageous should be considered as one, as they occurred within a minute. However, the two further 10° alterations were a clear violation of r 8(b) of the Collision Regulations. Although this breach did increase the risk of collision, he found insufficient evidence to establish causation on this aspect of the appellants’ negligence.
Further negligence by the appellants
33 The district judge also held that the appellants’ failure to make light and sound signals when making the last two 10° alterations was a negligent breach of rr 34(a) and (b) of the Collision Regulations, which state as follows:
(a) When vessels are in sight of one another, a power-driven vessel underway, when manoeuvring as authorised or required by these Rules, shall indicate that manoeuvre by the following signals on her whistle:
one short blast to mean “I am altering my course to starboard”;
two short blasts to mean “I am altering my course to port” …
(b) Any vessel may supplement the whistle signals prescribed in paragraph (a) by light signals, repeated as appropriate, whilst the manoeuvre is being carried out —
(i) these light signals shall have the following significance:
one flash to mean “I am altering my course to starboard”;
two flashes to mean “I am altering my course to port” …
34 However, the Prosecution did not allege that any of this was causative of the collision. The district judge was also not prepared to hold that the appellants were negligent in:
(a) failing to take compass bearings to assess the risk of collision;
(b) failing to infer from the decreasing CPA that the ANL must be altering course to starboard; and
(c) failing to switch the radar to a six-minute vector, 3nm range scale or a relative motion setting to assess the risk of collision more accurately.
The ANL’s contributory negligence
35 In response to the Courageous’ actions, the ANL had altered her course to starboard which was the correct direction as prescribed under the Collision Regulations. However, instead of making a bold alteration by manual steering, the ANL made two smaller alterations by autopilot. As a result, the district judge found that her eventual 22° alteration in course was not large enough to be readily apparent under r 8(b) of the Collision Regulations.
36 If the ANL had made a decisive alteration to starboard, the collision could have been avoided. Both elements of causation in fact and in law were satisfied. Nonetheless, drawing from the doctrine of novus actus interveniens, the district judge took the view that the ANL’s actions were not so unreasonable as to eclipse the appellant’s initial negligence in breaching r 14(a), which indisputably caused the collision and resulted in the deaths of the four RSN servicewomen. The appellants were accordingly convicted under s 304A of the PC.
37 The appellants raised three principal issues in their appeals:
(a) whether the alteration to port by the Courageous was a negligent act in breach of r 14(a) of the Collision Regulations;
(b) whether the alteration to port was the proximate and efficient cause of the collision under s 304A of the PC; and
(c) whether the second appellant, as a trainee OOW, should be held to the same standard of care as a qualified OOW.
Whether the alteration to port was a breach of r 14(a)
38 Rule 14(a) of the Collision Regulations would have applied only if the two vessels were meeting on reciprocal or nearly reciprocal courses (ie, approaching within 6° of each other), and they were approaching “so as to involve a risk of collision” at the material time. The appellants did not dispute that before the Courageous made her first alteration to port, she was meeting the ANL on a reciprocal course. Instead, they contended that there was no risk of collision, and the district judge’s finding to the contrary was against the weight of the evidence.
39 It is trite law that an appellate court will only disturb the trial judge’s findings of fact when they are plainly wrong or against the weight of the evidence: Lim Ah Poh v PP
“So as to involve risk of collision”
40 A preliminary point must first be made. Both in the trial court and before me, the parties appeared to take the position that the phrase “so as to involve a risk of collision” in r 14(a) was equivalent to the existence of a risk of collision. Their arguments therefore centred on whether a risk of collision existed at the material time. This approach was, strictly speaking, inaccurate as a matter of law. The Collision Regulations draw a clear distinction between situations where a risk of collision exists (for example, rr 18 and 19), and where the vessels are approaching each other “so as to involve risk of collision” (for example, rr 14 and 15): see Healey and Sweeney, The Law of Marine Collision ( supra) at 126–127. Support for this proposition can also be found in John Wheeler Griffin, The American Law of Collision (1949) which states at 23:
It will be noted that the phrase is not “so that there is risk of collision,” but “so as to involve risk of collision.” The words evidently refer to a time when there is not yet actual danger, but when the relation between the vessels is such that danger may shortly arise, if the rules are not obeyed. Thus Brett, MR, in the Beryl, 9 Prob Div 137, 5 Asp 321 , said that the rules are intended “not merely for the purpose of preventing a collision, but for the purpose of preventing even a risk of collision.” And Atkin, LJ, in the Gulf of Suez,  Prob Div 318, 332, said that the phrase “to involve risk of collision … indicates a period before the risk has, so to speak, actually matured; it means a period at which there is a probability that there will be a risk of collision if precautions are not taken.” [emphasis added]
41 The real question before me was not whether a risk of collision existed at the material time, but whether the situation simply involved a risk of collision. However, this subtle recharacterisation of the issue did not undermine the district judge’s decision, since his finding that a risk of collision existed at the time necessarily implied that a risk of collision was involved to begin with. The distinction was nonetheless crucial to these appeals, since the Prosecution did not have to prove that a risk of collision existed for r 14(a) to apply. All they had to show was that the situation was such that a risk of collision could possibly arise if the vessels did not act in accordance with the regulations. I evaluated the appellants’ contentions with this in mind.
Objective test for risk of collision
42 The appellants’ main argument was the fact that none of the navigators on the ANL or the Courageous perceived that a risk of collision existed at 11.30pm, just before the Courageous’ initial alteration to port. In this regard, I agreed with the Prosecution that when considering if a risk of collision was involved, the subjective opinions of the navigators could not be decisive. The test of whether a risk of collision is involved must be an objective one. Any other approach would completely subvert the Collision Regulations’ objective of promoting certainty and safety.
43 The Collision Regulations have force of law in Singapore by virtue of the fact that they are encapsulated in the Merchant Shipping (Prevention of Collisions at Sea) Regulations (Cap 179, Rg 10) of the Merchant Shipping Act. They are universally accepted principles of safe navigation that are designed to ensure that, whenever possible, ships will not reach a close-quarters situation in which a risk of collision will arise: The “Maloja II”  1 Lloyd’s Rep 48 at 50–51. As G P Selvam J stated emphatically in The Teng He
Certainty of the situation must be the primary objective. Optimistic assumptions must be avoided: pessimistic assumptions should be preferred. These are the requirements of the Collision Regulations. These are also common sense rules of caution for the avoidance of collisions at sea. Strict adherence to these rules cannot be overemphasized because of the huge losses non-adherence to them can cause to property and human lives.
44 The conservative approach of the Collision Regulations is also borne out by the contents of the rules themselves. For example, r 7(a) provides that if there is any doubt as to whether a risk of collision exists, such risk shall be deemed to exist. Given that the Collision Regulations are meant to regulate and “de-conflict” the actions of vessels approaching one another, an objective approach must be taken when deciding if a risk of collision is involved at the material time. I could not see how the regulations could effectively serve their function if, in the same situation, they would apply to one vessel but not the other, merely because their respective navigators held different opinions as to the risk of collision. I did take into consideration the opinions of the trained navigators on board the ANL and the Courageous. However, the determination of whether a risk of collision was involved is essentially an objective exercise, and their subjective opinions had to be weighed against the rest of the evidence.
45 Dealing first with the opinions of Botma and Koop on the ANL, I noted that they had made their assessment based solely on the fact that the Courageous was showing her port sidelights, indicating that the vessels would pass port to port. There was no evidence to suggest that Botma and Koop took any further measures, such as radar plotting, to ascertain if a risk of collision was imminent. This appeared to be an overly simplistic approach, bearing in mind r 7 of the Collision Regulations:
(a) Every vessel shall use all available means appropriate to the prevailing circumstances and conditions to determine if risk of collision exists. If there is any doubt such risk shall be deemed to exist.
(b) Proper use shall be made of radar equipment if fitted and operational, including long-range scanning to obtain early warning of risk of collision and radar plotting or equivalent systematic observation of detected objects.
(c) Assumptions shall not be made on the basis of scanty information …
46 Although r 7 only applies to the determination of whether a risk of collision exists, it underscores the conservative approach that should be taken by navigators whenever a close-quarters situation appears to be developing. Botma and Koop’s opinion that there was no risk of collision was therefore undermined to a certain extent by their rather crude method of assessing the risk involved.
47 Further, Botma and Koop had estimated that the CPA at the time was four to five cables. As the ANL’s CPA alarm was not switched on, this approximation could not be verified with certainty. In contrast, the CPA alarm on the Courageous was activated and clearly indicated that the CPA was three cables. As Botma’s and Koop’s evaluation of the situation was made in the context of an overestimation of the CPA, I found that the district judge was correct to have placed little weight on their opinions.
48 As for the appellants’ claim that there was no risk of collision, this had to be placed in the context of my earlier finding that a risk of collision did not have to exist at the material time for r 14(a) to apply. The threshold for the operation of r 14(a) was lower – all that was necessary was for the situation to involve a risk of collision.
49 Also, the appellants’ views that there was no risk of collision had to be balanced against PE’s considered opinion that there was such a risk. The appellants alleged that PE was tailoring his evidence to suit the Prosecution’s case, as he only suggested that there was a risk of collision in his Supplementary Expert Report after the appellants had adduced evidence to show that there was no such risk.
50 I found this attack on PE’s credibility to be disingenuous and completely unwarranted. The only evidence that the appellants produced to suggest that PE did not initially believe that there was a risk of collision was the following quote from his first Report:
From 2325 to 2329 hours, RC [the Courageous] was fine on the port bow of AI. At this stage, they were not on a collision course and would have passed each other port to port…
51 Saying that the vessels were not on a collision course is very different from taking the view that there was no risk of collision. The Courageous and the ANL were obviously not on a collision course at the material time because the CPA was not zero cables. The real question was whether the CPA of three cables was safe, or whether the situation was such that a risk of collision would arise if proper precautions were not taken. Although PE did not use the exact phrase “risk of collision” in his first Report, a brief perusal of the contents demonstrated that he had consistently maintained that a CPA of less than five cables was risky in the circumstances. At para 29 of the Report, he opined that the CPA of three cables was “unsafe and unjustifiable … as it would not allow the OOWs on both vessels to take corrective action(s) in time …” He reiterated this at para 34:
[A] small CPA of less than 5 cables at night (especially when going against the flow of traffic in the TSS) is risky and is certainly poor seamanship… [S]uch small passing distances is [sic] unacceptable.
52 The appellants sought to dispute PE’s findings by referring to the American case of Acacia Vera Navigation Co Ltd v Kezia Ltd 1996 AMC [American Maritime Cases] 2592. In this case, the vessels Omina and Blue Cloud were passing starboard to starboard at a distance equivalent to three cables when the Blue Cloud took a sharp starboard turn, causing her to collide twice with the Omina. In giving judgment for the owners of the Omina, the district court found that there would have been a distance of at least three cables between the two vessels, “a clear and sufficient distance in all respects”: see Acacia Vera at 2601. The decision was upheld on appeal, and the appellants argued that the case supported their contention that the CPA of three cables was safe.
53 I could not agree. The court in Acacia Vera had merely held that a CPA of three cables was sufficient in the circumstances of the case. As PE highlighted in his Report, the calculation of a safe CPA is necessarily dependent on the particular circumstances of the encounter. Factors such as traffic density, state of visibility, sea condition, manoeuvring characteristics of the vessel, speed and draft have to be taken into consideration. It follows from this that the court’s finding in Acacia Vera could not be transplanted in toto to the facts of the present case.
54 In any event, the Prosecution also cited a number of cases which found a risk of collision even though the CPA was more than three cables. This further substantiated my opinion that the appellants’ reliance on Acacia Vera was misplaced, as a safe CPA has to be determined according to the circumstances of every encounter.
Whether a CPA of three cables was safe in the circumstances
55 Therefore, the facts of the present case had to be scrutinised to determine if the ANL and the Courageous were approaching each other so as to involve a risk of collision. After evaluating the evidence, I found myself in firm agreement with the district judge that the CPA of three cables was unsafe in the circumstances. In coming to this conclusion, I took the following factors into account:
(a) the Courageous was travelling against the flow of traffic;
(b) a sufficient buffer had to be given to accommodate contingencies such as engine and steering failure and unexpected last minute manoeuvres;
(c) the vessels were closing in on each other rapidly;
(d) there were margins of error on the radars of both vessels that could assume significant portions at close range;
(e) visibility was obviously poorer at night; and
(f) the Courageous had failed to use any light or sound signals at any time.
56 Although the appellants were not negligent in navigating the Courageous against the flow of traffic, it was incumbent on them to take all the necessary safety measures including the maintenance of a sufficient CPA. Considering the circumstances highlighted above, and the relative approaches of the ANL and the Courageous, I had no doubt that the situation clearly involved a risk of collision. Therefore, I agreed with the trial judge that r 14(a) applied.
57 It has been established since my decision in Lim Poh Eng ( supra) that the civil standard of care applies to criminal cases as well. Since the appellants did not dispute that an unjustified breach of the Collision Regulations would constitute negligence, and they had plainly failed to alter to starboard as required by r 14(a), I found that the Prosecution had proven its case beyond a reasonable doubt.
58 In coming to this conclusion, I did take into account the fact that entries from the CO’s night order book indicated that the Courageous had kept a CPA of as low as two to three cables on a number of nights. Given that the applicable standard of care in this case was that of a reasonable naval OOW, the appellants argued that the entries in the night order book demonstrated that they did not fall below the standard required of them, even though the CPA on the night of the accident was only three cables.
59 This contention completely misses the point I had made earlier, which is that the minimum safe CPA to be kept at any given time must be determined on a case by case basis. Without any further evidence to suggest that the circumstances involved on the dates of these entries were similar or analogous to those on the night of 3 January 2003, I failed to see how this evidence advanced the appellants’ case. Even if I accepted that the practice on these specific occasions could be applied to the night of the accident, the success of the appellant’s argument rested on the assumption that the CO’s orders on the entries were not negligent to begin with. Not having adduced any evidence to support the soundness of such an assumption, the appellants could not hope to succeed on this point.
Whether the alteration to port was the proximate and efficient cause of the collision
60 The appellants’ second ground of appeal related to the issue of causation.
Test for causation
61 The established test for causation under s 304A of the PC is set out in Lee Kim Leng v R
In the Indian case of Emperor v Omkar Ram Pratap it was held that to impose criminal liability under section 304A of the Indian Penal Code (which is similar to our section 304A of the Penal Code) it is necessary that the death should have been the direct result of a rash and negligent act of the accused and that act must have been the proximate and efficient cause without the intervention of another’s negligence. It must have been the causa causans; it is not enough that it may have been the causa sine qua non. [emphasis added]
62 Although Chua J’s words are recognised to be inaccurate in so far as they suggest that death should be the direct result of a rash and negligent act (it being clear that the provision only requires the offending act to be rash or negligent), the rest of the decision is widely acknowledged as the correct approach to causation under s 304A. The Malaysian court in PP v Joseph Chin Saiko
63 Therefore, I agreed with the appellants that the district judge erred in importing principles of causation from civil negligence, such as the “but for” test and the doctrine of novus actus interveniens, into the operation of s 304A. Although I held in Lim Poh Eng ( supra) that the standard of negligence in criminal cases should be the civil standard of negligence, this was by no means a licence for the entire law of civil negligence to be transplanted into the criminal sphere. There is no doubt that the distinction between criminal and civil liability must be maintained. In the present context, the distinction between civil and criminal negligence is reflected in both the higher standard of proof for the Prosecution and the stricter test of causation that needs to be established for criminal liability to attach.
Whether ANL’s contributory negligence broke the chain of causation
64 Adopting Chua J’s test of causation in Lee Kim Leng, the question before me was whether the appellants’ negligent alteration to port in breach of r 14(a) of the Collision Regulations constituted the proximate and efficient cause of the collision, or whether the ANL’s undisputed contributory negligence intervened to break the chain of causation.
65 According to the appellants’ interpretation of Lee Kim Leng, the chain of causation is necessarily broken whenever another party’s negligence in fact intervenes, irrespective of the parties’ relative blameworthiness. Read literally, Chua J’s statement that the accused’s act “must have been the proximate and efficient cause without the intervention of another’s negligence” might support the appellants’ argument. However, I found such an approach to be so repugnant to common sense that it had to be rejected. Taking the appellants’ argument to its logical conclusion, criminal liability under s 304A would at all times be imposed on the person who committed the most recent act of negligence, even when the negligence of another was more substantially to blame. I could not accept that the authorities led inexorably to such an arbitrary conclusion.
66 To my mind, the words of Chua J in Lee Kim Leng can be interpreted in a much more reasonable and equitable fashion. In the same case, Chua J noted further in his judgment that to impose criminal liability, “it must be found as a fact that the collision was entirely or at least mainly due to the act of the appellant”: Lee Kim Leng,  supra, at 286. Proceeding on both principle and logic, it is evident that criminal liability under s 304A should attach to the person(s) whose negligence contributed substantially, and not merely peripherally, to the result. When Chua J observed that the accused’s act should be the proximate and efficient cause of the result without the intervention of another’s negligence, he was merely emphasising the point that the accused’s negligence, and not the negligence of any other person, should have contributed significantly to the result. If he meant to suggest that the chain of causation was necessarily broken by the very fact of a third party’s intervening negligence, then, with the greatest respect, I cannot agree. The particulars of the factual matrix, and the extent to which the third party’s negligence contributed to the deaths, have to be assessed as well. The court must ultimately direct its mind to whether the negligence of the accused contributed significantly or substantially to the result.
67 On the facts, the appellants were negligent in turning to port in the face of a potential risk of collision when r 14(a) of the Collision Regulations directed them to alter to starboard instead. On her part, the ANL had turned starboard in accordance with the Collision Regulations, but was also negligent in failing to alter her course boldly and quickly. Applying the above principles of causation to the present case, the question was whether the ANL’s contributory negligence had such causative potency that the appellant’s initial negligence could not be said to have contributed significantly to the collision.
68 The appellants cited a number of authorities to support their case. In Kurban Hussein Mohamedalli Rangawalla v State of Maharashtra AIR 1965 SC 1616, seven workers were killed in a fire at a paint factory. The fire was caused by the negligence of one worker, Hatim. While manufacturing paints through a heating process, he had poured turpentine too quickly into a hot rosin mixture, causing it to overflow. The heat that was generated subsequently caused some varnish and turpentine stored nearby to catch fire. The appellant, the managing partner of the factory, was initially convicted under the Indian equivalent of s 304A for causing the workers’ deaths, as he had allowed the heating process to take place in the same room where the varnish and turpentine were stored. The Supreme Court of India allowed his appeal, holding that the deaths were not the direct or proximate result of his negligence. Instead, they held that the direct or proximate cause of the fire was Hatim’s own negligence.
69 In Lee Lai Siew v PP
70 In my opinion, the facts in both cases were clearly distinguishable from those in the present appeals. In Lee Lai Siew, there was evidence to suggest that the overloading had absolutely no impact on the safety of the ship. The court took the view that the weather and the sea were the only causes of the passengers’ deaths. In Kurban Hussein, while the appellant’s negligence in storing the varnish and turpentine in the heating room did facilitate the fire, it was Hatim’s own independent act of negligence that ignited the materials in the first place. Hatim’s negligence was in no way related to or caused by the appellant’s initial negligence.
71 In the present case, I accepted the appellants’ point that the vessels would not have collided if the ANL had not been negligent. Nevertheless, it must be emphasised that the appellants were the ones who created the situation of danger to begin with. Although I agreed with the district judge that they were not negligent in navigating against the flow of traffic due to operational exigencies, the onus was on them to take the necessary extra precautions to avoid any potential risk of collision with other vessels proceeding in the correct direction on the TSS. Far from keeping a watchful eye on the situation, the appellants made a series of alterations to port in breach of the Collision Regulations. The ANL was obliged by the appellants’ actions to respond with her own, albeit negligent, manoeuvres. To my mind, the appellants’ negligence was clearly a substantial cause of the collision.
72 As Lord Pearce opined in The “Miraflores” and the “Abadesa”  1 Lloyd’s Rep 191 at 200 (sub nom Miraflores (Owners) v George Livanos (Owners)
It is axiomatic that a person who embarks on a deliberate act of negligence should, in general, bear a greater degree of fault than one who fails to cope adequately with the resulting crisis which is thus thrust upon him … For all humans can refrain from deliberately breaking well-known safety rules; but ’tis not in mortals to command the perfect reaction to a crisis; and many fall short at times of that degree which reasonable care demands. [emphasis added]
73 This analysis of the situation is also supported by Chua J’s decision in Lee Kim Leng itself. In this case, a pedestrian was killed by the impact of the appellant’s car colliding into the rear of a taxi, which had stopped suddenly at the pedestrian crossing. The appellant was convicted under s 304A of causing the death of the pedestrian, but his conviction was overturned on appeal. Chua J found that the appellant’s negligence was not wholly or mainly responsible for the collision, as the taxi driver was initially negligent in failing to see the pedestrian at the crossing until he was almost upon her. As the taxi had stopped suddenly without giving any signals, the appellant simply had no time to react and he could not therefore be held responsible for the subsequent collision.
74 Similarly, in this case, the ANL was forced to react in response to a risk of collision created by the appellants’ own negligent actions. I could not see how the appellants could seriously argue that their negligence had not contributed significantly to the collision and the consequent deaths of the four servicewomen.
Whether second appellant should be held to same standard as a qualified OOW
75 The final ground of appeal centred on the issue of the standard of care, and related only to the second appellant’s conviction. The second appellant argued that as a trainee OOW she should not be held to the same standard as a qualified OOW, and the district judge had erred in applying the case of Nettleship v Weston
Whether Nettleship v Weston should apply
76 In Nettleship v Weston, the plaintiff was injured as a result of the defendant’s negligent driving. Although the defendant was a learner driver, the English Court of Appeal found that she had to be held to the same standard of care as that of a reasonably competent driver. A doctrine of varying standards depending on the defendant’s experience was too uncertain to be viable. Megaw LJ (who dissented only on the issue of quantum of damages) explained the rationale for the court’s decision at 707:
The disadvantages of the resulting unpredictability, uncertainty and, indeed, impossibility of arriving at fair and consistent decisions outweigh the advantages. The certainty of a general standard is preferable to the vagaries of a fluctuating standard.
77 I acknowledged the second appellant’s point that Nettleship v Weston was influenced to some extent by extraneous considerations of insurance and risk allocation, none of which were relevant in the present case. Nonetheless, I found that the approach taken by the court was correct and ought to be adopted here – any other alternative was simply too ambiguous and uncertain.
78 Even if I were minded to accept the second appellant’s argument that she should be held to a lower standard because she was still a trainee, that still begged the question of what standard the second appellant should have been held to. The second appellant argued that all she had to do was take her training seriously. Assuming that I was prepared to agree with this, I could not see any logical reason to limit such an indulgence to the facts of this case. The court in Nettleship v Weston itself recognised that its decision would inexorably extend to a much wider sphere of negligence responsibility. Megaw LJ’s comments at 593 merit reproduction here:
… [I]f the principle of varying standards is to be accepted, why should it operate, in the field of driving motor vehicles, only up to the stage of the driver qualifying for a full licence? And why should it be limited to the quality of inexperience? If the passenger knows that his driver suffers from some relevant defect, physical or temperamental, which could reasonably be expected to affect the quality of his driving, why should not the same doctrine of varying standards apply? … Logically there can be no distinction.
79 At the end of the day, the duty of care should be tailored not to the actor, but rather to the act which he or she elects to perform: per Mustill LJ in Wilsher v Essex Area Health Authority
The criminal context
80 I recognised that Nettleship v Weston dealt with civil negligence, and, as I had emphasised earlier when discussing the appropriate test for causation under s 304A, the distinction between civil and criminal liability ought to be maintained. Nonetheless, I took the view that the court’s decision in Nettleship v Weston ought to be adopted in the present appeal.
81 I laid down the definitive standard of care for criminal negligence in the case of Lim Poh Eng ( supra). In this case, the appellant was a practitioner of traditional Chinese medicine who was charged under s 338 of the PC for causing grievous hurt to a patient who had visited his clinic for colonic washouts. The Prosecution alleged that the appellant had been criminally negligent in prescribing the colonic washouts without proper training in the procedure and use of equipment, and without any understanding of the risks and complications involved. One of the appellant’s arguments on appeal was that the standard of negligence in criminal cases should be higher than the civil standard.
82 After considering the case law and various policy concerns, I rejected the appellant’s contention and held that the standard of negligence required in criminal cases was the same as that in civil cases. This followed from the local High Court’s decisions in Woo Sing v R
83 In Lim Poh Eng, the appellant had proposed an intermediate standard of negligence, lower than the standard of gross negligence for the common law offence of manslaughter but higher than the civil standard of negligence. I found such an intermediate standard to be too elusive a concept to be workable, and held that the standard of negligence in criminal cases should be the civil standard of negligence.
84 It is therefore established that the criminal standard of negligence is the same as that in civil negligence. Nettleship v Weston was a case that dealt directly with the issue of standard of care. Since I took the view that Nettleship v Weston represented the correct approach to a trainee’s standard of care in civil negligence, it necessarily applied in the present criminal context. This flows strictly from my earlier decision in Lim Poh Eng.
85 I also noted that Lord Denning MR and Megaw LJ in Nettleship v Weston drew much support for their decision from the responsibility of a learner driver in criminal law. Lord Denning MR observed at 698:
In the criminal law it is no defence for a driver to say: “I was a learner driver under instruction. I was doing my best and could not help it.” Such a plea may go to mitigation of sentence, but it does not go in exculpation of guilt. The criminal law insists that every person driving a car must attain an objective standard measured by the standard of a skilled, experienced and careful driver.
86 The second appellant contended that the strict approach in Nettleship v Weston was inappropriate for a criminal charge because “the status of a trainee is itself diametrically opposed to the concept of mens rea … [as] a trainee is still be trained to know what is right and what is wrong.” With respect, I found counsel’s argument on this point difficult to follow. Presumably, he was suggesting that for criminal liability to attach in this case, the second appellant had to subjectively know that her actions were likely to cause death or injury.
87 In my opinion, this line of reasoning completely misses the point. Section 304A does not require proof of intention or knowledge. The relevant mens rea is rashness or negligence, which was defined by Holloway J in Re Nidamarti Nagabhushanam (1872) 7 MHC [Madras High Court] 119 at 120, and cited with approval by MPH Rubin JC (as he then was) in PP v Teo Poh Leng
Culpable rashness is acting with the consciousness that the mischievous and illegal consequences may follow, but with the hope that they will not, and often with the belief that the actor has taken sufficient precaution to prevent their happening. The imputability arises from acting despite the consciousness (luxuria). Culpable negligence is acting without the consciousness that the illegal and mischievous effect will follow, but in circumstances which show that the actor has not exercised the caution incumbent upon him, and that if he had he would have had the consciousness. The imputability arises from the neglect of the civic duty of circumspection. [emphasis added]
88 It was therefore legally inaccurate for the second appellant to argue that s 304A requires an accused person to possess knowledge that his act was likely to cause death or injury. The correct approach is to consider whether a reasonable man in the same circumstances would have been aware of the likelihood of damage or injury to others resulting from such conduct: Adnan bin Khamis v PP
89 Counsel for the appellant sought to impress upon me that this decision would have wider implications on the training regime of the RSN. In particular, future trainee OOWs taking the con of vessels as part of their training programme would be held to the same standard as a qualified OOW. I recognised the potential ramifications of my finding. Yet, to my mind, there was no reason for these trainees to be held to a lower standard of care simply because allowing them to take control of naval vessels was part and parcel of the RSN’s training regime. I had to consider not just the welfare of such RSN trainees, but also the wider interests of other RSN personnel as well as the other vessels and their crews at sea. To demand any less from these trainees would unduly place the lives and property of other innocent parties at risk.
90 For the reasons above, I dismissed both appeals against conviction. As neither party appealed against sentence, and I found the respective sentences imposed on the appellants to be entirely appropriate, I chose not to disturb the district judge’s orders on this matter.
Copyright © Government of Singapore.
This judgment text has undergone conversion so that it is mobile and web-friendly. This may have created formatting or alignment issues. Please refer to the PDF copy for a print-friendly version.
Version No 0: 13 Aug 2004 (00:00 hrs)